JOURNAL OF CLINICAL MICROBIOLOGY,0095-1137/98/$04.0010

Apr. 1998, p. 1050–1055 Vol. 36, No. 4

Copyright © 1998, American Society for Microbiology

Detection of Mycobacterium tuberculosis Complex in Cattle byPCR Using Milk, Lymph Node Aspirates, and Nasal Swabs

FABRIZIO VITALE,* GIUSEPPINA CAPRA, LETIZIA MAXIA, STEFANO REALE,GESUALDO VESCO, AND SANTO CARACAPPA

Istituto Zooprofilattico Sperimentale Della Sicilia, Palermo, Italy

Received 28 April 1997/Returned for modification 12 November 1997/Accepted 14 January 1998

The PCR technique was applied to the diagnosis of tuberculosis in live cattle, and both skin-test-negative andskin-test-positive animals were studied. DNA was taken from various sources including specimens of lymphnode aspirates, milk, and nasal swabs. After slaughter and visual inspection, tissues such as lymph nodes,lungs, and udders from tuberculin reactors were tested by the same technique. Specific oligonucleotide primersinternal to the IS6110 insertion element were used to amplify a 580-bp fragment. A 182-bp fragment wasobtained by designating a nested PCR from the first amplification product. This fragment was cloned andsequenced, and after being labeled it was employed in dot blot hybridization. A total of 100 cattle were tested,and PCR analysis was performed using nasal swab, milk, and lymph node aspirate. Sixty skin-test-positivecows were also tested to detect mycobacterial DNA in tissue samples from lymph nodes, lungs, and udders, andthe infection was confirmed in all of the animals. Using PCR analysis of tissue samples from slaughteredanimals as a “gold standard” we calculated 100% values for sensitivity, specificity, and positive and negativepredictive values for milk and lymph node aspirate samples. The respective values for nasal swab samples were58, 100, 100, and 28%. The respective values for all of the samples were 74, 100, 100, and 35%, while for visualinspection the values were 81, 100, 100, and 58%, respectively. PCR analysis of specimens of lymph nodeaspirates, milk, and nasal swabs from skin-test-negative animals showed that 52% of these skin test resultswere false negatives. These animals, not being removed from the farms, represent a potential source of furtherinfection.

Mycobacterium bovis, the cause of tuberculosis in cattle, isalso a pathogen for a large number of other animals, and itstransmission to humans constitutes a public health problem(10). The diagnosis of bovine tuberculosis in live animalsmainly depends on clinical manifestations of the disease, skintesting, and subsequent identification of the pathogen by bio-chemical testing. It is known that the skin test lacks sufficientsensitivity and specificity in many cases (8, 9, 16, 21). Neill andcoworkers (15) have reported that M. bovis may be isolatedfrom the secretions of skin-test-negative cattle and, further-more, that these animals were not anergic, as is sometimes thecase in the later stages of the disease.

Identification of the mycobacterium is based on the tradi-tional method with the Ziehl-Neelsen acid-fast stain and on thepigmentation, growth rate, and gross and microscopic colonymorphologies of cultures of the isolated causative organism.Biochemical methods such as tests for niacin, catalase, nitratereduction, and urease are used to identify different species.

The Ziehl-Neelsen stain is very rapid but lacks specificityand cannot be used to distinguish between the various mem-bers of the family Mycobacteriaceae, while the other proce-dures usually require 4 to 8 weeks to obtain good growth. Inorder to be certain of the diagnosis of tuberculosis postmortemhistopathological examination of organ lesions is carried out.

In the past few years molecular approaches to diagnosis havebeen transforming the investigation of tuberculosis, especiallyin human medicine. The introduction of PCR and nucleic acidhybridization has greatly reduced identification time (3), and

the use of PCR has improved the level of detection in clinicalspecimens. It has been previously reported (14) that by ampli-fying species-specific DNA sequences, and hybridizing the am-plified sequence with a labeled probe, 5 fg of mycobacterialDNA (corresponding to one mycobacterium) can be detectedin clinical samples. Rapid diagnosis by PCR with a number ofdifferent targets (11, 17), including the IS6110 insertion se-quence, has been previously described (2, 6, 7, 12, 23). IS6110has only been detected in species belonging to the M. tubercu-losis complex (M. tuberculosis, M. bovis, M. africanum, and M.microti) which present this sequence in multiple copies. In theclassical human M. tuberculosis variant, the IS6110 element isusually present in 8 to 20 copies. In M. bovis strains the IS6110element is present in two to six copies (4, 5, 24–26). Only M.bovis BCG has a single copy of IS6110, as has been demon-strated in many studies using restriction fragment length poly-morphism patterns (12, 13, 22).

The aim of this work was to evaluate the possible applicationof the PCR technique to the diagnosis of tuberculosis in livecattle. PCR analyses of biological samples such as milk, nasalswabs, and lymph node aspirates taken from animals withknown skin test reactions are described. PCR analysis was alsoperformed using tissue specimens from slaughtered skin-test-positive animals to confirm our results. The sensitivity andspecificity of PCR and nucleic acid hybridization methods werecompared with those of the skin test. The results indicate thatthese methods could become useful diagnostic tools especiallyfor the large-scale screening of cattle.

MATERIALS AND METHODS

Mycobacterial cultures. The following mycobacterial reference strains wereobtained from the collections of the Pasteur Institute (Paris, France): M. bovis(B7292), M. tuberculosis (140010002IP), M. avium (140310001IP), M. chelonae(140420003IP), M. phlei (141300001IP), and M. fortuitum (140410001IP). M.

* Corresponding author. Mailing address: Istituto ZooprofilatticoSperimentale Della Sicilia, Via Rocco Dicillo No. 4, 90129 Palermo,Italy. Phone: (39) 91-6565111. Fax: (39) 91-6565233. E-mail: izspa@interbusiness.it.

1050

on April 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

paratuberculosis (ATCC 19698) was obtained from the American Type CultureCollection. All the mycobacterial cultures were maintained on Lowenstein-Jensen agar slopes and were grown in 100 ml of Dubos medium enriched with10% Dubos medium albumin and 5% equine serum (Microbiological Diagnos-tici). The cultures were incubated for 25 to 28 days at 37°C. M. bovis was isolatedroutinely from sacrificed bovines submitted to the Istituto Zooprofilattico Speri-mentale della Sicilia in Palermo, Italy, where it was identified by conventionaltesting which included growth rate, gross and microscopic colony morphologies,and pigmentation of cultures and tests for niacin, catalase, nitrate reduction, andurease.

Preparation of mycobacterial DNA. Large-scale DNA extractions were per-formed as described by B. C. Ross et al. (18) with some modifications. Myco-bacterial cultures in 100 ml of Dubos medium were centrifuged for 15 min(3,500 3 g, 4°C) in a GPR Beckman centrifuge. The pellet was washed twice withSTE buffer (100 mM NaCl, 10 mM Tris-HCl, 1 mM EDTA [pH 8.0]) andsuspended in 4 ml of lysis buffer (50 mM Tris-HCl, 50 mM EDTA [pH 8.5] 15%[wt/vol], 4% sodium dodecyl sulfate [SDS]). Lysozyme (Boehringer Mannheim)was added to a final concentration of 2 mg/ml. The mixture was incubated in athermostatic bath at 37°C for 3 h. Proteinase K (Sigma Chemical Co.) was addedto a final concentration of 100 mg/ml, and incubation was continued at 37°C for1 h. After two rounds of phenol-chloroform extraction, the DNA was precipi-tated with ammonium acetate (final concentration, 0.3 M), overlaid with 2.5volumes of ice-cold ethanol, and mixed by inversion (19). Genomic DNA wasrecovered by centrifugation at 13,800 3 g for 30 min, washed with 70% (vol/vol)ethanol, and suspended in TE buffer (10 mM Tris-HCl, 1 mM EDTA [pH 8.0]).The DNA was incubated with RNase A (Boehringer Mannheim) (100 mg/ml) at37°C for 1 h and further purified with another phenol-chloroform extraction andprecipitation step. The concentration and purity of extracted DNA were calcu-lated by readings of A260 and A280 with a Hitachi U-1100 spectrophotometer. Thereagents were supplied by Sigma Chemical Co.

DNA purification from bovine samples. DNA was extracted from samples oflymph node aspirates, milk, and nasal swabs from live animals and from lymphnode, lung, and udder tissues taken from slaughtered animals. The extractionfrom the live animal samples was performed with a QIAamp Blood and TissueKit (QIAGEN). Nasal swabs were washed in 0.5 ml of phosphate-buffered salinesolution. A total of 200 ml of each sample was incubated with Proteinase K in 1volume of a suitable lysis buffer, and then the enzyme was inactivated by heatingto 70°C for 10 min. Ethanol (0.525 volumes) was added, and the mixture wasapplied onto a QIAamp spin column. After two rounds of washing, the DNA waseluted with 200 ml of the supplied buffer preheated to 70°C.

After slaughter, DNA extraction from tissue samples was performed by lysiswith chaotropic reagent in guanidinium thiocyanate (GuSCN) (Eastman KodakCompany), as described by Boom et al. (1). A small piece of tissue, about 25 mg,was lysed in 900 ml of GuSCN-containing lysis buffer (120 g of GuSCN in 100 mlof 100 mM Tris-HCl [pH 6.4] to which 20 ml of 36 mM EDTA [pH 8]–2%[wt/vol] Triton X-100 was added) with 40 ml of diatom suspension (Sigma Chem-ical Co.). The diatom-DNA complexes were rapidly collected by centrifugation at12,500 3 g for 30 s in a Beckman microcentrifuge. The pellet was washed twicewith 1 ml of GuSCN-containing washing buffer (120 g of GuSCN in 100 ml of 100mM Tris-HCl [pH 6.4]), twice with 1 ml of 70% (vol/vol) ethanol, and once with1 ml of acetone. After draining at 56°C for 10 min, the DNA was eluted with 100ml of TE buffer preheated to 56°C.

DNA amplification by PCR. The target DNA for amplification was a 580-bpfragment of IS6110, an insertion sequence-like element currently used to identifymembers of the M. tuberculosis complex.

The primers used were the oligonucleotides 295 up (59-dGGACAACGCCGAATTGCGAAGGGC-39) and 851 down (59-dTAGGCGTCGGTGACAAAGGCCACG-39), which correspond to base pairs 295 to 318 and 851 to 874 of theIS6110 insertion element, respectively. The oligonucleotide sequence was chosenbecause of its GC content by using the Mac-Vector 5.0 program sequenceanalysis software (Oxford Molecular Group).

To generate a sequence-specific probe, we designated an amplicon-nestedPCR product of 182 bp from the amplified 580-bp fragment. It was obtained byusing primer 505 up (59-dACGACCACATCAACCGGG-39) and primer 669down (59-dGAGTTTGGTCATCAGCCG-39), which correspond to base pairs505 to 523 and 669 to 686, respectively. All the oligonucleotides were supplied byCruachem Ltd. (Glasgow, United Kingdom). PCR amplification was carried outin 100-ml reaction mixtures containing (final concentrations) 2.0 mM MgCl2, 50mM KCl, 10 mM Tris-HCl [pH 9.3], 0.01% Triton X-100, 200 mM (each)deoxynucleoside triphosphate, 200 nM each primer, 72 ml of template DNAsolution, and 2.5 U of DNA Taq polymerase (Promega). The reactions wereperformed in an automated thermal cycler (Mini Cycler; MJ Research, Inc.). Theconditions were set as follows: denaturation at 95°C for 1 min, annealing at 65°Cfor 1 min, and extension at 72°C for 1 min. A 7-min extension period at 72°C wasadded after 30 cycles. A positive control containing 10 ng of M. bovis (B/29292)DNA and a negative control, without template DNA, were included.

Electrophoresis. Purified DNA and PCR products were analyzed by electro-phoresis through 0.8 and 1.5% neutral agarose gels, respectively, containing 0.1mg of ethidium bromide (Bio-Rad Laboratories)/ml in TBE buffer (0.089 MTris-HCl, 0.089 M boric acid, 0.002 M EDTA). The gels were visualized underUV light with a transilluminator (UV-GENTM; Bio-Rad Laboratories) and

photographed with Polaroid 667 film. The DNA markers used were l-HindIII orladder 100 (Pharmacia).

Cloning and sequencing of the specific DNA probe. The DNA probe, origi-nated by PCR, was cloned by using a TA cloning kit (Invitrogen). The recom-binant plasmid was obtained by ligation of 50 ng of pCR 2.1 vector with 10 ng offresh PCR product. This plasmid was employed to transform INVaF9 One Shotcompetent cells. Briefly, 2 ml of 0.5 M b-mercaptoethanol and 2 ml of the ligasereaction product were added to 50-ml vials of frozen cells. After 30 min on ice,the cells were heat shocked for 30 s in a 42°C water bath and incubated in 250ml of purchased SOC medium at 37°C for 1 h with rotary shaking at 225 rpm.Aliquots of 50 and 200 ml from each transformation were spread on Luria-Bertani agar plates containing 50 mg of ampicillin/ml and X-Gal (5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside). The plates were incubated for 18 h at37°C. The picked colonies were grown overnight in 20 ml of Terrific Broth. Thenucleotide sequence of the cloned fragment was determined by using pCR 2.1primer with the Sequenase kit from USB (20).

DNA labeling of the 182-bp DNA probe. The 182-bp amplified fragment waslabeled with digoxigenin-dUTP by using a DIG DNA Labeling Kit (BoehringerMannheim) according to the instructions of the manufacturer. For each reaction60 ng of DNA was labeled at 37°C for 2 h and employed in hybridization.

Dot blot assay. The amplified DNA was denatured for 5 min at 100°C and kepton ice. Samples of 5 ml were spotted on a positively charged nylon membrane(Boehringer Mannheim) and fixed by UV exposure. The blots were treated with0.4 M NaOH for 3 min and neutralized with 1 M Tris-HCl (pH 8.0) for 3 min.

Hybridization and detection. The filter was hybridized at 65°C overnight in anincubation bag. The prehybridization mix consisted of 43 SSC (13 SSC is 0.15M NaCl plus 0.015 M sodium citrate) 53 Denhardt solution (2% Ficoll, 2%albumin, bovine fraction V, 2% polyvinylpyrrolidone), and 1% SDS. About 4 mlof hybridization solution was used for the 100-cm2 membrane. The hybridizationsolution contained 15 ng of the freshly denatured (10 min, 100°C) digoxigenin-dUTP-labeled 182-bp probe/ml. After incubation the membrane was washedtwice in 13 SSC–1% SDS (100 ml/100-cm2 membrane) for 15 min and twice in13 SSC–0.1% SDS at 65°C for 15 min. The presence of a digoxigenin-labeledprobe was detected by using an alkaline phosphatase-conjugated antibody andCSPD substrate according to the instructions for the DIG-dUTP-DNA detectionkit (Boehringer Mannheim). The chemiluminescent signal was revealed on X-rayfilm (X AR Omat; Kodak) after exposure for 20 min at room temperature.

RESULTS

Specificity and sensitivity of PCR. The primers 295 up and851 down were used for PCR analysis of purified DNAs of theeight mycobacterial species listed in Materials and Methods. A580-bp product was found only in mycobacteria belonging tothe M. tuberculosis complex, as was confirmed by dot blotanalysis (data not shown).

The sensitivity of the PCR was determined by adding mix-tures containing decreasing amounts of M. bovis DNA in arange between 20 ng and 1 fg to the reaction vials. One fem-togram of DNA could be amplified and, in dot blot analysis,gave a detectable hybridization signal with the 182-bp probe(data not shown).

Detection of M. bovis DNA in different biological samples.During a 12-month period, 100 cattle were tested by PCR usingDNA extracted from lymph node aspirates, milk, and nasalswabs. All 60 of these animals which were skin test positivewere slaughtered, and DNA was purified from samples of lymphnode, lung, and udder tissues. PCR was carried out using theseDNA samples to confirm the diagnosis of M. bovis infection.

Gel electrophoresis analysis of representative examples ofPCR products is shown in Fig. 1. The 580-bp fragment ofIS6110 was amplified in all purified DNA from nasal swabs(Fig. 1A), milk (Fig. 1B) and lymph node aspirates (Fig. 1C) asdemonstrated by comparison with the positive control contain-ing 10 ng of M. bovis DNA (Fig. 1, lane 2). Successful ampli-fication of IS6110 fragments was also obtained with DNAextracted from tissue specimens. Figure 2 shows results forfour samples each of lymph node, lung, and udder tissues.These tissue samples were taken postmortem from four cowsthat were also used for the analysis shown in Fig. 1 (lanes 3 to6). Hybridization of the 182-bp probe with amplified DNAfrom specimens of nasal swabs, milk, and lymph node aspiratesis shown in Fig. 3. The dot blot of amplified DNA from lymph

VOL. 36, 1998 TUBERCULOSIS DIAGNOSIS IN CATTLE BY PCR 1051

on April 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

node, lung, and udder tissues is shown in Fig. 4, and only thelast spot, corresponding to the sample in lane 14 of Fig. 2, gaveno hybridization signal. Amplification of the 580-bp fragmentwhich was not detectable by gel analysis (Fig. 2, lane 13) wasproved by dot blot hybridization. These results show thegreater sensitivity of dot blot hybridization compared to gelelectrophoresis.

Samples were deemed positive when the 580-bp fragmenthybridized with the 182-bp probe in the dot blot test. Sampleswere considered negative when no detectable signal was foundin the dot blot hybridization with the labeled probe.

We calculated the sensitivity, specificity, and positive andnegative predictive values of the combined PCR-dot blot testperformed for the 60 animals that were skin test positive. Forthese calculations we used the results obtained by PCR usingboth samples from live cattle and tissue samples as a “goldstandard.” PCR analysis using 54 milk samples and 49 lymphnode aspirates gave the 100% values for sensitivity, specificity,and positive and negative predictive values. PCR using 50nasal swabs shown high specificity (100%) but a low sensi-tivity (58%); the positive predictive value was also 100%

while the negative predictive value was only 28% (Table 1).The lower values for nasal swabs account for the total sam-ple values of 74, 100, 100, and 35% for sensitivity, specificity, andpositive and negative predictive values, respectively (Table 2).

FIG. 1. Analysis of PCR-amplified 580-bp fragment by 1.5% agarose gelelectrophoresis. DNA was extracted from nasal swab (A), milk (B), and lymphnode aspirate (C) samples. Procedures for DNA preparation and PCR amplifi-cation, and sequences of primers used are given in the text. A total of 70 ml ofPCR products was analyzed. Lane 1, l phage digested with HindIII (panels Aand B) or ladder 100 (panel C) as a DNA molecular size marker; lane 2, positivecontrol amplified from 10 ng of M. bovis DNA. (A) Lanes 3 to 20, 580-bpamplified fragments from nasal swab samples from skin-test-positive cows; lane21, control without mycobacterial DNA. (B) Lanes 3 to 14, 580-bp amplifiedfragments from milk samples from skin-test-positive cows; lane 15, control with-out mycobacterial DNA. (C) Lanes 3 to 12, 580-bp amplified fragments fromlymph node aspirate samples from skin-test-positive cows; lane 13, control with-out mycobacterial DNA.

FIG. 2. Analysis of PCR-amplified 580-bp fragment by 1.5% agarose gelelectrophoresis. Procedures for DNA preparation and PCR amplification andsequences of primers used are given in the text. DNA was extracted from tissuesof four skin-test-positive cows that were also used for the experiment describedin Fig. 1 (lanes 3 to 6). Lane 1, ladder 100 as a DNA molecular size marker; lane2, target DNA amplified from 10 ng of M. bovis DNA; lanes 3 to 6, DNA fromlymph node tissue; lanes 7 to 10, DNA from lung tissue; lanes 11 to 14, DNAfrom udder tissue; lane 15, control without mycobacterial DNA.

FIG. 3. Analysis of PCR products by dot blot hybridization with 11-dUTP-labeled 182-bp probe. Procedures for PCR amplification, sequences of primersused, and description of cloning and sequencing of probe are given in the text.Samples are the same as those used in the experiment described in the legend toFig. 1. Spot 1a, target DNA amplified from 10 ng of M. bovis DNA; spots 2a to6c, PCR products from nasal swab samples; spot 7c, control without mycobac-terial DNA; spot 1d, target DNA amplified from 10 ng of M. bovis DNA; spots2d to 6e, PCR products from milk samples; spot 7e, control without mycobac-terial DNA; spot 1g, target DNA amplified from 10 ng of M. bovis DNA; spots2g to 5g, PCR products from lymph node aspirate samples; spot 6g, controlwithout mycobacterial DNA.

1052 VITALE ET AL. J. CLIN. MICROBIOL.

on April 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

As shown in Table 3, we compared the results of visualinspection with those of PCR detection using tissue samples.We observed 43 tissue specimens with typical lesions and 10showing no visible lesions during the veterinary inspection; allof these samples were PCR positive. The remaining sevenskin-test-positive cows were negative by PCR-dot blot hybrid-ization for all samples examined. We calculated the sensitivity,specificity, and positive and negative predictive values for vi-sual inspection after slaughtering as being 81, 100, 100, and58%, respectively.

The main goal of this study was to check M. bovis infectionin cattle with high precision and compare the results with thosefrom skin testing. We obtained interesting results especially inthe cases of dubious reactivity to the skin test and when anonspecific reaction occurred in animals in the absence oftypical clinical manifestations. This happened in seven skin-test-positive animals that, submitted to veterinary inspection,had not shown the typical M. bovis lesions. Two of these cowswere tested by PCR using milk, lymph node aspirates, lymphnode tissue, and lung tissue. One of these cows was positive tothe skin test, while the other showed a low reaction; however,being officially positive, both were slaughtered. The skin-test-positive cow was PCR and dot blot negative, while the resultsfor the cow with a dubious reaction were negative for all of thesamples tested (data not shown).

To establish if our test could detect false skin-test-negativesubjects, we examined 40 animals considered officially M. bovis

free. Of these a group of seven bulls and a group of seven cowson two different farms gave interesting results. All the animalshad been negative to the skin test performed 1 year before ourinvestigation. The dot blot hybridization of amplified DNAsamples from the bull lymph node aspirates and cow milkshowed that all of these animals were positive, except one bull(data not shown). One week later the animals were retested bythe skin test and were examined clinically, and all results werenegative. However, on the basis of our results, the skin test wasrepeated 6 months after our test, and again all the results werenegative. These animals are considered officially M. bovis free.

In Table 4 the results from the 100 animals tested by PCRduring the period of our investigation are summarized. Of the40 skin-test-negative animals, 19 were confirmed by PCR,while 21 resulted positive in at least two of the tests performedusing milk, lymph node aspirates, and nasal swabs.

DISCUSSION

The data presented here represent a successful attempt tosatisfy the need for a more sensitive, specific, and rapid test forthe diagnosis of tuberculosis in cattle. In particular, the utilityof PCR as a tool to test M. bovis infection in biological samplestaken from live animals was studied. The farms tested in thisstudy were randomly selected in Sicily, and the specimensexamined were taken and kept in sterile conditions to avoidcontamination between different cows from the same farm. Weperformed PCR using samples of milk, lymph node aspirates,and nasal swabs from 100 cows and on tissue samples, takenafter slaughter, 60 skin-test-positive animals. IS6110 was cho-sen as the target sequence for the PCR test because it isspecific for mycobacteria belonging to the M. tuberculosis com-plex, and the results obtained were in accordance with those ofother authors (12, 14, 23). Amplification of the 580-bp frag-ment, revealed by detection of the 182-bp hybridized probe,can reveal as little as 1 fg of DNA, corresponding to onemycobacterial genome (14). Excellent sensitivity, specificity,and positive and negative predictive values were found for thePCR-dot blot test performed using milk samples and lymphnode aspirates. So this kind of sample could be considereduseful in the screening for tuberculosis in cattle, having ahigher accuracy than the skin test. Sampling milk did not

FIG. 4. Analysis of PCR products by dot blot hybridization with 11-dUTP-labeled 182-bp probe. Procedures for PCR amplification, sequences of primersused, and description of cloning and sequencing of probe are given in the text.Samples are the same as those used for the experiment described in the legendto Fig. 2. Spot 1, target DNA amplified from 10 ng of M. bovis DNA; spots 2 to5, PCR products from lymph node tissue; spots 6 to 9, PCR products from lungtissue; spots 10 to 13, PCR products from udder tissue; spot 14, control withoutmycobacterial DNA.

TABLE 1. Detection of M. tuberculosis complex by PCR usingnasal swabs, milk, and lymph node aspirates taken from live

skin-test-positive cattle versus PCR detection using tissuesamples from slaughtered cattle

Kind of sample Result for PCR-testedbiological samples

Result (no.) for PCR-tested tissue samples

Positive Negative Total

Nasal swab Positive 25 0 25Negative 18 7 25Total 43 7

Milk Positive 47 0 47Negative 0 7 7Total 47 7

Lymph node aspirate Positive 42 0 42Negative 0 7 7Total 42 7

TABLE 2. Detection of M. tuberculosis complex by PCR usingsamples taken from live skin-test-positive cattle versus PCR

detection using tissue samples from slaughtered cattle

Result for PCR-testedsamples from live

cattle

Result (no.) for PCR-tested tissue samples

Positive Negative Total

Positive 114 0 114Negative 39 21 60Total 153 21

TABLE 3. Detection by visual inspection of typical M. tuberculosiscomplex lesions in tissue samples taken from slaughtered skin-test-positive cattle versus PCR detection using the same tissue samples

Type of tissue sampleResult (no.) for PCR-tested tissue samples

Positive Negative Total

With lesion 43 0 43Without lesion 10 7 17Total 53 7

VOL. 36, 1998 TUBERCULOSIS DIAGNOSIS IN CATTLE BY PCR 1053

on April 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

present difficulties, and we obtained samples from 54 of the 60skin-test-positive cows (the remaining 6 were pregnant). Atotal of 87% of these milk samples were positive by PCR, andthe results for the “no visible lesion reactors” were negative.This latter result was also confirmed by nasal swab and lymphnode aspirate examination, revealing 12% false positives to theskin test. It was hypothesized that this highly sensitive tech-nique could be employed to diagnose early tuberculosis incattle and to prevent the spread of infection. The nasal swabanalysis only permitted identification of 58% of infected ani-mals and 28% of noninfected cattle. Our results agree withknown clinical data that report few cases of open tuberculosisin cattle. Nevertheless PCR using nasal swabs has high speci-ficity and positive predictive value and could be used togetherwith PCR using other samples from the same subjects, giving aclear sign of airborne contamination within herds. Taking nasalswabs is easier and quicker than other more-invasive samplingmethods. It is therefore very useful particularly in the case ofgeneralized tuberculosis in which it is known that some animalsmay fail to respond to skin testing while M. bovis may bepresent in nasal fluid.

Another significant fact was the 81% sensitivity for the visualinspection test versus PCR using tissue samples. The absenceof macroscopic lesions does not exclude the presence of earlyinfection. On the other hand the 58% negative predictive valueindicates that a significant number of nonlesion reactors areslaughtered erroneously. The easy recognition of typical tuber-culosis lesions in slaughtered cattle gives 100% specificity andpositive predictive value.

Only 28 milk samples, 30 lymph node aspirates, and 26 nasalswabs from 33 cows and 7 bulls that were negative to the skintest were examined, because we could take only one or twokinds of sample from each animal. In fact on many farms,officially free from infection, complete sampling was not pos-sible. A total of 11 milk, 10 lymph node aspirate, and 13 nasalswab samples were positive by PCR, and the subjects for whichat least two samples were positive by PCR were deemed pos-itive. The data suggest that 52% of the skin-test-negative ani-mals tested in our study may be false negatives. The low sen-sitivity and specificity of the bovine skin test (8, 9, 16, 21) is thecause of decreased efficacy in eradication campaigns and leadsto a greater risk in public health programs and also to eco-nomic losses in the cattle industry.

The specific PCR analysis reported here can be performedon biological samples easily taken from animals on the farm.The simplicity of its application is typified by the fact that milksamples and lymph nodes aspirates, which give precise results,provide excellent material for diagnosis. The method is rapid,requiring only 48 to 72 h from sampling for the detection ofamplified DNA in dot blot hybridization. It may therefore bevery useful to do PCR in parallel with the officially approvedskin test, especially in the case of dubious reactions, anergy, orwhen in the presence of cross-reactivity with correlated anti-

genic determinants. Moreover, it may be possible to use thistest in epidemiological studies aimed at determining the prev-alence of bovine tuberculosis in areas in which the disease hasnot been eradicated.

ACKNOWLEDGMENT

This study was supported by grants from the Italian Ministry ofHealth.

REFERENCES

1. Boom, R., C. J. A. Sol, M. M. M. Salimans, C. L. Jansen, P. M. E. Wertheim-van Dillen, and J. van der Noordaa. 1990. Rapid and simple method forpurification of nucleic acids. J. Clin. Microbiol. 28:495–503.

2. Brisson-Noel, A., B. Gicquell, D. Lecossier, V. Levyfrebault, X. Nassif, andA. J. Hance. 1989. Rapid diagnosis of tuberculosis by amplification of my-cobacteria DNA in clinical samples. Lancet 4:1069–1071.

3. Clarridge, J. E., R. M. Shawar, T. M. Shinnick, and B. Plikaytis. 1993.Large-scale use of polymerase chain reaction for detection of Mycobacteriumtuberculosis in a routine mycobacteriology laboratory. J. Clin. Microbiol.31:2049–2056.

4. Collins, D. M., S. K. Erasmuson, D. M. Stephens, G. F. Yates, and G. W. DeLisle. 1993. DNA fingerprinting of Mycobacterium bovis strains by restrictionfragment analysis and hybridization with insertion elements IS1081 andIS6110. J. Clin. Microbiol. 31:1143–1147.

5. Cousins, D. V., S. N. Williams, B. C. Ross, and T. M. Eliis. 1993. Use ofrepetitive element isolated from Mycobacterium tuberculosis: a new tool forepidemiological studies of bovine tuberculosis. Vet. Microbiol. 37:1–17.

6. Eisenach, K. D., M. D. Cave, J. H. Bates, and J. T. Crawford. 1990. Poly-merase chain reaction amplification of a repetitive DNA sequence specificfor Mycobacterium tuberculosis. J. Infect. Dis. 161:977–981.

7. Eisenach, K. D., M. D. Sifford, M. D. Cave, J. H. Bates, and J. T. Crawford.1991. Detection of Mycobacterium tuberculosis in sputum samples using apolymerase chain reaction. Am. Rev. Respir. Dis. 144:1160–1163.

8. Francis, J., R. J. Seiler, I. W. Wilkie, D. O’Boyle, M. J. Lumsden, and A. J.Frost. 1978. The sensitivity and specificity of various tuberculin tests usingbovine PPD and other tuberculins. Vet. Rec. 103:420–425.

9. Gardner, I. A., and D. W. Hird. 1989. Environmental source of mycobacte-riosis in California swine herd. Can. J. Vet. Res. 53:33–37.

10. Hardie, R. M., and J. M. Watson. 1992. Mycobacterium bovis in England andWales: past, present and future. Epidemiol. Infect. 109:23–33.

11. Hawkey, P. M. 1994. The role of polymerase chain reaction in diagnosis ofmycobacterial infections. Rev. Med. Microbiol. 5:21–32.

12. Hermans, P. W. M., D. van Soolingen, J. W. Dale, A. R. J. Schuitema, R. A.McAdam, D. Catty, and J. D. A. van Embden. 1990. Insertion element IS986from Mycobacterium tuberculosis: a useful tool for diagnosis and epidemiol-ogy of tuberculosis. J. Clin. Microbiol. 28:2051–2058.

13. Hermans, P. W. M., D. van Soolingen, E. M. Bik, P. E. W. de Haas, J. W.Dale, J. D. A. van Embden. 1990. Insertion element IS987 from Mycobacte-rium bovis BCG is located in a hot-spot integration region for insertionelements in Mycobacterium tuberculosis complex strains. Infect. Immun. 59:2695–2705.

14. Kolk, A. H. J., A. R. J. Schuitema, S. Kuijper, J. van Leeuwen, P. W. M.Hermans, J. D. A. van Embden, and R. A. Hartskeerl. 1992. Detection ofMycobacterium tuberculosis in clinical samples by using polymerase chainreaction and a nonradioactive detection system. J. Clin. Microbiol. 30:2567–2075.

15. Neill, S. D., J. Hanna, D. P. Mackie, and T. G. D. Bryson. 1992. Isolation ofMycobacterium bovis from the respiratory tracts of skin-test-negative cattle.Vet. Rec. 131:45–47.

16. Neill, S. D., J. Cassidy, J. Hanna, D. P. Mackie, J. M. Pollock, A. Clements,E. Walton, and D. G. Bryson. 1994. Detection of Mycobacterium bovis infec-tion in skin-test-negative cattle with an assay for bovine interferon-gammaVet. Rec. 135:134–135.

TABLE 4. Correlation between PCR-dot blot and skin testa

PCR-dot blotresult

Result (no.) by skin test

Milk Lymph nodeaspirate Nasal swab Tissue with

lesionTissue without

lesion Total

1 2 1 2 1 2 1 2 1 2 1 2

1 47 11 42 10 25 13 43 10 53 212 7 17 7 18 25 13 7 7 19

a 1, positive; 2, negative.

1054 VITALE ET AL. J. CLIN. MICROBIOL.

on April 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

17. Pao, C. C., T. S. Yen, J. B. You, J. S. Maa, E. H. Fiss, and C. H. Chang. 1990.Detection and identification of Mycobacterium tuberculosis by DNA ampli-fication. J. Clin. Microbiol. 28:1877–1880.

18. Ross, B. C., K. Raios, K. Jackson, and B. Dwyer. 1992. Molecular cloning ofa highly repeated DNA element from Mycobacterium tuberculosis and its useas an epidemiological tool. J. Clin. Microbiol. 30:942–946.

19. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, ColdSpring Harbor, N.Y.

20. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing withchain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467.

21. Seiler, R. J. 1979. The non-diseased reactor: considerations on the interpre-tation of screening test results. Vet. Rec. 105:226–228.

22. Szewzyk, R., S. B. Svenson, S. E. Hoffner, G. Bolske, H. Wahlstrom, L.Englund, A. Engvall, and G. Kallenius. 1995. Molecular epidemiologicalstudies of Mycobacterium bovis infections in humans and animals in Sweden.J. Clin. Microbiol. 33:3183–3185.

23. Thierry, D., A. Brisson-Noel, V. Vincent-Levy-Frebault, S. Nguyen, J.-L.

Guesdon, and B. Gicquell. 1990. Characterization of a Mycobacterium tuber-culosis insertion sequence, IS6110, and its application in diagnosis. J. Clin.Microbiol. 28:2668–2673.

24. van Soolingen, D., P. E. W. de Haas, J. Haagsma, T. Eger, P. W. M.Hermans, V. Ritacco, A. Alito, and J. D. A. van Embden. 1994. Use of variousgenetic markers in differentiation of Mycobacterium bovis strains from ani-mals and humans and for studying epidemiology of bovine tuberculosis.J. Clin. Microbiol. 32:2425–2433.

25. van Soolingen, D., P. W. M. Hermans, P. E. W. de Haas, and J. D. A. vanEmbden. 1992. Insertion element IS1081-associated restriction fragmentlength polymorphisms in Mycobacterium tuberculosis complex species: a re-liable tool for recognizing Mycobacterium bovis BCG. J. Clin. Microbiol.30:1772–1777.

26. van Soolingen, D., P. W. M. Hermans, P. E. W. de Haas, D. R. Soll, andJ. D. A. van Embden. 1991. Occurrence and stability of insertion sequencesin Mycobacterium tuberculosis complex strains: evaluation of an insertionsequence-dependent DNA polymorphism as a tool in the epidemiology oftuberculosis. J. Clin. Microbiol. 29:2578–2586.

VOL. 36, 1998 TUBERCULOSIS DIAGNOSIS IN CATTLE BY PCR 1055

on April 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from