Rotavirus strain diversity in Eastern and Southern African countries before and after vaccine introduction

BackgroundThe African Rotavirus Surveillance Network has been detecting and documenting rotavirus genotypes in the African sub-continent since 1998 in anticipation of the rollout of rotavirus vaccination in routine Expanded Programme on Immunisation. This paper reports distribution of the rotavirus strains circulating in 15 Eastern and Southern African (ESA) countries from 2010-2015 as part of active World Health Organization (WHO) rotavirus surveillance, and investigates possibility of emergence of non-vaccine or unusual strains in six selected countries post-vaccine introduction.Material and methodsStool samples were collected from children <5 years of age presenting with acute gastroenteritis at sentinel hospitals pre- and post-rotavirus vaccine introduction. Samples were tested for group A rotavirus using an enzyme immunoassay by the national and sentinel laboratories. At the WHO Rotavirus Regional Reference Laboratory in South Africa, molecular characterisation was determined by PAGE (n = 4186), G and P genotyping (n = 6447) and DNA sequencing for both G and P types (n = 400).ResultsThe six-year surveillance period demonstrated that 23.8% of the strains were G1P[8], followed by G2P[4] (11.8%), G9P[8] (10.4%), G12P[8] (4.9%), G2P[6] (4.2%) and G3P[6] (3.7%) in 15 ESA countries. There was no difference in circulating strains pre- and post-rotavirus vaccine introduction with yearly fluctuation of strains observed over time. Atypical rotavirus G and P combinations (such as G1P[4], G2P[8], G9P[4] and G12P[4]) that might have arisen through inter-genogroup or inter-genotypes reassortment were detected at low frequency (2%). Close genetic relationship of African strains were reflected on the phylogenetic analysis, strains segregated together to form an African cluster in the same lineages/sub-lineage or monophyletic branch.ConclusionThere has been considerable concern about strain replacement post-vaccine introduction, it was not clear at this early stage whether observed cyclical changes of rotavirus strains were due to vaccine pressure or this was just part of natural annual fluctuations in the six ESA countries, long-term surveillance is required.     

Impact of rotavirus vaccine introduction and genotypic characteristics of rotavirus strains in children less than 5 years of age with gastroenteritis in Ethiopia: 2011–2016

IntroductionA monovalent rotavirus vaccine was introduced in the Ethiopian Expanded Program on Immunization from November 2013. We compared impact of rotavirus vaccine introduction on rotavirus associated acute diarrhea hospitalizations and genotypic characteristics of rotavirus strains pre-and post-vaccine introduction.MethodsSentinel surveillance for diarrhea among children <5 years of age was conducted at 3 hospitals in Addis Ababa, Ethiopia from 2011 to 2017. Stool specimens were collected from enrolled children and tested using an antigen capture enzyme immunoassay. Rotavirus positive samples (156 from pre- and 141 from post-vaccination periods) were further characterized by rotavirus genotyping methods to identify the predominant G and P types circulating during the surveillance era.ResultsA total of 788 children were enrolled during the pre- (July 2011-June 2013) and 815 children during the post-vaccination (July 2014-June 2017) periods. The proportion of diarrhea hospitalizations due to rotavirus among children <5 years of age declined by 17% from 24% (188/788) in the pre-vaccine period and to 20% (161/185) in post-vaccine introduction era. Similarly, a reduction of 18% in proportion of diarrhea hospitalizations due to rotavirus in children <12 months of age in the post (27%) vs pre-vaccine (33%) periods was observed. Seasonal peaks of rotavirus declined following rotavirus vaccine introduction. The most prevalent circulating strains were G12P[8] in 2011 (36%) and in 2012 (27%), G2P[4] (35%) in 2013, G9P[8] (19%) in 2014, G3P[6] and G2P[4] (19% each) in 2015, and G3P[8] (29%) in 2016.DiscussionFollowing rotavirus vaccine introduction in Ethiopia, a reduction in rotavirus associated hospitalizations was seen in all age groups with the greatest burden in children <12 months of age. A wide variety of rotavirus strains circulated in the pre- and post-vaccine introduction periods.     

Rotavirus strains circulating in Africa during 1996-1999: emergence of G9 strains and P[6] strains

Rotavirus infection is associated with 150000-200000 deaths annually in Africa. Although the withdrawal of the RotaShield vaccine has been a major setback in rotavirus vaccine development, new vaccine candidates are under development and approaching phase II and III trials. Before these trials could be conducted in Africa, a comprehensive survey of the circulating VP7 serotypes and VP4 genotypes is required. During the past 3 years, over 3000 rotavirus-positive specimens from several African countries have been analysed. RT-PCR techniques for the VP7 and VP4 genotypes and by monoclonal antibodies to the VP6 subgroup and VP7 serotype have been performed. Almost 75% of the strains were typed by the VP7 monoclonal antibodies or RT-PCR. VP4 genotyping was done in approximately half of these strains. The predominant strains circulating across Africa during 1996-1999 were P[6]G1 and P[6]G3 strains. Geographic differences were noted and West Africa displayed the most diverse strains with G3/8 and G1/3 "mosaic" viruses occurring commonly. G9 strains were identified in several countries indicating that the strain is emerging in Africa too. G9 was the predominant strain in certain countries during 1999. The circulating types observed will have implications for the new rotavirus vaccine candidates.     

Molecular characterization of rotavirus strains from pre- and post-vaccination periods in a country with low vaccination coverage: The case of Slovenia

Rotavirus vaccination started in Slovenia in 2007 on a voluntarily basis. The vaccination rate is relatively low (up to 27%) and no increasing trend is observed. We present rotavirus genotype distribution among children hospitalized for rotavirus gastroenteritis in Slovenia. Eight consecutive rotavirus seasons were followed, from 2005/06 to 2012/13, and 113 strains of the most common rotavirus genotypes were randomly selected for molecular characterization of rotavirus VP7 and VP4 (VP8¹) genome segments. During the vaccine introduction period, from 2007 to 2013, rotavirus genotype prevalences changed, with G1P[8] decreasing from 74.1% to 8.7% between 2007/08 and 2010/11 seasons, replaced by G4P[8] and G2P[4], with up to 52.0% prevalence. Comparable analysis of VP7 and VP8¹ genome fragments within G1P[8] genotype lineages revealed considerable differences for rotavirus strains circulating before and during the vaccination period. The G1P[8] rotavirus strains from the pre-vaccination period clustered in a phylogenetic tree within Rotarix®-like VP7 and VP8¹ lineages. However, since 2007, the majority of G1P[8] strains have shifted to distant genetic lineages with lower nucleotide (88.1–94.0% for VP7 and 86.6–91.1% for VP8¹) and amino acid (93.8–95.2% for VP7 and 85.3–94.6% for VP8¹) identities to the vaccine Rotarix® strain. This change also resulted in a different deduced amino acid profile at the major VP7 and VP8¹ antigenic epitopes.     

Genotypic linkages of VP4, VP6, VP7, NSP4, NSP5 genes of rotaviruses circulating among children with acute gastroenteritis in Thailand

Rotavirus is the main cause of acute viral gastroenteritis in infants and young children worldwide. Surveillance of group A rotavirus has been conducted in Chiang Mai, Thailand since 1987 up to 2004 and those studies revealed that group A rotavirus was responsible for about 20-61% of diarrheal diseases in hospitalized cases. In this study, we reported the continuing surveillance of group A rotavirus in 2005 and found that group A rotavirus was detected in 43 out of 147 (29.3%) stool samples. Five different G and P genotype combinations were detected, G1P[8] (27 strains), G2P[4] (12 strains), G9P[8] (2 strains), G3P[8] (1 strain), and G3P[10] (1 strain). In addition, analysis of their genotypic linkages of G (VP7), P (VP4), I (VP6), E (NSP4), and H (NSP5) genotypes demonstrated that the rotaviruses circulating in Chiang Mai, Thailand carried 3 unique linkage patterns. The G1P[8], G3P[8], and G9P[8] strains carried their VP6, NSP4, NSP5 genotypes of I1, E1, H1, respectively. The G2P[4] strains were linked with I2, E2, H2 genotypes, while an uncommon G3P[10] genotype carried unique genotypes of I8, E3 and H6. These findings provide the overall picture of genotypic linkage data of rotavirus strains circulating in Chiang Mai, Thailand.     

Epidemiological features and genetic characterization of virus strains in rotavirus associated gastroenteritis in children of Odisha in Eastern India

We have studied the clinical characteristics, severity and seasonality of rotavirus infection and prevalent genotypes in 652 non-rota vaccinated children in Odisha in eastern India. P genotypes were analysed for their association with host blood group antigens. P type of the virus is determined by the VP8* gene, and specific recognition of A - type of Histo - blood group antigen by P[14]VP8* has been reported. VP4, VP7 and VP6 genes of commonly identified G1P[8] strain were compared with genes of the same strain isolated from other parts of India, elsewhere and strains used for Rotarix and Rotateq vaccines.In 54.75% of children with gastroenteritis, rota virus was found. 9.65% of children had moderate, 78.07% severe, and 12.28% very severe disease as assessed using the Vesikari scoring system. The incidence of infection was highest during winter months. There was no association between any blood group and specific P genotypes. G1P[8] was the commonest cause of gastroenteritis, followed by G1P[11], G3P[8], G9P[8], G2P[4], G2P[6], G9P[4], G9P[11] and G1P[6]. Predominant G genotypes identified were G1 (72.9%), G9 (10.81%), G2 (8.10%) and G3 (8.10%). Sequence analysis of the VP7 gene, placed the G1P[8] strain in lineage 1 and of VP6 gene placed nine G1P[8] strains in subgroup II and one in subgroup I. The VP7 gene segment of two Odisha G1P[8] strains were found to cluster relatively close to the VP7 sequences of Rotarix vaccine. Antigenic differences were found with vaccine strains. Ten G1P[8] strains sequenced for the VP4 gene had 91–93% nucleotide and 92–96% amino acid identity with Rotateq vaccine P[8]). Rotarix vaccine VP4 had 89–91% nucleotide and 90–92% amino acid identity. Our findings indicate genetic variability of rotavirus strains circulating in the region and are significant, given the introduction of rota vaccination in the State.     

Anticipating rotavirus vaccines: epidemiology and surveillance of rotavirus in South Africa

Rotavirus infection is associated with acute infantile gastroenteritis in infants and young children globally. In South Africa, rotavirus infection has been shown to be associated with approximately one-quarter of all diarrhoeal admissions to hospital. Rotavirus infection predominantly occurs in infants less than 12 months of age (75%) and has a peak of shedding during the cooler, drier months of the year. A secondary peak during the spring has been observed. Multiple infections with rotavirus and at least one other microbial agent are common. The circulating VP7 serotypes and VP4 genotypes have been determined in various regions of South Africa and show a geographic specific distribution. A decade previously, P[8]G1 or G4 strains predominated, and P[4]G2 strains occurred in an epidemic pattern in one region. More recently, rotavirus strains with P[6] genotype have become common and novel VP7/VP4 genotype combinations are occurring across the country. G9 strains have been reported from Cape Town to Vendaland. The circulating rotavirus types observed in this study add to the knowledge of the natural history of rotavirus infection and provide the groundwork to consider future vaccine strategies.     

Impact of rotavirus vaccine on all-cause diarrhea and rotavirus hospitalizations in Madagascar

BackgroundRotavirus vaccine was introduced into the Extended Program on Immunization in Madagascar in May 2014. We analyzed trends in prevalence of all cause diarrhea and rotavirus hospitalization in children <5 years of age before and after vaccine introduction and assessed trend of circulating rotavirus genotypes at Centre Hospitalier Universitaire Mère Enfant Tsaralalàna (CHU MET).MethodsFrom January 2010 to December 2016, we reviewed the admission logbook to observe the rate of hospitalization caused by gastroenteritis among 19619 children <5 years of age admitted at the hospital. In June 2013–December 2016, active rotavirus surveillance was also conducted at CHUMET with support from WHO. Rotavirus antigen was detected by EIA from stool specimen of children who are eligible for rotavirus gastroenteritis surveillance at sentinel site laboratory and rotavirus positive specimens were further genotyped at Regional Reference Laboratory by RT-PCR.ResultsDiarrhea hospitalizations decreased after rotavirus vaccine introduction. The median proportion of annual hospitalizations due to diarrhea was 26% (range: 31–22%) before vaccine introduction; the proportion was 25% the year of vaccine introduction, 17% in 2015 and 16% in 2016. Rotavirus positivity paralleled patterns observed in diarrhea. Before vaccine introduction, 56% of stool specimens tested positive for rotavirus; the percent positive was 13% in 2015, 12% in 2016. Diverse genotypes were detected in the pre-vaccine period; the most common were G3P[8] (n = 53; 66%), G2P[4] (n = 12; 15%), and G1P[8] (n = 11; 14%). 6 distinct genotypes were found in 2015; the most common genotype was G2P[4] (n = 10; 67%), the remaining, 5, G12[P8], G3[P8], G1G3[P4], G3G12[P4][P8] and G1G3[NT] had one positive specimen each.ConclusionsFollowing rotavirus vaccine introduction all-cause diarrhea and rotavirus-specific hospitalizations declined dramatically. The most common genotypes detected in the pre-vaccine period were G3P[8] and G2P[4] in 2015, the post vaccine period.     

Molecular characterization of rotavirus strains detected during a clinical trial of a human rotavirus vaccine in Blantyre, Malawi

The human, G1P[8] rotavirus vaccine (Rotarix?) significantly reduced severe rotavirus gastroenteritis episodes in a clinical trial in South Africa and Malawi, but vaccine efficacy was lower in Malawi (49.5%) than reported in South Africa (76.9%) and elsewhere. The aim of this study was to examine the molecular relationships of circulating wild-type rotaviruses detected during the clinical trial in Malawi to RIX4414 (the strain contained in Rotarix?) and to common human rotavirus strains. Of 88 rotavirus-positive, diarrhoeal stool specimens, 43 rotaviruses exhibited identifiable RNA migration patterns when examined by polyacrylamide gel electrophoresis. The genes encoding VP7, VP4, VP6 and NSP4 of 5 representative strains possessing genotypes G12P[6], G1P[8], G9P[8], and G8P[4] were sequenced. While their VP7 (G) and VP4 (P) genotype designations were confirmed, the VP6 (I) and NSP4 (E) genotypes were either I1E1 or I2E2, indicating that they were of human rotavirus origin. RNA-RNA hybridization using 21 culture-adapted strains showed that Malawian rotaviruses had a genomic RNA constellation common to either the Wa-like or the DS-1 like human rotaviruses. Overall, the Malawi strains appear similar in their genetic make-up to rotaviruses described in countries where vaccine efficacy is greater, suggesting that the lower efficacy in Malawi is unlikely to be explained by the diversity of circulating strains.     

Distribution of rotavirus genotypes in Dhaka,Bangladesh, 2012–2016: Re-emergence of G3P[8] after over a decade of interval

Group A rotavirus causes a substantial proportion of diarrhoea related deaths worldwide among children under five years. We analyzed rotavirus prevalence and genotypes distribution among patients admitted with diarrhoea at icddr,b hospital in Dhaka during 2012–16. Stool specimens (n = 1110) were collected from diarrhoea patients and tested for RVA antigen using enzyme immunoassay. Rotavirus positive samples were G (VP7) and P (VP4) genotyped by RT-PCR and sanger sequencing. Data on clinical manifestations were collected from icddr,b hospital surveillance system. A total of 351 (32%) patients were positive for rotavirus antigen, about half of those were children under two years old. During the study period, G1P[8] (27%) was the most prevalent strain, followed by G12P[8] (15%) and G9[P4] (9%). Mixed G or P genotypes were identified in a substantial proportion (23%) with few strains of rare combinations such as G1P[4], G1P[6], G2P[6], G2P[8], G9P[6]. The genotypic fluctuation was noteworthy; G12P[8] was the major strain in 2012–14 but sharply decreased in 2015–16 when G1P[8] became the most common strain. G3P[8] re-emerged (17%) in 2016 after 11 years. Since the Government of Bangladesh has planned to include rotavirus vaccine in national immunization programme from 2018, our data will provide baseline information on rotavirus genotypes in the pre-vaccination era to observe the selection pressure on genotypes in the post vaccination epoch.     

Genotype distribution of Group A rotavirus from southern India, 2005–2016

Diarrheal disease due to Group A rotaviruses remain a leading cause of mortality and morbidity in the less developed parts of the world. India has started a phased roll out of rotavirus vaccine in the national immunization program. This analysis summarizes the rotavirus genotype strain distribution pre-vaccine introduction in Vellore, India from December 2005 to June 2016. Rotavirus was responsible for 32% of all diarrheal admission to the hospital. G2P[4] was the predominant strain in the initial years and was gradually replaced by G1P[8]. The emergence of G9P[4] replacing G9P[8], and the detection of G12 strains over several years were documented. There was no clear seasonality of disease. These data form the baseline to monitor genotype distribution post-vaccine introduction in Tamil Nadu.     

Molecular characterization of two rare human G8P[14] rotavirus strains,detected in Italy in 2012

Since 2007, the Italian Rotavirus Surveillance Program (RotaNet-Italy) has monitored the diversity and distribution of genotypes identified in children hospitalized with rotavirus acute gastroenteritis.We report the genomic characterization of two rare human G8P[14] rotavirus strains, identified in two children hospitalized with acute gastroenteritis in the southern Italian region of Apulia during rotavirus strain surveillance in 2012.Both strains showed a G8-P[14]-I2-R2-C2-M2-A11-N2-T6-E2-H3 genomic constellation (DS-1-like genomic background). Phylogenetic analysis of each genome segment revealed a mixed configuration of genes of animal and zoonotic human origin, indicating that genetic reassortment events generated these unusual human strains. Eight out of 11 genes (VP1, VP2, VP3, VP6, VP7, NSP3, NSP4 and NSP5) of the Italian G8P[14] strains exhibited close identity with a Spanish sheep strain, whereas the remaining genes (VP4, NSP1 and NSP2) were more closely related to human strains. The amino acid sequences of the antigenic regions of outer capsid proteins VP4 and VP7 were compared with vaccine and field strains, showing high conservation between the amino acid sequences of Apulia G8P[14] strains and human and animal strains bearing G8 and/or P[14] proteins, and revealing many substitutions with respect to the RotaTeq™ and Rotarix™ vaccine strains. Conversely, the amino acid analysis of the four antigenic sites of VP6 revealed a high degree of conservation between the two Apulia strains and the human and animal strains analyzed.These results reinforce the potential role of interspecies transmission and reassortment in generating novel rotavirus strains that might not be fully contrasted by current vaccines.     

Large-scale whole genome sequencing identifies country-wide spread of an emerging G9P[8] rotavirus strain in Hungary, 2012

With the availability of rotavirus vaccines routine strain surveillance has been launched or continued in many countries worldwide. In this study relevant information is provided from Hungary in order to extend knowledge about circulating rotavirus strains. Direct sequencing of the RT-PCR products obtained by VP7 and VP4 genes specific primer sets was utilized as routine laboratory method. In addition we explored the advantage of random primed RT-PCR and semiconductor sequencing of the whole genome of selected strains. During the study year, 2012, we identified an increase in the prevalence of G9P[8] strains across the country. This genotype combination predominated in seven out of nine study sites (detection rates, 45–83%). In addition to G9P[8]s, epidemiologically major strains included genotypes G1P[8] (34.2%), G2P[4] (13.5%), and G4P[8] (7.4%), whereas unusual and rare strains were G3P[8] (1%), G2P[8] (0.5%), G1P[4] (0.2%), G3P[4] (0.2%), and G3P[9] (0.2%). Whole genome analysis of 125 Hungarian human rotaviruses identified nine major genotype constellations and uncovered both intra- and intergenogroup reassortment events in circulating strains. Intergenogroup reassortment resulted in several unusual genotype constellations, including mono-reassortant G1P[8] and G9P[8] strains whose genotype 1 (Wa-like) backbone gene constellations contained DS1-like NSP2 and VP3 genes, respectively, as well as, a putative bovine–feline G3P[9] reassortant strain. The conserved genomic constellations of epidemiologically major genotypes suggested the clonal spread of the re-emerging G9P[8] genotype and several co-circulating strains (e.g., G1P[8] and G2P[4]) in many study sites during 2012. Of interest, medically important G2P[4] strains carried bovine-like VP1 and VP6 genes in their genotype constellation. No evidence for vaccine associated selection, or, interaction between wild-type and vaccine strains was obtained. In conclusion, this study reports the reemergence of G9P[8] strains across the country and indicates the robustness of whole genome sequencing in routine rotavirus strain surveillance.     

The laboratory test procedure to confirm rotavirus vaccine infection in severe complex immunodeficiency patients

The rotavirus vaccine is a live vaccine, and there is a possibility of infection by the virus strain used in the vaccine. We investigated the process of determining whether an infection was caused by the vaccine strain in a severe complex immunodeficiency (SCID) patient with rotavirus infection. The patient was vaccinated with RotaTeq prior to being diagnosed with SCID. The testing process was conducted in the following order: confirming rotavirus infection, determining its genotype, and confirming the vaccine strain. Rotavirus infection was confirmed through enzyme immunoassay and VP6 gene detection. G1 and P[8] were identified by multiplex polymerase chain reaction for the genotype, and G3 was further identified using a single primer. By detecting the fingerprint gene (WC3) of RotaTeq, it was confirmed that the detected virus was the vaccine strain. Genotypes G1 and P[8] were identified, and the infection was suspected of having been caused by rotavirus G1P[8]. G1P[8] is the most commonly detected genotype worldwide and is not included in the recombinant strains used in vaccines. Therefore, the infection was confirmed to have been caused by the vaccine strain by analyzing the genetic relationship between VP4 and VP7. Rotavirus infection by the vaccine strain can be identified through genotyping and fingerprint gene detection. However, genetic linkage analysis will also help to identify vaccine strains.     

Emergence of unusual rotavirus G9P[4] and G8P[8] strains during post vaccination surveillance in Argentina, 2017–2018

IntroductionIn 2015, Argentina included Rotarix™ monovalent vaccine for universal administration and it showed a sharp decline in all-cause and rotavirus-confirmed cases as well as an immediate predominance of the G2P[4] genotype. The aim of this study was to analyze the impact of rotavirus vaccination on laboratory-confirmed cases and genotype distribution in Argentina following its introduction.Material and methodsPrevalence and seasonality of laboratory-confirmed rotavirus cases data were assessed. Analyses of circulating genotypes were performed by conventional binary characterization (G and P typing). Phylogenetic study of VP7 gene was performed from emergent unusual strains.ResultsDuring 2017–2018, 1183 rotavirus cases (13.2%) were detected, and prevalence was uniform among different age subgroups. Weekly distribution showed a raise of confirmed cases around late July and early August. In 2017 the most frequently detected genotypes were G2P[4] and G3P[8]. However, in 2018 G12P[8] genotype increased and it was detected at a high rate. Noteworthy, the detection of uncommon G9P[4] and G8P[8] strains (bearing DS-1-like genetic backbones) was observed at moderate rates.DiscussionFollowing four years of universal vaccination, the prevalence of rotavirus remained low in children under 5 years of age with a shift of the seasonal peak in early spring. The emergence of uncommon genotypes was due to introduction of new strains rather than to reassortment of local strains. Continuous monitoring of rotavirus burden of disease and genotype distribution provides useful evidence to evaluate existing immunization strategies and to contribute in the development of new vaccines as well.