Development of a Rotavirus-Shedding Model in Rhesu
http://www.100md.com
病菌学杂志 2005年第2期
Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
Tulane National Primate Research Center, Covington, Louisiana
ABSTRACT
Although there are several reports on rotavirus inoculation of nonhuman primates, no reliable model exists. Therefore, this study was designed to develop a rhesus macaque model for rotavirus studies. The goals were to obtain a wild-type macaque rotavirus and evaluate it as a challenge virus for model studies. Once rotavirus was shown to be endemic within the macaque colony at the Tulane National Primate Research Center, stool specimens were collected from juvenile animals (2.6 to 5.9 months of age) without evidence of previous rotavirus infection and examined for rotavirus antigen. Six of 10 animals shed rotavirus during the 10-week collection period, and the electropherotypes of all isolates were identical to each other but distinct from those of prototype simian rotaviruses. These viruses were characterized as serotype G3 and subgroup 1, properties typical of many animal rotaviruses, including simian strains. Nucleotide sequence analysis of the VP4 gene was performed with a culture-grown isolate from the stool of one animal, designated the TUCH strain. Based on both genotypic and phylogenetic comparisons between TUCH VP4 and cognate proteins of representatives of the reported 22 P genotypes, the TUCH virus belongs to a new genotype, P[23]. A pool of wild-type TUCH was prepared and intragastrically administered to eight cesarean section-derived, specific-pathogen-free macaques 14 to 42 days of age. All animals were kept in a biocontainment level 2 facility. Although no diarrhea was observed and the animals remained clinically normal, all animals shed large quantities of rotavirus antigen in their feces after inoculation, which resolved by the end of the 14-day observation period. Therefore, TUCH infection of macaques provides a useful nonhuman primate model for studies on rotavirus protection.
INTRODUCTION
Rotaviruses are the primary cause of severe diarrhea in young children and are responsible for ca. 500,000 deaths in the world each year (35). Several vaccines against rotavirus disease have been developed, but none are available for routine immunization (48). All candidate vaccines evaluated in clinical trials have been live, attenuated rotaviruses that are delivered orally to mimic natural infection. Although natural rotavirus infection has been found to provide substantial protection against rotavirus disease, particularly severe disease, the mechanisms by which this occurs remain largely undetermined (45). A number of animal models have been developed to help understand the mechanisms of rotavirus immunity, including lambs (40), calves (4, 12, 33, 53, 54), piglets (3, 5, 17, 18, 56, 57), rabbits (7-10), and rats (6, 16), but the majority of the mechanistic studies have been conducted with the adult mouse model that we developed in 1990 (51). Even with these models, only a partial understanding of the mechanism of rotavirus protection has been achieved. Furthermore, an animal model that more closely resembles humans is clearly needed, both to understand the mechanisms of rotavirus immunity and for preclinical evaluations of novel rotavirus vaccine candidates.
Nonhuman primates are the animals most closely related to humans. Although several rotavirus challenge studies have been conducted with these animals, the first of which was reported in 1976 (55), very little is known about rotavirus infections in any nonhuman primate species. Several investigators have reported that oral inoculation of different nonhuman primates, including several types of monkeys as well as baboons, with either culture-adapted simian (SA11) or human (Wa) rotavirus or fecal preparations of human rotaviruses, will cause diarrheal illness in most animals during their first week of life (21, 23, 26, 31, 36, 41, 55). However, after that time, essentially no illness was observed, and most of the older animals neither shed virus nor seroconverted. The exception may have been the results found after inoculation of one chimpanzee that, when orally administered SA11 at 141 days of age, developed diarrhea and shed large amounts of rotavirus over a 9-day period (41).
A potential major limitation in these studies was the lack of a challenge virus that would reliably infect and produce illness in the nonhuman primate species under investigation. The only simian rotavirus used in these studies (SA11) was obtained from an asymptomatic vervet monkey (27) and has been repeatedly passaged in cultured cells, a method typically used for viral attenuation. The only other rotaviruses used were human strains. Although rotaviruses sometimes cross species barriers (32), their virulence is typically blunted in heterologous hosts. Thus, these human strains, even those from human fecal specimens, are likely to be naturally attenuated in nonhuman primates. There has been no report of experimental inoculation of a nonhuman primate being performed with a wild-type (fecally derived) simian rotavirus. Therefore, it was possible that wild-type simian rotaviruses may produce severe illness in nonhuman primates after the first week of life, perhaps even up to several years of age as occurs in humans. The first purpose of this study was to obtain and characterize such a virus from rhesus macaques to be used for challenge studies in nonhuman primates. The second purpose was to determine whether this new strain of simian rotavirus could consistently produce high levels of fecal rotavirus shedding and possibly elicit diarrheal disease in macaques when administered after the first week of life. If either or both outcomes could be attained, this model system could potentially provide a highly relevant substitute for humans in studies on active immunity against rotavirus.
MATERIALS AND METHODS
Rhesus macaques: sample collections and illness monitoring. (i) Conventional animals. A group of 16 juvenile rhesus macaques (Macaca mulatto), age 2.6 to 5.9 months at the time of their enrollment on 12 November 2002, were included in a study to obtain wild-type rotavirus from naturally infected animals. All animals were born at the Tulane National Primate Research Center (TNPRC) and were housed under biosafety level 2 conditions in accordance with standards of the Guide for the Care and Use of Laboratory Animals and the Association for Assessment and Accreditation of Laboratory Animal Care. Only simian retrovirus (simian immunodeficiency virus, simian retrovirus, and simian T-cell leukemia virus)-negative animals were used. Every animal was monitored for diarrhea and signs of lethargy on a daily basis. Initially, blood samples were collected and sera were harvested and stored (–20°C) until analyzed for rotavirus immunoglobulin G (IgG) and IgA. Once these were determined, the 10 animals with the lowest titers were monitored for rotavirus shedding from 25 November 2002 through 6 February 2003. For this, stool samples were collected twice per week from the bottom of the cage containing each animal and stored at –80°C until tested for the presence of rotavirus antigen. Monthly blood collections were also taken from these 10 animals, and the harvested sera were stored (–20°C) until analyzed for rotavirus IgG.
(ii) Cesarean section-derived animals. Two groups of cesarean section-derived macaques were used. All were lactogenic immunity deprived (hand fed and free of maternal secretory IgA) and housed in nursery (biosafety level 2) rooms that contained only specific-pathogen-free juvenile animals obtained in this manner. At the time of cesarean section, the level of maternal serum rotavirus IgG ranged between 300 and 5,600 U/ml, where the limit of detection was 5 U/ml. The first group contained five animals that were monitored for up to 7 months of age for evidence of a natural rotavirus infection based on increases in their titers of serum rotavirus IgG or IgA between blood collections. The second group (eight animals) were bled at the time of birth and just before the time of rotavirus challenge, and rotavirus IgG and IgA levels in these serum specimens were measured. Between 14 and 42 days after birth, these animals were transferred to a separate room in the facility and intragastrically challenged with 3 x 104 focus-forming units (FFU) of our new wild-type strain of macaque rotavirus following administration of 2 ml of 4% sodium bicarbonate to buffer stomach acidity. Stool specimens were collected between one and four times each day for the next 14 days, and the levels of rotavirus shedding were quantified. In addition, the consistencies of these stool specimens were graded for detection of diarrhea. During this period, the macaques were evaluated each day for signs of lethargy as an additional indicator of illness.
Quantitation of rotavirus IgG and IgA in serum specimens. The levels of rotavirus IgG and IgA were determined in serum specimens by enzyme-linked immunosorbent assays (ELISA) essentially as described previously (1, 46). In brief, 96-well microtiter plates were coated with purified, double-layered simian rotavirus SA11 or with mock-infected lysate purified in an identical fashion. This was followed by the stepwise addition of diluted specimen or control sera, biotinylated goat anti-monkey (rhesus) IgG (Research Diagnostics, Inc., Flandus, N.J.), peroxidase-conjugated avidin-biotin (Vector Laboratories, Inc., Burlingame, Calif.), and substrate (orthophenylenediamine [Sigma Aldrich, St. Louis, Mo.]). Optical density (OD) values were determined at 490 nm. The levels were then expressed as units of IgG per milliliter, based on a standard curve generated from a human serum pool arbitrarily assigned a level of 1,000 U/ml. The lower limit determinable by this assay under the conditions used was 5 U/ml. Rotavirus IgA was measured in the same manner, except that biotinylated goat anti-monkey IgA (Research Diagnostics, Inc.) was used. In this case, the control serum (human pool) was assigned a value of 250 U/ml, and the lower limit measurable was 10 U/ml.
Detection and quantitation of rotavirus antigen in stools of macaques. Frozen stools collected from the cages of individually housed macaques were thawed, made into 20% (wt/vol) suspensions with Earle's balanced salt solution (EBSS), and tested for the presence of rotavirus antigen by ELISA as described elsewhere (30, 46). For the group of conventional animals, shedding was quantified as optical density units (A490), and the highest OD value measurable in the rotavirus-positive specimens was 3.0. Therefore, the quantity of rotavirus antigen present in these specimens was expressed as OD measurements up to 3.0. For cesarean section-derived macaques inoculated with the new strain of wild-type macaque rotavirus, fecal shedding of rotavirus antigen was quantified by ELISA as nanograms per milliliter of 20% suspensions of stool, based on a standard curve generated by using purified double-layered particles of murine rotavirus strain EDIM.
Quantitation of infectious rotavirus in stool specimens. Rotaviruses present in 20% suspensions of macaque stools were quantified by a fluorescent-focus assay (47). Titers are expressed as FFU per milliliter of the 20% stool suspensions.
Cell culture adaptation of rotavirus from macaque stools. Rotaviruses in stool were adapted to grow in MA104 (monkey kidney) cells by using methods similar to those previously reported (30). In brief, 20% (wt/vol) stool suspensions and dilutions of these samples (in EBSS) were treated with trypsin (15 μg/ml; GIBCO Laboratories, Grand Island, N.Y.) (1:250) to activate the virus. Following centrifugation (1,200 x g, 20 min) to remove debris, the supernatants were added to tube cultures containing monolayers of MA104 cells. After adsorption (2 h) on a roller apparatus, the cultures were washed and medium without serum, but with trypsin, was added. The culture tubes were rolled for 4 days or until cytopathic effects reached 3+ (whichever came first) and then were stored at –20°C. Subsequent passages were conducted without trypsin pretreatment of the lysate from the previous passage. In this study, the presence and quantity of rotavirus were determined after two passages in MA104 cells, using the same ELISA procedure as described for detection of rotavirus in stool specimens.
Electrophoretic analysis of viral RNA segments. The double-stranded RNA genome segments of passage 2 of the culture-adapted macaque rotaviruses were extracted from viral lysates and analyzed by polyacrylamide gel electrophoresis (50). The electropherotype of one of the isolates was compared to those of prototype simian rotaviruses RRV and SA11, also grown in MA104 cells.
G serotyping of the culture-adapted macaque rotaviruses. The G serotypes of passage 2 of the culture-adapted macaque rotaviruses were determined by an ELISA with prototype G1 to G4 (Wa, DS-1, P, and ST3, respectively) human rotavirus strains as controls (15, 49). VP7-specific monoclonal antibodies (MAbs) against G1 (MAb 5E8), G2 (MAb 1C10), G3 (MAb 159), and G4 (MAb ST3:1) rotavirus strains were used to coat 96-well microtiter plates (capture antibodies), and hyperimmune guinea pig antisera against G1 to G4 prototype human rotaviruses were used as the detector antibodies. The MAbs against G1 to G3 strains were gifts of H. B. Greenberg (Stanford University, Palo Alto, Calif.) and the ST3:1 MAb was a gift of B. Coulson (University of Melbourne, Melbourne, Australia).
Subgroup analysis of the culture-adapted macaque rotaviruses. Subgroup determination of passage 2 of the culture adapted macaque rotaviruses was performed by an ELISA with subgroup-specific MAbs 225/60 (subgroup 1) and 631/9 (subgroup 2) (15, 29, 50, 52). Both MAbs were gifts of H. B. Greenberg. Control rotaviruses included in this assay were the simian RRV (subgroup 1 [SG1]) and human Wa (SG2) strains. Both control rotaviruses reacted only with their respective subgroup-specific MAb.
Nucleotide sequencing of the VP4 and VP6 genes of the TUCH strain of macaque rotavirus. Double-stranded cDNAs of genes 4 and 6 of CsCl gradient-purified, culture-adapted rotavirus (passage 3) obtained from the stool of macaque EA40 (specimen date, 2 January 2003), subsequently named the TUCH strain, were generated by reverse transcription-PCRs (RT-PCR) with the Thermo Script RT-PCR system (Invitrogen/Life Technologies, Carlsbad, Calif.) and Vent DNA polymerase (New England BioLabs, Beverly, Mass.). Briefly, RT was performed with 1 μg of purified double-stranded genomic RNA in 25 μl of distilled water containing an RNase inhibitor (20 U of RNaseOut [Invitrogen Life Technologies]) and a 200 nM concentration of a forward primer (gene 4, 5'-GGC TAT AAA ATG GCT TCG CTC-3' [corresponding to nucleotides 1 to 21]; gene 6, 5'-GGC TTT TAA ACG AAG TCT TC-3' [corresponding to nucleotides 1 to 20]) and a reverse primer (gene 4, 5'-GGT CAC ATC CTC TAG AAA TTA C-3' [corresponding to nucleotides 2342 to 2362]; gene 6, 5'-GGT CAC ATC CTC TCA CTA CGG CAT TC-3' [corresponding to nucleotides 1331 to 1356]) designed from the noncoding regions of the simian SA11 rotavirus genes. The reaction mixture was denatured (3 min, 94°C), and the primers were allowed to rapidly anneal to viral RNA by cooling the suspension on ice. Next, 4.0 μl of 5x cDNA synthesis buffer (Invitrogen Life Technologies), 1.0 μl of 0.1 M dithiothreitol, 1.0 μl of RNaseOut (40 U/μl), 1.0 μl of diethyl pyrocarbonate-treated water, and 1.0 μl of ThermoScript reverse transcriptase (15 U/μl) were added to the reaction mixture on ice. Reverse transcription was carried out at 50°C for 60 min in a heating block. The reaction was terminated by transferring the reaction mixture in a heating block set at 85°C for 5 min. The reaction mixture was cooled, 2 U of RNase H in 1 μl was added, and the reaction mixture was incubated at 37°C for 20 min. PCR was carried out in a 50-μl volume containing 2.0 μl of the RT-PCR products, 5 μl of 10x amplification buffer, 1.5 μl of 10 mM deoxynucleoside triphosphates, 1.0 μl of 50 mM MgSO4, 1.5 μl of forward and 1.5 μl of reverse primer (200 nM each), and 2.5 U of Platinum Pfx DNA polymerase (Invitrogen Life Technologies). The template was denatured (94°C, 2 min), and then 30 cycles of PCR were carried out. Each PCR cycle consisted of a denaturation step (94°C for 30 s), an annealing step (55°C for 30 s), and an extension step (68°C for 3 min for gene 4 and for 2 min for gene 6). After the PCR cycle, an additional incubation step was carried out (72°C, 7 min). PCR products were purified by using the QIAquick PCR purification kit (Qiagen Inc., Valencia, Calif.). The nucleotide sequences of the VP4 and VP6 genes were determined by using purified double-stranded cDNAs generated by RT-PCR as templates. Sequencing was performed by Cleveland Genomics (Cleveland, Ohio), using an ABI DNA sequencer (Applied Biosystems, Foster City, Calif.). The primer pair used for generating double-stranded DNA was also used to sequence the initial 500 nucleotides from the ends of the positive and the negative cDNA strands. New primer pairs were designed by using the sequence generated until cDNA strands were completely sequenced. The gene sequence was determined twice with PCR products generated in separate reactions, using both strands as templates.
Phylogenetic relationship of TUCH rotavirus VP4 and VP6 gene sequences with those of other rotavirus strains. CLUSTALW (http://www.ebi.ac.uk/clustalw/), a program that is commonly used for phylogeny reconstruction, was employed to analyze phylogenetic relationships of the deduced complete amino acid sequences of the TUCH VP4 and VP6 proteins with those of cognate proteins of other rotaviruses. This program predicts phylogenetic trees by comparing alignment scores for all possible pairs of the rotavirus VP4 and VP6 sequences. The rotavirus gene sequences used for pairwise alignment were first retrieved from protein databases by using WU-BLAST2 (http://BLAST.wustl.edu). The sequences were then aligned by using CLUSTALW to search for regions of similarity among the sequences. The fundamental unit algorithm output of BLAST is the high-scoring segment pair, which consists of two sequence fragments of arbitrary but equal length whose alignment is locally maximal and for which the alignment score meets or exceeds a threshold (the default cutoff score was chosen). In the analyses performed here, the default Blosum62 was used for the scoring matrix. Unrooted phylogenetic trees were constructed by using the neighbor-joining method (38). The trees are displayed with TreeView (34).
Preparation of a challenge pool of unpassaged TUCH rotavirus. To prepare a pool of unpassaged simian rotavirus to be used for subsequent challenge studies, the stool obtained on 2 January 2003 from macaque EA40 was processed. Although the rotaviruses from all five macaques examined had identical electropherotypes and therefore were likely to be very similar strains, only the rotavirus in this one stool specimen has been given the designation TUCH, and only this virus was used in subsequent experiments under that designation. This stool was used because only it and one other, of the seven stools examined, contained a high titer of infectious rotavirus, and it had the larger quantity of the two. A 5% (wt/vol) suspension of this entire stool (3.2 g) was made in EBSS by shaking (2 h, 4°C) with glass bends on an automated shaker. After centrifugation (1,200 x g, 20 min, 4°C) to clarify the suspension, the supernatant was filtered (0.2-μm-pore-size filter) and stored (–70°C) in 1-ml aliquots. The titer of the original 5% suspension was 3.4 x 105 FFU/ml, while the frozen aliquots contained 3.0 x 104 FFU/ml (9% recovery).
Nucleotide sequence accession numbers. The GenBank accession numbers for the TUCH VP4 and VP6 protein genes are AY596189 and AY594670, respectively.
RESULTS
Isolation of rotavirus from naturally infected rhesus macaques. (i) Identification of juvenile macaques with no evidence of previous rotavirus infection. Before attempting to obtain rotavirus from naturally infected rhesus macaques housed within open cages in containment rooms at the TNPRC, it was first necessary to demonstrate that rotavirus is endemic in this macaque population. For this, we relied on serum rotavirus IgG measurements. Upon examination of a group of 16 juvenile macaques age 2.6 to 5.9 months and housed within the facility, it was noted that 5 had high titers of rotavirus IgG (geometric mean titer, 704 U/ml) and that the other 11 had much lower levels (geometric mean titer, 13 U/ml) (Table 1). Because the average ages of the two groups of macaques were comparable (5.2 versus 5.0 months), it was assumed that at least the five animals with high titers of rotavirus IgG had experienced at least one natural rotavirus infection. It is likely that rotavirus IgG detected in the other 11 macaques was residual transplacental antibody. These suggestions were supported by the finding that the 5 macaques with the highest levels of rotavirus IgG had detectable quantities of serum rotavirus IgA, while only 1 of the 11 macaques with little or no detectable rotavirus IgG had detectable rotavirus IgA (Table 1). These results established that rotavirus infections were, indeed, endemic within this facility. They also identified animals that had no evidence of a previous rotavirus infection based on their low titers of rotavirus IgG and, therefore, that may be more susceptible to a natural infection within the following weeks.
(ii) Detection of rotavirus in stools of naturally infected macaques. To obtain rotavirus from naturally infected macaques, stool specimens were collected two times each week from the 10 macaques with the lowest titers of serum rotavirus IgG, and with undetectable rotavirus IgA, for a period of 10 weeks beginning 13 days after they were initially screened (12 November 2002) for rotavirus antibody. Blood specimens were also collected monthly over a 27-week period to detect changes in serum rotavirus IgG titers. Rotavirus antigen was detected in 6 of the 10 macaques in at least one stool specimen (Table 2). Corresponding increases in rotavirus IgG titers were detected in every case. No gastrointestinal illness (vomiting or diarrhea) was observed in any animal in association with rotavirus detection.
In every animal in which rotavirus was detected, the quantity shed was sufficiently large so that the optical density determined by ELISA was off the scale (>3.0) in at least one specimen (Table 2). To determine the recoveries of live rotavirus in these stools, the titers of infectious virus in the specimens with the highest optical densities obtained from the first five animals that shed rotavirus were measured by a fluorescent-focus assay. Based on the numbers of focus-forming units per milliliter of stool suspension (20%, wt/vol), these titers varied widely, and only two specimens had titers of >106 FFU/ml (Table 2). Differences in titers between specimens were probably due primarily to differences in specimen conservation, since the stools were obtained in various stages of dryness from the bottoms of the cages.
Characterization of rotaviruses obtained from stool specimens of naturally infected macaques. (i) Electropherotypes of culture-adapted rotaviruses. Eight stool specimens from five macaques for which the titers of infectious rotavirus were determined were also processed and used to infect MA104 cells in an attempt to culture adapt the rotaviruses in these specimens. A complete (4+) cytopathic effect was observed within 48 h in all culture tubes inoculated with the rotavirus-positive stool preparations, and, when evaluated for the presence of rotavirus by ELISA after two cell culture passages, all eight tissue culture lysates contained large quantities of virus (i.e., all had optical densities of >3.0, even the lysate obtained from the specimen with an infectivity titer of <2.8 x 102 FFU/ml). Thus, rotaviruses in these specimens required no adaptation to grow in cell culture.
When the double-stranded RNA genomes of these eight culture-adapted isolates (passage 2) were analyzed by polyacrylamide gel electrophoresis, all had identical electropherotypes (Fig. 1A). In addition, the electropherotype of a representative of the group (macaque EA40) was compared to those of simian rotaviruses SA11 and RRV, and all three electropherotypes were found to be distinctly different (Fig. 1B).
(ii) VP7 serotypes and VP6 subgroups of the culture-adapted rotaviruses obtained from macaques. The VP7 or G serotype of simian rotaviruses previously characterized has been found to be G3 (11). We utilized serotype-specific G1 to G4 monoclonal antibodies against VP7 in an ELISA format in an attempt to determine the G type of rotaviruses in the eight culture-adapted preparations (passage 2) and found that all eight reacted strongly and specifically with the G3 MAb 159 (results not shown).
The intermediate capsid layer of the rotavirus particle is composed of the VP6 protein, the group antigen (8). Within group A rotaviruses, four subgroups have been identified based on their reactivities with subgroup-specific MAbs that recognize epitopes within the VP6 protein (13, 15, 24, 25, 43). These include SG1, SG2, SG1/SG2, and non-SG1/SG2. It has also been reported that animal rotaviruses typically belong to SG1. In contrast, most human rotaviruses belong to SG2. In an ELISA format, passage 2 of the culture-adapted rotaviruses obtained from macaques reacted strongly with MAb 225/60 to SG1 and had no reaction with the SG2 MAb 631/9 (results not shown).
Based on these results, the rotaviruses obtained from macaques in this study were classified as serotype G3 and subgroup 1, as has been found for previously characterized simian rotavirus strains. The rotaviruses isolated from all five macaques characterized in this study appeared to be similar if not identical. However, in order to maintain consistency, all further studies were conducted with virus obtained from one specimen of one macaque, i.e., EA40. The virus obtained from this animal was named TUCH for Tulane University and Cincinnati Children's Hospital.
(iii) Phylogenetic analysis of its VP4 protein indicates that TUCH belongs to a new P genotype. Although the serotype and subgroup of the macaque rotavirus were found to be typical of simian rotaviruses and its electropherotype clearly distinguished it from prototype simian rotavirus strains SA11 and RRV, it was still unclear whether TUCH was a previously unidentified rotavirus strain. To ensure that it is distinct from the other rotavirus strains, nucleotide sequence analysis of the VP4 gene of a culture-grown isolate (three passages) from macaque EA40 was performed, and the deduced amino acid sequence was compared to those of representatives of the 22 established P genotypes by phylogenetic analyses (11, 28, 37). The greatest amino acid homologies between macaque rotavirus and the other VP4 proteins were with representative P[3] rotaviruses, which included the simian RRV strain (Table 3). Homologies with these strains (85.3 to 86.2%) were, however, very similar to those found with the P[10] 69 M human strain (85.4%) and the P[2] SA11 simian strain (84.4%). VP4 proteins of representatives of six other genotypes also showed >80% homology with the macaque rotavirus VP4, suggesting that the point of divergence from the macaque VP4 protein may have been similar for many rotavirus strains.
To further examine this possibility, phylogenetic analysis of the VP4 proteins of the TUCH macaque rotavirus and representatives of the 22 previously established P types was performed, followed by the construction of a phylogenetic tree (Fig. 2). From this analysis it appears that the TUCH VP4 gene diverged from those of other genotypes at a distant point in time. This, coupled with the fact that none of these representative strains share >89% amino acid homology with the VP4 protein of the new macaque rotavirus (the suggested requirement for inclusion within the same VP4 genotype [14]), indicates that the TUCH strain belongs to a new P genotype, which we tentatively classify as P[23]).
(iv) The gene encoding TUCH VP6 also diverges from VP6 genes of other rotavirus strains. The VP6 protein or group antigen is highly conserved, especially within group A mammalian rotaviruses (48). However, because of the divergence identified between the VP4 genes of TUCH and other rotavirus strains, it was of interest to determine whether this also occurred with another TUCH gene by using phylogenetic analysis. The greatest amino acid homology between TUCH VP6 and that of another rotavirus VP6 protein deduced from nucleotide sequence analysis was determined to be 95.5% (Table 4). This degree of homology, however, is the same or very similar to that shared with VP6 proteins of multiple rotavirus strains, again suggesting that the TUCH strain diverged from other rotaviruses at a distant point in time. A phylogenetic tree constructed from the amino acid sequences of these VP6 proteins indicates that the TUCH VP6 gene, like the VP4 gene of the virus, has diverged from all other rotavirus strains examined (Fig. 3).
Inoculation of young macaques with the wild-type TUCH strain to establish a nonhuman primate model for studies on active immunity. (i) Cesarean section-derived macaques can be maintained rotavirus free within the TNPRC. Because rotavirus was found to be endemic within the TNPRC, before conducting a rotavirus challenge study it was important to demonstrate that macaques housed in the TNPRC could be maintained without experiencing a natural rotavirus infection. It was already demonstrated that animals housed within conventional rooms within this facility became naturally exposed to rotavirus and that the TUCH strain was derived from one of these animals. To determine whether newborn macaques could be kept rotavirus free if derived and maintained in the absence of older animals, a group of five macaques was obtained by cesarean section and monitored until the ages of between 3.5 and 7 months for evidence of natural rotavirus infection based upon increases in serum rotavirus antibody levels. All animals experienced a steady decline in both serum rotavirus IgG and IgA from the date of birth (cord blood) until the end of the observation period (Table 5), thus suggesting that macaques could be maintained rotavirus free within the TNPRC if derived and raised under these conditions.
(ii) Inoculation of young, rotavirus-naive rhesus macaques with the TUCH strain. A 5% suspension of stool from macaque EA40 was filtered and stored frozen in 1-ml aliquots. The titer of this preparation was 3 x 104 FFU/ml, and 1 ml was given intragastrically to each of eight cesarean section-derived rhesus macaques between 14 and 42 days of age following intragastric administration of 2 ml of 4% sodium bicarbonate. The titers of serum rotavirus IgG on the day of challenge varied from animal to animal, but all had declined from the titers found on the day of birth (Table 6). Soft stools were detected during some collections in several animals during the 14 days after challenge, but none were indicative of diarrhea. Furthermore, the behavior of all macaques remained normal during this period, without evidence of lethargy. All eight macaques shed large quantities of rotavirus in their stools, beginning between 1 and 5 days after challenge and declining to undetectable or nearly undetectable levels by the end of the 14-day observation period (Fig. 4). Even so, only three of the eight inoculated animals had detectable increases in serum rotavirus IgG titer by 1 month after inoculation (Table 6), possibly due, at least in part, to maternal antibody.
DISCUSSION
Since nonhuman primates are the animals most related to humans, it follows that they should be the most reliable for model studies using human disease agents. Accordingly, many studies on the pathogenesis of and immunity to such agents have been performed with nonhuman primates. Very few studies, however, have been conducted with rotavirus in any nonhuman primate species (21, 23, 26, 31, 36, 41, 55). A major limitation for performing such studies has been the absence of a challenge strain of simian rotavirus that will consistently result in productive infections in a nonhuman primate species after oral challenge. The only simian rotaviruses in use today are the SA11 strain, reported by Malherbe and Harwin in 1963 (27), and the RRV strain, discovered by Stuker et al. in 1980 (42). Both have been passaged numerous times in cell culture, a method commonly used for viral attenuation, and no unpassaged (wild-type) simian rotavirus is presently available. The first purpose of this study was to obtain and characterize such a virus and prepare a challenge pool for use in a nonhuman primate model for rotavirus.
Although rotaviruses are not entirely species specific (19, 32), the vast majority of natural isolates obtained from any species, including humans, have been associated specifically with that species. Since the nonhuman primates most commonly used as experimental models are macaques, the goal was to obtain a challenge rotavirus strain that is endemic in macaques and fully adapted to infect this specific nonhuman primate. Therefore, we first determined whether rotavirus was endemic in rhesus macaques housed at the TNPRC. Based on the presence of high titers of serum rotavirus IgG in all macaques >1 year of age (15 total), rotavirus is clearly endemic in macaques in this colony.
Next, we identified rhesus macaques within this colony that were the most likely to experience a rotavirus infection during the following months. Even though repeat rotavirus infections are common in humans (2) and are likely to also occur in macaques, the most severe rotavirus disease in humans almost invariably occurs after their first infection (44), provided that this infection takes place during the period of greatest susceptibility, when a child is 6 months of age or older (22). Our plan, therefore, was to identify rhesus macaques of approximately this equivalent age that had no evidence of a previous rotavirus infection based on serum rotavirus antibody levels. Accordingly, a group of 16 macaques age 2.6 to 5.9 months were evaluated for the presence of serum rotavirus IgG and IgA, and based on the absence of serum rotavirus IgA and very low levels of serum rotavirus IgG in 10 of these animals, these were considered to be the most likely to be susceptible to a rotavirus infection. Stool specimens were collected over a 10-week period from these animals, and six shed rotavirus. Although most stools had relatively low titers of infectious rotavirus, probably due to viral inactivation prior to stool collection, one specimen each obtained from two of the animals (EA40 and EA42) contained titers of >106 FFU/ml, sufficient to utilize for the production of a challenge pool of unpassaged rotavirus.
The rotaviruses obtained from naturally infected macaques were then characterized. First it was shown that they were fully adapted to grow in cell culture and produced a complete cytopathic effect in MA104 cells during their first passage. This property is typical of some animal rotavirus strains and sharply contrasts with results found with human rotaviruses, where multiple cell culture passages are required to develop this phenotypic trait (50). The culture-grown rotaviruses obtained from macaques all had identical electropherotypes which were distinct from those of SA11 and RRV. However, these viruses were identified as serotype G3 and subgroup 1, both features characteristic of simian rotaviruses (22). Furthermore, a partial sequence comparison (amino acids 51 to 206) between the VP7 protein of one of the new macaque rotaviruses and the VP7 proteins of RRV and SA11 revealed that they shared 94.9 and 95.6% homology, respectively (results not shown), supporting the indication that the new macaque rotaviruses belong to serotype G3.
Because viruses obtained from all macaques were similar if not identical, the rotavirus obtained from only one specimen from one macaque (EA40) was utilized as the representative of the new rotavirus strain, which was named TUCH. Nucleotide analysis of the complete coding sequence of the VP4 gene from the TUCH strain revealed that it is genotypically dissimilar to those from rotaviruses belonging to the 22 established P types (11, 28, 37). Rotaviruses have been classified within the same P type if they share 89% homology (14). No rotavirus belonging to any established P type was found to share >86.2% homology with the VP4 protein of TUCH. The suggestion that the TUCH VP4 gene has evolved from those of other rotaviruses was supported by phylogenetic analysis. Therefore, the TUCH strain was tentatively classified as genotype P[23].
Because the VP4 gene of TUCH appeared to have greatly diverged from VP4 genes found in other rotavirus strains, it was of interest to determine whether this could also be demonstrated for another TUCH gene. For this we selected the VP6 gene, which is highly conserved among group A rotaviruses (22). Although the degree of difference between the TUCH VP6 protein and those of numerous others to which it was compared is much less than found for the VP4 proteins, the greatest homology was still only 95.5%. Furthermore, phylogenetic analysis of TUCH VP6 supported the hypothesis that the gene encoding this protein has also substantially diverged from cognate genes of other rotavirus strains. Therefore, although TUCH has retained epitopes required for retention of the VP7 serotype G3 and VP6 subgroup 1 specificities typical of simian rotaviruses, it is quite dissimilar to the standard simian rotavirus strains that are in use in many laboratories worldwide today. In fact, based on phylogenetic analysis with the VP4 and VP6 genes, TUCH is more closely related to mouse strains than it is to the simian SA11 or RRV strain. Possibly the TUCH strain arose from a reassortment event between a simian rotavirus and a rotavirus from another species at some distant point in time. If so, some genes, such as those for VP6 and VP4, may have been derived from the nonsimian strain. Independent segregation of rotavirus genes has been described in numerous publications, and even many human strains have been identified as reassortants between animal and human rotaviruses (see, e.g., reference 20). However, it is also possible that TUCH arose from a common ancestor of both simian and other G3 strains. Sequence comparisons between other gene segments of the TUCH strain and those from multiple rotaviruses could provide a more definitive explanation of the origin of the TUCH virus.
The second purpose of this study was to determine whether an unpassaged preparation of TUCH would replicate in previously uninfected macaques, resulting in consistently high levels of fecal rotavirus shedding and, possibly, diarrheal disease. Based on this study with eight cesarean section-derived macaques age 14 to 42 days, several of which contained high levels of maternal (transplacental) rotavirus IgG on their date of inoculation, the TUCH strain appears to be excellent for model studies on rotavirus immunity in macaques. In the age range examined, which is within a window sufficient to immunize and challenge, all eight animals consistently shed large quantities of rotavirus antigen. Presumably this age range could be readily extended, since the TUCH strain itself was derived from an animal that was >6 months of age. As was found with the mouse model that we established over a decade ago for studies on rotavirus vaccines and their mechanisms of protection (51), the TUCH strain will be useful for studies on both active and passive immunity. However, as was also found with rodent and rabbit models for rotavirus in their homologous hosts (6-10, 16, 51), the TUCH strain will not be useful for studies on active immunity against diarrheal disease in macaques. In place of this, fecal shedding of the TUCH virus can be used as a marker of rotavirus infection in macaques, as has been used extensively in mice and rabbits.
ADDENDUM
After this work was completed, we became aware of a publication (24a) that describes a porcine group A rotavirus (strain A34) belonging to a potential new P genotype based on sequence analysis of part (amino acids 13 to 250) of the VP8 region of the VP4 gene of this virus (GenBank accession no. AY174094). This region of the VP4 protein of the A34 strain was found to share 79.8% homology with the cognate region of the TUCH VP4 protein, thus confirming that the P type of the A34 strain is distinct from that of TUCH. If upon analysis of the complete sequence of the VP4 gene of the A34 strain, it is confirmed to belong to a new P genotype, the TUCH strain would then be tentatively classified as P[24] rather than P[23].
ACKNOWLEDGMENTS
This study was supported by NIAID grant AI056847-01 to A.H.-C.C. and by a Cincinnati Children's Hospital Medical Center pilot grant to R.L.W. Partial support was provided by TNPRC base grant RR00164.
REFERENCES
Bernstein, D. I., V. E. Smith, J. R. Sherwood, G. M. Schiff, D. S. Sander, D. DeFeudis, D. R. Spriggs, and R. L. Ward. 1998. Safety and immunogenicity of live attenuated human rotavirus vaccine 89-12. Vaccine 16:381-387.
Bernstein, D. I., and R. L. Ward. 2004. Rotaviruses, p. 2110-2111. In R. D. Feigin, J. D. Cherry, G. J. Demmler, and S. L. Kaplan (ed.), Textbook of pediatric infectious diseases, 5th ed. Saunders, Philadelphia, Pa.
Bishop, R. F., S. R. Tzipori, B. S. Coulson, J. E. Unicomb, M. J. Albert, and G. L. Barnes. 1986. Heterologous protection against rotavirus-induced disease in gnotobiotic piglets. J. Clin. Microbiol. 24:1023-1028.
Bridger, J. C., and G. Oldham. 1987. Avirulent rotavirus infections protect calves from disease with and without inducing high levels of neutralizing antibody. J. Gen. Virol. 68:2311-2317.
Bridger, J. C., G. I. Tauscher, and U. Desselberger. 1998. Viral determinants of rotavirus pathogenicity in pigs: evidence that the fourth gene of a porcine rotavirus confers diarrhea in the homologous host. J. Virol. 72:6929-6931.
Ciarlet, M., M. E. Conner, M. J. Finegold, and M. K. Estes. 2002. Group A rotavirus infection and age-dependent diarrheal disease in rats: a new animal model to study the pathophysiology of rotavirus infection. J. Virol. 76:41-57.
Ciarlet, M., S. E. Crawford, C. Barone, A. Bertolotti-Ciarlet, R. F. Ramig, M. K. Estes, and M. E. Conner. 1998. Subunit rotavirus vaccine administered parenterally to rabbits induces active protective immunity. J. Virol. 72:9233-9246.
Ciarlet, M., M. E. Estes, C. Barone, R. F. Ramig, and M. E. Conner. 1998. Analysis of host range restriction determinants in the rabbit model: comparison of homologous and heterologous rotavirus infections. J. Virol. 72:2341-2351.
Conner, M. E., S. E. Crawford, C. Barone, and M. K. Estes. 1993. Rotavirus vaccine administered parenterally induces protective immunity. J. Virol. 67:6633-6641.
Conner, M. E., M. K. Estes, and D. Y. Graham. 1988. Rabbit model for rotavirus infection. J. Virol. 62:1625-1633.
Estes, M. K. 2001. Rotaviruses and their replication, p. 1747-1785. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed., vol. 2. Lippincott Williams and Wilkins, Philadelphia, Pa.
Fernandez, F. M., M. E. Conner, D. C. Hodgins, A. V. Parwani, P. R. Nielsen, S. E. Crawford, M. K. Estes, and L. J. Saif. 1998. Passive immunity to bovine rotavirus in newborn calves fed colostrum supplements from cows immunized with recombinant SA11 rotavirus core-like particle (CLP) or virus-like particle (VLP) vaccines. Vaccine 16:507-516.
Gorziglia, M., Y. Hoshino, K. Nishikawa, W. L. Maloy, R. W. Jones, A. Z. Kapikian, and R. M. Chanock. 1988. Comparative sequence analysis of the genomic segment 6 of four rotaviruses each with a different subgroup specificity. J. Gen. Virol. 69:1659-1669.
Gorziglia, M., G. Larralde, A. Z. Kapikian, and R. M. Chanock. 1990. Antigenic relationships among human rotaviruses as determined by outer capsid protein VP4. Proc. Natl. Acad. Sci. USA 87:7155-7159.
Greenberg, H. B., V. McAuliffe, J. Valdesuso, R. Wyatt, J. Flores, A. Kalica, Y. Hoshino, and N. Singh. 1983. Serological analysis of subgroup protein of rotavirus, using monoclonal antibodies. Infect. Immun. 39:91-99.
Guerin-Danan, C., J. C. Meslin, F. Lambre, A. Charpilienne, M. Serezat, C. Bouley, J. Cohen, and C. Andrieux. 1998. Development of a heterologous model in germfree suckling rats for studies of rotavirus diarrhea. J. Virol. 72:9298-9302.
Hoshino, Y., L. J. Saif, K. Shien-Young, M. M. Sereno, W. K. Chen, and A. Z. Kapikian. 1995. Identification of group A rotavirus genes associated with virulence of a porcine rotavirus and host range restriction of a human rotavirus in the gnotobiotic piglet model. Virology 209:274-280.
Hoshino, Y., L. J. Saif, M. M. Sereno, R. M. Chanock. 1988. Infection immunity of piglets to either VP3 or VP7 outer capsid protein confers resistance to challenge with a virulent rotavirus bearing the corresponding antigen. J. Virol. 62:744-748.
Iturriza-Gomara. M., G. Kang, A. Mammen, A. K. Jana, M. Abraham, U. Desselberger, D. Brown, and J. Gray. 2004. Characterization of G10P(11) rotaviruses causing acute gastroenteritis in neonates and infants in Vellore, India. J. Clin. Microbiol. 42:2541-2547.
Iturriza-Gómara, M., C. Wong, S. Blome, U. Desselberger, and J. Gray. 2002. Molecular characterization of VP6 genes of human rotavirus isolates: correlation of genogroups with subgroups and evidence of independent segregation. J. Virol. 76:6596-6601.
Kalter, S. S., R. L. Heberling, A. R. Rodriguez, and T. L. Lester. 1983. Infection of baboons (Papio cynocephalus) with rotavirus (SA11). Dev. Biol. Stand. 53:257-261.
Kapikian, A. Z., Y. Hoshino, and R. M. Chanock. 2001. Rotaviruses, p. 1787-1833. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed., vol. 2. Lippincott Williams and Wilkins, Philadelphia, Pa.
Leong, Y. K., and A. Awang. 1990. Experimental group A rotaviral infection in cynomolgus monkeys raised on formula diet. Microbiol. Immunol. 34:153-162.
Liprandi, F., G. Lopez, I. Rodriguez, M. Hidalgo, J. E. Ludert, and N. Mattion. 1990. Monoclonal antibodies to the VP6 of porcine subgroup I rotaviruses reactive with subgroup I and non-subgroup I non-subgroup II strains. J. Gen. Virol. 71:1395-1398.
Liprandi, F., M. Gerder, Z. Bastidas, J. A. Lopez, F. H. Pujol, J. E. Ludert, D. B. Joelsson, and M. Ciralet. 2003. A novel type of VP4 carried by a porcine rotavirus strain. Virology 314:373-380.
Lopez, S., R. Espinosa, H. B. Greenberg, and C. F. Arias. 1994. Mapping the subgroup epitopes of rotavirus protein VP6. Virology 204:153-162.
Majer, M. F. Behrens, E. Weinmann, R. Mauler, G. Maass, H. G. Baumeister, and T. Luthardt. 1978. Diarrhea in newborn cynomolgus monkeys infected with human rotavirus. Infection 6:71-72.
Malherbe, H., and R. Harwin. 1963. The cytopathic affects of vervet monkey viruses. S. Afr. Med. J. 37:407-411.
Martella, V., M. Ciarlet, A. Camarda, A. Pratelli, M. Tempesta, G. Greco, A. Cavalli, G. Elia, N. Decaro, V. Terio, G. Bozzo, M. Camero, and C. Buonavoglia. 2003. Molecular characterization of the VP4, VP6, VP7, and NSP4 genes of lapine rotaviruses identified in Italy: emergence of a novel VP4 genotype. Virology 314:358-370.
McNeal, M. M., J. Belli, M. Basu, A. H.-C. Choi, and R. L. Ward. 2004. Discovery of a new strain of murine rotavirus that is consistently shed in large quantities after oral inoculation of adult mice. Virology 320:1-11.
McNeal, M. M., M. N. Rae, J. A. Bean, and R. L. Ward. 1999. Antibody-dependent and -independent protection following intranasal immunization of mice with rotavirus particles. J. Virol. 73:7565-7573.
Mitchell, J. D., L. A. Lambeth, L. Sosula, A. Murphy, and M. Albrey. 1977. Transmission of rotavirus gastroenteritis from children to a monkey. Gut 18:156-160.
Nakagomi, O., and T. Nakagomi. 1993. Interspecies transmission of rotaviruses studied from the perspective of genogroup. Microbiol. Immunol. 37:337-348.
Oldham, G., J. C. Bridger, C. J. Howard, and K. R. Parsons. 1993. In vivo role of lymphocyte subpopulations in the control of virus excretion and mucosal antibody responses of cattle infected with rotavirus. J. Virol. 67:5012-5019.
Page, R. D. M. 1996. Treeview: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12:357-358.
Parashar, U. D., E. G. Hummelman, J. S. Bresee, M. A. Miller, and R. I. Glass. 2003. Global illness and deaths caused by rotavirus disease in children. Emerg. Infect. Dis. 9:565-571.
Petschow, B. W., R. E. Litov, L. J. T. Young, and T. P. McGraw. 1992. Response of colostrum-deprived cynomolgus monkeys to intragastric challenge exposure with simian rotavirus strain SA11. Am. J. Vet. Res. 53:674-678.
Rao, C. D., K. Gowda, and B. S. Y. Reddy. 2000. Sequence analysis of VP4 and VP7 genes of nontypeable strains identifies a new pair of outer capsid proteins representing novel P and G genotypes in bovine rotaviruses. Virology 276:104-113.
Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
Sestak, K., M. M. McNeal, A. Choi, M. J. Cole, G. Ramesh, X. Alvarez, P. P. Aye, R. P. Bohm, M. Mohamadzadeh, and R. L. Ward. 2004. Defining T-cell-mediated immune responses in rotavirus-infected juvenile rhesus macaques. J. Virol. 78:10258-10264.
Snodgrass, D. R., and P. W. Wells. 1978. Passive immunity in rotaviral infections. J. Am. Vet. Med. Assoc. 173:565-568.
Soike, K. F., G. W. Gary, and S. Gibson. 1980. Susceptibility of nonhuman primate species to infection by simian rotavirus SA-11. Am. J. Vet. Res. 41:1098-1103.
Stuker, G., L. S. Oshiro, and N. J. Schmidt. 1980. Antigenic comparisons of two new rotaviruses from rhesus monkeys. J. Clin. Microbiol. 11:202-203.
Tang, B., J. M. Gilbert, S. M. Matsui, and H. B. Greenberg. 1997. Comparison of the rotavirus gene 6 from different species by sequence analysis and localization of subgroup-specific epitopes using site-directed mutagenesis. Virology 237:89-96.
Velazquez, F. R., D. O. Matson, J. J. Calva, M. L. Guerrero, A. L. Morrow, S. Carter-Campbell, R. I. Glass, M. K. Estes, L. K. Pickering, and G. M. Ruiz-Palacios. 1996. Rotavirus infection in infants as protection against subsequent infections. N. Engl. J. Med. 335:1022-1028.
Ward, R. L. 2003. Possible mechanisms of protection elicited by candidate rotavirus vaccines as determined with the adult mouse model. Virol. Immunol. 16:17-24.
Ward, R. L., D. I. Bernstein, R. Shukla, E. C. Young, J. R. Sherwood, M. M. McNeal, M. C. Walker, and G. M. Schiff. 1989. Effects of antibody to rotavirus on protection of adults challenged with a human rotavirus. J. Infect. Dis. 159:79-88.
Ward, R. L., D. I. Bernstein, E. C. Young, J. R. Sherwood, D. R. Knowlton, and G. M. Schiff. 1986. Human rotavirus studies in volunteers: determination of infectious dose and serological response to infection. J. Infect. Dis. 154:871-880.
Ward, R. L., H. F. Clark, P. A. Offit, and R. I. Glass. 2004. Live vaccine strategies to prevent rotavirus disease, p. 607-620. In M. M. Levine, J. B. Kaper, R. Rappuoli, M. A. Liu, and M. F. Good (ed.), New generation vaccines, 3rd ed. Marcel Dekker, Inc., New York, N.Y.
Ward, R. L., J. D. Clemens, D. A. Sack, D. R. Knowlton, M. M. McNeal, N. Huda, F. Ahmed, M. Rao, and G. M. Schiff. 1991. Culture adaptation and characterization of group A rotaviruses causing diarrheal illnesses in Bangladesh from 1985 to 1986. J. Clin. Microbiol. 29:1915-1923.
Ward, R. L., D. R. Knowlton, and M. J. Pierce. 1990. Efficiency of human rotavirus propagation in cell culture. J. Clin. Microbiol. 19:748-755.
Ward, R. L., M. M. McNeal, and J. F. Sheridan. 1990. Development of an adult mouse model for studies on protection against rotavirus. J. Virol. 64:5070-5075.
Ward, R. L., O. Nakagomi, D. R. Knowlton, M. M. McNeal, T. Nakagomi, N. Huda, J. D. Clemens, and D. A. Sack. 1991. Formation and selection of intergenogroup reassortants during cell culture adaptation of rotaviruses from dually infected subjects. J. Virol. 65:2699-2701.
Woode, G. N., N. E. Kelso, T. F. Simpson, S. K. Gaul, L. E. Evans, and L. Babiuk. 1983. Antigenic relationships among some bovine rotaviruses: serum neutralization and cross-protection in gnotobiotic calves. J. Clin. Microbiol. 18:358-364.
Woode, G. N., S. Zheng, B. I. Rosen, N. Knight, N. E. Kelso Gourley, and R. F. Ramig. 1987. Protection between different serotypes of bovine rotavirus in gnotobiotic calves: specificity of serum antibody and coproantibody responses. J. Clin. Microbiol. 25:1052-1058.
Wyatt, R. G., D. L. Sly, W. T. London, A. E. Palmer, A. R. Kalica, D. H. Van Kirk, R. M. Chanock, and A. Z. Kapikian. 1976. Induction of diarrhea in colostrum-deprived newborn rhesus monkeys with the human reovirus-like agent of infantile gastroenteritis. Arch. Virol. 50:17-27.
Yuan, L., C. Iosef, M. S. P. Azevedo, Y. Kim, Y. Qian, A. Geyer, T. V. Nguyen, K. O. Chang, and L. J. Saif. 2001. Protective immunity and antibody-secreting cell responses elicited by combined oral attenuated Wa human rotavirus and intranasal Wa 2/6-VLPs with mutant Escherichia coli heat-labile toxin in gnotobiotic pigs. J. Virol. 75:9229-9238.
Yuan, L., L. A. Ward, B. I. Rosen, T. L. To, and L. J. Saif. 1996. Systemic and intestinal antibody-secreting cell responses and correlates of protective immunity to human rotavirus in a gnotobiotic pig model of disease. J. Virol. 70:3075-3083.(Monica M. McNeal, Karol S)
Tulane National Primate Research Center, Covington, Louisiana
ABSTRACT
Although there are several reports on rotavirus inoculation of nonhuman primates, no reliable model exists. Therefore, this study was designed to develop a rhesus macaque model for rotavirus studies. The goals were to obtain a wild-type macaque rotavirus and evaluate it as a challenge virus for model studies. Once rotavirus was shown to be endemic within the macaque colony at the Tulane National Primate Research Center, stool specimens were collected from juvenile animals (2.6 to 5.9 months of age) without evidence of previous rotavirus infection and examined for rotavirus antigen. Six of 10 animals shed rotavirus during the 10-week collection period, and the electropherotypes of all isolates were identical to each other but distinct from those of prototype simian rotaviruses. These viruses were characterized as serotype G3 and subgroup 1, properties typical of many animal rotaviruses, including simian strains. Nucleotide sequence analysis of the VP4 gene was performed with a culture-grown isolate from the stool of one animal, designated the TUCH strain. Based on both genotypic and phylogenetic comparisons between TUCH VP4 and cognate proteins of representatives of the reported 22 P genotypes, the TUCH virus belongs to a new genotype, P[23]. A pool of wild-type TUCH was prepared and intragastrically administered to eight cesarean section-derived, specific-pathogen-free macaques 14 to 42 days of age. All animals were kept in a biocontainment level 2 facility. Although no diarrhea was observed and the animals remained clinically normal, all animals shed large quantities of rotavirus antigen in their feces after inoculation, which resolved by the end of the 14-day observation period. Therefore, TUCH infection of macaques provides a useful nonhuman primate model for studies on rotavirus protection.
INTRODUCTION
Rotaviruses are the primary cause of severe diarrhea in young children and are responsible for ca. 500,000 deaths in the world each year (35). Several vaccines against rotavirus disease have been developed, but none are available for routine immunization (48). All candidate vaccines evaluated in clinical trials have been live, attenuated rotaviruses that are delivered orally to mimic natural infection. Although natural rotavirus infection has been found to provide substantial protection against rotavirus disease, particularly severe disease, the mechanisms by which this occurs remain largely undetermined (45). A number of animal models have been developed to help understand the mechanisms of rotavirus immunity, including lambs (40), calves (4, 12, 33, 53, 54), piglets (3, 5, 17, 18, 56, 57), rabbits (7-10), and rats (6, 16), but the majority of the mechanistic studies have been conducted with the adult mouse model that we developed in 1990 (51). Even with these models, only a partial understanding of the mechanism of rotavirus protection has been achieved. Furthermore, an animal model that more closely resembles humans is clearly needed, both to understand the mechanisms of rotavirus immunity and for preclinical evaluations of novel rotavirus vaccine candidates.
Nonhuman primates are the animals most closely related to humans. Although several rotavirus challenge studies have been conducted with these animals, the first of which was reported in 1976 (55), very little is known about rotavirus infections in any nonhuman primate species. Several investigators have reported that oral inoculation of different nonhuman primates, including several types of monkeys as well as baboons, with either culture-adapted simian (SA11) or human (Wa) rotavirus or fecal preparations of human rotaviruses, will cause diarrheal illness in most animals during their first week of life (21, 23, 26, 31, 36, 41, 55). However, after that time, essentially no illness was observed, and most of the older animals neither shed virus nor seroconverted. The exception may have been the results found after inoculation of one chimpanzee that, when orally administered SA11 at 141 days of age, developed diarrhea and shed large amounts of rotavirus over a 9-day period (41).
A potential major limitation in these studies was the lack of a challenge virus that would reliably infect and produce illness in the nonhuman primate species under investigation. The only simian rotavirus used in these studies (SA11) was obtained from an asymptomatic vervet monkey (27) and has been repeatedly passaged in cultured cells, a method typically used for viral attenuation. The only other rotaviruses used were human strains. Although rotaviruses sometimes cross species barriers (32), their virulence is typically blunted in heterologous hosts. Thus, these human strains, even those from human fecal specimens, are likely to be naturally attenuated in nonhuman primates. There has been no report of experimental inoculation of a nonhuman primate being performed with a wild-type (fecally derived) simian rotavirus. Therefore, it was possible that wild-type simian rotaviruses may produce severe illness in nonhuman primates after the first week of life, perhaps even up to several years of age as occurs in humans. The first purpose of this study was to obtain and characterize such a virus from rhesus macaques to be used for challenge studies in nonhuman primates. The second purpose was to determine whether this new strain of simian rotavirus could consistently produce high levels of fecal rotavirus shedding and possibly elicit diarrheal disease in macaques when administered after the first week of life. If either or both outcomes could be attained, this model system could potentially provide a highly relevant substitute for humans in studies on active immunity against rotavirus.
MATERIALS AND METHODS
Rhesus macaques: sample collections and illness monitoring. (i) Conventional animals. A group of 16 juvenile rhesus macaques (Macaca mulatto), age 2.6 to 5.9 months at the time of their enrollment on 12 November 2002, were included in a study to obtain wild-type rotavirus from naturally infected animals. All animals were born at the Tulane National Primate Research Center (TNPRC) and were housed under biosafety level 2 conditions in accordance with standards of the Guide for the Care and Use of Laboratory Animals and the Association for Assessment and Accreditation of Laboratory Animal Care. Only simian retrovirus (simian immunodeficiency virus, simian retrovirus, and simian T-cell leukemia virus)-negative animals were used. Every animal was monitored for diarrhea and signs of lethargy on a daily basis. Initially, blood samples were collected and sera were harvested and stored (–20°C) until analyzed for rotavirus immunoglobulin G (IgG) and IgA. Once these were determined, the 10 animals with the lowest titers were monitored for rotavirus shedding from 25 November 2002 through 6 February 2003. For this, stool samples were collected twice per week from the bottom of the cage containing each animal and stored at –80°C until tested for the presence of rotavirus antigen. Monthly blood collections were also taken from these 10 animals, and the harvested sera were stored (–20°C) until analyzed for rotavirus IgG.
(ii) Cesarean section-derived animals. Two groups of cesarean section-derived macaques were used. All were lactogenic immunity deprived (hand fed and free of maternal secretory IgA) and housed in nursery (biosafety level 2) rooms that contained only specific-pathogen-free juvenile animals obtained in this manner. At the time of cesarean section, the level of maternal serum rotavirus IgG ranged between 300 and 5,600 U/ml, where the limit of detection was 5 U/ml. The first group contained five animals that were monitored for up to 7 months of age for evidence of a natural rotavirus infection based on increases in their titers of serum rotavirus IgG or IgA between blood collections. The second group (eight animals) were bled at the time of birth and just before the time of rotavirus challenge, and rotavirus IgG and IgA levels in these serum specimens were measured. Between 14 and 42 days after birth, these animals were transferred to a separate room in the facility and intragastrically challenged with 3 x 104 focus-forming units (FFU) of our new wild-type strain of macaque rotavirus following administration of 2 ml of 4% sodium bicarbonate to buffer stomach acidity. Stool specimens were collected between one and four times each day for the next 14 days, and the levels of rotavirus shedding were quantified. In addition, the consistencies of these stool specimens were graded for detection of diarrhea. During this period, the macaques were evaluated each day for signs of lethargy as an additional indicator of illness.
Quantitation of rotavirus IgG and IgA in serum specimens. The levels of rotavirus IgG and IgA were determined in serum specimens by enzyme-linked immunosorbent assays (ELISA) essentially as described previously (1, 46). In brief, 96-well microtiter plates were coated with purified, double-layered simian rotavirus SA11 or with mock-infected lysate purified in an identical fashion. This was followed by the stepwise addition of diluted specimen or control sera, biotinylated goat anti-monkey (rhesus) IgG (Research Diagnostics, Inc., Flandus, N.J.), peroxidase-conjugated avidin-biotin (Vector Laboratories, Inc., Burlingame, Calif.), and substrate (orthophenylenediamine [Sigma Aldrich, St. Louis, Mo.]). Optical density (OD) values were determined at 490 nm. The levels were then expressed as units of IgG per milliliter, based on a standard curve generated from a human serum pool arbitrarily assigned a level of 1,000 U/ml. The lower limit determinable by this assay under the conditions used was 5 U/ml. Rotavirus IgA was measured in the same manner, except that biotinylated goat anti-monkey IgA (Research Diagnostics, Inc.) was used. In this case, the control serum (human pool) was assigned a value of 250 U/ml, and the lower limit measurable was 10 U/ml.
Detection and quantitation of rotavirus antigen in stools of macaques. Frozen stools collected from the cages of individually housed macaques were thawed, made into 20% (wt/vol) suspensions with Earle's balanced salt solution (EBSS), and tested for the presence of rotavirus antigen by ELISA as described elsewhere (30, 46). For the group of conventional animals, shedding was quantified as optical density units (A490), and the highest OD value measurable in the rotavirus-positive specimens was 3.0. Therefore, the quantity of rotavirus antigen present in these specimens was expressed as OD measurements up to 3.0. For cesarean section-derived macaques inoculated with the new strain of wild-type macaque rotavirus, fecal shedding of rotavirus antigen was quantified by ELISA as nanograms per milliliter of 20% suspensions of stool, based on a standard curve generated by using purified double-layered particles of murine rotavirus strain EDIM.
Quantitation of infectious rotavirus in stool specimens. Rotaviruses present in 20% suspensions of macaque stools were quantified by a fluorescent-focus assay (47). Titers are expressed as FFU per milliliter of the 20% stool suspensions.
Cell culture adaptation of rotavirus from macaque stools. Rotaviruses in stool were adapted to grow in MA104 (monkey kidney) cells by using methods similar to those previously reported (30). In brief, 20% (wt/vol) stool suspensions and dilutions of these samples (in EBSS) were treated with trypsin (15 μg/ml; GIBCO Laboratories, Grand Island, N.Y.) (1:250) to activate the virus. Following centrifugation (1,200 x g, 20 min) to remove debris, the supernatants were added to tube cultures containing monolayers of MA104 cells. After adsorption (2 h) on a roller apparatus, the cultures were washed and medium without serum, but with trypsin, was added. The culture tubes were rolled for 4 days or until cytopathic effects reached 3+ (whichever came first) and then were stored at –20°C. Subsequent passages were conducted without trypsin pretreatment of the lysate from the previous passage. In this study, the presence and quantity of rotavirus were determined after two passages in MA104 cells, using the same ELISA procedure as described for detection of rotavirus in stool specimens.
Electrophoretic analysis of viral RNA segments. The double-stranded RNA genome segments of passage 2 of the culture-adapted macaque rotaviruses were extracted from viral lysates and analyzed by polyacrylamide gel electrophoresis (50). The electropherotype of one of the isolates was compared to those of prototype simian rotaviruses RRV and SA11, also grown in MA104 cells.
G serotyping of the culture-adapted macaque rotaviruses. The G serotypes of passage 2 of the culture-adapted macaque rotaviruses were determined by an ELISA with prototype G1 to G4 (Wa, DS-1, P, and ST3, respectively) human rotavirus strains as controls (15, 49). VP7-specific monoclonal antibodies (MAbs) against G1 (MAb 5E8), G2 (MAb 1C10), G3 (MAb 159), and G4 (MAb ST3:1) rotavirus strains were used to coat 96-well microtiter plates (capture antibodies), and hyperimmune guinea pig antisera against G1 to G4 prototype human rotaviruses were used as the detector antibodies. The MAbs against G1 to G3 strains were gifts of H. B. Greenberg (Stanford University, Palo Alto, Calif.) and the ST3:1 MAb was a gift of B. Coulson (University of Melbourne, Melbourne, Australia).
Subgroup analysis of the culture-adapted macaque rotaviruses. Subgroup determination of passage 2 of the culture adapted macaque rotaviruses was performed by an ELISA with subgroup-specific MAbs 225/60 (subgroup 1) and 631/9 (subgroup 2) (15, 29, 50, 52). Both MAbs were gifts of H. B. Greenberg. Control rotaviruses included in this assay were the simian RRV (subgroup 1 [SG1]) and human Wa (SG2) strains. Both control rotaviruses reacted only with their respective subgroup-specific MAb.
Nucleotide sequencing of the VP4 and VP6 genes of the TUCH strain of macaque rotavirus. Double-stranded cDNAs of genes 4 and 6 of CsCl gradient-purified, culture-adapted rotavirus (passage 3) obtained from the stool of macaque EA40 (specimen date, 2 January 2003), subsequently named the TUCH strain, were generated by reverse transcription-PCRs (RT-PCR) with the Thermo Script RT-PCR system (Invitrogen/Life Technologies, Carlsbad, Calif.) and Vent DNA polymerase (New England BioLabs, Beverly, Mass.). Briefly, RT was performed with 1 μg of purified double-stranded genomic RNA in 25 μl of distilled water containing an RNase inhibitor (20 U of RNaseOut [Invitrogen Life Technologies]) and a 200 nM concentration of a forward primer (gene 4, 5'-GGC TAT AAA ATG GCT TCG CTC-3' [corresponding to nucleotides 1 to 21]; gene 6, 5'-GGC TTT TAA ACG AAG TCT TC-3' [corresponding to nucleotides 1 to 20]) and a reverse primer (gene 4, 5'-GGT CAC ATC CTC TAG AAA TTA C-3' [corresponding to nucleotides 2342 to 2362]; gene 6, 5'-GGT CAC ATC CTC TCA CTA CGG CAT TC-3' [corresponding to nucleotides 1331 to 1356]) designed from the noncoding regions of the simian SA11 rotavirus genes. The reaction mixture was denatured (3 min, 94°C), and the primers were allowed to rapidly anneal to viral RNA by cooling the suspension on ice. Next, 4.0 μl of 5x cDNA synthesis buffer (Invitrogen Life Technologies), 1.0 μl of 0.1 M dithiothreitol, 1.0 μl of RNaseOut (40 U/μl), 1.0 μl of diethyl pyrocarbonate-treated water, and 1.0 μl of ThermoScript reverse transcriptase (15 U/μl) were added to the reaction mixture on ice. Reverse transcription was carried out at 50°C for 60 min in a heating block. The reaction was terminated by transferring the reaction mixture in a heating block set at 85°C for 5 min. The reaction mixture was cooled, 2 U of RNase H in 1 μl was added, and the reaction mixture was incubated at 37°C for 20 min. PCR was carried out in a 50-μl volume containing 2.0 μl of the RT-PCR products, 5 μl of 10x amplification buffer, 1.5 μl of 10 mM deoxynucleoside triphosphates, 1.0 μl of 50 mM MgSO4, 1.5 μl of forward and 1.5 μl of reverse primer (200 nM each), and 2.5 U of Platinum Pfx DNA polymerase (Invitrogen Life Technologies). The template was denatured (94°C, 2 min), and then 30 cycles of PCR were carried out. Each PCR cycle consisted of a denaturation step (94°C for 30 s), an annealing step (55°C for 30 s), and an extension step (68°C for 3 min for gene 4 and for 2 min for gene 6). After the PCR cycle, an additional incubation step was carried out (72°C, 7 min). PCR products were purified by using the QIAquick PCR purification kit (Qiagen Inc., Valencia, Calif.). The nucleotide sequences of the VP4 and VP6 genes were determined by using purified double-stranded cDNAs generated by RT-PCR as templates. Sequencing was performed by Cleveland Genomics (Cleveland, Ohio), using an ABI DNA sequencer (Applied Biosystems, Foster City, Calif.). The primer pair used for generating double-stranded DNA was also used to sequence the initial 500 nucleotides from the ends of the positive and the negative cDNA strands. New primer pairs were designed by using the sequence generated until cDNA strands were completely sequenced. The gene sequence was determined twice with PCR products generated in separate reactions, using both strands as templates.
Phylogenetic relationship of TUCH rotavirus VP4 and VP6 gene sequences with those of other rotavirus strains. CLUSTALW (http://www.ebi.ac.uk/clustalw/), a program that is commonly used for phylogeny reconstruction, was employed to analyze phylogenetic relationships of the deduced complete amino acid sequences of the TUCH VP4 and VP6 proteins with those of cognate proteins of other rotaviruses. This program predicts phylogenetic trees by comparing alignment scores for all possible pairs of the rotavirus VP4 and VP6 sequences. The rotavirus gene sequences used for pairwise alignment were first retrieved from protein databases by using WU-BLAST2 (http://BLAST.wustl.edu). The sequences were then aligned by using CLUSTALW to search for regions of similarity among the sequences. The fundamental unit algorithm output of BLAST is the high-scoring segment pair, which consists of two sequence fragments of arbitrary but equal length whose alignment is locally maximal and for which the alignment score meets or exceeds a threshold (the default cutoff score was chosen). In the analyses performed here, the default Blosum62 was used for the scoring matrix. Unrooted phylogenetic trees were constructed by using the neighbor-joining method (38). The trees are displayed with TreeView (34).
Preparation of a challenge pool of unpassaged TUCH rotavirus. To prepare a pool of unpassaged simian rotavirus to be used for subsequent challenge studies, the stool obtained on 2 January 2003 from macaque EA40 was processed. Although the rotaviruses from all five macaques examined had identical electropherotypes and therefore were likely to be very similar strains, only the rotavirus in this one stool specimen has been given the designation TUCH, and only this virus was used in subsequent experiments under that designation. This stool was used because only it and one other, of the seven stools examined, contained a high titer of infectious rotavirus, and it had the larger quantity of the two. A 5% (wt/vol) suspension of this entire stool (3.2 g) was made in EBSS by shaking (2 h, 4°C) with glass bends on an automated shaker. After centrifugation (1,200 x g, 20 min, 4°C) to clarify the suspension, the supernatant was filtered (0.2-μm-pore-size filter) and stored (–70°C) in 1-ml aliquots. The titer of the original 5% suspension was 3.4 x 105 FFU/ml, while the frozen aliquots contained 3.0 x 104 FFU/ml (9% recovery).
Nucleotide sequence accession numbers. The GenBank accession numbers for the TUCH VP4 and VP6 protein genes are AY596189 and AY594670, respectively.
RESULTS
Isolation of rotavirus from naturally infected rhesus macaques. (i) Identification of juvenile macaques with no evidence of previous rotavirus infection. Before attempting to obtain rotavirus from naturally infected rhesus macaques housed within open cages in containment rooms at the TNPRC, it was first necessary to demonstrate that rotavirus is endemic in this macaque population. For this, we relied on serum rotavirus IgG measurements. Upon examination of a group of 16 juvenile macaques age 2.6 to 5.9 months and housed within the facility, it was noted that 5 had high titers of rotavirus IgG (geometric mean titer, 704 U/ml) and that the other 11 had much lower levels (geometric mean titer, 13 U/ml) (Table 1). Because the average ages of the two groups of macaques were comparable (5.2 versus 5.0 months), it was assumed that at least the five animals with high titers of rotavirus IgG had experienced at least one natural rotavirus infection. It is likely that rotavirus IgG detected in the other 11 macaques was residual transplacental antibody. These suggestions were supported by the finding that the 5 macaques with the highest levels of rotavirus IgG had detectable quantities of serum rotavirus IgA, while only 1 of the 11 macaques with little or no detectable rotavirus IgG had detectable rotavirus IgA (Table 1). These results established that rotavirus infections were, indeed, endemic within this facility. They also identified animals that had no evidence of a previous rotavirus infection based on their low titers of rotavirus IgG and, therefore, that may be more susceptible to a natural infection within the following weeks.
(ii) Detection of rotavirus in stools of naturally infected macaques. To obtain rotavirus from naturally infected macaques, stool specimens were collected two times each week from the 10 macaques with the lowest titers of serum rotavirus IgG, and with undetectable rotavirus IgA, for a period of 10 weeks beginning 13 days after they were initially screened (12 November 2002) for rotavirus antibody. Blood specimens were also collected monthly over a 27-week period to detect changes in serum rotavirus IgG titers. Rotavirus antigen was detected in 6 of the 10 macaques in at least one stool specimen (Table 2). Corresponding increases in rotavirus IgG titers were detected in every case. No gastrointestinal illness (vomiting or diarrhea) was observed in any animal in association with rotavirus detection.
In every animal in which rotavirus was detected, the quantity shed was sufficiently large so that the optical density determined by ELISA was off the scale (>3.0) in at least one specimen (Table 2). To determine the recoveries of live rotavirus in these stools, the titers of infectious virus in the specimens with the highest optical densities obtained from the first five animals that shed rotavirus were measured by a fluorescent-focus assay. Based on the numbers of focus-forming units per milliliter of stool suspension (20%, wt/vol), these titers varied widely, and only two specimens had titers of >106 FFU/ml (Table 2). Differences in titers between specimens were probably due primarily to differences in specimen conservation, since the stools were obtained in various stages of dryness from the bottoms of the cages.
Characterization of rotaviruses obtained from stool specimens of naturally infected macaques. (i) Electropherotypes of culture-adapted rotaviruses. Eight stool specimens from five macaques for which the titers of infectious rotavirus were determined were also processed and used to infect MA104 cells in an attempt to culture adapt the rotaviruses in these specimens. A complete (4+) cytopathic effect was observed within 48 h in all culture tubes inoculated with the rotavirus-positive stool preparations, and, when evaluated for the presence of rotavirus by ELISA after two cell culture passages, all eight tissue culture lysates contained large quantities of virus (i.e., all had optical densities of >3.0, even the lysate obtained from the specimen with an infectivity titer of <2.8 x 102 FFU/ml). Thus, rotaviruses in these specimens required no adaptation to grow in cell culture.
When the double-stranded RNA genomes of these eight culture-adapted isolates (passage 2) were analyzed by polyacrylamide gel electrophoresis, all had identical electropherotypes (Fig. 1A). In addition, the electropherotype of a representative of the group (macaque EA40) was compared to those of simian rotaviruses SA11 and RRV, and all three electropherotypes were found to be distinctly different (Fig. 1B).
(ii) VP7 serotypes and VP6 subgroups of the culture-adapted rotaviruses obtained from macaques. The VP7 or G serotype of simian rotaviruses previously characterized has been found to be G3 (11). We utilized serotype-specific G1 to G4 monoclonal antibodies against VP7 in an ELISA format in an attempt to determine the G type of rotaviruses in the eight culture-adapted preparations (passage 2) and found that all eight reacted strongly and specifically with the G3 MAb 159 (results not shown).
The intermediate capsid layer of the rotavirus particle is composed of the VP6 protein, the group antigen (8). Within group A rotaviruses, four subgroups have been identified based on their reactivities with subgroup-specific MAbs that recognize epitopes within the VP6 protein (13, 15, 24, 25, 43). These include SG1, SG2, SG1/SG2, and non-SG1/SG2. It has also been reported that animal rotaviruses typically belong to SG1. In contrast, most human rotaviruses belong to SG2. In an ELISA format, passage 2 of the culture-adapted rotaviruses obtained from macaques reacted strongly with MAb 225/60 to SG1 and had no reaction with the SG2 MAb 631/9 (results not shown).
Based on these results, the rotaviruses obtained from macaques in this study were classified as serotype G3 and subgroup 1, as has been found for previously characterized simian rotavirus strains. The rotaviruses isolated from all five macaques characterized in this study appeared to be similar if not identical. However, in order to maintain consistency, all further studies were conducted with virus obtained from one specimen of one macaque, i.e., EA40. The virus obtained from this animal was named TUCH for Tulane University and Cincinnati Children's Hospital.
(iii) Phylogenetic analysis of its VP4 protein indicates that TUCH belongs to a new P genotype. Although the serotype and subgroup of the macaque rotavirus were found to be typical of simian rotaviruses and its electropherotype clearly distinguished it from prototype simian rotavirus strains SA11 and RRV, it was still unclear whether TUCH was a previously unidentified rotavirus strain. To ensure that it is distinct from the other rotavirus strains, nucleotide sequence analysis of the VP4 gene of a culture-grown isolate (three passages) from macaque EA40 was performed, and the deduced amino acid sequence was compared to those of representatives of the 22 established P genotypes by phylogenetic analyses (11, 28, 37). The greatest amino acid homologies between macaque rotavirus and the other VP4 proteins were with representative P[3] rotaviruses, which included the simian RRV strain (Table 3). Homologies with these strains (85.3 to 86.2%) were, however, very similar to those found with the P[10] 69 M human strain (85.4%) and the P[2] SA11 simian strain (84.4%). VP4 proteins of representatives of six other genotypes also showed >80% homology with the macaque rotavirus VP4, suggesting that the point of divergence from the macaque VP4 protein may have been similar for many rotavirus strains.
To further examine this possibility, phylogenetic analysis of the VP4 proteins of the TUCH macaque rotavirus and representatives of the 22 previously established P types was performed, followed by the construction of a phylogenetic tree (Fig. 2). From this analysis it appears that the TUCH VP4 gene diverged from those of other genotypes at a distant point in time. This, coupled with the fact that none of these representative strains share >89% amino acid homology with the VP4 protein of the new macaque rotavirus (the suggested requirement for inclusion within the same VP4 genotype [14]), indicates that the TUCH strain belongs to a new P genotype, which we tentatively classify as P[23]).
(iv) The gene encoding TUCH VP6 also diverges from VP6 genes of other rotavirus strains. The VP6 protein or group antigen is highly conserved, especially within group A mammalian rotaviruses (48). However, because of the divergence identified between the VP4 genes of TUCH and other rotavirus strains, it was of interest to determine whether this also occurred with another TUCH gene by using phylogenetic analysis. The greatest amino acid homology between TUCH VP6 and that of another rotavirus VP6 protein deduced from nucleotide sequence analysis was determined to be 95.5% (Table 4). This degree of homology, however, is the same or very similar to that shared with VP6 proteins of multiple rotavirus strains, again suggesting that the TUCH strain diverged from other rotaviruses at a distant point in time. A phylogenetic tree constructed from the amino acid sequences of these VP6 proteins indicates that the TUCH VP6 gene, like the VP4 gene of the virus, has diverged from all other rotavirus strains examined (Fig. 3).
Inoculation of young macaques with the wild-type TUCH strain to establish a nonhuman primate model for studies on active immunity. (i) Cesarean section-derived macaques can be maintained rotavirus free within the TNPRC. Because rotavirus was found to be endemic within the TNPRC, before conducting a rotavirus challenge study it was important to demonstrate that macaques housed in the TNPRC could be maintained without experiencing a natural rotavirus infection. It was already demonstrated that animals housed within conventional rooms within this facility became naturally exposed to rotavirus and that the TUCH strain was derived from one of these animals. To determine whether newborn macaques could be kept rotavirus free if derived and maintained in the absence of older animals, a group of five macaques was obtained by cesarean section and monitored until the ages of between 3.5 and 7 months for evidence of natural rotavirus infection based upon increases in serum rotavirus antibody levels. All animals experienced a steady decline in both serum rotavirus IgG and IgA from the date of birth (cord blood) until the end of the observation period (Table 5), thus suggesting that macaques could be maintained rotavirus free within the TNPRC if derived and raised under these conditions.
(ii) Inoculation of young, rotavirus-naive rhesus macaques with the TUCH strain. A 5% suspension of stool from macaque EA40 was filtered and stored frozen in 1-ml aliquots. The titer of this preparation was 3 x 104 FFU/ml, and 1 ml was given intragastrically to each of eight cesarean section-derived rhesus macaques between 14 and 42 days of age following intragastric administration of 2 ml of 4% sodium bicarbonate. The titers of serum rotavirus IgG on the day of challenge varied from animal to animal, but all had declined from the titers found on the day of birth (Table 6). Soft stools were detected during some collections in several animals during the 14 days after challenge, but none were indicative of diarrhea. Furthermore, the behavior of all macaques remained normal during this period, without evidence of lethargy. All eight macaques shed large quantities of rotavirus in their stools, beginning between 1 and 5 days after challenge and declining to undetectable or nearly undetectable levels by the end of the 14-day observation period (Fig. 4). Even so, only three of the eight inoculated animals had detectable increases in serum rotavirus IgG titer by 1 month after inoculation (Table 6), possibly due, at least in part, to maternal antibody.
DISCUSSION
Since nonhuman primates are the animals most related to humans, it follows that they should be the most reliable for model studies using human disease agents. Accordingly, many studies on the pathogenesis of and immunity to such agents have been performed with nonhuman primates. Very few studies, however, have been conducted with rotavirus in any nonhuman primate species (21, 23, 26, 31, 36, 41, 55). A major limitation for performing such studies has been the absence of a challenge strain of simian rotavirus that will consistently result in productive infections in a nonhuman primate species after oral challenge. The only simian rotaviruses in use today are the SA11 strain, reported by Malherbe and Harwin in 1963 (27), and the RRV strain, discovered by Stuker et al. in 1980 (42). Both have been passaged numerous times in cell culture, a method commonly used for viral attenuation, and no unpassaged (wild-type) simian rotavirus is presently available. The first purpose of this study was to obtain and characterize such a virus and prepare a challenge pool for use in a nonhuman primate model for rotavirus.
Although rotaviruses are not entirely species specific (19, 32), the vast majority of natural isolates obtained from any species, including humans, have been associated specifically with that species. Since the nonhuman primates most commonly used as experimental models are macaques, the goal was to obtain a challenge rotavirus strain that is endemic in macaques and fully adapted to infect this specific nonhuman primate. Therefore, we first determined whether rotavirus was endemic in rhesus macaques housed at the TNPRC. Based on the presence of high titers of serum rotavirus IgG in all macaques >1 year of age (15 total), rotavirus is clearly endemic in macaques in this colony.
Next, we identified rhesus macaques within this colony that were the most likely to experience a rotavirus infection during the following months. Even though repeat rotavirus infections are common in humans (2) and are likely to also occur in macaques, the most severe rotavirus disease in humans almost invariably occurs after their first infection (44), provided that this infection takes place during the period of greatest susceptibility, when a child is 6 months of age or older (22). Our plan, therefore, was to identify rhesus macaques of approximately this equivalent age that had no evidence of a previous rotavirus infection based on serum rotavirus antibody levels. Accordingly, a group of 16 macaques age 2.6 to 5.9 months were evaluated for the presence of serum rotavirus IgG and IgA, and based on the absence of serum rotavirus IgA and very low levels of serum rotavirus IgG in 10 of these animals, these were considered to be the most likely to be susceptible to a rotavirus infection. Stool specimens were collected over a 10-week period from these animals, and six shed rotavirus. Although most stools had relatively low titers of infectious rotavirus, probably due to viral inactivation prior to stool collection, one specimen each obtained from two of the animals (EA40 and EA42) contained titers of >106 FFU/ml, sufficient to utilize for the production of a challenge pool of unpassaged rotavirus.
The rotaviruses obtained from naturally infected macaques were then characterized. First it was shown that they were fully adapted to grow in cell culture and produced a complete cytopathic effect in MA104 cells during their first passage. This property is typical of some animal rotavirus strains and sharply contrasts with results found with human rotaviruses, where multiple cell culture passages are required to develop this phenotypic trait (50). The culture-grown rotaviruses obtained from macaques all had identical electropherotypes which were distinct from those of SA11 and RRV. However, these viruses were identified as serotype G3 and subgroup 1, both features characteristic of simian rotaviruses (22). Furthermore, a partial sequence comparison (amino acids 51 to 206) between the VP7 protein of one of the new macaque rotaviruses and the VP7 proteins of RRV and SA11 revealed that they shared 94.9 and 95.6% homology, respectively (results not shown), supporting the indication that the new macaque rotaviruses belong to serotype G3.
Because viruses obtained from all macaques were similar if not identical, the rotavirus obtained from only one specimen from one macaque (EA40) was utilized as the representative of the new rotavirus strain, which was named TUCH. Nucleotide analysis of the complete coding sequence of the VP4 gene from the TUCH strain revealed that it is genotypically dissimilar to those from rotaviruses belonging to the 22 established P types (11, 28, 37). Rotaviruses have been classified within the same P type if they share 89% homology (14). No rotavirus belonging to any established P type was found to share >86.2% homology with the VP4 protein of TUCH. The suggestion that the TUCH VP4 gene has evolved from those of other rotaviruses was supported by phylogenetic analysis. Therefore, the TUCH strain was tentatively classified as genotype P[23].
Because the VP4 gene of TUCH appeared to have greatly diverged from VP4 genes found in other rotavirus strains, it was of interest to determine whether this could also be demonstrated for another TUCH gene. For this we selected the VP6 gene, which is highly conserved among group A rotaviruses (22). Although the degree of difference between the TUCH VP6 protein and those of numerous others to which it was compared is much less than found for the VP4 proteins, the greatest homology was still only 95.5%. Furthermore, phylogenetic analysis of TUCH VP6 supported the hypothesis that the gene encoding this protein has also substantially diverged from cognate genes of other rotavirus strains. Therefore, although TUCH has retained epitopes required for retention of the VP7 serotype G3 and VP6 subgroup 1 specificities typical of simian rotaviruses, it is quite dissimilar to the standard simian rotavirus strains that are in use in many laboratories worldwide today. In fact, based on phylogenetic analysis with the VP4 and VP6 genes, TUCH is more closely related to mouse strains than it is to the simian SA11 or RRV strain. Possibly the TUCH strain arose from a reassortment event between a simian rotavirus and a rotavirus from another species at some distant point in time. If so, some genes, such as those for VP6 and VP4, may have been derived from the nonsimian strain. Independent segregation of rotavirus genes has been described in numerous publications, and even many human strains have been identified as reassortants between animal and human rotaviruses (see, e.g., reference 20). However, it is also possible that TUCH arose from a common ancestor of both simian and other G3 strains. Sequence comparisons between other gene segments of the TUCH strain and those from multiple rotaviruses could provide a more definitive explanation of the origin of the TUCH virus.
The second purpose of this study was to determine whether an unpassaged preparation of TUCH would replicate in previously uninfected macaques, resulting in consistently high levels of fecal rotavirus shedding and, possibly, diarrheal disease. Based on this study with eight cesarean section-derived macaques age 14 to 42 days, several of which contained high levels of maternal (transplacental) rotavirus IgG on their date of inoculation, the TUCH strain appears to be excellent for model studies on rotavirus immunity in macaques. In the age range examined, which is within a window sufficient to immunize and challenge, all eight animals consistently shed large quantities of rotavirus antigen. Presumably this age range could be readily extended, since the TUCH strain itself was derived from an animal that was >6 months of age. As was found with the mouse model that we established over a decade ago for studies on rotavirus vaccines and their mechanisms of protection (51), the TUCH strain will be useful for studies on both active and passive immunity. However, as was also found with rodent and rabbit models for rotavirus in their homologous hosts (6-10, 16, 51), the TUCH strain will not be useful for studies on active immunity against diarrheal disease in macaques. In place of this, fecal shedding of the TUCH virus can be used as a marker of rotavirus infection in macaques, as has been used extensively in mice and rabbits.
ADDENDUM
After this work was completed, we became aware of a publication (24a) that describes a porcine group A rotavirus (strain A34) belonging to a potential new P genotype based on sequence analysis of part (amino acids 13 to 250) of the VP8 region of the VP4 gene of this virus (GenBank accession no. AY174094). This region of the VP4 protein of the A34 strain was found to share 79.8% homology with the cognate region of the TUCH VP4 protein, thus confirming that the P type of the A34 strain is distinct from that of TUCH. If upon analysis of the complete sequence of the VP4 gene of the A34 strain, it is confirmed to belong to a new P genotype, the TUCH strain would then be tentatively classified as P[24] rather than P[23].
ACKNOWLEDGMENTS
This study was supported by NIAID grant AI056847-01 to A.H.-C.C. and by a Cincinnati Children's Hospital Medical Center pilot grant to R.L.W. Partial support was provided by TNPRC base grant RR00164.
REFERENCES
Bernstein, D. I., V. E. Smith, J. R. Sherwood, G. M. Schiff, D. S. Sander, D. DeFeudis, D. R. Spriggs, and R. L. Ward. 1998. Safety and immunogenicity of live attenuated human rotavirus vaccine 89-12. Vaccine 16:381-387.
Bernstein, D. I., and R. L. Ward. 2004. Rotaviruses, p. 2110-2111. In R. D. Feigin, J. D. Cherry, G. J. Demmler, and S. L. Kaplan (ed.), Textbook of pediatric infectious diseases, 5th ed. Saunders, Philadelphia, Pa.
Bishop, R. F., S. R. Tzipori, B. S. Coulson, J. E. Unicomb, M. J. Albert, and G. L. Barnes. 1986. Heterologous protection against rotavirus-induced disease in gnotobiotic piglets. J. Clin. Microbiol. 24:1023-1028.
Bridger, J. C., and G. Oldham. 1987. Avirulent rotavirus infections protect calves from disease with and without inducing high levels of neutralizing antibody. J. Gen. Virol. 68:2311-2317.
Bridger, J. C., G. I. Tauscher, and U. Desselberger. 1998. Viral determinants of rotavirus pathogenicity in pigs: evidence that the fourth gene of a porcine rotavirus confers diarrhea in the homologous host. J. Virol. 72:6929-6931.
Ciarlet, M., M. E. Conner, M. J. Finegold, and M. K. Estes. 2002. Group A rotavirus infection and age-dependent diarrheal disease in rats: a new animal model to study the pathophysiology of rotavirus infection. J. Virol. 76:41-57.
Ciarlet, M., S. E. Crawford, C. Barone, A. Bertolotti-Ciarlet, R. F. Ramig, M. K. Estes, and M. E. Conner. 1998. Subunit rotavirus vaccine administered parenterally to rabbits induces active protective immunity. J. Virol. 72:9233-9246.
Ciarlet, M., M. E. Estes, C. Barone, R. F. Ramig, and M. E. Conner. 1998. Analysis of host range restriction determinants in the rabbit model: comparison of homologous and heterologous rotavirus infections. J. Virol. 72:2341-2351.
Conner, M. E., S. E. Crawford, C. Barone, and M. K. Estes. 1993. Rotavirus vaccine administered parenterally induces protective immunity. J. Virol. 67:6633-6641.
Conner, M. E., M. K. Estes, and D. Y. Graham. 1988. Rabbit model for rotavirus infection. J. Virol. 62:1625-1633.
Estes, M. K. 2001. Rotaviruses and their replication, p. 1747-1785. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed., vol. 2. Lippincott Williams and Wilkins, Philadelphia, Pa.
Fernandez, F. M., M. E. Conner, D. C. Hodgins, A. V. Parwani, P. R. Nielsen, S. E. Crawford, M. K. Estes, and L. J. Saif. 1998. Passive immunity to bovine rotavirus in newborn calves fed colostrum supplements from cows immunized with recombinant SA11 rotavirus core-like particle (CLP) or virus-like particle (VLP) vaccines. Vaccine 16:507-516.
Gorziglia, M., Y. Hoshino, K. Nishikawa, W. L. Maloy, R. W. Jones, A. Z. Kapikian, and R. M. Chanock. 1988. Comparative sequence analysis of the genomic segment 6 of four rotaviruses each with a different subgroup specificity. J. Gen. Virol. 69:1659-1669.
Gorziglia, M., G. Larralde, A. Z. Kapikian, and R. M. Chanock. 1990. Antigenic relationships among human rotaviruses as determined by outer capsid protein VP4. Proc. Natl. Acad. Sci. USA 87:7155-7159.
Greenberg, H. B., V. McAuliffe, J. Valdesuso, R. Wyatt, J. Flores, A. Kalica, Y. Hoshino, and N. Singh. 1983. Serological analysis of subgroup protein of rotavirus, using monoclonal antibodies. Infect. Immun. 39:91-99.
Guerin-Danan, C., J. C. Meslin, F. Lambre, A. Charpilienne, M. Serezat, C. Bouley, J. Cohen, and C. Andrieux. 1998. Development of a heterologous model in germfree suckling rats for studies of rotavirus diarrhea. J. Virol. 72:9298-9302.
Hoshino, Y., L. J. Saif, K. Shien-Young, M. M. Sereno, W. K. Chen, and A. Z. Kapikian. 1995. Identification of group A rotavirus genes associated with virulence of a porcine rotavirus and host range restriction of a human rotavirus in the gnotobiotic piglet model. Virology 209:274-280.
Hoshino, Y., L. J. Saif, M. M. Sereno, R. M. Chanock. 1988. Infection immunity of piglets to either VP3 or VP7 outer capsid protein confers resistance to challenge with a virulent rotavirus bearing the corresponding antigen. J. Virol. 62:744-748.
Iturriza-Gomara. M., G. Kang, A. Mammen, A. K. Jana, M. Abraham, U. Desselberger, D. Brown, and J. Gray. 2004. Characterization of G10P(11) rotaviruses causing acute gastroenteritis in neonates and infants in Vellore, India. J. Clin. Microbiol. 42:2541-2547.
Iturriza-Gómara, M., C. Wong, S. Blome, U. Desselberger, and J. Gray. 2002. Molecular characterization of VP6 genes of human rotavirus isolates: correlation of genogroups with subgroups and evidence of independent segregation. J. Virol. 76:6596-6601.
Kalter, S. S., R. L. Heberling, A. R. Rodriguez, and T. L. Lester. 1983. Infection of baboons (Papio cynocephalus) with rotavirus (SA11). Dev. Biol. Stand. 53:257-261.
Kapikian, A. Z., Y. Hoshino, and R. M. Chanock. 2001. Rotaviruses, p. 1787-1833. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed., vol. 2. Lippincott Williams and Wilkins, Philadelphia, Pa.
Leong, Y. K., and A. Awang. 1990. Experimental group A rotaviral infection in cynomolgus monkeys raised on formula diet. Microbiol. Immunol. 34:153-162.
Liprandi, F., G. Lopez, I. Rodriguez, M. Hidalgo, J. E. Ludert, and N. Mattion. 1990. Monoclonal antibodies to the VP6 of porcine subgroup I rotaviruses reactive with subgroup I and non-subgroup I non-subgroup II strains. J. Gen. Virol. 71:1395-1398.
Liprandi, F., M. Gerder, Z. Bastidas, J. A. Lopez, F. H. Pujol, J. E. Ludert, D. B. Joelsson, and M. Ciralet. 2003. A novel type of VP4 carried by a porcine rotavirus strain. Virology 314:373-380.
Lopez, S., R. Espinosa, H. B. Greenberg, and C. F. Arias. 1994. Mapping the subgroup epitopes of rotavirus protein VP6. Virology 204:153-162.
Majer, M. F. Behrens, E. Weinmann, R. Mauler, G. Maass, H. G. Baumeister, and T. Luthardt. 1978. Diarrhea in newborn cynomolgus monkeys infected with human rotavirus. Infection 6:71-72.
Malherbe, H., and R. Harwin. 1963. The cytopathic affects of vervet monkey viruses. S. Afr. Med. J. 37:407-411.
Martella, V., M. Ciarlet, A. Camarda, A. Pratelli, M. Tempesta, G. Greco, A. Cavalli, G. Elia, N. Decaro, V. Terio, G. Bozzo, M. Camero, and C. Buonavoglia. 2003. Molecular characterization of the VP4, VP6, VP7, and NSP4 genes of lapine rotaviruses identified in Italy: emergence of a novel VP4 genotype. Virology 314:358-370.
McNeal, M. M., J. Belli, M. Basu, A. H.-C. Choi, and R. L. Ward. 2004. Discovery of a new strain of murine rotavirus that is consistently shed in large quantities after oral inoculation of adult mice. Virology 320:1-11.
McNeal, M. M., M. N. Rae, J. A. Bean, and R. L. Ward. 1999. Antibody-dependent and -independent protection following intranasal immunization of mice with rotavirus particles. J. Virol. 73:7565-7573.
Mitchell, J. D., L. A. Lambeth, L. Sosula, A. Murphy, and M. Albrey. 1977. Transmission of rotavirus gastroenteritis from children to a monkey. Gut 18:156-160.
Nakagomi, O., and T. Nakagomi. 1993. Interspecies transmission of rotaviruses studied from the perspective of genogroup. Microbiol. Immunol. 37:337-348.
Oldham, G., J. C. Bridger, C. J. Howard, and K. R. Parsons. 1993. In vivo role of lymphocyte subpopulations in the control of virus excretion and mucosal antibody responses of cattle infected with rotavirus. J. Virol. 67:5012-5019.
Page, R. D. M. 1996. Treeview: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12:357-358.
Parashar, U. D., E. G. Hummelman, J. S. Bresee, M. A. Miller, and R. I. Glass. 2003. Global illness and deaths caused by rotavirus disease in children. Emerg. Infect. Dis. 9:565-571.
Petschow, B. W., R. E. Litov, L. J. T. Young, and T. P. McGraw. 1992. Response of colostrum-deprived cynomolgus monkeys to intragastric challenge exposure with simian rotavirus strain SA11. Am. J. Vet. Res. 53:674-678.
Rao, C. D., K. Gowda, and B. S. Y. Reddy. 2000. Sequence analysis of VP4 and VP7 genes of nontypeable strains identifies a new pair of outer capsid proteins representing novel P and G genotypes in bovine rotaviruses. Virology 276:104-113.
Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
Sestak, K., M. M. McNeal, A. Choi, M. J. Cole, G. Ramesh, X. Alvarez, P. P. Aye, R. P. Bohm, M. Mohamadzadeh, and R. L. Ward. 2004. Defining T-cell-mediated immune responses in rotavirus-infected juvenile rhesus macaques. J. Virol. 78:10258-10264.
Snodgrass, D. R., and P. W. Wells. 1978. Passive immunity in rotaviral infections. J. Am. Vet. Med. Assoc. 173:565-568.
Soike, K. F., G. W. Gary, and S. Gibson. 1980. Susceptibility of nonhuman primate species to infection by simian rotavirus SA-11. Am. J. Vet. Res. 41:1098-1103.
Stuker, G., L. S. Oshiro, and N. J. Schmidt. 1980. Antigenic comparisons of two new rotaviruses from rhesus monkeys. J. Clin. Microbiol. 11:202-203.
Tang, B., J. M. Gilbert, S. M. Matsui, and H. B. Greenberg. 1997. Comparison of the rotavirus gene 6 from different species by sequence analysis and localization of subgroup-specific epitopes using site-directed mutagenesis. Virology 237:89-96.
Velazquez, F. R., D. O. Matson, J. J. Calva, M. L. Guerrero, A. L. Morrow, S. Carter-Campbell, R. I. Glass, M. K. Estes, L. K. Pickering, and G. M. Ruiz-Palacios. 1996. Rotavirus infection in infants as protection against subsequent infections. N. Engl. J. Med. 335:1022-1028.
Ward, R. L. 2003. Possible mechanisms of protection elicited by candidate rotavirus vaccines as determined with the adult mouse model. Virol. Immunol. 16:17-24.
Ward, R. L., D. I. Bernstein, R. Shukla, E. C. Young, J. R. Sherwood, M. M. McNeal, M. C. Walker, and G. M. Schiff. 1989. Effects of antibody to rotavirus on protection of adults challenged with a human rotavirus. J. Infect. Dis. 159:79-88.
Ward, R. L., D. I. Bernstein, E. C. Young, J. R. Sherwood, D. R. Knowlton, and G. M. Schiff. 1986. Human rotavirus studies in volunteers: determination of infectious dose and serological response to infection. J. Infect. Dis. 154:871-880.
Ward, R. L., H. F. Clark, P. A. Offit, and R. I. Glass. 2004. Live vaccine strategies to prevent rotavirus disease, p. 607-620. In M. M. Levine, J. B. Kaper, R. Rappuoli, M. A. Liu, and M. F. Good (ed.), New generation vaccines, 3rd ed. Marcel Dekker, Inc., New York, N.Y.
Ward, R. L., J. D. Clemens, D. A. Sack, D. R. Knowlton, M. M. McNeal, N. Huda, F. Ahmed, M. Rao, and G. M. Schiff. 1991. Culture adaptation and characterization of group A rotaviruses causing diarrheal illnesses in Bangladesh from 1985 to 1986. J. Clin. Microbiol. 29:1915-1923.
Ward, R. L., D. R. Knowlton, and M. J. Pierce. 1990. Efficiency of human rotavirus propagation in cell culture. J. Clin. Microbiol. 19:748-755.
Ward, R. L., M. M. McNeal, and J. F. Sheridan. 1990. Development of an adult mouse model for studies on protection against rotavirus. J. Virol. 64:5070-5075.
Ward, R. L., O. Nakagomi, D. R. Knowlton, M. M. McNeal, T. Nakagomi, N. Huda, J. D. Clemens, and D. A. Sack. 1991. Formation and selection of intergenogroup reassortants during cell culture adaptation of rotaviruses from dually infected subjects. J. Virol. 65:2699-2701.
Woode, G. N., N. E. Kelso, T. F. Simpson, S. K. Gaul, L. E. Evans, and L. Babiuk. 1983. Antigenic relationships among some bovine rotaviruses: serum neutralization and cross-protection in gnotobiotic calves. J. Clin. Microbiol. 18:358-364.
Woode, G. N., S. Zheng, B. I. Rosen, N. Knight, N. E. Kelso Gourley, and R. F. Ramig. 1987. Protection between different serotypes of bovine rotavirus in gnotobiotic calves: specificity of serum antibody and coproantibody responses. J. Clin. Microbiol. 25:1052-1058.
Wyatt, R. G., D. L. Sly, W. T. London, A. E. Palmer, A. R. Kalica, D. H. Van Kirk, R. M. Chanock, and A. Z. Kapikian. 1976. Induction of diarrhea in colostrum-deprived newborn rhesus monkeys with the human reovirus-like agent of infantile gastroenteritis. Arch. Virol. 50:17-27.
Yuan, L., C. Iosef, M. S. P. Azevedo, Y. Kim, Y. Qian, A. Geyer, T. V. Nguyen, K. O. Chang, and L. J. Saif. 2001. Protective immunity and antibody-secreting cell responses elicited by combined oral attenuated Wa human rotavirus and intranasal Wa 2/6-VLPs with mutant Escherichia coli heat-labile toxin in gnotobiotic pigs. J. Virol. 75:9229-9238.
Yuan, L., L. A. Ward, B. I. Rosen, T. L. To, and L. J. Saif. 1996. Systemic and intestinal antibody-secreting cell responses and correlates of protective immunity to human rotavirus in a gnotobiotic pig model of disease. J. Virol. 70:3075-3083.(Monica M. McNeal, Karol S)