An unusual presentation of pseudocowpox associated with an outbreak of pustular ulcerative vulvovaginitis in a Swedish dairy herd

Abstract

Species Pseudocowpox virus (PCPV; family Poxviridae) is known to cause pustular cutaneous disease in cattle. We describe an outbreak of pseudocowpox with an unusual clinical picture in a free-stall dairy herd of ~80 cows. Approximately 90% of the cows had vesicles, erosions, papules, and scabs on the vulva and vaginal mucosa. Histologic analysis of biopsy tissues indicated a primary, although not specified, viral infection. Transmission electron microscopy revealed parapoxvirus particles in both tissue and vesicular materials. Deep sequencing analysis of extracted DNA from swabbed vesicle areas gave a contig of nearly 120,000 nucleotides, matching the PCPV strain VR 634 with 100% identity. Analyses confirmed the absence of other potential causes of pustular vulvovaginitis such as bovine herpesvirus 1 and Ureaplasma diversum. A rolling cow brush was suspected to be the fomite.

Genus Parapoxvirus of the family Poxviridae includes viruses that cause diseases characterized by cutaneous lesions in small and large ruminants. Infections caused by parapoxvirus (PPV) are common and endemic worldwide.^{3,11,13,15,16} The bovine parapoxviruses (BPVs) of large ruminants include species Pseudocowpox virus (PCPV) and species Bovine papular stomatitis virus (BPSV). In small ruminants, orf virus (contagious pustular dermatitis virus) can cause serious disease in sheep and goats. Even though BPSV and PCPV are both bovine viruses, PCPV is genetically more closely related to orf virus than to BPSV.^9,13 PPVs are epitheliotropic viruses that cause infection of the epidermis and mucosa, with formation of papules that progress to pustules or vesicles and then to crusts or scabs.¹⁷ Infection with PPV occurs when virus enters host cells through damaged skin or mucosa and replicates.³

PPVs are usually transmitted within herds by direct contact between animals or indirectly through exposure to contaminated surfaces and/or equipment. Infection with BPSV can affect cattle of all ages, but is more common in younger animals.¹⁷ Lesions are generally located on the lips, in the oral cavity, and/or on the muzzle in younger animals, and on the teats and udders in cows.^8,16 Coalescing scabby erosions and ulcers can also be observed.⁵ The course of infection with PCPV exhibits clinical similarities with BPSV.^4,18 Formation of ring-shaped or horseshoe-like scabs is said to be typical for PCPV infections and considered pathognomonic for pseudocowpox.¹⁵ Unlike the case with BPSV, the information on confirmed cases of PCPV infections is very scarce.⁴ Until now, infection with PCPV has not been confirmed in Sweden, although PPV infections have been found in both cattle and sheep by electron microscopic analysis. Between 2010 and 2014, PPV infection was confirmed at the National Veterinary Institute in 2 cases in cattle and 9 cases in sheep.

We describe herein an outbreak of pseudocowpox with an unusual clinical presentation. The disease began in February 2014 in a dairy herd of ~80 Swedish Red-and-White milking cows in free-stall housing. Initially, nodular and small ulcerative lesions were observed in the vulvar region of ~10 cows with essentially unaffected general condition and no fever. Within a week, ~50% of the cows in the herd had the same lesions. In some of the cows, vaginal discharge and vesicles on the vaginal mucosa were observed. After another 2–3 wk, ~90% of the cows were affected. The lesions included multiple foci of inflammation with redness and swelling that evolved to vesicles, erosions, papules, and scabs on vulva (Fig. 1). Initial single circular lesions fused into larger areas. One cow also had crusts on the rear part of the udder. No signs of disturbed general condition were observed in any of the cows, and none of the calves showed any disease signs. Noteworthy is that no visible lesions could be seen on the teats of the cows.

Figure 1.

Pseudocowpox virus–induced lesions on the vulva of a milking cow. (Photo courtesy of Johanna Winberg.)

For laboratory examination, swab samples were collected from vesicle areas covered by scabs from 3 cows (A, B, C). Tissue biopsies from another 2 cows (D, E), both formalin-fixed and non-fixed, were taken from lesions. Swab samples used for DNA extraction were placed in test tubes and Tris–EDTA (TE) buffer (pH 8.0) was added to each tube. After shaking the tubes for ~10 min, 90 µL of the TE buffer containing dissolved sampling material was mixed with 10 µL of proteinase K (Sigma-Aldrich, Stockholm, Sweden). DNA was then extracted (Magnatrix 8000 extraction robot, Magnetic Biosolutions, Stockholm, Sweden; Bullet Stool kit 1.32.104, DiaSorin, Saluggia, Italy), following the manufacturers’ instructions. Because pathogens associated with pustular vulvovaginitis include bovine herpesvirus 1 (BoHV-1), extracted DNA was initially analyzed by PCR for the detection of BoHV-1 according to a published protocol,²⁰ with minor modifications. Also, analysis for BoHV-1 antibodies in bulk milk was performed (Svanovir IBR-Ab indirect ELISA, Boehringer Ingelheim Svanova, Uppsala, Sweden). Neither the virus nor antibodies against BoHV-1 were detected.

Sections of tissue biopsies from cows D and E were stained with hematoxylin and eosin (H&E) stain, periodic acid–Schiff (PAS) reaction, and Warthin–Starry (WS) stain for histologic examination.¹² Ulcerative dermatitis with intraepidermal microabscesses was found in the biopsies of both sampled cows. Ballooning degeneration of the keratinocytes and accumulations of neutrophils were seen within and on the epidermis. Infiltration of eosinophils, lymphocytes, and plasma cells could be demonstrated in the subcutis, mainly as perivascular infiltrates. No inclusion bodies were detected. The conclusion was that the reactions indicated a primary, although nonspecific, viral infection.

To identify virus, collected specimens were submitted to transmission electron microscopy (TEM) analysis (TecnaiG² Spirit TWIN/Bio TWIN, FEI, Hillsboro, OR). Pieces of the vaginal biopsies were placed in Eppendorf tubes with phosphate-buffered saline (PBS) and were crushed using a small plastic pestle. A drop of the biopsy suspension or TE buffer from the swab material was placed on a carbon/formvar-coated 400-mesh copper grid (Gilder Grids, Lincolnshire, England). After 1 min, excess fluid was removed, and the remaining specimen was negatively stained using 2% tungstophosphoric acid (Merck, Darmstadt, Germany) at pH 6. The TEM examination revealed the presence of PPV particles with typical ovoid helical structure in both tissue and vesicular swab material (Fig. 2).

Figure 2.

Poxvirus particles in a tissue biopsy from an affected vulva, processed for negative-staining transmission electron microscopy. Bar = 0.2 μm. (Photo courtesy of Elenor Hauzenberger)

Tissue and swab materials were used for virus isolation in cell cultures. Pieces of tissue samples were homogenized and suspended in PBS with Ca²⁺ and Mg²⁺ to 10% (weight per volume). Then 0.2 mL of homogenate was inoculated onto tubes of monolayers of bovine turbinate (TB) cells, Madin–Darby bovine kidney (MDBK) cells, African green monkey kidney epithelial (Vero) cells, and primary bovine skin cells. After 1 h of incubation at 37°C, the homogenate was poured out, and culture medium was added to the tubes for continued cultivation for 7 d. The cultivation was carried out for 3 passages of 7 d (MDBK cells) or 2 passages of 7 d (TB cells, Vero cells, and skin cells). An aliquot of 0.2 mL of suspended swab material in TE buffer was inoculated onto tubes of baby hamster kidney (BHK-21) cells and epithelial Georgia bovine kidney cells and cultivated for 3 passages of 7 d. All samples were considered negative for infectious virus given that no significant cytopathic effect was demonstrated in any of the inoculated cells.

To further characterize the PPV identified by TEM, extracted DNA from the 3 collected swab samples were used for deep sequencing analysis. The DNA library was prepared (Nextera XT DNA sample preparation kit, Illumina, San Diego, CA) according to the manufacturer’s protocol and thereafter sequenced (MiSeq sequencer, Illumina). Sequence data were analyzed (CLC Genomics Workbench 8.5, Qiagen, Aarhus C, Denmark). In total, ~1.2 million reads were recorded with 220,000–550,000 and 27,000–82,000 reads of bacterial and pox viral origin, respectively. Sequence homology was investigated using the BLAST algorithm with the non-redundant nucleotide database using DIAMOND software in sensitive mode.² The resulting BLAST tables, together with the sequence reads in FASTA format, were used to make taxonomic classifications and graphical displays utilizing MEGAN 5 software.¹⁴ In addition, the CLC Genomics Workbench was used to make de novo assemblies of the sequence reads. The longest contigs received were aligned with sequences in GenBank using the BLAST algorithm. A contig of almost 120,000 nucleotides was found to match PCPV strain VR 634 with 100% identity. The result that applied to extracted DNA from each of the 3 cows also showed that the PCPV was of the same strain.

In addition to the prevalence of PCPV, the high-throughput sequence data obtained from the swab samples indicated great variation in bacterial composition among the 3 cows (Fig. 3). The number of reads assigned to genus Porphyromonas varied widely between sampled cows, from 162,370 reads in 1 sample, representing almost one-third of all bacterial reads, to only 191–1,726 reads for the other 2 animals (not visible in Fig. 3 given the cutoff for displaying genera set to 10,000 reads). Differences among animals within a herd both in fecal bacterial flora as well as uterine bacterial flora have been observed in other studies.^6,19 This diversity among animals within one herd is not fully understood. The deep-sequencing data confirmed the absence of other pathogens that normally are considered causes of pustular vulvovaginitis, such as BoHV-1 and Ureaplasma diversum.¹⁰ The bacterium Porphyromonas levii has been suggested to play a role in bovine necrotic vulvovaginitis (BNVV), a syndrome characterized by erythema and hemorrhagic necrosis.^1,7 When diagnosed, BNVV has mainly been found postpartum in first-calf heifers. However, the bacterium has also been isolated postcalving from both heifers and multiparous cows without signs of BNVV.¹ In our outbreak, cows of different ages were affected, and the clinical signs and histologic picture differed from that seen in BNVV.

Figure 3.

Pie charts showing the relative amounts of MiSeq high-throughput sequencing reads originating from nucleic acids extracted from genital swabs from 3 investigated cows A–C. Only sequence reads assigned to bacterial and viral genera are shown; reads assigned to eukaryotes and unclassified and unassigned reads have been removed. The cutoff for displaying genera was set to 10,000 reads. Genera with fewer assigned reads are coalesced to the taxonomic phylum (clostridial Firmicutes), order (Gammaproteobacteria), or family level.

It is reasonable to assume that the bacterial presence noted by deep sequencing may be seen as an incidental finding rather than a causative factor. To what extent bacteria paved the way for, or contributed to establishment of viral infection is an open question.

The unusual vulvovaginal location of the PCPV infection found in the affected cows suggests a transmission of virus from contaminated equipment rather than transmission between animals. A mechanical vertical hanging cow brush, installed in 2011, was suggested to be the fomite. The brush rotates upon contact and follows the contour of the cow according to how the cow positions herself. Several observations were made as to how the cows used the brush on the vulva area. This would explain the wide spread of pseudocowpox within the herd as well as the unusual location of the PCPV infection. When the outbreak peaked, the cow brush was removed. After thorough cleaning and disinfection, the cow brush was re-installed in April 2014. Thereafter the brush was cleaned and disinfected regularly. In May, ~14 wk after the start of the outbreak, the signs had largely disappeared; 6 mo after the outbreak, only single skin lesions could be observed in a few animals.

When and how the PCPV was introduced to the herd was not determined. According to the owner, there were no obvious signs of a PCPV infection before the outbreak. Animals from other herds had not been introduced since January 2013. Thus, it is likely that the virus was introduced shortly before the outbreak. Fourteen heifers from the herd were grazed on a separate pasture during the summer of 2013. In mid-November, the heifers were brought home and kept in a separate stable. After calving in January 2014, the primiparous cows were moved to the building with the lactating cows. It is noteworthy that only 7 of 14 heifers suffered signs of pseudocowpox during the outbreak in February 2014. We hypothesize that the heifers contracted the viral infection on pasture and that they, at the time of transfer to the herd, were more or less protected by acquired immunity; some of them with an ongoing infection may have transferred virus to the common cow brush. This outbreak highlights the genital tropism of PCPV as well as the need to regularly clean and disinfect equipment to avoid mechanical transmission of pathogens.