Investigation of Cyprinid herpesvirus-3 genetic diversity by a multi-locus variable number of tandem repeats analysis

Journal of Virological Methods 173 (2011) 320–327

Contents lists available at ScienceDirect

Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet

Investigation of Cyprinid herpesvirus-3 genetic diversity by a multi-locus variable number of tandem repeats analysis Jean-Christophe Avarre a,∗ , Jean-Paul Madeira a , Ayi Santika b , Zakki Zainun b , Marine Baud c , Joëlle Cabon c , Domenico Caruso a , Jeannette Castric c , Laurent Bigarré c , Marc Engelsma d , Maskur Maskur b a

Institut des Siences de l’Evolution de Montpellier, UMR226 IRD-CNRS-UM2, 361 Rue Jean-Franc¸ois Breton, BP5095, 34196 Montpellier Cedex 05, France Main Center for Freshwater Aquaculture Development, Jalan Selabintana 37, P.O. Box 67, Kota Sukabumi 43114, Sukabumi, West Java, Indonesia c ANSES, Unité Pathologie Virale des Poissons, Technopole Brest-Iroise, BP70, 29280 Plouzané, France d Central Veterinary Institute of Wageningen UR, P.O. Box 65, 8200 AB Lelystad, The Netherlands b

a b s t r a c t Article history: Received 3 December 2010 Received in revised form 24 February 2011 Accepted 1 March 2011 Available online 8 March 2011 Key words: Cyprinid herpesvirus-3 MLVA Diversity Epidemiology

Cyprinid herpesvirus-3 (CyHV-3), or koi herpesvirus (KHV), is responsible for high mortalities in aquaculture of both common carp (Cyprinus carpio carpio) and koi carp (Cyprinus carpio koi) worldwide. The complete genomes of three CyHV-3 isolates showed more than 99% of DNA sequence identity, with the majority of differences located in short tandem repeats, also called VNTR (variable number of tandem repeats). By targeting these variations, eight loci were selected for genotyping CyHV-3 by multiple locus VNTR analysis (MLVA). CyHV-3 strains obtained after sequential in vivo infections exhibited identical MLVA profiles, whereas samples originating from a single isolate passaged 6 and 82 times in vitro exhibited mutations in two of the eight loci, suggesting a relatively slow genetic evolution rate of the VNTRs. The method was subsequently applied on 38 samples collected in Indonesia, France and the Netherlands. Globally, the isolates grouped in two main genetic clusters, each one divided in two subgroups including either CyHV-3-U/I or CyHV3-J. Interestingly, Indonesian strains were rather distant from CyHV-3-J isolate. The results of the present study indicate that these VNTR molecular markers are efficient in estimating the genetic diversity among CyHV-3 isolates and are therefore suitable for further molecular epidemiological studies. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Common carp (Cyprinus carpio) is an important human food source and is intensively bred worldwide with 2.9 million tons produced in 2007 (http://www.fao.org). Koi (Cyprinus carpio koi), a subspecies of the common carp selected for its striking colors, has created a worldwide market for the ornamental fish hobby (Balon, 1995). Intensive culture of common carp and large-scale movements of live koi, often in the absence of health certifications, have contributed to the rapid worldwide spread of a new disease responsible for mass mortalities in both fish (Haenen et al., 2004). In Indonesia, the first outbreak of the disease was reported in 2002 from East Java (Sunarto et al., 2002). Since then, it spread rapidly throughout Java Island, Bali, the southern part of Sumatera, East Kalimantan and Central Sulawesi. The disease caused very high mortality (80–95%) to both koi and common carp with estimated

∗ Corresponding author. Tel.: +33 467166404; fax: +33 467166440. E-mail address: [email protected] (J.-C. Avarre). 0166-0934/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2011.03.002

losses of more than US$15 M as of December 2003 (Sunarto et al., 2005). Observations by transmission electron microscopy, biological properties and complete genome sequencing led to the identification of a herpesvirus as the agent responsible for the disease. The new virus, very distinct from Cyprinid herpesvirus types 1 and 2 (CyHV-1 and CyHV-2), was initially designated koi herpesvirus (KHV) after the host from which it was first isolated (Hedrick et al., 2000), and was later renamed Cyprinid herpesvirus 3 (CyHV-3). It is now classified as a member of the Alloherpesviridae family (Waltzek et al., 2005). Interestingly, the genome of CyHV-3, a linear dsDNA of about 295 kbp, is the largest within the Herpesvirales order (Aoki et al., 2007). Comparison of the full sequences of three isolates (from Japan, the United States and Israel) revealed more than 99% identity (Aoki et al., 2007). This poor genetic variation has proved to be an advantage to set up sensitive and specific diagnostic tests based on PCR (Bercovier et al., 2005; Gilad et al., 2002; Gray et al., 2002; Yuasa et al., 2005), nested PCR (El-Matbouli et al., 2007), real-time PCR (Gilad et al., 2004) or loop-mediated isothermal amplification (Gunimaladevi et al., 2004; Yoshino et al., 2006, 2009). However,

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327

it unfortunately impeded both isolate tracing and phylogeographic analyses that may yet help to understand the emergence of this virus. Recently, an assay simultaneously targeting three variable regions was tested on strains from various countries and revealed slight sequence variations, some of them correlating to the geographical origin (Kurita et al., 2009). On the other hand, a duplex PCR assay targeting two other markers concluded to the absence of relation between genotypes and their origin (Bigarré et al., 2009). Therefore, the search for polymorphic markers among the scarce variable regions of the virus is needed to clarify the spatial and temporal spread of this emerging virus. Alignment of the three complete CyHV-3 sequences available revealed that, besides nucleotide mismatches, additions or deletions at single locations, most of sequence differences occurred at the level of tandemly repeated sequences, spread all over the viral genomes and varying in copy number. Variable number of tandem repeats (VNTR) can be an important source of genetic polymorphism, as the number of tandem repeats per locus may vary dramatically between strains within a given species (van Belkum et al., 1998). Multiple-locus variable number tandem repeat analysis (MLVA) is a PCR-based genotyping method relying on the polymorphic analysis of multiple VNTR loci (Lindstedt, 2005; van Belkum, 2007). Since its first use to type bacteria, MLVA has proven to be a high-resolution method for discrimination, and is now regarded as a reference typing method for many bacterial species (Li et al., 2009; van Belkum et al., 1998). Recently, VNTR polymorphism has shown great potential for differentiating isolates of large viral genomes such as herpes simplex virus type 1 (Deback et al., 2009), human adenovirus serotype 14 (Houng et al., 2009), white spot syndrome virus (John et al., 2010) and ostreid herpesvirus 1 (Segarra et al., 2010). With the aim of progressing in the molecular epidemiology of CyHV-3, the goal of this study was to develop an MLVA procedure and evaluate its potential and resolution for typing samples of different origins.

2. Materials and methods 2.1. Viral samples Samples used in this study came either from Indonesia, France or the Netherlands (NL) (Table 1). Indonesian samples were all collected from moribund cultured common carps, in Sukabumi district, West Java, either from natural or experimental infection. Samples from France and the Netherlands originated from koi carps following importation, but their exact origin was unknown (Bigarré et al., 2009). A total of 38 samples were investigated, all consisting of genomic DNA extracted either from the gills of infected carps, from carp homogenates or from the supernatant of infected Koi Fin (KF1) cells. Two negative controls were also used: the first one consisted of DNA extracted from a koi free of CyHV-3, whereas the second one corresponded to DNA from a CyHV-1 infected koi. DNA from Indonesian samples was extracted with the DNAzol kit (Invitrogen, Life Technologies, Carlsbad, USA), according to the manufacturer’s protocol, whereas DNA from the French and Dutch samples was extracted with a DNA tissue kit (Macherey-Nagel, Düren, Germany) as previously described (Bigarré et al., 2009). The viral DNA concentration in each sample was determined by real-time PCR, using the improved Sph I-5 primers (Yuasa et al., 2005). A standard curve was established with serial dilutions of a previously purified PCR product amplified with Sph I-5 primers and quantified with a spectrophotometer (NanoDrop 1000, Thermo scientific, Wilmington, USA). Amplification reactions contained 5 ␮l of 2X Sybr Green I Master mix (Roche Applied Science, Indianapolis, USA), 0.5 ␮M of each Sph I-5 primer, and 1 ␮l of 1:10 diluted tem-

321

plate DNA in a final volume of 10 ␮l. Amplifications were carried out in a LightCycler 480 (Roche) and cycling conditions were: initial denaturation at 95 ◦ C for 5 min; 45 cycles of amplification at 95 ◦ C for 10 s, annealing at 55 ◦ C for 20 s and elongation at 72 ◦ C for 20 s with a single fluorescence measurement; melting procedure from 65 to 97 ◦ C with a heating ramp of 0.1 ◦ C per second and a continuous fluorescence acquisition. 2.2. Experimental infections Ten moribund common carps were collected in December 2009, during a mortality outbreak in Cirata lake (West Java, Indonesia). Gills from these 10 moribund fishes were taken and homogenized in phosphate saline buffer (PBS) with a ratio of 1 g of gills in 9 ml of PBS. Part of the homogenates was used for testing the presence of CyHV-3 by PCR according to Yuasa et al. (2005). The rest of the homogenates was filtrated, and filtrates were mixed together and stored at −80 ◦ C for about one month. A 1:100 dilution in PBS of this stock filtrate was used for infecting 50 healthy carps by intramuscular injection. The gills of 20 moribund fish were collected individually 10–14 days post-injection, and the gills were treated exactly as described above, except that the new stock filtrate was kept at −80 ◦ C for 15 days only. Fifty healthy carps were again infected by intramuscular injection with the filtrate diluted 1:100 in PBS, and twenty moribund fish were collected individually 10–14 days post-injection. A new gill homogenate was prepared and immediately used for injecting 50 healthy carps. Six days later, 10 injected fish were placed in cohabitation with 100 healthy carps, and 20 moribund fish were collected 7–10 days post-infection. 2.3. In silico VNTR sequence identification and selection The Tandem Repeats Finder program (Benson, 1999) was used to search for VNTR sequences on the whole CyHV-3 genome sequences (strains J [GenBank AP008984], U [GenBank DQ657948] and I [GenBank DQ177346]). The output results were filtered to include only VNTR sequences with repeat units of less than 50 nt. In order to reduce the number of candidate loci, the following criteria were applied: VNTRs should contain at least 4 repetitions with 100% sequence identity. The occurrence of VNTR loci within coding or non-coding regions was determined from the annotation of CyHV-3-J genome. 2.4. VNTR amplification and analysis For each selected VNTR sequence, oligonucleotide primers targeting conserved 5 and 3 flanking regions were designed with the Gene Runner software (Hasting Software, Inc. 1994) and tested for specificity using BLAST program on the NCBI website. Primers were selected such that their length was ≥21 nt and their melting temperature was 60 ± 1 ◦ C; they were synthesized by Eurofins-MWG-Operon. PCR amplifications were performed with the puReTaq Ready-To-Go PCR Beads kit (GE Healthcare, Chalfont St Giles, UK), according to the manufacturer’s instructions. Reactions contained 200 nM of each forward and reverse primer and 1 ␮l of template DNA, in a final volume of 25 ␮l. Amplifications were carried out in a GeneAmp PCR System 2400 (Applied Biosystems, Life Technologies, Carlsbad, USA), under the following conditions: initial denaturation at 95 ◦ C for 5 min followed by 40 cycles of 94 ◦ C for 20 s, 57 ◦ C for 30 s, and 72 ◦ C for 30 s, and a final elongation step of 7 min at 72 ◦ C. Negative controls were included in each run, and the specificity of amplifications was evaluated both with a DNA extract from a non-infected koi (Koi− ) and a DNA extract from a koi naturally infected with CyHV-1. PCR products were analyzed by gel electrophoresis with 2% (w/v) high-melt agarose (Dutscher Scientific, Brumath, France) in tris-acetate EDTA, and visualized with

322

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327

Table 1 Origin of CyHV-3 and control samples analyzed in this study. Sample name

Host

Origin

Year of collection

Reference

I 08-4 I 08-6 I 09-2 I 09-4 I 09-5 I 10-1 I 10-2 I 10-3 I 09-i1 – I 09-i 5 I 09-2i1 – I 09-2i4 I 09-c1 – I 09-c3 NL 06011512 NL 06011562-2 NL 06012956-3 NL 06012956-5 NL 06021995 NL 06022482 NL 06022615 NL 06023188 NL 06023699 NL 06023818 NL 06023826 F 07/108b F Koi 2–F Koi4 F Koi 5 F 6P F 82P KoiCyHV-1 (F 09/017)

Common carp Common carp Common carp Common carp Common carp Common carp Common carp Common carp Common carpa Common carpb Common carpc Koi Koi Koi Koi Koi Koi Koi Koi Koi Koi Koi Koi Koid Koie KF1 supernatant (6 passages) KF1 supernatant (82 passages) Koi Koi

Indonesia (Sukabumi) Indonesia (Sukabumi) Indonesia (Bandung) Indonesia (Sukabumi) Indonesia (Sukabumi) Indonesia (Sukabumi) Indonesia (Sukabumi) Indonesia (Sukabumi) Indonesia (Cirata) Indonesia (Cirata) Indonesia (Cirata) Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown France France

2008 2008 2009 2009 2009 2010 2010 2010 2009 2009 2009 2006 2006 2006 2006 2006 2006 2006 2006 2006 2006 2006 2007 2008 2008 2008 2008 2009 2009

This study This study This study This study This study This study This study This study This study This study This study Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) Bigarré et al. (2009) This study This study This study This study

a b c d e

Experimentally infected by intramuscular injection. Experimentally infected by intramuscular injection. Experimentally infected by cohabitation. Experimentally infected by inoculation. Experimentally infected by cohabitation.

SYBR Safe DNA gel stain (Invitrogen) under a blue light transilluminator. For sequencing purposes, PCR products were purified from the gels by using the Gel/PCR DNA Fragments Extraction Kit (Geneaid, Taiwan) and eluted in 20 ␮l H2 O. Sequencing was realized in both directions, by Genoscreen Company (Lille, France).

ian distance and an average link clustering method. The confidence of clusters was calculated by a multiscale bootstrap resampling (B = 10,000) and expressed as approximately unbiased (au) p-values according to Shimodaira (2002, 2004). Simpson’s index of diversity was calculated according to Hunter and Gaston (1988), using the R package untb (Hankin, 2007).

2.5. Determination of VNTR sizes by capillary electrophoresis For capillary electrophoresis, PCRs were performed in the presence of forward primers labelled at the 5 end with one of the following fluorescent dyes: ATTO425, VIC and ATTO680 (EurofinsMWG-Operon, Ebersberg, Germany). PCR conditions were the same as described above except that primer concentration was 50 nM. One ␮l of each PCR product was diluted in distilled water (1:500) and 3 ␮l of this dilution were mixed with 0.2 ␮l of an internal size standard (GeneScan Liz-600, Applied Biosystems) and 12.2 ␮l of deionized formamide (Applied Biosystems). All products were analyzed on an ABI Prism 3310 XL DNA sequencer (Applied Biosystems), according to the manufacturer’s instructions. Fragment sizing was obtained with Peak Scanner Software version 1.0 (Applied Biosystems). Allele sizes were converted into repeat numbers using the following equation: number of repeats = [fragment size (bp) − flanking region sizes (bp)]/repeat size (bp). Fragment sizing accuracy was verified by running replicates of the same samples, and calculation of repeat numbers was validated by direct sequencing of some amplicons. 2.6. Data analysis VNTR data were compiled into an Excel table and missing data were specified as “NA”, for not amplified. Dendrograms were constructed using the R package pvclust (Suzuki and Shimodaira, 2006). Hierarchical cluster analysis was computed using the Euclid-

3. Results 3.1. Identification and characterization of VNTR loci from CyHV-3 genome Analysis of the three sequenced genomes with Tandem Repeat Finder (Benson, 1999) revealed the presence of a high number of VNTRs with a repeat unit size <50, i.e. 206 for CyHV-3-J, 193 for CyHV-3-U and 196 for CyHV-3-I. Only 21 VNTR loci displayed at least 4 repetitions with 100% sequence identity and were selected for subsequent analysis. The 21 corresponding primer pairs were tested by PCR on several CyHV-3 samples that consisted of DNA extracted from infected fish. Only 8 of them reproducibly amplified the expected product, as confirmed by direct sequencing, with no or limited additional non-specific products over replicates. When tested with DNA extracted from a CyHV-3-free koi (Koi− ) or from a CyHV-1-infected koi, these 8 primer pairs showed no amplification at all (Fig. 1), indicating a strong specificity of the primers for CyHV-3. The main characteristics of the corresponding VNTR loci are listed in Table 2. The repeat unit size ranged from 3 to 12 nucleotides, with a majority of 9-mers. All the loci were located in putative coding sequences (CDS) according to the automatic annotation of CyHV-3 genomes (Aoki et al., 2007). BLAST alignments of these putative CDS revealed poor hits, mainly with proteins of unknown function from various organisms. VNTR-

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327

323

Fig. 1. PCR amplification results for the 8 VNTR loci on F Koi5 (first lane), CyHV-1 (second lane) and Koi- (third lane) DNA samples. PCR products were resolved on a 2% agarose gel in the presence of Sybr Safe. M: 100-bp molecular weight ladder. Table 2 Characteristics of the selected VNTR loci. VNTR namea

Primer sequences

VNTR-11544

GGACTTCTTCTTGGACTTGTC GTCTGCAGAGAGGATAGAGCT GAAAGAGGAGGCAGAGGAGAA GTGTGGAACTCTAGGTTCTGCAT CAACAGTACAACCACAACATCGA GGTAACATTGGCGGTAGAACTA CAAGCCCTCTGAGGTTTCAAA TCCTCATCCACTTCTCCTTCTT ACCGCTGTTAAACTCACACCTA CAGCTTTACGCGAAGAGACAA AAGAAGGGGCAAGGATATATACC GCTGAAGGAAGTTTAGATGAGATG GGTAAGTCTCCAGCCATAGCG CCTCTCAGAGCATCCTTGATAA GTTCAGATTCTGGATCAGAGGAT TCTGCCTCCTCTGTCTCTGAG

VNTR-75785 VNTR-120160 VNTR-192410 VNTR-210654 VNTR-216279 VNTR-269058 VNTR-270261 a b

Copy numberb

Unit size

Location in the genome

Strand orientation

9

4

KHVJ007, N-term

+

9

5

KHVJ52, rather C-term

+

9

5

KHVJ73, N-term

+

12

5

KHVJ113, middle

+

9

6

KHVJ126, C-term

−

9

10

KHVJ131, C-term

+

3

10

KHVJ166, middle

−

9

5

KHVJ166, N-term

−

Names were given according to the position of the first base of the first repeat on KHV-J genome (accession number AP008984). Number of complete repeat units according to the sequence of KHV-J genome.

Table 3 Polymorphism and diversity of the VNTR loci. Locus Number of repeats Number of alleles Number of alleles with frequency ≤5% Simpson’s index of diversity Number of strains

VNTR-11544

VNTR-75785

VNTR-120160

VNTR-192410

VNTR-210654

VNTR-216279

VNTR-269058

VNTR-270261

3–7 5 3

5–8 4 1

4–5 2 0

3–6 3 0

4–9 6 3

4–12 8 4

7–12 6 3

5–7 3 1

0.586

0.656

0.488

0.680

0.645

0.820

0.666

0.396

39

37

41

41

41

34

38

40

269058 and VNTR-270261 were located in the same putative CDS (Table 2).

3.3. Stability of VNTR sequences during in vitro and in vivo infections

3.2. Polymorphism of VNTR sequences

The stability of the VNTR loci was evaluated through three series of CyHV-3 samples. The first series consisted in serial propagations of the CyHV-3 isolate 07/108b in KF1 cell cultures (Bigarré et al., 2009). Comparison of the allele profiles, or haplotypes, after 6 and 82 passages (samples 6P and 82P) revealed high allele stability for 6 out of the 8 loci. Only VNTR-11544 and VNTR-216279 displayed a mutation between the two isolates. The second series of samples corresponded to four fishes that succumbed to the virus after a bath infection with isolate 07/108b passaged 6 times (F 6P) and a maintenance of 2 months at 30 ◦ C followed by 28 days at 23 ◦ C (samples F Koi2, F Koi3 and F Koi4), or after contact with these latter 18 days after the temperature switch to 23 ◦ C (F Koi5) (see Bigarré et al. (2009) for more details). The MLVA profile of these 4 samples was strictly identical to that of F 6P sample. The third batch of samples included common carps serially infected by successive intramuscular injections and cohabitation (5 samples I 08-i, 4 samples I 08-2i and 3 samples I 08-c). Of all the loci that could be amplified (87 out of 96), allele profiles of these 12 analyzed samples were strictly identical (see Supplementary Table 1). Taken together, these results indicate that at least six, if not all of the

Polymorphism of the VNTR loci was evaluated over the 38 CyHV3 samples available in this study, by calculating the number of complete repeats at each locus from sizes obtained by capillary electrophoresis. All the 8 selected loci were successfully amplified for 27 samples; for the 11 remaining, 1, 2 or 3 loci were lacking, which explains the uneven number of strains analyzed per locus (Supplementary Table S1). Results, which also include the three available sequenced genomes, show that all the VNTR loci were polymorphic, with a number of alleles varying from 2 to 8 (Table 3). With the exception of VNTR-120160 and VNTR-192410, all the loci displayed at least one rare allele with a frequency ≤5%. A total of 20 haplotypes was observed over the 41 analyzed samples. The Simpson index of diversity ranged from 0.40 for VNTR-120160 to 0.82 for VNTR-216279 (Table 3). Sample distribution according to the number of repetitions was also very different for each locus. Altogether, these data suggest that all of the VNTR loci provide relevant and complementary information, by adding a specific contribution for the discrimination of CyHV-3 strains.

324

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327

Fig. 2. Dendrogram presenting the distance analysis of the 38 studied samples and the 3 sequenced CyHV-3 strains. Clustering confidence is expressed as approximately unbiased (au) p-values according to Shimodaira (2002) and Shimodaira (2004).

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327

VNTR loci, remain stable at the time-scale of tens of viral replication cycles. 3.4. Distance analysis between the studied samples From the data acquired with the present samples and with the three genomes previously sequenced, a distance tree was constructed by hierarchical clustering with the 41 MLVA profiles obtained (Fig. 2). Isolates from France formed a separate group supported by a high au value. Five samples from the Netherlands formed a separate cluster together with CyHV-3J isolate, while six others fell in a heterogeneous group that included CyHV-3-U and -I. In this group, samples NL 06023188, NL 06021995 and NL 06023826 were very closely related while samples NL 06023818 and NL 06022482 were yet far more distant. All but two Indonesian samples formed a large group, divided into two subgroups: one consisting of isolates collected in the same locality (Sukabumi, West Java) in 2009 and 2010, and one comprising the samples subjected to serial in vivo experimental infections (with a gill homogenate of fishes collected in 2009 in Cirata, West Java) in addition to a sample collected in Sukabumi (about 50 km away) the same year (I 08-6). The two last Indonesian strains (I 084 and I 09-2) were far less related to the others; they had unique haplotypes and differed from each other by 4 alleles (Table S1). Noticeably, samples I 08-4 and I 08-6, which were both collected in Sukabumi in the beginning of 2009, shared only 3 common alleles. 4. Discussion This study presents a first attempt to describe the genetic diversity of CyHV-3 strains using short tandem repeats. Sequence comparison of the three CyHV-3 isolates that were entirely sequenced (Aoki et al., 2007) revealed that most of their differences occurred at the level of such tandem repeats. Indeed, in silico exploration of the entire genomes resulted in a surprisingly high number of tandem repeats (e.g. more than 200 for CyHV-3J). This led to investigate the potential of an MLVA approach for characterizing the genetic variability of this large viral genome. However, VNTR loci may evolve too quickly to provide reliable phylogenetic relationships among closely related strains. In many organisms, the stability of VNTRs seems to be mainly associated with the number of repeated units, their length and their purity (i.e. sequence identity), the first factor being the most critical one (Legendre et al., 2007). In viruses, little is known about the variability of short tandem repeats. Yet, results of the present study showed that CyHV-3 strains obtained after sequential in vivo infections or in vitro propagations in cell culture exhibited identical or very similar MLVA profiles, respectively. This suggests that the selected VNTR loci vary at an appropriate rate for genotyping isolates during outbreaks. A similar approach had already been successfully applied to another herpes virus, herpes simplex virus type 1 (HSV-1) (Deback et al., 2009). In that case, HSV-1 strains from clinical patients could be accurately characterized with 10 VNTR sequences. Likewise, using a single repeated mononucleotide, Houng et al. (2009) were able to follow the dynamics of transmission of a human adenovirus during an epidemic. Therefore, as for other viral models, VNTRs constitute a potential powerful tool for epidemiological studies of the transmission routes and evolution of CyHV-3. Most of the samples that were analyzed consisted of DNA that was extracted from gill homogenates of carps or koi. Quantitation of the viral load by real-time PCR revealed highly heterogeneous concentrations between samples, varying from ∼500 (sample I 092) to ∼5 × 106 (sample F 82P) CyHV-3 copies/␮l (Supplementary

325

Table 2). The absence of VNTR amplification observed for some samples could mainly be explained by low viral loads. Meanwhile, only 17 alleles over 304 amplifications could not be resolved, which further demonstrates the robustness of the method. Moreover, the lowest viral concentration that enabled the amplification of all 8 VNTR loci was ∼500 copies/␮l and corresponded to sample I 09-2. Considering that 1 ␮l of template DNA was used for VNTR amplifications, this means that the potential sensitivity of the method is approximately 500 genome equivalents. Samples with lower viral loads failed to be amplified at the majority of the loci and were not included in the present study. Among the NL samples, the molecular relationships observed with the present markers were concordant with those previously obtained by Bigarré et al. (2009) with two polymorphic markers I and II. Nonetheless, as expected, MLVA revealed more discriminative. For illustration, sample NL 06022482 had the same profile as samples NL 06021995, NL 06022615, NL 06023188 and NL 06023826 with markers I and II. In contrast, it differed from those latter at three to four loci in the present study. NL samples were obtained from commercial traders who buy fish mostly in Israel and/or Japan. Though their individual origin is unknown, it is likely that samples NL 06023699, NL 06011512, NL 06012956.5, NL 06011562.2 and NL 06012956.3 were imported from Japan while the other ones originated from Israel. Concerning the Indonesian strains, the distance analysis showed that samples I 08-4 and I 08-6, which were both collected in Sukabumi (West Java province) at the end of 2008, shared only 3 common alleles out of 8. This demonstrates that genetically distinct viral strains can coexist in a same location of Indonesia and may suggest that these strains probably originate from different introduction events. This is supported by the position of the Indonesian strains I 084 and I 09-2 on the dendrogram, which seem very distant both from all the other Indonesian strains and between each other. Besides, none of the Indonesian strains cluster with neither the J nor the U/I strains. This indicates that the Indonesian samples investigated in this study may constitute separate genotype(s) slightly distant from the J and U/I ones. This is supported by the recent results of Sunarto et al. (2011) who identified, using markers I and II, a new genetic lineage among Indonesian isolates. Among the ten molecular markers tested to date (VNTRs and markers I and II), two showed some mutations during serial passages on KF1, demonstrating a possible genetic evolution of the virus in vitro. A viral evolution in vitro was suggested by others during the selection of an attenuated strain, subsequently used for a vaccine, but, in that case, the virus was submitted to UV irradiation (Perelberg et al., 2005). It should be noticed that CyHV-3-J, -U and -I have all been serially passaged in vitro before sequencing (3, 12 and 30 times respectively) (Aoki et al., 2007), raising the possibility that some of the genetic differences observed between the isolates may have arisen in vitro. This emphasizes the importance of characterizing samples directly from fish. Surprisingly, 17 of the 21 VNTR loci initially selected were located in putative open reading frames, either in the N-terminal, central or C-terminal portion. The 8 loci that were kept for the present analysis all belonged to putative genes encoding hypothetical proteins. A blastx analysis revealed significant homologies with fragments of predicted proteins of various organisms for all the corresponding ORFs, with the exception of that containing VNTR-11544. In other biological models, it is now known that VNTR expansions or contractions in genes affect the corresponding protein function and/or expression (Gemayel et al., 2010; Li et al., 2004). In bacteria, there is growing evidence indicating that VNTRs play an important role in many aspects of adaptive behavior in relation with pathogenesis and virulence (Li et al., 2004;

326

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327

van Belkum et al., 1998). The effects of VNTR polymorphism on the virulence of CyHV-3, and of viruses in general, are still largely unknown. Yet, it is reasonable to expect that the use of such VNTR loci for the genetic characterization of CyHV-3 strains may help to establish links between genetic profiles and the biological properties of this deadly virus. This might be useful not only to get insight into the evolution and pathogenesis of CyHV-3, but also in view of establishing a vaccine by selecting appropriate strains. CyHV-3 has emerged and spread on all continents in less than 15 years due to fish trade of latently infected koi carps (Haenen et al., 2004). Carps surviving an outbreak are potential carriers of the virus for a long time and may disseminate the virus to naïve individuals in farms or in the wild at very large scales (Uchii et al., 2009). In spite of an important international effort for rapid knowledge acquisition on this new viral threat, little is known about the diversity and the origin of this virus. It is suspected to have been derived from an innocuous virus of C. carpio or of another cyprinid fish, adapting to the conditions of intensive aquaculture via increased virulence (Aoki et al., 2007). Extensive comparisons of CyHV-3 isolates will thus help testing this hypothesis and should undoubtedly bring insights into the evolutionary mechanisms that lead to the emergence of such pathogenic strains from natural reservoirs. The recent detection of highly related herpesviruses, yet distinct, from CyHV-3 (Engelsma, personal communication) may signify that numerous fish herpesviruses are yet to be discovered, especially in the context of the rapid worldwide expansion of intensive fish culture. Recently, a severe outbreak linked to a herpes-like virus has been reported in farmed sturgeon, another economically important fish species (Shchelkunov et al., 2009). It is therefore essential to understand better the transmission routes and pathogenesis of the virus. The method described in this study will contribute to this aim, by providing valuable information on the diversity and spread of viral genotypes. Acknowledgements This work was supported by the Institut de Recherche pour le Développement (IRD). We are especially grateful to Dr. Emmanuel Paradis for his precious help in data analyses under R package and his critical reading of the manuscript. This is a publication IRDDIVA-ISEM 2011-027. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jviromet.2011.03.002. References Aoki, T., Hirono, I., Kurokawa, K., Fukuda, H., Nahary, R., Eldar, A., Davison, A.J., Waltzek, T.B., Bercovier, H., Hedrick, R.P., 2007. Genome sequences of three koi herpesvirus isolates representing the expanding distribution of an emerging disease threatening koi and common carp worldwide. J. Virol. 81, 5058–5065. Balon, E.K., 1995. Origin and domestication of the wild carp Cyprinus carpio – from Roman gourmets to the swimming flowers. Aquaculture 129, 3–48. Benson, G., 1999. Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 27, 573–580. Bercovier, H., Fishman, Y., Nahary, R., Sinai, S., Zlotkin, A., Eyngor, M., Gilad, O., Eldar, A., Hedrick, R.P., 2005. Cloning of the koi herpesvirus (KHV) gene encoding thymidine kinase and its use for a highly sensitive PCR based diagnosis. BMC Microbiol. 5, 13. Bigarré, L., Baud, M., Cabon, J., Antychowicz, J., Bergmann, S.M., Engelsma, M., Pozet, F., Reichert, M., Castric, J., 2009. Differentiation between Cyprinid herpesvirus type-3 lineages using duplex PCR. J. Virol. Methods 158, 51–57. Deback, C., Boutolleau, D., Depienne, C., Luyt, C.E., Bonnafous, P., Gautheret-Dejean, A., Garrigue, I., Agut, H., 2009. Utilization of microsatellite polymorphism for differentiating herpes simplex virus type 1 strains. J. Clin. Microbiol. 47, 533–540.

El-Matbouli, M., Saleh, M., Soliman, H., 2007. Detection of cyprinid herpesvirus type 3 in goldfish cohabiting with CyHV-3-infected koi carp (Cyprinus carpio koi). Vet. Rec. 161, 792–793. Gemayel, R., Vinces, M.D., Legendre, M., Verstrepen, K.J., 2010. Variable tandem repeats accelerate evolution of coding and regulatory sequences. Annu. Rev. Genet. 44, doi:10.1146/annurev-genet-072610-155046. Gilad, O., Yun, S., Andree, K.B., Adkison, M.A., Zlotkin, A., Bercovier, H., Eldar, A., Hedrick, R.P., 2002. Initial characteristics of koi herpesvirus and development of a polymerase chain reaction assay to detect the virus in koi, Cyprinus carpio koi. Dis. Aquat. Organ. 48, 101–108. Gilad, O., Yun, S., Zagmutt-Vergara, F.J., Leutenegger, C.M., Bercovier, H., Hedrick, R.P., 2004. Concentrations of a Koi herpesvirus (KHV) in tissues of experimentally infected Cyprinus carpio koi as assessed by real-time TaqMan PCR. Dis. Aquat. Organ. 60, 179–187. Gray, W.L., Mullis, L., LaPatra, S.E., Groff, J.M., Goodwin, A., 2002. Detection of koi herpesvirus DNA in tissues of infected fish. J. Fish Dis. 25, 171–178. Gunimaladevi, I., Kono, T., Venugopal, M.N., Sakai, M., 2004. Detection of koi herpesvirus in common carp Cyprinus carpio L., by loop-mediated isothermal amplification. J. Fish Dis. 27, 583–589. Haenen, O.L.M., Way, K., Bergmann, S.M., Ariel, E., 2004. The emergence of koi herpesvirus and its significance to European aquaculture. Bull. Eur. Assoc. Fish Pathol. 24, 293–307. Hankin, R.K.S., 2007. Introducing untb, an R package for simulating ecological drift under the unified neutral theory of Biodiversity. J. Stat. Software 22, 1–15. Hedrick, R.P., Gilad, O., Yun, S., Spangenberg, J.V., Marty, G.D., Nordhausen, R.W., Kebus, M.J., Bercovier, H., Eldar, A., 2000. A herpesvirus associated with mass mortality of juvenile and adult koi, a strain of common carp. J. Aquat. Anim. Health 12, 44–57. Houng, H.S.H., Lott, L., Gong, H.P., Kuschner, R.A., Lynch, J.A., Metzgar, D., 2009. Adenovirus microsatellite reveals dynamics of transmission during a recent epidemic of human adenovirus serotype 14 infection. J. Clin. Microbiol. 47, 2243–2248. Hunter, P.R., Gaston, M.A., 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson’s index of diversity. J. Clin. Microbiol. 26, 2465–2466. John, K.R., George, M.R., Iyappan, T., Thangarani, A.J., Jeyaseelan, M.J.P., 2010. Indian isolates of white spot syndrome virus exhibit variations in their pathogenicity and genomic tandem repeats. J. Fish Dis. 33, 749–758. Kurita, J., Yuasa, K., Ito, T., Sano, M., Hedrick, R.P., Engelsma, M.Y., Haenen, O.L.M., Sunarto, A., Kholidin, E.B., Chou, H.Y., Tung, M.C., de la Pena, L., Lio-Po, G., Tu, C., Way, K., Iida, T., 2009. Molecular epidemiology of koi herpesvirus. Fish Pathol. 44, 59–66. Legendre, M., Pochet, N., Pak, T., Verstrepen, K.J., 2007. Sequence-based estimation of minisatellite and microsatellite repeat variability. Genome Res. 17, 1787–1796. Li, W.J., Raoult, D., Fournier, P.E., 2009. Bacterial strain typing in the genomic era. FEMS Microbiol. Rev. 33, 892–916. Li, Y.C., Korol, A.B., Fahima, T., Nevo, E., 2004. Microsatellites within genes: structure, function, and evolution. Mol. Biol. Evol. 21, 991–1007. Lindstedt, B.A., 2005. Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis 26, 2567–2582. Perelberg, A., Ronen, A., Hutoran, M., Smith, Y., Kotler, M., 2005. Protection of cultured Cyprinus carpio against a lethal viral disease by an attenuated virus vaccine. Vaccine 23, 3396–3403. Segarra, A., Pépin, J.F., Arzul, I., Morga, B., Faury, N., Renault, T., 2010. Detection and description of a particular ostreid herpesvirus 1 genotype associated with massive mortality outbreaks of Pacific oysters, Crassostrea gigas, in France in 2008. Virus Res. 153, 92–99. Shchelkunov, I.S., Shchelkunova, T.I., Shchelkunov, A.I., Kolbassova, Y.P., Didenko, L.V., Bykovsky, A.P., 2009. First detection of a viral agent causing disease in farmed sturgeon in Russia. Dis. Aquat. Organ. 86, 193–203. Shimodaira, H., 2002. An approximately unbiased test of phylogenetic tree selection. Syst. Biol. 51, 492–508. Shimodaira, H., 2004. Approximately unbiased tests of regions using multistepmultiscale bootstrap resampling. Ann. Stat. 32, 2616–2641. Sunarto, A., McColl, K.A., Crane, M.S., Sumiati, T., Hyatt, A.D., Barnes, A.C., Walker, P.J., 2011. Isolation and characterization of koi herpesvirus (KHV) from Indonesia: identification of a new genetic lineage. J. Fish. Dis. 34, 87–101. Sunarto, A., Rukyani, A., Itami, T., 2005. Indonesian experience on the outbreak of koi herpesvirus in koi and carp (Cyprinus carpio). Bull. Fish. Res. Agency (Suppl. 2), 15–21. Sunarto, A., Taukhid, Rukyani, A., Koesharyani, I., Supriyadi, H., Huminto, H., Agungpriyono, D.R., Pasaribu, F.H., Widodo, Herdikiawan, D., Rukmono, D., Prayitno, S.B., 2002. Field investigations on a serious disease outbreak among koi and common carp (Cyprinus carpio) in Indonesia. In: 5th Symposium on Diseases in Asian Aquaculture , Gold Cost, Australia. Suzuki, R., Shimodaira, H., 2006. Pvclust: an R package for assessing the uncertainty in hierarchical clustering. Bioinformatics 22, 1540–1542. Uchii, K., Matsui, K., Iida, T., Kawabata, Z., 2009. Distribution of the introduced cyprinid herpesvirus 3 in a wild population of common carp, Cyprinus carpio L. J. Fish Dis. 32, 857–864. van Belkum, A., 2007. Tracing isolates of bacterial species by multilocus variable number of tandem repeat analysis (MLVA). FEMS Immunol. Med. Microbiol. 49, 22–27. van Belkum, A., Scherer, S., van Alphen, L., Verbrugh, H., 1998. Short-sequence DNA repeats in prokaryotic genomes. Microbiol. Mol. Biol. Rev. 62, 275–293.

J.-C. Avarre et al. / Journal of Virological Methods 173 (2011) 320–327 Waltzek, T.B., Kelley, G.O., Stone, D.M., Way, K., Hanson, L., Fukuda, H., Hirono, I., Aoki, T., Davison, A.J., Hedrick, R.P., 2005. Koi herpesvirus represents a third cyprinid herpesvirus (CyHV-3) in the family Herpesviridae. J. Gen. Virol. 86, 1659–1667. Yoshino, M., Watari, H., Kojima, T., Ikedo, M., 2006. Sensitive and rapid detection of koi herpesvirus by LAMP method. Fish Pathol. 41, 19–27.

327

Yoshino, M., Watari, H., Kojima, T., Ikedo, M., Kurita, J., 2009. Rapid, sensitive and simple detection method for koi herpesvirus using loop-mediated isothermal amplification. Microbiol. Immunol. 53, 375–383. Yuasa, K., Sano, M., Kurita, J., Ito, T., Iida, T., 2005. Improvement of a PCR method with the Sph 1-5 primer set for the detection of koi herpesvirus (KHV). Fish Pathol. 40, 37–39.