Molecular analysis of partial VP-2 gene amplified from rectal swab samples of diarrheic dogs in Pakistan confirms the circulation of canine parvovirus genetic variant CPV-2a and detects sequences of feline panleukopenia virus (FPV)

A total of 40 fecal samples/rectal swabs were collected from suspected dogs i.e., domestic as well as stray dogs from Veterinary Teaching Hospital, Department of Clinic Medicine and Surgery, University of Agriculture Faisalabad Pakistan and Civil Veterinary Hospital Faisalabad regardless of their vaccination status, as it was not available. In Pakistan more than 90% dogs are stray dogs with no vaccination what so ever. The samples were collected from dogs from different areas of Faisalabad district over the period of two years, (2014 and 2015). Five fecal samples/rectal swabs were collected from healthy dogs as negative control. The collected samples were emulsified in 0.1 M PBS and centrifuged at 3000 g for 10 min at 4C. The supernatant was used for further analysis.

All the collected samples were checked for CPV Ag with the help of FASTest PARVO Strip (https://​jfa.​no/​produkt/​fastest-parvo-strip-2stk/​), according to the instructions of manufacturer).

Viral DNA from fecal samples of infected and healthy dogs was extracted as described by [15]. To reduce the residual effect of DNA polymerase activity inhibitors the supernatants were diluted to a ratio of 1:10 using distilled water. A single step polymerase chain reaction (PCR) was performed to amplify viral DNA from clinical specimens. Gene specific primer pair Hfor/Hrev [16] were used to amplify 630 bp fragment of the capsid protein gene (VP2). The PCR-amplification of partial VP2 gene was carried out using 100 ng template DNA, in PCR mix as described earlier [16]. The PCR amplified products were run on 1.0% agarose gel containing ethidium bromide in Tris acetate EDTA (TAE) buffer and then visualized under UV transilluminator (Syngene, UK).

PCR amplified fragments were cloned in pTZ57R/T vector (Fermentas) and completely sequenced by chain termination method of sequencing on automated sequencer Applied Biosystem 3100. Sequences were analyzed and assembled by the Lasergene DNA analysis package (v8; DNA-STAR Inc., Madison, WI, USA). Phylogenetic trees were generated, first by aligning the molecules using CLUSTAL-W, followed by Maximum likelihood method of phylogenetic tree construction with 1000 bootstrap replications in MEGA7 program [17]. MegAlign program in DNA STAR was used to produce pairwise comparisons for amino acid sequence similarities. The accession numbers for CPV and FPV isolated in this study are provided in Tables 1 and 2. The other viral sequences used for amino acid sequence comparisons and phylogenetic analyses were downloaded from GenBank. The branch lengths were measured in terms of the number of substitutions per site.

From 40 CPV suspected cases, fecal samples were collected and with the help of rapid immunochromatographic technique, 33 were found positive (82%) while 7 were negative (17%), 5 samples taken as negative control were found negative when analyzed using immunochromatographic technique. Of the total 33 positive samples, 25 samples, 75% (25/33), were confirmed positive for CPV as a fragment of 630 bp was amplified in all 25 samples. Amplified product from all samples were successfully cloned in pTZ57R/T (Fermentas) vector and completely sequenced.

VP2 gene amplified from representative isolates (total of 25) selected from diverse geographical locations of district Faisalabad was partially sequenced. After sequencing BLAST analysis showed 25 sequences from 25 different samples as the CPV new 2a variant and 3 sequences from 3 different samples, as the FPV. Three samples in which FPV was identified were also infected with the new CPV-2a variant. Of the 25 samples in which VP2 was amplified 100% (25/25) confirmed sequences with CPV 2a new variant and in 12% (3/25) sequences of FPV were identified. The partial VP2 nucleotide sequences from this study were analyzed using DNASTAR software, revealing homologies of 99.5–99.9% and 99.9% within the new CPV-2a strain solates and between the FPV isolates, previously reported from Asiatic region, mainly China and Europe. Similarly low nucleotide sequence similarity between the reference CPV 2a strain (AY742953), FPV (M38246) and CPV new 2a, FPV strains isolated in this study was found. All new CPV-2a clones had a nucleotide substitution at position 3675 (T-G); 3685 (C-G) and an identical nucleotide (T) at position 4064 (Table 1), which are characteristics of nucleotide variations of new CPV-2a (CPV-2a with nucleotide variation T-G at position 3675 or CPV-2a with amino acid variation Ser297-Ala). Multiple amino acid sequence alignment was performed to find sequence homology of the prevalent strains with each other and with reference CPV 2a (M24003), and a new CPV-2a (AY742953) and FPV (M38246) strains. Based on the amino acid residue at the 426 position of the VP2 gene, the isolates were classified into CPV-2a type (426-Asn) (Table 1). Additional amino acid changes were also found at 324 and 440 positions of the VP2 gene (Table 1). All of the new CPV 2a isolates in the present study had amino acid substitution Ser297Ala in VP2 (Table 1), indicating that the isolates were either new CPV-2a or new CPV-2b. All the new CPV 2a isolates reported during 2014–2015 had amino acid Tyr324Ile substitution due to mutation of TAT to ATT of the VP2 gene (Table 1). Residue 324 is a part of an important antigenic epitope and has ability to alter viral biology; it is under positive selection pressure [13] and arising independently in different geographic locations. Sequence analysis showed that this mutant is predominant in Asian countries like Korea, China and India. Emergence of this new variant is attributed to possible importation of foreign strains or could be the evolution of the existing CPV-2a strain. Further investigations are required to distinguish between these two possibilities; else this particular viral variant may spread throughout the Pakistan in the coming years. The neighboring residue 323, is known for binding to canine transferrin receptor TfR [18] along with residue 93 and have influence on viral host range and tissue tropism. Mutation in residue 323 from Asn to Asp causes changes in interactions between residues of elaborated loops within the same VP2 molecule or the other three fold VP2 molecule and hence could jeopardize virus multiplication in canine cells [19]. Twenty out of twenty-five isolates had an additional amino acid change 440 replacing Thr with Ala and further two out of those 20 isolates showed amino acid substitution Asp434Gly. Residue 440 gains the importance antigenically as it is present in GH loop of the VP2 protein which makes the surface of the virus capsid [20]. Residue 440 is present at the top of the three fold spike, which is considered to be the main antigenic site of the virus and also counts as one of the few VP sites that undergo positive selection [21]. Three isolate were found as Feline panleukopenia virus (Table 2), which mostly infects cats and cause gastroenteritis, was detected in three infected canine fecal samples. Previously it was known that two of CPV specific sequence amino acid at position 93-Asn and 323-Asn if introduced in FPV will allow it to multiply in-vitro canine cell line. These two amino acids along with 103-Val allow CPV to efficiently multiply in canine cells [19], but in the present study residue 323 remain FPV specific (Asp) and the remaining two amino acids were not determined in present sequence analysis. In addition, among the five amino acids in Table 2, the only difference between FPV and CPV from Pakistan as well as other regions is at the amino acid residue 297 (Ala in FPV and Ser in CPV isolate). On the other side, a study have shown that in-vivo inoculation of the FPV in dogs, unexpectedly enabled the virus to replicate in thymus and bone marrow cells but not in the intestine or mesenteric lymph nodes [22]. But, a recombinant virus of CPV and FPV was shown to multiply in high titers in canine intestine and mesenteric lymph nodes [23]. Two non-synonymous mutations, found in the present study, resulting in amino acid change Thr390Ala in (FPV sequence MF182903), Asp434Gly in (New CPV-2a sequence MF182909, MF182911, MF182917), which were previously not been reported in any other country.

Maximum likelihood method was used to construct phylogenetic tree to determine the evolutionary relationship of our analyzed new CPV-2a strains with other strains of different geological areas along with vaccine strains. Three strains of FPV comprising 630 bp of partial VP2 gene from Pakistan along with other FPV strains of different geological areas were also included in the tree (Fig. 1). The parvoviruses sequenced in this study clustered in three different monophyletic clades in the phylogenetic tree at the antigenic region of VP2 gene though there is a very high sequence similarity among these sequences (Fig. 1). Total 25 isolates from 2014 to 15, could be typed as new CPV- 2a strains as shown in Fig. 1. These strains appeared to cluster with other CPV isolates from China, India, Japan, Thailand and Uruguay. These observations suggest that CPV-2a in Pakistan has maximum similarity with other strains of Asiatic origin. High nucleotides similarity between Indian, Chinese and Pakistani isolates of CPV-2a is an indication that this disease is endemic in the region. High nucleotides similarity with Indian isolates is not surprising, because there are similar environmental conditions in both the countries and it remained the same region for centuries. New CPV-2a isolates clustered with isolates of 2a or new 2a types from India, Uruguay and China. Thus the important observation, which is inferred from this phylogenetic analysis is that, the recent CPV isolates are clustering away from the commercial vaccine strains. Interestingly three FPV strains showed immediate ancestral relationship with viral strain reported from USA, Asia and Europe. The close clustering with these isolates suggests that FPV is highly conserved and should be addressed at global level. The mixed infection of FPV and CPV is surprising and interesting, as it has not been previously reported. The interesting observation is that the FPV alone was not detected in any of the samples. It was only detected in CPV positive samples. So, it might be possible that FPV was somehow supported by CPV. FPV alone might not be able to replicate in dogs. Further detailed experimentation is required to explain this phenomenon. There is another possibility to explain this observation. Dogs analyzed in this study were stray animals and due to coprophagous behavior in dogs it is very much likely that FPV detection was due to faeces ingestion and they are of no clinical relevance. However, further experiments are needed to support this observation.