Screening of Neonatal Calves for Persistent Infection with Bovine Viral Diarrhea Virus by Immunohistochemistry on Skin Biopsy Samples

Detection and elimination of cattle that are persistently infectedwith bovine viral diarrhea virus (BVDV) is important for controllingthe transmission of this virus. Colostrum-derived antibodiesmake the detection of persistently BVDV-infected neonatal calvescumbersome and expensive. The objective of this study was toevaluate the use of immunohistochemical staining of skin biopsysamples from neonatal calves as a method for the early detectionof persistent BVDV infection. Three hundred thirty-two 1- to4-week-old dairy calves were screened for BVDV as part of aroutine control program. Formalin-fixed skin biopsy sampleswere stained for BVDV antigen by immunohistochemistry (IHC),and the results were compared to those of virus isolation (VI)from white blood cell preparations. Six calves were positiveby both IHC and VI and remained positive for BVDV upon subsequentfollow-up testing; thus, they were classified as persistentlyinfected with BVDV. One calf was positive by VI but negativeby IHC. On subsequent testing, the calf was negative by VI,suggesting that the initial VI result was due to an acute BVDVinfection. One calf was positive by IHC but negative by VI.This calf remained negative by VI on follow-up testing. Immunohistochemicalstaining of skin biopsy samples is a reliable method for screeningneonatal calves for persistent BVDV infection and would be auseful management tool as an aid for controlling and preventingBVDV infection.

INTRODUCTION

Top

Introduction

Bovine viral diarrhea virus (BVDV) is a significant infectiousviral agent of cattle (1, 8). The primary sources of virus transmissionwithin and between cattle herds are cattle that are persistentlyinfected with BVDV (13). Persistent infection occurs with thedevelopment of immunotolerance to BVDV, which can result iffetal infection with BVDV occurs prior to the fourth month ofgestation (14, 16). A key management strategy that is used tocontrol and prevent BVDV infection is the detection and eliminationof cattle that are persistently infected with BVDV. Accurateand inexpensive methods for the detection soon after birth ofcattle persistently infected with BVDV would be ideal, sinceit would allow prevention of BVDV dissemination on farms. Manymethods have been used to screen cattle for persistent infection,including virus isolation (VI), enzyme-linked immunosorbentassay (ELISA), and PCR analysis of blood samples (5, 7). However,the detection of BVDV in neonatal calves is complicated by thepresence of colostrum-derived passive antibodies which can neutralizevirus in serum for up to 4 months (6, 19), thus making the useof VI and an ELISA using serum unreliable prior to this point(5, 7). VI from and PCR analysis of white blood cell preparationsfrom neonatal calves have been used successfully to detect BVDVin neonatal calves (5, 7), but the costs of these proceduresfor routine screening may be prohibitive. Immunohistochemicalstaining of skin biopsy samples has recently been shown to bea useful method for screening cattle for persistent BVDV infection(11, 18, 20) and has the potential to be a relatively inexpensiveand accurate method for screening cattle for persistent BVDVinfection. The use of immunohistochemistry (IHC) for detectingneonatal calves with persistent infection has not been evaluated.The objective of this study was to evaluate IHC as a tool fordetecting neonatal calves persistently infected with BVDV.

MATERIALS AND METHODS

Top

Materials and Methods

Calves. Holstein calves between 1 and 4 weeks of age from four differentdairy farms were used in this project. Samples were collectedbetween June and August 2000. These farms routinely screenedneonatal calves for BVDV by VI from white blood cell preparationsas part of a BVDV control program.

Samples. Triangular skin biopsy samples (ear notches) measuring approximately12 by 5 mm were obtained from the ventral pinna of each calfby using a pig ear notcher. Hemostasis at the biopsy site wasaccomplished by applying a spring paper clip over cotton gauzefor 5 min. All tissue specimens were fixed in 10% buffered formalinfor 24 to 48 h, followed by embedment in paraffin blocks. Concurrentlywith tissue sampling, 10 ml of whole blood was collected fromeach animal in EDTA Vacutainer tubes. Buffy coat cells wereisolated from the whole-blood samples, washed three times in0.1 M phosphate-buffered saline (pH 7.4), and lysed by freezingat -70°C (21).

VI. White blood cell lysates were inoculated onto bovine turbinatecell cultures and incubated for 3 days in 5% CO₂ at 37°C.Virus was detected in cell monolayers with an immunoperoxidasemonolayer assay (17) using polyclonal anti-BVDV (National VeterinaryServices Laboratory, Ames, Iowa) of swine origin as the detectionantibody.

IHC. Five-micrometer tissue sections were mounted on silane-treatedslides and stained for BVDV by using an automated procedureadapted from a previously described technique (10). The preparationof samples for staining included warming in an oven at 56°Cfor 30 min and deparaffinization in 4 changes of limonene-basedsolvent for 5 min each. Samples were rehydrated in decreasingconcentrations of ethanol and treated with a 3% H₂O₂ in methanolsolution to inactivate any endogenous peroxidase. Tissue wasdigested with protease for 10 min and then treated with 5% horseserum to block nonspecific antibody binding sites. Antigen detectionwas achieved by using the BVDV monoclonal antibody 15C5 (EdDubovi, Cornell University) as the primary antibody and a commerciallyavailable immunoperoxidase detection system (Vectastain EliteABC; Vector Laboratories, Burlingame, Calif.). Control slidesconsisting of known BVDV-positive and -negative samples werestained concurrently in each batch to monitor quality control.To distinguish nonspecific binding, serial sections of eachblock were treated with an irrelevant monoclonal antibody directedagainst bovine coronavirus. Blinded to the VI results, eachauthor independently evaluated every sample. A skin sample wasdeclared positive for BVDV if cell-specific cherry-red stainingwas present.

Data analysis. Calves were classified as persistently infected with BVDV ifthey remained positive by VI for 2 weeks following the initialtest. Sensitivity, specificity, positive predictive value, negativepredictive value, and their respective confidence levels werecalculated for the skin IHC assay to detect calves persistentlyinfected with BVDV. The correlation between the two tests wasdetermined by using the kappa statistic.

RESULTS

Top

Results

The diagnostic accuracy of skin IHC as a screening test forpersistent BVDV infection in neonatal calves is compared inTable 1 to that of VI from white blood cells. A total of 332calves were tested during the course of the study. Of these,6 (1.8%) were found to be persistently infected with BVDV. Infectionin each of these calves was detected by both VI and IHC. Atthe initial screening, virus was detected in the white bloodcells of one calf but was not detected in that calf by skinIHC. On follow-up 2 weeks later, virus could not be detectedin the white blood cells. In another calf, BVDV was detectedby skin IHC but could not be detected by VI from white bloodcells. Several attempts to isolate virus from this calf overthe next 3 weeks yielded negative results.

View this table:
[in this window]
[in a new window]
TABLE 1. Comparison of IHC with VI from white blood cells as a diagnostic test for detecting neonatal calves persistently infected with BVDV^a

DISCUSSION

Top

Discussion

Early detection of cattle that are persistently infected withBVDV would be of significant benefit to cattle producers whowish to implement BVDV control programs. Control programs thatscreen neonatal calves soon after birth have been advocated(4, 12). By strategically identifying and removing calves thatare persistently infected with BVDV as early as possible, thelength of time that these virus reservoirs are on a farm, servingas a source of virus dissemination, is significantly reduced.The cost to raise calves is significant, and it would be economicallybeneficial if a BVDV control strategy could be implemented inwhich cattle that are persistently BVDV infected were identifiedand eliminated at an age as young as possible, i.e., prior tothe investment of significant resources in them. The most commonlyused technique to screen neonatal calves for BVDV is VI fromwhite blood cells. However, the cost of the routine use of thistechnique as a screening tool for BVDV may be prohibitive. Theuse of less costly screening assays such as ELISA or VI fromserum is not reliable in neonatal calves because of the presenceof colostrum-derived passive antibodies to BVDV that neutralizefree virus. These assays can be used as passive antibodies decayand free virus begins to reappear in serum. The reappearanceof virus in serum is variable and may occur in cattle that rangefrom 2 to 6 months of age (6, 19). Many BVDV control programsscreen calves at 4 to 6 months of age. However, by this timeproducers have incurred a significant cost in raising thesecalves, and any calf that is persistently infected with BVDVis a potential source of virus transmission.

In this study, the use of IHC to detect BVDV in skin samplesfrom neonatal calves was shown to be an accurate method foridentifying persistent BVDV infection. We sampled calves between1 and 4 weeks of age. It would be advantageous to sample calvesimmediately after birth and eliminate persistently infectedcalves as soon as possible. However, as was the case in thisstudy, it may be logistically simpler to regularly sample allnew calves born during a set interval. Virus can readily beidentified in most tissues of cattle that are persistently infectedwith BVDV, including the skin (2, 3, 18, 20). The use of skinbiopsy samples has been reported to be an accurate method foridentifying cattle persistently infected with BVDV (18). However,these studies did not focus on neonatal calves, in which passivelyderived antibodies to BVDV are likely to be present.

In this study, the results of IHC and VI from white blood cellsfor two calves were in disagreement. In one calf, virus antigenwas detected by IHC but not by VI. Staining was limited to smallareas in the epidermal layer of the skin only. Several attemptsto isolate virus from this calf over the next 3 weeks were unsuccessful.Staining by IHC was still present on a skin biopsy sample taken1 week after the initial test. It is likely that this calf hadundergone an acute infection prior to the initial test and thatresidual viral antigen was still detectable in tissues. A cohortcalf that was persistently infected with BVDV was identifiedthe previous week, thus providing a virus reservoir for theinfection of susceptible calves. Viral antigen may be detectablein tissues for an extended time after the virus has been clearedfrom the blood (10, 15, 21). In addition, virus can be detectedin the skin of cattle that have undergone an acute infection,although the amount and distribution of staining are reportedto be different than those found in cattle persistently infectedwith BVDV (9, 18). In a second calf, BVDV was isolated fromwhite blood cells on initial screening but was not detectedby IHC on the skin. On follow-up testing 1 and 2 weeks later,virus was not isolated from white blood cells or found by IHCon the skin. Again, it is likely that this calf was undergoingan acute BVDV infection and was viremic at the time of initialtesting. The detection of BVDV in skin by IHC following acuteinfection has been reported to be inconsistent (9, 18), so thefailure to detect virus in this calf is not surprising. Thesetwo cases illustrate the importance of follow-up testing toconfirm positive results that are found in the initial screeningof calves for persistent BVDV infection.

The screening of neonatal calves for persistent BVDV infectionwhen implementing a BVDV control program is economically andstrategically advantageous. In this study, IHC on skin sampleswas shown to be an effective method for screening neonatal calvesfor persistent BVDV infection and could be used as a managementtool to control and prevent BVDV infection in cattle herds.

ACKNOWLEDGMENTS

Eric Keilen was supported in part through the Michigan StateUniversity College of Veterinary Medicine Merck Merial SummerResearch Program.

FOOTNOTES

* Corresponding author. Mailing address: Michigan State University, College of Veterinary Medicine, Department of Large Animal Clinical Sciences, A100 VTH, East Lansing, MI 48824. Phone: (517) 432-1494. Fax: (517) 432-1042. E-mail: groomsd{at}cvm.msu.edu.

REFERENCES

Top