International Journal of Medical Microbiology 304 (2014) 275–283

Contents lists available at ScienceDirect

International Journal of Medical Microbiology

journal homepage: www.elsevier.com/locate/ijmm

A molecular scheme for Yersinia enterocolitica patho-serotyping

derived from genome-wide analysis
Debora Garzetti a , Rosa Susen a , Angelika Fruth b , Erhard Tietze b , Jürgen Heesemann a ,
Alexander Rakin a,∗
a
Max von Pettenkofer-Institute for Hygiene and Medical Microbiology, Ludwig Maximilians-University, Munich, Germany
b
Robert Koch Institute, Wernigerode Branch, Division Enteropathogenic Bacteria and Legionella, National Reference Centre for Salmonellae and Other
Bacterial Enteric Pathogens, Wernigerode, Germany

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

Article history: Yersinia enterocolitica is a food-borne, gastro-intestinal pathogen with world-wide distribution. Only 11
Received 20 September 2013 serotypes have been isolated from patients, with O:3, O:9, O:8 and O:5,27 being the serotypes most
Received in revised form 17 October 2013 commonly associated with human yersiniosis. Serotype is an important characteristic of Y. enterocol-
Accepted 20 October 2013
itica strains, allowing differentiation for epidemiology, diagnosis and phylogeny studies. Conventional
serotyping, performed by slide agglutination, is a tedious and laborious procedure whose interpretation
Keywords:
tends to be subjective, leading to poor reproducibility. Here we present a PCR-based typing scheme for
Yersinia enterocolitica
molecular identification and patho-serotyping of Y. enterocolitica. Genome-wide comparison of Y. ente-
Molecular patho-serotyping
O-antigen
rocolitica sequences allowed analysis of the O-antigen gene clusters of different serotypes, uncovering
Genome-wide comparison their formerly unknown genomic locations, and selection of targets for serotype-specific amplification.
Two multiplex PCRs and one additional PCR were designed and tested on various reference strains and
isolates from different origins. Our genotypic assay proved to be highly specific for identification of Y.
enterocolitica species, discrimination between virulent and non-virulent strains, distinguishing the main
human-related serotypes, and typing of conventionally untypeable strains. This genotyping scheme could
be applied in microbiology laboratories as an alternative or complementary method to the traditional
phenotypic assays, providing data for epidemiological studies.
© 2013 Elsevier GmbH. All rights reserved.

Introduction for full virulence expression, and several typical chromosomally-

Yersinia enterocolitica is a Gram-negative, globally distributed,
foodborne human pathogen, which causes gastrointestinal diseases
by transmission via the fecal–oral route. Importantly for humans,
Y. enterocolitica has a high prevalence in animal and food sources,
in particular the consumption of pork products has been associated
with yersiniosis (Sabina et al., 2011). Biochemical and genetic fea-
tures allow classification of Y. enterocolitica strains into 6 biotypes
(1A, 1B, 2–5) (Bottone, 1997), more than 70 serotypes (Wauters
et al., 1987), and 2 subsp. (Y. enterocolitica subsp. enterocolitica
and subsp. palearctica) (Neubauer et al., 2000a). Y. enterocolitica
strains are further divided into 3 groups according to their virulence
degree: highly virulent strains of biotype 1B, low-virulence strains
of biotypes 2–5 and non-virulent strains of biotype 1A. Strains
belonging to biotypes 1B and 2–5 carry the plasmid pYV, essential
∗ Corresponding author at: Max von Pettenkofer-Institute for Hygiene and Medical
Microbiology, Ludwig Maximilians-University, Pettenkoferstr. 9A, 80336 Munich,
Germany. Tel.: +49 089 2180 72875; fax: +49 089 2180 72923.
E-mail address: rakin@mvp.uni-muenchen.de (A. Rakin).
encoded virulence markers, including ail, inv, and ystA. In turn,
biotype 1A strains lack the pYV-associated virulence determinants,
but possess the virulence marker ystB. Clinical Y. enterocolitica iso-
lates from humans predominantly belong to serotypes O:3, O:9,
O:8 and O:5,27; only 11 serotypes have been associated with
human yersiniosis, with a certain variability among different con-
tinents (Bottone, 1997; Fàbrega and Vila, 2012). Strains belonging
to bioserotype 4/O:3 are the most frequently detected pathogenic
Y. enterocolitica over the world, with swine accepted as their reser-
voir (Sabina et al., 2011). Despite being traditionally considered
non-virulent, strains of biotype 1A have been associated with gas-
trointestinal infections and isolated from humans, animals and food
(McNally et al., 2004; Ratnam et al., 1982). Among the six Y. ente-
rocolitica biotypes, 1A is the most heterogeneous group, including
more than 17 different serotypes (Wauters et al., 1987).
Serotype is one of the main characteristics of Y. enterocolitica
useful for epidemiological, diagnostic and phylogenetic purposes,
as strains enteropathogenic for humans are limited to a few
serotypes. The classification in serotypes of Y. enterocolitica is
mainly determined by the O-antigen (O-ag) moiety of the cell sur-
face lipopolysaccharide (LPS) (Skurnik and Bengoechea, 2003). The

1438-4221/$ – see front matter © 2013 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.ijmm.2013.10.007
276 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283

chemical structure of the O-ag, which consists of a sugar chain of Table 1

Bacterial strains used as references for the development of the PCR assays.
repeated O-units, has been characterized for some Y. enterocolitica
serotypes (Gorshkova et al., 1985, 1986; Oertelt et al., 2001; Ovodov Yersinia enterocolitica strains
et al., 1992; Radziejewska-Lebrecht et al., 1994). The O-ag genetic
Strain Bioserotype Genome accession no. Reference
clusters and their genomic locations, however, are partly solely
8081 1B/O:8 NC 008800 (Thomson et al., 2006)
known for serotypes O:3, O:8 and O:9 (Skurnik and Bengoechea,
WA-314 1B/O:8 AKKR00000000 (Garzetti et al., 2012)
2003). Conventional serotyping is performed by slide agglutina- Y11 4/O:3 NC 017564 (Batzilla et al., 2011a)
tion using O-ag specific typing antisera, obtained after a laborious W22703 2/O:9 FR718488–FR718797a (Fuchs et al., 2011)
process from rabbits immunized with the corresponding strains Y5.27P 3/O:5,27 CACW00000000 (Batzilla et al., 2011a)
(Aleksic et al., 1986). This serotyping method requires subjective NF-O 1A/O:5 CACY00000000 (Batzilla et al., 2011b)
IP2222 1A/O:36 CACZ00000000 (Batzilla et al., 2011b)
interpretation of the results. Furthermore the reliability of a set of
typing antisera is defined in only a few laboratories in the world, Non-Yersinia enterocolitica strains
which are able to offer the complete serotyping of Y. enterocolit-
Species Strain Serotype/Biotype
ica. Importantly, isolates that lose expression of the O-ag in certain
conditions have a rough LPS and are thus not typeable with the Y. frederiksenii H56-36/81 O:60
traditional agglutination test. Cross-reactions may also occasion- Y. intermedia H9-36/83 O:17
Y. kristensenii H17-36/83 O:12,25
ally occur between Yersinia and other members of the family of
Y. rohdei H274-36/78 O:76
Enterobacteriaceae, as has been demonstrated for Brucella and Y. Y. bercovieri H632-36/85 O:16
enterocolitica O:9 strains (Godfroid et al., 2002). Moreover, vir- Y. mollaretii H279-36/86 O:59
ulence cannot be inferred by serotyping itself, as, for example, Y. pseudotuberculosis IP32953 O:1
Y. pestis KIM Medievalis
Y. enterocolitica strains of serotype O:8 can belong to both non-
virulent 1A and virulent 1B biotypes (Bottone, 1997).
Non-Yersinia strains
Molecular techniques can provide alternative serotyping
tests with more standardized procedures and, therefore, higher Species Strain Serotype
efficiency and specificity. Previously published assays allow inde- Escherichia coli UTI89 O18:K1:H7
pendent detection of Y. enterocolitica strains belonging to the O:3 Escherichia coli CFT073 O6:H1
Salmonella enterica ssp. I C5 Typhimurium
or O:9 serotypes through amplification of O-ag-encoded genes
Salmonella enterica ssp. I ATCC 14028 Typhimurium
(Jacobsen et al., 2005; Wannet et al., 2001; Weynants et al., 1996). Campylobacter jejuni L11 Lior 11
Different PCR assays have also been designed for detection of Y. Campylobacter jejuni G23 n.a.
enterocolitica from laboratory cultures and clinical or environmen- Legionella pneumophila 2020 n.a.
tal samples (Fredriksson-Ahomaa et al., 2006). To date, a valid a
Complete genome of Y. enterocolitica serotype O:9 was analyzed using the
and comprehensive molecular method for specific and concurrent sequence from strain 105.5R(r) (Accession number CP002246.1) (Wang et al., 2011).
identification and serotyping of Y. enterocolitica is missing. The
availability of multiple Y. enterocolitica genome sequences from Bacterial genomic DNA extraction
strains of the most common serotypes (Batzilla et al., 2011a,b;
Garzetti et al., 2012; Thomson et al., 2006; Wang et al., 2011) Genomic DNA from E. coli and S. enterica strains were obtained
facilitates comparison among O-ag gene clusters and identifica- from Prof. B. Stecher (MvPI). For the other strains, preparation
tion of serotype-specific genes. The aim of this study was to of genomic DNA for testing primer specificity and for assay vali-
apply this genome-wide information for development and vali- dation was based on the silica-membrane technology, using the
dation of a PCR-based method for simultaneous identification and NucleoSpin Tissue kit (Macherey-Nagel, Düren, Germany). Briefly,
patho-serotyping of Y. enterocolitica. Genomic location and genetic isolates were grown on CIN-, LB- (lysogeny broth) or blood-agar
organization of the O-ag clusters of the analyzed serotypes (O:8, at 27 ◦ C (Yersinia species) or 37 ◦ C (non-Yersinia species). For each
O:3, O:9, O:5,27, O:5) have also been studied. strain, genomic DNA was isolated from 900 ␮l of an overnight cul-
ture, which was grown from a single colony in LB broth, following
the manufacturer’s instructions. Pure DNA was obtained in two
Materials and methods elution steps with a total volume of 100 ␮l of low-salt buffer. The
concentration of the extracted DNA was photometrically measured
Bacterial strains at 260 nm and DNA was diluted with distilled water to a final con-
centration of 100 ␮g/ml, in order to obtain standardized results.
To set up the conditions and evaluate the specificity of the three In colony PCR experiments, template DNA was isolated directly
PCR assays developed in the present study, 15 Yersinia strains and from bacterial cells by the boiling method. In brief, a single colony
7 bacterial strains representative of different zoonotic and envi- was inoculated in 50 ␮l of 25 mM NaOH and the bacterial mixture
ronmental agents were used (Table 1). The PCR method was then was heated at 100 ◦ C for 10 min. After neutralization with 50 ␮l of
validated on 60 Y. enterocolitica strains from the culture collec- 80 mM Tris/HCl buffer (pH 7.5) and centrifugation at 20,000 × g,
tion of the Max von Pettenkofer-Institute (MvPI) and on 73 strains 4 ◦ C, for 5 min, the supernatant was transferred to a new tube and
obtained from the Robert Koch Institute (RKI). Strains were iso- 1 ␮l used as template for PCR.
lated from humans, swine, food and the environment mainly in
Germany but also from other countries. Yersinia isolates were sub- Target selection and primer design
cultured onto selective Cefsulodin-Irgasan-Novobiocin (CIN) agar
and incubated at 30 ◦ C for 18 h. Classification into biotypes was Publicly available genomic sequences of different Y. enterocol-
based predominantly on biochemical reactions and performed as itica serotypes were used as references for identification of O-ag
previously described (Bockemühl and Roggentin, 2004; Wauters gene clusters and for selection of regions useful for the serotyping
et al., 1987). Serotyping was performed using mono-specific test multiplex PCR (Table 1). For both multiplex PCRs, four primer pairs
sera (SIFIN GmbH, Berlin, Germany) for the identification of Y. were manually designed (Table 2) in order to fulfill the following
enterocolitica O-antigens (O:8, O:3, O:9, O:5, O:27) by slide agglu- criteria: (a) primer length is 18–24 bp with GC content of 40–60%
tination, according to the manufacturer’s instructions. and similar annealing temperature; (b) primer sequences in one
 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283 277

Table 2 Bioinformatics analysis

Primer sequences and amplicon sizes of the developed PCR assays.

Target Primers Primer sequences Amplicon Identification of O-ag serotype-specific gene clusters was per-
(5 →3 ) size (bp) formed by BLAST analysis, whole genome alignment with Mauve
Multiplex 1: identification (Darling et al., 2010) and comparative genome analysis with Edgar
inv inv FW TGGCATCAATCTCGTGATTTCG 1009 (Blom et al., 2009) and BRIG (Alikhan et al., 2011). General sequence
inv RV GTTGCCCCTGAATATCTAAAGTGAC analysis was performed with CLC DNA Workbench (CLC bio, Aarhus,
ail ail Yen FW TGTTAATGTGTACGCTGCGAGTa 431
Denmark) and BioEdit (Hall, 1999). Genome Accession Numbers of
ail Yen RV GTTTGGAGTATTCATATGAAGCGTCa
16S rRNA 16S Yen FW AATACCGCATAACGTCTTCGGAb 330 the analyzed sequences are given in Table 1 with the respective
16S Yen RV CTTCTTCTGCGAGTAACGTCAATb strain information.
ystB ystB FW AACTTTTTGGACACCGCACAG 208
ystB RV GTCTGAGTATCGCACGCT

Multiplex 2: serotyping Results

per (O:9) per FW TCCTTCTCCAAATATATAGGTGCCA
per RV ATGCGGCATTAGATGAGATGGA
wzt
wzt FW GTTAGTTCCTGCATCTGATCGCC
(O:5,27) wzt RV ATCCAGCATCCATGGCTCC
wbbU (O:3) wbbU FW ACCTCGTATTTTTGAAGATGATCGC
wbbU RV GTACTCAATAACTTGCTGTTCGGA
wbcA (O:8) wbcA FW TGATGAACGAGGCGAGTTTGTT
wbcA RV TACTCCGTCTGTTATGCGGATTTAG
a
Primer sequences modified from Wannet et al. (2001).
b
Primer sequences modified from Neubauer et al. (2000b).
reaction do not form self- or hetero-dimers; (c) primer sequences
specifically align to the target region (d) the produced amplicons
have unambiguously different sizes.
PCR amplification
For both identification and serotyping multiplex PCRs, a 50 ␮l
reaction mixture contained 0.2 mM dNTPs, 2.0 mM MgCl2 , 75 mM
Tris–HCl [pH 8.5], 20 mM (NH4 )2 SO4 , 0.1% Tween 20® , 1.25 U of
Amplicon Taq DNA polymerase (VWR International) and 0.4 ␮M of
each primer (Eurofins MWG Operon, Ebersberg, Germany). Reac-
tions were prepared in two master mixes and 100 ng of DNA
template, or 1 ␮l of supernatant when performing colony PCR
experiments, were added. All amplifications were performed in
a Veriti® 96-well Thermal Cycler (Applied Biosystem) using the
following amplification protocol: initial denaturation at 95 ◦ C for
5 min, 30 cycles of denaturation at 95 ◦ C for 40 s, primer annealing
at 58 ◦ C for 40 s, extension at 72 ◦ C for 60 s, and final extension at
72 ◦ C for 8 min.
To differentiate between Y. enterocolitica 1A/O:5 and O:5,27,
a single PCR with primers RM O5,27 FW (5 -TTCCAGCACAC-
GTCGAACAAGTTC-3 ) and RM O5,27 RV (5 -AGGAAGATATCC-
AGTGCCGCT-3 ) was performed using the same reaction and
amplification protocol described above. An amplicon of 627 bp is
expected only in strains of serotype O:5,27.
After PCR amplification, 7 ␮l of products were separated by elec-
trophoresis on horizontal 1.5% agarose gels in TAE buffer (40 mM
Tris–acetate, 1 mM EDTA). Gels were stained with ethidium bro-
mide and visualized under UV-light transillumination.
Bacteria identification
Amplification of the 16s rRNA gene for bacterial species
identification was performed with the universal primers
fD1 (5 -CGATATCTCTAGAAGAGTTTGATCCTGGCTCAG-3 ), fD2
(5 -CGATATCTCTAGAAGAGTTTGATCATGGCTCAG-3 ) and rP1 (5 -
GATATCGGATCCACGGTTACCTTGTTACGACTT-3 ). PCR products
were purified with the peqGOLD Gel Extraction Kit (PEQLAB).
Sequencing of the fragments was obtained using the primer
27f (5 -AGAGTTTGATCMTGGCTCAG-3 ) (Eurofins MWG Operon).
Nucleotide sequences were then searched against public sequence
databases with BLAST (Basic Local Alignment Search Tool) (Altschul
et al., 1990) to identify the bacterial species.
837
Genomic location of serotype-specific O-ag gene clusters
662
463 O-antigens of Y. enterocolitica serotypes O:8, O:3 and O:9 have
been cloned and largely characterized (Lubeck et al., 2003; Skurnik
269
and Bengoechea, 2003; Zhang et al., 1993, 1997). The availabil-
ity of various whole genome sequences enabled us to determine
the previously unknown genomic location of the O-ag region of
serotypes O:3 and O:9 and to identify the O-ag gene cluster of
serotypes O:5,27 and O:5 (Figs. 1 and 2). The genes involved in the
O-ag biosynthesis of Yersinia are generally clustered between the
hemH and gsk genes (Skurnik, 1999). In the analyzed Y. enterocolitica
strains, this is only applicable to serotype O:8. Indeed, the hemH-
gsk locus of serotypes O:5 and O:5,27 is occupied by the outer core
gene cluster, as previously described for Y. enterocolitica serotypes
O:9 and O:3 (Skurnik et al., 1995).
The O-ag genetic region of Y. enterocolitica serotype O:9
(locus tags YE105 C1500-1511 in Y. enterocolitica strain 105.5R(r)
genome) lies between the gnd and galF genes (Fig. 1), similarly to
E. coli and S. enterica (Samuel and Reeves, 2003). The genes per,
gmd, manB and manC are present, confirming the O-ag biochem-
ical structure containing perosamine (4,6-dideoxy-D-mannose)
(Caroff et al., 1984). The Y. enterocolitica serotype O:3 O-ag gene
cluster (locus tags Y11 16641–16781 in Y. enterocolitica strain
Y11 genome) is also located in this genomic region between
the galU gene and a transposase-encoding gene (Fig. 1). A sec-
ond transposase gene could be identified within this cluster. It is
located between the genes involved in synthesizing the nucleotide
sugar precursors and processing the O-ag (wbbS-wbbX), and the
glycosyltransferase-encoding genes. Interestingly, in serotype O:8
this region is interrupted by several transposase genes (Fig. 1).
Together with the atypical G + C content in the O-ag gene clusters
of Y. enterocolitica serotypes O:3 and O:9 (38% and 35%, respec-
tively, compared to a typical Y. enterocolitica genomic G + C content
of 47%), this indicates a probable inter-species lateral gene transfer
in this potential hot-spot region.
The chemical structures of the specific repeating O-unit of
Y. enterocolitica serotypes O:5,27 and O:5 have been elucidated
(Gorshkova et al., 1986; Perry and MacLean, 1987) and a similar
sugar composition between the two O-antigens was found. Never-
theless, the chemical structure for the factor 27 was not resolved.
Similarly, no significant genetic differences could be determined
between the O-ag clusters of Y. enterocolitica serotypes O:5,27 and
O:5, nor was a specific gene cluster found for the factor 27. The
O-ag gene clusters of serotypes O:5,27 and O:5 occupy different
genomic locations (Fig. 2). In Y. enterocolitica serotype O:5,27, the
O-ag genes are situated near a phage-remnant and a transposase-
encoding gene, in a region where also Y. enterocolitica serotypes O:8
and O:3 have transposase and phage-related genes (Fig. 2). The O-
ag gene cluster of Y. enterocolitica serotype O:5 is inserted between
the gor and yapE genes (locus tags YE4057-4058 in the genome of
Y. enterocolitica strain 8081) and has 97.3% of DNA sequence simi-
larity with the respective O-ag gene cluster of serotype O:5,27. This
278 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283

Fig. 1. Representation of the O-antigen genetic clusters of Y. enterocolitica serotypes O:3 and O:9 and their corresponding genomic location in Y. enterocolitica serotype O:8.
Genes chosen as targets for the serotyping multiplex are in black squares. GT-A: glycosyl transferase group A. Drawn to scale.

high genetic similarity prevented designing of specific primers for
distinguishing Y. enterocolitica serotype O:5,27 from serotype O:5
(see below).
Selection of Y. enterocolitica-specific genes for the identification
multiplex
virulence-associated ystB gene, on the other hand, is specifically
carried by strains of biotype 1A (Kot et al., 2010; Ramamurthy
et al., 1997; Stephan et al., 2013; Thoerner et al., 2003), and is
therefore a distinguishing marker for non-virulent strains. The
ystB gene of biotype 1A strains is 76.5% similar to the ystA gene
encoded by genomes of virulent strains.

To unambiguously detect Y. enterocolitica isolates at the species
level, two target regions specifically conserved in this species
were selected for PCR amplification: a 330 nt fragment of the Y.
enterocolitica 16S rRNA gene and a 1009 nt fragment of the inv
gene. The inv gene has been detected by molecular methods in Y.
enterocolitica strains with 100% frequency, irrespective of isolation
source (humans, animals or food), serotype or pathogenicity level
(Falcão et al., 2006; Thoerner et al., 2003; Zheng et al., 2008).
Accordingly, the inv sequence can be used as a specific PCR
marker for Y. enterocolitica identification, in combination with the
genetically stable 16S rRNA gene. Regions within the ail and ystB
genes were added as amplification targets to differentiate between
virulent (biotypes 1B and 2–5) and non-virulent (biotype 1A) Y.
enterocolitica strains, respectively. The chromosomally encoded ail
gene has been recognized as a stable marker for virulent strains
and has been widely applied in molecular detection of pathogenic
Y. enterocolitica strains (Fredriksson-Ahomaa et al., 2006; Harnett
et al., 1996; Kwaga et al., 1992; Stephan et al., 2013; Thisted
Lambertz and Danielsson-Tham, 2005; Wannet et al., 2001). The
Selection of O-ag serotype-specific regions for the serotyping
multiplex
In order to design a multiplex PCR able to differentiate between
the major Y. enterocolitica virulent serotypes (O:3, O:9, O:8 and
O:5,27), serotype-specific regions were selected (Table 2). It has
been shown that a specific detection of Y. enterocolitica serotype O:3
can be obtained by amplifying the rfbC/wbbU gene (Weynants et al.,
1996). Our genomic analysis confirmed that this gene (locus tag:
Y11 16731 in strain Y11) can be an appropriate target for serotype
O:3-specific detection, since no homology with genes in up-to-date
Yersinia and no-Yersinia genomes was detected by BLAST analysis.
In a similar way, the per gene (locus tag: YE105 C1505 in strain
105.5R(r)) in the O-ag gene cluster of serotype O:9 was selected
as specific target for Y. enterocolitica serotype O:9, as previously
described (Jacobsen et al., 2005). Importantly, the per genes of Y.
enterocolitica serotype O:9 and Brucella spp. have 64% nucleotide
sequence similarity. In order to avoid cross-reaction with Brucella
species, primers which specifically anneal to the per gene of Y.

Fig. 2. Genetic organization of the O-ag gene cluster of Y. enterocolitica serotypes O:5,27 (upper part) and O:5 (lower part), with their corresponding locations in Y. enterocolitica
serotypes O:8 and O:3 genomes. Genes chosen as targets for the serotyping multiplex are in black squares. Drawn to scale.
Table 3
Comparison between results obtained by conventional methods (slide agglutination and biochemical tests) and molecular methods (multiplex PCRs and 16S rRNA gene sequencing) tested on 133 Y. enterocolitica strains. Strains
which gave discordant results are marked by *. The results of the re-serotyping are shown in parenthesis.

Number of strains Multiplex PCR assays Identification and serotyping

inv ail 16S rRNA ystB wbcA/O8 wbbU/O3 wzt/O5-O5,27 per/O9 RM/O5,27 Conventional methods Molecular methods

D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283

31 + + + − − + − − Y. enterocolitica 4/O:3 Y. enterocolitica O:3
1 + + + − − + − − Y. enterocolitica 3/O:3 Y. enterocolitica O:3
16 + + + − − + − − Y. enterocolitica O:3 Y. enterocolitica O:3
3 + + + − − + − − Y. enterocolitica Y. enterocolitica O:3
1* + + + − − + − − Y. enterocolitica O:1,2a,3 Y. enterocolitica O:3
1* + + + − − + − − Y. enterocolitica O:2a,2b,3 Y. enterocolitica O:3
8 + + + − − − − + Y. enterocolitica 3/O:9 Y. enterocolitica O:9
1 + + + − − − − + Y. enterocolitica 2/O:9 Y. enterocolitica O:9
1 + + + − − − − + Y. enterocolitica 4/O:9 Y. enterocolitica O:9
1 + + + − − − − + Y. enterocolitica O:9 Y. enterocolitica O:9
5 + + + − − − − + Y. enterocolitica 3/rough Y. enterocolitica O:9
3 + + + − + − − − Y. enterocolitica 1B/O:8 Y. enterocolitica 1B/O:8
3 + + + − + − − − Y. enterocolitica O:8 Y. enterocolitica 1B/O:8
5 + − + + + − − − Y. enterocolitica 1A/O:8 Y. enterocolitica 1A/O:8
1* + − + + + − − − Y. enterocolitica 1A/O:7,8 Y. enterocolitica 1A/O:8
6 + + + − − − + − + Y. enterocolitica 3/O:5,27 Y. enterocolitica O:5,27
10 + + + − − − + − + Y. enterocolitica 2–3/O:5,27 Y. enterocolitica O:5,27
2 + + + − − − + − + Y. enterocolitica 4/O:5,27 Y. enterocolitica O:5,27
10 + − + + − − + − − Y. enterocolitica 1A/O:5 Y. enterocolitica 1A/O:5
1* + − + + − − − − − Y. enterocolitica O:5,27 (rough) Y. enterocolitica 1A
10 + − + + − − − − Y. enterocolitica 1A/rough Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:8 (rough) Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:36 Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:6,30 Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:4,33 Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:10 Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:48 Y. enterocolitica 1A
1 + − + + − − − − Y. enterocolitica 1A/O:41,43 Y. enterocolitica 1A
2* − − − − − − − − Y. enterocolitica 4/O:3 (O:3) Y. frederiksenii
1* − − + − − − − − Y. enterocolitica 3/rough (rough) Y. kristensenii
1 + + + − − − − − Y. enterocolitica 1B/O:13 Y. enterocolitica virulent
1 + + + − − − − − Y. enterocolitica 1B/O:20 Y. enterocolitica virulent
1 + + + − − − − − Y. enterocolitica 1B/O:21 Y. enterocolitica virulent

279
280 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283

enterocolitica were designed. BLAST search of the primer sequences

against Brucella sequences revealed no significant matches. To
unambiguously detect Y. enterocolitica strains belonging to the
highly virulent serotype O:8 the rfbC/wbcA gene (locus tag: YE3085
in strain 8081) was selected, due to its low sequence similarity
(51.7%) with the orthologous rfbC/wbbU gene in Y. enterocolitica
O:3. Target regions present in the Y. enterocolitica serotype O:5,27
O-ag gene cluster but absent from that of serotype O:5 could not
be found. Hence, a gene (rfbE/wzt) had to be selected as specific
amplification target for both serotypes O:5,27 and O:5. The two
gene sequences show 97.2% similarity, and no polymorphism could
be used to design primers specific for either serotype according to
the chosen criteria (see Section “Materials and methods”). Widen-
ing the search for a suitable target for detection of serotype O:5,27
to a whole genome level, a restriction modification (RM) system
absent from Y. enterocolitica serotype O:5 was found. This clus-
ter is also absent from serotype O:3 and O:9 genomes and from
Y. enterocolitica strain 8081 (1B/O:8), but is 95.2% similar to a RM
system detected in Y. enterocolitica strain WA-314 (1B/O:8) as a
strain-specific region (Garzetti et al., 2012). Therefore, primers tar-
geting this region cannot be applied to the serotyping multiplex,
since they would give a positive signal for both strains of serotype
O:5,27 and some strains of serotype O:8. They can still be used in a
single PCR reaction to discriminate between Y. enterocolitica strains
of serotype O:5,27 and O:5 with a positive signal for the wzt gene. Fig. 3. Results from the multiplex assays for identification (top) and patho-
serotyping (bottom) of Y. enterocolitica reference strains. MW, molecular weight
The specificity of this third PCR assay has been verified on 29 Y. ente-
standard (1-kb DNA ladder, Fermentas). NC, negative control.
rocolitica strains of either serotype O:5 or serotype O:5,27 (Table 3),
confirming a 100% prevalence of the selected RM cluster in strains
of serotype O:5,27 and its absence from all the strains of serotype amplification of the wzt (O:5,27) target occurs, as expected, also in
O:5. our reference serotype O:5 strain, distinguishing Y. enterocolitica
strains of serotype O:5 from strains of serotype O:5,27 is possi-
Specificity of the PCR assays ble combining the results of the two multiplexes. Indeed, being
non-virulent, strains of serotype O:5 would have pattern inv(+), 16S
Once the genetic regions to be targeted by the PCR assays had rRNA(+), ail(−), ystB(+), wzt(+). On the other hand, pattern inv(+),
been selected, primers with 18 to 24 nt in length, 40% to 58% of GC 16S rRNA(+), ail(+), ystB(−), wzt(+) would result from strains of
content and with melting temperature (TM ) between 54.5 ◦ C and serotype O:5,27. An unmistakable distinction derives by the appli-
59.7 ◦ C were designed. Primer sequences did not form hairpins nor cation of the single PCR targeting the RM system in Y. enterocolitica
self- or hetero-dimers, and thus were suitable for multiplex PCRs. serotype O:5,27, resulting in amplification of the 627 bp product
Primer pairs were first tested in single PCRs at different annealing only for serotype O:5,27 strains.
temperatures. Amplification products having the expected lengths,
without non-specific amplicons and giving clear bands on agarose Validation of the assay on various Y. enterocolitica strains
gel, were obtained with an annealing temperature of 58 ◦ C (data
not shown). The two multiplex PCRs were then tested on our Y. The developed PCR assays were validated on 133 Y. enterocolitica
enterocolitica reference strains (Table 1). The amplification product strains, either of known or unknown serotypes, isolated from dif-
sizes adequately differed from each other, allowing unambiguous ferent sources and countries. Genotypic and phenotypic methods
identification of the products depending on their length (Fig. 3). for Y. enterocolitica species identification gave concordant results
As expected, the identification multiplex PCR produced different in 130 out of 133 strains (Table 3). One of the 3 discrepant strains,
patterns according to the virulence group of the tested strains: a rough Y. enterocolitica biotype 3 strain, was completely negative
the 16S rRNA and the inv targets were amplified in all Y. entero- in the PCR, while the 16S rRNA fragment was weakly amplified in
colitica strains, while the ail product was positive in the virulent the other 2 strains, which are pig-isolates classified as bioserotype
strains and negative in the non-virulent strains, with an opposite 4/O:3 by conventional methods. According to the designed method,
result obtained from the ystB gene amplification. Therefore, viru- the absence of the inv amplicon suggested that these 3 strains
lent strains would give pattern inv(+), 16S rRNA(+), ail(+), ystB(−), do not belong to Y. enterocolitica species. Indeed, upon sequenc-
while pattern inv(+), 16S rRNA(+), ail(−), ystB(+) would be obtained ing of the 16S rRNA gene for universal bacterial identification,
from non-virulent strains. To test the specificity of the multiplex the strains were found to be non-Y. enterocolitica, confirming that
PCR at a species level, non-Y. enterocolitica and non-Yersinia strains the identification multiplex is 100% specific. Results from molec-
were analyzed. No detectable PCR amplification was observed with ular and conventional serotyping methods were concordant in
non-Y. enterocolitica strains, while weak non-specific bands were 126 out of 133 examined strains (Table 3). Of the 7 discordant
amplified by C. jejuni and S. enterica strains (data not shown). strains, 2 were the serotype O:3 pig-isolates mentioned above,
The serotyping multiplex PCR is designed in order to amplify a identified as non-Y. enterocolitica by molecular techniques. A com-
single PCR product for each serotype (O:3, O:9, O:8, O:5,27) and pletely negative amplification was obtained from both strains by
identify it based on the amplicon size (Table 2). As clearly shown the serotyping multiplex PCR, suggesting a dissimilar sequence of
in Fig. 3, the goal is successfully achieved and only specific prod- the wbbU gene in the O-ag gene cluster of these strains. The serotype
ucts are amplified. Undoubtedly, Y. enterocolitica strains belonging of another strain, conventionally classified as bioserotype 1A/O:8,
to serotypes not recognized by this test will give a negative result, could not be assigned, since no amplification of any product was
as for strain IP2222, which belongs to serotype O:36. Although the observed. Having pattern inv(+), 16S rRNA(+), ail(−), ystB(+), this
 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283
strain was genotyped as Y. enterocolitica biotype 1A. A strain tra-
ditionally designated to serotype O:5,27 produced negative results
from the serotyping multiplex and from the O:5,27-specific single
PCR. Pattern inv(+), 16S rRNA(+), ail(−), ystB(+) was obtained by the
identification multiplex and, therefore, this strain was classified as
non-virulent (biotype 1A). Re-serotyping of these four discrepant
strains confirmed the previously determined serotype O:3 for the
two pig isolates and gave negative reaction for the other two strains,
supporting the molecular serotyping results. The tested strain of
bioserotype 1A/O:7,8 gave a O:8-like PCR-pattern, with amplifi-
cation of the wbcA (O:8) product. Producing pattern inv(+), 16S
rRNA(+), ail(+), ystB(−), wbcA (+), strains of serotype O:7,8 would
thus be genotyped as Y. enterocolitica bioserotype 1A/O:8. Amplifi-
cation of the wbbU (O:3) product resulted from strains of serotypes
O:1,2a,3 and O:2a,2b,3, leading to an inaccurate genotyping of these
rare strains as virulent Y. enterocolitica strains of serotype O:3.
These results indicate that the developed assay needs further devel-
opment for strains of serotypes O:7,8, O:1,2a,3 and O:2a,2b,3. Y.
enterocolitica strains of serotypes O:13, O:20 and O:21, belonging to
the virulent biotype 1B, were also tested. As expected, the ail target
was amplified and no amplicons were obtained from the serotyping
multiplex. They can therefore be identified as virulent Y. enteroco-
litica. A group of 14 rough and non-typeable strains denoted as Y.
enterocolitica was included in the validation process. One of these
strains turned out to be non-Y. enterocolitica (Y. kristensenii) while
8 strains, genotyped as biotype 1A, gave negative results from the
serotyping multiplex. Importantly, 5 rough strains were serotyped
by our assay as Y. enterocolitica serotype O:9, revealing a notable
advantage of the molecular test over the conventional serotyping
agglutination method.
The detection limit of the designed multiplex PCRs was tested,
performing serial dilutions from reference and various Y. enteroco-
litica strains. The results indicated that 1 ng of template DNA was
necessary for detection of all the products with a 30-cycle ampli-
fication protocol. Only the 16S rRNA amplicon could be detected
with 10 pg of DNA (data not shown).
In order to remove the DNA extraction step and reduce the work-
ing time, the use of bacterial colonies as templates was examined.
This method gave clear and reproducible results for the identifica-
tion multiplex, while non-specific amplification was observed with
the serotyping PCR. Nevertheless, the expected bands were brighter
than the non-specific smear and could allow identification of the
strain serotype (data not shown).
Discussion
Whole genome sequencing and comparative genomics have
opened new windows on molecular microbiology. Next-generation
sequencing has altered the way infectious disease are studied,
allowing, among others, epidemiological analysis, molecular geno-
typing and evolution studies. Through comparison of genome
sequences obtained from Y. enterocolitica strains of various
serotypes, the O-ag gene clusters of the 4 serotypes most commonly
isolated from humans were examined. In contrast to Y. pseudotu-
berculosis and Y. pestis, where the O-ag cluster possesses a highly
conserved genetic organization and is located in the same chromo-
somal location between the hemH and gsk genes (Bogdanovich et al.,
2003), Y. enterocolitica continues its heterogenous nature even in
the genomic position of the genes involved in the O-ag biosynthesis
(Fig. 4). Indeed, only the O-ag genes of Y. enterocolitica serotypes
O:3 and O:9 share the same location, while the O-ag of serotype
O:8 shows homology to the O-ag clusters of Y. pseudotuberculo-
sis and Y. pestis. Interestingly, in spite of the high DNA sequence
similarity, the genetic clusters of serotypes O:5,27 and O:5 are car-
ried in different locations. No genetic cluster responsible for the
281
synthesis of the O-ag factor 27 could be identified in the genome
of Y. enterocolitica strain Y5.27P, raising doubts on the validity of
the classification of Y. enterocolitica serotypes O:5 and O:5,27 into
two different serotypes. As demonstrated for other Gram-negative
bacteria, the atypical chromosomal positions and the enormous
genetic diversity of Y. enterocolitica O-antigens may be explained
by recent inter-species lateral gene transfer (Samuel and Reeves,
2003). Several transposase-encoding genes have been, in fact, iden-
tified in the analyzed O-ag clusters, supporting this hypothesis.
This study introduces a three PCR-based test for specific simul-
taneous identification and serotyping of Yersinia enterocolitica. The
first assay aims at the molecular identification at species- and
subspecies-levels through amplification of four gene targets: 16S
rRNA and inv genes as species-specific products, ail as marker
for virulence (biotypes 1B and 2–5), and ystB as marker for non-
virulent strains (biotype 1A). Only chromosomally-encoded genes
have been selected due to the known instability of the pYV plasmid
of Yersiniae. The second assay allows geno-serotyping of the most
commonly isolated human Y. enterocolitica of serotypes O:3, O:9,
O.8 and O:5,27. O-ag genetic clusters were identified through anal-
ysis and comparison of available Y. enterocolitica genome sequences
in order to select serotype-specific targets: rfbC/wbbU for serotype
O:3, rfbC/wbcA for serotype O:8, per for serotype O:9 and rfbE/wzt
for serotype O:5,27. Amplification of the latter target occurred
also in strains belonging to the non-virulent serotype O:5, causing
misidentification. Indeed, the O-ag gene clusters of Y. enterocolitica
serotypes O:5 and O:5,27 were found to be highly similar. Differ-
entiation between serotypes O:5,27 and O:5 is possible from the
identification multiplex patterns ail(+), ystB(−) and ail(−), ystB(+),
respectively. Nevertheless, serotype O:5,27 may be unambiguously
confirmed by the third assay, through amplification of a genetic
region in a RM system absent from serotype O:5 strains.
The test presented here is able to detect Y. enterocolitica
strains at a species-level and to identify all the 4 clinically pre-
vailing serotypes. It can also differentiate between virulent and
non-virulent strains, detecting the genetic targets ail and ystB,
respectively. However, some Y. enterocolitica biotype 1A isolates
from pigs and pork products carry the ail gene, as demonstrated by
recent epidemiological studies (Bonardi et al., 2013; Paixao et al.,
2013). By contrast, the frequency of Y. enterocolitica biotype 1A clin-
ical samples carrying the ail gene is insignificant (Stephan et al.,
2013). The developed identification multiplex would therefore cor-
rectly classify as “non-virulent” the biotype 1A strains from human
samples, while a misclassification may occur when testing isolates
from animals which have acquired the ail gene. A similar situa-
tion has been observed concerning the ystB gene, which has not
been detected in 27.6% non-virulent Y. enterocolitica isolates from
pigs (Bonardi et al., 2013), while in fecal samples from humans the
ystB frequency is 95% (Stephan et al., 2013). The presented genetic
scheme can therefore be used to identify ail-positive and ystB-
negative biotype 1A strains, which should be further studied for
presence of other virulence-associated factors.
Our molecular Y. enterocolitica identification and patho-
serotyping method can be applied in diagnostic and research
laboratories, avoiding the subjectivity of the traditional aggluti-
nation test and the possible misinterpetation of the results, and
enabling higher efficiency, sensitivity and specificity. This assay
may be especially useful for typing Y. enterocolitica isolates which
are conventionally serotyped as “rough” due to lost expression of
the O-ag. Starting from cultivated bacteria, results may be available
in 24–36 h, with the extraction of the genomic DNA from liquid
culture, or in 20 h, using a colony PCR protocol. Ideally, this PCR
assay should be directly applicable to feces samples, reducing the
time for the bacterial identification and allowing a fast treatment of
the patients. The recent enhancements in whole genome sequenc-
ing technologies have made them suitable for typing of bacteria,
282 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283
Fig. 4. Location of the O-ag gene clusters of the analyzed Y. enterocolitica serotypes, using a circular map of the genome sequence of Y. enterocolitica serotype O:8 strain
8081 as reference. The inner ring shows the G + C content, while the outer ring highlights the O-ag genetic clusters (in black) and the main features of the genome (in silver)
(Thomson et al., 2006).
although PCR methods are cheap, fast and straightforward, hence
accessible to a higher number of routine clinical laboratories. The
developed typing scheme can be theferore considered as a start-
ing point for further improvements. Availability of new genome
sequences from Y. enterocolitica strains belonging to less-common
serotypes will also allow a broader molecular serotyping, adding
new targets to the present scheme.
Acknowledgements
We would like to thank Prof. B. Stecher for supplying the
genomic DNA of E. coli and S. enterica, A. Kessler for providing
the L. pneumophila strain and S. Brugiroux for suggestions on the
universal bacterial sequencing. We are grateful to Dr. R.C. Bader
for re-serotyping of strains and to Dr. C. Harrison for revising the
manuscript.
This work was supported by grants 01KI1012H and 01KI1012F
(Food-Borne Zoonotic Infections) from the German Federal Min-
istry of Education and Research (BMBF).
References
Aleksic, S., Bockemuhl, J., Lange, F., 1986. Studies on the serology of flagellar antigens
of Yersinia enterocolitica and related Yersinia species. Zentralblatt fur Bakteri-
ologie, Mikrobiologie, und Hygiene. Series A, Med. Microbiol. Infect. Dis. Virol.
Parasitol. 261, 299–310.
Alikhan, N.F., Petty, N.K., Ben Zakour, N.L., Beatson, S.A., 2011. BLAST Ring Image
Generator (BRIG): simple prokaryote genome comparisons. BMC Genom. 12,
402.
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local align-
ment search tool. J. Mol. Biol. 215, 403–410.
Batzilla, J., Antonenka, U., Höper, D., Heesemann, J., Rakin, A., 2011a. Yersinia entero-
colitica palearctica serobiotype O:3/4 – a successful group of emerging zoonotic
pathogens. BMC Genom. 6, 348.
Batzilla, J., Heesemann, J., Rakin, A., 2011b. The pathogenic potential of Yersinia
enterocolitica 1A. Int. J. Med. Microbiol. 301, 556–561.
Blom, J., Albaum, S.P., Doppmeier, D., Puhler, A., Vorholter, F.J., Zakrzewski, M., Goes-
mann, A., 2009. EDGAR: a software framework for the comparative analysis of
prokaryotic genomes. BMC Bioinform. 10, 154.
Bockemühl, J., Roggentin, P., 2004. Intestinal yersiniosis. Clinical importance,
epidemiology, diagnosis, and prevention. Bundesgesundheitsblatt Gesundheits-
forschung Gesundheitsschutz 47, 685–691.
Bogdanovich, T., Carniel, E., Fukushima, H., Skurnik, M., 2003. Use of O-antigen
gene cluster-specific PCRs for the identification and O-genotyping of Yersinia
pseudotuberculosis and Yersinia pestis. J. Clin. Microbiol. 41, 5103–5112.
Bonardi, S., Bassi, L., Brindani, F., D’Incau, M., Barco, L., Carra, E., Pongolini, S.,
2013. Prevalence, characterization and antimicrobial susceptibility of Salmonella
enterica and Yersinia enterocolitica in pigs at slaughter in Italy. Int. J. Food Micro-
biol. 163, 248–257.
Bottone, E.J., 1997. Yersinia enterocolitica: the Charisma Continues. Clin. Microbiol.
Rev. 10, 257–276.
Caroff, M., Bundle, D.R., Perry, M.B., 1984. Structure of the O-chain of the phenol-
phase soluble cellular lipopolysaccharide of Yersinia enterocolitica serotype O:9.
Eur. J. Biochem./FEBS 139, 195–200.
Darling, A.E., Mau, B., Perna, N.T., 2010. Progressive Mauve: multiple genome align-
ment with gene gain, loss and rearrangement. PLoS ONE 5, e11147.
Fàbrega, A., Vila, J., 2012. Yersinia enterocolitica: pathogenesis, virulence and antimi-
crobial resistance. Enferm. Infecc. Microbiol. Clin. 30, 24–32.
Falcão, J.P., Falcão, D.P., Pitondo-Silva, A., Malaspina, A.C., Brocchi, M., 2006. Molecu-
lar typing and virulence markers of Yersinia enterocolitica strains from human,
animal and food origins isolated between 1968 and 2000 in Brazil. J. Med. Micro-
biol. 55, 1539–1548.
Fredriksson-Ahomaa, M., Stolle, A., Korkeala, H., 2006. Molecular epidemiology of
Yersinia enterocolitica infections. FEMS Immunol. Med. Microbiol. 47, 315–329.
Fuchs, T., Brandt, K., Starke, M., Rattei, T., 2011. Shotgun sequencing of Yersinia
enterocolitica strain W22703 (biotype 2, serotype O:9): genomic evidence for
oscillation between invertebrates and mammals. BMC Genom. 12, 1–14.
Garzetti, D., Bouabe, H., Heesemann, J., Rakin, A., 2012. Tracing genomic variations
in two highly virulent Yersinia enterocolitica strains with unequal ability to com-
pete for host colonization. BMC Genom. 13, 1–15.
Godfroid, J., Saegerman, C., Wellemans, V., Walravens, K., Letesson, J.J., Tibor, A., Mc
Millan, A., Spencer, S., Sanna, M., Bakker, D., Pouillot, R., Garin-Bastuji, B., 2002.
How to substantiate eradication of bovine brucellosis when aspecific serological
reactions occur in the course of brucellosis testing. Vet. Microbiol. 90, 461–477.
Gorshkova, R.P., Kalmykova, E.N., Isakov, V.V., Ovodov, Y.S., 1985. Structural studies
on O-specific polysaccharides of lipopolysaccharides from Yersinia enterocolitica
serovars O:1,2a,3, O:2a,2b,3 and O:3. Eur. J. Biochem./FEBS 150, 527–531.
Gorshkova, R.P., Kalmykova, E.N., Isakov, V.V., Ovodov, Y.S., 1986. Structural studies
on O-specific polysaccharides of lipopolysaccharides from Yersinia enterocolitica
serovars O:5 and O:5,27. Eur. J. Biochem./FEBS 156, 391–397.
Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and
analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98.
Harnett, N., Lin, Y.P., Krishnan, C., 1996. Detection of pathogenic Yersinia entero-
colitica using the multiplex polymerase chain reaction. Epidemiol. Infect. 117,
59–67.
 D. Garzetti et al. / International Journal of Medical Microbiology 304 (2014) 275–283
Jacobsen, N.R., Bogdanovich, T., Skurnik, M., Lubeck, P.S., Ahrens, P., Hoorfar, J., 2005.
A real-time PCR assay for the specific identification of serotype O:9 of Yersinia
enterocolitica. J. Microbiol. Methods 63, 151–156.
Kot, B., Piechota, M., Jakubczak, A., 2010. Analysis of occurrence of virulence
genes among Yersinia enterocolitica isolates belonging to different biotypes and
serotypes. Pol. J. Vet. Sci. 13, 13–19.
Kwaga, J., Iversen, J.O., Misra, V., 1992. Detection of pathogenic Yersinia enterocolitica
by polymerase chain reaction and digoxigenin-labeled polynucleotide probes. J.
Clin. Microbiol. 30, 2668–2673.
Lubeck, P.S., Hoorfar, J., Ahrens, P., Skurnik, M., 2003. Cloning and characterization
of the Yersinia enterocolitica serotype O:9 lipopolysaccharide O-antigen gene
cluster. Adv. Exp. Med. Biol. 529, 207–209.
McNally, A., Cheasty, T., Fearnley, C., Dalziel, R.W., Paiba, G.A., Manning, G., Newell,
D.G., 2004. Comparison of the biotypes of Yersinia enterocolitica isolated from
pigs, cattle and sheep at slaughter and from humans with yersiniosis in Great
Britain during 1999–2000. Lett. Appl. Microbiol. 39, 103–108.
Neubauer, H., Aleksic, S., Hensel, A., Finke, E., Meyer, H., 2000a. Yersinia entero-
colitica 16S rRNA gene types belong to the same genospecies but form three
homology groups. Int. J. Med. Microbiol. 290, 61–64.
Neubauer, H., Hensel, A., Aleksic, S., Meyer, H., 2000b. Identification of Yersinia ente-
rocolitica within the genus Yersinia. Syst. Appl. Microbiol. 23, 58–62.
Oertelt, C., Lindner, B., Skurnik, M., Holst, O., 2001. Isolation and structural charac-
terization of an R-form lipopolysaccharide from Yersinia enterocolitica serotype
O:8. Eur. J. Biochem./FEBS 268, 554–564.
Ovodov, Y.S., Gorshkova, R.P., Tomshich, S.V., Komandrova, N.A., Zubkov, V.A.,
Kalmykova, E.N., Isakov, V.V., 1992. Chemical and immunochemical studies on
lipopolysaccharides of some Yersinia species: a review of some recent investi-
gations. J. Carbohydr. Chem. 11, 21–35.
Paixao, R., Moreno, L.Z., Sena de Gobbi, D.D., Raimundo, D.C., Hofer, E., Matte, M.H.,
Ferreira, T.S., de Moura Gomes, V.T., Costa, B.L., Moreno, A.M., 2013. Char-
acterization of Yersinia enterocolitica biotype 1A strains isolated from swine
slaughterhouses and markets. Sci. World J. 2013, 769097.
Perry, M.B., MacLean, L.L., 1987. Structure of the lipopolysaccharide O-chain of
Yersinia enterocolitica serotype O:5,27. Biochemistry and cell biology/Biochimie
et biologie cellulaire 65, 1–7.
Radziejewska-Lebrecht, J., Shashkov, A.S., Stroobant, V., Wartenberg, K.,
Warth, C., Mayer, H., 1994. The inner core region of Yersinia ente-
rocolitica Ye75R (0:3) lipopolysaccharide. Eur. J. Biochem./FEBS 221,
343–351.
Ramamurthy, T., Yoshino, K., Huang, X., Balakrish Nair, G., Carniel, E., Maruyama, T.,
Fukushima, H., Takeda, T., 1997. The novel heat-stable enterotoxin subtype gene
(ystB) of Yersinia enterocolitica: nucleotide sequence and distribution of the yst
genes. Microb. Pathog. 23, 189–200.
Ratnam, S., Mercer, E., Picco, B., Parsons, S., Butler, R., 1982. A nosocomial outbreak of
diarrheal disease due to Yersinia enterocolitica serotype 0:5, biotype 1. J. Infect.
Dis. 145, 242–247.
Sabina, J., Rahman, A., Chandra Ray, R., Montet, D., 2011. Yersinia enterocolitica: mode
of transmission, molecular insights of virulence, and pathogenesis of infection.
J. Pathog., http://dx.doi.org/10.4061/2011/429069.
283
Samuel, G., Reeves, P., 2003. Biosynthesis of O-antigens: genes and pathways
involved in nucleotide sugar precursor synthesis and O-antigen assembly. Car-
bohydr. Res. 338, 2503–2519.
Skurnik, M., 1999. Molecular genetics of Yersinia lipopolisaccharide. In: Goldberg, J.
(Ed.), Genetics of Bacterial Polysaccharide. CRC Press, Boca Raton, Fl, USA.
Skurnik, M., Bengoechea, J.A., 2003. The biosynthesis and biological role of
lipopolysaccharide O-antigens of pathogenic Yersiniae. Carbohydr. Res. 338,
2521–2529.
Skurnik, M., Venho, R., Toivanen, P., al-Hendy, A., 1995. A novel locus of Yersinia ente-
rocolitica serotype O:3 involved in lipopolysaccharide outer core biosynthesis.
Mol. Microbiol. 17, 575–594.
Stephan, R., Joutsen, S., Hofer, E., Säde, E., Björkroth, J., Ziegler, D., Fredriksson-
Ahomaa, M., 2013. Characteristics of Yersinia enterocolitica biotype 1A strains
isolated from patients and asymptomatic carriers. Eur. J. Clin. Microbiol. Infect.
Dis. 32, 869–875.
Thisted Lambertz, S., Danielsson-Tham, M.L., 2005. Identification and characteriza-
tion of pathogenic Yersinia enterocolitica isolates by PCR and pulsed-field gel
electrophoresis. Appl. Environ. Microbiol. 71, 3674–3681.
Thoerner, P., Bin Kingombe, C.I., Bogli-Stuber, K., Bissig-Choisat, B., Wassenaar, T.M.,
Frey, J., Jemmi, T., 2003. PCR detection of virulence genes in Yersinia enterocolitica
and Yersinia pseudotuberculosis and investigation of virulence gene distribution.
Appl. Environ. Microbiol. 69, 1810–1816.
Thomson, N.R., Howard, S., Wren, B.W., Holden, M.T.G., Crossman, L., Challis, G.L.,
Churcher, C., Mungall, K., Brooks, K., Chillingworth, T., Feltwell, T., Abdellah,
Z., Hauser, H., Jagels, K., Maddison, M., Moule, S., Sanders, M., Whitehead, S.,
Quail, M.A., Dougan, G., Parkhill, J., Prentice, M.B., 2006. The complete genome
sequence and comparative genome analysis of the high pathogenicity Yersinia
enterocolitica strain 8081. PLoS Genet. 2, e206.
Wang, X., Li, Y., Jing, H., Ren, Y., Zhou, Z., Wang, S., Kan, B., Xu, J., Wang, L., 2011.
Complete genome sequence of a Yersinia enterocolitica old world (3/O:9) strain
and comparison with the new world (1B/O:8) strain. J. Clin. Microbiol. 49,
1251–1259.
Wannet, W.J., Reessink, M., Brunings, H.A., Maas, H.M., 2001. Detection of pathogenic
Yersinia enterocolitica by a rapid and sensitive duplex PCR assay. J. Clin. Microbiol.
39, 4483–4486.
Wauters, G., Kandolo, K., Janssens, K., 1987. Revised biogrouping scheme of Yersinia
enterocolitica. Contrib. Microbiol. Immunol. 9, 14–21.
Weynants, V., Jadot, V., Denoel, P.A., Tibor, A., Letesson, J.J., 1996. Detection of Yersinia
enterocolitica serogroup O:3 by a PCR method. J. Clin. Microbiol. 34, 1224–1227.
Zhang, L., al-Hendy, A., Toivanen, P., Skurnik, M., 1993. Genetic organization and
sequence of the rfb gene cluster of Yersinia enterocolitica serotype O:3: simi-
larities to the dTDP-l-rhamnose biosynthesis pathway of Salmonella and to the
bacterial polysaccharide transport systems. Mol. Microbiol. 9, 309–321.
Zhang, L., Radziejewska-Lebrecht, J., Krajewska-Pietrasik, D., Toivanen, P., Skurnik,
M., 1997. Molecular and chemical characterization of the lipopolysaccharide O-
antigen and its role in the virulence of Yersinia enterocolitica serotype O:8. Mol.
Microbiol. 23, 63–76.
Zheng, H., Sun, Y., Mao, Z., Jiang, B., 2008. Investigation of virulence genes in clinical
isolates of Yersinia enterocolitica. FEMS Immunol. Med. Microbiol. 53, 368–374.