Food Control 35 (2014) 94e100

Contents lists available at SciVerse ScienceDirect

Food Control
journal homepage: www.elsevier.com/locate/foodcont

Prevalence of Yersinia enterocolitica from food and pigs in selected

states of Malaysia
Lai Kuan Tan a, b, Peck Toung Ooi c, Kwai Lin Thong a, b, *
a
Microbiology Division, Institute of Biological Sciences, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Laboratory of Biomedical Science and Molecular Microbiology, UMBIO Research Cluster, Institute of Graduate Studies, University of Malaya,
50603 Kuala Lumpur, Malaysia
c
Clinical Veterinary Studies, Faculty of Veterinary Medicine, University Putra Malaysia, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study was to determine the prevalence of Yersinia enterocolitica and its bioserotypes from
Received 17 April 2013 food and pigs in Malaysia. Fifty-eight raw porcine (raw pork meat, internal organs and other parts) and
Received in revised form 48 non-porcine food (raw beef, poultry products, seafood, vegetables, tofu, and pasteurised milk) from
18 June 2013
wet markets located in Kuala Lumpur, Selangor, Perak, and Pahang were examined for the presence of
Accepted 25 June 2013
Y. enterocolitica. Specimens (nasal, oral and rectal swabs) from 165 pigs (from nine farms) located at
central and northern parts of Malaysia were also collected for Y. enterocolitica detection. Presumptive
Keywords:
isolates were characterised biochemically and further confirmed by PCR. Out of 58 raw porcine food,
Food microbiology
Pig
Y. enterocolitica was detected in 7 (12.1%) samples in which raw pork meat (whole meat) had the highest
Pork prevalence 5/21 (23.8%), followed by raw pork liver 1/5 (20.0%) and raw pork intestine 1/8 (12.5%). No
Yersinia enterocolitica Y. enterocolitica was isolated from the 48 non-porcine foods. Overall, two pathogenic (bioserotypes 3
variant/O:3 and 1B/O:8) and one non-pathogenic (bioserotype 1A/O:5) Y. enterocolitica strains were
isolated from food. Out of 165 pigs examined, 3 (1.8%) pigs were carriers for Y. enterocolitica. All 3 pigs
were asymptomatic grower pigs from Penang, carried Y. enterocolitica bioserotype 3 variant/O:3. Post-
enrichment PCR approach gave a higher prevalence, 60.3%, 41.7% and 27.9% for porcine food, non-
porcine food and pigs, respectively. Both pathogenic and non-pathogenic Y. enterocolitica were present
in our domestic pigs and food. Improper food handling and processing may cause cross contamination of
this pathogen to humans, affirms a potential risk for public health.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Young children and infants are the most susceptible age group
(Rosner, Stark, & Werber, 2010).
Yersinia enterocolitica is a bacterium which belongs to Enter- In United States, it is estimated that Y. enterocolitica causes over
obacteriaceae that is widely found in natural environments. It is 115,000 infections annually (Scallan et al., 2011). In Europe, there
psychrotrophic and has the capability to survive and multiply in were 6776 reported yersiniosis cases in humans during 2012
cold (Annamalai & Venkitanarayanan, 2005; Neuhaus, Francis, (European Food Safety Authority & European Centre for Disease
Rapposch, Görg, & Scherer, 1999). It is enteropathogenic and cau- Prevention and Control, 2012). Reports on the incidence of yersi-
ses gastrointestinal problems such as acute enteritis with fever, niosis in Southeast Asian countries are few. However, in the recent
bloody diarrhoea and pseudo appendicitis, which frequently leads report of Ananchaipattana et al. (2012a, 2012b), Y. enterocolitica was
to unnecessary laparotomy in humans (Vlachaki, Tselios, Tsapas, & reported present in some Thai food (beef, shrimp and tofu). This
Klonizakis, 2007). Y. enterocolitica is notified as the third most suggests a risk of human infection when such contaminated food is
important foodborne enteric pathogen in Europe after campylo- consumed in this area.
bacteriosis and salmonellosis (European Food Safety Authority & Y. enterocolitica is ubiquitous in the nature and is routinely
European Centre for Disease Prevention and Control, 2012). isolated from animals, food and environment (Fredriksson-Ahomaa
& Korkeala, 2003). Among the sources, swine is reported as a major
reservoir for Y. enterocolitica. The bacterium is often present in the
* Corresponding author. Microbiology Division, Institute of Biological Sciences, oral cavity of pigs especially tonsils, intestinal content, faeces and
Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia. Tel.: þ60 3
lymph nodes (Fondrevez et al., 2010; Gutler, Alter, Kasimir,
79675836; fax: þ60 3 79675908.
E-mail address: [email protected] (K.L. Thong).
Linnebur, & Fehlhaber, 2005; Liang et al., 2012; Nesbakken,

0956-7135/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodcont.2013.06.053
 L.K. Tan et al. / Food Control 35 (2014) 94e100 95

Eckner, Hridal, & Rrtterud, 2003). Besides the pigs, strains of All raw food in this study were transferred in sterile plastic bags
Y. enterocolitica have been frequently isolated in raw pork too and transported in ice box to the laboratory.
(European Food Safety Authority, & European Centre for Disease
Prevention and Control, 2012). This is due to the cross contami- 2.1.2. Isolation and enumeration of Y. enterocolitica from raw
nation of the organism via oral cavity, faeces, and intestinal con- porcine food
tents during slaughtering, cutting, further processing and Food samples were analysed for the presence of Y. enterocolitica
distribution of fresh pork (Martínez et al., 2011; Terentjeva & by conventional culture methods and post-enrichment polymerase
Berzins, 2010). Other possible food vehicles of yersiniosis are chain reaction (PCR) screening. Enumeration of Y. enterocolitica was
ruminant and ruminant products (Fukushima, Hoshina, Itogawa, & performed using a 3  3 most probable number (MPN) method.
Gomyoda, 1997), poultry (Dallal et al., 2010), vegetables Five g of raw food sample was cut into small pieces, added to
(Xanthopoulos, Tzanetakis, & Litopoulou-Tzanetaki, 2010), milk and 45 ml of selective enrichment broth in sterile plastic bag and
dairy products (Yucel & Ulusoy, 2006), ready-to-eat food homogenised manually by hand. Enrichment broths used were
(Xanthopoulos et al., 2010) and chitterlings (Lee et al., 1990). phosphate buffered saline (PBS, Sigma, Germany), Yersinia selective
In Malaysia, there is limited study on Y. enterocolitica and the enrichment broth according to OSSMER (YSEO, Merck, Germany),
bacterium is not routinely isolated. The first case of human yersi- and irgasan-ticarcillin-potassium chlorate (ITC) broth [ITC broth
niosis in Malaysia was reported by Jegathesan, Paramasivam, base (Fluka, Germany) supplemented with ticarcillin supplement
Rajagopalan, & Loo (1984) where Y. enterocolitica serotype O:3 (Fluka) and potassium chlorate supplement (Fluka)]. Food ho-
was isolated from a 34-year-old female. The only food related mogenates in ITC and PBS were incubated at 25  C for 2 days and
prevalence report in Malaysia was from unpublished study of 4  C for 3 weeks, respectively, and food homogenate in YSEO was
Dzomir (2005), Y. enterocolitica bioserotype 1A/O:52,53 and 1A/ used for MPN enrichment for food safety enumeration.
O:41,42 were isolated from beef burger and chicken burger meats. Food particles in YSEO were left to settled down and the fluid
Due to the limited study of this bacterium in Malaysia, the potential was dispensed into a 3  3 MPN system consisting of 10 ml of
complications of yersiniosis in the country remain unknown. The undiluted fluid in each of three 10 ml test tubes (level A), 1 ml of
actual incidence of this bacterium in various type of food is not well fluid in 9 ml YSEO broth in each three 10 ml test tubes (a 1:10
documented too. Therefore, the aims of this study were: (i) to dilution, level B), and 1 ml of a 1:10 dilution of the fluid in 9 ml
determine the prevalence of Y. enterocolitica from porcine food, YSEO broth in each three 10 ml test tubes (a 1:100 dilution, level C),
non-porcine food, and life pigs; and (ii) to determine the bio- incubated at 25  C for 18 h (Hudson et al., 2008).
serotype of Y. enterocolitica that is currently present in Malaysia. A loopful of each enriched samples was streaked onto
cefsulodin-irgasan-novobiocin (CIN) agar [Yersinia Selective Agar
2. Materials and methods Base supplemented with Yersinia Selective Supplement (Oxoid,
UK)] and incubated at 25  C for 24e48 h. In parallel, sample was
2.1. Isolation of Y. enterocolitica from raw porcine food plated onto CIN agar immediately after alkaline treatment in which
0.5 ml of enriched culture was transferred into 4.5 ml of 0.25%
2.1.1. Sampling potassium hydroxide (KOH): 0.50% sodium chloride (NaCl) solution
Between June 2010 to March 2011, 58 raw porcine samples were (ISO standard 10273:2003; Hudson et al., 2008; Johnson, 1998). The
sampled from wet markets at selected states in Peninsular Malaysia alkaline treatment is recommended for samples which are nor-
(Kuala Lumpur, Perak and Pahang). The raw porcine samples were mally highly contaminated with background microbiota.
further grouped into three categories as raw pork meats (n ¼ 25), Y. enterocolitica can resist weak alkaline treatment, while back-
raw pork internal organs (n ¼ 23) and other parts (n ¼ 10) (Table 1). ground microbiota such as Pseudomonas, Proteus and Serratia can

Table 1
Prevalence of Y. enterocolitica from raw porcine food determined by cultural method and post-enrichment PCR screening.

Food type No. of samples No of positives (%)a Isolation rate by strain typeb (%)

PCR Cultural Pathogenic Non-pathogenic

Raw pork meat 25 20 (80.0) 5 (20.0) 4/5 (80.0) 1/5 (20.0)

Whole meat 21 18 (85.7) 5 (23.8) 4/5 (80.0) 1/5 (20.0)
Minced meat 3 2 0 0 0

Raw pork internal organs 23 14 (60.9) 2 (8.7) 2/2 (100.0) 0

Liver 5 3 (60.0) 1 (20.0) 1/1 (100.0) 0
Intestine 8 7 (87.5) 1 (12.5) 1/1 (100.0) 0
Heart 5 3 (60.0) 0 0 0
Kidney 4 1 (25.0) 0 0 0
Throat 1 0 0 0 0

Other parts 10 1 (10.0) 0 0 0

Skin 4 1 0 0 0
Foot 2 0 0 0 0
Fat tissue 1 0 0 0 0
Ear 1 0 0 0 0
Eye tissue 1 0 0 0 0
Nose 1 0 0 0 0

Total 58 35 (60.3) 7 (12.1) 6/7 (85.7) 1/7 (14.3)

a
The values are the number of positives detected in either of the enriched samples (YSEO, ITC, and PBS).
b
The isolation rate refers to the number of positive samples determined by cultural methods.
96 L.K. Tan et al. / Food Control 35 (2014) 94e100

be suppressed (Schiemann, 1983). Typical Y. enterocolitica isolates (nasal, oral and rectal swabs) were collected from each pig and each
(red bull’s eyes) were picked and tested with biochemical tests, i.e. maintained in Cary-Blair transport medium (Oxoid, UK) in ice box
oxidase test, Gram staining, and citrate test. All Gram negative, before being processed in laboratory.
oxidase negative and citrate negative isolates were further char- Specimens were processed by two methods, i.e. (i) direct
acterised by using the API 20E strips (bio-MérieuxÒ SA, France). The streaking on selective agar plates and (ii) enrichment in ITC and PBS
strips were inoculated and read according to the recommendations broths (as described in 2.1) followed by streaking on selective agar
of the manufacturer, with the exception of incubation at 28  C plates. Selective agars used were CIN or modified CIN. Modified CIN
(Archer, Schell, Pennell, & Wick, 1987). Identity of Y. enterocolitica was made by adding 1% L-arginine (Sigma, Germany), 0.8 g/L ferric
isolates was confirmed by using PCR targeting Y. enterocolitica- ammonium citrate (BDH Prolabo, UK), 6.8 g/L sodium thiosulphate
specific 16S rRNA gene (Y1, 50 -AATACCGCATAACGTCTTCG-30 ; Y2, 50 - (BDH Prolabo), and 2.0 g/L DL-phenylalanine (Sigma) at pH
CTTCTTCTGCGAGTAACGTC-30 ) (Neubauer, Hensel, Aleksic, & Meyer, 7.4  0.02 into the CIN agar. The same procedures used for isolation,
2000). The virulence of Y. enterocolitica was determined by PCR confirmation and post-enrichment PCR screening as stated in 2.1.
targeting yst gene (forward, 50 -GTTAATGCTGTCTTCATTTGGAGC-30 ; except the MPN step.
reverse, 50 -GACATCCCAATCACTACTGACTTC-30 ) (Gómez-Duarte,
Bai, & Newell, 2009). Amplicons of selected PCR products were
2.4. Biogrouping and serotyping of Y. enterocolitica
purified using MEGAquick-spinÔ PCR & agarose gel DNA extraction
system (iNtRON Biotechnology, Korea) and sequenced to validate
The biogroup of Y. enterocolitica was determined by using
the identity.
biochemical tests as described by Wauters, Kandolo, & Janssens
(1987). The biochemical tests included lipase test, esculin hydro-
2.1.3. Post-enrichment PCR screening from enriched food
lysis, salicin utilisation, indole test, xylose utilisation, trehalose
homogenates
utilisation, nitrate reduction, pyrazinamidase test, b-D-glucosidase
DNA templates were prepared from 1 ml of each enriched
test, VogeseProkauer test, and DNase test. The serotype of
samples (YSEO, ITC, and PBS) using the cell lysate method. The
Y. enterocolitica was determined by using the O-Antisera “SEIKEN”
presence of Y. enterocolitica was screened by PCR using the
set purchased (DENKA SEIKEN Co., Ltd, Japan).
Y. enterocolitica-specific 16S rRNA gene primers (Neubauer et al.,
2000).
3. Results
2.1.4. MPN calculation
The three digits for each level in the 3  3 MPN system were 3.1. Prevalence of Y. enterocolitica from raw porcine food
determined based on the post-enrichment PCR screening results
(the YSEO enriched tubes). The MPN/g value was calculated using The prevalence of Y. enterocolitica from raw porcine food was
the Microsoft Excel spreadsheet provided by Institute of Environ- low by cultural method. Out of 58 food tested, 7 (12.1%) were
ment Science and Research (ESR), New Zealand (Hudson et al., naturally contaminated by Y. enterocolitica; i.e. raw pork meat
2008). The range over which these nine tubes MPN system oper- (whole meat) 5/21 (23.8%), raw pork liver 1/5 (20.0%) and raw pork
ates is between 0.30 MPN/g (LCI of 0.07 with one positive at level A) intestine 1/8 (12.5%) (Table 1). Twenty-six isolates of
to 44.84 MPN/g (UCI of 198.70). Y. enterocolitica (PCR confirmed) were isolated from the 7 positive
samples. Twenty-three isolates were pathogenic (from 4 raw pork
2.2. Isolation of Y. enterocolitica from non-porcine food meats, 1 raw pork intestine and 1 raw pork liver) and 3 isolates
were non-pathogenic (from 1 raw pork meat). Bioserotyping re-
Forty-eight non-porcine foods were purchased from wet mar- sults revealed that the pathogenic Y. enterocolitica isolates were
kets (located in Kuala Lumpur, Selangor, and Pahang) and examined either bioserotyped as 1B/O:8 (n ¼ 3) or 3 variant/O:3 (VP negative
for the presence of Y. enterocolitica. Food types purchased included variant strain, n ¼ 20), and all non-pathogenic Y. enterocolitica
raw vegetables (n ¼ 19), raw seafood (n ¼ 11), raw poultry products isolates were bioserotyped as 1A/O:5. All positive samples were
(n ¼ 9), raw beef (n ¼ 6), tofu (n ¼ 2), and pasteurised milk (n ¼ 1). from the same hawker stall.
The same procedures used for isolation, confirmation and post- MPN/g of Y. enterocolitica in food was determined by using the
enrichment PCR screening as stated in Section 2.1. were per- YSEO enriched tubes and the results are tabulated in Table 3. Our
formed except the MPN step. results showed that the concentration of Y. enterocolitica in the
positive samples ranged from <0.30 MPN/g to >43.84 MPN/g.
2.3. Isolation of Y. enterocolitica from pigs Although there was no specific requirement for the levels of

The presence of Y. enterocolitica from pigs in Malaysia was

Table 2
investigated during the period of October 2010 to September 2011. Prevalence of Y. enterocolitica from non-porcine food determined by cultural method
Random selection of participating pig farms was not possible in this and post-enrichment PCR screening.
observational study as selection was subjected to management and
Food type No. of samples No of positivesa (%)
practices of the commercial pig farms. A total of nine pig farms
located in three states in middle- and north- western part of PCR Cultural
Peninsular Malaysia, i.e., Selangor (Farms A, B, C), Perak (Farms D, E, Raw beef 6 4 (66.7) 0
F), and Penang (Farms G, H, I) were enrolled in this study. The three Raw poultry products 9 5 (55.6) 0
Raw seafood 11 5 (45.5) 0
states are the top three largest pig-producing states in Malaysia
Raw vegetables 19 6 (31.6) 0
(Department of Veterinary Services, Malaysia, 2011). A stratified Raw tofu 2 0 0
random sampling was performed in categorising the pigs based on Pasteurised milk 1 0 0
the health condition, i.e. healthy (pigs without disease symptoms)
and unhealthy (sick, weak, and dead). A total of 165 pigs were Total 48 20 (41.7) 0
selected (farms A, n ¼ 9; B, n ¼ 14; C, n ¼ 30; D, n ¼ 20; E, n ¼ 20; F, a
The values are the number of positive detected in either of the enriched samples
n ¼ 20; G, n ¼ 16; H, n ¼ 20; I, n ¼ 16) and three major specimens (YSEO, ITC, and PBS).
 L.K. Tan et al. / Food Control 35 (2014) 94e100 97

Table 3
The MPN/g values (calculated using the results of post-enrichment PCR) and the background information of food samples.

Isolation time Sample code Food type Location Bioserotype MPNa/g UCIb LCIc

June 2010 M1 Raw pork meat KLd 3 variant/O:3 <0.30 ee 0.07

June 2010 M3 Raw pork meat KL 3 variant/O:3 <0.30 e 0.07
July 2010 M5 Raw pork meat KL e 0.30 1.36 0.07
Aug 2010 M9 Raw pork meat KL e 0.30 1.36 0.07
Sept 2010 M12 Raw pork meat KL e 1.58 7.16 0.35
Sept 2010 M14 Raw pork meat KL e 7.60 34.44 1.68
Sept 2010 M13 Raw pork meat KL 3 variant/O:3 1.90 8.60 0.42
Sept 2010 M15 Raw pork meat KL e 0.62 2.81 0.14
Jul 2010 I2 Raw pork intestine KL e 0.30 1.36 0.07
Jul 2010 I3 Raw pork intestine KL e 0.36 1.63 0.08
Jul 2010 D1 Raw pork heart KL e 0.30 1.36 0.07
Jul 2010 D2 Raw pork heart KL e 0.60 2.73 0.13
Jul 2010 L1 Raw pork liver KL e 3.12 14.14 0.69
Aug 2010 D4 Raw pork heart KL e 18.98 86.04 4.19
Sept 2010 I5 Raw pork intestine KL e 0.30 1.36 0.07
Jan 2010 I7 Raw pork intestine KL e 1.90 8.60 0.42
Jan 2011 M16 Raw pork meat KL 1A/O:5 >43.84 198.70 e
Jan 2011 M17 Raw pork meat KL e >43.84 198.70 e
Jan 2011 M18 Raw pork meat KL e >43.84 198.70 e
Jan 2011 M19 Raw pork meat KL e >43.84 198.70 e
Jan 2011 M20 Raw pork meat KL e 2.71 12.29 0.60
Jan 2011 YE032 Raw pork liver KL 1B/O:8 0.36 1.63 0.08
Jan 2011 YE036 Raw pork meat KL 1B/O:8 0.36 1.63 0.08
Jan 2011 YE037 Raw pork intestine KL 3 variant/O:3 0.36 1.63 0.08
Mar 2011 K3 Raw pork kidney Taiping e 0.73 3.33 0.16
Mar 2011 L3 Raw pork liver Taiping e 4.57 20.72 1.01
a
MPN, most probable number.
b
UCI, upper confidence level.
c
LCI, lower confidence level.
d
KL, Kuala Lumpur.
e
Value cannot be calculated or identity cannot be determined. For the rest samples which were not mentioned in table, Y. enterocolitica was negative in both cultural and
post-enrichment PCR method in all nine tubes of the 3  3 YSEO enrichment tubes.

Y. enterocolitica in food under FDA Food Code (Lawley, Curtis, & indicated that Y. enterocolitica was most frequently found in nasal
Davis, 2012), majority of the positive samples had low MPN/g swabs, 29/165 (17.6%) followed by oral swabs, 25/165 (15.2%) and
values (18.98 MPN/g), except for four samples (>43.84 MPN/g). rectal swabs, 21/165 (12.7%) (Table 5). Overall, Y. enterocolitica was
A food sample was counted as PCR positive when either one of PCR detected in the swine hosts in the 3 states (Selangor, Perak, and
the 3 enriched cultures showed presence of amplicon. Post- Penang).
enrichment PCR detection showed a higher incidence of
Y. enterocolitica (35/58) as compared to the cultural methods (raw 3.4. Confirmation of Y. enterocolitica by DNA sequencing
pork meat, n ¼ 20; raw pork internal organs, n ¼ 14; skin, n ¼ 1).
DNA sequences of the 330 bp and 145 bp amplicons
3.2. Prevalence of Y. enterocolitica in non-porcine food (Y. enterocolitica 16S rRNA and yst genes, respectively) were
checked using the Basic Local Alignment Search Tool (http://blast.
Overall, no Y. enterocolitica was isolated via cultural methods ncbi.nlm.nih.gov/) and the analyses confirmed the isolates were
from the 48 non-porcine food tested (Table 2). However, the post- as Y. enterocolitica (99% and 96% homology, respectively).
enrichment PCR screening indicated Y. enterocolitica was present in
20/48 (41.7%) of food samples; i.e. 4/6 (66.7%) raw beef, 5/9 (55.6%) 4. Discussion
raw poultry products, 5/11 (45.5%) raw seafood, and 6/19 (31.6%)
raw vegetables. The results showed that PCR-based detection method was more
sensitive than the conventional cultural method, indicating that
3.3. Prevalence of Y. enterocolitica in live pigs use of selective medium might underestimate the real prevalence
of this pathogen in our local food and pigs. This result concurred
Based on cultural methods, only 3 out of 165 pigs (1.8%) har- with many other reports that the PCR detection was more sensitive
boured Y. enterocolitica (Table 4). All the positive pigs were healthy than the cultural method (Bhaduri, Wesley, & Bush, 2005;
grower pigs (asymptomatic) fed in the same pen from Farm I in Johannessen, Kapperud, & Kruse, 2000; Messelhäusser et al.,
Penang. All of them carried pathogenic Y. enterocolitica bioserotype 2011). For example, Messelhäusser et al. (2011) reported that 18%
3 variant/O:3. These isolates were from all 3 swab types (nasal, oral, of the pork samples analysed was PCR- positive as compared to 10%
rectal) of 2 carrier pigs. For the other carrier pig, the bacterium was positive by cultural method. However, these percentages are not
only identified in the nasal swab (total positive swabs ¼ 7). Overall, necessarily comparable due to the different methods used in the
a total of 72 isolates were isolated (nasal swab, n ¼ 28; oral swab, detection of Y. enterocolitica. The primers used in the PCR assay are
n ¼ 20,; rectal swab, n ¼ 24). specific which amplified the targeted gene of Y. enterocolitica. In
On the other hand, PCR detected a higher prevalence of contrast, detection by cultural method is less sensitive as the
Y. enterocolitica, in 46 pigs (27.9%), i.e. 11/22 (50.0%) healthy method is based on the physiology and biochemical activities of
growers, 14/44 (31.8%) unhealthy weaners, 20/68 (29.4%) healthy bacteria. Besides, DNA templates for PCR assay were prepared from
weaners, and 1/11 (9.1%) healthy growers (Table 4). Our PCR results bacterial cells that were concentrated from 1 ml of enriched
98 L.K. Tan et al. / Food Control 35 (2014) 94e100

Table 4
Prevalence of Y. enterocolitica based on the age and health condition of pigs determined by cultural method and post-enrichment PCR screening.

Health group na No of positives (%)

Selangor (n ¼ 53) Perak (n ¼ 60) Penang (n ¼ 52) Total positive

b
PCR C PCR C PCR C PCR C

Hc pigletd 4 0 0 ee e e e 0 0
UHf piglet 4 0 0 e e e e 0 0
H weanerg 68 4 (5.9) 0 8 (11.8) 0 8 (11.8) 0 20 (29.4) 0
UH weaner 44 5 (11.3) 0 8 (18.2) 0 1 (2.3) 0 14 (31.8) 0
H growerh 22 0 0 3 (13.6) 0 8 (36.4) 3 (13.6) 11 (50.0) 3 (13.6)
UH grower 11 0 0 1 (9.1) 0 e e 1 (9.1) 0
H finisheri 10 0 0 e e e e 0 0
H sow 2 e e 0 0 e e 0 0

Nj 165 9 (5.5) 0 20 (12.1) 0 17 (10.3) 3 (1.8) 46 (27.88) 3 (1.8)

a
n, number of pigs within each health group.
b
C, cultural method.
c
H, healthy.
d
Piglet, <1 month old.
e
e, sample not collected from this group.
f
UH, unhealthy.
g
Weaner, 1e2 months old.
h
Grower, 2e4 months old.
i
Finisher, 4e6 months old.
j
N, number of pigs from each state.

homogenate, thus increasing the probability in getting more DNA of decreased the workload for biochemical tests and the usage of API
Y. enterocolitica. Moreover, the PCR can detect all kinds of cells in 20E kits. Further evaluation of this modified CIN needs to be carried
regardless of dead cells or viable including non-culturable cells out.
which might not grow on artificial medium (Parker, 2002; Singh & We observed that the use of alkaline treatment prior to
McFeters, 1987). streaking onto the CIN or modified CIN agars eliminated a large
During the isolation of Y. enterocolitica from food and swine amount of background microbiota on the agar plates during the
specimens, a high amount of natural background microbiota had isolation, such as that reported in Schiemann (1983). However, we
been isolated as presumptive Y. enterocolitica from the CIN medium observed no difference between both methods in terms of positive
(colony appearance similar to Y. enterocolitica). Most of these bac- recovery according to the sample size (with alkaline treatment,
teria were later identified as Citrobacter spp., Providencia rettgeri, n ¼ 12; without alkaline treatment, n ¼ 12). This concurred with
Aeromonas hydrophila and Enterobacter cloacae by API 20E. The study of Ratnam, Looi, & Patel (1983) who noted that alkaline
similarity in the colonies’ appearance caused difficulty in isolation treatment did not significantly enhance the isolation rates.
of true Y. enterocolitica from these highly contaminated samples. In In this study, Y. enterocolitica was isolated from both raw porcine
the later stage in our study, we modified the CIN medium by adding food (12.7%) and pigs (1.8%). Three Y. enterocolitica bioserotypes
L-arginine, ferric ammonium citrate, sodium thiosulphate, and DL- were identified, 3 variant/O:3, 1B/O:8 and 1A/O:5. The results
phenylalanine to improve the differentiation of Y. enterocolitica confirmed that Y. enterocolitica was present in our local porcine
from these bacteria. Colony appearance of Y. enterocolitica on food and pigs. Interestingly, pathogenic Y. enterocolitica (3 variant/
modified CIN remained unchanged (Y. enterocolitica utilise neither O:3) was isolated from three healthy grower pigs. Similar finding
one of these substrates) as it was on CIN agar, but each of the was reported in Jos, Nigeria, where Y. enterocolitica was isolated
microbiota (Citrobacter spp., P. rettgeri, A. hydrophila, and E. cloacae) from healthy pigs (Okwori et al., 2009). The infected pigs appeared
appeared with a distinct colony appearance (utilised at least one of asymptomatic due to the colonisation of Y. enterocolitica in the
the substrates, and caused colour changes). We observed that lesser lymphoid tissue, particularly in tonsils (Horter, Yoon, &
Yersinia-like bacteria (but not Y. enterocolitica) were miss-identified Zimmerman, 2003). The colonisation caused the identification of
as presumptive Y. enterocolitica during the isolation since they were asymptomatic carrier animals difficult in disease control and/or
screened out on the modified CIN medium. This indirectly pathogen elimination. These asymptomatic pigs serve as food for

Table 5
Distribution of the number of positive swab samples of pigs from Selangor, Perak and Penang using post-enrichment PCR screening and cultural methods.

Swab type na No of positives (%)

Selangor (n ¼ 53) Perak (n ¼ 60) Penang (n ¼ 52) Total positive

PCR Cb PCR C PCR C PCR C

Nasal 165 5 (9.4) 0 14 (23.3) 0 10 (19.2) 3 (5.8) 29 (17.6) 3 (1.2)

Rectal 165 5 (9.4) 0 8 (13.3) 0 8 (15.4) 2 (3.8) 21 (12.7) 2 (1.2)
Oral 165 4 (7.5) 0 13 (21.7) 0 8 (15.4) 2 (3.8) 25 (15.2) 2 (1.2)

Nc 495 14 (2.8) 0 35 (7.1) 0 26 (5.3) 7 (1.4) 75 (9.9) 7 (1.4)

a
n, number of samples within each swab type.
b
C, cultural method.
c
N, number of pigs from each state.
 L.K. Tan et al. / Food Control 35 (2014) 94e100 99

humans when they are matured to be sold. Cross-contamination of Y. enterocolitica had no clinical importance in this region. However,
Y. enterocolitica from pigs’ oral cavity, intestine and faeces to meat is this pathogen was recently isolated from unpacked tofu
possible during the slaughtering and dressing operations through (Ananchaipattana et al., 2012a), beef and shrimp samples
the slaughtering tools and containers (Gill & Jones, 1995; (Ananchaipattana et al., 2012b) in Thailand. These recent reports
Nesbakken, 1988; Skjerve, Lium, Nielsen, & Nesbakken, 1998). Be- from Thailand indicated that there is a risk of human infection. In
sides that, cross contamination may happen during food storage. Malaysia, yersiniosis is rarely reported. The under-reporting in
For example, the interior surfaces of household refrigerators Malaysia could be due to several possibilities: (i) Malaysians prefer
(Jackson, Blair, McDowell, Kennedy, & Bolton, 2007) or surfaces of well cooked food to raw or undercooked meats (dietary prefer-
storage containers. Y. enterocolitica may be transferred from the ence), (ii) Y. enterocolitica is not the routine pathogen monitored in
contaminated surfaces to other food items, especially the higher our diarrhoeal patients, or (iii) most of the food poisoning due to
risk ready-to-eat foods. Improper food handling, processing and Y. enterocolitica cases are self-limiting. Although there is no other
storing practices such as undercooked meats or cross contamina- official report on yersiniosis in Malaysia since 1984, we should be
tion of contaminated meats or surfaces to other food or water are cognisant that this bacterium could be another potential agent to
risk factors for yersiniosis in humans. contribute to the incidence of food poisoning cases in our country
Besides humans, companion animals such as dogs and cats are since this pathogen was confirmed present in our local food and
exposed to the risk of infection of Y. enterocolitica. This is because pigs. Improper food handling and processing may cause cross
dogs and cats are more free-ranging and more likely to come in contamination of this pathogen to humans and therefore affirms a
contact with the materials contaminated with Y. enterocolitica, potential risk for the consumers.
especially stray dogs and cats in the vicinity of wet markets or pig’s
abattoirs. Dogs and cats may be at risk if they were fed with 5. Conclusion
contaminated meat. Several reports had indicated that companion
animals are infected after the consumption of contaminated pork This is the first report on the prevalence of Y. enterocolitica in
or by-products (Byun, Yoon, Lim, Lee, & Jung, 2011; Fredriksson- pigs and food from Malaysia. Y. enterocolitica was isolated from raw
Ahomaa, Korte, & Korkeala, 2001; Greene, 2006). porcine food and pigs. The prevalence of Y. enterocolitica in raw
Besides consumption of contaminated food, humans might be porcine food and pigs were 12.1% and 1.8%, respectively. The most
infected through direct contact with sick animals. Farmers who common bioserotypes isolated were 3 variant/O:3, followed by 1B/
work in pig farms or abattoirs may be infected through animal bits, O:8 and 1A/O:5. The presence of pathogenic Y. enterocolitica in raw
saliva or faeces of infected pigs. Moreover, companion animals such porcine food and pigs affirms a potential risk for the consumers.
as dogs and cats may also be another potential route for animal to Improper food handling and processing may cause cross contami-
human transmission. Infected dogs and cats can cause human nation of this pathogen to humans.
yersiniosis when they are in contact with humans, i.e. through
contact with animals’ excreta such as saliva and faeces (Fenwick, Acknowledgements
Madie, & Wilks, 1994; Stamm, Hailer, Depner, Kopp, & Rau, 2013;
Wang et al., 2010). Human infection from companion animals is This work was supported by University of Malaya through UM PPP
believed to be more serious than swine and consumption of grant (PS316/2010B) and University of Malaya High Impact Research
contaminated food, especially people with poor hygiene practices, Grant (UM.C/625/1/HIR/MOHE/CHAN-02). Lai Kuan Tan is supported
children, elderly, and immunocompromised patient (Plaut, by a fellowship from University of Malaya. Special thanks to Dr.
Zimmerman, & Goldstein, 1996). Hudson A. from Institute of Environment Science & Research (ESR)
More than 50 serotypes and 6 biogroups of Y. enterocolitica have Limited, New Zealand in providing the Microsoft Excel spreadsheet
been identified currently and their geographical distributions are for the MPN values calculation; and Dr. Gómez-Duarte O. G. in
diverse. In Europe, Y. enterocolitica particularly bioserotype 4/O:3 providing Y. enterocolitica positive control strain for our PCR test.
has been frequently isolated in humans, pig husbandry and food,
followed by the less common bioserotype, 2/O:9 and 2/O:5,27
References
(European Food Safety Authority & European Centre for Disease
Prevention and Control, 2012; Fondrevez et al., 2010; Fredriksson- Ananchaipattana, C., Hosotani, Y., Kawasaki, S., Pongsawat, S., Latiful, B. M., Isobe, S.,
Ahomaa, Gerhardt, & Stolle, 2009). In USA, O:8 is the primary in- et al. (2012a). Bacterial contamination of soybean curd (tofu) sold in Thailand.
fectious serotype, followed by O:5,27, O:13a, 13b, O:20, O:9 Food Science and Techonology Research, 18(6), 843e848.
Ananchaipattana, C., Hosotani, Y., Kawasaki, S., Pongsawat, S., Latiful, B. M., Isobe, S.,
(Bottone, 1997; Kwaga, Iversen, & Misra, 1992). Y. enterocolitica also et al. (2012b). Prevalence of foodborne pathogens in retailed foods in Thailand.
was frequently isolated in pigs from China in which the bio- Foodborne Pathogens and Disease, 9(9), 835e840.
serotypes isolated were 2/O:9, 4/O:3, 3/O:3, 1A/O:5, 1A/O:8. Among Annamalai, T., & Venkitanarayanan, K. (2005). Expression of major cold shock
proteins and genes by Yersinia enterocolitica in synthetic medium and foods.
the three bioserotypes identified in this study, Y. enterocolitica Journal of Food ProtectionÒ, 68(11), 2454e2458.
bioserotypes 3 variant/O:3 was the most common. In the past, the Archer, J. R., Schell, R. F., Pennell, D. R., & Wick, P. D. (1987). Identification of Yersinia
bioserotype 3 variant/O:3 had been reported in imported pork and spp. with the API 20E system. Journal of Clinical Microbiology, 25(12), 2398e2399.
Bhaduri, S., Wesley, I. V., & Bush, E. J. (2005). Prevalence of pathogenic Yersinia
chicken to Japan and food animals (pigs, rats and rabbits) in China enterocolitica strains in pigs in the United States. Applied and Environmental
(Fukushima, et al., 1997; Zheng & Xie, 1996). Y. enterocolitica bio- Microbiology, 71(11), 7117e7121.
serotype 3/O:3 was the major bioserotype in pigs, particularly from Bottone, E. J. (1997). Yersinia enterocolitica: the charisma continues. Clinical
Microbiology Reviews, 10(2), 257e276.
Jiangxi and Fujian Provinces with warmer climate (Liang et al., Byun, J. W., Yoon, S. S., Lim, S. K., Lee, O. S., & Jung, B. Y. (2011). Hepatic yersiniosis
2012; Wang et al., 2009). All these reports showed that this caused by Yersinia enterocolitica 4:O3 in an adult dog. Journal of Veterinary
particular bioserotype 3/O:3 was frequently isolated from the re- Diagnostic Investigation, 23(2), 376e378.
Dallal, M. M. S., Doyle, M. P., Rezadehbashi, M., Dabiri, H., Sanaei, M., Modarresi, S.,
gions with warmer climate. Therefore, Y. enterocolitica bioserotype
et al. (2010). Prevalence and antimicrobial resistance profiles of Salmonella
3/O:3 or 3 variant/O:3 strains might be the common strains in serotypes, Campylobacter and Yersinia spp. isolated from retail chicken and beef,
warm regions. Tehran, Iran. Food Control, 21(4), 388e392.
Lim & Tay (1992) attempted to isolate Y. enterocolitica from over Dzomir, M. A. Z. (2005). Isolation, identification and characterisation of Yersinia spp.
from meat and meat products. Doctoral thesis of University Putra Malaysia.
6000 samples from patients with diarrhoea in Singapore, but no European Food Safety Authority, & European Centre for Disease Prevention and
Y. enterocolitica was detected. They also suggested that Control. (2012). The European Union summary report on trends and sources of
100 L.K. Tan et al. / Food Control 35 (2014) 94e100

zoonoses, zoonotic agents and food-borne outbreaks in 2010. EFSA Journal, 10, Nesbakken, T. (1988). Enumeration of Yersinia enterocolitica O:3 from the porcine
1e442. oral cavity, and its occurrence on cut surfaces of pig carcasses and the envi-
Fenwick, S., Madie, P., & Wilks, C. (1994). Duration of carriage and transmission of ronment in a slaughterhouse. International Journal of Food Microbiology, 6(4),
Yersinia enterocolitica biotype 4, serotype 0:3 in dogs. Epidemiology and Infec- 287e293.
tion, 113(03), 471e477. Nesbakken, T., Eckner, K., Hridal, H. K., & Rrtterud, O. J. (2003). Occurrence of Yer-
Fondrevez, M., Labbé, A., Houard, E., Fravalo, P., Madec, F., & Denis, M. (2010). sinia enterocolitica and Campylobacter spp. in slaughter pigs and consequences
A simplified method for detecting pathogenic Yersinia enterocolitica in slaugh- for meat inspection, slaughtering, and dressing procedures. International Jour-
tered pig tonsils. Journal of Microbiological Methods, 83(2), 244e249. nal of Food Microbiology, 80(3), 231e240.
Fredriksson-Ahomaa, M., Gerhardt, M., & Stolle, A. (2009). High bacterial contam- Neubauer, H., Hensel, A., Aleksic, S., & Meyer, H. (2000). Identification of Yersinia
ination of pig tonsils at slaughter. Meat Science, 83(2), 334e336. enterocolitica within the Genus Yersinia. Systematic and Applied Microbiology,
Fredriksson-Ahomaa, M., & Korkeala, H. (2003). Low occurrence of pathogenic 23(1), 58e62.
Yersinia enterocolitica in clinical, food, and environmental samples: a method- Neuhaus, K., Francis, K. P., Rapposch, S., Görg, A., & Scherer, S. (1999). Pathogenic
ological problem. Clinical Microbiology Reviews, 16(2), 220e229. Yersinia species carry a novel, cold-inducible major cold shock protein tandem
Fredriksson-Ahomaa, M., Korte, T., & Korkeala, H. (2001). Transmission of Yersinia gene duplication producing both bicistronic and monocistronic mRNA. Journal
enterocolitica 4/O:3 to pets via contaminated pork. Letters in Applied Microbi- of Bacteriology, 181(20), 6449e6455.
ology, 32(6), 375e378. Okwori, A., Agina, S., Olabode, A., Ochapa, A., Turay, A., & Lombin, L. (2009). Profiles
Fukushima, H., Hoshina, K., Itogawa, H., & Gomyoda, M. (1997). Introduction into of Yersinia enterocolitica isolated from apparently healthy pigs in Jos, Nigeria.
Japan of pathogenic Yersinia through imported pork, beef and fowl. Interna- The Internet Journal of Microbiology, 6(1). http://dx.doi.org/10.5580/23ec.
tional Journal of Food Microbiology, 35(3), 205e212. Parker, L. V. (2002). Long-term survival of enteric microorganisms in frozen
Gill, C., & Jones, T. (1995). The presence of Aeromonas, Listeria and Yersinia in carcass wastewater. Cold Regions Research and Engineering Laboratory (ERDC/CRREL
processing equipment at two pig slaughtering plants. Food Microbiology, 12, TR-02-16).
135e141. Plaut, M., Zimmerman, E. M., & Goldstein, R. A. (1996). Health hazards to
Gómez-Duarte, O. G., Bai, J., & Newell, E. (2009). Detection of Escherichia coli, Sal- humans associated with domesticated pets. Annual Review of Public Health,
monella spp., Shigella spp., Yersinia enterocolitica, Vibrio cholerae, and 17(1), 221e245.
Campylobacter spp. enteropathogens by 3-reaction multiplex polymerase chain Ratnam, S., Looi, C. L., & Patel, T. R. (1983). Lack of efficacy of alkali treatment for
reaction. Diagnostic Microbiology and Infectious Disease, 63(1), 1e9. isolation of Yersinia enterocolitica from feces. Journal of Clinical Microbiology,
Greene, C. E. (2006). Infectious diseases of the dog and cat. WB Saunders, Elsevier 18(5), 1092e1097.
Science. Rosner, B. M., Stark, K., & Werber, D. (2010). Epidemiology of reported Yersinia
Gutler, M., Alter, T., Kasimir, S., Linnebur, M., & Fehlhaber, K. (2005). Prevalence enterocolitica infections in Germany, 2001-2008. BMC Public Health, 10(1), 337.
of Yersinia enterocolitica in fattening pigs. Journal of Food ProtectionÒ, 68(4), Scallan, E., Hoekstra, R. M., Angulo, F. J., Tauxe, R. V., Widdowson, M. A., Roy, S. L.,
850e854. et al. (2011). Foodborne illness acquired in the United States-major pathogens.
Horter, D. C., Yoon, K. J., & Zimmerman, J. J. (2003). A review of porcine tonsils in Emerging Infectious Disease, 17(1), 1e15.
immunity and disease. Animal Health Research Reviews, 4(2), 143e155. Schiemann, D. A. (1983). Alkatolerance of Yersinia enterocolitica as a basis for se-
Hudson, J. A., King, N. J., Cornelius, A. J., Bigwood, T., Thom, K., & Monson, S. (2008). lective isolation from food enrichments. Applied and Environmental Microbi-
Detection, isolation and enumeration of Yersinia enterocolitica from raw pork. ology, 46(1), 22e27.
International Journal of Food Microbiology, 123(1e2), 25e31. Singh, A., & McFeters, G. A. (1987). Survival and virulence of copper-and chlorine-
Jackson, V., Blair, I., McDowell, D., Kennedy, J., & Bolton, D. (2007). The incidence of stressed Yersinia enterocolitica in experimentally infected mice. Applied and
significant foodborne pathogens in domestic refrigerators. Food Control, 18(4), Environmental Microbiology, 53(8), 1768e1774.
346e351. Skjerve, E., Lium, B., Nielsen, B., & Nesbakken, T. (1998). Control of Yersinia enter-
Jegathesan, M., Paramasivam, T., Rajagopalan, K., & Loo, L. (1984). Yersinia enter- ocolitica in pigs at herd level. International Journal of Food Microbiology, 45(3),
ocolitica infection: first case report from Malaysia. Tropical and Geographical 195e203.
Medicine, 36(2), 207. Stamm, I., Hailer, M., Depner, B., Kopp, P. A., & Rau, J. (2013). Yersinia enterocolitica in
Johannessen, G. S., Kapperud, G., & Kruse, H. (2000). Occurrence of pathogenic diagnostic fecal samples of European dogs and cats: identification by FT-IR and
Yersinia enterocolitica in Norwegian pork products determined by a PCR method MALDI-TOF MS. Journal of Clinical Microbiology, http://dx.doi.org/10.1128/
and a traditional culturing method. International Journal of Food Microbiology, JCM.02506-12.
54(1), 75e80. Terentjeva, M., & Berzins, A. (2010). Prevalence and antimicrobial resistance of
Johnson, J. L. (1998). Isolation and identification of pathogenic Yersinia enterocolitica Yersinia enterocolitica and Yersinia pseudotuberculosis in slaughter pigs in Latvia.
from meat and poultry products. In USDA/FSIS Microbiology Laboratory Guide- Journal of Food ProtectionÒ, 73(7), 1335e1338.
book (3rd ed.). (pp. 1e28). Vlachaki, E., Tselios, K., Tsapas, A., & Klonizakis, J. (2007). Yersinia enterocolitica O:3
Kwaga, J., Iversen, J., & Misra, V. (1992). Detection of pathogenic Yersinia enter- mesenteric lymphadenopathy in an apparently healthy adult. The Netherlands
ocolitica by polymerase chain reaction and digoxigenin-labeled polynucleotide Journal of Medicine, 65(8), 311e312.
probes. Journal of Clinical Microbiology, 30(10), 2668e2673. Wang, X., Cui, Z., Jin, D., Tang, L., Xia, S., Wang, H., et al. (2009). Distribution of
Lawley, R., Curtis, L., & Davis, J. (2012). The food safety hazard guidebook. Royal pathogenic Yersinia enterocolitica in China. European Journal of Clinical Micro-
Society of Chemistry. biology & Infectious Diseases, 28(10), 1237e1244.
Lee, L. A., Gerber, A. R., Lonsway, D. R., Smith, J. D., Carter, G. P., Puhr, N. D., et al. Wang, X., Cui, Z., Wang, H., Tang, L., Yang, J., Gu, L., et al. (2010). Pathogenic strains of
(1990). Yersinia enterocolitica O:3 infections in infants and children, associated Yersinia enterocolitica isolated from domestic dogs (Canis familiaris) belonging
with the household preparation of chitterlings. New England Journal of Medicine, to farmers are of the same subtype as pathogenic Y. enterocolitica strains iso-
322(14), 984e987. lated from humans and may be a source of human infection in Jiangsu Province,
Liang, J., Wang, X., Xiao, Y., Cui, Z., Xia, S., Hao, Q., et al. (2012). Prevalence of Yersinia China. Journal of Clinical Microbiology, 48(5), 1604e1610.
enterocolitica in pigs slaughtered in Chinese abattoirs. Applied and Environ- Wauters, G., Kandolo, K., & Janssens, M. (1987). Revised biogrouping scheme
mental Microbiology, 78(8), 2949e2956. of Yersinia enterocolitica. Contributions to Microbiology and Immunology, 9,
Lim, Y., & Tay, L. (1992). Failure to isolate Yersinia enterocolitica from patients with 14e21.
diarrhoea in Singapore. Journal of Diarrhoeal Diseases Research, 10, 159e160. Xanthopoulos, V., Tzanetakis, N., & Litopoulou-Tzanetaki, E. (2010). Occurrence and
Martínez, P. O., Fredriksson-Ahomaa, M., Pallotti, A., Rosmini, R., Houf, K., & characterization of Aeromonas hydrophila and Yersinia enterocolitica in mini-
Korkeala, H. (2011). Variation in the prevalence of enteropathogenic Yersinia in mally processed fresh vegetable salads. Food Control, 21(4), 393e398.
slaughter pigs from Belgium, Italy, and Spain. Foodborne Pathogens and Disease, Yucel, N., & Ulusoy, H. (2006). A Turkey survey of hygiene indicator bacteria and
8(3), 445e450. Yersinia enterocolitica in raw milk and cheese samples. Food Control, 17(5),
Messelhäusser, U., Kämpf, P., Colditz, J., Bauer, H., Schreiner, H., Höller, C., et al. 383e388.
(2011). Qualitative and quantitative detection of human pathogenic Yersinia Zheng, X., & Xie, C. (1996). Note: isolation, characterization and epidemiology of
enterocolitica in different food matrices at retail level in Bavaria. Foodborne Yersinia enterocolitica from humans and animals. Journal of Applied Microbiology,
Pathogens and Disease, 8(1), 39e44. 81(6), 681e684.