Seroprevalence of bovine leukemia virus in dairy cattle in Argentina: comparison of sensitivity and specificity of different detection methods

Abstract

Bovine leukemia virus (BLV) is a retrovirus that induces a chronic infection in cattle, which develop in three possible pathological forms: asymptomatic course, persistent lymphocytosis (PL) and lymphosarcoma. Once infected, cattle remain virus carriers for life and start to show a serological reaction within a few weeks after infection. Eradication and control of the disease is based on early diagnostic and segregation of the carriers. The agar gel immunodiffusion (AGID) test has been the serological test of choice for routine diagnosis of serum samples. Nevertheless, in more recent years, the enzyme-linked immunosorbent assay (ELISA) has replaced the AGID for large scale testing. Although Argentina has over 60 million cattle population, no nationwide studies have been conducted yet to determine the prevalence of the infection. To estimate the rate of BLV infection in dairy cattle in Argentina, a survey for specific antibodies in >10,000 serum samples from animals over 18 months old, belonging to 363 different herds from the largest dairy production areas of the country, was carried out in our laboratory, along 1999. For this purpose, we developed an ELISA to detect serum antibodies against the BLV virus. The cut-off of the ELISA was established over 339 serum samples, using polymerase chain reaction and southern blot (PCR–SB) as confirmatory test. The sensitivity and specificity of the ELISA was of 97.2 and 97.5%, respectively, while the local official AGID test showed a sensitivity of 79.7% and specificity of 99.0%. To know the seroprevalence of BLV on dairy herds, and also the incidence of the infection within the herd, the serologica1 survey was based on individual serum samples. The results show that the prevalence of infected individuals is of 32.85%, while the percentage of infected herds, harboring one or more infected animals, is of 84%. These results indicate a medium level of seropositive animals when taken individually, but a high prevalence of infected farms, which has been notoriously increased in the last 15 years as shown when compared with previous data from particular geographic areas, indicating that BLV constitutes a serious sanitary problem for dairy producers in Argentina. They also indicate the poor sensitivity of the official AGID test used in the country.

Introduction

The causal agent of the enzootic bovine leukosis (EBL), the bovine leukemia virus (BLV), is a retrovirus which naturally infects cattle. BLV induces a chronic infection in cattle that develop to persistent lymphocytosis (PL) in 20–30% of the cases (Burny et al., 1980), while a small percentage (1–5%) of seropositive cattle develop malignant lymphoma, the fatal clinical form of BLV infection (reviewed in Johnson and Kaneene, 1991). The disease may also follow an asymptomatic course, in which the infected animals act as carriers without showing any symptom of disease. In addition to losses caused by death due to lymphosarcoma, live cattle as well as semen and ova from seropositive cattle, are ineligible for export to many countries (Burny et al., 1980). Moreover, production efficiency may be affected more by subclinical than by clinical disease (Radostits et al., 1994), in fact, there have been conflicting reports in the literature regarding the possible detriments to the production and/or health in BLV seropositive animals (Miller, 1980, Brenner et al., 1989, Jacobs et al., 1991, Pollari et al., 1992).

Once infected, cattle remain infected, and start showing a serological reaction within a few weeks after infection. Since there is no vaccine available, antibody presence is an accurate indicator of natural exposure to the virus. Eradication and control of the disease is exclusively based on early diagnostic and segregation of the carriers, being the sensitivity of the testing strategy a critical consideration, as false-negative test results may unnecessarily prolong the eradication efforts. For a number of years, the agar gel immunodiffusion (AGID) test has been the prescribed test for international trade (Hoff-Jorgensen, 1989). Nevertheless, diagnosis by AGID is time-consuming and require skilled observers, and they may result in false positive and negative reactors between readings, depending upon the technique and the observer. In more recent years, the enzyme-linked immunosorbent assay (ELISA) has replaced the AGID in eradication programs (Mammerickx et al., 1985, Molloy et al., 1990, Klintevall et al., 1991, Gibson, 1995) and there are several ELISA kits commercially available to detect antibodies against the virus, mainly the glycoprotein gp51, which appear early in the course of the immune response.

Clinical cases of lymphosarcoma are regularly reported in Argentina, and partial studies done in 1997 (Ghezzi et al., 1997) using ELISA on bulk milk samples in a small geographic area of Buenos Aires province (Mar y Sierras), indicated a prevalence of 68.5% infected herds. Even though that study did not analyzed individual samples, 49.4% of the tested farms were classified as having probably <15% of infected cattle and 31.5% of the farms were free of infection. Although being an enzootic disease in Argentina, there are not data of the nationwide seroprevalence of the disease. The objective of this study was to obtain the first nationwide seroprevalence data on dairy cattle in Argentina. Although the AGID test is still the official test for individual animal eligibility for export, and is characterized by a high level of specificity (99.8%) (Monke et al., 1992), it does not lend itself to large scale testing trials. Thus, we developed an ELISA to detect BLV specific antibodies, and analyzed over 10,000 serum samples obtained from cattle on the main dairy production areas of Argentina. The results indicate that the overall prevalence is 32.85% for individual samples and that 84% of the tested herds are infected. The analysis of the infection rates within the herds indicated that 19.6% of them presented <15% of animals infected, 17.1% showed between 15 and 30% of seropositive individuals and 47.3% of farms have >30% of the animals infected. The incidence, by geographical areas, indicate a range between 100% of the herds infected (Entre Rı́os west) and <15% in others (Córdoba center).

These results indicate by the first time, the seroprevalence of the BLV in the dairy herds of Argentina, and provide the basis for discussing the importance of establishing eradication campaigns and control policies to protect the less infected areas.