ELSEVIER Clinical and Diagnostic Virology

3 (1995) 29-38

Direct detection and characterization of rotavirus into subgroups by dot blot hybridization and correlation with

‘long’ and ‘short’ electropherotypes

Shobha Broor*, Mohammad Husain, Biswaroop Chatterjee, Anita Chakraborty, Pradeep Seth

Section o.f Virology, Department of' Microbiology, All India Institute qf Medical Sciences and Department of Microbiology Maulana Azad Medical College New Delhi, India

Received 8 December 1993; revised 8 April 1994; accepted 13 April 1994

Abstract

Background: Enzyme-linked immunosorbent assay (ELISA) and polyacrylamide gel electro- phoresis (PAGE) of viral RNA are well-established methods for detection of rotavirus in stool samples. Dot-blot hybridization has also been found to be a sensitive and specific technique for detection and characterization of rotaviruses.

Objectives: To compare the performance of dot blot hybridization with ELBA and PAGE for detection of rotavirus in stool samples. To assess the use of dot blot hybridization for characterization of rotaviruses into subgroups.

Study design: Stool samples were collected from 214 children presenting to the hospital with acute diarrhoea. These were assayed for rotavirus by ELBA and PAGE. Dot-blot hybridization was done with full length cloned radiolabelled c-DNA probes of gene segment 6 of SA-11 (subgroup I) and Wa (subgroup II) rotaviruses.

Results: Out of 214 stool samples 134 were found to be positive for rotavirus by one of the three methods. Among these 134 positive specimens 114 were positive by dot blot hybridiza- tion, this included 18 specimens which were positive only by dot blot assay. One-hundred- and-twelve of these 114 specimens could be subgrouped. Fifteen of these were classified as subgroup I, 97 as subgroup II and two had a dual subgroup specificity. Three subgroup 1 strains had a ‘long‘ RNA pattern, whereas one subgroup II strain had a ‘short‘ RNA pattern which has not been reported earlier for human rotaviruses.

Conclusion: Dot blot hybridization as described here is a sensitive detection and subgrouping of rotaviruses. However, as there is a diversity among rotaviruses, the panel should include probes from all segment 6.

Kevwords: Rotavirus; Dot blot hybridization; Electropherotype

and specific assay for considerable genomic the genotypes of gene

*Corresponding author. Fax: +91 I l-6862663.

092%0197/95/$9.50 0 1995 Elsevier Science B.V. All rights reserved SSDI 0928-0197(94)00020-U

30 S. Broor et aLIClinica1 and Diagnostic Virology 3 (1995) 29-38

1. Introduction

Diarrhoeal diseases are responsible for 5-10 million deaths per year in young children in the developing countries. Rotavirus has been recognised as the major cause of severe diarrhoea in infants and young children throughout the world (Kapikian and Chanock, 1985). Various methods including electron microscopy (EM), enzyme-linked immunosorbent assay (ELISA), polyacrylamide gel electro- phoresis (PAGE), nucleic acid hybridization and more recently polymerase chain reaction (PCR) has been reported for detection of rotavirus in stool samples (Brandt et al., 1981; Grauballe et al., 1981; Herring et al., 1982; Gouvea et al., 1990; Wilde et al, 1990). However, none of these methods has proved to be entirely satisfactory for detection of rotaviruses in stool specimens. While EM has a low sensitivity (Selb et al., 1985), ELISA using polyclonal sera may give false positive results and requires validation by confirmatory or blocking ELISA (Cuker and Blacklow, 1984). The detection of rotavirus by PAGE followed by silver staining though a highly specific technique lacks sensitivity as a minimun of 3-4 ng of viral RNA is needed for detection (Herring et al., 1982). PCR has the disadvantage that inhibitory substances present in stool specimens may interfere with PCR (Gouvea et al., 1990). On the other hand, dot blot hybridization using radiolabelled cDNA probes has proved to be sensitive and specific for rotavirus detection (Flores et al., 1983; Pedley and McCrae, 1984).

The rotavirus has two distinct antigenic specificities subgroup and serotype. The subgroup specificity is present on inner capsid protein, VP6 and is encoded by gene segment 6. The serotype specificity is mainly determined by VP7 antigen (G sero- types) encoded by gene segment 7,8 or 9 depending on the virus strain (Estes and Cohen, 1989). Serotype specificity is also determined to a limited extent by VP4 antigen (P serotypes), (Estes and Cohen, 1989). Largely two subgroups (I and II) have been described but the presence of a third subgroup has also been proposed (Svensson et al., 1988). In addition, the presence of dual subgroup specificity has also been reported (Urasawa et a!., 1990). The specific cDNA probes may be used not only for detection but also for characterization of human rotaviruses into subgroup and serotypes (Dimitrov et al., 1985). So far 14 G serotypes of rotavirus have been described of which G 1-4, 8 and 9 commonly cause human infections (Clark et al., 1987). Since there are 6 serotypes infecting humans, it is not feasible to use serotype specific probes for detection of rotavirus. On the other hand since there are only two subgroups, subgroup specific probes are easy to use for simulta- neous detection and characterization of rotaviruses into subgroups. In the present study we describe the use of gene 6 cDNA probes for detection and characterization of rotavirus into subgroups.

2. Material and methods

2.1. Patients and stool specimens

Stool specimens were collected over a one-year period (Dec., 1990 - Dec., 1991) from 214 children under 5 years of age who presented with acute diarrhoea at Lok

S. Broor et al./Clinical and Diagnostic Virology 3 (1995) 29-38 31

Nayak Jai Prakash Narayan Hospital, New Delhi. Some of these children were hospitalized but a large majority were seen in the out-patient clinic. A 10% suspension of stool samples was made in PBS pH 7.2 and lysis buffer (0.1 M sodium acetate with 0.5% SDS, pH 5.2) and stored at -20°C till further processing. These specimens were tested by enzyme-linked immunosorbent assay (ELISA) using reagents from Dakopatts (Dakopatts A/S, Denmark) for the presence of rotavirus antigen and by RNA electrophoresis in 7.5% polyacrylamide gel followed by silver staining (Herring et al., 1982).

2.2. Viruses

SAl 1 and Wa (Subgroup I and II respectively) rotaviruses obtained from National Institute of Health Maryland, (USA) were used as prototype strains. These strains were propagated in MA104 cells in the presence of crystalline trypsin as described by Urasawa and his colleagues ( 198 1) .

2.3. Extraction of RNA from stool specimens or tissue culture lysate

Confluent monolayers of MA104 cells in Roux bottles infected with SA-11 or Wa rotaviruses showing 3-4+ cytopathic effect were freeze-thawed three times to pre- pare tissue culture lysates. The lysate was centrifuged at 10,000 x g for 15 min at 4°C to remove cell debris. The supernatant was ultracentrifuged at 100,000 x g for 2 h at 4°C to pellet the virus. The virus pellet was suspended in 500 ul of lysis buffer. The viral RNA was extracted from stool suspensions in lysis buffer or from tissue culture lysates with equal volume of phenol/choloroform and precipitated with 2.5 volumes of ethanol in the presence of 0.3 M sodium acetate (pH 5.2). The RNA was resuspended in 100 ul of 5 mM Tris-HCl and 5 mM EDTA buffer (pH 8.0)

2.4. Preparation of 32-P labelled cDNA probes

SAll gene 6 and Wa gene 6 cloned in pBR322 plasmid DNA were obtained by courtesy of Dr. Mary K. Estes (Houston, USA) and Dr. G.W. Both (CSlRO, Sydney, Australia) respectively. These cloned plasmids were introduced into HBlOl strain of Escherichia coli and extracted by alkali lysis method (Maniatis, 1982) from the transformed E. coli. Rotavirus-specific cDNA was excised from pBR322 vector by PstI digestion and resolved in 1% low melting agarose gel in lx TBE buffer (Maniatis, 1982). The gel slice containing the insert was suspended in 0.5 ml of sterile distilled water and DNA was extracted by phenol/choloroform and precipi- tated with two-and-a-half volumes of ethanol. DNA was dissolved in 10 mM Tris- HCl and 1 mM EDTA buffer (pH 8.0). Each insert cDNA was labelled with alpha 32-p dTTP/dCTP (DuPont, USA) using random primer labelling kit from Promega (USA) to a specific activity between 2.6 x 10’ and 5.5 x lo8 cpm/ug of DNA.

32 S. Broor et al. JClinical and Diagnostic Virology 3 (1995) 29-38

2.5. Dot blot hybridization

The viral RNA extracted from stool specimens and infected tissue culture cells was denatured by boiling for 5 min and immediately chilled on ice for 5 min. Equal volume of chilled 10 x SSC (1 x SSC =0.15 M NaCl and 0.015 M Sodium Citrate) was added in each sample then it was applied onto a nitrocellulose membrane (Schleicher and Schuell, Germany) by using a manifold filteration apparatus (Biorad, USA). Duplicate membranes were prepared from the same set of specimens. The filters were vacuum baked at 80°C for 2 h. DNA from E. coli, pBR322 plasmid and nucleic acid from uninfected MA104 cells were included as unrelated controls.

The blotted filters were prehybridized for 5 h at 55°C in a solution containing 50% formamide, 6 x SSC, 5 x Denhardt’s solution (0.1% each of bovine serum albumin, ployvinylpyrrolidone and Ficoll), 0.2% SDS and 100 yg/ml of denatured salmon sperm DNA. The hybridization was carried out for 16-20 h at 55°C in the prehybridization solution containing 32-P labelled heat denatured probe (in boiling water bath for 10 min) at a concentration of 1 x lo6 cpm/ml. After hybridization, the filters were washed twice with 2 x SSC for 5 min each at room temperature followed by two washes with 2 x SSC containing 0.5% SDS for 15 min each at 65°C and two final washes were done with 0.1 x SSC for 10 min each at room temperature. The membranes were autoradiographed by exposing to Indu QX16 X-ray films (India) with intensifying screens at - 70°C and the films were developed after 48 h.

3. Results

3. I. Standardization of dot blot hybridization

The sensitivity of hybridization was determined at 55°C in the presence of 50% formamide as these conditions were found to be optimal. Ten-fold dilutions (20 ng to 2 pg) of SAll RNA were blotted on nitrocellulose membrane and hybridized with SAll gene 6 cDNA probe at a concentration of 1 x 106cpm/ml. It was seen that as little as 20 pg of total double-stranded homologus RNA could be detected (Fig. 1). However with Wa gene 6 cDNA probe the hybridization signal was very poor even with 5 ng of SAl 1 RNA. (Fig. 2a). The specifiity of SAll and Wa gene 6 cDNA probes was tested by hybridization with heterologous nucleic acid such as E. coli DNA, pBR322 DNA and nucleic acid extracted from uninfected MA104 cells. No hybridization signals were observed with these heterologous nucleic acids (Fig. 2a,b).

3.2. Detection of rotavirus in stool specimens

The ability to detect rotavirus in clinical samples by dot hybridization using both SA-11 and Wa gene 6 cDNA probes was evaluated on 214 specimens. The results of ELISA, dot blot hybridization assay and PAGE are presented in Table 1. A

S. Broor et al/Clinical and Diagnoslic Virology 3 (1995) 29-38 33

2 3 5 6

Fig. 1. Sensitivity of rotavirus dot blot assay. Different concentrations ranging from 20 ng to 2 pg (lane l-6) of ds-SAll RNA were blotted on nitrocellulose membrane and hybridized with 32-P labelled SAl 1 gene segment 6 probe at 55°C with 50% formamide; autoradiography was done for 48 h. As little as 20 pg of viral RNA was detected (lane 4).

1 2 3 4 a

A

6

C

D

b A

B

C

D

Fig. 2. Specificity of rotavirus dot blot hybridization. Blots a and b are identical blots hybridized with Wa gene 6 (a) and SAll gene 6 (b) cDNA probes. Lane 1: A, B, C and D represent Wa dsRNA, SAll dsRNA, Wa gene 6 insert and SAll gene6 insert. Lane 2: A represents RNA extracted from normal MA104 cells and B, C and D represent RNA extracted from stool samples Lane 3: A represents pBR322 plasmid DNA, B, C and D represent RNA from stool samples. Lane 4: A represents E. coli DNA, B, C and D represent RNA from stool specimens. Hybridization was done overnight at 55°C in the presence of of 50% formamide.

34 S. Broor et aL/Clinical and Diagnostic Virology 3 (1995) 29-38

Table 1 Results of rotavirus detection in stool samples by dot blot hybridization, ELBA and PAGE

Tests No. of specimens

85 10 1 18 9 11

Dot blot + + + + _ ELBA + + + + PAGE + + +

Table 2 Comparison of PAGE and ELBA with dot blot hybridization for detection of rotavirus in stool samples

Dot blot hybridization result

No. of PAGE results No. of ELISA results

+ _ Total + _ Total

-t 86 28 114* 95 19 114 _ 11 88 99 20 80 100

Total 97 116 214 115 99 214

*In one sample PAGE was not done. Sensitivity of dot blot versus PAGE, 88.6%; specificity of dot blot versus PAGE, 76%; positive predictive value, 0.75; negative predictive value, 0.88; sensitivity of dot blot versus ELBA, 83%; specificity of dot blot versus ELISA, 80.8%; positive predictive value, 0.83; negative predictive value, 0.80.

comparison of the results of dot blot hybridization with those of ELBA and PAGE are shown in Table 2.

3.3. Subgrouping of rotavirus by dot blot assay

Subgrouping was done using SAll and Wa gene 6 cDNA probes. Except 2 samples which hybridized with both the probes, the rest of the 112 samples selectively hybridized with either Wa or SAll probes (Fig. 3). Ninety-seven of these were characterized as subgroup I and 1.5 as subgroup II. Of these 112, in 86 samples RNA electrophoresis patterns were descernible. The correlation of subgroups with electropherotypes is shown in Table 3.

4. Discussion

Rotavirus infection can be diagnosed by the detection of viral antigen or viral genomic RNA. Although RNA detection assays have generally not been proved to be more sensitive than immunoassay procedures (Dimitrov et al., 1985) but these have an advantage that viral RNA can be detected even in absence of immunoreactive antigens. Moreover, RNA detection methods have the potential for characterization

S. Broor et al.lClinical and Diagnostic Virology 3 (1995) 29-38 35

a 1234567 8 9 10 11 12

A

D

E

F

G r)

H

b A

F

G

H

Fig. 3. Subgrouping results on stool samples. Two representative identical blots a and b were hybridized with Wa gene 6 and SAll gene 6 cDNA probes respectively. Lane 12: row E represents Wa dsRNA, row F and G represent SAl 1 ds RNA and SAl 1 gene 6 cDNA insert respectively and row H represents pBR322 plasmid DNA. Row D in lane 12 represents a sample which has a dual specificity. Blot (b) lane 1, row H represents a sample with subgroup I specificity and short RNA pattern. Rest all dots represent RNA extracted from stool specimens.

of viral genotype (Dimitrov et al., 1985). Serotype specific probes of gene segment 7 have been used to characterise rotaviruses into serotypes using dot blot hybridiza- tion (Flores et al., 1989). The present paper describes the use of dot blot assay using gene segment 6 c-DNA probes for detection and characterization of rotaviruses into subgroups.

Dot blot hybridization using radiolabelled cDNA probes has been reported for detection of rotaviruses on clinical samples in earlier studies (Pedley and McCrae,

36 S. Broor et al./Clinical and Diagnostic Virology 3 (1995) 29-38

Table 3 Correlation of subgrouping results by dot blot hybridization with RNA PAGE patterns

No. of samples PAGE pattern Subgroup as determined by dot blot hybridization

74 Long II 1 Short II 8 Short I 3 Long I

1984; Dimitrov et al., 1985). The sensitivity and specificity of dot blot hybridization varies with the type of probe and stringency of hybridization. The sensitivity of dot blot assay as reported here for homologous RNA is comparable to earlier reports in which probes specific for gene segments 7 and 6 or 9 were used (Pedley and McCrae, 1984; Dimitrov et al., 1985).

On comparing the results of rotavirus detection by ELISA, PAGE and dot blot hybridization, it was observed that in 85 samples the results of the three assays were similar. Ten samples were negative by PAGE but positive by ELISA and dot blot hybridization. This seems to be due to the low quantity of viral RNA which could not be visualized by PAGE (the sensitivity limit of PAGE followed by silver staining is 3-4 ng of viral genomic RNA (Herring et al., 1982). Both ELISA and dot blot hybridization have higher sensitivity than PAGE (Arens and Swierkosz, 1989). The failure to detect rotavirus by ELISA in 19 samples positive by dot blot assay may be due to the presence of blocking factors in the stool samples (Watanabe et al., 1978) or due to the loss of immunoreactive antigens on either long term storage or repeated freeze-thawing. The quantity of RNA in these samples may also be insuffi- cient and thus only one of these 19 samples was positive by PAGE.

Most of the human rotaviruses can be classified into sugbroup I or II but some strains of non-subgroup I and II specificity (Svensson et al., 1988) and others with both subgroup I and II specificities (Urasawa et al., 1990) have been described. The most likely explanation for those samples which were negative by dot blot hybridiza- tion but positive by ELISA or PAGE may be that they belonged to non-subgroup I and II specificity. Gene 6 sequence shows about 78% homology (Both et al., 1984), but under high stringency conditions, there is minimal or no cross-hybridization between the different subgroups. The detection limit for heterolgous RNA has been reported to be 0.5 ng to 31 ng (Dimitrov et al., 1985). In our study, the detection limit for heterologous RNA was approximately 5 ng. Due to this lower detection limit, heterologous RNA samples may be false negative in dot blot hybridization. Thus, cDNA probes from non-subgroup I and II specificity should also be included to detect these strains.

Most of the human rotavirus belonging to subgroup I have ‘short’ electrophero- type pattern characterized by slow migration of 10th and 11th RNA segments, whereas subgroup II rotaviruses have ‘long’ RNA pattern characterized by relatively faster migration of 10th and 11 th RNA segments (Kalica et al., 1981). Exceptions

S. Broor et al./Clinical and Diagnostic Virologq 3 (1995) 29.-38 37

of subgroup-electropherotype linkage have been reported in earlier studies. Subgroup I with ‘long’ RNA pattern have been reported from Phillipines and India (Kobayashi et al.. 1989; Ghosh and Naik, 1989). In the present study, 3 samples characterized as subgroup I by dot blot hybridization had a ‘long’ RNA pattern. There is no report to describe subgroup II strains with ‘short’ electropherotype pattern. We report here a strain which was characterised as subgroup II by dot blot hybridization but had ‘short’ electropherotype pattern. It is not surprising that subgroup specifi- cities do not conform to the well described electropherotype patterns since subgroup specificity is determined by gene 6, whereas electropherotype depends upon the migration of 10th and 1 lth RNA segments.

Dot blot hybridization as described here appears to be sensitive and specific technique for detection and subgrouping of rotavirus in stool samples. However, a number of probes from all the genotypes of a particular gene segment should be included in the panel. The difficulties in getting subgroup-specific monoclonal anti- bodies for ELISA may be overcome by using dot blot hybridization for characteriza- tion of rotavirus in terms of both speed and accuracy. The modification of our procedure using non-radioactive or synthetic probes may further make this method simpler and cheaper for use in developing countries.

Acknowledgements

We are grateful to Dr. Mary K. Estes and Dr. G.W. Both for providing us rotavirus SAll gene 6 and Wa gene 6 cloned plasmids respectively. This work was supported by a grant from Department of Science and Technology, Govt. of India.

References

Arens, M. and Swierkosz, E.M. (1989) Detection of rotavirus by hybridization with a non-radioactive synthetic DNA probe and comparison with commercial enzyme immunoassays and silver-stained polyacrylamide gels. J. Clin. Microbial. 27, 127771279.

Both, G.W., Siegman, L.J., Bellamy, A.R., Ikegami, N., Shatkin, A.J. and Furuichi, Y. (1984) Comparative sequence analysis of rotavirus genomic segment 6 the gene specifying viral subgroups 1 and 2. J. Virol. 51, 97-101.

Brandt, C.D., Kim. H.W., Rodriguez, W.J., Thomas, L., Yolken, R.H.. Arrobio. J.O., Kapikian. A.Z., Parrott. R.H. and Chanock, R.M. ( 1981) Comparison of direct electron microscopy, immune electron microscopy and rotavirus enzyme-linked immunosorbent assay for detection of gastroenteritis viruses in children. J. Chn. Microbial. 13, 976-981.

Clark, H.F., Hoshino, Y., Bell, L.M., Groff, J., Hess, G., Buchman, P. and Offit, P.A. (1987) Rotavirus isolate Wl61 representing a presumptive new human serotype. J. Clin. Microbial. 25, 175771762

Cuker, G. and Blacklow, N.R. (1984) Human viral gastroenteritis. Microbial Rev. 48, 1577179. Dimitrov, D.H., Graham, D.Y. and Estes, M.K. (1985) Detection of rotavirus by nucleic acid

hybridization with cloned DNA of simian rotavirus SAl 1 genes. J. Infect. Dis. 152, 293 -300. Estes, M.K. and Cohen, J. (1989) Rotavirus gene structure and function. Microbial. Rev. 53. 410-449. Flores, J., Purcell. R.H., Perez, I., Wyatt, R.G., Boeggman, E., Sereno, M., White, L., Chanock, R.M.

and Kapikian. A.Z. (1983) A dot hybridization assay for detection of rotavirus. Lancet i, 555~ 559. Flores, J., Green, K.Y., Garcia, D., Sears, J., Perez-Schael, I., Avendano. L.F., Rodriguez. W.B.,

38 S. Broor et al./Clinical and Diagnostic Virology 3 (199.5) 29-38

Taniguchi, K., Urasawa, S. and Kapikian, A.Z. (1989) Dot hybridization for distinction of rotavirus serotypes. J. Clin. Microbial. 27, 29-34.

Ghosh, SK. and Naik, T.N. (1989) Detection of a large number of subgroup 1 human rotaviruses with a ‘long’ RNA electropherotype. Arch. Virol. 105, 119-127.

Gouvea, V., Glass, RI., Woods, P., Taniguchi, K., Clark, H.F., Forrester, B. and Fang, Z.-Y. (1990) Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens. J. Clin. Microbial. 28, 276-282.

Grauballe, P.C., Vestergaard, B.F., Meyling, A. and Genner, J. (1981) Optimized enzyme-linked immunosorbent assay for detection of human and bovine rotavirus in stools: comparison with electron-microscopy, immunoelectro-osmophoresis and fluorescent antibody techniques. J. Med. Virol. 7, 29940.

Herring, A.J., Inglis, N.F., Ojeh, C.K., Snodgrass, D.R. and Menzies, J.D. ( 1982) Rapid diagnosis of rotavirus infection by direct detection of viral nucleic acid in silver stained polyacrylamide gels. J. Clin. Microbial. 16, 4733477.

Kalica, A.R., Greenberg, H.B., Espejo, R.T., Flores, J., Wyatt, R.G., Kapikian, A.Z. and Chanock, R.M. (1981) Distinctive ribonucleic acid patterns of human rotavirus subgroup 1 and 2. Infect. Immunol. 33, 9588961.

Kapikian, A.Z. and Chanock, R.M. (1985) Rotaviruses. In: B.N. Fields, D.N. Knipe, R.M. Chanock, J.L. Melnick, B. Roizman and R.E. Shope (Eds.), Virology. Raven Press, New York. pp. 863-906.

Kobayashi, N., Lintag, I.C., Urasawa, T., Taniguchi, K. Saniel, M.C. and Urasawa, S. (1989) Unusual human rotavirus strains having subgroup I specificity and ‘long’ RNA electropherotype. Arch. Virol. 109, 11-23.

Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning. A Laboratory Manual, 1st ed., Cold Spring Harbor Laboratory, pp. 86.

Pedley, S. and McCrae, M.A.(1984) A rapid assay for detecting RNA species in field isolates of rotaviruses. J. Virol. Methods 9, 173-181.

Selb, B., Baumeister, H.G., Maass, G. and Doerr, H.W.( 1985) Detection of rotavirus by nucleic acid analysis in comparison to enzyme-linked immunoassay and electron microscopy. Eur. J. Clin. Microbial. 4, 41-45.

Svensson, L., Grahnquist, L., Petterson, C., Grandien, M., Stintzing, G. and Greenberg, H.B. (1988) Detection of human rotaviruses which do not react with subgroup I- and II-specific monoclonal antibodies. J. Clin. Microbial. 26, 123881240.

Urasawa, T., Urasawa, S. and Taniguchi, K. (1981) Sequential passages of human rotavirus in MA-104 cells, Microbial. Immunol. 25, 102551035.

Urasawa, T., Taniguchi, K., Kobayashi, N., Wakasugi, F., Oishi, I., Minekawa, Y., Oseto, M., Ahmed, M.U. and Uraswa, S. (1990) Antigenic and genetic analyses of human rotavirus with dual subgroup specificity. J. Clin. Microbial. 28, 283772841.

Watanabe, H., Gust, I.D. and Holmes, I.H. (1978) Human rotavirus and its antibody: their coexistence in feces of infants. J. Clin. Microbial. 7, 405-409.

Wilde, J., Eiden, J. and Yolken, R. (1990) Removal of inhibitory substances from human fecal specimens for detection of group A rotaviruses by reverse transcriptase and polymerase chain reaction. J. Clin. Microbial. 28, 1300-1307.