Comparative analysis of VP8* sequences from rotaviruses possessing M37-like VP4 recovered from children with and without diarrhoea

Rotavirus strains belonging to G types 1 to 4 and having a P3 genotype (M37-like VP4) were recovered from children with symptomatic and asymptomatic infections. Partial sequences of their VP4 genes were determined in an attempt to characterize these strains further. The genomic regions encoding VP8*, the connecting and putative fusion peptides and three other regions in VP5* were sequenced. The deduced amino acid sequences were compared with rotavirus strains belonging to different P genotypes that had been previously reported. High degrees of identity were found between the VP8* fragment of all human P3 strains (92.7 to 99.7%) suggesting that they belong to the same genotype, regardless of differences in their virulence. Furthermore, based on comparative sequence analysis, we did not identify any amino acid(s) that differ appreciably between symptomatic and asymptomatic strains and could therefore be associated with virulence. The results suggest that the P3 genotype, although frequently associated with asymptomatic infections, may not be the single determining factor in attenuation of symptoms.

VP4 genotyping of human rotavirus in the United States

The VP4 (P) and VP7 (G) types of 171 rotavirus isolates obtained from children with diarrhea in the United States were characterized by PCR typing assays. Strains P1G1 predominated (71%); this was followed by strains P1G3 (20%) and P2G2 and P1G4 (2% each). Mixed types were identified in five (3%) specimens. Two (1%) strains bearing the P3 genotype (P3G1 and P3G2) were found in children with severe dehydrating diarrhea, although the P3 genotype has been regarded as a possible marker for virus attenuation.

Amino acid sequence similarity of the VP7 protein of human rotavirus HCR3 to that of canine and feline rotaviruses

The sequence of the VP7 gene of human rotavirus strain HCR3 was determined and its predicted amino acid sequence was compared with that of other rotavirus strains. The VP7 gene is 1062 nucleotides long and contains a single open reading frame of 981 nucleotides capable of encoding a protein of 326 amino acids. The VP7 amino acid sequence similarity of strain HCR3 to those of various human and animal G3 serotypes ranged from 88.7 to 99.4%, and from 60.4 to 88.3% to strains representing each of the other 13 G serotypes. Alignment of four variable regions [VR4, VR5(A), VR7(B) and VR8(C)] of HCR3 with those of G3 strains of different host species showed that HCR3 possesses a sequence almost identical to that of canine rotaviruses and feline rotavirus strain CAT97 in all four regions. A considerable divergence in regions VR4, VR7(B) and VR8(C) was found with strains of human, mouse, monkey, horse and rabbit rotaviruses. This observation together with results of our previous study on VP4 indicated that human rotavirus HCR3 is genetically more closely related to animal rotaviruses than to other human rotaviruses.

Nucleotide sequence of the VP4-encoding gene of an unusual human rotavirus (HCR3)

Human rotavirus strain HCR3 was isolated from the stool of a clinically normal infant and identified as a serotype G3 rotavirus; however, it could not be grouped into any known human VP4 genetic groups by a polymerase chain reaction assay. The fourth gene of strain HCR3, which encodes the outer capsid protein VP4, was sequenced. This gene is 2362 nucleotides in length and contains one open reading frame capable of encoding a protein of 776 amino acids. The VP4 protein of strain HCR3 shared 67.5-73.5% amino acid identity with those of strains KU, RV-5, 1076, and K8, representing four human genetic groups, and relatively high homology (84.7%) with a fifth genetic group represented by strain 69M, whose VP4 shows more similarity to animal than to human strains. Strain HCR3 shared higher VP4 amino acid homology with various animal rotaviruses, ranging from 74.5 to 89.4%. These observations suggest that the VP4 outer capsid protein of strain HCR3 represents a new VP4 genetic group that is more closely related to animal rotaviruses than to human rotaviruses.

Murine rotavirus genes encoding outer capsid proteins VP4 and VP7 are not major determinants of host range restriction and virulence

Simian rotavirus (RRV) and murine rotavirus (EDIM-RW) differ dramatically in the oral inoculum required to cause diarrheal disease in neonatal mouse pups and in their ability to spread and cause disease in uninoculated littermates. A genetic approach was used to explore the molecular basis of these differences. Reassortant viruses were produced in vivo by coinfecting infant mice with RRV and EDIM-RW. Reassortant viruses were isolated by plaque purification of progeny virus obtained from mouse pup intestines on MA104 cells. The plaque-purified reassortants were evaluated for 50% diarrhea dose (DD50) and for the ability to spread and cause diarrhea in uninoculated littermates. The parental RRV strain had a DD50 of 10(5) PFU per animal, while the EDIM-RW parental strain had a DD50 of less than 1 PFU per animal. RRV never spreads from inoculated to uninoculated littermates and causes disease. Twenty-three reassortants were tested. Of great interest were the reassortants D1/5 and C3/2, which derived genes 4 and 7 (encoding VP4 and VP7) from RRV. These viruses had a DD50 similar or identical to that of EDIM-RW and spread efficiently from inoculated mouse pups to uninoculated pups. We conclude that the major outer capsid proteins VP4 and VP7 are not primarily responsible for virulence or host range restriction in the mouse model using a homologous murine rotavirus.

Group A rotaviruses produce extrahepatic biliary obstruction in orally inoculated newborn mice

Extrahepatic biliary atresia is a devastating disease occurring in 1 in 10,000 to 14,000 infants annually in the United States. We have recently described preliminary data suggesting an association of group C rotavirus with biliary atresia in two infants. However, a group C rotavirus animal model of biliary atresia is not presently available. On the other hand, some strains of the better-characterized and much more common group A rotaviruses produce hepatobiliary disease in infant mice. This disease shares many characteristics of the human infection. The present report describes extrahepatic biliary obstruction in immunocompetent BALB/c infant mice infected with a human or animal strain of group A rotavirus. Two-d-old BALB/c mice orally inoculated with hepatobiliary tropic rotavirus were shown to have active virus replication in the biliary tract and liver as early as 48 h postinoculation. At approximately 7 d postinoculation, between one fourth and one half of infant mice, depending on the virus strain, showed signs of inflammation and swelling in the bile ducts. The obstruction was complete in about one half of symptomatic animals. Although there was no obvious atresia as described in human infants, the obstruction was irreversible about 50% of the time, and the resulting fibrosis and bile ductular proliferation in the liver were strikingly similar to those seen in the liver of the human infant with biliary atresia.

Restriction endonuclease analysis of the vp7 genes of human and animal rotaviruses

The vp7 genes of 194 strains of group A rotaviruses representing all known G types were analyzed with three restriction enzymes by direct digestion of amplified cDNA copies or by deduction of the restriction patterns from known sequences. Mammalian rotavirus strains were classified into 28 restriction patterns consisting of combinations of the 6 profiles (s1 to s6) obtained by digestion with Sau96I endonuclease, 9 profiles (h1 to h9) obtained with HaeIII, and 15 profiles (b1 to b15) obtained with BstYI. Digestion with Sau96I and HaeIII identified restriction sites common to all, or almost all, rotavirus strains studied, whereas BstYI was the most discriminating among rotavirus strains. A clear correlation between some restriction patterns or individual profiles and G type and/or host species of origin was found. Several discriminatory restriction sites consisted of type-specific nucleic acid sequences that encoded conserved amino acid residues. Although not directly involved in antigenic diversity, these sites appear to indicate the G type of the isolate. The technique permits rapid comparison of a large number of virus isolates directly from fecal specimens and provides useful markers for investigating the evolution of rotavirus vp7 genes and tracing vaccine virus and interspecies transmission.

Prevalence of serum neutralizing antibody to serotype 9 rotavirus WI61 in children from South America and central Europe

Neutralizing serum antibody to serotype 9 rotavirus WI61 was detected in 41% of 870 Ecuadorian children and 26% of 140 German children. In both areas an age-related prevalence increase was observed. We identified 11 serum samples from Ecuadorian children which neutralized exclusively serotype 9 rotavirus. Thirteen of 71 (18%) German children hospitalized with serologically defined primary rotavirus gastroenteritis showed a seroconversion to serotype 9 rotavirus; however, in 10 of these 13 patients, the infecting serotype could be identified as serotype 1, 3, or 4. Furthermore, all 13 patients showed fourfold increases in titer to at least one further serotype.

Serotypes and electropherotypes of human rotavirus in the USA: 1987-1989

The epidemiology of rotavirus gastroenteritis was investigated for two consecutive seasons (1987-1988 and 1988-1989) in seven locales in the continental USA. The 281 representative fecal samples obtained from children with diarrhea were electropherotyped and serotyped by an enzyme immunoassay with serotype-specific monoclonal antibodies and a new amplification typing technique (polymerase chain reaction typing). Serotype 1 was predominant in both years, particularly in the North and East; serotype 3 was second in frequency and found most often in the South; serotype 2 was detected only occasionally; serotypes 4, 8, and 9 were never found. Rotavirus strains were grouped into five major electropherotypes, each corresponded to a single serotype, and the relative migration of the gene segments 7-9 could be used to distinguish serotype 1 from serotype 3. The amplification typing technique proved to be of great value in typing the 17% of rotavirus-positive specimens untypable by the serologic technique.

Serotype 1 reassortant of bovine rotavirus WC3, strain WI79-9, induces a polytypic antibody response in infants

A reassortant rotavirus, strain W179-9, was constructed bearing gene 9 of serotype 1 rotavirus strain WI79 and all other genes derived from bovine (serotype 6) rotavirus strain WC3. The antigenic phenotype of WI79-9 is bivalent: serotype 1 and serotype 6. WI79-9 administered orally at a dose of 10(7.5) p.f.u. induced no adverse effects in 48 infants of age 2-11 months. Serotype 1- and serotype 6-specific serum neutralizing antibody titres were induced with approximately equal frequency in these infants. Serotype 1-specific antibody responses were inhibited in infants previously seropositive to type 1. The immune response rate was enhanced by administration of a second, 'booster dose'.

Immune protection of infants against rotavirus gastroenteritis by a serotype 1 reassortant of bovine rotavirus WC3

The safety and protective efficacy of a serotype 1 reassortant of bovine rotavirus WC3, disignated strain WI79-9, was evaluated in a double-blind placebo-controlled trial. Rotavirus reassortant WI79-9 contains a gene segment 9 coding for the surface structural protein vp7 of a human serotype 1 rotavirus, with all other gene segments derived from WC3 rotavirus, which had previously been shown to be safe and immunogenic in infants. Infants 2-11 months of age were given two doses of vaccine (10(7.3) plaque-forming units/dose) or of placebo 28 days apart. Adverse reactions to the vaccine were not detected. The incidence of serum plaque reduction neutralization antibody responses to two doses of vaccine was serotype 6, 97%; serotype 3, 68%; and serotype 1, 22%. Active surveillance during the subsequent rotavirus season revealed 8 cases of rotavirus gastroenteritis in 39 placebo control infants and no cases in 38 WI79-9 vaccine recipients (protection = 100%, P = .003). Six cases of rotavirus gastroenteritis were caused by type 1 and two by type 3 virus. Although vaccination with WI79-9 affected only the incidence of rotavirus gastroenteritis, the vaccinated infants exhibited a significantly reduced incidence of total days of diarrhea, fever, and illness associated with gastroenteritis in general.

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

The rotavirus gene segment coding for the major outer capsid glycoprotein vp7 was amplified directly from stool specimens by the polymerase chain reaction (PCR). Double-stranded RNA extracted from stool samples was used as the template for reverse transcription, which was followed immediately and in the same reaction mix with amplification, using the Taq polymerase. Various conditions were examined to optimize the yield of the amplified gene. The concentrations of MgCl2, dimethyl sulfoxide, and template RNA were critical. The choice of primer pairs allowed amplification of the entire segment or specific portions. By using type-specific primers derived from distinct regions on the gene, we devised a PCR typing method in which each human serotype virus produced a characteristic segment size, readily identifiable in agarose gels. The PCR typing method was applied to 10 rotavirus reference strains, including all 6 known human serotypes (serotypes 1, 2, 3, 4, 8, and 9), and to 34 stool specimens previously serotyped by an enzyme immunoassay with monoclonal antibodies. An absolute correlation was found between the molecular and serologic methods. In addition, 14 stool specimens nonserotypable by an enzyme immunoassay with monoclonal antibodies could be typed by the PCR method. Besides the application for rotavirus detection and typing directly from stools, the PCR method provides a rapid and efficient means of obtaining large quantities of cDNA suitable for sequencing, cloning, and other genetic studies, precluding the need for cell culture and virus purification.

Reactogenicity and immunogenicity of rotavirus WC3 vaccine in 5-12 month old infants

Rotavirus is a major cause of acute gastroenteritis in infants worldwide and there is need for an effective vaccine. Rotavirus Wistar calf 3 (WC3) is a strain of bovine origin attenuated by 12 passages in cell culture. A lyophilized candidate vaccine containing 1 x 10(7) PFU of WC3 has been developed. An oral dose was given to 25 French infants 5-12-months old (mean age 8.6 months). No diarrhoea was observed within 2 weeks after vaccination. Unexplained vomiting was reported once and isolated fever greater than 37.8 degrees C was reported 3 times during the first week. One month later, a neutralizing antibody response to serotypes tested was shown in 88% of cases, with heterotypic responses to human serotype 3 (SA11 strain) in 72% and to type 1 (WA strain) in 48%. The percentage of immune response was similar whether the infant had antibody prior to immunization or not, but a booster effect was observed in children who had pre-immunization rotavirus antibodies. Considering these promising results, efficacy trials are in the planning in different parts of the world.

Discrepant rotavirus results in two laboratories using the same enzyme immunoassay

Rotavirus testing was performed on 66 stool samples in two separate laboratories using the same enzyme immunoassay. Sixteen of 46 of the positive results reported by the reference laboratory were negative in the hospital laboratory. In addition, RNA gel electrophoresis had negative results in the 16 discrepant samples. This indicates the need to repeat or confirm positive results.

Protective effect of WC3 vaccine against rotavirus diarrhea in infants during a predominantly serotype 1 rotavirus season

We used a double-blind, placebo-controlled trial to study the efficacy of WC3 rotavirus vaccine administered to 104 infants (ages, three to 12 months) before the rotavirus season. Forty-nine infants received vaccine; 55 received placebo. Rotavirus disease during this season was predominantly caused by a serotype 1 strain. In placebo recipients there were 14 cases of rotavirus diarrhea (attack rate, 25%); 11 were moderate to severe (attack rate, 20%). Vaccinees experienced only three cases of rotavirus disease (attack rate, 6.1%), all mild. When all cases (whether associated with rotavirus or not) of clinically significant diarrhea (CSD) were evaluated, WC3 vaccine provided statistically significant (P less than .01) protection against the total number of episodes of CSD and reduced the number of days of CSD-associated diarrhea, vomiting, fever, or illness. Seventy-one percent of the WC3-vaccinated infants had serum antibody responses to the vaccine. The 14 placebo recipients who experienced natural disease predominantly had antibody responses to serotype 1. Sera taken after the rotavirus season revealed a nearly identical rate (40%) of natural rotavirus infection in the vaccinated and placebo groups.

Rotavirus serotype-specific neutralizing activity in human milk

A plaque reduction neutralization assay was used to determine rotavirus serotype-specific neutralizing activity in human breast milk from 25 mothers of upper socioeconomic background and 20 mothers of a lower socioeconomic status. Levels of neutralizing activity, as well as those of rotavirus-specific antibodies detected by enzyme-linked immunosorbent assay (ELISA), were comparable for each socioeconomic group. Overall, neutralizing activity in human milk was detected in the majority of samples and can be increased many months postpartum. The prevalence of neutralization antibodies (titers, greater than or equal to 1:10) was 77% against Wa (serotype 1), 86% against SA-11 (serotype 3), and 75% against NCDV (bovine) rotavirus. Rotavirus-specific IgA and IgG antibodies detected by ELISA (titers, greater than or equal to 1:10) were present in 35% and 55% of breast milks, respectively. Sequential analysis of repeated breast milk samples from five individual mothers revealed that rotavirus neutralizing activity fluctuated over time, with high activity observed in one mother's milk at 18 months postpartum. Mothers who breast-fed for six months or more tended to have higher milk neutralizing titers against rotavirus.

Rotavirus isolate WI61 representing a presumptive new human serotype

A virus (strain WI61) representing a presumptive new human serotype was isolated from an 18-month-old child with gastroenteritis admitted to Children's Hospital of Philadelphia in February 1983. The WI61 virus was clearly distinguished by cross-neutralization tests from human rotaviruses of serotypes 1, 2, 3, and 4, human 69M, and representative bovine (NCDV), porcine (OSU), and chicken (Ch2) rotaviruses. Antisera generated in guinea pigs hyperimmunized to WI61 virus displayed a partial cross-reactivity with rotaviruses of human serotypes 1, 2, 3, and 4. By means of studies with reassortant rotaviruses, it was presumptively determined that the WI61 virus cross-reactive antigenic determinants are localized on the vp3 surface polypeptide coded by gene segment 4. The characteristic RNA genome electropherotype of WI61 virus was observed in 5 of 59 cases of infant gastroenteritis detected in 1983 and 1984 but has not been observed in a subsequently at Children's Hospital. Serotype WI61-specific neutralizing antibodies were observed in a majority of sera of normal adults and infants of less than 4 or greater than 12 months of age collected in the Philadelphia area. Median antibody titers to WI61 equaled or exceeded those to rotaviruses of serotype 1 or 3. Each of seven samples of commercial cow's milk exhibited neutralizing antibodies to WI61 virus at a titer greater than or equal to that to serotype 1 or 3 or bovine (strain NCDV) rotavirus. However, WI61 rotavirus did not induce disease or a specific serum-neutralizing antibody response when fed to a caesarean-derived colostrum-deprived newborn calf. WI61 rotavirus caused diarrhea in newborn mice with a 50% diarrhea-inducing dose of 10(7.0) PFU.

Gastroenteritis caused by human rotaviruses (serotype three) in a suckling mouse model

The pathogenic potential of human rotaviruses of serotypes 1 through 4 was evaluated in suckling mice. Oral inoculation of three different human rotaviruses of serotype 3 into 5-6 day old CD-1 mice caused disease characterized by diarrhea and dehydration. The mean 50% diarrhea inducing dose (DD50) was 5 X 10(5) pfu. Histopathological examination of small intestines revealed villus epithelial cell vacuolization localized to the distal one-third of the villus. Only Serotype 3 rotaviruses exhibited a rapid phase of viral growth in the intestine between 7 and 12 hours post-inoculation. Larger inocula of rotavirus serotypes 1, 2, and 4 did not cause disease or typical histopathologic changes. However, immunoperoxidase staining for rotavirus antigen was positive in all serotypes tested indicating that infection can occur without apparent disease and is not serotype specific. This convenient in-vivo model can be used to evaluate attenuation of human origin vaccine candidates of serotype 3.

Reassortant rotaviruses containing structural proteins vp3 and vp7 from different parents induce antibodies protective against each parental serotype

Genetic studies of reassortant rotaviruses have demonstrated that gene segments 4 and 9 each segregate with the serotype-specific neutralization phenotype in vitro. Reassortant rotaviruses derived by coinfection of MA-104 cells with the simian strain SA11 and the antigenically distinct bovine strain NCDV were used to determine which viral genes coded for proteins which induced a protective immune response in vivo. In addition, reassortant rotaviruses containing only the gene segment 4 or 9 protein products (vp3 and vp7, respectively) from SA11 or NCDV were used to determine the serotypic specificities of both vp3 and vp7 in several mammalian rotavirus strains. vp3 and vp7 from the murine strain Eb were shown to be indistinguishable from the corresponding proteins from strain SA11. Adult mice orally inoculated with strain Eb developed neutralizing antibodies to both vp3 and vp7. The two naturally occurring bovine rotavirus strains NCDV and UK were shown to contain antigenically similar vp7 but distinct vp3 proteins. Mouse dams orally immunized with a reassortant virus containing only gene 9 from NCDV passively protected their progeny against UK challenge, whereas mouse dams orally immunized with a reassortant virus containing only gene 4 from NCDV did not. Finally, we constructed reassortant viruses that immunized against rotaviruses of two distinct serotypes. SA11 X NCDV reassortants that contained vp3 and vp7 from different parents induced a protective immune response against both parental serotypes. vp3 and vp7 were independently capable of inducing a protective immune response after oral immunization. An understanding of the serotypic specificities of both vp3 and vp7 of human rotavirus isolates will be necessary for the development of successful strategies to protect infants against severe rotavirus infections.

Immune response of infants and children to low-passage bovine rotavirus (strain WC3)

A bovine rotavirus (strain WC3) was isolated from a calf in Pennsylvania and adapted to growth in continuous Cercopithecus cell line CV1. A pool for human vaccine trials was produced at the 12th cell culture passage level. After preliminary testing in adults and older children, a dose of 3 X 10(7) plaque-forming units was given by mouth to 52 infants and children aged 5 months to 6 years. No clinical sequelae were detected, and shedding in feces was detected in only 30% of tested infants. A serum-neutralizing antibody response was induced in 95% of 21 infants aged 5 to 11 months; response rates were slightly reduced in older infants. The antibody response was primarily directed toward bovine rotavirus, but a response to human serotype 3 rotavirus was also observed in approximately 50% of vaccinees. After vaccination with WC3, infants with preexisting antibody to rotaviruses of human serotype 1 or 3 frequently exhibited a booster response to those serotypes. WC3 is a candidate rotaviral vaccine deserving larger trials in children.

Response of adult human volunteers to oral administration of bovine and bovine/human reassortant rotaviruses

Small groups of adult volunteers, in sequence, were inoculated orally with inactivated purified bovine rotavirus of strain NCDV, with live NCDV purified or unpurified and with two different NCDV X human rotavirus reassortant viruses. One of five volunteers given 200 micrograms of ultravioletinactivated NCDV developed a virus-neutralizing (VN) and a binding antibody response detected by enzyme-linked immunosorbent assay (ELISA). Four of 10 volunteers given from 1 X 10(6) to 1 X 10(8) p.f.u. of live NCDV developed VN antibody, but nine of 10 responded when ELISA, HAI and radioimmuno-precipitation tests for serum antibody were also considered. Two different NCDV X human serotype 1 Wa strain virus reassortants, each containing Wa gene segment 9 and the serotype 1 neutralization phenotype, were administered orally in doses up to 10(6) p.f.u. The reassortants were relatively ineffective in eliciting a serum antibody response at the dosage level employed.

Molecular basis of rotavirus virulence: role of gene segment 4

Bovine rotavirus NCDV and simian rotavirus SA-11 exhibited markedly different patterns of gastrointestinal tract disease when inoculated orally into newborn mice. A genetic approach was used to define the molecular basis of these differences. The SA-11 strain of rotavirus was more virulent than the NCDV strain when inoculated orally into newborn mice; the dose of SA-11 required to cause diarrhea in 50% of infant mice was 50-fold less than that required for NCDV. Nineteen reassortant viruses were derived by coinfection of MA-104 cells in vitro with the SA-11 and NCDV strains. The parental origin of reassortant virus double-stranded RNA segments was determined by gene segment migration differences in polyacrylamide gels and hybridization with radioactively labeled parental viral transcripts. The neutralization antigen phenotype of reassortant viruses was determined by plaque reduction neutralization. We found that the dose of SA-11 and NCDV rotavirus required to induce gastroenteritis in newborn mice was determined by gene segment 4. The results suggest that rotavirus virulence may be manipulated by modification or reassortment of gene segment 4.

Rapid inactivation of rotaviruses by exposure to acid buffer or acidic gastric juice

Inactivation rates of three bovine and several primate-origin rotaviruses were determined during exposure to acid buffers at pH 2.0, pH 3.0 or pH 4.0. Each rotavirus was inactivated at pH 2.0 (the acidity most resembling the normal fasting stomach) very rapidly, with half-lives for infectivity determined to be 1 min or less. Each rotavirus was inactivated at a much slower rate at pH 3.0; inactivation at pH 4.0 was minimal. No remarkable differences in acid resistance between different rotavirus strains were detected. Although these determinations were performed at room temperature (23 degrees C), experiments at diverse temperatures indicated an even more rapid rate of viral inactivation by acid at normal body temperature (37 degrees C). Studies of rotavirus exposed to natural human gastric juice at pH 1.8 or pH 2.1 revealed a rate of virus inactivation similar to that observed with glycine buffer of identical pH.