Pathogenic bacteria load and safety of retail marine fish

first-Abstract This study was conducted to determine the pathogenic bacteria load of 14 species of marine fish obtained from two suppliers (in Bitlis city, Turkey), which provide fish for fish markets, and to reveal the safety of the marine fish in terms of microbiological quality. The counts of total mesophilic aerobic bacteria (TMAB), Staphylococcus aureus , and Escherichia coli , the presence of Salmonella spp. and Listeria monocytogenes were determined in anchovy, horse mackerel, salmon, red mullet, gilthead seabream, bonito, pilchard, common sole, sand smelt, axillary seabream, seabass, Mediterranean horse mackerel, bluefish, and garpike. It was determined that common sole, axillary seabream, seabass, bluefish and Mediterranean horse mackerel obtained from both suppliers were unacceptable in terms of the counts of TMAB. Twenty-four samples exceeded the critical limit of S. aureus and all the samples were unacceptable according to the critical limit of E. coli . While L. monocytogenes was isolated from 50.0% of the samples, Salmonella spp. was isolated from 39.3% of the samples. These results showed that the pathogenic bacteria load of the analyzed marine fish was quite high and they were unsafe in terms of microbiological quality.


Introduction
Marine fish are important for the human diet due to its high nutrition value.However, it has also a great risk for pathogens in spite of this advantage.The number of pathogens is affected by microbial flora in the marine.Salt amount and temperature of the water, its pollution rate, catching methods and chilling situations are other factors that affect the microbial flora of products.Marine fish may contain many pathogens that cause foodborne diseases if they are caught from contaminated waters or people do not follow hygiene rules during its transportation and storage.Pathogenic bacteria in marine fish constitute majority of these pathogenic agents (Feldhusen, 2000).
There is no information from Bitlis, Turkey about the pathogenic bacteria load or how safe commercially-sold marine fish.Bitlis, which is an inland city, is located 20-24 hours away from all of the coastal cities (Istanbul, Samsun, Izmir and Iskenderun) which are the main suppliers of marine fish.If the cold chain damaged during transportation, it is inevitable that the existing microbial load in the marine fish will gradually increase and the hygienic quality of it will gradually deteriorate until it will reach to Bitlis.So, the aim of this study was to determine the load of some pathogenic bacteria (S. aureus, E. coli, Salmonella spp., L. monocytogenes) in 14 species of marine fish obtained from two suppliers, which provide fish for fish markets, and to reveal the safety of the marine fish in terms of microbiological quality.

Materials
Fourteen species of marine fish obtained from the two suppliers between January and February 2017 were used as the material.The suppliers in Bitlis transport the fish in the boxes containing crushed ice from Istanbul, Samsun, Izmir or Iskenderun to Bitlis for 20-24 hours.

Methods
To find out whether or not the samples contained any pathogenic bacteria, counts were done for each of the total mesophilic aerobic bacteria (TMAB), S. aureus, E. coli, Salmonella spp.and L. monocytogenes.Each analysis was carried out twice.Colonies were counted using a colony counter.

Sample preparation
Twenty-five gram samples was taken from each fish flesh and placed in sterile Stomacher bags (Seward Medical, London, UK).Two hundred twenty-five milliliters of 0.1% peptone water (Merck, 107228) was added onto each sample.The mixture was then homogenized using a 400 mL lab blender (Stomacher, degree contamination, secondary contamination occurs in processing, transportation, and marketing stages.Fish contain high moisture and protein and the pH values close to neutral; thus, microorganisms can develop in a very favorable media (Inal, 1992).
If the contaminated microorganism is a pathogen (disease-causing) and its count increases, it first causes economic losses due to the degradation of the product and then causes invasive infection at the consumer side.Moreover, if the pathogenic microorganism produces also toxin, more serious foodborne poisonings (intoxication and toxicoinfection) occur (Sahin and Basoglu, 2011).
Staphylococcus aureus causes intoxications by forming toxins (Kutlu et al., 2011) and the enterotoxin produced by S. aureus bacteria leads to Staphylococcal poisoning.Since the cooking process does not destroy enterotoxin, food poisoning symptoms occur in the person consuming the food (Durgac, 2006).Since the proteins and their amino acids in fish are disintegrated into peptides with low molecular weight, the growth of S. aureus is provided.S. aureus is not found in freshly caught seafood; contamination only takes place after the seafood has been caught.Workers carry enterotoxigenic S. aureus via skin infections, alongside their throats and noses, hence allowing it to easily reach its way into seafood.The contamination could be the cause of a combination of cross-contamination, unhygienic or incorrect processing and incorrect storage conditions (Simon and Sanjeev, 2007).
Escherichia coli, which is a typical host microorganism found in the human intestinal tract, may also be an agent causing intestinal infections (Ivnitski et al., 1999).Therefore, it is often used as the indicator of the fecal contamination.Escherichia coli is the part of the natural microflora in unpolluted warm tropical waters (Feldhusen, 2000).
Salmonella spp., which causes foodborne gastrointestinal diseases across the world, is one of the most important microorganisms (USDA, 2015).Fish caught from offshore seas are not contaminated with Salmonella spp.under normal circumstances.Salmonella spp.can be found especially in the fresh fish caught from waters in polluted coastal areas.The contamination of Salmonella spp. in the marine depends on various factors.One such factor is pollution from nearby poultry farms (Feldhusen, 2000).Salmonellosis based on Salmonella spp. is still an important infection that affects human life in many countries.Fifty-nine percent of Salmonellosis outbreaks could be traced to a specific food vehicle between 1973 and 1987 (Ivnitski et al., 1999).
In 1929, L. monocytogenes was deemed as a pathogen in humans.In 1981 it was also accepted as a foodborne pathogen (Farber and Losos, 1988).Listeria monocytogenes causes foodborne infections including diarrhea, bacteremia, meningitis, brain abscess, endocarditis, osteomyelitis or pneumonia.Listeriosis based on Listeria spp.has a high mortality rate although its prevalence is between 2-10 per million (Kilinc, 2001;Midi et al., 2005;Weinstein and Ortiz, 2004).The bacterial load of the environment affects the bacterial flora of the fish and Listeria species are contaminated by contaminated waters (Ben Embarek, 1994).

Enumeration of TMAB
Zero point one milliliter of each serial dilution was spread onto Plate Count Agar (PCA, Biomark B298), and incubated for 24-48 h at 37 °C.All colonies that developed afterwards were TMAB (Maturin and Peeler, 2001).

Enumeration of E. coli
Zero point one milliliter of each serial dilution was spread onto Violet Red Bile Agar Methylumbelliferyl-b-D-glucuronide (VRBA MUG, Biolife, 4021862), and then incubated for 18 h at 37 °C.After incubation, dark red colonies that were 1-2 mm in diameter were checked under a UV lamp and the fluorescent colonies from these colonies were evaluated as E. coli (Feng et al., 2020).

Enumeration of S. aureus
Zero point one milliliter of each serial dilution was spread onto Baird Parker Agar Base (Merck, 1.05406) containing Egg Yolk Tellurite Emulsion (Merck, 103785) and incubated for 48 h at 35 °C.Black glossy colonies with transparent zones that were 1-1.5 mm in diameter and developed after incubation were regarded as S. aureus (Tallent et al., 2016).

Isolation of Salmonella spp.
ISO 6579 was used to isolate Salmonella spp.First, a pre-enrichment culture was prepared.A twenty-five gram sample was homogenized in 225 mL of buffered peptone water (Merck 1.07228) and incubated for 16-20 hours at 35-37 °C for non-selective pre-enrichment purpose.After incubation, 0.1 mL and 10 mL of the pre-enrichment culture were transferred to 10 mL of Rappaport Vassiliadis Soy (RVS) Broth (Merck 1.07700) and 100 mL of Selenite Cystine (SC) Broth (Merck 1.07709), respectively, for selective enrichment.The RVS Broth was incubated for 24 hours at 42/43 °C, while the SC Broth was incubated for 24 hours at 37 °C.Later, the selective enrichment culture was streaked onto Brilliant Green Phenol Red Lactose Sucrose Agar (Merck 1.10747) as well as onto Xylose Lysine Tergitol-4 (XLT-4) Agar (Merck 1.13919) to which XLT-4 Supplement (Liofilchem, 80410) was added.Both were then left to incubate aerobically for 24 hours at 37 °C.To confirm pink-red colored suspicious colonies surrounded by a bright red zone on Brilliant Green Phenol Red Agar and black suspicious colonies on XLT-4 Agar were indeed Salmonella ssp., the culture was then streaked onto Triple Sugar Iron Agar (Merck 1.03915) and Lysine Iron Agar (Merck 1.11640) and incubated for 24 hours at 37 °C.It was then inoculated into Urea Broth (Merck 1.08483), and incubated again for 48 hours at 37 °C.Last, a Salmonella Latex Test Kit (Oxoid FT0203A) was used for serological confirmation (Andrews et al., 2016).

Isolation of L. monocytogenes
Listeria monocytogenes was isolated based on method of FDA's Bacteriological Analytical Manual (Hitchins et al., 2016).First, a pre-enrichment culture was prepared.A twenty-five gram sample was homogenized in 225 mL of Buffered Listeria Enrichment Broth (LAB, LAB138) and incubated at 30 °C for 4 hours.After incubation, Listeria Enrichment Selective Supplement (LAB, LABX139) was added and incubated at the same temperature for additional 44 hours.Afterwards, the selective enrichment culture was streaked onto Palcam Listeria-Selective Agar (Merck, 1.11755) containing Listeria Palcam Antimicrobic Supplement (Biolife, 4240042) as well as Oxford Agar (Merck, 1.07004) containing Oxford Selective Supplement (Merck, 1.07006).In order to confirm that suspicious colonies growing on Palcam Listeria-Selective Agar, being olive green-gray colored, having sometimes black centers but always with black zones and suspicious blackish green brown colonies with black zones and sunken centers growing on Oxford Agar, were indeed L. monocytogenes, colonies were inoculated on Tryptone Soya Agar (Oxoid, CM013B) containing Yeast Extract (Merck, 1.03753) and incubated at 30-37 °C for 24-48 hours (Hitchins et al., 2016).
Suspected isolates which matched to all identification parameters (Gram staining, catalase activity, motility test, fermentation of maltose, rhamnose, mannitol, and xylose, hydrolyzation of esculin, reduction of nitrate) according to reference method were evaluated as L. monocytogenes CAMP, S. aureus and Rhodococcus equi tests were also applied to all suspected samples (Hitchins et al., 2016).

Counting TMAB, E. coli and S. aureus, and indicating the presence/the absence of Salmonella spp. and L. monocytogenes
The TMAB, E. coli, and S. aureus counts were indicated as a logarithm of colony-forming units per gram (log cfu/g) of the sample according to the number of colonies, dilution factor and cultivation amount (Bell et al., 2005).The results for Salmonella spp.and L. monocytogenes were expressed as present (+) / absent (-) in 25 g (European Union, 2005).

Statistical analysis
The data was analyzed using Statistical Package for the Social Sciences (IBM SPSS Statistics, Version 25.0).The results were expressed in mean ± standard deviation.The one-way analysis of variance was used to establish whether or not there was any difference between means according to microbial load.As the data were parametric and homogeneity of variance was provided, the Tukey test was conducted to find the sources of the differences between the groups.A P value of < 0.05 was accepted as a significant difference for each fish (Sumbuloglu and Sumbuloglu, 2002).

Results
Table 1 shows the variations in the microbiological load of marine fish obtained from two suppliers.The count of TMAB is the most important analysis method for the overall microbiological quality of the seafood.The highest number of TMAB (6.78 ± 0.01 log cfu/g) was found in P. saltatrix obtained from the first supplier (P<0.05)(Table 1).
The difference between the majority of the tested fish species in terms of the count of TMAB was significant (P<0.05).But, considering counts of TMAB in the some fish species, S. salar from the first supplier, S. pilchardus from the second supplier, A. hepsetus from the second supplier and T. mediterraneus from the second supplier; T. trachurus from the first supplier and S. salar from the second supplier; M. barbatus from the second supplier and S. sarda from the second supplier; S. pilchardus from the first supplier and A. hepsetus from the first supplier; P. lascaris from the first and second supplier were similar (P>0.05)(Table 1).
The highest number of S. aureus (4.42 ± 0.01 log cfu/g) was determined in P. saltatrix obtained from the first supplier (P<0.05)(Table 1).The difference between the majority of the tested fish species in terms of the count of S. aureus was significant (P<0.05).But, considering counts of S. aureus in the some fish species, T. trachurus from the first supplier and A. hepsetus from the first supplier; S. salar from the first and second supplier; M. barbatus from the second supplier and S. sarda from the second supplier; S. pilchardus from the first supplier and D. labrax from the first supplier; P lascaris from the first supplier and B. belone from the first supplier; T. trachurus from the second supplier, S. aurata from the second supplier and P. saltatrix from the second supplier; S. aurata from the second supplier and S. pilchardus from the second supplier; S. aurata from the first supplier and S. sarda from the first supplier; T. trachurus from the second supplier, P. lascaris from the second supplier, D. labrax from the first supplier and P. saltatrix from the second supplier; S. sarda from the first supplier and D. labrax from the second supplier were similar (P>0.05)(Table 1).
The highest number of E. coli (5.52 ± 0.01 log cfu/g) was determined in P. lascaris obtained from the first supplier (P<0.05)(Table 1).
The difference between the majority of the tested fish species in terms of the count of E. coli was significant (P<0.05).But, considering counts of E. coli in the some fish species, S. aurata from the second supplier, S. pilchardus from the first supplier, A. hepsetus from the first supplier and P. acarne from the second supplier; M. barbatus from the second supplier, S. sarda from the first supplier and T. mediterraneus from the second supplier; S. aurata from the first supplier, B. belone from the first supplier and B. belone from the second supplier; S. sarda from the second supplier and P. acarne from the first supplier; P. lascaris from the first supplier and P. saltatrix from the first supplier; S. sarda from the first supplier and P. lascaris from the second supplier; S. sarda from the second supplier, S. pilchardus from the first supplier, A. hepsetus from the first supplier and P. saltatrix from the second supplier; S. aurata from the second supplier and T. mediterraneus from the first supplier; P. acarne from the second supplier and T. mediterraneus from the first supplier were similar (P>0.05)(Table 1).
While Salmonella spp. was isolated from 39.3% of the samples, L. monocytogenes was isolated from 50.0% of the samples (Table 1).

Discussion
The Turkish Food Codex Regulation on Microbiological Criteria (Turkey, 2011) states no limit values for the microbiological values of fresh chilled fish, but only the limit value for histamine level.For the count of total TMAB, Turkish Seafood Regulation (Turkey, 1995) states the critical limit as 6-7 log cfu/g for frozen fish.No sample exceeded the limit of 7 log cfu/g in the present study.
For newly caught fish and fish products, the total bacterial load of 2.00-6.00log cfu/g is considered as normal and most of the consumer safety standards state that, the total bacterial load of 6.00 log cfu/g is acceptable (Olafsdottir et al., 1997).Pegusa lascaris obtained from both suppliers as well as D. labrax, P. saltatrix, T. mediterraneus obtained from the first supplier exceeded the limit of 6.00 log cfu/g (Table 1).
The count of TMAB was found to be 3.14 ± 0.01 log cfu/g in fresh bonito (S. sarda) samples and 5.37 ± 0.02 log cfu/g in fresh anchovy (E.encrasicholus) samples in January 2015 (Corapci, 2018).These values are quite different from the values of the same fish species detected in the current study (Table 1).
The count of TMAB in E. encrasicholus obtained from the second supplier (Table 1) is compatible with a previous study reporting that the average count of TMAB was 3.10 ± 1.10-3.42± 1.96 log cfu/g in anchovy (E.encrasicholus) samples sold at fish markets in three regions of Istanbul, Turkey (Mol and Tosun, 2011).The average count of TMAB was 3.36 ± 1.22-4.04± 1.21 log cfu/g in horse mackerel (T.trachurus) in the same study (Mol and Tosun, 2011); whereas, it was 4.62 ± 0.01-4.81± 0.01 log cfu/g in the current study (Table 1).Bektas (2013) determined that the count of TMAB in gilthead seabream (S. aurata) samples taken from 4 different retail sale points in Isparta, Turkey was 3.000 ± 0.685-5.810± 0.469 log cfu/g for 4 weeks.Mol and Tosun Brazilian Journal of Biology, 2022, vol.82, e262735 5/9 Pathogenic bacteria load of marine fish (2011) found that the count of TMAB was 3.97 ± 1.16-5.74± 0.42 log cfu/g in seabass (D. labrax) and 3.72 ± 0.21-5.68± 0.31 log cfu/g in gilthead seabream (S. aurata).In the current study, the count of TMAB was 4.37 ± 0.01 log cfu/g and 5.25 ± 0.01 log cfu/g in S. aurata and 6.07 ± 0.01 log cfu/g and 5.93 ± 0.01 log cfu/g in D. labrax (Table 1).Papadopoulos et al. (2003) found that the initial mesophylic bacteria load of fresh seabass (D. labrax) was 4 log cfu/g and evaluated the fish quality as "good".This level was well below the TMAB value of the seabass (D. labrax) samples in the current study (Table 1).Kaba et al. (2013) determined the count of TMAB as 4.45 ± 0.00 cfu/g (0.65 ± 0.00 log cfu/g) in fresh bonito (S. sarda) samples analyzed before making the ball from smoked bonitos.The count of TMAB was 4.48 ± 0.29 log cfu/g in fresh bonito (S. sarda) sold at the fish market in Giresun, Turkey, between September and December 2015 and stored at 4 °C (Kulcu, 2017).The count of TMAB was found at high Ogur, S. levels (5.16 ± 0.01 log cfu/g and 5.80 ± 0.01 log cfu/g) in S. sarda in the current study (P<0.05)(Table 1).
Total viable count of fresh Mediterranean horse mackerel (T.mediterraneus) caught in the gulf of Chalkidiki (North Greece) found to be 4.09 log cfu/g and 3.97 log cfu/g, in December and in January, respectively (Tzikas et al., 2007).The count of TMAB contained in the same fish species was higher than this value in the current study (5.35 ± 0.01 log cfu/g and 6.53 ± 0.01 log cfu/g) (Table 1).
Many people carry S. aureus in their face, nose, and mouth.It is an opportunistic pathogenic bacterium which causes many infections in humans, especially with a weak immune system.S. aureus contaminates the hands, face and nose and then the seafood during the processes (catching, placement in the boxes, transporting, selling, etc.).Then, it grows and produces enterotoxin until the seafood becomes ready to eat (Kocatepe et al., 2013).
The presence of S. aureus is not seen in the native flora of marine fish caught from the unpolluted marine (Hernandez-Herrero et al., 1999).Some infections may occur when S. aureus, which is transmitted from workers to food and transmitted to consumers from food products prepared or stored under bad conditions, exceeds 5 log cfu/g (Varnam and Evans, 1991).
The count of S. aureus in frozen fish should not be more than the level of 3 log cfu/g (Turkey, 1995).The limit of S. aureus in fresh fish is also accepted as 1000 cfu/g (3 log cfu/g) by ICMSF (1986).Twenty-four samples exceeded this critical limit, except for E. encrasicolus obtained from the both suppliers as well as P. lascaris, B. belone obtained from the first supplier (Table 1).
The vegetative cells of S. aureus can be inactivated with heat treatment done for 2-50 min at 60 °C (Ash, 1997).However, the enterotoxin produced by S. aureus is heat resistant and therefore does not disintegrate when cooked (Azanza et al., 2001).A heat treatment at higher temperatures is required to inactive staphylococcal enterotoxin.The norms of heat treatment vary from several hours at 80-100 °C to 5-10 min at 121 °C depending on the suspending medium and toxin type (Mossel et al., 1995).
Staphylococcal poisoning may occur with S. aureus bacteria producing enterotoxin.It was stated that the number of pathogens should reach 10 5 cells/g for S. aureus to form enterotoxins (Aytaç and Taban, 2011).Therefore, one can deduce that none of the fish samples in the study was risky in terms of enterotoxins -meaning that S. aureus could form (Table 1).
Escherichia coli bacteria were not found in gilthead seabream (S. aurata) samples sold at four points of sale (Bektas, 2013).The count of E. coli was found to be 3.47 ± 0.01 log cfu/g and 4.42 ± 0.01 log cfu/g in S. aurata and 4.18 ± 0.01 log cfu/g and 4.51 ± 0.01 log cfu/g in D. labrax in the current study (P<0.05)(Table 1).
Escherichia coli is accepted as a hygiene indicator which shows whether or not fecal contamination is found in the product.And furthermore, it is an agent of the dangerous foodborne diseases (Sahin and Basoglu, 2011).The count of E. coli in frozen fish should not be higher than 0.95-1.08log cfu/g (Turkey, 1995).But, the samples were approximately 2-3 times higher, except for E. encrasicolus obtained from the first supplier (Table 1).
The standard recommended by ICMSF (ICMSF, 1986) is <3.0 MPN/g for E. coli in fish.According to this determination, all samples were unacceptable in the current study (Table 1).
Listeria monocytogenes and Salmonella spp.both should be not found in 25 g of fish (Turkey, 1995;ICMSF, 1986).Bektas (2013) determined the average count of Salmonella spp. in gilthead seabream (S. aurata) samples sold at the two points of sale between 3.477 ± 0.733 log cfu/g and 5.810 ± 0.595 log cfu/g, but Salmonella spp. was not found in gilthead seabream (S. aurata) samples sold at the other two points of sale.In the current study, the development of Salmonella spp. was not also observed in S. aurata, P. acarne, D. labrax, P. saltatrix, T. mediterraneus and B. belone obtained from both suppliers (Table 1).
Although Salmonella isn't psychrotrophic or native in the aquatic environment, it has been isolated from fish, sometimes (Youssef et al., 1992).If fish or other seafood is caught from fecally contaminated water, Salmonella spp. is found naturally in these foods (Banwart, 1981).The cause behind the presence of Salmonella spp. in some fish may be human-based contamination due to unhygienic applications (use of ice obtained from contaminated water for chilling fish, dirty boxes where fish are kept, inadequate hand hygiene of personnel, etc.).Varnam and Evans (1991) reported that a minimum internal temperature of 74 °C is required for disintegrating Salmonella spp. in foods with high a w like analyzed fish samples.
Listeria monocytogenes was determined in S. salar, M. barbatus, S. aurata and T. mediterraneus obtained from both suppliers (Table 1).
The incidence of L. monocytogenes in some aquatic products has been reported in different researches (Rodassuarez et al., 2006;Rodrigues et al., 2015).Ben Embarek (1994) stated that the prevalence of L. monocytogenes in fresh fish and seafood varied from 4-12% in temperate areas and was lower (0-2%) in tropical areas.Fifty percent prevalence for L. monocytogenes (Table 1) determined in the current study was not similar to Ben Embarek (1994)'s study.
The presence of Listeria spp. was not observed in anchovy (E.encrasicholus) (Erdem et al., 2010).Listeria monocytogenes was not detected in E. encrasicholus obtained from the second supplier, but it was isolated in E. encrasicholus sample obtained from the first supplier in the current

Table 1 .
Microbial Load of Retail Marine Fish Lower case letters indicate the difference between lines in the same column (p<0.05).The difference between the mean values indicated by the same letter is insignificant (p>0.05).+It indicate the presence of the microorganism in 25 g sample.-It indicate the absence of the microorganism in 25 g sample.