1. Introduction
Yersinia enterocolitica (
Ye) are zoonotic psychrotropic bacteria, which cause acute gastroenteritis and, occasionally, more serious diseases in humans. Yersiniosis was the fourth zoonosis reported in humans in 2018, with 6699 confirmed cases reported in Europe (EU). The trend of human cases was stable in 2014–2018, confirming the trend in 2019; however, in 2017 the number of confirmed cases was 2.8% lower than in 2016, and represented the lowest recorded rate in the last 5 years [
1,
2].
Ye was the most common species reported to be isolated from human cases, in food and in animals. The more widespread serotypes in human yersiniosis were O:3, followed by O:9 and O:8. Moreover, the biotype prevalent in 2016 was biotype 4, followed by biotype 2 and 3. In 2018, biotypes and serotypes of
Ye were rarely reported [
1].
Ye species forms a heterogeneous group of non-pathogenic and pathogenic strains.
Ye comprises six biotypes: 1A, 1B, 2, 3, 4 and 5, based on metabolic differences which are further classified into numerous serotypes [
3,
4,
5,
6]. Biotype 1A is often recognized to be avirulent; however, some biotype 1A strains can be a cause of gastrointestinal symptoms and sporadic extraintestinal infections [
3,
7,
8,
9]. The virulence of the strains belonging to biotypes 1B and 2–5 depends on the presence of both chromosomal and plasmid-borne genes [
10]. The presence of a high pathogenicity island (HPI) encoding for the yersinia bactin siderophore system determines the high pathogenicity of biotype 1B strains infection in the mouse model, while biotypes 2–5 constitute low–moderate pathogenic lineages [
4,
5,
11,
12,
13]. Only few serotypes are known to be injurious to humans and were associated with different clinical manifestations. Bioserotypes 1B/O:8, 2/O:5,27, 2/O:9, 3/O:3 and 4/O:3 (in order of increasing frequency) are the most frequently isolated pathogenic strains in Europe [
14,
15,
16]. The pathogenicity of
Ye is often associated with chromosomal virulence genes that comprise
Attachment and invasion locus (
ail),
Invasin (
inv),
Mucoid Yersinia factor (
myf),
Host-responsive element (
hreP) and
Yersinia stable toxin (
yst) [
17,
18]. Additionally,
Yersinia-modulating protein(
ymoA) is an important chromosomal gene encoding for the YmoA protein, which negatively regulates the expression of various genes; it inhibits the expression of
inv and
Yersinia stable toxin A (
ystA) [
19].Regarding virulence plasmid genes (
pYV), there are known genes:
Adhesin A (
yadA), whose product is involved in autoagglutination, serum resistance and adhesion [
20];
Transcriptional regulator (
virF), which encodes transcriptional activators of the yop regulon [
21], and is therefore fundamental for the type-III secretion system. Biotype 1A is recognized as non-pathogenic, since they do not have
pYV plasmid and some chromosomal virulence genes, e.g.,
ystA and
myfA [
9]. Although
inv is present, it seems to be non-functional in most 1A strains [
22]. However, the 1A strains carry other virulence genes, such as
ystB and
hreP, and some biotype 1A strains that were involved in human infections [
23]. The epidemiology of the infection and the distribution of serotypes need to be further understood [
2,
24]. It is known that infection occurs after the consumption of poorly cooked pork meat or vegetables, and healthy pigs are the principal reservoir of
Ye [
25,
26]. Indeed, the pathogen is often isolated from the tonsils, intestines or faeces of swine. Furthermore,
Ye can be isolated from soil, water, and the environment, in which it is able to survive for a long period [
27]. Moreover, there is evidence of the link between pigs, pork carcasses, wild boars and associated products [
28,
29,
30,
31]. Indeed,
Ye are sometime isolated in wild boars, and recent papers showed a prevalence between 3.5% and11% in European wild boars [
31,
32,
33]. However, data on their pathogenicity and antimicrobial resistance are still lacking. This is particularly concerning in highly urbanized areas like Liguria region (Italy), where the increased human–wildlife interface facilitates the spread of infectious diseases from wildlife to domestic animals and humans. The aim of our study was to investigate the presence and thedetection of biotypes of
Ye in the wild boars hunted in Liguria region from 2013 to 2018, and to evaluate the presence of chromosomic genes of pathogenicity (GoP) and
Ye antimicrobial resistance.
3. Discussion
Yersinia enterocolitica is a zoonotic pathogen which causes acute gastroenteritis and, occasionally, more serious diseases in humans [
27]. Today, there is no harmonized surveillance of
Ye in the EU: recorded data are not comparable between member states and extreme caution is needed when interpreting results at the EU level; nevertheless, yersiniosis is the fourth most reported zoonosis in the EU [
1]. There was a decreasing trend in reported confirmed human cases of yersiniosis in the EU/EEA from 2008 to 2018, but the trend did not show any significant increase or decrease in the past 7 years (2013–2019). The highest country-specific notification rates were observed in northeastern European member states.
Ye was the most common reported pathogens in these states, and it was identified in 11food-borne outbreaks. The most common bioserotype was 4/O:3, followed by 2/O:9 and 2/O:5,27. Very few European member states reported food and animal data on
Yersinia occurrence or prevalence in 2017; indeed, reporting this kind of data is not mandatory. These scarce data preclude meaningful observations at the EU level. According to the last EFSA report,
Yersinia has been isolated mainly in pork fresh meat (8.3%), in meat products from sheep (16%), beef cattle (6.3%) and in living animals (pigs 4.4 %, other animals 3.5%) [
1].
In wildlife, European authors reported a prevalence between 33.3% and 1.3%, in Spain and Poland, respectively [
34,
35]. Other studies have highlighted the influence of seasonality on the prevalence; these authors reported a prevalence of 17.1% in Germany and 20% in Sweden, with the highest values recorded in cold seasons (winter and spring) [
31,
36].
In our study,
Ye was isolated on 2.9% of animals; thesedata are in accordance with the 3.5% prevalence reported in the EFSA report of the 2014 [
37], and with the study of Bancerz-Kisiel [
35]. Despite the low prevalence, it is worth remembering that the wild boar population increased significantly in the last century, both in European and Italian territories. As a result, the species hasspread to new areas and contact with humans and livestock increased simultaneously with the risk to public safety [
35]. In this condition, it may be useful to characterize the strains of the wild boar populations in order to know their serotype, biotype and, above all, the pathogenic potential.
Ye is classified by the heat-resistant somatic antigen O (seventy serotypes described) and by the biotype. To date six biotypes are known: 1A, 1B, 2, 3, 4, 5. Strains O:3 and O:9 are often isolated from swine that are considered the main reservoir, and strains O:8 are isolated from water, vegetables and dairy food. In our study, 33.3% of the strains was not characterized: 38.1% was O:8 serotype and10.3% was O:5 serotype, which wereboth associated to human gastroenteritis cases [
38]. Our results differ from data reported by Kamińs kaand Sadkowska-Todys [
39], which highlighted the circulation of O:3 (88%), O:8 (6.9%) and O:9 (5.2%) strains. These differences could be due to a non-correlation between serotypes and geographical distribution [
40]. Reports on the
Ye presence in wild boars are rare, and the epidemiological link between wild boars and domestic pigs is still unknown [
31,
41]. In our study, the higher frequency of serotype O:8 suggests that, in our region, the major source of
Ye in wild boars is anthropogenic; moreover, we can speculate that water could be the link between humans and wild boars [
42]. Indeed, most
Ye isolates in water belong to non-biotype 1A or to
Ye-like bacteria [
42].
In regard to the biotypes, the biotype 4 (serotype O:3) and 2 (serotype O:9) are more frequently associated with human yersiniosis: strains 1A were isolated from environment, foods, and human and animal faeces [
25]. 1A strains are not supposed to be pathogens and are not presumed to represent a risk to public safety [
19]. Our results highlight the high prevalence of the 1A biotype; indeed, we detected 117 (92.9%) strains of the biotype 1A, 8 strains of the biotype 1B and 1 strain of the biotype 2. Among the European states, Bancerz-Kisiel2016 [
43], have reported a 1A strains prevalence of 15.4% in Poland.In Italy, there are no data other than ours on wild boars’
Ye 1A prevalence. However,
Ye detection in swine has been reported by Bonardi [
44]. In particular, 11.2% of samples (19 amygdales out of 170) obtained from 19 different farms, located in the province of Mantua, Brescia, Reggio Emilia, Verona, Parma and Cuneo were found to be positive. Most of them belonged to serogroup O:3 biotype 4 (13/19, 68.4%), while 15.8% (3/19) belonged to bio-serotype 1A/O:8, 10.5% (2/19) at bio-serotype 1A/O:5 and only 5.2% (1/19) at bio-serotype 4/O:8. The distribution of genes associated with virulence appeared to vary, with a prevalence, within the bio-serotype 4/O:3, of positive strains for both
ail and
ystA genes, and positive strains for the three sequences
yadA,
ail and
ystA.Some studies suggested that few 1A strains may be the cause of intestinal infection. According to Liang and colleagues [
45], the 1A isolates lack the
pYV plasmid and are therefore considered non-virulent. Although, some studies have demonstrated
Ye 1A ability to invade the epithelial cells and to cause symptomatology indistinguishable from that caused by pathogenic biotypes (1B, 2–5) [
46]. Moreover, our recent study demonstrated the ability of different strains of
Ye 1A to adhere to and penetrate enterocytes, causing an innate immune response characterized by a strong pro-inflammatory response [
47].
In the present study, to evaluate the pathogenic potential of the isolates, we assessed the presence of chromosomic genes of virulence and 69.8% of the strains was positive for
ystB,
45.2% positivefor
ymoA, 43.6% was positive for
ail and 19.8% was positive for
ystA. A small percentage of the isolates was positive for
myfA and
inv (11.9 and 7.9%, respectively). In our study, we outlined a major presence of
ail and
yst genes with respect to the study conducted by Younis and coworkers in Egypt [
48]. Our results showed the presence of
ystB both in 1A and 1B biotypes; therefore, on the basis of Liang et al.’sassumptions [
45], it could be speculated that both 1A and 1B biotype strains are pathogenic. More in vitro and in vivo investigations are needed to assess that event.
Data regarding the strains’ pathogenicity are of high interest if correlated to antimicrobial resistance or tolerance to biocides. In this respect, a recent study demonstrated the ability of
Ye to acquire tolerance to biocides and to increase its antibiotic resistance after exposure to sub-MICs of such disinfectants [
49].
In our study, 61.9% of the isolates showed antimicrobial resistance, with an increase from 2013 to 2017. Similar results were obtained in Egypt [
48]. In 31 Bavarian farms from 2000 to 2004, a study on
Ye strains showed that 77% of 4/O:3 strains were sensitive to 14 antimicrobial drugs [
50]. A sporadic resistance was observed against Amoxicillin and Clavulanic acid (5%), Streptomycin (9%), Sulfamethoxazole (9%) and Tetracycline (1%). Similar levels in swine isolates resistance were found in Switzerland and Brasil [
51,
52].
According to Italian studies, a large spread of multi-drug-resistant isolates has been detected, and strains resistant to three or more antimicrobial drugs were detected in 91% of the isolates 4/O:3. The resistance against Chloramphenicol was detected in all the strains [
53]. In a study conducted in China, high levels of antimicrobial resistance were found on
Ye 4/O:3 strains: Sulfonamide (91%), Streptomycin (64%) and Chloramphenicol (55%) [
45]. Bhaduri [
54] has tested the resistance profile in isolates both
pYV positive and negative and reported a high resistance to Tetracycline in USA (27%); Simonova [
50] has reported resistance to the Nalidixic acid (4%) and to Chloramphenicol (4%).In the Czech Republic, the presence of
pYV gene has not been associated to the profile of resistance, as described by Bhaduri [
54]; moreover, no correlation has been found between the different
Ye 4/O:3 genotypes and the models of antimicrobial resistance [
55]. It is known that
Ye can produce two chromosomal β-lactamases, BlaA and BlaB. The latter can induce a broad spectrum cephlaosporinase that has a different activity in
Ye biotypes; in particular, a recent study suggest that BlaB is more inducible in biotypes 2 and 4, than in biotypes 1 A and 1B [
55,
56].
The possible causes of the development of antimicrobial resistance are: (i) massive treatments carried out for the prophylaxis of bacterial infections; (ii) inaccurate dosage of antimicrobial drugs; (iii) inadequateduration of the treatments; (iv) usage of antimicrobial as growth promoters.
The detection of antimicrobial resistance in strains isolated in wild boar could be associated to two main factors: transfer of the antimicrobial resistance occurring, by means of plasmids, between strains of the same or different species (hosted in both swine and wild boar) [
57] and the rapid increase in the wild boar population, which is causing more frequent contacts with domestic livestock (mainly pigs). Whatever the cause, the presence of antimicrobial-resistant strains isolated in wildlife samples is an important aspect to consider due to its impact on public health [
58].