Characterisation of non-polio enteroviruses identified in disease biomes in South Africa

Abstract

Human enteroviruses (family Picornaviridae) consist of 106 serotypes and are divided into four species: Human enterovirus (HEV)–A, B, C, and D. Enteroviruses cause a variety of clinical symptoms from severe (e.g. acute flaccid paralysis) to less severe (e.g. hand-foot-and-mouth disease). Whilst there is currently no antiviral treatment, viral genotyping allows for: identification of increased virulence, identification of new enteroviruses, correlation of virus types with immunity, epidemiological investigations and provides information on viral inter-relationships. A comprehensive study is underway to determine the prevalence and type of circulating non-polio enteroviruses in South Africa, specifically for those involved in recent outbreaks. This study investigated the prevalence of non-polio enteroviruses circulating in South African between 2010 and 2012 using samples obtained from 2 national surveillance programs conducted at the National Institute for Communicable Diseases: Acute Flaccid Paralysis (AFP) and Rotavirus. Typing was performed using a Real-Time PCR (RT-PCR) assay, followed by Sanger sequencing. 832 samples were tested to date (562 from the Rotavirus and 270 from the AFP surveillance programs, respectively). 446 positive enterovirus samples were detected from which 308 samples were successfully sequenced. Specimens from the AFP program yielded mostly HEV-B serotypes (90.40%), whereas samples typed directly from the Rotavirus program stools yielded mostly HEV-C serotypes (47.20%). 92.8% of typed samples were from patients under 5 years. Despite most detections being HEV-B (56.55%), the most commonly detected virus was Enterovirus 99 (8.63%) from the HEV-C species. RT-PCR and sequencing, whilst more expensive, have proven more accurate than cell culture and neutralization assays for typing enteroviruses. In South Africa, HEV-B viruses were predominant, and in comparison to studies from other countries, a larger proportion of HEV-C viruses were detected. Detecting EV directly from stool yielded a larger diversity of the viruses, and while disease associated viruses were detected, they did not contribute significantly to the associated disease burden.