|Home | About | Journals | Submit | Contact Us | Français|
Rotavirus G10P strains have long been associated with asymptomatic neonatal infections in some parts of India. We have previously reported G10P strains associated with both asymptomatic infections and severe gastrointestinal disease in neonates from Vellore in southern India, with >90% partial nucleotide and amino acid identity to the VP4, VP6, VP7 and NSP4 genes of the exclusively asymptomatic G10P strain I321.
In this study, the whole genome of a G10P isolate (N155) from a neonate with severe gastrointestinal disease was characterized to determine whether there were significant differences in its genetic makeup in comparison to G10P strain I321 and to establish the origin of the G10P strains in Vellore.
PCR amplification and complete genome sequencing was carried out for all 11 gene segments of symptomatic G10P rotavirus isolate N155. Nucleotide and amino acid sequence similarity with I321, other human and bovine strains for each gene segment were determined. The origin of each gene was determined based on the degree of identity to bovine or human rotavirus strains.
N155 was found to be a reassortant between human and bovine rotaviruses. With the exception of NSP2, gene sequences of strain N155 showed >90% identity to published sequences of I321. Gene segments encoding NSP1, 2 and 3 were of human rotavirus origin for both strains; however, phylogenetic analysis of NSP2 sequences indicated that the human parental strain that led to the origin of these bovine-human reassortant strains was different. There were no significant differences between NSP2 sequences of strains from symptomatic and asymptomatic neonates in the same setting.
The study shows that the difference in clinical presentations in neonates may not be due to the limited variability in the genome sequence of G10P strains and that G10P strains in different parts of India could have evolved through reassortment of different parental strains.
Rotaviruses have an 11 segmented double-stranded RNA genome encoding 6 structural and 6 non-structural proteins.1 Classification into 19 G and 27 P genotypes is based on variability in genes encoding outer capsid proteins VP7 and VP4.1–6
Bovine–human reassortant G10P rotaviruses were first described in asymptomatic neonatal infections in Bangalore and Mysore in southern India.7 A significant reduction in rotavirus diarrhea hospitalizations was reported in parallel with high rates of asymptomatic neonatal rotavirus infection.8 The strain I321, with all genes of bovine rotavirus origin except NSP1 and NSP39 was developed as a candidate vaccine, but later withdrawn because of poor immunogenicity.10 The other Indian candidate vaccine 116E is also a natural bovine–human reassortant (human G9 VP7 and bovine P VP4) isolated from neonates.11,12
I321-like G10P strains causing both symptomatic and asymptomatic infections have been described in neonates from Vellore, southern India.13,14 These strains were seen in neonates with diarrhea, vomiting and other gastrointestinal symptoms. Partial genome analysis of NSP4, VP6, VP7 and VP4 did not reveal significant differences between G10P strains from symptomatic and asymptomatic infections or from strain I321.13
All genes from N155, a G10P rotavirus isolate from a symptomatic child, were compared with published data on I321, a culture-adapted asymptomatic strain developed as a vaccine, to study potential determinants of virulence or attenuation.
Stool samples from 245 neonates in the nurseries of the Christian Medical College were screened for rotavirus using an enzyme immunoassay (EIA) for VP6 antigen (Rota IDEIA, Oxoid, Basingstoke, UK).14 Viral RNA was extracted from 20% fecal suspensions in minimum essential medium (MEM) using the guanidine isothiocyanate-silica based method.15 Complementary DNA (cDNA) was generated using random primers (hexamers; Pd(N)6, Pharmacia Biotech, UK) and reverse transcription with 400 units of Moloney murine leukemia virus reverse transcriptase (M-MLV RT) (Invitrogen, UK). The sample were genotyped by RT-PCR for VP7 and VP4 genes as previously described.16–20 Briefly, for G-typing, the first round PCR used primers VP7-F and VP7-R primers to amplify an 881 bp region of the VP7 gene. The nested multiplex PCR incorporated primers specific for amplification of genotypes G1, G2, G3, G4, G8, G9 and G10. Primers Con2 and Con3 were used in the first round PCR to amplify an 876 bp fragment of the VP4 gene. The second round PCR included primers specific for genotypes P, P, P, P, P and P. Ninety samples were positive for rotavirus and >80% of strains were genotyped G10P. Polyacrylamide gel electrophoresis of a subset of G10P strains from symptomatic and asymptomatic neonates did not show any differences in RNA migration pattern.
N155 RNA was obtained from the stool sample of a term, male baby born by normal vaginal delivery at 34 + 1 weeks of gestation, who was referred to the nursery at 67 h of age with abdominal distension and feed intolerance. An abdominal radiograph revealed dilated bowel loops and NEC. The baby was given intravenous fluids and kept nil oral for three days after which time feeds were initiated and well tolerated. The sample was screened for other enteropathogens, including Norovirus genogroups I and II, enteric adenovirus, Cryptosporidium spp. and diarrhoeagenic E. coli, including enteropathogenic, enterohaemorrhagic, enteroinvasive, diffusely adherent, enteroaggregative and enterotoxigenic E. coli using published oligonucleotide primers and methods.21
RNA extraction and random priming reverse transcription for cDNA synthesis was carried out as described previously. Primers for the amplification of individual genes were designed using primer3 (http://frodo.wi.mit.edu/) and BioEdit softwares (http://www.mbio.ncsu.edu/BioEdit/bioedit.html) based on corresponding genes sequences available in GenBank. Primer pairs were designed to amplify short fragments of each gene ranging from 387 to 1355 bp, with each fragment overlapping the next one by about 100 bp. RT-PCR was carried out with an initial denaturation at 95 °C for 5 min and 35 cycles of amplification using 5U of Taq polymerase (Invitrogen). High-fidelity Taq polymerase (Invitrogen) was used for the amplification of three genes (VP6, VP7 and NSP3) whose amplicon sizes were >1000 bp. The final sequence was determined by assembling consensus genome sequences from all fragments for the gene. The primer sequences used for sequencing the beginning and end of each gene in this study are given in Table 1. All primers used in the study and their sequences are listed in the supplementary table.
The 5′ and 3′ sequences of the 11 gene segments were determined by using the single-primer amplification method.22 Extraction of viral RNA was followed by ligation of a modified amino-linked oligonucleotide (5′-PO4-TTCCTTATGCAGCTGATCACTCTGTGTCA-C6-NH2-3′) to the 3′ ends of both strands of the viral dsRNA using T4 RNA Ligase (Promega, USA). One step RT-PCR using Superscript III RT-PCR kit (Invitrogen, UK) was carried out with a primer complementary to the modified amino-linked oligonucleotide (5′-TGACACAGAGTGATCAGC-3′) and a gene specific internal primer.
PCR amplicons were purified using PCR cleanup filter plates (Millipore, USA) and sequenced with ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction kit (Applied Biosystems, USA). Sequences were resolved in the ABI PRISM 310 Genetic Analyzer (Applied Biosystems, USA). Deduced amino acid (aa) sequences were derived using the Expert Protein Analysis System (ExPASy) proteomics server of the Swiss Institute of Bioinformatics (SIB). The nucleotide (nt) and deduced amino acid sequences of all genes were compared with sequences available in the NCBI (National Center for Biotechnology Information) GenBank database using BLAST (Basic Local Alignment Search Tool). Multiple alignments and phylogenetic analysis of nt and deduced aa sequences were performed using Clustal W and neighbor-joining algorithms of the BioEdit (version 188.8.131.52) or Bionumerics (Applied Maths, Kortrijk, Belgium) software packages. Dendrograms were confirmed by neighbor-joining and maximum parsimony methods. Genetic lineages were confirmed by bootstrap values of >95% using 1000 pseudoreplicate runs.
A total of 18,242 nucleotides encoding 5781 amino acids of the eleven segments of the G10P strain N155 were sequenced. The complete coding sequences (cds) were determined for all genes except VP6 and NSP3, in which the first two and three amino acids, respectively, of the open reading frames could not be ascertained.
N155 sequences were compared with published sequences of corresponding genes, with origin determined based on the degree of identity to bovine or human rotavirus strains. The closest sequence match and similarity with I321, human and bovine strains for each gene of N155 is shown in Table 2. Sequence data of I321 was available only for genes VP4, VP6, VP7, NSP1, NSP2, NSP3 and NSP4. Comparison of nt sequences of N155 showed >93% identity for all genes except NSP2.
The VP1, VP2 and VP3 genes of N155 were of bovine origin, with 93%, 94% and 92% identity, respectively, with bovine strains NCDV or UK. The VP6 sequence of strain N155 clustered with SGI rotavirus strains and was more closely related to I321 (97% identity at the nucleotide (nt) level) and VP6 SGI sequences derived from bovine strains than SGI sequences of common human rotavirus strains (Table 2). On translation, there were 4 amino acid (aa) substitutions between I321 and N155 at positions 58, 179, 190 and 244, none of which are believed to be subgroup-defining (data not shown).23 The genes encoding VP4 and VP7 were more closely related to strain I321 (95% and 93% at the nt level, respectively) than to other bovine or human P and G10 strains (Table 2).
The gene encoding NSP1 of N155 was 97% identical to the corresponding sequence of I321, and likely of human origin (Table 2). Human rotavirus strains clustered in four genetic lineages with strains N155 and I321 in the same lineage (Fig. 1). The derived aa sequences showed 15 substitutions between the two strains, one of which was located in the cysteine rich zinc finger motif (I321, R55K N155). Mutations in cysteine and histidine residues in the zinc finger motif are important in NSP1 stability and interferon regulatory factor 3 (IRF3) binding,24 and multipoint substitutions of arginine–lysine residues increase enzyme thermostability.25,26 However, a single substitution of these highly similar basic amino acids is unlikely to result in a significant difference in the strain. This is supported by the finding that this substitution was conserved among symptomatic and asymptomatic G10P strains from Vellore (data not shown).
The NSP2 encoding gene of N155 was closely related to sequences from human rotaviruses showing >95% nt and 98% aa identity with recent human isolates from Vellore and Bangladesh of common genotypes (G1P and G9P), and >92% with human prototype strain Wa. Only 80.9% nt identity was seen with I321 and 82.8–89.1% with bovine strains (Table 2). Phylogenetic analysis of the NSP2 sequences showed that most human rotaviruses segregated into 2 genetic clusters and N155 and I321 clustered in different lineages of human rotaviruses (Fig. 2).
In order to assess the role of this difference in virulence, partial nucleotide sequence analysis of the region encoding aa 15–291 was carried out from 5 additional G10P strains (2 symptomatic and 3 from asymptomatic infections). No significant differences were seen among NSP2 sequences of these isolates (Fig. 3). Terminal sequencing reactions were not carried out for the additional G10P sequences as these regions are known to be highly conserved.
The NSP3 gene of N155 was similar to I321 (>95% identity at the nt level) and was highly related to human rotaviruses (Table 2). Two distinct genetic lineages were observed among human rotavirus NSP3 gene sequences (Fig. 4).
The NSP4 sequence of N155 showed 95% identity with I321 and other G10P strains isolated in Vellore (Table 2), and were likely to originate from a bovine strain.13 One aa substitution (I321 C129R N155) in the enterotoxin coding region and substitutions at aa residues 97, 94 and 136 described as associated with virulence among lapine strain BAP, porcine strains Gottfried and OSU, respectively27,28 were seen; however, these changes in N155 were conserved across isolates from both symptomatic and asymptomatic G10P strains from Vellore.13
Gene segment 11 of strain N155 showed 95% identity at the nt level to bovine strain VMRI (Table 2).
Sequence analysis of the complete genome of strain N155 isolated from a neonate with symptomatic rotavirus infection confirms that G10P strains found in humans in Vellore are bovine–human reassortant strains. Eight out of 11 gene segments are likely to have originated from bovine strains. With the exception of NSP2, a high degree of identity was seen between N155 and I321 in six of seven genes for which sequence data was available for both strains. Sequence data on I321 for genes encoding VP1, VP2, VP3 and NSP5 were not available for comparison although these have been reported to be of bovine origin through whole genome hybridization studies.9
This study showed that the NSP2 gene of both I321 and N155 originated from human rotavirus strains, but the strains belonged to two distinct genetic lineages with <80% identity between them. These findings would suggest that independent reassortment events between human and bovine strains led to the emergence of G10P strains in Vellore and Bangalore. Both strains have three genes of human rotavirus origin, but the gene encoding NSP2 clearly originated from different human parental strains. These differences however do not explain the differences in virulence, as shown through more extensive analysis of NSP2 sequences from additional G10P strains from symptomatic and asymptomatic infections.
Early studies on I321 had revealed strong cross hybridization of NSP1, NSP2 and NSP3 genes with human rotavirus strain Wa.9 Subsequent sequence analysis showed that NSP1 and NSP3 genes were of human origin,29,30 and NSP2 was assigned to a bovine origin. The current study has demonstrated that the NSP2 of strains I321 clusters within a different genetic lineage of human rotaviruses. The whole genome hybridization methods for tracking the origin of rotavirus gene segments may be insensitive, probably due to difficulties in standardizing appropriately stringent conditions.
Based on the whole genome characterization of G10P, the 3 genes of human rotavirus origin—NSP1, NSP2 and NSP3 are probably essential for efficient replication of a bovine strain in a human host. The role of these genes in host restriction has previously been discussed in other studies but the data are inconclusive.29,31–33 In this study, segregation of NSP1 and NSP3 into genetic lineages did not correlate with the host of origin but there was more clear segregation into genetic lineages and clusters according to the host of origin for NSP2.
Complete genome sequences are available from few strains throughout the world34–36 and there is lack of information on co-circulating animal rotaviruses. The comparison of N155 with I321 is important because G10P strains in Vellore are associated with disease,14 while I321 was developed as a vaccine candidate because of reports of exclusive asymptomatic infection.7 Further studies involving complete genome characterization and amino acid modifications for each gene and their role in structure and function from a number of symptomatic and asymptomatic G10P strains will help with assessment of virulence determinants. Co-surveillance of animal and human strains and more comprehensive sequence analyses would provide a better understanding of the emergence and evolution of rotavirus strains. Host factors such as difference in susceptibility and response to infection, maternal protection and innate immune response may also play a role in determining disease.
The authors would like to thank Dr. Vito Martella and Dr. Jelle Matthijnssens for their help in 5′ and 3′ terminal sequencing. This work was supported by grants from the Indian Council for Medical Research and the Wellcome Trust Trilateral Initiative for Infectious Diseases, grant no. 063144.
Appendix ASupplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jcv.2009.05.003.