Virus Research 99 (2004) 121–129

Genetic characterization of a mumps virus isolate during passagingin the amniotic cavity of embryonated chicken eggs�

Jelena Ivancic∗, Dubravko Forcic, Tanja Kosutic Gulija, Renata Zgorelec, Leonida Repalust,Marijana Baricevic, Majda Mesko-Prejac, Renata Mazuran

Molecular Biomedicine Unit, Department of Research and Development, Institute of Immunology Inc., Rockefellerova 10, 10 000 Zagreb, Croatia

Received 22 July 2003; received in revised form 23 October 2003; accepted 3 November 2003

Abstract

The aim of this study was the molecular characterization of a historical mumps isolate (an alleged individual sample). After RNA extrac-tion and cDNA synthesis, selective nested PCR amplification with specific primers, automated DNA sequencing and RFLP analyses wereperformed. The relative ratios of the detected virus sequences were determined by GeneScan electrophoresis. Phylogenetic tree based on the316 nucleotide region of the SH gene of the mumps virus was generated by the neighbor-joining method.

Results obtained by the described molecular approach show: (a) there are two mumps virus variants, A and B, detected in the fourth passageof wild type virus in the amniotic cavity of embryonated chicken eggs (ECE); (b) variants A and B belong to different genotypes; (c) variantsA and B differ in the HN and NP genes which code for amino acid sequences comprising immunogenic epitopes; (d) variant B contains oneor more minor variants.

We discuss whether the observed differences between the two variants are a consequence of natural heterogeneity or of laboratory contam-ination in the early passages.© 2003 Elsevier B.V. All rights reserved.

Keywords:Mumps; Genotypes; Antigenic epitopes; Subtypes

1. Introduction

Mumps is a systemic viral illness spread by virus-containing droplets from the upper respiratory tract. Thedisease is normally characterized by parotitis and mild non-specific symptoms. The causative agent, the mumps virus,is a member of the genusRubulavirus, family Paramyx-oviridae. The single-stranded, negative-sense RNA genomeis 15 384 nucleotides long and contains seven genes en-coding nucleocapsid protein (NP), phospho protein (P),matrix protein (M), fusion protein (F), small hydrophobicprotein (SH), hemagglutinin-neuraminidase (HN) and RNApolymerase (L) (Carbone and Wolinsky, 2001).

The RNA-dependent RNA polymerase of RNA viruseslacks proofreading capabilities (Domingo et al., 1985), and

� Presented in part at the 12th International Conference on NegativeStrand RNA Viruses, Pisa, Italy, 14–19 June 2003.

∗ Corresponding author. Tel.:+385-1-4684-500;fax: +385-1-4684-303.

E-mail address:jivancic@imz.hr (J. Ivancic).

variant mutants are spontaneously generated during viralgenome replication. Depending on growth conditions bothin vitro and in vivo, a mumps virus isolate contains a hetero-geneous mixture of closely related genomic RNA moleculeswith different nucleotide sequences (quasispecies) (Afzalet al., 1993; Boriskin et al., 1993; Brown et al., 1996; Yateset al., 1996). Studying the genetic heterogeneity in a singleindividual may be fundamental for the understanding of howgenomic changes influence the ability of a virus to causea disease and the characteristics of the pathogenic process(Jones et al., 2002). For example, the Urabe AM9 mumpsvaccine strain is a mixture of viruses differing at amino acid335 of the hemagglutinin-neuraminidase protein with oneform being associated with the development of meningitis(Wright et al., 2000). It may also be of value in elucidatingthe evolution of a virus, designing vaccines and/or improv-ing diagnostic assays.

Comparison of the nucleotide sequences of the SH gene(the most variable part of the mumps genome) from mumpsvirus strains collected worldwide has shown the existenceof 10 genotypes named A–J (Tecle et al., 2001). Different

0168-1702/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.virusres.2003.11.002

122 J. Ivancic et al. / Virus Research 99 (2004) 121–129

virus genotypes have been shown to co-circulate in the samecountry, and the distribution of genotypes may vary, evenbetween regions in close neighbourhood (Takahashi et al.,2000).

The aim of our investigation was the molecular charac-terization of a historical mumps virus isolate (an alleged in-dividual sample).

2. Materials and methods

2.1. The isolation of the wild virus, passage history inembryonated chicken eggs (ECE) and subcultivation onMRC-5 cells

In 1969, a throat washing was taken from a 6-year-oldchild from Zagreb, Croatia, with clinical symptoms ofmumps infection. A flow-chart of the wild virus passagehistory is given inFig. 1. Five 7-day-old ECE were inocu-lated intra-amniotically with 0.2 ml of the specimen. Twoeggs died within 24 h and were discarded, three were incu-bated at 37◦C. The amniotic fluids were harvested six daysafter inoculation, and a hemagglutination assay (HA) wascarried out for each of the three individual samples sepa-rately. Two out of the three amniotic fluids were weaklypositive by HA. The positive amniotic fluids were pooledand inoculated into ten 7-day-old ECE (the second passage).Again, two out of the ten eggs died within 24 h after inocu-lation and were discarded; for the rest, HA was carried outfor each individual amniotic fluid separately 6 days later.

Fig. 1. Passage history of wild mumps virus. Eggs that died within 24 h after inoculation are marked with×. Eggs that were positive in the hemagglutinationassay are indicated in grey and eggs that were HA negative are shown as not filled.

The results showed increased HA titres (within the rangeof 18–162 HA units ml−1). In the third passage, 0.2 ml ofeach individual amniotic fluid of the second passage wereinoculated into five to seven 8-day-old ECE; only two am-niotic fluids from the second passage were pooled, and thepool was inoculated, as individual volumes were too smallfor further subcultivation. Six days later, the amniotic fluidof each egg was harvested, and HA titres were determined.Some of the samples were used for further subcultivationand some were preserved at−70◦C.

In 2002, five of the preserved historical samples were cho-sen for further subcultivation: four were the third passageof individual amniotic fluids and the fifth was the third pas-sage of pooled amniotic fluids from the second passage. Foramniotic cavity inoculation, five groups of 8-day-old em-bryonated SPF eggs were inoculated (30 in total); within thegroup eggs were inoculated with 0.2 ml per specimen andcontrol (uninoculated) eggs were included. The eggs wereincubated at 35◦C. The eggs that died within 24 h werediscarded (six in total). The amniotic fluids were harvested6 days after inoculation (the fourth passage), and HA wascarried out for the individual samples separately for eachgroup. In the first, second and fifth group, the HA titres ofthe seven positive amniotic fluids ranged between 640 and2560 HA units ml−1. Amniotic fluids positive in HA wereanalysed at the molecular level. According to the sequenceanalysis, the amniotic fluids were pooled into two separatepools, named pools A and B. Pool A contained amnioticfluids from five eggs from the first and second group ofeggs inoculated with individual historical samples. Pool B

J. Ivancic et al. / Virus Research 99 (2004) 121–129 123

contained amniotic fluids from two eggs from the fifth groupof eggs inoculated with the pooled historical sample.

The mumps virus was further subcultivated on humandiploid cells MRC-5 (National Institute for BiologicalStandardization and Control, UK). Infected cells were incu-bated in MEM (AppliChem, Germany) with Hanks’ salts,neomycin (Invitrogen Life Technologies, UK) and 10%FCS (Moregate BioTech, Australia).

2.2. RNA extraction and cDNA synthesis

The total RNA was extracted as reported by Chomczyn-ski and Mackey (Chomczynski and Mackey, 1998). Briefly,300�l of solution D (4 M guanidinium thiocyanate, 25 mMsodium citrate, pH 7, 0.5% sarcosyl, 0.1 M 2-mercapto-ethanol), 0.2 M sodium acetate (1:0.1), phenol (1:1) andchloroform (1:0.4) were added to 200�l of amniotic fluid.RNA was precipitated from the aqueous phase with 1 vol-ume of isopropanol and the pellet was washed twice in300�l of 75% ethanol. Finally, the RNA pellet was dis-solved in 10�l of DNase, RNase free water (Sigma, USA).

The RNA was reversely transcribed into cDNA at 42◦C/20 min in a reaction mixture containing 2.5 U/�l MuLVreverse transcriptase, 1× PCR buffer (50 mM KCl, 1.5 mMMgCl2, 10 mM Tris–HCl, pH 9.0) 2.5�M random hex-amers as primers, 15 mM MgCl2 and 1 mM of dNTP mixin a total volume of 24�l. All the chemicals for reversetranscription were purchased from Amersham PharmaciaBiotech, Sweden.

2.3. PCR amplification

Full length or partial sequences of the NP, HN, F and SHgenes were amplified from cDNA by using specific primers(primer sequences are available upon request). Each PCRmixture included 4 U of Taq DNA polymerase, 1xPCR buffer(50 mM KCl, 1.5 mM MgCl2, 10 mM Tris-HCl, pH 9.0),20 pmol of each primer and 2 mM dNTP mix in a total vol-ume of 100�l. All the PCR chemicals were purchased fromAmersham Pharmacia Biotech, Sweden. The initial denatu-ration step was performed at 94◦C for 2 min, followed bytwo rounds of cycles (as recommended in “PCR Applica-tions Manual”, Boehringer Mannheim, Germany). The firstround of 10 cycles involved denaturation at 94◦C for 1 min,primer annealing at 50◦C for 30 s and elongation at 72◦Cfor 5 min. The second round consisted of 30 cycles withthe same thermal profile, except that the elongation for eachsuccessive cycle was prolonged by 20 s. The final elongationstep was performed at 72◦C for 10 min.

By using generated long PCR products as a templateand one of ten different primer pairs, we reamplified PCRproducts for direct sequencing. PCR reamplification wasconducted through 35 cycles in a final volume of 50�l con-taining 5�l of the long PCR reaction mix, 20 pmol of eachprimer, 1× PCR buffer, 2.1 mM MgCl2, 0.21 mM of eachof dNTP and 2.5 U Taq DNA polymerase. After an initial

denaturation step of 94◦C for 3 min, 35 cycles of 94◦Cfor 30 s, 50◦C for 50 s and 72◦C for 1 min with a terminalelongation step at 72◦C for 10 min were performed.

Amplification of PCR fragments for restriction fragmentlength polymorphism (RFLP) and GeneScan electrophoresiswas performed by using FAM fluorescent dye labeled F3forward primer and F4 reverse primer. The amplificationswere carried out after the initial denaturation step at 94◦C for3 min: 40 cycles of 94◦C for 30 s, 54◦C for 1 min and 72◦Cfor 2 min were performed, followed by a terminal elongationstep at 72◦C for 10 min. The reaction mixture containedcDNA reaction mix (denaturated at 94◦C for 3 min), 1×PCR buffer, 20 pmol of each primer, 0.21 mM dNTP and2.5 U Taq polymerase in the final volume of 50�l.

To investigate the possible presence of mumps virus vari-ant from pool A (variant A) in supernatants of MRC-5 cellculture inoculated with pool B (variant B) and reversely, weused selective nested RT-PCR amplifications with primersthat bind specifically to variant A (primers A1–A4) or vari-ant B (primers B1–B4) (primer sequences are available uponrequest). These primers were designed to differ by at leasttwo nucleotides at the 3′ end to prevent them from bindingto the cDNA of the other variant.

The first step of nested PCR for both sets of primers wasperformed in a final volume of 50�l containing the cDNAreaction mix (denaturated at 94◦C for 3 min), 20 pmol ofeach of outer primers, 1× PCR buffer, 1.25 mM MgCl2, and2.5 U Taq DNA polymerase. The amplification step con-sisted of 35 cycles and was performed as follows: denatu-ration at 94◦C for 30 s, annealing of primers at 50◦C for30 s (for primers A1 and A2) or 48◦C for 30 s (for primersB1 and B2) and extension at 72◦C for 1 min. The terminalelongation step was performed at 72◦C for 10 min.

In the second round of nested PCR, 5�l of the first re-action were amplified through 35 cycles in a final volumeof 50�l containing 20 pmol of each inner primer, 1× PCRbuffer, 2.1 mM MgCl2, 0.21 mM each of dNTP and 2.5 UTaq DNA polymerase. After an initial denaturation step of94◦C for 3 min, 30 cycles of 94◦C for 30 s, 50◦C for 30 s(for primers A3 and A4) or 49◦C for 30 s (for primers B3and B4) and 72◦C for 1 min, with a terminal elongation stepat 72◦C for 10 min, were performed.

Rigorous precautions were taken to avoid PCR productcarry-over and sample-to-sample contamination (Kwok andHiguchi, 1989; Kitchin and Bootman, 1993).

2.4. Automated DNA sequencing

The PCR products were purified from agarose gel us-ing QIAEX II Gel Extraction Kit (QIAGEN GmbH, Ger-many), and readable overlapping 600–700 bp fragmentswere directly sequenced using specific primers. Nucleotidesequences were determined using the BigDye terminatorschemistry sequencing kit (Applied Biosystems, USA) in anABI Prism 377 automatic DNA sequencer (Applied Biosys-tems, USA). The nucleotide and amino acid sequences were

124 J. Ivancic et al. / Virus Research 99 (2004) 121–129

analysed using Clone Manager Suite software (Scientific &Educational Software, USA).

Generated sequence data were deposited in NCBIdatabase under accession numbers AY376470–AY376473.

2.5. RFLP analysis

Fluorescently labeled 736 bp amplicons of the F geneFAM-F3F4 were purified from agarose gel and testedfor RFLP using the restriction endonucleaseBamHI(5′G↓GATCC3′) (Amersham Pharmacia Biotech, Sweden).FAM-F3F4 amplicons were incubated at 37◦C for 12–16 hwith 15 U of BamHI in a final volume of 25�l. Reactionproducts were analysed by GeneScan electrophoresis.

2.6. GeneScan electrophoresis

FAM-F3F4 amplicons were diluted 1:10 with deionisedwater and 1�l aliquots were mixed with 1�l GS-500TAMRA size standard (DNA fragments of known size la-beled with ABI PRISM dyeN,N,N′-tetra-methyl-6-carboxy-rhodamine (TAMRA)) (Perkin-Elmer, USA), 1�l 2×agarose loading buffer and 2�l deionised formamide. Thesamples were denaturated by heating at 90◦C for 2 min fol-lowed by cooling on ice. Electrophoresis was performed ona 4% denaturing polyacrylamide gel. Analyses of PCR prod-ucts were performed by GeneScan software (Perkin-Elmer,USA). Relative ratios of the detected virus sequences weredetermined by comparison of peak area values for each ofthe detected fragments.

2.7. Phylogenetic analysis

The SH genes of virus variants A and B were com-pared with previously published mumps virus sequences

Fig. 2. Sequence alignment of the SH gene of mumps virus variants A and B.

by ClustalX 1.81 software that uses the neighbor-joiningmethod of Saitou and Nei (Saitou and Nei, 1987). The num-ber of bootstrap trials was set to 1 000. The phylogenetictree was displayed by Njplot.

3. Results

3.1. Sequence analyses of mumps virus genes after fourpassages in ECE

Seven samples of amniotic fluids that were positive in thehemagglutination test were characterized at the molecularlevel. Sequence analyses of the SH gene showed the presenceof two different dominant mumps virus variants: five am-niotic fluids contained a variant named A and two amnioticfluids contained a variant named B (Fig. 2). The comparisonsbetween the two virus variants, based on sequencing resultsand deduced amino acid sequences of HN, F and NP protein,are presented inTable 1. andFig. 3. There was a total of

Table 1Nucleotide and amino acid differences between virus variants A and B

Gene coding region

SH HN F N

Number of nucleotides 174 1749 1617 1650Number of nucleotide

differences10 50 38 50

Nucleotide differences (%) 5.75 2.86 2.35 3.03Number of amino acids 57 582 538 549Number of amino

acid differences7 9 5 7

Amino acid differences (%) 12.28 1.55 0.93 1.28Missense to silent ratio 4 0.22 0.16 0.19

J. Ivancic et al. / Virus Research 99 (2004) 121–129 125

148 nucleotide differences (2.85%), 30 of which resulted inamino acid substitutions. The highest percentage of aminoacid substitutions (12.28%) was found between the SH pro-teins of the two variants (Table 1). The missense to silent ra-tio ranged from 0.16 (F protein) to 4.0 (SH protein) (Table 1).The comparison of deduced amino acid sequences showed

Fig. 3. Comparison of deduced amino acid sequences of the (a) NP, (b) HN and (c) F proteins of mumps virus variants A and B. Squared sequencescontain important functional or antigenic sites.

6 amino acid differences between variant A and variant B inthe C-terminal region of the NP gene (Fig. 3a). Analysis ofamino acid homology in HN protein showed one amino aciddifference in each of the three crucial HN epitopes, position287 (valine in variant A and isoleucine in variant B), po-sition 336 (leucine in variant A and serine in variant B)

126 J. Ivancic et al. / Virus Research 99 (2004) 121–129

Fig. 3. (Continued).

and position 354 (glutamine in variant A and lysine in vari-ant B) (Fig. 3b). Within the deduced amino acid sequencesof F proteins, we did not observe any differences in func-tionally important domains between variant A and variantB, except one substitution within 25 amino acids of the N-terminal region (serine in variant A and proline in variant B)(Fig. 3c).

3.2. Phylogenetic analysis

A mumps virus phylogenetic tree (Fig. 4) based on the 316nucleotide region of the SH gene (Fig. 2) was generated bythe neighbor-joining method. It shows that variant A belongsto genotype D with a very strong bootstrap probability whilevariant B does not belong to any of the established mumpsvirus genotypes.

3.3. RFLP and GeneScan electrophoresis of detectedviruses

Sequence analyses of FAM-F3F4 PCR amplicons of thetwo mumps virus variants showed that aBamHI restric-tion site is present only in the variant B sequence, posi-tion 428–433 of DNA fragment (results not shown). Afterincubation of FAM-F3F4 PCR amplicons withBamHI re-striction enzyme, GeneScan electrophoresis of the obtainedfragments showed the presence of viral sequences withouta BamHIsite in the samples of variant B but there were noviral sequences with aBamHIsite in the samples of variantA. In variant B, the ratio of the sequences with aBamHI

site to the sequences without aBamHI site is 10:1 (resultsnot shown).

3.4. Selective nested PCR amplification

In order to confirm the results of the RFLP analysis,primers specific for variant A or variant B were used. PrimersA1–A4 (specific for variant A) do not bind to variant B andprimers B1–B4 (specific for variant B) do not bind to vari-ant A. By using these selective primers, we did not detectvariant B in the supernatant of cells inoculated with viralpool A nor variant A in the supernatant of cells inoculatedwith viral pool B (results not shown).

4. Discussion

The aim of our investigation was the molecular charac-terization of a historical mumps virus isolate (an allegedindividual sample). The analysis of the results obtained bythe described molecular approach showed: (a) two mumpsvirus variants, A and B, were found in the fourth passage ofthe wild virus in the amniotic cavity of ECE; (b) variants Aand B belong to different genotypes; (c) variant A and vari-ant B differ in the HN and NP genes which code for aminoacid sequences marked by other authors (Tanabayashi et al.,1990; Kövamees et al., 1990; Yates et al., 1996; Örvell et al.,1997b; Nöjd et al., 2001; Cusi et al., 2001) as immunogenicepitopes; (d) variant B contains one or more minor variants;(e) neither the presence of variant B in samples of variant A

J. Ivancic et al. / Virus Research 99 (2004) 121–129 127

Fig. 4. Mumps virus phylogenetic tree generated by the neighbor-joining method, based on the 316 nucleotide region of the SH gene. All sequences exceptZgA Croatia 69 (variant A) and ZgB Croatia 69 (variant B) (positions marked by open arrows) were obtained from the NCBI GenBank (accession nos.:Wlz1, Z77158; Wlz2, MVW1Z2; Wlz3, MVW1Z3; Wsh1, Z77160; Wsh2, Z81005; Tay, AF142774; L-Zagreb, AJ272363; ZgB, AY376471; Hoshino,AB003414; Miyahara, D90234; Matsuyama, D90233, MP 80-J, AB003423; MP 89-K, AB003422; MP 85-S, AB003416; Urabe AM9, AF314559;Bristol1, X63713; Po10s, Y08214; Ylb 92, U35848; Goa1, AF142765; Ed 4, X63710; Ed 2, X63711; Belfast, X63709; AS 97-1, AF180375; DD 98-40,AF180378; IS 98-50, AF180381; IS 98-58, AF180384; IS 98-53, AF180382; CS 98-2, AF180377; KJ 98-29, AF180388; IS 98-48, AF180380, MP93-AK, AB003424; Cir1, AF142761; Manchester1, AF142771; Ylb 95, U35849; Stratf1, AF142773; Enders, D90231; Rubini, X72944; Kilham, X63706;JL2, AF345290; JL5, AF338106; Kent1-3, AF142767; London1, AF142769; Islip1, AF142766; Po3s, Y08212; ZgA, AY376470; MP 93-N, AB003415;Dud1-2, AF142763; Wrek1, AF142775; York1, AF142776; Kent4, AF142768; Pok1–4, AF142772; Glou1, AF142764). Branch lengths are proportionalto nucleotide changes. The numbers are bootstrap probabilities determined for 1000 iterations.

nor the presence of variant A in samples of variant B wereconfirmed using specific primers.

The SH gene of the mumps virus is the most variablepart of the mumps virus genome. Comparison of the nu-cleotide sequences of the SH gene has been used in geno-typing mumps virus strains (Örvell et al., 1997a; Jin et al.,1999). We determined that variant A belongs to genotype D,while variant B does not show a strong similarity with any ofthe established genotypes. Genotype J was not included inour phylogenetic analysis because we based our study on en-tire SH gene sequences (316 nucleotides), and the availableSH gene sequences of virus strains belonging to genotype Jare only 271 nucleotides long. However, if the phylogenetictree is based on a shorter segment of the SH gene and ifstrains belonging to genotype J are included, it is clear thatvariant B does not belong to this genotype either (data notshown).

Variants A and B differ by 5.75% at the nucleotide level(SH gene) and by 12.28% at the amino acid level (SHprotein). Complete sequences of two viral substrains thatare components of live attenuated Jeryl Lynn vaccine, pro-duced from a single clinical isolate, were recently published(Amexis et al., 2002). These substrains differ by 6.89% atthe nucleotide level (SH gene) and by 13.55% at the aminoacid level (SH protein) (Amexis et al., 2002).

The suggested level of sequence diversity used as a cut-offvalue for inter-genotype discrimination is not precisely de-fined. It has been proposed to be 8% (Wu et al., 1998)or recently 6% because of the low-end variation in geno-type B versus genotype I (Johansson et al., 2002). Thus, thedifference between variants A and B is more likely to beinter-genotypic than intra-genotypic. This is also confirmedby the phylogenetic analysis (Fig. 4). Furthermore, the in-vestigators used different numbers of nucleotides in their

128 J. Ivancic et al. / Virus Research 99 (2004) 121–129

analyses. Some analysed exactly 316 nt of SH gene (5′ non-coding region, coding region and 3′ noncoding region) (Wuet al., 1998) as we did in our phylogenetic analysis, whileothers used only 174 nt of SH gene coding region (Johanssonet al., 2002).

A heterogeneous mixture of genomic RNA (quasispecies)in mumps virus samples was confirmed by others (Afzalet al., 1993; Boriskin et al., 1993; Brown et al., 1996; Yateset al., 1996). The difference between the two variants ofJeryl Lynn vaccine, JL2 and JL5, could have originated inthe course of passaging of a common isolate, but in viewof the large number of differences between their sequences,it seems more likely that they derived from two separatewild-type strains that belonged to the same cluster (geno-type A) and originated from a common epidemiological re-gion (Afzal et al., 1993). With respect to these previouslypublished data, we investigated the possibility that our indi-vidual clinical sample also contained two virus strains.

The phylogenetic relationship illustrated inFig. 4showedthat variants A and B are clearly two separate wild-typestrains which do not belong to the same cluster. The RFLPanalysis of the samples containing variant B showed thepresence of approx. 10% of virus sequences without aBamHI site. This indicated that variant A could be presentin samples of variant B. However, the samples of variantA did not contain detectable amounts of sequences witha BamHI site. Sensitive and specific nested PCR analysesconfirmed the lack of one virus variant in the samples ofanother. With respect to these results, the samples of variantB probably contain a third virus substrain. Our starting hy-pothesis that the presence of two variants is a consequenceof natural heterogeneity was not confirmed because it is un-likely that variant B vanished from the viral pool after only4 passages in ECE if the original clinical isolate containedboth of them.

Our explanation of the results is laboratory contaminationin early passaging. Due to the lack of material, we are notable to perform the molecular analysis of samples from theearliest passages and thus cannot either confirm or reject thehypothesis.

We also compared the nucleotide sequences of the cod-ing regions of the HN, F and NP genes of variants A andB. These proteins are very important in the biology of themumps virus and for development of protective immunity tothe mumps virus. Studies with monoclonal antibodies sug-gest that the HN protein is the major target for the humoralimmune response upon mumps virus infection (Wolinskyet al., 1985). Investigations of potential neutralizing epitopesand their localization on the mumps virus HN protein haveresulted in the determination of at least three important HNpeptides (aa 264–288, 329–340 and 352–360) (Kövameeset al., 1990; Örvell et al., 1997b; Cusi et al., 2001). Com-parison of these segments between HN proteins of variantsA and B showed amino acid differences in all three epi-topes (Fig. 3b). Investigation of the antigenic variation ofthe NP protein among mumps virus strains determined the

C-terminal region as a very important part of mumps virusNP (Tanabayashi et al., 1990; Jin et al., 2000). Analysesof amino acid differences in the NP of variants A and Bshowed 6 amino acid differences in the C-terminal region(Fig. 3a). F protein analyses showed that genotype A differsmore from genotypes B, C and D than the latter genotypesdiffer between themselves (Tecle et al., 2000). However, Fproteins from different virus strains belonging to the samegenotype are relatively conserved structurally and antigeni-cally (Tecle et al., 2000). Analysis of virus variants A andB showed one difference at position 5 (serine in variant Aand proline in variant B) at the N-terminal region of the Fprotein. The first 19 amino acids of the N-terminal region ofF protein comprise a “signal peptide” that is important forF protein function (Tecle et al., 2000).

A lack of similarity between circulating wild type andvaccine mumps virus strains in recent outbreaks in differ-ent European countries has frequently been observed (Afzalet al., 1997; Cohen et al., 1999; Jin et al., 1999; Montes et al.,2002). These observations strongly suggest that the immu-nity provided by a specific strain may be ineffective againststrains from other genotypes (Nöjd et al., 2001; Montes et al.,2002; Örvell et al., 2002). Therefore, molecular analyses ofvaccine strains and circulating genotypes of mumps virus arevery important. Unfortunately, the genotypes of the mumpsvirus wild strains circulating in Croatia are unknown. Onlysporadic cases of mumps infection occur in non-vaccinatedpersons annually in Croatia (data not published). Therefore,we assume that the neutralizing epitopes of the circulatingwild type viruses and vaccine strain now in use in Croatia(L-Zagreb) do not differ, yet this remains to be proven.

Acknowledgements

This study has been carried out with financial supportfrom the Croatian Ministry of Science and Technology spe-cific TEST programme “The development of the new mumpsvaccine strain”, code no. TP-01/0021-01.

References

Afzal, M.A., Pickford, A.R., Forsey, T., Heath, A.B., Minor, P.D., 1993.The Jeryl Lynn vaccine strain of mumps virus is a mixture of twodistinct isolates. J. Gen. Virol. 74, 917–920.

Afzal, M.A., Buchanan, J., Dias, J.A., Cordeiro, M., Bentley, M.L.,Shorrock, C.A., Minor, P.D., 1997. RT-PCR based diagnosis andmolecular characterisation of mumps viruses derived from clinicalspecimens collected during the 1996 mumps outbreak in Portugal. J.Med. Virol. 52, 349–353.

Amexis, G., Rubin, S., Chizhikov, V., Pelloquin, F., Carbone, K., Chu-makov, K., 2002. Sequence diversity of Jeryl Lynn strain of mumpsvirus: quantitative mutant analysis for quality control. Virology 300,171–179.

Boriskin, Y.S., Kaptsova, T.I., Booth, J.C., 1993. Mumps virus variantsin heterogeneous mumps vaccine. Lancet 341, 318–319.

J. Ivancic et al. / Virus Research 99 (2004) 121–129 129

Brown, E.G., Dimock, K., Wright, K.E., 1996. The Urabe AM9 mumpsvaccine is a mixture of viruses differing at amino acid 355 of thehemagglutinin-neuraminidase gene with one form associated with dis-ease. J. Infect. Dis. 174, 619–622.

Carbone, K.M., Wolinsky, J.S., 2001. Mumps virus. In: Fields, B.N.,Howley, P.M., Griffin, D.E., Lamb, R.A., Martin, M.A., Roizman, B.,Straus, S.E., Knipe, D.M. (Eds.), Fields Virology. Lippincott Williams& Wilkins, Philadelphia, pp. 1381–1400.

Chomczynski, P., Mackey, K., 1998. Single-step method of total RNAisolated by acid guanidine phenol extraction. In: Celis, J.E. (Ed.), CellBiology: A Laboratory Handbook. Academic Press, New York, pp.221–224.

Cohen, B.J., Jin, L., Brown, D.W., Kitson, M., 1999. Infection with wild-type mumps virus in army recruits temporally associated with MMRvaccine. Epidemiol. Infect. 123, 251–255.

Cusi, M.G., Fischer, S., Sedlmeier, R., Valassina, M., Valensin, P.E.,Donati, M., Neubert, W.J., 2001. Localization of a new neutralizingepitope on the mumps virus hemagglutinin-neuraminidase protein.Virus Res. 74, 133–137.

Domingo, E., Martinez-Salas, E., Sobrino, F., de la Torre, J.C., Portela,A., Ortin, J., Lopez-Galindez, C., Perez-Brena, P., Villanueva, N.,Najera, R., VandePol, D., Steinhauser, D., DePolo, N., Holland, J.J.,1985. The quasispecies (extremely heterogenous) nature of viral RNAgenome populations: biological relevance—review. Gene 40, 1–8.

Jin, L., Beard, S., Brown, D.W., 1999. Genetic heterogeneity of mumpsvirus in the United Kingdom: identification of two new genotypes. J.Infect. Dis. 180, 829–833.

Jin, L., Beard, S., Hale, A., Knowles, W., Brown, D.W., 2000. Thegenomic sequence of a contemporary wild-type mumps virus strain.Virus Res. 70, 75–83.

Johansson, B., Tecle, T., Örvell, C., 2002. Proposed criteria for classifi-cation of new genotypes of mumps virus. Scand. J. Infect. Dis. 34,355–357.

Jones, L.R., Zandomeni, R., Weber, E.L., 2002. Quasispecies in the 5′untranslated genomic region of bovine viral diarrhoea virus from asingle individual. J. Gen. Virol. 83, 2161–2168.

Kitchin, P.A., Bootman, J.S., 1993. Quality control of the polymerasechain reaction. Rev. Med. Virol. 3, 107–114.

Kövamees, J., Rydbeck, R., Örvell, C., Norrby, E., 1990. Hema-gglutinin-neuraminidase (HN) amino acid alternations in neutraliza-tion escape mutants of Kilham mumps virus. Virus Res. 17, 119–129.

Kwok, S., Higuchi, R., 1989. Avoiding false positives with PCR. Nature339, 237–238.

Montes, M., Cilla, G., Artieda, J., Vincente, D., Basterretxea, M., 2002.Mumps outbreak in vaccinated children in Gipuzkoa (Basque Country).Spain. Epidemiol. Infect. 129, 551–556.

Nöjd, J., Tecle, T., Samuelsson, A., Örvell, C., 2001. Mumps virus neu-tralizing antibodies do not protect against reinfection with a heterol-ogous mumps virus genotype. Vaccine 19, 1727–1731.

Örvell, C., Kalantari, M., Johansson, B., 1997a. Characterization of fiveconserved genotypes of the mumps virus small hydrophobic (SH)protein gene. J. Gen. Virol. 78, 91–95.

Örvell, C., Alsheikhly, A.R., Kalantari, M., Johansson, B., 1997b. Charac-terization of genotype-specific epitopes of the HN protein of mumpsvirus. J. Gen. Virol. 78, 3187–3193.

Örvell, C., Tecle, T., Johansson, B., Saito, H., Samuelson, A., 2002. Anti-genic relationships between six genotypes of the small hydrophobicprotein gene of mumps virus. J. Gen. Virol. 83, 2489–2496.

Saitou, N., Nei, M., 1987. The neighbor-joining method: a new methodfor reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425.

Takahashi, M., Nakayama, T., Kashiwagi, Y., Takami, T., Sonoda, S.,Yamanaka, T., Ochiai, H., Ihara, T., Tajima, T., 2000. Single genotypeof measles virus is dominant whereas several genotypes of mumpsvirus are co-circulating. J. Med. Virol. 62, 278–285.

Tanabayashi, K., Takeuchi, K., Hishiyama, M., Yamada, A., Tsurudome,M., Ito, Y., Sugiura, A., 1990. Nucleotide sequence of the leader andnucleocapsid protein gene of mumps virus and epitope mapping withthe in vitro expressed nucleocapsid protein. Virology 177, 124–130.

Tecle, T., Johansson, B., Yun, Z., Örvell, C., 2000. Antigenic and geneticcharacterization of the fusion (F) protein of mumps virus strains. Arch.Virol. 145, 1199–1210.

Tecle, T., Böttiger, B., Örvell, C., Johansson, B., 2001. Characterization oftwo decades of temporal co-circulation of four mumps virus genotypesin Denmark: identification of a new genotype. J. Gen. Virol. 82,2675–2680.

Wolinsky, J.S., Waxham, M.N., Server, A.C., 1985. Protective effects ofglycoprotein-specific monoclonal antibodies on the course of experi-mental mumps virus meningoencephalitis. J. Virol. 53, 727–734.

Wright, K.E., Dimock, K., Brown, E.G., 2000. Biological characteristicsof genetic variants of Urabe AM9 mumps vaccine virus. Virus Res.67, 49–57.

Wu, L., Bai, Z., Li, Y., Rima, B.K., Afzal, M.A., 1998. Wild type mumpsviruses circulating in China establish a new genotype. Vaccine 16,281–285.

Yates, P.J., Afzal, M.A., Minor, P.D., 1996. Antigenic and genetic variationof the HN protein of mumps virus strains. J. Gen. Virol. 77, 2491–2497.