The Role of Aquatic Ecosystems (River Tua, Portugal) as Reservoirs of Multidrug-Resistant Aeromonas spp.

The inappropriate use of antibiotics, one of the causes of the high incidence of antimicrobial-resistant bacteria isolated from aquatic ecosystems, represents a risk for aquatic organisms and the welfare of humans. This study aimed to determine the antimicrobial resistance rates among riverine Aeromonas spp., taken as representative of the autochthonous microbiota, to evaluate the level of antibacterial resistance in the Tua River (Douro basin). The prevalence and degree of antibiotic resistance was examined using motile aeromonads as a potential indicator of antimicrobial susceptibility for the aquatic environment. Water samples were collected from the middle sector of the river, which is most impacted area by several anthropogenic pressures. Water samples were plated on an Aeromonas-selective agar, with and without antibiotics. The activity of 19 antibiotics was studied against 30 isolates of Aeromonas spp. using the standard agar dilution susceptibility test. Antibiotic resistance rates were fosfomycin (FOS) 83.33%, nalidixic acid (NA) 60%, cefotaxime (CTX) 40%, gentamicin (CN) 26.67%, tobramycin (TOB) 26.67%, cotrimoxazole (SXT) 26.67%, chloramphenicol (C) 16.67%, and tetracycline (TE) 13.33%. Some of the nalidixic acid-resistant strains were susceptible to fluoroquinolones. Multiple resistance was also observed (83.33%). The environmental ubiquity, the natural susceptibility to antimicrobials and the zoonotic potential of Aeromonas spp. make them optimal candidates for studying antimicrobial resistance (AMR) in aquatic ecosystems. Aquatic environments may provide an ideal setting for the acquisition and dissemination of antibiotic resistance because anthropogenic activities frequently impact them. The potential risk of multi- and pan-resistant bacteria transmission between animals and humans should be considered in a “One Health—One World” concept.


Introduction
Extensive use of water and anthropogenic activities contribute to water body pollution. Agricultural, urban, and animal waste, often characterized by numerous toxic and carcinogenic chemicals, pathogenic bacteria, and antibiotics, as well as antibiotic resistance genes (ARGs), loaded with microflora, can contaminate water and enter the food chain, posing a considerable danger to public health [1,2].
Antibiotic resistance is rising to dangerously high levels worldwide. New resistance mechanisms are emerging and spreading globally, threatening our ability to treat common infectious diseases in humans and animals [3]. ARGs are found in the clinical and natural environments and are linked to antibiotic-resistant bacteria (ARB) [4][5][6][7][8][9]. Antibiotics, ARBs, The Tua River basin's hydrological regime presents significant interannual variability, with an average annual runoff of 988.1 hm 3 . Intra-annual variability also occurs, with high values in winter and low values in summer (on average it varies between 4 hm 3 in August and 277 hm 3 in January). The basin area in Portugal is 3,122.80 km 2 and in Spain is 690.742 km 2 , making a total of 3,813.540 km 2 . The Tua basin is subject to relatively high organic loads, especially the Tua River, coming from agriculture, urban agglomerations, and industrial activities (factories producing and processing food oils, and others that have set up at an agro-industrial complex-wool, nuts, and a regional slaughterhouse), resulting in high biological oxygen demand (BOD), chemical oxygen demand (COD) and total suspended solids (TSS). In this river basin, there is only one small town (Valpaços) upstream from the T1 site (approximately 12 km), with 16,882 inhabitants. The T2 site is influenced by the city of Valpaços (Rabaçal River), Vinhais (9066 inhabitants, Tuela River) and Mirandela (23,850 inhabitants, Tua River).
Phenomena of eutrophication and oil stains over large areas of its waters have been recurrent since 2017 (Portuguese Environment Agency-APA). The fires of 2013 and 2016 also left 5000 tons of hazardous waste from the agro-industrial complex. Additionally, the presence of several hydroelectric dams results in loss of connectivity for aquatic communities and inflow regulation, which, especially in the summer, can lead to the intensification of eutrophication phenomena.

Sample Processing and Isolation
The water samples were collected at two sites located in the Tua River basin (Rabaçal River T1: 41 • 30 45.821 N; 7 • 12 32.92 W, and Tua River T2: 41 • 24 18.69 N; 7 • 9 38.93 W), in two seasons, the summer and autumn of 2018 ( Figure 1). Two replicates of water were collected in 1 L sterile glass bottles, stored in a cold container, and transported to the Laboratory of Medical Microbiology in Trás-os-Montes e Alto Douro University (UTAD), Vila Real, Portugal. For each replicate sample, three volumes of each were analyzed. The detection and quantification of bioindicators was performed by the filter membrane method. Briefly, 100 mL of water samples were filtered with nitrocellulose membrane filters (0.45 µm pore size) (Millipore, Watford, UK). The filter was put on solid culture media with selective and differential for growth and development of Aeromonas spp. and Pseudomonas spp. (Glutamate Starch Phenol red (GSP) agar, Oxoid Thermo Scientific, Oxoid, UK), supplemented with 2 µg/mL of Imipenem (GSP-IMP) to select for potential carbapenemases producers. For each sampling site (T1 and T2) membranes were also placed on plates without antibiotic in order to estimate the proportion of resistant Aeromonas. Plates were incubated at 30 • C (GSP plates) for 18 to 24 h. From each plate, up to two to five colonies were identified based on colony morphology on GSP and GSP-IMP agar, with Aeromonas spp. presenting a yellow colour and Pseudomonas spp. presenting a red colour. Yellow colonies were selected and purified for further confirmation. Once confirmed, they were stored in aliquots of Brain Heart Infusion (BHI) medium with 17% (v/v) glycerol at −80 • C.

Statistical Analysis
Differences between antimicrobial resistance rates of Aeromonas spp. isolates from different river locations (sites T1 and T2), and the frequency of resistance isolates among different seasons (summer and autumn) were evaluated through Tukey's test using IBM SPSS statistics v.22.0 software (SPSS Inc., Chicago, IL, USA). Statistical calculations were based on a confidence level of ≥95%; p-value < 0.05 was considered statistically significant.

Results and Discussion
Although the Aeromonas genus is autochthonous in the aquatic environment, and easy to detect in any water type, they have received increasing attention as opportunistic pathogens [2,[35][36][37]. Aeromonas spp. are important pathogens of fish and can cause diseases in humans, as well as in wild animals [19,37].
In the present study, 30 Aeromonas isolates were collected from the sampling survey in the Tua River basin, in summer and in autumn seasons (2018). Specifically, 13 strains were isolated from the site T1 and 17 from the downstream site T2 ( Figure 1). All isolates were tested for resistance profile to a panel of 19 antibiotics, namely 20 Aeromonas isolates obtained from GSP agar plates, and 10 Aeromonas isolates from GSP-IMP agar plates. The proportions of antimicrobial resistance of 30 Aeromonas strains are detailed in Figure 2.

Results and Discussion
Although the Aeromonas genus is autochthonous in the aquatic environment, and easy to detect in any water type, they have received increasing attention as opportunistic pathogens [2,[35][36][37]. Aeromonas spp. are important pathogens of fish and can cause diseases in humans, as well as in wild animals [19,37].
Nevertheless, the combination of an aminopenicillin and a carboxipenicillin with a β-lactamases inhibitor was effective in reducing resistance, as shown by the decrease in the proportion of resistant strains: 93.33% (amoxicillin) versus 73.33% (amoxicillin/clavulanic acid); and 83.33% (ticarcillin) versus 56.67% (ticarcillin/clavulanic acid) ( Figure 2). Some of the nalidixic acid-resistant strains were also susceptible to fluoroquinolones (20%). Indeed, the resistance profile of β-lactams occurs because enzymes, named β-lactamases, produced by a large variety of Gram-negative and -positive bacteria are capable of hydrolyzing the β-lactam ring of antibiotics such as penicillins, cephalosporins and aztreonam [44]. Thus, the combination of aminopenicillin with clavulanic acid increases the effectiveness (reduces the resistance profile) by inhibiting the β-lactamases, as evidenced in other studies [45].
Resistance to a first-generation cephalosporin (cephalothin) was measured in 70% of the isolates and for cefotaxime, a third-generation cephalosporin, resistance was measured in 40%. Aztreonam, a monobactam antibiotic, was more effective against these bacteria, with only 33.33% of the isolates being resistant. It should be noted that the occurrence of resistance to imipenem, an antibiotic belonging to the carbapenem group, was observed in 43.33% of isolates. Among aminoglycosides, resistant strains were between 10% (kanamycin) to 26.67% (gentamicin and tobramycin). Although the susceptibility of Aeromonas strains to fluoroquinolones has been observed in other studies [19,46], in the present study, resistance up to 50% to ciprofloxacin was observed. Likewise, more than 70% of the Aeromonas isolates were found to be susceptible to chloramphenicol, trimethoprimsulfamethoxazole and tetracycline (Figure 2).
The proportions of antimicrobial-resistant Aeromonas spp. isolated from the T1 (n = 13; nine grown in GSP and four in GSP-IMP) and T2 (n = 17; 11 grown in GSP and six in GSP-IMP) river locations are summarized in Figure 3. Generally, the pattern of antimicrobialresistant Aeromonas spp. showed that high values were observed on site T2. Indeed, the T2 site was the one where the most resistant isolates were observed (32.46%), when compared with the T1 site (14.39%). Likewise, the percentages of fosfomycin resistance were markedly higher among isolates derived from the T2 site than among those obtained from the T1 site. It must be pointed out that from antibiotics employed in clinical practice, for human and veterinary medicine fosfomycin is classified as a "critically important antimicrobial" and a "highly important antimicrobial", respectively [47,48]. It is noteworthy that it is precisely this site (T2) that presents the greatest anthropogenic pressure, namely, related to the production and refining of food oils and the agro-industrial complex, and all with the urban effluent discharged by the Mirandela city. The T2 site was also heavily affected by the fires of 2013 and 2016 in the agro-industrial complex that left 5000 tons of hazardous waste that continues to be released into the aquatic ecosystem whenever rainfall occurs.  Rivers are considered reservoirs of MDR aeromonads, since they receive water from a wide range of environments, like wastewater treatment plants (WWTPs), industrial effluents, agricultural activities, hospital sewage or animal production effluents [19,37]. The incidence of Aeromonas spp. in wastewater was observed to be high, providing an ideal Rivers are considered reservoirs of MDR aeromonads, since they receive water from a wide range of environments, like wastewater treatment plants (WWTPs), industrial effluents, agricultural activities, hospital sewage or animal production effluents [19,37]. The incidence of Aeromonas spp. in wastewater was observed to be high, providing an ideal setting for the acquisition and dissemination of antibiotic resistance mechanisms [29,30,39,40].
Some differences were observed in site T1, between Aeromonas spp. isolates at summer (four strains grown in GSP and two grown in GSP-IMP) and autumn (five grown in GSP and two grown in GSP-IMP) seasons, although they were not statistically significant (Figure 4). In the summer season, a lower percentage of antimicrobial resistance to βlactams was observed in site T1 (19.25 ± 15.93%) than site T2 (31.17 ± 5.60%) (Figure 4). On the other hand, in site T1 none of the isolates were resistant to aminoglycosides (K) and amphenicos (C) in autumn season (Figure 4). Conversely, in T2 (in summer, four strains grown in GSP and three grown in GSP-IMP; in autumn, seven strains grown in GSP and three grown in GSP-IMP) this absence of resistance was observed in summer. In this work, the percentage of CIP unsusceptible Aeromonas spp. strains was generally lower in comparison with previous studies [29,49,50].
Comparing two sampling periods (summer vs. autumn), it was observed that the resistance to antibiotics increased in autumn, being 1.3-fold higher ( Figure 4). However, it should be noted that the analyzed summer period corresponded to the end of spring/early summer when the flow rates were still relatively high after a very rainy April, a relatively dry May and a very rainy June (Portuguese Institute of the Sea and the Atmosphere-IPMA). In contrast, September (the month of autumn sampling) was classified as extremely hot and extremely dry. According to the IPMA, this month was the warmest since 1931 and the second driest in the last 30 years. Thus, and as would be expected, the absence of precipitation, with a direct consequence in the drastic reduction in flows, was the main factor that contributed to a greater concentration of pollution in the river (less capacity for self-purification). Therefore, a high percentage of resistant bacteria could be found, increasing the risk to public health due to the direct use of water in its different applications (human and animal consumption, irrigation, bathing). To highlight this, it is precisely in the months of July, August and part of the month of September that the use of this essential resource and pollution are felt most intensely, due to a seasonal increase in the emigrant population, which is more pronounced in the northern and central regions of Portugal, often doubling the resident population.
However, Knapp et al. [51] showed that it is precisely in periods of greater precipitation that the potential for human exposure to ARB increases, since it is just in these seasons that the ARGs are found to be more equally distributed, both in the watercourse and in the water column. A wide variety of human activities are generally associated with strong impacts on the environment. The stretches of rivers that run through cities are often used as receiving bodies for treated and untreated urban wastewater worldwide [52][53][54]. Some of these rivers, contaminated by man-made sewers and animal effluents, are among the most extreme examples of ecosystems disturbed by anthropogenic activities. General treatment processes to improve water quality in WWTPs, using primary and secondary or biological treatments, are not very efficient in reducing Aeromonas spp. concentration [9,55,56]. Nevertheless, using tertiary treatments, such as chemical (ozone, chlorination), physical (ultraviolet radiation), and natural tertiary treatments such as lagooning, these bacteria can be completely eliminated [57][58][59]. Thus, it would be important to carry out further studies to assess antibiotic resistance variation with the seasons because of the results obtained here. Comparing two sampling periods (summer vs. autumn), it was observed that the resistance to antibiotics increased in autumn, being 1.3-fold higher (Figure 4). However, knowledge on the MDR in the aquatic environmental bacteria, 25 out of 30 Aeromonas isolates showed an MDR profile to 19 antibiotics, where 83.33% of the strains were resistant to more than three antibiotic classes, indicating a high level of multi-resistance in the Tua River basin ( Figure 5). Previous studies [18,[60][61][62][63][64] reported β-lactams and quinolone-resistant isolates of Aeromonas spp. recovered from freshwater, fishes, and humans. To our knowledge, this was the first report that showed multidrug-resistant Aeromonas spp. isolated from the Tua River, representing a potential risk to the population since an appreciable part of the population residing in the Tua River Basin still uses fish and river mussels in their daily diet. Foodborne diseases (FD) are a priority in public health issues around the world due to their incidence and mortality [64]. The genus Aeromonas is regarded not only as an important disease-causing pathogen of cold-blooded species but also as the etiologic agent responsible for a variety of infectious complications in both immunocompetent and immunocompromised persons [23]. Carbapenems are often used as "last-line agents" or "antibiotics of last resort" for the treatment of severe infections due to multidrug-resistant hospital-acquired bacteria [50]. However, the appearance of MDR seriously threatens this class of lifesaving antimicrobials. The presence of multidrug-resistant bacteria in water has resulted from the indiscriminate use of antibiotics in the last decades, which has exerted a selective pressure on bacteria from the environment, together with effluent treatment systems that are inefficient in the antibiotic removal process [65][66][67][68]. Antibiotics and resistant bacteria are entering our local waterways and have the potential to influence biotic processes. Another important point is associated with the transfer of resistant genes from not-pathogenic bacteria to pathogenic bacteria, and on to humans interacting with the aquatic environment [69][70][71].

Conclusions
From the results of this study, it was clear that there was a pool of MDR aeromonads in the Tua River basin. These results, together with the resistance patterns of Aeromonas spp. to antibiotic tests, suggest that Aeromonas spp. can be effective bioindicator organisms for monitoring antimicrobial resistance in rivers. This knowledge is essential to manage and mitigate potential risks to human health, emphasizing the need for predicting and preventing the spread of antibiotic-resistant pathogenic aeromonads. A continuous monitoring surveillance in aquatic systems is imperative, considering the interactions between the key elements (geographical, ecological, human activities and the food-agricultural components) within the "One Health-One World" approach.
In view of the results obtained in this study, it would be important to assess, in the long term, the seasonal variation in antibiotic resistance. Further studies are also needed to investigate aquatic animals for the possible presence of antibiotic resistant bacteria, as The presence of multidrug-resistant bacteria in water has resulted from the indiscriminate use of antibiotics in the last decades, which has exerted a selective pressure on bacteria from the environment, together with effluent treatment systems that are inefficient in the antibiotic removal process [65][66][67][68]. Antibiotics and resistant bacteria are entering our local waterways and have the potential to influence biotic processes. Another important point is associated with the transfer of resistant genes from not-pathogenic bacteria to pathogenic bacteria, and on to humans interacting with the aquatic environment [69][70][71].

Conclusions
From the results of this study, it was clear that there was a pool of MDR aeromonads in the Tua River basin. These results, together with the resistance patterns of Aeromonas spp. to antibiotic tests, suggest that Aeromonas spp. can be effective bioindicator organisms for monitoring antimicrobial resistance in rivers. This knowledge is essential to manage and mitigate potential risks to human health, emphasizing the need for predicting and preventing the spread of antibiotic-resistant pathogenic aeromonads. A continuous monitoring surveillance in aquatic systems is imperative, considering the interactions between the key elements (geographical, ecological, human activities and the food-agricultural components) within the "One Health-One World" approach.
In view of the results obtained in this study, it would be important to assess, in the long term, the seasonal variation in antibiotic resistance. Further studies are also needed to investigate aquatic animals for the possible presence of antibiotic resistant bacteria, as well as to understand the role of ARG and the mobile genetic element relatedness of these isolates.
Author Contributions: In the overall context of this paper, the contributions are as follows: A.T., S.V. and M.J.S. made the sampling collection. S.G., C.F., S.M. and M.J.S. isolated the bacteria and performed the laboratory measurements. C.F., S.V. and M.J.S. made substantial contributions to conception and design and drafted the manuscript. S.G., C.F., S.V. and M.J.S. edited the draft. E.C. and A.T. provided additional information, local insights, and supplementary references. All authors provided critical feedback and helped shape the research, analysis, and manuscript. All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.