Isolation and Molecular Detection of Bovine Parainfluenza Virus Type 3 in Cattle in Serbia

Abstract The presence of bovine parainfluenza virus type 3 (BPIV3) was examined in 119 nasal swabs collected from cattle with severe respiratory infection. All samples were conducted for virus isolation on the MDBK cell line. The cytopathic effect was observed after 48h to 72h in cells inoculated with eight samples (8/119; 6.7%). The confirmation of isolated strains of BPIV3 was done by the virus-neutralization test. In addition, all samples of bovine nasal swabs were also examined for the presence of BPIV3 virus using RT-PCR with primers specific for the part of HN gene. The presence of BPIV3 was detected in eight samples (8/119; 6.7%) that were also positive upon virus isolation. The molecular characterization based on nucleotide sequencing of the part of the HN gene showed that all BPIV3 isolates belonged to genotype C of BPIV3. They branched in one distinct cluster with three different branches, but these branches were very similar to each other (98.1% to 99.8%). Serbian BPIV3c isolates were most similar to the Chinese BPIV3c isolates SD0805, SD0809 and SD0835 (from 97.92% to 99.7%), and to South Korean (12Q061), Japanese (HS9) and American (TVMDL16 and TVMDL20) BPIV3c strains (from 97.1% to 98.8%), and distinct from American (TVMDL15and TVMDL17) and Australian (Q5592) BPI3V genotype B strains (only 79.9% to 82.3% similarity), as well as from the genotype A BPIV3 strains from different countries published in GenBank.


INTRODUCTION
Bovine parainfl uenza virus type 3 (BPIV3) is an important pathogen of the known viruses associated with bovine respiratory disease complex (BRDC), which is a major health problem of cattle worldwide. BPIV3 is one of the most common causative agents of respiratory infection of cattle and sheep, especially in youngsters in intensive and extensive livestock production. Primary infection is usually followed by secondary bacterial infections which can lead to death. Clinical signs associated with BPIV3 infections are cough, anorexia, pyrexia, nasal and ocular discharges, dyspnea and sometimes diarrhoea [1]. Severe BRDC including BPIV3 infections can even lead to cardiac complications [2]. Unfavourable environmental factors such as low temperature and inadequate nutrition may infl uence the course of the disease in cattle [3]. BPIV3 belongs to genus Respirovirus, subfamily Paramyxovirinae and family Paramyxoviridae. The genome of the virus consists of a single -stranded, non -segmented RNA molecule and a helicoidal capsid surrounded with an outer envelope [1,4]. Hemagglutinin-neuraminidase (HN) antigen of the outer envelope of parainfl uenza virus type 3 represents a tetramer consisting of a two disulfi de linked dimer. This is a multifunctional molecule with three different and crucial activities for cell infection: receptor binding activity, neuraminidase activity and fusion promotion activity [5]. Three genotypes: A (BPIV3a), B (BPIV3b) and C (BPIV3c) have been described, based on genetic and phylogenetic analysis [6]. Previously, the complete genome analysis of the BPIV3 isolates from Australia and America indicated that their viral strains were classifi ed into two distinct genotypes, BPIV3 genotype A and BPIV3 genotype B. Recently, four Chinese isolates of BPIV3 were found to be distinct from other reported BPIV3 strains and classifi ed as genotype C [1]. Currently several traditional and molecular laboratory methods are in use for diagnosis of BPIV3. The gold standard in the laboratory diagnosis of BPIV3 is the virus isolation method followed by virus-neutralization test (VN test). Some other tests can also be used, such as hemagglutination test (HA test), hemagglutination -inhibition test (HI test), ELISA and direct or indirect immunofl uorescence [7]. The molecular methods often used for diagnosis of BPIV3 infection are RT-PCR and real-time RT-PCR [8]. The aim of our study was to investigate if BPIV3 was present as an etiological agent of respiratory disorders in cattle in Serbia and to characterize (if present) the virus isolate.

Samples
In total, 119 samples of nasal swabs were collected from calves and cows with severe respiratory infection from 20 herds of intensive and small backyard livestock productions from different areas of the Republic of Serbia (mainly central parts of Serbia -surroundings of City of Belgrade (Municipalities of Obrenovac and Mladenovac), cities of Jagodina and Kruševac and Municipality of Svilajnac). Immediately after swabbing the nasal cavity of animals, the cotton part of the swabs were cut and immersed into 1.5 mL of PBS 7.2 supplemented with antibiotics and chilled on ice during transport to the laboratory. In the laboratory, the nasal swabs samples in PBS were vortexed vigorously multiple times and used immediately for testing or freezed on -20 ºC pending testing.

Isolation of virus in cell culture
For virus isolation, Madin-Darby bovine kidney -MDBK cell line (ATCC CCL-22, IZSBS, Brescia, Italy) was used. As the positive control, SF4 strain of BPIV3 (ATCC-VR 281, American Bioresearch, USA) with a titre of 6 log 10 TCID 50 /ml was used during the examinations. All samples of nasal swabs, prepared as previously described, were individually inoculated into 24 hours old and 80% confl uent monolayer of MDBK cell line in 24-well microtiter plates. Each well, after the aspiration of growing media, was inoculated with 100μl of sample and incubated for 1h at 37°C in an environment with 5% CO 2 . After one hour, 500μl of minimum essential medium (MEM, Biochrom, Germany) supplemented with 2% fetal calf serum (PAA, Austria) was added. Plates were then incubated in the above mentioned conditions and observed daily for the appearance of cytopathic effect of virus (CPE). If CPE was not observed after 7 days, the plates with the inoculated cell cultures were frozen and thawed three times and passaged for two more times at fi ve days intervals. If CPE appeared in the monolayer of MDBK cells, the cells were frozen and the virus isolates were further identifi ed [9].

Virus -neutralization test (VN -test)
Twofold serial dilutions of heat -inactivated specifi c immune serum against BPIV3 (Scientifi c Institute of Veterinary Medicine of Serbia, Belgrade, Serbia; titer 128) were incubated for 1h at 37 o C with equal volumes of 100 TCID /50 of the previously titrated supernatants from inoculated MDBK cells with CPE in the wells of microplates. SF4 strain of BPIV3 (in working dilution of 100 TCID /50 ) served as a positive control in the assay. After the end of the incubation period, the aliquots prepared from the mixture of supernatants and specifi c immune sera were transferred to the MDBK cells monolayers in 96-well microplates and incubated for three days at 37 o C [10]. The absence of CPE was considered as a positive result of virus isolate neutralization.

RT-PCR detection
The extraction of viral RNA from the nasal swab samples, prepared as previously described, was done using Pure Link RNA Mini Kit (Ambion, Life Technology, USA) according to the manufacturer's instruction. Primers for the specifi c part of hemagglutininneuraminidase (HN) gene of BPIV3 were used for one step reverse transcriptionpolymerase chain reaction (RT-PCR) (forward 5'GAATGACTCATGATAGAGGTA3' and reverse 5'AGGACAACCAGTTGTATTACAT3') [11]. For amplifi cation, One- Step RT-PCR kit (Qiagen, Germany) was used according to the manufacturer's instructions. Briefl y, the reaction was performed in a total volume of 25 μl as follows: 12 μl of nuclease free water, 5 μl of 5 x PCR buffer, 1 μl of dNTP mix (containing 10 mM of each dNTP), 1 μl of 20 μM solution of each primer, 1 μl of one step RT-PCR enzyme mix and 5 μl of RNA template. Virus positive controls and negative controls were also performed for each reaction. RT-PCR protocol comprised reverse transcription at 50 o C for 30 minute followed with initial denaturation at 93 o C for 3 minutes. After this step, PCR consisted of 35 repeated cycles of denaturation at 94 o C for 45 sec, annealing at 49 o C for 45 sec and elongation at 72 o C for 60 sec. The process was fi nalised with a fi nal elongation at 72 o C for 10 minutes. The presence of specifi c 647bp product of BPIV3 detected by agarose gel electrophoresis was considered as a positive result [11].

BPIV3 HN gene direct sequencing
PCR products of partial BPIV3 HN gene nucleotide sequences were purifi ed with QIA quick Purifi cation Kit (Qiagen, Valencia, CA, USA). The purifi ed PCR products were subsequently sequenced with Big Dye Terminator Cycle Sequencing Kit (PE Applied Biosystems, Foster City, CA, USA) using PCR primers as sequencing primers. Direct sequencing of BPIV3 isolates nucleic acid sequence was performed by Sanger sequencing method. Cycle sequencing protocol comprised of initial denaturation at 96 °C for 2 minutes, followed by 40 repeated cycles of denaturation at 96 °C for 10 sec, annealing at 50 °C for 5 sec and elongation at 60 °C for 4 minutes. The sequencing products were purifi ed by isopropanol and denatured at 95°C for 2 minutes. Sequencing reactions were analysed on the ABI Prism 310 Genetic Analyser (Applied Biosystems, USA). The obtained nucleotide sequences were aligned and compared with documented virus sequences available in the GenBank database using BLAST tool (http://www.ncbi.nlm.nih.gov/BLAST). Evolutionary analyses were conducted in MEGA 6 software [12]. The phylogenetic tree of BPIV3 virus isolates from this study and strains obtained from NCBI GenBank was constructed using Neighbor-Joining algorithm with 1000 bootstrap replicates. The evolutionary distance computed using Maximum Composite Likelihood method [13].

Virus isolation
The cytopathic effect was observed 48h -72h after inoculation into the MDBK cell line for eight out of 119 tested (8/119; 6.7%) nasal swab samples. All isolates were confi rmed as BPIV3 by virus neutralization test. Six out of 8 BPIV3 isolates were obtained from 3 to 4 month old calves and two isolates were from cows. All virus positive animals had a severe respiratory infection at the time of sampling. Details like virus isolates names, origin, time of sampling and location of positive materials are shown in Table 1. The area from where the isolates were obtained has a diameter of approximately 200 km in the central part of the Republic of Serbia.

RT-PCR
All samples of bovine nasal swabs were also examined for the presence of BPIV3 using RT-PCR methods with primers for the specifi c part of the HN gene. The presence of BPIV3 was detected in eight samples (8/119; 6.7%) that were also positive by virus isolation. This result additionally confi rms the result of virus isolation and the presence of BPIV3 in cattle population in the Republic of Serbia.

BPIV3 isolates phylogenetic analysis
The phylogenetic analysis of the HN gene nucleotide sequences showed that all eight analysed BPIV3 isolates from Serbia detected in this study belong to genotype C of BPIV3 (BPIV3c). Nucleotide sequences of 577 bp long HN gene of all 8 BPIV3 isolates were submitted to the NCBI GenBank (Accession No: KX579933 -KX579940) ( Table 1; Fig. 1). Regarding the differences that were found between Serbian isolates, they can be separated in one distinct cluster with three different branches (Fig. 1) (Fig. 1).

DISCUSSION
The presence of BPIV3 in Serbia was already known through routine diagnostic testing, for many years in the past, but the virus isolates were never fully characterized. In this study we confi rmed that BPIV3 is present as an etiological agent of respiratory disorders in cattle in Serbia. Out of 119 tested nasal swabs from animals with severe respiratory infection, reared in both intensive and backyards livestock productions, BPIV3 was detected and confi rmed, by both virus isolation and RT-PCR method, in nasal swabs of eight animals (8/119) or 6.7% of tested samples. The samples were collected from the territory of the central part of Serbia (diameter of approximately 200 km between the outermost cases), and the distribution of BPIV3 positive animals confi rmed that this virus was present on the whole tested area, and probably all over the country.
In order to detect BPIV3 presence in nasal swabs of animals with acute and severe respiratory infection, the virus isolation on MDBK cell line was used. The gold standard in laboratory diagnosis of BPIV3 is still the method of virus isolation in cell lines followed by virus neutralization test, and MDBK cell line is a very sensitive cell culture for propagation of BPIV3 [1,14]. In our study, cytopathic effects on MDBK cells were observed 48 hours and 72 hours after inoculation, in eight inoculated nasal swabs. The confi rmation of eight isolated BPIV3 was done by the virus neutralization test. Similar studies were also described in the literature. Gafer and co-authors [15] isolated three strains of BPIV3 from 205 nasal swabs of sheep and goats collected in seven different regions in Egypt, by using the same methodology as we did in the present study. They also performed the virus isolation on MDBK cell line and confi rmed the isolated strains by virus neutralization test and RT-PCR.
Having in mind that the HN protein of the outer envelope of the virus plays a crucial role for receptor binding activity, neuraminidase activity and fusion promotion activity, phylogenetic analysis of nucleotide sequences of a part of the HN gene was performed. The results of phylogenetic analysis of the 577 bp long nucleotide sequences of the HN gene showed that all eight Serbian BPIV3 isolates obtained in this study belonged to the genotype C of BPIV3 (BPIV3c). The BPIV3c strains were recently discovered in China, when Zhu and collaborators [11] found that some of the detected and phylogenetically analysed BPIV3 strains isolated during 2008 were distinct from previously characterized virus strains. Phylogenetic analysis of partial HN gene for four isolates and the complete genome for the SD0835 isolate implicated that the four Chinese BPIV3 strains were distinct from the previously reported genotype A (BPIV3a) and genotype B (BPIV3b) strains. Thus the authors suggested that those strains should be considered as genotype C BPIV3 strains (BPIV3c) [11]. A similar strategy for detection and molecular characterization of BPIV3 was reported by a number of authors [1,3,4,6,11,15,16].
Previously Horwood and co-authors [16] [17] reported complete genome sequences of the BPIV3 BN-1 strain, isolated in Japan that was phylogenetically analysed and classifi ed into genotype A of BPIV3.
All BPIV3 isolates phylogenetically analysed in this study were classifi ed as genotype C strains. The analysed isolates were obtained from animals widely distributed in different parts of central Serbia with a distance of about 200 km between the outermost cases.
In addition, BPIV3 were isolated from animals from commercial cattle farms as well as from backyard animals (one or a few animals per household). This result strongly indicated that BPIV3c strains were probably widely distributed in the whole country, and that probably the BPIV3c strains were the main virus genotype present in Serbia in this moment. Regarding the differences found between partial HN gene nucleotide sequences of Serbian isolates obtained in this study, they can be separated in one distinct cluster with three different branches, however isolates from these branches are very similar to each other (98.1% to 99.8%). It should be consider that the analysis was done on only 577 bp long nucleotide sequences. Also, it must be emphasized that there is a very limited number of BPIV3 nucleotide sequences available in GenBank. Just a few whole genome and HN gene nucleotide sequences of BPIV3 were published in NCBI GenBank. Those viruses were from China, South Korea, Japan and USA. There is no whole genome or HN gene nucleotide sequence of BPIV3 from Europe that are published in NCBI GenBank. So, it is very diffi cult to discuss the possible similarities and differences of Serbian BPIV3c isolates in their geographic surroundings. The Serbian BPIV3c isolates branched together with Chinese, South Korean, Japanese and some American BPIV3 strains. Serbian BPIV3c isolates were the most similar to Chinese BPIV3c isolates SD0805, SD0809 and SD0835 (from 97.92% to 99.7%), and then to South Korean (12Q061), Japanese (HS9) and American (TVMDL16 and TVMDL20) BPIV3c strains (from 97.1% to 98.8%). We could not fi nd any direct connection either by trade of animals or by any kind of contacts in the near past that could be the reason of virus transmission and consequently the similarity between Serbian BPIV3 strains on one side, and Chinese, South Korean and American BPIV3 strains on the other side, especially without data from other European countries, so it could be assumed that genotype C BPIV3 strains were present in Republic of Serbia for a long time.
Serbian BPIV3c isolates were found to be very distinct from American (TVMDL15 and TVMDL17) and Australian (Q5592) BPIV3 genotype B strains (only 79.9% to 82.3% similarity), as well as from the genotype A BPIV3 strains from different counties published in GenBank (similarity from 79.3% to 81.9%).
We could stress that virus isolation and RT-PCR detection with consecutive molecular typing are very suitable methods for BPIV3 detection and characterization. Due to the results from this study and considering that the presented data represents one of the fi rst reports of the molecular detection and phylogenetic analysis of BPIV3 isolates in the Republic of Serbia, more data and research is needed on virus isolation and molecular characterization of BPIV3 from cattle in north and south parts of Serbia to defi ne if BPIV3 genotype A or B strains are present in the country.