Prevalence and molecular characterization of some circulating strains of the peste-des-petits-ruminants virus in Saudi Arabia between 2014–2016

Vertebrate animal study

This project was approved by the King Faisal University deanship of scientific research animal ethics committee) and approval reference number (KFU-DSR-project No: 150074).

Study population

A surveillance campaign was conducted to assess the current situation of PPRV infection among small ruminants in KSA during 2014–2016. A total of 97-PPRV outbreaks, including 75 in sheep and 22 in goats, were observed during the duration of our surveillance (Table 1). The average size of these herds is 10–20 animals. We used the following criteria for animal sampling. First, we tried to collect specimens from at least 10% per each of the investigated flocks. However, the number of sampled animals was primarily affected by the consent from owners. Second, we selected the most severely affected animals that showed clinical signs of infection, specifically, respiratory, enteric, and ocular manifestations. We applied the purposeful sampling approach, as previously described (Patton, 2002). Specimens from 223 animals from confirmed PPRV outbreaks (including oral, nasal and lachrymal swabs, as well as tissues including lung, liver, spleen, lymph nodes and intestines) were collected from flocks of small ruminants across the KSA between Jan 2014 and Oct 2016 (Fig. 1). We also sampled some outwardly appearing healthy animals.

Tissue specimens

Tissue specimens were collected from selected animals that were showing typical clinical signs of PPRV during the necropsy examination during each outbreak. We collected ten grams from each organ that showed gross pathological changes (lung, liver, spleen, intestine, and some regional lymph nodes,especially the mesenteric and retropharyngeal). Samples were collected under complete aseptic conditions. We prepared 10% tissue suspensions for each tissue, as previously described, with some modifications (Troung et al., 2014). Briefly, one gram of tissue specimen was mixed with 9 ml of PBS. Sterile sand was added, and the tissue was triturated using a sterile mortar and pestle. The tissue suspensions obtained were centrifuged for 10 min at 4 °C in a cooled centrifuge. We then collected the tissue supernatants and stored them at −80 °C for further use.

Swabs

A total of 23 swabs (oral, nasal, and lachrymal) were collected from several suspected PPRV outbreaks. Swabs were obtained as previously described, with some modifications (Troung et al., 2014). The collected swabs were transferred to a viral transport medium containing Dulbecco’s Modified Eagle Medium (DMEM; Gibco™) that included 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA) and an antibiotic cocktail (penicillin, 10,000 units/mL, and streptomycin, 10,000 g/mL). Swabs were vortexed and centrifuged at 5000 RPM for 10 min at 4 °C. The supernatants were collected and stored at (−80 °C) for further testing.

RNA extraction and reverse transcription

Total viral RNA was extracted from both swabs and tissue specimens using a viral RNA extraction kit (RNeasy Mini Kit, QIAGEN, cat. nos. 74104 and 74106). RNA extraction was performed according to the manufacturer’s instructions, as previously described elsewhere (Kumar et al., 2014). The eluted viral RNAs were stored at −80 °C for further testing.

Detection of PPRV by TaqMan real-time RT-PCR assay (qPCR)

The qPCR amplification of the PPRV genome was carried out by using a Pestes des Petits Ruminants (PPRV) ©2013 Genekam Biotechnology AG, Ref; K229, Germany). Amplification was performed using qPCR (Light cycler 2.0, Serial No: 1416238, Roche, Switzerland). Each reaction of 20 µl contained 2.4 µl of Mg CL solution, 2 µl of Roche Master (LightCycler Fast Start DNA Master HybProbe Roche, Germany, Cat. No. 03 003 248 001), 2 µl of reagent mix (specific reagent containing oligonucleotides and probes), 2 µl of IC mix (reagent containing oligonucleotides, probes, and cDNAs), and 6.6 µl of PCR-grade water. A 15-µl aliquot of each reaction mixture was added to light cycler capillaries. Finally, 5 µl of the cDNA was added to the light cycler capillaries. The kit’s amplification conditions were applied to test for the presence of the PPRV genome in the samples.

We used one set of oligonucleotides that amplify the partial PPRV-F gene as previously described elsewhere (Ozkul et al., 2002) (the PPRV-F, 5′-AGTACAAAAGATTGCTGAT- CACAGT-3′, and the PPRV-R, 5′-GGGTCTCGAAGGCTAGGCCCGAATA-3′).

Detection of PPRV by classical PCR assay

A conventional PCR assay was used to amplify the partial PPRV-F gene using the oligonucleotides as previously described (Ozkul et al., 2002); these oligonucleotides are listed above. The first-strand (cDNA) synthesis was performed according to the protocol of the kit (Roche, cat. No. 04 379 012 001). The PCR amplification conditions were as follows: denaturation (1 min at 95 °C), annealing (1 min at 60 °C), and extension (1 min at 72° C). These steps were repeated for 40 cycles, followed by a final extension at 72  °C for 10 min. We used both negative (non-template DNA), and positive (Nigeria 75/1strain) controls alongside each reaction. The RT-PCR products were resolved in 1% agarose gels containing SYBR^® Safe DNA Gel Stain (Life Technologies). The amplified DNA fragments were visualized under ultraviolet light, and photographs were captured with a gel documentation system (Bio-Rad Laboratories, Inc., Hercules, California, USA). The target amplified partial PPRV-F gene bands were excised from the gel and purified using a QIAquick Gel Extraction Kit (Cat No/ID: 28704) according to the protocol of the kit manufacturer. The purified reactions were eluted in a 50-µl elution buffer.

Sanger method of sequencing and sequencing analysis

We selected some RT-PCR-positive specimens that were collected from animals in Hafar Al-Batin (a region located along the border of Saudi Arabia and other Gulf countries such as Kuwait). We sequenced these specimens by the Sanger method to confirm the identity of the amplicons. The purified PCR products were sequenced in both directions using the original oligonucleotides used in the PCR amplification. We assembled the obtained sequences into one contig and performed a nucleotide BLAST through NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Web&PAGE_TYPE=BlastHome). The obtained sequences were aligned and compared with the PPRV control vaccine strain, as well as the other PPRV genome sequences available in GenBank.

Next-generation sequencing (NGS)

To obtain the full-length genome sequence of the PPRV, we used another positive PPRV specimen from sheep lung tissue that was isolated from Riyadh and sequenced it by the NGS approach described below. Briefly, the short ds-cDNA fragments were joined by specific adaptor sequences. The desired amplicons were obtained using agarose gel electrophoresis. Then, qPCR amplification of the obtained RNA libraries was conducted. Quantification of these libraries was performed by the qPCR. Reactions were performed according to the kit manufacturers’ instructions (San Diego, CA, USA). Further qualification was conducted using a Bioanalyzer (Agilent2100). The cDNA library was sequenced using the HiSeqTM 2000 platform (Illumina; Macrogen Corp., Maryland, USA). The quality of the sequences was checked by the FastQC (v0.10.0) program (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Data analysis was conducted by filtering the raw sequences and removing scrambled sequences using Trimmomatic software (Bolger, Lohse & Usadel, 2014). Trimmed reads were then assembled using the Trinity program, as previously described (Grabherr et al., 2011). All obtained sequences were assembled into one contig, and the expression level of each contig was determined by the fragments/ kilobase. For annotation, the consensus sequences were searched against the NCBI-Nucleotide (NT) database using blastn (2.2.26+version).

PPRV isolation

PPRV was isolated using Vero cell lines (ATCC, CCL-81), as previously described (Hegde et al, 2009). We selected ten samples from seven laboratory-confirmed animals for the purpose of PPRV isolation. These samples included six tissue suspensions (lung, liver, spleen) and four swabs (oral and lachrymal). In some cases, we used swabs and one tissue from one animal. The rationale behind this was the presence of some pathological changes in this organ suggestive for the PPRV infection. These specimens were selected based on their low Ct values during the qPCR testing, which represented the high RNA concentration in these specimens. We used 100 µl of the tested samples (tissue suspension or swab) to inoculate 90% confluent cells. After one hour, the virus inoculum was removed, and cells were washed three times with PBS. This was followed by adding fresh DMEM maintenance media that contained fetal bovine serum (FBS; Invitrogen, Carlsbad, CA). The inoculated cells were incubated at 37 °C and 5% CO_2. Cells were monitored on a daily basis under a light microscope for the development of cytopathic effects (CPE) for up to 7 days post-infection. The PPRV vaccine (Nigeria 75/1) strain was obtained from the Veterinary Vaccines Production Center in Riyadh, KSA. The lyophilized vaccine was diluted in 1 ml of DMEM, and 100 µl of the diluted vaccine was inoculated into the Vero cell culture flasks and monitored as described above.

Sequencing and phylogenetic analyses

The phylogenetic trees (the maximum likelihood) were constructed based on the obtained sequences of the individual PPRV partial genes (F, N, M, H, L, and P). We selected some sequences representing a high degree of identity to our sequences and representing all lineages to build up our trees. Multiple alignments of these sequences with other sequences in Genebank were made using the Mega-7 package. The phylogeny was conducted using the best-fit model determination, and at least 1,000 bootstraps replicate for the maximum likelihood method, as previously described (Kumar et al., 2014).

Statistical analysis

We conducted both a descriptive and inferential statistical analysis for our samples. We used Chi-square to test the significance of the difference between frequencies. Descriptive statistical analysis performed to analyses the basic demographics. A p-value cut off point of 0.05 at 95% CI used to determine statistical significance. We used the Statistical Packages for Social Sciences [SPSS] version 21.

Clinical and necropsy findings

We conducted molecular surveillance of PPRV among small ruminant flocks across the KSA from Jan 2014 to Oct 2016. A total of 223 samples (tissues and swabs) representing 97 suspected PPRV outbreaks were tested by qPCR (Table 1). A KSA map that shows the geographical distribution of the alleged PPRV outbreaks in 10 different regions across the KSA is shown in Fig. 1 (Riyadh, Dammam, Al-Hasa, Hafar Al-Batin, Qasseem, Arar, Tabuk, Madinah, Taif, and Asir). The molecular prevalence of PPRV was mainly based on the testing of 223 samples that represented the 97 outbreaks (Table 1). The observed PPRV outbreaks in small ruminant animals were associated with a wide range of clinical signs. These clinical signs included coughing, sneezing, and nasal discharge (Fig. 2A). Some animals developed erosion and ulceration in and around the corners of the mouth (Fig. 2B). Some animals showed conjunctivitis, excessive tears, lacrimation, corneal opacity, and discharge from one eye. This discharge and conjunctivitis may progress to affect both eyes within 2–3 days of infection onset (Figs. 2C and 2D). Some animals showed ocular infections that progressed to complete blindness one week post-infection (Figs. 2E and 2F). Additionally, some animals showed arching of the back, tails that were soiled with fecal matter, diarrhea, emaciation, and ultimately death at the end of the course of infection (Figs. 2G and 2H). Necropsy examination of some of the PPRV-infected animals revealed congestion and consolidation of the lungs (Fig. 3).

Real-time and conventional PCR for PPRV detection

We also conducted molecular surveillance to evaluate the PPRV status in small ruminants across the KSA between Jan 2014 and Oct 2016. The same 97-suspected PPRV outbreaks were investigated by RT-PCR. A total of 223 specimens representing 97 outbreaks were tested. Our results show that only 27.8% of the suspected outbreaks were PPRV-positive, with 70% of the tested outbreaks being negative. Meanwhile, 29.1% of sheep and 24% of goats were PPRV genome-positive (Table 1). Meanwhile, 11 of 13 sheep swabs and 8 of 10 goat swabs were positive for the PPRV genome (Table 1). The overall prevalence of PPRV among the tested specimens (tissues and swabs) was 24.8% and 25.8% in sheep and goats, respectively (Table 1).

Six out of the ten regions reported positive PPRV outbreaks (Table 1). The total number of outbreaks in sheep was much higher than that in goat (p-value <  0.05). Meanwhile, the prevalence of the PPRV outbreaks was much higher in sheep during 2014 than 2015 in various locations (Table 1). The overall prevalence of the suspected PPRV outbreaks was much higher in small ruminants in the Riyadh region compared to other regions in the Kingdom (Table 1).

Virus isolation

We used Vero cell lines to propagate and isolate the currently circulating PPRV strains in the KSA. We then selected ten samples from seven of the outbreaks to isolate the PPRV. These samples included swabs and various tissues collected from positive PPRV infected animals. Successful isolation of the PPRV was assessed by the appearance of CPE in the inoculated cells. The presence of the virus was then confirmed by qPCR after three subsequent passages. PPRV-induced CPE in the inoculated Vero cells included rounding of the cells, detachment of the cells from the confluent monolayer sheet, floating of the cells in the cell culture supernatant, and cell death. PPRV isolation was successful for only two out of the ten samples tested. Specimens that did not show CPE were subcultured in parallel for three subsequent passages as well. The first sample was obtained from an RT-PCR-positive sheep lung that was collected from a small ruminant herd in the Riyadh region. The second sample was obtained from a lachrymal swab of a PPRV-positive sheep herd that was also in the Riyadh region (Table 1). The positive lung specimens induced CPE in the three subsequent passages, starting from the first passage. The Ct values for these three passages were 31, 32, and 34, respectively. Furthermore, the positive lachrymal swab induced CPE in the three subsequent passages, with the Ct values of those passages being 32, 34, and 38. We included a negative control (PBS-inoculated) cell culture as well as a positive control (PPRV vaccine Nig-75). The negative control-infected cells did not show any CPE and were RT-PCR-negative in the three subsequent passages. However, the positive control inoculated cells showed CPE in the inoculated cells. The Ct values of the positive control inoculated groups were 34, 33, and 34, respectively (Table 1). The trend of lowering the Ct values with the subsequent passage may be due to the cell death of the virus-infected cells.

Phylogenetic analysis

A portion of the samples that were positive for the PPRV genome, according to qPCR, was tested by agarose gel electrophoresis. The amplified amplicon size was 448 nucleotides in length. We further selected another PPRV-positive sample showing low Ct value to be sequenced by the NGS method. We tested the collected tissue specimens for the presence of PPRV genomes by qRT-PCR. We used the primers listed above to amplify the partial PPRV-F gene. The size of these amplicons was 448 base-pairs in length. We sequenced several tissue specimens from the suspected outbreaks. We confirmed the identity of the amplified partial PPRV-F gene (448 base pairs) by blasting the sequences against other PPRV isolates in GenBank. All the obtained sequences were identical. We used only one sequence, as represented in our phylogenetic analysis. We analyzed one PPRV-positive sequence per PPRV outbreak in Riyadh. The 448 base-pair sequence of the partial PPRV-F gene originated in PPRV-infected sheep intestine and was deposited in GenBank (accession no: KY798143). This sequence revealed high nucleotide identity to other available PPRV sequences from GenBank, especially to the Ethiopian strain (99%) (accession no: KJ867541.1). We used this Ethiopian strain as a reference sequence for the PPRV genome in our data analysis. Although these specimens were sequenced using NGS to decode the full-length PPRV genome, we were able to obtain only partial PPRV genome sequences from NGS (approximately 45% of the full-length genome of the virus). However, sequencing the PPRV-infected lung tissues from CPE in the Riyadh outbreak by the NGS approach revealed several fragments per gene of the full-length PPRV genome. We assembled a fragment from the PPRV-F gene that spanned 416 nucleotides (accession no: KY679139) from one outbreak in the Riyadh region. The sequence obtained from the partial PPRV-N gene spans a region of 618 nucleotides (accession no: KY679139). Furthermore, another fragment of the M gene that spans 518 nucleotides (accession no: MT421967) was also obtained. This is in addition to the fragments from the PPRV-L, PPRV-P, and PPRV-H genes that span 398, 244, and 229 nucleotides, respectively (received accession numbers: KY684035, KY679140, and KY679136, respectively).