Relationship between electropherotypes and VP7/VP4 genotypes of group A rotaviruses detected between 2000 and 2007 in Tunisian children

Results Twelve different electropherotypes were visualized, eight with a long profile (186 cases) and four with a short one (87 cases). Concerning VP7 types, G2 viral strains were found to be predominant and were detected in 91 specimens (32.7%). Strains with G1, G3, G4, G8 and G9 specificities were detected in 62 (22.3%), 82 (29.5%), 13 (4.7%), two (0.7%) and seven cases (2.5%), respectively. The results of VP4 genotyping showed a predominance of P[8] genotype which comprised half of the strains identified (139 cases, 50%). VP4 P[4], P[6] and P[11] were found in 83 (29.9%), 31 (11.1%) and 11 (4.0%) specimens, respectively. A high rate of mixed strains was also found (1.8% mixed electropherotypes, 7.6% G-mixed and 5% P-mixed strains). Electropherotype pattern of rotavirus strains was significantly correlated with VP7 genotype ( p = 0.018) and with VP4 genotype specificities ( p < 0.001). Résumé But de l’étude Les rotavirus sont les principaux agents étiologiques responsables de diarrhées infantiles au niveau mondial. La diversité génomique des rotavirus du groupe A peut être analysée par électrophorèse sur gel de polyacrylamide (PAGE) des 11 fragments d’ARN double brin génomiques, ou mieux par détermination des spécificités géniques VP7 et VP4, permettant ainsi une classification en génotypes G et P, respectivement. Le but du présent travail était de rechercher une association entre électrophorétypes et génotypes G/P des souches de rotavirus. Matériel et méthodes Ont été analysés 278 prélèvements positifs en rotavirus par PAGE puis génotypés par RT-PCR semi-nichées multiplex. L’analyse statistique a été réalisée par tests de corrélation de Pearson. Résultats Douze électrophorétypes distincts ont été mis en évidence, huit avec un profil long (186 cas) et quatre avec un profil court (87 cas). Les souches de génotype VP7 G2 étaient prédominantes (91 cas, soit 32,7 %). Des souches de génotypes G1, G3, G4, G8 et G9 ont également été détectées dans 62 (22,3 %), 82 (29,5 %), 13 (4,7 %), deux (0,7 %) et sept cas (2,5 %), respectivement. Les résultats du typage VP4 ont montré une cocirculation des génotypes P[8], P[4], P[6] et P[11] respectivement retrouvés dans 139 (50 %), 83 (29,9 %), 31 (11,1 %) et 11 (4,0 %) prélèvements. Un nombre important de souches mixtes (1,8 % électrophorétypes mixtes, 7,6 % VP7-mixtes et 5 % VP4-mixtes) a également été détecté. Les profils électrophorétiques des souches de rotavirus étaient corrélés de façon significative avec les génotypes VP7 ( p = 0,018) et les génotypes VP4 ( p < 0,001) de ces souches. Mots clés Rotavirus Électrophorétype Génotype VP7 Génotype VP4 PAGE Keywords Rotavirus Electropherotype VP7 genotype VP4 genotype PAGE 1 Introduction Group A rotavirus is the most common cause of gastroenteritis among infants and young children. It is estimated that, each year, rotavirus gastroenteritis causes the deaths of 500,000 infants and young children worldwide [1] . Rotavirus infection is the single greatest cause of diarrhoea-related deaths among children. Although more than 90% of these deaths occur in the poorest countries, virtually every child will experience at least one case of rotavirus gastroenteritis during the first few years of life. In both developed and developing countries alike, these cases will lead to several million hospitalisations related to dehydration and diarrhoea [2] . The rotaviral genome is formed by 11 double-stranded RNA (dsRNA) segments. Analysis of the electrophoretic mobility of the 11 segments by polyacrylamide gel electrophoresis (PAGE) yields a pattern which is both constant for each rotavirus group and characteristic for a particular rotavirus isolate [3] . Viral particles are nonenveloped but encapsulated by a triple-layered capsid. The major protein in the central layer of the viral capsid is VP6, which determines seven different groups of rotaviruses (A–G). The outer layer of the viral capsid is composed of two structural proteins: VP4, encoded by gene 4, and VP7, encoded by gene 7, 8 or 9 (depending on the strain). A classification system has been established for group A rotaviruses, with the VP7 glycoprotein defining G-types and the protease-sensitive VP4 defining P-types. At least 15 G-types and 23 P-types have thus far been characterized, and various combinations of G- and P-types exist. Nested PCR, based on type-specific primers, is used to determine G and P genotypes. As VP7 and VP4 genes can and do segregate independently, a dual typing system is necessary in order to characterize the strains of rotaviruses cocirculating during different seasons in different locations [3] . The aim of the present study was to report a molecular characterization of rotaviruses strains in Tunisia and to determine the relationship between electropherotype pattern (long, short or mixed) and molecular characteristics of the rotavirus strains (VP7 genotype, VP4 genotype). 2 Material and methods Stool samples were collected from Tunisian children with diarrhoea from January 2000 to December 2007. Group A rotavirus-positive specimens were detected by ELISA method. All specimens were analyzed by PAGE and VP7/VP4 genotyping. Samples were included in the present study only if they were both positive by PAGE and genotyping. 2.1 Group A rotavirus detection All samples were processed for rapid diagnosis by a qualitative enzyme immunoassay (IDEIA Rotavirus, Dako ® , Denmark). This assay is a direct solid-phase sandwich enzyme-linked immunosorbent assay which utilises a polyclonal antibody to detect group-specific antigen present in group A rotaviruses. Break-apart microwells are coated with a rotavirus-specific polyclonal antibody. Faecal suspension or control sample was added to the microwell and incubated simultaneously with a rotavirus-specific polyclonal antibody conjugated to horseradish peroxidase. Rotavirus antigen present in the sample was captured between antibody on the solid phase and the enzyme conjugated antibody. After 60 min incubation at room temperature, the microwells were washed with working strength wash buffer to remove excess specimen and any unbound enzyme-labelled antibody. A chromogen was added to the microwells and incubated for 10 min at room temperature. The presence of specifically bound enzyme-labelled antibody in the microwells resulted in a colour change, which was stopped by the addition of acid. Colour intensity significantly above background levels was indicative of the presence of rotavirus antigen in the specimen or control. 2.2 PAGE The rotavirus-positive faecal specimens were analyzed by PAGE to identify the rotavirus strains in circulation. In brief, faecal suspensions (10 to 20% suspensions of the faecal specimens in phosphate-buffered saline) were mixed with an equal volume of phenol–chloroform to disrupt the viral particles and release the viral double-stranded RNA (dsRNA) genome. After centrifugation at 1200 ×  g for 5 min, the aqueous phase containing the dsRNA was precipitated in absolute ethanol overnight at –20 °C. Following centrifugation, the dsRNA pellet was resuspended in 0.01 M Tris-EDTA buffer for electrophoresis in 10% polyacrylamide slab gels at 100 V for 16 to 18 h at room temperature. The gels were stained using silver nitrate [4] . 2.3 VP7 and VP4 genotyping Initially, the viral RNA was extracted and purified from 100 μl of 10% faecal suspensions in phosphate-buffered saline using the TRIzol method. The extracted RNA was used for a reverse transcription-polymerase chain reaction (RT-PCR) by use of specific VP7 and VP4 consensus primer pairs ( Tables 1 and 2 ) [5,6] . The purified RNA was reverse transcribed with AMV reverse transcriptase at 43 °C for 25 min, in the presence of primers to the terminal sequences of the VP7 gene (mixture of the primer pair Beg9/End9). The cDNA was then amplified with the same primers during 30 cycles, each consisting of heating up to 95 °C for 1 min to denature the cDNA, followed by cooling to 48 °C for 2 min to anneal the primers and finally heating to 72 °C for 1 min to extend the strands. The final extension was lengthened to 3 min to promote full-length amplicons. A nested, multiplex PCR reaction was then performed using different primers specific for genotype-specific regions of the VP7 gene ( Table 1 ) [6] . The type-specific primers could detect G1, G2, G3, G4, G8 and G9 VP7 specificities. The size of the resultant amplicon allows determination of the VP7 genotype ( Table 1 ). The VP4 genotype was determined in the same way using an RT-PCR system described by Gentsch et al. [5] . Specific primers (consensus primers Con2 and Con3) were used to amplify the VP8* gene which was then differentiated into VP4 genotypes by a cocktail of primers designed for the common human VP4 genotypes (P[6], P[4] and P[8]) and for the less common (P[9], P[10] and P[11]) genotypes ( Table 2 ). 2.4 Statistical analysis Pearson's correlation tests were used to determine the relationship between electropherotype pattern (long, short, mixed) and molecular characteristics of the rotavirus strains (VP7 genotype, VP4 genotype). Significance level of P < 0.05 was used for all analyses. 3 Results In total, 278 rotavirus-positive samples were also positive by both PAGE and genotyping. Concerning PAGE results, all 278 faecal specimens presented an electropherotype with a characteristic 4-2-3-2 migration pattern typical of group A rotaviruses. Group A rotavirus strains generally possess either the short (S) or long (L) RNA electropherotype. In the present study, 12 different electropherotypes were visualized, eight with a long profile (L1 to L8) and four with a short one (S1 to S4). Rotavirus strains presented long profile in 186 cases (66.9%), short profile in 87 cases (31.3%), and mixed profiles (M) in five cases (1.8%) ( Figs. 1 and 2 ). Moreover, strains were genotyped by a nested RT-PCR assay for the VP7 and VP4 genotypes. Concerning VP7 types, G2 viral strains were found to be predominant and were detected in 91 specimens (32.7%). Strains with G1, G3, G4, G8 and G9 specificities were detected in 62 cases (22.3%), 82 cases (29.5%), 13 cases (4.7%), two cases (0.7%) and seven cases (2.5%), respectively. A high rate of mixed strains (21%) was also found. The results of VP4 genotyping showed a predominance of P[8] genotype which comprised half of the strains identified (139 cases, 50%). The second most commonly seen VP4 genotype was P[4], detected in 83 cases (29.9%). VP4 P[6] and P[11] were found in 31 (11.1%) and 11 (4.0%) specimens, respectively. Fourteen mixed VP4 types (5%) were detected. Seventeen different G–P combinations of rotavirus strains cocirculated in Tunisia between 2000 and 2007, reflecting the high level of genomic diversity ( Fig. 3 ). The vast majority of strains were typed as G2P[4] (71 cases, 25.5%), G3P[8] (71 cases, 25.5%) and G1P[8] (34 cases, 12.2%). Electropherotype pattern (long or short) of rotavirus strains was significantly correlated with VP7 genotype ( P = 0.018) and with VP4 genotype specificity ( P < 0.001). No significant correlation was found between mixed genotypes and mixed electropherotypes. 4 Discussion Since, in general, each rotavirus strain displays a dsRNA migration pattern (electropherotype) on polyacrylamide gels distinct from that of other strains, analysis of such genomic polymorphism as determined by PAGE has been routinely employed to assign the parental gene origin of a reassortant. Although the standard PAGE analysis of rotavirus dsRNA is simple, easy and cost-effective to perform, various factors and running conditions of PAGE have been reported to affect the electropherotype of each strain which include the concentration of acrylamide/bisacrylamide, type of running buffer, the magnitude of running voltage and purity of dsRNA analyzed [7] . Concerning the results of rotavirus VP7 genotyping it was unexpected to find strains with G8 (two cases) and G9 (seven cases) specificity, as the previous Tunisian study conducted between 1995 and 1998 did not report any unconventional genotype [8] . Although G9 strains have recently emerged as the fifth most common VP7 genotype, human G8 strains are more commonly found in Africa than elsewhere [9–11] . The emergence in developing countries of genotypes such as G8, classically isolated from cattle, reflects the exposure to an environment contaminated with human and animal faeces. In the present study, P[6] specificity accounted for 11% of P-types. Thus, as previously reported, the presence of the P[6] genotype seems to have rapidly expanded into all regions investigated in Africa. The proportion of mixed infections in the present study (G-type mixed infections: 21%; P-type mixed infections: 5%) was high as reported previously in developing countries such as Brazil and Bangladesh [12,13] . Close contact with animals in the domestic environment is a factor that is likely to promote mixed infections [11,14,15] . The high rate of mixed infections is considered a prerequisite for the occurrence of rotavirus reassortment events, where two strains of rotavirus infect the same patient with the possibility of genetic reassortment of the segmented viral genome [12] . Such events may lead to the appearance of new strains and new variants. The observations of increasing rotavirus strain diversity and high rate of mixed infections in Tunisia define a profile characteristic of developing countries, as previously described in studies conducted in Brazil or Bangladesh [12,13] . Electropherotype pattern of rotavirus strains was significantly correlated with VP7 genotype ( P = 0.018) and with VP4 genotype specificities ( P < 0.001). During the period of study, strains with G2 VP7 specificity presented in 91% of cases short electropherotypes, whereas G1, G3, G4, G8 and G9 strains showed long profiles in 98% of cases after PAGE. Moreover, strains with P[4] VP4 specificity presented in 83% of cases short electropherotypes, whereas P[8] strains showed long profiles in 96% of cases. Similar results especially concerning G1 to G4 genotypes, have already been described elsewhere [8,16–21] . The diversity of strains circulating in different regions may be important for vaccine development and clinical evaluation. Indeed, in 2006, two rotavirus vaccines were licensed: an oral, live pentavalent human-bovine (WC3) reassortant rotavirus vaccine (RotaTeq, Merck) containing antigens of rotavirus G1, G2, G3, G4 and P[8] genotypes, and a live attenuated human rotavirus vaccine containing the RIX4414 strain of G1P[8] specificity (Rotarix, GlaxoSmithKline Biologicals) [2,22] . Although G1-G4 genotypes are the most common circulating genotypes of group A rotaviruses worldwide, G8 and G9 genotypes, which were not included in the vaccines, have more regularly appeared in the last few years. Because of the theoretical possibility of uncommon genotypes and the lack of data on the efficacy of new vaccines against strains such as G8 and G9, initiation of strain surveillance systems in countries prior to and following vaccine introduction is crucial. Such studies will allow health officials to monitor the impact of vaccination programs on genotype prevalence and the emergence of uncommon strains and to provide information on the ability of vaccination to protect against diarrhoea caused by genotypes other than G1 to G4. 5 Conclusion The present study has extended the database for rotavirus strain characterization and diversity across the continent, especially in North Africa where only Libyan [23] and Tunisian [8,24–26] data are available to date. The incidence and distribution of human group A rotavirus genotypes varies between geographical areas during a rotavirus season and from one season to the next. Because serotype-specific immunity may be more protective than heterologous immunity, continued vigilance for changing trends in identified strains, as well as for new rotavirus strains, is needed before and after introduction of any rotavirus vaccine in any country. Acknowledgements This present study was supported by grant V27/181/167 from the World Health Organization (EMRO). The authors declare no conflict of interest. We thank the staff of the laboratories of Microbiology of Sahloul for their technical help and the nursing staff of the Paediatric Units of all hospitals included in Tunisian Rotavirus Network for assisting in sample collection. References [1] U.D. Parashar C.J. Gibson J.S. Bresee R.I. Glass Rotavirus and severe childhood diarrhea Emerg Infect Dis 12 2006 304 306 [2] J.S. Bresee E. Hummelman E.A.S. Nelson R.I. Glass Rotavirus in Asia: the value of surveillance for informing decisions about the introduction of new vaccines J Infect Dis 192 2005 S1 5 [3] M.K. Estes A.Z. Kapikian D.M. Knipe P.M. Howley Fields virology fifth ed. 2007 Lippincott-Raven Press Philadelphia 1917 1974 [4] A.J. Herring N.F. Inglis C.K. Ojeh D.R. Snodgrass J.D. Menzies Rapid diagnosis of rotavirus infection by direct detection of viral nucleic acid in silver-stained polyacrylamide gels J Clin Microbiol 16 1982 473 477 [5] J.R. Gentsch R.I. Glass P. Woods V. Gouvea M. Gorziglia J. Flores Identification of group A rotavirus gene 4 types by polymerase chain reaction J Clin Microbiol 30 1992 1365 1373 [6] V. Gouvea R.I. Glass P. Woods K. Taniguchi H.F. Clark B. Forrester Polymerase chain reaction amplification for typing of rotavirus nucleic acid from stool specimens J Clin Microbiol 28 1990 276 282 [7] J. Ross E.N. Ostlund D. Cao M. Tatsumi Y. Hoshino Acrylamide concentration affects the relative position of VP7 gene of serotype G2 strains as determined by polyacrylamide gel electrophoresis J Clin Virol 42 2008 374 380 [8] A. Trabelsi I. Peenze C. Pager M. Jeddi D. Steele Distribution of rotavirus VP7 serotypes and VP4 genotypes circulating in Sousse, Tunisia, from 1995 to 1999 : emergence of natural human reassortants J Clin Microbiol 38 2000 3415 3419 [9] R.H. Asmah J. Green G.E. Armah C.I. Gallimore J.J. Gray M. Iturriza-Gómara Rotavirus G and P genotypes in rural Ghana J Clin Microbiol 39 2001 1981 1984 [10] A.A. Cunliffe J.S. Gondwe R.L. Broadhead M.E. Molyneux P.A. Woods J.S. Bresee Rotavirus G and P types in children with acute diarrhea in Blantyre, Malawi, from 1997 to 1998 J Med Virol 57 1999 308 312 [11] A.D. Steele S.P. Parker I. Peenze C.T. Pager M.B. Taylor W.D. Cubitt Comparative studies of human rotavirus G8 strains recovered in South Africa and the United Kingdom J Gen Virol 80 1999 3029 3034 [12] M.C. Timenetsky N. Santos V. Gouvea Survey of rotavirus G and P types associated with human gastroenteritis in Sao Paulo, Brazil, from 1986 to 1992 J Clin Microbiol 32 1994 2622 2624 [13] L.E. Unicomb G. Podder J.R. Gentsch P.A. Woods K.Z. Hasan A.S. Faruque Evidence of high-frequency genomic reassortment of group A rotavirus strains in Bangladesh: Emergence of type G9 in 1995 J Clin Microbiol 37 1999 1885 1891 [14] M. Riepenhoff-Talty P.C. Lee P.J. Carmody H.J. Barrett P.L. Ogra Age-dependant rotavirus enterocyte interactions Proc Soc Exp Biol Med 170 1982 146 154 [15] I. Silberstein L.M. Shulman E. Mendelson I. Shif Distribution of both rotavirus VP4 genotypes and VP7 serotypes among hospitalized and nonhospitalized Israeli children J Clin Microbiol 33 1995 1421 1422 [16] J. O’Mahony B. Foley S. Morgan VP7 and VP4 genotyping of rotavirus samples recovered from infected children in Ireland over a 3-year period J Clin Microbiol 37 1999 1699 1703 [17] G.E. Armah C.T. Pager R.H. Asmah F.R. Anto A.R. Oduro F. Binka Prevalence of unusual human rotavirus strains in Ghanaian children J Med Virol 63 2001 67 71 [18] V. Jain B.K. Das M.K. Bhan R.I. Glass J.R. Gentsch Great diversity of group A rotavirus strains and high prevalence of mixed rotavirus infections in India J Clin Microbiol 39 2001 3524 3529 [19] M. Zeng Q.R. Zhu Y. Zhang G.H. Li D.M. Chen Y.X. Ding Molecular epidemiologic survey of rotaviruses from infants and children with diarrhea in Shanghai Zhonghua Er Ke Za Zhi 42 2004 10 15 [20] P. Saravanan S. Ananthan M. Ananthasubramanian Rotavirus infection among infants and young children in Chennai, South India Ind J Med Microbiol 22 2004 212 221 [21] M.M. Patel L.H. De Oliveira A.M. Bispo J. Gentsch U.D. Parashar Rotavirus P[4]G2 in a vaccinated population, Brazil Emerg Infect Dis 14 2008 863 864 [22] K.B. Phua S.H. Quak B.W. Lee S.C. Emmanuel P. Goh H.H. Han Evaluation of RIX4414, a live, attenuated rotavirus vaccine, in a randomized, double-blind, placebo-controlled phase 2 trial involving 2464 Singaporean infants J Infect Dis 192 2005 S6 S16 [23] N.A. Cunliffe W. Dove J.E. Bunn M. Ben Ramadam J.W. Nyangao R.L. Riveron Expanding global distribution of rotavirus serotype G9: detection in Libya, Kenya, and Cuba Emerg Infect Dis 7 2001 890 892 [24] K. Sdiri-Loulizi H. Gharbi-Khelifi A. De Rougemont Acute infantile gastroenteritis associated with human enteric viruses in Tunisia J Clin Microbiol 46 2008 1349 1355 [25] A. Chouikha I. Fodha S. Noomen L. Bouzid M. Mastouri I. Peenze Group A rotavirus strains circulating in the eastern center of Tunisia during a 10-year period (1995–2004) J Med Virol 79 2007 1002 1008 [26] I. Fodha A. Chouikha I. Peenze M. De Beer J. Dewar A. Geyer Identification of viral agents causing diarrhea among children in the eastern center of Tunisia J Med Virol 78 2006 1198 1203