Материал: Bovine Viral Diarrhea Virus Diagnosis, Management, and Control

Xue W, Zhang S, Minocha HC: 1997, Characterization of a putative receptor protein for bovine viral diarrhea virus. Vet Microbiol 51:105–118.

Yu H, Isken O, Grassmann CW, Behrens S-E: 2000, A stem-loop motif formed by the immediate 5’ terminus of the bovine viral diarrhea virus genome modulates translation as well as replication of the viral RNA. J Virol 74:5825–5835.

Zhang G, Aldridge S, Clarke MC, McCauley JW: 1996, Cell death induced by cytopathic bovine viral

diarrhoea virus is mediated by apoptosis. J Gen Virol 77:1677–1681.

Zhong W, Gutshall L, Del Vecchio AM: 1998, Identification and characterization of an RNAdependent RNA polymerase activity within the nonstructural protein 5B of bovine viral diarrhea virus. J Virol 72:9365–9369.

INTRODUCTION

Infections with bovine viral diarrhea virus (BVDV) are endemic in many countries, leading to heavy economic losses for the cattle industry. Sweden was one of the first countries to introduce a national BVDV control program in 1993, which now forms the basis for control programs in many other countries (Moennig and Greiser-Wilke, 2003). The primary aim of BVDV control programs in Scandinavian countries is the identification of BVDV-free herds and prevention of reinfection of these herds so that there is a gradual decrease in the number of infected herds. A crucial requirement for the success of these programs is the availability of rapid, economical, and simple diagnostic methods that are highly sensitive and specific. In the Scandinavian model, different diagnostic tests are used for the detection of infection at various levels. Initially, bulk tank milk is screened for BVDV or anti-BVDV antibody using a reverse transcription-polymerase chain assay (RT-PCR) or an enzyme-linked immunosorbent assay (ELISA), respectively. If positive, identification of individual, persistently infected (PI) animals is undertaken. For this purpose, an ELISA for the demonstration of virus in blood has proved reliable (Bitsch and Ronsholt, 1995).

The Scandinavian model may work well in countries where cattle density is low and vaccination is not allowed. It may not work, however, in countries where both virus prevalence and cattle densities are high and where vaccination is permitted. Economic losses in these countries can be minimized by lowering the infection pressure (Moennig and GreiserWilke, 2003). No matter what control program is used, the identification and elimination of PI animals is of the utmost importance (Schelp and Greiser-Wilke, 2003), which is possible only if the diagnostic tests are highly efficient and reliable.

The diagnosis of BVDV infection can sometimes be made on the basis of history and clinical signs. However, clinical signs following BVDV infection are highly variable depending on viral strain, age, and immune status of the animal; reproductive status of the animal; and the presence of other pathogens. Thus, BVDV infection may result in subclinical acute infections; severe acute infections characterized by fever, leukopenia, and thrombocytopenia; persistent infections; reproductive disease presenting as congenital defects, repeat breeding, abortion, or mummification; enteric disease; respiratory disease; and immunosuppression. Because so many different types of clinical presentation are associated with BVDV infection, a diagnosis on the basis of history, clinical signs, and postmortem examination of dead animals can only be considered presumptive. Accurate and definitive detection of BVDV infection depends on laboratory diagnosis.

The availability of accurate and rapid diagnostic tests is necessary not only for control programs but also for prognosis, monitoring, and epidemiology of BVDV infection. Another area in which accurate BVDV diagnosis is important is the presence of BVDV or its antibody in biologics of bovine origin, such as fetal bovine serum, cell cultures grown in BVDV-contaminated fetal bovine serum, and even stocks of viruses prepared in BVDV-contaminated cell cultures. For example, Nakamura et al. (1993) isolated noncytopathic (ncp) BVDV from three different stocks of cytopathic (cp) BVDV using a reverse plaque formation method based on intrinsic interference phenomenon.

As stated above, fetal bovine serum (FBS) and calf serum are extensively used in cell cultures as nonspecific nutrients and are often contaminated with BVDV (Bolin et al., 1991; Potts et al., 1989). The use of contaminated sera can result in contami-

nation of cell cultures affecting the production of biological reagents and the results of diagnosis. Bolin et al. (1994) examined 41 cell lines in the ATCC collection and found viral antigen or RNA in 13 of them by immunohistochemistry (IHC) and RT-PCR. The use of contaminated cells may result in contaminated vaccines, which may lead to seroconversion or disease in the vaccinated animals or humans (Levings and Wessman, 1991; Wessman and Levings, 1999). Hypervirulent outbreaks of BVDV in the Netherlands and Italy were attributed to BVDV contamination of bovine herpesvirus type 1 marker vaccine (Falcone et al., 1999). The use of contaminated serum or cell cultures may also interfere with the diagnosis of viral infections by interfering with the growth of other viruses.

Genital tracts are often obtained from abattoirs to harvest cumulus-oocyte complexes and coculture of feeder cells (oviduct epithelial cells and granulose cells) for use in vitro bovine embryo production systems. A certain number of these tissues are likely to be contaminated with BVDV and their use represents a risk of BVDV transmission. Hence, all materials of animal origin used in the production of bovine embryos for in vitro fertilization should be screened for BVDV (Givens et al., 2002).

It is important, therefore, to continually monitor FBS, cell cultures, seed viruses, and live vaccines prepared in cell cultures for BVDV. Both cp and ncp strains of BVDV should be looked for although most of the contamination is associated with ncp strains, which infect cells without morphological alterations, inducing problems that arise after several cell generations. Gamma irradiation, exposure to ultraviolet light, and inactivation with beta propiolactone have been used to cure BVDV from FBS (Zabal et al., 2000).

A number of different tests are available for the detection of antigen, antibody, and viral components (antigen and nucleic acid) of BVDV. Each method has its advantages, disadvantages, and applicability. Factors that can affect the efficiency of a particular diagnostic method include antigenic and/or genetic diversity of the virus, variation in virus load, and interference from maternal antibodies obtained through colostrum. Methods using monoclonal antibodies (Mabs) can be used to differentiate pestiviruses (BVDV, border disease virus of sheep, and classical swine fever virus of pigs). Monoclonal antibodies prepared against NS2-3 protein are panpestivirus because they recognize highly conserved epitopes (Mignon et al., 1992). Methods are also available to classify BVDV into subtypes 1a, 1b, and 2.

DIRECT ANTIGEN DETECTION

Methods for direct antigen detection in clinical samples are rapid and are often as sensitive as some of the other methods. However, the presence of viral antigen in tissues is often not associated with lesions, particularly in subclinical and persistent infections. When lesions appear, they are seen primarily in lymphoid tissues, where the presence of viral antigen is associated with lymphoid depletion (Liebler-Tenorio et al., 2002). In persistent infections and mucosal disease, virus can be isolated from almost all tissues. Similarly, infection with a virulent BVDV 2 usually results in a widespread dissemination of viral antigen in the host tissues. The tests that can be used for direct antigen detection in fresh, frozen, or fixed tissues include ELISA, IHC (including immunoperoxidase staining of peripheral blood leukocytes and skin biopsies), and immunofluorescence.

IMMUNOFLUORESCENCE

In this procedure, cryostat sections of fresh tissues or smears of buffy coat cells are stained with fluo- rescein-conjugated anti-BVDV antibody and then examined under a fluorescent microscope. The presence of apple green fluorescence indicates a positive test. This method is known as direct fluorescent antibody (DFA) or direct immunofluorescence test. Although DFA on buffy coat cells has been advocated for the detection of PI animals (Bezek et al., 1988), in our hands, the technique was not very successful (Werdin et al., 1989a).

IMMUNOHISTOCHEMISTRY OF PERIPHERAL

BLOOD LEUKOCYTES

An indirect immunoperoxidase test has been used to detect BVDV in smears of buffy coat cells (Saino et al., 1994). Deregt and Prins (1998) developed a Mabbased immunoperoxidase monolayer assay (IPMA) for the detection of BVDV and compared it with a bovine polyclonal antibody (Pab)-based IPMA. A pool of five Mabs (four Mabs against BVDV 1 and one Mab against BVDV 2) was employed. These Mabs were chosen because of their broad cross-reactivity, antigenic avidity, reactivity to different BVDV proteins, and lack of competition for binding sites or binding to unusual BVDV isolates. The Mab-IPMA was found to outperform the Pab-IPMA in staining, ease of reading test results, and relative sensitivity.

IMMUNOHISTOCHEMISTRY OF SKIN BIOPSIES

Immunohistological testing of skin biopsies (ear notch samples) has been used to detect PI animals.

The technique can also be applied to dead animals using thyroid gland, skin, oral mucosa, esophagus, and abomasum as samples for IHC (Thur et al., 1996). The IHC staining of formalin-fixed, paraffinembedded tissues is an efficient method for the detection of BVDV and is often considered to be better than histopathology (Haines and Ellis, 1994; Hewicker-Trautwein et al., 1995). In a study of 41 cell lines for the presence of BVDV antigen or RNA, Bolin et al. (1994) found an excellent correlation between IHC and RT-PCR. The presence of colostrum-derived antibodies did not interfere with IHC of skin biopsy (Grooms and Keilen, 2002).

To compare IHC with virus isolation (VI), Grooms and Keilen (2002) screened samples from 332 calves. Formalin-fixed skin biopsy samples were stained by IHC and virus was isolated from buffy coat cells. Six calves were positive by both techniques. One was VI-positive and IHC-negative due probably to acute infection since IHC does not detect acute infections but VI does (Ridpath et al., 2002). In another study, skin from 41 of 42 calves, known to be PI by repeated virus isolation, were found to be positive by IHC (Njaa et al., 2000).

The Mab used for IHC should be chosen carefully because only one of 32 Mabs against BVDV proteins and glycoproteins was able to detect BVDV in formalin-fixed tissues by IHC (Haines et al., 1992). This Mab (designated 15C5) is widely employed to detect viral antigen by IHC (Baszler et al., 1995; Ellis et al., 1995).

ENZYME-LINKED IMMUNOSORBENT ASSAY

Many antigen-capture ELISAs have been developed for the direct detection of BVDV antigen in buffy coat cells, serum, and ear notch samples. The basic principle consists of the use of monoclonal antibodies to capture viral antigen followed by detection of antigen-antibody complex with enzymeconjugated antibody (Bottcher et al., 1993; Ludeman and Katz, 1994). A test that can detect all currently circulating strains of BVDV is the most desirable. The most commonly used antigen capture ELISA (AC-ELISA) uses Mab directed against a conserved antigenic domain of a nonstructural protein (NS2/3) of pestiviruses. The captured antigen is then detected with a pestivirus-specific polyclonal peroxidase conjugate (Gottschalk et al., 1992). Serum is a good sample for the detection of PI animals by antigen-capture ELISA. Acute animals are rarely detected because the virus is present in the blood of an acutely infected animal only for a short time. Monoclonal antibodies against NS2-3

have been used in antigen capture ELISA tests. These tests have been found to yield results comparable to those of virus isolation (Sandvik and Krogsrud, 1995; Greiser-Wilke et al., 1992). Entrican et al. (1995) developed a double Mab ELISA for the detection of viral antigen in blood samples. Two Mabs against p125/p80 were used to capture viral antigen from blood and another two Mabs were used to detect the captured antigen. The Mab ELISA was found to be more sensitive than Pab ELISA.

VIRUS ISOLATION

The most reliable method for the detection of BVDV infection has been the isolation of BVDV in cell cultures followed by identification of the viral isolate by immunofluorescence or immunoperoxidase monolayer assay (IPMA; Meyling, 1984; Werdin et al., 1989a) or RT-PCR (Ridpath et al., 2002). During viremia, virus can be isolated from nasal discharge, PBL, lungs, and feces. Semen, blood, serum, fetus, and feces can be used for virus isolation. However, the presence of anti-BVDV antibody may interfere with virus isolation from serum and buffy coat samples.

Many different cells of bovine origin support the growth of BVDV but bovine turbinate (BT) cells are the most widely used for virus isolation because they are more sensitive to BVDV-induced cytopathic effects, which makes it easier to differentiate cp from ncp strains.. A comparative study was carried out to determine the susceptibility of five different cell types to BVDV. The cell systems used were swine testicle (ST), mink lung (ML), bovine turbinate (BT), porcine kidney (PK15), and equine dermal (ED) cells. The titers obtained on day 8 postinfection were 101.13, 103.25, 104.13, 100.00, and 100.00, in ST, ML, BT, PK15, and ED cells, respectively, indicating that BT and ML cells are optimal for the propagation of BVDV (Onyekaba et al., 1987). In another study, primary bovine embryo kidney (pBEK) cells and two cell lines originating from bovine embryonic trachea (EBTr) and buffalo lung (IMR-31) were found to be equally susceptible to BVDV (Ferrari, 1985).

Isolated virus can be confirmed by DFA (direct fluorescent antibody assay), immunoperoxidase, antigen capture ELISA, or RT-PCR. For DFA, infected cells are rinsed in PBS and fixed in anhydrous acetone for 10 minutes. Fluorescein (FITC)-conjugated anti-BVDV conjugate is allowed to react at 37°C for 25 minutes. After rinsing lightly, the infected cells and soaked in pH 9 carbonate-bicarbonate buffer

with 0.05% Tween-20 for 10 minutes. The stained cells are examined under a fluorescent microscope. Positive cells are characterized by the appearance of apple green fluorescence. In immunoperoxidase tests, a peroxidase-labeled conjugate is used instead of an FITC-labeled one and the stained cells are examined by light microscopy.

Saliki et al. (1997) compared two techniques for the identification of BVDV isolated in cell cultures. Serum samples were inoculated in indicator cells contained in 96-well microtiter plates followed by immunostaining of infected cells with a pool of Mabs using either immunoperoxidase monolayer assay (IPMA) or the monolayer enzyme-linked immunosorbent assay (M-ELISA). In IPMA, positive samples developed a red intracellular precipitate; a yellow color appears in solution in m-ELISA. The optimal time for staining was determined to be 4 days postinoculation. Although both tests were sensitive and specific, the authors preferred M-ELISA because of its rapidity and greater objectivity. The IPMA virus isolation-immunoperoxidase test (IPX) was also found to be sensitive for the detection of BVDV in another study (Castro et al., 1997). The increased sensitivity of these tests is due to virus isolation component associated with them (Saliki et al., 1997). Pooled Mabs are often used in IPMA and ELISA to detect BVDV after amplification in cell cultures because all Mabs do not give equivalent results. For example, some of the E2-specific monoclonal antibodies generated with BVDV 1 are crossreactive with BVDV 2 and others are not (Deregt and Prins, 1998; Ridpath et al., 1994).

ANTIBODY DETECTION

An indirect measure of virus infection is the detection of virus-specific antibodies in the sera of animals. Unfortunately, it is often difficult to differentiate among antibodies produced in response to acute infection, vaccination, or transfer of maternal antibodies from dam to offspring. In cattle, calves are usually born without antibody but seroconvert after colostrum consumption. These passive antibodies wane after 3–8 months. Hence, the presence of antibody in colostrum-deprived calves can be due only to active infection (either in utero or postnatal) or vaccination. Seroconversion of sentinel animals can be used as an evidence for possible exposure to PI animals. Many tests are available for the detection of anti-BVDV antibodies—namely, virus-neutralization (VN), indirect immunofluorescence (IIF) assay, indirect immunoperoxidase (IIP), and ELISA tests (Muvavarirwa et al., 1995).

VIRUS NEUTRALIZATION TEST (VN)

The virus neutralization (VN), also known as serumneutralization (SN), is considered to be the gold standard test for the detection of anti-BVDV antibodies and is used worldwide (Rossi and Kiesel, 1971). In this test, twofold serial dilutions of serum sample are incubated with a constant amount (200–500 TCID50) of the virus for 1 hour followed by the addition of indicator cells. The test is read after 4–5 days of incubation at 37°C. The highest dilution of the serum that inhibits virally induced cytopathic effects in approximately 50% of inoculated cells is considered to be the antibody titer of the serum. The test can be used for the detection of antibodies against BVDV 1 or BVDV 2 depending upon the virus used in the test. In most situations, cp strains of BVDV are used in the test so that the presence of neutralizing antibodies can be detected by inhibition of viral infectivity as detected by the absence of viral cytopathology. However, the test can also be used with ncp strains, in which case the inhibition of viral infectivity is measured by immunoperoxidase staining of infected cells (Fulton et al., 1997).

Cross-neutralization tests can be used to characterize antigenic differences among pestiviruses (Dekker et al., 1995), and titers due to active infection can be differentiated from vaccination titers by demonstrating a fourfold rise in antibody titers using paired serum samples. Virus neutralizing antibodies usually appear 3–4 weeks after infection and persist for years. Titers induced by vaccination may also persist for a long time (Oguzoglu et al., 2003). Passive antibodies decline at 105–230 days (but may persist for more than a year).

ENZYME-LINKED IMMUNOSORBENT

ASSAY

Various ELISA tests have been developed for the detection of anti-BVDV antibodies in serum samples. The antigens used in ELISA tests include whole virus antigen, nonstructural protein, monoclonal antibodies, and peptides. Several factors can influence the results of an ELISA test—e.g., antigen, conjugated antibody, test sample, etc. (Schrijver and Kramps, 1998). The procedure used to prepare whole virus antigen can also affect the specificity and sensitivity of the ELISA test. For example, Pilinkiene et al. (1999) found that antigens prepared by mild treatment showed the most specificity and activity. Cho et al. (1991) prepared antigen from MDBK-grown BVDV. The antigen was solubilized