Whole genome analysis of multiple rotavirus strains from a single stool specimen using sequence-independent amplification and 454® pyrosequencing reveals evidence of intergenotype genome segment recombination

https://doi.org/10.1016/j.meegid.2011.09.023Get rights and content

Abstract

Infection of a single host cell with two or more different rotavirus strains creates conditions favourable for evolutionary mechanisms like reassortment and recombination that can generate novel strains. Despite numerous reports describing mixed rotavirus infections, whole genome characterisation of rotavirus strains in a mixed infection case has not been reported. Double-stranded RNA, exhibiting a long electropherotype pattern only, was extracted from a single human stool specimen (RVA/Human-wt/ZAF/2371WC/2008/G9P[8]). Both short and long electropherotype profiles were however detected in the sequence-independent amplified cDNA derived from the dsRNA, suggesting infection with more than one rotavirus strain. 454® pyrosequencing of the amplified cDNA revealed co-infection of at least four strains. Both genotype 1 (Wa-like) and genotype 2 (DS-1-like) were assigned to the consensus sequences obtained from the nine genome segments encoding NSP1–NSP5, VP1–VP3 and VP6. Genotypes assigned to the genome segments encoding VP4 were P[4] (DS-1-like), P[6] (ST3-like) and P[8] (Wa-like) genotypes. Since four distinct genotypes [G2 (DS-1-like), G8, G9 (Wa-like) and G12] were assigned to the four consensus nucleotide sequences obtained for genome segment 9 (VP7), it was concluded that at least four distinct rotaviruses were present in the stool. Intergenotype genome recombination events were observed in genome segments encoding NSP2, NSP4 and VP6. The close similarities of some of the genome segments encoding NSP2, VP6 and VP7 to artiodactyl rotaviruses suggest that some of the infecting strains shared common ancestry with animal strains, or that interspecies transmission occurred previously. The sequence-independent genome amplification technology coupled with 454® pyrosequencing used in this study enabled the characterisation of the whole genomes of multiple rotavirus strains in a single stool specimen that was previously assigned single genotypes, i.e. G9P[8], by sequence-dependent RT-PCR.

Highlights

► The whole genomes of multiple rotavirus strains were characterized from a single stool sample. ► Sequence-independent amplification and 454® pyrosequencing were used. ► Four distinct rotavirus strains were identified. ► Intergenotype recombination occurred in VP6, NSP2 and NSP4 encoding genome segments. ► Some genome segments encoding VP6, VP7 and NSP2 shared ancestry with animal rotaviruses.

Introduction

Human rotaviruses are the main cause of severe infant gastroenteritis. Each year 527 000 rotavirus-associated deaths occur, mostly in developing countries (Parashar et al., 2009). Rotaviruses represent a genus in the Reoviridae virus family. The mature rotavirus particle contains a 11-segmented double-stranded RNA (dsRNA) genome that is surrounded by three layers of capsid proteins. With the exception of genome segment 11 that encodes two proteins in some group A rotaviruses, each genome segment encodes one protein. There are six structural (VP1–VP4, VP6 and VP7) and six non-structural (NSP1–NSP5) viral proteins. The commonly used dual rotavirus classification system is based on the properties of the two outer capsid proteins, VP4 and VP7 (Estes and Kapikian, 2007). The P and G serotypes/genotypes refer to VP4 (protease-sensitive) and VP7 (glycoprotein), respectively. To date, 35 P and 27 G group A rotavirus genotypes have been defined (Matthijnssens et al., 2011), illustrating the diversity of rotavirus strains.

Point mutations, genome reassortment and recombination are thought to be the main evolutionary forces driving rotavirus strain diversity (Desselberger et al., 2001). Rotaviruses are prone to point mutations as they utilise a viral RNA-dependent RNA polymerase during replication that lacks proof-reading activity (Estes and Kapikian, 2007). One nucleotide mutation is estimated to occur per replication of one full rotavirus genome (Blackhall et al., 1996). Despite increasing the genetic diversity of rotaviruses (Espínola et al., 2008), such nucleotide variations have implications for the characterisation of rotaviruses as it affects serotype reactivity of rotavirus strains to monoclonal antibodies (Ciarlet et al., 1997) and primer binding sites used for sequence-dependent genotyping (Martella et al., 2004). The segmented dsRNA genome enables reassortment events between distinct strains infecting a single cell simultaneously (Gouvea and Brantly, 1995). Furthermore, it is believed that zoonotic transmission contribute significantly towards rotavirus strain diversity (Tsugawa and Hoshino, 2008, Ghosh et al., 2010, Martella et al., 2010). Some animal rotavirus strains are thought to be directly introduced into humans via interspecies transmission (Matthijnssens et al., 2006a, Tsugawa and Hoshino, 2008), while others are generated through apparent single or multiple-genome segment reassortment events between human and animal rotaviruses (Matthijnssens et al., 2008b). Several other evolutionary mechanisms like intragenic genome recombination (both inter-lineage and inter-sub-lineage) further expands rotavirus strain diversity (Suzuki et al., 1998, Parra et al., 2007, Phan et al., 2007).

Several approaches have been used to determine the origin and the genetic relationships of rotavirus strains. Recently, the full genome classification scheme based on the nucleotide sequence identity of all 11 rotavirus genome segments has proved to be an excellent tool for identifying and studying the evolution of unusual rotavirus strains (Matthijnssens et al., 2008a). It was shown that the G12 porcine RU172 strain could have been generated through reassortment of human and porcine rotaviruses (Ghosh et al., 2010). Interspecies transmission cases were also confirmed where canine and feline rotavirus strains were directly introduced into humans (Tsugawa and Hoshino, 2008), and Matthijnssens et al. (2006a) demonstrated that a strain RVA/Human-wt/BEL/B4106/2000/G3P[14] that contains an entire lapine genome caused severe disease in humans. Similarly, Matthijnssens et al. (2006b) and Esona et al. (2009) used the whole genome classification scheme to show that G8 human rotaviruses isolated across Africa share ancestral relationships with DS-1-like and animal rotavirus strains, respectively. Furthermore, Matthijnssens et al. (2008a) also illustrated that DS-1- and Wa-like rotaviruses share a common origin with bovine and porcine strains, respectively.

For genome reassortment or recombination to take place, a single host cell must be infected by two or more different rotavirus strains (Estes and Kapikian, 2007). Several molecular epidemiology rotavirus studies suggest that human infection with multiple rotavirus strains, known as mixed infection, is common (Arguelles et al., 2000, Nielsen et al., 2005, Iturriza-Gómara et al., 2009, Iturriza-Gómara et al., 2011, Esteban et al., 2010, Han et al., 2010, Mwenda et al., 2010, Potgieter et al., 2010). These mixed infections may create conditions favourable for the generation of novel strains through genome segment reassortment or recombination. Despite numerous reports describing mixed rotavirus infections, no study has yet attempted to characterise the whole genomes of rotavirus strains in a mixed infection. In this study, both short and long electropherotype rotavirus profiles were detected in the cDNA synthesised from the dsRNA of strain RVA/Human-wt/ZAF/2371WC/2008/G9P[8]. Initially a long electropherotype pattern was assigned and sequence-dependent RT-PCR using a cocktail of G1, G2, G3, G8, G9, G12, P[4], P[6] and P[8] genotype-specific primers (Gouvea et al., 1990, Gentsch et al., 1992, Das et al., 1994, Iturizza-Gómara et al., 2004), assigned G9P[8] to the sample. The consensus nucleotide sequences of all 11 genome segments of the multiple rotavirus strains in the sample were generated through sequence-independent genome amplification and 454® pyrosequencing. Sequence analysis suggested the possible origins for the genome segments of the infecting strains, and the role mixed infections could play in influencing evolutionary mechanisms such as genome recombination in the generation of novel rotaviruses.

Section snippets

Rotavirus strain

Strain RVA/Human-wt/ZAF/2371WC/2008/G9P[8] was obtained from the Viral Gastroenteritis Unit (VGU), National Institute for Communicable Diseases (NICD), South Africa. The stool sample was analysed as part of the routine surveillance activities within VGU which was approved by an ethics committee (protocol number M060449). The sample was collected in 2008 from a 27 month old male child presenting with diarrhoea at Gatesville Hospital, Western Cape Province, South Africa. No additional personal or

Whole genome amplification of strain RVA/Human-wt/ZAF/2371WC/2008/G9P[8]

Strain RVA/Human-wt/ZAF/2371WC/2008/G9P[8] was typed by sequence-dependent RT-PCR as G9P[8] unequivocally. Its dsRNA displayed a long electropherotype pattern (Fig. 1a). When all the 11 genome segments of strain RVA/Human-wt/ZAF/2371WC/2008/G9P[8] were amplified, its cDNA displayed two distinct genome segment 11 profiles that are characteristic of both long and short electropherotype patterns (Fig. 1b) that are commonly associated with Wa- and DS-1-like rotaviruses, respectively (Estes and

Discussion

The specimen characterised in this study was part of a surveillance study of rotavirus strains circulating in the Western Cape Province of South Africa. Sequence-dependent RT-PCR that used genotype-specific primers only assigned the P[8] and G9 genotypes to the genome segments 4 and 9 of the study strain, respectively. In addition, the dsRNA profile had only a long electropherotype profile. Upon synthesising and amplifying the cDNA from the dsRNA extracted from the stool sample with the

Authors’ contributions

KCJ was involved in the design of the study, laboratory assays, data analysis and manuscript writing. NAP was involved in the collection of study specimen and manuscript writing. LM was involved in data analysis and manuscript writing. HGO and AAvD were involved in the study design, data analysis and manuscript writing. All authors read and approved the final manuscript.

Acknowledgements

We thank the Viral Gastroenteritis Unit, National Institute for Communicable Diseases (NICD) for providing the rotavirus strain characterised in this study. This study was partially funded by the North-West University, the South African National Research Foundation (Grants FA2005031700015 and UID 63427) and the Poliomyelitis Research Foundation of South Africa (Grant No. 09/34).

References (68)

  • N.M. Nielsen et al.

    Characterisation of rotavirus strains among hospitalised and non-hospitalised children in Guinea-Bissau, 2002 A high frequency of mixed infections with serotype G8

    J. Clin. Virol.

    (2005)
  • G.I. Parra et al.

    Diversity of group A rotavirus strains circulating in Paraguay from 2002 to 2005: detection of an atypical G1 in South America

    J. Clin. Virol.

    (2007)
  • Y. Suzuki et al.

    Intragenic recombinations in rotaviruses

    FEBS Lett.

    (1998)
  • T. Tsugawa et al.

    Whole genome sequence and phylogenetic analyses reveal human rotavirus G3P[3] strains Ro1845 and HCR3A are examples of direct virion transmission of canine/feline rotaviruses to humans

    Virology

    (2008)
  • M.I. Adah et al.

    Close relationship between G8-serotype bovine and human rotaviruses isolated in Nigeria

    J. Clin. Microbiol.

    (2003)
  • M.U. Ahmed et al.

    Analysis of human rotavirus strains prevailing in Bangladesh in relation to nationwide floods brought by the 1988 monsoon

    J. Clin. Microbiol.

    (1991)
  • H. Arguelles et al.

    VP7 and VP4 genotyping of human group A rotavirus in Buenos Aires

    Argentina. J. Clin. Microbiol.

    (2000)
  • K. Bányai et al.

    Sequencing and phylogenetic analysis of the coding region of six common rotavirus strains: evidence for intragenogroup reassortment among co-circulating G1P[8] and G2P[4] strains from the United States

    J. Med. Virol.

    (2011)
  • D. Cao et al.

    Porcine rotavirus bearing an aberrant gene stemming from an intergenic recombination of the NSP2 and NSP5 genes is defective and interfering

    J. Virol.

    (2008)
  • M. Ciarlet et al.

    Single point mutations may affect the serotype reactivity of serotype G11 porcine rotavirus strains: a widening spectrum?

    J. Virol.

    (1997)
  • M. Ciarlet et al.

    Genomic characterization of a novel group A lamb rotavirus isolated in Zaragoza, Spain

    Virus Genes

    (2008)
  • B.K. Das et al.

    Characterization of rotavirus strains from newborns in New Delhi

    India J. Clin. Microbiol.

    (1994)
  • U. Desselberger et al.

    Rotavirus epidemiology and surveillance

    Novartis Found Symp.

    (2001)
  • Donker, N.C., Boniface, K., Kirkwood, C.D., 2011. Phylogenetic analysis of rotavirus A NSP2 gene sequences and evidence...
  • M.D. Esona et al.

    Genomic characterization of human rotavirus G8 strains from the African rotavirus network: relationship to animal rotaviruses

    J. Med. Virol.

    (2009)
  • L.E. Esteban et al.

    Molecular epidemiology of group A rotavirus in Buenos Aires, Argentina 2004–2007: reemergence of G2P[4] and emergence of G9P[8] strains

    J. Med. Virol.

    (2010)
  • M.K. Estes et al.

    Rotaviruses

  • J.R. Gentsch et al.

    Identification of group A rotavirus gene 4 type by polymerase chain reaction

    J. Clin. Microbiol.

    (1992)
  • S. Ghosh et al.

    Full genomic analysis and possible origin of a porcine G12 rotavirus strain RU172

    Virus Genes

    (2010)
  • V. Gouvea et al.

    Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens

    J. Clin. Microbiol.

    (1990)
  • T.A. Hall

    BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT

    Nucleic Acids Symp. Ser.

    (1999)
  • T.H. Han et al.

    Genetic characterization of rotavirus in children in South Korea from 2007 to 2009

    Arch. Virol.

    (2010)
  • C.Q. He et al.

    Intragenic recombination as a mechanism of genetic diversity in bluetongue virus

    J. Virol.

    (2010)
  • E.M. Heiman et al.

    Group A human rotavirus genomics: evidence that gene constellations are influenced by viral protein interactions

    J. Virol.

    (2008)
  • Cited by (32)

    • Whole genome analyses of DS-1-like Rotavirus A strains detected in children with acute diarrhoea in southern Mozambique suggest several reassortment events

      2019, Infection, Genetics and Evolution
      Citation Excerpt :

      Genome assembly of DS-1-like rotavirus strains detected in the Maputo province of Mozambique. The percentage genome segment length and percentage length of the open reading frame (ORF) of the Mozambican sequences were based on RVA/Human-tc/USA/DS-1/1976/G2P[4] (Mlera et al. 2011) except for the following: segment 5, genotype 2 (A2) and segment 7, genotype 2 (T2) were based on strain RVA/Human-wt/ZAF/3203WC/2009/G2P[4] (1066 bp) (Jere et al., 2011). The percentage identity was determined in ViPR during genotyping.

    • Emergence and spread of G3P[8] rotaviruses possessing an equine-like VP7 and a DS-1-like genetic backbone in the Basque Country (North of Spain), 2015

      2016, Infection, Genetics and Evolution
      Citation Excerpt :

      Next Generation Sequencing was performed in six atypical G3 strains selected on the basis of having sufficient material leftover and the presence of high viral load inferred from the presence of high intensity bands of VP4 and VP7 amplicons when visualized under ultraviolet light in agarose electrophoresis gel. The dsRNA genomes were extracted following previously described methods with minor modifications (Jere et al., 2011). Briefly, the total RNA was extracted adding 1 ml of TRI-REAGENT–LS (Molecular Research Centre, Ohio, USA) in 300 μl of previously suspended stool.

    • Whole genomic analysis of human and bovine G8P[1] rotavirus strains isolated in Nigeria provides evidence for direct bovine-to-human interspecies transmission

      2016, Infection, Genetics and Evolution
      Citation Excerpt :

      We next constructed phylogenetic trees using the full-genome sequence for each of the 11 gene segments because phylogenetic analysis of RVA nucleotide sequences furnishes direct evidence of their relatedness to those of other strains, even within the same genotype (Matthijnssens et al., 2008a; Komoto et al., 2014b, 2016). The VP7 genes of strains HMG035 and NGRBg8 exhibited complete nucleotide sequence identity (100%), and the maximum nucleotide sequence identity (97.8%) with that of South African human strain 2371WCVP7C (G8P[x]) (Jere et al., 2011). On phylogenetic analysis, strains HMG035 and NGRBg8 were shown to be closely related with strain 2371WCVP7C in a common branch with several human and bovine-like human strains from sub-Saharan Africa in G8 lineage-6, in which the majority of African human G8 strains are clustered (Agbemabiese et al., 2015; Nakagomi et al., 2013) (Fig. 1a).

    • Intrasegmental recombination does not contribute to the long-term evolution of group A rotavirus

      2015, Infection, Genetics and Evolution
      Citation Excerpt :

      Little is known about whether genetic recombination within segments occurs. Several reports have identified sequences with intrasegmental recombination (Donker et al., 2011; Jere et al., 2011; Martínez-Laso et al., 2009; Parra, 2004; Phan et al., 2007a,b; Suzuki et al., 1998). Importantly, intrasegmental recombination has been reported to occur between the epitopes of the dominant antigenic protein, VP7 (Phan et al., 2007b).

    View all citing articles on Scopus
    View full text