1. Introduction
Staphylococci are a major cause of clinical or subclinical mastitis in dairy sheep, and among them
Staphylococcus aureus is well known to be the primary causal agent [
1,
2].
In small ruminants, especially sheep,
S. aureus often causes severe mastitis, and it is the principal cause of gangrenous mastitis [
3]. For this reason, small ruminants are often vaccinated against
S. aureus with commercial vaccines, but in some countries, autologous vaccines are also used [
4]. However,
S. aureus can also cause subclinical mastitis in sheep, resulting in long-term infection that can evolve into chronic forms, with the consequent loss of milk production [
5].
The severity of staphylococcal infections depends on the relationship between the host’s defense abilities and the strain’s virulence [
6]. The production of virulence factors, such as toxins, antigens, and resistance proteins, allows
S. aureus strains to cause mastitis [
7].
Many virulence factors that allow the bacterium to adapt to the mammary gland environment [
8] have been described in
S. aureus (
Table 1), and several of them have been identified in ruminant mastitis [
8,
9,
10]. These virulence factors include more than 40 known exotoxins that can be classified into three groups based on their known functions: cytotoxins, superantigens, and cytotoxic enzymes [
11]. After the adhesion phase,
S. aureus produces hemolysins and exoenzyme exotoxins that allow the invasion and destruction of the mammary tissue, with the degradation of the epithelium of the cistern, duct, and alveoli [
10].
Most of the staphylococcal enterotoxins (SEA, SEB, SECn, SED, SEE, and SEH) belong to staphylococcal pyrogenic exotoxin (PTSAgs) [
9]. They have the ability to inhibit host immune responses to
S. aureus [
9], but SAgs were originally termed staphylococcal enterotoxins (SEs) for their ability to cause typical symptoms of
S. aureus food poisoning, such as vomiting and diarrhea [
15]. The α-, β-, γ-, and δ-hemolysins are identified as important virulence factors, produced respectively by
hla,
hlb,
hlg and
hld genes, which allow
S. aureus to escape the host immune response and the invasion of the mammary tissue [
28,
29]. They create pores in host cell membranes or dissolve cell wall components, lysing erythrocytes [
30]. Among them, α-hemolysin is considered one of the main pathogenicity factors of
S. aureus with hemolytic, dermonecrotic, and neurotoxic effects [
28]. All these genes can be used to characterize
S. aureus strains. Numerous studies investigated virulence factors in bovine mastitis [
31,
32,
33,
34,
35,
36,
37,
38,
39,
40,
41,
42,
43,
44,
45,
46] and some of them also explored their geographical variability [
8,
47,
48]. In contrast, the presence of
S. aureus virulence genes has been less investigated in small ruminants [
5,
7,
49,
50,
51,
52,
53], and the studies on their geographical distribution are rare [
49].
From an epidemiological point of view, investigating the virulence factors’ profiles of
S. aureus isolates involved in the etiology of mastitis is highly relevant, provided that the distribution of the bacterial strains in a given territory is constantly monitored [
48]. In parallel, conducting such investigations on isolated
S. aureus strains from sheep herds with mastitis could provide relevant information for the development of vaccines [
10].
The principal aim of the present study was to characterize S. aureus strains, collected during sheep mastitis episodes in a defined territory and used to produce autologous vaccines, by investigating the toxins’ patterns and their geographical distribution to deepen the knowledge on the circulating genetic lineages among the sheep population with mastitis. A second aim was the characterization of strains isolated from a sheep herd with confirmed subclinical mastitis due to S. aureus to investigate the toxins’ patterns of this bacterium.
3. Results
All the analyzed strains tested positive for the 23S gene, and
coa and
nuc genes, confirming the
S. aureus identification. Regarding the enterotoxin genes, the presence of
sea,
sed, and
sej genes was not detected in any strain (
Table 3). The
sec and
sel genes were detected in 74.55% of analyzed strains (82 strains), while
seg,
seh, and
sei genes were observed in only one strain (0.91% of prevalence). The
hld hemolysin gene was observed in 89.09% of the strains (98 strains), followed by
hlgAC with a prevalence of 87.27% (96 strains), and
hla 81.82% (90 strains). Only the 15.45% of the strains tested positive for the
hlb gene.
Considering all the 110 strains analyzed, 20 virulent gene profiles were found (SA1-SA20) (
Table 4). The most common profiles were SA16 (
sec+,
sel+,
hla+,
hld+,
hlgAC+) with a prevalence of 43.64 % and SA4 (
hla+,
hld+,
hlgAC+) with a prevalence of 17.27%.
The phylogenetic analysis evidenced that SA1, SA2, SA5, SA10, SA11, SA12, SA15, SA16, SA17, SA18, and SA19 clustered together (
Figure 1). The ATCC strains (ATCC 23235, 700699, 19095 called respectively SA ATCC21, SA ATCC22, SA ATCC23) clustered with SA3, SA4, SA6, and SA20, while the last strains SA7, SA8, SA9, SA13 and SA14 clustered together.
When only the enterotoxin genes were considered, four profiles were observed (SAET1-SAET4) (
Table 5), but SAET 2 (
sec+,
sej+) and SAET1 (no virulence genes) were the most frequent profiles with a prevalence of 74.55% and 23.64%, respectively, while just one sample resulted for the profile SAET3 (
seh+) and SAET4 (
seg+,
sei+).
The four enterotoxin gene profiles clustered with SA ATCC21, while the SA ATCC22, SA ATCC23 clustered together (
Figure 2).
The analysis of the hemolysin genes revealed 12 profiles (
Table 6). Most of the profiles clustered in the same group while SAEM10 clustered with ATCC SA22 and ATCC SA23 and SAEM12 clustered with ATCC SA21 (
Figure 3). The most representative profile (60.91% of the strains) was SAEM10 (
hla+,
hld+,
hlgAC+).
The 14 strains of
S. aureus, isolated from half udders of ewes with subclinical mastitis at the same farm during a
S. aureus eradication program, corresponded to six different virulence gene profiles (SA 7, SA9, SA 10, SA11, SA16, SA17). In this case, the most identified profile was SA16 (46.82%), followed by SA10 (21.43%) and SA11 (14.29%) (
Table 6). The enterotoxin gene profiles were the same for all isolated strains (SAET 2:
sec+,
sel+), but not for the hemolysin genes that showed six profiles: SAEM 1 (no hemolysin genes), SAEM3 (
hlb+), SAEM4 (
hlgAC+), SAEM5 (
hlgAC+,
hld+), SAEM 10 (
hla+,
hlgAC+,
hld+), and SAEM 11 (
hla+,
hlb+,
hlgAC+). The frequency of the profiles was the same as the corresponding profiles when all genes were considered (
Table 7). Not all profiles clustered together (
Figure 1 and
Figure 3).
The isolated strains were collected from ewes reared in 86 dairy sheep farms located in the provinces of Siena (36/86; 41.86%) and Grosseto (50/86; 58.14%) in central Italy (
Figure 4). These provinces are located in the southern part of the region of Tuscany, which is considered the fourth Italian region in terms of dairy sheep herds (219.424; 6.95%) after Sardinia, Sicily, and Lazio (data retrieved from the national farms’ registry). In Tuscany, dairy sheep farming is mainly concentrated in these two provinces (171.105; 77.97%) because of the land’s suitability for extensive sheep farming.
Spatial distribution of the gene profiles and hemolysin gene profiles did not show the presence of specific patterns but only reflected the frequency of the various profiles within each province (
Figure 5,
Figure 6 and
Figure 7).
4. Discussion
Very few studies have investigated the virulence gene profiles of S. aureus in sheep mastitis and its distribution in a geographic area. Considering the possible severity of this udder infection in sheep and the use of commercial and autologous vaccines, more study will be recommended. In the present study, enterotoxin and hemolysin genes were investigated in 96 strains, isolated in clinical mastitis, and distributed in a specific area, and 14 strains were isolated during an eradication program in sheep subclinical mastitis.
There is no consensus on analysis of virulence genes of
S. aureus to investigate the relationship with clinical mastitis in cattle. Several studies suggested that enterotoxin genes act as virulence genes and they were used to classify some strains isolated in bovine subclinical mastitis and to correlate them with clinical observation [
32,
35,
36]. The presence of
S. aureus, which produces these toxins in the milk, can also have severe consequences for human health because these toxins remain stable in milk, causing food poisoning in humans [
5].
Using RS–PCR genotyping of the 16S–23S rRNA intergenic spacer region, Fournier and collaborators associated some genotypes with the presence of some enterotoxin genes [
35]. Particularly, genotype B (GTB) was characterized by the presence of the
sea,
sed, and
sej genes, and genotype C (GTC) by
sec and
seg. The remainder were grouped as other genotypes (GTOG) because they were rarely found [
35]. In our study, no profiles corresponded to GTB or GTC. The 74.55% of our strains had
sec and
sel gene profiles, but no
sed gene was detected, and one strain showed the
seg gene but in association with
sei. However, in the literature, some studies on bovine mastitis found
seg and
sej genes as the most frequent genes in
S. aureus [
59] while others found
sed,
seg, and
sei [
60], or
sea, as in a study on 238
S. aureus isolates from cow milk in two China regions [
59]. The comparison with enterotoxin gen’ profiles in sheep is very difficult because the same genes were not amplified in the few studies published in literature. In a study on
S. aureus isolated from bulk-tank milk samples of goat and sheep, the only gene in the strains isolated in 30 ewes’ milk was
sec [
53]. Strains positive for
seg and
sei genes’ presence were found in 18 goat milk samples, while no strain positive for
sec and
sej was found [
53]. In another study on 20
S. aureus isolated from sheep milk samples with subclinical mastitis,
sea and
seb were observed in combination in three strains,
sea and
sec in one strain, and
sec alone in another strain, but only
sea,
seb,
sec, and
sed enterotoxin genes were amplified [
5]. A study conducted in Italy evidenced the combination of
seb and
sec as the most isolated genetic profile in
S. aureus from sheep milk in Italy, but
sec was the most frequent gene [
61], as it was in a study of sheep milk in Algeria [
62]. Another study conducted in Greece confirmed the
sec profile as the most frequent in
S. aureus strains from sheep milk, and a
seg- and
sej- positive profile was found in two strains [
7]. In a study in Turkey,
sell had the highest frequency in sheep milk
S. aureus strains, followed by
sec, and the
sec and
sell profile was the unique combination found [
50]; additionally,
sec and
sel were the most detected genes in Switzerland [
63].
Few studies investigated hemolysin genes in
S. aureus strains isolated from mastitis milk of dairy species and there is no consensus on the genes found in the strains, as for enterotoxin genes. Aslantaş et al. (2022) [
50] observed a frequency of
hla,
hlb,
hld, and
hlg2 of 100, 95.2, 98.4, and 61.3%, respectively, in strains obtained from mastitic sheep milk. In another study on
S. aureus strains from bovine and goat milk, the frequencies of
hla and
hlb were lower in goat strains (
hla 51.85%,
hlb 48.15%) respect to bovine strains (
hla 95.93%,
hlb 93.50%) [
49]. In contrast, Moraveij et al. (2014) reported the presence of
hlb only in one of 20 isolates of
S. aureus from bovine mastitis milk, while the majority of the strains showed
hla and
hld, and no strains had the
hlg gene [
64]. In a study on 229 bovine
S. aureus isolates in Belgium
hla,
hlb,
hld, and
hlg showed respectively a frequency of 98.7, 99.1, 100.0, and 78.6% [
42]. In our study, the frequency of
hla,
hld, and
hlgAC was 81.82, 89.09, and 87.27%, respectively, while the
hlb gene was found with a lower frequency (15.45%). In contrast, a study conducted in Italy on strains from mastitis sheep milk found a higher frequency of
hlb-positive strains (34.9%) and a lower frequency of
hlg (2.7%) [
61]. However, we observed a greater variability of the frequency of hemolysin genes in
S. aureus strains than enterotoxin genes, with 12 and 4 different profiles, respectively. Particularly, in the farm with subclinical mastitis, all the strains showed the same profiles for enterotoxin genes, while six hemolysin gene profiles were observed. The diversity of toxin profiles of strains that cause mastitis is the major obstacle in the development of an effective vaccine [
10], and the presence of several strains in the same farm could be a further obstacle.
In human isolates, it has been observed that the prevalence of hemolysin genes in multiple antibiotic-resistant strains was higher than in susceptible strains [
65]. For this reason, the identification of isolates with hemolytic genes in mastitis milk samples could have a great importance [
65]. However, the level of expression of genes during
S. aureus infection is reported to be profoundly different from that of in vitro study [
10], and the presence of virulence genes does not always correspond with the toxins’ expression [
29]. We found high prevalence of
hla,
hld, and
hlgAC, but further studies should be conducted to investigate the expression of these hemolysin genes during sheep
S. aureus mastitis to understand their role in this disease.
It is also important to implement the spatial analysis of gene profiles to identify specific trends and geographical patterns over time. In this study, no specific trend or pattern was detected but profiles were apparently randomly distributed. This might be due to the small number of the analyzed strains in contrast with the prevalence of staphylococcal sheep mastitis or to the considered genes in the present study. However, in a previous study on spatial distribution of the virulence profiles of
S. aureus strains isolated from cow and goat milk samples in three different geographical regions of the state Pernambuco, Brazil, no pattern in the frequency was observed by geographical area [
49].
Therefore, the collection and analysis of further strains is encouraged to reach an adequate level of representativeness of the dairy sheep population in Tuscany, while the increase of the number of amplified genes could identify different profiles with a spatial pattern.
The antibiotic treatment of small ruminant mastitis is widely spread [
66]. This has led to the development of bacterial resistance, and the presence of methicillin-resistant
S. aureus (MRSA) strains in animals with the possibility of transferring resistance genes to human strains is an emerging problem [
10]. In addition, the use of antibiotics for treating mastitis can lead to residues in milk with risks for human health [
66]. The prevention of
S. aureus mastitis through vaccination could be a valid alternative to antibiotic treatment [
10]. However, the development of a vaccine against
S. aureus is a challenge because of its complex pathogenesis, which involves numerous virulence factors [
67]. The best vaccine should have an array of virulence factors [
68]. The commercially available vaccines against
S. aureus mastitis are unable to provide a complete protection [
66], for this reason, autologous vaccines, which contain whole bacterial cells, are commonly used in dairy ruminant species [
69], and some studies showed the benefit of the use of these vaccines in cow herds [
69]. The autologous vaccine is often used in combination with commercial vaccines, especially in those cases where the latter did not provide sufficient protection against
S. aureus and a consequent outbreak in the herd has occurred. The study of virulence factors of the strains used to produce these autologous vaccines, as in our study, could help to identify the virulence factors involved in this economically important disease and to develop an effective vaccine for this species. In the present study, the presence of staphylococcal enterotoxins and hemolysins was investigated, but the number of possible virulence factors involved in mastitis is higher and further study should be conducted to extend the number of investigated genes in the studied strains. However, most of the autologous vaccines are produced using a single clonal type of bacteria isolated from farms [
70]. In contrast, we evidenced that more than one strain of
S. aureus could be present on the same farm. The investigation of this aspect could help to make autologous vaccines more effective.