A One-Year Systematic Study to Assess the Microbiological Profile in Oysters from a Commercial Harvesting Area in Portugal

As filter-feeding animals farmed in water bodies exposed to anthropogenic influences, oysters can be both useful bioremediators and high-risk foodstuffs, considering that they are typically consumed raw. Understanding the dynamic of bacterial and viral load in Pacific oyster (Crassostrea gigas) tissues, hemolymph, outer shell surface biofilm, and farming water is therefore of great importance for microbiological risk assessment. A one-year survey of oysters collected from a class B production area (Canal de Mira, on the Portuguese western coast) revealed that these bivalve mollusks have a good depurating capacity with regard to bacteria, as Salmonella spp. and viable enterococci were not detected in any oyster flesh (edible portion) samples, despite the fact that these bacteria have regularly been found in the farming waters. Furthermore, the level of Escherichia coli contamination was clearly below the legal limit in oysters reared in a class B area (>230–≤4600 MPN E. coli/100 g). On the contrary, norovirus was repeatedly detected in the digestive glands of oysters sampled in autumn, winter, and spring. However, their presence in farming waters was only detected during winter.


Introduction
As a seafood product with high nutritional value, the Pacific oyster (Crassostrea gigas) is farmed across the globe, being highly appreciated in the southern European markets [1,2]. This species is also the most produced oyster in Portugal, particularly in Canal de Mira [3].
Oysters are a very particular foodstuff and one of the few animal foods that are consumed whole and raw. Furthermore, adult oysters are capable of filtering approximately 200 L of water per day, retaining many bacteria and other suspended particles [4,5]. Thus, when oysters farmed in water bodies are exposed to anthropogenic influence, their bodies concentrate chemical pollutants and fecal microorganisms, some of which can constitute a risk to human health [5][6][7]. Among pathogenic microorganisms, Vibrio spp., norovirus (NoV), Salmonella spp., and Listeria monocytogenes (L. monocytogenes) are the ones most frequently associated with foodborne zoonosis outbreaks [8,9]. Other important zoonotic

Sampling and Processing
Throughout a complete seasonal cycle, summer (July 2016), autumn (November 2016), winter (January 2017), and spring (May 2017), samples of 35 cultivated Pacific oysters (C. gigas) (with size 9-11 cm and weight 70-90 g) and the respective farming water (1 L collected three times within 60 min intervals in three different sample points) were collected from Canal de Mira (40 • 38 N, 8 • 45 W) (Figure 1), in the western Portuguese coast. At the time of collection, this production area was rated as class B, meaning that live oysters from this production site could only be placed on the market for human consumption after treatment in a purification center or after relaying [16].
Samples were transported within 3 h in temperature-controlled food boxes and immediately processed upon arrival at the laboratory. The oysters were thoroughly washed with sterile seawater to remove sand, mud, and slime before the measurement of their weight, length, height, and width. They were then divided into six pools, as illustrated in Figure 2. The edible content (flesh, hemolymph, and intra-valvular liquid) of five pools, composed of five oysters each, was transferred to sterile stomacher bags and homogeneously suspended in 1/10 buffered peptone water (BPW, Biokar, Allonne, France). The sixth pool (10 oysters) was used to perform microbiological analysis on the superficial biofilm (outer shell surface), intra-valvular liquid, and hemolymph; virological analysis on the digestive gland; and metabarcoding analysis on the hemolymph. The superficial biofilm was collected by washing the shells with 100 mL of BPW using a pair of sterile toothbrushes. Intra-valvular fluid was collected into a sterile falcon after filtration through a sterile gaze. The hemolymph was collected with a sterile syringe, followed by the dissection of the digestive glands.
washed with sterile seawater to remove sand, mud, and slime before the measurement of their weight, length, height, and width. They were then divided into six pools, as illustrated in Figure 2. The edible content (flesh, hemolymph, and intra-valvular liquid) of five pools, composed of five oysters each, was transferred to sterile stomacher bags and homogeneously suspended in 1/10 buffered peptone water (BPW, Biokar, Allonne, France). The sixth pool (10 oysters) was used to perform microbiological analysis on the superficial biofilm (outer shell surface), intra-valvular liquid, and hemolymph; virological analysis on the digestive gland; and metabarcoding analysis on the hemolymph. The superficial biofilm was collected by washing the shells with 100 mL of BPW using a pair of sterile toothbrushes. Intra-valvular fluid was collected into a sterile falcon after filtration through a sterile gaze. The hemolymph was collected with a sterile syringe, followed by the dissection of the digestive glands.

Bacterial Analysis
The microbiological analysis of flesh and intra-valvular liquid samples was performed in compliance with the European Union microbiological criteria for live washed with sterile seawater to remove sand, mud, and slime before the measurement of their weight, length, height, and width. They were then divided into six pools, as illustrated in Figure 2. The edible content (flesh, hemolymph, and intra-valvular liquid) of five pools, composed of five oysters each, was transferred to sterile stomacher bags and homogeneously suspended in 1/10 buffered peptone water (BPW, Biokar, Allonne, France). The sixth pool (10 oysters) was used to perform microbiological analysis on the superficial biofilm (outer shell surface), intra-valvular liquid, and hemolymph; virological analysis on the digestive gland; and metabarcoding analysis on the hemolymph. The superficial biofilm was collected by washing the shells with 100 mL of BPW using a pair of sterile toothbrushes. Intra-valvular fluid was collected into a sterile falcon after filtration through a sterile gaze. The hemolymph was collected with a sterile syringe, followed by the dissection of the digestive glands.

Bacterial Analysis
The microbiological analysis of flesh and intra-valvular liquid samples was performed in compliance with the European Union microbiological criteria for live

Bacterial Analysis
The microbiological analysis of flesh and intra-valvular liquid samples was performed in compliance with the European Union microbiological criteria for live bivalve mollusks [17], taking also into account the potential microbiological hazards of raw oyster consumption. The analysis included the enumeration of total aerobic microorganisms at 7 • C and 30 • C, marine heterotrophic bacteria at 21 • C, E. coli, Pseudomonas spp., Clostridium perfringens (C. perfringens), coagulase-positive Staphylococcus, Enterococcus spp., molds and yeasts, and the detection of Salmonella spp. and L. monocytogenes. For bacterial enumeration, a pooled sample comprising 25 oysters was used. Regarding the detection of Salmonella spp. and L. monocytogenes, five pools comprising five oysters each were prepared. Total aerobic microorganisms at 30 • C, marine heterotrophic bacteria at 21 • C, E. coli, and Enterococcus spp. were also assessed on the superficial biofilm, intra-valvular liquid, and hemolymph samples (pool of 10 oysters). Finally, the total counts of aerobic microorganisms at 22 • C and 37 • C, marine heterotrophic bacteria at 21 • C, E. coli, Enterococcus spp., and Salmonella spp. were also evaluated in the farming water samples.
International Organization for Standardization (ISO) methods were used for the enumeration/detection of microorganisms:  (Table S1). The detection of both Pseudomonas spp. and marine heterotrophic bacteria was performed using an internal laboratory method (Table S1). Briefly, to detect marine heterotrophic bacteria, the pour-plate method was performed. In total, 1 mL of each sample was plated in marine agar medium (Condalab, Madrid, Spain) and incubated at 21 • C for 48 h. The detection of Pseudomonas spp. was performed using serial dilutions and spreading 100 µL of each sample on cephaloridin fucidin cetrimide (CFC) agar (Oxoid, Basingstoke, UK), and the plates were incubated at 30 • C for 48 h.
In addition, bacteria total counts on samples of superficial biofilm, intra-valvular liquid, and hemolymph collected during summer were also estimated by fluorescence in situ hybridization (FISH), as previously described by [24][25][26][27]. Eco440, PseaerA, and GV probes (MWG-Biotech, Ebersberg, Germany) were used to detect E. coli, P. aeruginosa, and Vibrio spp., respectively. The slides were mounted using Vectashield ® Mounting Medium (Vector Laboratories, Newark, CA, USA) and immediately observed in a Nikon Eclipse E400 microscope (Nikon Instruments, Amsterdam, The Netherlands) at 1000× magnification with an oil immersion objective (HCX PLAN APD). All samples were analyzed in triplicate, and the data are presented as cell/milliliter.

Antimicrobial Susceptibility Testing
The antimicrobial susceptibility of all E. coli and Enterococcus spp. isolated from the farming waters, and from the flesh, superficial biofilm, intra-valvular liquid, and hemolymph of oysters was tested and interpreted according to the Clinical and Laboratory Standards Institute guidelines (CLSI, 2018), using the Kirby-Bauer method. A panel of 18 and 15 antimicrobial agents was used for the antimicrobial susceptibility testing of E. coli and Enterococcus spp. strains, respectively ( Table 1). All antimicrobial disks were from Oxoid (Oxoid, Basingstoke, UK). Isolates resistant to at least one antibiotic agent of three or more antibacterial classes were considered multidrug-resistant (MDR) bacteria [28].

Detection of Food-and Waterborne Viruses
The detection of norovirus (NoV), hepatitis E virus (HEV), and hepatitis A virus (HAV) was performed on both the farming waters and the oysters' digestive gland samples from all seasons ( Figure 2) following ISO/TS 15216-1:2017 'Microbiology of food and animal feed-Horizontal method for determination of hepatitis A virus and norovirus in food using real-time RT-PCR-Part 1: Method for quantification' and as previously described. [29,30]. Briefly, viral extraction was carried out from the homogenates of each sample and mixed with 2 mL of proteinase K (0.1 mg/mL). This mixture was spiked with 10 µL of a virus used to control extraction efficiency, the murine norovirus (MNV-1; 2.7 × 10 9 RNA copies/µL), followed by agitation for 1 h at 37 • C at 320 osc/min (ELMI DOS-10 M Digital Orbital Shaker, ELMI, Riga, Letonia). Then, it was incubated for 15 min at 60 • C and centrifuged for 5 min at 3000× g at room temperature. In total, 500 µL of supernatant was recovered and used for RNA extraction using an NZY Total RNA Isolation Kit (NZYTech, Lisbon, Portugal), according to the manufacturer's instructions. RNA was eluted in 50 µL of RNAfree sterile water and stored at −80 • C until further analysis. NoV GI and GII and HAV were quantified using the primers/probes described in ISO 15216-1:2017. The detection and quantification of HEV were performed by an RTqPCR assay targeting the ORF3 region with the primers/probes previously described [30,31]. The RTqPCR assays were performed using the iTaq Universal PROBES One-Step Kit (Bio-Rad Laboratories, Hercules, CA, USA) in a final volume of 20 µL reaction mixture according to the manufacturer's and run in a CFX Connect Real-Time System (Bio-Rad Laboratories).
The presence of oyster herpesvirus type 1 (OsHV-1) was also evaluated on oyster edible portions from all seasons ( Figure 2). The DNA extraction of the oyster edible portions was performed using a QIAamp cador Pathogen Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions. Briefly, 50 mg of tissue and fluids were subjected to 'Pretreatment T2-Enzymatic Digestion of Tissue', followed by 'Pretreatment B1-for Difficult-to-lyse Bacteria in whole blood or Pre-treated Tissue' and finally 'Purification of Pathogenic Nucleic acids from Fluid Samples'. Eluted DNA was stored at −80 • C until further analysis. OsHV-1 quantification was performed following an improved protocol published by Martenot et al., using a Taqman probe and primers that target the B region of the OsHV-1 genome. qPCR was performed using SsoAdvanced Universal PROBES Supermix (Bio-Rad Laboratories) [32].

Metabarcoding Analysis for Microbiome Composition
DNA extraction for metabarcoding analysis was performed using a QIAamp cador Pathogen Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions. For edible samples, 'Pretreatment T2-Enzymatic Digestion of Tissue' was used, followed by 'Pretreatment B1-for Difficult-to-lyse Bacteria in whole blood or Pre-treated Tissue', and for hemolymph samples, 'Pretreatment B2-for Difficult-to-lyse Bacteria in Cell-free Fluids' was used.
The metabarcoding analysis was carried out in edible portion samples collected in the four seasons and hemolymph samples collected in autumn and spring, using nextgeneration sequencing (GATC Microbiome Profiling (Combined Analysis)) (GATC Biotech, Constance, Germany). This amplicon-based method targeted the V1-V8 variable region of the 16S rRNA gene, using the primers 27F (AGAGTTTGATCCTGGCTCAG) and BS-R1407 (GACGGGCGGTGWGTRC), resulting in a fragment of 1381 bp. The data were checked for chimeras using UCHIME, and the corresponding sequences were removed from further analysis. Non-chimeric, unique sequences were then subjected to BLASTn analysis using non-redundant 16S rRNA reference sequences with an E-value cutoff 1 × 10 6 . Reference 16S rRNA sequences were obtained from the Ribosomal Database Project. Only good quality and unique 16S rRNA sequences that have a taxonomic are considered and used as a reference database to assign operational taxonomic unit (OTU) status to the sequences. Taxonomic classification was based on NCBI Taxonomy [5]-http://www.ncbi.nlm.nih. gov/taxonomy (accessed on 15 December 2017). Except for the E-value cutoff (1 × 10 6 ), no other thresholds were used during the BLAST analysis. All the hits to reference the 16S rRNA database were considered, and specific filters were applied to the hits to remove false positives. Further, the best hit and multiple hits per sequence were analyzed separately to determine the discriminatory power of the sequences with respect to the assigned OTUs. Finally, the classification of OTU sequences was consolidated to compute relative abundancies (percentage composition).

Morphological Parameters
In each season, four morphological parameters were evaluated individually: total weight, height, length, and width ( Figure S1 and Table S2). The total weight varied between 56.9 g ± 5.0 (spring; mean body weight ± S.D.) and 84.3 g ± 18.5 (winter). Considering the total height measurements, the minimum values recorded were 2.6 cm ± 0.4 (spring), and the maximum values were 3.0 cm ± 0.3 (winter). Regarding the total length measurements, the values varied between 8.4 cm ± 0.8 (spring and summer) and 10.2 cm ± 1.6 (winter). Finally, the total width values varied between 4.7 cm ± 0.6 and 5.2 cm ± 0.5, where the highest and lowest widths were observed in the winter and spring, respectively.

Bacterial Analysis
In the present study, the microbiological quality of oysters and their farming waters was examined in four seasonal sampling surveys (Table 2). Salmonella spp. and L. monocytogenes were not found in the flesh or intra-valvular liquid. The level of E. coli contamination was found to be between 20 (summer and spring) and 92 (winter) MPN E. coli/100 g in the edible portion. Furthermore, viable enterococci were not detected in any flesh or intra-valvular liquid samples. On the contrary, Salmonella spp., Enterococcus spp., and E. coli were detected in the farming waters. E. coli was detected in all seasons, whereas Enterococcus spp. was detected in summer, autumn, and winter, and Salmonella spp. was only detected in summer and winter in the farming water samples. The highest concentration of heterotrophic marine bacteria in the farming water (1.5 × 10 4 CFU/100 mL) was found in the sample collected in winter.   S A : Satisfactory according to [17]; S B /U B : Satisfactory/Unsatisfactory according to [33,34]; S C : Satisfactory according to [35]. CFU: colony-forming unit, MPN: most probable number.
The number of total microorganisms on superficial biofilms (covering the outer shell) seems to have followed their abundance in the farming water, particularly in the samples collected in winter and spring. On the contrary, the number of marine heterotrophic bacteria on the surface biofilm of the oysters and their feeding waters was less articulated: whereas similar values were found in autumn, in summer and in spring, the difference exceeded two logarithms.
Regarding intra-valvular liquid samples, there were differences between the number of microorganisms detected in this physiological fluid and the quantity found in the farming water column. A higher concentration of microorganisms in the intra-valvular liquid was found compared to the farming water column during summer and autumn. Despite this, the number of fecal bacteria (E. coli and Enterococci) was generally higher in water than in the intra-valvular liquid, similarly to what was observed with the superficial biofilm. Likewise, marine heterotrophic bacteria in the intra-valvular liquid showed the same dynamics when compared to the superficial biofilm. Indeed, the highest value observed for this group of microorganisms was with the sample collected during summer. On the other hand, the hemolymph showed an increased number of marine heterotrophic bacteria in the summer and spring samples.
Analysis of the superficial biofilm, intra-valvular liquid, and hemolymph samples by the FISH protocol revealed the presence of Vibrio spp. in 100% of the samples ( Table 3). The hemolymph was the most contaminated material (median, 4.1 × 10 6 cells/g), and the highest values of Vibrio spp. cells were observed during summer. The FISH method allowed the detection of E. coli cells in 75% of the superficial biofilm samples (summer, winter, and spring), whereas the bacteriological method only detected viable E. coli in the winter sample. A clear contrast between the traditional plating method and the FISH cell counting was also observed with regard to Pseudomonas spp. Data are expressed as cells/mL (intra-valvular liquid and hemolymph) and as cells/g (superficial biofilm).

Antimicrobial Susceptibility Testing
In this study, we have evaluated the antimicrobial susceptibility of 30 E. coli isolates that were obtained from the farming water (n = 18), edible portion (n = 10), intra-valvular liquid (n = 1), and superficial biofilm (n = 1), and 20 Enterococcus spp. isolated from the farming water (n = 10), superficial biofilm (n = 9) and intra-valvular liquid (n = 1). Overall, 27% of the E. coli and 1% of the enterococci isolates were susceptible to all of the antimicrobial drugs tested. The remaining 22 E. coli and 19 Enterococcus spp. isolates showed resistance to at least one antimicrobial drug. The frequency of antimicrobial susceptibility to each antimicrobial drug on E. coli and Enterococcus spp. isolates was calculated and is presented in Figures 3 and 4, respectively.
Moreover, 30 susceptibility profiles of E. coli isolates and 20 susceptibility profiles of Enterococcus spp. isolates were analyzed, where 37% and 35% were revealed to be MDR E. coli isolates and MDR Enterococcus spp. isolates, respectively. The resistance profiles of MDR strains are shown in Tables 3 and 4. MDR E. coli was isolated from the farming water (n = 6), edible portion (n = 4), and superficial biofilm (n = 1) (Table 4), and MDR Enterococcus spp. was detected in the farming water (n = 6) and superficial biofilm (n = 1) ( Table 5).
Regarding seasonality, MDR E. coli was found in farming water in summer, autumn, and winter samples, and the edible content in autumn and winter samples. On the other hand, during the winter, eight samples were found to be contaminated with MDR E. coli in the edible portion and superficial biofilm. The water contaminated with MDR Enterococcus spp. was collected during summer (n = 1), autumn (n = 3), and winter (n = 3). Furthermore, the superficial biofilm containing MDR Enterococcus spp. was collected during winter (n = 1).
Microorganisms 2023, 11, x FOR PEER REVIEW 9 of 17 nitrofurantoin revealed the highest prevalence of resistance, followed by linezolid, rifampicin, and tetracycline (16.7%), ampicillin (11.1%), doxycycline and quinupristin-dalfopristin (5.6%). It is worth mentioning that neither the third-generation of cephalosporinresistant E. coli nor vancomycin-resistant Enterococcus were found. However, the high frequency of resistance to aminopenicillins, second-generation cephalosporins, and nalidixic acid in E. coli isolates, and the resistance levels against ampicillin and linezolid in enterococci, deserve to be highlighted.   nitrofurantoin revealed the highest prevalence of resistance, followed by linezolid, rifampicin, and tetracycline (16.7%), ampicillin (11.1%), doxycycline and quinupristin-dalfopristin (5.6%). It is worth mentioning that neither the third-generation of cephalosporinresistant E. coli nor vancomycin-resistant Enterococcus were found. However, the high frequency of resistance to aminopenicillins, second-generation cephalosporins, and nalidixic acid in E. coli isolates, and the resistance levels against ampicillin and linezolid in enterococci, deserve to be highlighted.

Detection of Food-and Waterborne Viruses
The analysis of foodborne viral contamination was performed, and the results are shown in Table 6. NoV was detected in the digestive gland in the spring and summer samples, as well as in the farming water in spring. HEV was detected in the farming water in spring. HAV was not detected in any digestive gland or water samples. OsHV-1 was also not detected in any edible portion sample.

Metabarcoding Analysis
Metabarcoding analysis revealed that the microbiome of the edible portion and hemolymph throughout seasons were dominated by Vibrio spp. (22.3%), excluding the edible portion in the winter sample (O3C) ( Figure 5). The most predominant microorganisms belong to the genus Vibrio followed by Psychrilyobacter (Table 7).

Metabarcoding Analysis
Metabarcoding analysis revealed that the microbiome of the edible portion and hemolymph throughout seasons were dominated by Vibrio spp. (22.3%), excluding the edible portion in the winter sample (O3C) ( Figure 5). The most predominant microorganisms belong to the genus Vibrio followed by Psychrilyobacter (Table 7).

Discussion
Presently, official controls to prevent food poisoning associated with raw oyster consumption are based on the classification of their harvesting areas. Oysters examined in this study were harvested on the Canal de Mira, which is under threat of organic pollution and limited water renewal as it is a long narrow inlet of the seacoast, where freshwater from the Vouga River mixes with seawater from the Atlantic Ocean [3]. Therefore, the low rainfall and the increase in tourism during summer [36] are possible contributors to the rise of heterotrophic marine bacteria and Salmonella contamination in the farming water. Indeed, the higher prevalence of Salmonella in Portugal during the summer months [37] might help its spread into the aquatic environment. On the other hand, the detection of this pathogenic bacterial species during winter is most likely due to rainfall or surface runoff [38]. Enterococci and E. coli monitoring confirmed that this oyster farming area is exposed to fecal pollution, although the level of contamination was not as high as expected, considering that Canal de Mira is under anthropogenic pressure and receives treated/untreated sewage discharges [39].
Regardless of the sampling period, the edible portion of oysters showed compliance with the microbiological safety criteria set out in [16,17,[33][34][35]. The level of E. coli contamination was clearly below the legal limit for E. coli contamination in oysters reared in a class B area (>230-≤4600 NMP E. coli/100 g). However, samples collected in the rainy seasons of autumn and winter showed the highest total microorganisms, Pseudomonas and C. perfringens contamination. In the spring, the bacterium C. perfringens was found below the detection level of 10 CFU/g, and the MPN of E. coli per 100 g of flesh was the lowest and only comparable to that obtained in the summer sampling. However, the biometric measurements during this study suggested that oysters should be harvested during the winter due to their greater growth during this season. Previous work [40] found that the summer months have a negative impact on oyster growth and their immunological parameters as the oysters are exposed to high temperatures and low food availability, recovering during the autumn and winter months.
Nevertheless, taking into account only the bacteriological assessment, oysters could be harvested at any time of the year, as the microorganisms of greatest concern (Salmonella spp. and L. monocytogenes) were not detected in any of the samples collected, and fecal indicator bacteria contamination levels were also low compared to those reported by other authors [7,[41][42][43]. Furthermore, E. coli contamination levels were clearly below the European Union legal end product standard (230 MPN/100 g) [17] and enterococci, which are broadly recognized by their resistance to environmental stress [44], were not found (<10 UFC/g) in any flesh sample included in this study.
Despite having been proven that microbial colonization of oyster outer shell is shaped by the number and nature of microorganisms present in the farming water [45,46], neither Salmonella spp. nor enterococci were found, and E. coli was only found in the winter sample. Similarly, hemolymph analysis did not show contamination with the fecal bacteria that were detected in the farming water and the intra-valvular liquid.
As filter feeders, oysters developed a highly sophisticated innate immune system that is able to recognize and eliminate various microorganisms via an array of orchestrated immune reactions [47,48]. This "depuration capacity" has been previously reported in Anodonta cygnea for enterococci and E. coli [49], and also in C. gigas for Salmonella Newport [50]. Hemolymph is pivotal in oyster immune defense, and hemocytes are the main effector cell population, capable of selectively recognizing, adsorbing, internalizing, and inactivating non-symbiotic microorganisms [51,52]. Indeed, oysters' hemolymph is not sterile, being a rich microbial environment (10 2 -10 5 bacteria per g) composed mainly of organisms of the genera Vibrio, Pseudomonas, Aeromonas, and Alteroromas [51].
Analysis of the hemolymph by the FISH protocol revealed the presence of Vibrio spp., P. aeruginosa, and E. coli cells in 100%, 75%, and 25% of the samples, respectively, whereas the bacteriological method was unable to detect any colony-forming E. coli in hemolymph or pseudomonas in the flesh. These contrasting data were most likely due to the presence of viable but non-culturable (VBNC) bacterial cells, which are characterized by having a better fitness for survival under stressful conditions. In 2021, Wagley et al. [53] showed that V. parahaemolyticus VBNC cells could be resuscitated (100% revival) under favorable conditions.
In this study, we observed that the peak of Vibrio spp. cells on the surface of shells and intra-valvular liquid observed in summer was most likely the result of the proliferation of this genus with warmer water temperatures [6,54,55]. Metabarcoding analysis revealed high levels of Vibrio spp. in both the flesh and hemolymph during summer and autumn. Vibrio spp. plays an important role in oyster welfare, but also in public health, as it could be either an oyster pathogen, associated with mass summer mortalities of Crassostrea gigas or a zoonotic pathogen, including V. parahaemolyticus (the principal causes of seafood-borne disease linked to the consumption of shellfish) and V. vulnificus, which may cause serious wound infections [54,56,57]. Moreover, this study showed that hemolymph contained more Vibrio spp. compared to the edible content, which could be explained by the immunological function of hemolymph. Indeed, the overall microbiome of oysters displays a seasonal influence, also mentioned by Scannes et al. (2021) [58]. This is, to our knowledge, the first study that the occurrence of norovirus (NoV), hepatitis A virus (HAV), hepatitis E virus (HEV), and oyster herpesvirus type 1 (OsHV-1) in the Canal de Mira production area. NoV and HEV were both detected, but NoV was more frequent and the only one found simultaneously in the digestive gland and water samples collected during winter. According to Lowther et al. (2012) [43], this seasonality is typical in Europe and it might be explained by the convergence of several factors: the higher prevalence of noroviruses in the human population, the greater persistence of viral particles under winter environmental conditions (low temperature and low solar irradiation), and lower viral clearance in oysters due to the slowing of the metabolism. In the present investigation, HAV and OsHV-1 were not found in any of the samples analyzed. Since 2008, OSHV-1 has been causing epidemics with high mortality in C. gigas throughout Europe. To the best of our knowledge, OSHV-1 has only been detected in one sample of C. gigas harvested in Portugal, although the authors reported that this animal could have been imported from France.
Antimicrobial resistance remains a serious global health concern, being considered one of the most pressing global issues by the World Health Organization (2020) [59]. Paradoxically, wastewater treatment can favor the emergence and spreading of antimicrobial resistance (AMR) as resident bacterial communities are exposed to sub-inhibitory concentrations of antimicrobials (due to the elimination of these substances in the feces and urine of medicated individuals), favoring the transfer of genes between bacteria and their subsequent dissemination into aquatic environments [60,61]. The consequences of these events were found in this research, as evidenced by the isolation of both E. coli and Enterococcus spp. multidrug-resistant strains and the high frequency of resistance to important classes of antimicrobial drugs.

Conclusions
The present study was performed with a limited number of samples, which may result in a misestimation of prevalence. However, this is the first report assessing a wide range of microbiological parameters of oysters and their farming waters, combining genomics and classical plating methods to both commensal and microorganisms of great concern. In common with previous studies, the contrast between the results for the presence of E. coli and norovirus demonstrates the limitations of using E. coli to estimate and manage the risk of human enteric virus in oysters.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/microorganisms11020338/s1, Figure S1: (a) Total weight variation in grams; (b) total height variation in centimeters; (c) total length variation in centimeters; (d) total width variation in centimeters of oysters during each season; Table S1: Summary of the methodology used for bacteriologic analysis; Table S2: Summary of morphological parameters throughout the seasons.  seafood products: meeting local challenges and opportunities", founded by the Northern Regional Operational Programme (NORTE 2020) through the European Regional Development Fund (ERDF); and funded by the project OCEAN3R (NORTE-01-0145-FEDER-000064), supported by the North Portugal Regional Operational Program (NORTE2020), under the PORTUGAL 2020 Partnership Agreement and through the European Regional Development Fund (ERDF).

Data Availability Statement:
The original contributions presented in the study are included in the article. Further enquiries can be directed to the corresponding author.