Genetic Studies Of Rotavirus Pathogenesis And Development Of Rotavirus Vaccines

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications, trials & patents

Abstract

In the past year, we have advanced the objectives of this project as follows:[unreadable] (1) It is well known that multiple serial passages of a wild-type virulent virus in cell culture often results in the loss of virulence of the virus in a homologous host, which has been a traditional strategy in generating attenuated live viral vaccines including Rotarix vaccine. However, mechanisms underlying this phenomenon are not well understood. The purpose of this study was to monitor phenotypic and molecular characteristics of (i) a mouse-adapted virulent murine rotavirus following serial passages in cell cultures and (ii) a cell culture-adapted avirulent murine rotavirus following serial passages in mouse pups. [unreadable] We passaged a cell culture-adapted avirulent (non-diarrheagenic) murine rotavirus EB strain serially in 4-5 day-old CD-1 mice until the virus became virulent (diarrheagenic). This virulent virus was then inoculated onto primary African green monkey kidney (AGMK) cells and passaged serially 18 times. The 18th cell culture-passaged virus that was shown to be avirulent in mouse pups became virulent again during subsequent serial passage in CD-1 mice. This virulent virus when passaged again serially 18 times in primary AGMK cells became avirulent in mice. Sequencing of all 11 genes of selected virulent and avirulent viruses generated in this study, revealed that virulent viruses when compared with avirulent viruses bore (i) distinct mutations (residues 385, 537, 538 and 558) in gene 4 (encoding outer capsid VP4) and in gene 10 (residues 21 and 37) (encoding viral enterotoxin NSP4) and (ii) no significant mutations in other genes. In addition, no significant difference in virus replication efficiency in mice was observed between virulent and avirulent viruses. We generated 3 baculovirus recombinants expressing the NSP4 protein derived from high mouse-passaged virulent EB virus (sample A), low mouse-passaged virulent EB virus (sample B), or cell culture-passaged avirulent EB virus (sample C). Following oral inoculation of various recombinant NSP4 proteins in mouse pups, diarrhea developed in 91.6% (11/12) in sample A group; 12.5% (1 of 8) in sample B group; and 0% (0 of 12) in sample C group, indicating that the NSP4 protein played an important role in pathogenesis in this model. [unreadable] (2) Rotaviruses are ubiquitous in a wide variety of mammalian and avian species and, in general, highly species-specific. In addition, the genome RNA homology among rotavirus strains in a given animal species has been reported to be much higher than that of rotavirus strains derived from different animal species as demonstrated by RNA-RNA hybridization. However, there are a number of reports of atypical rotavirus strains isolated from humans and other animal species that share genetic and antigenic features of virus strains derived from heterologous species. Epizootiologic studies have demonstrated that rotaviruses bearing G3:P3 specificity are common in both dogs and cats. Interestingly, there have been at least 6 reports describing the detection of G3:P3 strains in humans: Ro1845, HCR3A and HCR3B, CMH222, PA260/97 and CRI33594. Among them, the Ro1845 and HCR3A strains, which were isolated in the mid-1980s, were analyzed fairly extensively with various assays (e.g., RNA-RNA hybridization, restriction fragment length polymorphism assay and sequencing of VP4 and VP7 genes). RNA-RNA hybridization assays, which demonstrated that each of these two human G3:P3 strains formed 10 or 11 hybrid bands versus certain canine and feline rotavirus strains bearing a G3:P3 specificity, suggested a possible transmission of whole rotavirus virions of canine/feline origin to humans. We sequenced and analyzed all 11 genes of strains Ro1845 and HCR3A as well as 3 canine strains (CU-1, K9 and A79-10, each with G3:P3 specificity) and 2 feline strains (Cat97 with G3:P3 specificity and Cat2 with G3:P9 specificity), and demonstrated that (i) these 2 human G3P3 strains were indeed examples of whole virion transmissions of canine/feline rotaviruses to humans; and (ii) each of 11 cognate genes of the 6 G3:P3 strains shared a characteristically high nucleotide identity and formed a distinct phylogenetic cluster suggesting that canine and feline rotaviruses with G3:P3 specificity bore highly conserved species-specific genomes that were distinct from genomes of human, simian, bovine, porcine or avian rotavirus origin. [unreadable] Although each of 11 cognate genes of the 6 G3:P3 strains analyzed in the present study displayed high sequence identity (94-99% nt and 94-100% aa), the clinical symptoms of the hosts from which these strains were isolated varied depending upon the strain. The HRV Ro1845, CRV K9 and FRV Cat97 strains were derived from diarrheal hosts whereas the HRV HCR3A and CRV CU-1 strains were isolated from asymptomatic hosts. By using a genetic approach, we and others showed previously that at least 6 genome segments (i.e., genes encoding VP3, VP4, VP7, NSP1, NSP3 and NSP4) were involved in pathogenesis of rotaviruses. Therefore, it was of interest to analyze the involvement of these 6 virulence-associated genes in the observed differences in clinical signs of the hosts by comparing whole genome nucleotide sequences between 3 symptomatic (Ro1845, K9 and Cat97) and 2 asymptomatic (HCR3A and CU-1) strains. Analyses revealed the existence between the two groups of (i) 11 mutations in the VP7 gene, (ii) 6 mutations in the VP3 gene, (iii) 4 mutations in the NSP4 gene, (iv) 2 mutations each in the VP1, VP4 and NSP3 genes, (v) 1 mutation each in the VP2, NSP1 and NSP5 genes, and (vi) no mutations in the VP6 and NSP2 genes. Although the numbers of symptomatic and asymptomatic infections were too few to draw a significant association, comparative whole genome sequence analyses were in agreement with previous genetic studies, since there were several mutations in the virulence-associated genes and a few mutations or no mutations at all in the genes not associated with virulence. [unreadable] The Ro1845 and HCR3A strains (which were shown to be of canine/feline origin in this study) were recovered from a heterologous host (i.e., humans) whereas the CU-1, K9, A79-10 and Cat97 strains were recovered from a homologous host (i.e., dogs or cats). In order to analyze whether the canine/feline rotaviruses acquired any mutation(s) during the replication in a heterologous host, we compared whole genome nucleotide sequences of the 4 strains (i.e., CU-1, K9, A79-10 and Cat97) recovered from a homologous host to those of the 2 strains (i.e., Ro1845 and HCR3A) isolated from a heterologous host. The following mutations were detected in the 2 groups: (i) 2 each in the VP2, VP3 and NSP4 gene, (ii) 1 in the NSP1 and NSP3 gene, and (iii) none in the VP1, VP4, VP6, VP7, NSP2 and NSP5 genes. Since no mutations were detected in the VP1, VP4, VP6, VP7, NSP2 and NSP5 genes in the 2 groups, these proteins did not appear to restrict human (heterologous) infection by canine/feline G3:P3 viruses.