Prevalence and Antimicrobial Resistance of Bacteria Isolated from Marine and Freshwater Fish in Tanzania

This study aimed to determine the prevalence and antimicrobial resistance of bacteria isolated from retail fish and shrimp in Tanzania. A total of 92 fish and 20 shrimp samples were analyzed. Fish samples consisted of 24 Nile tilapia, 24 Nile perch, and 24 red snapper. The isolates were identified by their morphological characteristics, conventional biochemical tests, and analytical profile index test kits. The antibiotic susceptibility of selected bacteria was determined by the disc diffusion method. Out of the 92 samples analyzed, 96.7% were contaminated with 7 different bacterial species. E. coli was the most prevalent bacteria (39%), followed by Klebsiella spp. (28%) and Salmonella spp. (16%). Other species isolated from this study were Staphylococcus spp. (8%), Citrobacter (4%), Shigella spp. (3%), and Pseudomonas spp. (1%). All samples were analyzed for Campylobacter spp.; however, none of the samples tested were positive for Campylobacter spp. Fish from the open-air market were contaminated by six bacterial species: E. coli (40%), Klebsiella spp. (26%), Salmonella spp. (24%), Shigella spp. (6.7%), Citrobacter spp. (6.5%), and Pseudomonas spp. (2%), while E. coli (37%), Klebsiella spp. (33%), Staphylococcus spp. (23%), and Shigella spp. (2%) were isolated in supermarket samples. According to the International Commission on Microbiological Specifications for Foods criteria, 54 (58.7%) and 38 (41.3%) samples were good and marginally acceptable, respectively. E. coli isolates were resistant to penicillin (PEN), erythromycin (ERY), gentamicin (GEN), azithromycin (AZM), and tetracycline (TET), while Salmonella spp. isolates exhibited resistance to gentamicin (CN), tetracycline (TET), penicillin (PEN), and erythromycin (ERY). These results suggest that the presence of these bacteria might cause a health risk/hazard to human beings and may cause disease to susceptible individuals, especially immune-compromised consumers.


Introduction
Fish are a vital source of food for people and are the most important source of high-quality protein. Fish provide approximately 16% of the animal protein consumed by the world's population [1]. Approximately 60% of developing countries derive 30% of their annual protein from fish [2]. Fish provide 19% of protein and are one of the cheapest sources of protein in Africa. In Tanzania, per capita, fish consumption is 8.0 kilograms, and fish make up 19.7% of the country's animal protein intake [3]. However, fish consumption is much higher in coastal regions and Zanzibar (almost 20 kg per capita), as well as along the coasts of Lake Tanganyika and Lake Victoria. Apart from being a rich source of protein, fish also play a unique role in providing a range of micronutrients and essential fatty acids, especially long-chain polyunsaturated fatty acids, which cannot be easily substituted by other food commodities [4,5]. Despite these numerous advantages of fish, they are susceptible to a wide variety of bacterial pathogens, most of which can cause disease and are considered by some to be saprophytic in nature [1]. Fish and fishery products are known to be sources of transmitting foodborne infections and intoxication if not properly handled/processed [6].
It has been reported that the microbiological diversity of fresh fish muscle depends on the fishing grounds and environmental factors around it [7]. According to Clucas and Ward [8], the type of microorganisms that are associated with a particular fish depends on their habitat.
Bacterial pathogens associated with fish are classified as indigenous and nonindigenous. Examples of nonindigenous bacterial pathogens include Escherichia coli, Clostridium botulinum, Shigella dysenteriae, Staphylococcus aureus, Listeria monocytogenes, and Salmonella. ese nonindigenous organisms contaminate the fish or the habitat in one way or the other. Examples of indigenous bacterial pathogens are Vibrio species and Aeromonas species, which are found naturally living in fish habitats [9,10]. e bacteria from fish only become pathogens when there are stressors, such as poor water quality and overstocking, which allow opportunistic bacterial infections to prevail [1]. Additionally, studies have reported the presence of indicator microorganisms of fecal contamination, opportunistic, and pathogenic bacteria to humans in fish samples [11,12].
Human infections caused by pathogens transmitted from fish are widespread and depend on the season, the patients' contact with fish and related environments, dietary habits, and the immune system status of the individual [13]. Pathogens can be transmitted through food or the handling of fish. Foodborne illnesses such as dysentery and diarrhea resulting from the consumption of contaminated fish can result in economic losses. Microbial associations with fish compromise the safety and quality of human consumption; this is particularly critical when microorganisms are opportunistic and/or pathogenic in nature [11,13].
Poor fish handling, processing, and packaging methods and practices in retail markets are still common in Tanzania [14]. ere are limited studies on the prevalence of bacteria isolated from retail fish and shrimps in Tanzania. e risk of getting foodborne diseases or food poisoning by fish consumers and handlers may be high in Tanzania. erefore, this study investigates the prevalence of bacteria isolated from retail fish and shrimps in Tanzania.
Several antibiotics used in agriculture and aquaculture are critical for human medicine by the World Health Organization (WHO), including the antibiotic classes of tetracyclines, quinolones, and penicillin [15].
Resistance to all antibiotic classes has been observed in a wide range of bacteria, including those pathogenic to humans [16,17].
Antimicrobial resistance is widespread in Tanzania, and there is widespread misuse of antimicrobials in the livestock and aquaculture industries. However, there is a scarcity of information regarding antibiotic resistance in Tanzania [18]. e scarcity of data may have contributed to the use of drugs without a prescription or prescribing drugs without carrying out a laboratory test to identify the pathogens and test their susceptibility to drugs for effective treatment.
Hence, another goal of this study was to determine the prevalence and antimicrobial resistance of foodborne pathogens isolated from retail fish and shrimps in Tanzania.

Sample Collection.
A total of 92 fresh fish samples of both marine and freshwater were purchased from two different open-air fish markets (n � 60) and two supermarkets (n � 32) in the Ilala district, Dar es Salaam, Tanzania. Out of the 92 samples collected, twenty-four samples were of Nile tilapia, Nile perch, red snapper, and twenty shrimp. After the collection, samples were aseptically and immediately transported in a sterile polypropylene bag placed in a cooler box. e cooler box contained crushed ice, and the temperatures were between 4°C and 8°C during transportation. e samples were transported to the Open University Laboratory, and for further analysis, the samples were analyzed at the College of Veterinary and Biomedical Sciences, Sokoine University of Agriculture.
2.2. Culture, Isolation, and Identification. For isolation and identification of bacteria, culturing was performed by chopping a piece of fish flesh aseptically and spreading it over the surface of blood agar and MacConkey agar before incubation between 24 and 48 hours at 37°C. For Campylobacter spp. identification, a piece of fish samples was first enriched in Bolton broth overnight and later inoculated onto modified charcoal cefoperazone deoxycholate agar (mCCDA) (Oxoid Ltd., Basingstoke, Hampshire, England) containing the Campylobacter mCCDA selective supplement SR155 E (Oxoid Ltd., Basingstoke, Hampshire, England). Incubation was performed as previously described by Kurekci et al. [19] at 37°C for 48 h under microaerophilic conditions generated by CampyGenTM gas sachets (Oxoid, Basingstoke, England, UK).
Gram staining was done to differentiate organisms based on the structure of their cell walls [20]. Furthermore, the classical identification of bacterial colonies and biotyping were performed according to the methods described by Abbott et al. [20] and El Deen et al. [21] with slight modifications. Briefly, the isolates were conventionally studied for their macro-and micromorphological characteristics and then by biochemical assays that consisted of 21 phenotypic characteristics tests. e assays included lactose, raffinose, trehalose, dulcitol, maltose, mannose, D-mannitol, melibiose, sucrose, citrate, urea, indole, catalase, motility, ampicillin resistance, m-inositol, oxidase, nitrate, cellobiose, and xylose. Triple sugar iron agar and IMViC were also used for the characterization of members of the family Enterobacteriaceae.

Microbial Load by Total Viable Count.
To assess the level of microbial contamination, 10 g of fish pieces was rinsed in 10 mL of sterile normal saline. en, 1 mL of the rinse was serially diluted 10-fold using 10 universal bottles containing sterile normal saline. In each dilution, 1 mL was poured on the plate count of medium Petri dishes in duplicate. e plates were incubated at 37°C for 24 h. en, colonies were counted, and the average colony counts were used to establish the colony-forming units (CFU/mL or CFU/g).

Antimicrobial Assay.
Antimicrobial susceptibility testing was performed using the disc diffusion method. e isolates were tested against quinolones, namely, ciprofloxacin (CIP); macrolides, erythromycin (E), and azithromycin (AZM); and aminoglycosides, gentamicin (CN) (0.06-64 µg/mL), and tetracycline (TE). Other antimicrobials were clindamycin (DA) and penicillin (P). All antimicrobials were supplied by Sigma-Aldrich (St. Louis, MO, USA). For antimicrobial susceptibility assays, a pool of bacterial colonies was used to prepare suspensions corresponding to 0.5 McFarland standards (1.5 × 10 8 CFU/mL) using normal saline, and then bacteria were spread on top of Müller-Hinton agar using a sterile swab. Discs were placed on top of the medium, and the plates were then incubated at 37°C for 24 h. Zones of inhibition were measured by means of a simple ruler, and the diameter was recorded in millimeters (mm).
Isolates were defined as susceptible, intermediate, or resistant in accordance with the CLSI [22] Enterobacteriaceae breakpoints.

Data Analysis.
Data from counts (in log CFU/g) were analyzed (means, standard deviations, and standard errors) using the Statistical Package SPSS v21 (IBM Corporation, Armonk, NY, USA). Descriptive statistics were used to describe the results of the prevalence analysis. e prevalence was estimated as the number of samples detected positive for Salmonella spp. Campylobacter spp., E. coli, Shigella spp., and Staphylococcus spp. isolation from the total sample analyzed.

Total Viable Count (TVC) and Microbiological Quality
Data.
e average TVC for the 92 samples was 4.57 log CFU/g, ranging from 3.82 to 4.70 and 5.22 log CFU/g in tilapia, Nile perch, and red snapper, respectively. A significant difference (P < 0.05) was observed between Nile tilapia and red snapper samples, as shown in Table 1. When comparing total samples (Table 1), red snapper had the highest mean TVC at 5.22 log CFU/g, and Nile tilapia had the lowest at 3.82 log CFU/g (P < 0.05). ere is no significant difference (P > 0.05) in TVC log CFU/g between samples from the supermarket and retail market. However, samples from the supermarket have a lower TVC at 3.77 log CFU/g compared to samples from the open-air market (4.99 log CFU/g) (data not shown).
Based on the International Commission on Microbiological Specifications for Foods criteria, 54 (58.7%) and 38 (41.3%) samples were good and marginally acceptable, respectively (Table 1). Nile tilapia had a higher good quality sample, followed by shrimp, Nile perch, and red snapper. None of the samples had unacceptable microbiological quality.

Prevalence of Campylobacter, E. coli, Listeria, Salmonella, and Shigella.
e prevalence and contamination rates of open-air market fish with pathogenic bacteria are presented in Table 2). Seven bacterial species were identified from all analyzed samples. e most prevalent bacterial species recovered in the samples was E. coli (39%), followed by Klebsiella spp. (28%) and Salmonella spp. (16%). Other species isolated from this study were Staphylococcus spp. (8%), Citrobacter (4%), Shigella spp. (3%), and Pseudomonas spp. (1%). Nile tilapia were highly contaminated by E. coli (48%), followed by shrimp (45%). Nile perch were highly contaminated by Salmonella spp., followed by Nile tilapia. Only shrimp and Nile perch were contaminated by Shigella spp., while Staphylococcus spp. were isolated in Nile tilapia and Nile perch only.
All samples were screened for Campylobacter, but no Campylobacter was recovered by culture.
Samples from the open-air market were contaminated by six bacterial species, whereas four species were detected in supermarket samples (Figure 1)
As observed in Figure 2, bacterial isolates from marine fish were more resistant to antibiotics than those isolated from freshwater fish. Shigella spp. isolated from marine fish were resistant to tetracycline 28 (17%), gentamicin, and ciprofloxacin, while no isolates from freshwater were resistant to antibiotics.
Salmonella spp. isolates recovered from marine fish showed resistance to six antimicrobial agents comprising gentamicin, tetracycline, penicillin, erythromycin, azithromycin, and ciprofloxacin. Salmonella spp. isolates recovered from freshwater fish exhibited resistance to only three antimicrobial agents comprising gentamicin, tetracycline, and erythromycin.

Discussion
Most outbreaks of food poisoning related to fish are derived from the consumption of raw or insufficiently heat-treated fish and cross-contamination during handling. According to Huss et al. [24] and Aberoumand [25], approximately 12% of foodborne outbreaks related to the consumption of fish are caused by bacteria.     In this study, a total of 7 different species of bacteria were isolated. e most prevalent bacteria were E. coli (39%), followed by Klebsiella spp. (28%) and Salmonella spp. (16%). Other bacteria isolated in this study are Staphylococcus spp. (8%) and Shigella spp. (3%). Citrobacter (4%) and Pseudomonas spp. (1%) were also isolated in this study.

International Journal of Microbiology
Salmonella spp. was detected in fish samples (16%) collected from different retail markets in this study. Higher results were reported by Kumar et al. [26], Jegadeeshkumar et al. [27], and Budiati et al. [28] at rates of 30%, 43.8%, and 90% in fish samples, respectively. In contrast, other studies from Trinidad and Turkey detected Salmonella spp. at 0% and 5% in fish samples, respectively [29,30]. e presence of some bacterial species isolated from this study can cause food poisoning, diarrhea, typhoid fever, and shigellosis. Sichewo et al. [1] suggested that, when present in food, pathogens such as S. aureus, Salmonella, Shigella, and Pseudomonas are most likely to cause foodborne diseases. e presence of Salmonella, Shigella, and E. coli in fish indicates faecal and environmental contamination [1]. Coliforms such as E. coli are not normal bacteria in fish and are commonly present where there has been faecal contamination from warmblooded animals [31]. E. coli is known to be a reliable indicator of fecal contamination in small numbers and large numbers, and it is an indicator of mishandling [32]. e presence of E. coli in this study could be attributed to poor mishandling of fish by traders. e presence of these bacteria isolated from the open-air market samples might also possibly be due to unhygienic handling during transportation and storage. e use of contaminated water for cleaning and processing fish in the fish market is presumably the cause of secondary contamination. e lack of proper drainage facilities and heavy fly infestation in these fish markets also promotes tertiary contamination to a great extent. e isolation rate of Staphylococcus spp. in the retail fish samples in the present study (8%) is relatively low compared with previous reports, which had rates ranging from 19.9% to 61.7% [33][34][35][36]. However, in this study, Staphylococcus spp. were isolated in samples from supermarkets only, which might be associated with inappropriate handling of fish. e findings of this study show that the mean TVC for 92 samples was 4.57 log CFU/g, ranging from 3.82 to 4.70 and 5.22 log CFU/g in Nile tilapia, Nile perch, and red snapper, respectively. is is similar to the study of Wang et al. [37], who found the TVCs for salmon and tilapia to be 5.58 and 4.37 log CFU/g, respectively. According to the International Commission on Microbiological Specifications for Foods criteria, the microbiological quality of the fish examined in this study is acceptable and does not pose a potential risk to public health. e results from this study also showed that the enteric bacteria isolated from fish were resistant to some antibiotics. It was observed that a small percentage of Salmonella spp. were resistant to gentamicin, tetracycline, penicillin, and erythromycin. Similar to our results (except erythromycin), Ponce et al. [38] found that the majority of isolates were resistant to ampicillin, tetracycline, chloramphenicol, gentamicin, sulfisoxazole, streptomycin, and kanamycin. International Journal of Microbiology 5 A total of 29,25,17,15, and 13% of the E. coli isolates exhibited resistance to penicillin, erythromycin, gentamicin, azithromycin, and tetracycline, respectively.
is is in contrast to the results obtained by Kibret and Abera [39], where erythromycin and tetracycline were highly resistant to E. coli isolates at 89.4 and 72.6%, respectively. e high prevalence of resistance to tetracycline, ampicillin, and cotrimoxazole in E. coli in the region has also been reported by Sifuna et al. [40], in which E. coli demonstrated resistance mostly to ampicillin and tetracycline. According to Sifuna et al. [40], the resistance pattern reported in their study can also be linked to the use of these drugs in veterinary practice and lead to resistance in humans.
In this study, bacteria isolated from marine fish were more resistant to antibiotics than those from freshwater fish. Salmonella spp. isolates recovered from marine samples were resistant to six antimicrobial agents, while isolates recovered from freshwater fish were resistant to three antimicrobial agents. Shigella isolates recovered from marine fish were resistant to tetracycline, gentamicin, and ciprofloxacin. However, Shigella isolates recovered from freshwater samples were not resistant to any antimicrobial agents. In contrast, a study on the antimicrobial resistance in isolates from marine species showed a high occurrence of antibiotic resistance, with 68% of isolates demonstrating resistance to at least one antibiotic, which is higher than the occurrence of antibiotic resistance observed in the current study [41]. Another study examining antimicrobial resistance in marine mammals reported E.coli was resistant to doxycycline (5%), amoxicillin (4%), and gentamicin (1%). Lockwood et al. [42] carried out a study on bacteria isolated from stranded harbor seals over a 12-year period. e author reported that bacterial isolates exhibit resistance most frequently to ampicillin (74% resistant) and cephalothin (64% resistant). Antibioticresistant microbes and genes within the aquatic environment have been documented in various marine species from cephalopods to marine mammals and elasmobranchs, such as sharks [43][44][45][46][47].
Rose et al. [41] and Ahmed et al. [48] concluded that the presence of antibiotic-resistant bacteria from animals, and marine animals, in particular, indicates not only the widespread presence of the microbes but often a significant percentage of the bacteria demonstrating resistance to multiple antibiotics.
Oates et al. [49] reported that when wastewater discharge and runoff from agricultural activity, which carries antibiotics and resistant bacteria into terrestrial waterways, finds its way to marine coastlines, it may cause disease in marine organisms, contributing to antibiotic resistance. is, along with other anthropogenic contributions, may contribute to the elevation of natural background levels of antibiotic resistance genes in aquatic environments, encouraging their transfer into pathogens or serving as a means for antibiotic resistance propagation [44]. When resistant bacteria are introduced to animals or their environment, the animals may become sick, or resistance traits may be transferred to other bacterial species, or they may become a reservoir that transfers the bacteria and resistance back to humans and the environment [45]. e antibiotics used by the majority of the Tanzanian population are unregulated and used indiscriminately, and statistics are rarely collected, hence increasing the risk to the environment and human and animal health. erefore, due to the widespread acquisition of resistance, it is important that susceptibility tests be routinely performed to guide antibiotic treatment and policy.

Conclusion
In conclusion, this study demonstrated the presence of highly pathogenic agents such as Salmonella and Shigella species and opportunistic pathogens, and some carried antimicrobial resistance. eir presence is a potential health risk/hazard to human beings and may cause disease to susceptible individuals, especially immune-compromised consumers.
is study has therefore proven the need for the adoption of proper hygienic measures from farm to fork and hygienic education for fish handlers/traders and consumers.
In addition, bacterial isolates recovered from marine fish samples were more resistant to antibiotics than isolates recovered from freshwater fish samples. e environment might be a reservoir of resistance genes and dispersal vectors due to the influence of anthropogenic activities in marine environments. e resistance patterns observed in this study imply that some components of resistance are likely related to environmental origins and may spread without the selective pressure of antibiotic use.
ere is a need for research on antibiotic susceptibility surveillance in aquatic environments where fresh fish and marine fish are obtained for human consumption.

Data Availability
Data will be available on request. e author wishes to gratefully thank Mr. Mahona and Mr. Lazack of the chemistry and biology laboratory at the Open University of Tanzania for their assistance in the field and laboratory work. e author is also grateful to Dr. Alexander Mzula, Mr. Erick Oswald, and Mr. Mikidadi Abdallah, who assisted in the laboratory work at the College of Veterinary Medicine and Biomedical Sciences at the Sokoine University of Agriculture. e author also thank fish traders who participated in this study by offering the fish samples.