EVALUATION OF A RECOMBINANT RIFT VALLEY FEVER VIRUS NUCLEOCAPSID PROTEIN AS A VACCINE AND AN IMMUNODIAGNOSTIC REAGENT Petrus Jansen van Vuren A thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg, October 2011 Promoter: Prof. Janusz T. Paweska Co-promoter: Prof. Caroline T. Tiemessen Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page ii P. Jansen van Vuren DECLARATION I, Petrus Jansen van Vuren declare that this thesis is my own work. It is being submitted for the degree of Doctor of Philosophy in the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination at this or any other university. _______________________________ Petrus Jansen van Vuren ______ day of ______________, _________ Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page iii P. Jansen van Vuren For my loving wife Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page iv P. Jansen van Vuren PUBLICATIONS AND PRESENTATIONS ARISING FROM THE THESIS Publications Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2011). Anti-nucleocapsid protein immune responses counteract pathogenic effects of Rift Valley fever virus infection in mice. PLoS One 6(9): e25027 Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2010). Evaluation of a recombinant Rift Valley fever virus subunit nucleocapsid protein as an immunogen in mice and sheep. Open Vaccine Journal 3, 114-126 Van Vuren, P. J. & Paweska, J. T. (2010). Comparison of enzyme-linked immunosorbent assay-based techniques for the detection of antibody to Rift Valley fever virus in thermochemically inactivated sheep sera. Vector-Borne and Zoonotic Diseases 10, 697-9. Paweska, J.T., van Vuren, P.J., Kemp, A., Swanepoel, R., Buss, P., Bengis, R.G., Gakuya, F., Breimann, R.F. and Njenga, K. (2010). A recombinant nucleocapsid-based indirect ELISA for serodiagnosis of Rift Valley fever in African wildlife. In Symposium Proceedings: Odongo, N.E., Garcia, M. And Viljoen, G.J. (eds). Sustainable Improvement of Animal Production and Health. Vienna, Austria, 8 ? 11 June 2009. Food and Agriculture Organization of the United Nations: FAO/IAEA, 309-12. ISBN 978-92-5-106697-3. Jansen van Vuren, P. & Paweska, J. T. (2009). Laboratory safe detection of nucleocapsid protein of Rift Valley fever virus in human and animal specimens by a sandwich ELISA. Journal of Virological Methods 157, 15-24. Paweska, J. T., van Vuren, P. J., Kemp, A., Buss, P., Bengis, R. G., Gakuya, F., Breiman, R. F., Njenga, M. K. & Swanepoel, R. (2008). Recombinant nucleocapsid-based ELISA for detection of IgG antibody to Rift Valley fever virus in African buffalo. Veterinary Microbiology 127, 21-8. Paweska, J. T., Jansen van Vuren, P., and Swanepoel, R. (2007). Validation of an indirect ELISA based on a recombinant nucleocapsid protein of Rift Valley fever virus for the detection of IgG antibody in humans. Journal of Virological Methods 146(1-2), 119-24. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page v P. Jansen van Vuren Presentations Jansen van Vuren, P., Tiemessen, C.T., Paweska, J.T. (2011). Aberrant host gene expression in liver and spleen of mice during acute Rift Valley fever virus infection. Virology Africa 2011, Cape Town, South Africa, 29 November ? 2 December 2011. (Oral presentation). Jansen van Vuren, P., Tiemessen, C.T., Paweska, J.T. (2011). Rift Valley fever pathogenesis in a mouse model ? host gene expression in liver of mice immunized with a recombinant nucleocapsid protein. 30th World Veterinary Congress, Cape Town, South Africa, 10-14 October 2011. (Oral presentation). Jansen van Vuren, P., Tiemessen, C.T., Paweska, J.T. (2011). Anti-nucleocapsid immune responses counteract the immunopathologic effects of severe Rift Valley fever virus infection in mice. ARBO-ZOONET third annual meeting, St. Raphael, France, 5 October 2011. (Poster presentation). Jansen van Vuren, P., Tiemessen, C.T., Paweska, J.T. (2011). Evaluation of a recombinant RVFV nucleocapsid protein as a recombinant vaccine and immunodiagnostic reagent. National Institute for Communicable Diseases Research Forum, Johannesburg, South Africa, 23 March 2011. (Oral presentation). Jansen van Vuren, P., Kemp, A., Le Roux, C., Grobbelaar, A., Leman, P., Weyer, J., Paweska, J.T. (2010). The 2010 Rift Valley fever outbreak in humans in South Africa: A laboratory perspective. Annual meeting of the ARBO-ZOONET, Rabat, Morocco, 22-24 November 2010. (Oral presentation). Jansen van Vuren, P., Kemp, A., Le Roux, C., Grobbelaar, A., Leman, P., Weyer, J., Paweska, J.T. (2010). The 2010 Rift Valley fever outbreak in South Africa from a laboratory perspective. University of the Witwatersrand, Faculty of Health Sciences research day, Johannesburg, South Africa, 22 September 2010. (Oral presentation). Paweska, J.T., van Vuren, P.J., Kemp, A., Swanepoel, R., Buss, P., Bengis, R., Gakuya, F., Breiman, R., & Njenga, M.K. (2009). Serodiagnosis of Rift Valley fever in African wildlife using a recombinant nucleocapsid-based indirect ELISA. FAO/IAEA International Symposium on Sustainable Improvement of Animal Production and Health, Vienna, Austria, 8 ? 11 June 2009. (Oral presentation) Jansen van Vuren, P., Paweska, J.T. (2009). A comparative evaluation of ELISA-based techniques for serodiagnosis of Rift Valley fever. Annual meeting of the ARBO-ZOONET network, St. Raphael, France, 30 September 2009. (Oral presentation). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page vi P. Jansen van Vuren Jansen van Vuren, P., Paweska, J.T. (2009). Comparison of ELISA-based techniques for serodiagnosis of Rift Valley fever. 5th European Meeting on Viral Zoonoses, St. Raphael, France, 26 ? 29 September 2009. (Poster presentation). Jansen van Vuren, P. & Paweska, J.T. (2009). Safe detection of Rift Valley fever virus in human and animal specimens by a sandwich ELISA. International Meeting on Emerging Diseases and Surveillance, Vienna, Austria, 13 ? 16 February 2009. (Poster presentation). Jansen van Vuren, P. & Paweska, J.T. (2009). Preliminary evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as an immunogen in combination with different adjuvants in mice and sheep. FAO/IAEA International Symposium on Sustainable Improvement of Animal Production and Health, Vienna, Austria, 8 ? 11 June 2009. (Poster presentation). Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2009). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine immunogen in combination with four adjuvants. International Meeting on Emerging Diseases and Surveillance, Vienna, Austria, 13 ? 16 February 2009. (Poster presentation). Jansen van Vuren, P. & Paweska, J.T. (2008). A safe antigen detection ELISA for rapid diagnosis of Rift Valley fever. University of the Witwatersrand, Faculty of Health Sciences research day, Johannesburg, South Africa, 20 August 2008. (Oral presentation). Jansen van Vuren, P. & Paweska, J.T. (2008). Laboratory safe detection of nucleocapsid protein of Rift Valley fever virus in human and animal specimens by a sandwich ELISA. National Institute for Communicable Diseases Academic Day, 11 November 2008. (Poster presentation). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page vii P. Jansen van Vuren ABSTRACT The serodiagnosis of Rift Valley fever (RVF) relies on the use of inactivated whole virus based reagents which present biosafety, financial and operational constraints. There are no vaccines for humans, the availability of animal vaccines is limited and they have several drawbacks. The aim of this study was to evaluate a bacterially expressed recombinant RVF virus (RVFV) nucleocapsid protein (recNP) as a safe immunodiagnostic reagent, and an immunogen in a mouse and host animal model. Several enzyme-linked immunosorbent assays (ELISAs) were developed in this study, enabling sensitive and specific detection of antibodies and RVFV antigen in human and animal specimens. The recNP was combined with different adjuvants and used to immunize mice and sheep subsequently challenged with a virulent wild type RVFV strain. Depending on the recNP/adjuvant combination, protection against disease in mice ranged between 17 and 100%, with sterilizing immunity elicited in some experimental groups, compared to 100% morbidity/mortality and excessive viral replication in adjuvant and PBS control mice. Immunization with recNP combined with Alhydrogel, an adjuvant that biases immunity towards Th2 humoral immunity, that yielded 100% protection, induced an earlier and stronger type I interferon response in mice after challenge, compared to repression of the same gene in adjuvant and PBS control mice. There was massive activation of pro-inflammatory responses and genes with pro-apoptotic effects in the livers of control mice at the acute phase of infection, accompanied by high viral replication, possibly contributing to the pathology of the liver. There was also evidence of activation and repression of several genes involved in activation of B- and T-cell immunity in control mice, some indicating possible immune evasion by the challenge virus. Immunization of sheep with the same recNP/adjuvant combinations were, however, not able to decrease replication of challenge virus. The recNP based ELISAs are an important addition to and improvement of the currently available serodiagnostic tests for RVF. The mechanism by which recNP immunization protects mice from developing severe disease during the acute phase of infection is now better understood, but the mechanism for earlier clearance of the virus needs further investigation. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page viii P. Jansen van Vuren ACKNOWLEDGEMENTS I would like to thank the following persons for their assistance, advice and contributions to specific aspects of this project: ? Professor Janusz Paweska for being a patient and supporting supervisor who guided me through this entire process of starting off my research career. ? Professor Caroline Tiemessen for guidance and advice as a co-supervisor. ? The National Health Laboratory Service for granting me a one year bursary to complete this study towards my PhD. ? The Poliomyelitis Research Foundation for partial funding of the project (PRF grant number 08/14) ? The National Institute for Communicable Diseases for granting me the opportunity to utilize their infrastructure for completion of this study. ? The Special Pathogens Unit staff members for their continued input and assistance throughout the project. ? Mrs. Busi Mogodi and her team of animal care workers for their valuable assistance with the experiments involving animal work. ? The South African Vaccine Producers for their assistance with the immunization experiments. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page ix P. Jansen van Vuren TABLE OF CONTENTS Page DECLARATION ii PUBLICATIONS AND PRESENTATIONS iv ABSTRACT vii ACKNOWLEDGEMENTS viii TABLE OF CONTENTS ix LIST OF FIGURES xii LIST OF TABLES xiv CHAPTER ONE 1 INTRODUCTION AND LITERATURE REVIEW 15 1.1 Virus classification and characteristics 15 1.2 Epidemiology 17 1.3 Rift Valley fever in animals 18 1.4 Rift Valley fever in humans 19 1.5 Pathogenesis 19 1.6 Diagnostic techniques 22 1.6.1 Virus / antigen detection 22 1.6.2 Molecular biology 23 1.6.3 Serology 23 1.7 Vaccines and antivirals 24 1.7.1 Inactivated virus vaccines 25 1.7.2 Attenuated virus vaccines 25 1.7.3 Recombinant viruses by reverse genetics 26 1.7.4 DNA vaccines 27 1.7.5 Virus vectored / replicon vaccines 28 1.7.6 Virus like particles as vaccines 29 1.7.7 Recombinant subunit vaccines 30 1.8 Study objectives 31 CHAPTER TWO 2 RECOMBINANT NUCLEOCAPSID PROTEIN AS IMMUNODIAGNOSTIC REAGENT ? SEROLOGY 33 2.1 Detection of IgG antibody to Rift Valley fever virus in wild ruminants 33 2.1.1 Introduction 33 2.1.2 Materials and methods 34 2.1.3 Results 37 2.1.4 Discussion 42 2.2 Conjugation of the recombinant nucleocapsid protein with horseradish peroxidase for detection of IgM antibody to Rift Valley fever virus in humans 44 2.2.1 Introduction 44 2.2.2 Materials and methods 44 2.2.3 Results 45 2.2.4 Discussion 49 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page x P. Jansen van Vuren Page 2.3 Comparative evaluation of ELISA-based techniques for detection of antibodies against RVFV 51 2.3.1 Introduction 51 2.3.2 Materials and methods 52 2.3.3 Results 55 2.3.4 Discussion 56 2.4 Detection of IgG antibody to Rift Valley fever virus in humans 57 2.4.1 Introduction 57 2.4.2 Materials and methods 57 2.4.3 Results 58 2.4.4 Discussion 63 CHAPTER THREE 3 EVALUATION OF ANTI-RECOMBINANT NUCLEOCAPSID PROTEIN RABBIT AND SHEEP POLYCLONAL SERA AS IMMUNODIAGNOSTIC REAGENTS IN AN ANTIGEN DETECTION SANDWICH ELISA 65 3.1 Introduction 65 3.2 Materials and methods 66 3.2.1 Generation of rabbit and sheep hyperimmune sera against the RVFV recNP 66 3.2.2 Sandwich ELISA procedure 66 3.2.3 Inactivation of specimens and safety testing 67 3.2.4 Antigen detection in animal specimens 67 3.2.5 Antigen detection in human specimens 68 3.2.6 Monitoring viral growth in vitro 68 3.2.7 Antigen detection in decomposing tissues 69 3.2.8 ELISA performance, cut-off selection and IQC 69 3.3 Results 71 3.4 Discussion 80 CHAPTER FOUR 4 EVALUATION OF A RECOMBINANT NUCLEOCAPSID PROTEIN AS AN IMMUNOGEN IN A MOUSE MODEL 84 4.1 Introduction 84 4.2 Immunogenicity of the recombinant nucleocapsid protein alone and in combination with four adjuvants 86 4.2.1 Materials and methods 86 4.2.2 Results 87 4.2.3 Discussion 92 4.3 In vivo neutralizing ability of anti-nucleocapsid immune sera in mice 93 4.3.1 Materials and methods 93 4.3.2 Results 94 4.3.3 Discussion 95 4.4 Rift Valley fever virus challenge of mice immunized with the recombinant nucleocapsid protein 96 4.4.1 Materials and methods 96 4.4.2 Results 97 4.4.3 Discussion 102 4.5 Conclusion 104 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page xi P. Jansen van Vuren CHAPTER FIVE Page 5 RECOMBINANT NUCLEOCAPSID PROTEIN AS AN IMMUNOGEN IN SHEEP 105 5.1 Introduction 105 5.2 Immunogenicity of the recombinant nucleocapsid protein alone and in combination with four adjuvants in sheep 106 5.2.1 Materials and methods 106 5.2.2 Results 107 5.2.3 Discussion 108 5.3 Rift Valley fever virus challenge of immunized sheep 109 5.3.1 Materials and methods 109 5.3.2 Results 110 5.3.3 Discussion 113 5.4 Conclusion 115 CHAPTER SIX 6 HOST GENE EXPRESSION IN MICE IMMUNIZED WITH RECOMBINANT NUCLEOCAPSID PROTEIN AND CONTROL MICE AFTER RIFT VALLEY FEVER VIRUS INFECTION 116 6.1 Introduction 116 6.2 Materials and methods 120 6.2.1 Immunization and Rift Valley fever virus challenge of mice 120 6.2.2 Measuring up- and downregulation of genes using qRT-PCR 122 6.3 Results 126 6.4 Discussion 145 CHAPTER SEVEN 7 CONCLUSIONS 151 APPENDIX 1 156 Ethics approvals APPENDIX 2 162 Reprints of publications originating from this study REFERENCES 223 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page xii P. Jansen van Vuren LIST OF FIGURES Page Chapter one Figure 1.1. Rift Valley fever virus genome 15 Figure 1.2. Rift Valley fever virus structure 16 Chapter two Figure 2.1.1. Upper and lower internal quality limits for PP values of high positive, low-positive, negative serum and conjugate controls in recNP-based IgG I-ELISA 37 Figure 2.1.2. Dose response curves of individual African buffalo sera in the recNP I-ELISA 38 Figure 2.1.3. Dose response curves of wildlife sera in the recNP I-ELISA 39 Figure 2.1.4. Optimisation of the cut-off for the recNP I-ELISA in African buffalo using the misclassification cost term option of the two-graph receiver operating characteristic analysis 40 Figure 2.1.5. Distribution of recNP IgG I-ELISA PP values in African buffalo tested positive and negative in the virus neutralization test 42 Figure 2.2.1. Titration curves of three different recNP-HRP preparations with strong positive human RVF IgM controls 46 Figure 2.2.2. Titration curves of recNP-HRP with three different human IgM controls 47 Figure 2.2.3. Optimisation of the cut-off for the recNP-HRPO IgM ELISA in humans using the misclassification cost term (MCT) option of the two-graph receiver operating characteristic analysis (TG-ROC) 49 Figure 2.3.1. Comparison of mean immune responses in three experimentally infected sheep as measured by testing na?ve versus thermo-chemically inactivated sera 55 Figure 2.4.1. Upper and lower internal quality limits for PP values of high positive, low-positive, negative serum and conjugate controls in recNP-based IgG I-ELISA 60 Figure 2.4.2. Cross-reactivity of recombinant nucleocapsid protein of RVFV in I-ELISA with mouse IgG antibody against selected viruses of the family Bunyaviridae 61 Figure 2.4.3. The effect of different ELISA cut-off values on the discrimination between positive and negative human sera 63 Chapter three Figure 3.1. Upper and lower internal quality control limits for PP values of high-, low-positive and negative antigen controls 72 Figure 3.2. Dose response curves of recNP, RVFV Ar20368 RSA 81 and control antigen 73 Figure 3.3. Dose response curves of human and animal sera, animal tissue and mosquito homogenates spiked with RVFV and their corresponding non-spiked controls 75 Figure 3.4. RVFV replication kinetics in Vero cells inoculated with different concentrations of the virus measured by antigen detection ELISA 76 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page xiii P. Jansen van Vuren Page Figure 3.5. Detection of infectious virus and viral nucleocapsid antigen in tissue homogenates spiked with RVFV followed by incubation at 37?C 77 Figure 3.6. Detection of nucleocapsid protein by ELISA in RVFV-infected buffalo foetus and mouse organ tissues. 78 Figure 3.7. Monitoring of viremia, antigenemia, anti-nucleocapsid IgM and IgG responses in a sheep experimentally infected with RVFV 79 Figure 3.8. Blocking effect of increasing levels of anti-RVFV specific antibody on antigen detection in viremic sheep serum 79 Chapter four Figure 4.1. Detection of total IgG, IgG1 and IgG2A specific antibodies against the RVFV recNP in individual mice after recNP immunization 88-92 Figure 4.2. In vivo neutralization of RVFV with anti-recNP immune sera in mice 95 Figure 4.3. Protection of mice immunized with the 35?g and 70?g recNP doses against disease or death after RVFV challenge compared to placebo control mice 99 Figure 4.4. Mean RVFV TCID50/gram of tissues from dead or sick mice compared to healthy mice 100 Chapter five Figure 5.1. Average anti-NP responses in sheep after recNP first and booster immunizations 108 Figure 5.2. Mean IgM responses in sheep after RVFV challenge on day 37 or 168 111 Figure 5.3. Mean virus neutralizing antibody responses in sheep after RVFV challenge on day 37 or 168 112 Chapter six Figure 6.1. Detection of total IgG, IgG1 and IgG2A specific antibodies against the RVFV recNP in mice after recNP immunization with Alhydrogel 127 Figure 6.2. Mean viral loads in livers and spleens of RVFV infected 131 mice. Figure 6.3. Fold changes in expression of IL10, IFN? and IFN? genes in 134 tissues of mice after RVFV infection. Figure 6.4. Volcano plot displaying fold changes in expression of 84 genes in livers of recNP immunized mice 139 Figure 6.5. Volcano plot displaying fold changes in expression of 84 genes in livers of Alhydrogel control mice 140 Figure 6.6. Volcano plot displaying fold changes in expression of 84 genes in livers of PBS control mice 140 Figure 6.7. The fold changes in expression of genes, with specific immune outcomes, in the liver of experimental groups at 72 hours after RVFV infection 141 Figure 6.8. Heat maps showing fold changes in liver and spleen at 72 hours after RVFV infection 142 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page xiv P. Jansen van Vuren LIST OF TABLES Page Chapter Two Table 2.1.1. Number of field-collected wildlife sera tested in the virus neutralization test 35 Table 2.1.2. Diagnostic accuracy of Rift Valley fever recNP-based I-ELISA in African wildlife 41 Table 2.2.1. Diagnostic accuracy of Rift Valley fever recNP-HRPO IgM ELISA in humans 48 Table 2.3.1. Antibody detection ELISAs for RVF diagnosis in humans 53 Table 2.3.2. Antibody detection ELISAs for RVF diagnosis in livestock 54 Table 2.4.1. Internal quality control data and repeatability estimates for Rift Valley fever IgG I-ELISA based on recombinant nucleocapsid antigen 59 Table 2.4.2. Diagnostic accuracy of Rift Valley fever IgG I-ELISA based on recombinant nucleocapsid antigen 62 Chapter Three Table 3.1. Identification, year of isolation, origin and concentration of RVFV strains used to evaluate the sAg-ELISA 70 Table 3.2. Internal quality control limits and repeatability estimates of RVF s-Ag ELISA 71 Table 3.3. Diagnostic accuracy of the sAg-ELISA for the detection of nucleocapsid protein of RVFV in sheep and human sera 77 Chapter Four Table 4.1. Mouse survival rates after RVFV challenge 100 Table 4.2. Viral load data in mice after RVFV challenge 101 Chapter Five Table 5.1. Group assignments and immunization schedules of sheep 106 Table 5.2. Viremia in sheep after RVFV challenge 114 Chapter Six Table 6.1. Genes analyzed by Quantitect RT-PCR 123 Table 6.2. Fold changes in gene expression, and viral loads in liver tissues from all treated groups at different time points 128 Table 6.3. Fold changes in gene expression, and viral loads in spleen tissues from all treated groups at different time points 129 Table 6.4. Fold changes in gene expression, and viral loads in brain tissues from all treated groups at different time points 130 Table 6.5. Genes analyzed by SABiosciences PCR array, including fold changes and P-values 136-138 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 15 P. Jansen van Vuren CHAPTER ONE INTRODUCTION AND LITERATURE REVIEW 1.1 Virus classification and characteristics Rift Valley fever virus (RVFV) is a mosquito borne member of the Phlebovirus genus in the Bunyaviridae family of viruses (Bishop et al., 1980). The Bunyaviridae family consists of spherical shaped enveloped viruses classified in five genera: Hantavirus, Nairovirus, Orthobunyavirus, Phlebovirus and Tospovirus. RVFV has a diameter of up to 120 nanometres (nm) with short glycoprotein spikes projecting through the lipid envelope (Gerdes, 2004). Even though strains of RVFV differ in their pathogenicity, they are structurally and serologically indistinguishable (Rice et al., 1980). The RVFV genome consists of three single-stranded ribonucleic acid (RNA) segments; large (L), medium (M) and small (S). The L segment, consisting of 6404 bases, has negative polarity and encodes the viral RNA-dependent RNA polymerase. The M segment, consisting of 3885 bases, has negative polarity and encodes the precursor of the viral envelope glycoproteins Gn and Gc, a 78- kilodalton (kDa) non-structural glycoprotein and a non-glycosylated 14-kDa protein. The S segment consists of 1690 bases and encodes the viral nucleocapsid protein (NP) and a non-structural protein NSs using an ambisense coding strategy (Ihara et al., 1984, Giorgi et al., 1991). The NP (length: 245 amino acid residues, weight: 27,431-kDa) is encoded by 738 bases of subgenomic viral- complementary messenger (m) RNA. The NSs protein (length: 265 amino acid residues, weight: 29,903-kDa) is encoded by 798 bases of subgenomic viral-sense mRNA (Suzich et al., 1990, Billecocq et al., 2004). Mature viral particles, however, have been shown to not only contain negative sense viral RNA but also a fraction of RNA complementary to viral RNA (cRNA) (Ikegami et al., 2005), allowing the virulence factor NSs to be expressed immediately after the virus enters the cell (Bouloy and Weber, 2010). Figure 1.1. Rift Valley fever virus genome. The gene encoding the NP of Phleboviruses is highly conserved (Giorgi et al., 1991, Vialat et al., 1997). The NP and viral polymerase proteins of negative sense RNA viruses associate with the Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 16 P. Jansen van Vuren virus genome and form ribonucleoproteins (RNPs) which are necessary for transcription and replication (Gauliard et al., 2006). RNP molecules appear circular because of the complementary sequences of 5? and 3? non-coding regions which cause the formation of panhandle structures (Le May et al., 2005). It was recently shown that RNPs interact with the cytoplasmic tail of the glycoproteins, supposedly enabling successful packaging of the genome into virus particles (Bouloy and Weber, 2010). The NP is the most abundant protein in infected cells, and thus the immunodominant antigen during infections with viruses from the Bunyaviridae family (Swanepoel et al., 1986a, Magurano and Nicoletti, 1999, Gauliard et al., 2006). The RVFV NP is the first viral protein to be synthesized (Ikegami et al., 2005) and it has been shown that it can be released from infected cells independently of the glycoproteins (Liu et al., 2008). The NSs is the most variable protein among Phleboviruses (Sall et al., 1997). The RVFV NSs protein is different from those of other Bunyaviruses in that it is phosphorylated and found in the nucleus of infected cells, which is unique because all stages of the viral life cycle occurs exclusively in the cytoplasm. The NSs protein forms filamentous structures in the host nucleus and interacts with cellular nuclear proteins (Swanepoel and Blackburn, 1977, Yadani et al., 1999, Le May et al., 2004, Ikegami et al., 2006, Le May et al., 2008). The structural glycoproteins, Gn and Gc, are responsible for attachment and entry of the virus to cells and carries neutralizing epitopes (Besselaar and Blackburn, 1991, Besselaar and Blackburn, 1992, Bouloy and Weber, 2010). Figure 1.2. Rift Valley fever virus structure. 1.2 Epidemiology Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 17 P. Jansen van Vuren Rift Valley fever virus was first isolated in 1931 in the Rift Valley of Kenya near Lake Naivasha from the blood of a newborn lamb during an outbreak in domestic livestock, after which the first human infections were also noted (Daubney et al., 1931). The first report of extensive human disease caused by RVFV was in 1951 in South Africa where an estimated 20 000 persons were infected (Mundel and Gear, 1951, House et al., 1996). Large outbreaks then occurred in Egypt 1977/1978 (between 18 000 and 200 000 human infections, 598 deaths), Mauritania 1987 (approximately 200 human deaths), East Africa and Madagascar 1991 (89 000 infections and more than 500 human deaths) and East Africa 1998 (98 000 humans infected and 250 deaths) (Swanepoel and Coetzer, 2004). The disease spread outside its endemic range, across the Arabian Peninsula, to Saudi Arabia and Yemen in 2000 affecting 882 humans and causing 124 deaths (Balkhy and Memish, 2003). More recently outbreaks were recorded in East Africa (2006-07), Sudan (2007/2008), Madagascar (2008) and South Africa (2008, 2009-2011)(Anonymous, 2007, Mohamed et al., 2010, Nguku et al., 2010). Although no RVF outbreaks have been confirmed in a number of African countries such as Mali, Gabon, Congo, Chad, Botswana, Angola, Nigeria, Uganda, antibodies to the virus were found in humans and livestock from these countries (Gerdes, 2004, Pourrut et al., 2010). Antibodies have also been found in domestic livestock from Senegal, Cameroon, Togo, Benin, Ivory Coast and Burkina-Faso (Zeller et al., 1995). Outbreaks of RVF are usually associated with above average rainfall, occurring at irregular intervals of about 10 years (Swanepoel and Coetzer, 2004). Several possible explanations have been put forward to explain the survival and circulation of the virus between outbreaks. Initially it was thought that the virus was endemic in forests where it circulated between mosquitoes and vertebrate hosts, and spilled over into domestic livestock to cause outbreaks (Smithburn et al., 1948, Smithburn et al., 1949a). However, the RVF outbreak in South Africa in the 1950s occurred in grassland country where shallow and poorly drained depressions (pans) are abundant. These pans are usually dry but during heavy rains they fill up and allow mosquito eggs, possibly laid months or years before, to hatch. After RVFV was isolated from mosquitoes that were reared from eggs collected in a pan in South Africa, as well as unfed male and female mosquitoes in Kenya during an inter-epidemic period, it hinted that the virus might be transmitted transovarially by aedine mosquitoes (Linthicum et al., 1985). RVFV is transmitted mostly by Aedes and Culex mosquitoes, but other mosquitoes (Anopheles, Eretmapoites, and Mansonia) have also been shown to be potential vectors of the virus (Smithburn et al., 1948, Smithburn et al., 1949a, Turell and Bailey, 1987, Turell et al., 1990, Turell et al., 2007, Turell et al., 2008a, Turell et al., 2008b, Sang et al., 2010). Interepidemic transmission of RVF in domestic livestock (Scott et al., 1956, Linthicum et al., 1985) and wildlife (Evans et al., 2008) have been shown, leading to the current theory that RVFV is maintained in interepidemic periods by Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 18 P. Jansen van Vuren transovarial transmission and cycling between mosquitoes and ungulates. Outbreaks then occur after abnormally high rainfall due to an explosion in competent mosquito vector numbers. 1.3 Rift Valley fever in animals Sheep are primarily affected by RVF but other ruminants including cattle, goats, camels and wildlife animals are also susceptible to infection (Swanepoel and Coetzer, 2004). Antibodies against RVFV have been found in African buffalo, black rhino, lesser kudu, impala, African elephant, Thompson?s gazelle, gerenuk and waterbuck (Evans et al., 2008). It has been proposed that wild vertebrates might play a role in maintenance of the virus between epizootics, since wild animals repeatedly test positive for antibodies to RVFV in locations unrelated to any documented livestock or human outbreaks (Bengis et al., 2004). Fatality rates reach up to 100% in young lambs whereas 20- 30% of infected adult sheep die. Up to 90% of pregnant ewes abort after being infected. This can be attributed to very high levels of viremia, which also increases the chances of transmission to humans that handle infected tissues (Woods et al., 2002). In non-pregnant adult animals clinical signs include listlessness, abdominal pain, vomiting, diarrhoea, jaundice hepatitis, icterus, nasal discharge and death in some cases. Onset of abortions and high neonatal mortality are characteristic signs of a RVF outbreaks (Swanepoel and Coetzer, 2004). Animals like buffalo and camels do not exhibit disease but pregnant animals can abort when infected (Gerdes, 2004). It was recently shown that a European breed of sheep is also susceptible to RVFV infection, with experimentally infected sheep showing no mortality but mostly pyrexia and corneal opacity despite relatively high viral loads in their blood (Busquets et al., 2010). Infected hepatocytes are probably the major source of high plasma viremia observed in infected animals (Ritter et al., 2000). RVFV has been found to cause extensive necrosis in the livers of aborted lambs, whereas livers from adult sheep are not affected as much. The spleen is not enlarged and is not affected by extensive haemorrhaging as seen in the liver, which is an important feature which distinguishes it from other sheep diseases with similar symptoms to RVF. Other organs also show signs of congestion or haemorrhaging including kidneys and the lymphatic system (Daubney et al., 1931). Non-human primates are susceptible to RVF. Monkeys have been successfully infected with the RVFV ZH-501 pathogenic strain in the laboratory, resulting in poor appetite, anorexia and petechiation, as well as reduced activity 3 to 5 days after infection. The virus grew to titres reaching as high as 6.7log10 PFU/ml between 3 and 5 days post infection, and could be isolated from brain, spinal cord, liver, spleen and mesenteric lymph node tissues of infected monkeys (Morrill and Peters, 2003). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 19 P. Jansen van Vuren 1.4 Rift Valley fever in humans Humans are highly susceptible to infection with RVFV. Transmission of RVFV to humans can occur through bites from infected mosquitoes, contact with contaminated meat or through aerosols created during slaughtering. When the virus was isolated for the first time in 1931 in Kenya, all four European scientists/veterinarians involved in the outbreak were infected while working with aborted fetuses and other infected tissues. None of the infections were fatal but it showed clearly how easily humans can become infected. The group described their own symptoms as malaise at onset of the disease followed by rigors, headache, fever for a period of 12 to 36 hours and joint pains, with symptoms generally disappearing within four days. One of the scientists developed a second latent reaction which included headache and defective vision for a few weeks afterwards (Daubney et al., 1931). Human infection may take on four forms: i) Uncomplicated, febrile, influenza-like illness; ii) hemorrhagic fever with liver involvement, thrombocytopenia, icterus and bleeding; iii) encephalitis following a febrile episode with confusion and coma or even death; iv) ocular involvement with reported blurred vision and loss of visual acuity due to retinal haemorrhage and macular oedema (Al- Hazmi et al., 2003, Gerdes, 2004, Mohamed et al., 2010). The more severe complications occur in up to 5% of the cases. Risk factors recognized for Rift Valley fever infection include consuming or handling products from sick animals and caretaking of animals, whereas touching aborted foetuses is associated with severe RVF complications and consuming or handling products from sick animals associated with death (Anyangu et al., 2010). Examination of blood from infected individuals show leucopaenia, elevated blood enzymes because of liver damage and thrombocytopenia. The main sites of viral replication are the liver, spleen and the brain (Ritter et al., 2000). The incubation period of RVF ranges from 12 hours to six days in young and adult humans, with illness lasting up to eight days. The mortality rate is ?1%, but a 15% rate has been observed in hospitalized patients (Al-Hazmi et al., 2003, Gerdes, 2004). RVFV can be transmitted to a newborn child from its infected mother, as was shown in the 2000 Saudi Arabia outbreak when a 5-day-old newborn died from RVF, after onset of disease on day 2 after birth and presence of IgM antibodies (Arishi et al., 2006). The infant had sepsis, enlarged liver, coagulopathy, anemia and abnormal liver function. 1.5 Pathogenesis The pathogenesis of RVFV infection after mosquito bite infection has been proposed to follow the following sequence: the virus spreads from the skin to draining lymph nodes where it initially replicates in the macrophages and then spreads to the circulatory system (Smith et al., 2010). The liver Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 20 P. Jansen van Vuren is the first and major site of major RVFV replication after infection (Anderson et al., 1987, Shieh et al., 2010) where large amounts of virus is produced. This results in necrosis, apoptosis of the hepatic cells as early as day 2 after infection which, together with coagulation, is probably responsible for the hemorrhagic manifestation of the disease. Liver enzymes become elevated in infected mice on day 3 post infection (p.i.), and remain elevated until day 8 p.i. Infection also causes significant increases (neutrophils, red blood cells, eosinopils, basophils) or decreases (lymphocytes) in important circulating blood cells (Smith et al., 2010). The virus is also able to cross the blood-brain barrier as early as day 5 p.i. and cause meningoencephalitis and retinitis (Gonzalez-Scarano et al., 1991, Smith et al., 2010). Apart from these major organs, the virus has also been proved to show tropism to a variety of other organs including spleen, lymph nodes, heart, kidney, lung, pancreas and adrenal glands in a mouse model (Smith et al., 2010). Specific host genetic factors seem to play a very important role in the type and severity of disease caused by RVFV as shown by different responses to infection in different strains of inbred rats (Peters and Slone, 1982, Anderson et al., 1987). Interferons (IFN) are a family of secreted proteins with many functions including antiviral defence, cell growth, and regulation and activation of the immune response. Interferons are regarded as a powerful defence mechanism and its effectiveness has led many viruses to develop mechanisms to counteract the production and actions of these proteins. In some degree all viruses that successfully infect an animal or human host must have some measure of evading IFN. Therefore the interaction between virus and host IFN is an important determinant of pathogenicity (Goodburn et al., 2000). Type I interferons (IFN-?/?) are produced in direct reaction to virus infection and are the products of two gene families. Leukocytes are mostly responsible for the production of the IFN-? multigene family, whereas IFN-? is synthesized in most cell types but predominantly in fibroblasts. On the other hand, type II IFN is the product of the IFN-? gene and is produced by T lymphocytes and natural killer (NK) cells in response to the recognition of infected cells. Both type I and II IFN function by activating an antiviral state in target cells which interferes with viral and cellular processes. They also slow growth of the cells which induces apoptosis and thus limits the spread of a virus, and stimulate the acquired immune response (Goodburn et al., 2000). There are two main pathways leading to the expression of IFN genes (Haller et al., 2006). The classical pathway is utilized by most cells in the body including fibroblasts, hepatocytes and conventional dendritic cells. Viral components are detected in the cytoplasm by intracellular sensors which lead to activation of interferon regulatory transcription factors (IRF-3) and nuclear factor kappa beta (NF-kB) which activates IFN-? expression. IFN-? is secreted as an initial response and IFN-? only later as a secondary response. Toll-like receptors (TLRs) are expressed on the surface of plasmacytoid dendritic cells or in endosomes where they sense extracellular or engulfed virus material. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 21 P. Jansen van Vuren The regulatory factor, IRF-7, serves as a regulator for IFN-?/? expression, and plasmacytoid dendritic cells express mainly IFN-?. Type I IFN activate a whole range of genes involved in antiviral activity which are grouped into three main systems, namely Protein Kinase R (PKR), 2-5 OAS/RNaseL and Mx protein systems. Viruses have found ways to interfere with the cellular IFN response in order to survive, which include interference with IFN induction, basic transcription, IFN signalling and IFN effectors (Haller et al., 2006). Rift Valley fever virus is sensitive to interferon (Peters et al., 1986). The importance of IFN in the pathogenesis of Phleboviruses has been shown previously in experimental RVFV infection of rats and hamsters (Anderson and Peters, 1988, Perrone et al., 2007). It was found that some RVFV strains that were more capable of killing rats than other strains were not as sensitive to the antiviral effects of IFN in cell culture. The suppression of type I interferon by Punta Toro virus NSs protein results in uncontrolled viral replication and hamster death (Perrone et al., 2007). In rhesus monkeys it was found that IFN-?, administered as a prophylactic, was able to suppress viremia and subsequently disease (Morrill et al., 1990). Thus viruses or their specific strains that are able to counteract the induction of, or evade the action of, type 1 interferon (IFN-?/?) will be the more virulent ones. In a study to evaluate the effects of varying virulence of RVFV in mice, it was found that strains with lower virulence were better IFN -?/? inducers, with IFN being detected at a very early stage of infection, than more virulent strains where IFN was only detectable shortly before death (Higashihara et al., 1972). A study recently showed that a specific laboratory mouse breed MBT/Pas was more susceptible to RVFV infection and its embryonic fibroblasts (MEF) were able to propagate virus to higher titres when compared to a traditional laboratory mouse breed BALB/cByJ (do Valle et al., 2010). Through gene expression profiling of RVFV infected MEF cells from MBT/Pas mice it was shown that a delayed and incomplete type I interferon response was responsible for the increased virulence in these mice compared to BALB/cByJ MEF cells. This study showed the important role the innate immune response plays in susceptibility to RVFV infection (do Valle et al., 2010). Understanding of the mechanism by which the virus inhibits the antiviral attacks of the host on the molecular level is important for the development of vaccines and anti-viral treatments. A large deletion in the NSs gene of the RVFV Clone 13 strain resulted in attenuation. This observation resulted in recent research exploring the role of the NSs in virulence and pathogenicity of RVFV. The determinants for RVFV virulence lies in the S segment (Vialat et al., 2000, Bouloy et al., 2001), with the NSs protein acting as an interferon antagonist. The NSs protein, which is neither stimulatory nor inhibitory to viral replication, inhibits the type I IFN response of the host by blocking virus-induced IFN-?/? production (Bouloy et al., 2001, Ikegami et al., 2006). Transcription factor II H (TFIIH), a basal transcription factor involved in deoxyribonucleic acid (DNA) repair and cell cycle regulation, is a Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 22 P. Jansen van Vuren target of the NSs protein which interacts with the p44 subunit of TFIIH, blocking its assembly. This inhibits its helicase activities responsible for separating nucleic acid strands prior to transcription and the phosphorylation of RNA polymerase II, resulting in transcription shut down. (Dasgupta, 2004). The NSs protein also interacts with the host protein, Sin3A Associated Protein 30 (SAP30), which is an important regulator of IFN-? gene expression (Le May et al., 2008), thus blocking IFN-?/? production at the transcriptional level (Billecocq et al., 2004). The NSs also promotes the post-transcriptional downregulation of PKR, thus inhibiting the phosphorylation of important proteins in the interferon response (Habjan et al., 2009b, Ikegami et al., 2009, Bouloy and Weber, 2010). The cytopathic effect of many viruses can be attributed to apoptosis, which can be regulated by viral gene products, host immune response or double stranded RNA mediated cell responses. The aim of apoptosis is to eliminate infected cells to limit further spread of the virus. The RVFV non-structural proteins from the M segment are not necessary for virus replication (Won et al., 2006, Bird et al., 2007a) but rats infected with a RVFV mutant lacking expression of both NSm proteins (14 kDa and 78 kDa) showed attenuated virulence which implies their possible involvement in RVF pathogenesis. The RVFV NSm proteins have been implicated as anti-apoptotic agents since it was found that NSm suppressed staurosporine (STP)-induced apoptosis in the absence of the other RVFV proteins, and cells infected with mutant virus lacking NSm genes underwent apoptosis earlier than cells infected with wild type virus (Won et al., 2007). 1.6 Diagnostic techniques Rift Valley fever can be diagnosed by detecting antibodies, viral antigens or genetic material. Rapid diagnosis of RVF is essential in endemic areas but even more important in RVF-na?ve countries at risk of introduction of the disease. Various classical laboratory methods for RVF diagnosis have been developed but are costly, time consuming and require biocontainment facilities since live virus is used, which hamper quick diagnosis. These tests also pose safety risks to laboratory personnel. There is an increasing demand for safe, accurate and simple diagnostic tools for RVF diagnosis because of the possibility of the virus spreading to previously non-endemic areas such as Europe and north-America where competent mosquito vectors are present. 1.6.1 Virus / antigen detection The golden standard method for RVF virus isolation is intracranial (i.c.) inoculation of suckling mice which yields positive results within seven days after inoculation. The virus can also easily be propagated in a variety of mammalian cell lines including Vero (African green monkey Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 23 P. Jansen van Vuren kidney) and BHK (baby hamster kidney) cells (Swanepoel and Coetzer, 2004). These methods, however, are time consuming and involve the propagation of live virus, thus necessitating the use of biocontainment facilities. Viral antigen can also be detected by immunofluorescence, complement fixation, immunodiffusion or enzyme-linked immunosorbent assay (ELISA) (Niklasson et al., 1983, Peters et al., 1989, Swanepoel and Coetzer, 2004, Zaki et al., 2006). These methods are all based on whole virus antigen which still requires biocontainment facilities for production of antigens and performaing assays. The short duration of viremia in experimentally infected sheep (usually days 2 ? 4 post infection) and monkeys (days 3 ? 5 post infection) should, however, be taken into account when interpreting virus isolation, antigen detection or vRNA detection results (Olaleye et al., 1996, Morrill and Peters, 2003, Bird et al., 2011). 1.6.2 Molecular biology Various molecular techniques have been developed and evaluated for the detection of RVFV RNA and have been found to be highly specific, sensitive and rapid. These assays include conventional and nested polymerase chain reaction (PCR) (Ibrahim et al., 1997, Jupp et al., 2000, Sall et al., 2001, Sall et al., 2002), real-time PCR (Garcia et al., 2001, Drosten et al., 2002, Njenga et al., 2009) and most recently loop-amplification-mediated-PCR (LAMP) (Peyrefitte et al., 2008, Le Roux et al., 2009). The molecular methods mentioned here have been optimized for the detection of RVF genetic material in livestock, human and insect specimens. Detection of virus genetic material, however, is of limited value once the virus has been cleared from the infected individual?s system, especially for viruses with a short viremia when only blood or serum is available for testing. Molecular techniques are also highly specialized and might not be ideal for use in developing countries or in the field where conditions are less than ideal. 1.6.3 Serology Methods that detect antigen or viral genetic material are highly sensitive shortly after infection but because of the short viremia in RVFV infected individuals, these methods have no use once the virus is cleared. Serological assays in general play a very important role in the field of infectious diseases. They are used for the diagnosis of suspected cases, sero-epidemiological studies, import- export certification of animals, disease eradication programmes and monitoring of vaccine efficacy. It is therefore critical to be able to detect antibodies against RVFV to show recent or past infections. In experimentally infected sheep IgM and IgG antibodies against RVFV can be detected as early as 4 days post infection. IgM responses are, however, transient and usually wane below detectable levels by day 60 post infection, whereas IgG does not have the same transient nature. A recent infection can Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 24 P. Jansen van Vuren therefore be confirmed by detection of virus specific immunoglobulin-M (IgM) in serum as a result of the early production and transient nature of IgM, or by the indication of a four-fold increase in virus specific immunoglobulin-G (IgG). A past infection can be confirmed by the detection of virus specific IgG since infection of an individual with RVFV induces life-long immunity. The gold standard serological technique for RVF is the virus neutralization test (VNT) (Swanepoel et al., 1986a). The VN test detects neutralizing antibodies which are solely directed against the virus? glycoproteins, but cannot distinguish between IgM and IgG. It also involves the propagation of live virus. Validation of a diagnostic assay essentially refers to the process of determining the fitness of the test for its intended use including assay accuracy, repeatability, reproducibility and stability. Once a diagnostic assay is validated, it can be used to identify the presence or absence of the specific analyte with high confidence. When taken seriously, validation is not a once-off experiment, but it is rather an ongoing process. However, certain steps have to be followed during initial validation which include i) a feasibility study; ii) development and standardization of reagents and protocols; iii) determination of assay performance by testing large numbers of reference or well characterized samples and calculation of cut-off values and accuracy estimates and iv) ongoing evaluation of performance (Crowther et al., 2006). Proper validation of diagnostic assays is therefore important since it results in the determination of reliable estimates of diagnostic specificity and sensitivity which in turn is important factors to take into account for disease diagnosis, risk-assessment and risk-factor studies (Paweska et al., 2003a). Various ELISAs have been developed and validated for the diagnosis of RVF in humans (Niklasson et al., 1984, Swanepoel et al., 1986a, Paweska et al., 2005a, Paweska et al., 2005b, Jansen van Vuren et al., 2007) and animals (Paweska et al., 1995, Paweska et al., 2003a, Paweska et al., 2003b, Paweska et al., 2005b, Fafetine et al., 2007, Jansen van Vuren et al., 2007, Cetre-Sossah et al., 2009, McElroy et al., 2009). Most of these ELISAs are based on whole inactivated virus antigens which still poses safety risks, but recently a few ELISAs were developed using recombinant antigens which are completely safe (Fafetine et al., 2007, Jansen van Vuren et al., 2007, McElroy et al., 2009). The ELISAs based on the recombinant nucleocapsid protein (recNP) were found to be highly sensitive and specific but has not yet been extensively validated. It would be important to develop and validate ELISAs based on recombinant antigens for detection of IgG and IgM in humans and also for detection of antibodies in animals because of the important role they play in disease maintenance and transmission. 1.7 Vaccines and antivirals In RVFV enzootic areas vaccination is the only practical method of preventing the disease. Various types of vaccines have been developed for RVFV, but none are commercially available for human use and those available for animal use have several drawbacks. Recently a broad-spectrum Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 25 P. Jansen van Vuren antiviral (so called ?LJ001?, derived from an organic compound called rhodanine), was shown to inhibit virus-cell fusion of lipid enveloped viruses without inhibiting host cell-cell fusion (Wolf et al., 2010). It was specifically shown to prevent death in 100% of RVFV challenged mice that had been pre-treated with the compound. 1.7.1 Inactivated virus vaccines Inactivated vaccines are relatively expensive to produce, require multiple inoculations because of weak immunogenicity and pose a safety risk due to possible incomplete inactivation of the virus. A laboratory adapted neurotropic RVF virus strain and the pantropic Entebbe RVF virus strain were used to generate a range of formalin inactivated experimental vaccines in chick embryo, mouse brain and monkey kidney cell cultures (Randall et al., 1962, Randall et al., 1964). The experimental vaccine produced in monkey kidney cells from the pantropic Entebbe strain was safe and relatively effective for the immunization of mice, monkeys and humans. This vaccine (TSI-GSD 200) produced under strict quality control conditions, had been used to immunize many veterinarians and laboratory workers at risk of infection (Niklasson et al., 1985, Pittman et al., 1999). Only one batch of TSI-GSD 200 was, however, produced by the US Army (USAMRID) and it is therefore in short supply and very expensive. It also requires three initial inoculations and a booster after 6 months, making it impractical (Bouloy and Flick, 2009). The same formalin-inactivated vaccine used for humans (TSI-GSD 200) was evaluated in sheep (Harrington et al., 1980). It induced neutralizing immunity that resulted in protection from disease and decreased viral replication after challenge. A formalin inactivated RVF vaccine is available from Onderstepoort Biological Products (OBP, South Africa) for use in livestock (Barnard and Botha, 1977, Barnard, 1979, Bouloy and Flick, 2009). It induces protective responses but is not very immunogenic, neccesitating multiple inoculations which could be problematic during outbreaks when rapid induction of protective immunity is required. 1.7.2 Attenuated virus vaccines Attenuated vaccines are less expensive to produce, and more immunogenic compared to inactivated vaccines, but still carry a safety risk because of possible reversion to virulence and possible spread by mosquito vectors (Swanepoel and Coetzer, 2004). The attenuated Smithburn neuroadapted strain is also commercially available from OBP (South Africa) for use in livestock but has adverse side effects like teratology, liver pathology and abortions in pregnant animals (Coetzer and Barnard, 1977, Botros et al., 2006, Kamal, 2009). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 26 P. Jansen van Vuren Another attenuated RVF strain, MP12, which was generated by mutagenesis with 5- fluorouracil of the ZH548 strain, has been extensively evaluated as a possible vaccine candidate (Caplen et al., 1985, Morrill et al., 1991a, Hunter et al., 2002, Morrill and Peters, 2003). The MP12 strain contains nine, 12 and four mutations in the L, M and S segments of the virus? genome respectively, with at least one mutation in each segment playing a role in attenuation (Takehara et al., 1989, Vialat et al., 1997). The MP12 attenuated strain induced strong neutralizing antibody responses but is also teratogenic and abortogenic in sheep when administered in the first trimester of pregnancy (Morrill et al., 1991a, Hunter et al., 2002). Despite these negative effects in sheep, MP12 was evaluated in rhesus monkeys and shown to be markedly attenuated, causing only minor neurovirulence comparable to that seen with the widely used 17D yellow fever vaccine (Morrill and Peters, 2003). Humans immunized with MP12 remained asymptomatic and 95% developed neutralizing antibodies against RVFV (Bettinger et al., 2009, Bouloy and Flick, 2009). Clone 13, a small plaque naturally attenuated RVFV strain that was isolated in the Central African Republic from a human patient, lacks approximately 70% of the open reading frame coding for the NSs protein preventing it from evading the host IFN pathway (Muller et al., 1995, Bouloy et al., 2001, Billecocq et al., 2004). It was shown to be highly immunogenic in animals, elicited protective immune responses against subsequent challenge and did not cause teratogenesis in sheep during early pregnancy (Muller et al., 1991, Swanepoel and Coetzer, 2004, Bouloy and Flick, 2009, Dungu et al., 2010). Clone 13 was very recently commercialized and is currently under mass production at Onderstepoort Biological Products. Mice immunized with a reassortant virus (R566), containing the S segment of clone 13 and the L and M segments of MP-12, were protected from subsequent viral challenge and sheep did not show any side effects or abortions as a result of vaccination (Bouloy and Flick, 2009). 1.7.3 Recombinant viruses by reverse genetics The development of various reverse genetics systems for the rescue of recombinant RVF viruses has allowed for the determination of certain factors/proteins involved in pathogenesis and virulence of the virus (Billecocq et al., 2004, Won et al., 2006, Bird et al., 2007a, Gerrard et al., 2007, Won et al., 2007, Billecocq et al., 2008, Habjan et al., 2008a, Habjan et al., 2008b). The 14-kDa (NSm) and 78-kDa non-structural proteins encoded by the M segment were shown to be dispensable for replication and recombinant virus without these proteins remained highly virulent in vivo even though it suppressed virus-induced apoptosis (Bird et al., 2007a, Gerrard et al., 2007, Won et al., 2007). The NSs is the major virulence factor of RVFV. Various recombinant viruses have been generated with the NSs completely omitted, or where it was replaced by a reporter gene like the green Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 27 P. Jansen van Vuren fluorescent protein(GFP) (Billecocq et al., 2004, Ikegami et al., 2006, Bird et al., 2008, Habjan et al., 2008b, Ikegami et al., 2009). Only one recombinant virus generated by reverse genetics has been evaluated as a possible vaccine candidate (Bird et al., 2008). A ZH501 strain derived mutant virus, lacking the NSm gene and carrying a GFP gene in the place of the NSs gene (rRVF-?NSs:GFP- ?NSm), was shown to be highly attenuated but still immunogenic in rats, and protected rats from lethal RVFV challenge. Additionally this vaccine candidate allows for differentiation of naturally infected and vaccinated animals (DIVA) because of the missing viral genes and the insertion of a non-viral gene (Bird et al., 2008). This vaccine candidate shows much promise but has not been evaluated in a host animal model. Attenuated viruses, either naturally or by reverse genetics, still pose a possible safety risk due to recombination in nature where the missing gene(s) conferring attenuation could be replaced again by a gene from wild type virus, thus reverting it back to virulence. Therefore vaccines of this nature would not be ideal during outbreaks and should be used before seasonal activity of mosquito vectors. 1.7.4 DNA vaccines DNA-based vaccines are completely safe and are usually more immunogenic when compared to inactivated or subunit vaccines since their target proteins are expressed in vivo by the host cells resulting in correct protein folding, they result in longer term expression of target proteins compared to once-off inoculation with subunit proteins and they are able to induce cellular and humoral immunity (Lorenzo et al., 2008). Various DNA vaccine candidates have been developed against RVF. A DNA vaccine, administered by gene gun and expressing the M segment without the NSm gene, was highly immunogenic in mice after three inoculations and elicited 100% protection against lethal challenge (Spik et al., 2006). Immunization of sheep with a DNA construct expressing the M segment and the NP was not able to elicit detectable humoral responses, but low level antigen-specific cellular responses were induced (Lorenzo et al., 2008). A construct expressing only the NP, however, was able to induce strong anti-NP IgG1 isotype responses as well as cellular responses in sheep (Lorenzo et al., 2008), although the protective ability of this response is not known. DNA constructs based on a pCMV (cytomegalovirus) vector backbone expressing both the RVFV glycoproteins (pCMV-M4), or the nucleocapsid protein (pCMV-N), were evaluated as vaccines in a transgenic mouse model with an impaired type I interferon response using the attenuated RVFV MP12 strain as a challenge virus (Lorenzo et al., 2010). The mice vaccinated with pCMV-M4 were completely protected from challenge, whereas mice immunized with pCMV-N, or pCMV-N combined with pCMV-M4, were not. Gene-gun immunization of mice with a cDNA construct encoding the RVFV NP induced high anti-NP antibody titres and strong proliferative cellular responses, but no neutralizing antibodies Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 28 P. Jansen van Vuren (Lagerqvist et al., 2009). Fifty percent of NP cDNA immunized mice were protected from viral challenge, most likely due to cell-mediated immunity. Immunization with a cDNA construct expressing both glycoproteins (Gn/Gc) simultaneously resulted in neutralizing antibody responses and 62.5% protection from RVFV challenge (Lagerqvist et al., 2009). A DNA vaccine expressing the RVFV Gn protein, coupled to the molecular adjuvant C3d, induced increased neutralizing antibody titres compared to one expressing only Gn, and also increased survival from lethal challenge (Bhardwaj et al., 2010). DNA vaccines have an added advantage of not needing stringent shipping conditions because of high stability, making it ideal for use in countries with sub-optimal infrastructure. The disadvantage, however, is that DNA vaccines require multiple immunizations to induce protective responses. 1.7.5 Virus vectored / replicon vaccines Virus vectored and replicon vaccines are capable of expressing high levels of inserted genes, can be easily produced in large quantities and are highly immunogenic. The earliest report of an unrelated virus being used to express RVFV proteins shows the insertion of RVF glycoprotein genes into vaccinia virus (Collett et al., 1987). Mice inoculated with the live recombinant RVF/vaccinia virus developed strong neutralizing responses and protection of 90-100% of challenged mice, depending on immunization route. Despite these promising results and the fact that mice immunized with the recombinant RVF/vaccinia virus did not develop any side effects from immunization, this vaccine candidate has not been further evaluated or commercialized possibly due to the fact that vaccinia virus poses a threat to immunocompromised individuals. A recombinant lumpy skin disease virus (LSDV) containing both RVF glycoprotein genes yielded strong neutralizing responses and 100% protection from subsequent lethal challenge in mice, and 100% protection from clinical disease in sheep (Wallace and Viljoen, 2005, Wallace et al., 2006). It also allows for differentiation of RVF naturally infected and vaccinated animals since only RVF glycoproteins are expressed by the construct. Despite these promising results, this vaccine candidate has not been commercialized. Alphaviruses, such as Sindbis virus (SINV) and Venezualan equine encephalitis (VEEV), have been evaluated as virus vectors, or replicons, to express RVFV glycoprotein genes (Gorchakov et al., 2007). VEEV was able to express glycoproteins to such a level that protective immunity against lethal challenge was induced in mice (Gorchakov et al., 2007). A SINV replicon based RVF vaccine produced neutralizing responses in mice and sheep, and protected mice from lethal challenge (Heise et al., 2009). Alphaviruses are, however, widespread and it is not known to what effect the background immunity against these viruses could have an influence on efficacy of vaccines. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 29 P. Jansen van Vuren Recently the RVF glycoprotein genes were incorporated into a non-replicating complex adenovirus (CAdVax) vector and evaluated as a possible vaccine in mice (Holman et al., 2009). The CAdVax-RVF construct induced long-lasting humoral immunity that protected 100% of mice against lethal challenge. Background immunity against adenovirus does, however, have an effect on protection with only 25-75% of mice with pre-existing adenoviral immunity surviving challenge (Holman et al., 2009). A capripoxvirus recombinant expressing the two structural RVFV glycoproteins was evaluated as a possible vaccine in a mouse and sheep model (Soi et al., 2010), and was shown to induce up to 100% protection against lethal challenge in mice, depending on the dose, route and number of immunizations, and induced sterilizing immunity in sheep. A recombinant Newcastle disease virus (NDV) expressing the RVFV glycoproteins has been evaluated as a vaccine in mice, lambs and calves, showing promising protection against disease/death and viral replication as a result of the development of neutralizing antibodies (Kortekaas et al., 2010a, Kortekaas et al., 2010b). 1.7.6 Virus like particles as vaccines Virus-like particles (VLP) are formed when the virus? structural proteins self-assemble into replication deficient particles that resemble wild type virus in structure. VLPs are usually more immunogenic when compared to subunit recombinant proteins since the conformational epitopes of structural proteins are presented in a more natural way, similar to wild type virus. Immune responses against VLPs, therefore, are also thought to more accurately represent the immune responses elicited against natural infections (Noad and Roy, 2003, Grgacic and Anderson, 2006). VLP production can be easily upscaled and VLPs have been successfully evaluated as vaccines for various other viruses including Bunyaviruses and Filoviruses (Grgacic and Anderson, 2006). Indeed, various groups have recently successfully generated Rift Valley fever VLPs in mammalian and insect cell systems (Liu et al., 2008, Habjan et al., 2009a, Mandell et al., 2009, Naslund et al., 2009, de Boer et al., 2010, Mandell et al., 2010b). Immunization of mice with RVFV VLPs containing glycoproteins and the NP, produced in mammalian cells (293T) resulted in high neutralizing titres and 50-92% protection from lethal challenge, depending on VLP dose (Habjan et al., 2009a, Naslund et al., 2009). Moreover, none of these mice developed any detectable anti-NP antibodies after immunization, despite the NP being present in the VLP. Chimeric RVFV VLPs containing the RVF NP, glycoproteins and the Moloney murine leukaemia gag protein were shown to be highly immunogenic in two murine animal models, yielding high neutralizing titres and strong cytokine responses which protected mice (68%) and rats Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 30 P. Jansen van Vuren (100%) from lethal RVFV challenge (Mandell et al., 2009). Interestingly VLPs that were modified to actively express the NP were also able to induce 100% protection against lethal challenge (Pichlmair et al., 2010). These results clearly indicate that the NP does play a role in protection despite it having no neutralizing epitopes. Recently, however, VLPs based on both glycoproteins but without NP, produced in Drosophila insect cells, were shown to be immunogenic in mice and conferred 100% protection from lethal RVFV challenge (de Boer et al., 2010). Because of the advantages offered by VLPs, they seem like ideal vaccine candidates for RVF. However, efforts to prepare RVFV VLPs were only very recently undertaken and therefore are in the very early stages of evaluation as vaccine candidates. Based on promising results obtained in murine models, it would be interesting to see how efficient they would be as immunogens in a RVF host animal or humans. 1.7.7 Recombinant subunit vaccines Recombinant subunit immunogens are the least explored method for producing vaccines against RVF. This is most likely due to the fact that recombinant subunit proteins are generally weak immunogens, requiring multiple immunizations and the use of adjuvants, and are usually expressed as inclusion bodies thus not presenting conformational epitopes in a natural way (O'Hagan et al., 2001). Partially purified recombinant Gn protein, bacterially expressed as inclusion bodies and subsequently solubilized with urea, induced low neutralizing titres in mice which protected 56-70% of animals from lethal RVF challenge, whereas the Gc protein did not elicit a neutralizing response or protection against challenge at all (Collett et al., 1987). Recombinant Gn protein, expressed using Autographa californica nuclear polyhedrosis viral recombinants in SF9 insect cells, or a combination of Gc/Gn is immunogenic and resulted in protection from lethal challenge after two immunizations. Passive immunization of mice with mouse anti-serum generated against these recombinant antigens protected a proportion of the animals, further indicating that humoral antibodies against the glycoproteins play a major role in protection (Schmaljohn et al., 1989). Recombinant Gn protein, expressed in Drosophila insect cells and adjuvanated with Stimune, was shown to be immunogenic in mice and conferred 100% protection against lethal challenge (de Boer et al., 2010). Immunity, however, was not sterilizing since anti-NP antibodies were detected in the mice sera after challenge, an evidence of viral replication. In a preliminary study using a limited number of experimental animals, Wallace et. al. evaluated a bacterially expressed RVF nucleocapsid protein as an immunogen in mice (Wallace et al., 2006). The protein was expressed as inclusion bodies and solubilized with urea. Mice immunized with Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 31 P. Jansen van Vuren the protein and QuilA adjuvant did not develop any detectable antibodies, yet 60% were protected from lethal RVFV challenge. Immunization of animals with recombinant subunit N proteins from related Bunyaviruses resulted in complete protection against viral challenge (Schmaljohn et al., 1990, Maes et al., 2008). Recombinant subunit immunogens are easy to produce, relatively cheap, completely safe and production can be easily upscaled. Their disadvantages of weak immunogenicity and expression as incorrectly folded proteins can be overcome by using adjuvants (O'Hagan et al., 2001, Lautze et al., 2007) and optimizing expression conditions respectively. A bacterially expressed recombinant RVFV nucleocapsid protein (recNP) was recently produced in a completely soluble form, and thus assumed to be correctly folded (Jansen van Vuren et al., 2007). The fact that it recognizes RVFV specific antibodies from naturally or experimentally infected individuals with very high efficiency is further proof that conformational epitopes are presented correctly (Jansen van Vuren et al., 2007, Paweska et al., 2007, Evans et al., 2008). 1.8 Study objectives This study had two major objective. The first objective was to evaluate the recNP as an immunodiagnostic reagent. From the literature it became apparent that most serological techniques for RVF diagnosis are based on reagents that are expensive and time consuming to produce, and pose safety risks to laboratory personnel. In order to address this it was decided to develop and validate safe ELISAs, based on the recNP, for the detection of RVFV specific antibodies and antigens in animals and humans. To achieve this, laboratory animals were immunized to generate immune sera and clinical specimens from RVF cases were tested to evaluate diagnostic accuracy of the assays. This study undertook the first development of an ELISA based on a horseradish-peroxidase conjugated RVFV recombinant nucleocapsid protein, and extensive validation of recombinant nucleocapsid protein based ELISAs for use in RVF serodiagnosis in humans and wildlife. The second objective was to evaluate a bacterially expressed RVFV nucleocapsid protein (recNP) as an immunogen in two animal models and attempt to understand the mechanism of protection against viral challenge. From the literature it became apparent that the role of the anti-NP response in immunity against viral infection is not well understood. A preliminary study in mice by another group showed that the RVFV recNP was able to induce protection against lethal RVFV challenge despite the absence of neutralizing antibodies. The same was shown for related viruses from the Bunyaviridae family. It was decided to expand the investigation on immunogenicity in mice using the soluble recNP and the protection it confers against lethal challenge when used in combination with different adjuvants that enhance immune responses by varying mechanisms. In an effort to understand Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 32 P. Jansen van Vuren the mechanism of protection elicited by recNP immunization the expression levels of selected genes in immunized versus na?ve mice after viral challenge were compared. The recNP was also evaluated as an immunogen in a host animal species, sheep. This study is the first evaluation of a subunit RVFV nucleocapsid protein as an immunogen in a RVF host species. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 33 P. Jansen van Vuren CHAPTER TWO RECOMBINANT NUCLEOCAPSID PROTEIN AS IMMUNODIAGNOSTIC REAGENT ? SEROLOGY 2.1 Detection of IgG antibody to Rift Valley fever virus in wild ruminants* * Partially published as: Paweska, J. T., van Vuren, P. J., Kemp, A., Buss, P., Bengis, R. G., Gakuya, F., Breiman, R. F., Njenga, M. K. & Swanepoel, R. (2008). Recombinant nucleocapsid-based ELISA for detection of IgG antibody to Rift Valley fever virus in African buffalo. Veterinary Microbiology 127, 21-8. Paweska, J.T., van Vuren, P.J., Kemp, A., Swanepoel, R., Buss, P., Bengis, R.G., Gakuya, F., Breimann, R.F. and Njenga, K. (2010). A recombinant nucleocapsid-based indirect ELISA for serodiagnosis of Rift Valley fever in African wildlife. In Symposium Proceedings: Odongo, N.E., Garcia, M. And Viljoen, G.J. (eds). Sustainable Improvement of Animal Production and Health. Vienna, Austria, 8 ? 11 June 2009. Food and Agriculture Organization of the United Nations: FAO/IAEA, 309-12. ISBN 978-92-5-106697-3. * Partially presented at an international conference as: Paweska, J.T., van Vuren, P.J., Kemp, A., Swanepoel, R., Buss, P., Bengis, R., Gakuya, F., Breiman, R., & Njenga, M.K. (2009). Serodiagnosis of Rift Valley fever in African wildlife using a recombinant nucleocapsid-based indirect ELISA. FAO/IAEA International Symposium on Sustainable Improvement of Animal Production and Health, Vienna, Austria, 8 ? 11 June 2009. (Oral presentation) 2.1.1 Introduction Antibodies against the virus have been found in many wildlife species including African buffalo, black rhino, lesser kudu, impala, African elephant, Thompson?s gazelle, gerenuk and waterbuck (Anderson and Rowe, 1998, Fischer-Tenhagen et al., 2000, Evans et al., 2008). Experimental infection of African buffalo with RVFV results in fever, malaise and abortion as a result of transient viremia (Davies and Karstad, 1981). The role of wildlife in the epidemiology of RVF is not well understood but they are thought to maintain the virus together with mosquitoes in a sylvatic cycle during inter-epizootic periods (Swanepoel and Coetzer, 2004, Evans et al., 2008). Because there is low level transmission of the virus during inter-epizootic periods, and most infected animals are probably Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 34 P. Jansen van Vuren asymptomatic, this transmission goes undetected without the aid of proper diagnostic techniques. Properly validated assays that are safe, cheap and easy to use could be very useful in gaining more information about the prevalence of RVF in wildlife during inter-epizootic periods. Various ELISAs have been developed and validated recently for the detection of antibodies in domestic livestock and humans (Paweska et al., 2003a, Paweska et al., 2005a, Paweska et al., 2007), and one for humans, domestic livestock and African buffalo (Paweska et al., 2005b). At the time of this study no ELISA had been validated for the detection of anti-RVFV antibodies in wildlife, with the exception of African buffalo. The ELISA available for detection in African buffalo is, however, based on inactivated whole virus antigen which requires it to be produced in biocontainment, making its production expensive and unsafe (Paweska et al., 2005b). To address the needs highlighted above, an ELISA was developed and validated for the detection of IgG antibodies in wildlife. 2.1.2 Materials and methods 2.1.2.1 Serum controls and internal quality control (IQC) Freeze-dried, gamma-irradiated serum controls from experimentally infected sheep produced previously were used (Paweska et al., 2003a). Internal quality control upper and lower control limits for the controls were established as described previously (Paweska et al., 2003a) by testing each control 24 times on five plates on five separate occasions (24 x 5 x 5 = 600 determinants). The upper control limit (UCL) for the controls (high positive C++, low positive C+, negative C- and conjugate control) was determined by calculating the percentage positivity (PP) from the mean optical density (OD) value from the 600 replicates, plus two standard deviations (+ 2 S.D.). PP values were calculated as follows: PP = (net OD serum / net mean OD C++) x 100. The lower control limit (LCL) was determined similarly by calculating the mean values minus 2 standard deviations (- 2 S.D.). During routine runs of the assay, four replicates of each control (high positive C++, low positive C+, negative C- and conjugate control CC) were included on each plate. The means and standard deviations of OD values and PP values were calculated from the replicates on 40 routine runs of the assay over a period of three months (4 replicates x 40 runs = 160 determinants) to assess intra- and interplate variation. The coefficient of variation (CV = standard deviation of replicates / mean of replicates x 100) was determined for positive control sera. Assay repeatability was determined from these results. 2.1.2.2 Serum specimens A total of 1900 individual wildlife sera collected in 1978-2008 in Kenya, South Africa and Zimbabwe were used. Sera which tested negative in the virus neutralization test were regarded as a Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 35 P. Jansen van Vuren reference panel from non-infected animals, and sera which tested positive as a reference panel from animals infected with RVFV (Table 2.1.1). Table 2.1.1. Number of field-collected wildlife sera tested in the virus neutralization test. Species Total Tested VNT-a VNT+b African buffalo 1023 946 77 Black rhinoceros 43 29 14 Common zebra 24 24 0 Elephant 73 69 4 Giraffe 81 81 0 Grevy zebra 78 77 1 Warthog 49 47 2 Eland 66 63 3 Gerenuk 6 1 5 Hartebeest 10 10 0 Impala 324 315 9 Kudu 73 66 7 Waterbuck 42 40 2 Thomson gazelle 8 1 7 Grand Total 1900 1769 131 a Number of sera tested negative in virus neutralization test b Number of sera tested positive in virus neutralization test 2.1.2.3 Virus neutralization test Duplicates of serial two-fold dilutions of sera inactivated at 56?C for 30 min were tested as previously described (Paweska et al., 2003a). Titres were expressed as the reciprocal of the serum dilution that inhibited ? 75 % of viral cytopathic effect. A serum sample was considered positive when it had a titre of ? log10 1.0, equivalent to a serum dilution ? 1:10. 2.1.2.4 ELISA antigen production and recNP I-ELISA procedure Production of the recNP and the assay procedure was carried out as described previously with minor modifications (Jansen van Vuren et al., 2007). Maxisorb immunoplates (Nunc, Denmark) were coated with stock antigen, diluted 1:2000 in carbonate?bicarbonate buffer pH 9.6 and incubated overnight at 4?C. After washing three times with a washing buffer consisting of phosphate-buffered saline (PBS) pH 7.2 and 0.1% Tween 20, the plates were blocked with 200 ?l of 10% fat-free milk Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 36 P. Jansen van Vuren powder in PBS and incubated in a moist chamber for 1 h at 37?C and then washed as described before. Control and test sera were diluted 1:400 in PBS containing 2% milk powder (diluting buffer) and 100 ?l of diluted sera was added to the plates. Each test serum was assayed in duplicate and each internal control was tested in quadruplicate. After incubation in a moist chamber for 1 h at 37?C, plates were washed three times with the washing buffer and 100 ?l of a 1:5000 dilution of the horseradish peroxidase (HRPO) conjugated Protein G (Zymed Laboratories, Inc.) was added. Plates were incubated for 1 h at 37?C, washed three times, and 100 ?l of 2,2?-azino di-ethyl-benzothiazoline-sulfonic acid substrate was added to each well. Plates were then incubated in the dark at room temperature for 30 minutes. The reactions were stopped by the addition of 100 ?l of 1% sodium dodecyl sulphate (SDS) and OD values were determined at 405 nm. The results were expressed as PP values. PP values were calculated as follows: PP = (net OD serum / net mean OD C++) x 100. 2.1.2.5 Selection of cut-off values and determination of ELISA diagnostic accuracy Cut-off values at 95 % accuracy level were optimised using the misclassification cost term option of the two-graph receiver operating characteristics (TG-ROC) analysis (Greiner, 1996). Optimization of cut-off values was based on the following equation: misclassification cost term = (1 - p) (1 - Sp) + rp (1 - Se), where p (prevalence) = 0.5 and r (costs of false-positive and false-negative results) = 1.0. In addition, cut-off values were determined by mean plus 2S.D.s. and by mean plus 3S.D.s derived from PP values in uninfected animals. Estimates of diagnostic sensitivity and specificity and other measures of combined diagnostic accuracy were calculated as previously described (Paweska et al., 2003a). Sensitivity (D ? Se) = [Tp/(Tp + Fn)] ? 100; specificity (D ? Sp) = [Tn/(Tn+Fp)]?100; Youden?s index (J) = [Sn+(Sp?1)]; efficiency (Ef) = [PSe + Sp(1 ? P)]; positive predictive value (PPV) = Pse/[Pse + (1 ? P)(1 ? Sp)] ? 100; negative predictive value (NPV) = [(1 ? P)Sp]/[(1 ? P)Sp + P(1 ? Se)] ? 100; apparent prevalence (AP) = [(Tp + Fp)/N] ? 100; and true prevalence (TP) = [AP + (Sp?1)]/[Sn+(Sp?1)] ?100, where Tp is the true-positive sera, Fn the false-negative sera, Tn the true-negative sera, Fp the false- positive sera, P the prevalence, and N the number of sera tested. 2.1.2.6 Antibody dose response curves Antibody dose response curves were generated by testing increasing dilutions of wildlife sera known to be positive or negative for anti-RVFV antibodies, based on VNT results, using the recNP based IgG indirect ELISA (I-ELISA). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 37 P. Jansen van Vuren 2.1.3 Results 2.1.3.1 Assay repeatability and internal quality control (IQC) The recNP IgG I-ELISA was able to differentiate clearly between the internal controls used and generated minimal background. Variation between and within runs were minimal (Figure 2.1.1). 2.1.3.2 Antibody dose response curves Dose response curves using different dilutions of sera known to be positive or negative in the virus neutralisation test had the expected analytical slope and the recNP IgG I-ELISA clearly differentiated between different levels of specific IgG antibody against RVFV in African buffalo (Figure 2.1.2) and other wildlife (Figure 2.1.3). 0 20 40 60 80 100 120 1 6 11 16 21 26 31 36 Nummber of plates EL IS A PP va lu e Figure 2.1.1 Upper (?) and lower (- - -) internal quality limits for PP values of high positive (C++), low-positive (C+), negative serum (C-) and conjugate (Cc) controls in recNP-based IgG I-ELISA and the results for these controls (mean ? SD) on 40 plates during routine runs of the assay over a period of 3 months. Each plate includes four replicates of each of the internal controls. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 38 P. Jansen van Vuren 0 20 40 60 80 100 120 140 160 2.6 2.9 3.2 3.5 3.8 4.1 4.4 4.7 5.0 5.3 Log10 two-fold dilution EL IS A PP va lu e Figure 2.1.2 Dose response curves of individual African buffalo sera in the recNP I-ELISA. Sera were collected from 16 animals in RVF endemic areas in Kenya and South Africa of which 8 tested positive and 8 negative in virus neutralization test. Positive sera (?) represent different levels of virus neutralizing antibody ranging in titres from log10101.6 (?) to log10103.1 (?). Mean ? SD of 8 negative sera (- - -) with VNT titres < log10101. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 39 P. Jansen van Vuren 0 20 40 60 80 100 120 140 160 180 2.6 2.9 3.2 3.5 3.8 4.1 4.4 4.7 5 5.3 Log10 two-fold serum dilution EL IS A PP v al u e Figure 2.1.3 Dose response curves of wildlife sera in the recNP I-ELISA. Sera were collected from individuals that tested positive (?) or negative (- - -) in the virus neutralization test (VNT): Black rhinoceros (?), eland (?), gerenuk (?), kudu (?), impala (?), Thomson gazelle (?). VNT titres in positive sera ranging from log10101.9 (?) to log10103.1 (?). 2.1.3.3 Cut-off values and diagnostic accuracy Threshold values for the recNP IgG I-ELISA were derived from data sets dichotomised according to the results of the VN test (Table 2.1.1). The effect of differently determined cut-off values on distinguishing between sera which tested negative or positive in this assay, and consequently on estimates of sensitivity, specificity, and other estimates of diagnostic accuracy is given in Table 2.1.2 and figure 2.1.5. Optimisation of cut-off values using the misclassification cost term option of the TG- ROC analysis was based on the non-parametric programme option (Greiner, 1996) due to departure from a normal distribution of data sets analysed. Graphical presentation of the TG-ROC analysis for African buffalo is shown in figure 2.1.4. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 40 P. Jansen van Vuren 0.0 0.1 0.2 0.3 0.4 0.5 0 20 40 60 80 100 120 140 Cut-off (PP) M CT Figure 2.1.4. Optimisation of the cut-off for the recNP I-ELISA in African buffalo using the misclassification cost term (MCT) option of the two-graph receiver operating characteristic analysis (TG-ROC). The two curves represent MCT values based on non-parametric (?) or parametric (- - -) estimates of sensitivity and specificity derived from data sets in field- collected sera. Optimisation of the cut-off value was based on the non-parametric (?) program option due to departure from a normal distribution of data sets analysed. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 41 P. Jansen van Vuren Table 2.1.2. Diagnostic accuracy of Rift Valley fever recNP-based I-ELISA in African wildlife. Species Cut-offa D-Snb D-Spc Yd Efe PPVf NPVg African buffalo 26.94h 98.7 99.4 0.98 99.3 92.5 99.9 17.73i 100 94.2 0.94 94.6 55.0 100 22.23j 100 97.8 0.98 97.9 77.9 100 Black rhinoceros 33.6 100 100 1 100 100 100 27.5 100 91.3 0.91 93.7 81.5 100 35.5 100 100 1 100 100 100 Common zebra -k - - - - - - 13.9 - 100 - - - - 17.9 - 100 - - - - Elephant 28 100 100 1 100 100 100 10.6 100 95.8 0.96 99.7 99.7 100 13.6 100 97.2 0.97 99.8 99.8 100 Giraffe - - - - - - - 11.7 - 100 - - - - 14.3 - 100 - - - - Gravy zebra - - - - - - - 17.3 - 100 - - - - 22.5 - 100 - - - - Warthog 27.7 100 100 1 100 100 100 13.5 100 95.9 0.96 96 49.7 100 17.5 100 97.9 0.98 98 66.4 100 Antelopesl 20.4 100 99.8 0.99 99.7 95.6 100 8.4 100 88.1 0.88 88.7 32.4 100 14.4 100 97.0 0.97 97.1 67.9 100 Animals were categorized according to the results of virus neutralization test (VNT) a Cut-off value expressed as percentage positivity (PP) of an internal high-positive serum control. b Diagnostic sensitivity (%). c Diagnostic specificity (%). d Youden?s index. e Efficiency (%). f Positive predictive value (%). g Negative predictive value (%). h Cut-off value optimised by TG-ROC analysis. i Cut-off value based on mean + 2 S.D. of ELISA PP values in VNT-negative population. j Cut-off value based on mean + 3 S.D. of ELISA PP values in VNT-negative population. k Not determined due to unavailability or very limited number of VNT-positive sera l Eland, gerenuk, hartebeest, impala, kudu, Thomson gazelle, waterbuck Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 42 P. Jansen van Vuren 0 20 40 60 80 100 120 140 160 1 101 201 301 401 501 601 701 801 901 Number of sera tested EL IS A PP va lu e Figure 2.1.5. Distribution of recNP IgG I-ELISA PP values in African buffalo tested positive (n = 77, gray area) and negative (n = 946, black area) in the virus neutralisation test. Sera ordered according to ELISA PP values. Horizontal lines indicate the ELISA cut-off values determined by the TG-ROC analysis (?), and as a mean plus three (- - -) and two (? ? ?) standard deviations of the ELISA PP values observed in the VNT-negative population. 2.1.4 Discussion The VN test is the golden standard test for serological diagnosis of RVF because it can be used for any species and is highly sensitive and specific (Swanepoel et al., 1986a). It is, however, laborious, expensive and requires the amplification of live virus which poses a significant safety risk to laboratory personnel and thus restricts its use to high biocontainment facilities. For these reasons it is not widely used which creates the need for alternative, safer assays. The ELISA offers a quick, safe and less expensive alternative to the VNT. In particular the I-ELISA is one of the simplest immunoassays for antibody detection but requires highly pure antigen preparations for coating plates. A recNP based I- ELISA was recently developed and so-far validated for the detection of RVF specific IgG antibodies humans (Jansen van Vuren et al., 2007, Paweska et al., 2007). Because of the important role wildlife seem to play in the inter-epizootic maintenance of RVFV, there was a need for a safe, reliable and properly validated assay for detection of IgG in wildlife animals. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 43 P. Jansen van Vuren In this study the recNP IgG I-ELISA had high estimates of sensitivity and specificity for the detection of antibodies in wildlife animals. The diagnostic performance of the assay was consistently lower when using traditional methods for cut-off determination (mean values of known negatives plus 2 or 3 standard deviations) when compared to the TG-ROC method. It was necessary to determine separate cut-off values for each species, or group of similar animals (e.g. antelopes), to optimize diagnostic accuracy for a targeted wildlife species. Establishing a single cut-off for all species would make the assay less complex to interpret and use but would have a negative effect on diagnostic performance which is more important. Apart from assay sensitivity and specificity, other statistical parameters were also used to evaluate assay performance. The Youden?s index is another statistic that aims to capture assay performance by including sensitivity and specificity. Positive and negative predictive values indicate the ability of the test to accurately identify true positives and negatives as positives and negatives respectively. Although diagnostic sensitivity and specificity is the most used parameters in diagnostic assays, these other statistical values can give further confidence in the performance of the assay. Based on the results from this study, the recNP IgG I-ELISA has the potential to be a safe, quick and cheap tool for the detection of IgG antibodies in African wildlife species to aid in the monitoring of inter-epizootic transmission of RVFV. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 44 P. Jansen van Vuren 2.2 Conjugation of the recombinant nucleocapsid protein with horseradish peroxidase for detection of IgM antibody to Rift Valley fever virus in humans 2.2.1 Introduction The symptoms of RVF in humans are not very specific and can therefore not be easily diagnosed clinically which emphasizes a need for quick and accurate techniques for RVF diagnosis in humans. Molecular techniques are sensitive and rapid but viremia in RVFV infected individuals is transient and therefore they are only useful for a limited time and very shortly after infection when testing blood samples. It is therefore recommended that serological tests for detection of specific IgM be done concurrently with antigen/RNA detection techniques. RVFV-infected patients usually develop anti-RVFV IgM antibodies within 6 days of exposure, making IgM detection a useful tool for diagnosis of recent infections (Paweska et al., 2005a). To expand on the successful development and validation of an indirect IgG ELISA based on the RVFV recNP (refer to chapter 2.4), this chapter describes the development and evaluation of an IgM detecting ELISA based on the recNP antigen conjugated to the HRPO enzyme. 2.2.2 Materials and methods 2.2.2.1 Horseradish peroxidase conjugation of the recNP The recNP was produced as described before (section 2.1.2.4) but additionally the protein was concentrated and salts removed by using a Vivaspin 5000 kDa molecular weight cut-off (MWCO) ultrafiltration spin column (Sartorius-Stedim Biotech, Germany) as recommended by the manufacturer. The recNP, at 0.8 mg/ml in PBS buffer, pH 7.2, was conjugated to the horseradish peroxidase enzyme by using the LYNX Rapid HRP conjugation kit (ABD Serotec, United Kingdom) as prescribed by the manufacturer. Briefly, LYNX Modifier reagent was added to the recNP at a ratio of 1:10 (v/v) and gently mixed. The resulting mixture was added to lyophilized HRPO at different ratios to determine the optimal conjugation ratio [4:1, 2:1 and 1:1 (weight recNP/weight HRPO)]. The mixture was incubated at room temperature for 4 hours, and after incubation LYNX Quencher reagent was added at a ratio of 1:10 (v/v) of the original recNP volume, mixed and stored at 4?C until use. Once the optimal recNP/HRPO ratio was determined, conjugation was repeated to produce bulk recNP-HRPO for further testing. 2.2.2.2 Enzyme-linked immunosorbent assay (recNP-HRP IgM ELISA) Maxisorb immunoplates (Nunc, Denmark) were coated with 100?l goat anti-human IgM ?- chain (Zymed Laboratories, Inc.) diluted 1:500 in phosphate-buffered saline (PBS) pH 7.2 and Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 45 P. Jansen van Vuren incubated overnight at 4?C. After washing three times with a washing buffer consisting of phosphate- buffered saline (PBS) pH 7.2 and 0.1% Tween 20, the plates were blocked with 200 ?l of 10% fat-free milk powder in PBS and incubated in a moist chamber for 1 h at 37?C and then washed as described before. Control and test sera were diluted 1:400 in PBS containing 2% milk powder (diluting buffer) and 100 ?l of diluted sera was added to the plates. Each test serum was assayed in duplicate and each internal control was tested in quadruplicate. After incubation in a moist chamber for 1 h at 37?C, plates were washed six times with the washing buffer and 100 ?l of a 1:100 dilution of the recNP-HRPO added to the plates. Plates were incubated for 1 h at 37?C, washed six times, and 100 ?l of 2,2?-azino di-ethyl-benzothiazoline-sulfonic acid substrate was added to each well. Plates were then incubated in the dark at room temperature for 30 minutes. The reactions were stopped by the addition of 100 ?l of 1% SDS and OD values were determined at 405 nm. OD values of test sera were converted into percentages of the high-positive control serum (PP value). PP values were calculated as follows: PP = (OD serum / mean OD C++) x 100. 2.2.2.3 Serum controls Internal serum controls were prepared as described previously (Paweska et al., 2005a). 2.2.2.4 Human serum specimens A total of 257 individual human sera collected in Kenya in 2007, and South Africa in 2008/09 were used. Sera which tested negative in the RVF IgM Capture ELISA (Paweska et al., 2005a) were regarded as a reference panel from non-infected individuals (n = 219 humans), and sera which tested positive as a reference panel from individuals recently infected with RVFV (n = 38 humans). 2.2.2.5 Selection of cut-off values and determination of diagnostic performance Cut-off values of the recNP-HRP IgM ELISA was determined as described before (section 2.1.2.5). The following criteria were used to evaluate the diagnostic performance: sensitivity (D-Sn) = [Tp/(Tp + Fn)] ? 100; specificity (D-Sp) = [Tn/(Tn+Fp)]?100; Youden?s index (Y) = [Sn+(Sp?1)]; efficiency (Ef) = (D-Sn+D-Sp)/2; positive predictive value (PPV) = TP/(TP + FP); negative predictive value (NPV) = TN/(FN + TN) where TP is true positives, FP is false positives, FN is false negatives and TN is true negatives. 2.2.3 Results 2.2.3.1 Optimal conjugation ratio and recNP-HRP dilution for ELISA Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 46 P. Jansen van Vuren The optimal ratio at which to conjugate recNP with the peroxidase enzyme was determined to be 1:1 (weight recNP / weight HRP) (Figure 2.2.1). Dilution of the recNP-HRP at 1:100 yielded the best discrimination between high positive, low positive and negative control sera (Figure 2.2.2). 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1:100 1:200 1:400 1:800 1:1600 1:3200 1:6400 1:12800 recNP-HRP dilution EL IS A OD v al u e (40 5 n m ) Figure 2.2.1. Titration curves of three different recNP-HRP preparations with RVF IgM strong positive human serum. The recNP/HRP ratios during conjugation (w/w) were as follows: ratio 4:1 (??? ); ratio 2:1 (--?--); and ratio 1:1 (???). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 47 P. Jansen van Vuren 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1:100 1:200 1:400 1:800 1:1600 1:3200 recNP-HRP dilution EL IS A OD v al u e (40 5n m ) C++ C+ C- Figure 2.2.2. Titration curves of recNP-HRP with three different human IgM controls: C++ strong positive control (???); C+ low positive control (???); and negative control (--x--). 2.2.3.2 Cut-off values and diagnostic accuracy Threshold values for the recNP IgG I-ELISA were derived from data sets dichotomised according to the results of the IgM capture ELISA (Paweska et al., 2005a). The effect of differently determined cut-off values on distinguishing between sera which tested negative or positive in this assay, and consequently on estimates of sensitivity, specificity, and other estimates of diagnostic accuracy is given in table 2.2.1. Optimisation of cut-off values using the misclassification cost term option of the TG-ROC analysis was based on the non-parametric programme option (Greiner, 1996) due to departure from a normal distribution of data sets analysed. Graphical presentation of the TG- ROC analysis is shown in figure 2.2.3. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 48 P. Jansen van Vuren Table 2.2.1. Diagnostic accuracy of Rift Valley fever recNP-HRPO IgM ELISA in positive (n = 38) and negative (n = 219) human specimens as catagorized according to the results of the IgM capture ELISA. Cut-offa D-Snb D-Spc Yd Efe PPVf NPVg 27.06h 81.58 95.90 0.77 88.73 80.85 96.90 29.40i 78.95 97.72 0.77 88.33 88.37 96.48 36.10j 71.05 98.17 0.69 84.61 90.48 95.22 a Cut-off value expressed as a percentage positivity (PP) of an internal high-positive serum control. b Diagnostic sensitivity (%). c Diagnostic specificity (%). d Youden?s index. e Efficiency (%). f Positive predictive value (%). g Negative predictive value (%). h Cut-off value optimised by TG-ROC analysis. i Cut-off value based on mean + 2 S.D. of ELISA PP values in negative population. j Cut-off value based on mean + 3 S.D. of ELISA PP values in negative population. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 49 P. Jansen van Vuren Figure 2.2.3. Optimisation of the cut-off for the recNP-HRPO IgM ELISA in humans using the misclassification cost term (MCT) option of the two-graph receiver operating characteristic analysis (TG-ROC). The two curves represent MCT values based on non-parametric (?) or parametric (- - -) estimates of sensitivity and specificity derived from data sets in field-collected sera. Optimisation of the cut-off value was based on the non-parametric (?) program option due to departure from a normal distribution of data sets analysed. 2.2.4 Discussion A capture ELISA format was developed recently for the confirmation of a recent infection with RVFV in humans by detection of IgM. This capture ELISA was validated against the golden standard method, virus neutralization, and displayed high diagnostic sensitivity and specificity (Paweska et al., 2005a). It is, however, based on whole virus antigen which needs to be prepared in high biocontainment laboratory. An indirect ELISA based on a completely safe recombinant antigen of RVFV, the nucleocapsid protein, was validated recently for the detection of IgG in humans (Paweska et al., 2007). Although the indirect ELISA format, with recNP coated on the plate, was initiall evaluation for the detection of IgM in human sera, it yielded a high false-positivity rate, probably because of interfering rheumatoid factor (results not shown). This study describes attempts to conjugate the recNP with the HRPO enzyme and the subsequent development and preliminary validation of a RVFV recNP-HRPO IgM ELISA. In this format IgM is first captured from the sera to Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 50 P. Jansen van Vuren minimize rheumatoid factor interference, after which HRPO conjugated recNP antigen is added as a detection system. The best specificity of the assay was achieved when using the mean plus 3 standard deviations method, but this was at the cost of lower sensitivity. The method yielding the most sensitive cut-off was the misclassification cost term option (MCT) of the two-graph receiver operating characteristic analysis (TG-ROC) method but this was at the cost of specificity. Although this assay has some promise as a diagnostic assay based on its high specificity, it still requires further optimization to improve the sensitivity. The assay can be further improved by further improving the conjugation of the RVFV recNP to HRPO to obtain a more concentrated product, or using monoclonal antibodies to capture human IgM from specimens. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 51 P. Jansen van Vuren 2.3 Comparative evaluation of ELISA-based techniques for detection of antibodies against RVFV* * Partially published as: Van Vuren, P. J. & Paweska, J. T. (2010). Comparison of enzyme-linked immunosorbent assay-based techniques for the detection of antibody to Rift Valley fever virus in thermochemically inactivated sheep sera. Vector-Borne and Zoonotic Diseases 10, 697-9.0 * Presented at international conferences as: Jansen van Vuren, P., Paweska, J.T. (2009). A comparative evaluation of ELISA-based techniques for serodiagnosis of Rift Valley fever. Annual meeting of the ARBO-ZOONET network, St. Raphael, France, 30 September 2009. (Oral presentation). Jansen van Vuren, P., Paweska, J.T. (2009). Comparison of ELISA-based techniques for serodiagnosis of Rift Valley fever. 5th European Meeting on Viral Zoonoses, St. Raphael, France, 26 ? 29 September 2009. (Poster presentation). 2.3.1 Introduction In recent years numerous new RVF diagnostic techniques have been developed and validated, including molecular assays (Le Roux et al., 2009) for RNA detection and ELISAs for antibody detection (Paweska et al., 2003a, Paweska et al., 2005a, Paweska et al., 2005b, Jansen van Vuren et al., 2007, Paweska et al., 2007). The abovementioned ELISAs were all developed and validated separately and direct comparison was never undertaken. The diagnostic performance various ELISAs, as taken from the published literature, is summarized in table 2.3.1 (ELISAs for human diagnosis) and table 2.3.2 (ELISAs for livestock diagnosis). These ELISAs are based on gamma-irradiated reagents and/or recombinant antigens and are thus regarded safe. A simple thermo-chemical inactivation method for RVFV was developed (section 3.2.3) which would render these tests completely safe to conduct outside biocontainment facilities. The effect of the inactivation on detectable antibodies, however, is not known. This sub-chapter describes the direct comparison of four livestock ELISAs for anti-RVFV antibody detection using a well characterized panel of sera collected from experimentally infected sheep, as well as an evaluation of the effect of a thermo-chemical RVFV inactivation step on detectable antibodies. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 52 P. Jansen van Vuren 2.3.2 Materials and methods 2.3.2.1 Enzyme-linked immunosorbent assays The following ELISAs were directly compared using serial bleeds from experimentally infected sheep: IgG-sandwich ELISA (Paweska et al., 2003a), IgM-capture ELISA (Paweska et al., 2003a), inhibition ELISA (Paweska et al., 2005b) and an indirect ELISA based on the recombinant RVFV N protein (Jansen van Vuren et al., 2007) (also see section 2.2 of this thesis). 2.3.2.2 Experimental sheep sera Serial sera were obtained from three sheep experimentally infected with wild type RVF virus as described previously (Le Roux et al., 2009). 2.3.2.3 Thermo-chemical inactivation of sera Sheep sera were inactivated as described in chapter 3 (section 3.2.3). Briefly, an equal volume of 1% Tween20 in PBS was added to each serum and incubated at 56?C for 1 hour. Inactivated sera were tested for complete inactivation on 24 - 48h old Vero cell monolayers and in 2-3 day old suckling mice. Cells were monitored for cytopathic effect (CPE) until 14 days after inoculation and mice until 10 days p.i. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 53 P. Jansen van Vuren a Indicates the complete immunocomplex formed on the ELISA plate in the case of a positive reaction bMouse anti-serum against: Saint-Floris (Phlebovirus), Gordil (Phlebovirus), Arumowot (Phlebovirus), Gabek Forest (Phlebovirus), Nairobi sheep disease (Nairovirus), Hazara (Nairovirus), Crimean-Congo hemorrhagic fever (Nairovirus), Akabane (Orthobunyavirus), Bunyamwera (Orthobunyavirus), Shuni (Orthobunyavirus) and Bhanja viruses (unassigned, Bunyaviridae family). c Cut-off value at 95% accuracy level optimized using the misclassification cost term option (Greiner, 1996) of the two-graph receiver operating characteristics analysis (Greiner, 1995; Greiner, Sohr and G?bel, 1995). d Se = [Tp/(Tp + Fn)] x 100 where Tp is true positives and Fn is false negatives. e Sp = [Tn/(Tn + Fp)] x 100 where Tn is true negatives and Fp is false positives. f PPV = [(P)(Se)]/[(P)(Se)] + [(1 ? P)(1 ? Sp)] x 100 where P is the prevalence g NPV = [(1 ? P)(Sp)]/[(1 ? P)(Sp)] + [(P)(1 ? Se)] x 100 h Percentage positivity i Percentage inhibition Table 2.3.1 Antibody detection ELISAs for RVF diagnosis in humans ELISA set-up Indirect ELISA Sandwich ELISA Capture ELISA Inhibition ELISA Diagrama Species specific, Anti-IgG HRPO conjugated antibody ? IgG antibody in specimen ? Pure RVFV antigen bound to plate surface Species specific, Anti-IgG HRPO conjugated antibody ? IgG antibody in specimen ? Crude RVFV antigen ? Hyperimmune anti-RVFV serum Species specific, Anti-IgG HRPO conjugated antibody ? Hyperimmune anti-RVFV serum ? Crude RVFV antigen ? IgM antibody in specimen ? Anti-IgM capturing antibody Species specific, Anti- IgG HRPO conjugated antibody (unbound) Hyperimmune anti- RVFV serum (out-competed) IgG/IgM antibody in specimen ? Crude RVFV antigen ? Hyperimmune anti- RVFV capturing serum Antigen Bacterially expressed recombinant nucleocapsid Sucrose acetone extracted whole virus from mouse liver Sucrose acetone extracted whole virus from mouse liver RVF infected Vero cell supernatant Antibody detected IgG IgG IgM Total Antibody Number of specimens n = 2969 n = 2400 n = 1396 n = 1367 Cross-reactivity evaluatedb Yes No No No TG-ROC cut-offc 28.98 PPh 13.21 PP 7.1 PP 38.6 PIh Sensitivityd 99.72 % 100.0 % 96.47 % 99.47 % Specificitye 99.62 % 99.95 % 99.44 % 99.66 % Positive predictive valuef 97.14 % 99.50 % 96.55 % 97.97 % Negative predictive valueg 99.66 % 100.0 % 99.54 % 99.91 % Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 54 P. Jansen van Vuren Table 2.3.2 Antibody detection ELISAs for RVF diagnosis in livestock Species Ovine Caprine Bovine Camel ELISA set-up Sandwich ELISA Capture ELISA Inhibitio n ELISA Sandwich ELISA Capture ELISA Inhibition ELISA Sandwich ELISA Capture ELISA Inhibition ELISA Inhibition ELISA Antigen Sucrose acetone whole virus Sucrose acetone whole virus RVF Vero cell supernata nt Sucrose acetone whole virus Sucrose acetone whole virus RVF Vero cell supernatant Sucrose acetone whole virus Sucrose acetone whole virus RVF Vero cell supernatant RVF Vero cell supernatant Antibody detected IgG IgM Total Antibody IgG IgM Total Antibody IgG IgM Total Antibody Total Antibody Number of specimens n =1321 n =1321 n = 493 n =1459 n =1459 n = 806 n = 997 n = 997 n = 694 n = 156 TG-ROC cut-off 13.2 PP 7.9 PP 38.4 PI 18.8 PP 9.5 PP 41.4 PI 16.4 PP 14.3 PP 41.9 PI 36.1 PI Sensitivity 99.05 % 100.0 % 100.0 % 100.0 % 97.40 % 99.56 % 96.34 % 99.20 % 100.0 % 100.0 % Specificity 99.10 % 99.40 % 99.29 % 99.90 % - 99.65 % 99.67 % - 99.52 % 100.0 % Positive predictive value 91.19 % - 95.78 % 99.69 % - 99.15 % 96.61 % - 95.35 % 100.0 % Negative predictive value 99.90 % - 100.0 % 100.0 % - 99.81 % 99.66 % - 100.0 % 100.0 % Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 55 P. Jansen van Vuren 2.3.3 Results The immune responses in sheep after experimental infection with wild-type RVFV were monitored using the IgG sandwich, indirect IgG, IgM capture and an inhibition ELISA. There was no significant difference in detection of antibodies between untreated versus inactivated serum using any of the ELISAs but these assays differed in their ability to detect the early humoral responses to infection with RVFV (Figure 2.3.1). The IgM-capture ELISA was able to detect seroconversion on day 4 post-infection (p.i.) compared to day 5 p.i. with the IgG-sandwich ELISA. The inhibition ELISA yielded false-positive results on day 2 and 3 p.i. as a result of the capturing of viral antigen in highly viremic sera on days 2 and 3 p.i. (circle in figure 2.3.1). The recombinant N protein-based IgG ELISA, using Protein G HRPO, was less sensitive in detecting seroconversion (day 9 p.i.) as compared to the IgG-sandwich ELISA (day 5 p.i.). This problem was alleviated when replacing Protein G with anti- sheep IgG HRPO (Figure 2.3.1). -20 0 20 40 60 80 100 120 140 160 180 200 220 240 260 0 1 2 3 4 5 6 7 9 12 14 16 19 23 26 40 55 70 84 114 Days post infection EL IS A PP o r PI v al u e (% ) False-positive result (Inhibition ELISA) Figure 2.3.1. Comparison of mean immune responses in three experimentally infected sheep as measured by testing na?ve (solid lines) versus thermo-chemically inactivated (dotted lines) sera by IgM- capture ELISA (?), inhibition ELISA (x), indirect ELISA Protein G HRPO (?), indirect ELISA anti-sheep IgG HRPO (?), and IgG- sandwich ELISA (?). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 56 P. Jansen van Vuren 2.3.4 Discussion The thermo-chemical inactivation step had no adverse effect on the detection of antibody in any of the ELISAs tested and renders RVFV completely inactive, as evidenced by safety testing in suckling mice and tissue culture. This makes it a practical method for use in the field or in laboratories with limited biocontainment facilities. The most sensitive assay for detection of antibodies early after infection was the IgM capture ELISA. This was expected since the early humoral response to infection is IgM. The IgM capture ELISA would therefore be the most suitable test for diagnosis of recent RVFV infections. There was a slight decrease in sensitivity of the recNP based indirect ELISA when the pseudoimmunogen Protein G was used as an HRPO conjugate in the assay instead of the species specific conjugate. It is possible that the early IgG antibody has lower afinity for Protein G. Using Protein G, which is not species specific, instead of a species specific HRPO conjugate, however, offers the opportunity to use the same assay format for detection of antibodies in various species. In addition the likelyhood of testing a large number of samples that were collected so early after infection is low. The recNP based indirect ELISA would therefore be suitable for serological surveys, testing of immune status after vaccination and import/export testing. An additional advantage of the recNP based I-ELISA is that it can be used for the differentiation of naturally infected and vaccinated animals (DIVA) when vaccines are based on the other structural proteins of RVFV (McElroy et al., 2009, de Boer et al., 2010). The RVFV inhibition ELISA, which detects total antibodies IgG/IgM, was recently evaluated by a European group and found to be highly sensitive and specific for testing European ruminant sera (Cetre-Sossah et al., 2009). It was surprising, however, to note false-positive results with the inhibition ELISA when testing sera collected early after infection of sheep. Upon closer inspection of the assay format, it was concluded that the ELISA set up allows for capture any RVFV antigens present in the experimental sera during viremia and thus it will yield false-positive results before seroconvertion. This intrinsic characteristic of the test has an impact on how the results should be interpreted, especially if it is used during RVF outbreaks where viremic individuals are likely to be encountered. It also stresses the importance of understanding the basis of each assay for correct interpretation of results. However, taking into account the short viremia during RVFV infection the practical consequence of this seems to be neglible. This is the first direct comparison of validated ELISA techniques for RVF serological diagnosis which highlights the difference in characteristics and design of each assay and their impact on the interpretation of results. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 57 P. Jansen van Vuren 2.4 Detection of IgG antibody to Rift Valley fever virus in humans* * Published as: Paweska, J. T., Jansen van Vuren, P., and Swanepoel, R. (2007). Validation of an indirect ELISA based on a recombinant nucleocapsid protein of Rift Valley fever virus for the detection of IgG antibody in humans. Journal of Virological Methods 146(1-2), 119-24. 2.4.1 Introduction An IgG-sandwich ELISA was recently developed and validated using extended panels of well- characterized human sera (Paweska et al., 2005a). This ELISA is based on sucrose acetone extracted RVF whole virus and hyperimmune sera that were generated against live RVFV. Although these reagents are gamma-irradiated for safety before usage, they still need to be prepared in high biocontainment facilities before inactivation. The use of a recombinant antigen circumvents some of the issues that hamper the safe production of immunoreagents for serological diagnostic assays. The indirect ELISA is also a less time consuming and simple assay when compared to the sandwich format, which makes it more user-friendly and cost-effective. 2.4.2 Materials and methods 2.4.2.1 ELISA serum controls and internal quality control Freeze-dried, gamma-irradiated serum controls were produced as described previously (Paweska et al., 2005a). To assess inter- and intra-plate variation the means and standard deviations (S.D.) of the ELISA optical density and percentage positivity values (PP) were determined from replicates of the internal controls included in each plate and run of the assay during validation. Coefficient of variation values (CV%) were also determined for the positive serum controls [CV% = (S.D. of replicates / means of replicates) x 100]. Estimates of the assay repeatability and the upper and lower control limits for each internal control were determined from the resultant data. During routine runs of the assay each plate had four replicates of high positive (C++), low positive (C+), negative (C-) and the conjugate control (Cc). 2.4.2.2 Human serum panels A total of 2967 sera collected in Kenya (n = 982), South Africa (n = 1255), Tanzania (n = 360), Uganda (n = 210) and Zimbabwe (n = 160) were used. The South African and Zimbabwean sera represented post RVF outbreak specimens collected in the late 1970s and routine diagnostic submissions to the Special Pathogens Unit of the National Institute for Communicable Diseases (National Health Laboratory Services) for the period 1999 to 2005. East African sera were taken to Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 58 P. Jansen van Vuren monitor the 1997-98 outbreak of RVF in the region (Woods et al., 2002). Sera which tested negative in the virus neutralization test were regarded as a reference panel from non-infected individuals, whereas sera which tested positive as a reference panel from previously infected individuals. Cut-off value calculation and diagnostic accuracy determination were done using the IgG I-ELISA results obtained from these field collected sera. 2.4.2.3 Mouse ascetic fluids for cross-reactivity testing Hyperimmune mouse ascetic fluids generated against viruses representing the genus Phlebovirus, Nairovirus, Orthobunyavirus and Bhanja virus of the family Bunyaviridae as described before (Burt et al., 1993) were obtained from the serum bank of the Arbovirus section of the SPU- NICD/NHLS. 2.4.2.4 Virus neutralization test The virus neutralization test was done as described previously (section 2.1.2.3) 2.4.2.5 Antigen production and IgG I-ELISA procedure Antigen was produced and the I-ELISA procedure done as described previously (section 2.1.2.4), except that HRPO conjugated to goat anti-human IgG (H+L chain) was used for human specimens and HRPO conjugated to recombinant Protein G (Zymed Laboratories, Inc.) was used for mouse ascitic fluid. 2.4.2.6 Selection of cut-off values and determination of ELISA diagnostic accuracy Cut-off values were determined as described previously (section 2.1.2.5). Parameters of diagnostic accuracy were determined as described previously (section 2.1.2.5). 2.4.3 Results 2.4.3.1 Internal quality control and repeatability The upper and lower internal quality control limits and estimates of repeatability of the assay are summarized in Table 2.4.1. There was no excessive variation within and between routine runs of the assay, and the internal controls were constantly within upper and lower control limits during routine runs of the assay (Figure 2.4.1). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 59 P. Jansen van Vuren 2.4.3.2 Cross-reaction with hyperimmune mouse ascitic fluid The I-ELISA optical density (OD) value of the mouse IgG positive RVFV ascitic fluid was 1.52 while that of normal mouse ascitic fluid and the conjugate control was 0.072 and 0.068 respectively. The OD readings of hyperimmune ascitic fluids from mice experimentally infected with different viruses from the genus Phlebovirus, Nairovirus, Orthobunyavirus and Bhanja virus of the family Bunyaviridae were within the OD values for negative controls (Figure 2.4.2). These results demonstrate highly specific binding affinity of mouse IgG antibody against RVFV and the recNP of the virus and lack of cross-reaction between the recNP and IgG antibody against other Bunyaviruses assayed. Table 2.4.1. Internal quality control data and repeatability estimates for Rift Valley fever IgG I-ELISA based on recombinant nucleocapsid antigen IQC parameters IQCa limits UCLb LCLc OD C++ 1.7 0.81 PPd C++ 117 82 PP C+ 39 24 PP C- 9.8 5.4 PP Cc 7.2 3 Repeatabilitye Intra-plate variation C++ 5.98 ? 2.4 S.D. (2.72-9.83)f C+ 6.19 ? 3.25 S.D. (2.73-13.65) Inter-plate variation C++ 6.01 ? 1.54 S.D. (3.35-7.7) C+ 6.19 ? 2.33 S.D. (3.98-11.03) a Internal quality control (IQC) data were calculated from the mean ? 2 S.D. of 420 replicates of each control over seven runs including five plates. b Upper control limit c Lower control limit d Percentage positivity e Repeatability estimates for high positive (C++) and low positive (C+) serum controls were calculated as the %CV. f Range of %CV values Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 60 P. Jansen van Vuren 2.4.3.3 Cut-off values and diagnostic accuracy Cut-off values were optimized using the TG-ROC as described before (section 2.1.3.3). At a cut-off value of 28.98 PP the overall misclassification costs were minimal under assumption of 50% disease prevalence and equal costs of false-positive and false-negative test results. Graphical presentation of the effect of three differently determined threshold values on distinguishing between positive or negative sera is shown in Figure 2.4.3. At a cut-off optimized by TG-ROC at 95% accuracy level, the diagnostic sensitivity of the I-ELISA was 99.72% and diagnostic specificity 99.62% while estimates for the J and Ef were 0.993 and 99.62% respectively. When cut-off values were determined by traditional statistical approaches, the diagnostic sensitivity was 100% but estimates of J, Ef, PPV and NPV values were lower compared to those based on the TG-ROC cut-off (Table 2.4.2). 0 20 40 60 80 100 120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Number of runs EL IS A PP va lu e Figure 2.4.1. Upper (____) and lower (----) internal quality control limits for PP values of high- positive (?), low positive (?), negative (?) serum controls and conjugate control (?) and means ? S.D. for these controls during 27 routine runs of the assays over a period of 12 weeks. Two or three plates were used during each run with four replicates of each control on each plate. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 61 P. Jansen van Vuren 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 a b c d e f g h i j k l m n ELISA controls (a - c) and mouse antisera against selected bunyaviruses (d - n) EL IS A O D va lu e Figure 2.4.2. Cross-reactivity of recombinant nucleocapsid protein of RVFV in I-ELISA with mouse IgG antibody against selected viruses of the family Bunyaviridae. (a) Mouse IgG anti RVFV (Phlebovirus), (b) normal mouse ascitic fluid and (c) conjugate control. Mouse IgG anti: (d) Saint- Floris (Phlebovirus), (e) Gordil (Phlebovirus), (f) Arumowot (Phlebovirus), (g) Gabek Forest (Phlebovirus), (h) Nairobi sheep disease (Nairovirus), (i) Hazara (Nairovirus), (j) Crimean-Congo hemorrhagic fever (Nairovirus), (k) Akabane (Orthobunyavirus), (l) Bunyamwera (Orthobunyavirus), (m) Shuni (Orthobunyavirus) and Bhanja viruses (not assigned to a recognized genera of the family Bunyaviridae). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 62 P. Jansen van Vuren Table 2.4.2. Diagnostic accuracy of Rift Valley fever IgG I-ELISA based on recombinant nucleocapsid antigen Measurea Cut-off 28.98 PPb Cut-off 17.18 PPc Cut-off 21.38 PPd Sensitivity (%) VNT+=350e 99.72 100 100 Specificity (%) VNT-=2617f 99.62 95.3 97.54 Youden?s index 0.993 0.953 0.975 Efficiency (%) 99.62 95.84 97.82 Positive predictive value (%) 97.14 73.23 84.12 Negative predictive value (%) 99.66 100 100 Individuals were categorized according to the results of the virus neutralization test (VNT). b Cut-off value optimized by the misclassification cost term option of the two-graph receiver operating characteristics analysis at 95% accuracy level. c Cut-off value determined by mean plus two standard deviations derived from PP values in uninfected reference population. d Cut-off value determined by mean plus three standard deviations derived from PP values in uninfected reference population. e Number of sera tested positive in the VN test. f Number of sera tested negative in the VN test. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 63 P. Jansen van Vuren 0 20 40 60 80 100 120 140 160 180 200 1 351 701 1051 1401 1751 2101 2451 Number of sera tested EL IS A PP va lu es Figure 2.4.3. The effect of different ELISA cut-off values on the discrimination between human sera tested negative or positive in the VN test for antibodies against RVFV. Distribution of IgG I-ELISA PP values in human sera tested positive (n = 350, ?) or negative (n = 2617, ?) in the VN test. Sera ordered according to ELISA PP values. Horizontal lines: (?) cut-off value of 28.98 PP determined by TG-ROC analysis, (---) cut-off value of 21.37 PP determined by mean plus 3 S.D., and (???) cut-off value of 17.18 PP determined by mean plus 2 S.D. of ELISA PP values observed in negative sera. 2.4.4 Discussion The indirect ELISA is one of the simplest immunoassay techniques for the detection of antibodies, but its routine application is impeded by non-specific signals arising from the use of crude or semi-purified antigens (Gravell et al., 1977, Frazier and Shope, 1979). RVFV does replicate to high titres in cell cultures but production of purified and concentrated virus stocks by classical virological methods is expensive, time consuming and requires high biocontainment. The results from this chapter confirm earlier findings (Jansen van Vuren et al., 2007) that the recNP of RVFV binds readily to ELISA plates, generates minimal background and effectively differentiates sera with varying concentrations of IgG antibodies to the virus in humans. The I-ELISA presented here achieved high repeatability estimates within statistically pre-determined IQC limits. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 64 P. Jansen van Vuren In this study the virus neutralization test was used to classify individuals according to their RVF infection status. Infection with RVFV induces life long virus neutralizing antibody in humans (Findlay and Howard, 1951), and only one serotype of RVFV is known (Swanepoel and Coetzer, 2004). The NP is one of the most immunodominant viral proteins and appears to be highly conserved among members of the Bunyaviridae family (Swanepoel et al., 1986b, Vapalahti et al., 1995, Schwarz et al., 1996, Magurano and Nicoletti, 1999, Gauliard et al., 2006). The serological cross-reactivity results from this study indicates that infection with related African Phleboviruses and other viruses from the Bunyaviridae family should not hamper serodiagnosis of RVF based on the recNP, which corresponds with previous reports on cross-reaction between known African Phleboviruses (Swanepoel et al., 1986b). The antigenic specificities of antibodies measured by ELISA and the virus neutralization test differ (Swanepoel et al., 1986b), therefore it is expected that the ELISA based on inactivated whole virus will be more sensitive than the virus neutralization test which detects only antibodies against the RVFV glycoproteins. This may explain the slightly lower specificity of the I-ELISA in this study compared to virus neutralization test in the study population. The recNP I-ELISA has been shown to be more sensitive than the virus neutralization test in detecting early immune responses in experimentally infected sheep (Jansen van Vuren et al., 2007). The recNP I-ELISA had a higher estimate of diagnostic sensitivity when cut-offs were determined by traditional statistical methods (mean + 2S.D. and 3S.D.), but lower specificity and combined measurements of assay performance characteristics when compared to the cut-off determined by TG-ROC analysis. In conclusion, the I-ELISA based on the recNP is highly accurate and robust for the detection of specific IgG antibody against RVFV in human sera and can be used in the diagnosis of infection and sero-epidemiological studies. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 65 P. Jansen van Vuren CHAPTER THREE EVALUATION OF ANTI-RECOMBINANT NUCLEOCAPSID PROTEIN RABBIT AND SHEEP POLYCLONAL SERA AS IMMUNODIAGNOSTIC REAGENTS IN AN ANTIGEN DETECTION SANDWICH ELISA * * Published as: Jansen van Vuren, P. & Paweska, J. T. (2009). Laboratory safe detection of nucleocapsid protein of Rift Valley fever virus in human and animal specimens by a sandwich ELISA. Journal of Virological Methods 157, 15-24. * Partially presented at international and local conferences as: Jansen van Vuren, P. & Paweska, J.T. (2009). Safe detection of Rift Valley fever virus in human and animal specimens by a sandwich ELISA. International Meeting on Emerging Diseases and Surveillance, Vienna, Austria, 13 ? 16 February 2009. (Poster presentation). Jansen van Vuren, P. & Paweska, J.T. (2008). A safe antigen detection ELISA for rapid diagnosis of Rift Valley fever. University of the Witwatersrand, Faculty of Health research day, Johannesburg, South Africa, 20 August 2008. (Oral presentation). Jansen van Vuren, P. & Paweska, J.T. (2008). Laboratory safe detection of nucleocapsid protein of Rift Valley fever virus in human and animal specimens by a sandwich ELISA. National Institute for Communicable Diseases Academic Day, 11 November 2008. (Poster presentation). 3.1 Introduction Various methods exist for the detection of antibodies against RVFV, but detection of antibodies only become possible after seroconversion which is usually between 2-4 days after the onset of viremia. There is therefore a period after RVFV infection where individuals would test negative with serological methods when in fact they are infected. Virus isolation is expensive, laborious and requires the propagation of live virus, necessitating the use of biocontainment facilities. Various molecular techniques have been developed to enable detection of virus genetic material (Jupp et al., 2000, Drosten et al., 2002, Peyrefitte et al., 2008, Le Roux et al., 2009). Molecular methods are highly sensitive and specific but also expensive and highly specialized. ELISA is a robust technique that requires less specialized equipment and training, and is probably in use in most diagnostic laboratories around the world. Some ELISAs for the detection of RVFV antigens have been reported (Niklasson et al., 1983, Meegan et al., 1989, Zaki et al., 2006) but these assays are based on reagents that are difficult and expensive to produce, pose a biohazard risk to Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 66 P. Jansen van Vuren laboratory personnel, and have not been properly validated. This chapter describes the first development and validation of a sandwich ELISA for antigen detection based on polyclonal antibodies specifically generated against the major viral antigen, the nucleocapsid protein, and the use of thermo- chemical inactivation of specimens to ensure safety of laboratory personnel. 3.2 Materials and methods 3.2.1 Generation of rabbit and sheep hyperimmune sera against the RVFV recNP The recNP was produced as described before (Chapter 2.1.2.4). The pET32 control antigen was produced in the same way but using wild-type pET32(a)+ vector without the RVFV NP-gene insert. Three New Zealand white rabbits were immunized subcutaneously (s.c.) with 140?g recNP emulsified in an equal volume of TiterMax Gold? adjuvant (Sigma-Aldrich, USA) according to the manufacturer?s instructions. Rabbits received an identical booster inoculation 14 days later followed by another booster of 375?g recNP without adjuvant on day 33 after the first immunization. Two Dorper cross sheep were immunized s.c. with 350?g recNP in TiterMax Gold adjuvant, emulsified as described above. Sheep received an identical booster inoculum on day 21 after the first immunization. Blood was taken regularly from immunized animals to monitor their responses to immunization. When the animals? responses yielded optical density (OD at 405nm) readings higher than 2.0 on a recNP I-ELISA at 1:400 dilution they were regarded as hyperimmune and bulk serum collected. Bulk sera from individual animals were pooled together to obtain homogenous un-purified preparations of polyclonal rabbit- and sheep anti-recNP respectively. 3.2.2 Sandwich ELISA procedure The top half of a high protein binding plate (Maxisorb, Nunc, Denmark) was coated with sheep anti-recNP hyperimmune serum (capture antibody) and the bottom half with normal sheep serum, both at dilution 1:400 in PBS, pH 7.2 and incubated overnight at 4?C. After washing three times with washing buffer (PBS, pH 7.2 and 0.1% Tween-20), plates were blocked with 200?l of 10% fat-free milk powder (?Elite?, Clover SA, Pty, Ltd.) in PBS, incubated in a moist chamber at 37?C for 60 minutes and washed as described above. RVFV recNP stock antigen diluted 1:3000 in 2% milk powder (diluting buffer) was used as a high positive control; 100?l of the diluted antigen was added in quadruplicate to the top and bottom half of the plates. RVFV recNP stock antigen, diluted 1:30,000 was used as a low positive control antigen and pET32 antigen diluted 1:3000 as negative control antigen; 100?l of each was added in duplicate to the top and bottom halves of each plate. A volume of 100?l of each specimen, inactivated as determined in 3.2.3, was added undiluted and in duplicate to the Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 67 P. Jansen van Vuren top and bottom halves of each plate. The plates containing internal controls and specimens were incubated at 37?C for 60 minutes in a moist chamber, washed as before and 100?l of hyperimmune rabbit anti-recNP serum (detecting antibody), diluted 1:3000 in diluting buffer, added to each well. After incubation at 37?C for 60 minutes in a moist chamber, plates were washed as before and 100?l of goat anti-rabbit IgG (H+L) HRPO conjugate (Zymed Laboratories, USA) diluted 1:8000 added to each well. After incubating as before plates were washed as before and 100?l of 2?2- azinodiethylbenzthiazoline sulfonic acid (ABTS, KPL Laboratories, USA) peroxidase substrate added to each well. Plates were incubated at room temperature in the dark for 30 minutes after which the reaction was stopped by the addition of 100?l of 1% sodium dodecyl sulphate (SDS, Sigma-Aldrich, USA) to each well. Optical density (OD) was determined at 405nm wavelength and results expressed as percentage positivity of the mean high-positive control antigen (PP) using the formula: (mean net OD of duplicate test specimen/mean net OD of high positive control) x 100. 3.2.3 Inactivation of specimens and safety testing Three regularly used laboratory detergents were evaluated together with heat for their ability to increase antigen detection efficiency in the sAg-ELISA. The detergents TritonX-100, NP40 and Tween-20 were mixed, each at 1%, with PBS pH 7.2 or carbonate/bicarbonate buffer pH 9.6 and evaluated as inactivation buffers. Normal sheep serum was then spiked with RVFV Ar20368 RSA 81 strain to a final concentration of 105.8 TCID50/ml and inactivated with each inactivation buffer (equal volumes of spiked serum and inactivation buffer) for 60 minutes at 56?C. Spiked serum was also inactivated at 56?C for 60 minutes without the presence of detergent. As a no-treatment control, RVFV spiked serum was added to an equal volume of PBS without detergent and not subjected to heat inactivation. As a negative control, negative serum without virus was treated the same as the no- treatment control. These preparations were tested on the sAg-ELISA. The optimal inactivation protocol (1% Tween-20 in PBS + 56?C for 60 minutes) was safety tested in 2-3 days old suckling mice and Vero cell monolayers. Mice were monitored for clinical symptoms until day 10 p.i. and Vero cells were monitored for CPE for a period of 14 days p.i. To control for all conditions, the following controls were also set-up: RVFV spiked sheep serum inactivated just by heat and spiked serum treated just with Tween-20. 3.2.4 Antigen detection in animal and insect specimens Heart, lung, liver, kidney and brain tissues were harvested from three female BALB/c mice on day 2 after subcutaneous inoculation with the SPU22/118 KEN 07 strain of RVFV. The same tissues were harvested from a mock inoculated BALB/c mouse. Diagnostic submissions of liver, heart, kidney, Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 68 P. Jansen van Vuren lung, and brain tissues from three aborted buffalo fetuses during the 2008 RVF outbreak in South Africa (Paweska et al., 2008a) were used. Animal tissues were homogenized as 10% (w/v) suspensions in Eagles Minimal Essential Medium (EMEM) (BioWhitaker,MD, USA) containing L-glutamine, non- essential amino acids and antibiotics (100 IU penicillin, 100?g streptomycin, and 0.25?g amphotericin B). After centrifugation at 3000?g, supernatants were harvested and stored at ?70?C. Homogenates (10%, w/v) of uninfected ovine and bovine liver and spleen tissues, prepared as described above, and uninfected human, sheep and cattle sera were spiked with the Ar20368 RSA 81 strain of RVFV to a final virus concentration of 106.5 TCID50/ml. Half log10 dilutions of these preparations were used to determine the sAg-ELISA analytical detection limit in tissues and sera. As controls, the unspiked homogenates and sera were used with EMEM in place of virus suspension. Homogenates (10%, w/v) of mosquito pools, each containing 100 individuals of Anopheles arabiensis, A. gambiae and A. funestus, obtained from laboratory mosquito colonies at Vector Control Unit of the National Institute for Communicable Diseases, were prepared in EMEM and spiked with the Ar20368 RSA 81 strain of RVFV as described above. A total of 105 sheep sera were used of which 20 were from sheep inoculated subcutaneously with the SPU22/118 KEN 07 strain of RVFV, and the remaining 85 were from na?ve sheep. 3.2.5 Antigen detection in human specimens A total of 130 human sera submitted to the Special Pathogens Unit of the National Institute for Communicable Diseases, Sandringham, South Africa (SPU-NICD) for routine testing were used; 70 specimens were from suspected RVF cases sampled during the 2006?2008 disease outbreaks in Southern Africa. 3.2.6 Monitoring viral growth in vitro Tenfold dilutions of the Ar20368 RSA 81 strain of RVFV in EMEM (from 105.8 to100.8 TCID50/ml) were used for inoculation of 25cm2 tissue culture flasks containing 48 h confluent Vero cell monolayers. Inoculated flasks were incubated on a rotating platform for 1h at 37?C. Two mock inoculated flasks were included as controls. After 1 h of incubation, inoculated flasks were removed, cells were washed with PBS and supplemented with 10 ml of EMEM containing 1% fetal calf serum and antibiotics. Inoculated cells were maintained at 37?C in a CO2 incubator. One ml aliquots of tissue culture medium were collected hourly for the first 8 h, and thereafter at 12, 24, 30, 48, 54, 72, 78, 96 and 102 h after inoculation for testing on the sAg-ELISA. The collected aliquots of tissue culture medium were replaced each time with the same volume of fresh medium. Appearance of cytopathic effect (CPE) was documented at each collection time. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 69 P. Jansen van Vuren 3.2.7 Antigen detection in decomposing tissues To mimic clearance of NP protein in decomposing tissues from RVFV-infected animals, homogenates (5%, w/v) of fresh normal ovine and bovine liver tissues were prepared as described above and mixed with an equal volume of tissue culture supernatant containing 107.9 TCID50/ml of RVFV Zim688/78 strain, and then incubated at 37?C for a period of 8 days during which aliquots were taken for testing immediately after mixing, and 5, 24, 48, 72, 168 and 192 h thereafter. Supernatants were collected after centrifugation at 3000?g at 4?C and tested by sAg-ELISA and virus titration. Virus titrations of clinical and laboratory generated specimens were performed as described previously (Swanepoel et al., 1986a). Briefly, four 100?l replicates of 10-fold dilutions (10?1 to 10?7) of specimens in EMEM were transferred into flat bottomed 96- well cell culture microplates (Nunc, Denmark) and equal volumes of Vero cell suspension in EMEM, containing 2?105 cells/ml, 8% fetal bovine serum/ml (Gibco) and antibiotics, were added. The inoculated microplates were incubated at 37?C in a CO2 incubator and observed under a microscope for CPE for 10 days post-infection (p.i.). Virus concentrations, calculated by the method of K?rber (K?rber, 1931), were expressed as median tissue culture infectious dose (TCID50) per ml of specimen. 3.2.8 ELISA performance, cut-off selection and IQC Seventeen RVFV isolates recovered over a period of 53 years (1955?2008) in African countries, Madagascar and Saudi Arabia (Table 3.1), four African Phleboviruses (Arumowot, Gabek Forest, Gordil and Saint-Floris) and two other members of the family Bunyaviridae (Akabane and Bunyamwera viruses) were used to evaluate the analytical sensitivity and specificity of the sAg- ELISA. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 70 P. Jansen van Vuren Table 3.1. Identification, year of isolation, origin and concentration of RVFV strains used to evaluate the sAg-ELISA. Strain Year of isolation Source Country of origin Concentration Log10TCID50/ ml ELISA PP valuea Lunyo UGA 55 1955 Mosquito Uganda 106.3 112.0 ZH 501 1977 Human Egypt 105.8 121.9 Zim 688/78 1978 Bovine Zimbabwe 108.5 137.8 VRL-825-ZIM79 1979 Bovine foetus Zimbabwe 107.0 123.3 Ar20368 RSA 81 1981 Mosquito South Africa 106.8 110.3 Ank-6087 1984 Bat Guinea 107.0 108.1 ArD38661 SEN 83 1983 Mosquito Senegal 107.3 134.2 ArD38388 BF 83 1983 Mosquito Burkina Faso 107.5 126.1 CAR R1662 1985 Human Central African Republic 106.5 131.5 SPU-204-ANGL85 1985 Human Angola 106.0 115.0 900085MAU88 1988 Human Mauritania 107.3 125.3 An991-MAD91 1991 Bovine Madagascar 107.5 121.7 SPU12002-SOM98 1998 Caprine Somalia 108.0 124.6 AR21229-SA00 2000 Mosquito Saudi Arabia 107.3 117.2 SPU 77/04 2004 Human Namibia 107.0 132.3 SPU22.118KEN 07 2007 Human Kenya 106.8 131.2 AR 52/08 2008 Human South Africa 106.8 127.5 a Percentage positivity of ELISA high positive antigen control. The internal quality control data were generated as described before (2.1.2.1). Means and standard deviations (S.D.) of the ELISA optical density readings and the percentage positivity (PP) of high-positive antigen control were calculated from replicates of all internal controls in each plate and each run of the assay to assess intra- and inter-plate variation. Additionally, coefficients of variation (CV = standard deviation of replicates/mean of replicates?100) were calculated for positive antigen controls. Data obtained from this analysis were used to estimate the assay repeatability and to establish the upper and lower control limits for each of the internal controls. Upper and lower control limits together with CV values (?10 for high-positive serum and ?15 for low-positive serum) were applied as IQC rules for further analysis. During routine runs of the ELISA each plate had four replicates of high- positive antigen control (Ag++) and two replicates each of low-positive antigen (Ag+) and negative control antigen (Ag?). Cut-off values at the 95% accuracy level were optimized using the misclassification cost term option (Greiner, 1996) of the two-graph receiver operating characteristics (TG-ROC) analysis (Greiner, 1995). Additionally, cut-off values were determined by mean plus 2 standard deviations (S.D.s) and Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 71 P. Jansen van Vuren mean plus 3 S.D.s derived from PP values in known RVFV-negative human and sheep sera. Estimates of diagnostic sensitivity, specificity, and efficiency were calculated as described before (2.1.2.5). 3.3 Results 3.3.1 Internal quality control and assay repeatability The s-Ag ELISA was able to differentiate clearly between the internal controls used and generated minimal background. Variation between and within runs were minimal (Figure 3.1). CV values for intra- and inter-plate runs were below 5% (Table 3.2) demonstrating high repeatability of the assay. Table 3.2 Internal quality control limits and repeatability estimates of RVF s-Ag ELISA a Internal quality control - data were calculated from the mean ? 2 S.Ds. of 180 replicates of high positive antigen control (Ag++), 90 replicates of low positive (Ag+) and negative (Ag-) antigen controls over 5 runs each including 3 plates. b Upper control limit. c Lower control limit. d Percentage positivity of high positive antigen control. e Repeatability estimates for high (Ag++) and low (Ag+) positive antigen controls were calculated as the percentage coefficient of variation [%CV = (PP S.D. of replicates / PP mean of replicates) x 100] Internal controls IQCa limits Repeatabilitye UCLb LCLc Intra-plate variation Inter-plate variation OD Ag++ 2.0 1.1 PPd Ag++ 129 71 1.04 ? 0.9 S.D. 1.28 ? 0.63 S.D. PP Ag+ 51 28 1.85 ? 1.83 S.D. 2.55 ? 1.72 S.D. PP Ag- 4.9 1.4 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 72 P. Jansen van Vuren 0 20 40 60 80 100 120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Number of runs EL IS A PP va lu e Figure 3.1. Upper (j1) and lower (---) internal quality control limits for PP values of high- positive (?), low-positive (?), negative (?) antigen controls and means ?2 S.D. for these controls during 15 routine runs of the assays over a period of 16 weeks. Two to six plates were used during each run with four replicates of high positive, two replicates of low positive and negative controls on each plate. 3.3.2 Efficiency of different inactivation buffers and safety testing The 1% Tween-20 in PBS inactivation buffer together with 56?C for 60 minutes yielded the highest signal of antigen detection (results not shown). Mice and tissue cultures inoculated with spiked samples that were inactivated according to this method did not develop any signs of infection in the specified monitoring period, indicating that it renders samples completely safe. In comparison, spiked samples only inactivated by heat or Tween-20 separately were not rendered safe since Vero cell monolayers developed CPE and suckling mice developed typical RVFV clinical signs after inoculation. 3.3.3 Analytical detection limit, specificity and sensitivity The sAg-ELISA was able to detect as little as 110 pg of recNP, corresponding to 102.2 TCID50 of RVFV per 100?l of Vero-derived infectious tissue culture supernatant (Figure 3.2). Analysis of the ELISA readings for sera and tissue homogenates spiked with different concentrations of RVFV shows Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 73 P. Jansen van Vuren that the detection limit in most assayed samples was approximately log10102.2 TCID50 per ELISA reaction volume (100?l) except for bovine liver where the detection limit was 10 times lower (Figure 3.3) at log10103.2 TCID50 per ELISA reaction volume (100?l). When testing infectious tissue culture supernatant containing related African Phleboviruses (Arumowot, Gabek Forest, Gordil and Saint-Floris) and two other members of the family Bunyaviridae (Akabane and Bunyamwera viruses), ELISA readings ranged from 0 to 0.63 PP (mean 0.41?0.26) (results not shown) whereas non-specific background noise of normal tissue culture fluid was 0.44 PP. These results demonstrate the highly specific binding affinity of anti-recNP RVFV polyclonal hyperimmune sheep and rabbit sera and the absence of detectable cross-reactions between these anti- sera and nucleocapsid proteins of the other Bunyaviruses assayed. All of the 17 RVFV strains were easily detected by ELISA (Table 3.1). 0 20 40 60 80 100 120 140 220 110 55 27.5 13.75 6.88 3.44 1.72 0.86 0.43 0.22 0.11 0.05 0.03 Concentration of recNP (ng/100?l) EL IS A PP va lu e 0 0.5 1 1.5 2 2.5 3 5.5 5.2 4.9 4.6 4.3 4 3.7 3.4 3.1 2.8 2.5 2.2 1.9 1.6 Log10TCID50/reaction volume EL IS A O D v al u e Figure 3.2. Dose response curves of recNP (PP ?, OD ?), RVFV Ar20368 RSA 81 (PP ?, OD ?), and control antigen (PP +, OD -). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 74 P. Jansen van Vuren 3.3.4 Monitoring viral growth in-vitro The detection of antigen in infected tissue culture supernatants after specific incubation times are shown in figure 3.4. Irrespective of the inoculum used, the sAg-ELISA yielded positive results earlier than CPE was observed. For example, in the flasks inoculated with 105.8 and 100.8 TCID50/ml RVFV the ELISA detected antigen 8 and 48 h after inoculation respectively, whereas CPE could only be observed 16 and 24 h later. 3.3.5 Antigen detection in decomposing tissues The sAg-ELISA was able to detect nucleocapsid antigen equally until the last collection time (192 hours) in spiked sheep liver incubated at 37?C, whereas antigen detection ability decreased in spiked bovine liver from 168 hours onwards (Figure 3.5). In contrast the simulated decomposition of tissues resulted in rapid inactivation of infectious virus particles as shown by negative results in the same organs after 48 hours incubation at 37?C by virus titration. 3.3.6 Diagnostic cut-off values and accuracy The effect of three differently determined cut-off values on the estimates of diagnostic sensitivity, specificity, and efficiency of the sAg-ELISA in human and sheep sera are given in Table 3.3. The highest diagnostic accuracy for human and sheep serum data sets was achieved when threshold values (5.6 PP and 1.23 PP) determined as mean plus 3 S.D. were used. However, estimates of the assay diagnostic performance based on cut-off determined as mean plus 2 S.D or derived from the TG-ROC analysis were similar (Table 3.3). In ELISA positive human sera at the optimal cut-off, mean TCID50/ml of the virus was 5.6?0.83, and in ELISA negative sera, it was 3.7?1.61. At cut-off determined as mean plus 2 S.D. derived from PP values of normal mice and ruminant tissues (2.4 PP), the sAg-ELISA had 100% sensitivity and specificity in detecting the nucleocapsid protein of RVFV in various tissues of experimentally infected mice and naturally infected buffalo foetuses (Figure 3.6). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 75 P. Jansen van Vuren 0 20 40 60 80 100 120 140 5.2 4.7 4.2 3.7 3.2 2.7 2.2 1.7 1.2 0.7 Log10TCID50 per reaction volume EL IS A PP v al u e Figure 3.3. Dose response curves of human and animal sera, animal tissue and mosquito homogenates spiked with RVFV and their corresponding non-spiked controls. Virus-spiked samples: human (???), sheep (- -?- -), bovine (- -?- -) serum; sheep (???), bovine (???) spleen; sheep (--?--), bovine (--?--) liver; Anopheles mosquito (???). Uninfected samples: human (???), sheep (- -?- -), bovine (- -?- -) serum; sheep (???), bovine (???) spleen; sheep (--?--), bovine (--?--) liver; Anopheles mosquito (???). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 76 P. Jansen van Vuren Figure 3.4. RVFV replication kinetics in Vero cells inoculated with different concentrations of the virus measured by antigen detection ELISA. Log10TCID50 virus concentrations in 1 ml of inoculum were 105.8 (?), 104.8 (?), 103.8 (x), 102.8 (?), 101.8 (?) and 100.8 (+), mock control (*). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 77 P. Jansen van Vuren 0 1 2 3 4 5 6 7 0 hr 5 hr 24 hr 48 hr 72 hr 168 hr 192 hr Incubation time at 37oC Lo g1 0T CI D 50 /m l 0 20 40 60 80 100 120 EL IS A PP Va lu e Figure 3.5. Detection of infectious virus and viral nucleocapsid antigen in tissue homogenates spiked with RVFV followed by incubation at 37?C. TCID50/ml in bovine (?) and sheep (?), and ELISA readings in bovine (?) and sheep (?) liver homogenates. Table 3.3. Diagnostic accuracy of the sAg-ELISA for the detection of nucleocapsid protein of RVFV in sheep and human sera Measurea Human sera ? cut-off (PP)b V+c = 31, Vmaqafd = 99 Sheep sera ? cut-off (PP)b V+c = 20, Vmaqafd = 85 1.96e 4.11f 5.60g 0.86e 0.91f 1.23g D-Se (%)h 77.4 67.7 67.7 70.0 70.0 70.0 D-Sp (%)i 81.8 95.95 97.97 97.65 97.65 100.0 Efficiency (%)j 79.6 81.83 82.84 83.83 83.83 85.0 a Sera were categorized according to the results of virus isolation. b Cut-off value expressed as a percentage positivity of high positive antigen control. c Number of sera tested positive for RVFV. d Number of sera tested negative for RVFV. e Cut-off value optimised by the misclassification cost term option of the two-graph receiver operating characteristics analysis at 95% accuracy level. f Cut-off value determined as mean plus two standard deviation derived from PP values in RVFV-negative sera. g Cut-off value determined as mean plus three standard deviation derived from PP values in RVFV-negative sera. h Diagnostic sensitivity. i Diagnostic specificity. j Efficiency. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 78 P. Jansen van Vuren Liver Kidney Heart Liver Kidney Brain Liver Kidney Brain 0 20 40 60 80 100 120 140 EL IS A PP v al u e 103.6 87.1 88.0 102.9 82.1 28.5 0.5 0.6 1.2 Figure 3.6. Detection of nucleocapsid protein by ELISA in RVFV-infected buffalo foetus (?) and mouse organ tissues (?). Non-infected mouse organ tissues (?) are included as a control. Mean PP values were determined for the liver, heart and kidney tissues of the three infected buffalo foetuses. Mean PP values were determined for the liver, brain and kidney tissues of three infected mice. Control PP values were obtained from one uninfected mouse. Cut-off of 2.4 PP (---) was determined as the mean plus 2 S.D. of PP values in uninfected mouse and ruminant organ tissues. 3.3.7 Interference on antigen detection in serum by the presence of anti-RVFV antibodies The ELISA yielded negative results in sera taken from an experimentally infected sheep on day 5 and onwards post infection (p.i.) despite relatively high levels of viremia detected on days 5 and 6 p.i. The negative results coincided with the appearance of the first detectable anti-nucleocapsid IgM and IgG antibody on day 5 p. i. (Figure 3.7). To confirm the blocking effect of RVFV-immune sera in the sAg-ELISA, viremic sheep serum was mixed with an increasing concentration of known sheep immune serum. The inhibitory effect of increasing levels of specific antibodies on the ELISA specific signal in highly viremic sheep serum is shown in Figure 3.8. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 79 P. Jansen van Vuren 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10 Days post inoculation Lo g 1 0T CI D 50 /m l 0 50 100 150 200 250 300 350 EL IS A PP v al u e Figure 3.7. Monitoring of viremia (?), antigenemia (?), anti-nucleocapsid IgM (?) and IgG (x) responses in a sheep experimentally infected with RVFV. 91.8 88.6 83.3 72.6 54.5 36.9 14.4 1.6 0.2 0.3 0.6 0 10 20 30 40 50 60 70 80 90 100 A A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 EL IS A PP v al u e Figure 3.8. Blocking effect of increasing levels of anti-RVFV specific antibody on antigen detection in viremic sheep serum. Viremic serum free of antibody against nucleocapsid of RVFV (A, ?). Mixture (v/v) of viremic serum and immune serum with two-fold increase of anti- RVFV specific antibodies (A1- 1:1024 to A10-1:2, ?). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 80 P. Jansen van Vuren 3.4 Discussion A sandwich ELISA using hyperimmune mouse and rabbit antisera reported by Niklasson et al. (1983) had a sensitivity of 105 plaque forming unit (PFU)/ml in detecting RVFV in supernatant fluids from infected Vero cell cultures. However, marked differences between levels of antigen and the virus infectivity in experimentally infected hamsters and rhesus monkeys were noted. While ELISA could reliably detect 106 PFU/ml of virus in viremic hamsters, rhesus monkeys with viremia of 103.4 PFU/ml tested positive (Niklasson et al., 1983). A sandwich ELISA utilizing a biotin-avidin labelled mouse monoclonal antibody as a detector system had a sensitivity of 29.3% in viremic human sera collected during the 1987 RVF epidemic in West Africa (Meegan et al., 1989). This estimate appears to be rather low compared to 76.9% sensitivity of the assay reported by Niklasson et al. (1983) in sera from experimentally infected rhesus monkeys. Viral titres were demonstrated to be significantly correlated with quantity of viral antigen as measured by ELISA in orally infected Egyptian Culex pipiens (Niklasson and Gargan, 1985). Moreover, when comparing results of the infectivity assay with the ELISA, the latter had similar sensitivity (100% vs. 93%) and specificity (94% vs. 94%) in detecting mosquitoes capable of transmitting virus to susceptible hamsters. These early studies clearly demonstrated potential field applications of a sandwich ELISA, however, its wider use for routine laboratory detection of RVFV have been limited due to a number of considerations: testing of RVFV- infected specimens by ELISA pose laboratory biohazards because of intensive pipeting and washing procedures, the use of OD readings for interpretation of ELISA results is not currently recommended, lack of non-infectious and well-characterized internal antigen controls hampered adequate standardization, and evaluation of its performance within and between laboratories. Recently immunofluorescence assays which utilize a pool of mouse IgG monoclonal conjugates reacting with a combination of virus specific antigens (Gs, Gn, N, NSs) were reported (Zaki et al., 2006). Although it was demonstrated to be highly reliable in detecting RVFV in patient sera, its use requires tissue culture amplification and handling of live virus. ELISA offers an affordable and simple alternative to traditional and molecular techniques for detection of RVFV, but as an open bench system might contribute to laboratory infections when samples containing live virus are analysed. A number of laboratory infections with RVFV were recorded under circumstances which indicate the virus to be highly infectious for man (Findlay, 1932, Kitchen, 1934, Smithburn, 1949, Smithburn et al., 1949b). To address this problem, a sandwich ELISA based on an entirely safe procedure was developed, including a set of internal controls based on recNP protein for monitoring of assay routine performance which increases its utility in surveillance and diagnosis in non-endemic areas. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 81 P. Jansen van Vuren RVFV has been shown to be extremely stable when stored in infected plasma at low temperature but is unstable at higher temperatures. For example, infected sheep plasma retained infectivity after 8 years of cold storage (Easterday, 1965) but viremic blood became non-infective after 40 minutes of incubation at 56?C in phosphate buffer, pH 7.2 (Findlay, 1932), contrary to the results presented in this chapter. The presence of lipid bilayer in the virus envelop makes it highly sensitive to lipid solvents (Bishop et al., 1980). The simple and inexpensive thermo-chemical inactivation procedure used in this study effectively inactivated RVFV present in test specimens as demonstrated by negative virus isolation results in vivo and in vitro systems. Using a similar procedure West Nile virus can be successfully inactivated at 37?C for 30 minutes in the presence of 0.05% Tween-20 (Mayo and Beckwith, 2002). In addition to effective virus killing, the inactivation protocol used increases ELISA specific signal compared to that in non-inactivated specimens. This is likely the result of enhanced trapping of nucleocapsid protein by capture antibody in sAg-ELISA after viral envelope had been disrupted by Tween-20. The analytical sensitivity of log10103.2 TCID50/ml determined by testing infective supernatant of Vero cell culture fluid seems to be at least 10-fold higher compared to that reported by Niklasson et al., (1983). Using different concentrations of RVFV we demonstrated that the NP antigen can be detected in supernatant fluids from infected Vero cell cultures as early as 8 h post inoculation and 12 to 30 h before CPE could be microscopically observed. This ability renders the assay very useful for rapid identification of RVFV when primary isolation from clinical specimens is attempted in vitro. Live virus became undetectable much earlier than the nucleoprotein by sAg-ELISA when incubated in animal tissues at adverse temperature. This ELISA can therefore be used to diagnose RVF by using decomposing tissues of ruminants that might have been found dead in the field. Analytical detection limit established in virus-spiked samples which mimicked diagnostically relevant submissions, was the same except in bovine liver homogenate for which it was 10 times less (log10104.2 TCID50/ml). An estimate of ELISA diagnostic sensitivity derived from results in viremic human sera was much higher (67.7%) compared to that (29.3%) reported by Megan et al. (1989) but similar (76.9%) to that in viremic rhesus monkeys sera (Niklasson et al., 1983). Diagnostic accuracy of a sandwich ELISA in viremic human sera but also in other specimens is likely to be dependent on a number of factors, including origin and type-specificity of capture and trapping antibody, their purity (monoclonal vs. polyclonal), titres and spectrum of reactivity to RVFV structural proteins (Meegan et al., 1989). Serum specimens are commonly used for RVF diagnosis. Viremia titres ranging from 105.6 to 109.0 of mouse median lethal doses per ml have been recorded in domestic ruminants (Daubney et al., 1931, Barnard and Botha, 1977, Harrington et al., 1980, Swanepoel et al., 1986a, Morrill et al., 1987), 108.6 in humans (Peters and Meegan, 1981) and 105.4 TCID50/ml in adult African buffalo (Davies and Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 82 P. Jansen van Vuren Karstad, 1981). Although viremia in RVFV-infected individuals reaches high titres, it is of short duration which limits the application of viral detection systems for RVF diagnosis when using blood samples. Moreover, results obtained in serial sera from an experimentally infected sheep and ELISA blocking experiments indicate that the appearance of specific antibodies during viremia hampers the assay results despite the presence of relatively high concentrations of the virus. These findings seem to indicate that differences in ELISA diagnostic performance might also be due to variations in immune status among viremic individuals at the time of sampling. Therefore, attempts to detect recent RVFV infection by ELISA should include a combination of assays which target both viral antigens and IgM antibody. It should be noted that high viremia frequently occurs in the absence of severe illness. Consequently, in the absence of noticeable clinical signs and adequate diagnostic procedures, considerable geographic dispersal of RVFV might occur before an outbreak is recognized (McIntosh et al., 1973). On the other hand, the South African outbreaks of 1950-51 and the Egyptian outbreaks of 1977-78 were not recognized as RVF until several months had elapsed with deaths of thousand of animals, and, in the Egyptian outbreaks, many deaths in humans. Delays in recognition of these outbreaks occurred because the disease was previously unknown in those geographical areas and the possibility of RVF was not at first considered. In this study very high estimates of diagnostic accuracy (100%) where obtained when testing various tissue homogenates of experimentally infected mice and naturally infected African buffalo foetuses. RVFV can persist at high titres for 21 days in ovine brain and liver, and up to 30 days in spleen (Yedloutschnig et al., 1981). High diagnostic accuracy of the sAg-ELISA in detection of RVFV in infected tissues which usually contain virus concentrations at least 10 to 100-fold times above (Easterday et al., 1962, Easterday and Murphy, 1963, Harrington et al., 1980, Morrill et al., 1987) the detection limits determined in this study, indicate that the assay will be highly reliable for testing specimens from aborted fetuses and fatal cases. Massive abortion and high fatality rates in young animals are one of the characteristic features of RVF outbreaks. The ELISA format reported here allows for assaying relatively large numbers of specimens within a short period of time. The assay throughput, if required, could be easily increased by using semi- or fully automated ELISA workstations. The ability of a diagnostic assay to produce consistent results within the tolerable analytical error limits is one of requirements for any diagnostic device to be accepted for routine applications. While the antigen internal controls based on the recombinant NP protein achieved very high repeatability estimates within the IQC limits, the reproducibility of the sAg-ELISA remains to be addressed for more comprehensive inter-laboratory evaluation. Antibody- and antigen-binding levels should be expressed in relative rather than absolute terms. One of the advantages of converting ELISA Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 83 P. Jansen van Vuren OD data into PP values relative to a known standard is that this method does not assume uniform background activity, and therefore enables inter-laboratory standardisation (Wright et al., 1993). The polyclonal hyperimmune anti-recNP rabbit and sheep antisera did not cross-react with other members of the Bunyaviridae family, including four African Phleboviruses tested in this study. No cross-reactivity with other members of the sand fly fever virus group, including sand fly Naples, sand fly Sicilian, Arumowot, Punta Toro, Gordil, Karimabad, Gabek Forest, and Saint Floris, was detected in sandwich ELISA using hyperimmune mouse and rabbit antisera by Niklasson et al. (1983). The nucleocapsid protein appears to be highly conserved among members of the Bunyaviridae family (Swanepoel et al., 1986a, Vapalahti et al., 1995, Schwarz, 1996, Magurano and Nicoletti, 1999, Gauliard et al., 2006) and antigenic cross-reactivity studies in animals (Davies, 1975, Swanepoel, 1976, Swanepoel et al., 1986b) and indirect ELISA based on recNP protein (Paweska et al., 2007) failed to provide any evidence that other African phleboviruses could obscure the reliable serodiagnosis of RVF. The sAg-ELISA detected nucleocapsid proteins of a wide range of geographically distinct RVFV isolates collected over 53 years, which represent three major lineages of the virus, namely Egyptian, Western African, and Central, Eastern and Southern African. These results are expected since the RVFV genome, and especially the gene encoding the N protein, is highly conserved (Bird et al., 2007b). In conclusion the sAg-ELISA procedure developed and evaluated in this study is safe, highly accurate in detection of RVFV NP antigen in diagnostically relevant concentrations, rapid and robust and therefore can be utilized in diagnosis and surveillance in both endemic and non-endemic RVF areas. It offers a less complicated alternative to nucleic acid techniques when large numbers and clinical variety of specimens have to be tested in a short period of time. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 84 P. Jansen van Vuren CHAPTER FOUR EVALUATION OF A RECOMBINANT NUCLEOCAPSID PROTEIN AS AN IMMUNOGEN IN A MOUSE MODEL* * Partially published as: Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2010). Evaluation of a recombinant Rift Valley fever virus subunit nucleocapsid protein as an immunogen in mice and sheep. The Open Vaccine Journal 3, 114-126 * Partially presented at international conferences as: Jansen van Vuren, P. & Paweska, J.T. (2009). Preliminary evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as an immunogen in combination with different adjuvants in mice and sheep. FAO/IAEA International Symposium on Sustainable Improvement of Animal Production and Health, Vienna, Austria, 8 ? 11 June 2009. (Poster presentation). Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2009). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine immunogen in combination with four adjuvants. International Meeting on Emerging Diseases and Surveillance, Vienna, Austria, 13 ? 16 February 2009. (Poster presentation). 4.1 Introduction Until recently there were only two vaccines commercially available for use in livestock: a live attenuated vaccine based on the Smithburn strain and a formalin inactivated vaccine ? both available exclusively available from Onderstepoort Biological Products (Pretoria, South-Africa). However, Clone 13 was very recently commercialized by Onderstepoort Biological Products and widely used for the vaccination of livestock in South Africa during the 2010/2011 RVF season (Paweska, J.T., personal communication)(Dungu et al., 2010). There are no commercially available RVF vaccines for human use, but an experimental inactivated vaccine has been used in the past to vaccinate veterinarians, scientists and other personnel at risk of exposure (Bouloy and Flick, 2009). Because limited amounts of this vaccine (TSI-GSD 200) were produced under strict quality controlled conditions at the USAMRID facility, it is currently in short supply and very expensive. As discussed in the literature review (Chapter 1), various vaccine candidates generated by classical virological methods have been evaluated to counter this problem. These, however, are expensive, laborious to produce and not completely safe to use. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 85 P. Jansen van Vuren Recently molecular biological methods enabled research into safer, more effective and less expensive RVFV vaccine candidates. These include attenuated viruses generated by reverse genetics (Bird et al., 2008, Habjan et al., 2008b), DNA vaccines (Spik et al., 2006, Lorenzo et al., 2008, Lagerqvist et al., 2009), virus vectored vaccines (Collett et al., 1987, Wallace and Viljoen, 2005, Wallace et al., 2006, Heise et al., 2009, Soi et al., 2010), virus like particles (Habjan et al., 2009a, Mandell et al., 2009, Naslund et al., 2009, Pichlmair et al., 2010) and recombinant subunit vaccines (Collett et al., 1987, Schmaljohn et al., 1989, Wallace et al., 2006). Most of these constructs are aimed at inducing immunity against the glycoproteins that carry neutralizing determinants, with the exception of a vaccine candidate evaluated by Wallace et al. (2006) which consisted of a preliminary experiment with a recombinant RVFV N protein, expressed as an insoluble protein. The nucleocapsid protein is the major antigen of RVFV and strong immune responses against this protein have been shown after natural and experimental infections with the virus (Fafetine et al., 2007, Jansen van Vuren et al., 2007). The 60% protection rate from lethal challenge achieved in mice in the Wallace et al. (2006) preliminary experiment with a recombinant RVFV N protein needed further investigation, especially taking into account good protection rates achieved by using the N proteins of related viruses as immunogens (Schmaljohn et al., 1990, Maes et al., 2006, Maes et al., 2008). Recombinant protein subunits are generally weak immunogens (O'Hagan et al., 2001, Lautze et al., 2007) and require administration with adjuvants to enhance their immunogenicity (Dasgupta, 2004). Adjuvants promote the uptake of antigens by antigen presenting cells (APC), contribute to the delivery of antigen to lymph nodes, and stimulate cytokine release or expression of co-stimulatory signals on APC which are needed to prime T helper cells for B cell proliferation and induction of cytotoxic T lymphocytes (O'Hagan et al., 2001, O'Hagan and Singh, 2003). Some of the more commonly tested and/or used adjuvants are saponins, alum and water-in-oil adjuvants. Saponin adjuvant, a surface active agent isolated from the Chilean soap bark tree (Quillaja saponaria), modulates humoral (Th-2) as well as cellular immunity (Th-1) and biases immune responses towards the Th-1 phenotype and can induce strong CD8+ cytotoxic T-cell responses (Kensil, 1996, Cribbs et al., 2003). CD8+ T cells are able to kill virus-infected cells by inducing apoptosis, and kill infected cells directly in the lymph nodes draining infected sites (Xu et al., 2007). Aluminium hydroxide gel (Alhydrogel), commonly known as alum allows for a depot effect at the inoculation site, and has also been found to promote the release of IL-4 which results in the increased expression of MHC II molecules on monocytes, consequently increasing antigen uptake by APC (Mannhalter et al., 1985, Ulanova et al., 2001, O'Hagan and Singh, 2003). Alum does not induce the cytokines IL-2 and IFN-? which are involved in the Th-1 type response, but might directly activate NF-kB, that is involved in regulating the cellular response to infections (Ulanova et al., 2001). The NF-kB is required for positive Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 86 P. Jansen van Vuren selection of memory CD8+ T cells (Hettmann and Leiden, 2000, Hettmann et al., 2003). Montanide ISA50 adjuvant is based on a mannide oleate in mineral oil solution, and contributes to the establishment of a depot effect, transportation of emulsified antigen to distant sites through the lymphatic system, and interaction with mononuclear cells such as APC. ISA50 has been shown to direct the immune response against specific antigens towards the Th-2 type response, involved in humoral immunity (O'Hagan et al., 2001). TiterMax Gold (TMG) is a water-in-oil adjuvant that contains a metabolizable oil (squalene), sorbitan monooleate and an immunostimulatory copolymer. It has been shown to induce mixed Th-1/Th-2 responses against specific antigens, but these responses were more directed towards Th-2, indicating humoral immunity (Cribbs et al., 2003). In this study the immunogenicity of a bacterially expressed recombinant subunit RVFV N protein was evaluated alone, and combined with four different adjuvants. The protection against subsequent viral challenge was studied in a mouse model. 4.2 Immunogenicity of the recombinant nucleocapsid protein alone and in combination with four adjuvants 4.2.1 Materials and methods 4.2.1.1 Mouse immunization The recombinant RVFV nucleocapsid protein (recNP) was produced as described in section 2.1.2.4. Four-week old female BALB/c inbred mice were used as an experimental animal model. The low dose vaccination group (M-I) consisted of 48 mice divided in 4 sub-groups of 12 mice each which were immunized with a 100?l inoculum containing 35?g RVFV recNP in combination with ISA-50 adjuvant (Seppic, France), TiterMax-Gold adjuvant (TMG)(Sigma, U.S.A.), Alhydrogel (Sigma) or SaponinQ (60?g, Sigma), respectively. The high dose vaccination group (M-II) consisted of 48 mice which were subdivided as the M-I group but immunized with 200?l of inoculum containing 70?g of recNP in combination with the adjuvants as described above. The neat recNP group (M-N) consisted of 12 mice immunized with 70?g recNP in PBS buffer. The adjuvant control group consisted of 36 mice divided in 3 sub-groups of 12 mice each which were respectively inoculated only with ISA-50, Alhydrogel or SaponinQ. The placebo control group consisted of 12 mice which were immunized with PBS buffer. All mice were inoculated subcutaneously (s.c) and received identical booster immunizations at 14 days after the initial immunization. A mouse from each group was sacrificed and heart-bled every seven days after primary and booster immunizations to monitor immune responses. Adjuvants ISA50, TMG and Alhydrogel were used as suggested by the manufacturers. The dose of SaponinQ adjuvant (Sigma, U.S.A.) was determined by titration in BALB/c mice and by selecting the highest non-toxic Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 87 P. Jansen van Vuren dose at 60?g (results not shown). The selection of recNP doses were determined by recNP concentration and feasible mouse inoculum sizes. 4.2.1.2 Monitoring mouse immune responses Immunoplates (Maxisorb, Nunc, Denmark) were coated with RVFV recNP antigen at a dilution of 1:2000 in Carbonate-Bicarbonate buffer (pH 9.6) and incubated overnight at 4?C. After washing three times with a washing buffer consisting of PBS pH7.2 and 0.1% Tween-20, the plates were blocked with 200?l of 10% fat free milk powder (?Elite?, Clover SA, Pty, Ltd.) in PBS at 37?C for 1h and then washed as before. Test sera were diluted 1:400 in diluent buffer consisting of 2% fat free milk powder in PBS, 100?l added to each well and incubated for 1h at 37?C. Samples were tested in duplicate for each isotype-specific HRPO conjugate used. After washing as before, 100 ?l of goat anti-mouse IgG (H+L), goat anti-mouse IgG1 or goat anti-mouse IgG2a HRPO conjugate (Zymed Laboratories, Invitrogen, U.S.A.) at 1:2000 dilution was added to respective plates testing for the same serum specimens in parallel. After 1h incubation at 37?C plates were washed as before and 100 ?l of 2,2?-azinodiethylbenzthiazoline sulfonic acid (ABTS) (KPL Laboratories, Inc., USA) added to each well. After 30 min incubation in the dark the reaction was stopped by the addition of 100 ?l of 1% sodium dodecyl sulphate (SDS) to each well. Optical density (OD) was determined at 405nm and the results expressed as the mean OD value for the duplicates tested. In addition, a virus neutralization test (VNT) was performed on sera collected from mice after immunization. The VNT was performed as described previously (section 2.1.2.3). 4.2.2 Results Serial cardiac bleeds taken from one mouse from each group on days 0, 7 and 14 after the primary immunization, and on days 7, 14 and 21 after the booster immunization were analyzed for the presence of total IgG, IgG1 and IgG2A antibodies anti-recNP. The representative mouse from each vaccinated group had produced detectable total IgG anti-N antibodies on day 7 after primary vaccination (Figure 4.1 a-i). The total IgG antibodies increased steadily over the monitoring period in all groups. The IgG1 isotype antibodies developed in a similar fashion to the total IgG in all groups, following the same response kinetics. All immunized groups, except those immunized with recNP/SaponinQ combinations, developed weaker IgG2a isotype responses as compared to total IgG and IgG1. The mice from the recNP neat and recNP/Alhydrogel groups developed weak IgG2A responses as compared to IgG1 in the same mice. Only the mice from the recNP/SaponinQ and recNP/ISA50 groups had detectable anti- recNP IgG2A specific antibodies after the first immunization whereas the other groups only developed Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 88 P. Jansen van Vuren IgG2A after the booster inoculum. As expected mice from the adjuvant and placebo control groups did not develop any anti-recNP responses (results not shown). Anti-recNP immune sera from mice were not able to neutralize the virus in-vitro (results not shown). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 89 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 90 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 91 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 92 P. Jansen van Vuren Figure 4.1(a-i). Detection of total IgG (???), IgG1 (?x?) and IgG2A (???) isotype antibodies against the RVFV recNP in mice after recNP immunization alone or in combination with adjuvants. Individual mice were tested at each time point so means and standard deviations could not be calculated. 4.2.3 Discussion The nucleocapsid protein induces production of high levels of anti-NP specific IgG and IgM responses in host animals after natural or experimental infection (Fafetine et al., 2007, Jansen van Vuren et al., 2007). The RVFV recNP used in this study easily detects anti-NP antibodies in previously infected individuals (Jansen van Vuren et al., 2007) and it was therefore assumed that it would be immunogenic. The recNP was indeed immunogenic in BALB/c mice, even in the absence of adjuvant, but more immunogenic with adjuvant. With all recNP/adjuvant combinations the total IgG responses to immunization were similar, but delayed and lower without adjuvant. The IgG1 isotype antibodies followed similar kinetics to total IgG in all recNP/adjuvant immunized mice, indicating strong humoral Th-2 response activation by all adjuvants. Only the recNP/SaponinQ combination, irrespective of recNP dose, induced strong IgG2A isotype responses comparable to their IgG and IgG1 responses, indicating strong activation of cellular Th-1 immunity by SaponinQ adjuvant. The combination of recNP with Alhydrogel was unable to generate a strong IgG2A response, indicating that this adjuvant might not induce strong cellular Th-1 immunity. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 93 P. Jansen van Vuren Interferon-gamma (IFN-?) and interleukin-2 (IL-2) are two cytokines known to be involved in the Th-1 type cellular response resulting in increased IgG2A responses. Interferon gamma (IFN-?) is secreted by natural killer cells (NK) and CD8+ cytotoxic T lymphocytes amongst others and has direct antiviral activity but also acts as an immunoregulatory factor. Interleukin 2 (IL-2) is secreted by T helper cells (Th) and acts by stimulating the growth, differentiation and survival of antigen specific CD8+ cytotoxic T lymphocytes (Kensil, 1996, Cribbs et al., 2003). The fact that there are no known neutralizing epitopes on the RVFV N protein suggests that a strongly biased humoral response against the N protein would not play a role in protection against viral challenge. Therefore it would seem that only recNP/adjuvant combinations that induced strong cellular Th-1 immunity would confer strong protection against RVF viral challenge. The lack of in vitro neutralizing ability of the anti-NP response has been shown before (Lorenzo et al., 2008, Lagerqvist et al., 2009) and confirmed in this study. To evaluate whether humoral antibodies play any role in vivo in protection against infection, the in vivo neutralizing ability of anti-recNP hyperimmune sera was tested in mice (see section 4.3). 4.3 In vivo neutralizing ability of anti-nucleocapsid immune sera in mice 4.3.1 Materials and methods 4.3.1.1 Cells and virus Vero cells were cultivated in Eagles Minimal Essential Medium (EMEM) (BioWhitaker, MD, USA) containing L-Glutamine, non-essential amino acids, antibiotics (100 IU penicillin, 100 ?g streptomycin and 0.25 ?g amphotericin B) and 10% foetal bovine serum (Gibco) and maintained at 37?C in 5% CO2 incubator. The SPU22/118 KEN 07 strain of RVFV was isolated from a RVF human case during the 2007 Kenyan epidemic. Second passage of the virus, propagated in Vero cells, was used for the challenge. 4.3.1.2 Inoculation of mice with virus/hyperimmune sera mixtures The ability of anti-recNP antibodies to passively confer immunity was evaluated using polyclonal antisera generated in sheep, rabbits and mice. Mice were immunized with recNP as described 4.2.1.1, and antisera from different recNP/adjuvant experimental groups were respectively pooled before testing. Polyclonal anti-recNP antisera in rabbits and sheep were produced as described previously (Chapter 3, section 3.2.1). All polyclonal sera were mixed to a final dilution of 1:10 with Vero-derived virus preparation containing 107.0 TCID50/ml of the 2007 Kenya RVFV isolate, and the mixture incubated at 37?C for 30 min before inoculation. As controls, sera from na?ve sheep, rabbits Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 94 P. Jansen van Vuren and mice were mixed identically with RVFV. To control the effects of non-related compounds in serum, sterile PBS was mixed to a 1:10 with the virus. A total of 42 BALB/c 3-4 weeks old female mice, were divided into groups of 6 animals each and inoculated s.c. with 200 ?l of the following mixtures: a) virus and mouse anti-recNP, b) virus and sheep anti-recNP, c) virus and rabbit anti-recNP, d) virus and naive mouse serum, e) virus and na?ve sheep serum, f) virus and na?ve rabbit serum, and g) virus and PBS. Mice were examined twice daily clinically and those displaying severe signs of illness were euthanized. Surviving mice were monitored for 22 days post infection. 4.3.1.3 Statistical methods Survival proportions in mice receiving virus/hyperimmune sera mixtures versus control mice after challenge were compared using Fisher?s exact test (Soper, 2009). 4.3.2 Results Anti-recNP immune sera did not neutralize virus in-vivo (Figure 4.2). No significant decrease in mortality/morbidity could be shown in any of the groups: a\ virus and mouse anti-recNP (survival 1/6, 17%, p = 0.500), b\ virus and sheep anti-recNP (survival 2/6, 33%, p = 0.227), c\ virus and rabbit anti-recNP (survival 0/6, 0%, p = 1.000), d\ virus and naive mouse serum (survival 0/6, 0%), e\ virus and na?ve sheep serum (survival 0/6, 0%) f\ virus and na?ve rabbit serum (survival 0/6, 0%), and g\ virus and PBS (survival 0/6, 0%). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 95 P. Jansen van Vuren 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 Days post infection Pe rc en ta ge su rv iv al (% ) Figure 4.2. In vivo neutralization of RVFV with anti-recNP immune sera in mice. RVFV mixed with immune sera are indicated with solid lines as follows: mouse anti-recNP (?), sheep anti- recNP (?) and rabbit anti-recNP (?). Corresponding normal sera from these animals are indicated with dotted line and the same symbols. RVFV mixed with PBS is indicated with a dotted line and (+). 4.3.3 Discussion Neutralization of viruses can be mediated by various mechanisms including aggregation of virions, virus structure destabilization, inhibition of attachment to cell receptors and blocking the release of virions from infected cells (Reading and Dimmock, 2007). It is widely accepted that the RVFV nucleocapsid protein does not contain neutralizing epitopes (Lorenzo et al., 2008, Lagerqvist et al., 2009) but this has only been evaluated in vitro using the cell culture based virus neutralization test. The neutralization of viruses in vivo is, however, more complex as it also involves interaction of antibodies with cells and molecules of the innate immune system (Reading and Dimmock, 2007). In this study it was confirmed that the anti-NP antibodies are not neutralizing in vitro. However, to evaluate whether anti-NP specific antibodies conferred some form of antibody mediated immunity independent of the various neutralization mechanisms known, an in vivo neutralization experiment was conducted in mice. A recent report shows that human and murine antibodies against the nucleocapsid protein of Toscana virus, a virus of the Phlebovirus genus and thus closely related to RVFV, has low neutralizing Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 96 P. Jansen van Vuren ability (Gori Savellini et al., 2008). In this study, however, the high levels of anti-recNP specific antibodies from mice, sheep and rabbits were ineffective in neutralizing RVFV in vitro and in vivo. No significant decrease in morbidity or mortality could be shown in the experimental groups when compared to the control groups. These results strongly suggest that humoral anti-recNP antibodies do not play a role in protection against viral infection. Cell free immune serum was, however, used for the in vivo neutralization experiment and therefore it is not known whether immune cells, and other factors not present in serum, from immunized individuals could play a role in protection. To evaluate the ability of the complete immune response, including innate/cellular/humoral, to protect against viral infection and morbidity/mortality, immunized mice were challenged with RVFV. 4.4 Rift Valley fever virus challenge of mice immunized with the recombinant nucleocapsid protein 4.4.1 Materials and methods 4.4.1.1 Cells and virus Cells and virus were cultured as described before (section 4.3.1.1). 4.4.1.2 RVFV challenge The mice remaining in each group after the immunization period (5 to 7 animals depending on group) were challenged with RVFV on day 32 after the booster immunization. Mice were inoculated subcutaneously (s.c.) with a 100 ?l inoculum containing 107.0 TCID50/ml RVF challenge virus, and after challenge examined twice daily for signs of clinical illness. Animals displaying severe illness were euthanized and organs collected. Organs were also collected at regular intervals from healthy, sick and dead mice to monitor challenge virus replication. Surviving mice were monitored for 22 days post infection. A control group was mock inoculated with EMEM free of the virus. 4.4.1.3 Determination of viral loads in mouse tissues Mouse liver, kidney and brain tissues were homogenized as 10% (w/v) suspensions in EMEM containing L-Glutamine, non-essential amino acids and antibiotics (100 IU penicillin, 100 ?g streptomycin and 0.25 ?g amphotericin B). After centrifugation at 3000 x g, 4?C for 15 minutes, supernatants were collected and stored at -70?C until tested. Virus titrations of mouse tissue homogenates were performed as described before (section 3.2.7). Briefly, four 100?l replicates of 10-fold dilutions (10-1 to 10-8) of homogenates were transferred into flat bottomed 96-well cell culture microplates (Nunc, Denmark) and equal volumes of Vero cell suspension in EMEM containing 2 x 105 cells/ml, 8% FBS and antibiotics were added. The plates Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 97 P. Jansen van Vuren were incubated at 37?C in CO2 and observed microscopically for cytopathic effects (CPE) for 10 days post inoculation. Virus titres, calculated by the K?rber method (K?rber, 1931) were expressed as median tissue culture infectious dose (TCID50) per gram of tissue. 4.4.1.4 Real-time reverse transcriptase PCR (qRT-PCR) Real time PCR was performed only on tissue homogenates that yielded negative results by virus titration. Viral RNA was extracted from 140?l of tissue homogenates using the QIAmp? Viral RNA Kit (QIAgen, Germany) according to the instructions of the manufacturer. The qRT-PCR was performed as described previously (Le Roux et al., 2009). Briefly, amplifications were carried out in 20?l reaction mixtures containing 5?l of the extracted vRNA using the LightCycler RNA Amplification Hybprobe kit (Roche, Germany) and the Roche LightCycler instrument. Primers and a labelled probe targeting the Gn glycoprotein gene of RVFV were used. 4.4.1.5 Statistical methods Survival proportions in immunized mice versus control mice after challenge were compared using the Fisher exact test (Soper, 2009). Viral load results in mouse organs are based on TCID50 titrations of virus in tissues from 3 or more animals and given as means. 4.4.2 Results All mice in the adjuvant and PBS placebo control groups died or developed severe symptoms by day six after the experimental infection with RVFV, indicating severe challenge. In contrast, mice that were immunized with recNP/adjuvant combinations and challenged identically to the control mice were fully or partially protected from death and severe symptoms, depending on the recNP/adjuvant combination and dose (Figure 4.3). Clinical signs in sick animals included loss of appetite and consequent weight loss, scruffy coat, decreased alertness, decreased mobility, loss of balance, shallow and irregular breathing, and hunched posture. Interestingly, clinical signs in immunized animals that were partially protected were delayed by four to nine days as compared to controls and were more neurological in nature (partial paralysis and loss of balance). Only immunization with 35?g recNP/ISA50, 35 and 70?g recNP/Alhydrogel and 35 and 70?g recNP/SaponinQ yielded significant protection from disease/death (Table 4.1). The best protection from disease/death (100%) was achieved by immunization with 35 and 70?g recNP/Alhydrogel, and 70?g recNP/SaponinQ. The least effective protection achieved was 17% after immunization with 70?g recNP and no adjuvant. The least effective Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 98 P. Jansen van Vuren protection achieved when using adjuvant with the recNP was 40% after immunization with 70?g recNP/ISA50. Mock inoculated mice did not develop any clinical signs during the experiment. Despite full or partial protection of some immunized mice from disease/death after RVFV challenge, the virus still replicated in immune mice, but to lower levels in liver and kidney tissues when compared to unvaccinated control mice. Generally, however, the amount of virus detected in brain tissues from immunized mice after challenge was higher when compared to control mice, with the exception of mice immunized with 70?g recNP combined with Alhydrogel or SaponinQ (Table 4.2). Mice that developed severe disease or succumbed had higher viral loads in liver, kidney and brain tissues compared to mice that were apparently healthy (Figure 4.4.). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 99 P. Jansen van Vuren A 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Days post infection Pe rc en ta ge s u rv iv a l (% ) B 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Days post infection Pe rc e n ta ge s u rv iv a l (% ) Figure 4.3. Protection of mice immunized with the 35?g (A) and 70?g recNP doses (B) against disease or death after RVFV challenge compared to placebo control mice (A, B). Immunized mice are indicated by solid lines (?) with different adjuvant/recNP combinations indicated as (*) for ISA50, (?) for TiterMax Gold, (+) for Alhydrogel, (?) for SaponinQ and (o) for 70?g recNP without adjuvant. Placebo control groups are indicated by dashed lines (---) and (*) for ISA50, (+) for Alhydrogel, (?) for SaponinQ and (?) for PBS placebo mice respectively. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 100 P. Jansen van Vuren Table 4.1. Mouse survival rates after RVFV challenge. Group Survivors/Total (% survival)a Significance of protectionb 35?g recNP/ISA50 4/6 (67%) p = 0.0303 70?g recNP/ISA50 2/5 (40%) p = 0.1818 35?g recNP/TMG 4/7 (57%) p = 0.0489 70?g recNP/TMG 3/5 (60%) p = 0.0606 35?g recNP/Alhydrogel 6/6 (100%) p = 0.0011 70?g recNP/Alhydrogel 5/5 (100%) p = 0.0022 35?g recNP/SaponinQ 4/6 (67%) p = 0.0303 70?g recNP/SaponinQ 6/6 (100%) p = 0.0011 70?g recNP 1/6 (17%) p = 0.500 Adjuvant control group ISA50 0/5 (0%) N/A Adjuvant control group Alhydrogel 0/5 (0%) N/A Adjuvant control group SaponinQ 0/6 (0%) N/A Placebo control group (PBS) 0/6 (0%) N/A a Survivors/total number of mice ratio. (%) is percentage survival b Significance of protection calculated using Fisher?s Exact test. P-values < 0.050 are considered significant. 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 Liver Kidneys Brain Lo g 1 0T CI D 5 0/g tis su e Figure 4.4. Mean RVFV TCID50/gram of tissues from dead or sick mice (grey bars) compared to healthy mice (white bars) from all experimental groups. Error bars indicate the standard deviations from the mean values. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 101 P. Jansen van Vuren Table 4.2. Viral load data in immunized and control mice after RVFV challenge. Group Organ tissue Number of mice Mean viral loada Range (S.D.)b 35?g recNP/ISA50 Liver 3 Negative * N/A Kidney 3 1.3 0.0 ? 4.0 (2.3) Brain 3 6.6 5.8 ? 8.0 (1.2) 70?g recNP/ISA50 Liver 4 Negative N/A Kidney 4 Negative * N/A Brain 4 7.1 6.0 ? 8.0 (0.7) 35?g recNP/TMG Liver 4 Negative N/A Kidney 4 0.9 0.0 ? 3.8 (1.9) Brain 4 7.4 7.0 ? 8.0 (0.5) 70?g recNP/TMG Liver 3 Negative * N/A Kidney 3 Negative * N/A Brain 3 6.5 0.0 ? 7.0 (0.5) 35?g recNP/Alhydrogel Liver 3 Negative N/A Kidney 3 1.3 0.0 ? 4.0 (2.3) Brain 3 4.8 3.0 ? 7.0 (2.1) 70?g recNP/Alhydrogel Liver 3 Negative N/A Kidney 3 Negative N/A Brain 3 1.3 0.0 ? 4.0 (2.3) 35?g recNP/SaponinQ Liver 4 1.0 0.0 ? 4.0 (2.0) Kidney 4 Negative * N/A Brain 4 3.9 0.0 ? 8.5 (4.5) 70?g recNP/SaponinQ Liver 3 Negative * N/A Kidney 3 1.3 0.0 ? 3.8 (2.2) Brain 3 1.5 0.0 ? 4.5 (2.6) 70?g recNP Liver 6 0.8 0.0 ? 4.5 (1.8) Kidney 6 3.0 0.0 ? 5.3 (2.4) Brain 6 6.0 0.0 ? 8.5 (3.1) Adjuvant control group (ISA50) Liver 5 3.5 0.0 ? 5.3 (2.0) Kidney 5 5.0 3.8 ? 5.3 (0.8) Brain 5 2.9 0.0 ? 4.3 (1.7) Adjuvant control group (Alhydrogel) Liver 5 4.5 3.8 ? 5.3 (0.9) Kidney 5 5.7 5.0 ? 6.5 (0.6) Brain 5 4.6 3.0 ? 6.5 (1.3) Adjuvant control group (SaponinQ) Liver 6 5.3 3.8 ? 6.8 (1.3) Kidney 6 5.8 4.5 ? 6.3 (0.7) Brain 6 4.2 3.0 ? 5.8 (1.1) Placebo control group (PBS) Liver 6 5.9 4.0 ? 7.3 (1.3) Kidney 6 5.3 4.3 ? 6.0 (0.6) Brain 6 2.5 0.0 ? 3.3 (1.2) Organs were collected from sick and healthy mice between day 2 and 15 after infection. a Viral loads are given as mean log10TCID50/g tissue b Range of log10TCID50 values and standard deviation from the mean *Indicates where qRT-PCR positives were detected in virus negative tissues Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 102 P. Jansen van Vuren 4.4.3 Discussion It was shown in this study that the bacterially expressed recombinant RVFV nucleocapsid protein (recNP) is highly immunogenic in BALB/c mice when used with different adjuvants (section 4.2). These antibodies, however, are not neutralizing in vitro or in vivo (section 4.3). To evaluate whether immunization with recNP could induce protective immunity in mice by a mechanism other than neutralization of the virus, immunized mice were challenged with virulent RVFV. The recNP/SaponinQ combination was best able to induce a strong IgG2A isotype response. Increased IgG2A isotype antibodies indicate activation of the Th-1 response, which indicates secretion of IFN-? and IL-2, two cytokines known to be involved in the cellular immune response. Interferon gamma (IFN-?) is secreted by natural killer cells (NK) and CD8+ cytotoxic T lymphocytes amongst others and has direct antiviral activity but also acts as an immunoregulatory factor. Interleukin 2 (IL-2) is secreted by T helper cells (Th) and acts by stimulating the growth, differentiation and survival of antigen specific CD8+ cytotoxic T lymphocytes. These factors might have played a role in significant protection of recNP/SaponinQ immunized mice from death/disease (67 ? 100% protection depending on recNP dose) and in decreasing replication of challenge virus. Irrespective of the recNP dose used, immunization with recNP/Alhydrogel resulted in 100% protection from morbidity/mortality and the lowest levels of virus replication after challenge. Interestingly, and contrary to findings in recNP/SaponinQ groups, the recNP/Alhydrogel antigen/adjuvant combination induced a low IgG2A response, even though IgG1 and total IgG levels were still comparable to those in other groups. This would indicate that recNP/Alhydrogel was indeed immunogenic but that the response was strongly biased towards the Th-2 humoral response. These results are consistent with previous findings showing no up-regulation of IFN-? and IL-2, and thus decreased IgG2A isotype antibodies by alum adjuvant (Ulanova et al., 2001). Alhydrogel, however, might directly activate NF-kB, a protein complex found in almost all cell types and that is involved in regulating cellular responses to an infection (Ulanova et al., 2001). Studies have shown that NF-kB is required for the positive selection of memory CD8+ T cells (Hettmann and Leiden, 2000, Hettmann et al., 2003). CD8+ T cells are able to kill virus-infected cells by inducing apoptosis, and have recently been shown to kill infected cells directly in the lymph nodes draining the infected site (Xu et al., 2007). This does not prevent infection but acts by limiting the spread of virus to organs, as it is the case for viruses that cause systemic infections such as RVFV. Likely the combination of recNP/Alhydrogel was able to induce a RVFV N-protein specific memory cellular response independently from the Th1/Th2 pathway by inducing NF-kB transcription factors. This in turn might have resulted in the production of N-protein specific memory CD8+ T cells recognizing specific fragments of the virus? N protein displayed on the surface of APC or other infected cells. Killing of those cells would curb the spread of Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 103 P. Jansen van Vuren the virus and lower its replication level. This assumption is supported by our observations of drastically lower viral loads in target organs from recNP/Alhydrogel vaccinated mice during the early stages after infection, as compared to virus concentration in placebo control mouse organs. RVFV initially infects the white blood cells in the proximity of the inoculation site and circulates in these cells for 2-3 days, after which large amounts of the virus is released into the bloodstream to infect other organs. Once the complete virus particles are released into the bloodstream the non-neutralizing anti- NP response becomes ineffective since the NP antigen cannot be accessed by cell receptors. However, during the initial stage when white blood cells are infected, these cells would present processed fragments of the N protein antigen on their surface which would make them targets for apoptosis if specific memory cytotoxic T cells are present. It is therefore proposed that recNP/Alhydrogel immunizations were able to protect mice from severe symptoms by inducing the generation of NP- specific memory CD8+ T cells that were able to limit the spread of virus by killing infected white blood cells before they could propagate and release large amounts of virus. ISA50 have been shown previously to direct the immune response against a specific antigen towards the Th-2 type response, involved in humoral immunity, rather than Th-1 (O'Hagan et al., 2001). We obtained similar results with our antigen, showing that mice immunized with recNP/ISA50 developed very strong IgG1 type responses compared to lower, but still respectable IgG2A responses. The protection rates of 40 ? 67% noted in this experimental group is probably due to the fact that Th-1 immunity was induced by the adjuvant, but to a lower level compared to SaponinQ. Mice immunized with both doses of recNP/TiterMax Gold generated very high levels of IgG1, but also intermediate IgG2A responses against recNP, which is consistent with previous findings (Cribbs et al., 2003). The increased Th-2 response over Th-1 pattern was very similar to that found when using ISA50 as adjuvant with recNP, and subsequently also resulted in similar protection rates of 57 ? 60%. The distribution and titres of virus in organs from recNP/ISA50 and recNP/TMG immunized mice were also almost identical indicating that these two adjuvants function by very similar mechanisms. Despite the recNP without adjuvant being able to induce a detectable humoral immune response after a booster immunization, it was not able to protect mice against morbidity/mortality after RVFV challenge. This is most probably due to absence of adjuvant to direct the response against the recNP towards cellular immunity, partly shown by a weak IgG2A response. This result confirms the importance of selecting an adjuvant that directs the immunity towards the correct type of response for the specific antigen used. Interestingly, all recNP/ISA50 or TMG vaccinated mice had similar, or in most cases higher viral loads in brain tissues from day 6 p.i. onwards when compared to placebo control mice, whereas during the first 5 days p.i. placebo control mice had high titres of replicating virus in all tissues tested. This organ specific tropism might explain the clinical observation of some Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 104 P. Jansen van Vuren vaccinated mice developing delayed severe neurological symptoms, as opposed to more general severe symptoms in placebo control mice during the early days after infection. From these results it seems that recNP/ISA50 or TMG vaccinated mice were successful in averting initial disease or death by decreasing viral loads in other target organs, but were not able to prevent RVFV from crossing the blood-brain barrier and viral replication to damaging levels. 4.5 Conclusion This chapter describes efforts to evaluate whether the response against the nucleocapsid protein of RVFV plays a role in protection against viral infection. To achieve this, different adjuvants with different mechanisms of enhancing immunity was used with a bacterially expressed RVFV nucleocapsid protein to immunize mice. The recNP was highly immunogenic, even in the absence of adjuvant. These anti-recNP humoral antibodies, however, had no neutralizing ability, either in vitro or in vivo. Despite this, mice immunized with recNP were protected from morbidity/mortality caused by challenge RVFV and depending on adjuvant used, 50 ? 100% protection was achieved. Two recNP/adjuvant combinations resulted in significant protection (100%) and reduction of challenge virus replication in organs. The results show that one of these protective antigen/adjuvant combinations (recNP/SaponinQ) induced the strongest IgG2A isotype response compared to other adjuvants in this study, indicating activation of Th-1 cellular immunity. The assays used were not able to indicate activation of cellular immunity by the other protective antigen/adjuvant combination (recNP/Alhydrogel), but this combination might have induced the production of NP-specific memory CD8+ T-cells by a separate pathway, as indicated by results from previous studies with alum adjuvant (Ulanova et al., 2001). The RVFV recNP combined with adjuvants that are known to bias responses towards Th-2 humoral immunity (ISA-50, TiterMax Gold) (O'Hagan et al., 2001, Cribbs et al., 2003) induced weaker IgG2A responses compared to SaponinQ, protected only 40 ? 67% of mice from morbidity and mortality, and were not able to sufficiently curb viral replication in important organs. Results from this study suggest that the anti-NP response play a role in protection of mice against RVFV infection. Even though obtained results indicate that cellular immunity is the major role player further experiments were needed to support this hypothesis. This was attempted and results are presented in chapter 6. Based on the promising results obtained in a mouse model, it was decided to evaluate the recNP as an immunogen in a host animal species. Therefore, additional studies have been undertaken, of which results are presented in chapter 5. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 105 P. Jansen van Vuren CHAPTER FIVE RECOMBINANT NUCLEOCAPSID PROTEIN AS AN IMMUNOGEN IN A RVF HOST ANIMAL SPECIES* * Partially published as: Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2010). Evaluation of a recombinant Rift Valley fever virus subunit nucleocapsid protein as an immunogen in mice and sheep. The Open Vaccine Journal 3, 114-126 * Partially presented at international conferences as: Jansen van Vuren, P. & Paweska, J.T. (2009). Preliminary evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as an immunogen in combination with different adjuvants in mice and sheep. FAO/IAEA International Symposium on Sustainable Improvement of Animal Production and Health, Vienna, Austria, 8 ? 11 June 2009. (Poster presentation). Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2009). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine immunogen in combination with four adjuvants. International Meeting on Emerging Diseases and Surveillance, Vienna, Austria, 13 ? 16 February 2009. (Poster presentation). 5.1 Introduction The advent of molecular biology has enabled development of various novel vaccine candidates for RVF. These next generation vaccines offer advantages over classical vaccines in that they are safe, easy and less expensive to produce. All the recent vaccine candidates have been evaluated in mice or rats, but very few have been further evaluated in a host ruminant animal species. A recombinant lumpy skin disease virus (LSDV) containing both RVF glycoprotein genes induces strong neutralizing responses and 100% protection from clinical disease in sheep (Wallace et al., 2006). A SINV replicon-based RVF vaccine produces neutralizing responses in sheep but protection against viral challenge was not evaluated (Heise et al., 2009). Immunization of sheep with a DNA construct expressing the RVFV M segment and the nucleocapsid protein was not able to elicit detectable humoral responses but low level antigen-specific cellular responses were induced (Lorenzo et al., 2008). A construct expressing only the nucleocapsid protein, however, is able to induce strong anti-NP IgG1 isotype responses as well as cellular responses in sheep (Lorenzo et al., 2008), although protection against viral challenge was not evaluated. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 106 P. Jansen van Vuren The role of the anti-NP response in the protection of a host animal species against RVF viral challenge has not been evaluated before. The recNP used in this study induced protective immune responses in mice; therefore it was decided to evaluate its protective ability against RVFV challenge in sheep using the same recNP/adjuvant combinations as used for mice. 5.2 Immunogenicity of the recombinant nucleocapsid protein alone and in combination with four different adjuvants in sheep 5.2.1 Materials and methods 5.2.1.1 Immunization of sheep The recombinant RVFV nucleocapsid protein (recNP) was produced as described before (section 2.1.2.4). Sheep were pre-screened for antibodies against RVFV using enzyme linked immunosorbent assay (ELISA). Twenty three adult female Dorper cross sheep, younger than one year, were used. The sheep were divided into groups as described in table 5.1. All sheep were inoculated subcutaneously (s.c.) and received identical booster inoculations as described in table 5.1. Serum was collected at regular intervals, as given in table 5.1, for monitoring of immune responses. Table 5.1. Group assignments and immunization schedules of sheep. Group number Number of sheep RVFV recNP dose (?g) Adjuvant/Inoculum Immunization schedule Blood collection schedule (day after immunization/booster) 1a n = 2 175 ISA50 Day 0 (initial) Day 21 (booster) Day 0, Day 14, Day 26 Day 14 after booster 1b n = 2 350 ISA50 2a n = 2 175 Alhydrogel 2b n = 2 350 Alhydrogel 3a n = 2 175 TiterMax Gold? 3b n = 2 350 TiterMax Gold? 4a n = 2 175 SaponinQ 4b n = 2 350 SaponinQ 5 n = 4 0 ISA50, Alhydrogel, TiterMax Gold? or SaponinQ Day 0 (initial mock) Day 21 (mock booster) 6 n = 3 0 PBS 5.2.1.2 Monitoring of immune responses Immune responses in sheep after immunization was monitored by an indirect ELISA based on the recNP as follows: Immunoplates (Maxisorb, Nunc, Denmark) were coated with RVFV recNP antigen at a dilution of 1:2000 in Carbonate-Bicarbonate buffer (pH 9.6) and incubated overnight at 4?C. After washing three times with a washing buffer consisting of phosphate buffered saline (PBS) pH7.2 and 0.1% Tween-20, the plates were blocked with 200?l of 10% fat free milk powder (?Elite?, Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 107 P. Jansen van Vuren Clover SA, Pty, Ltd.) in PBS at 37?C for 1h and then washed as before. Control and test sera were diluted 1:400 in diluent buffer consisting of 2% fat free milk powder in PBS, 100?l added to each well and incubated for 1h at 37?C. Sheep internal controls were generated as described by Paweska et al. in 2003 [36]. Controls were tested in quadruplicate and test samples in duplicate. After washing as before, 100 ?l of rabbit anti-sheep IgG HRPO (Zymed Laboratories, Invitrogen, USA) at 1:6000 dilution was added to plates. After 1h incubation at 37?C plates were washed as before and 100 ?l of 2,2?- azinodiethylbenzthiazoline sulfonic acid (ABTS, KPL Laboratories, Inc.) added to each well. After 30 min incubation in the dark the reaction was stopped by the addition of 100 ?l of 1% sodium dodecyl sulphate (SDS) to each well. Optical density (OD) was determined at 405nm and the results expressed as mean OD values. Means and standard deviations from the means were determined based on two animals per group. 5.2.2 Results All sheep immunized with recNP combined with adjuvants produced detectable anti-NP IgG responses by day 14 after one immunization (Figure 5.1). The anti-NP antibodies in immunized sheep were consistently equal to or higher than the same antibodies in the high positive control serum from an experimentally infected sheep (dotted vertical line in figure 5.1). The second immunization of all sheep had the desired effect of boosting immune responses. The combination of recNP with Alhydrogel (group 2 and 6) was the least immunogenic but still induced strong responses compared to the positive control. The larger recNP dose (350?g) did not induce much stronger responses in any of the immunized groups when compared to the lower dose (175?g). As expected, the adjuvant and PBS control groups did not develop any anti-NP responses during the immunization period. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 108 P. Jansen van Vuren 1a 1a 1a 1a 2a 2a 2a 2a 3a 3a 3a 3a 4a 4a 4a 4a 1b 1b 1b 1b 2b 2b 2b 2b 3b 3b 3b 3b 4b 4b 4b 4b 5 5 5 56 6 6 6 0 0.5 1 1.5 2 2.5 3 Day 0 Day 14 Day 26 Day 14 Boost EL IS A OD v al u e (40 5n m ) Figure 5.1. Average anti-NP responses in sheep after recNP first and booster immunizations; error bars indicate standard deviations from the means. 1) 175?g recNP/ISA50 (n=2); 2) 175?g recNP/Alhydrogel (n=2); 3) 175?g recNP/TiterMax Gold (n=2); 4) 175?g recNP/SaponinQ (n=2); 5) 350?g recNP/ISA50 (n=2); 6) 350?g recNP/Alhydrogel (n=2); 7) 350?g recNP/TiterMax Gold (n=2); 8) 350?g recNP/SaponinQ; 9) Adjuvant controls (n = 4); 10) PBS controls (n=3); 11) 350?g recNP (n=2). The vertical dotted line indicates the average OD reading for a high positive control serum. The solid vertical line indicates the average OD reading for a negative control serum. 5.2.3 Discussion The recNP combined with adjuvants was highly immunogenic in sheep, even after a single immunization. The strength of the responses did not depend so much on the dose of recNP used (175 vs. 350?g), but seemed more dependent on the adjuvant used. The highest immunogenicity was achieved when combining recNP with SaponinQ or TiterMax Gold. Alhydrogel seemed to be the least effective of the adjuvants in inducing humoral anti-recNP responses. None of the anti-recNP responses in sheep were neutralizing. The responses measured were IgG antibodies, and from earlier results in this study it became clear that humoral immunity against the RVFV NP does not play a role in protection against infection. Some of the adjuvants used are known to induce cellular immunity and although the cellular response was not measured in this study, based on the strength of the humoral responses one would think they Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 109 P. Jansen van Vuren were activated. In a recent study it was shown that sheep, immunized with a DNA construct expressing the RVFV NP, developed strong humoral as well as lymphoproliferative responses (IFN-?) (Lorenzo et al., 2008). 5.3 Rift Valley fever virus challenge of immunized sheep 5.3.1 Materials and methods 5.3.1.1 Cells and virus Cells and virus were cultured as described before (section 4.3.1.1). 5.3.1.2 RVFV challenge All sheep were challenged s.c. with 2 ml challenge virus (1 ml on both sides of the neck). Sheep were challenged at different times as follows: one sheep from each sub-group (group 1a,b ? 4a,b), all sheep from group five and one sheep from group six were challenged on day 37 after the booster immunization (total = 13 sheep); the remaining sheep were challenged on day 168 after the booster immunization (total = 10 sheep). Sheep were monitored daily for the first two weeks after challenge and blood taken daily for the first seven days, and at regular intervals thereafter to monitor viremia and immune responses until day 70. 5.3.1.3 Immune response monitoring after RVFV challenge Immune responses in sheep after challenge were monitored by IgM capture ELISA as described previously (Paweska et al., 2003a). A virus neutralization test (VNT) was performed as described before (section 2.1.2.3). Means and standard deviations for IgM ELISA percentage positivity values and VNT titres were based on data from minimum two animals per group. 5.3.1.4 Virus titrations Virus titrations of sheep sera collected after challenge were performed as described before (section 3.2.7). Means and standard deviations from the means were determined based on two or more animals per group. 5.3.1.5 Statistical methods The significance of differences between immune responses and viremia in sheep was confirmed using the Fisher F-test giving a two-tailed probability value (Excel, Microsoft Office). P- values lower than 0.01 were considered to be significant. Mean values and standard deviations from the means were calculated using at least two sheep per group. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 110 P. Jansen van Vuren 5.3.2 Results 5.3.2.1 Immune responses in sheep after RVFV challenge Because the dose of recNP did not have a significant impact on the strength of the humoral responses, groups that were immunized with the same recNP/adjuvant combination, regardless of dose, were grouped together for the RVFV challenge experiment. All adjuvant control sheep were regarded as one group, and all PBS control sheep were regarded as one group, regardless of when they were challenged with RVFV. The sheep IgM responses after challenge are shown in figure 5.2a and in figure 5.2b. None of the immunized or control sheep had any detectable RVFV specific IgM antibodies on the day of challenge. High levels of RVFV specific IgG, however, was detected in all immunized sheep on the day of challenge, but as expected not in control sheep (results not shown). Control sheep developed typical IgM responses after RVFV challenge. All immunized sheep developed lower IgM responses after challenge when compared to control sheep, except the recNP/Alhydrogel immunized sheep that were challenged on day 168, which developed elevated IgM responses when compared to other immunized sheep. Incidentally, these recNP/Alhydrogel immunized sheep had decreased RVFV specific IgG on the day 168 when they were challenged (results not shown). The virus neutralizing antibody responses after challenge are shown in figure 5.3a and figure 5.3b. Immunization did not have significant effect on decreasing the development of virus neutralizing antibodies when compared to PBS control sheep: recNP/ISA50 (day 37, p = 0.883; day 168, p = 0.825 Fisher F-test), recNP/Alhydrogel (day 37, p = 0.920; day 168, p = 0.850), recNP/TiterMax Gold (day 37, p = 0.881; day 168, p = 0.975) and recNP/SaponinQ (day 37, p = 0.682; day 168, p = 0.858). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 111 P. Jansen van Vuren Figure 5.2. Mean IgM responses in sheep after RVFV challenge on day 37 (a) or 168 (b). Sheep groups are indicated as recNP/ISA50 (--+--), recNP/Alhydrogel (???), recNP/TiterMax Gold (???), recNP/SaponinQ (?x?), adjuvant controls (???) and PBS controls (???). Error bars indicate standard deviations from the means of two or more sheep per group. 0 50 100 150 200 250 300 Day 0 Day 4 Day 6 Day 14 Day 70 Days post infection (challenge day 37) M ea n EL IS A pe rc en ta ge po si tiv ity (% ) 0 50 100 150 200 250 300 Day 0 Day 4 Day 6 Day 14 Day 70 Days post infection (challenge day 168) M ea n EL IS A pe rc en ta ge po si tiv ity v al u e (% ) A B Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 112 P. Jansen van Vuren Figure 5.3. Mean virus neutralizing antibody responses in sheep after RVFV challenge on day 37 (a) or 168 (b). Sheep groups are indicated as recNP/ISA50 (--+--), recNP/Alhydrogel (???), recNP/TiterMax Gold (???), recNP/SaponinQ (?x?), adjuvant controls (???) and PBS controls (???). Error bars indicate standard deviations from the means of two or more sheep per group. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Day 0 Day 4 Day 6 Day 14 Day 70 Days post infection (challenge day 37) M ea n Lo g1 0 VN T tit er 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Day 0 Day 4 Day 6 Day 14 Day 70 Days post infection (challenge day 168) M ea n Lo g1 0 VN T tit er A B Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 113 P. Jansen van Vuren 5.3.2.2 Viremia in sheep after RVFV challenge The viremia in sheep after RVFV challenge is shown in Table 5.2. Immunization of sheep did not result in significant decrease of viral loads in sera when compared to PBS control sheep. Viremia was, however, of two to four days duration whereas one PBS control sheep developed prolonged viremia up to day seven. None of the sheep, including controls, displayed any clinical signs. 5.3.3 Discussion Despite the recNP being highly immunogenic in sheep when combined with adjuvants (section 5.2), immunity against recNP was not able to decrease the replication of challenge virus in sheep. The IgM responses in immunized sheep after challenge was considerably lower when compared to na?ve sheep that were challenged with the same virus. However, this is most probably due to the fact that the IgM ELISA detects antibodies against the whole RVFV, including the glycoproteins. When taking into consideration that virus neutralizing antibody titres, which consists solely of anti-glycoprotein antibodies, were equal in immunized and control sheep, it appears that the decrease in IgM responses in immunized sheep versus control sheep was because anti-NP IgM antibodies were not produced by immunized sheep since they had already been exposed to this antigen during immunization. This is further substantiated by the fact that there was no significant decrease in viremia in serum of immunized sheep after challenge when compared to control sheep. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 114 P. Jansen van Vuren Table 5.2. Viremia in immunized and control sheep after RVFV challenge. Inoculum Number of sheep Days post infection Viremia Mean Log10TCID50/ml?Standard deviation (Range) Significance of decreased viremiaa recNP/ISA50 (1a-b) (175 and 350?g combined) Challenge day 37 Challenge day 168 2 1 4.5?0.4 (4.3 to 4.8) p = 0.55 2 4.4?0.2 (4.3 to 4.5) 3 0.8?1.1 (0.0 to 1.5) 4-7 Negative 2 1 3.6?0.5 (3.3 to 4.0) p = 0.37 2 6.1?0.5 (5.8 to 6.5) 3-7 2.0?1.8 (0.8 to 3.3) recNP/Alhydrogel (2a-b) (175 and 350?g combined) Challenge day 37 Challenge day 168 2 1 2.6?2.7 (0.8 to 4.5) p = 0.37 2 2.0?2.8 (0.0 to 4.0) 3 1.1?1.6 (0.0 to 2.3) 4-7 Negative 2 1 3.4?0.2 (3.3 to 3.5) p = 0.79 2 4.4?2.7 (2.5 to 6.3) 3 2.4?3.4 (0.0 to 4.8) 4 0.5?0.7 (0.0 to 1.0) 5-7 Negative recNP/TiterMax Gold (3a-b) (175 and 350?g combined) Challenge day 37 Challenge day 168 2 1 4.3?1.1 (3.5 to 5.0) p = 0.88 2 2.6?1.2 (1.8 to 3.5) 3-7 Negative 2 1 4.3?0.4 (4.0 to 4.5) p = 0.49 2 5.0?1.1 (4.3 to 5.8) 3 1.9?1.2 (1.0 to 2.8) 4-7 Negative recNP/SaponinQ (4a-b) (175 and 350?g combined) Challenge day 37 Challenge day 168 2 1 4.4?0.5 (4.0 to 4.8) p = 0.52 2 4.8?0.4 (4.5 to 5.0) 3 1.6?2.3 (0.0 to 3.3) 4-7 Negative 2 1 1.9?2.7 (0.0 to 3.8) p = 0.19 2 2.0?2.8 (0.0 to 4.0) 3 0.4?0.5 (0.0 to 0.8) 4-7 Negative Adjuvant control (ISA50, Alhydrogel, TiterMax Gold and SaponinQ combined) Challenge day 37 4 1 5.1?0.3 (4.8 to 5.5) p = 0.40 2 4.8?1.1 (3.8 to 6.0) 3 0.8?1.5 (0.0 to 3.0) 4-7 Negative PBS control Challenge day 37 and 168 3 1 4.5?0.3 (4.3 to 4.8) Control group 2 5.3?2.3 (3.3 to 7.8) 3 2.3?4.0 (0.0 to 7.0) 4 1.9?3.3 (0.0 to 5.8) 5 1.8?3.2 (0.0 to 5.5) 6 1.5?2.6 (0.0 to 4.5) 7 0.8?1.4 (0.0 to 2.5) a Indicates the statistical significance of decrease in viral load as compared to PBS control sheep, as calculated by the Fisher F-test. P-values ? 0.01 indicate a statistically significant decrease in viremia. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 115 P. Jansen van Vuren 5.4 Conclusion This chapter describes the first evaluation of the recNP, combined with adjuvants, as an immunogen in a RVF host species, sheep, and protection against viral replication after RVFV challenge. Vaccination against arthropod borne viruses should ideally aim to decrease morbidity and mortality, but even more importantly it should stop the spread of the virus by inducing sterilizing immunity. The only study on record where immune responses against specifically the NP were evaluated (Lorenzo et al., 2008) reported on humoral and lymphoproliferative responses induced after immunization with a DNA construct expressing the NP. The protective ability of these responses was not evaluated. The findings from this study highlight some important aspects that should be considered for future research and development of vaccine candidates for RVF. Firstly the anti-nucleocapsid response alone, although protective in mice, does not seem to play a role in the protection of an actual host species against RVFV infection. It must be noted though that in mice protection was evaluated as survival and decrease of viral replication in organs, whereas in sheep protection was based solely on viremia in the blood. The reason for this is that it is almost impossible to produce disease in sheep older than a few months with experimental RVFV infection. For this reason viremia, which is also an important factor in protection since high viremia would lead to subsequent infection of feeding mosquitos, was measured in this study. Therefore it would be more accurate to state that anti-NP responses were not able to curb viremia in the blood of sheep. Secondly, the results show that results in mice cannot necessarily be extrapolated to a host species with the expectation of achieving the same level of protection. It must be remembered that mice are merely used because in some cases the disease caused by viral infection mimicks some of the symptoms seen in actual host species, in addition to their ease of handling and low cost. Therefore RVFV candidate vaccines should always be evaluated in a host species first before conclusions can be drawn about its efficacy. Although the target proteins of choice for RVFV vaccines are glycoproteins because of inducing neutralizing antibody, RVFV vaccine candidates targeting the glycoproteins which were evaluated in mice have also yielded inconsistent protection against challenge (Wallace et al., 2006, Mandell et al., 2009, Naslund et al., 2009). In a recent study it was shown that immunization with VLPs combining the glycoproteins and nucleocapsid protein yielded better protection (Mandell et al., 2009). Therefore it appears that vaccine candidates combining glyco- and nucleocapsid proteins should be further investigated. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 116 P. Jansen van Vuren CHAPTER SIX HOST GENE EXPRESSION IN MICE IMMUNIZED WITH RECOMBINANT NUCLEOCAPSID PROTEIN AND CONTROL MICE AFTER RIFT VALLEY FEVER VIRUS INFECTION* * Partially published: Jansen van Vuren, P., Tiemessen, C.T. & Paweska, J.T. (2011). Anti-nucleocapsid protein immune responses counteract pathogenic effects of Rift Valley fever virus infection in mice. PLoS One 6(9): e25027 * Presented as: Jansen van Vuren, P., Tiemessen, C.T., Paweska, J.T. (2011). Rift Valley fever pathogenesis in a mouse model ? host gene expression in liver of mice immunized with a recombinant nucleocapsid protein. 30th World Veterinary Congress, Cape Town, South Africa, 10-14 October 2011. (Oral presentation). 6.1 Introduction As shown in Chapters 4 and 5, the bacterially expressed recombinant RVFV N-protein was highly immunogenic in mice and sheep, and elicited 100% protection against lethal challenge in mice when combined with certain adjuvants. However, the mechanism by which immunization with this protein elicits protection is still not clear, however. A deeper look into these mechanisms might also reveal the discrepant results between mice and sheep in this study. The N-protein is not a viral surface protein and therefore does not play any role in virus entry into host cells. It is therefore not surprising that anti-recNP humoral antibodies from immunized mice and sheep in this study were not able to neutralize the virus in vitro or in vivo (Chapter 4). The N-protein is the main RVFV immunogen and strong humoral responses have been detected in various species against the RVFV recNP (Jansen van Vuren et al., 2007, Paweska et al., 2007, Paweska et al., 2008b) which has led some to suggest that N- protein acts as a decoy protein (Lorenzo et al., 2008). This theory is, however, not supported by the successful protection of mice by recNP immunization in this study. Low level N-protein specific cellular responses have been noted in sheep after immunization with a DNA vaccine expressing NP (Lorenzo et al., 2008). However, in this study the recNP/adjuvant combination that induced the weakest IgG2a isotype response (indicative of weak Th1 cellular immunity), but still a strong IgG1 isotype response (indicative of activation of Th2 humoral immunity) (Figure 4.1 e-f), in mice after immunization resulted in the best protection against RVFV challenge. This indicates that the activation Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 117 P. Jansen van Vuren of cellular immunity against NP might not play such a substantial role in protection against RVFV infection. The host defence against viral infection is a complex response consisting of two categories: innate and adaptive immunity. The innate immune system is the first line of defence against infections but is not pathogen specific and is mainly comprised of the complement system, cytokines, natural killer cells, macrophages and apoptosis (Strauss and Strauss, 2008). Cytokines are a family of proteins that also have a regulatory role in the host adaptive immune system, especially interleukins, whereas interferons also have direct antiviral action. Type I interferons, IFN? and IFN?, are especially important in vertebrates for controlling viral infections and are produced by almost all cell types in the host. Type I interferons are induced mainly by the detection of double-stranded RNA, an intermediate product in viral replication, by Toll-like receptors on the surface of cells or helicases within the cell (Strauss and Strauss, 2008, Haller and Weber, 2009). Once induced, type I interferons not only stimulate the adaptive immune response by increasing production of class I major histocompatibility complex (MHC I) molecules, but they also create an antiviral state in host cells, thus preventing or decreasing viral replication in those cells. This antiviral state involves antiviral pathways such as protein Kinase R (PKR), 2-5 OAS/RNaseL and the Mx proteins (Strauss and Strauss, 2008, Haller and Weber, 2009) that interferes with viral mRNA translation. RVFV is sensitive to the actions of type I interferons (Anderson and Peters, 1988, Morrill et al., 1990, Sandrock et al., 2001, Peterss et al., 2009) but the virus has developed several mechanisms by which it counteracts the actions of thereof. The NSs protein of RVFV forms filaments in the nuclei of infected cells and interacts with a repressor complex (Sin3A/NCoR/HDAC) inhibiting transcriptional activation of the IFN? gene (Le May et al., 2008, Bouloy and Weber, 2010). A more general shutdown of cellular gene expression is also caused by interaction of NSs with the p44 subunit of the TFIIH basal transcription factor, resulting in reduced transcriptional activity in RVFV infected cells (Le May et al., 2004). The NSs has also been shown to act on a post-translational level by degrading PKR, a protein responsible for the shutdown of translation of viral proteins (Habjan et al., 2009b). Interferon gamma (IFN?) is the only type II interferon and is produced by natural killer T cells (NKT) and NK cells as part of the innate immune response, and by cytotoxic T lymphocytes (CTL) and Th-1 cells as a part of the memory cellular response (Strauss and Strauss, 2008). IFN? is a strong immunoregulator altering the transcription of a number of genes which, amongst others, leads to increased production of MHC I and II molecules, suppression of Th-2 humoral immunity and activation of NK cells. It also has direct antiviral properties. However, the role that IFN? plays in protection of the host against RVFV infection is debatable. A study in rhesus monkeys showed that prophylactic treatment with recombinant human IFN? before RVFV infection protected monkeys from clinical disease, and decreased viremia Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 118 P. Jansen van Vuren significantly (Morrill et al., 1991b). A recent study, however, showed that there was no marked difference in pathogenicity of RVFV MP-12 or Clone-13 in wild type mice compared to mice deficient in IFN? receptor (IFNGR-/-) suggesting that IFN? only plays a negligible role in RVFV attenuation (Bouloy et al., 2001). Another non-structural protein of RVFV, encoded by the M segment (NSm), was recently implicated in the pathogenesis of RVF by acting as an anti-apoptotic protein (Won et al., 2007). Apoptosis is a controlled process in the host and a way of eliminating cells that are infected before they can produce a large progeny of virus (Strauss and Strauss, 2008). Another antigen-independent mechanism of killing infected cells is carried out by NK cells. These cytolytic cells express two separate sets of receptors. One of these sets of receptors interacts with MHC I molecules on host cells, which inhibits killing of the host cell. The other set of NK receptors interacts with activating molecules on infected hosts cells, resulting in the stimulation of NK cells to kill the target cell. NK cells also kill any cells that are not expressing MHC I (or low amounts thereof) as a result of the ability of some pathogens to inhibit MHC I expression in infected cells to evade CTL responses (Strauss and Strauss, 2008). NK cells can also play a role in adaptive immunity by means of antibody-dependent cell- mediated cytotoxicity (Weiner and Adams, 2000). This is a process by which antigen specific antibodies form a bridge between a virus infected cell displaying that viral antigen on its surface, and NK cells which bind the Fc portion of antibodies and results in lysis of the target cell. The adaptive immune response consists of a cellular arm, making use of cytotoxic T lymphocytes (CTL), and a humoral arm, making use of B-lymphocytes (B-cells) that secrete antibodies, with helper T-lymphocytes (Th-1 or Th-2) activating these cells and directing responses (Strauss and Strauss, 2008). Humoral immunity is important against extracellular pathogens, such as for example RVFV circulating in the bloodstream of its host. A B-cell displaying an antibody on its surface would recognize an extracellular viral antigen, leading to activation, followed by a second signal from a Th-2 cell after recognition of a fragment of the antigen displayed by MHC II molecule, causing the activated B-cell to proliferate and produce more cells capable of producing and secreting the same antibody (Strauss and Strauss, 2008). Antibodies play an important role in the control of viral infections by direct inhibition of virus entry into host cells (neutralization), coating of virus for subsequent removal by macrophages and activation of the complement cascade that leads to opsonisation, phagocytosis, chemotaxis or lysis. As discussed earlier, strong humoral responses against the RVFV NP is elicited after RVFV infection but these antibodies are not neutralizing in vitro or in vivo (Jansen van Vuren et al., 2010, Lorenzo et al., 2010), most probably because the NP plays no role in viral entry into host cells and is not found on the envelope of the virus and thus cannot be coated with antibodies for removal by macrophages. Humoral responses generated against the RVFV Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 119 P. Jansen van Vuren glycoproteins are, however, neutralizing and a correlate of protection against viral challenge (Wallace et al., 2006, Heise et al., 2009, Kortekaas et al., 2010a, Lorenzo et al., 2010, Mandell et al., 2010a). Cellular immunity is important against intracellular pathogens, such as for example RVFV infecting host cells. Infected cells present peptide fragments of viral proteins in a class I MHC context, which is recognized by T-cell receptors (TCR) on CD8+ T cells (cytotoxic T cells) that consequently become activated when a second co-stimulatory signal is present (Strauss and Strauss, 2008). An activated CTL requires further stimulation by cytokines (i.e. IL-2, IFN?), supplied by Th-1 cells, to enable it to proliferate and mount a vigorous cellular response against infection. The activation of the cellular arm of the adaptive immune response to RVFV infection has not been shown. A recent study, evaluating RVF virus like particles (VLP) containing both glycoproteins and NP as a vaccine, showed the secretion of some cytokines from RVFV induced spleen cells of mice 31 days post immunization with VLP (Mandell et al., 2010a). These cytokines included those associated with Th1 cellular immunity (IL-2, IFN? and IL-12) and Th2 humoral immunity (IL-4 and IL-5). A few recent studies have utilized microarrays or quantitative PCRs to analyze host gene expression in response to arboviral infections (Venter et al., 2005, Calzavara-Silva et al., 2009, Nascimento et al., 2009, do Valle et al., 2010, Momose et al., 2010). One study in particular utilized microarray and quantitative PCR to show a critical role for host innate immunity in resistance to RVF (do Valle et al., 2010). The study showed that a specific strain of mice, BALB/cByJ, was more resistant to RVFV infection when compared to a wild mouse strain, MBT/Pas. The study analyzed the expression of genes involved in the innate immune response by infecting mouse embryonic fibroblasts (MEF) from both mouse strains with RVFV in vitro, with results indicating a more significant type I IFN response in the BALB/cByJ MEFs. The results from this study are a further indication of the involvement of innate immunity, especially type I IFN, in the host?s fight against RVFV infection. The activation of adaptive immunity after RVFV infection on a gene expression level, and in vivo in a known RVFV target organ, has not been shown yet. A better understanding of the activation of memory humoral and cellular immune responses might provide some useful information for future RVFV vaccine developments, or even show some genes that might be targets of gene therapy or antivirals. The role that anti-NP responses play in the protection of vaccinated individuals against RVFV infection is also not well understood. In an attempt to elucidate the protective mechanism of anti-recNP responses, the regulation of expression of certain genes involved in the activation of T- and B-cell immunity, and innate immunity, were analysed by Real-Time PCR and relative quantification in three organs known to be important in RVFV pathogenesis. Relative quantification is a method by which expression levels of genes in treated subjects are related to expression levels of the same genes in untreated subjects (Livak and Schmittgen, Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 120 P. Jansen van Vuren 2001, Schmittgen and Livak, 2008). In this study the three treated groups, consisting of recNP/Alhydrogel immunized mice, Alhydrogel mock-immunized mice and PBS mock-immunized mice, were challenged with RVFV. The untreated group did not receive any immunization and was mock-challenged with tissue culture supernatant instead of RVFV. In this study we report that mice immunized with recNP combined with Alhydrogel adjuvant induced a strong IgG1 immune response, but weak IgG2A, indicating that the response was biased towards Th-2 humoral immunity rather than Th-1 cellular immunity. This humoral immunity still succeeded in protecting mice against clinical disease and decreasing viral load up to 104 fold in liver, the main target organ during RVFV infection, compared to control mice, which is consistent with previous results (Jansen van Vuren et al., 2010). The expression of type I IFN is upregulated in the liver of immunized mice shortly after RVFV challenge, compared to an initial downregulation and subsequent delayed upregulation of the same gene in the liver of non-immunized mice. In the acute phase of liver infection, however, a massive upregulation of type I and II interferon occurs in the presence of high viral titres in non-immunized mice, compared to immunized mice. It also shows the up- and downregulation of several genes involved in the activation of B- and T-cells in liver of non-immunized mice at the acute phase of RVFV infection, confirming that both cellular and humoral immunity are activated during RVFV infection in a mouse model. Some of these genes are involved in other immune functions as well. Various genes with pro-apoptotic effects were strongly upregulated, and anti-apoptotic genes downregulated in non-immunized mice. There was also upregulation of several genes involved in pro- inflammatory responses in liver of non-immunized mice. 6.2 Materials and Methods 6.2.1 Immunization and Rift Valley fever virus challenge of mice 6.2.1.1 Mouse immunization The recombinant RVFV nucleocapsid protein (recNP) was produced as described in section 2.1.2.4. Four-week old female BALB/cOlaHsd (Harlan Laboratories, U.K LTD) mice were used as an experimental animal model. The immunized group (MS-1) consisted of 85 mice each immunized with a 200?l inoculum containing 70?g RVFV recNP in combination with Alhydrogel (Sigma, U.S.A). The adjuvant control group (MS-2) consisted of 40 mice ?mock?-immunized with Alhydrogel in PBS. The placebo control group (MS-3) consisted of 40 mice which were ?mock?-immunized with PBS buffer. The normal control group (MS-4) consisted of 40 mice and was not inoculated with anything at this point but kept as a control group for the challenge experiment. All mice (except for MS-4) were inoculated subcutaneously (s.c) and received identical booster immunizations at 14 days after the initial immunization. Three mice from the immunized group MS-1 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 121 P. Jansen van Vuren was sacrificed and heart-bled on the following days to monitor immune responses: day 0, 3, 5, 7, 10 and 12 after the immunization and day 0, 3, 5, 7, 12, 18, 21 and 27 after the booster. 6.2.1.2 Cells and virus Cells and virus were cultured as described before (section 4.3.1.1). 6.2.1.3 RVFV challenge The remaining mice in each group (MS-1 = 40 mice; MS-2 = 38 mice and MS-3 = 40 mice) after the immunization period were challenged with RVFV on day 28 after the booster immunization. Mice were inoculated subcutaneously (s.c.) with a 100 ?l inoculum containing 106.5 TCID50/ml RVF challenge virus, and after challenge examined twice daily for signs of clinical illness. The normal control group (MS-4 = 39 mice) was ?mock?-infected with tissue culture medium (EMEM) without virus to act as an untreated control group for comparison of up- and downregulation of genes. Animals displaying severe illness were euthanized and organs collected. Three (3 biological replicates) mice were euthanized from each group and liver, spleen and brain tissues collected into RNAlater (QIAgen, Germany), to preserve RNA integrity, at the following time points after infection/?mock?-infection: 3 hours, 6 hours, 12 hours, 24 hours, 72 hours and 120 hours. The same tissues were also collected for virus titration (thus not into RNAlater). All tissue samples were stored at -70?C until RNA extraction (tissues in RNAlater) or virus titration (tissues not in RNAlater). 6.2.1.4 Determination of viral loads in mouse tissues Mouse liver, spleen and brain tissues were homogenized as 10% (w/v) suspensions in EMEM containing L-Glutamine, non-essential amino acids and antibiotics (100 IU penicillin, 100 ?g streptomycin and 0.25 ?g amphotericin B). After centrifugation at 3000 x g, 4?C for 15 minutes, supernatants were collected and stored at -70?C until tested. Three mice per group were analyzed per time point, and average values calculated. Virus titrations of mouse tissue homogenates were performed as described before (section 3.2.7). Briefly, four 100?l replicates of 10-fold dilutions (10-1 to 10-8) of homogenates were transferred into flat bottomed 96-well cell culture microplates (Nunc, Denmark) and equal volumes of Vero cell suspension in EMEM containing 2 x 105 cells/ml, 8% FBS and antibiotics were added. The plates were incubated at 37?C in CO2 and observed microscopically for cytopathic effects (CPE) for 10 days post inoculation. Virus titres, calculated by the K?rber method (K?rber, 1931) were expressed as median tissue culture infectious dose (TCID50) per gram of tissue. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 122 P. Jansen van Vuren 6.2.1.5 Monitoring mouse immune responses after immunization Mouse immune responses after immunization were measured by a recNP based indirect ELISA. Immunoplates (Maxisorb, Nunc, Denmark) were coated with RVFV recNP antigen at a dilution of 1:2000 in Carbonate-Bicarbonate buffer (pH 9.6) and incubated overnight at 4?C. After washing three times with a washing buffer consisting of PBS pH7.2 and 0.1% Tween-20, the plates were blocked with 200?l of 10% fat free milk powder (?Elite?, Clover SA, Pty, Ltd.) in PBS at 37?C for 1h and then washed as before. Test sera were diluted 1:400 in diluent buffer consisting of 2% fat free milk powder in PBS, 100?l added to each well and incubated for 1h at 37?C. Samples were tested in duplicate for each isotype-specific HRPO conjugate used. After washing as before, 100 ?l of goat anti-mouse IgG (H+L), goat anti-mouse IgG1 or goat anti-mouse IgG2a HRPO conjugate (Zymed Laboratories, Invitrogen, U.S.A.) at 1:2000 dilution was added to respective plates testing for the same serum specimens in parallel. After 1h incubation at 37?C plates were washed as before and 100 ?l of 2,2?-azinodiethylbenzthiazoline sulfonic acid (ABTS) (KPL Laboratories, Inc., USA) added to each well. After 30 min incubation in the dark the reaction was stopped by the addition of 100 ?l of 1% sodium dodecyl sulphate (SDS) to each well. Optical density (OD) was determined at 405nm and the results expressed as the mean OD value for the biological triplicates tested. 6.2.2 Measuring up- and downregulation of genes using qRT-PCR 6.2.2.1 Selection of genes to be analyzed (Quantitect RT-PCR) A total of 5 genes, involved in the immune response against viral infections, were chosen as target genes for measuring up- and downregulation (Table 6.1). An additional housekeeping gene was included for normalization of data. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 123 P. Jansen van Vuren Table 6.1. Genes analyzed by Quantitect RT-PCR Gene Gene name Function of transcript Accession number Gapdh glyceraldehyde-3- phosphate dehydrogenase Housekeeping gene NM_008084 Nfkb1 nuclear factor of kappa light polypeptide gene enhancer in B-cells 1, p105 Transcription factor involved in immunity NM_008689 Casp3 caspase 3 Involved in cell apoptosis NM_009810 Il10 interleukin 10 [cytokine synthesis inhibitory factor (CSIF)] Anti-inflammatory cytokine NM_010548 Ifng interferon gamma Cytokine involved in innate/adaptive immunity against viral infections, direct antiviral activity, activation of macrophages NM_008337 Ifnb1 Interferon beta 1, fibroblast Cytokine with antiviral activity involved in innate immunity NM_010510 Optimized and validated primer sets were ordered specifically for use with the QIAgen SYBR Green-based real-time RT-PCR kit (Quantitect primer assay, QIAgen, Germany) (https://www.qiagen.com/geneglobe/default.aspx). 6.2.2.2 RNA extraction from mouse tissues (Quantitect RT-PCR) Liver, spleen and brain tissues were transferred from RNAlater directly into 2ml centrifugation tubes containing 800?l buffer RLT (RNeasy Mini kit, QIAgen, Germany) and one 5mm stainless steel bead for tissue disruption and homogenization using the Tissuelyser II (QIAgen, Germany) as recommended by the manufacturer (4 min, 25 Hz). Homogenates were centrifuged for 3 minutes at 13 200 rpm, and the supernatant transferred to a new tube. RNA was extracted from these supernatants using the RNeasy Mini kit (QIAgen, Germany) as suggested by the manufacturer. As suggested, 50% ethanol was used for extraction from liver tissues to increase RNA yield, and 70% ethanol for spleen and brain tissues. The optional on-column DNase digestion was performed using the RNase-free DNase set (QIAgen, Germany) as suggested by the manufacturer, to remove genomic DNA. The RNA was eluted in the supplied RNase-free water, the concentration determined using a NanoDrop ND-1000 spectrophotometer (Thermo Scientific, U.S.A) and stored at -70?C until further testing. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 124 P. Jansen van Vuren 6.2.2.3 Determination of RT-PCR efficiency using different primer sets Determination of RT-PCR efficiencies for the different primer sets was necessary to enable data analysis as described below (6.2.2.5). RNA was extracted as described above (6.2.2.2) from the liver, spleen and brain of a mouse collected before RVFV challenge to use as tissue specific RNA standards. Dilution series were prepared and the following amounts of RNA standard from all organs tested in duplicate using all primer sets: 30 ng; 15 ng, 7.5 ng, 3.75 ng and 1.875 ng. The real time RT-PCR reactions were performed as described by the manufacturer (Quantifast SYBR Green RT-PCR kit, QIAgen, Germany) using a LightCycler 1.5 (Roche, Germany). Shortly, a reaction mix was prepared by mixing 2 x Quantifast SYBR Green RT-PCR Master Mix (HotStarTaq Plus DNA Polymerase, Quantifast SYBR Green RT-PCR buffer, dNTP mix and ROX passive reference dye), 10 x Quantitect Primer sets (Gapdh, Nfkb1, Casp3, Il10, Ifng or Ifnb1), Quantifast RT- Mix (Omniscript RT and Sensiscript RT), template RNA (30 ng; 15 ng, 7.5 ng, 3.75 ng or 1.875 ng) and RNase-free water to a final volume of 20?l per reaction. This mix was transferred to 20?l LightCycler Capillaries (Roche, Germany) and run on the LightCycler 1.5 using the following cycles: 1 x reverse transcription (10 min, 50?C), 1 x hotstart PCR activation (5 min, 95?C) and 40 x cycles of denaturation (10 sec, 95?C) and annealing/extension (30 sec, 60?C), with fluorescence data collection just after the annealing/extension step. The threshold cycle (CT) values were determined using the second derivative maximum method (LightCycler Data Analysis Software version 3.5.28, Roche). The CT values and their corresponding template amount values were then used to determine the PCR reaction efficiencies using the Relative Expression Software Tool (REST, QIAgen, (http://www.qiagen.com/Products/REST2009Software.aspx?r=8042) (Pfaffl et al., 2002). The calculated PCR efficiencies were subsequently used to calculate fold changes in gene expression as described in 6.2.2.5. 6.2.2.4 Quantitect RT-PCR on mouse tissues collected at different time points Quantifast SYBR Green RT-PCR reactions were performed on RNA extracts from mouse tissues collected at different time points (3, 6, 12, 24, 72 and 120 hours p.i.) as described above, except that 10 ng of RNA was used as template for all reactions. Three mice per group were analyzed per time point, and average values calculated. 6.2.2.5 Data analysis (Quantitect RT-PCR) Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 125 P. Jansen van Vuren Threshold values (Ct) for the five different genes analyzed as determined for the immunized mice, adjuvant control and PBS control mice were firstly normalized to the Ct values of the housekeeping gene in the same samples, and then normalized to the normalized Ct values from the non-infected control mice to determine the relative changes in gene expression compared to age related normal mice. This is called the 2-??Ct method. Fold changes in gene expression of specific tissues at specific collection times were calculated using CT values and the REST software. Fold change values equal, or close to 1, indicate no change in gene expression. Values > 1 indicate up-regulation, and values < 1 indicate down-regulation. The software determines fold changes by using the 2-??Ct method (Livak and Schmittgen, 2001, Pfaffl, 2001). The software also uses randomizations and a hypothesis test, P(H1), to determine the statistical significance of fold changes in gene expression (Pfaffl et al., 2002). Genes were only regarded as statistically up-regulated when the following requirements were met: p-value ? 0.05 and fold change ? 2.0. For values < 1 the negative inverted value was determined, and genes were only regarded as statistically down-regulated when these values were ? -2.0 and the p- value ? 0.05. Where fold changes were ? 2.0 or ? -2.0, but p-values not ? 0.05, the results were regarded as indicative of an up- or downregulated trend but not statistically significantly so. 6.2.2.6 RNA extraction (SABiosciences PCR Array) RNA that was extracted from mouse livers collected at 72 hours (Group MS1,2,3 and 4) after infection as described above (6.2.2.2) were cleaned up further using the RT2 qPCR-Grade RNA Isolation Kit (SABiosciences, QIAgen, U.S.A) as recommended by the manufacturer. On-column DNase treatment was performed to remove genomic DNA. RNA was eluted in RNase-free H2O and RNA concentration determined. All RNA extracts were diluted to 150 ng/?l in nuclease-free water. 6.2.2.7 Mouse T- and B-cell activation PCR array (SABiosciences) Complementary DNA (cDNA) was prepared from the RNA extracted in 6.2.2.6 using the RT2 First Strand kit (SABiosciences, QIAgen, U.S.A) as described by the manufacturer. A total of 1.2 ?g of each RNA preparation was mixed with 5 x genomic DNA Elimination buffer and the reaction incubated at 42?C for 5 minutes (total volume 10 ?l). After the incubation the reactions were placed on ice immediately and subsequently an equal volume of RT-cocktail mix added (5 x RT buffer, primers and external control mix, RT-enzyme mix and RNase-free water). These reactions were then incubated at 42?C for 15 minutes and 95?C for 5 minutes. The resultant cDNA of each preparation was then diluted 1:10 with nuclease-free water and stored at -20?C until the assays were run. The diluted cDNA was then mixed with the master mix (2 x SABiosciences RT2 qPCR Master Mix) and nuclease-free water, and aliquoted onto the PCR array plates containing primer pairs (25 ?l Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 126 P. Jansen van Vuren of reaction mix per well). Plates were run on an ABI 7500 cycler (Applied Biosystems, U.S.A). The following cycling program was used: 1 x 95?C for 10 minutes, 40 x 95?C for 15 seconds and 60?C for 1 minute, followed by the default melting curve program. Fluorescence was measured just after the 1 minute / 60?C step. The cycle threshold (CT) values were determined using the cycler software and an automatic baseline adjustment (ABI 7500 Software Version 2.0.1, Applied Biosystems, U.S.A). Three mice per group were analyzed per time point, and average values calculated. 6.2.2.8 Data analysis (SABiosciences PCR Array) Data from the immunized mice, as well as adjuvant control and PBS control mice, were normalized to the data from the non-infected control mice to determine the relative changes in gene expression compared to age related normal mice. For analysis of data from the PCR array plates, the CT values were exported into Microsoft Excel from the ABI software and subsequently copied into the SABiosciences PCR Array Data Analysis Template Excel Utility, which is freely available on the manufacturer?s website (http://sabiosciences.com/pcrarraydataanalysis.php). This template calculates fold changes in gene expression using the 2-??Ct method, as well as the statistical significance (p-value) of results by using a T-test. Fold change values equal, or close to 1, indicate no change in gene expression. Values > 1 indicate up-regulation, and values < 1 indicate down-regulation. Genes were only regarded as up-regulated when the following requirements were met: p-value ? 0.05 and fold change ? 2.0. For values < 1 the negative inverted value was determined, and genes were only regarded as down-regulated when these values were ? -2.0 and the p-value ? 0.05. Where fold changes were ? 2.0 or ? -2.0, but p-values not ? 0.05, the results were regarded as indicative of an up- or downregulated trend but not statistically significantly so. 6.3 Results After recNP/Alhydrogel immunization serum was collected from three mice at each collection point to monitor immune responses. Immunization of mice with recNP combined with Alhydrogel adjuvant yielded an almost identical profile of total IgG, IgG1 and IgG2a responses as shown in Chapter 4 (Figure 4.1 f). Strong total IgG and IgG1 responses were elicited, with a much weaker IgG2a response. High levels of total IgG and IgG1 were still detectable on day 27, a day before RVFV challenge (Figure 6.1). The expression levels of 5 genes involved in the immune response against viral infections, normalized to a housekeeping gene (Gapdh), were analyzed in liver, spleen and brain tissues of recNP/Alhydrogel immunized mice, adjuvant control mice, PBS control mice and uninfected normal Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 127 P. Jansen van Vuren control mice, by qRT-PCR at 3, 6, 12, 24, 72 and 120 hours post infection. Data from the other three groups were normalized against the data from the uninfected normal control group to show the changes in gene expression relative to age-related normal mice. Viral loads were also determined in the corresponding tissues. The results of the gene expression analysis and virus titration are shown in Table 6.3 (liver), Table 6.4 (spleen) and Table 6.5 (brain) for the following collection times: 3, 6, 12, 24, 72 and 120 hours. Statistically significant results (p-value ? 0.05) are indicated by an asterisk. Figure 6.1. Detection of total IgG (???), IgG1 (??) and IgG2A (--?--) specific antibodies against the RVFV recNP in mice after recNP immunization with Alhydrogel. Error bars indicate standard deviation from the mean (3 mice per time point). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 128 P. Jansen van Vuren Table 6.2. Fold changes in gene expression and viral loads in liver specimens from all treated groups at different time points. Gene recNP immunized mice Alhydrogel control mice PBS control mice Mean fold change Standard error P-value Up/down regulated Mean fold change Standard error P-value Up/down regulated Mean fold change Standard error P-value Up/down regulated 3 h o u r s NFKB -1.3 -1.7 to 1.0 0.30 - 1.2 1.0 to 1.5 0.40 - 1.1 -1.1 to 1.3 0.65 - CASP3 1.3 1.0 to 1.8 0.20 - 1.1 -1.3 to 1.3 0.56 - 1.0 -1.4 to1.3 0.85 - IL10 -1.3 -3.3 to 2.3 0.60 - 4.5 2.1 to 11.2 0.00 UP* 2.3 -1.4 to7.5 0.38 UP IFN? -1.85 -2.3 to -1.3 0.07 - -1.1 -2.5 to 1.8 0.85 - 1.3 1.0 to 1.9 0.20 - IFNb1 7.7 5.0 to 10.8 0.03 UP* -2.0 -5.0 to 1.3 0.38 DOWN -2.5 -10.0 to 1.6 0.69 DOWN 6 h o u r s NFKB 1.2 -1.3 to 1.8 0.71 - 1.3 -1.3 to 2.3 0.40 - 1.0 -1.7 to 1.7 0.78 - CASP3 1.9 1.5 to 2.2 0.01 - 1.4 1.1 to 1.7 0.07 - 2.0 1.7 to 2.4 0.02 UP* IL10 -1.3 -2.5 to 1.5 0.58 - -3.3 -10.0 to 1.5 0.48 DOWN -3.3 -5.0 to -1.3 0.03 DOWN* IFN? 17.2 4.9 to 94.2 0.00 UP* 11.7 3.3 to 76.9 0.00 UP* 9.3 2.4 to 46.3 0.00 UP* IFNb1 34.9 6.6 to 263.8 0.00 UP* 4.5 1.2 to 10.9 0.05 UP* 2.0 -1.7 to 5.1 0.31 UP TCID50 0 0 - - 0 0 - - 0 0 - - 1 2 h o u r s NFKB 1.1 -1.3 to 1.5 0.80 - -1.1 -1.7 to 1.1 0.46 - 1.2 -1.3 to 1.7 0.42 - CASP3 1.1 -1.4 to 1.7 0.65 - 1.0 -1.7 to 1.5 0.90 - 1.3 -1.1 to 1.9 0.31 - IL10 -1.1 -1.4 to 1.1 0.73 - 1.6 -1.4 to 3.1 0.33 - 1.0 -2.0 to 1.6 0.94 - IFN? 11.0 6.1 to 25.0 0.06 UP 5.6 3.4 to 10.5 0.03 UP* 5.8 3.2 to 11.1 0.08 UP IFNb1 5.9 2.3 to 12.8 0.07 UP 1.0 -2.0 to 1.8 0.95 - 2.0 1.5 to 2.8 0.04 UP* TCID50 0 0 - - 0 0 - - 0 0 - - 2 4 h o u r s NFKB 1.2 -1.3 to 1.5 0.55 - 1.1 -1.3 to 1.5 0.67 - 1.4 -1.1 to 1.8 0.34 - CASP3 1.0 -1.7 to 1.8 0.97 - -1.3 -2.0 to 1.7 0.69 - -1.3 -2.0 to 1.5 0.55 - IL10 -1.3 -3.3 to 1.9 0.64 - 1.0 -2.5 to 2.2 0.93 - 3.8 1.7 to 8.6 0.09 UP IFN? 5.3 4.1 to 7.0 0.04 UP* -2.0 -5.0 to 1.0 0.10 DOWN -1.7 -2.5 to -1.3 0.02 - IFNb1 1.0 -2.5 to 2.2 0.94 - 1.2 -2.0 to 2.6 0.72 - -3.3 -10 to -1.4 0.09 DOWN TCID50 0 0 - - 102.1 101.8 - - 0 0 - - 7 2 h o u r s NFKB 1.3 1.1 to 1.6 0.10 - 11.7 5.3 to 20.2 0.00 UP* 8.5 4.7 to 14.9 0.0 UP* CASP3 -1.1 -1.3 to 1.1 0.44 - 2.5 1.9 to 3.0 0.07 UP 1.3 -1.1 to 1.7 0.37 - IL10 24.4 10.6 to 75.0 0.00 UP* 808.3 275.8 to 2521.9 0.03 UP* 243.2 107.9 to 844.4 0.0 UP* IFN? 8.4 3.4 to 30.9 0.03 UP* 5.8 2.4 to 15.0 0.03 UP* 4.4 1.5 to 10.0 0.03 UP* IFNb1 -2.0 -3.3 to -1.1 0.16 DOWN 2171.7 1044.3 to 4208 0.03 UP* 2524.1 1016 to 6214 0.02 UP* TCID50 100.92 101.59 - - 104.58 100.38 - - 105.25 100.43 - - 1 2 0 h o u r s NFKB -1.3 -2.0 to 1.1 0.33 - 1.3 1.0 to 2.3 0.54 - 1.6 1.3 to 2.0 0.02 - CASP3 1.3 1.0 to 1.4 0.23 - 1.1 -1.1 to 1.4 0.56 - 1.5 1.2 to 1.9 0.09 - IL10 24.3 7.4 to 54.7 0.03 UP* 54.7 14.1 to 321.6 0.00 UP* 99.3 29.9 to 244.4 0.03 UP* IFN? 23.4 10.2 to 65.3 0.03 UP* 224.0 92.6 to 531.0 0.00 UP* 237.0 100.3 to 655.4 0.03 UP* IFNb1 -2.0 -3.3 to 1.0 0.24 DOWN 8.7 5.2 to 16.3 0.05 UP* 9.0 5.1 to 20.3 0.00 UP* TCID50 0 0 - - 102.5 102.17 - - 103.75 100.25 - - Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 129 P. Jansen van Vuren Table 6.3. Fold changes in gene expression and viral loads in spleen specimens from all treated groups at different time points. Gene recNP immunized mice Alhydrogel control mice PBS control mice Mean fold change Standard error P-value Up/down regulated Mean fold change Standard error P-value Up/down regulated Mean fold change Standard error P-value Up/down regulated 3 h o u r s NFKB -1.7 -2.5 to 1.0 0.28 - 1.2 1.0 to 1.4 0.38 - -1.1 -1.4 to 1.1 0.60 - CASP3 -1.4 -1.7 to 1.0 0.26 - 1.1 -1.3 to 1.6 0.86 - 1.0 -1.4 to 1.2 1.00 - IL10 1.1 -1.3 to 1.7 0.71 - 1.1 -1.3 to 1.5 0.71 - -1.1 -1.7 to 1.4 0.55 - IFN? -2.5 -3.3 to 0.9 0.05 DOWN* -1.4 -2.0 to 1.0 0.21 - -1.7 -2.0 to -1.3 0.05 - IFNb1 12.0 3.6 to 56.9 0.03 UP* 1.5 -1.7 to 4.1 0.44 - 1.4 -2.0 to 5.5 0.62 - 6 h o u r s NFKB -1.4 -2.0 to -1.1 0.12 - -1.1 -1.4 to 1.3 0.35 - -1.4 -1.7 to -1.1 0.08 - CASP3 -1.4 -1.7 to -1.1 0.21 - -1.1 -1.4 to 1.1 0.35 - 1.1 -1.1 to 1.2 0.59 - IL10 1.0 -1.7 to 1.4 0.79 - -1.3 -1.4 to 1.1 0.37 - -1.1 -1.3 to 1.2 0.39 - IFN? -1.7 -2.5 to -1.1 0.12 - -1.3 -1.7 to 1.0 0.09 - -1.7 -2.5 to 1.1 0.22 - IFNb1 5.7 2.5 to 15.9 0.06 UP 10.6 2.1 to 44.6 0.06 UP 9.7 5.0 to 29.3 0.02 UP* TCID50 0 0 - - 0 0 - - 0 0 - - 1 2 h o u r s NFKB 1.0 -1.7 to 1.4 0.95 - 1.2 -1.1 to 1.8 0.54 - 1.1 -1.3 to 1.5 0.69 - CASP3 1.2 1.0 to 1.3 0.10 - 1.1 1.0 to 1.3 0.45 - 1.1 -1.3 to 1.5 0.96 - IL10 1.2 -1.3 to 1.8 0.50 - -1.1 -1.4 to 1.5 0.80 - 1.1 -1.1 to 1.3 0.84 - IFN? -1.25 -1.4 to 1.0 0.00 - -1.3 -1.4 to 1.0 0.00 - 1.1 1.0 to 1.3 0.00 - IFNb1 1.4 -1.1 to 2.5 0.52 - 2.1 1.0 to 4.0 0.26 UP 1.2 -2.0 to 3.1 0.56 - TCID50 0 0 - - 0 0 - - 0 0 - - 2 4 h o u r s NFKB -2.0 -2.5 to -1.7 0.01 DOWN* -2.0 -2.0 to -1.7 0.02 DOWN* -1.1 -1.4 to 1.0 0.00 - CASP3 1.3 1.1 to 1.5 0.17 - 1.1 -1.1 to 1.3 0.34 - 1.2 -1.1 to 1.6 0.54 - IL10 -1.1 -1.4 to 1.3 0.64 - 1.1 -1.1 to 1.6 0.79 - 1.7 1.1 to 2.3 0.11 - IFN? -2.0 -3.3 to -1.7 0.07 DOWN 1.0 -1.4 to 1.2 0.92 - -1.1 -1.7 to 1.4 0.47 - IFNb1 -10.0 -10.0 to1.2 0.19 DOWN -3.3 -10 to -1.7 0.00 DOWN* -2.5 -10.0 to 1.5 0.31 DOWN TCID50 0 0 - - 102.33 102.08 - - 0 0 - - 7 2 h o u r s NFKB -2.0 -10 to -1.4 0.00 DOWN* -1.7 -3.3 to 1.6 0.37 - -2.5 -5.0 to 1.0 0.29 DOWN CASP3 1.3 1.1 to 1.6 0.18 - -3.3 -5.0 to -3.3 0.00 DOWN* -2.5 -2.5 to -2.0 0.04 DOWN* IL10 -1.4 -3.3 to 1.8 0.50 - 46.0 25.2 to 119.2 0.00 UP* 55.0 30.2 to 139.9 0.00 UP* IFN? 20.0 1.2 to 214.4 0.10 UP -1.4 -3.3 to 1.8 0.68 - -1.3 -5.0 to 3.4 0.68 - IFNb1 16.1 10.7 to 26.6 0.07 UP 542.3 372.2 to 808.5 0.00 UP* 510.3 336.7 to 714.3 0.00 UP* TCID50 103.58 100.52 - - 105.58 100.52 - - 105.50 100.66 - - 1 2 0 h o u r s NFKB -10.0 -10 to -3.3 0.00 DOWN* -5.0 -10 to -1.7 0.10 DOWN -5.0 -10 to -1.3 0.13 DOWN CASP3 -5.0 -10 to -2.0 0.00 DOWN* -2.0 -5.0 to- 1.1 0.10 DOWN -3.3 -10 to 1.2 0.20 DOWN IL10 -5.0 -10 to -3.3 0.03 DOWN* -1.4 -3.3 to 1.4 0.55 - 5.2 1.4 to 17.3 0.20 UP IFN? -1.3 -5.0 to 2.6 0.81 - 2.3 1.2 to 5.6 0.00 UP* 1.5 -2.0 to 3.0 0.57 - IFNb1 4.4 1.5 to 25.4 0.03 UP* 2.1 -1.4 to 11.8 0.81 UP 2.3 -1.4 to 15.4 0.66 UP TCID50 0 0 - - 105.25 100.25 - - 105.17 100.29 - - Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 130 P. Jansen van Vuren Table 6.4. Fold changes in gene expression and viral loads in brain specimens from all treated groups at different time points. Gene recNP immunized mice Alhydrogel control mice PBS control mice Mean fold change Standard error P-value Up/down regulated Mean fold change Standard error P-value Up/down regulated Mean fold change Standard error P-value Up/down regulated 2 4 h o u r s NFKB 1.1 -1.1 to 1.4 0.69 - -1.1 -1.4 to 1.2 0.82 - 1.1 1.0 to 1.4 0.12 - CASP3 1.3 1.0 to 2.0 0.24 - 1.1 -1.1 to 1.3 0.40 - 1.0 -1.3 to 1.3 0.97 - IL10 2.2 1.5 to 3.5 0.13 UP 2.7 1.7 to 4.4 0.07 UP 1.3 -2.5 to 3.4 0.71 - IFN? 2.2 1.8 to 2.6 0.05 UP* 1.4 1.1 to 1.7 0.14 - 1.8 1.5 to 2.1 0.03 - IFNb1 6.0 3.4 to 11.1 0.08 UP 4.5 3.2 to 6.7 0.05 UP* 1.6 -1.4 to 3.6 0.45 - TCID50 0 0 - - 101.08 101.88 - - 0 0 - - 7 2 h o u r s NFKB -1.1 -1.7 to 1.1 0.61 - -2.5 -2.5 to -2.0 0.00 DOWN* -2.0 -2.5 to -1.7 0.04 DOWN* CASP3 1.1 1.0 to 1.4 0.50 - -1.3 -1.4 to -1.1 0.10 - -1.7 -2.0 to -1.4 0.08 - IL10 1.3 -1.3 to 2.4 0.79 - 2.7 1.8 to 5.5 0.08 UP 7.0 4.1 to 13.5 0.09 UP IFN? 2.2 1.5 to 2.7 0.03 UP* 1.3 1.0 to 1.8 0.25 - 1.1 -1.3 to 2.0 0.88 - IFNb1 2.4 1.0 to 4.2 0.20 UP 4.6 2.1 to 9.3 0.06 UP 4.9 2.0 to 8.2 0.03 UP* TCID50 100.92 101.59 - - 103.67 100.14 - - 103.92 100.52 - - 1 2 0 h o u r s NFKB -1.7 -2.5 to -1.3 0.12 - 1.0 -1.4 to 1.5 0.82 - -1.1 -1.7 to 1.4 0.76 - CASP3 -1.7 -2.5 to -1.3 0.04 - -1.7 -2.0 to -1.3 0.06 - -2.5 -3.3 to -1.7 0.02 DOWN* IL10 1.5 -1.3 to 3.2 0.38 - 1.4 -2.0 to 3.1 0.64 - 1.8 -1.3 to 3.4 0.45 - IFN? 4.0 2.2 to 7.1 0.07 UP 4.1 2.9 to 6.1 0.09 UP 3.9 2.6 to 5.3 0.07 UP IFNb1 -3.3 -10 to -1.4 0.13 DOWN -1.3 -2.0 to 1.2 0.49 - -2.5 -10.0 to 1.4 0.55 DOWN TCID50 0 0 - - 103.83 101.04 - - 104.08 100.14 - - Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 131 P. Jansen van Vuren The virus was detected at the earliest in liver, spleen and brain tissues of Alhydrogel control mice at 24 hours post infection (table 6.3, 6.4 and 6.5)(figure 6.2 a ? b). The average titre of virus detected was 102.1 TCID50/g tissue. Replication of challenge virus was consequently detected in livers of all infected groups at 72 hours (3 days) post infection. Average viral load was drastically lower (?4000 to 20 000 fold lower) in recNP immunized mice (100.92 TCID50/g tissue) compared to Alhydrogel control (104.58 TCID50/g tissue) and PBS control mice (105.25 TCID50/g tissue). At 120 hours (5 days) post infection viremia was undetectable in recNP immunized mice, whereas infectious virus could still be detected in Alhydrogel control (102.5 TCID50/g tissue) and PBS control mice (103.75 TCID50/g tissue) at lower titres compared to 72 hours. Figure 6.2 (a-b). Mean viral loads in livers and spleens of RVFV infected mice. Groups, consisting of 3 mice per group per time point, are indicated as: RecNP immunized (???), adjuvant control (?o?) and PBS control mice (???). Livers are indicated in panel a, and spleens in panel b. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 132 P. Jansen van Vuren At 24 hours post infection virus to an average titre of 102.33 TCID50/g tissue was detected in spleens of Alhydrogel control mice. Replication of challenge virus was consequently detected in spleen tissues of all infected groups at 72 hours (3 days) post infection. Average viral load was lower (?100 fold) but still high in recNP immunized mice (103.58 TCID50/g tissue) compared to Alhydrogel control (105.58 TCID50/g tissue) and PBS control mice (105.5 TCID50/g tissue). At 120 hours (5 days) post infection viremia was undetectable in recNP immunized mice, whereas infectious virus could still be detected in Alhydrogel control (105.25 TCID50/g tissue) and PBS control mice (105.17 TCID50/g tissue) at similar titres compared to 72 hours. At 24 hours post infection virus to an average titre of 101.08 TCID50/g tissue was detected in brains of Alhydrogel control mice. Replication of challenge virus was consequently detected in brain tissues of all infected groups at 72 hours (3 days) post infection. Average viral load was drastically lower (? 500 to 1000 fold) in recNP immunized mice (100.92 TCID50/g tissue) compared to Alhydrogel control (103.67 TCID50/g tissue) and PBS control mice (103.92 TCID50/g tissue). At 120 hours (5 days) post infection viremia was undetectable in recNP immunized mice, whereas infectious virus could still be detected in Alhydrogel control (103.83 TCID50/g tissue) and PBS control mice (104.08 TCID50/g tissue) at similar titres compared to 72 hours. Survival rates of mice in this specific study could not be compared due to the regular euthanasia of mice for collection of organs. It was noted, however, that remaining mice from the recNP immunized group remained healthy throughout all collection time points. On the other hand mice from the PBS control group displayed illness from day 2 post infection, and 4 mice from the same group were found dead on day 3 post infection, and another on day 4 (these mice were excluded from further gene expression experiments). Three mice from the Alhydrogel control mouse group were also found dead on day 3 post infection. Remaining mice from these two groups continued displaying sings of illness due to RVFV infection until the end of the experiment, but not sick enough to warrant unscheduled euthanasia. At 3 hours post infection (p.i.) the expression of the IFN? gene was upregulated with statistical significance (7.7 fold, p = 0.03) in liver tissue of recNP immunized mice, whereas expression was decreased in adjuvant control (-2.0 fold, p = 0.38) and PBS control (-2.3 fold, p = 0.69) groups, but not with statistical significance (figure 6.3). Expression of the gene remained upregulated in immunized mice until 12 hours p.i., after which it waned and subsequently decreased. In adjuvant control mice IFN? was briefly upregulated at 6 hours p.i. (4.5 fold, p = 0.048), but levelled out after that until 72 hours p.i. when it was significantly upregulated (2171.7 fold, p = 0.03). In PBS control mice on the other hand expression was increased at 6 hours (2.0 fold, p = 0.313) and 12 hours p.i. (2 fold, p = 0.036), with a sudden significant decrease in expression at 24 hours p.i. (-3.6 fold, p = 0.09), and a Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 133 P. Jansen van Vuren sudden upregulation at 72 hours post infection (2524.1 fold, p = 0.02). At 120 hours p.i. IFN? expression was still upregulated in control mice (adjuvant control 8.7 fold, p = 0.032; PBS control (9 fold, p < 0.01). The expression of IL-10 was upregulated significantly (4.5 fold, p < 0.01) in liver tissue of adjuvant control mice, and higher than normal in PBS control mice (2.3 fold, p = 0.38), but unaffected in recNP immunized mice (-1.2 fold, p = 0.6) at 3 hours p.i (figure 6.3). Just three hours later the same gene was decreased in control mice (adjuvant -3.6 fold, p = 0.48; PBS -3.1 fold, p = 0.03), returning to normal at 12 hours p.i. At 72 and 120 hours p.i., however, IL-10 expression was upregulated in the adjuvant (72 hours, 808.3 fold, p = 0.033; 120 hours, 55.0 fold, p < 0.01) and PBS control groups (72 hours, 243.3 fold, p < 0.01; 120 hours, 99.3 fold, p = 0.034), and upregulated in immunized mice (72 hours, 24.4 fold, p < 0.01; 120 hours, 24.3 fold, p = 0.033). The expression of IFN? was unaffected in liver tissue of all mice at 3 hours, but upregulated in all groups at 6 hours (recNP immunized, 17.2 fold, p < 0.01; adjuvant control, 11.7 fold, p < 0.01; PBS control, 9.3 fold, p < 0.01) and 12 hours post infection (recNP immunized, 11.0 fold, p = 0.06; adjuvant control, 5.6 fold, p = 0.03; PBS control, 5.8 fold, p = 0.08) (figure 6.3). Expression remained upregulated with relative stability in recNP immunized mice until 120 hours p.i., with a slight increase at 120 hours (23.4 fold, p = 0.03). In adjuvant and PBS control mice, however, there was a decrease in expression at 24 hours p.i. (adjuvant control, -2.2 fold, p = 0.1; PBS control, -1.7 fold, p = 0.02), followed by upregulation at 72 hours, and 120 hours (adjuvant control, 224.0 fold, p < 0.01; PBS control, 237.0 fold, p = 0.029). The expression of the transcription factor NF-kB was stable in liver tissue of all mice at all time points up to 72 hours. At 72 hours, however, there was a significant upregulation of the gene in adjuvant control (11.7 fold, p < 0.01) and PBS control mice (8.5 fold, p < 0.01) while expression remained unaffected in immunized mice throughout all collection points. At 120 hours expression returned to constitutive levels in control mice too. The expression of the gene encoding Caspase-3 was relatively normal in liver tissue at all time points in all groups. There was only a brief upregulation in PBS control mice at 6 hours (2.0 fold, p = 0.02), and in adjuvant control mice at 72 hours (2.5 fold, p = 0.07). Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 134 P. Jansen van Vuren Figure 6.3 (a-f). Fold changes in expression of IL10, IFN? and IFN? genes in tissues of mice after RVFV infection. RecNP immunized mice (n = 3 per time point) are indicated by solid black bars, adjuvant control mice (n = 3 per time point) by grey bars and PBS control mice (n = 3 per time point) by white bars. The horizontal dotted lines indicate the cut-off values for upregulation (+2) or downregulation (22). The asterisk (*) indicates where the P-value is smaller than or equal to 0.05 (statistically significant results). Standard error values are indicated by the error bars. Note the differences in the Y-axis scales. The following time points are indicated: 3, 6, 12, 24, 72 and 120 hours. The expression of IFN? was also significantly upregulated in spleen tissue of recNP immunized mice at 3 hours p.i. (12.0 fold, p = 0.03), while expression was normal in adjuvant control (1.5 fold, p = 0.44) and PBS control mice (1.4 fold, p = 0.62). By 6 hours p.i. expression of IFN? was upregulated in control mice as well (recNP immunized, 5.7 fold, p = 0.06; adjuvant control, 10.6 fold, p = 0.06; PBS control, 9.7 fold, p = 0.02). At 24 hours p.i. there was a sudden decrease in IFN? expression in all groups (recNP immunized, -10.0 fold, p = 0.19; adjuvant control, -3.3 fold, p < 0.01; PBS control, -2.5 fold, p = 0.31), after which the gene was upregulated in recNP immunized mice (16.1 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 135 P. Jansen van Vuren fold, p = 0.07) and upregulated in adjuvant control (542.3 fold, p < 0.01) and PBS control mice (510.3 fold, p < 0.01). Expression was still upregulated at 120 hours p.i., but to much lower levels compared to 72 hours (recNP immunized, 4.4 fold, p = 0.03; adjuvant control, 2.1 fold, p = 0.81; PBS control, 2.3 fold, p = 0.66). The expression of the IL-10 gene remained normal in spleen tissue of all groups until 72 hours p.i. when it was upregulated significantly in adjuvant control (46.0 fold, p < 0.01) and PBS control mice (55.0 fold, p < 0.01). Interestingly its expression was downregulated significantly in recNP immunized mice at 120 hours p.i. (-5.0 fold, p = 0.03). There was almost no change in the expression of IFN? in spleen tissue of control mice throughout the experiment, until 120 hours p.i. when it was upregulated in adjuvant control mice (2.3 fold, p < 0.01). In immunized mice, however, its expression was inconsistent, being downregulated shortly after infection (3 hours, -2.5 fold, p = 0.05) and again higher than normal at 72 hours p.i. (20.0 fold, p = 0.1). The expression of NF-kB in spleen tissue was not different between immunized and control groups, being downregulated at 24 hours (recNP immunized, -2.0 fold, p = 0.01, adjuvant control, -2.0 fold, p = 0.02), 72 hours (recNP immunized -2.0 fold, p < 0.01; PBS control, -2.5 fold, p = 0.29) and 120 hours p.i. (recNP immunized, -10.0 fold, p < 0.01; adjuvant control, -5.0 fold, p = 0.1; PBS control, -5.0 fold, p = 0.13). Expression of Caspase-3 remained normal in spleen tissue of all mice until 72 hours p.i. when it was downregulated significantly in adjuvant control (-3.3 fold, p < 0.01) and PBS control mice (-2.5 fold, p = 0.04), but unaffected in recNP immunized mice. At 120 hours p.i. expression was significantly downregulated in recNP immunized mice (-5.0 fold, p < 0.01) and decreased in adjuvant control (-2.0 fold, p = 0.1) and PBS control mice (-3.3 fold, p = 0.2). The expression of IFN? was upregulated in brain tissue of recNP immunized mice (6.0 fold, p = 0.08) and adjuvant control mice (4.5 fold, p = 0.05) at 24 hours p.i., but normal in PBS control mice. At 72 hours p.i. its expression was upregulated significantly in PBS control mice (4.9 fold, p = 0.03) and increased in recNP immunized (2.4 fold, p = 0.2) and adjuvant control mice (4.6 fold, p = 0.06). The expression of IFN? was upregulated in brain tissue of recNP immunized mice at 24 hours (2.2 fold, p = 0.05), 72 hours (2.2 fold, p = 0.03) and 120 hours (4.0 fold, p = 0.07), but only increased at 120 hours p.i. in adjuvant control (4.1 fold, p = 0.49) and PBS control mice (3.9 fold, p = 0.07). The expression of Caspase-3 was only significantly downregulated in brain tissue of PBS control mice at 120 hours p.i. (-2.5 fold, p = 0.02) and normal at all other time points and other groups. The expression of NF-kB was significantly downregulated in brain tissue of adjuvant control (-2.5 fold, p < 0.01) and PBS control mice (-2.0 fold, p = 0.04) at 72 hours p.i. but normal at all other time points and in immunized mice. The relative expression levels of 84 genes involved in various facets of the immune response against viral infections, normalized to four housekeeping genes, were analyzed in liver of Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 136 P. Jansen van Vuren recNP/Alhydrogel immunized mice, adjuvant control mice, PBS control mice and uninfected normal control mice, by qRT-PCR at 72 hours post infection. Data from the other three groups were normalized against the data from the uninfected normal control group to show the changes in gene expression relative to age-related normal mice. The results of the gene expression analysis of the three different experimental groups are shown in Figures 6.2, 6.3 and 6.4. Table 6.5. Genes analyzed by SABiosciences PCR array (T- and B-cell activation). Fold change in expression of 84 genes involved in activation of B- and T-cell immunity and other immune functions in immunized mice versus control mice after RVFV challenge at 72 hours in liver, relative to expression in an age-related control group of mice. * = statistically significant result. Gene Gene name Transcript accession number recNP Immunized Fold change and P-value Adjuvant control Fold change and P-value PBS control Fold change and P-value Ap3b1 Adaptor-related protein complex 3, beta 1 subunit NM_009680 1.65 p = 0.21 -1.09 p = 0.83 1.13 p = 0.58 Bad BCL2-associated agonist of cell death NM_007522 1.06 p = 0.62 -2.19* p = 0.0004 -1.64* p = 0.001 Cxcr5 Chemochine (C-X-C motif) receptor 5 NM_007551 1.56 p = 0.88 5.53 p = 0.18 5.23 p = 0.23 Cblb E3 Ubiquitin Ligase Casitas B-lineage lymphoma b NM_001033238 1.07 p = 0.68 7.16* p = 0.006 8.65* p = 0.001 Ccnd3 Cyclin D3 NM_007632 1.02 p = 0.86 1.12 p = 0.29 1.27* p = 0.02 Cd1d1 CD1d1 antigen NM_007639 1.27 p = 0.41 -6.27* p = 0.005 -6.88* p = 0.003 Cd2 CD2 antigen NM_013486 1.02 p = 0.90 2.56* p = 0.02 2.66* p = 0.04 Cd28 CD28 antigen NM_007642 6.48 p = 0.13 22.33* p = 0.002 21.41* p = 0.005 Cd3d CD3 antigen, delta polypeptide NM_013487 1.15 p = 0.23 2.39 p = 0.07 1.86 p = 0.06 Cd3e CD3 antigen, epsilon polypeptide NM_007648 1.58 p = 0.30 3.44 p = 0.06 3.28* p = 0.02 Cd3g CD3 antigen, gamma polypeptide NM_009850 1.34 p = 0.20 1.14 p = 0.57 -1.47 p = 0.12 Cd4 CD4 antigen NM_013488 1.06 p = 0.74 -1.89 p = 0.45 -1.01 p = 0.92 Cd40 CD40 antigen NM_011611 2.91 p = 0.06 29.40* p = 0.002 25.87* p = 0.003 Cd40lg CD40 ligand NM_011616 1.41 p = 0.48 1.42 p = 0.44 1.83 p = 0.11 Cd74 CD74 antigen (invariant polypeptide of major histocompatibility complex, class II antigen-associated) NM_010545 2.03 p = 0.12 1.39* p = 0.012 1.36 p = 0.13 Cd81 CD81 antigen NM_133655 1.04 p = 0.68 -2.17* p = 0.002 -1.77* p = 0.002 Cd8a CD8 antigen, alpha chain NM_001081110 2.33 p = 0.40 4.88 p = 0.13 2.81 p = 0.23 Cd8b1 CD8 antigen, beta chain 1 NM_009858 -1.05 p = 0.99 4.08 p = 0.14 2.39 p = 0.40 Cd93 CD93 antigen NM_010740 1.14 p = 0.85 83.91 p = 0.06 144.67* p = 0.001 Cdkn1a Cyclin-dependent kinase inhibitor 1A (P21) NM_007669 9.02* p = 0.02 127.19* p = 0.0000 145.01* p = 0.0006 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 137 P. Jansen van Vuren Clcf1 Cardiotrophin-like cytokine factor 1 NM_019952 1.27 p = 0.89 33.92* p = 0.01 52.10* p = 0.002 Cr2 Complement receptor 2 NM_007758 -2.0 p = 0.15 1.07 p = 0.74 1.61 p = 0.10 Csf2 Colony stimulating factor 2 (granulocyte-macrophage) NM_009969 -1.56 p = 0.66 2.37 p = 0.45 2.09 p = 0.58 Cxcl12 Chemokine (C-X-C motif) ligand 12 NM_021704 1.24* p = 0.03 -2.21* p = 0.02 -1.64* p = 0.008 Cxcr4 Chemokine (C-X-C motif) receptor 4 NM_009911 1.66* p = 0.006 22.12* p = 0.01 22.84* p = 0.002 Dock2 Dedicator of cyto-kinesis 2 NM_033374 1.11 p = 0.43 4.85* p = 0.04 3.47* p = 0.003 Egr1 Early growth response 1 NM_007913 -5.00* p = 0.01 3.88* p = 0.005 4.54* p = 0.004 Flt3 FMS-like tyrosine kinase 3 NM_010229 -1.06 p = 0.87 12.62* p = 0.004 9.19* p = 0.03 Gadd45g Growth arrest and DNA-damage- inducible 45 gamma NM_011817 7.26 p = 0.26 5.35* p = 0.002 4.37* p = 0.008 Glmn Glomulin, FKBP associated protein NM_133248 1.08 p = 0.59 -1.03 p = 0.82 1.05 p = 0.73 H2-Aa Histocompatibility 2, class II antigen A, alpha NM_010378 1.84 p = 0.11 2.01* p = 0.014 1.56 p = 0.16 H60a Histocompatibility 60a NM_010400 -1.51 p = 0.70 16.50* p = 0.046 13.99* p = 0.036 Hdac5 Histone deacetylase 5 NM_010412 1.05 p = 0.65 -1.46 p = 0.04 -2.18* p = 0.02 Hdac7 Histone deacetylase 7 NM_019572 -1.07 p = 0.68 4.83* p = 0.012 5.58* p = 0.0007 Hells Helicase, lymphoid specific NM_008234 1.34 p = 0.46 2.15 p = 0.08 2.64* p = 0.007 Hsp90aa1 Heat shock protein 90, alpha (cytosolic), class A member 1 NM_010480 1.08 p = 0.67 1.33 p = 0.13 1.48* p = 0.012 Icosl Icos ligand NM_015790 2.33 p = 0.30 3.48 p = 0.25 3.29 p = 0.07 Ifng Interferon gamma NM_008337 4.05 p = 0.32 4.90 p = 0.09 3.10 p = 0.12 Igbp1 Immunoglobulin (CD79A) binding protein 1 NM_008784 -1.23 p = 0.15 -2.39* p = 0.002 -2.37* p = 0.002 Igbp1b Immunoglobulin (CD79A) binding protein 1b NM_015777 1.39 p = 0.37 7.26* p = 0.04 4.53 p = 0.10 Il10 Interleukin 10 NM_010548 3.99 p = 0.21 17.64* p = 0.04 10.65* p = 0.02 Il11 Interleukin 11 NM_008350 1.93 p = 0.20 275.80 p = 0.13 772.47* p = 0.007 Il12b Interleukin 12B NM_008352 2.59 p = 0.25 11.75* p = 0.02 20.35* p = 0.005 Il15 Interleukin 15 NM_008357 1.12 p = 0.48 -1.64* p = 0.026 -1.54* p = 0.04 Il18 Interleukin 18 NM_008360 1.21 p = 0.12 -3.52* p = 0.0002 -4.91* p = 0.0001 Il27 Interleukin 27 NM_145636 3.25 p = 0.63 5.73 p = 0.32 11.42 p = 0.10 Il2ra Interleukin 2 receptor, alpha chain NM_008367 3.18 p = 0.09 5.66* p = 0.04 4.41 p = 0.06 Il4 Interleukin 4 NM_021283 2.04 p = 0.35 21.42 p = 0.24 8.73* p = 0.03 Il7 Interleukin 7 NM_008371 1.64* p = 0.05 -1.70 p = 0.06 -2.38* p = 0.009 Impdh1 Inosine 5'-phosphate dehydrogenase 1 NM_011829 2.13* p = 0.004 3.26 p = 0.25 4.92* p = 0.02 Impdh2 Inosine 5'-phosphate dehydrogenase 2 NM_011830 1.18 p = 0.15 1.00 p = 0.99 1.10 p = 0.48 Inha Inhibin alpha NM_010564 -1.16 p = 0.74 4.49 p = 0.06 3.12 p = 0.17 Irf4 Interferon regulatory factor 4 NM_013674 1.51 p = 0.96 2.59 p = 0.39 1.95 p = 0.79 Jag2 Jagged 2 NM_010588 1.67 p = 0.18 6.79* p = 0.02 7.46* p = 0.02 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 138 P. Jansen van Vuren Ms4a1 Membrane-spanning 4-domains, subfamily A, member 1 NM_007641 1.39 p = 0.37 7.26* p = 0.04 4.53 p = 0.10 Nkx2-3 NK2 transcription factor related, locus 3 (Drosophila) NM_008699 1.10 p = 0.72 1.56 p = 0.26 1.16 p = 0.59 Nos2 Nitric oxide synthase 2, inducible NM_010927 1.05 p = 0.97 25.71 p = 0.06 32.07* p = 0.001 Pawr PRKC, apoptosis, WT1, regulator NM_054056 1.06 p = 0.57 2.68* p = 0.0003 3.05* p = 0.0006 Pdcd1lg2 Programmed cell death 1 ligand 2 NM_021396 1.90 p = 0.1 12.42* p = 0.0004 9.47 p = 0.08 Pik3cd Phosphatidylinositol 3-kinase catalytic delta polypeptide NM_008840 3.59* p = 0.04 5.77 p = 0.13 7.53* p = 0.002 Pik3r1 Phosphatidylinositol 3-kinase, regulatory subunit, polypeptide 1 (p85 alpha) NM_001024955 1.10 p = 0.63 -3.80* p = 0.007 -3.63* p = 0.005 Prkcd Protein kinase C, delta NM_011103 1.30 p = 0.1 3.73* p = 0.03 4.01* p = 0.002 Prkcq Protein kinase C, theta NM_008859 1.92 p = 0.18 4.63 p = 0.09 3.89 p = 0.06 Prlr Prolactin receptor NM_011169 1.15 p = 0.42 -62.79* p = 0.0003 -59.16* p = 0.0003 Ptprc Protein tyrosine phosphatase, receptor type, C NM_011210 2.07 p = 0.57 8.14 p = 0.055 3.69 p = 0.13 Rag1 Recombination activating gene 1 NM_009019 -1.05 p = 0.98 4.98* p = 0.05 3.11 p = 0.13 Relb Avian reticuloendotheliosis viral (v-rel) oncogene related B NM_009046 2.50* p = 0.04 12.30* p = 0.002 15.85* p = 0.002 Rgs1 Regulator of G-protein signalling 1 NM_015811 1.67 p = 0.25 41.67* p = 0.006 24.76* p = 0.016 Sftpd Surfactant associated protein D NM_009160 1.30 p = 0.41 20.40* p = 0.03 26.35* p = 0.0006 Sit1 Suppression inducing transmembrane adaptor 1 NM_019436 1.92 p = 0.24 3.64 p = 0.07 2.96 p = 0.25 Sla2 Src-like-adaptor 2 NM_029983 1.08 p = 0.66 2.38 p = 0.16 2.30* p = 0.007 Socs5 Suppressor of cytokine signalling 5 NM_019654 -1.01 p = 0.90 2.55* p = 0.006 3.63* p = 0.0007 Spp1 Secreted phosphoprotein 1 NM_009263 -1.10 p = 0.62 29.12* p = 0.003 26.97* p = 0.001 Tlr1 Toll-like receptor 1 NM_030682 1.82 p = 0.2 8.68 p = 0.07 5.55 p = 0.06 Tlr4 Toll-like receptor 4 NM_021297 3.46 p = 0.06 14.13* p = 0.03 8.28* p = 0.005 Tlr6 Toll-like receptor 6 NM_011604 1.30 p = 0.43 2.11 p = 0.14 1.77 p = 0.12 Tnfrsf13b Tumor necrosis factor receptor superfamily, member 13b NM_021349 -1.35 p = 0.76 6.67* p = 0.0006 4.27 p = 0.15 Tnfrsf13c Tumor necrosis factor receptor superfamily, member 13c NM_028075 1.23 p = 0.68 2.88 p = 0.12 2.46 p = 0.15 Tnfsf13b Tumor necrosis factor (ligand) superfamily, member 13b NM_033622 1.32 p = 0.60 8.74* p = 0.014 5.75* p = 0.005 Tnfsf14 Tumor necrosis factor (ligand) superfamily, member 14 NM_019418 2.14 p = 0.08 24.66* p = 0.04 16.80* p = 0.001 Traf6 Tnf receptor-associated factor 6 NM_009424 1.53 p = 0.06 1.72* p = 0.03 2.34 p = 0.06 Vav1 Vav 1 oncogene NM_011691 2.20 p = 0.13 7.50* p = 0.03 8.00* p = 0.02 Was Wiskott-Aldrich syndrome homolog (human) NM_009515 1.53 p = 0.50 4.33 p = 0.07 4.64* p = 0.004 Wwp1 WW domain containing E3 ubiquitin protein ligase 1 NM_177327 -1.07 p = 0.38 -9.16* p = 0.0001 -11.99* p = 0.0001 Only one gene, Egr1 (early growths response 1 protein) was downregulated with statistical significance in the recNP immunized mouse group relative to the untreated control group (Table 6.2) (upper left quadrant of Figure 6.2). A total of four genes were upregulated with statistical significance Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 139 P. Jansen van Vuren in the same group (Table 6.2) (upper right quadrant of Figure 6.2). A total of nine (9) genes were downregulated with statistical significance in the Alhydrogel control group relative to the untreated control group (Table 6.2) (upper left quadrant of Figure 6.3). A total of 28 genes were upregulated with statistical significance in the Alhydrogel control group relative to the untreated control group (Table 6.2) (upper right quadrant of Figure 6.3). A total of eight (8) genes were downregulated with statistical significance in the PBS control group relative to the untreated control group (Table 6.2) (upper left quadrant of Figure 6.4). A total of 37 genes were upregulated with statistical significance in the PBS control group relative to the untreated control group (Table 6.2) (upper right quadrant of Figure 6.4). Selected genes that were significantly up- or downregulated in any of the experimental groups and indicating a clear difference or interesting similarity between immunized and control mice were grouped according to known effects on specific immune functions and shown in Figure 6.5 (a-f). Note that some genes are involved in multiple immune functions and are thus present in more than one figure. Figure 6.4. Volcano plot displaying average Log2-fold changes in expression of 84 genes in the livers of three recNP immunized mice at 72 hours post infection, relative to an untreated control group of mice. Different genes are indicated by black dots (?). Black dots above the horizontal double line (representing a p-value of 0.05) indicate statistically significant results. The vertical dotted lines indicate the fold change margins, with dots to the left representing downregulated genes, and dots to the right indicating upregulated genes. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 140 P. Jansen van Vuren Figure 6.5. Volcano plot displaying average Log2-fold changes in expression of 84 genes in the livers of three Alhydrogel control mice at 72 hours post infection, relative to an untreated control group of mice. Different genes are indicated by black dots (?). Black dots above the horizontal double line (representing a p-value of 0.05) indicate statistically significant results. The vertical dotted lines indicate the fold change margins, with dots to the left representing downregulated genes, and dots to the right indicating upregulated genes. Figure 6.6. Volcano plot displaying average Log2-fold changes in expression of 84 genes in the livers of three PBS control mice at 72 hours post infection, relative to an untreated control group of mice. Different genes are indicated by black dots (?). Black dots above the horizontal double line (representing a p-value of 0.05) indicate statistically significant results. The vertical dotted lines indicate the fold change margins, with dots to the left representing downregulated genes, and dots to the right indicating upregulated genes. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 141 P. Jansen van Vuren Figure 6.7 (a-f). Changes in expression of gene in the liver of experimental groups at 72 hours after RVFV infection. RecNP immunized mice are indicated by solid black bars, adjuvant control mice by grey bars and PBS control mice by white bars. The horizontal dotted lines indicate the cut- off values for upregulation (+2) or downregulation (-2). The asterisk (*) indicates where the P-value is smaller than or equal to 0.05 (statistically significant results). Standard deviation from the mean fold changes are indicated by the error bars. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 142 P. Jansen van Vuren Figure 6.8 (a-b). Heat maps showing fold changes in liver and spleen at 72 hours after RVFV infection. Expression of genes in RecNP immunized, adjuvant control and PBS control mice are organized according to function. Livers are indicated in panel a, and spleens in panel b. The genes shown in orange are upregulated, those in blue are downregulated and those in black or darker shades of orange and blue have fold-change values between -2 and 2 and/or have p-values.0.05. There was a significant upregulation of several genes that have pro-apoptotic effects in adjuvant and PBS control mice, whereas these genes were normally expressed in recNP immunized Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 143 P. Jansen van Vuren mice (figure 6.7a and 6.8a) (Cho et al., 2006). One gene, encoding the early growth response 1 protein (Egr-1) which is a transcription factor involved in proliferation, differentiation and activation of cell death pathways (Kiebala et al., 2010), that was upregulated in adjuvant (3.9 fold, p = 0.005) and PBS control mice (4.5 fold, p = 0.004) was downregulated in recNP immunized mice (-5.0 fold, p = 0.01). Only one gene encoding the transcription factor RelB, part of the NF-KB family of proteins and responsible for counter-regulating the effects of NF-KB, was upregulated in immunized (2.5 fold, p = 0.04) and control mice (adjuvant control, 12.3 fold, p = 0.002; PBS control, 15.9 fold, p = 0.002) (figure 6.7a) (Yu-Lee, 2002, Kong et al., 2004, Fry and Mackall, 2005, Kittipatarin and Khaled, 2007, He et al., 2008, Jackson et al., 2008, Li et al., 2008, Lomonosova and Chinnadurai, 2008, Ajay et al., 2010). Several genes with anti-apoptotic effects were downregulated in the control mice but normal in recNP immunized mice (figure 6.7b) (Li et al., 2008). Most notably of these were the genes encoding the prolactin receptor (Prlr) (recNP immunized, 1.15 fold, p = 0.42; adjuvant control, -62.8 fold, p = 0.0003; PBS control, -59.2 fold, p = 0.003), an anti-inflammatory protein known to promote proliferation, protect against apoptosis and enhance cell survival, and the WW domain containing E3 ubiquitin protein ligase 1 (Wwp1) (recNP immunized, -1.1 fold, p = 0.38; adjuvant control, -9.2 fold, p = 0.0001; PBS control, -12.0 fold, p = 0.0001), an anti-apoptotic protein playing a role in proliferation (Senaldi et al., 1999, Denhardt et al., 2001, Coleman, 2002, Mazzali et al., 2002, Senaldi et al., 2002, Curnow et al., 2004, Zhou et al., 2004, Dalakas et al., 2005, Leth-Larsen et al., 2005, Cho et al., 2006, Prince et al., 2007, Pritchard et al., 2007, Guo et al., 2008, Lee et al., 2008, Peterss et al., 2009). There was also evidence of severe liver inflammation in adjuvant and PBS control mice, but not in recNP immunized mice (figure 6.7c) (Denhardt et al., 2001, Mazzali et al., 2002). Despite this the gene encoding the anti-inflammatory cytokine interleukin-10 (IL-10) was upregulated (recNP immunized, 4.0 fold, p= 0.21; adjuvant control, 17.6 fold, p = 0.04; PBS control, 10.7 fold, p = 0.02). The expression of osteopontin (gene Spp1), important for tissue damage healing, was upregulated significantly in control mice, compared to normal expression in immunized mice (Table 1) (Choi et al., 2001, Gartel and Radhakrishnan, 2005). The gene expressing the Cyclin-dependent kinase inhibitor P21 (Cdkn1a), a protein with pro- or anti-apoptotic effects and normally upregulated in response to liver injury, was upregulated in immunized (9 fold) and control mice (127 to 145 fold) (Table 1) (Lowenstein and Padalko, 2004). The gene expressing the inducible nitric oxide synthase (Nos2), an effector of the innate immune system targeting viral proteases and inhibiting viral replication, was normal in recNP immunized mice (1.1 fold, p = 0.97), increased in adjuvant control mice (25.7 fold, p = 0.06) and highly upregulated in PBS control mice (32.1 fold, p = 0.001) (Li et al., 1998, Senaldi et al., 1999, Senaldi et al., 2002, Kong et al., 2004, Norsworthy et al., 2004, Bohlson et al., 2005, Ceredig et al., 2006, Kunisaki et al., 2006, Kasler and Verdin, 2007, Qiao et al., 2007, Zhu et al., 2007, Couper Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 144 P. Jansen van Vuren et al., 2008, Qiao et al., 2008, Zhang et al., 2008, Boesteanu and Katsikis, 2009, Duttagupta et al., 2009, Khan, 2009, Miletic et al., 2009, Peterss et al., 2009). Both arms of the adaptive immune response, humoral (Th2) and cellular (Th1), were activated in control mice, but normal in immunized mice, at 72 hours (Durand et al., 2009). The gene encoding the Phosphatidylinositol 3-kinase catalytic delta polypeptide (Pik3cd), involved in the regulation of B- cells and antibody production, was upregulated in recNP immunized mice (3.6 fold, p = 0.04) and PBS control mice (7.5 fold, p = 0.002), and increased in adjuvant control mice (5.8 fold, p = 0.13) (Renukaradhya et al., 2005, Kunisaki et al., 2006, Kasmar et al., 2009). The genes encoding the Dedicator of cyto-kinesis 2 (Dock2) protein (recNP immunized, 1.1 fold, p = 0.43; adjuvant control, 4.9 fold, p = 0.04; PBS control, 3.5 fold, p = 0.003) and interleukin-12b (recNP immunized, 2.6 fold, p = 0.25; adjuvant control, 11.8 fold, p = 0.02; PBS control, 20.4 fold, p = 0.005), which are involved in the development and induction of NKT cells, were upregulated in control mice. Some important genes were, however, downregulated in control mice (figure 6.7 d-e). The gene encoding the Cd1d1 antigen, which is important for the presentation of antigens to, and activation of NKT cells, was downregulated with statistical significance in control mice (recNP immunized, 1.3 fold, p = 0.42; adjuvant control, - 6.3 fold, p = 0.005; PBS control, -6.9 fold, p = 0.003) (Figure 4 d) (Palmer et al., 2008). The expression of the gene encoding interleukin-7, necessary for B- and T-cell and NK cell survival, was downregulated in PBS control mice (-2.4 fold, p = 0.009) (LeVine et al., 2001). The expression of surfactant protein D, a member of the collectin family, important role player in innate immunity and inhibitor of T lymphocyte proliferation, was upregulated in control mice (adjuvant control, 20.4 fold, p = 0.03; PBS control, 26.4 fold, p = 0.0006) but normal in immunized mice (1.3 fold, p = 0.41) (figure 6.7 d) (Qiao et al., 2007, Qiao et al., 2008, Zhang et al., 2008). The expression of the gene encoding the E3 Ubiquitin Ligase Cbl-b, capable of negatively regulating T-cell activation, was upregulated in control mice (adjuvant control, 7.2 fold, p = 0.006; PBS control, 8.7 fold, p = 0.001) but unaffected in immunized mice (1.1 fold, p = 0.68) (Figure 4 d) (Gracie et al., 2003). The expression of the gene encoding interleukin-18, responsible for biasing immunity towards Th-1 cellular immunity and enhancing T-cell cytotoxicity, was downregulated in control mice (adjuvant control, -3.5 fold, p = 0.0002; PBS control, -4.9 fold, p = 0.0001) but unaffected in immunized mice (1.2 fold, p = 0.12) (figure 6.7 d) (Seki et al., 2002). The expression of the suppressor of cytokine signalling 5 (Socs5), part of a family of proteins that negatively regulate cytokine signalling (Kong et al., 2004), was upregulated in control mice (adjuvant control, 2.6 fold, p = 0.006; PBS control, 3.6 fold, p = 0.0007) but normal in immunized mice (-1.01 fold, p = 0.9). The gene encoding the immunoglobulin binding protein 1 (Igbp1), a component of receptor cell signalling in B- and T-cells (Meroni et al., 2007, Palmer et al., 2008), was downregulated in control mice (adjuvant control, -2.4 fold, p = 0.002; PBS control, -2.4 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 145 P. Jansen van Vuren fold, p = 0.002) but unaffected in immunized mice (-1.23 fold, p = 0.15). Other genes (Cd81 and Pik3r1) involved in activation, signalling and differentiation of B- and T-cells were also downregulated in control mice but normal in immunized mice (Table 1) (Li et al., 1998, Mavoungou et al., 2005, O'Connor et al., 2006). An important role player in innate immunity, NK cells, was also activated in control mice (figure 6.7 f) (O'Connor et al., 2006). The gene encoding the histocompatibility 60 A protein, a ligand for an activating receptor on NK cells (Gracie et al., 2003, Mavoungou et al., 2005), was upregulated in control mice (adjuvant control, 16.5 fold, p = 0.046; PBS control, 14.0 fold, p = 0.036) but unaffected in recNP immunized mice (-1.5 fold, p = 0.7). However, expression of two important genes in NK cell activation and maturation, interleukin-18 (Il-18) and the prolactin receptor (Prlr), were downregulated in control mice but normal in immunized mice (Janssens and Beyaert, 2003). The expression of three genes encoding Toll-like receptors, a component of the innate immune system responsible for recognizing conserved structures, were analyzed with only Tlr4, responsible for recognizing patterns present on viral antigens, being upregulated in control mice (recNP immunized, 3.5 fold, p = 0.06; adjuvant control, 14.1 fold, p = 0.03; PBS control, 8.3 fold, p = 0.005) (Bouloy et al., 2001, Billecocq et al., 2004, Le May et al., 2004). 6.4 Discussion The immune evasion mechanisms known for RVFV are all directed against the innate immune response, more specifically the type I interferon response (Won et al., 2007) and programmed cell death (apoptosis) (do Valle et al., 2010). Despite the proven involvement of the NSs protein of RVFV in inhibition of the type I interferon response, and thus increase in pathogenicity, it was recently shown in vitro that RVFV is not able to completely inhibit the expression of type I interferon (Lorenzo et al., 2008). It was shown that mice displaying an earlier and stronger type I interferon response were less susceptible to RVFV infection than mice with a delayed and partial response. The fact that RVFV does not cause a complete inhibition of the type I interferon response would suggest that the virus must have some additional evasive or regulatory effects, such as other innate immune mechanisms or adaptive immunity, to enable sufficient unhindered replication. To gain some knowledge on the effects of RVFV on these other immune functions it was decided to test the expression of genes involved mainly in the activation of the B- and T-cell immunity, but also other immune functions, in vivo in the liver of mice experimentally infected with RVFV at the time of acute infection. The expression of five genes important in RVFV pathogenicity and involved in innate and adaptive immunity (nuclear factor kappa-beta, caspase-3, interleukin-10 and interferon- Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 146 P. Jansen van Vuren gamma and -beta) was also tested at consecutive time points early and late after RVFV infection in mouse liver, spleen and brain to gain some insight into the dynamics of the expression of these genes during the extent of infection. At the same time the expression of all the genes mentioned above were compared between experimentally infected mice and mice that were vaccinated with a recombinant RVFV nucleocapsid protein (recNP), combined with the adjuvant Alhydrogel, before experimental infection. As shown previously, immunization of mice with recNP combined with Alhydrogel resulted in complete protection against disease and significant reduction in viral replication (Chapter 4). The RVFV NP does not have any neutralizing epitopes and a previous study suggested that a cellular (Th- 1) response to the RVFV NP might be responsible for protection (do Valle et al., 2010). However, results shown in Chapter 4 suggested that the response after recNP/Alhydrogel immunization was biased towards Th-2 humoral immunity. The results presented in this chapter shows that recNP immunized mice were able to launch a stronger and earlier, but more controlled later type I interferon response compared to non-immunized mice, most probably contributing to the protection of immunized mice. More importantly the results show activation of several genes with pro-apoptotic and pro-inflammatory effects, but suppression of anti-apoptotic genes, at the time of acute RVFV infection in the liver of infected mice, possibly contributing to hepatic damage which is the main pathological feature of RVF. Also, the expression of several important genes involved in the activation and function of Natural Killer cells (innate immunity) and B- and T-lymphocytes (adaptive immunity) were suppressed in infected mice, indicating possible additional immune evasion tactics of RVFV. Immunization of mice with recNP combined with Alhydrogel resulted in a strong IgG1 subclass response, compared to a weak IgG2A response, confirming that the immune response was biased towards Th-2 humoral immunity (Figure 6.1). Although tracking of the development of clinical disease in immunized and non-immunized mice after RVFV challenge was already previously shown (Chapter 4) and thus not a priority of the study presented in this chapter, it is worth noting that none of the recNP/Alhydrogel immunized mice developed any clinical illness during the course of the study, compared to non-immunized mice which displayed typical signs from day 3 p.i. to the end of the study (day 5). This corresponds to the viral loads detected in the different experimental groups (Table 6.3, 6.4 and 6.5) which shows that recNP/Immunized mice developed a very short and low viremia on day 3 post infection compared to much higher and extended viremia in non-immunized control mice (day 3 ? 5). These results also confirmed that day 3 was indeed the point in acute infection with the highest viral replication in these mice and that analysis of the expression of genes at this point the most applicable. The induction of expression of IFN? has been shown to occur in vitro at 3 ? 6 hours p.i, and it is therefore surprising that the results presented here show induction in vivo in liver and spleen tissue Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 147 P. Jansen van Vuren of immunized mice already at 3 hours p.i., taking into account that after host infection the virus is likely conveyed to lymph nodes where it first replicates before it can spread to the liver and other organs (Haller and Weber, 2009). The expression of the same gene was, however, decreased (but not significantly downregulated) in liver tissue, and at constitutive levels in spleen tissue, of adjuvant and PBS control mice at 3 hours p.i. This decrease might already have been a result of the action of NSs on the type I interferon response of the host. The upregulation of IFN? expression in immunized mice cannot be a direct result of anti-recNP memory since the innate response is general and not antigen specific, and this needs to be further investigated. It might, however, be as a result of some indirect actions. The fact that the anti-recNP response was largely humoral and that these antibodies are not neutralizing might indicate that some other form of antibody dependent mechanism, such as antibody- dependent cell-mediated cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC), was responsible. Antibody-dependent cell-mediated cytotoxicity is an NK cell mediated mechanism making use of specific antibodies, rather than memory cytotoxic T-cells, to form a link between the effector cell and an infected cell presenting an antigen on its surface, leading to lysis of infected cells. Complement-dependent cytotoxicity relies on the interaction of the C1q molecule binding to IgG or IgM already bound to an antigen. The lysis of infected cells that would have otherwise produced progeny virus, because of the inhibitory action of NSs, by these mechanisms might have resulted in the activation of the type I interferon response in neighbouring uninfected cells as a result of the release of dsRNA (Morrill et al., 1990, do Valle et al., 2010). The expression of IFN? remained upregulated in immunized mice up to 12 hours p.i., compared to a brief and much lower upregulation in control mice between 6 and 12 hours p.i. At 72 and 120 hours p.i. IFN? expression in liver tissue was decreased but still within the constitutive range in immunized mice compared to over expression in control mice liver and spleen tissue. This over expression of IFN? in control mice liver and spleen tissue was not able to curb the replication of the virus and was probably more detrimental than valuable, contributing to the pathology of the liver. The sudden downregulation of IFN? expression in spleen tissue of all mice at 24 hours probably contributed to the inability to control viral replication, as shown by high viral titres at 72 hours p.i. in spleen tissue, and was probably a result of the type I interferon inhibitory effect of the NSs protein. It has been shown previously that an early type I interferon response is protective against RVFV infection (Bouloy et al., 2001) and it is thus highly likely that the early and correctly regulated expression of IFN? in immunized mice contributed to effective viral clearance and protection from liver pathology. The fact that IFN? expression was upregulated in brain tissue of adjuvant control mice at 24 hours p.i., yet still the virus replicated to high titres, indicates that IFN? did not play such an important role in innate immunity against RVFV infection in the brain. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 148 P. Jansen van Vuren There was not such a striking difference in the expression of type II interferon (IFN?) between immunized and non-immunized control mice early after infection in liver and spleen tissue. Its expression was more or less similar up to 24 hours after which there was a sudden decline of expression in control mice liver tissue, while expression remained upregulated in immunized mice. This drop preceded the peak of viremia in control mice so it might be that the decreased IFN? expression at this critical time point during the acute infection might have contributed, or even have been a result of, uncontrolled viral replication. Interestingly the expression of IFN? in immunized and control mice was again similar at 72 hours, but at 120 hours expression was upregulated in control mice which might have contributed to the pathology of the liver. Interestingly, IFN? expression was upregulated much earlier in the brain tissue of recNP immunized mice compared to control mice, also corresponding to much less viral replication, which might indicate that IFN? plays an important role in protection against RVFV infection in brain tissue. It has been suggested that the role of IFN? in RVFV pathogenesis is negligible (Morrill et al., 1991b) but this suggestion was based on in vitro results. It has been shown in vivo that IFN? does indeed play a role in the attenuation of RVFV (Couper et al., 2008). There was a dysregulation of IL-10 expression in liver tissue of control mice very early after infection, with expression being upregulated at 3 hours p.i., and again downregulated at 6 hours. The dysregulated expression of IL-10 during a viral infection might actually contribute to immune escape since IL-10 is an anti-inflammatory cytokine that inhibits the actions of Th-1 cells, NK cells, decreases antigen presentation by cells and limits the production of various important cytokines (i.e. IL-12, IL-18 and TNF-?) (Bai et al., 2009). It has been shown that IL-10 is upregulated in vivo and in vitro after West Nile virus (WNV) infection, and that IL-10 deficient mice are less susceptible to WNV infection than mice expressing the gene constitutively (Ubol et al., 2010). Dengue virus has also been shown to replicate less efficiently in vitro when IL-10 expression is suppressed (Hsu et al., 1990, Spencer, 2007, van Putten et al., 2009). Some viruses even express IL-10 homologs to enable them to modulate the host immune system and escape viral clearance (Couper et al., 2008). At 72 and 120 hours p.i. there was again over expression of IL-10 in control mice, which might have led to immune escape by RVFV, although the anti-inflammatory effects of IL-10 might also have been an attempt by the host to counteract severe inflammation of the liver. On the other hand, expression of IL-10 seemed to follow a constitutive pattern of expression in liver tissue of recNP immunized mice, with the gene only being upregulated at 72 hours p.i. when there was viral replication in the liver, and onwards. The interplay between IL-10 and IFN?, which are counter regulatory of each other, also seemed to be at constitutive levels in recNP immunized mice, with IFN? being upregulated early to counter viral infection, and IL- 10 only being upregulated later to curb immunopathology of the liver (Afford et al., 2001). The upregulation of IL-10 expression at 72 hours p.i. in spleen tissue of control mice is possibly an Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 149 P. Jansen van Vuren indication of inflammation of the spleen and the hosts attempt counteract inflammation, but might have contributed to the virus being able to escape immune detection and thus replicate efficiently. The downregulation of IL-10 expression in spleen tissue of recNP immunized mice at 120 hours p.i., and the absence of replicating virus at the same time, is an indication that decreased IL-10 expression is advantageous to the host. The activation of several genes with pro-apoptotic and pro-inflammatory effects, and the suppression of several genes with ant-apoptotic effects, in the liver of control mice 72 hours p.i. most probably contributed to severe hepatic disease. The overexpression of CD40, a member of the TNF receptor superfamily and potent activator of nuclear factor kappa beta, is of particular importance to apoptosis in the liver. CD40 has been shown to induce apoptosis in hepatocytes via a FAS dependent mechanism, the key mechanism for hepatocyte death in the liver (Gold et al., 2003). Mice deficient in the expression of CD40 has been shown to have improved survival during bacterial sepsis as a result of decreased induction of IL-6, IL-10, IL-12 and IFN? expression (Peterss et al., 2009). Apart from its role in apoptosis of hepatocytes, CD40 plays a very important role in mediating B- and T-cell responses, thus the hosts attempt to launch an adaptive immune response to RVFV might actually contribute to immunopathology (Matsui et al., 2002, Anand et al., 2006). The expression of a TNF receptor superfamily ligand (Tnfsf14) was also upregulated in control mice. This protein is able to block TNF? mediated apoptosis but not FAS mediated apoptosis, and is a co-stimulatory factor that enhances T-cell mediated immunity leading to severe inflammation (Denhardt et al., 2001, Mazzali et al., 2002). The upregulation of the genes expressing osteopontin (Spp1) and the Cyclin-dependent kinase inhibitor P21 (Cdkn1a) is evidence of the host?s attempt to counteract the damaging effects of the infection. Osteopontin is a cell survival factor and influences tissue repair at sites of severe inflammation (Gartel and Radhakrishnan, 2005). The Cyclin-dependent kinase inhibitor P21 (Cdkn1a) is a protein that plays a role in cell cycle control, with overexpression of P21 leading to cell cycle arrest (Choi et al., 2001, Gartell and Tyner, 2002). The protein has pro- or anti-apoptotic effects, has the ability to inhibit proliferation of cells and is upregulated in response to tissue injury (Dong et al., 2005). P21 interacts, amongst others, with the growth arrest and DNA damage-inducible gene 45 (Gadd45) (Chung et al., 2003), which was also upregulated in control mice. Gadd45 has been implicated in DNA repair, apoptosis, regulation of signal transduction and cell cycle control (Mansuroglu et al., 2010). The NSs protein of RVFV has been shown to interact with some specific regions of host cell DNA, causing defects in host chromosome structure and segregation (Gracie et al., 2003). Therefore it might be that these DNA damage inducible proteins are upregulated in an attempt to arrest the cell cycle of affected cells and prevent apoptosis. The fact that P21 is upregulated in Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 150 P. Jansen van Vuren healthy immunized mice (9 fold) is probably as a result of the very low level of viral replication in their livers. Despite several genes involved in the activation, differentiation and proliferation of B- and T- cells being upregulated in control mice and indicating activation of adaptive immunity, several genes with important functions in immune activation were also downregulated. Interleukin-18 (Il-18) is involved in the maturation and cytotoxicity of T-cells and NK cells, which are important cells in cellular and innate immunity respectively, and was downregulated in control mice (Mavoungou et al., 2005). The prolactin receptor is important in the functioning of NK cells but was downregulated in control mice (Kasmar et al., 2009). The expression of Cd1d1 antigen, which activates Natural Killer T- cells, was downregulated in control mice (Palmer et al., 2008). Interleukin-7, which is important for the development and survival of B-cells, T-cells and NK-cells were also downregulated in PBS control mice. The decreased expression of these important genes, and overexpression of other genes that have the ability to negatively regulate T-cell responses (Cblb and Sftpd), might have contributed to the inability of the control mice to clear virus from their liver despite the activation of other genes. In summary this study shows that expression of IFN? is upregulated later and to a lesser extent in the liver of non-immunized mice compared to immunized mice after RVFV challenge, but over expressed in control mice during acute infection of the liver. This expression pattern of the type I interferon, which is very important in the pathogenicity of RVF, possibly resulted in immunized mice being able to clear the viral infection much more efficiently than non-immunized mice. The over expression of IFN? during acute liver infection was probably detrimental to the health of the control mice too. The expression of interleukin-10 was also irregular in the liver of control mice, possibly leading to the virus escaping detection to a certain extent and thus leading to excessive viral replication. The results also indicate activation of apoptosis in infected liver at the acute stage of infection, and severe inflammation which probably contributed to the pathology of the infection. The results for some of the genes also indicate inactivation of the induction of important immune cells which could have contributed to immune evasion. The results from this study will be useful for the evaluation of future candidate vaccines as genes that are regulated in na?ve mice in response to RVFV infection have been identified. Host gene responses identified in this study may serve as potential targets for development of therapeutic interventions by suppressing inflammatory and apoptotic effects of RVFV infection in the liver. It would also be interesting to determine whether the same patterns of expressed genes are achieved in a host animal model such as sheep. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 151 P. Jansen van Vuren CHAPTER SEVEN CONCLUSIONS At the advent of this project, all available techniques for the detection of antibodies against RVFV relied on the use of reagents that were prepared from infectious virus, thus requiring high biocontainment facilities and/or vaccinated personnel which restricts their production to a very limited number of laboratories in the world. In this study a recombinant RVFV nucleocapsid protein was utilized to develop several ELISA-based assays for rapid and safe diagnosis of RVF. An ELISA was developed and validated for the detection of IgG antibodies in wildlife species. There is still some controversy regarding the maintenance of RVFV during long inter-epidemic periods. One school of thought is that the virus is transmitted transovarially to mosquito eggs where it can survive for many years until the next sufficient flooding occurs. However, evidence to support this phenomenon is very limited and, in fact, transovarial transmission has been demonstrated only once and could not be reproduced ever since. The more likely explanation for natural transmission mechanism of RVFV is that there is low level inter-epidemic circulation of the virus between mosquitoes and wildlife that are not as closely monitored by farmers or game wardens as domestic livestock. Exceptional rainfall might then lead to spill-over of RVFV into domestic livestock and humans. This hypothesis seems to be supported by the presence of anti-RVFV antibodies in various wildlife species. The ELISA for IgG detection in wildlife developed in this study will be a valuable tool for facilitating cost-effective, large- scale sero-surveys which data might contribute to better understanding of RVF epidemiology, including natural transmission cycles. Understanding the inter-epidemic transmission of RVFV will contribute to more effective control of the disease. Another useful ELISA that was developed as part of this study is the human anti-RVFV IgM detection test based on a horseradish peroxidase labelled recombinant RVFV nucleocapsid protein. This test is an improvement on the traditional whole RVFV based ELISA, which carries some safety risks and is expensive to produce. An IgM detection ELISA is an essential part of the repertoire of tests necessary for successful diagnosis of a viral hemorrhagic fever, since it is indicative of a recent infection. An ELISA was also developed for the safe detection of RVFV antigen in human and animal specimens, another important addition to the array of tests for RVF diagnosis. The assay is based on a completely safe set of reagents and thus does not have to be prepared in a high containment facility. It enables the detection of antigen in decaying tissue long after infectious virus has been inactivated, which is an important tool for example when dead animals or foetuses are not immediately discovered in the field. A study was done to compare the diagnostic accuracy of the assays developed as part of this study to traditional assays, and it was found that the recombinant NP based ELISAs performed satisfactorily. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 152 P. Jansen van Vuren There is no vaccine available for commercial human use, and the vaccines available for animal use are only available from a single facility in Africa, making supply inconsistent, have potential serious drawbacks in terms of adverse effects on animals when live-attenuated vaccines are administered or require multiple immunizations, in case of inactivated vaccines, to enable protection. Recombinant DNA technology and reverse genetics have been used extensively in recent times to develop vaccine candidates. Many of these candidates have shown promising results in preliminary studies but needs further evaluation before they can be used for large-scale vaccination. Most of these candidates have focussed on the glycoproteins of RVFV, because they contain neutralizing epitopes and neutralizing antibodies against RVFV is a known protective correlate. Studies on related viruses from the Bunyaviridae family have, however, shown that immunity generated against the nucleocapsid proteins of these viruses also offered some protection against subsequent viral challenge. A preliminary study was also done using a recombinant RVFV nucleocapsid as an immunogen in mice, which resulted in partial protection against lethal RVFV challenge. Therefore it was decided to do a more in-depth evaluation of a recombinant nucleocapsid protein of RVFV as an immunogen in a mouse model, but also in an actual host animal model, namely sheep. The results from this study pointed out several important findings that could aid future RVFV vaccine development. It was shown that the choice of adjuvant is very important when using subunit immunogens, not only for enhancing immune responses but more importantly to bias immune responses towards a certain branch of the host immune system. Immunization of mice with recNP and certain adjuvants resulted in complete protection from lethal RVFV challenge. Two commercially available adjuvants, Saponin and Aluminium hydroxide gel (alum), yielded the best protection. Saponin is known to activate Th1 and Th2 immunity against antigens, and this was confirmed in this study with strong IgG1 and IgG2A responses against RVFV NP measured in mice. Although mice immunized with a high dose of recNP combined with Saponin yielded 100% protection, a lower dose only partially protected mice against lethal RVFV challenge. When recNP was combined with alum, known to strictly bias responses towards Th2 humoral immunity, even the lower dose resulted in 100% protection against challenge. The fact that recNP combined with alum in this study also activated Th2 humoral rather than Th1 cellular immunity, yet still yielded 100% protection, was surprising since anti- NP responses are not neutralizing and the common hypothesis was that it was cellular responses against the NP that was responsible for protection. There are, however, mechanisms known in addition to physical neutralization of virus by which antibodies can target and destroy pathogens. These are antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC). These mechanisms work by using pre-existing antibodies against a foreign antigen to link adaptive immunity to innate immune mechanisms such as the complement cascade and NK cells. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 153 P. Jansen van Vuren Whereas physical neutralization requires antibodies to bind to surface proteins of the virus, glycoproteins in the case of RVFV, blocking epitopes necessary for entry into host cells by specific receptors, ADCC and CDC utilizes the recognition of antigens presented on the surface of host cells, which would include antigens not necessarily present on the virus surface, and subsequent lysis of those infected cells before virus replication. Therefore it might be that pre-existing anti-NP antibodies in recNP immunized mice resulted in the lysis of infected cells by either of these mechanisms before RVFV could replicate efficiently and cause a systemic infection, resulting in protection. This hypothesis is based on elimination of other possible mechanisms and not on concrete scientific data, and thus needs further investigation. This was the first study showing complete protection in a mouse model using the RVFV NP as vaccine candidate, and it confirmed the importance of including NP in RVFV vaccine candidates. Although the recNP/adjuvant combinations were highly immunogenic in sheep, this immunity was not able to significantly decrease viral replication. The reason why the same recNP/adjuvant combinations that protected mice from RVFV infection could not protect sheep is unclear, but it is important to point out that mice were considered as protected once they did not develop clinical disease and did not have viral replication in specific organ tissues, whereas in sheep we could only use absence of virus in the blood as an indication of protection. Although it might therefore seem that a proper comparison wasn?t done between mice and sheep, this was not the aim of the study. Because RVFV is a mosquito borne virus, the decrease in concentration of infectious virus in the blood is a very important indicator of the effectiveness of a vaccine candidate since the aim of the vaccine would not only be to protect the immunized host animal from development of disease, but more importantly to decrease the risk of transmitting virus amongst susceptible vertebrate hosts and competent mosquito vectors. Based on this information, the specific recNP/adjuvant combinations and concentrations used in this study can therefore not be considered as a potential vaccine candidate in RVF host species. Whether increasing the recNP doses, increasing the number of immunizations or changing the route of immunization would result in better protection is unknown, but the results from this study highlighted the very important fact that any potential RVF vaccine candidate should be properly evaluated in a host species before any conclusions can be made regarding its efficacy. To enable more efficient development of vaccines and treatments against any infectious disease, one first needs a better understanding of the pathogenesis of that disease. Until recently the pathogenesis of RVF was poorly understood but reverse genetics enabled scientists to determine the RVFV NSs protein as the virulence marker of the virus. This protein acts by counteracting the effects of the very important innate immune system component, type I interferon. Another RVFV protein, NSm, was also recently shown to counteract apoptosis, another important innate immune mechanism. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 154 P. Jansen van Vuren Not much, however, was known about the activation or repression of other immune functions in vivo during RVFV infection. Despite the ability of NSs to counteract type I interferon, it was recently shown that this system is not completely repressed in vitro during RVFV infection, and it is therefore logical to assume that there must be some additional mechanisms by which the virus can evade the host immune response to enable sufficient viral replication. Although RVFV is a cytolytic virus that readily causes systemic infections, it was not clear whether immunopathology might also play a role in RVF disease progression and to what extent anti-NP responses might play a role in protection in this regard. In this study the expression of several genes involved in innate and adaptive immunity was evaluated in important target organs of the virus, at several time points after RVFV infection, and compared between recNP/Alhydrogel-immunized mice and two control groups consisting of Alhydrogel and PBS inoculated mice. The results clearly indicate that recNP immunized mice were able to mount an earlier and stronger innate immune response compared to both control groups where these responses were repressed initially. The immunized mice were also able to control expression of genes with anti- inflammatory effects that might result in immune evasion when incorrectly regulated more appropriately than non-immunized mice early after infection. This resulted in replication being kept to a minimum in immunized mice compared to non-immunized mice that had excessive replication of virus in their target organs. During the acute phase of infection this excessive replication of virus in non-immunized mice was accompanied by massive upregulation of pro-inflammatory responses and genes with pro-apoptotic effects in their livers. These effects that very likely contributed to the pathology of the liver in non-immunized mice, the main target organ of RVFV, were not demonstrable in immunized mice. In addition to these immunopathological effects in non-immunized mice, there was also evidence of up- and downregulation of several important genes that could have translated into dysregulation of the activation of adaptive immunity, which very likely contributed to immune evasion. In conclusion, this study not only expands the repertoire of safe and validated diagnostic methods for RVF diagnosis, but also contributes to a better understanding of the role of the NP in protection against RVF and the pathogenesis of the disease on a molecular level. There are still, however, many challenging issues regarding this neglected but emerging disease requiring improvements to be made in the serodiagnosis of RVF, for example the development of a point-of- care, or penside tests for rapid detection of antibodies or antigen. This would not only be a valuable diagnostic tool during suspected RVF outbreaks in remote areas where laboratories are not readily available, but also for the rapid screening of livestock at import/export stations to enable safe transportation between endemic and non-endemic regions. The development of ELISAs utilizing other Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 155 P. Jansen van Vuren structural and non-structural recombinant RVFV antigens would also be an important addition to the current array of tests. This might not only further improve sensitivity and specificity of detection, but also enable differentiation between vaccinated and non-vaccinated animals, and better understanding of humoral responses in naturally infected and vaccinated individuals. Although this study showed earlier activation of innate immunity in recNP immunized mice, the memory mechanism responsible for this protection needs further investigation. The possibility that antibody-dependent cell-mediated cytotoxicity or complement-dependent cytotoxicity is responsible for this protection needs to be elucidated. Because of the bi-phasic nature of RVF in some infected individuals, as a result of virus infection of the brain, it would be interesting to gain insight into the expression of apoptotic and inflammatory genes in this non-regenerative tissue. It will also be important to investigate how the virus crosses the blood-brain barrier, and why only a limited number of individuals develop encephalitis as a result of central nervous system infection. Future RVFV vaccine candidates should include glycoproteins and the nucleocapsid protein to induce the optimal humoral and cellular protection mechanisms against infection with field strains of the virus. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 156 P. Jansen van Vuren APPENDIX 1 ETHICS APPROVALS Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 157 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 158 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 159 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 160 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 161 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 162 P. Jansen van Vuren APPENDIX 2 REPRINTS OF PUBLISHED PAPERS Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 163 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 164 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 165 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 166 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 167 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 168 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 169 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 170 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 171 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 172 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 173 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 174 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 175 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 176 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 177 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 178 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 179 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 180 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 181 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 182 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 183 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 184 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 185 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 186 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 187 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 188 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 189 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 190 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 191 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 192 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 193 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 194 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 195 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 196 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 197 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 198 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 199 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 200 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 201 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 202 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 203 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 204 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 205 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 206 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 207 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 208 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 209 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 210 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 211 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 212 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 213 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 214 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 215 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 216 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 217 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 218 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 219 P. Jansen van Vuren Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 220 P. Jansen van Vuren Gene Gene name Transcript accession number recNP Immunized Fold change and P-value Adjuvant control Fold change and P-value PBS control Fold change and P-value Ap3b1 Adaptor-related protein complex 3, beta 1 subunit NM_009680 1.65 p = 0.21 -1.09 p = 0.83 1.13 p = 0.58 Bad BCL2-associated agonist of cell death NM_007522 1.06 p = 0.62 -2.19* p = 0.0004 -1.64* p = 0.001 Cxcr5 Chemochine (C-X-C motif) receptor 5 NM_007551 1.56 p = 0.88 5.53 p = 0.18 5.23 p = 0.23 Cblb E3 Ubiquitin Ligase Casitas B-lineage lymphoma b NM_001033238 1.07 p = 0.68 7.16* p = 0.006 8.65* p = 0.001 Ccnd3 Cyclin D3 NM_007632 1.02 p = 0.86 1.12 p = 0.29 1.27* p = 0.02 Cd1d1 CD1d1 antigen NM_007639 1.27 p = 0.41 -6.27* p = 0.005 -6.88* p = 0.003 Cd2 CD2 antigen NM_013486 1.02 p = 0.90 2.56* p = 0.02 2.66* p = 0.04 Cd28 CD28 antigen NM_007642 6.48 p = 0.13 22.33* p = 0.002 21.41* p = 0.005 Cd3d CD3 antigen, delta polypeptide NM_013487 1.15 p = 0.23 2.39 p = 0.07 1.86 p = 0.06 Cd3e CD3 antigen, epsilon polypeptide NM_007648 1.58 p = 0.30 3.44 p = 0.06 3.28* p = 0.02 Cd3g CD3 antigen, gamma polypeptide NM_009850 1.34 p = 0.20 1.14 p = 0.57 -1.47 p = 0.12 Cd4 CD4 antigen NM_013488 1.06 p = 0.74 -1.89 p = 0.45 -1.01 p = 0.92 Cd40 CD40 antigen NM_011611 2.91 p = 0.06 29.40* p = 0.002 25.87* p = 0.003 Cd40lg CD40 ligand NM_011616 1.41 p = 0.48 1.42 p = 0.44 1.83 p = 0.11 Cd74 CD74 antigen (invariant polypeptide of major histocompatibility complex, class II antigen-associated) NM_010545 2.03 p = 0.12 1.39* p = 0.012 1.36 p = 0.13 Cd81 CD81 antigen NM_133655 1.04 p = 0.68 -2.17* p = 0.002 -1.77* p = 0.002 Cd8a CD8 antigen, alpha chain NM_001081110 2.33 p = 0.40 4.88 p = 0.13 2.81 p = 0.23 Cd8b1 CD8 antigen, beta chain 1 NM_009858 -1.05 p = 0.99 4.08 p = 0.14 2.39 p = 0.40 Cd93 CD93 antigen NM_010740 1.14 p = 0.85 83.91 p = 0.06 144.67* p = 0.001 Cdkn1a Cyclin-dependent kinase inhibitor 1A (P21) NM_007669 9.02* p = 0.02 127.19* p = 0.0000 145.01* p = 0.0006 Clcf1 Cardiotrophin-like cytokine factor 1 NM_019952 1.27 p = 0.89 33.92* p = 0.01 52.10* p = 0.002 Cr2 Complement receptor 2 NM_007758 -2.0 p = 0.15 1.07 p = 0.74 1.61 p = 0.10 Csf2 Colony stimulating factor 2 (granulocyte- macrophage) NM_009969 -1.56 p = 0.66 2.37 p = 0.45 2.09 p = 0.58 Cxcl12 Chemokine (C-X-C motif) ligand 12 NM_021704 1.24* p = 0.03 -2.21* p = 0.02 -1.64* p = 0.008 Cxcr4 Chemokine (C-X-C motif) receptor 4 NM_009911 1.66* p = 0.006 22.12* p = 0.01 22.84* p = 0.002 Dock2 Dedicator of cyto-kinesis 2 NM_033374 1.11 p = 0.43 4.85* p = 0.04 3.47* p = 0.003 Egr1 Early growth response 1 NM_007913 -5.00* p = 0.01 3.88* p = 0.005 4.54* p = 0.004 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 221 P. Jansen van Vuren Flt3 FMS-like tyrosine kinase 3 NM_010229 -1.06 p = 0.87 12.62* p = 0.004 9.19* p = 0.03 Gadd45g Growth arrest and DNA-damage- inducible 45 gamma NM_011817 7.26 p = 0.26 5.35* p = 0.002 4.37* p = 0.008 Glmn Glomulin, FKBP associated protein NM_133248 1.08 p = 0.59 -1.03 p = 0.82 1.05 p = 0.73 H2-Aa Histocompatibility 2, class II antigen A, alpha NM_010378 1.84 p = 0.11 2.01* p = 0.014 1.56 p = 0.16 H60a Histocompatibility 60a NM_010400 -1.51 p = 0.70 16.50* p = 0.046 13.99* p = 0.036 Hdac5 Histone deacetylase 5 NM_010412 1.05 p = 0.65 -1.46 p = 0.04 -2.18* p = 0.02 Hdac7 Histone deacetylase 7 NM_019572 -1.07 p = 0.68 4.83* p = 0.012 5.58* p = 0.0007 Hells Helicase, lymphoid specific NM_008234 1.34 p = 0.46 2.15 p = 0.08 2.64* p = 0.007 Hsp90aa1 Heat shock protein 90, alpha (cytosolic), class A member 1 NM_010480 1.08 p = 0.67 1.33 p = 0.13 1.48* p = 0.012 Icosl Icos ligand NM_015790 2.33 p = 0.30 3.48 p = 0.25 3.29 p = 0.07 Ifng Interferon gamma NM_008337 4.05 p = 0.32 4.90 p = 0.09 3.10 p = 0.12 Igbp1 Immunoglobulin (CD79A) binding protein 1 NM_008784 -1.23 p = 0.15 -2.39* p = 0.002 -2.37* p = 0.002 Igbp1b Immunoglobulin (CD79A) binding protein 1b NM_015777 1.39 p = 0.37 7.26* p = 0.04 4.53 p = 0.10 Il10 Interleukin 10 NM_010548 3.99 p = 0.21 17.64* p = 0.04 10.65* p = 0.02 Il11 Interleukin 11 NM_008350 1.93 p = 0.20 275.80 p = 0.13 772.47* p = 0.007 Il12b Interleukin 12B NM_008352 2.59 p = 0.25 11.75* p = 0.02 20.35* p = 0.005 Il15 Interleukin 15 NM_008357 1.12 p = 0.48 -1.64* p = 0.026 -1.54* p = 0.04 Il18 Interleukin 18 NM_008360 1.21 p = 0.12 -3.52* p = 0.0002 -4.91* p = 0.0001 Il27 Interleukin 27 NM_145636 3.25 p = 0.63 5.73 p = 0.32 11.42 p = 0.10 Il2ra Interleukin 2 receptor, alpha chain NM_008367 3.18 p = 0.09 5.66* p = 0.04 4.41 p = 0.06 Il4 Interleukin 4 NM_021283 2.04 p = 0.35 21.42 p = 0.24 8.73* p = 0.03 Il7 Interleukin 7 NM_008371 1.64* p = 0.05 -1.70 p = 0.06 -2.38* p = 0.009 Impdh1 Inosine 5'-phosphate dehydrogenase 1 NM_011829 2.13* p = 0.004 3.26 p = 0.25 4.92* p = 0.02 Impdh2 Inosine 5'-phosphate dehydrogenase 2 NM_011830 1.18 p = 0.15 1.00 p = 0.99 1.10 p = 0.48 Inha Inhibin alpha NM_010564 -1.16 p = 0.74 4.49 p = 0.06 3.12 p = 0.17 Irf4 Interferon regulatory factor 4 NM_013674 1.51 p = 0.96 2.59 p = 0.39 1.95 p = 0.79 Jag2 Jagged 2 NM_010588 1.67 p = 0.18 6.79* p = 0.02 7.46* p = 0.02 Ms4a1 Membrane-spanning 4-domains, subfamily A, member 1 NM_007641 1.39 p = 0.37 7.26* p = 0.04 4.53 p = 0.10 Nkx2-3 NK2 transcription factor related, locus 3 (Drosophila) NM_008699 1.10 p = 0.72 1.56 p = 0.26 1.16 p = 0.59 Nos2 Nitric oxide synthase 2, inducible NM_010927 1.05 p = 0.97 25.71 p = 0.06 32.07* p = 0.001 Pawr PRKC, apoptosis, WT1, regulator NM_054056 1.06 p = 0.57 2.68* p = 0.0003 3.05* p = 0.0006 Pdcd1lg2 Programmed cell death 1 ligand 2 NM_021396 1.90 p = 0.1 12.42* p = 0.0004 9.47 p = 0.08 Pik3cd Phosphatidylinositol 3-kinase catalytic delta polypeptide NM_008840 3.59* p = 0.04 5.77 p = 0.13 7.53* p = 0.002 Pik3r1 Phosphatidylinositol 3-kinase, regulatory subunit, polypeptide 1 (p85 alpha) NM_001024955 1.10 p = 0.63 -3.80* p = 0.007 -3.63* p = 0.005 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 222 P. Jansen van Vuren Prkcd Protein kinase C, delta NM_011103 1.30 p = 0.1 3.73* p = 0.03 4.01* p = 0.002 Prkcq Protein kinase C, theta NM_008859 1.92 p = 0.18 4.63 p = 0.09 3.89 p = 0.06 Prlr Prolactin receptor NM_011169 1.15 p = 0.42 -62.79* p = 0.0003 -59.16* p = 0.0003 Ptprc Protein tyrosine phosphatase, receptor type, C NM_011210 2.07 p = 0.57 8.14 p = 0.055 3.69 p = 0.13 Rag1 Recombination activating gene 1 NM_009019 -1.05 p = 0.98 4.98* p = 0.05 3.11 p = 0.13 Relb Avian reticuloendotheliosis viral (v-rel) oncogene related B NM_009046 2.50* p = 0.04 12.30* p = 0.002 15.85* p = 0.002 Rgs1 Regulator of G-protein signaling 1 NM_015811 1.67 p = 0.25 41.67* p = 0.006 24.76* p = 0.016 Sftpd Surfactant associated protein D NM_009160 1.30 p = 0.41 20.40* p = 0.03 26.35* p = 0.0006 Sit1 Suppression inducing transmembrane adaptor 1 NM_019436 1.92 p = 0.24 3.64 p = 0.07 2.96 p = 0.25 Sla2 Src-like-adaptor 2 NM_029983 1.08 p = 0.66 2.38 p = 0.16 2.30* p = 0.007 Socs5 Suppressor of cytokine signaling 5 NM_019654 -1.01 p = 0.90 2.55* p = 0.006 3.63* p = 0.0007 Spp1 Secreted phosphoprotein 1 NM_009263 -1.10 p = 0.62 29.12* p = 0.003 26.97* p = 0.001 Tlr1 Toll-like receptor 1 NM_030682 1.82 p = 0.2 8.68 p = 0.07 5.55 p = 0.06 Tlr4 Toll-like receptor 4 NM_021297 3.46 p = 0.06 14.13* p = 0.03 8.28* p = 0.005 Tlr6 Toll-like receptor 6 NM_011604 1.30 p = 0.43 2.11 p = 0.14 1.77 p = 0.12 Tnfrsf13b Tumor necrosis factor receptor superfamily, member 13b NM_021349 -1.35 p = 0.76 6.67* p = 0.0006 4.27 p = 0.15 Tnfrsf13c Tumor necrosis factor receptor superfamily, member 13c NM_028075 1.23 p = 0.68 2.88 p = 0.12 2.46 p = 0.15 Tnfsf13b Tumor necrosis factor (ligand) superfamily, member 13b NM_033622 1.32 p = 0.60 8.74* p = 0.014 5.75* p = 0.005 Tnfsf14 Tumor necrosis factor (ligand) superfamily, member 14 NM_019418 2.14 p = 0.08 24.66* p = 0.04 16.80* p = 0.001 Traf6 Tnf receptor-associated factor 6 NM_009424 1.53 p = 0.06 1.72* p = 0.03 2.34 p = 0.06 Vav1 Vav 1 oncogene NM_011691 2.20 p = 0.13 7.50* p = 0.03 8.00* p = 0.02 Was Wiskott-Aldrich syndrome homolog (human) NM_009515 1.53 p = 0.50 4.33 p = 0.07 4.64* p = 0.004 Wwp1 WW domain containing E3 ubiquitin protein ligase 1 NM_177327 -1.07 p = 0.38 -9.16* p = 0.0001 -11.99* p = 0.0001 Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 223 P. Jansen van Vuren REFERENCES AFFORD, S. C., AHMED-CHOUDHURY, J., RANDHAWA, S., et al. (2001) CD40 activation- induced, Fas-dependent apoptosis and NF-kappaB/AP-1 signaling in human intrahepatic biliary epithelial cells. Faseb J, 15, 2345-54. AJAY, A. K., UPADHYAY, A. K., SINGH, S., et al. (2010) Cdk5 phosphorylates non-genotoxically overexpressed p53 following inhibition of PP2A to induce cell cycle arrest/apoptosis and inhibits tumor progression. Mol Cancer, 9, 204. AL-HAZMI, M., AYOOLA, E. A., ABDURAHMAN, M., et al. (2003) Epidemic Rift Valley fever in Saudi Arabia: a clinical study of severe illness in humans. Clin Infect Dis, 36, 245-52. ANAND, S., WANG, P., YOSHIMURA, K., et al. (2006) Essential role of TNF family molecule LIGHT as a cytokine in the pathogenesis of hepatitis. J Clin Invest, 116, 1045-51. ANDERSON, E. C. & ROWE, L. W. (1998) The prevalence of antibody to the viruses of bovine virus diarrhoea, bovine herpes virus 1, rift valley fever, ephemeral fever and bluetongue and to Leptospira sp in free-ranging wildlife in Zimbabwe. Epidemiol Infect, 121, 441-9. ANDERSON, G. W., JR. & PETERS, C. J. (1988) Viral determinants of virulence for Rift Valley fever (RVF) in rats. Microb Pathog, 5, 241-50. ANDERSON, G. W., JR., SLONE, T. W., JR. & PETERS, C. J. (1987) Pathogenesis of Rift Valley fever virus (RVFV) in inbred rats. Microb Pathog, 2, 283-93. ANONYMOUS (2007) Rift Valley fever outbreak--Kenya, November 2006-January 2007. MMWR Morb Mortal Wkly Rep, 56, 73-6. ANYANGU, A. S., GOULD, L. H., SHARIF, S. K., et al. (2010) Risk factors for severe Rift Valley fever infection in Kenya, 2007. Am J Trop Med Hyg, 83, 14-21. ARISHI, H. M., AQEEL, A. Y. & AL HAZMI, M. M. (2006) Vertical transmission of fatal Rift Valley fever in a newborn. Ann Trop Paediatr, 26, 251-3. BAI, F., TOWN, T., QIAN, F., et al. (2009) IL-10 signaling blockade controls murine West Nile virus infection. PLoS Pathog, 5, e1000610. BALKHY, H. H. & MEMISH, Z. A. (2003) Rift Valley fever: an uninvited zoonosis in the Arabian peninsula. Int J Antimicrob Agents, 21, 153-7. BARNARD, B. J. (1979) Rift Valley fever vaccine--antibody and immune response in cattle to a live and an inactivated vaccine. J S Afr Vet Assoc, 50, 155-7. BARNARD, B. J. & BOTHA, M. J. (1977) An inactivated rift valley fever vaccine. J S Afr Vet Assoc, 48, 45-8. BENGIS, R. G., LEIGHTON, F. A., FISCHER, J. R., et al. (2004) The role of wildlife in emerging and re-emerging zoonosis. Revue Scientifique Et Technique - Office International Des Epizooties, 23, 497-511. BESSELAAR, T. G. & BLACKBURN, N. K. (1991) Topological mapping of antigenic sites on the Rift Valley fever virus envelope glycoproteins using monoclonal antibodies. Arch Virol, 121, 111-24. BESSELAAR, T. G. & BLACKBURN, N. K. (1992) The synergistic neutralization of Rift Valley fever virus by monoclonal antibodies to the envelope glycoproteins. Arch Virol, 125, 239-50. BETTINGER, G. E., PETERS, C. J., PITTMAN, P., et al. (2009) Rift Valley fever MP-12 vaccine: a university, government, & industry collaborative development. Rift Valley Fever Workshop. BHARDWAJ, N., HEISE, M. T. & ROSS, T. M. (2010) Vaccination with DNA plasmids expressing Gn coupled to C3d or alphavirus replicons expressing gn protects mice against Rift Valley fever virus. PLoS Negl Trop Dis, 4, e725. BILLECOCQ, A., GAULIARD, N., LE MAY, N., et al. (2008) RNA polymerase I-mediated expression of viral RNA for the rescue of infectious virulent and avirulent Rift Valley fever viruses. Virology, 378, 377-84. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 224 P. Jansen van Vuren BILLECOCQ, A., SPIEGEL, M., VIALAT, P., et al. (2004) NSs protein of Rift Valley fever virus blocks interferon production by inhibiting host gene transcription. J Virol, 78, 9798-806. BIRD, B. H., ALBARINO, C. G., HARTMAN, A. L., et al. (2008) Rift valley fever virus lacking the NSs and NSm genes is highly attenuated, confers protective immunity from virulent virus challenge, and allows for differential identification of infected and vaccinated animals. J Virol, 82, 2681-91. BIRD, B. H., ALBARINO, C. G. & NICHOL, S. T. (2007a) Rift Valley fever virus lacking NSm proteins retains high virulence in vivo and may provide a model of human delayed onset neurologic disease. Virology, 362, 10-5. BIRD, B. H., BAWIEC, D. A., KSIAZEK, T. G., et al. (2007b) Highly sensitive and broadly reactive quantitative reverse transcription-PCR assay for high-throughput detection of Rift Valley fever virus. J Clin Microbiol, 45, 3506-13. BIRD, B. H., MAARTENS, L. H., CAMPBELL, S., et al. (2011) Rift Valley fever virus vaccine lacking the NSs and NSm genes is safe, non-teratogenic, and confers protection from viremia, pyrexia, and abortion following lethal challenge in adult and pregnant sheep. J Virol. BISHOP, D. H., CALISHER, C. H., CASALS, J., et al. (1980) Bunyaviridae. Intervirology, 14, 125? 143. BOESTEANU, A. C. & KATSIKIS, P. D. (2009) Memory T cells need CD28 costimulation to remember. Semin Immunol, 21, 69-77. BOHLSON, S. S., ZHANG, M., ORTIZ, C. E., et al. (2005) CD93 interacts with the PDZ domain- containing adaptor protein GIPC: implications in the modulation of phagocytosis. J Leukoc Biol, 77, 80-9. BOTROS, B., OMAR, A., ELIAN, K., et al. (2006) Adverse response of non-indigenous cattle of European breeds to live attenuated Smithburn Rift Valley fever vaccine. J Med Virol, 78, 787- 91. BOULOY, M. & FLICK, R. (2009) Reverse genetics technology for Rift Valley fever virus: current and future applications for the development of therapeutics and vaccines. Antiviral Res, 84, 101-18. BOULOY, M., JANZEN, C., VIALAT, P., et al. (2001) Genetic evidence for an interferon- antagonistic function of rift valley fever virus nonstructural protein NSs. J Virol, 75, 1371-7. BOULOY, M. & WEBER, F. (2010) Molecular biology of rift valley Fever virus. Open Virol J, 4, 8- 14. BURT, F. J., SWANEPOEL, R. & BRAACK, L. E. (1993) Enzyme-linked immunosorbent assays for the detection of antibody to Crimean-Congo haemorrhagic fever virus in the sera of livestock and wild vertebrates. Epidemiol Infect, 111, 547-57. BUSQUETS, N., XAVIER, F., MARTIN-FOLGAR, R., et al. (2010) Experimental infection of young adult European breed sheep with Rift Valley fever virus field isolates. Vector Borne Zoonotic Dis, 10, 689-96. CALZAVARA-SILVA, C. E., GOMES, A. L., MAIA, R. C., et al. (2009) Early molecular markers predictive of dengue hemorrhagic fever. An Acad Bras Cienc, 81, 671-7. CAPLEN, H., PETERS, C. J. & BISHOP, D. H. (1985) Mutagen-directed attenuation of Rift Valley fever virus as a method for vaccine development. J Gen Virol, 66 ( Pt 10), 2271-7. CEREDIG, R., RAUCH, M., BALCIUNAITE, G., et al. (2006) Increasing Flt3L availability alters composition of a novel bone marrow lymphoid progenitor compartment. Blood, 108, 1216-22. CETRE-SOSSAH, C., BILLECOCQ, A., LANCELOT, R., et al. (2009) Evaluation of a commercial competitive ELISA for the detection of antibodies to Rift Valley fever virus in sera of domestic ruminants in France. Prev Vet Med, 90, 146-9. CHO, S. J., KANG, M. J., HOMER, R. J., et al. (2006) Role of early growth response-1 (Egr-1) in interleukin-13-induced inflammation and remodeling. J Biol Chem, 281, 8161-8. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 225 P. Jansen van Vuren CHOI, Y. L., PARK, S. H., JANG, J. J., et al. (2001) Expression of the G1-S modulators in hepatitis B virus-related hepatocellular carcinoma and dysplastic nodule: association of cyclin D1 and p53 proteins with the progression of hepatocellular carcinoma. J Korean Med Sci, 16, 424-32. CHUNG, H. K., YI, Y. W., JUNG, N. C., et al. (2003) CR6-interacting factor 1 interacts with Gadd45 family proteins and modulates the cell cycle. J Biol Chem, 278, 28079-88. COETZER, J. A. & BARNARD, B. J. (1977) Hydrops amnii in sheep associated with hydranencephaly and arthrogryposis with wesselsbron disease and rift valley fever viruses as aetiological agents. Onderstepoort J Vet Res, 44, 119-26. COLEMAN, J. W. (2002) Nitric oxide: a regulator of mast cell activation and mast cell-mediated inflammation. Clin Exp Immunol, 129, 4-10. COLLETT, M. S., KEEGAN, K., HU, S. L., et al. (1987) Protective subunit immunogens to Rift Valley fever virus from bacteria and recombinant vaccinia virus. IN MAHY, B., KOLAKOFSKY, D. (Ed.) The Biology of Negative Strand Viruses. New York, Elsevier. COUPER, K. N., BLOUNT, D. G. & RILEY, E. M. (2008) IL-10: the master regulator of immunity to infection. J Immunol, 180, 5771-7. CRIBBS, D. H., GHOCHIKYAN, A., VASILEVKO, V., et al. (2003) Adjuvant-dependent modulation of Th1 and Th2 responses to immunization with beta-amyloid. Int Immunol, 15, 505-14. CROWTHER, J. R., UNGER, H. & VILJOEN, G. J. (2006) Aspects of kit validation for tests used for the diagnosis and surveillance of livestock diseases: producer and end-user responsibilities. Rev Sci Tech, 25, 913-35. CURNOW, S. J., WLOKA, K., FAINT, J. M., et al. (2004) Topical glucocorticoid therapy directly induces up-regulation of functional CXCR4 on primed T lymphocytes in the aqueous humor of patients with uveitis. J Immunol, 172, 7154-61. DALAKAS, E., NEWSOME, P. N., HARRISON, D. J., et al. (2005) Hematopoietic stem cell trafficking in liver injury. Faseb J, 19, 1225-31. DASGUPTA, A. (2004) Targeting TFIIH to inhibit host cell transcription by Rift Valley Fever Virus. Mol Cell, 13, 456-8. DAUBNEY, R., HUDSON, J. R. & GARNHAM, P. C. (1931) Enzootic hepatitis or Rift Valley fever: an undescribed virus disease of sheep, cattle, and man from East Africa. Journal of Pathology, 34, 545-557. DAVIES, F. G. (1975) Observations on the epidemiology of Rift Valley fever in Kenya. J Hyg (Lond), 75, 219-30. DAVIES, F. G. & KARSTAD, L. (1981) Experimental infection of the African buffalo with the virus of Rift Valley fever. Trop Anim Health Prod, 13, 185-8. DE BOER, S. M., KORTEKAAS, J., ANTONIS, A. F., et al. (2010) Rift Valley fever virus subunit vaccines confer complete protection against a lethal virus challenge. Vaccine, 28, 2330-9. DENHARDT, D. T., NODA, M., O'REGAN, A. W., et al. (2001) Osteopontin as a means to cope with environmental insults: regulation of inflammation, tissue remodeling, and cell survival. J Clin Invest, 107, 1055-61. DO VALLE, T. Z., BILLECOCQ, A., GUILLEMOT, L., et al. (2010) A new mouse model reveals a critical role for host innate immunity in resistance to Rift Valley fever. J Immunol, 185, 6146- 56. DONG, C., LI, Q., LYU, S. C., et al. (2005) A novel apoptosis pathway activated by the carboxyl terminus of p21. Blood, 105, 1187-94. DROSTEN, C., GOTTIG, S., SCHILLING, S., et al. (2002) Rapid detection and quantification of RNA of Ebola and Marburg viruses, Lassa virus, Crimean-Congo hemorrhagic fever virus, Rift Valley fever virus, dengue virus, and yellow fever virus by real-time reverse transcription- PCR. J Clin Microbiol, 40, 2323-30. DUNGU, B., LOUW, I., LUBISI, A., et al. (2010) Evaluation of the efficacy and safety of the Rift Valley Fever Clone 13 vaccine in sheep. Vaccine, 28, 4581-7. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 226 P. Jansen van Vuren DURAND, C. A., HARTVIGSEN, K., FOGELSTRAND, L., et al. (2009) Phosphoinositide 3-kinase p110 delta regulates natural antibody production, marginal zone and B-1 B cell function, and autoantibody responses. J Immunol, 183, 5673-84. DUTTAGUPTA, P. A., BOESTEANU, A. C. & KATSIKIS, P. D. (2009) Costimulation signals for memory CD8+ T cells during viral infections. Crit Rev Immunol, 29, 469-86. EASTERDAY, B. C. (1965) Rift valley fever. Adv Vet Sci, 10, 65-127. EASTERDAY, B. C., MC, G. M., ROONEY, J. R., et al. (1962) The pathogenesis of Rift Valley fever in lambs. Am J Vet Res, 23, 470-9. EASTERDAY, B. C. & MURPHY, L. C. (1963) Studies on Rift Valley Fever in Laboratory Animals. Cornell Vet, 53, 423-33. EVANS, A., GAKUYA, F., PAWESKA, J. T., et al. (2008) Prevalence of antibodies against Rift Valley fever virus in Kenyan wildlife. Epidemiol Infect, 136, 1261-9. FAFETINE, J. M., TIJHAAR, E., PAWESKA, J. T., et al. (2007) Cloning and expression of Rift Valley fever virus nucleocapsid (N) protein and evaluation of a N-protein based indirect ELISA for the detection of specific IgG and IgM antibodies in domestic ruminants. Vet Microbiol, 121, 29-38. FINDLAY, G. M. (1932) The infectivity of Rift Valley fever for monkeys. Transcripts of the Royal Society of Tropical Medicine and Hygiene, 26, 161-168. FINDLAY, G. M. & HOWARD, E. M. (1951) Notes on Rift Valley fever. Arch Gesamte Virusforsch, 4, 411-23. FISCHER-TENHAGEN, C., HAMBLIN, C., QUANDT, S., et al. (2000) Serosurvey for selected infectious disease agents in free-ranging black and white rhinoceros in Africa. J Wildl Dis, 36, 316-23. FRAZIER, C. L. & SHOPE, R. E. (1979) Detection of antibodies to alphaviruses by enzyme-linked immunosorbent assay. J Clin Microbiol, 10, 583-5. FRY, T. J. & MACKALL, C. L. (2005) The many faces of IL-7: from lymphopoiesis to peripheral T cell maintenance. J Immunol, 174, 6571-6. GARCIA, S., CRANCE, J. M., BILLECOCQ, A., et al. (2001) Quantitative real-time PCR detection of Rift Valley fever virus and its application to evaluation of antiviral compounds. J Clin Microbiol, 39, 4456-61. GARTEL, A. L. & RADHAKRISHNAN, S. K. (2005) Lost in transcription: p21 repression, mechanisms, and consequences. Cancer Res, 65, 3980-5. GARTELL, A. L. & TYNER, A. L. (2002) The role of the cyclin-dependent kinase inhibitor p21 in apoptosis. Mol Cancer Ther, 1, 639-49. GAULIARD, N., BILLECOCQ, A., FLICK, R., et al. (2006) Rift Valley fever virus noncoding regions of L, M and S segments regulate RNA synthesis. Virology, 351, 170-9. GERDES, G. H. (2004) Rift Valley fever. Rev Sci Tech, 23, 613-23. GERRARD, S. R., BIRD, B. H., ALBARINO, C. G., et al. (2007) The NSm proteins of Rift Valley fever virus are dispensable for maturation, replication and infection. Virology, 359, 459-65. GIORGI, C., ACCARDI, L., NICOLETTI, L., et al. (1991) Sequences and coding strategies of the S RNAs of Toscana and Rift Valley fever viruses compared to those of Punta Toro, Sicilian Sandfly fever, and Uukuniemi viruses. Virology, 180, 738-53. GOLD, J. A., PARSEY, M., HOSHINO, Y., et al. (2003) CD40 contributes to lethality in acute sepsis: in vivo role for CD40 in innate immunity. Infect Immun, 71, 3521-8. GONZALEZ-SCARANO, F., ENDRES, M. J. & NATHANSON, N. (1991) Bunyaviridae: Pathogenesis. Curr Top Microbiol Immunol, 169, 217-49. GOODBURN, S., DIDCOCK, L. & RANDALL, R. E. (2000) Interferons: cell signalling, immune modulation, antiviral responses and virus countermeasures. Journal of General Virology, 81, 2341-2364. GORCHAKOV, R., VOLKOVA, E., YUN, N., et al. (2007) Comparative analysis of the alphavirus- based vectors expressing Rift Valley fever virus glycoproteins. Virology, 366, 212-25. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 227 P. Jansen van Vuren GORI SAVELLINI, G., DI GENOVA, G., TERROSI, C., et al. (2008) Immunization with Toscana virus N-Gc proteins protects mice against virus challenge. Virology, 375, 521-8. GRACIE, J. A., ROBERTSON, S. E. & MCINNES, I. B. (2003) Interleukin-18. J Leukoc Biol, 73, 213-24. GRAVELL, M., DORSETT, P. H., GUTENSON, O., et al. (1977) Detection of antibody to rubella virus by enzyme-linked immunosorbent assay. J Infect Dis, 136 Suppl, S300-3. GREINER, M. (1995) Two-graph receiver operating characteristics (TG-ROC)?a Microsoft-Excel template for the selection of cut-off values in diagnostic tests. Journal of Immunological Methods, 185, 145-146. GREINER, M. (1996) Two-graph receiver operating characteristics (TG-ROC): update version supports optimisation of cut-off values that minimize overall misclassification costs Journal of Immunological Methods, 191, 93-94. GRGACIC, E. V. & ANDERSON, D. A. (2006) Virus-like particles: passport to immune recognition. Methods, 40, 60-5. GUO, C. J., ATOCHINA-VASSERMAN, E. N., ABRAMOVA, E., et al. (2008) S-nitrosylation of surfactant protein-D controls inflammatory function. PLoS Biol, 6, e266. HABJAN, M., ANDERSSON, I., KLINGSTROM, J., et al. (2008a) Processing of genome 5' termini as a strategy of negative-strand RNA viruses to avoid RIG-I-dependent interferon induction. PLoS One, 3, e2032. HABJAN, M., PENSKI, N., SPIEGEL, M., et al. (2008b) T7 RNA polymerase-dependent and - independent systems for cDNA-based rescue of Rift Valley fever virus. J Gen Virol, 89, 2157- 66. HABJAN, M., PENSKI, N., WAGNER, V., et al. (2009a) Efficient production of Rift Valley fever virus-like particles: The antiviral protein MxA can inhibit primary transcription of bunyaviruses. Virology, 385, 400-8. HABJAN, M., PICHLMAIR, A., ELLIOTT, R. M., et al. (2009b) NSs protein of rift valley fever virus induces the specific degradation of the double-stranded RNA-dependent protein kinase. J Virol, 83, 4365-75. HALLER, O., KOCHS, G. & WEBER, F. (2006) The interferon response circuit: Induction and suppression by pathogenic viruses. Virology, 344, 119-130. HALLER, O. & WEBER, F. (2009) The interferon response circuit in antiviral host defense. Verh K Acad Geneeskd Belg, 71, 73-86. HARRINGTON, D. G., LUPTON, H. W., CRABBS, C. L., et al. (1980) Evaluation of a formalin- inactivated Rift Valley fever vaccine in sheep. Am J Vet Res, 41, 1559-64. HE, X., ZHU, Z., JOHNSON, C., et al. (2008) PIK3IP1, a negative regulator of PI3K, suppresses the development of hepatocellular carcinoma. Cancer Res, 68, 5591-8. HEISE, M. T., WHITMORE, A., THOMPSON, J., et al. (2009) An alphavirus replicon-derived candidate vaccine against Rift Valley fever virus. Epidemiol Infect, 137, 1309-18. HETTMANN, T. & LEIDEN, J. M. (2000) NF-?B is required for the positive selection of CD8+ thymocytes. J. Immun., 165, 5004-5010. HETTMANN, T., OPFERMAN, J. T., LEIDEN, J. M., et al. (2003) A critical role for NF-kappaB transcription factors in the development of CD8+ memory-phenotype T cells. Imm. Lett., 85, 297-300. HIGASHIHARA, M., KOYAMA, H. & IGARASHI, Y. (1972) Induction of virus-inhibiting factor or interferon in mice by strains with different virulence of Rift Valley fever virus. Kitasato Arch Exp Med, 45, 33-43. HOLMAN, D. H., PENN-NICHOLSON, A., WANG, D., et al. (2009) A complex adenovirus-vectored vaccine against Rift Valley fever virus protects mice against lethal infection in the presence of preexisting vector immunity. Clin Vaccine Immunol, 16, 1624-32. HOUSE, C., ALEXANDER, K. A., KAT, P. W., et al. (1996) Serum antibody to Rift Valley fever virus in African carnivores. Ann N Y Acad Sci, 791, 345-9. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 228 P. Jansen van Vuren HSU, D. H., DE WAAL MALEFYT, R., FIORENTINO, D. F., et al. (1990) Expression of interleukin- 10 activity by Epstein-Barr virus protein BCRF1. Science, 250, 830-2. HUNTER, P., ERASMUS, B. J. & VORSTER, J. H. (2002) Teratogenicity of a mutagenised Rift Valley fever virus (MVP 12) in sheep. Onderstepoort J Vet Res, 69, 95-8. IBRAHIM, M. S., TURELL, M. J., KNAUERT, F. K., et al. (1997) Detection of Rift Valley fever virus in mosquitoes by RT-PCR. Mol Cell Probes, 11, 49-53. IHARA, T., AKASHI, H. & BISHOP, D. H. (1984) Novel coding strategy (ambisense genomic RNA) revealed by sequence analyses of Punta Toro Phlebovirus S RNA. Virology, 136, 293-306. IKEGAMI, T., NARAYANAN, K., WON, S., et al. (2009) Dual functions of Rift Valley fever virus NSs protein: inhibition of host mRNA transcription and post-transcriptional downregulation of protein kinase PKR. Ann N Y Acad Sci, 1171 Suppl 1, E75-85. IKEGAMI, T., WON, S., PETERS, C. J., et al. (2005) Rift Valley fever virus NSs mRNA is transcribed from an incoming anti-viral-sense S RNA segment. J Virol, 79, 12106-11. IKEGAMI, T., WON, S., PETERS, C. J., et al. (2006) Rescue of infectious rift valley fever virus entirely from cDNA, analysis of virus lacking the NSs gene, and expression of a foreign gene. J Virol, 80, 2933-40. JACKSON, L. N., LARSON, S. D., SILVA, S. R., et al. (2008) PI3K/Akt activation is critical for early hepatic regeneration after partial hepatectomy. Am J Physiol Gastrointest Liver Physiol, 294, G1401-10. JANSEN VAN VUREN, P., POTGIETER, A. C., PAWESKA, J. T., et al. (2007) Preparation and evaluation of a recombinant Rift Valley fever virus N protein for the detection of IgG and IgM antibodies in humans and animals by indirect ELISA. J Virol Methods, 140, 106-14. JANSEN VAN VUREN, P., TIEMESSEN, C. T. & PAWESKA, J. T. (2010) Evaluation of a recombinant Rift Valley fever virus subunit nucleocapsid protein as an immunogen in mice and sheep. The Open Vaccine Journal, 3, 114-26. JANSSENS, S. & BEYAERT, R. (2003) Role of Toll-like receptors in pathogen recognition. Clin Microbiol Rev, 16, 637-46. JUPP, P. G., GROBBELAAR, A. A., LEMAN, P. A., et al. (2000) Experimental detection of Rift Valley fever virus by reverse transcription-polymerase chain reaction assay in large samples of mosquitoes. J Med Entomol, 37, 467-71. KAMAL, S. A. (2009) Pathological studies on postvaccinal reactions of Rift Valley fever in goats. Virol J, 6, 94. K?RBER, G. (1931) Beitrag zur kollectiven Behandlung pharmakologischer Reihensversuche. Arch. Exp. Path. Pharma., 162, 480?483. KASLER, H. G. & VERDIN, E. (2007) Histone deacetylase 7 functions as a key regulator of genes involved in both positive and negative selection of thymocytes. Mol Cell Biol, 27, 5184-200. KASMAR, A., VAN RHIJN, I. & MOODY, D. B. (2009) The evolved functions of CD1 during infection. Curr Opin Immunol, 21, 397-403. KENSIL, C. R. (1996) Saponins as vaccine adjuvants. Crit Rev Ther Drug Carrier Syst, 13, 1-55. KHAN, W. N. (2009) B cell receptor and BAFF receptor signaling regulation of B cell homeostasis. J Immunol, 183, 3561-7. KIEBALA, M., POLESSKAYA, O., YAO, Z., et al. (2010) Nuclear factor-kappa B family member RelB inhibits human immunodeficiency virus-1 Tat-induced tumor necrosis factor-alpha production. PLoS One, 5, e11875. KITCHEN, S. F. (1934) Laboratory infections with the virus of Rift Valley fever. American Journal of Tropical Medicine and Hygiene, 14, 547?564. KITTIPATARIN, C. & KHALED, A. R. (2007) Interlinking interleukin-7. Cytokine, 39, 75-83. KONG, M., FOX, C. J., MU, J., et al. (2004) The PP2A-associated protein alpha4 is an essential inhibitor of apoptosis. Science, 306, 695-8. KORTEKAAS, J., DE BOER, S. M., KANT, J., et al. (2010a) Rift Valley fever virus immunity provided by a paramyxovirus vaccine vector. Vaccine, 28, 4394-401. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 229 P. Jansen van Vuren KORTEKAAS, J., DEKKER, A., DE BOER, S. M., et al. (2010b) Intramuscular inoculation of calves with an experimental Newcastle disease virus-based vector vaccine elicits neutralizing antibodies against Rift Valley fever virus. Vaccine, 28, 2271-6. KUNISAKI, Y., TANAKA, Y., SANUI, T., et al. (2006) DOCK2 is required in T cell precursors for development of Valpha14 NK T cells. J Immunol, 176, 4640-5. LAGERQVIST, N., NASLUND, J., LUNDKVIST, A., et al. (2009) Characterisation of immune responses and protective efficacy in mice after immunisation with Rift Valley Fever virus cDNA constructs. Virol J, 6, 6. LAUTZE, J., MCCARTNEY, M., KIRSHEN, P., et al. (2007) Effect of a large dam on malaria risk: the Koka reservoir in Ethiopia. Trop Med Int Health, 12, 982-9. LE MAY, N., DUBAELE, S., PROIETTI DE SANTIS, L., et al. (2004) TFIIH transcription factor, a target for the Rift Valley hemorrhagic fever virus. Cell, 116, 541-50. LE MAY, N., GAULIARD, N., BILLECOCQ, A., et al. (2005) The N terminus of Rift Valley fever virus nucleoprotein is essential for dimerization. J Virol, 79, 11974-80. LE MAY, N., MANSUROGLU, Z., LEGER, P., et al. (2008) A SAP30 complex inhibits IFN-beta expression in Rift Valley fever virus infected cells. PLoS Pathog, 4, e13. LE ROUX, C. A., KUBO, T., GROBBELAAR, A. A., et al. (2009) Development and evaluation of a real-time reverse transcription-loop-mediated isothermal amplification assay for rapid detection of Rift Valley fever virus in clinical specimens. J Clin Microbiol, 47, 645-51. LEE, C. G., HARTL, D., MATSUURA, H., et al. (2008) Endogenous IL-11 signaling is essential in Th2- and IL-13-induced inflammation and mucus production. Am J Respir Cell Mol Biol, 39, 739-46. LETH-LARSEN, R., GARRED, P., JENSENIUS, H., et al. (2005) A common polymorphism in the SFTPD gene influences assembly, function, and concentration of surfactant protein D. J Immunol, 174, 1532-8. LEVINE, A. M., WHITSETT, J. A., HARTSHORN, K. L., et al. (2001) Surfactant protein D enhances clearance of influenza A virus from the lung in vivo. J Immunol, 167, 5868-73. LI, J., NISHIZAWA, K., AN, W., et al. (1998) A cdc15-like adaptor protein (CD2BP1) interacts with the CD2 cytoplasmic domain and regulates CD2-triggered adhesion. Embo J, 17, 7320-36. LI, Y., ZHOU, Z. & CHEN, C. (2008) WW domain-containing E3 ubiquitin protein ligase 1 targets p63 transcription factor for ubiquitin-mediated proteasomal degradation and regulates apoptosis. Cell Death Differ, 15, 1941-51. LINTHICUM, K. J., DAVIES, F. G., KAIRO, A., et al. (1985) Rift Valley fever virus (family Bunyaviridae, genus Phlebovirus). Isolations from Diptera collected during an inter-epizootic period in Kenya. J Hyg (Lond), 95, 197-209. LIU, L., CELMA, C. C. & ROY, P. (2008) Rift Valley fever virus structural proteins: expression, characterization and assembly of recombinant proteins. Virol J, 5, 82. LIVAK, K. J. & SCHMITTGEN, T. D. (2001) Analysis of relative gene expression data using real- time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 25, 402-8. LOMONOSOVA, E. & CHINNADURAI, G. (2008) BH3-only proteins in apoptosis and beyond: an overview. Oncogene, 27 Suppl 1, S2-19. LORENZO, G., MARTIN-FOLGAR, R., HEVIA, E., et al. (2010) Protection against lethal Rift Valley fever virus (RVFV) infection in transgenic IFNAR(-/-) mice induced by different DNA vaccination regimens. Vaccine, 28, 2937-44. LORENZO, G., MARTIN-FOLGAR, R., RODRIGUEZ, F., et al. (2008) Priming with DNA plasmids encoding the nucleocapsid protein and glycoprotein precursors from Rift Valley fever virus accelerates the immune responses induced by an attenuated vaccine in sheep. Vaccine, 26, 5255-62. LOWENSTEIN, C. J. & PADALKO, E. (2004) iNOS (NOS2) at a glance. J Cell Sci, 117, 2865-7. MAES, P., CLEMENT, J., CAUWE, B., et al. (2008) Truncated recombinant puumala virus nucleocapsid proteins protect mice against challenge in vivo. Viral Immunol, 21, 49-60. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 230 P. Jansen van Vuren MAES, P., KEYAERTS, E., BONNET, V., et al. (2006) Truncated recombinant Dobrava hantavirus nucleocapsid proteins induce strong, long-lasting immune responses in mice. Intervirology, 49, 253-60. MAGURANO, F. & NICOLETTI, L. (1999) Humoral response in Toscana virus acute neurologic disease investigated by viral-protein-specific immunoassays. Clin Diagn Lab Immunol, 6, 55- 60. MANDELL, R. B., KOUKUNTLA, R., MOGLER, L. J., et al. (2009) A replication-incompetent Rift Valley fever vaccine: chimeric virus-like particles protect mice and rats against lethal challenge. Virology, 397, 187-98. MANDELL, R. B., KOUKUNTLA, R., MOGLER, L. J., et al. (2010a) A replication-incompetent Rift Valley fever vaccine: chimeric virus-like particles protect mice and rats against lethal challenge. Virology, 397, 187-98. MANDELL, R. B., KOUKUNTLA, R., MOGLER, L. J., et al. (2010b) Novel suspension cell-based vaccine production systems for Rift Valley fever virus-like particles. J Virol Methods, 169, 259-68. MANNHALTER, J. W., NEYCHEV, H. O., ZLABINGER, G. J., et al. (1985) Modulation of the human immune response by the non-toxic and non-pyrogenic adjuvant aluminium hydroxide: effect on antigen uptake and antigen presentation. Clin. Exp. Immunol., 61, 143?151. MANSUROGLU, Z., JOSSE, T., GILLERON, J., et al. (2010) Nonstructural NSs protein of rift valley fever virus interacts with pericentromeric DNA sequences of the host cell, inducing chromosome cohesion and segregation defects. J Virol, 84, 928-39. MATSUI, H., HIKICHI, Y., TSUJI, I., et al. (2002) LIGHT, a member of the tumor necrosis factor ligand superfamily, prevents tumor necrosis factor-alpha-mediated human primary hepatocyte apoptosis, but not Fas-mediated apoptosis. J Biol Chem, 277, 50054-61. MAVOUNGOU, E., BOUYOU-AKOTET, M. K. & KREMSNER, P. G. (2005) Effects of prolactin and cortisol on natural killer (NK) cell surface expression and function of human natural cytotoxicity receptors (NKp46, NKp44 and NKp30). Clin Exp Immunol, 139, 287-96. MAYO, D. R. & BECKWITH, W. H. (2002) Inactivation of West Nile virus during serologic testing and transport. Journal of Clinical Microbiology, 40, 3044?3046. MAZZALI, M., KIPARI, T., OPHASCHAROENSUK, V., et al. (2002) Osteopontin--a molecule for all seasons. Qjm, 95, 3-13. MCELROY, A. K., ALBARINO, C. G. & NICHOL, S. T. (2009) Development of a RVFV ELISA that can distinguish infected from vaccinated animals. Virol J, 6, 125. MCINTOSH, B. M., DICKINSON, D. B. & DOS SANTOS, I. (1973) Rift Valley fever. 3. Viraemia in cattle and sheep. 4. The susceptibility of mice and hamsters in relation to transmission of virus by mosquitoes. J S Afr Vet Assoc, 44, 167-9. MEEGAN, J., LE GUENNO, B., KSIAZEK, T., et al. (1989) Rapid diagnosis of Rift Valley fever: a comparison of methods for the direct detection of viral antigen in human sera. Res Virol, 140, 59-65. MERONI, L., MILAZZO, L., MENZAGHI, B., et al. (2007) Altered expression of the tetraspanin CD81 on B and T lymphocytes during HIV-1 infection. Clin Exp Immunol, 147, 53-9. MILETIC, A. V., GRAHAM, D. B., SAKATA-SOGAWA, K., et al. (2009) Vav links the T cell antigen receptor to the actin cytoskeleton and T cell activation independently of intrinsic Guanine nucleotide exchange activity. PLoS One, 4, e6599. MOHAMED, M., MOSHA, F., MGHAMBA, J., et al. (2010) Epidemiologic and clinical aspects of a Rift Valley fever outbreak in humans in Tanzania, 2007. Am J Trop Med Hyg, 83, 22-7. MOMOSE, H., IMAI, J., HAMAGUCHI, I., et al. (2010) Induction of indistinguishable gene expression patterns in rats by Vero cell-derived and mouse brain-derived Japanese encephalitis vaccines. Jpn J Infect Dis, 63, 25-30. MORRILL, J. C., CARPENTER, L., TAYLOR, D., et al. (1991a) Further evaluation of a mutagen- attenuated Rift Valley fever vaccine in sheep. Vaccine, 9, 35-41. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 231 P. Jansen van Vuren MORRILL, J. C., CZARNIECKI, C. W. & PETERS, C. J. (1991b) Recombinant human interferon- gamma modulates Rift Valley fever virus infection in the rhesus monkey. J Interferon Res, 11, 297-304. MORRILL, J. C., JENNINGS, G. B., CAPLEN, H., et al. (1987) Pathogenicity and immunogenicity of a mutagen-attenuated Rift Valley fever virus immunogen in pregnant ewes. Am J Vet Res, 48, 1042-7. MORRILL, J. C., JENNINGS, G. B., JOHNSON, A. J., et al. (1990) Pathogenesis of Rift Valley fever in rhesus monkeys: role of interferon response. Arch Virol, 110, 195-212. MORRILL, J. C. & PETERS, C. J. (2003) Pathogenicity and neurovirulence of a mutagen-attenuated Rift Valley fever vaccine in rhesus monkeys. Vaccine, 21, 2994-3002. MULLER, R., ARGENTINI, C., BOULOY, M., et al. (1991) Completion of the genome sequence of Rift Valley fever phlebovirus indicates that the L RNA is negative sense and codes for a putative transcriptase-replicase [corrected]. Nucleic Acids Res, 19, 5433. MULLER, R., SALUZZO, J. F., LOPEZ, N., et al. (1995) Characterization of clone 13, a naturally attenuated avirulent isolate of Rift Valley fever virus, which is altered in the small segment. Am J Trop Med Hyg, 53, 405-11. MUNDEL, B. & GEAR, J. (1951) Rift valley fever; I. The occurrence of human cases in Johannesburg. S Afr Med J, 25, 797-800. NASCIMENTO, E. J., BRAGA-NETO, U., CALZAVARA-SILVA, C. E., et al. (2009) Gene expression profiling during early acute febrile stage of dengue infection can predict the disease outcome. PLoS One, 4, e7892. NASLUND, J., LAGERQVIST, N., HABJAN, M., et al. (2009) Vaccination with virus-like particles protects mice from lethal infection of Rift Valley Fever Virus. Virology, 385, 409-15. NGUKU, P. M., SHARIF, S. K., MUTONGA, D., et al. (2010) An investigation of a major outbreak of Rift Valley fever in Kenya: 2006-2007. Am J Trop Med Hyg, 83, 5-13. NIKLASSON, B., GRANDIEN, M., PETERS, C. J., et al. (1983) Detection of Rift Valley fever virus antigen by enzyme-linked immunosorbent assay. J Clin Microbiol, 17, 1026-31. NIKLASSON, B., PETERS, C. J., BENGTSSON, E., et al. (1985) Rift Valley fever virus vaccine trial: study of neutralizing antibody response in humans. Vaccine, 3, 123-7. NIKLASSON, B., PETERS, C. J., GRANDIEN, M., et al. (1984) Detection of human immunoglobulins G and M antibodies to Rift Valley fever virus by enzyme-linked immunosorbent assay. J Clin Microbiol, 19, 225-9. NIKLASSON, B. S. & GARGAN, T. P., 2ND (1985) Enzyme-linked immunosorbent assay for detection of Rift Valley fever virus antigen in mosquitoes. Am J Trop Med Hyg, 34, 400-5. NJENGA, M. K., PAWESKA, J., WANJALA, R., et al. (2009) Using a field quantitative real-time PCR test to rapidly identify highly viremic rift valley fever cases. J Clin Microbiol, 47, 1166- 71. NOAD, R. & ROY, P. (2003) Virus-like particles as immunogens. Trends Microbiol, 11, 438-44. NORSWORTHY, P. J., FOSSATI-JIMACK, L., CORTES-HERNANDEZ, J., et al. (2004) Murine CD93 (C1qRp) contributes to the removal of apoptotic cells in vivo but is not required for C1q-mediated enhancement of phagocytosis. J Immunol, 172, 3406-14. O'CONNOR, G. M., HART, O. M. & GARDINER, C. M. (2006) Putting the natural killer cell in its place. Immunology, 117, 1-10. O'HAGAN, D. T., MACKICHAN, M. L. & SINGH, M. (2001) Recent developments in adjuvants for vaccines against infectious diseases. Biomol Eng, 18, 69-85. O'HAGAN, D. T. & SINGH, M. (2003) Microparticles as vaccine adjuvants and delivery systems. Expert Rev Vaccines, 2, 269-83. OLALEYE, O. D., TOMORI, O., FAJIMI, J. L., et al. (1996) Experimental infection of three Nigerian breeds of sheep with the Zinga strain of the Rift Valley Fever virus. Rev Elev Med Vet Pays Trop, 49, 6-16. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 232 P. Jansen van Vuren PALMER, M. J., MAHAJAN, V. S., TRAJMAN, L. C., et al. (2008) Interleukin-7 receptor signaling network: an integrated systems perspective. Cell Mol Immunol, 5, 79-89. PAWESKA, J., BLUMBERG, L., WEYER, J., et al. (2008a) Rift Valley fever outbreak in South Africa. NICD-NHLS Communicable Diseases Surveillance Bulletin, 6, 1-2. PAWESKA, J. T., BARNARD, B. J. & WILLIAMS, R. (1995) The use of sucrose-acetone-extracted Rift Valley fever virus antigen derived from cell culture in an indirect enzyme-linked immunosorbent assay and haemagglutination-inhibition test. Onderstepoort J Vet Res, 62, 227- 33. PAWESKA, J. T., BURT, F. J., ANTHONY, F., et al. (2003a) IgG-sandwich and IgM-capture enzyme-linked immunosorbent assay for the detection of antibody to Rift Valley fever virus in domestic ruminants. J Virol Methods, 113, 103-12. PAWESKA, J. T., BURT, F. J. & SWANEPOEL, R. (2005a) Validation of IgG-sandwich and IgM- capture ELISA for the detection of antibody to Rift Valley fever virus in humans. J Virol Methods, 124, 173-81. PAWESKA, J. T., JANSEN VAN VUREN, P. & SWANEPOEL, R. (2007) Validation of an indirect ELISA based on a recombinant nucleocapsid protein of Rift Valley fever virus for the detection of IgG antibody in humans. J Virol Methods, 146, 119-24. PAWESKA, J. T., MORTIMER, E., LEMAN, P. A., et al. (2005b) An inhibition enzyme-linked immunosorbent assay for the detection of antibody to Rift Valley fever virus in humans, domestic and wild ruminants. J Virol Methods, 127, 10-8. PAWESKA, J. T., SMITH, S. J., WRIGHT, I. M., et al. (2003b) Indirect enzyme-linked immunosorbent assay for the detection of antibody against Rift Valley fever virus in domestic and wild ruminant sera. Onderstepoort J Vet Res, 70, 49-64. PAWESKA, J. T., VAN VUREN, P. J., KEMP, A., et al. (2008b) Recombinant nucleocapsid-based ELISA for detection of IgG antibody to Rift Valley fever virus in African buffalo. Vet Microbiol, 127, 21-8. PERRONE, L. A., NARAYANAN, K., WORTHY, M., et al. (2007) The S segment of Punta Toro virus (Bunyaviridae, Phlebovirus) is a major determinant of lethality in the Syrian hamster and codes for a type I interferon antagonist. J Virol, 81, 884-92. PETERS, C. J., ENNIS, W. H., TURELL, M. J., et al. (1989) Rapid detection of Rift Valley fever antigen in the serum of infected lambs. Res Virol, 140, 43-6. PETERS, C. J. & MEEGAN, J. M. (1981) Rift Valley fever. IN BERAN, G. (Ed.) CRC Handbook Series in Zoonosis. Boca Raton, Florida., CRC Press Inc. PETERS, C. J., REYNOLDS, J. A., SLONE, T. W., et al. (1986) Prophylaxis of Rift Valley fever with antiviral drugs, immune serum, an interferon inducer, and a macrophage activator. Antiviral Res, 6, 285-97. PETERS, C. J. & SLONE, T. W. (1982) Inbred rat strains mimic the disparate human response to Rift Valley fever virus infection. J Med Virol, 10, 45-54. PETERSS, A. L., STUNZ, L. L. & BISHOP, G. A. (2009) CD40 and autoimmunity: the dark side of a great activator. Semin Immunol, 21, 293-300. PEYREFITTE, C. N., BOUBIS, L., COUDRIER, D., et al. (2008) Real-time reverse-transcription loop-mediated isothermal amplification for rapid detection of rift valley Fever virus. J Clin Microbiol, 46, 3653-9. PFAFFL, M. W. (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res, 29, e45. PFAFFL, M. W., HORGAN, G. W. & DEMPFLE, L. (2002) Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res, 30, e36. PICHLMAIR, A., HABJAN, M., UNGER, H., et al. (2010) Virus-like particles expressing the nucleocapsid gene as an efficient vaccine against Rift Valley fever virus. Vector Borne Zoonotic Dis, 10, 701-3. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 233 P. Jansen van Vuren PITTMAN, P. R., LIU, C. T., CANNON, T. L., et al. (1999) Immunogenicity of an inactivated Rift Valley fever vaccine in humans: a 12-year experience. Vaccine, 18, 181-9. POURRUT, X., NKOGHE, D., SOURIS, M., et al. (2010) Rift Valley fever virus seroprevalence in human rural populations of Gabon. PLoS Negl Trop Dis, 4, e763. PRINCE, J. M., MING, M. J., LEVY, R. M., et al. (2007) Early growth response 1 mediates the systemic and hepatic inflammatory response initiated by hemorrhagic shock. Shock, 27, 157- 64. PRITCHARD, M. T., ROYCHOWDHURY, S., MCMULLEN, M. R., et al. (2007) Early growth response-1 contributes to galactosamine/lipopolysaccharide-induced acute liver injury in mice. Am J Physiol Gastrointest Liver Physiol, 293, G1124-33. QIAO, G., LEI, M., LI, Z., et al. (2007) Negative regulation of CD40-mediated B cell responses by E3 ubiquitin ligase Casitas-B-lineage lymphoma protein-B. J Immunol, 179, 4473-9. QIAO, G., LI, Z., MOLINERO, L., et al. (2008) T-cell receptor-induced NF-kappaB activation is negatively regulated by E3 ubiquitin ligase Cbl-b. Mol Cell Biol, 28, 2470-80. RANDALL, R., BINN, L. N. & HARRISON, V. R. (1964) Immunization against Rift Valley Fever Virus. Studies on the Immunogenicity of Lyophilized Formalin-Inactivated Vaccine. J Immunol, 93, 293-9. RANDALL, R., GIBBS, C. J., JR., AULISIO, C. G., et al. (1962) The development of a formalin- killed Rift Valley fever virus vaccine for use in man. J Immunol, 89, 660-71. READING, S. A. & DIMMOCK, N. J. (2007) Neutralization of animal virus infectivity by antibody. Arch Virol, 152, 1047?1059. RENUKARADHYA, G. J., WEBB, T. J., KHAN, M. A., et al. (2005) Virus-induced inhibition of CD1d1-mediated antigen presentation: reciprocal regulation by p38 and ERK. J Immunol, 175, 4301-8. RICE, R. M., ERLICK, B. J., ROSATO, R. R., et al. (1980) Biochemical characterization of Rift Valley fever virus. Virology, 105, 256-60. RITTER, M., BOULOY, M., VIALAT, P., et al. (2000) Resistance to Rift Valley fever virus in Rattus norvegicus: genetic variability within certain 'inbred' strains. J Gen Virol, 81, 2683-8. SALL, A. A., DE, A. Z. P. M., ZELLER, H. G., et al. (1997) Variability of the NS(S) protein among Rift Valley fever virus isolates. J Gen Virol, 78 ( Pt 11), 2853-8. SALL, A. A., MACONDO, E. A., SENE, O. K., et al. (2002) Use of reverse transcriptase PCR in early diagnosis of Rift Valley fever. Clin Diagn Lab Immunol, 9, 713-5. SALL, A. A., THONNON, J., SENE, O. K., et al. (2001) Single-tube and nested reverse transcriptase- polymerase chain reaction for detection of Rift Valley fever virus in human and animal sera. J Virol Methods, 91, 85-92. SANDROCK, M., FRESE, M., HALLER, O., et al. (2001) Interferon-induced rat Mx proteins confer resistance to Rift Valley fever virus and other arthropod-borne viruses. J Interferon Cytokine Res, 21, 663-8. SANG, R., KIOKO, E., LUTOMIAH, J., et al. (2010) Rift Valley fever virus epidemic in Kenya, 2006/2007: the entomologic investigations. Am J Trop Med Hyg, 83, 28-37. SCHMALJOHN, C. S., CHU, Y. K., SCHMALJOHN, A. L., et al. (1990) Antigenic subunits of Hantaan virus expressed by baculovirus and vaccinia virus recombinants. J Virol, 64, 3162-70. SCHMALJOHN, C. S., PARKER, M. D., ENNIS, W. H., et al. (1989) Baculovirus expression of the M genome segment of Rift Valley fever virus and examination of antigenic and immunogenic properties of the expressed proteins. Virology, 170, 184-92. SCHMITTGEN, T. D. & LIVAK, K. J. (2008) Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc, 3, 1101-8. SCHWARZ, T. F. (1996) Imported vector- and rodent-borne virus infections--an introduction. Arch Virol Suppl, 11, 3-11. SCHWARZ, T. F., GILCH, S., PAULI, C., et al. (1996) Immunoblot detection of antibodies to Toscana virus. J Med Virol, 49, 83-6. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 234 P. Jansen van Vuren SCOTT, G. R., WEDDELL, W. & REID, D. (1956) Preliminary findings on the prevelance of Rift Valley fever in Kenyan cattle. Bulletin of Epizootic Diseases of Africa, 4, 17-25. SEKI, Y., HAYASHI, K., MATSUMOTO, A., et al. (2002) Expression of the suppressor of cytokine signaling-5 (SOCS5) negatively regulates IL-4-dependent STAT6 activation and Th2 differentiation. Proc Natl Acad Sci U S A, 99, 13003-8. SENALDI, G., STOLINA, M., GUO, J., et al. (2002) Regulatory effects of novel neurotrophin-1/b cell-stimulating factor-3 (cardiotrophin-like cytokine) on B cell function. J Immunol, 168, 5690-8. SENALDI, G., VARNUM, B. C., SARMIENTO, U., et al. (1999) Novel neurotrophin-1/B cell- stimulating factor-3: a cytokine of the IL-6 family. Proc Natl Acad Sci U S A, 96, 11458-63. SHIEH, W. J., PADDOCK, C. D., LEDERMAN, E., et al. (2010) Pathologic studies on suspect animal and human cases of Rift Valley fever from an outbreak in Eastern Africa, 2006-2007. Am J Trop Med Hyg, 83, 38-42. SMITH, D. R., STEELE, K. E., SHAMBLIN, J., et al. (2010) The pathogenesis of Rift Valley fever virus in the mouse model. Virology, doi:10.106/j.virol.2010.08.016. SMITHBURN, K. C. (1949) Rift Valley fever; the neurotropic adaptation of the virus and the experimental use of this modified virus as a vaccine. Br J Exp Pathol, 30, 1-16. SMITHBURN, K. C., HADDOW, A. J. & GILLETT, J. D. (1948) Rift Valley fever; isolation of the virus from wild mosquitoes. Br J Exp Pathol, 29, 107-21. SMITHBURN, K. C., HADDOW, A. J. & LUMSDEN, W. H. (1949a) Rift Valley fever; transmission of the virus by mosquitoes. Br J Exp Pathol, 30, 35-47. SMITHBURN, K. C., MAHAFFY, A. F. & ET AL. (1949b) Rift Valley fever; accidental infections among laboratory workers. J Immunol, 62, 213-27. SOI, R. K., RURANGIRWA, F. R., MCGUIRE, T. C., et al. (2010) Protection of sheep against Rift Valley fever virus and sheep poxvirus with a recombinant capripoxvirus vaccine. Clin Vaccine Immunol, 17, 1842-9. SOPER, D. S. (2009) The free statistics calculators website. http://www.danielsoper.com/statcalc/. SPENCER, J. V. (2007) The cytomegalovirus homolog of interleukin-10 requires phosphatidylinositol 3-kinase activity for inhibition of cytokine synthesis in monocytes. J Virol, 81, 2083-6. SPIK, K., SHURTLEFF, A., MCELROY, A. K., et al. (2006) Immunogenicity of combination DNA vaccines for Rift Valley fever virus, tick-borne encephalitis virus, Hantaan virus, and Crimean Congo hemorrhagic fever virus. Vaccine, 24, 4657-66. STRAUSS, J. H. & STRAUSS, E. G. (2008) Viruses and Human Disease, 2nd Edition, Oxford, Elsevier. SUZICH, J. A., KAKACH, L. T. & COLLETT, M. S. (1990) Expression strategy of a phlebovirus: biogenesis of proteins from the Rift Valley fever virus M segment. J Virol, 64, 1549-55. SWANEPOEL, R. (1976) Studies on the epidemiology of Rift Valley Fever. J S Afr Vet Assoc, 47, 93- 4. SWANEPOEL, R. & BLACKBURN, N. K. (1977) Demonstration of nuclear immunofluorescence in Rift Valley fever infected cells. J Gen Virol, 34, 557-61. SWANEPOEL, R. & COETZER, J. A. W. (2004) Rift Valley fever. IN COETZER, J. A. W., TUSTIN, R.C. (Ed.) Infectious Diseases of Livestock. Cape Town, Oxford University Press. SWANEPOEL, R., STRUTHERS, J. K., ERASMUS, M. J., et al. (1986a) Comparison of techniques for demonstrating antibodies to Rift Valley fever virus. J Hyg (Lond), 97, 317-29. SWANEPOEL, R., STRUTHERS, J. K., ERASMUS, M. J., et al. (1986b) Comparative pathogenicity and antigenic cross-reactivity of Rift Valley fever and other African phleboviruses in sheep. J Hyg (Lond), 97, 331-46. TAKEHARA, K., MIN, M. K., BATTLES, J. K., et al. (1989) Identification of mutations in the M RNA of a candidate vaccine strain of Rift Valley fever virus. Virology, 169, 452-7. TURELL, M. J. & BAILEY, C. L. (1987) Transmission studies in mosquitoes (Diptera: Culicidae) with disseminated Rift Valley fever virus infections. J Med Entomol, 24, 11-8. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 235 P. Jansen van Vuren TURELL, M. J., DOHM, D. J., MORES, C. N., et al. (2008a) Potential for North American mosquitoes to transmit Rift Valley fever virus. J Am Mosq Control Assoc, 24, 502-7. TURELL, M. J., LEE, J. S., RICHARDSON, J. H., et al. (2007) Vector competence of Kenyan Culex zombaensis and Culex quinquefasciatus mosquitoes for Rift Valley fever virus. J Am Mosq Control Assoc, 23, 378-82. TURELL, M. J., LINTHICUM, K. J. & BEAMAN, J. R. (1990) Transmission of Rift Valley fever virus by adult mosquitoes after ingestion of virus as larvae. Am J Trop Med Hyg, 43, 677-80. TURELL, M. J., LINTHICUM, K. J., PATRICAN, L. A., et al. (2008b) Vector competence of selected African mosquito (Diptera: Culicidae) species for Rift Valley fever virus. J Med Entomol, 45, 102-8. UBOL, S., PHUKLIA, W., KALAYANAROOJ, S., et al. (2010) Mechanisms of immune evasion induced by a complex of dengue virus and preexisting enhancing antibodies. J Infect Dis, 201, 923-35. ULANOVA, M., TARKOWSKI, A., HAHN-ZORIC, M., et al. (2001) The common vaccine adjuvant aluminum hydroxide up-regulates accessory properties of human monocytes via an interleukin- 4-dependent mechanism. Inf. Imm., 69, 1151-1159. VAN PUTTEN, S. M., WUBBEN, M., HENNINK, W. E., et al. (2009) The downmodulation of the foreign body reaction by cytomegalovirus encoded interleukin-10. Biomaterials, 30, 730-5. VAPALAHTI, O., KALLIO-KOKKO, H., NARVANEN, A., et al. (1995) Human B-cell epitopes of Puumala virus nucleocapsid protein, the major antigen in early serological response. Journal of Medical Virology, 46, 293?303. VENTER, M., MYERS, T. G., WILSON, M. A., et al. (2005) Gene expression in mice infected with West Nile virus strains of different neurovirulence. Virology, 342, 119-40. VIALAT, P., BILLECOCQ, A., KOHL, A., et al. (2000) The S segment of rift valley fever phlebovirus (Bunyaviridae) carries determinants for attenuation and virulence in mice. J Virol, 74, 1538- 43. VIALAT, P., MULLER, R., VU, T. H., et al. (1997) Mapping of the mutations present in the genome of the Rift Valley fever virus attenuated MP12 strain and their putative role in attenuation. Virus Res, 52, 43-50. WALLACE, D. B., ELLIS, C. E., ESPACH, A., et al. (2006) Protective immune responses induced by different recombinant vaccine regimes to Rift Valley fever. Vaccine, 24, 7181-9. WALLACE, D. B. & VILJOEN, G. J. (2005) Immune responses to recombinants of the South African vaccine strain of lumpy skin disease virus generated by using thymidine kinase gene insertion. Vaccine, 23, 3061-7. WEINER, L. M. & ADAMS, G. P. (2000) New approaches to antibody therapy. Oncogene, 19, 6144- 51. WOLF, M. C., FREIBERG, A. N., ZHANG, T., et al. (2010) A broad-spectrum antiviral targeting entry of enveloped viruses. Proc Natl Acad Sci U S A, 107, 3157-62. WON, S., IKEGAMI, T., PETERS, C. J., et al. (2006) NSm and 78-kilodalton proteins of Rift Valley fever virus are nonessential for viral replication in cell culture. J Virol, 80, 8274-8. WON, S., IKEGAMI, T., PETERS, C. J., et al. (2007) NSm protein of Rift Valley fever virus suppresses virus-induced apoptosis. J Virol, 81, 13335-45. WOODS, C. W., KARPATI, A. M., GREIN, T., et al. (2002) An outbreak of Rift Valley fever in Northeastern Kenya, 1997-98. Emerg Infect Dis, 8, 138-44. WRIGHT, P. F., NILSSON, E., VAN ROOIJ, E. M. A., et al. (1993) Standardisation and validation of enzyme-linked immunosorbent assay techniques for the detection of antibody in infectious disease diagnosis. Revue Scientifique Et Technique - Office International Des Epizooties, 12, 435-450. XU, F., LIU, D., NUNES, M. R., et al. (2007) Antigenic and genetic relationships among Rift Valley fever virus and other selected members of the genus Phlebovirus (Bunyaviridae). Am J Trop Med Hyg, 76, 1194-200. Evaluation of a recombinant Rift Valley fever virus nucleocapsid protein as a vaccine and an immunodiagnostic reagent. Page 236 P. Jansen van Vuren YADANI, F. Z., KOHL, A., PREHAUD, C., et al. (1999) The carboxy-terminal acidic domain of Rift Valley Fever virus NSs protein is essential for the formation of filamentous structures but not for the nuclear localization of the protein. J Virol, 73, 5018-25. YEDLOUTSCHNIG, R. J., DARDIRI, A. H., MEBUS, C. A., et al. (1981) Abortion in vaccinated sheep and cattle after challenge with Rift Valley fever virus. Vet Rec, 109, 383-4. YU-LEE, L. Y. (2002) Prolactin modulation of immune and inflammatory responses. Recent Prog Horm Res, 57, 435-55. ZAKI, A., COUDRIER, D., YOUSEF, A. I., et al. (2006) Production of monoclonal antibodies against Rift Valley fever virus Application for rapid diagnosis tests (virus detection and ELISA) in human sera. J Virol Methods, 131, 34-40. ZELLER, H. G., BESSIN, R., THIONGANE, Y., et al. (1995) Rift Valley fever antibody prevalence in domestic ungulates in Cameroon and several west African countries (1989-1992) following the 1987 Mauritanian outbreak. Res Virol, 146, 81-5. ZHANG, R., ZHANG, N. & MUELLER, D. L. (2008) Casitas B-lineage lymphoma b inhibits antigen recognition and slows cell cycle progression at late times during CD4+ T cell clonal expansion. J Immunol, 181, 5331-9. ZHOU, L., NAZARIAN, A. A. & SMALE, S. T. (2004) Interleukin-10 inhibits interleukin-12 p40 gene transcription by targeting a late event in the activation pathway. Mol Cell Biol, 24, 2385- 96. ZHU, R., DIEM, S., ARAUJO, L. M., et al. (2007) The Pro-Th1 cytokine IL-12 enhances IL-4 production by invariant NKT cells: relevance for T cell-mediated hepatitis. J Immunol, 178, 5435-42.