Listeria monocytogenes, Escherichia coli and Coagulase Positive Staphylococci in Cured Raw Milk Cheese from Alentejo Region, Portugal

Traditional cheeses are part of the Portuguese gastronomic identity, and raw milk of autochthonous species is a common primary ingredient. Here, we investigated the presence of Listeria monocytogenes, Coagulase Positive Staphylococci (CPS) and pathogenic Escherichia coli, as well as of indicator microorganisms (E. coli and other Listeria spp.) in 96 cured raw milk cheeses from the Alentejo region. Whole genome sequencing (WGS) of pathogenic E. coli and Listeria spp. as well as antimicrobial resistance (AMR) screening of E. coli isolates was also performed. L. monocytogenes, CPS > 104 cfu/g and Extraintestinal E. coli were detected in 15.6%, 16.9% and 10.1% of the samples, respectively. Moreover, L. monocytogenes > 102 cfu/g and Staphylococcal enterotoxins were detected in 4.2% and 2.2% of the samples, respectively. AMR was observed in 27.3% of the E. coli isolates, six of which were multidrug resistant. WGS analysis unveiled clusters of high closely related isolates for both L. monocytogenes and L. innocua (often correlating with the cheese producer). This study can indicate poor hygiene practices during milk collection/preservation or during cheese-making procedures and handling, and highlights the need of more effective prevention and control measures and of multi-sectoral WGS data integration, in order to prevent and detect foodborne bacterial outbreaks.


Introduction
As defined in European Regulation No. 853/2004, "raw milk" means "milk produced by the secretion of the mammary gland of farmed animals that has not been heated to more than 40 • C or undergone any treatment that has an equivalent effect" [1].
Due to the absence of a heat treatment, raw milk can harbor a diverse microbial flora, including both spoilage and pathogenic microorganisms, with the potential of causing human illness [2].
Data published by Eurostat concerning milk production in 2021 [3] shows that of the total raw milk (161.0 million tonnes) produced in European Union (EU) farms, the largest portion (155.2 million tonnes) was cows' milk, the rest being ewes' milk (3 million tonnes), goats' milk (2.5 million tonnes) and buffalos' milk (0.3 million tonnes). Concerning the final use of the total produced milk, only a small portion (10.4 million tonnes) was used on farms (consumed by the farmers and their families, sold directly to consumers, used and non-pathogenic) antimicrobial resistance (AMR) was also studied.

Sampling
Ninety-six cured raw milk cheeses from different batches, corresponding to 30 brands produced by 20 identified producers located in the Alentejo region of Portugal were analyzed. The Alentejo region is one of Portugal regions with the most appreciated traditional cheeses [5] and there is a lack of studies carried out on their microbiological quality. The Alentejo Regional Coordination and Development Commission divides Alentejo region into 4 sub-regions: Alto Alentejo, Alentejo Central, Alentejo Litoral and Baixo Alentejo. Figure 1 shows the number of brands tested, by region/sub-region. Cheeses were purchased from different hypermarkets, supermarkets, local markets and grocery stores around the Lisbon region from June 2021 to May 2022. Samples were stored at refrigeration temperature (2 °C to 4 °C) from the time of purchasing until processing, within 24 h after collection, and were analyzed during their assigned shelf-life period.

Microbiological Analysis
All microbiological analyses were performed according to the general requirements and guidance for microbiological examinations described in ISO 7218:2007 [9].
The 96 collected samples were examined for the presence of Listeria spp. in 25 g. L. monocytogenes enumeration was performed in the positive samples (result: L. monocytogenes detected in 25 g). Eighty-nine (89) out of the 96 samples were also tested for E. coli and Coagulase Positive Staphylococci (CPS) detection and enumeration. For those samples with a CPS concentration ≥4.9 × 10 4 cfu/g, Staphylococcal enterotoxins (SE) detection was also performed.

E. coli and Coagulase Positive Staphylococci (CPS) Detection and Enumeration
Each cheese sample (test portion of 25 g) was added to 225 mL of sterile Buffered peptone water (BPW-Oxoid, Basingstoke, Hampshire, UK) and homogenized at 230 rpm for 1 min using a stomacher (Stomacher, 400 Circulator, London, UK). Appropriate decimal dilutions to 10 −3 were prepared in Tryptone salt diluent (Biokar Diagnostics, Pantin, France). Detection and enumeration of E. coli and Coagulase Positive Staphylococci were performed by the AFNOR validated TEMPO ® EC and TEMPO ® STA automated most probable number (MPN) system (bioMérieux, Marcyl l'Etoile, France), respectively, following the manufacturer's instructions.
Simultaneously to TEMPO ® EC E. coli enumeration, the initial suspension 1/10 Cheese/BPW mixture was incubated at 37 • C during 24 h ± 2 h. E. coli plating-out was performed by streaking a loopful of this culture medium on the surface of Chromogenic Coliform Agar (CCA, Biokar Diagnostics) plates and incubated at 37 • C during 24 h± 2 h. E. coli colonies were selected and sub-cultured on Columbia Agar + 5% Sheep Blood (COS; bioMérieux) and incubated at 37 • C during 24 h ± 2 h, where hemolytic activity was determined. The identification of presumptive isolates was confirmed by biochemical identification on VITEK ® 2 compact system (bioMérieux). All positive isolates were stored at -80 • C in broth with 20% glycerol. For L. monocytogenes detection, the ISO 11290-1 horizontal method [10] was followed in parallel with the alternative method-VIDAS ® LMO2 (bioMérieux). A primary enrichment was prepared with 25 g of cheese sample in 225 mL of half-Fraser broth (bioMérieux), homogenized in a stomacher for 1 min and incubated at 30 • C during 25 h ± 1 h. One hundred microliters of the incubated suspension (primary enrichment) were transferred to 10 mL of secondary enrichment medium Fraser broth (bioMérieux) and incubated at 37 • C during 24 h ± 2 h. After incubation, 0.5 mL of the culture was tested in the VIDAS ® LMO2 automated system, according to the manufacturer's instructions.
For L. monocytogenes enumeration, the ISO/11290-2 horizontal method [11] was followed. One milliliter of a 1:10 homogenized initial suspension (10 g of cheese + 90 mL of BPW) was spread in equal parts on the surface of three Microinstant ® Listeria Agar (Ottaviani e Agosti) (Biokar Diagnostics) plates and incubated at 37 • C during 48 h ± 2 h. L. monocytogenes presumptive colonies (blue colored surrounded by an opaque halo) were counted and subsequently isolated on Columbia Agar + 5% Sheep Blood (COS; bioMérieux) at 37 • C during 24 h ± 2 h, where hemolytic activity was determined. Biochemical identification of the isolates was performed on VITEK ® 2 compact system (bioMérieux), following the manufacturer's instructions. All positive isolates were stored at −80 • C in Tryptone Soy Broth (TSB; Biokar Diagnostics) with 20% glycerol.
Other Listeria spp. colonies (blue colonies without an opaque halo), when present, were also transferred to COS agar and identity of the isolates confirmed by biochemical identification on VITEK ® 2 compact system.

Staphylococcal Enterotoxins (SE) Detection
A staphylococcal enterotoxins (SE) detection was performed in all cheese samples that presented Coagulase Positive Staphylococci levels ≥4.9 × 10 4 cfu/g. For the detection of SE, ISO 19020:2017 [12] was followed.
Briefly, 25 g of cheese (10% of the shell and 90% of the inner part) suspended in 40 mL of distilled water at 38 • C ± 2 • C were homogenized in a stomacher, for 1 min and then shaken in an VXR basic Vibrax orbital shaker (Ika ® , Staufen, Germany) at room temperature for 30 to 60 min to allow toxin diffusion. The pH of the slurry was adjusted between 3.5 and 4.0 with HCl and centrifuged at 3130× g for 15 min at 4 • C. The supernatant was collected and the pH adjusted to 7.5 ± 0.1 with NaOH and centrifuged again as described above. The supernatant was concentrated on a dialysis membrane with a molecular cut-off of 6000-8000 Da (Spectrum Laboratories, Rancho Dominguez, CA, USA) against 30% (w/v) of polyethylene glycol 20,000 (Merck, Darmstadt, Germany), overnight, at 4 • C. SE detection was performed using the alternative automated method VIDAS ® Staph enterotoxin II (SET 2) (bioMérieux).

Interpretation of Microbiological Results
The criteria for the interpretation of microbiological results are listed on Table 1 and were based on the following references: Commission Regulation (EC) No 2073/2005 of 15 November 2005 on microbiological criteria for foodstuffs [13], on the Luxembourg Microbiological criteria applicable to foodstuffs [14] and on the Health Protection Agency (HPA) guidelines for assessing the microbiological safety of ready-to-eat foods placed on the market [15]. According to these criteria, one brand was classified as: • Satisfactory, when the results of all the analyzed samples were classified as satisfactory; • Borderline, when none of the samples were unsatisfactory and the results of at least one sample was classified as borderline; • Unsatisfactory/potentially injurious to health, when at least one of the samples was classified as unsatisfactory/ potential injurious to health.

Pathogenic E. coli Identification and Antimicrobial Susceptibility Testing
Potential pathogenic E. coli isolates were identified by screening for the presence of some characteristic virulence genes (eae, aggR, elt, estp, and ipaH) by multiplex PCR (modified from Persson 2007, Boisen 2012and Fujioka 2013) and for the presence of Shiga toxins stx1 and stx2 [19], as previously described [20]. An E. coli isolate was classified as potentially pathogenic (STEC; EPEC, Enteropathogenic E. coli; EAEC, Enteroaggregative E. coli; ETEC, Enterotoxigenic E. coli; EIEC, Enteroinvasive E. coli) when at least one of the pathotype-specific genes was detected.
E. coli pathogenicity was also defined after sequence analysis (some of the presumptive non-pathogenic E. coli isolates were sequenced because were multidrug resistant (MDR) or presented hemolytic activity). In this case, the presence of two or more Extraintestinal pathogenic E. coli (ExPEC) typical virulence genes were used for this pathotype classification [21].
Sequencing reads were deposited on the European Nucleotide Archive (ENA) under the bioprojects PRJEB31216 (Listeria spp.) and PRJEB54735 (E. coli). Accession numbers for each isolate are listed in Supplementary Material, Table S2.
For L. innocua, in the absence of a cgMLST schema, a core-genome alignment (enrolling polished assemblies of 20 out of 21 isolates with sequencing data) was constructed with Parsnp v.1.7.4 implemented on Harvest suite [35], using the default parameters, with exception of parameter -C, which was adjusted to 2000 in order to maximize the resolution. The core-genome SNP-based clustering analysis was performed with ReporTree v.1.0.1 "https://github.com/insapathogenomics/ReporTree (accessed on 19 November 2022)" [32] Microorganisms 2023, 11, 322 7 of 17 using GrapeTree (MSTreeV2 method) [33], with clusters of closely related isolates being determined and characterized at SNP thresholds of 1, 4, 7 and 15 SNPs. This core-genome SNP-based clustering analysis relied on a core-genome alignment (comprising 93% of the L. innocua genome size) involving a total 170 variant sites.

Microbiological Quality
Of the 89 samples tested for all the parameters, 44 (49.4%) were classified as unsatisfactory/potentially injurious to health, 30 (33.7%) as borderline and 15 (16.9%) as satisfactory ( Table 2 and Table S1). The classification of unsatisfactory/potentially injurious to health samples was related with diverse results, the most common the detection of E. coli being at a level >10 4 cfu/g ( Table 2). Most of the unsatisfactory samples (37/44, 84.0%) were also borderline regarding several other results, eight of them also containing L. monocytogenes (Table S1).
Concerning the 30 borderline samples, the reason for the attributed classification was also highly variable (Table 2), the most common, once more, being related to the presence of E. coli > 10 cfu/g and ≤10 4 cfu/g.

Pathogenic E. coli Identification and Antimicrobial Susceptibility Testing
None of the 89 cheeses tested for E. coli were considered pathogenic based on PCR for the tested virulence genes (eae, aggR, elt, estp, invE, stx1 and stx2).
Antimicrobial susceptibility testing (AST) was performed in all hemolytic E. coli (n = 3) and in a subset of presumptive non-pathogenic E. coli isolates (n = 52), and in a total of 55 isolates. Fifteen of the 55 (15/55, 27.3%) isolates were resistant to at least one of the 18 tested antimicrobials, six of which were classified as MDR (Table 3, Table 4 and Table S1).
Genomic analysis of hemolytic (n = 3) and MDR (n = 6) E. coli isolates showed that all of them were classified as ExPEC (Table 3,Table 4 and Table S1).
Ten of the 15 E. coli AMR isolates were detected in cheese samples also containing L. monocytogenes and/or Coagulase Positive Staphylococci. In the case of Coagulase Positive Staphylococci, some of these samples contained concentrations > 10 4 cfu/g and in one sample, it was possible to identify the enterotoxin producer staphylococci (Table 4).
Among the 15 L. monocytogenes isolates, six ST were identified: eight isolates were identified as belonging to ST788 (six isolated from brands 17, 18, 20 and 21, from producer L, one from brand 23, producer N, and one from brand 24, producer O); three as ST378 (two from brand 3, producer B, and one from brand 17, producer L); one as ST1 (from brand 19, producer L); one as ST9 (from brand 19, producer L); one as ST666 (from brand 24, producer O); and one as ST87 (from brand 28, producer R) (Table S2).

Core-Genome Clustering Analysis of Listeria spp. Isolates
In order to assess the genetic relatedness among L. monocytogenes and L. innocua food isolates, and its correlation with cheese producer/brands, a core-genome clustering analysis was performed ( Figure 2). For L. monocytogenes, the cgMLST analysis (comprising 15 isolates) revealed two genetic clusters of high closely related isolates (≤ 7 ADs): cluster A (enrolling isolates 12 and 13, both from producer B) and cluster B (enrolling 6 isolates-1, 4, 6, 7, 14 and 15 from producer L, and the isolate 3 from producer N) (Figure 2A; Table  S2). Of note, we found more than one L. monocytogenes strain in cheeses from the same producer, namely two strains (belonging to sequence types ST788 and ST666) from producer O and four strains (belonging to sequence types ST1, ST9, ST378 and ST788) from producer L. For L. innocua (20 sequenced isolates, all belonging to ST1085), the genetic clustering perfectly correlated with the producer ( Figure 2B; Table S2), with same-producer isolates being interconnected by ≤ 12 SNPs. Notably, similarly to L. monocytogenes, most L. innocua isolates were linked to producer L.
(enrolling isolates 12 and 13, both from producer B) and cluster B (enrolling 6 isolates-1, 4, 6, 7, 14 and 15 from producer L, and the isolate 3 from producer N) (Figure 2A; Table  S2). Of note, we found more than one L. monocytogenes strain in cheeses from the same producer, namely two strains (belonging to sequence types ST788 and ST666) from producer O and four strains (belonging to sequence types ST1, ST9, ST378 and ST788) from producer L. For L. innocua (20 sequenced isolates, all belonging to ST1085), the genetic clustering perfectly correlated with the producer ( Figure 2B; Table S2), with same-producer isolates being interconnected by ≤ 12 SNPs. Notably, similarly to L. monocytogenes, most L. innocua isolates were linked to producer L.  [30]. Each circle (node) contains the strain's designation and represents a unique allelic profile, with numbers on the connecting lines representing allelic distances (AD) between nodes. Straight and dotted lines reflect nodes linked with ADs below and above a threshold of seven ADs, which can provide a proxy to the identification of genetic clusters with potential epidemiological concordance [34]. The traditional seven-loci MLST classification is also indicated. (B) For L. innocua (20 isolates), the MST was constructed based on a core-genome SNP-based alignment (comprising 93% of the L. innocua genome size) involving a total 170 variant sites. Each circle (node) contains the strain's designation and represents a unique SNP profile, with numbers on the connecting lines  [30]. Each circle (node) contains the strain's designation and represents a unique allelic profile, with numbers on the connecting lines representing allelic distances (AD) between nodes. Straight and dotted lines reflect nodes linked with ADs below and above a threshold of seven ADs, which can provide a proxy to the identification of genetic clusters with potential epidemiological concordance [34]. The traditional seven-loci MLST classification is also indicated. (B) For L. innocua (20 isolates), the MST was constructed based on a core-genome SNP-based alignment (comprising 93% of the L. innocua genome size) involving a total 170 variant sites. Each circle (node) contains the strain's designation and represents a unique SNP profile, with numbers on the connecting lines representing SNP distances between nodes. Straight and dotted lines reflect nodes linked with a SNP distance below and above a threshold of 15 SNPs. For both panels, data visualization was adapted from GrapeTree dashboard [33], with the node colors reflecting the producer.

Discussion
According to the EU's data from monitoring foodborne outbreaks, between 2015 and 2020, several outbreaks were associated with cheese consumption: five were caused by L. monocytogenes, with 47 human cases, 43 hospitalizations and 11 deaths; 73 were caused by S. aureus toxins, with 1040 human cases, 108 hospitalizations and no deaths; and 8 were caused by STEC, with 53 human cases, 24 hospitalizations and 2 deaths [36]. These data show that L. monocytogenes, S. aureus and pathogenic E. coli are bacteria capable of surviving, multiplying and/or producing toxins throughout different stages in farm, production (cheese-making process) and consumer levels, constituting a microbiological risk and potentially causing disease after cheese consumption.
In fact, in this study, all these pathogens were often found in the 96 analyzed cheeses and, in the majority of them, in concentrations that classified them as unsatisfactory/ potentially injurious to health or borderline, from a microbiological point of view.
The prevalence of L. monocytogenes found by other authors in cheese samples around Europe are diverse. Most of the studies reported the prevalence of Listeria monocytogenes as lower than the one found in this study: Little et al. [37] analyzed 1819 raw milk cheeses, from United Kingdom (UK), and detected Listeria monocytogenes in 17 (0.9%), one in concentrations above 100 cfu/g; O'Brien et al. [38] studied 351 cheeses from 15 Irish producers, and reported a prevalence of L. monocytogenes of 6%; Rudol et al. [39] reported a prevalence of 6.4% of L. monocytogenes after analyzing 329 European red smear cheese samples; and Almeida et al. [40] examined 70 raw milk Portuguese cheeses, and encountered L. monocytogenes in 8 (11.4%), one in concentrations of >100 cfu/g. Moreover, in some cases, authors could not detect L. monocytogenes in the tested cheese samples [41][42][43][44]. However, there is at least one European study that reported a value of the prevalence of L. monocytogenes similar to the one found in this work (15.6%); Coroneo et al. [45] tested 87 samples of Ricotta Salata, produced in Sardinia, and stated that 17.2% of the samples were positive for the presence of L. monocytogenes. Also, in accordance with the results found in this work, other authors also reported the presence of Listeria species, other than L. monocytogenes, in the evaluated cheese samples [39,46]. These species, although not considered pathogenic, are important indicators of the possible presence of L. monocytogenes, and its presence should be considered.
Recently, cases of listeriosis are increasingly at a multinational level and are frequently related to the consumption of cheeses [47]. In EU, at least five recent listeriosis outbreaks were correlated with the consumption of this foodstuff: a commercial cheese (acid curd) made from pasteurized milk in Germany, in 2006-2007 [48]; a quargel cheese in Austria, Germany and Czech Republic in 2009-2010 [49]; a hard cheese made with pasteurized milk in Belgium in 2011 [50]; a Latin-style fresh cheese made from pasteurized milk in Spain in 2012 [51] and a fresh cheese made from pasteurized cow and goat milk in Portugal in 2009-2012 [52].
Similar to L. monocytogenes, the prevalence values of S. aureus and E. coli found in cheese samples around Europe are also divergent, and are sometimes difficult to compare due to the distinct cut-off values applied among studies. Giammanco et al. [44] analyzed 50 Pecorino Siciliano (PS) "primosale" cheeses in Italy and reported a prevalence of S. aureus coagulase positive in concentrations >10 5 cfu/g of 4% and of E. coli ≥ 10 3 cfu/g of 44%; Little et al. [37] in the UK detected S. aureus > 10 4 cfu/g in 13/1819 (0.7%) of the analyzed raw milk cheeses and E. coli ≥ 10 3 cfu/g in 1.4% of the samples; Almeida et al. [40] in Portugal identified S. aureus > 10 4 cfu/g in 5.7% and E. coli > 10 4 cfu/g in 21.4% of the samples and Rosengren et al. [42] in Sweden described S. aureus >10 5 cfu/g in 10.9% and E. coli ≥10 5 cfu/g in 3.6% of the samples. Moreover, in accordance with the results presented in this study, several studies more focused on the detection of S. aureus in cheeses reported not only high prevalence values of this microorganism but also the presence of staphylococcal enterotoxins [43,53].
The prevalence values encountered in this study, as well as the ones reported in other studies around Europe, clearly demonstrate that E. coli, S. aureus and Listeria spp. are microorganisms that are frequently detected in raw milk cheeses and are sometimes present in concentrations above the normative levels, and consequently may potentially cause disease.
Although several studies in Europe have already described the presence of STEC isolates in cheese samples [54][55][56][57], in this study, we did not find this pathotype in the tested samples. However, nine ExPEC strains were isolated. It is important to notice that many ExPEC strains found in humans with urinary tract infection, sepsis and other extraintestinal infections, particularly the most resistant to antimicrobials, may have a food animal source and may be transmitted via the food supply [58]. In fact, six out of the nine identified ExPEC isolates were MDR. Moreover, the detection of E. coli isolates resistant to antimicrobials in 15 cheeses, and the concomitant presence of at least one of the other tested microorganisms in ten of them, highlights the potential horizontal transfer of antibiotic resistance genes among these cohabiting bacteria and also, eventually, to other gut bacteria, through cheese consumption. Bacterial antibiotic resistance, in particular MDR, has become a global challenge, threatening human and animal health [59]. It is estimated that by 2050, the number of deaths accounted for by MDR will be higher than the ones due to cancer [60].
Moreover, the six MDR ExPEC isolates belonging to three STs (ST10, ST58, ST69) are already associated to human disease. The E. coli ST10 clonal complex is among the emerging ExPEC lineages. Although commonly encountered as an antimicrobial-susceptible low-virulence human intestinal colonizer, it has also been associated with human infections [58]. E. coli ST58 has emerged as a prominent sequence type and a globally disseminated uropathogen that often progresses to sepsis [61]. E. coli ST69 accounted for 4% of the E. coli isolates causing extraintestinal infections in Spain, two of them being also characterized as belonging to O15:H18 serotype, the one detected in our study [62].
Regarding L. monocytogenes typing, ST1, ST9, and ST87 clonal complexes, found in this study in five L. monocytogenes isolates, were already reported in human clinical isolates in at least one of two large WGS studies regarding the characterization of L. monocytogenes isolates in foodstuffs and human samples [30,63]. ST1 and ST87 were also the two most frequent sequence types reported in a study performed in Gipuzkoa in Northern Spain, aiming to describe the clinical features and the molecular epidemiology of human listeriosis over the 2010-2020 period [64].
WGS techniques, when combined with epidemiological information, have the potential to attribute relatedness among studied strains and thus to establish links between human disease cases and causative suspect food vehicles. Regarding the cgMLST analysis of the 15 L. monocytogenes isolates, it is noteworthy that the detection of genetic clusters of high closely related isolates (one of them involving two producers), as well as the identification of highly genetically distant strains, were linked to the same producer(s) (Figure 2A). These results suggest that L. monocytogenes cheese contamination may be related with bad manufacturing and hygienic practices during cheese production or transportation, since all these cheeses were purchased in different locations and belong to different batches. The identification of one isolate from producer N in Cluster B may be justified by the fact that producers L and N were located on the same street and may share suppliers, distribution chain, etc. When integrating the cgMLST results of the L. monocytogenes isolates found in this study, in the global WGS L. monocytogenes collection of the National Institute of Health database, it was possible to verify that three of the L. monocytogenes cheese isolates potentially matched with clinical isolates from 2009 to 2022 (data not shown). These results suggest a potential relatedness among these L. monocytogenes strains, the cheeses from which they were isolated and the reported human listeriosis cases. These results were communicated to the relevant Portuguese authorities and are subsequently under investigation.
The core-genome SNP-analysis of L. innocua isolates reinforces the idea that Listeria spp. cheese contamination is related to bad manufacturing and hygienic practices during cheese production or transportation. Three different clusters were detected, all of them producerspecific ( Figure 2B).
The results revealed that in Alentejo's cheese factories, the investment in training in food safety procedures should be reinforced and the analysis for microbial control are not sufficient or not carried out with the desirable periodicity.

Conclusions
In conclusion, despite the existence of European regulation applicable to raw milk cheeses during the production process and when placed in the market, the contamination detected in a significant number of the cheese samples analyzed within our study alerts for the need of improving the compliance with the good manufacturing and hygienic practices along the different levels of the food chain (farm, artisanal production and consumer).
Considering the possible exposure of the consumer to the above-mentioned pathogenic microorganisms in dairy products made from raw milk, appropriate risk communication on the consumption of these products, particularly to vulnerable populations, is recommended.
It is also crucial to develop enhanced strategies, controlling the initial microbial load and the presence of pathogenic microorganisms in raw milk and the dairy farm environment, therefore monitoring potential hazards along the manufacturing of artisanal cheeses in order to contribute to the prevention of foodborne diseases involving these types of traditional Portuguese products. In addition, this study shows the need for a systematic integration of genomic data at a multi-sectorial level towards an enhanced routine surveillance and outbreak investigation of foodborne diseases.