The Veterinary Journal 2002, 163, 299±305
doi:10.1053/tvjl.2001.0681, available online at http://www.idealibrary.com on
Detection of Brucella Species in the Milk of Infected
Cattle, Sheep, Goats and Camels by PCR
M. E. R. HAMDY* and A. S. AMINy
*Animal Health Affairs, Department of Agricultural Development, Doha, State of Qatar; yAnimal Reproduction Research Institute (ARRI),
Al-Haram, Giza, Egypt
SUMMARY
One hundred and three milk samples were collected from 52 cows, 21 ewes, 18 goats and 12 camels. The
animals tested positive to at least one of the following: (1) standard tube agglutination test (SAT); (2) Rose
Bengal plate test (RBPT); (3) milk ring test (MRT). All milk samples were examined by culture and single-step
polymerase chain reaction (PCR) techniques for detection of Brucella species. The PCR assay amplified BrucellaDNA from 29 bovine milk samples, 10 from sheep, 13 from goats and one from a camel. The direct culture
method detected Brucella organisms from 24 samples of cows' milk, 12 from sheep, 10 from goats and failed
to detect any Brucella organisms from camels' milk. PCR detected up to 100 colony forming units (CFU) of
B. abortus per millilitre of milk in 100% of diluted milk samples, and 1000 CFU of B. melitensis from 70% of milk
samples. Although the overall sensitivity of the PCR was higher than the culture method, it should be possible to
increase the sensitivity to detect lower numbers of Brucella organisms in field samples. The speed and sensitivity
of the PCR assay suggest that this technique could be useful for detection of Brucella organisms in bovine milk,
# 2002 Elsevier Science Ltd. All rights reserved.
as well as in sheep, goat, and camels milk.
KEYWORDS: Brucella; milk; PCR; dairy animals; infection.
INTRODUCTION
Brucellosis is a zoonosis, which still threatens public
and animal health in many countries of the world. It
infects a variety of domestic and wild animals and
man, causing incapacitating disease. In dairy animals, the organism localizes in the supra-mammary
lymph nodes and mammary glands of 80% of infected animals, and these can continue to excrete the
pathogen in milk throughout their lives (Cordes &
Carter, 1979; Morgan & Mackinnon, 1979). Man can
contract the disease through consumption of contaminated milk and/or milk products. Consequently, control of the disease in animals should
lead to decreased incidence in humans.
Diagnosis of brucellosis is the cornerstone of any
control program and is based on bacteriological and
*Correspondence to: Mahmoud E. R. Hamdy, Animal Health
Affairs, Department of Agricultural Development, Ministry of
Municipal Affairs and Agriculture, Doha, State of Qatar.
Tel.: ‡ 966 2 6817369; Fax: ‡ 966 2 6482195;
E-mail: merhamdy@hotmail.com
immunological findings. The use of serological tests
is recommended as a means of indirectly diagnosing
the disease. However, many current serological tests
have proved to be either too sensitive, giving falsepositive results, or too specific, giving false-negative
results (Morgan & Mackinnon, 1979; Farina, 1985).
In addition, the presence of antibodies does not
always mean an active case of brucellosis, since vaccinated animals tend to yield persistent post-vaccinal
immune responses, and other gram-negative bacteria such as Yersinia enterocolotica may cross-react
with smooth Brucella spp. (Corbel, 1985; Diaz &
Moriyon, 1989).
The milk ring test (MRT) is the most widely used
test for screening and monitoring brucellosis in
dairy cattle (Alton et al., 1988). Although the sensitivity of the MRT is overemphasized (Huber &
Nicoletti, 1986), its specificity has been questioned
when prevalence is low (Rolfe & Sykes, 1987). In
addition, false-positive reactions may be given when
milk is tested on the day of collection or taken from
cows with mastitis (Morgan & Mackinnon, 1979).
1090-0233/02/$ ± see front matter # 2002 Elsevier Science Ltd. All rights reserved.
300
THE VETERINARY JOURNAL, 163, 3
The most specific diagnostic test involves isolation of
the causative organism, but this suffers from the
drawback of requiring a long incubation period and
low sensitivity, especially in the chronic stage of the
disease. Moreover, the culture material must be
handled carefully, as the Brucella organism is a class
III pathogen (Alton et al., 1988). Because of these
difficulties, the development of new diagnostic tests
for the direct detection of Brucella species in milk is
increasingly drawing interest.
Recently, the polymerase chain reaction (PCR)
has been shown to be a valuable method for detecting DNA from different fastidious and non-cultivable
agents (Brikenmeyer & Mushahwar, 1991). Although
there are several studies on Brucella-DNA detection
by PCR from pure culture (Fekete et al., 1990;
Herman & Ridder, 1992), only a few studies have
been performed with clinical or field samples and
most of these were in cattle (Fekete et al., 1992; Amin
et al., 1995; Leal-Klevezas et al., 1995; Romero et al.,
1995). In addition, not enough data are available
to assess the performance of the PCR assay on
milk samples from farm animals other than cattle.
Milk from other animal species such as sheep, goats,
and camels (Camelus dromedarius) is an important
source of human brucellosis, particularly in those
parts of the world were B. melitensis prevails (Alton,
1990).
The aim of this study was to evaluate the diagnostic performance of PCR on milk samples taken
from cattle, sheep, goats, and camels naturally
infected with different Brucella species.
MATERIALS AND METHODS
Samples
Milk and blood samples were collected from 52
imported Friesian cows and Egyptian native breeds
of sheep (n ˆ 21), goats (n ˆ 18) and camels
(n ˆ 12). All animals were from farms with a known
history of brucellosis. The samples were taken under
strict hygienic conditions, kept on ice and sent to our
laboratory as soon as possible. The animals all tested
positive to at least one of the following: (1) standard
tube agglutination test (SAT); (2) Rose Bengal plate
test (RBPT); (3) milk ring test (MRT) (Alton et al.,
1988). Positive samples to the SAT were those
with titres > 1/40 (50%) according to the European
technique (Alton et al., 1988). Samples were collected from a further 50 cows from Brucella-free dairy
herds, as negative controls. Milk samples used for
MRT or bacteriological examination were stored at
4 C, while milk for PCR assay was stored at ÿ20 C
till used.
To evaluate the PCR limit of detection in milk,
ten-fold dilution series of Brucella abortus 544
(reference strain) and B. melitensis 16 M (reference
strain) were added to sterile cow's milk (ten samples
per dilution) to get final concentrations of each
Brucella strain as follows: 1 105 CFU/mL, 1 104 CFU/mL, 1 103 CFU/mL, 1 102 CFU/mL,
and 1 10 CFU/mL. The samples were stored at
ÿ20 C until assayed by PCR.
Brucella reference strains were obtained from
Department of Brucellosis Research, Animal Health
Research Institute (AHRI) Dokki, Egypt.
Bacteriological examination
The cream and sediment mixture obtained after
centrifugation of 20 mL milk samples at 3000 g for
15 min at 4 C were cultured on duplicated plates of
tryptose soy agar medium supplemented with different antibiotics. The following concentrations of
antibiotics were added per litre of media: cycloheximide (100 mg), bacitracin (25 000 units), polymyxin
B sulphate (5000 units), vancomycin (20 mg),
nalidixic acid (5 mg) and nystatin (100 000 units)
(Oxoid). The plates were then incubated in a 10%
CO2 incubator at 37 C for at least seven days; suspected colonies were identified according to the
methods adopted by Alton et al. (1988). Typing
of Brucella isolates was done according to CO2 requirement, H2S production, growth in the presence
of dyes (thionin and basic fuchsin ± 20 mg/mL),
reaction with mono-specific sera (A&M), and lyses by
Tblizi and Iz phages (Alton et al., 1988).
Extraction of Brucella DNA from milk samples
The Brucella DNA extraction and purification
method in this study was that described by LealKlevezes et al. (1995). Briefly, 400 mL of lysis solution
(2% Triton X-100, 1% sodium dodecyl sulphate,
100 mM NaCl, 10 mM Tris-HCl [pH 8.0]) and 10 mL
of proteinase K (10 mg/mL) were added to 400 mL
taken from the fatty top layer of each milk sample.
The contents were mixed thoroughly and incubated
for 30 min at 50 C. Then, 400 mL of saturated phenol
(liquid phenol containing 0.1% 8-hydroxyquinoline,
saturated and stabilized with 100 mM Tris-HCl [pH
8.0] and 0.2% of 2-mercaptoethanol) were added.
The contents were mixed thoroughly and centrifuged at 8000 g for 5 min and the aqueous layer
was transferred to a fresh tube. An equal volume of
chloroform-isoamyl alcohol (24 : 1) was added, mixed
thoroughly and centrifuged at 8000 g for 5 min. The
DETECTION OF BRUCELLA SPECIES IN MILK SAMPLES BY PCR
upper layer was again transferred to fresh tube, and
200 mL of 7.5 M ammonium acetate was added and
mixed thoroughly. The contents were kept on ice for
10 min and centrifuged at 8000 g for 5 min. Before
the aqueous content was transferred to a fresh tube,
two volumes of 95% ethanol were added. The contents were mixed and the tubes were stored at
ÿ20 C. DNA was recovered by centrifuging the
samples at 8000 g for 5 min, the pellets were rinsed
with 1 mL of 70% ethanol, dried and resuspended in
20 mL of TE buffer (10 mM Tris-HCl [pH 8.0], 1 mM
disodium EDTA). DNA concentrations were determined by measuring their wavelengths at A260.
Finally the DNA extraction was stored at ÿ20 C until
they were processed in the thermocycler.
Oligonucleotide primers
B. abortus and B. melitensis primers sequences used
were previously described (Bricker & Halling, 1994).
The primers were synthesized using DNA synthesizer
(Institute of Molecular Biology and Genetic Engineering). The sequences of the oligonucleotide
primers were:
5 0 GAC, GAA, CGG, AAT, TTT, TCC, AAT, CCC 3 0
(B. abortus-specific primer);
5 0 AAA, TCG, CGT, CCT, TGC, TGG, TCT, GA 3 0
(B. melitensis-specific primer); and
5 0 TGC, CGA, TCA, CTT, AAG, GGC, CTT, CAT 3 0
(IS711-specific primer).
DNA amplification and detection of PCR products
Before amplification, DNA samples were fully denaturated by boiling for 10 min and then quenched on
ice. PCR assay conditions were performed as described by Bricker and Halling (1994): in 50 mL of a
reaction mixture containing final concentrations of
60 mM Tris HCl (pH 9.0), 1.5 mM MgCl2, 15 mM
(NH4)2SO4, 250 mM (each) of the four deoxynucleotide triphosphates (Pharmacia), 0.2 mM
(each) primer, 1 unit of Taq polymerase (Promega)
and 200 mL of extracted DNA. The PCR mixtures
were overlaid with 40 mL of paraffin oil (Sigma) and
amplified in a DNA thermal cycler (Coy Corporation). Cycling conditions consisted of 35 cycles of
1.15 min at 95 C, 2 min at 55.5 C and 2 min at 72 C,
followed by a final incubation at 72 C for 5 min.
Positive control (reactions with B. abortus 544 and
B. melitensis 16 M, reference strains DNA) and negative control (Brucella-free milk) were included.
Amplification products were fractionated in a
1.5% (wt/vol) agarose gel containing 1 TBE
(100 mM Tris-Hcl [pH 0.8], 90 mM boric acid, 1 mM
301
disodium EDTA), stained with an ethidium bromide
solution (0.5 mg/mL) and visualized under an ultraviolet transilluminator and photographed (Maniatis
et al., 1984). Visible bands of appropriate sizes of
(498 bp) for B. abortus and (731 bp) for B. melitensis
were considered positive reactions.
RESULTS
Comparison of PCR and culture with
the serological tests
One hundred and three milk samples were collected
from 52 cows, 21 sheep, 18 goats and 12 camels.
These animals were positive to SAT, RBPT or MRT
(Table I). Milk samples were examined by culture
and PCR techniques for detection of Brucella species.
Forty seven (46%) different Brucella strains were
isolated from milk samples of cattle (n ˆ 24), sheep
(n ˆ 12) and goats (n ˆ 11) but no isolates were
obtained from camels' milk. All isolated Brucella
strains except two were obtained from animals serologically positive to SAT and RBPT (Table I). PCR
assay amplified 53 (52%) of Brucella DNA from different milk samples from cattle (n ˆ 29), sheep
(n ˆ 10), goats (n ˆ 13) and from one camel-milk
sample. PCR-positive samples were obtained from
serologically positive animals to SAT and RBPT,
except one sample obtained from ovine milk which
was negative by SAT and RBPT (Table I). All milk
samples positive to PCR and culture were positive
to MRT (Table I, Figs 1 and 2). PCR assay succeeded
in amplifying B. melitensis DNA from one camel milk
sample (Fig. 2). On the other hand, direct culture
methods detected two isolates from sheep milk,
which were negative by PCR assay (Table I). Isolated
strains recovered from cows' milk proved to be
B. abortus biovar 1, while isolated strains cultured
from sheep and goats milk were typed as B. melitensis
biovar 3.
Fifty cows' milk samples obtained from a Brucellafree herd were used as control and these all tested
negative by PCR and culture techniques.
PCR limit of detection in inoculated milk
To assess the limit of detection of the employed PCR
assay in milk, sterile bovine milk was inoculated with
a known number of either B. abortus 544 (B. abortus
biovar reference strain) and B. melitensis 16 M
(B. melitensis biovar 1 reference strain) and processed
subsequently for PCR amplification and culture.
A positive PCR result was obtained with different
aliquots containing at least 100 CFU of B. abortus
302
THE VETERINARY JOURNAL, 163, 3
Table I
Serological, bacteriological and PCR results of milk samples taken from cattle, sheep, goats, and camels
Animal species
Animals (n)
PCR
SAT
52
Sheep
21
Goats
18
Camels
12
Total
103
MRT
Culture
ÿ
‡
ÿ
‡
ÿ
‡
ÿ
(n ˆ 29)
(n ˆ 23)
(n ˆ 10)
(n ˆ 11)
(n ˆ 13)
(n ˆ 5)
(n ˆ 1)
(n ˆ 11)
29
18
9
9
13
4
1
9
0
5
1
2
0
1
0
2
29
21
9
11
13
4
1
9
0
2
1
0
0
1
0
2
29
6
10
6
13
2
1
6
0
17
0
5
0
3
0
5
24b
0
10c
2
11
0
0
0
5
23
0
9
2
5
1
11
‡ (n ˆ 53)
92
11
97
6
73
30
47
56
‡
Cattle
RBPT
a
‡
ÿ
‡
ÿ
‡
ÿ
‡
ÿ
SAT: serum agglutination test; RBPT: Rose Bengal Plate test; MRT: Milk ring test; ‡: positive; ÿ: negative; aPositive samples
to the SAT were those with titres > 1/40 (50%); bAll isolated strains cultured from cows' milk were proved to be B. abortus
biovar 1; cAll isolated strains cultured from sheep and goats milk were proved to B. melitensis biovar 3.
Lane
1
2
3
4
5
6
7
8
9
10
Fig. 1.
PCR products amplified from B. melitensis DNA
extracted from infected sheep milk (Lanes 2±5) and goats
milk (Lanes 6±8); Lane 1: molecular weight marker;
Lane 9: negative control; Lane 10: positive control.
organisms per milliliter of milk, while direct culture
method detected up to 1000 CFU of the same
organism (Fig. 3). On the other hand, 1000 CFU
of B. melitensis organisms were detected per milliliter
of milk using both the PCR and direct culture
method (Fig. 4). However, amplification signals were
obtained in 30% of the aliquots containing up to
100 CFU of B. melitensis per milliliter of milk.
DISCUSSION
The reliability of the PCR assay was assessed by
determining the sensitivity and specificity of the
assay. It was evident from our results (Table I, Figs 1
and 2) that PCR assay detected more positive samples (n ˆ 53) from the milk of different animals
(except sheep) than the culture method (n ˆ 47).
Lane 1
2
3
4
5
6 7 8 9 10 11 12 13
Fig. 2.
PCR products amplified from B. abortus DNA
extracted from infected cows milk (Lanes 2±6); Lane 10:
PCR products amplified from B. melitensis DNA extracted
from infected camel milk; Lanes 9 and 11: negative camel
milk samples; Lane 1: molecular weight marker; Lanes 8
and 12: negative control; Lanes 7 and 13: positive controls.
This indicated that the sensitivity of the PCR was
higher than that of the culture method. The same
conclusion was reached by Leal-Klevezas et al. (1995)
and Romero et al. (1995) and may be attributed to
the fact that PCR detects living and dead organisms,
since it is based on detection of Brucella DNA, while
culture detects only living organisms. PCR could
detect fewer numbers of Brucella organisms per
millilitre of milk than could be detected by direct
culture. It is noteworthy to mention that all milk
samples collected from Brucella-free herds tested
negative by PCR; a finding which indicated satisfactorily the specificity of the assay.
The detection limit of B. abortus and B. melitensis in
milk was carried out using the same DNA-extraction
protocol and amplification process performed on
field samples according to the protocol adopted by
DETECTION OF BRUCELLA SPECIES IN MILK SAMPLES BY PCR
*Culture
+
+
+
–
–
Lane 1
2
3
4
5
6
*Culture
7
8
Lane
1
303
+
+
+
+
+
+
2
3
4
5
6
7
Fig. 3.
Visualization of PCR, amplified B. abortus DNA
on an agarose gel by ethidium bromide staining following
electrophoresis. Lane 1: molecular weight marker; Lanes
2±6: containing amplified DNA corresponding to 1 105,
1 104, 1 103, 1 102 and 1 101 CFU/mL of B. abortus
544 (reference strain) respectively, as determined by viable
counts; Lane 7: negative control; Lane 8: B. abortus-positive
control. *The results of B. abortus isolation from different
dilutions of milk samples.
Fig. 4.
Visualization of PCR amplified products of
B. melitensis DNA on an agarose gel by ethidium bromide
staining following electrophoresis. Lane 1: molecular
weight marker; Lanes 2 and 3: containing B. melitensis
DNA corresponding to 1 105 CFU/mL; Lanes 4 and 5:
containing B. melitensis DNA corresponding to 1 104
CFU/mL; Lanes 6 and 7: containing B. melitensis DNA
corresponding to 1 103 CFU/mL as determined by viable
count. *The results of B. melitensis isolation from different
dilutions of milk samples.
Leal-Klevezas et al. (1995). We found that a positive
PCR result was always obtained with different
aliquots containing at least 100 CFU of B. abortus
organisms per millilitre of milk, while direct culture
method detected up to 1000 CFU of the same
organism (Fig. 3). On the other hand, 1000 CFU of
B. melitensis organisms were detected per millilitre of
milk using both the PCR assay and the direct culture
method (Fig. 4). However, amplification signals were
obtained in only 30% of the aliquots containing up
to 100 CFU of B. melitensis per millilitre of milk. PCR
detected more positive cases from field samples of
cows' milk than from sheep milk (Table I), which
could be attributed to the fact that PCR detects lower
numbers of B. abortus than B. melitensis per millilitre
of milk (Figs 3 and 4). Similar results were obtained
by Romero et al. (1995), who detected approximately
170 CFU of B. abortus per millilitre of milk and 1700
CFU of B. melitensis per millilitre of milk. However, in
a recent study (Romero & Lopez-Goni, 1999), lower
numbers of Brucella organisms were detected in milk
using different extraction protocols. This indicated
that the sensitivity of PCR is primarily affected by the
effectiveness of the DNA-extraction protocol and the
amount of sample processed by the assay. Since low
numbers of Brucella organisms are able to transmit
the disease, the sensitivity of the PCR should be
increased, particularly in the ovine species, to detect
lower numbers of Brucella organisms in field
samples. It was evident from the results obtained that
Brucella organisms were not detected by culture or
PCR from all sero-positive animals and this was
expected, as the excretion of the organism in milk is
intermittent (Morgan & Mackinnon, 1979; Alton
et al., 1988). In the present study, the PCR did fail to
detect Brucella DNA from two ovine milk samples
from which the organism was isolated by direct culture. This could be attributed either to the presence
of PCR inhibitors in sheep milk, or to the lower limit
of detection of the PCR to B. melitensis.
The serological tests employed in this study (SAT
and RBPT) gave negative results in one sheep sample
from which the organism was detected by culture
and PCR assay. Similar results were obtained by LealKlevezas et al. (1995); who amplified B. melitensis
DNA from two milk samples out of 15 serologically
negative goats. The finding was, however, alarming
and substantiates the belief that the sensitivity of the
RBPT for examining sheep blood samples should be
increased (Alton et al., 1988). Although previous
recommendations (Blasco et al., 1994) to increase
the sensitivity of the RBPT in examining ovine serum
by increasing the amounts of tested serum from
0.030 mL to 0.075 mL, this is still not satisfactory.
Nevertheless, modification of RBPT antigen by
decreasing the pH or decreasing cell concentrations
304
THE VETERINARY JOURNAL, 163, 3
of the antigen may serve to increase the sensitivity of
the test using ovine sera.
It was evident that all samples which were positive
to culture and PCR assay were positive also to the
MRT. Although the MRT is sensitive, it detected
lower numbers of positive animals (73) than those
detected by blood serological tests. This could be
due either to the stage of infection where the levels
of the agglutinins were not high enough to be
excreted in milk, or to the irregularity in the filtration of the agglutinins from blood to milk (Pat &
Panigrahi, 1965).
Biotyping of the isolated strains showed that all
strains isolated from cattle were B. abortus biovar 1,
and all strains isolated from sheep and goats were
B. melitensis biovar 3 (Table I). Although no Brucella
isolates were obtained from camels' milk, PCR
detected B. melitensis DNA from one sample of camel
milk (Fig. 2). Detection of B. melitensis from camel
milk has public health significance, since farmers
from nomadic areas believe that raw camel and goat
milk has a curative effect on the digestive system
(Elberg, 1981; Alton, 1990; Leal-Klevezas et al., 1995)
and its consumption has led to a high percentage of
human brucellosis (Kiel & Khan 1989). Isolation of
B. melitensis from camels has been previously reported from Mediterranean countries (Hamdy, 2000;
Radwan et al., 1995).
The DNA extraction protocol described in this
study has been shown to be successful in the detection of Brucella organisms in milk samples taken
from infected cattle, sheep, goats and camels.
According to the available data this is the first report
of the application of PCR for detection of Brucella
organisms from camels' milk. The PCR assay described has several advantages over the traditional culture method, since it was shown to be faster and
more sensitive. In addition, the amount of milk used
for PCR is much more smaller than that required for
bacteriological methods. Furthermore, live Brucella
organisms are not necessary for the assay, which
enhances safety and prevents the risk of transmission
of the disease to laboratory workers. Although we
recommend the use of the PCR assay as a supplemental diagnostic tool for detection and identification of Brucella organisms in milk of different animal
species, the need for new additional primers for
detection of different biovars of Brucella species, and
to differentiate the vaccinal strains from field strains,
is desirable to achieve the utmost benefits from the
PCR assay.
REFERENCES
ALTON, G. G. (1990). Brucella melitensis, 1887 to 1987. In
Animal Brucellosis, eds. K. Nielsen and J. R. Duncan.
Boston: CRC Press Inc.
ALTON, G. G., JONES, L. M., ANGUS, R. D. & VERGER, J. M.
(1988). Techniques for the Brucellosis Laboratory. Paris:
INRA.
AMIN, A. S., HUSSEINEN, H. S., RADWAN, G. S., SHALABY, M. N.
& EL-DANAF, N. (1995). The polymerase chain reaction
assay as a rapid and sensitive test for detection of
Brucella antigen in field samples. Journal of the Egyptian
Veterinary Medical Association 55, 761±7.
BLASCO, J. M., GARIN-BASTUJI B., MARIN, C. M., GERBIER, G.,
FANLO, J., JIMENEZ DE BAGUES, M. P. & CAU, C. (1994).
Efficacy of different rose bengal and complement fixation antigens for the diagnosis of Brucella melitensis
infection in sheep and goats. Veterinary Record 134,
415±20.
BRICKER, B. J. & HALLING, S. M. (1994). Differentiation of
Brucella abortus bv. 1, 2, and 4, Brucella melitensis, Brucella
ovis, and Brucella suis bv. 1 by PCR. Journal of Clinical
Microbiology 32, 1660±6.
BRIKENMEYER, L. G. & MUSHAHWAR, I. K. (1991). DNA probe
amplification methods. Journal of Virological Methods 35,
117±26.
CORBEL, M. J. (1985). Recent advances in the study of
Brucella antigens and their serological cross-reaction.
Veterinary Bulletin 55, 927±42.
CORDES, D. O. & CARTER, M. E. (1979). Persistency of
Brucella abortus infection in six herds of cattle under
brucellosis eradication. New Zealand Veterinary Journal
27, 255±9.
DIAZ, R. & MORIYON, I. (1989). Laboratory techniques
in the diagnosis of human brucellosis. In Brucellosis:
Clinical and Laboratory Aspects. eds. E. J. Young and
M. J. Corbel. pp. 73±83. Boca Raton, Fla, CRC Press, Inc.
ELBERG, S. S. (1981) Editor. A guide to the diagnosis, treatment
and prevention of human brucellosis. Unpublished document VPH/81.31 Rev. 1. Geneva: WHO.
FARINA, R. (1985). Current serological methods in
B. melitensis diagnosis. In Brucella melitensis, eds
J. M. Verger and M. Plommet. Nouzilly: INRA.
FEKETE, A., BANTLE, J. A., HALLING, S. M., & SANBORN, M. R.
(1990). Preliminary development of diagnostic test for
Brucella using PCR. The Journal of Applied Bacteriology 69,
216±27.
FEKETE, A., BANTLE, J. A. & HALLING, S. M. (1992). Detection of Brucella by polymerase chain reaction in bovine
fetal and maternal tissues. Journal of Veterinary Diagnostic
Investigation 4, 79±83.
HAMDY, M. E. R. (2000). Evaluation of indirect ELISA test
in diagnosis of brucellosis in camels. Veterinary Medical
Journal, Giza 48, 467±77.
HERMAN, I. & RIDDER, H. D. (1992). Identification of
Brucella spp. by using the polymerase chain reaction.
Applied and Environmental Microbiology 58, 2099±101.
HUBER, J. D. & NICOLETTI, P. (1986). Comparison of the
results of card, Rivanol, comlement-fixation and milk
ring test with the isolation rate of Brucella abortus from
cattle. American Journal of Vetrinary Research 47, 1529±31.
DETECTION OF BRUCELLA SPECIES IN MILK SAMPLES BY PCR
KIEL, F. W. & KHAN, M. Y. (1989). Brucellosis in Saudi
Arabia. Journal of Social Science and Medicine 29,
999±1001.
LEAL-KLEVEZAS, D. S., MARTINEZ-VAZQUEZ, I. O., LOPEZMERINO, A. & MARTINEZ-SORIANO, J. P. (1995). Singlestep PCR for detection of Brucella spp. from blood and
milk of infected animals. Journal of Clinical Microbiology
33, 3087±90.
MANIATIS, T., FRIT, E. F. & SAMBROOK, J. (1984). Molecular
Cloning: a Laboratory Manual. Cold Spring Harbor, NY:
Cold Spring Harbor Laboratory.
MORGAN, W. J. B. & MACKINNON, D. J. (1979). Brucellosis.
In Fertility and Infertility in Domestic Animals, ed.
J. A. Laing. pp. 171±98 ELBS, London: Bailliere Tindall.
PAT, K. V., & PANIGRAHI, B. (1965). Comparative study of
the Abortus Bang Ring Test and SAT in cows and buffaloes. A preliminary report. Indian Veterinary Journal 42,
748±50.
RADWAN, A. I., BEKAIRI, S. I., MUKAYEL, A. A., AL-BOKMY,
A. M., PRASAD, P. V., AZAR, F. N., & COLOYAN, E. R.
305
(1995). Control of Brucella melitensis infection in a large
camel herd in Saudi Arabia using antibiotherapy and
vaccination with Rev. 1 vaccine. Revue Scientifique et
Technique, Office International Des Epizooties (OIE) 14,
719±32.
ROLFE, D. C. & SYKES, W. E. (1987). Monitoring of dairy
herds for Brucella abortus infection when prevalence is
low. Australian Veterinary Journal 64, 97±100.
ROMERO, C. & LOPEZ-GONI, I. (1999). Improved method for
purification of bacterial DNA from bovine milk for
detection of Brucella Spp. by PCR. Applied and Environmental Microbiology 65, 3735±7.
ROMERO, C., PARDO, M., GRILLO, M. J., DIAZ, R., BLASCO, J. M.
& LOPEZ-GONI, I. (1995). Evaluation of PCR and indirect-ELISA on milk samples for the diagnosis of brucellosis in dairy cattle. Journal of Clinical Microbiology 33,
3198±200.