International Dairy Journal 84 (2018) 46e53

Contents lists available at ScienceDirect

International Dairy Journal

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

Detection, virulence and antimicrobial resistance of Yersinia

enterocolitica in bulk tank milk in Italy
S. Bonardi a, b, *, A.S. Le Guern c, C. Savin c, G. Pupillo a, L. Bolzoni d, M. Cavalca e,
S. Pongolini d
a
Department of Veterinary Science, Unit of Food Inspection, University of Parma, Via del Taglio 10, 43126, Parma, Italy
b
Multidisciplinary Interdepartmental Dairy Center e MILC, University of Parma, Parma, Italy
c
Yersinia Research Unit and National Reference Laboratory for Yersinia, Institut Pasteur, 28 Rue du Docteur Roux, 75724, Paris, France
d
Risk Analysis Unit, Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia-Romagna, Sezione di Parma, Strada dei Mercati 13/A, 43126, Parma,
Italy
e
National Veterinary Service, Local Unit of Parma, Via Vasari 13/A, 43126, Parma, Italy

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

Article history: In 2016 and 2017, 509 raw milk samples were collected from bulk milk tanks in 315 dairy farms in Parma
Received 15 February 2018 province, Italy. Yersinia pseudotuberculosis was never detected. Yersinia enterocolitica was isolated from 17
Received in revised form (3.1%) bulk milk samples by culture methods. Estimated apparent prevalence was 3.91% in 2016, and
4 April 2018
2.63% in 2017, with no significant difference between the two years (df ¼ 1; p-value ¼ 0.419). Five BT 2/
Accepted 5 April 2018
Available online 13 April 2018
O:5,27 and 12 BT 1A isolates were detected. The most common virulence genes (ail, inv, virF, yadA, ystA,
ystB, myfA, irp2, fyuA) were screened and the genes ail, virF, yadA, ystA were harboured only by 2 O:5,27
isolates. The Italian strains were very close by SNP analysis. The 2 O:5,27 isolates showed a unique
antimicrobial resistance pattern, being resistant to amoxicillin, cefoxitin, cephalexin and amoxicillin -
clavulanic acid. The study suggests that consumption of raw or improperly pasteurised milk may be
hazardous for the consumers.
© 2018 Published by Elsevier Ltd.

1. Introduction Y. entomophaga (Hurst, Becher, Young, Nelson, & Glare, 2011) and
Y. wautersii (Savin et al., 2014).
The genus Yersinia comprises 18 species and belongs to the Y. enterocolitica is a food-borne pathogen responsible for acute
Enterobacteriaceae family (Hurst Becher, Young, Nelson, & Glare, gastroenteritis in humans, but more invasive syndromes such as
2011). Yersinia pestis, Yersinia pseudotuberculosis and Yersinia terminal ileitis, mesenteric lymphadenitis mimicking appendicitis
enterocolitica are pathogens responsible for human and animal and septicaemia in immunocompromised hosts may occur. Post-
diseases. The pathogenic potential for humans of Yersinia wautersii infection sequelae include erythema nodosum, arthritis, and
is still unclear (Savin et al., 2014). While Yersinia ruckeri is a fish glomerulonephritis (Bottone, 1999, 2015).
pathogen and Yersinia entomophaga is an insect pathogen, all the Unlike other pathogenic enterobacteria, Y. enterocolitica is a
other species are non-pathogenic, environmental isolates (Wanger, psychotrophic microorganism able to grow well at refrigeration
2007). In the last few years, new members have been added to the temperatures (Petsios, Fredriksson-Ahomaa, Sakkas, & Papado-
genus, namely Yersinia aleksiciae (Sprague & Neubauer, 2005), poulou, 2016). This ability is of great significance in food safety
Yersinia massiliensis (Merhej, Ade
kambi, Pagnier, Raoult, & because Y. enterocolitica can grow at refrigeration temperature in
Drancourt, 2008), Yersinia similis (Sprague, Scholz, Amann, Busse, vacuum packed or modified atmosphere foods with a prolonged
& Neubauer, 2008), Yersinia nurmii (Murros-Kontiainen et al., shelf-life (Doherty et al., 1995; Fredriksson-Ahomaa, 2012; Nes-
2011a), Yersinia pekkanenii (Murros-Kontiainen et al., 2011b), bakken, 2006). The psychotrophic nature of Y. enterocolitica enables
its survival and growth also in contaminated milk, with subsequent
health risk for the consumers (Jamali, Paydar, Radmehr, & Ismail,
2015). Among consumers, children are particularly at risk for
* Corresponding author. Tel.: þ39 0521 032744. Y. enterocolitica infections (Hoogkamp-Korstanje & Stolk-Engelaar,
E-mail address: [email protected] (S. Bonardi). 1995).

https://doi.org/10.1016/j.idairyj.2018.04.003
0958-6946/© 2018 Published by Elsevier Ltd.
 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53 47

Biochemical and serological heterogeneity are a hallmark of the 2. Materials and methods
microorganism (Gupta, Gulati, Bhagat, Dhar, & Virdi, 2015). Ac-
cording to Wauters, Kandolo, & Janssens (1987), Y. enterocolitica is During MarcheMay 2016 and MarcheMay 2017, 509 raw milk
biochemically divided into six biovars (1A, 1B, 2, 3, 4, 5) and strains samples were collected from bulk milk tanks in 315 dairy farms in
of biovars 1B, 2, 3, 4, and 5 possess the pYV plasmid (plasmid for Parma province, Northern Italy. A total of 281 samples were
Yersinia virulence) and several chromosomally encoded genes that collected in 2016, and 228 samples in 2017. In the two-year period,
are essential for virulence (Skurnik et al., 2010). According to their 212 (67.3%) farms were tested twice. Sample collection was part of
pathogenicity, the biotypes are classified into nonpathogenic the official sampling for brucellosis and leucosis tests in dairy
biotype 1A, weakly pathogenic biotypes 2e5 and highly pathogenic Italian farms. The farms to be tested were selected by using a simple
biotype 1B, which is the only one to harbour the chromosomal randomisation system, including 30% of the farms of the province.
high-pathogenicity island (HPI) (Carniel, 1999; Cornelis et al., Apparent prevalence was estimated using generalised linear
1989a). Nevertheless, a few cases of human infections due to models with binomial (GLM) error distribution with the detection
Y. enterocolitica biotype 1A have been suspected (Tennant, Grant, & of Y. enterocolitica as response variable. In addition, we used the
Robins-Browne, 2003). According to some researchers, the biotype GLM to test whether significant differences in Y. enterocolitica
1A may represent indeed more than one genetic group, with prevalence exist between the two years of sampling through a
different pathogenic potential (Sihvonen et al., 2012). Serologically, likelihood ratio test against the null model. Analyses were per-
over 70 serotypes of Y. enterocolitica have been identified (Skurnik formed using package “MASS” in statistical software R (R
et al., 2010), with the invasive strains predominantly belonging to Development Core Team, 2016).
serotypes O:3, O:9, O:5,27 and O:8 (Kirjavainen et al., 2008). Milk samples were collected in sterile containers and refriger-
Yersiniosis is the third most common reported zoonosis in the ated until being processed within 24 h.
European Union (EU), with a notification rate of 1.8 cases per
100,000 inhabitants in 2016. The highest country-specific notifi- 2.1. Detection of Y. enterocolitica and Y. pseudotuberculosis
cation rates were observed in north-eastern Europe; Finland, Czech
Republic and Lithuania reported 7.4, 5.8 and 5.4 cases per 100,000 Milk samples were tested for Y. enterocolitica and
inhabitants, respectively. Y. enterocolitica was the most commonly Y. pseudotuberculosis presence by cold enrichment. For
reported species, responsible for 99.1% of the confirmed cases at EU Y. enterocolitica detection 10 mL were diluted 1:10 in phosphate
level. Thirty-nine percent of the isolates were serotyped, and the buffered saline containing 2% sorbitol and 1.5% bile salts (PSB;
most common serotype was O:3 (84.6%), followed by O:9 (11.8%) Biolife, Milano, Italy). For Y. pseudotuberculosis detection 10 mL
and O:8 (1.7%). Biotype data were reported for 4.6% of the were diluted 1:10 in PBS containing 1% mannitol and 0.15% bile salts
confirmed cases and the most common was biotype 4 (79.6%), (PMB). The suspensions were incubated at 4 ± 1  C for 14 and 21
followed by biotype 2 (16.9%) and biotype 3 (2.5%) (EFSA & ECDC, days. Thereafter, 10 mL were streaked onto Cefsulodin-Irgasan-
2017). Y. pseudotuberculosis was reported as causative agent of Novobiocin Yersinia-selective agar (CIN agar; Oxoid, Basingstoke,
0.9% of the yersiniosis cases in 2016, mainly by Finland and the UK) plates, incubated at 30 ± 1  C for 48 h. In parallel, the enriched
United Kingdom (EFSA & ECDC, 2016). PSB and PMB cultures were treated with alkali before plating onto
Y. enterocolitica is recovered from a broad range of animal spe- CIN agar, mixing 0.5 mL of the broth cultures with 4.5 mL of 0.5%
cies, including livestock animals, domestic pets and wild animals potassium hydroxide (KOH) solution for 20 s. After mixing, 10 mL of
(Le Guern, Martin, Savin, & Carniel, 2016; Wang et al., 2009). the alkali treated cultures were plated onto CIN agar plates, incu-
Among farm animals, pigs are considered the most important bated at 30 ± 1  C for 48 h. Flat, not mould colonies with entire edge
reservoir of the microorganism, which finds a niche in their having a red centre (“bull's-eye”) surrounded by a translucent,
lymphatic tissue and may be shed by the faecal route (Bonardi et al., transparent or milk-white zone were considered suspect
2013; Fredriksson-Ahomaa, Stolle, & Stephan, 2007). Cattle may be Y. enterocolitica colonies. Purple and small (<1 mm diameter) col-
infected by Y. enterocolitica, although prevalence is generally lower onies on CIN agar were considered suspect Y. pseudotuberculosis
than in sheep or pigs (McNally et al., 2004) and infection is mild or colonies.
asymptomatic (Schmid, Messelh€ ausser, Ho€ rmansdorfer, Sauter- Characteristic colonies were subcultured in tryptone soy agar
Louis, & Mansfeld, 2013). Pathogenic strains belonging to bio- (TSA; Oxoid, Basingstoke, UK) at 25 ± 1  C for 24 h. Preliminary
serotype 3/O:5,27 were shared by cattle and humans in Great identification was performed by seeding pure colonies in Kligler
Britain, thus indicating that cattle may play a role in the epidemi- iron agar (Biolife, Milano, Italy) and Christensen's urea agar (Biolife,
ology of yersiniosis (Bonardi et al., 2007; McNally et al., 2004). Milano, Italy) incubated at 25 ± 1  C for 24 h. Lactose-negative and
Furthermore, among the different serotypes, Y. enterocolitica O:9 is urease-positive colonies were further tested for melibiose, sorbitol,
of concern in cattle because of the false-positive serological re- sucrose and rhamnose fermentation at 25 ± 1  C for 24 h
actions it can cause in brucellosis tests (Perry & Bundle, 1990). (Fredriksson-Ahomaa, 2007). Species identification was performed
Transmission of pathogenic Y. enterocolitica by raw milk, improp- using the API® 20E system (bioMe
rieux, Marcy l’Etoile, France)
erly pasteurised milk and dairy products has been reported from incubated at 25 ± 1  C for 48 h.
several countries worldwide such as the United Kingdom (Barrett,
1986; Greenwood, Hooper, & Rodhouse, 1990), Sweden 2.2. Characterisation of Y. enterocolitica
(Alsterlund et al., 1995), Australia (Butt, Gordon, Lee-Archer, Moritz,
& Merrell, 1991), the United States (Ackers et al., 2000; Black et al., Biotyping and serotyping of Y. enterocolitica isolates were car-
1978; Longenberger et al., 2014) and India (Abraham et al., 1997). ried out by the French Yersinia National Reference Laboratory
The present study investigated the prevalence of Y. enterocolitica (YNRL), Institut Pasteur, Paris, France. Isolates were biotyped us-
and Y. pseudotuberculosis in bulk tank cow milk in dairy farms of ing metabolic tests: API® 20 E and API® 50CH systems (bio-
Parma province (Northern Italy). Virulence characterisation, anti- Me
rieux), pyrazinamidase (Kandolo & Wauters, 1985) and Tween
microbial susceptibility of the isolates and genetic relatedness esterase activity (Lee, 1977), according to the biogrouping scheme
evaluation of pathogenic strains contributed to the evaluation of of Wauters et al. (1987). Serotyping was carried out by slide
the risk posed by Y. enterocolitica through the consumption of raw agglutination using commercial antisera O:3 (Bio-Rad, Segrate
or improperly pasteurised milk in this province. (MI), Italy) and O:5,27 (Sifin, Berlin, Germany) and a panel of 45 O:
48 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53

antigen-specific antisera prepared at the Institut Pasteur by 2017). Mueller-Hinton agar and commercial antimicrobial suscep-
immunisation of rabbits. tibility discs (Bio-Rad, Segrate (MI), Italy) were used. Plates were
incubated at 28 ± 1  C for 18e24 h. Pathogenic Yersinia isolates
2.3. Typing of Y. enterocolitica were tested for amoxicillin (20 mg), ticarcillin (75 mg), amoxicillin
and clavulanic acid (20 mg and 10 mg, respectively), cephalexin
The whole genome of the isolates was sequenced with NEXTERA (30 mg), cefoxitin (30 mg), ceftriaxone (30 mg), nalidixic acid (30 mg),
XT protocol using a NextSeq® 500 sequencer (Illumina, San Diego, ciprofloxacin (5 mg), tetracycline (30 mg), sulphonamide (200 mg)
USA). Sequencing allowed the collection of paired-end reads of 150 and trimethoprim (5 mg). Susceptibility results were categorised as
cycles. Based on the raw reads, de novo assembly of the genomes susceptible, intermediate or resistant according to the 2017 EUCAST
with SPAdes 3.6 was performed as described by Saraka et al. (2017). recommendations.
Genomes of the Y. enterocolitica isolates were tested for the
presence of plasmid- and chromosome-borne virulence genes
(Table 1). The plasmid pYV was detected by yadA and virF. The 3. Results
chromosomal virulence genes tested comprised ail (attachment
invasion locus), inv (invasin), ystA (Yersinia stable toxin A), ystB Y. enterocolitica was isolated from 17 bulk milk samples using
(Yersinia stable toxin B), myfA (mucoid Yersinia factor A), irp2 (iron- culture methods (3.1%). One additional isolate, previously identified
repressible protein 2) and fyuA (ferric yersiniabactin uptake). as Y. enterocolitica, but presenting atypical biochemical character-
The virulence genes were detected by BLAST of reference se- istics, was reclassified as Y. massiliensis. In 2016, 11 of 281 samples
quences on the genomes. The reference sequences were recovered were positive to Y. enterocolitica; the estimated apparent preva-
from the database of the National Center for Biotechnology Infor- lence was 3.91% (95% CI: 2.05e6.61%). In 2017, 6 of 228 were pos-
mation (NCBI, Bethesda, Maryland, USA): ail (Accession number: itive to Y. enterocolitica; the estimated apparent prevalence was
HF933425), inv (Accession number: HF933425), virF (Gene ID: 2.63% (95% CI: 1.05e5.26%). Despite a larger number of positive
4711913), yadA (Gene ID: 4711901), irp2 (Gene ID: 4714263), fyuA
(Gene ID: 4714268), myfA (Accession number: HF33424). Reference
sequences for ystA and ystB were determined as described by Table 2
Thoerner et al. (2003) from genes published in the NCBI: ystA Bio-serotyping and virulence gene typing of 17 Y. enterocolitica isolates from bulk
(Accession number: CP009846.1, 96-176), and ystB (Accession tank milk.a
number: D88145.1, 66-211). Isolate Biotype Serotype Virulence genes
The genetic relatedness of pathogenic Y. enterocolitica 2/O:5,27 number
ail inv virF yadA myfA ystA ystB irp2 fyuA
isolates was evaluated by a whole genome single nucleotide poly-
morphism (SNP) analysis with Bionumerics 7.6 (Applied Maths, Saint- IP39242 2 O:5,27 þ þ þ þ þ e e e e
IP39243 2 O:5,27 þ þ þ þ þ e e e e
Martens-Latem, Belgium). Y. enterocolitica 3/O:5,27 genome of YE149/ IP39244 2 O:5,27 þ þ e e þ e e e e
02 strain (accession number: HF933424) was used as reference. All IP39245 2 O:5,27 þ þ þ þ þ e e e e
the Y. enterocolitica pathogenic strains with O:5,27 serotype from IP39265 2 O:5,27 þ þ þ þ þ þ e e e
Reuter et al. (2014) were included in the comparison: they consist in IP39246 1A O:5 e þ e e e e þ e e
IP39247 1A O:5 e þ e e e e þ e e
16 strains mostly isolated in Europe: 9 from United Kingdom, 3 from
IP39250 1A O:5 e þ e e e e þ e e
France, 1 from Germany, 1 from Greece, 1 from New Zealand and 1 IP39251* 1A O:5 e þ e e e e þ e e
from an unknown origin. Genomic data publicly available were IP39256* 1A O:5 e þ e e e e þ e e
downloaded and used for the whole genome SNP analysis. IP39257 1A O:5 e þ e e e e þ e e
IP39259 1A O:5 e þ e e e e e e e
IP39248 1A O:6,30e6,31 e þ e e e e e e e
2.4. Antimicrobial susceptibility of Y. enterocolitica IP39253 1A O:6,30e6,31 e þ e e þ e þ e e
IP39252x 1A O:7,8-13 e þ e e e e þ e e
Antibiotic susceptibility was tested using the Kirby Bauer disc- IP39258x 1A O:7,8-13 e þ e e e e þ e e
diffusion method, according to the recommendations of the Euro- IP39255 1A O:10-34 e þ e e þ e e e e
a
pean committee on antimicrobial susceptibility testing (EUCAST, The symbols* and x identify pairs of isolates coming from the same farms.

Table 1
Virulence-associated genes in Y. enterocolitica investigated in the current study.a

Genes Gene product/function References

C
inv Invasin (outer membrane protein essential for bacteria translocation across the Grassl, Bohn, Müller, Bühler, & Autenrieth., 2003; Pepe & Miller, 1993
intestinal epithelium)
C
ail Adhesin (outer membrane protein contributing to adhesion, invasion and Bliska & Falkow, 1992; Pederson & Pierson, 1995
resistance to complement-mediated lysis)
virF P
Transcriptional activator controlling the Yop regulon €lin, Forsberg, Norlander, Skurnik, & Wolf-Watz, 1988; Cornelis,
Bo
Sluiters, De Rouvroit, & Michiels, 1989b
yadAP Yersinia adhesin A (adhesion to host ileocaecal epithelium) Bottone, 1997; Schulze-Koops Burkhardt, Heesemann, Von Der Mark, K.,
& Emmrich, 1992; Schulze-Koops et al., 1993
myfAC Mucoid Yersinia factor (fimbrial antigen and putative adhesin) Diaz, Urra, Toyos, & Moriyon, 1985; Iriarte et al., 1993
ystAC Enterotoxin (Yersinia stable toxin A which may contribute to the pathogenesis of Delor, Kaeckenbeeck, Wauters, & Cornelis, 1990
diarrhoea)
ystBC Enterotoxin (Yersinia stable toxin B which may contribute to the pathogenesis of Ramamurthy et al., 1997
diarrhoea)
irp2H Iron-repressible high-molecular-weight protein HMWP2 presumably involved Carniel, Guiyoule, Guilvout, & Mercereau-Puijalon, 1992;
in the production of yersiniabactin Lucier,Fetherston, Brubaker, & Perry, 1996
fyuAH yersibactin receptor FyuA (ferric yersiniabactin uptake) Rakin, Saken, Harmsen, & Heesemann, 1994
a
Abbreviations are: C, chromosome-borne gene; P, Plasmid-borne gene; H, high-pathogenicity-island gene.
 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53 49

samples in 2016, no significant difference in Y. enterocolitica prev- years, the same biotype was detected (5 and 1A) (Table 2). Then, the
alence between the two years of sampling was observed (df ¼ 1; p- apparent prevalence of farms positive to biotypes 1A and 2 were
value ¼ 0.419). In two farms Y. enterocolitica was detected in bulk 3.17% (95% CI: 1.60e5.50%) and 1.59% (95% CI: 0.57e3.38%),
tank milk samples both in 2016 and 2017. It follows that 15 farms respectively.
(over 315) had at least one sample of bulk tank milk positive to The 17 genomes were screened for the presence of the most
Y. enterocolitica; then, the apparent prevalence of Y. enterocolitica- common virulence genes (Table 2). The ail, virF, yadA, ystA genes
positive farms was 4.76% (95% CI: 2.77e7.49%). were harboured only by pathogenic 2/O:5,27 isolates. However, one
Y. pseudotuberculosis was never detected in the bulk tank milk 2/O:5,27 isolate (IP39244) may lack the pYV plasmid as it lacks virF
samples tested. and yadA sequences, and four isolates lacked the ystA gene. The
Bio-serotyping of Y. enterocolitica isolates identified five patho- myfA gene was present in all pathogenic isolates and in two non-
genic isolates (2/O:5,27) and 12 non-pathogenic isolates (biotype pathogenic isolates (IP39253 and IP39255). The irp2 and fyuA
1A) (Table 2). In farms where Y. enterocolitica was isolated in both genes located on the high pathogenicity island of the highly

Fig. 1. SNP analysis of Y. enterocolitica 2/O:5,27 isolates from different countries.

50 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53

pathogenic 1B isolates were never present. Among 1A isolates, the slaughter were positive for Y. enterocolitica in rectal samples.
ail, yadA, virF, and ystA, genes were never present. The inv and ystB Among the isolates, only two were pathogenic 3/O:5,27 strains,
genes were present in 100% and 75% of the 1A isolates, respectively. whilst the other were biotype 1A (McNally et al., 2004). Faecal
Two 1A isolates harboured the myfA gene (Table 2). contamination of milk may occur on farm during milking practice.
The genomes of the five pathogenic 2/O:5,27 isolates along However, also storage of milk in contaminated tanks may represent
with other 16 strains with public genomes were compared by a risk. Refrigeration of milk at maximum 6e8  C on farm and at
whole genome SNP analysis (Fig. 1). Excluding the Italian strains, maximum 10  C during transportation to the dairy factories
there was a minimum distance between strains of 29 SNPs (Regulation EC No. 853/2004: EC, 2004) may favour growth of this
(IP26249 from France versus H582/87 with an unknown origin) psychotropic microorganism compared to other pathogenic mes-
and a maximum distance of 112 SNPs (YE153/02 from United ophilic bacteria.
Kingdom versus ERL032126 from New Zealand) indicative of an Among food production animals, Y. enterocolitica O:5,27 strains
overall lack of relatedness amongst the isolates analysed. Strains are mainly isolated from pigs (Bottone, 1997; Fukushima, Gomyoda,
from the United Kingdom are spread over the whole minimum Aleksic, & Tsubokura, 1993), but detection in cattle may occur
spanning tree suggesting that they have no close genetic rela- (Fukushima et al., 1993; McNally et al., 2004). In Parma province,
tionship. An exception appears to be regarding YE04/03 and YE08/ pigs and cattle are reared in separate farms, therefore, we consider
03 strains that had no SNPs between them, meaning they are transmission of Y. enterocolitica 2/O:5,27 between the two species
closely related. The 3 strains from France are also spread across unlikely.
different branches of the tree meaning there is no close rela- SNPs analysis showed that the 2/O:5,27 isolates are genetically
tionship between the French isolates. Italian isolates’ closest very close (maximum of 1 SNP) and suggest that the same strain
relative is ERL032126 strain from New Zealand (65 SNPs) circulates between the dairy farms. The hypothesis is supported by
excluding a close genetic relationship based on the geographical the study of Saraka et al. (2017) who demonstrated the circulation
origin. The whole genome SNP comparison between Italian iso- of a Y. enterocolitica 4/O:3 strain between pig farms in Co^te d’Ivoire
lates revealed the presence of two groups with no SNPs inside. by a whole genome SNP analysis: four isolates collected in 3 farms
Those groups are separated by only one SNP: one group of three exhibited a maximum distance of 7 SNPs.
isolates (IP39242, IP39244 and IP39265) and the second of two The pathogenic nature of Y. enterocolitica serotype O:5,27 is
isolates (IP39243 and IP39245). This finding suggests that the five well-known, being able to adhere to and invade human intestinal
isolates are genetically very close and that the same strain circu- epithelial cells, survive and replicate in macrophages and induce a
lates between the investigated dairy farms. cytokine response in the host (Schaake et al., 2013). After invasion
The Italian 2/O:5,27 Y. enterocolitica isolates showed a unique and transmigration through the M cells of the intestinal layer,
antimicrobial resistance pattern. They were sensitive to tetracy- yersiniae are encountered by neutrophils and macrophages, which
cline, ticarcillin, ciprofloxacin, nalidixic acid, ceftriaxone, trimeth- are part of the first immune response. The ability of the microor-
oprim and sulfamethoxazole and resistant to amoxicillin, cefoxitin, ganisms to actively invade, persist, and replicate within macro-
cephalexin and amoxicillin - clavulanic acid. phages is of the greatest importance during early stages of infection
(Pujol & Bliska, 2005). Y. enterocolitica serotype O:5,27 belongs to
4. Discussion biotype 2 (indole-positive strains) or biotype 3 (indole-negative
strains) (Fukushima et al., 1993) and it is responsible for human
The culture methods conventionally employed for the detection yersiniosis cases worldwide (Bottone, 1997). In 2015, 1.6% of the
of Y. enterocolitica in food include cold-enrichment, plating of human isolates typed in the EU were serotype O:5,27, whilst 82%
enriched samples onto selective media and identification at species were O:3 and 11% were O:9 (EFSA & ECDC, 2016).
level of the isolates by biochemical tests (Gupta et al., 2015). A new In our study, the pathogenic 2/O:5,27 isolates harboured the
version of the International Organisation for Standardisation (ISO) well-established Yersinia virulence genes. The virF and yadA genes,
10273 method has been recently published (ISO, 2017), but it was which are located on the plasmid of virulence pYV, were present in
not available at the time of the study. The previous version of the four out of the five isolates, and one isolate lacked the plasmid.
method (ISO, 2003) included two different enrichments of the However, the loss of pYV is common after a few subcultures in
samples using peptone, sorbitol and bile salts (PSB) broth as a 10- laboratory, thus the isolate likely hosted the pYV at the moment of
fold dilution and irgasan, ticarcillin and potassium-chlorate (ITC) sampling and must actually be considered as pathogenic. The ystA
broth as a 100-fold dilution, followed by plating of the enriched gene, which codes a heat-stable enterotoxin in pathogenic
samples onto two selective media after potassium hydroxide (KOH) Y. enterocolitica strains, was found in only one 2/O:5,27 isolate. The
treatment and without KOH treatment. However, the complexity of four isolates that lacked ystA also lacked ystB, suggesting they had
the method was not balanced by high recovery rates of lost the whole yst gene. The loss of the same gene in four isolates
Y. enterocolitica (Van Damme, Habib, & De Zutter, 2010; Van detected in four different farms is an uncommon event. In addition,
Damme et al., 2013) and for this reason it was not applied in our the five O:5,27 isolates are genetically very close. This suggests that
study. the five isolates may have the same source. The five O:5,27 isolates
Most investigations on Y. enterocolitica in EU countries are also harboured ail, inv and myfA which are usually found in path-
related to pork and products thereof (EFSA & ECDC, 2016). In 2015, ogenic Yersinia.
investigations on Yersinia in milk and dairy products were reported In the current study, prevalence of farms harbouring
by a few EU countries and all of them with less than 10 samples; Y. enterocolitica biotype 1A in raw milk was 3.17%. This biotype is
positive results were found for bovine and goat milk, but not for mostly considered non-pathogenic because it lacks the pYV plasmid
dairy products (EFSA & ECDC, 2016). In 2016, only a very few data and major chromosomal virulence genes (Bhagat & Virdi, 2011).
were reported and no positive milk samples were found (EFSA & Nevertheless, some biotype 1A strains have occasionally been asso-
ECDC, 2017). In this context, our results from a random sample of ciated with food-borne gastroenteritis in humans, causing symptoms
315 dairy farms can contribute important information on the role of indistinguishable from those produced by pathogenic biotypes
raw milk in foodborne human yersiniosis. (Bhagat & Virdi, 2011; Tennant et al., 2003). In 2015, several cases of
Faecal carriage of Y. enterocolitica by cattle is well recognised. A human infections by biotype 1A were reported from Denmark, rep-
national survey in Great Britain found that 6.3% of 891 cattle at resenting almost half of the Y. enterocolitica isolates typed by this
 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53 51

country (EFSA & ECDC, 2016). Biovar 1A strains are also known to 5. Conclusions
invade epithelial cells and show resistance to phagocytosis by mac-
rophages (Grant, Bennett-Wood, & Robins-Browne, 1999). In our study, prevalence of farms harbouring pathogenic
Different studies have been carried out to investigate which Y. enterocolitica 2/O:5,27 in raw cow milk was 1.59%. Since data
factors might contribute to pathogenicity in biovar 1A strains. about Y. enterocolitica contamination of raw milk are scarce in the
Intraphagocytic survival was evaluated in two clones of biovar 1A, EU countries, our findings are hardly comparable with other Eu-
which varied for distribution of virulence genes like those encoding ropean surveys. In the geographical area of the survey, cow milk is
insecticidal toxins (tccC), mucoid Yersinia factor (myfA), subtilisin/ mainly used to produce Parmesan cheese (Parmigiano-Reggiano
kexin-like protease (hreP), ferric enterochelin receptor (fepA) and cheese), which is an Italian hard cheese aged for 12e36 months.
Yersinia stable toxin (ystB) (Bhagat & Virdi, 2007). Both clones of Different pathogenic bacteria may potentially contaminate raw
biovar 1A were able to adhere to epithelial cells in vitro, but inva- milk, such as Salmonella, Y. enterocolitica, Campylobacter jejuni,
sion was low if compared with other biovars (Dhar & Virdi, 2013), Staphylococcus aureus and Bacillus cereus, but their growth and
probably because biovar 1A strains lack YadA (Yersinia adhesion) toxin production is normally inhibited by lactic acid bacteria and
and Ail (adhesion and invasion locus), which are known to aid the low pH (Giffel & Wells-Bennik, 2010; Hill & Kethireddipalli, 2013).
microorganism in adhesion and invasion to the host cells (Bhagat & As a consequence, microbiological hazards are absent in the aged
Virdi, 2007; El Tahir & Skurnik, 2001). cheese and Enterobacteriaceae are usually no more detectable after
Survival in macrophages was higher in the strains carrying the the first 24 h of ripening (Coppola et al., 2000). For these reasons,
febA gene, which should be further investigated to ascertain the true Y. enterocolitica 2/O:5,27 in raw milk used to produce Parmigiano-
pathogenic potential of biovar 1A strains (Dhar & Virdi, 2013). Reggiano cheese does not represent a risk for the consumer.
Another important observation highlights that Y. enterocolitica bio- Conversely, consumption of raw milk or improperly pasteurised
var 1A strains show great phenotypic heterogeneity, in contrast to milk and products might be potentially hazardous (Longenberger
their relatively limited genotypic heterogeneity, thus suggesting et al., 2014).
that the variations in macrophage response may depend on the role
of urease, superoxide dismutase and the ferric enterochelin receptor Acknowledgements
(Dhar & Virdi, 2013). Nevertheless, genes related to iron acquisition
and storage were detected both in clinical and non-clinical 1A iso- The authors gratefully acknowledge Dr. Gisella Pizzin and Mrs.
lates, thus suggesting they might not be responsible for a better Ida Poli (University of Parma) for technical assistance. The study
ability to cause infection in humans (Kanaujia, Bajaj, & Virdi, 2015). was funded by Fondazione Cariparma, Parma, Italy (Project No
In the current study, the 1A isolates detected in raw milk were 2013.0237).
tested for the most important virulence determinants. Two out of
12 isolates carried the myfA gene, whose role in Y. enterocolitica 1A References
adhesion to epithelial cells has not been demonstrated (Dhar &
Virdi, 2013). Interestingly, 1A strains isolated from children with Abraham, M., Pai, M., Kang, G., Asokan, G. V., Magesh, S. R., Bhattacharji, S., et al.
diarrhoea harboured the myfA gene (Kot, Piechota, & Jakubiak, (1997). An outbreak of food poisoning in Tamil Nadu associated with Yersinia
enterocolitica. Indian Journal of Medical Research, 106, 465e468.
2017). All 1A isolates harboured the inv gene. However, inv has Ackers, M. L., Schoenfeld, S., Markman, J., Smith, M. G., Nicholson, M. A., DeWitt, W.,
been shown non-functional in the non-pathogenic strains of et al. (2000). An outbreak of Yersinia enterocolitica O: 8 infections associated
Y. enterocolitica because of transcriptional default (Pierson & with pasteurized milk. The Journal of Infectious Diseases, 181, 1834e1837.
Alsterlund, R., Danielsson-Tham, M. L., Karp, G., Eden, T., De Jong, B., Nilsson, P. O.,
Falkow, 1990). The ystB gene is frequently carried by 1A isolates et al. (1995). An outbreak of Yersinia enterocolitica infection on the Bja €rred
(Ramamurthy et al., 1997) and was harboured by the majority of the Peninsula. Indications for risks of refrigerated food. Lakartidningen, 92,
isolates detected in cow milk. Nevertheless, pathogenicity attri- 1213e1214.
Barrett, N. J. (1986). Communicable disease associated with milk and dairy products
bution to Y. enterocolitica biovar 1A strains is still difficult.
in England and Wales: 1983e1984. Journal of Infection, 12, 265e272, 275.
The 2/O:5,27 Y. enterocolitica isolates showed a unique antimi- Baumgartner, A., Küffer, M., Suter, D., Jemmi, T., & Rohner, P. (2007). Antimicrobial
crobial resistance pattern. They were sensitive to tetracycline, resistance of Yersinia enterocolitica strains from human patients, pigs and retail
pork in Switzerland. International Journal of Food Microbiology, 115, 110e114.
ticarcillin, ciprofloxacin, nalidixic acid, ceftriaxone, trimethoprim
Bhaduri, S., Wesley, I., Richards, H., Draughon, A., & Wallace, M. (2009). Clonality
and sulfamethoxazole and resistant to amoxicillin, cefoxitin, and antibiotic susceptibility of Yersinia enterocolitica isolated from U.S. market
cephalexin and amoxicillin - clavulanic acid. weight hogs. Foodborne Pathogens and Disease, 6, 351e356.
The present study confirms the sensitivity of Y. enterocolitica 2/ Bhagat, N., & Virdi, J. S. (2007). Distribution of virulence-associated genes in Yersinia
enterocolitica biovar 1A correlates with clonal groups and not the source of
O:5,27 pathogenic strains to fluoroquinolones and third generation isolation. FEMS Microbiology Letters, 266, 177e183.
cephalosporins, which are recommended for therapeutic use in Bhagat, N., & Virdi, J. S. (2011). The enigma of Yersinia enterocolitica biovar 1A.
case of human illness (Fa brega & Vila, 2012). Susceptibility to Critical Reviews in Microbiology, 37, 25e39.
Black, R. E., Jackson, R. J., Tsai, T., Medvesky, M., Shayegani, M., Feeley, J. C., et al.
ticarcillin of the serotype O:5,27 is due to the absence of enzyme A, (1978). Epidemic Yersinia enterocolitica infection due to contaminated chocolate
a broad spectrum chromosomal b-lactamase (Cornelis & Abraham, milk. New England Journal of Medicine, 298, 76e79.
1975), and it is a characteristic of the serotype (Pham, Bell, Martin, Bliska, J. B., & Falkow, S. (1992). Bacterial resistance to complement killing mediated
by the Ail protein of Yersinia enterocolitica. Proceedings of the National Academy
& Carniel, 2000). of Sciences, 89, 3561e3565.
Resistance to ampicillin and first generation cephalosporins in €lin, I., Forsberg, A., Norlander, L., Skurnik, M., & Wolf-Watz, H. (1988). Identifi-
Bo
Y. enterocolitica is well documented (Baumgartner, Küffer, Suter, cation and mapping of the temperature-inducible, plasmid-encoded proteins of
Yersinia spp. Infection and Immunity, 56, 343e348.
Jemmi, & Rohner, 2007; Bhaduri, Wesley, Richards, Draughon, & Bonardi, S., Bassi, L., Brindani, F., D'Incau, M., Barco, L., Carra, E., et al. (2013).
Wallace, 2009; Pham et al., 2000). Resistance to cefoxitin, Prevalence, characterization and antimicrobial susceptibility of Salmonella
together with resistance to ampicillin, was found in 2/O:5,27 strains enterica and Yersinia enterocolitica in pigs at slaughter in Italy. International
~o et al., 2004). Journal of Food Microbiology, 163, 248e257.
from fresh water in Brazil (Falca
Bonardi, S., Paris, A., Bacci, C., D'Incau, M., Ferroni, L., & Brindani, F. (2007). Detection
Resistance to amoxicillin - clavulanic acid seems to be common and characterization of Yersinia enterocolitica from pigs and cattle. Veterinary
in Y. enterocolitica 2/O:5,27 strains, as it was observed in 2/O:5,27 Research Communications, 31, 347e350.
isolates in the USA, UK, Australia, France and Japan (Pham et al., Bottone, E. J. (1997). Yersinia enterocolitica: The charisma continues. Clinical
Microbiology Reviews, 10, 257e276.
2000), as well as in all the 2/O:5,27 strains characterized by the Bottone, E. J. (1999). Yersinia enterocolitica: Overview and epidemiologic correlates.
French YNRL between 2012 and 2017 (Le Guern, unpublished data). Microbes and Infection, 1, 323e333.
52 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53

Bottone, E. J. (2015). Yersinia enterocolitica: Revisitation of an enduring human Hurst, M. R., Becher, S. A., Young, S. D., Nelson, T. L., & Glare, T. R. (2011). Yersinia
pathogen. Clinical Microbiology Newsletter, 37, 1e8. entomophaga sp. nov., isolated from the New Zealand grass grub Costelytra
Butt, H. L., Gordon, D. L., Lee-Archer, T., Moritz, A., & Merrell, W. H. (1991). Rela- zealandica. International Journal of Systematic and Evolutionary Microbiology, 61,
tionship between clinical and milk isolates of Yersinia enterocolitica. Pathology, 844e849.
23, 153e157. Iriarte, M., Vanooteghem, J. C., Delor, I., Diaz, R., Knutton, S., & Cornelis, G. R.
Carniel, E. (1999). The Yersinia high-pathogenicity island. International Microbiology, (1993). The Myf fibrillae of Yersinia enterocolitica. Molecular Microbiology, 9,
2, 161e167. 507e520.
Carniel, E., Guiyoule, A., Guilvout, I., & Mercereau-Puijalon, O. (1992). Molecular ISO. (2003). Microbiology of food and animal feeding stuffs e horizontal method for the
cloning, iron-regulation and mutagenesis of the irp2 gene encoding HMWP2, a detection of presumptive pathogenic Yersinia enterocolitica. EN ISO 10273:2003.
protein specific for the highly pathogenic Yersinia. Molecular Microbiology, 6, Geneva, Switzerland: International Standardisation Organisation.
379e388. ISO (International Organization for Standardization). (2017). Microbiology of the food
Coppola, R., Nanni, M., Iorizzo, M., Sorrentino, A., Sorrentino, E., Chiavari, C., et al. chain. Horizontal method for the detection of pathogenic Yersinia enterocolitica
(2000). Microbiological characteristics of Parmigiano Reggiano cheese during ISO 10273:2017. Geneve, Switzerland.
the cheesemaking and the first months of the ripening. Lait, 80, 479e490. Jamali, H., Paydar, M., Radmehr, B., & Ismail, S. (2015). Prevalence, characterization,
Cornelis, G., & Abraham, E. (1975). nbsp;-lactamases from Yersinia enterocolitica. and antimicrobial resistance of Yersinia species and Yersinia enterocolitica iso-
Journal of General Microbiology, 87, 273e284. lated from raw milk in farm bulk tanks. Journal of Dairy Science, 98, 798e803.
Cornelis, G. R., Biot, T., Rouvroit, C. L., Michiels, T., Mulder, B., Sluiters, C., et al. Kanaujia, P. K., Bajaj, P., & Virdi, J. S. (2015). Analysis of iron acquisition and storage-
(1989a). The Yersinia yop regulon. Molecular Microbiology, 3, 1455e1459. related genes in clinical and non-clinical strains of Yersinia enterocolitica biovar
Cornelis, G., Sluiters, C., De Rouvroit, C. L., & Michiels, T. (1989b). Homology between 1A. Apmis, 123, 858e866.
virF, the transcriptional activator of the Yersinia virulence regulon, and AraC, the Kandolo, K., & Wauters, G. (1985). Pyrazinamidase activity in Yersinia enterocolitica
Escherichia coli arabinose operon regulator. Journal of Bacteriology, 171, and related organisms. Journal of Clinical Microbiology, 21, 980e982.
254e260. Kirjavainen, V., Jarva, H., Biedzka-Sarek, M., Blom, A. M., Skurnik, M., & Meri, S.
Delor, I., Kaeckenbeeck, A., Wauters, G., & Cornelis, G. R. (1990). Nucleotide (2008). Yersinia enterocolitica serum resistance proteins YadA and Ail bind the
sequence of yst, the Yersinia enterocolitica gene encoding the heat-stable complement regulator C4b-binding protein. PLoS Pathogens, 4. Article
enterotoxin, and prevalence of the gene among pathogenic and nonpatho- e1000140.
genic yersiniae. Infection and Immunity, 58, 2983e2988. Kot, B., Oiechota, M., & Jakubiak, K. (2017). Virulence genes and antibiotic resistance
Dhar, M. S., & Virdi, J. S. (2013). Interaction of Yersinia enterocolitica biovar 1A with of Yersinia enterocolitica strains isolated from children. European Journal of
cultured cells in vitro does not reflect the two previously identified clonal Biological Research, 7, 366e373.
groups. Journal of Medical Microbiology, 62, 1807e1814. Le Guern, A. S., Martin, L., Savin, C., & Carniel, E. (2016). Yersiniosis in France:
Diaz, R., Urra, E., Toyos, J., & Moriyon, I. (1985). Characterization of a Yersinia Overview and potential sources of infection. International Journal of Infectious
enterocolitica antigen common to enterocolitis-associated serotypes. Journal of Diseases, 46, 1e7.
Clinical Microbiology, 22, 1035e1039. Lee, W. H. (1977). Two plating media modified with Tween 80 for isolating Yersinia
Doherty, A., Sheridan, J. J., Allen, P., McDowell, D. A., Blair, I. S., & Harrington, D. enterocolitica. Applied and Environmental Microbiology, 33, 215e216.
(1995). Growth of Yersinia enterocolitica O: 3 on modified atmosphere packaged Longenberger, A. H., Gronostaj, M. P., Yee, G. Y., Johnson, L. M., Lando, J. F.,
lamb. Food Microbiology, 12, 251e257. Voorhees, R. E., et al. (2014). Yersinia enterocolitica infections associated with
EC. (2004). Regulation (EC) No 853/2004 of the European Parliament and of the improperly pasteurized milk products: Southwest Pennsylvania, March-
Council of 29 April 2004 laying down specific hygiene rules for on the hygiene eAugust, 2011. Epidemiology and Infection, 142, 1640e1650.
of foodstuffs. Official Journal of the European Union, p.22. L226, 25/06/2004. Lucier, T. S., Fetherston, J. D., Brubaker, R. R., & Perry, R. D. (1996). Iron uptake and
EFSA, & ECDC. (2016). The European Union summary report on trends of zoonoses, iron-repressible polypeptides in Yersinia pestis. Infection and Immunity, 64,
zoonotic agents and food-borne outbreaks in 2015. EFSA Journal, 14, 4634. 3023e3031.
EFSA, & ECDC. (2017). The European Union summary report on trends of zoonoses, McNally, A., Cheasty, T., Fearnley, C., Dalziel, R. W., Paiba, G. A., Manning, G., et al.
zoonotic agents and food-borne outbreaks in 2016. EFSA Journal, 15, 5077. (2004). Comparison of the biotypes of Yersinia enterocolitica isolated from pigs,
El Tahir, Y., & Skurnik, M. (2001). YadA, the multifaceted Yersinia adhesin. Interna- cattle and sheep at slaughter and from humans with yersiniosis in Great Britain
tional Journal of Medical Microbiology, 291, 209e218. during 1999e2000. Letters in Applied Microbiology, 39, 103e108.
EUCAST. (2017). Breakpoint tables for interpretation of MICs and zone diameters. Merhej, V., Ade
kambi, T., Pagnier, I., Raoult, D., & Drancourt, M. (2008). Yersinia
Version 7.0. http://www.eucast.org. massiliensis sp. nov., isolated from fresh water. International Journal of System-
brega, A., & Vila, J. (2012). Yersinia enterocolitica: Pathogenesis, virulence and
Fa atic and Evolutionary Microbiology, 58, 779e784.
antimicrobial resistance. Enfermedades Infecciosas y Microbiología Clínica, 30, Murros-Kontiainen, A., Fredriksson-Ahomaa, M., Korkeala, H., Johansson, P.,
24e32. Rahkila, R., & Bjo €rkroth, J. (2011a). Yersinia nurmii sp. nov. International Journal
Falca~o, J. P., Brocchi, M., Proença-Mo
dena, J. L., Acrani, G. O., Corre ^a, E. F., & of Systematic and Evolutionary Microbiology, 61, 2368e2372.
Falca~o, D. P. (2004). Virulence characteristics and epidemiology of Yersinia Murros-Kontiainen, A., Johansson, P., Niskanen, T., Fredriksson-Ahomaa, M.,
enterocolitica and Yersiniae other than Y. pseudotuberculosis and Y. pestis iso- Korkeala, H., & Bjo €rkroth, J. (2011b). Yersinia pekkanenii sp. nov. International
lated from water and sewage. Journal of Applied Microbiology, 96, 1230e1236. Journal of Systematic and Evolutionary Microbiology, 61, 2363e2367.
Fredriksson-Ahomaa, M. (2007). Yersinia enterocolitica and Yersinia pseudotubercu- Nesbakken, T. (2006). Yersinia infections. In H. P. Riemannn, & D. O. Cliver (Eds.),
losis. In S. Simjee (Ed.), Foodborne diseases (pp. 79e113). Totowa, NJ, USA: Foodborne infections and intoxications (3rd edn, Chapt. 8. pp. 289e312).
Humana Press. Amsterdam, The Netherlands: Elsevier Academic Press.
Fredriksson-Ahomaa, M. (2012). Isolation of enteropathogenic Yersinia from non- Pederson, K. J., & Pierson, D. E. (1995). Ail expression in Yersinia enterocolitica is
human sources. Advances in Experimental Medicine and Biology, 954, 97e99. affected by oxygen tension. Infection and Immunity, 63, 4199e4201.
Fredriksson-Ahomaa, M., Stolle, A., & Stephan, R. (2007). Prevalence of pathogenic Pepe, J. C., & Miller, V. L. (1993). Yersinia enterocolitica invasin: A primary role in the
Yersinia enterocolitica in pigs slaughtered at a Swiss abattoir. International initiation of infection. Proceedings of the National Academy of Sciences USA, 90,
Journal of Food Microbiology, 119, 207e212. 6473e6477.
Fukushima, H., Gomyoda, M., Aleksic, S., & Tsubokura, M. (1993). Differentiation of Perry, M. B., & Bundle, D. R. (1990). Lipopolysaccharide antigens and carbohydrates
Yersinia enterocolitica serotype O:5,27 strains by phenotypic and molecular of Brucella. In G. L. Adams (Ed.), Advances in brucellosis research (pp. 76e88).
techniques. Journal of Clinical Microbiology, 31, 1672e1674. College Station, TX, USA: Texas A & M University Press.
Giffel, M. C., & Wells-Bennik, M. H. J. (2010). Good hygienic practice in milk pro- Petsios, S., Fredriksson-Ahomaa, M., Sakkas, H., & Papadopoulou, C. (2016). Conven-
duction and processing. In M. W. Griffiths (Ed.), Improving the safety and quality tional and molecular methods used in the detection and subtyping of Yersinia
of milk. Vol. 1. Milk production and processing (pp. 179e193). Cambridge, UK: enterocolitica in food. International Journal of Food Microbiology, 237, 55e72.
Woodhead Publishing Ltd. Pham, J. N., Bell, S. M., Martin, L., & Carniel, E. (2000). The b-lactamases and b-
Grant, T., Bennett-Wood, V., & Robins-Browne, R. M. (1999). Characterization of the lactam antibiotics susceptibility of Yersinia enterocolitica. Journal of Antimicro-
interaction between Yersinia enterocolitica biotype 1A and phagocytes and bial Chemiotherapy, 46, 951e957.
epithelial cells in vitro. Infection and Immunity, 67, 4367e4375. Pierson, D. E., & Falkow, S. (1990). Nonpathogenic isolates of Yersinia enterocolitica
Grassl, G. A., Bohn, E., Müller, Y., Bühler, O. T., & Autenrieth, I. B. (2003). Interaction do not contain functional Inv-homologous sequences. Infection and Immunity,
of Yersinia enterocolitica with epithelial cells: Invasin beyond invasion. Inter- 58, 1059e1064.
national Journal of Medical Microbiology, 293, 41e54. Pujol, C., & Bliska, J. B. (2005). Turning Yersinia pathogenesis outside in: Subversion
Greenwood, M. H., Hooper, W. L., & Rodhouse, J. C. (1990). The source of Yersinia spp. of macrophage function by intracellular yersiniae. Clinical Immunology, 114,
in pasteurized milk: An investigation at a dairy. Epidemiology and Infection, 104, 216e226.
351e360. R Development Core Team. (2016). R: A language and environment for statistical
Gupta, V., Gulati, P., Bhagat, N., Dhar, M. S., & Virdi, J. S. (2015). Detection of Yersinia computing. R Foundation: The R project for statistical computing.
enterocolitica in food: An overview. European Journal of Clinical Microbiology and Rakin, A., Saken, E., Harmsen, D., & Heesemann, J. (1994). The pesticin receptor of
Infectious Diseases, 34, 641e650. Yersinia enterocolitica: A novel virulence factor with dual function. Molecular
Hill, A. R., & Kethireddipalli, P. (2013). Dairy products: Cheese and yoghurt. In Microbiology, 13, 253e263.
N. A. M. Eskin, & F. Shahidi (Eds.), Biochemistry of foods (pp. 319e351). London, Ramamurthy, T., Yoshino, K. I., Huang, X., Nair, G. B., Carniel, E., Maruyama, T., et al.
UK: Academic Press. (1997). The novel heat-stable enterotoxin subtype gene (ystB) of Yersinia
Hoogkamp-Korstanje, J. A., & Stolk-Engelaar, V. M. (1995). Yersinia enterocolitica enterocolitica: Nucleotide sequence and distribution of the yst genes. Microbial
infection in children. Pediatric Infectious Disease Journal, 14, 771e775. Pathogenesis, 23, 189e200.
 S. Bonardi et al. / International Dairy Journal 84 (2018) 46e53 53

Reuter, S., Connor, T. R., Barquist, L., Walker, D., Feltwell, T., Harris, S. R., et al. Skurnik, M., Rådstro € m, P., Knutsson, R., Segerman, B., Hallanvuo, S., Thisted
(2014). Parallel independent evolution of pathogenicity within the genus Lambertz, S., et al. (2010). Yersinia. In D. Liu (Ed.), Molecular detection of food-
Yersinia. Proceedings of the National Academy of Sciences USA, 111, borne pathogens (pp. 501e520). Boca Raton, FL, USA: CRC Press.
6768e6773. Sprague, L. D., & Neubauer, H. (2005). Yersinia aleksiciae sp. nov. International Journal

, B., Koffi, E., Cabanel, N., et al. (2017). Yersinia
Saraka, D., Savin, C., Kouassi, S., Cisse of Systematic and Evolutionary Microbiology, 55, 831e835.
enterocolitica, a neglected cause of human enteric infections in Co ^ te d'Ivoire. Sprague, L. D., Scholz, H. C., Amann, S., Busse, H. J., & Neubauer, H. (2008). Yersinia
PLoS Neglected Tropical Diseases, 11. Article e0005216. similis sp. nov. International Journal of Systematic and Evolutionary Microbiology,
Savin, C., Martin, L., Bouchier, C., Filali, S., Chenau, J., Zhou, Z., et al. (2014). The 58, 952e958.
Yersinia pseudotuberculosis complex: Characterization and delineation of a Tennant, S. M., Grant, T. H., & Robins-Browne, R. M. (2003). Pathogenicity of Yersinia
new species, Yersinia wautersii. International Journal of Medical Microbiology, enterocolitica biotype 1A. FEMS Immunology and Medical Microbiology, 38, 127e137.
304, 452e463. Thoerner, P., Kingombe, C. B., Bo €gli-Stuber, K., Bissig-Choisat, B., Wassenaar, T. M.,
Schaake, J., Kronshage, M., Uliczka, F., Rohde, M., Knuuti, T., Strauch, E., et al. (2013). Frey, J., et al. (2003). PCR detection of virulence genes in Yersinia enterocolitica
Human and animal isolates of Yersinia enterocolitica show significant serotype- and Yersinia pseudotuberculosis and investigation of virulence gene distribution.
specific colonization and host-specific immune defense properties. Infection Applied and Environmental Microbiology, 69, 1810e1816.
and Immunity, 81, 4013e4025. Van Damme, I., Berkvens, D., Botteldoorn, N., Dierick, K., Wits, J., Pochet, B., et al.
Schmid, A., Messelha €usser, U., Ho€ rmansdorfer, S., Sauter-Louis, C., & Mansfeld, R. (2013). Evaluation of the ISO 10273: 2003 method for the isolation of human
(2013). Occurrence of zoonotic Clostridia and Yersinia in healthy cattle. Journal of pathogenic Yersinia enterocolitica from pig carcasses and minced meat. Food
Food Protection, 76, 1697e1703. Microbiology, 36, 170e175.
Schulze-Koops, H., Burkhardt, H., Heesemann, J., Kirsch, T., Swoboda, B., Bull, C., Van Damme, I., Habib, I., & De Zutter, L. (2010). Yersinia enterocolitica in slaughter
et al. (1993). Outer membrane protein YadA of enteropathogenic yersiniae pig tonsils: Enumeration and detection by enrichment versus direct plating
mediates specific binding to cellular but not plasma fibronectin. Infection and culture. Food Microbiology, 27, 158e161.
Immunity, 61, 2513e2519. Wang, X., Cui, Z., Jin, D., Tang, L., Xia, S., Wang, H., et al. (2009). Distribution of
Schulze-Koops, H., Burkhardt, H., Heesemann, J., Von Der Mark, K., & Emmrich, F. pathogenic Yersinia enterocolitica in China. European Journal of Clinical Micro-
(1992). Plasmid-encoded outer membrane protein YadA mediates specific biology and Infectious Diseases, 28, 1237e1244.
binding of enteropathogenic yersiniae to various types of collagen. Infection and Wanger, A. (2007). Yersinia. In P. R., E. J. Murray, J. H. Baron, M. L. Jorgensen, Landry,
Immunity, 60, 2153e2159. & M. A. Pfaller (Eds.), Manual of clinical microbiology (pp. 688e697). Washing-
Sihvonen, L. M., Jalkanen, K., Huovinen, E., Toivonen, S., Corander, J., Kuusi, M., et al. ton, DC, USA: American Society for Microbiology.
(2012). Clinical isolates of Yersinia enterocolitica biotype 1A represent two Wauters, G., Kandolo, K., & Janssens, M. (1987). Revised biogrouping scheme
phylogenetic lineages with differing pathogenicity-related properties. BMC of Yersinia enterocolitica. Contributions to Microbiology and Immunology, 9,
Microbiology, 12. Article 208. 14e21.