GENETIC DIVERSITY of OVINE HERPESVIRUS 2 STRAINS OBTAINED FROM MALIGNANT CATARRHAL FEVER CASES in EASTERN TURKEY

Journal Pre-proof GENETIC DIVERSITY of OVINE HERPESVIRUS 2 STRAINS OBTAINED FROM MALIGNANT CATARRHAL FEVER CASES in EASTERN TURKEY Turhan Turan, Hakan Isidan, Mustafa Ozan Atasoy, ˙Ibrahim Sozdutmaz, Hakan Bulut

PII:

S0168-1702(19)30103-0

DOI:

https://doi.org/10.1016/j.virusres.2019.197801

Reference:

VIRUS 197801

To appear in:

Virus Research

Received Date:

12 March 2019

Revised Date:

18 October 2019

Accepted Date:

23 October 2019

Please cite this article as: Turan T, Isidan H, Atasoy MO, Sozdutmaz ˙I, Bulut H, GENETIC DIVERSITY of OVINE HERPESVIRUS 2 STRAINS OBTAINED FROM MALIGNANT CATARRHAL FEVER CASES in EASTERN TURKEY, Virus Research (2019), doi: https://doi.org/10.1016/j.virusres.2019.197801

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

GENETIC DIVERSITY of OVINE HERPESVIRUS 2 STRAINS OBTAINED FROM MALIGNANT CATARRHAL FEVER CASES in EASTERN TURKEY Turhan TURAN1, Hakan ISIDAN1, Mustafa Ozan ATASOY1, İbrahim SOZDUTMAZ2, Hakan BULUT3 1

Sivas Cumhuriyet University Faculty of Veterinary Medicine, Department of Veterinary Virology,

58140, Sivas/TURKEY 2

ro of

Erciyes University Faculty of Veterinary Medicine, Department of Veterinary Virology, 38280,

Kayseri/TURKEY 3

Namik Kemal University Faculty of Veterinary Medicine, Department of Veterinary Virology, 59030,

-p

Tekirdag/TURKEY

Phone: +905365786448

Phone: +905416102995

lP

Hakan Isidan: [email protected]

re

Turhan Turan: [email protected]

Correspondent Author: Mustafa Ozan ATASOY

+905548996943

na

[email protected]

ur

ORCID: 0000-0003-0096-6297

Ibrahim Sozdutmaz: [email protected]

Jo

Phone: +90508055274

Hakan Bulut: [email protected] +905334809670

Highlights 

Genetic diversity of ovine gammaherpesvirus 2 was revealed in Eastern Turkey. 1



Glycoprotein, RTA and FGARAT coding genes were used to determine genetic variability across the genomes of Ovine gammaherpesvirus 2.



Ov9.5 gene region based grouping yields consistent results for type detection, while ORF50 and ORF27 partial gene sequences were provided sufficient data to support subgrouping.



This study proves that multi locus approach based classification is valid for the monitoring infection cycle.

ro of

Abstract

Malignant Catarrhal Fever (MCF) is a generalized, definitive lethal disease affecting the epithelial and lymphoid tissues of the respiratory and digestive tract, mainly cattle and some wild ruminants such

-p

as deer, buffalo or antelope. The sheep-related form of MCF is known to be present in Turkey and is caused by ovine herpesvirus 2 (OvHV-2). The aim of this study was to reveal the genetic diversity of

re

OvHV-2 strains obtained from MCF cases in Eastern Turkey where the livestock industry has an

lP

important impact on economic activities. For this purpose, RTA (Replication and transcription activator), FGARAT (formylglycineamide ribotide amidotransferase) and some of glycoprotein genes (Ov7, Ov8 ex2, ORF27 and Ov9.5) were investigated in blood samples from 24 cattles, clinically

na

diagnosed with MCF. Genomic data of chosen samples were furthermore used to characterize and undergo combined phylogenetic analysis to determine possible alleles and subvariants. The results

ur

showed that high level of OvHV-2 diversity existed in selected genes and strains carrying allelic variants might circulate both in two geographically distinct regions and in a region itself. Moreover, three

Jo

different OvHV-2 types and various subtypes were identified based on multi locus approach. This study provides important data to epidemiological research and thereby helps to determine the source of the virus and understand the spread of the disease.

Keywords: Malignant Catarrhal Fever, Eastern Turkey, Molecular Characterization, Glycoprotein, FGARAT, RTA

2

ro of

1. Introduction Malignant catarrhal fever (MCF) is a certain fatal disease of cattle and some wild ruminants, such as deer, buffalo or antelope, that primarily affects the lymphoid tissues and epithelial cells of the

-p

respiratory and gastrointestinal tracts (Plowright, 1990). Causative agents of this disease are basically six viruses classified under the name of the "MCF group", and four of them officially belong to the Macavirus

of

the

subfamily

Gammaherpesvirinae

according

to

ICTV

re

genus

(https://talk.ictvonline.org/taxonomy/). Members of the MCF group are alcelaphine herpesvirus 1 and

lP

2 (AlHV-1 and -2), caprine herpesvirus 2 and 3 (CpHV-2 and -3), ovine herpesvirus 2 (OvHV-2) and the ibex malignant catarrhal fever virus (Alhajri et al., 2018).

na

There are two geographically distinct forms of MCF. In Eastern and Southern Africa, where the wildebeest (Connochaetes taurinus) population is mostly concentrated, AlHV-1 is transmitted from

ur

subclinically-infected wildebeests to susceptible hosts. Other forms of this disease occur worldwide, including Turkey (Dabak & Bulut, 2003), wherever OvHV-2-infected sheep and susceptible species are

Jo

kept together. These forms are referred as the wildebeest-associated and sheep-associated forms (Russell et al., 2009). Reservoir hosts shed virus particles in nasal and ocular discharges. Clinical manifestations of MCF are high fevers, depression and lymphadenopathy; nasal and ocular secretions, corneal opacity, skin lesions and multifocal necrotic lesions of the gums, tongue and palate are typical for the head and eye form (Russell, 2013a).

3

Since complete genomic sequences of AlHV-1 and OvHV-2 have been revealed and compared, it was determined that the genomic structure and unique genes share close sequence identities to each other. These unique genes were named A1-A10 and Ov1-Ov10 for the viruses (Russell et al., 2009). Recently, gene-sparse areas located between A9-A10 and Ov9-Ov10 have been shown to encode a novel spliced gene possessing immunomodulatory properties. These genes are A9.5 and Ov9.5 and share only 33% of their nucleotide conservation. Despite the overall low conservation percentage, cysteine residues and predicted N-glycosylation sites have been shown to become strictly conserved, and these two

ro of

proteins are thought to have the same function (Russell et al., 2013b) In addition, Taus et al. (2007) demonstrated that the amino acid similarity of coding sequences between secretion-derived strains ranged between 98% and 100% for OvHV-2. Ov9.5 and the repetitive

90%, respectively (Russell et al., 2014; Taus et al., 2007).

-p

region of ORF73 are known as two exceptional genes whose segments of identity differ by 60% and 70-

re

Multiple locus-based studies have been successfully conducted to determine the sources of these viruses and to clarify the virus–infection relations. Despite limitations in the number of sequence data

lP

and the studies in restricted geographical area, multiple locus approaches have also been applied to MCF viruses in recent years. The main purpose of the research was to investigate the sequence data of six

na

gene regions of OvHV-2 samples obtained from sporadic cases in Eastern Turkey and compare possible variants amongst other isolates being reported before. For this purpose, a RTA (replication and

ur

transcription activator) and FGARAT (formylglycineamide ribonucleotide amidotransferase) encoding genes, some partial- or whole-coding sequences of glycoprotein genes (Ov7, Ov8 ex2, ORF27 and

Jo

Ov9.5) and a latency associated nuclear antigen gene (ORF73) were selected. Outputs of this research would contribute towards the limited number of epidemiological studies to determine the infection sources and to illuminate the transmission chain of OvHV-2.

2. Materials and Methods 2.1. Clinical Samples

4

In this study, twenty four blood samples had been obtained from MCF suspected animals brought to veterinary faculties of Cumhuriyet University, Sivas and Firat University, Elazig located in Eastern Turkey since 2003. All samples were tested for the OvHV-2 presence with use of primers for diagnostic PCR targeting ORF75 region (Baxter et al., 1993) and these twenty four positive samples were stored at -80ºC until use. DNA was extracted from 200 µl aliquots of samples using GF-1 Viral Nucleic Acid

ro of

Extraction Kit (Vivantis Technologies, Malaysia) according to manufacturer’s instructions.

2.2. Amplification of Selected Genes

-p

Primer pairs used in the study were designed to be nested for Ov9.5 and ORF50; semi-nested for ORF75 and conventional primers for Ov7, Ov8 ex2, ORF27 and ORF73. Nested and conventional PCR

re

primers were named as set I and set II, respectively (see Table 1).

PCR mixture was prepared as a 25 µl final volume containing 5 U of Taq DNA polymerase

lP

(Vivantis Technologies, Malaysia), 75 ng of genomic DNA, and 0.2 pmol/µL of each set of sense/antisense primers for each gene. PCR was conducted under the following conditions: for the first

na

round, 1 cycle at 94°C for 30 s; 40 cycles at 94°C for 30 s, 55°C for 30 s, and 68°C for 60 s; and a final elongation step at 68°C for 5 min. A 2 µl aliquot from the first round of PCR was used in the second

ur

reaction with same as first round, except 0.4 pmol/µL of each set of sense/antisense primers. For the conventional PCR conditions, Taq DNA Polymerase (5 U), MgSO4 (0.5 mM), dNTP (0.3 mM), forward

Jo

and reverse primers (0.3 μM each) and the DNA template (50 ng per reaction) were used. Conventional PCR reactions for each gene were conducted in order to regime described before by Doboro et al. (2016). Expected PCR amplicons were confirmed by using standard gel electrophoresis representation

under UV light. 2.3. Sequencing and Phylogenetic Analysis

5

Amplicons were selected based on yielding single, strong band for each two sets for further analysis. Selected amplicons were purified with Wizard SV Gel and PCR Clean-Up System (Promega, USA) and sequenced bidirectionally by ABI 3100 Automated Sequencer (Applied Biosystems, USA) with use of BigDye Terminator Cycle Sequencing Kit (Applied Biosystems, USA). All sequence reads flanked by PCR primers were assessed in order to give high quality of traces in each direction of the read. Output data of every samples were assembled using Geneious software v. 11.1.4 (Kearse et al., 2012).

ro of

Obtained sequence data for each sample were aligned independently using the MUSCLE (Edgar, 2004). For Ov9.5, introns and exons were determined by inspection, based on the Ov9.5*01 reference sequence which were previously described by Russell et al., (2014). Phylogenetic analyses for each gene regions were also conducted using Geneious software v. 11.1.4. For this purpose, obtained sequences

-p

from gene regions were merged with publicly available sequences in GenBank (NCBI: https://www.ncbi.nlm.nih.gov/) and phylogenetic trees were constructed using the maximum likelihood

re

method (PhyML 3.2.2 package) (Guindon et al, 2010) with TN93 substitution model (Tamura K. and

lP

Nei M., 1993). The tree was drawn to scale, with branch lengths measured in the number of substitutions per site and was bootstrapped with 500 replicates.

na

Nucleotide sequences successfully amplified for this study were deposited to GenBank database and accession numbers were given between MK336185 and MK336231 (Table 2).

ur

2.4. Assignment of Virus Alleles and Combined Phylogenetic Analysis Determining the alleles and subvariants were carried out according to the method described in

Jo

Russell et al., (2014). Briefly, aligned nucleotide and predicted amino acid sequences for each gene were evaluated and substitutions were determined by comparison of each codon in every sequence with the respective consensus sequence at that position. Major alleles were described and numbered based on non-synonymous (NS) mutations, while diversity originating from synonymous mutations (Syn) of genes were described as minor alleles and numbered accordingly. For Ov9.5 gene, major allele

6

nomenclature was based on phylogenetic classification, while each subvariants were determined according to amino acid changes in Ov9.5 coding sequence. Combined phylogenetic analysis was conducted by combining exons of Ov9.5 and ORF50 for nine of twenty four samples. The main tree topology was constructed according to variations of Ov9.5 genes. Diversity in the rest of the genes were considered to determine possible subdivisions of samples. Mutations in ORF50 genes became a primary determinant for separating identical samples based on Ov9.5 alleles, while other genes were counted on in case of available data. ORF75 sequences showed

ro of

no amino acid differences amongst samples; therefore, were discriminated from combined phylogenetic analysis. Main types were indicated with use of numbers, while possible subvariants were denoted by one additional lower cases.

3.1. Detection of MCF virus DNA in samples

-p

3. Results

re

A PCR analysis of DNA samples extracted from the blood samples was conducted by nested

lP

diagnostic PCR assays. From the 24 investigated samples, 11 were for ORF50 and 13 were for Ov9.5 genes; six for Ov7, 10 for Ov8 ex2 and 11 for ORF27 were successfully amplified with using the Set I and Set II primer sets. On the other hand, the ORF73 gene amplification trials failed and were, therefore,

na

discarded from the study.

ur

3.2. Sequence Variations in ORF50 and ORF75 Genes Of the 24 samples investigated, nine of the ORF50 and ORF75 amplicons were successfully

Jo

sequenced. Multiple sequence alignments with other published data were executed with 372 nucleotides and 124 derived amino acid sequences (306-429th positions) for ORF50, and 177 nucleotides and 59 amino acid sequences (1187-1254th positions) were for the ORF75 region. A phylogenetic analysis based on nucleotide sequences revealed that the Turkish samples displayed the closest relation with the reference strain, BJ1035 (AY839756.1), along with another UK strain (HG813090.1). These two sequences showed the closest identity with MEA-8. Moreover, the 7

phylogenetic analysis showed that all of the Turkish samples clustered under the same node shared with the ORF50*0401 allele strains consisting of the UK strain and BJ1035 (Figure 1). Sequence alignment similarity showed that identical samples existed between samples, therefore five smaller groups were generated based on amino acid variations in the ORF50 region within the Turkish samples. The most common Turkish allele of ORF50 was found in three identical samples, EA4, EA-7 and EA-20, which were grouped under the allele name ORF50*0701. Two other samples, EA14 and EA-11, which differed from ORF50*0701 by a single amino acid substitution (T417A), were

ro of

therefore classified as subtypes of ORF50*08. These two samples differed by a single synonymous difference (C-T within S362), which was used to define the allele subgroups: EA-14 was named ORF50*0801, and EA-11 was named ORF50*0802. Sample MEA-8 differed from ORF50*0701 by two amino acid changes (T417A, R335G) and was called ORF50*0901. Other alleles were assigned for

-p

MEA-15 (ORF50*1001) with three substitutions (T417A, R335G, C396G); EA-2 and EA-21

re

(ORF50*1101) were assigned with four substitutions (T417A, R335G, S402N, A248V) (Table 2). Based on the ORF75 region, eight of the nine sequences, and the reference strain BJ1035, were

lP

found to be identical, except for EA-21, as a consequence of a single syn substitution mutation (C→T substitution in S1189). Thus, EA-21 was named as ORF75*0102, while the rest of the samples were the

na

same as BJ1035 allele, ORF75*0101.

ur

3.3. Sequence analysis of Ov9.5 Gene Raw data for the Ov9.5 gene sequence from the nine samples varied between 808 bp and 958 bp

Jo

in length. Intron sequences, according to spliced cDNA sequences, were neglected; thereby, 477 nucleotides and 159 derived amino acid sequences were obtained. A phylogenetic tree of nine samples was constructed according to four exons determined in the Ov9.5 and A9.5 regions. Since a similarity matrix was computed both with and without introns, it was revealed that the nucleotide identity, including the intron regions between groups, showed around 58.49% to 85.70%. In addition, predicted spliced cDNA sequences were also highly variable; ranged between 61.08% to 99.88% identities, except for identical samples. 8

Phylogenetic analyses for the nine selected samples were conducted based on a predicted cDNA sequence of the Ov9.5 genes, with the results that were clustered into three main groups. Samples MEA8 and MEA-15 from Sivas were grouped with the Ov9.5*02 allele, while rest of the samples collected from Elazig were clustered into two different alleles: EA-2, EA-4, EA-20 and EA-21 were clustered into the Ov9.5*03 allele; EA-7 and EA-14 were clustered into the Ov9.5*01 allele (Figure 2). EA-7 and EA-14, which were identical to each other, showed the closest nucleotide identity (99.58%) to the reference sequence BJ1035 (AY839756) and clustered in the same node. These two

ro of

samples had a single amino acid substitution (H110T) and were, therefore, named *0102 after the reference sequence, *0101.

Two identical samples, EA-2 and EA-4 along with EA-20 and EA-21, were clustered with two

-p

published sequences: the Calabria strain from Italy (KX838867) and an UK strain (Ov9.5*0301; HG813102); the latter was set as a reference of allele Ov9.5*03 to properly determine the possible

re

variations. In this consideration, EA-20 was diverged from the UK strain by two amino acid residues (N99K and K101G), while EA-2, EA-4 and EA-21 had two more amino acid substitutions (S4F and

lP

N24B). Furthermore, two syn mutations were detected in EA-2 and EA-4 (C→T and G→A substitutions at the F3 and Q23 residues, respectively). Considering these NS mutations, the alleles were named as

na

follows: EA-20, Ov9.5*0302; EA-2, EA-4 and EA-21, Ov9.5*0303. On the other hand, a phylogenetic analysis revealed that these four samples had a sequence identity with the UK strain of 98.53% to

ur

99.37% and were clustered in a separate clade. Allele Ov9.5*02 included three UK strains, along with an US strain (GenBank accession no.

Jo

DQ198083), obtained from the nasal secretions of sheep. Two samples, MEA-8 and MEA-15, showed the highest identity with the UK strains, varying between 98.85% and 99.37%. Sample 8 had a single amino acid substitution (H101Y) while MEA-15 had two (H101T, L154F), both of which were also present in the US strain. Therefore, MEA-8 and MEA-15 were named *0204 and *0205, respectively. EA-11 from Elazig was also clustered together with samples from the Sivas region, and it showed the closest amino acid similarity (153 out of 159). Moreover, amino acid similarity between the EA-11 9

and UK strains belonging to both Ov9.5*02 alleles and the single strain of Ov9.5*04 allele varied between 94.96–95.59% and 94.96%, respectively. Considering the amino acid similarities, EA-11 was defined as the *0206 allele.

3.4. Variations in Ov7, ORF27 and Ov8 ex2 Glycoprotein Genes 3.4.1. Sequence analysis of Ov7 gene Six sequences were obtained from the samples, and two available gene sequences, the BJ1035

ro of

strain (AY839756) and US strain (DQ198083.1), were aligned based on the 343 bp nucleotide and 121 derived amino acid data. A comparison of all the data based on the Ov7 sequences demonstrated that the most common Turkish allele of Ov7 was found in four identical samples, EA-2, EA-4, MEA-15 and EA-21, which were identical with both the BJ1035 and US strain and grouped under the allele name

-p

ORF50*0101. MEA-8 and EA-20 showed important variations, particularly in the epitope motifs that were defined by Doboro et al. (2016). MEA-8 presented a single amino acid substitution (Y119C) that

re

was located in the E5 motif (SEDSWYYDS); it was, therefore, called Ov7*0201. EA-20 had two amino

Ov7*0301 allele (Figure 3a).

lP

acid changes located in its epitopes, S98T in E4 (DEYSDPTV) and S121D in E5 were named after the

na

3.4.2 Sequence Variations in Ov8 ex2 Gene

Multiple alignments of the nine obtained sequences and two reference gene sequences were

Jo

acid data.

ur

conducted, and a similarity matrix was constructed based on the 201 bp nucleotide and 67 derived amino

Partial nucleotide and amino acid sequence data of two references showed that the BJ1035

(AY839756) and US strain (DQ198083.1) differed from each other by a single amino acid substitution (V47A); hence, they were classified as subtypes of Ov8ex2*0101 and Ov8ex2*0201, respectively. The EA-2 and US strain shared identical amino acid sequences, however EA-21 had a single guanine – adenine syn mutation and was, therefore, named as Ov8ex2*0202 allele. On the other hand, EA-4, MEA8, EA-11, EA-14 and MEA-15 were identical to each other and closely related to the BJ1015 strain. 10

These five strains differed from the BJ1035 strain by two amino acid substitutions, B26E and V47A; the latter was in a single epitope E1 (PHDLFPPEDLTTP; Figure 3b) and was defined by Doboro et al. (2016). Such sequences were, therefore, classified as Ov8ex2*0301. Allele Ov8ex2*0401 was assigned for MEA-12 and MEA-13 with three amino acid substitutions (B26E, Q34B and V47A).

3.4.3 Sequence Variations in ORF27 Gene Multiple alignments of the five obtained sequences and two published gene sequences were conducted, and a similarity matrix was constructed based on the 843 bp nucleotide and 281 derived

ro of

amino acid data. The number of similar amino acid presented a variety between 273 and 280 out of 281 residues. A phylogenetic analysis based on the nucleotide sequence revealed that the BJ1035 strain was clustered with the EA-4, MEA-8 and EA-21, while the EA-14, MEA-15 and US strain were outside of

-p

this group (Figure 4).

The BJ1035 (AY839756) and US (DQ198083.1) strains differed by six substitutions in the amino

re

acid sequences (C112W, K264E, E272V, D275E, A276D and V278K) which were used to define the

lP

allele subgroups: ORF27*0101 and ORF27*0201, respectively. EA-4 and MEA-8 shared the closest identity with the BJ1035 strain, and each one diverged by a single amino acid substitution (V53M for EA-4; V264M for MEA-8). Therefore, EA-4 was named as ORF27*0301, and MEA-8 was named as

ur

ORF27*0501.

na

ORF27*0401. EA-21 had two amino acid substitutions (Q34H and V278K) and was called allele

On the other hand, the phylogenetic analysis showed that EA-14 and MEA-15 were discriminated

Jo

from the main group by amino acid changes. Therefore, other alleles were assigned for MEA-15 (ORF27*0601) with six substitutions (I110F, H111Y, E272V, D275E, A276D, V278K), and EA-14 (ORF27*0701) with eight substitutions (Q34H, S161T, W166G, K264E, E272V, D275E, A276D, V278K). Notably, prominent variations for these three sequences were located between the 264 and 278th positions where the “KGPECVPREQPDADV” motif existed.

11

Four predicted epitope sites of the glycoprotein gene were defined by Doboro et al. (2016); these were evaluated, and three of them (E1, E3 and E4) did not show any variation. However, MEA-14 and EA-21 showed a single substitution (G34H) in the E2 motif (QAGKAKQPPDAES).

3.5. Combined Phylogenetic Analysis According to whole genomic reference, BJ1035 (AY839756.1) was described previously as a “1a” type (Russell et al., 2014). Nucleotide alignment based on the Ov9.5 gene showed that EA-7 and EA-14 from Elazig were identical with BJ1035 and clustered in the same branch. However, these

ro of

diverged from BJ1035 based on their amino acid substitutions located in the ORF50 allele (G335R, V377A and L419P). Furthermore, EA-14 had one additional amino acid polymorphism (thr417ala) in same gene. Thus, EA-7 and EA-14 were sub-grouped as 1b and 1c in addition to BJ1035, respectively.

-p

According to the Ov9.5 sequences, samples EA-2, EA-4, EA-20 and EA-21 from Elazig were clustered under type 3 with a UK strain (HG813102.1) and an Italian strain (KX838867.1). Since EA-2

re

and EA-21 shared the same allele at ORF50 genes, they were defined as subvariant “e”. This subvariant

lP

was validated by the available data for the Ov8ex2 and Ov7 gene regions which also share the same alleles (See Figure 5; in rectangular box). On the other hand, EA-4 and EA-20 had the same Ov9.5 and ORF50 alleles; however, a partial Ov7 glycoprotein sequence existed in distinct alleles. As a result,

na

these were named 3f and 3g, respectively.

ur

MEA-8, MEA-15 and EA-11 presented the closest similarity with the US strain (DQ198083) amongst the regional samples; therefore, they were identified as type 2. Alleles based on the ORF50

Jo

sequences showed a high diversity for each sample: *0901, *1001 and *0802, respectively. This diversity was validated by the existence of glycoprotein sequence data, except for the Ov7 gene sequence in the EA-11. Thus, types and subvariants were defined as follows: MEA-8, 2e; EA-11, 2f; MEA-15, 2g. Overall results for each sample are given in Figure 5.

4. Discussion

12

The mutations and variations of viral genomes provide researchers access to characterize and classify virus strains. In case of partial genome sequence currently being studied, regions with sufficient numbers of mutations must be determined (Arens, 1999). In this study, we aimed to reveal genetic diversity from the OvHV-2 field strains in Eastern Turkey. We evaluated blood samples of 24 clinicallydiagnosed MCF cases that have been collected since 2003; these were all previously corrected with the ORF75 diagnostic primer sets (Baxter et al., 1993). The OvHV-2 reference strain, BJ1035 (AY839756.1) genome, is 135135 bp in length (Hart et al.,

ro of

2007). In order to gain insight about the level of genetic variation, we determined specific gene regions according to two recently published approaches. Russell et al. (2014) described a method in which OvHV-2 virus types and subtypes were determined through Ov9.5, ORF75 and ORF50 gene sequences encoding virion enzyme, formylglycineamide ribonucleotide amidotransferase and a replication

-p

transcription factor, respectively. On the other hand, Doboro et al. (2016) focused on the glycoprotein

re

genes (Ov8ex2, Ov7 and ORF27), whose nucleotide identities ranged between 92.5% and 99.7%, and a latency associated nuclear antigen gene, ORF73, that was relatively divergent for monitoring OvHV-2

lP

virus circulation in South Africa. Since these two research methodologies were applied together, a sufficient number of genes (approx. 2.5% of genomic data in length) for OvHV-2 virus were revealed

na

for the evaluation of genetic diversity.

Despite PCR conditions being applied, as described in published studies, only a certain number

ur

of amplicons for each gene was able to be obtained. In this study, 11 of ORF50 (45.8%), 13 of Ov9.5 (54.1%) and 24 (100.0%) of ORF75 were for Set I; six of Ov7 (25.0%), 10 of Ov8 ex2 (41.6%) and 11

Jo

of ORF27 (45.8%) for Set II were successfully amplified. DNA extracts that were unable to obtain amplicons had been studied multiple times; however, PCR reactions failing to produce amplicon DNA could be attributed to genetic variability of primer binding sites. We concluded that the Set I primers had relatively higher success rate than the Set II primers, and success rates could be improved by designing more sensitive and universal primers.

13

A sequence alignment and distance matrix of Ov9.5 genes revealed that nucleotide identity, including intron regions between groups, showed around 58.4% to 85.7%. The predicted Ov9.5 spliced cDNA sequences were also highly variable; these showed a 61.2%–87.0% nucleotide identity. Moreover, the predicted amino acid sequences presented similarity and only varied between 50 to 51 out of the 159 amino acid residues in the coding regions of all three OvHV-2 virus types of Turkish samples. Russell et al. reported that the coding regions of spliced cDNA sequences presented 62% identity, while the predicted amino acid translation of the Ov9.5 region also had high polymorphism with a 50% similarity (Russell et al., 2014). Likewise, a 46% nucleotide identity was detected in the

ro of

Southern Italy strains involved in water buffalo MCF outbreaks by Amoroso et al. (2017). The Ov9.5 data obtained from this study seemed compatible with these two published results. On the other hand, a phylogenetic analysis based on Ov9.5 gene sequence showed that samples obtained from the Sivas

-p

region (MEA- samples) were only clustered with strain BJ1035 (type 1); however, the Elazig samples (EA- samples) were distributed into separate branches (type 2 and 3). Considering such a heterogeneity

re

for the Elazig samples, we inferred that the distribution of distinct OvHV-2 virus strains might depend on different dynamics in different areas of Eastern Turkey. Dabak et al. (2003) emphasized that livestock

lP

markets where the ruminants were kept together were the one of the main sources for virus transmission in the Elazig region. Further information is needed to reveal the association between animal trading and

na

virus circulation in Eastern Turkey.

The ORF50 partial gene sequences of samples were characterized according to nucleotide and

ur

amino acid sequences, and they shared 98.92% to 99.73% nucleotide identity. We determined five amino acid substitutions (T417A, R355G, S402N, A428V, and C396G) between groups according to their

Jo

ORF50 amino acid translation residues. These amino acid substitutions were selected as the baseline for each sample to define the alleles. We divided nine samples into five alleles, ORF50*06 to *11. Notably, samples gathered in the same alleles showed significant differences according to their Ov9.5 genes, except for EA-2 and EA-21. This illustrates that the ORF50 gene provides significant data about the genetic diversity of OvHV-2 when it is collocated with an Ov9.5 gene. On the other hand, ORF75 was the most well-conserved gene between samples, and it was identical in eight out of nine samples and

14

also two strains of the ORF75*0301 allele (AY839756 and HG813096). One remaining sample, EA-21, was the most divergent sample; it showed a 99.57% identity, and divergence was created by a single nucleotide substitution. Despite any variations being not detected from the sequences in the obtained samples of this study, a V1179L substitution in the ORF75 gene was reported from the Kars region of Eastern Turkey (Yildirim et al., 2012). These findings suggested that variations ORF75 could still be considered as one of the determinants in the classification. We analysed the variations of three other glycoprotein genes along with Ov9.5 to reveal any

ro of

possible genetic diversity: Ov7, Ov8 ex2 and ORF27. Ov7 and Ov8ex2 are thought to function as cellular binding glycoproteins (Hart et al., 2007), with the potential to serve as serodiagnostic antigens and vaccine candidates (Doboro et al., 2016). Furthermore, validation of the Ov8-based ELISA assay was demonstrated by Alhajri et al. (2018). Ov8 ex2 and Ov7 showed relatively narrow ranges of

-p

variation, limited to a 1.10% and 1.49% maximum nucleotide diversity, respectively. We revealed that,

re

along with Ov8ex2, the Ov7 gene region was also relatively well-conserved and could be considered as a subject of researches in the serology-based detection. In contrast, the ORF27 was the second-most

lP

variable glycoprotein gene after Ov9.5, displaying a nucleotide identity ranging between 98.46% and 99.53% among samples. Doboro et al. (2016) detected a series of substitution and deletion mutations on the ORF27 gene, making variation ranged from 96.30% to 97.60% in its sequence identity. Considering

na

the success of the amplification, we recommend that ORF27 genes to be included in the molecular

ur

epidemiology researches for OvHV-2.

Doboro et al. (2016) implemented a B cell epitope prediction analysis in glycoprotein genes and

Jo

showed the existence of conserved B cell epitope sites ranging from one to five. In this study, epitope E5 for Ov7 gene and epitope E2 for ORF27 gene were in disagreement among the samples as a consequence of amino acid polymorphisms. In general, OvHV-2 is able to generate genetic diversity, depending on various geographical factors (Doboro et al., 2016; Suttle, 2016). These factors have also been proven capable of creating mutations in the members of gammaherpesviruses, such as equine gammaherpesvirus 2 (Thorsteinsdóttir et al., 2013) or bovine gammaherpesvirus 6 (de Oliveira et al., 2014). On the other hand, sequential disagreements in predicted B cell epitopes may affect the continuity 15

of epitopes, thereby exhibiting a diminished reactivity with antibodies (Griffiths et al., 2000). To conclude, environmental conditions might lead to genetic differences in OvHV-2 strains circulating in Eastern Turkey, and these must be considered before deciding on the rationale of designing vaccines or serodiagnostic materials. Russell et al. (2014) determined five OvHV-2 virus types according to Ov9.5 genes and classify subtypes based on a multi locus approach. In order to determine genetic variability across the genomes of samples, we also implemented a combined phylogenetic analysis to determine possible types and

ro of

subvariants. In this study, we determined eight subgroups from nine samples dispersed into three main clades. Samples MEA-8 and MEA-15 from Sivas were clustered with type 2 strains, while the Elazig samples were with type 1, type 2 and type 3. In addition, the presence of several subvariants were also identified. Notably, the Elazig samples presented more diversity in aspects of OvHV-2 virus types and

-p

subvariants than the Sivas samples. Amoroso et al. (2017) also demonstrated how to obtain genetically-

re

distinct isolates from tissue samples of water buffaloes during MCF outbreaks in two regions of Southern Italy, Campania and Calabria, using the same methodology. Similarly, homologue genes for

lP

AlHV-1, A9.5 and ORF50 were used to examine variations in wildebeest and cattle samples by Lankester et al., (2015). These results indicated that determining a polymorphism of Ov9.5 and ORF50 allele have enabled researchers to characterize OvHV-2 virus; however, evaluations of various

ur

of characterization.

na

additional glycoprotein genes, such as the ORF27, gene would greatly improve the validity and accuracy

5. Conclusion

Jo

Determining the common source of infection and distribution of OvHV-2 is the crucial step to

decrease the incidence of MCF cases worldwide. With this study, we have reasonably demonstrated the genetic diversity of OvHV-2 obtained from two regions which represented the Eastern Turkey. We yielded the most consisting results from multi-locus approach by combining Ov9.5, ORF50 and ORF27 genes. Considering the limited number of studies on the genetic characterization of MCF viruses, further research is still required to describe other possible variants circulating in different areas. Thus, we

16

believe that our study provides important data on understanding the distribution and limits of the MCF and may serve as reference for novel studies.

6. Acknowledgements This study did not receive any financial support.

7. Compliance

authors. The authors declare that they have no conflict of interest.

8. References

ro of

This article does not contain any studies with human or animal subjects performed by any of the

-p

Alhajri, S.M., Cunha, C.W., Knowles, D.P., Li, H., Taus, S., 2018. Evaluation of glycoprotein Ov8 as a potential antigen for an OvHV-2-specific diagnostic assay. PLoS One. 13(7): e02001308, 2–13.

re

https://doi.org/10.1371/journal.pone.0200130

lP

Amoroso, M.G., Galiero, G., Fusco, G., 2017. Genetic characterization of ovine herpesvirus 2 strains involved in water buffaloes malignant catarrhal fever outbreaks in Southern Italy. Vet.

na

Microbiol. 199, 31–35. https://doi.org/10.1016/j.vetmic.2016.12.020 Arens, M., 1999. Methods for subtyping and molecular comparison of human viral genomes. Clin.

ur

Microbiol. Rev. 12, 612–626.

Baxter, S.I.F., Pow, I., Bridgen, A., Reid, H.W., 1993. PCR detection of the sheep-associated agent of

Jo

malignant catarrhal fever. Arch. Virol. 132, 145–159. https://doi.org/10.1007/BF01309849

Barlow DJ, Edwards MS, Thornton JM., 1986. Continuous and discontinuous protein antigenic determinants. Nature. Aug 21-27;322(6081),747-748. https://doi.org/10.1038/322747a0 de Oliveira, CH., de Oliveira, FG., Gasparini, MR., Galinari, GC., Lima, GK., Fonseca, AA., Barbosa, JD., Barbosa-Stancioli, EF., Leite, RC., Dos Reis, JK., 2015. Bovine herpesvirus 6 in buffaloes

17

(Bubalus bulalis) from the Amazon region, Brazil. Trop. Anim Health. Prod. Feb; 47(2), 465468. https://doi.org/10.1007/s11250-014-0733-z. Dabak, M., Bulut, H., 2003. Outbreak of malignant catarrhal fever in cattle in Turkey, Vet. Rec. https://doi.org/10.1136/vr.152.8.240 Doboro, F.A., Njiro, S., Sibeko-Matjila, K., Vuuren, M. Van, 2016. Molecular analysis of South African ovine herpesvirus 2 strains based on selected glycoprotein and tegument genes. PLoS

ro of

One. 11, 123–150. https://doi.org/10.1371/journal.pone.0147019 Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32(5), 1792-1797. https://doi.org/10.1093/nar/gkh340

-p

Griffiths A.J.F., Miller JH, Suzuki D.T., Lewontin R.C., Gelbart W.M., 2000. An Introduction to Genetic Analysis, 7th edition New York: W. H. Freeman. ISBN-10: 0-7167-3520-2

re

Guindon S, Dufayard J.F., Lefort V, Anisimova M, Hordijk W, Gascuel O., 2010. New algorithms and

lP

methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst. Biol. 59(3), 307-321. doi: https://doi.org/10.1093/sysbio/syq010

na

Hart J, Ackermann M, Jayawardane G, Russell G, Haig DM, Reid H, Stewart JP., 2007. Complete sequence and analysis of the ovine herpesvirus 2 genome. J. Gen. Virol. 88(1), 28-39. doi:

ur

https://doi.org/10.1099/vir.0.82284-0

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A.,

Jo

Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., Drummond, A., 2012. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28, 1647–1649. https://doi.org/10.1093/bioinformatics/bts199

Lankester F., Lugelo A., Mnyambwa N., Ndabigaye A., Keyyu J., Kazwala R., Grant DM., Relf V., Haig DM., Cleaveland S., Russell GC., 2015. Alcelaphine Herpesvirus-1 (Malignant Catarrhal Fever Virus) in Wildebeest Placenta: Genetic Variation of ORF50 and A9.5 Alleles. PLoS One. 18

May 13;10(5), e0124121. doi: https://10.1371/journal.pone.0124121. Plowright, W., 1990. Malignant Catarrhal Fever Virus, in: Virus Infections of Ruminants. Elsevier, pp. 123–150. https://doi.org/10.1016/B978-0-444-87312-5.50023-0 Russell, G., 2013a. Malignant Catarrhal Fever. Man. Diagnostic Tests Vaccines Terr. Anim. 1–12. Russell, G.C., Scholes, S.F., Twomey, D.F., Courtenay, A.E., Grant, D.M., Lamond, B., Norris, D., Willoughby, K., Haig, D.M., Stewart, J.P., 2014. Analysis of the genetic diversity of ovine

ro of

herpesvirus 2 in samples from livestock with malignant catarrhal fever. Vet. Microbiol. 172, 63– 71. https://doi.org/10.1016/j.vetmic.2014.04.011

Russell, G.C., Stewart, J.P., Haig, D.M., 2009. Malignant catarrhal fever: A review. Vet. J.

-p

https://doi.org/10.1016/j.tvjl.2007.11.007

Russell, G.C., Todd, H., Deane, D., Percival, A., Dagleish, M.P., Haig, D.M., Stewart, J.P., 2013b. A

re

novel spliced gene in alcelaphine herpesvirus 1 encodes a glycoprotein which is secreted in vitro.

lP

J. Gen. Virol. 94, 2515–2523. https://doi.org/10.1099/vir.0.055673-0 Suttle, CA., 2016. Environmental microbiology: Viral diversity on the global stage. Nat Microbiol.

na

26;1(11), 16205. https://doi.org/10.1038/nmicrobiol.2016.205. Tamura, N and Nei, M., 1993. Estimation of the number of nucleotide substitutions in the control

ur

region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10(3), 512–526. doi:

Jo

https://doi.org/10.1093/oxfordjournals.molbev.a040023 Taus, N.S., Herndon, D.R., Traul, D.L., Stewart, J.P., Ackermann, M., Li, H., Knowles, D.P., Lewis, G.S., Brayton, K.A., 2007. Comparison of ovine herpesvirus 2 genomes isolated from domestic sheep (Ovis aries) and a clinically affected cow (Bos bovis). J. Gen. Virol. 88, 40–45. https://doi.org/10.1099/vir.0.82285-0 Thorsteinsdóttir, L., Torfason, E. G., Torsteinsdóttir, S., Svansson, V., 2013. Genetic diversity of

19

equine gammaherpesviruses (γ-EHV) and isolation of a syncytium forming EHV-2 strain from a horse in Iceland. Res. Vet. Sci. 94(1), 170–177. https://doi.org/10.1016/j.rvsc.2012.07.011 Yildirim, Y., Bilge Dağalp, S., Yilmaz, V., Faraji Majarashin, A., 2012. Molecular characterisation of ovine herpesvirus type 2 (OvHV-2) in Turkey. Acta Vet. Hung. 60, 521–527.

-p

ro of

https://doi.org/10.1556/AVet.2012.046

re

Figure 1: The phylogenetic trees of the ORF50 nucleotide sequences of OvHV-2 strains and their

lP

closely related sequences obtained from GenBank. The evolutionary history based on 372 bp nucleotide sequences was constructed using the maximum likelihood algorithm and Tamura Nei parameter correction and were bootstrapped 1000 times. (Tamura and Nei, 1993). The tree is drawn to scale, with

Jo

ur

na

branch lengths measured in the number of substitutions per site.

20

ro of -p re lP na ur

Figure 2: Phylogenetic analysis of Ov9.5 gene sequences based on 477 nucleotides coding four exons.

Jo

The evolutionary history was inferred by using the Maximum Likelihood method based on the TamuraNei model with bootsrap (1000 replicates) (Tamura and Nei, 1993). The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Different colors indicates main alleles of OvHV-2

21

ro of -p re lP

Figure 3: Alignment of the a) 121 aa residues Ov7 and b) 67 aa residues Ov8ex2 sequences aligned by

na

MUSCLE (Edgar, 2004). Conserved residues are indicated above the alignment, with those identity percentages were indicated as heatmap. Epitope prediction sites previously defined by Doboro et al.,

Jo

ur

(2016) were shown in rectangles.

22

ro of

Figure 4: Phylogenetic tree of samples based on ORF27 gene sequences. The evolutionary history was

-p

inferred by using the Maximum Likelihood method based on the Tamura-Nei model with bootsrap (1000 replicates) (Tamura and Nei, 1993). The tree is drawn to scale, with branch lengths measured in the

Jo

ur

na

lP

re

number of substitutions per site.

23

ro of -p re

Figure 5: Overall combined phylogenetic analysis results for nine samples. Main genotypes were

lP

determined according to phylogenetic tree of Ov9.5 as is shown in the left of the figure 5, whilst subgroups were further divided based on amino acid variants of Ov9.5 and ORF50 genes; additionally glycoprotein alleles in case of existing data. Alleles of set I and set II genes were shown in the right of

Jo

ur

na

the figure. Samples with identical genotypes were indicated in the rectangle.

24

25

ro of

-p

re

lP

na

ur

Jo

Set 2

Sequence (5’-3’) AAAGACACATGCATCAAACTCT GGGTAAGTACATGGTATAAAGCAG TGAAAAACTGGCCACATAAA AAGAACCCTGATAAACTCCAGA CCCCAACAAGTCAGCATTTT TCAGTCGAATGCTGTTGGAG GGACCTCTCATCTCTTCTGCAA ATGGCAAAGTCACAGGGATG TTCTGGGGTAGTGGCGAGCGAAGGCTTC AAGATAAGCACCAGTTATGCATCTGATAAA TTCTGGGGTAGTGGCGAGCGAAGGCTTC AGTCTGGGTATATGAATCCAGATGGCTCTC CACTATGCCCAACTGTATATTGC CATAAGCTAGGTGCTTGC GCTAGCACAAGGCTGGCGAGTCTAAAC TTACTCGGTTAAACACAGGAC GTATGGTGGGCATACAGAGACTAATC GCACTACACACAGCCAGGTTTTTC GTATCCTATTGTTGGTTAAAAGGTAAAGAT GGTGCTTTTACGAAGTGG

Size (bp)

Reference

954 893 (Russell et al., 2014) 600 444 422 (Baxter et al., 1993) 238 495

ro of

Set 1

Primer Name Ov9.5 L1 Ov9.5 R1 Ov9.5 L2 Ov9.5 R2 OHVorf50_F1 OHVorf50_R1 OvHV2orf50_F2 OvHV2orf50_R2 556 755 556 555 Ov7-F Ov7-R Ov8Ex2-F Ov8Ex2-R ORF27-F ORF27-R ORF73-F ORF73-R

253

(Doboro et al., 2016)

999

1490/1649

-p

Sets

Jo

ur

na

lP

re

Table 1: Primers used in this study.

26

Table 2: Table presents the amplification status for each gene from collected samples for this study. For each sample, at least three successful amplification along with Ov9.5 gene amplicon were chosen for sequencing. Positive amplifications, negative amplifications and failed sequence reads were indicated in “p”, “n” and “f”, respectively. Successfully amplified and sequenced genes were shown with their accession numbers, alleles and colored according to their proposed types. Genes Set I

EA-7 MEA-8 EA-9 EA-10 EA-11 MEA-12 MEA-13 EA-14 MEA-15 EA-16 EA-17 MEA-18 MEA-19 EA-20 EA-21

Type 1

p

p

ORF75*0101

MK336223

MK336190

MK336199

n

n

p

ORF50*0701

ORF75*0101

MK336224

MK336191

MK336200

n n

n n

p p

ORF50*0701

ORF75*0101

MK336192

MK336201

Ov9.5*0303

Ov9.5*0102 MK336225

Ov9.5*0204

ORF50*0901

ORF75*0101

MK336193

MK336202

n n

n n

p n

Ov9.5*0206

ORF50*0802

ORF75*0101

MK336227

MK336194

MK336203

f f

f f

f f

Ov9.5*0102

ORF50*0801

ORF75*0101

MK336195

MK336204

MK336228

Ov9.5*0205

n

n

n

n

n

ORF50*1001

ORF75*0101

MK336196

MK336205

n n n n

n n n n

p p p p

Ov9.5*0302

ORF50*0701

ORF75*0101

MK336230

MK336197

MK336206

Ov9.5*0303

ORF50*1101

ORF75*0101

MK336231

MK336198

MK336207

p p n

n n n

p p p

27

p

n

Ov8ex2*0201

Ov7*0101

MK336214

MK336208

n

n

Ov8ex2*0301

Ov7*0101

MK336185

MK336215

MK336209

n n

n n

n n

n

n

n

ORF27*0401

Ov8ex2*0301

Ov7*0201

MK336186

MK336216

MK336210

n n

n n

n n

n n

n

n

Ov8ex2*0301 MK336217

n

f f

f f

ORF27*0701

Ov8ex2*0301

MK336187

MK336220

ORF27*0601

Ov8ex2*0301

MK336188

MK336221

n n n n

n n n n

n n n n

n

f

n

ORF27*0501

Ov8ex2*0201

Ov7*0101

MK336189

MK336222

MK336213

n n p

n n n

n n p

n n

n n n n

Type 3

Ov7

ORF27*0301

n

n n

MK336229

Type 2

n

n n

MK336226

Jo

EA-22 EA-23 MEA-24

n

n

ORF50*1101

Set II Ov8 ex2

ro of

EA-5 EA-6

ORF27

re

EA-4

ORF73

lP

EA-3

ORF75

na

EA-2

n

Ov9.5*0303

ur

MEA-1

ORF50

-p

Ov9.5

n n n

Ov7*0101 MK336211

n n n n

Ov7*0301 MK336212

28

ro of

-p

re

lP

na

ur

Jo