Persistent Herpesvirus and Retrovirus Infections of Dairy Cattle and Bison Malignant Catarrhal Fever: Polymerase Chain Reaction Survey for Ovine Herpesvirus 2 and Other

Using a polymerase chain reaction (PCR) test for sequences of ovine herpesvirus 2 (OHV2), this virus was shown to be significantly associated with sheep-associated malignant catarrhal fever (SA-MCF) in terminal cases of disease in 34 cattle and 53 bison. Ovine herpesvirus 2 was not detected in cattle (38) and bison (10) that succumbed to other diseases. Other persistent herpesviruses, retroviruses, and pestivirus, some of which have been previously isolated from cases of SA-MCF, were not associated with the disease. These included bovine herpesvirus 4 (BHV4), bovine lymphotrophic herpesvirus (BLHV), bovine syncytial virus (BSV, also known as bovine spumavirus), bovine immunodeficiency virus (BIV), and bovine viral diarrhea virus (BVDV). A PCR survey for OHV2 in DNA from individual cow’s peripheral blood lymphocytes in 4 dairies showed that the 1 dairy that was in close contact to sheep had a prevalence of OHV2 of 21.3%, whereas the 3 other dairies had no OHV2. Prevalence of the other herpesviruses and retroviruses in the dairy cows was variable, ranging from 2% to 51% for BHV4, 52% to 78.7% for BLHV, and 10% to 34% for BSV. Bovine lymphotrophic herpesvirus and BSV were also found in a few (1–4 of 21 tested) cases of terminal SA-MCF, but BIV and BVDV were not found in either the dairy cows sampled, or in the cases of SA-MCF. No significant correlation was found between the presence of any 2 viruses (OHV2, BHV4, BLHV, BSV) in the dairy cows or terminal cases of SA-MCF. Malignant catarrhal fever (MCF) is an acute lymphoproliferative disease of cattle, bison, deer, and exotic ruminants.20 This disease is characterized by high fever, corneal opacity, generalized lymphadenopathy, lymphoid vasculitis, and severe inflammatory and degenerative lesions in the mucosa of the upper respiratory tract and/or gastrointestinal tract.10,13,14,20,21 Two epidemiologic forms of MCF diseases are recognized: wildebeest-derived MCF (WD-MCF), found in Africa and in zoological parks of North America and Europe, and sheep-associated MCF (SA-MCF), which has been reported worldwide.14,20 The etiologic agent responsible for WD-MCF was identified in 1960 as a cellassociated herpesvirus, classified in the Rhadinovirus genus of the subfamily Gammaherpesvirinae, and referred to as Alcelaphine herpesvirus-1 (AHV1).6,14,20 Ovine herpesvirus 2 (OHV2) is considered to be the cause of SA-MCF, but the virus has been refractory to culture. Sequences from a single region of the OHV2 genome with partial sequence homology to AHV1 From the Diagnostic Laboratory (Collins, Vermedahl, Schiebel, Jessen, Schultheiss), and the Departments of Microbiology (Collins), Environmental Health (Bruns), Pathology (Schultheiss, DeMartini), and Clinical Sciences (Anderson, Dinsmore, Callan), College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO 80523. Received for publication October 11, 1999. have been identified and used to establish a polymerase chain reaction (PCR) test to detect viral DNA.1,2 The PCR test has been successfully applied to detect OHV2 DNA in peripheral blood leukocytes1 and in tissues from terminal cases of SA-MCF in bison24 and cattle.4 Previous investigations of SA-MCF resulted in the detection of a cell-associated cytomegalovirus,30 and retroviruses.12,27 In the present study, cases of SA-MCF in cattle and bison were analyzed retrospectively for OHV2 and for other persistent bovine herpesviruses, retroviruses, and pestivirus. Polymerase chain reaction assays were performed for OHV2, bovine herpesvirus 4 (BHV4), bovine lymphotrophic herpesvirus (BLHV), bovine syncytial virus (BSV, also known as bovine spumavirus), and bovine immunodeficiency virus (BIV). In addition, a prospective survey was performed to determine the prevalence of the 3 herpesviruses and 3 retroviruses in cows from dairies with varying history of SA-MCF. Materials and methods Cases of MCF. Fifty-three cases of MCF in bison (25 females, 16 males, 12 unrecorded) were diagnosed on the basis of gross and histopathologic lesions24 over a 10-yr period (1990–1999). Forty-two animals received a complete necropsy at the Colorado State University (CSU) Diagnostic 407 Association of OHV2 with SA-MCF Table 1. Characteristics of dairies selected for virus surveillance. Dairy no. Housing Exposure to sheep No. cows milked MCF* history 1 2 3 4 dry lot dry lot free stall free stall large feedlot within 1.6 km none known within 1.6 km 850 470 400 2,000 about 25 cases in past 5 yr none in past 5 yr (only 1 case ever) none in history of dairy about 5 cases in past 5 yr * MCF 5 malignant catarrhal fever. Laboratories, 3 were necropsied at the Wyoming State Veterinary Laboratory, and 2 were necropsied at the Texas Veterinary Medical Diagnostic Laboratory, Amarillo; 6 bison received field necropsies with appropriate tissues submitted by veterinarians. All animals were from bison ranches with the exception of 3 feedlot bison; all ranches experienced multiple losses (from 1% to 95% of bison); ranches were located in Colorado (17), Wyoming (2), and Kansas, New Mexico, Nevada, and Utah (1 each). Ages at the time of death were distributed evenly through 7 yr of age; 95% of bison died during November through April. Thirty cases of MCF in cattle (22 cows, 7 steers, 1 bull) were diagnosed from gross and histopathologic lesions13,19 over a 16-yr period (1983–1999). Twenty-six animals received a full necropsy at the CSU Diagnostic Laboratory and 4 received field necropsies with tissues forwarded by a veterinarian. All cattle cases were from Colorado dairies (12), feedlots (9), cow– calf ranches (8), or a bull farm (1); in all but 1 instance, cases were singular. Ages at the time of death were distributed evenly through 6 yr of age, except for yearlings in feedlots; 80% of the cattle died during November through May, but deaths occurred in every month of the year. Tissues were fixed in 10% buffered formalin, processed routinely, and sectioned at 5 mm. Sections were either stained with hematoxylin and eosin or extracted for PCR, as described previously.24 Cases of other diseases of cattle and bison. Cases of other terminal diseases in cattle and bison were collected for the purpose of virus surveillance. Thirty-eight bovine cases included 10 pneumonias of various etiology; 3 mucosal disease (bovine viral diarrhea); 3 pulmonary hemorrhage; 2 necrotizing enteritis; 2 emaciation; and 1 each of liver abscess, salmonellosis, suppurative esophagitis, metritis, osteomyelitis, pulmonary vasculitis, uterine prolapse, encephalitis, high mountain disease, dehydration, mastitis, emphysema, myocardial degeneration, pleuritis, hepatic necrosis, and rumenitis. Ten cases in bison included 5 unknown causes of death, esophagitis, emaciation, hepatic lipidosis, encephalitis, and 1 normal bull harvested for meat. Dairies. Four dairies were selected in northeastern Colorado, and blood was collected from approximately 50 random milking cows on each dairy farm. The cows were in different stages of lactation and of variable age, from 2 to 9 yr old. Three of the dairies had a history of MCF disease (dairies 1, 2, and 4), and 1 had no previous history of the disease (dairy 3) (Table 1). Dairy 1 milked 850 cows, lost 4 to 5 cows per year to MCF over the past 5 yr, and was located next to a large sheep feedlot. Dairy 2 milked 470 cows, had no case of MCF in over 5 yr, and had only 1 case of MCF in its history. Sheep were sporadically present in low numbers (20–40) less than 1.6 km away. Dairy 3 milked 400 cows, had no recorded cases of MCF in its history, and no sheep were located within 8 km of the farm. Dairy 4 milked 2,000 cows, had no case of MCF in the previous 2 yr, and had only 5 cases in the 3 yr before that. A small herd (40–50) of sheep was within 1.6 km of the dairy. Polymerase chain reaction. The DNA was extracted from 200 ml of buffy coat using a commercial kita and suspended in 200 ml of kit buffer. Five microliters (0.01–0.05 A260 unit) was used for each PCR, except for the BSV PCR (see below), for which 0.5 ml was used. Each DNA specimen was first confirmed for amplifiabilty (99.4% of DNA preparations) using 0.5 ml and a PCR test for the bovine actin gene.22 The DNA from fixed tissue was extracted as described previously24 using a commercial extraction reagent.b The PCR test for each virus was performed as described (Table 2): OHV2,1,9,24 BLHV,22 BHV4,5 BIV,28 and BSV.26 Positive controls for each PCR were included in every test, as were negative controls (water) every fifth sample. Positive controls consisted of virus or purified DNA as follows: BHV4, strain DN599c; BLHV, plasmid pBLHVd; BIV, plasmid BIV–R29127e; BLV, test kit antigenf; and OHV2 and BSV, buffy coat lymphocyte DNA from field cases. Identity of the OHV2 amplicon was confirmed by sequencing of 4 amplicons from bovine MCF cases and 8 amplicons from bison cases and comparing to GenBank accession HSVOVSEQ.1 One case each of BHV4, BLHV, and BSV infection was confirmed by comparison to sequence data cited in the published reports given above. Tests performed for bovine viral diarrhea virus (BVDV) and other ruminant viruses. Virus isolation (VI),3 immunofluorescence (FA),3 and immunoperoxidase (IP)7 tests were conducted for BVDV and other ruminant viruses on 26 of the bison MCF cases and on 24 of the bovine MCF cases. Virus isolation was performed on alimentary tract, lung, and/ or lymphoid tissues from 5 bison and 5 cattle; FA testing for BVDV was performed on multiple tissues including lymphoid tissues from 7 bison and 4 cattle; and IP testing for BVDV was performed on multiple tissues including lymphoid tissues from 15 bison and 15 cattle. An antigen-capture enzyme-linked immunosorbent assayg for the detection of BVDV persistently infected cattle was performed on plasma from dairy cows. Statistical analysis. Chi-square and Pearson correlation coefficients were performed as described.23