Arch Virol (2007) 152: 199–208DOI 10.1007/s00705-006-0831-yPrinted in The Netherlands

Genetic analysis of an ADRV-N-like novel rotavirusstrain B219 detected in a sporadic case of adult

diarrhea in Bangladesh

Brief Report

M. M. Alam1, N. Kobayashi2, M. Ishino2, M. S. Ahmed1, M. U. Ahmed1,S. K. Paul3, B. K. Muzumdar4, Z. Hussain4, Y.-H. Wang5, and T. N. Naik6

1Department of Veterinary Medicine, Faculty of Veterinary Science,Bangladesh Agricultural University, Mymensingh, Bangladesh

2Department of Hygiene, Sapporo Medical University School of Medicine,Sapporo, Japan

3Department of Microbiology, Mymensingh Medical College,Mymensingh, Bangladesh

4Department of Pediatrics, Sher e Bangla Medical College, Barisal, Bangladesh5Virology section, Wuhan Centers for Disease Prevention and Control, Wuhan, China

6Division of Virology, National Institute of Cholera and Enteric Diseases,Kolkata, India

Received May 8, 2006; accepted June 30, 2006Published online August 11, 2006 c© Springer-Verlag 2006

Summary. An unusual human rotavirus strain B219 was detected in a stoolspecimen from a 65-year old patient with diarrhea in Bangladesh during April2002. Cloning and sequence analysis of five genes of the B219 strain indicatedthat this virus is genetically closely related to the ADRV-N strain, which caused anadult diarrhea outbreak in China, but distinct from groups A, B, and C rotavirusesknown to cause diarrheal diseases in humans. Accordingly, rotavirus strains B219and ADRV-N were considered to belong to a novel group of human rotavirus, andthe ADRV-N-like novel human rotaviruses were suggested to be distributed to ageographically wider area.

∗Members of the genus Rotavirus of the family Reoviridae, have been recognized ascausative agents of gastroenteritis in human and animals.As a genome, rotaviruseshave 11 segments of double-stranded (ds) RNA encoding 6 structural proteins and6 nonstructural proteins [3]. The rotavirus particle is constituted by three layers,

200 M. M. Alam et al.

i.e., the outer capsid layer consisting of VP4 and VP7, the inner capsid layer withVP6 and VP2, and the inner core consisting of VP1–3.

The genus Rotavirus has been differentiated into seven species (groups), i.e.Rotaviruses A–G (groups A–G) based on antigenicity of VP6 and genetic char-acteristics, although groups F and G are tentative species [9]. Among them, as acause of diarrheal diseases in humans, groups A, B, and C rotaviruses have beendetected so far [9, 13]. Group A rotavirus is the most common cause of diarrhea ininfants and young children worldwide. Group C rotavirus also causes diarrhea inchildren, although its prevalence is considerably lower than group A rotavirus.In contrast, group B rotavirus causes severe diarrhea, primarily in adults, and hasbeen detected only in China, India, and Bangladesh [6, 11, 14]. In addition to thesethree rotavirus species, a novel human rotavirus designatedADRV-N was reportedas a cause of a diarrheal outbreak among adults in China in 1997 [18]. Sequenceanalysis of the three RNA segments encoding VP6, NSP1, and NSP3 suggestedthat ADRV-N might be genetically distinct from groups A to G throughout [19],although its species category (group) has not yet been formally assigned.

Since the outbreak of adult diarrhea in China due to ADRV-N, similar virushad never been detected in China or in other countries until our present report.Although ADRV-N is a notable rotavirus with respect to pathogenicity in adultsand novelty in viral taxonomy, its distribution in and outside China and prevalenceand significance in adult diarrheal diseases are still unknown. To elucidate epi-demiogical and virological features of ADRV-N, further detection and analysis ofADRV-N-like viruses are indispensable. Recently, we detected an ADRV-N-likerotavirus from an adult diarrheal case in Bangladesh, for the first time after thereport of ADRV-N. In the present study, we analyzed the newly detected ADRV-N-like rotavirus genetically and elucidated its relatedness to ADRV-N as well asgroups A, B, and C rotaviruses.

The epidemiologic study on the prevalence of group B rotavirus in adultdiarrhea in Bangladesh was conducted during a period between March 2000 andDecember 2002 [1]. In this period, a total of 959 stool specimens were collectedfrom adult diarrhea patients who visited S. K. Hospital or Mymensingh MedicalCollege Hospital in Mymensingh, Bangladesh, and the presence of rotavirus RNA

�Fig. 1. Genetic profiles of rotavirus strain B219. a RNA pattern of rotavirus strain KU(group A, 1), B219 (novel rotavirus, 2), and Bang373 (group B, 3) after polyacrylamidegel electrophoresis and silver staining. RNA segment numbers on the left are assigned toB219 RNA segments. b Phylogenetic tree for the VP6 genes of representative rotavirusesbelonging to groups A, B, and C, and novel rotavirus strains B219 and ADRV-N. The treewas constructed by the neighbor-joining method using the ClustalW program. The scale barindicates the branch length that corresponds to 0.1 nucleotide substitutions per nucleotide site.Bootstrap probabilities of 1000 trials are indicated at diverging points of branches. The VP6gene sequences of following strains were taken from the GenBank database (accession nos.):US1205 (AF079357), CJN (AF461757), Wa (K02086), RV-3 (U04741), I321 (X94618), IS2(X94617), Jajeri (AF325805), Bristol (X59843), Preston (M94156), ADRV-N (AY632080),Bang373 (AY238389), ADRV (M55982), CAL-1 (AB037931), and IDIR (M84456)

ADRV-N-like novel rotavirus strain in Bangladesh 201

segments was investigated by polyacrylamide gel electrophoresis of extractedRNA, followed by silver staining of gels [5]. Among the specimens, the RNApattern of group B rotavirus was detected in 12 specimens, and an unusual RNApattern was found in one specimen. Group A rotavirus was detected in 2.8% ofall the specimens. The unusual RNA pattern from the rotavirus designated B219

202 M. M. Alam et al.

was evidently different from those of known rotavirus groups (Fig. 1(a)), while itwas similar to the RNA pattern of ADRV-N [19].

The rotavirus B219 was found in a stool specimen from a 65-year old femalepatient exhibiting severe diarrhea in April 2002. This adult diarrheal case occurredsporadically, and gastroenteritis symptoms were not present in two other members(daughters) of the patient’s family. The patient lived in a residential area ofMymensingh city and kept no domestic animals, and there was no informationindicating contact with animals. The patient recovered from the diarrheal diseaseafter rehydration treatment.

Monoclonal antibody to group A rotavirus VP6 common antigen did not reactwith B219 in ELISA assay [15]. Furthermore, none of the groups A, B, or Crotavirus-specific gene sequences was detected in B219 by RT-PCR [4, 16]. Tissueculture of rotavirus B219 was attempted using MA104 cells, CV-1 cells, and Caco-2 cells, but propagation of the virus was not observed, although ADRV-N wasreported to be propagated in primary human embryo kidney cells [8, 19].

To know the genetic characteristics of the rotavirus B219, viral gene segmentswere cloned by the single primer amplification method as described previously[12, 17]. As a result, five viral genes were cloned, and their sequences were de-termined by the dideoxynucleotide chain termination method using an automatedsequencer (ABI Model 310). For each gene, at least three independent cloneswere isolated and sequenced. In the single primer amplification method, there isa possibility that a nonspecific nucleotide sequence is amplified because of thepresence in the specimen of non-rotavirus dsRNA, nonspecific reaction of thesingle DNA primer to DNA in the specimen, or ligation of the single primer toDNA by T4 RNA ligase despite its extremely slow reaction rate [2]. Accordingly,attention was given to extraction and purification of viral RNA from the fecalsample, and genetic analysis of the sequence data obtained by this method.

Sequence data were analyzed by using the Gene Works software package(IntelliGenetics, Inc., Mountain View, CA), and multiple sequence alignment andphylogenetic analysis with other rotavirus sequences were performed using theClustalW program. Based on the gene size and sequence homology with rotavirusgenes published previously [19], the gene segments 5 and 6 were considered tobe the NSP1 and VP6 genes, respectively, and the other genes were consideredto encode NSP2, NSP3, and VP7, corresponding to one of segments 7, 8, or 9.The nucleotide sequences of B219 genes analyzed in this study were depositedin GenBank under the following accession nos.: VP6 gene, DQ168033; VP7gene, DQ168034; NSP1 gene, DQ168032; NSP2 gene, DQ168035; NSP3 gene,DQ168036.

A terminal GG-sequence common to members of the genus Rotavirus [3, 13]was located at the 5′-end of all the cloned genes, while at the 3′-end of the genes,two similar sequences, -AUACACCC (NSP1 and NSP3 genes) and -AUAUACCCC (VP6 and NSP2 genes) were found. The latter 3′-end sequence was identicalto that reported for ADRV-N genes, but different from those of groups A, B, and Crotaviruses [19]. The NSP1 gene (1307 nucleotides),VP6 gene (1287 nucleotides),and NSP3 gene (932 nucleotides) of B219 contain ORFs encoding products of

ADRV-N-like novel rotavirus strain in Bangladesh 203

Table 1. Sequence identities of B219 gene segments to representative rotavirus strains fromdifferent groups

Rotavirus strains Nucleotide (amino acid) sequence identity (%) of B219 genescompared (origin)

VP6 gene VP7 gene NSP1 gene* NSP2 gene NSP3 gene

Group AKU (human) 41.8 (11.9) 31.8 (17.7) 34.7 (13.6) 34.2 (15.8) 24.5 (5.8)UK (bovine) 43.4 (13.5) 33.2 (17.1) 21.8 (10.3) 44.3 (14.2) 26.1 (14.2)

Group BADRV (human) 54.8 (35.9) 47.9 (20.8) 37.9 56.5 (40.2) 23.8 (19.1)CAL-1 (human) 54.5 (37.0) 41.8 (21.5) 38.7 56.2 (44.6) 22.6 (18.4)IDIR (murine) 54.9 (37.2) 53.5 (19.5) 41.3 57.1 (45.9) 37.3 (22.1)WD653 (bovine) –** 44.7 (19.6) – – –DB101 (bovine) – 33.0 (21.2) 42.7 – –KB63 (ovine) – – 41.1 – –

Group CBristol (human) 41.1 (14.0) 11.2 (15.5) 36.7 (13.5) 47.5 (12.8) 21.9 (10.7)Cowden (porcine) 14.5 (14.3) 17.6 (10.3) 40.5 (9.4) 46.4 (13.8) 23.1 (11.7)

Novel groupADRV-N (human) 93.9 (98.7) – 94.9 (94.9) – 87.1 (93.5)

*Amino acid sequence identity to group B rotavirus NSP1 was not calculated becausethere are two or three overlapping ORFs in the NSP1 gene, unlike other rotavirus groupshaving a single ORF

**Sequence identity was not calculated because the sequence data of rotavirus strainscompared were not available in GenBank database

395, 396, and 262 amino acids, respectively, which are identical amino acid num-bers to those reported for the cognate genes of ADRV-N [19]. Protein products ofNSP2 and VP7 genes are considered to be comprised of 297 and 258 amino acids,respectively.

Nucleotide and deduced amino acid sequence identities between B219 andrepresentative strains of groups A, B, and C rotaviruses and the ADRV-N strainare shown in Table 1. Although nucleotide sequences of five B219 genes showedless than 60% identity (less than 50% identity at amino acid level) to those ofgroups A, B, and C rotaviruses, sequence identities with three ADRV-N genes(VP6, NSP1, and NSP3 genes) were extremely high (more than 90% at aminoacid level). Among viral genes from groups A, B, and C rotaviruses, the VP6and NSP2 genes from group B rotaviruses exhibited slightly higher sequenceidentities to the B219 genes (54.5–57.1%) than other gene segments. However,similar levels of sequence identities were found between group A rotaviruses(strains KU, UK) and group C rotaviruses (strains Bristol, Cowden) (56.1–57.2%in VP6 genes, and 54.9–57.9% in NSP2 genes), when published sequence data(GenBank accession nos.: VP6 gene, AB022768 (KU), X53667 (UK), X59843(Bristol), M94157 (Cowden); NSP2 gene, AB022770 (KU), J02420 (UK),

204 M. M. Alam et al.

AJ132205 (Bristol), X65939 (Cowden)) were analyzed in the present study. In thephylogenetic tree based on the VP6 sequence (Fig. 1(b)), B219 and ADRV-N areclassified into a single cluster independent of groupsA, B, and C rotavirus clusters.Therefore, together with ADRV-N, B219 probably belongs to a novel rotavirusgroup (species) which is genetically distinct from groups A, B, or C rotaviruses.

Fig. 2 (continued)

ADRV-N-like novel rotavirus strain in Bangladesh 205

Fig. 2. Alignment of deduced amino acid sequences of rotavirus inner capsid protein VP6(a) and outer capsid protein VP7 (b) among group B rotaviruses (ADRV and CAL-1) andnovel rotaviruses (B219 and ADRV-N). VP6 sequences of ADRV, CAL-1, and ADRV-N wereobtained from GenBank, accession nos. M55982, AB037931, and AY632080, respectively,and VP7 sequences of ADRV and CAL-1 were from M33872 and AF184083, respectively.* and :, conserved and homologous amino acid, respectively. a #, Position of different aminoacid betweenADRV-N and B219. @, Proline residue conserved in all the sequences compared.Highly conserved regions which are comprised by more than 10 consecutive amino acids andcontain more than 65% conserved amino acids are indicated by underlines. b # and @,conserved cysteine and proline residues, respectively. Amino acid sequences indicated byunderlines represent putative N-glycosylation sites (B219, amino acid nos. 42–44; CAL-1and ADRV, amino acid nos. 45–57, 91–93, and 105–107). Dotted lines above and below thealignment represent putative hydrophobic regions of CAL-1 and B219 strains, respectively,

predicted by a hydropathicity profile using the Kyte-Doolittle method

Alignment of VP6 from novel rotaviruses (B219 andADRV-N), and two groupB human rotaviruses (ADRV and CAL-1) is shown in Fig. 2(a). VP6 amino acidsequences of B219 and ADRV-N are different at five positions, indicated by # inFig. 2(a), where amino acids from the two strains are structurally and/or function-ally similar. Accordingly, VP6 proteins of B219 and ADRV-N are considered asvirtually identical. Between the novel rotaviruses and group B rotaviruses, amino

206 M. M. Alam et al.

acid sequence similarity of VP6 was as high as 59%, although sequence identitywas 35%. Over the whole VP6 sequence, nine proline residues were conservedbetween these two rotavirus groups, and three regions near the 5′-terminus andone region at 3′-terminus were highly conserved. In addition, the hydropathicityprofile of B219 (ADRV-N) VP6 predicted by the Kyte-Doolittle method is similarto that of group B rotavirus ADRV in two regions (amino acid nos. 1–120, and210–310), while similarity to group A and C rotavirus VP6 was not evident (datanot shown). These findings suggested that the VP6 of novel rotaviruses B219 andADRV-N may be genetically and functionally related to that of group B rotavirus.

Deduced amino acid sequence of B219 VP7 and alignment with VP7 of groupB rotavirusesADRV and CAL-1 are shown in Fig. 2(b). In the B219VP7 sequence,a highly hydrophobic region of a putative signal peptide is located at N-terminus,as in the other known rotaviruses. Between B219 and ADRV (or CAL-1), aminoacid sequence similarity was 43%, while sequence identity was 21–22%. Fivecysteine residues and a single proline residue are conserved among VP7 of B219,ADRV, and CAL-1, and putative hydrophobic regions obtained by hydropathicityprofile were mostly consistent between the novel rotavirus and group B rotaviruses.B219 VP7 has only one putative N-glycosylation site (amino acid nos. 42–44),which is located at a position (amino acid nos. 55–57) close to one of the two N-glycosylation sites inADRV and CAL-1.These findings suggested that the putativeVP7 product from B219 has common characteristics to group B rotavirus VP7.

Our present study on the rotavirus strain B219 indicated that it is closelyrelated genetically to the Chinese novel adult diarrhea rotavirus ADRV-N. Thenovel rotaviruses B219 and ADRV-N are considered to belong to a new rotavirusgroup (species) which is distinct from groupsA, B, and C rotaviruses. The group ofthese novel rotaviruses is suggested to also be different from other rotavirus groups(D–G), because RNA patterns of these rotavirus groups are evidently differentfrom those of B219 and ADRV-N [9], although gene sequence data of groupsD–G are not available. Although we have not yet succeeded in establishing tissueculture of B219, and therefore antisera to this virus is not available, serologicalverification may be needed to discriminate the proposed group from other knownrotavirus groups. For this purpose, construction of virus-like particles producedin insect cells using baculovirus recombinant expression system may be useful inthe near future.

It is noteworthy that the rotavirus B219, as well as ADRV-N, is geneticallyrelated to group B rotavirus to some extent, although the novel rotaviruses can beclearly distinguished from group B rotaviruses based on sequence identities. B219showed generally higher sequence identities to group B rotaviruses than groups Aand C rotaviruses, and structural similarities were observed in VP6 and VP7 asdescribed above. Furthermore, both novel rotaviruses (B219 and ADRV-N) andgroup B rotaviruses have been detected mainly in adult diarrheal cases, in Chinaand the Indian Subcontinent. It is also interesting that sequence identities (threegene segments) between ChineseADRV-N and Bangladeshi B219 (87–95%) weresimilar to those observed for group B rotaviruses between those two geographicalregions (90–94%, nine gene segments) [1, 10, 20]. From these findings, it is

ADRV-N-like novel rotavirus strain in Bangladesh 207

suggested that the B219-like novel rotaviruses may be ecologically similar togroup B human rotaviruses, and it is also possible that these two rotavirus groupsmight have originated from a common ancestral virus. Although the ecological“niche” of the ADRV-N/B219 is not clear at present, an animal reservoir (e.g., ratsand pigs) has been considered for group B rotavirus in China [7]. Thus, using theanalogy of group B rotavirus, there is a possibility that ADRV-N/B219 may alsoexist in certain animal(s) and sometimes cause infection in humans via zoonotictransmission.

Our present study indicated that a human rotavirus belonging to a novelgroup that was first reported in China has also been disseminated in Bangladesh.The means for detection of this novel rotavirus are still limited, and extensiveepidemiologic study will be necessary to determine the prevalence of this rotavirusand the impact it will have on diarrheal diseases in adults.

Acknowledgement

This study was supported in part by a Grant-in-Aid for Scientific Research (Grant no.18406018) from the Ministry of Education, Culture, Sports, Science, and Technology ofJapan, and a Grant for International Health Cooperation Research (18-5) from the Ministryof Health, Labour, and Welfare of Japan.

References1. Ahmed MU, Kobayashi N, Wakuda M, Sanekata T, Taniguchi K, Kader A, Naik TN,

Ishino M, Alam MM, Kojima K, Mise K, Sumi A (2004) Genetic analysis of groupB human rotaviruses detected in Bangladesh in 2000 and 2001. J Med Virol 72:149–155

2. Brennan CA, Manthey AE, Gumport RI (1983) Using T4 RNA ligase with DNAsubstrates. Methods Enzymol 100: 38–52

3. Eates MK (2001) Rotaviruses and their replication. In: Knipe DM, Howley PM et al. (eds)FieldsVirology, 4th edn. Lippincott Williams & Wilkins, Philadelphia, PA, pp 1747–1785

4. Gouvea V, Allen JR, Glass RI, Fang ZY, Bremont M, Cohen J, McCrae MA, Saif LJ,Sinarachatanant P, Caul EO (1991) Detection of group B and C rotaviruses by polymerasechain reaction. J Clin Microbiol 29: 519–523

5. Herring AJ, Inglis NF, Ojeh CK, Snodgrass DR, Menzies JD (1982) Rapid diagnosisof rotavirus infection by direct detection of viral nucleic acid in silver-stainedpolyacrylamide gels. J Clin Microbiol 16: 473–477

6. Hung T, Chen G, Wang C, Fang Z, Chou Z, Chang X, Liong X, Yao H, Chao T, Ye W,Den S, Chang W (1984) Waterborne outbreak of rotavirus diarrhoea in adults in Chinacaused by a novel rotavirus. Lancet 1: 1139–1142

7. Hung T (1988) Rotavirus and adult diarrhea. Adv Virus Res 35: 193–2188. Ji S, BiY,Yang H,Yang F, Song J, Tao X, Cui X (2002) Cultivation and serial propagation

of a new rotavirus causing adult diarrhea in primary human embryo kidney cells.Zhonghua Yi Xue Za Zhi (Natl Med J China) 82: 14–18 (in Chinese; English abstract)

9. Kapikian AZ, Hoshino Y, Chanock RM (2001) Rotaviruses. In: Knipe DM, Howley PMet al. (eds) Fields Virology, 4th edn. Lippincott Williams & Wilkins, Philadelphia, PA,pp 1787–1833

10. Kobayashi N, Naik TN, Kusuhara Y, Krishnan T, Sen A, Bhattacharya SK, Taniguchi K,Alam MM, Urasawa T, Urasawa S (2001) Sequence analysis of genes encoding structural

208 M. M. Alam et al.: ADRV-N-like novel rotavirus strain in Bangladesh

and nonstructural proteins of a human group B rotavirus detected in Calcutta, India.J Med Virol 64: 583–589

11. Krishnan T, Sen A, Choudhury JS, Das S, Naik TN, Bhattacharya SK (1999) Emergenceof adult diarrhoea rotavirus in Calcutta, India. Lancet 353: 380–381

12. Lambden PR, Cooke SJ, Caul EO, Clarke IN (1992) Cloning of noncultivatable humanrotavirus by single primer amplification. J Virol 66: 1817–1822

13. Mackow ER (1995) Group B and C rotavirus. In: Blaser MJ, Smith PD, Ravdin JI,Greenberg HB, Guerrant RL (eds), Infections of the Gastrointestinal Tract. Raven Press,New York, pp 983–1008

14. Sanekata T, Ahmed MU, Kader A, Taniguchi K, Kobayashi N (2003) Human group Brotavirus infections cause severe diarrhea in children and adults in Bangladesh. J ClinMicrobiol 41: 2187–2190

15. Taniguchi K, Urasawa T, Urasawa S, Yasuhara T (1984) Production of subgroup-specificmonoclonal antibodies against human rotaviruses and their application to an enzyme-linked immunosorbent assay for subgroup determination. J Med Virol 14: 115–125

16. Taniguchi K,Wakasugi F, PongsuwannaY, Urasawa T, Ukae S, Chiba S, Urasawa S (1992)Identification of human and bovine rotavirus serotypes by polymerase chain reaction.Epidemiol Infect 109: 303–312

17. Wakuda M, Pongsuwanna Y, Taniguchi K (2005) Complete nucleotide sequences of twoRNA segments of human picobirnavirus. J Virol Methods 126: 165–169

18. Yang H, Chen S, Ji S (1998) A novel rotavirus causing large scale of adult diarrhea inShi Jiazhuang. Zhonghua Liu Xing Bing Xue Za Zhi (Chin Epidemiol) 19: 336–338 (inChinese; English abstract)

19. Yang H, Makeyev EV, Kang Z, Ji S, Bamford DH, van Dijk AA (2004) Cloning andsequence analysis of dsRNA segments 5, 6 and 7 of a novel non-group A, B, C adultrotavirus that caused an outbreak of gastroenteritis in China. Virus Res 106: 15–26

20. Yang J-H, Kobayashi N, Wang Y-H, Zhou X, Li Y, Zhou D-J, Hu Z-H, Ishino M, AlamMM, Naik TN, Ahmed MU (2004) Phylogenetic analysis of a human group B rotavirusWH-1 detected in China in 2002. J Med Virol 74: 662–667

Author’s address: Nobumichi Kobayashi, Department of Hygiene, Sapporo MedicalUniversity School of Medicine, S-1 W-17 Chu-ku, Sapporo 060-8556, Japan; e-mail:nkobayas@sapmed.ac.jp