Pasteurella multocida causes a wide range of important diseases
in domestic animals, being responsible for pneumonia in cattle and sheep
(Chanter and Rutter, 1989; Frank, 1989) and haemorrhagic septicemia in
cattle and buffalo (Carter and de Alwis, 1989). Five serogroups (A, B,
D, E and F), based on an capsular antigens were recognized in P. multocida
(Harper et al., 2006). Townsend et al. (2001) developed
a multiplex PCR system based on capsular genes loci amplification as the
alternative way to the IHA test.
Mutters et al. (1985) classified P. multocida into three
subspecies on DNA-DNA hybridization and named them as P. multocida
subspecies multocida, P. multocida subspecies septica,
P. multocida subspecies galicida and also demonstrated that
dulcitol and sorbitol fermentation could be used to distinguish these
three subspecies. An advanced method called microplate fermentation was
described by Blackall et al. (1995) as an advanced technique for
phenotyping of P. multocida.
Recently, a new method was suggested for epidemiological studies in P.
multocida on the basis of a method that is called virulence genotyping
by its advocators examined for capsular and 14 virulence associated genes
(Ewers et al., 2006). This research group showed association between
ToxA and swine diseases as well as TbPA and pftA and bovine
diseases (Ewers et al., 2006). According to our knowledge association
between Pfha1, HgbB, TbPA and ToxA genes and
disease incidence in sheep till remains unclear. In order to achieve this
goal, In this study we investigated the prevalence of Pfha1, HgbB,
TbPA and ToxA genes along with genotype and phenotype of
P. multocida in healthy and diseased sheep.
MATERIALS AND METHODS
This study was carried out in North West of Iran covering East Azerbaijan,
West Azerbaijan and Ardabil provinces. A total from 518 samples were collected
from nasal discharge, pneumonic lung swab of diseased sheep (388) and
nasal swab of clinically healthy ones (130), starting from January 2005
till February 2006 (Table 1). Specimens collected from
diseased animals were fallen into two groups; group A: those collected
from nasal discharge of sheep with respiratory infection (269) and group
B: those collected from pneumonic lung of slaughtered sheep (119). All
samples were plated onto 10% Sheep Blood Agar (SBA) and incubated at 37°C
Bacterial Isolation and Identification
Culture and morphological identification of suspected P. multocida
isolates were carried out according to standard biochemical tests (Barrow
and Feltham, 2006). Following incubation at 37°C for 24 h, small glistening
mucoid dewdrop-like colonies were appeared on blood agar medium. Microscopic
observations revealed that all isolates were gram-negative coccobacilli
and biochemical analysis confirmed that the isolates were indol, catalase
and oxidase positive but, citrate, MR, VP and gelatin liquefaction negative.
Growth test on MacCankey agar was negative with no motile and non-hemolytic
effects on blood agar. The cells were stored in Brain Heart Infusion (BHI)
with 30% glycerol at –70°C.
All isolates were subjected to full biochemical characterization based
on microplate fermentation method described by Blackall et al.
(1995). Briefly, characterization was done using fermentation of ten carbohydrates
and results of ornithine decarboxylase (ODC) and Ortho-nitrophenyl-b-D-
||Primers for detection of virulence genes and capsular
*: These primers were originally developed by mentioned
references, but were modified in this study to improve PCR amplification
||Distribution and source of collected of P. multocida
galactopyranoside (ONPG) tests. The carbohydrates used in this part included
L-arabinose, dulcitol, D-glucose, D-lactose, maltose,D-mannitol, D-sorbitol,
D-sucrose, D-trehalose and D-xylose.
Capsular Typing by Multiplex-PCR
Capsular typing was conducted with multiplex PCR (Townsend et al.,
2001) using further modified primers designed on the basis of capsular
gene sequences (Table 2). The PCR amplification was
conducted directly on bacterial culture stock composed of BHI (70%) and
glycerol (30%) without genomic DNA extraction step. Each 25 μL reaction
contained 0.4 μL bacterial glycerol stock as DNA template, 1 U Taq
DNA polymerase, 3.2 mM from each primer, 200 μM of each dNTP, 1 x
PCR buffer and 2 mM MgCl2. Amplification was carried out for
35 cycles, each cycle consisting of DNA denaturation at 94°C for 30
sec, annealing at 54°C for 30 sec, extension at 72°C for 30 sec.
The cycles were preceded by an initial denaturation at 94°C for 5
min and followed by a final extension at 72°C for 5 min. The resulting
PCR products were electrophoresed in 2% agarose gel and stained with ethidium
bromide and imaged. Distilled water without any DNA was used as negative
Virulence Genes Detection Using PCR Analyses
Virulence genes detection was carried out using a new multiplex PCR
for Pfha1, HgbB, TbPA and ToxA genes. For
this multiplex PCR system, we designed new or modified previously developed
primers for each gene to prepare suitable band size and PCR program (Table
2). This method comes in company with satisfied results in compare
with PCR method for each gene separately. For all PCR reactions, 0.8 μL
of bacterial culture stock composed of BHI (70%) and glycerol (30%) without
genomic DNA extraction step were taken as template DNA and added to the
reaction mixture (50 mL) containing 3.2 mM of each primer pair, 200 μM
from the four dNTP, 5 μL of 10 x PCR buffer, 1.5 μL of 50 mM
Magnesium chloride and 1 U of Taq-Polymerase. The samples were subjected
to 35 cycles of amplification in a thermal cycler. The primers used in
this study are listed in Table 2. Each cycle consisting
of DNA denaturation at 94°C for 45 sec, annealing at 54°C for
50 sec, extension at 72°C for 50 sec. The cycles were preceded and
followed by an initial denaturation at 95°C for 5 min and final extension
at 72°C for 10 min, respectively. Amplification products were analyzed
by gel electrophoresis on a 1% agarose gel, stained with ethidium bromide
and photographed at UV exposure. Statistical analyses were performed using
software SPSS 12.0.
Of the 518 specimens originated from different healthy (130) and diseased
(388) hosts and investigated for P. multocida, 47 (9.07%) samples
were positive. Of 130 specimens collected from healthy sheep 11 (8.46%)
and of 388 specimens collected from diseased hosts 36 (9.27%) were identified
as positive samples (Table 1).
All positive samples (47 isolates) were biochemically phenotyped using
microplate fermentation method. As shown in Table 3,
11 (23.4%) samples displayed biovar seven fermentation phenotype.
||Biochemical property of isolates of P. multocida
||Capsular typing of isolates of P. multocida
Consequently, these isolates were characterized as P. multocida subspecies
septica. The remaining 36 (76.6%) samples were assigned as P.
multocida subspecies multocida. According to fermentation
results, P. multocida subspecies multocida isolates fell
into six biovars. 2 (4.25%) isolate was classified as biovar two, 10 (21.27%)
biovar three, 9 (19.14%) biovar four, 1 (2.1%) biovar five, 13 (27.65%)
biovar six and 1 (2.1%) biovar eleven. This observation indicates that
the majority of P. multocida isolates belong to biovar six and
seven with 27.65 and 23.4% prevalence, respectively. Additionally, the
high prevalence of biovars 4 (100%), 6 (92%) and 7 (81.8%) imply role
of this biovars in disease status.
Capsular Typing by Multiplex-PCR
Capsular genotyping was conducted based on amplification of five different
capsular groups using multiplex PCR in the presence of each capsule`s
specific primers. A pair of P. multocida specific primers was also
added into the reaction for species confirmation of the isolates.
The presence of a DNA band with about 460 bp size further established
the identification of the isolates as P. multocida. As seen, amplified
DNA products of ~1044 and ~657 bp corresponding to P. multocida
capsular groups A and D were observed, respectively. The band sizes of
~760, ~511 and ~854 bp expected to be produced corresponding to P.
multocida capsular serogroups B, E and F, respectively were not observed.
As shown in Table 4, two genotypes (A and D) were found
among both Multocida and Septica subspecies. Of the samples,
39 (83%) isolates were classified as capsular type A and 3 (6.38%) as
type D. No amplicon addressing to groups B, E and F was found. All two
types (A and D) were found among healthy sheep and sheep with disease
status. Five isolates (10.6%) from sheep with disease status were untyped.
Virulence Genes Detection Using PCR Analyses
Virulence genes detection was conducted based on amplification of
four virulence factor genes using multiplex PCR in the presence of specific
primers. Amplification of DNA bands with about 275, 540, 728 and 846 bp
sizes was addressed to the presence of Pfha1, HgbB, TbPA
and ToxA genes in the isolates, respectively.
As shown in Table 5, TbPA (69.4%) and ToxA
(72.2%) genes have the highest prevalence rate among diseased cases. We
did not detect any ToxA(+) isolate among healthy sheep and Phfa1(+)
isolate among diseased samples sheep. Additionally, prevalence rate of
TbPA among healthy isolates was low. The prevalence of ToxA
and TbPA among biovars 4, 6 and 7 are higher than other biovar.
Prevalence ToxA among biovars 4, 6 and 7 is 84.4, 80 and 80% and
TbPA 84.6, 70 and 90%, respectively.
||Virulence factor detection of isolates of P. multocida
This study as the first report describes phenotyping, capsular typing
and virulence factor profile of ovine P. multocida. The results
of capsular typing except a few items are in accordance with some of previous
studies described in the literature (Fussing et al., 1999; Weiser
et al., 2003; Ewers et al., 2006). Two genotypes (A and
D) were detected among the isolates. Type D was found only in diseased
cases, while type A was found in diseased and healthy samples. Prevalence
rate of type A (83%) was higher than type D (6.38%). The results suggest
that type A strain are the most common in Iran independent of disease
In this study we show the remarkable high prevalence of P. multocida
ToxA(+) (86.2%) in nasal swabs of sheep with respiratory infection
that is noticeable. Other groups reported that dermonecrotic toxin encoded
by ToxA gene, is expressed mainly by serogroup D strains responsible
for the clinical and pathological signs of atrophic rhinitis (Harper
et al., 2006) and with disease in goats (Baalsrud, 1987; Zamir-Saad
et al., 1996). Although Ewers et al. (2006) found high prevalence
of ToxA(+) strains among sheep population, but due to some limitation
they did not identify the association of this gene with disease status.
We found high prevalence of ToxA(+) strains among sheep with respiratory
disorder (nasal discharge and pneumonia) (p<0.05) and absence of this
gene in healthy sheep. This finding could address the important role of
P. multocida ToxA(+) in respiratory infection in sheep. In addition,
we found ToxA(+) strains among capsular type A(i.e., 24 out of
26 ToxA(+) isolate display capsular type A genotype). 2 out of
3 type D strains isolated from disease animals showed ToxA(+)(data
not shown). However high prevalence of this capsular type A ToxA(+)
suggested a more important role of this strain, in sheep diseases, which
should be kept mind in future study.
On the other hand, some papers reported the isolation of P. multocida
type D ToxA(+) from bighorn sheep introduced by the feral goats.
it is suggested that goat may be served as reservoir of Pasteurella
strains that is likely to be virulent in bighorn sheep (Rudolph et
al., 2003; Weiser et al., 2003). Considering very close relation
and nose to nose contact between sheep and goat in Iranian small ruminant
herds; it is though that P. multocida type A ToxA(+) may
be introduced to domestic sheep population through domestic goats similar
to P. multocida type D, ToxA(+) to bighorn sheep through
feral goats in the wild life. Comparative study based on genetic approach
is necessary to clarify this relationship.
Two iron acquisition related genes, TbPA and HgbB, were
studied in this study. The role of iron in pathogenesis of P. multocida
is important, two independent non sidrophore-mediated acquisition
of iron mechanism have been identified in P. multocida. The first
mechanism involves iron-binding proteins expressed on the outer membrane
of the bacterial cell, interacting directly with host iron binding glycoproteins.
The second mechanism includes bacterial proteins that bind hemoglobin
and hemoglobin complexed to the host glycoprotein (Cox et al.,
2003). Two iron acquisition related genes, TbPA and HgbB,
were studied in this paper associated to first and second mechanism of
iron acquisition, respectively. Both of these genes were introduced as
and epidemiological markers (Ewers et al., 2006).
A high prevalence of TbPA among P. multocida isolates from
respiratory infection in sheep was found as well. Previous studies reported
the presence of TbPA in bovine isolates of P. multocida
associated with pneumonia and hemorrghic septicemia (Veken et al.,
1994; Ogunnariwo and Schryvers, 2001). Present results confirmed the presence
of a significant association between this gene and ovine disease (p<0.05).
The total presence of HgbB gene in diseased sheep is lower than
its prevalence rate in healthy animals. In contrast of bovine strain,
this gene may be not important in ovine disease and it is also not valuable
as epidemiological marker. However, present data show different results
between animal with nasal discharges and pneumonic lungs. The absence
of HgbB gene in isolates originated from animals with nasal discharge
and high prevalence of that gene in pneumonic animals is noticeable and
further studies are desired.
Additionally, despite of the report describing Pfha1 as an epidemiological
marker in cattle isolates, prevalence rate of Pfha1 gene among
P. multocida isolates from sheep was very low. This finding indicates
that Pfha1 is not important in virulence of sheep isolates and
is not likely to be a suitable gene candidate in epidemiological studies
in sheep. Probably the low rate of Pfha1 gene among sheep population
especially infected ones paled the importance of filamentous hemagglutinin
encoded by this genes in colonization of ovine strain of P. multocida
and there might be another factor such as type 4 fimbriae (PtfA) that
high prevalence among ovine isolates has an important role in this processes
(Ewers et al., 2006).
The PCR amplification using bacterial stockculture as the template DNA
in this study provided satisfied results that has already been results
obtained from colony touched PCR amplification used by Townsend et
al. (2001) and DNA extraction used by Lichtensteiger et al.
(1996) and Ewers et al. (2006). This experience shows that direct
PCR capsular typing utilizing P. multocida glycerol stock
is a reliable and applicable method for capsular typing that also significantly
decreases the PCR typing time due to deletion of re-culture bacterial
Biochemical phenotyping, according to biotyping system of Blackall et
al. (1995) allowed identification of nine biovar groups. This report
is first experiment of microplate fermentation method about ovine isolates,
showing the higher prevalence of biovars 3, 6 and 7. High percentage of
biovars six and four (p<0.05) among disease cases showed the relation
between these biovars and disease status. This finding is not in accordance
with the results of other studies from Australia reported biovar 3 has
the highest among avian and swine isolates (Fegan et al., 1995;
Blackall et al., 1997).
No relevance was found between biochemical properties and capsular PCR
typing. But high prevalence of TbPA and ToxA genes between
biovars six and seven is higher than the other biovars which have association
with ovine disease. Although the number of samples in this study was not
too high, but according of the results outcome from this collection, in
spite of the important role of capsular and epidemiological markers in
swine and bovine disease (Ewers et al., 2006), it seems that in
sheep the contemporary use of phenotyping and virulence epidemiological
marker gene is useful method for epidemiological studies.
In summary, among four genes that is showed as epidemiological markers
in swine and bovine diseases, only two genes, TbPA and ToxA,
have significant association with diseases in sheep. Also, in attention
to low variation in capsular typing system among ovine strains of P.
multocida, it suggested that the use of alternative system with high
variation among ovine isolates instead of capsular typing in epidemiological
studies. Although phenotyping system is demonstrated as a suitable alternative
system, more studies are necessary to introduce other systems on based
According to the result of this article, we can conclude that the role
of TbPA and ToxA genes in epidemiological study of sheep
respiratory disorder is significant. Additionally, P. multocida type
A ToxA(+) plays an important role in sheep disease.
We are very grateful to Professor P. J. Blackall for his generous and
helpful comments during this study and Mr. R. Azarbaijani and J. Dolgari
Sharaf for kind technical assistance. Many thanks to Islamic Azad University
(IAU) of Shabestar and Tabriz University of Medical Sciences for supporting