Scw

Ra

b

a

A

KGVVSEV

1

nIdtosgt

h0

Vaccine 32S (2014) A75–A83

Contents lists available at ScienceDirect

Vaccine

j our na l ho me page: www.elsev ier .com/ locate /vacc ine

equence analysis of VP7 and VP4 genes of G1P[8] rotavirusesirculating among diarrhoeic children in Pune, India: A comparisonith Rotarix and RotaTeq vaccine strains

uta Kulkarnia, Ritu Aroraa, Rashmi Arorab, Shobha D. Chitambara,∗

Enteric Viruses Group, National Institute of Virology, 20-A, Dr. Ambedkar Road, Pune-411001, IndiaDivision of Epidemiology and Communicable Diseases, Indian Council of Medical Research, Ansari Nagar, New Delhi-110029, India

r t i c l e i n f o

rticle history:

eywords:1P[8] rotavirusP7P4ubgenotypic lineagepitopeaccine

a b s t r a c t

Background: The G1P[8] rotaviruses are a common cause of rotavirus diarrhoea among children in India.Two rotavirus vaccines licensed in India, Rotarix and RotaTeq, contain strains with G1 and P[8] genotypes.A comparative analysis of these genotypes in the live rotavirus vaccines with circulating rotavirus strainsis essential for assessment of rotavirus diversity.Methods: G1P[8] strains detected during rotavirus surveillance among diarrhoeic children hospitalizedin Pune in 1992–1993 and 2006–2008, were included in the study. Amplification, sequencing and phylo-genetic analysis of the VP7 and VP4 genes were carried out for identification of the G1 and P[8] lineages,respectively. Antigenic epitopes of VP7 and VP4 encoded proteins were compared to determine thedifferences between the G1P[8] strains from Pune and the vaccine strains.Results: G1-Lineage 1, P[8]-Lineage 3 strains were predominant in Pune during 1992–1993 and2006–2008. Strains of G1-Lineage 2, P[8]-Lineage 3 and G1-Lineage 1, P[8]-Lineage 4 were detectedat low levels during 2006–2008. The G1-Lineage 1, P[8]-Lineage 3 strains showed up to eight aminoacid changes, each in the VP7 and VP4 epitopes, with respect to the Rotarix vaccine strain (G1-Lineage2, P[8]-Lineage 1) and the G1 (Lineage-3) and P[8] (Lineage 2) components of the RotaTeq vaccine. TheG1-Lineage 2 strains were closer to both vaccine strains with no or only two amino acid substitutions inthe VP7 epitopes. The divergent P[8]-Lineage 4 (OP354-like) strains showed fourteen and fifteen amino

acid differences, with Rotarix and RotaTeq vaccine strains, respectively, in the VP4 epitopes.Conclusion: The differences between the G1P[8] strains in Pune and the G1 and P[8] components ofthe vaccine strains need to be described for appropriate evaluation of vaccine shedding. Continuousmonitoring of the G1P[8] subgenotypic lineages would be necessary to study any long term impact ofvaccine use on G1P[8] strain evolution.

© 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

. Introduction

Rotavirus is the most important cause of severe diarrhoeal ill-ess in infants and young children, worldwide [1]. Annually, in

ndia, rotaviral diarrhoea causes an estimated 122,000–153,000eaths, 457,000–884,000 hospitalizations and 2,000,000 outpa-ient visits in children less than five years of age [2]. Rotaviruses,f the family Reoviridae, are triple-layered particles (TLPs) con-

isting of the outer capsid, inner capsid and core. The rotavirusenome consists of 11 dsRNA segments which code for the six struc-ural (VP1-VP4, VP6, VP7) and five non-structural (NSP1-NSP5)

∗ Corresponding author. Tel.: +91 020 26006251; fax: +91 020 26122669.E-mail address: drshobha.niv@gmail.com (S.D. Chitambar).

ttp://dx.doi.org/10.1016/j.vaccine.2014.03.080264-410X/© 2014 Published by Elsevier Ltd. This is an open access article under the CC B

(http://creativecommons.org/licenses/by-nc-nd/3.0/).

proteins. The outer capsid proteins, VP7 and VP4, serve as viralattachment proteins and neutralization antigens [3]. VP4 is acti-vated by proteolytic cleavage into two fragments—VP8* and VP5*.VP8* forms a globular attachment domain at the tip of the VP5*stalk [4]. A binary system classifies group A rotaviruses into 27G and 37 P types [5,6], a classification initially based on neu-tralization specificities of VP7 (Glycoprotein) and VP4 (Proteasesensitive protein). Globally, G1P[8], G2P[4], G3P[8], G4P[8] andG9P[8] genotype combinations of rotavirus strains are the mostcommon cause of human infections [7]. Of these, G1P[8] strainsare most predominant (37.7%) [7]. These strains exhibit diver-

sity in the form of 11 G1 and 4 P[8] subgenotypic lineages [8,9].According to a multi-centre hospital-based study carried out inIndia from 2005 to 2009, G1P[8] strains were highly prevalent[10].

Y-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

A ccine 3

lv8bgWtpd

wIP(wasd2n

2

2

rcs(s

2

uftBgaP3caa

aNrt

2

aDFJJ1sH

76 R. Kulkarni et al. / Va

Two rotavirus vaccines, Rotarix and RotaTeq, are currentlyicensed in many countries including India. Rotarix is a monovalentaccine containing the attenuated human G1P[8] rotavirus strain9-12. RotaTeq is a pentavalent vaccine containing five human-ovine reassortant rotavirus strains, each representing one humanenotype—WI79-9 (G1), SC2-9 (G2), WI78-9 (G3), BrB-9 (G4) andI79-4 (P[8]). Studies from different countries have revealed that

he G1 and P[8] subgenotypic lineages included in these vaccines,revalent at the time of vaccine development (1980s), are not pre-ominant today [8,9,11–23].

Earlier, we have reported identification of different lineagesithin VP7 gene of G1 rotaviruses circulating in Pune, western

ndia [24]. The study did not include analysis of the corresponding[8] lineages of VP4 genes and the rotavirus vaccine strains, 89-12Rotarix G1P[8]), WI79-9 (RotaTeq G1) and WI79-4 (RotaTeq P[8])ere not compared due to the unavailability of their sequence data

t the time. The aim of the present study was to assess the diver-ity of G1P[8] rotavirus strains circulating among children withiarrhoea in Pune during the two time periods, 1992–1993 and006–2008, and compare sequences with the G1 and P[8] compo-ents of vaccines.

. Materials and methods

.1. Specimen selection

A surveillance program for rotavirus disease and strains was car-ied out in children (<5 years), hospitalized for diarrhoea in Puneity during 1990s and 2000s [10,25] (Table 1). The G1P[8] rotavirustrains identified during the years 1992 (n = 8), 1993 (n = 11), 2006n = 21), 2007 (n = 29) and 2008 (n = 13) were selected in the presenttudy for further characterization.

.2. Amplification and sequencing of the VP7 and VP4 genes

Viral RNA was isolated from 30% (w/v for solid and v/v for liq-id) stool suspensions of the selected G1P[8] rotavirus positiveaecal specimens using TRIzol (Invitrogen, U.S.A). Amplification ofhe complete VP7 gene (1062 bp) was carried out using the primerseg9 and End9 [26] as described previously [24]. The partial VP4ene (VP8* region: 10 to 729 bp) was amplified with primers con2nd con3 [27] using One-step RT-PCR kit (Qiagen, Germany). TheCR conditions involved an initial reverse transcription step of0 min at 45 ◦C, followed by PCR activation at 95 ◦C for 15 min, 40ycles of amplification (1 min at 94 ◦C, 1 min at 50 ◦C and 2.5 mint 70 ◦C) with a final extension of 7 min at 70 ◦C. The VP7 and VP8*mplicons were sequenced as reported previously [24].

Sequencing of the complete VP4 genes was carried outs described earlier [28] for six G1P[8] strains (NIV-0613158,IV-06361, NIV-061060, NIV-0715880, NIV-07523, NIV-083375)

epresenting each of the two P[8] lineages (P[8]-3 and P[8]-4) iden-ified in Pune on the basis of VP8* sequences.

.3. Nucleotide sequence accession numbers

The VP7 sequences were submitted to GenBank under theccession numbers DQ886943-46, DQ886953-56, DQ886958,Q886959, DQ886962, DQ886964-68, DQ886972, DQ875602,J948829-55, JN192054-55, JN192060-61, JN192063-64,N192068-69, JN192071-75, JN192079, JN192082-83, JN192086,

N192089, JN192093-96, JN192098-99, JN192100-01, JN192112-3, JN192115-16, JN192119-26 and JN192128-31. The VP4equences were submitted under the accession numbersQ881499 to HQ881575, EU984107 and HM467806-08.

2S (2014) A75–A83

2.4. Phylogenetic analysis

The VP7 and VP4 sequences of the G1P[8] reference strains [8,9]representing each of the 11 G1 and 4 P[8] subgenotypic lineagesand the sequences of the Rotarix and RotaTeq vaccine strains wereretrieved from GenBank. The sequences available in GenBank forG1P[8] strains from other cities [Kolkata (n = 8), Delhi (n = 3) andManipur (n = 4)] included in the study were classified into lineagesduring comparative analysis.

Multiple sequence alignments were conducted using theClustalW implementation in MEGA 5.05 [29]. Phylogenetic treeswere constructed using the neighbour joining algorithm andKimura 2-parameter model in MEGA 5.05. The statistical signif-icance of the genetic relationships was estimated by bootstrapresampling analysis (1000 replications). Nucleotide and amino aciddistances were calculated using Kimura 2-parameter model andP-distance model, respectively.

3. Results

3.1. Distribution of the G1P[8] rotavirus strains circulating inPune into subgenotypic lineages

Phylogenetic analysis of the VP7 (Fig. 1(A)) and VP4 genes(Fig. 1(B)) showed clustering of the G1P[8] strains from Pune intoG1-Lineage 1 or 2 and P[8]-Lineage 3 or 4 (Fig. 2). All the strains fromthe years 1992 (8/8, 100%) and 1993 (11/11, 100%) were placedinto G1-Lineage 1, P[8]-Lineage 3. In the year 2006, the G1P[8]strains from Pune were distributed into G1-Lineage 1, P[8]-Lineage3 (20/21, 95.2%) and G1-Lineage 2, P[8]-Lineage 3 (1/21, 4.8%). In2007, while the G1-Lineage 1, P[8]-Lineage 3 strains continued topredominate (23/29, 79.3%), the prevalence of G1-Lineage 2, P[8]-Lineage 3 strains increased (5/29, 17.2%). In addition, one strain ofG1-Lineage 1, P[8]-Lineage 4 (1/29, 3.5%) was detected. In 2008, theG1P[8] strains from Pune were distributed into G1-Lineage 1, P[8]-Lineage 3 (12/13, 92.3%) and G1-Lineage 1, P[8]-Lineage 4 (1/13,7.7%).

Phylogenetic analysis of the G1P[8] strains from other cities inIndia (Fig. 1(A) and (B)) revealed circulation of the same subgeno-typic lineages as in Pune. All G1P[8] strains from Kolkata (8/8,2008–2009) and Delhi (3/3, 2000s) clustered into G1-Lineage 1,P[8]-Lineage 3. The G1P[8] strains from Manipur (2006–2007) weredistributed into G1-Lineage 1, P[8]-Lineage 3 (2/4) and G1-Lineage1, P[8]-Lineage 4 (2/4).

The Rotarix vaccine strain, 89-12, clustered into G1-Lineage 2,P[8]-Lineage 1. The WI79-9 (G1) strain of RotaTeq vaccine wasplaced in G1-Lineage 3 while the WI79-4 (P[8]) strain was classifiedin P[8]-Lineage 2 (Fig. 1(A) and (B)).

3.2. Comparative sequence analysis of VP7 and VP4 genes ofrotavirus G1P[8] strains from Pune and rotavirus vaccines

The G1-Lineage 1 strains showed 92.8–95.2% nucleotide and92.9–95.4% amino acid identity with the Lineage 2 of G1 Rotarixvaccine strain and 89.9–92.0% nucleotide and 92.0–94.4% aminoacid identity with the Lineage 3 of the G1 strain in RotaTeq vaccine.The G1-Lineage 2 strains were closer to the Rotarix VP7 of the samelineage (97.3–97.5% nucleotide and 97.2–97.5% amino acid iden-tity) than to the RotaTeq VP7 of Lineage 3 (92.1–92.2% nucleotideand 94.4–94.8% amino acid identity).

The VP8* of the P[8]-Lineage 3 strains were more similar to the

RotaTeq P[8] (92.3–93.9% nucleotide and 92.9–95.8% amino acididentity) than to Rotarix VP8* (89.5–91.4% nucleotide and 90.8-93.3% amino acid identity). The divergent P[8]-Lineage 4 strainsshowed lower identities with both the vaccine strains (Table 2).

R. Kulkarni et al. / Vaccine 32S (2014) A75–A83 A77

Fig. 1. Phylogenetic dendrogram of the nucleotide sequences representing G1P[8] rotavirus strains circulating in Pune, India.(A) VP7 (coding region: 49–1026 bp) and (B) VP4 (VP8* region: 10–729 bp)The strains from Pune are indicated by a solid triangle while the vaccine strains are indicated by a solid circle. Bootstrap values ≥70% are shown. The GenBank accessionnumbers for the VP7 gene sequences of the reference G1 rotavirus strains are AY631049 (Dhaka 8-02), DQ512979 (Thai-804), U26370 (Cos-70), K02033 (Wa), U26378(Kor-64), DQ377573 (PA5/90), AB018697 (AU19), U26366 (Ban-59), U26373 (Egypt-7), DQ377574 (PA32/90), AF426162 (SW20/21), L24164 (C60). The GenBank accessionnumbers for the VP4 gene sequences of the reference P[8] rotavirus strains are L34161 (Wa), U30716 (F45), AJ605320 (Hun9), AJ302148 (OP354). The GenBank accessionnumbers for the sequences of the vaccine strains are JN849114 (Rotarix VP7), JN849113 (Rotarix VP4), GU565057 (RotaTeq G1 VP7), GU565044 (RotaTeq P[8] VP4). Rep-resentative strains from Kolkata, Delhi and Manipur have also been included in the dendrogram. The VP7 gene sequences for the strains from Kolkata are available underthe accession numbers AB561878 (GRAVP736), AB553330 (GRAVP732), AB553329 (GRAVP731), AB553328 (GRAVP730), AB551663 (GRAVP728), AB550709 (GRAVP727),AB550233 (GRAVP720), AB547709 (GRAVP718) while the VP4 sequences are available under the accession numbers AB561881 (GRAVP436), AB553327 (GRAVP432),AB553326 (GRAVP431), AB553325 (GRAVP430), AB551661 (GRAVP428), AB550708 (GRAVP427), AB544058 (GRAVP420), AB544056 (GRAVP418). The accession numbers forthe VP7/VP4 sequences of the strains from Delhi are FJ827595/FJ827605 (Dan279), FJ827593/FJ861662 (Dan114), FJ827592/FJ827602 (Dan103) while those for the strainsfrom Manipur are GQ229043/GQ240617 (mani-375/07), GQ229042/GQ240616 (mani-365/07), GQ229041/GQ240614 (mani-140/06), GQ229040/GQ240612 (mani-63/06).

A78 R. Kulkarni et al. / Vaccine 32S (2014) A75–A83

Fig. 1. (Continued ).

R. Kulkarni et al. / Vaccine 32S (2014) A75–A83 A79

Table 1Year wise details of the rotavirus surveillance carried out among children (<5 years) hospitalized for diarrhoea in Pune city during 1992–1993 and 2006–2008.

Year No. of faecal specimenscollected

No. (%) of rotaviruspositive specimens

No. of specimens subjectedto genotypinga

No. (%) of specimens positivefor G1P[8] rotavirus

1992 144 32 (22.2) 16 8 (50.0)b

1993 335 104 (31.0) 38 11 (28.9)b

2006 301 96 (31.9) 96 27 (28.1)2007 311 137 (44.1) 137 64 (46.7)2008 231 65 (28.1) 65 13 (20.0)

a Genotyping has been carried out by multiplex PCR.b Number of specimens subjected to genotyping was considered as denominator for pe

Fl

Br

3G

7[da(stwm27Itoa

2008, represented a highly divergent subgenotypic lineage and

TP

ig. 2. Distribution of the G1P[8] rotaviruses circulating in Pune into subgenotypicineages.

oth P[8] lineages showed higher amino acid divergence in VP8*egion than in VP5* region (Table 2).

.3. Comparison of the VP7 and VP4 antigenic epitopes of the1P[8] rotavirus strains from Pune and rotavirus vaccines

The rotavirus VP7 protein consists of two antigenic epitopes:-1 (7-1a and 7-1b) and 7-2 encompassing 29 amino acid residues30]. The G1-Lineage 1 strains from Pune showed 3–6 amino acidifferences with the G1-Lineage 2 strain of Rotarix and 5–8 aminocid differences with the G1-Lineage 3 strain of RotaTeq vaccineTable 3). The majority (92.1–100%) of the G1-Lineage 1 strainshowed three and one amino acid differences, respectively, in epi-opes 7-1a (N94S, S123N, K291R) and 7-2 (M217T/I) in comparisonith both vaccine strains. All amino acid differences were com-on to the G1-Lineage 1 strains of both periods (1992–1993 and

006–2008) with the exception of the substitution L148F in epitope-2 that was restricted to seven strains from the years 2006–2008.

n addition, all G1-Lineage 1 strains had the substitutions D97E (epi-ope 7-1a) and S147N (epitope 7-2) when compared to the G1 strainf RotaTeq vaccine. Strain specific differences were noted at aminocid positions 125, 129 (epitope 7-1a), 212, 213 (epitope 7-1b) and

able 2ercent nucleotide and amino acid identities between the VP4 genes of G1P[8] rotavirus

Rotavirus straina/P[8]lineage

Percent nucleotide/amino acid identity with VP4 g

Rotarix

VP8* VP5* Comp

NIV-06361/P[8]-3 90.9/92.1 88.6/95.1 89.4/9NIV-061060/P[8]-3 91.0/92.9 89.2/95.3 89.7/9NIV-0613158/P[8]-3 90.9/92.1 90.3/95.6 90.5/9NIV-0715880/P[8]-3 90.2/91.2 89.2/95.6 89.5/9NIV-07523/P[8]-4 87.0/87.0 88.4/94.7 88.0/9NIV-083375/P[8]-4 86.8/87.0 88.6/94.9 88.1/9

a Six representative strains for which complete VP4 sequencing was done, are shown i

rcentage calculation.

221 (epitope 7-2) in a few (1–3) of the G1-Lineage 1 strains oncomparison with both vaccine strains.

The G1-Lineage 2 strains detected in Pune during 2006 and2007 showed 100% sequence identity to the Rotarix vaccine strain(G1-Lineage 2) in the VP7 epitopes. With respect to the RotaTeq vac-cine strain, the G1-Lineage 2 strains showed only two amino aciddifferences–D97E (epitope 7-1a) and S147N (epitope 7-2) (Table 3).Overall, the epitopes 7-1a and 7-2 were more prone to variationsthan epitope 7-1b among all G1 strains.

The VP4 protein of rotavirus consists of nine antigenicepitopes—four (8-1 to 8-4) in VP8* and five (5-1 to 5-5) in VP5*,which together include 37 amino acids [31,32]. The P[8]-Lineage3 strains from Pune showed 5-8 amino acid differences with theP[8]-Lineage 1 strain of Rotarix and 2-5 amino acid differences withthe P[8]-Lineage 2 strain of RotaTeq vaccine in the VP8* antigenicepitopes (Table 4A). These comprised S146G, S190N and N196Gin epitope 8-1 and N113D, S125N, S131R, N135D in epitope 8-3 ascompared with Rotarix vaccine strain. With regard to the P[8] strainof RotaTeq vaccine, the P[8]-Lineage 3 strains of this study showedthree and one amino acid differences, respectively, in epitopes 8-1(S146G, N190S, D196G) and 8-3 (N113D). Strain specific differenceswere noted at the amino acid positions 192, 193, 195 (epitope 8-1),and 114, 115,116 (epitope 8-3) in a few (1-5) of the P[8]-Lineage3 strains on comparison with both vaccine strains. Epitopes 8-2and 8-4 were completely conserved. The amino acid substitutionsin VP8* region were common to all P[8]-Lineage 3 strains at bothtime points (1992–1993 and 2006–2008).

To compare VP5* epitopes of the P[8]-Lineage 3 strains, we usedcomplete VP4 sequences available for four P[8]-Lineage 3 strains,NIV-0613158, NIV-06361, NIV-061060, NIV-0715880 (Table 4B).These strains showed 1-2 amino acid differences (Y386D in all fourstrains, S388N in one strain, NIV-061060) with Rotarix and 2-3amino acid differences (R384S, H386D in all four strains, S388Nin NIV-061060) with RotaTeq in epitope 5-1. Epitopes 5-2 to 5-5showed no variations (Table 4B).

The P[8]-Lineage 4 strains, detected in Pune during 2007 and

showed fourteen amino acid differences (twelve in VP8* and twoin VP5*) with the Rotarix vaccine strain and fifteen amino acid dif-ferences (twelve in VP8* and three in VP5*) with the P[8] strain of

strains from Pune and rotavirus vaccines.

ene from vaccine strains

RotaTeq

lete VP4 VP8* VP5* Complete VP4

4.2 93.9/95.8 91.5/95.8 92.2/95.94.4 93.1/95.0 91.4/95.6 91.8/95.44.6 92.8/94.1 93.1/96.0 93.0/95.54.2 92.6/94.1 92.0/96.4 92.1/95.62.5 87.8/88.7 87.5/93.9 87.7/92.52.6 87.6/88.7 87.3/93.7 87.4/92.4

n the table.

A80 R. Kulkarni et al. / Vaccine 32S (2014) A75–A83

Table 3Alignment of the amino acid residues in VP7 antigenic epitopes of the G1P[8] rotavirus strains circulating in Pune and rotavirus vaccines.

Table includes representative G1P[8] strains from children in Pune. The G1P[8] strains from adolescents and adults in Pune also showed the same lineage-specific aminoacid substitutions (data not shown). Representative G1P[8] strains from Kolkata (GRAVP732, GRAVP731, GRAVP727, GRAVP718), Delhi (Dan279, Dan103) and Manipur( rencea iffer fra ] as ne

RL8c

P[wsf

GMaseP(4a

4

re≥al

mani-375/07, mani-140/06) are also included for comparison. The amino acid diffecid residues that differ from Rotarix are shown in yellow colour while those that dnd RotaTeq are in green colour. The amino acid positions described previously [30

otaTeq vaccine (Table 4A and B). The variability between the P[8]-ineage 4 and the vaccine strains was restricted to the epitopes 8-1,-2, 8-3 and 5-1 while the epitopes 8-4, 5-2 to 5-5 were completelyonserved.

Comparison of the VP7 and VP4 epitopes of the G1-Lineage1,[8]-Lineage 3 strains reported from adolescents and adults in Pune33,34], showed the same amino acid variations (data not shown)ith respect to the vaccine strains as were noted in the present

tudy (Tables 3 and 4) for the G1-Lineage 1, P[8]-Lineage 3 strainsrom children in Pune.

Classification (Fig. 1(A) and (B)) and epitope analysis of the1P[8] strains from Kolkata (2008–2009), Delhi (2000s) andanipur (2006–2007) showed the same lineage-specific amino

cid substitutions in the VP7 and VP4 epitopes, as noted for thetrains from Pune (Tables 3 and 4). An earlier study of P[8] lin-ages of G1P[8] strains from Kolkata has described the circulation of[8]-Lineages 3 and 4 during 2004–2005 [35]. These P[8]-Lineage 3ISO115, ISO114, ISO113, 27B3) and P[8]-Lineage 4 (ISO117, ISO116,7B3) strains also showed the same lineage-specific sequence vari-tions in the VP8* epitopes (Table 4A).

. Discussion

The World Health Organization has recommended inclusion ofotavirus vaccines in national immunization programs worldwide,

specially in countries like India where diarrhoea is responsible for10% mortality in children [36]. Two vaccines, Rotarix and RotaTeqre currently licensed for use against rotavirus. In India, Rotarix wasaunched in 2008 and RotaTeq in 2011. Both vaccines are available

s between Rotarix and RotaTeq vaccine strains are shown in pink colour. The aminoom RotaTeq are in blue colour. The amino acid residues different from both Rotarixutralization escape mutation sites are shown in red colour.

through the private sector. However, they have not been introducedinto the national immunization program [37]. The Indian Academyof Paediatrics Committee on Immunization (IAPCOI) recommendsadministration of either of the vaccines to children with consentfrom the parents [38]. According to a nationally representative sur-vey carried out during 2009–2010, 9.7% of sampled paediatriciansin India reported routine administration of rotavirus vaccine [39].However, given that the majority of childhood immunization isdelivered by the public sector, data on rotavirus vaccine coveragein India is not currently available.

The mechanisms of protection against rotavirus after vaccina-tion are not fully understood. This has resulted in the adoption ofdifferent approaches to the development of broadly protective vac-cines. The RotaTeq vaccine (pentavalent) is based on the conceptthat genotype specific neutralizing antibodies against the rotaviralouter capsid proteins VP7 and VP4 are the primary determinants ofprotection and thus includes VP7 and VP4 components of the majorhuman rotavirus genotypes [40]. The Rotarix vaccine (monovalentG1P[8]), on the other hand, is based on the theory that protec-tive immune response could be stimulated by B- or T-cell epitopespresent on any rotaviral protein, and these epitopes may be con-served among different rotavirus VP7 and VP4 genotypes [40]. Boththe vaccines have demonstrated efficacy against a range of geno-types in the developed countries [41–43].

The success of the rotavirus vaccines in India will depend on

their ability to provide protection against the rotavirus strainsprevalent in the country. G1P[8] rotavirus strains are predominantin India and are represented in both the current vaccines. In thisstudy, we investigated the intragenotypic differences between the

R. Kulkarni et al. / Vaccine 32S (2014) A75–A83 A81

Table 4Alignment of the amino acid residues in (A) VP8* and (B) VP5* antigenic epitopes of the G1P[8] rotavirus strains circulating in Pune and rotavirus vaccines.

Table includes representative G1P[8] strains from children in Pune. The G1P[8] strains from adolescents and adults in Pune also showed the same lineage-specific aminoacid substitutions (data not shown). Representative G1P[8] strains from Kolkata (GRAVP432, GRAVP431, GRAVP427, GRAVP418, ISO114, 47B3), Delhi (Dan279, Dan103) andManipur (mani-375/07, mani-140/06) are also included for comparison. The amino acid differences between Rotarix and RotaTeq vaccine strains are shown in pink colour.T those

b reviou

GRWbfwIm

he amino acid residues that differ from Rotarix are shown in yellow colour while

oth Rotarix and RotaTeq are in green colour. The amino acid positions described p

1P[8] strains in India and the G1, P[8] components of Rotarix andotaTeq vaccines, by comparison of the VP7 and VP4 sequences.e found that the G1P[8] rotavirus strains circulating in India

elong to the G1-Lineage 1, P[8]-Lineage 3 (Figs. 1 and 2) and differ

rom the subgenotypic lineages of vaccine strains. On comparisonith vaccine strains, the G1-Lineage 1, P[8]-Lineage 3 strains from

ndia show amino acid variations at known neutralization escapeutation sites [30–32] within the VP7 and VP4 antigenic epitopes

that differ from RotaTeq are in blue colour. The amino acid residues different fromsly [31, 32] as neutralization escape mutation sites are shown in red colour.

(Tables 3 and 4). Such amino acid variations between the differ-ent subgenotypic lineages warrant further investigation as theymay ultimately affect vaccine efficacy, particularly if protectionis mediated primarily by VP7 and VP4 genotype specific immune

responses.

Antigenic differences have been reported previously betweenthe G1-Lineage 2 and Lineage 3 strains which share 95.9–96.5%amino acid identity in VP7 protein and differ at the amino acid

A ccine 3

pGlqpot

bonpHooam

4Taisv

l(ivVGevaicitsscidwR

tGIttnretr

C

A

o

[

[

[

[

[

[

[

[

[

[

[

[

[

82 R. Kulkarni et al. / Va

ositions 97 and 147 in the VP7 epitopes. Antisera raised against the1-Lineage 3 strain, D, neutralized another strain (Wa) of the same

ineage more efficiently than G1-Lineage 2 strains [44]. This raisesuestions of antigenic variability between the G1-Lineage 1 strainsrevailing in India and G1-Lineages 2 (Rotarix) and 3 (RotaTeq)f rotavirus vaccine strains and the immune response induced byhem.

A study conducted to examine the antigenic differencesetween the strain MX08-659 of P[8]- Lineage 3 and the Wa strainf P[8]-Lineage 1, has described the use of truncated recombi-ant VP8* peptides from each of these strains and suggested theresence of conserved epitopes in the VP8* variable region [45].owever, in the present study, comparison of the VP8* epitopesf the P[8]-Lineage 3 strains from India with the vaccine strainsf P[8]-Lineage 1 (Rotarix) or Lineage 2 (RotaTeq) revealed aminocid differences (Table 4A and B) at known neutralization escapeutation sites [31,32].Rotavirus strains belonging to the G1-Lineage 1, P[8]-Lineage

(Figs. 1 and 2) have been identified in India during the 2000s.he antigenic properties of the P[8]-Lineage 4 or OP354-like strainsre not well understood. The P[8]-Lineage 4 strains are beingncreasingly detected worldwide [13,16,17,20,21,46–48] leading topeculation about the long term protective effect of the currentaccines against this divergent lineage.

The G1-Lineage 1, P[8]-Lineage 3 strains, indicating the sameineage-specific amino acid substitutions noted in the present studyTables 3 and 4), are currently in circulation worldwide [8,9,11–23]ncluding in Europe and America wherein the efficacy of rotavirusaccines is high [41–43]. Thus, sequence differences in VP7 andP4 encoding genes, between the circulating G1P[8] strains and the1, P[8] components of vaccine strains, do not seem to render anyffect as yet on vaccine efficacy in these countries. In fact, Rotarixaccine (monovalent G1P[8]) has been shown to be effective evengainst non-G1P[8] rotavirus strains [42,43]. Antibodies againstmmunorecessive neutralization epitopes in VP7 and VP4 which areonserved between genotypes and/or immune responses againstnternal or non-structural proteins are thought to be responsible forhis apparent cross-protection [40]. Although both vaccines havehown substantial utility in Europe and America to date, it has beenuggested that their long term use may result in selection of strainsapable of escaping vaccine-induced immunity [49]. It is worth not-ng that, after the introduction of Rotarix vaccine in Belgium, theecrease of G1P[8] strains belonging to lineages closer to Rotarixas more than the decrease of G1P[8] strains distantly related tootarix [50].

In conclusion, the present study describes differences betweenhe G1P[8] rotavirus strains circulating in Pune, India and the1 and P[8] components of the Rotarix and RotaTeq vaccines.

n order to understand the significance of these differences andheir influence if any, on vaccine efficacy, further investigation ofhe intragenotype antigenic variability and the protective mecha-ism of vaccines would be necessary. Any increase in use of theotavirus vaccines in India, may have long term effects on strainvolution leading to emergence of novel strains. This warrants con-inuous monitoring of the subgenotypic lineages within the diverseotavirus G1P[8] strains.

onflict of interest statement

The authors have no conflicts of interest to report.

cknowledgements

The authors thank Dr. D.T. Mourya, Director, National Institutef Virology, Pune for his support. The work presented here involves

[

[

2S (2014) A75–A83

utilization of some of the specimens collected during 2005–2009under a multicentric study on rotavirus surveillance coordinatedand funded by Division of Epidemiology and Communicable Dis-eases, ICMR Headquarters, New Delhi and CDC, Atlanta. (Grantnumber: 5/8-1(183)/TF/2002/NIV(1)-ECD-II dated 07/18/07/2005).

References

[1] Tate JE, Burton AH, Boschi-Pinto C, Steele AD, Duque J, Parashar UD, et al. 2008estimate of worldwide rotavirus-associated mortality in children younger than5 years before the introduction of universal rotavirus vaccination programmes:a systematic review and meta-analysis. Lancet Infect Dis 2012;12:136–41.

[2] Tate JE, Chitambar S, Esposito DH, Sarkar R, Gladstone B, Ramani S, et al.Disease and economic burden of rotavirus diarrhoea in India. Vaccine2009;27(5):F18–24.

[3] Estes MK, Kapikian AZ. Rotaviruses. In: Knipe DM, Howley PM, editors. FieldsVirology. 5th ed. Philadelphia, PA: Lippincott Williams and Wilkins; 2007. p.1918–74.

[4] Settembre EC, Chen JZ, Dormitzer PR, Grigorieff N, Harrison SC. Atomic modelof an infectious rotavirus particle. EMBO J 2011;30:408–16.

[5] Matthijnssens J, Ciarlet M, McDonald SM, Attoui H, Baı́nyai K, Brister JR, et al.Uniformity of rotavirus strain nomenclature proposed by the Rotavirus Classi-fication Working Group (RCWG). Arch Virol 2011;156(8):1397–413.

[6] Trojnar E, Sachsenröder J, Twardziok S, Reetz J, Otto PH, Johne R. Identification ofan avian group A rotavirus containing a novel VP4 gene with a close relationshipto those of mammalian rotaviruses. J Gen Virol 2013;94(Pt 1):136–42.

[7] Bányai K, László B, Duque J, Steele AD, Nelson AS, Gentsch JR, Parashar UD.Systematic review of regional and temporal trends in global rotavirus straindiversity in the pre rotavirus vaccine era: insights for understanding the impactof rotavirus vaccination programs. Vaccine 2012;30S:A122–30.

[8] Phan TG, Khamrin P, Quang TD, Dey SK, Takanashi S, Okitsu S, et al. Detectionand genetic characterization of group A rotavirus strains circulating amongchildren with acute gastroenteritis in Japan. J Virol 2007;81:4645–53.

[9] Espinola EE, Amarilla A, Arbiza J, Parra GI. Sequence and phylogenetic analysis ofthe VP4 gene of human rotaviruses in Paraguay. Arch Virol 2008;153:1067–73.

10] Kang G, Desai R, Arora R, Chitamabar S, Naik TN, Krishnan T, et al. Diversityof circulating rotavirus strains in children hospitalized with diarrhea in India,2005–2009. Vaccine 2013;31(27):2879–83.

11] Arista S, Giammanco GM, De Grazia S, Ramirez S, Biundo CL, Colomba C, et al.Heterogeneity and temporal dynamics of evolution of G1 human rotavirusesin a settled population. J Virol 2006;80:10724–33.

12] Diwakarla CS, Palombo EA. Genetic and antigenic variation of capsid pro-tein VP7 of serotype G1 human rotavirus isolates. J Gen Virol 1999;80:341–4.

13] Zeller M, Patton JT, Heylen E, De Coster S, Ciarlet M, Van Ranst M, MatthijnssensJ. Genetic analyses reveal differences in the VP7 and VP4 antigenic epitopesbetween human rotaviruses circulating in Belgium and rotaviruses in Rotarixand RotaTeq. J Clin Microbiol 2012;50:966–76.

14] Maranhão AG, Vianez-Júnior J, Benati FJ, Bisch PM, Santos N. Polymorphism ofrotavirus genotype G1 in Brazil: in silico analysis of variant strains circulatingin Rio de Janeiro from 1996 to 2004. Infect Genet Evol 2012;12:1397–404.

15] Mladenova Z, Iturriza-Gomara M, Esona MD, Gray J, Korsun N. Genetic char-acterization of Bulgarian rotavirus isolates and detection of rotavirus variants:challenges for the rotavirus vaccine program? J Med Virol 2011;83:348–56.

16] Cashman O, Collins PJ, Lennon G, Cryan B, Martella V, Fanning S, et al. Molecularcharacterization of group A rotaviruses detected in children with gastroenteri-tis in Ireland in 2006–2009. Epidemiol Infect 2012;140:247–59.

17] Pietsch C, Schuster V, Liebert UG. A hospital based study on inter- andintragenotypic diversity of human rotavirus A VP4 and VP7 gene segments,Germany. J Clin Virol 2011;50:136–41.

18] László B, Kónya J, Dandár E, Deák J, Farkas A, Gray J, et al. Surveillance of humanrotaviruses in 2007–2011 Hungary: exploring the genetic relatedness betweenvaccine and field strains. J Clin Virol 2012;55:140–6.

19] Chan-it W, Thongprachum A, Dey SK, Phan TG, Khamrin P, Okitsu S, et al. Detec-tion and genetic characterization of rotavirus infections in non-hospitalizedchildren with acute gastroenteritis in Japan, 2007–2009. Infect Genet Evol2011;11:415–22.

20] Nakagomi T, Nakagomi O, Dove W, Doan YH, Witte D, Ngwira B, et al. Molecularcharacterization of rotavirus strains detected during a clinical trial of a humanrotavirus vaccine in Blantyre, Malawi. Vaccine 2012;30S:A140–51.

21] Khananurak K, Vutithanachot V, Simakachorn N, Theamboonlers A, Chongsri-sawat V, Poovorawan Y. Prevalence and phylogenetic analysis of rotavirusgenotypes in Thailand between 2007 and 2009. Infect Genet Evol 2010;10:537–45.

22] McDonald SM, McKell AO, Rippinger CM, McAllen JK, Akopov A, Kirkness EF,et al. Diversity and relationships of cocirculating modern human rotavirusesrevealed using large-scale comparative genomics. J Virol 2012;86(17):

9148–62.

23] Afrad MH, Hassan Z, Farjana S, Moni S, Barua S, Das SK, et al. Changing profile ofrotavirus genotypes in Bangladesh, 2006–2012. BMC Infect Dis 2013;13:320.

24] Arora R, Chhabra P, Chitambar SD. Genetic diversity of genotype G1 rotavirusesco-circulating in western India. Virus Res 2009;146(1–2):36–40.

ccine 3

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

R. Kulkarni et al. / Va

25] Zade JK, Chhabra P, Chitambar SD. Characterization of VP7 and VP4genes of rotavirus strains: 1990–1994 and 2000–2002. Epidemiol Infect2009;137:936–42.

26] Gouvea V, Glass RI, Woods P, Taniguchi K, Clark HF, Forrester B, et al. Polymerasechain reaction amplification and typing of rotavirus nucleic acid from stoolspecimens. J Clin Microbiol 1990;28:276–82.

27] Gentsch JR, Glass RI, Woods P, Gouvea V, Gorziglia M, Flores J, et al. Identificationof group A rotavirus gene 4 types by polymerase chain reaction. J Clin Microbiol1992;30:1365–73.

28] Arora R, Chitambar SD. Full genomic analysis of Indian G1P[8] rotavirus strains.Infect Genet Evol 2011;11:504–11.

29] Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. MEGA5: molecularevolutionary genetics analysis using maximum likelihood evolutionary dis-tance, and maximum parsimony methods. Mol Biol Evol 2011;28(10):2731–9.

30] Aoki ST, Settembre EC, Trask SD, Greenberg HB, Harrison SC, Dormitzer PR.Structure of rotavirus outer-layer protein VP7 bound with a neutralizing Fab.Science 2009;324:1444–7.

31] Dormitzer PR, Sun ZY, Wagner G, Harrison SC. The rhesus rotavirus VP4 sialicacid binding domain has a galectin fold with a novel carbohydrate binding site.EMBO J 2002;21:885–97.

32] Dormitzer PR, Nason EB, Prasad BV, Harrison SC. Structural rearrangementsin the membrane penetration protein of a non-enveloped virus. Nature2004;430:1053–8.

33] Tatte VS, Chitambar SD. Diversity in the VP7 encoding genes of rotavirus strainsisolated from adolescent and adult cases of acute gastroenteritis. J Med Virol2012;84:1481–8.

34] Tatte VS, Chitambar SD. Intragenotypic diversity in the VP4 encoding genes ofrotavirus strains circulating in adolescent and adult cases of acute gastroen-teritis in Pune Western India: 1993 to 1996 and 2004 to 2007. J Gen Mol Virol2011;3(4):53–62.

35] Samajdar S, Ghosh S, Dutta D, Chawla-Sarkar M, Kobayashi N, Naik TN. Humangroup A rotavirus P[8] Hun9-like and rare OP354-like strains are circulatingamong diarrhoeic children in Eastern India. Arch Virol 2008;153:1933–6.

36] World Health Organization. Meeting of the immunization Strategic AdvisoryGroup of Experts, April 2009–conclusions and recommendations. Wkly Epi-demiol Rec 2009;84:220–36.

37] Indian Academy of Pediatrics Committee on Immunization (IAPCOI). Consen-sus recommendations on immunization and IAP immunization timetable 2012.Indian Pediatr 2012;49(7):549–64.

38] Indian Academy of Pediatrics Committee on Immunization (IAPCOI). Consensusrecommendations on immunization, 2008. Indian Pediatr 2008;45(8):635–48.

[

2S (2014) A75–A83 A83

39] Gargano LM, Thacker N, Choudhury P, Weiss PS, Pazol K, Bahl S, et al. Predictorsof administration and attitudes about pneumococcal, Haemophilus influenzaetype b and rotavirus vaccines among pediatricians in India: a national survey.Vaccine 2012;30:3541–5.

40] Ward RL, Clark HF, Offit PA. Influence of potential protective mecha-nisms on the development of live rotavirus vaccines. J Infect Dis 2010;202:S72–9.

41] Vesikari T, Matson DO, Dennehy P, Van Damme P, Santosham M, RodriguezZ, et al. Safety and efficacy of a pentavalent human-bovine (WC3) reassortantrotavirus vaccine. N Engl J Med 2006;354:23–33.

42] Ruiz-Palacios GM, Pérez-Schael I, Velázquez FR, Abate H, Breuer T, ClemensSC, et al. Safety and efficacy of an attenuated vaccine against severe rotavirusgastroenteritis. N Engl J Med 2006;354:11–22.

43] Vesikari T, Karvonen A, Prymula R, Schuster V, Tejedor JC, Cohen R, et al. Efficacyof human rotavirus vaccine against rotavirus gastroenteritis during the first 2years of life in European infants: randomised, double-blind controlled study.Lancet 2007;370:1757–63.

44] Jin Q, Ward RL, Knowlton DR, Gabbay YB, Linhares AC, Rappaport R, et al. Diver-gence of VP7 genes of G1 rotaviruses isolated from infants vaccinated withreassortant rhesus rotaviruses. Arch Virol 1996;141:2057–76.

45] Contreras JF, Menchaca GE, Infante R, Hernandez CE, Rodriguez C, Tamez RS.Evidence of conserved epitopes in variable region of VP8* subunit of VP4protein of rotaviruses of P[8]-1 and P[8]-3 lineages. Acta Virologica 2011;55:273–8.

46] Nagashima S, Kobayashi N, Paul SK, Alam MM, Chawla-Sarkar M, Krishnan T.Characterization of full-length VP4 genes of OP354-like P[8] human rotavirusstrains detected in Bangladesh representing a novel P[8] subtype. Arch Virol2009;154:1223–31.

47] Wang YH, Zhou X, Ghosh S, Zhou DJ, Pang BB, Peng JS, et al. Prevalence of humanrotavirus genotypes in Wuhan, China, during 2008–2011: changing trend ofpredominant genotypes and emergence of strains with the P[8]b subtype ofthe VP4 gene. Arch Virol 2011;156:2221–31.

48] Kuzuya M, Fujii R, Hamano M, Kida K, Kanadani T, Nishimura K, et al. Molec-ular characterization of OP354-like P[8](P[8]b subtype) human rotaviruses Aspecies isolated in Japan. J Med Virol 2012;84:697–704.

49] Matthijnssens J, Bilcke J, Ciarlet M, Martella V, Baı́nyai K, Rahman M, et al.

Rotavirus disease and vaccination: impact on genotype diversity. Future Micro-biol 2009;4(10):1303–16.

50] Zeller M, Matthijnssens J, Rahman M, Van Ranst M. Possible immune evasion byG1 lineage I after vaccine introduction in Belgium? Third European RotavirusBiology Meeting. 2009.