Host factors and broadly neutralizing antibodies in South African women infected with HIV-1 subtype C

Date
2016-02-22
Authors
Scheepers, Cathrine
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Broadly neutralizing antibodies are capable of neutralizing a large number of HIV-1 strains and have shown to be protective against infection in non-human primate models. These antibodies are likely to play an important role in an effective vaccine against HIV. Eliciting them by vaccination has thus far been unsuccessful and their unusual features such as long CDHR3 lengths and high levels of somatic hypermutation make this a particularly challenging task. Approximately 15-30% of chronically HIV-1 infected individuals develop these types of antibodies but an understanding of the underlying mechanisms is limited. The aim of this study, therefore, was to investigate host factors associated with the development of broadly neutralizing antibodies in HIV-1 subtype C infected individuals. In particular we analysed genetic variation of the genes encoding the variable region of antibodies, the evolution of an HIV-1 specific antibody lineage and glycan-binding profiles of serum antibodies. The human heavy chain variable region genes (IGHV) are the largest and most variable of all human immunoglobulin genes and encode the major antigen-binding region. These genes are divided into seven subgroups, each subgroup contains numerous genes and alleles. Using genomic DNA from 28 HIV-1 subtype C infected individuals we performed next generation sequencing using both Illumina MiSeq and Roche 454 technologies. Included were 13 individuals who developed broadly neutralizing antibodies, 13 who did not despite chronic HIV-1 infection and two intermediate neutralizers. We found no genetic differences in the IGHV genes between these two groups. However, we identified 85 novel alleles and 38 alleles that had previously only been observed in rearranged antibody sequences. Of these alleles, eight were used by functional antibodies, two of which were HIV-1 specific. This study highlights the importance of a fully comprehensive database for inferring germline gene usage and the unmutated common ancestors of antibody lineages. In addition it showed that everyone has the same genetic potential of developing broadly neutralizing antibodies, which has positive implications for vaccine development. A number of studies have demonstrated the importance of strain-specific antibodies in the development of broadly neutralizing antibodies. In addition to being the forerunners of broadly neutralizing antibodies, strain-specific antibodies can help shape the viral populations that elicit different broadly neutralizing antibody lineages. We therefore studied the evolution of a strain-specific HIV antibody lineage (CAP88-CH06) in an individual who failed to develop neutralization breadth even after 5 years of infection, to understand why some strain-specific antibodies remain limited to autologous viruses. CAP88-CH06 was previously isolated as an IgA1 antibody using the IGHV4-39 gene with 8.8% divergence from donor CAP88 at 34 weeks post-infection, which mapped to the C3- V4 region of gp120. IgA and IgG antibodies using the same germline IGHV were sequenced on an Illumina MiSeq from 5 to 121 weeks post-infection. IgA sequences identical to that of the fully matured antibody with 8.8% divergence were detected from early infection and throughout until after 2 years, well after viral escape. This was consistent with plasma neutralization of the C3-V4 region within CAP88. However, very little evidence of evolution was seen within the IgA sequences. A group of related IgG sequences were also identified between 11 and 34 weeks but not at other time-points. Interestingly, within the 11 week transcripts we identified an identical sequence as both an IgA and IgG isotype, which likely gave rise to these transient IgG antibodies. The lack of neutralization breadth in this individual could therefore be the result of both limited evolution of the IgA isotype and well as the disappearance of the IgG isotype. The HIV envelope is surrounded by glycans, known as the glycan shield. These glycans contribute towards the structural integrity of the envelope and serve as protection against immune responses to conserved regions. However, glycans often form targets for broadly neutralizing antibodies. Thus we studied the glycan-binding profiles of HIVnegative and HIV-positive individuals (including 12 individuals who develop broadly neutralizing antibodies and 13 who did not despite chronic infection) to determine whether glycan-binding was specific to individuals who develop broadly neutralizing antibodies. Longitudinal samples were taken yearly for three years from all 47 individuals and their serum IgG levels were tested on glycan microarrays. We observed fluctuations in glycanbinding over time within the HIV-negative individuals and these were used to establish baseline values. The HIV-positive individuals were found to have elevated levels of antibodies targeting high mannose N-linked glycans, Tn-peptides and glycolipids during infection. Binding to Tn-peptides and glycolipids were elevated throughout infection, whereas high mannose N-linked glycans were elevated from 2-3 years post-infection. We observed no differences in these glycans between the individuals who developed broadly neutralizing antibodies compared to those who did not despite chronic HIV infection. This data suggests that the elevated levels of glycan-binding serum antibodies were a consequence of infection rather than specific to broadly neutralizing antibodies. Since glycan-binding antibodies against Tn-peptides and glycolipids were detected earlier than high mannose N-linked glycans and antibodies targeting these glycans were elevated during infection, they might warrant further investigation with respect to immunogen design. Collectively this study has contributed to a greater understanding of the role of various host factors in the development of broadly neutralizing antibodies to HIV. This includes showing that there were no differences in the IGHV genes between individuals who did and did not develop broadly neutralizing antibodies as well as providing a wealth of new data on human antibody genes that will have benefits beyond the field of HIV. Furthermore our study has reinforced the essential role of somatic hypermutation in developing neutralization breadth and the need for further co-evolution studies on strainspecific lineages to understand this roadblock. Finally our study using glycan arrays has highlighted that glycan-binding antibodies are induced in all HIV-infected individuals even though only a minority go on to develop broadly neutralizing antibodies. Overall these data suggest that all humans have the ability to develop broadly neutralizing antibodies but a vaccine capable of eliciting such protective responses remains a major hurdle.
Description
A thesis submitted to the Faculty of Health Sciences, School of Pathology, University of the Witwatersrand in fulfilment of the degree of Doctor of Philosophy (PhD) Johannesburg, September 2015
Keywords
Citation
Collections