Journal of Medical Virology

Characterization of Rubella Virus GenotypesAmong Pregnant Women in Northern Vietnam,2011–2013

Son Van Le,1,2 Duc Hoang Le,1,2 Huong Thi Hoang,3 Ha Hoang,2 Nam Trung Nguyen,1,2 andHa Hoang Chu2*1Laboratory of Applied DNA Technology, Institute of Biotechnology, Vietnam Academy of Science and Technology,Hanoi, Vietnam2National Key Laboratory of Gene Technology, Institute of Biotechnology, Vietnam Academy of Science andTechnology, Hanoi, Vietnam3National Hospital of Obstetrics and Gynecology, Hanoi, Vietnam

Rubella virus (RV) infection is an unresolvedclinical complication that affects children indeveloping countries including Vietnam. RVinfection during the first trimester of pregnan-cy causes severe birth defects known as con-genital rubella syndrome. This study reportson the genomic characterization of RV strainscirculating in northern Vietnam during 2011–2013. RV-IgM positive amniotic fluid specimenswere collected from 38 women from northernVietnam who presented with clinical rubella atthe National Hospital of Obstetrics and Gyne-cology in Hanoi, Vietnam. The RV genes weredetermined by nested PCR with primers ampli-fying the 739-nucleotide coding region of theE1 gene. The sequences from the amplifiedDNA fragments were phylogenetically analyzedand compared to reference RV strains. Seven-teen out of 38 samples are positive for RVdetecting. All new RV isolates are clustered togenotype 2B. Eighteen amino acid mutationswere found in the T and B cell epitopes. Theseresults suggest that genotype 2B RV strainsfrequently circulate in northern Vietnam. Thesedata describe the RV genotype in Vietnam withthe aim of improving maternal and child healthin this country. J. Med. Virol.# 2014 Wiley Periodicals, Inc.

KEY WORDS: genotype; molecular; phyloge-netic

INTRODUCTION

Rubella virus (RV) causes serious birth defects incases of maternal infection during pregnancy. RVinfection might result in congenital rubella syndrome(CRS) with symptoms that include cataract, glauco-

ma, and spontaneous abortion [Banatvala andBrown, 2004]. RV, the only member of the Rubivirusgenus of the Togaviridae family [Prasad et al., 2013],has a single stranded, positive sense RNA genomethat is 9,762 nucleotides in length [Hobman andChantler, 2007]. Rubella virus is an enveloped virus,and the virions are approximately 600–800A

�[Battisti

et al., 2012]. RV genome contains three structuralproteins including capsid (C) protein, E1 and E2envelope glycoproteins, and two non-structuralproteins including p90 and p150 [Frey, 1994]. TheC protein functions in the structure, transcription,and replication of the virus [Chen and Icenogle,2004]. The glycoprotein E1 plays a crucial role inbinding to the host cell receptors and bears anantigenic determinant for neutralizing antibodies[Terry et al., 1988; Cusi et al., 1989; Chayeet al., 1992, 1993; Mitchell et al., 1992; Ou et al.,1993]. The immunogenic role of E2 is unclear[Abernathy et al., 2013]. The two non-structuralproteins are involved in viral replication [Hobmanand Chantler, 2007].The 739-nucleotide sequence genotyping window in

E1 gene has been widely used for genotyping and forphylogenetic analyses of RV strains [Frey andAbernathy, 1993; Zheng et al., 2003; WHO, 2005,2006]. According to the systematic nomenclature

Grant sponsor: Institute of Biotechnology CS11-11

Conflict of Interest: None declared.�Correspondence to: Ha Hoang Chu, PhD, National Key

Laboratory of Gene Technology, Institute of Biotechnology,Vietnam Academy of Science and Technology, 18 Hoang QuocViet Road, Cau Giay district, Hanoi, Vietnam.E-mail: chuhoangha@ibt.ac.vn

Accepted 15 July 2014

DOI 10.1002/jmv.24049Published online in Wiley Online Library(wileyonlinelibrary.com).

�C 2014 WILEY PERIODICALS, INC.

adopted by the World Health Organization (WHO),two viral clades have been identified by geneticcharacterization. Clade 1 is divided into six genotypes(1B, 1C, 1D, 1E, 1F, and 1G) and four provisionalgenotypes (1a, 1h, 1i, and 1j), and clade 2 is dividedinto three genotypes (2A, 2B, and 2C) [WHO,2005, 2006, 2007]. Previous reports showed that ninegenotypes (1B, 1C, 1D, 1E, 1F, 1G, 2A, 2B, and 2C)and four provisional genotypes 1a, 1h, 1i, and 1j arepresent in Europe, North America, South America,and Asia whereas genotype 2B frequently circulatesin Asian countries [Katow et al., 1997; Freyet al., 1998; Zheng et al., 2003].In Vietnam, rubella vaccination is offered by the

private sector; although vaccination programs havereduced RV infection rates, reports of RV infectionscontinue [Tran et al., 2011, 2012; Pham et al., 2013].Phylogenetic analysis is a powerful tool that iscommonly used to study the genetic divergence andphylogenetic relationship between circulating strainsof emerging viruses such as rubella, foot-and-mouth,West Nile, and influenza A; the analysis contributesto a better understanding of the distribution of thediseases caused by these viruses [Malirat et al., 2007;Tran et al., 2011, 2012; A~nez et al., 2013; Phamet al., 2013; Kim et al., 2014]. The aim of this studyis to characterize rubella virus genes from pregnantwomen in northern Vietnam to increase levels ofknowledge regarding the genetic characterization ofRV in Vietnam.

MATERIALS AND METHODS

Specimens

Specimens were collected from the RV-IgM positiveamniotic fluid of 38 women who presented withclinical rubella at the National Hospital of Obstetricsand Gynecology (Hanoi, Vietnam). The women residedin northern Vietnam provinces including Bac Giang,Hai Phong, Hung Yen, Quang Ninh, Hai Duong,Nghe An, Tuyen Quang, Phu Tho, Nam Dinh, andThanh Hoa. According to clinical records from theNational Hospital of Obstetrics and Gynecology, noneof the women in this study had been vaccinatedagainst rubella. Informed consent was obtained fromall the patients or a family of each patient. The studywas performed in accordance with protocols approvedby the ethnics committee of the National Hospital ofObstetrics and Gynecology.

RNA Extraction and RT-PCR

For RNA genome extracted directly from clinicalspecimens, 150ml of specimen was used. The RNAgenome was extracted with Trizol (Invitrogen, Carls-bad, CA) according to the manufacturer’s instruc-tions. The RT-PCR reactions were performed aspreviously described [Tran et al., 2012] with minormodifications. Briefly, cDNA was synthesized fromthe RNA genome using the First Strand cDNA

Synthesis Kit (Fermentas, Vilnius, Lithuania). Theprotocol of this kit uses RevertAidTM M-MuLV Re-verse Transcriptase, which has lower RNase H activi-ty than AMV reverse transcriptase, with random(dT)18 primers, which bind non-specifically to theRNA template, to synthesize the cDNA from all theRNAs in the total RNA population. The protocol usedin cDNA synthesis is 25˚C for 5min, 42˚C for 60minand 70˚C for 5min. The nested PCR was imple-mented with primer pairs designed for the 739-nucleotide E1 gene following the WHO-recommendedsequences RubF1 50-CCC ACC GAC ACC GTG ATGA -30; RubR1 50-CCA GGT CTG CCG GGT CTC -30;RubF2 50-GTG ATG AGC GTG TTC GCC C-30;RubR2 50-GCD GTG GTG TGT GTG CC -30. For thefirst round, the total volume of 25ml of the PCRreaction consisted of 2.5ml of 10X PCR buffer, 1.5mlof 25mM MgCl2, 2.5ml of dNTP, 0.5ml of 10 pmolRubF1, 0.5ml of 10 pmol RubR1, 1ml of 5U/ml TaqDNA polymerase, 3ml of cDNA, and 13.5ml of dis-tilled water. The PCR reaction was performed with40 cycles of 94˚C for 1min, 56˚C for 1min, 72˚C for90 sec, and an extension of 72˚C for 7min and thenkept at 4˚C. For the nested PCR, 1ml of the first-round product was used as the template in a totalvolume of 25ml using the primer pairs, RubF2/RubR2(10 pmol), at an annealing temperature of 55˚C. ThePCR products were run on 1% agarose gel, and thebands were detected by ethidium bromide under UVlight. The expected band size was 875 bp.

Nucleotide Sequencing and PhylogeneticAnalysis

The products of the nested PCR were purified usingthe Qiagen kit (Qiagen, Hilden, Germany), insertedinto a cloning vector and then sequenced using an ABIPRISM1 3100 Avant Genetic Analyzer with BigDyeTerminator v3.1 Cycle Sequencing Kit (Applied Bio-systems, Foster City, CA). The nucleotide sequenceswere analyzed using BLAST (NCBI) and compared toreference strains and strains from a database of all thegenotypes. The sequences and phylogenesis were ana-lyzed using BioEdit [Hall, 1999] v.7.0.5.3 and MEGA v5.0 software [Tamura et al., 2011]. The sequences ofthe RV strains were submitted to the GeneBank andwere assigned accession numbers KJ095587-KJ095603for the RV E1 gene. The RV strains were namedaccording to the WHO systematic nomenclature for theRV [WHO, 2006].

RESULTS

Screening RV and Analyzing RV Sequences

Using reverse transcription PCR (RT-PCR), the RVRNA of 17 positive samples was detected in 38 testedsamples. The 739-nucleotide fragments encoding ami-no acid in the E1 gene recommended by WHO havebeen analyzed and compared to the E1 sequences ofthe vaccine strain, WHO reference strains and data-base strains. In this study, the 739-nucleotide

J. Med. Virol. DOI 10.1002/jmv

2 Van Le et al.

sequence of the region encoding amino acid residues159–404 within the E1 glycoprotein was analyzedas recommended by the WHO for the 17 positiveVietnamese strains in comparison with the vaccinestrain Rvi/USA/64 (RA27/3US64), three WHO 2Breference strains and the database strains [Tranet al., 2012] (Fig. 1). Compared to the vaccinestrains Rvi/USA/64, the 17 Vietnamese strainsshowed 85 variable positions in the nucleotidesequence and 18 in the amino acid sequence (Fig. 2).The analysis of the nucleotide changes demonstratedthat there are nine changes at the first base andsecond base and 67 changes at the third base. Threecodons have nucleotide substitutions at the first andthe third position simultaneously. The amino acidslocated in the T and B cell epitopes described by Chaye

et al. [1993] were analyzed. Four amino acid mutationswere present in the sequence (H210Y, A333V, T337A,L377V) of most (15/17 (88%)) of the strains. Twostrains, including Rvs/Haiphong and Rvs/Hungyen, didnot contain the change at the A333 position. Inaddition, seven amino acid mutations were present inthe sequence of various strains including S216P,N224D, V332L, R336H, E359K, N367S, and V386A.

Phylogenetic Analysis

The phylogenetic tree of the RV strains based on the739-nucleotide sequence of the E1 gene is shown inFigure 3. The strains circulating in northern Vietnambelonged to the group 2B genotype as well as withthose from southern Vietnam (i.e., Rvs/HoChiMinh.

Fig. 1. The percentage of similarity and divergence of the RV strains from northern Vietnam,vaccine strain Rvi/USA/64 and the WHO 2B reference strains with Bioedit software.

Fig. 2. The comparison of the amino acid sequence of a 739-nucleotide window within the E1gene (nucleotide 8731-9469, amino acid 159-404) of 17 Vietnamese strains, the vaccine strainRvi/USA/64 and three WHO 2B reference strains including Rvi/TELAVIV.ISR/68, Rvi/WA.USA/16.00, and Rvi/ANHUI.CHN/00.

J. Med. Virol. DOI 10.1002/jmv

Rubella Virus Genotypes in Vietnam 3

Fig. 3. The phylogenetic tree of the sequences of the RV strains from northern Vietnam,vaccine strain Rvi/USA/64 and the WHO 2B reference strains with MEGA software. Bootstrapvalues of more than 70% are shown at the branch nodes. The RV strains in this study areunderlined. The WHO 2B reference strains are indicated by circles. The strains from southernVietnam are marked with asterisks. The vaccine strain is indicated by a solid square.

J. Med. Virol. DOI 10.1002/jmv

4 Van Le et al.

VNM23.11-2B) and other 2B genotype referencestrains from the database (i.e., Rvi/Aichi.JPN/18.13/2-2B). The nucleotide difference between the Vietnam-ese 2B strains ranged from 0.1% to 1.6%. The strainsRvs/Hanoi.VNM/35.11 and Rvs/QuangNinh.VNM/35.11 have the highest nucleotide divergence, with amean of 1.6%. The strains Rvs/BacGiang.VNM/33.11,Rvs/Hanoi.VNM/30.11, and Rvs/Hanoi.VNM/34.11have the highest nucleotide similarity, reaching amean of 100%. The mean divergence within all theVietnamese viral sequences was 0.5% and 1.4% rela-tive to the WHO 2B reference strains. The phyloge-netic analysis of the 17 sequences demonstrated thatthe circulating strains in northern Vietnam are the2B genotype with a bootstrap value of 70% or higher(Fig. 3). The Vietnamese strains detected in northernVietnam represent the same cluster as the RV strainsfound in Japan during 2011–2013 (Fig. 3).

DISCUSSION

The RV strains from northern Vietnam werephylogenetically analyzed, and they showed theregional epidemiology of this virus during 2011 to2013. In Vietnam, the presence and spread of RV stillremains, despite the effort to control the virus with avaccination campaign by the private sector. Thesesamples were harvested from provinces in northernVietnam, a large region that includes 10 provinces.According to the RV genotypes established by theWHO, there are nine RV genotypes and four provi-sions. It has been reported recently that the RVstrains from southern Vietnam belonged to the 2Bgenotype [Tran et al., 2012; Pham et al., 2013].Additionally, the RV strains found in northernVietnam in this study belonged to the 2B genotype.These results suggested that the genotype 2B virusescirculate predominantly in two regions of Vietnam.Combined with the data of the RV strains from thesouthern region of the country, this study providescomprehensive data regarding the RV genotypescirculating in Vietnam.Compared to the vaccine strain, 80/98 (90.8%)

similar nucleotides and 18 similar amino acid varia-tions were identified. In general, the nucleotide andamino acid sequences of RV were highly conserved[Bosma et al., 1996; Hubschen et al., 2007]. In thisstudy, in which samples were harvested from pa-tients from various provinces in northern Vietnam,the regional difference appears to be an importantfactor that affects the epidemic characteristics of RVand might cause additional variations in the aminoacid sequence. The mutations found in these Viet-namese strains appear in the T and B cell epitope ofthe immuno-reactive region in the E1 glycoprotein.The mutations in the amino acids of the RV strainsfrom northern Vietnam require further investigationto reveal the differences in the immunogenic functionof RV strains.

The possibility of spreading RV from Southeast Asiancountries including Vietnam to Japan and other coun-tries has been reported [Pham et al., 2013]. Sugishitafound an ongoing rubella outbreak among adults inTokyo, Japan during 2012 to 2013 [Sugishitaet al., 2013]. In this study, the RV sequences found inJapan during that time period were associated with anidentical cluster of Vietnamese strains, with a nucleo-tide sequence similarity of 99%. These results suggestthat the strains circulating in Japan may have origi-nated in Vietnam. However, confirmation of RV distri-bution in neighboring countries requires moreevidence.This study contributes to the understanding of the

molecular epidemiological characterization of the RVstrains circulating in northern Vietnam. Althoughfurther studies are necessary for complete characteri-zation of RV in Vietnam, these data are helpful forcontrolling the spread of RV.

ACKNOWLEDGMENTS

We thank the physicians and nurses of the Nation-al Hospital of Obstetrics and Gynecology (Hanoi,Vietnam) for their help in the patient counseling andsample collection.

REFERENCES

Abernathy E, Chen MH, Bera J, Shrivastava S, Kirkness E, ZhengQ, Bellini W, Icenogle J. 2013. Analysis of whole genomesequences of 16 strains of rubella virus from the United States,1961-2009. Virol J 10:32.

A~nez G, Grinev A, Chancey C, Ball C, Akolkar N, Land KJ,Winkelman V, Stramer SL, Kramer LD, Rios M. 2013. Evolu-tionary dynamics of West Nile virus in the United States, 1999-2011: Phylogeny, selection pressure and evolutionary time-scaleanalysis. PLoS Negl Trop Dis 7:e2245.

Banatvala JE, Brown DWG. 2004. Rubella. Lancet 363:1127–1137.

Battisti AJ, Yoder JD, Plevka P, Winkler DC, Prasad VM, KuhnRJ, Frey TK, Steven AC, Rossmann MG. 2012. Cryo-electrontomography of rubella virus. J Virol 86:11078–11085.

Bosma TJ, Best JM, Corbett KM, Banatvala JE, Starkey WG. 1996.Nucleotide sequence analysis of a major antigenic domain of theE1 glycoprotein of 22 rubella virus isolates. J Gen Virol 77:2523–2530.

Chaye H, Chong P, Tripet B, Brush B, Gillam S. 1992. Localizationof the virus neutralizing and hemagglutinin epitopes of E1glycoprotein of rubella virus. Virology 189:483–492.

Chaye H, Ou D, Chong P, Gillam S. 1993. Human T- and B-cellepitopes of E1 glycoprotein of rubella virus. J Clin Immunol13:93–100.

Chen MH, Icenogle JP. 2004. Rubella virus capsid protein modu-lates viral genome replication and virus infectivity. J Virol78:4314–4322.

Cusi MG, Metelli R, Valensin PE. 1989. Immune responses to wildand vaccine rubella. Arch Virol 106:63–72.

Frey TK. 1994. Molecular biology of rubella virus. Adv Virus Res44:69–160.

Frey TK, Abernathy ES. 1993. Identification of strain-specific nucleo-tide sequences in the RA 27/3 rubella virus vaccine. J Infect Dis168:854–864.

Frey TK, Abernathy ES, Bosma TJ, Starkey WG, Corbett KM, BestJM, Katow S, Weaver SC. 1998. Molecular analysis of rubellavirus epidemiology across three continents, North America,Europe and Asia, 1961-1997. J Infect Dis 178:642–650.

Hall TA. 1999. BioEdit: A user-friendly biological sequence align-ment editor and analysis program for Windows 95/98/NT. NuclAcids Symp Ser 41:95–98.

J. Med. Virol. DOI 10.1002/jmv

Rubella Virus Genotypes in Vietnam 5

Hobman T, Chantler J. 2007. Rubella virus. In: Knipe DM, HowleyPM, editors Fields virology. 5th edition. Philadelphia, PA:Lippincott Williams, and Wilkins. pp 1069–1100.

Hubschen JM, Yermalovich M, Semeiko G, Samoilovich E, BlatunE, De Landtsheer S, Muller CP. 2007. Co-circulation of multiplerubella virus strains in Belarus forming novel genetic groupswithin clade 1. J Gen Virol 88:1960–1966.

Katow S, Minahara H, Fukushima M, Yamaguchi Y. 1997. Molecu-lar epidemiology of rubella by nucleotide sequences of therubella virus E1 gene in three East Asian countries. J Infect Dis176:602–616.

Kim JI, Lee I, Park S, Lee S, Hwang MW, Bae JY, Heo J, Kim D,Jang SI, Kim K, Park MS. 2014. Phylogenetic analysis of aswine influenza A(H3N2) virus isolated in Korea in 2012. PLoSOne 9:e88782.

Malirat V, de Barros JJ, Bergmann IE, Campos Rde M, Neitzert E,da Costa EV, da Silva EE, Falczuk AJ, Pinheiro DS, de VergaraN, Cirvera JL, Maradei E, Di Landro R. 2007. Phylogeneticanalysis of foot-and-mouth disease virus type O re-emerging infree areas of South America. Virus Res 124:22–28.

Mitchell LA, Zhang T, Tingle AJ. 1992. Differential antibodyresponses to rubella virus infection in males and females. JInfect Dis 166:1258–1265.

Ou D, Chong P, Tingle AJ, Gillam S. 1993. Mapping T-cell epitopesof rubella virus structural proteins E1, E2, and C recognized byT-cell lines and clones derived from infected and immunizedpopulations. J Med Virol 40:175–183.

Pham VH, Nguyen TV, Nguyen TT, Dang LD, Hoang NH, NguyenTV, Abe K. 2013. Rubella epidemic in Vietnam: characteristic ofrubella virus genes from pregnant women and their fetuses/new-borns with congenital rubella syndrome. J Clin Virol 57:152–156.

Prasad VM, Willows SD, Fokine A, Battisti AJ, Sun S, Plevka P,Hobman TC, Rossmann MG. 2013. Rubella virus capsid protein

structure and its role in virus assembly and infection. Proc NatlAcad Sci U S A 110:20105–20110.

Sugishita Y, Takahashi T, Hori N, Abo M. 2013. Ongoing rubellaoutbreak among adults in Tokyo, Japan, June 2012 to April2013. West Pac Surveill Response J 4:37–41.

Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S.2011. MEG A5: Molecular evolutionary genetics analysis usingmaximum likelihood, evolutionary distance, and maximum par-simony methods. Mol Biol Evol 8:2731–2739.

Terry GM, Ho-Terry L, Londesborough P, Rees KR. 1988. Localiza-tion of the rubella E1 epitopes. Arch Virol 98:189–197.

Tran DN, Pham NT, Tran TT, Khamrin P, Thongprachum A,Komase K, Hayakawa S, Mizuguchi M, Ushijima H. 2012.Phylogenetic analysis of rubella viruses in Vietnam during 2009-2010. J Med Virol 84:705–710.

Tran DN, Vu MP, Ha MT, Giang TP, Komase K, Mizuguchi M,Ushijima H. 2011. Viral molecular characterization of the firstcongenital rubella syndrome case in Vietnam. Clin Lab 57:397–401.

World Health Organization (WHO). 2005. Standardization of thenomenclature for genetic characteristics of wild-type rubellaviruses. Wkly Epidemiol Rec 80:126–132.

World Health Organization (WHO). 2006. Global distribution ofmeasles and rubella genotypes-update. Wkly Epidemiol Rec81:474–479.

World Health Organization (WHO). 2007. Update of standardnomenclature for wild-type rubella viruses. Wkly Epidemiol Rec82:209–224.

Zheng DP, Frey TK, Icenogle J, Katow S, Abernathy ES, Song KJ,Xu WB, Yarulin V, Desjatskova RG, Aboudy Y, Enders G,Croxson M. 2003. Global distribution of rubella virus genotypes.Emerg Infect Dis 9:1523–1530.

J. Med. Virol. DOI 10.1002/jmv

6 Van Le et al.