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

with MEGA-10 (decanonyl-N-methylglucamide) followed by the removal of hydrophobic proteins with Triton X-100 treatment. Compared to VN, this test was 100% specific and 97% sensitive. Moennig et al. (1991) described the development of an ELISA using the nonstructural protein p125/80 of BVDV as antigen. The results were comparable to those obtained by the VN test.

The specificity of serodiagnosis has been enhanced greatly by the use of monoclonal antibody in competitive ELISA systems, and further improvements are possible with the use of defined antigens derived by recombinant DNA techniques (Haines and Ellis, 1994). Lecomte et al. (1990) prepared a panel of monoclonal antibodies that recognized 80 kDa (NS3) antigen of cytopathic BVDV strains. A competitive ELISA was developed using monoclonal antibodies and recombinant protein expressed in Escherichia coli as a fusion protein with ß-galac- tosidase. Beaudeau et al. (2001) developed a blocking ELISA using monoclonal antibody against NS2- 3 and used this test for mass screening of milk and serum samples with sensitivity and specificity of approximately 97% as compared to VN.

Langedijk et al. (2001) developed a solid phase antibody ELISA using peptides deduced from the C- terminal end (residues 191–227) of pestivirus envelope protein Erns. This ELISA was cross-reactive for several types of pestiviruses and could be used for general detection of pestivirus antibodies. To detect type-specific antibody, a liquid phase ELISA using a labeled specific CSFV peptide, and an unlabeled BVDV peptide (to block cross-reactivity) was used. This test can potentially be used for the differentiation of vaccinated animals from infected ones if vaccination is based on another envelope protein (E2).

A single serum dilution can be used in an ELISA test to quantitate antibodies. Graham et al. (1997) standardized a commercial ELISA test for detection of serum antibodies to BVDV so that a single serum dilution could be tested and the results expressed quantitatively using a standard curve. Various dilutions of known sera were tested and their endpoint titers calculated by an algebraic method directly from a plot of each titration series and also from a regression line fitted to this plot.

IMMUNOPEROXIDASE AND

IMMUNOFLUORESCENCE

Indirect immunoperoxidase and indirect immunofluorescence tests have also been used for the detection of anti-BVDV antibodies. However, ELISA and VN tests appear to have replaced them.

REVERSE TRANSCRIPTIONPOLYMERASE CHAIN REACTION (RT-PCR) ASSAYS

Several molecular diagnostic tests have been described for the detection of BVDV infection. Lewis et al. (1991) used 32P end-labeled, synthetic oligonucleotide probes as a tool for the detection of BVDV RNA. A probe originating nearest the 5’ end of the viral RNA detected 86% of the 22 viral isolates tested. Brock (1991) compared DNA dot blot hybridization and RT-PCR for the detection of BVDV in serum and blood samples. RT-PCR was found to provide clearer identification of PI animals than DNA hybridization.

Due to its high sensitivity, RT-PCR is considered as an alternative to current standard methods for detecting BVDV especially in pooled samples such as bulk tank milk. Because the nucleotide sequence of BVDV is highly variable, it is important to carefully select and test primers (Ridpath et al., 1993). The detection of virus by RT-PCR has been found to be more sensitive and rapid than virus isolation. In addition, contrary to virus isolation, RT-PCR is not affected by the presence of antibodies in serum samples. Weinstock et al. (2001) were able to detect the presence of viral RNA when a single viremic sample was pooled with 100 BVDV-negative sera. Horner et al. (1995) compared antigen capture ELISA, RT-PCR, and virus isolation-IPMA. RTPCR was the most sensitive and ELISA was the least sensitive. The use of RT-PCR is not without disadvantages, however. For instance, the test does not differentiate between nucleic acid from live or inactivated virus and may yield false positive results. In a study by Givens et al. (2002), RT-PCR was positive for BVDV for fetal bovine serum and primary uterine tubal cells used for in vitro embryo fertilization. However, no virus was detectable by virus isolation, and the recipients of washed embryos did not seroconvert.

The prolonged stability of viral nucleic acid as compared to the virus itself has led to a simple method for the collection, storage, transport, and testing of blood samples. In this procedure, 10 µl of blood or serum is applied to a Whatman No. 1 paper, the sample is air dried and then tested. Using this procedure, Vilcek et al. (2001) were able to detect viral RNA for up to 6 months by RT-PCR while virus isolation was positive for a few days only.

A prerequisite for the performance of RT-PCR is an efficient and simple method for sample preparation. A study was conducted to compare the efficiency of three commercial kits for RNA extraction.

The most sensitive RT-PCRs were obtained when samples were prepared by acidic guanidinium- isothiocyanate-phenol-chloroform extraction with the Trizol (Gibco) reagent. A kit based on the binding of RNA to silica membrane in a spin column had somewhat lower sensitivity, and the kit using salt precipitation of DNA and proteins was unsuitable for the isolation of viral RNA (Scheibner et al., 2000).

Several singleand two-tube RT-PCR assays have been described for the detection of BVDV RNA in serum, buffy coat cells, and fresh and formalin-fixed tissues (including ear notches). Most of these tests are based on the detection of 5’ untranslated region (5’ UTR) and E2 gene of BVDV. Many different formats of RT-PCR have been described including nested PCR (RT-nPCR) and multiplex PCR. Both virus isolation and ELISA tests are unreliable in the presence of high levels of antibody in the sample but RT-PCR test is not affected (Zimmer et al., 2004).

For retrospective studies, it may be necessary to isolate and amplify viral RNA from formalin-fixed, paraffin-embedded tissues. Gruber et al. (1993) used proteinase K digestion of deparaffinized tissue sections followed by nPCR. This procedure consistently detected an 803 bp fragment of the gene coding for the nonstructural protein p125 of BVDV and was also able to detect BVDV RNA from fresh brain tissue after 10 days of autolysis. Bhudevi and Weinstock (2003) described a TaqMan test for the detection of BVDV in formalin-fixed tissues. The viral nucleic acid was detectable in freshly fixed tissues as well as tissues that were 7 years old. To determine the effect of temperature on RNA degradation, freshly collected tissues from a PI animal were stored at 4°C or at room temperature prior to formalin fixation. Although there was a mild drop in signal strength, tissues stored at 4°C were positive for up to 1 week prior to formalin fixation, but those stored at room temperature were positive until 74 hours only.

NESTED PCR

Gruber et al. (1994) recommended nested RT-PCR over single-step RT-PCR for archival studies because the former was found to be 100-fold more sensitive than the latter. Deregt et al. (2002) described rPCR (conventional RT-PCR after RNA extraction) and dPCR (direct RT-PCR without RNA extraction) and found that both were comparable to virus isolation.

TAQMAN

Open-tube RT-PCRs, in which PCR product is identified by agarose gel electrophoresis, are subject to

contamination with amplicons and are not suitable for high throughput of samples. Several one-tube, real-time formats of PCRs have been developed to minimize sample manipulation. These assays utilize fluorescent signals from oligonucleotide probes to detect amplification of nucleic acid. Hamel et al. (1995) described a simple, one-tube RT-PCR for the rapid detection of BVDV. Total RNA was extracted directly from whole blood and tissue samples. Drew et al. (1999) also described a single step, single-tube RT-PCR test to detect the presence of BVDV in somatic cells from bulk milk samples. The test was highly specific and sensitive, detecting one PI animal in a herd of 162 lactating animals.

Bhudevi and Weinstock (2001) developed a sin- gle-tube fluorigenic RT-PCR assay that was found to be 10–100fold more sensitive than the two-tube TaqMan and the standard single-tube RT-PCR assays. They were also able to differentiate between BVDV 1 and BVDV 2 by using two different genespecific labeled fluorigenic probes for the 5’ UTR region. A single-tube RT-PCR, using panpestivirus 324/326 primers targeting the 5’ UTR region was developed by Rossmanith et al. (2001).

Mahlum et al. (2002) reported a closed-tube format of nucleic acid amplification and detection. For the development of TaqMan RT-PCR, the forward and reverse primers to 5’ UTR (El-Kholy et al., 1998) were lengthened, and degeneracy at specific nucleotide sites was included to enable them to anneal to a wider range of BVDV strains. These primers (forward: 5’-GGGNAGTCGTCARTG- GTTCG-3’; reverse: 5’GTGCCATGTACAGCA- GAGWTTTT-3’) amplify a fragment of approximately 190 bases in BVDV types 1 and 2. The TaqMan probe (5’-6-FAM-CCAYGTGGAC- GAGGGCAYGC-TAMRA-3’) included a 6-FAM (6-carboxy-fluorescein) fluorescent reporter molecule at the 5’ end and a TAMRA (6-carboxy- tetramethyl-rhodamine) quencher molecule at the 3’ end. This test was found to be more sensitive than virus isolation and IPMA in detecting BVDV in sera and more sensitive than virus isolation and IHC in detecting BVDV in tissue samples. This test is now routinely used at the Minnesota Veterinary Diagnostic Laboratory and is considered to be rapid, sensitive, and specific.

STRAIN DIFFERENTIATION

The antigenic and genetic variability of BVDV isolates has direct implications on sensitivity of BVDV diagnosis. The diagnostic tests (such as ELISA and RT-PCR) should detect all prevailing strains of

BVDV in the area. A number of investigators have developed RT-PCRs for the identification and differentiation of BVDV and classical swine fever virus. Sullivan and Akkina (1995) developed an nPCR for the identification and differentiation of BVDV 1, BVDV 2, and BDV. Ridpath et al. (1994) designed a PCR to differentiate between types 1 and 2 of BVDV. Later, Ridpath and Bolin (1998) described a differential PCR (based on phylogenetic analysis of 5’ UTR) for the segregation of types 1a, 1b, and 2 of BVDV. Harpin et al. (1995) used PCR followed by restriction endonuclease analysis of PCR product to type BVDV. Single-strand conformation polymorphism (SSCP) analysis of polymerase chain reaction (PCR) products, a genetic screening technique for rapid detection of nucleotide substitutions in PCRamplified genomic DNA or cDNA, has also been employed to identify and differentiate among BVDV isolates (Jones and Weber, 2001).

Fulton et al. (1999) developed a 2-step RT-PCR assay for typing ruminant pestiviruses—e.g., BVDV 1, BVDV 2, and border disease virus (BDV). The first PCR (consensus PCR) detected all three viruses. The amplification of the consensus PCR product by a second (nested) PCR, which uses typespecific primers, resulted in the differentiation of all three viruses. Gilbert et al. (1999) described a nested multiplex PCR for genotyping of BVDV with or without RNA extraction.

Letellier and Kerkhofs (2003) developed a quantitative real-time PCR for the detection and genotyping of BVDV in genotypes 1 and 2. They used a primer pair that was specific for highly conserved regions of the 5’ UTR and two TaqMan probes labeled with FAM and VIC. The probe sequences differed by three nucleotides, allowing differentiation between genotypes 1 and 2. Using this procedure they detected 1000 and 100 copies of BVDV 1 and BVDV 2, respectively.

MILK AS A DIAGNOSTIC SAMPLE

Screening of bulk milk samples for antibody or antigen is a promising tool for the detection of PI animals. The milk is centrifuged to remove fat, and undiluted skim milk is tested most commonly in an ELISA test. Because of the ease of obtaining sample, milk sampling is preferable in large scale epidemiological studies and for herd screening. According to the scheme of Alenius, milk samples can be divided into four classes on the basis of antibody levels. Samples belonging to class 0–1 are rarely virus-positive and those in class 2–3 are likely to be positive (Meier et al., 2003). Commercial

ELISAs detecting antibodies against BVDV have been used as a screening test (Bottcher et al., 2003; Meier et al., 2003; Sandvik, 1999). Niskanen et al. (1991) found an excellent correlation between the level of antibodies in the bulk tank milk as detected by an indirect ELISA and the prevalence of BVDV antibody-positive cows. Analysis of bulk milk samples for BVDV antibodies is now routinely used in Scandinavian countries as a tool in the diagnosis of BVDV infections in dairy herds.

To minimize the number of animals tested, Houe (1994) conducted a study to determine whether 10 young stock (8–18 months of age) from a herd can be used as a reliable screening test. The herds were divided into “slightly” to “heavily” infected if less than 3 or more than 8 samples, respectively, were positive for BVDV antibody. No PI animal was found in 24 slightly infected herds and 5 of 18 heavily infected herds were positive for PI animals in a follow-up whole-herd test. In addition, bulk tank milk titers were also generally higher in heavily infected herds.

RT-PCR assay to screen bulk milk samples for BVDV has proven to be a sensitive and economic method for the detection of a single PI animal within a group of several hundred cows (Renshaw et al., 2000) because virus titers are usually higher in milk than in serum samples (Radwan et al., 1995). In this procedure, viral RNA is extracted from somatic cells purified from whole milk using a guanidinium isothiocyanate and phenol/chloroform extraction method. Oligonucleotide primers are selected from the 5’ UTR and NS3 region of BVDV genome. RTPCR of somatic cells was fifteenfold more sensitive than virus isolation, and the presence of antibodies did not interfere with the test. Using RT-PCR on milk samples, Drew et al. (1999) were able to detect one of 162 PI animals. In a comparative study, BVDV could be detected by both virus isolation and RT-PCR when milk from a single PI animal was diluted 1:600 with milk from a herd of BVDVnegative animals. The correlation between the two assays was 95.9% (Renshaw et al., 2000).

SCREENING FOR PERSISTENTLY INFECTED (PI) ANIMALS

The PI animals are considered “virus factories” for their herdmates. They shed large amounts of virus throughout their lifetime and hence should be identified and removed from the herd as soon as possible. After removal of PI animals and cleaning of the herd, all “brought-in” stock should be isolated unless found to be BVDV-free. Although PI animals are an important source of virus on a farm (Duffel

and Harkness, 1985), immunocompetent animals that undergo acute infections can also be a significant source of the virus and should not be ignored. Most of the diagnostic tests are designed and validated based on the detection of PI animals. Since acute infections may also lead to positive reactions from samples of serum, buffy coat, semen, and skin biopsy, it is important to distinguish between acute and persistent infections (Njaa et al., 2000; Ridpath et al., 2002). Confirmation of persistent infection can be done by obtaining positive results from two different serum samples taken 3–4 weeks apart.

Immunohistochemical staining of ear notch samples from calves is considered an effective method to detect PI animals. The ear punch samples are easy to obtain and are relatively stable, and the test is not affected by the presence of maternal antibodies. However, sample pooling is not possible. In addition, tests based on the detection of viral antigen and nucleic acid (IHC and RT-PCR) do not yield a viral isolate that can be further characterized and studied. In a landmark study, Njaa et al. (2000) used IHC staining of formalin-fixed, paraffin-embedded skin biopsy specimens to detect the presence of PI animals. Of the 42 samples that were shown to be persistently infected by repeated virus isolation, 41 were positive by IHC of skin biopsy. Based on the site of immunohistochemical staining, these authors were able to distinguish between persistent and acute infections. Staining in PI animals was most pronounced in keratinocytes, hair follicle epithelium, matrix cells of the hair bulb, and the dermal papilla; in acute infection, small foci of staining were confined to the nonfollicular epidermis and follicular ostia (Njaa et al., 2000).

Buffy coats (PBL) have been used for the isolation of BVDV for PI detection. In conventional method of buffy coat isolation, heparanized blood is sedimented, the plasma layer containing mononuclear and polymorphonuclear leukocytes is aspirated, and then centrifuged at 160 xg for 5–10 minutes. The cell pellet is suspended in 0.5–1.0 mL of Eagle’s MEM followed by inoculation in cell cultures for virus isolation. To maximize leukocyte yield, Ficoll-Paque/Macrodex (F-P/M) is often used. In this procedure, heparanized blood is diluted 1:1 with saline solution and layered on FP solution in a ratio of 3:1 and then centrifuged at 400 xg for 30 minutes at room temperature. The mononuclear cell fraction lays on top of the FP gradient. This layer is removed, washed twice in MEM followed by resuspension in 2 mL of MEM containing 5% fetal bovine serum (Howell et al., 1979).

Serum is sometimes preferred as a specimen for virus isolation because the virus survives longer in serum and because serum can be collected and stored easily as compared to buffy coat cells (Saliki et al., 2000). However, the presence of antibodies in serum may interfere with virus isolation. In some instances, pooled samples of serum may be the specimen of choice. Serum pools are used in the voluntary control program initiated in the German federal state Saxony-Anhalt in 2002 (Gaede et al., 2003). Initially, pools of sera were tested with real-time RT-PCR in a Light-Cycler system. Individual PI animals in positive pools were detected by using commercial Erns- antigen-ELISA. The prevalence of PI animals was found to be 0.2% during a 3-year period.

A prenatal test that can detect PI animals is not available although fetal fluids obtained percutaneously from late stages of gestation have been tested by virus isolation (Callan et al., 2002). Unfortunately, the procedure is labor-intensive and is not very safe. Lindberg et al. (2001) used an indirect ELISA to identify dams pregnant with PI calves and claimed good sensitivity and specificity after 7 months of gestation.

Some practitioners use paired serology to screen for PI animals. The procedure is to bleed a herd, identify animals with no or low VN titers, revaccinate, and rebleed the whole herd. The animals that do not seroconvert in response to vaccination or that die from mucosal disease are considered to be PI animals. It must be emphasized, however, that this is not a reliable method to screen for PI animals; if the vaccine strain is sufficiently different from the persisting strain, the PI animal may seroconvert to the vaccine strain and be declared non-PI although it still is persistently infected (Werdin et al., 1989a, 1989b; Ferreira et al., 2000). Such animal will not be removed from the herd and will continue to be a source of the virus for its herdmates. Other factors can also result in failure to detect PI animals—e.g., maternal antibodies may interfere with virus isolation, virus load may not be enough for detection, and detection by PCR and ELISA may not be successful if the test is not universal.

SUMMARY AND CONCLUSIONS

A tentative diagnosis of BVDV infection can often be made on the basis of history, clinical signs, and postmortem examination. For definite diagnosis, however, laboratory investigations using rapid and reliable tests are necessary. The methods for the detection of BVDV infections have greatly improved recently; several methods are available for the detec-

tion of virus, antibody, or viral components (antigen and nucleic acid). High throughput methods are now available for whole herd screening to identify and remove PI animals. Bulk milk screening for BVDV or anti-BVDV antibody is used to identify infected herds in Sweden. However, this procedure may not be successful in countries with high cattle densities and in which vaccination is permitted.

Bulk milk and serology of young animals by ELISA test is good. Herds with no or low antibody titers in milk and negative serology in young animals are considered free of PI animals. On the other hand, high titers in milk with or without positive titers in young animals is taken as the evidence that the herd is positive and that further investigation should be undertaken to detect and remove PI animals. The use of serum samples for RT-PCR and of skin biopsies for IHC or RT-PCR has greatly improved BVDV diagnosis. Using these new tools, it should now be possible to consider national control programs for BVDV.

REFERENCES

Baszler TV, Evermann JF, Kaylor PS, et al.: 1995, Diagnosis of naturally occurring bovine viral diarrhea virus infections in ruminants using monoclonal antibody-based immunohistochemistry. Vet Pathol 32:609–618.

Beaudeau F, Belloc C, Seegers H, et al.: 2001, Informative value of an indirect enzyme-linked immunosorbent assay (ELISA) for the detection of bovine viral diarrhoea virus (BVDV) antibodies in milk. J Vet Med B Infect Dis Vet Public Health

48:705–712.

Bezek DM, Baker JC, Kaneene JB: 1988, Immunofluorescence of bovine virus diarrhea viral antigen in white blood cells from experimentally infected immunocompetent calves. Can J Vet Res 52:288–290.

Bhudevi B, Weinstock D: 2001, Fluorogenic RT-PCR assay (TaqMan) for detection and classification of bovine viral diarrhea virus. Vet Microbiol 83:1–10.

Bhudevi B, Weinstock D: 2003, Detection of bovine viral diarrhea virus in formalin fixed paraffin embedded tissue sections by real time RT-PCR (Taqman). J Virol Methods 109:25–30.

Bitsch V, Ronsholt L: 1995, Control of bovine viral diarrhea virus infection without vaccines. Vet Clin North Am Food Anim Pract 11:627–640.

Bolin SR, Matthews PJ, Ridpath JF: 1991, Methods for detection and frequency of contamination of fetal calf serum with bovine viral diarrhea virus and antibodies against bovine viral diarrhea virus. J Vet Diagn Invest 3:199–203.

Bolin SR, Ridpath JF, Black J, et al.: 1994, Survey of cell lines in the American Type Culture Collection for bovine viral diarrhea virus. J Virol Methods 48:211–221.

Bottcher J, Gottschalk E, Greiser-Wilke I, et al.: 1993, Diagnosis of bovine virus diarrhoea by two en- zyme-linked immunosorbent assays. Rev Sci Tech 12:461–469.

Bottcher J, Meier N, Stengel KH, et al.: 2003, Feasibility of serological bulk milk testing as a method for BVD surveillance. Berl Munch Tierarztl Wochenschr 116:244–251.

Brock KV: 1991, Detection of persistent bovine viral diarrhea virus infections by DNA hybridization and polymerase chain reaction assay. Arch Virol Suppl 3:199–208.

Callan RJ, Schnackel JA, Van Campen H, et al.: 2002, Percutaneous collection of fetal fluids for the detection of bovine viral diarrhea virus infection in cattle. J Am Vet Med Assoc 220:1348–1352.

Castro MD, Stoffregen WC, Brigman GP, Hillard KA: 1997, A method to detect bovine viral diarrhea virus contamination in cell cultures using immunoperoxidase staining. J Vet Diagn Invest 9:427–431.

Cho HJ, Masri SA, Deregt D, et al. : 1991, Sensitivity and specificity of an enzyme-linked immunosorbent assay for the detection of bovine viral diarrhea virus antibody in cattle. Can J Vet Res 55:56–59.

Dekker A, Wensvoort G, Terpstra C: 1995, Six antigenic groups within the genus pestivirus as identified by cross neutralization assays. Vet Microbiol 47:317–329.

Deregt D, Carman PS, Clark RM, et al.: 2002, A comparison of polymerase chain reaction with and without RNA extraction and virus isolation for detection of bovine viral diarrhea virus in young calves. J Vet Diagn Invest 14:433–437.

Deregt D, Prins S: 1998, A monoclonal antibodybased immunoperoxidase monolayer (microisolation) assay for detection of type 1 and type 2 bovine viral diarrhea viruses. Can J Vet Res 62:152–155.

Drew TW, Yapp F, Paton D: 1999, The detection of bovine viral diarrhoea virus in bulk milk samples by the use of a single-tube RT-PCR. Vet Microbiol 64:145–154.

Duffell SJ, Harkness JW: 1985, Bovine virus diar- rhoea-mucosal disease infection in cattle. Vet Rec 117:240–245.

Ellis JA, Martin K, Norman GR, Haines DM: 1995, Comparison of detection methods for bovine viral diarrhea virus in bovine abortions and neonatal death. J Vet Diagn Invest 7:433–436.