The Prevalence of Bacteria Commonly Related to the Production of Mussels and Oysters in Saldanha Bay

,


Introduction
Shellfsh farming is becoming an important sector for the South African government as it creates much-needed job opportunities for the coastal communities.Te sustainability and safety of shellfsh growing areas are essential in terms of protection from contaminants and preventing contaminants from reaching the bivalves produced [1].Bivalve molluscs play important ecological functions in aquatic ecosystems as well as being highly nutritious.Tey can be found at the bottom of the sea or attached to hard surfaces or on one another.Teir flter-feeding nature assists in purifying the surrounding waters and increases the penetration of sunlight [2].Furthermore, they provide micronutrients to other marine organisms increasing primary production and nutrient recycling, coastal habitat conservation, and restoration [3,4].Tese characteristics and ability to bioaccumulate materials in their soft tissues make the bivalve molluscs suitable aquatic species for biomonitoring of environmental conditions.Bioaccumulation of materials is not selective, as both benefcial and harmful materials are equally accrued [5].Besides being highly nutritious compared to beef, chicken, and pork, bivalves are highly perishable and require proper handling from farm to fork.Failure to adhere to food safety best practices could lead to an increased risk of illness from pathogens, including bacteria, viruses, and protozoa [6,7].
Aquatic environments are home to various microbiota, some indigenous and some introduced through anthropogenic activities around these environments.Te presence of pathogens in aquatic ecosystems is a risk to the shellfsh production industry and poses a public health threat.Several foodborne outbreaks have been reported globally due to the consumption of contaminated shellfsh [8][9][10].Zgouridou et al. [11] in their study indicated that mussels of the genus Mytilus are primarily the bivalve species that are a public health risk to consumers, as well as oysters (Ostrea edulis) and clams (Venus verrucosa).Bioaccumulation and bioconcentration of pathogens vary according to host species and seasonality.Both mussels and oysters can concentrate pathogens in their body tissues.However, oysters are an important medium for infecting humans with these pathogens as they are eaten raw or partially cooked [12].Several studies have been conducted worldwide to determine the bacterial prevalence in shellfsh-growing waters [13][14][15].To date, similar studies have not been conducted in Saldanha Bay except for the microbiological monitoring undertaken by the South African Department of Forestry, Fisheries, and Environment.Tis created a need to investigate the bacterial communities, especially the disease-causing ones that may be present in this Bay.Tis study investigated pathogens commonly associated with shellfsh-related foodborne disease outbreaks, such as Salmonella, Vibrio parahaemolyticus, Vibrio vulnifcus, Vibrio cholera, and the prevalence of Escherichia coli as an indicator species.Te results did not conform to prior expectations, as bacteria such as the Enterobacter cloacae complex, Citrobacter freundii, Klebsiella pneumoniae spp.pneumoniae, Aeromonas sobria, Vibrio alginolyticus, and Sphingomonas paucimobilis were confrmed through biochemical characterisation.
Te study applied an experimental design, and interpretations were formed using a multimethod, quantitative strategy for three sampling occasions.Te researcher collected samples during warm, cold, and rainy periods as informed by the obtained literature.Data collection and analysis techniques are detailed under materials and methods [16].

Study Area. Saldanha Bay harbour on the West Coast of
South Africa (latitude: −33.027699, longitude: 17.917631) houses the biggest port in Southern Africa operating as an international port for the export of iron ore.Te Bay's water depth is approximately 23.7 m.Construction of a 4 km long iron ore jetty has divided the Inner bay into Small Bay and Big Bay [17], and a 1.7 km long breakwater separated the Inner Bay from the Outer Bay.Small Bay is sheltered from ofshore swells and has constrained water circulation, while Big Bay is semiexposed to wave energy with better circulation compared to Small Bay.Te Outer Bay, which is located at the mouth of the Bay, is regarded as the less polluted site [18].Te Bay is exposed to the disposal of treated and untreated sewage from the nearby wastewater treatment plant, which discharges into the Bok river.Several sewage pumps, ballast water, dredging, stormwater discharge, and ship trafc are some of the pollutants sources close to Small Bay.Two mussels species are farmed, the indigenous black mussels (Choromytilus meridionalis), which are not a preferred species for farming due to the dark fesh colour of the female species.Te exotic Mediterranean mussels (Mytilus galloprovincialis) and the Pacifc oysters (Crassostrea gigas) are also farmed.Figure 1 shows fve sampling points where mussels and seawater were collected.

Sample Collection.
A total of 27 shellfsh and seawater (mussels (n � 12) and seawater (n � 13) and oysters (n � 2)) samples were collected from various sampling sites.Samples were collected in the morning between 8:00 am and 11:00 am during low tides in order to reach all sites especially the ofshore ones.Oysters were collected from the harbour deck immediately after harvesting by the farmers.Five sampling points were used for the collection of seawater and mussels.Tree of them were located in Small Bay (SP1, SP2, and SP3), one in Big Bay (SP4), and the last one in Outer Bay (SP5) (Figure 1).Samples were collected in March (warm period), July (winter-before heavy rainfall), and August (winter-after heavy rainfall).During sampling, 30 oysters and 30 mussels were hand-picked and stored in sterile whirl-pack bags (Nasco, US).Seawater samples were collected (2 meters below the surface) in 1 liter sterile Schott bottles (Schott, UK) and mussels from a hanging rope.Physicochemical parameters (i.e., water temperature ( °C), salinity (psu), and dissolved oxygen (ppm)) were measured during sampling at each sampling point using a Hanna HI9810-6 multimeter.Samples were transported to the laboratory, in a cooler box packed with ice packs maintaining a temperature between 2 and 8 °C, within 2 hours, and microbiological analyses were performed immediately.

Sample Preparation.
Upon arrival at the laboratory, the mussel and oyster samples were scrubbed under running tap water to remove shell debris, and attached algae and the shells were opened aseptically with a sterile chucking knife.Approximately, 300 g of fesh and intravalvular liquid of mussels and oysters were stored in 500 g sterile beakers and then transferred aseptically into stomacher bags (circulator 400, Seward, Worthing, UK).Samples were homogenised with 200 ml sterile phosphate water at 230 rpm speed for 2 minutes.

Most Probable Number (MPN) (Mussel, Oyster, and
Seawater Samples), Escherichia coli.Lauryl Tryptose Broth (LTB) (Merck, Germany), Brilliant Green Broth (BGBB) (Merck, Germany), and Tryptone Water (TW) (Merck, Germany) were prepared following the manufacturer's instructions.Te method was conducted using the method described by Leuta [19] Concentrated mussel and oyster homogenate extracted from the mussel samples was used as stock to conduct the fve-tube MPN technique.Serial dilutions of 10 −1 to 10 −5 of the mussel and oyster homogenate and seawater samples, respectively, was performed before inoculation of 1 ml of each diluted sample into LTB tubes containing Durham tubes.Durham tubes provide a visual indication of gas 2 Aquaculture Research production.Te inoculated test tubes were incubated for 48 hours at 37 °C (indicating all gas-producing organisms).
All tubes showing gas formation after a 48-hour incubation period were regarded as a positive presumptive test, and the presumptive total MPN count was read of De Man's tables [20].For each positive presumptive LTB tube, a 10 ml Brilliant Green Bile Broth (BGBB) tube and 10 ml Tryptone Water (TW) tubes were prepared.One hundred microliters (μl) of the sample from each positive LTB tube were reinoculated into BGBB and TW tubes, respectively, according to the guidelines set out by the South African Bureau of Standards [21].Tese guidelines also incorporate the standard methods set out by the American Public Health Association, for the examination of seawater and shellfsh as well as the methods for the examination of water and wastewater (American Society for [22]; American Society for [23].Tese tubes were incubated in a 44.5 °C waterbath for 24 hours (44 °C-44.5 °C) and has the specifc advantage of detecting E. coli, as it is the only faecal coliform present in water capable of producing indole at this temperature).With the observation of positive gas production in the BGBB tubes (indicating faecal coliforms (FC)) after 24 hours, a few drops of Ehrlich's reagent (LabChem, USA) were added to the corresponding TW tubes.Te presence of E. coli was confrmed with a colour change from clear to pink or red in the Tryptone Water tubes.

Detection and Isolation of Salmonella. Salmonella spp.
was detected according to the protocol based on ISO 6579-1:2007 [24].Bufered Peptone Water (BPW) (Merck, Germany), Selenite Cysteine broth (SCB) (Merck, Germany), and Salmonella shigella agar (SS agar) (Oxoid, UK) were prepared according to the manufacturer's instructions.Twenty-fve grams (25 g) of mussels and oysters homogenate were aseptically weighed and placed into a 225 ml sterile BPW to prepare a pre-enrichment culture.
Te mixture was then incubated at 37 °C for 16-20 h.After incubation, the sample was gently mixed, and 1 ml of the BPW was added into a sterile McCartney bottle.Ten milliliters (10 ml) of SCB were added to the sample to prepare an enrichment culture and incubated at 37 °C for 24 h.After the 24 h incubation period, the enriched culture was streaked onto SS agar and incubated inverted at 37 °C for 24 h.Te plates were examined (typical pinkish-red colonies) for the absence or presence of Salmonella spp.Subsequently, Gram stains were performed on the obtained colonies.Te observation of Gram-negative, nonsporeforming colonies confrmed the presence of Salmonella, while biochemical identifcation was carried out using VITEK 2 compact Gram-negative (GN) ID cards (bio Mérieux, France).Salmonella typhimurium (NCTC 12023) strain was used as a positive control.

Statistical Analysis.
A Pearson correlation was conducted to determine the relationship between seawater samples, shellfsh samples, and physicochemical parameters (temperature, salinity, and dissolved oxygen).For all the tests, the criterion for statistical signifcance was p < 0.05.

Results and Discussion
3.1.Physicochemical Parameters.Physicochemical parameters recorded in Table 1 did not show a signifcant variation, where the recorded temperature ranged between 12 °C and 19 °C, salinity ranged between 33.91 psu to 35.45 psu, and dissolved oxygen between 0.71 and 2.96 ppm showing prevailing hypoxic conditions, which are often associated with pollution due to anthropogenic activities [26].Seawater temperature, salinity, pH, dissolved oxygen, turbidity, and organic matter become water quality stressors when available in excessive amounts [27].Tese stressors may infuence the survival, health, and growth of shellfsh, which depend on the water quality of their growing environments.Poor water quality increases the risk of shellfsh contamination with disease-causing pathogens [28].
An increase in salinity during warm periods and a decrease during cold periods were observed throughout the study and correlated with similar fndings reported by Lamine et al. [29].Te lowest rainfall was observed in July and the highest rainfall in August.Colaiuda et al. [30] found in their study that the amount of rainfall and the increased E. coli concentrations in shellfsh depend on the specifc area where the samples were collected.Chahouri et al. [31] and Padovan et al. [32] found that high precipitation increases levels of faecal coliform.In this study, no clear indication of the infuence of rainfall on E. coli levels was detected.Similar Figure 2: (a) Inoculation of undiluted and diluted samples into LTB tubes to obtain the positive presumptive test (adopted and adapted from [19]).(b) Reinoculation of positive LTB tubes into BGBB and TW tubes (adopted and adapted from [19]).(c) Enumeration of faecal coliforms (from BGBB tubes) and E. coli (from TW tubes) in water samples (adopted and adapted from [19]).
Aquaculture Research results were observed by Sampson et al. [33], where no association was found between precipitation and bacterial concentrations.Tabanelli et al. [34] included the infuence of the fow rate of the river feeding into the coastal area of their study and concluded that meteorological events could bring a substantial amount of contaminated fresh water into coastal water.Tis could be the case with the Bok river, which feeds into Saldanha Bay.During heavy rainfall, the Bok river fow rate increase is suspected, which could wash down all the runof from upstream agricultural areas and runof from roads and residential areas [31].

Prevalence of Faecal Coliforms and Escherichia coli in
Mussels and Oysters.Oyster harvesting did not take place during the March sampling occasion.Sampling sites SP4 (during March and July) and SP5 (during March) could not be reached due to high tides.In addition, no mussels were available during August at the SP4 site (Table 2).Te total MPN count per 100 ml of mussel samples were between 4.9 and 4700 microorganisms/100 ml and for oysters were 18 and 1000 microorganisms/100 ml in (July and August), respectively.Increased total MPN counts of 400 microorganisms/100 ml were observed in mussel samples collected at SP1 in July.Of the recorded total MPN count at this site, FC and E. coli counts of <0.18 microorganisms/100 ml, respectively, were observed.In the August sampling run, mussels collected at SP2 recorded a total MPN count of 4700 microorganisms/100 ml, while a total MPN count of 1000 microorganisms/100 ml) were recorded in oyster samples at the Harbour Deck site.In comparison, the FC and E. coli concentrations at these respective sites were 0.2 microorganisms/100 ml, respectively (mussels), and <0.18 microorganisms/100 ml, respectively (oysters).Mussels and oysters can accumulate and retain suspended particles of phytoplankton size and pathogenic microorganisms in their bodies due to their flter-feeding nature [35,36].Tis creates a public health concern, especially for oysters, as oysters are consumed raw or partially cooked [37,38].Te spikes observed in mussels and oysters suggest possible contamination due to heavy rainfall or pollution sources including the sewage pump stations, stormwater drains, and a sewage discharge point that is located in close proximity to the afected sampling sites [39,40].Saldanha Bay Municipality recently made remarkable improvements to their sewage treatment plants and diverted the majority of treated efuent for the irrigation of sports grounds and use by interested local businesses.However, the little that is being discharged together with efuent from fsh factory industries, untreated stormwater discharge, and ballast water should not be underestimated.According to Clark et al. [41], the shipping trafc has increased in the harbour, which brings large volumes of ballast discharge.All of these need to be monitored closely.Several studies in various parts of the world seem to agree on the fact that microbial contaminants are the results of treated and untreated sewage being discharged into shellfsh growing waters, sewage overfow during rainfall periods, and runof from agricultural areas [42,43].Sewage is loaded with nutrients that, in excessive amounts, could stimulate microbial growth, production of harmful algal blooms, and eutrophication, ultimately afecting the viability of shellfsh mariculture [44].Even though oyster samples were not taken from the farm but at the loading area of the harbour, i.e., the Harbour Deck, the samples came from the same farming area as the mussels.

Prevalence of Faecal Coliforms and Escherichia coli in
Seawater.Sampling site SP5 could not be reached due to high tides in March and July (Table 3).Te total MPN count/ 100 ml in seawater ranged from <0.18 to 1.3 microorganisms/100 ml, with a high spike recorded at SP2 in August (2400 microorganisms/100 ml).Faecal coliforms and E. coli concentrations were the same (<0.18microorganisms/ 100 ml) at all sampling sites.Te high increase in the total MPN count observed in SP2 in the seawater sample correlates with a spike in mussels collected during the same period.Tis could be attributed to heavy rainfall, stormwater drain discharges and sewage discharges, and the location and proximity of the sampling site to pollution sources.Sampling site SP2 is located in Small Bay, which is subjected to various sources of pollution including a sewage discharge outfall.Understanding the causes of faecal contamination in areas where shellfsh are grown is essential for assessing the associated health risks and determining the way forward to address the problem [45].
During high tide episodes, pollutants can be transported rapidly from the areas where they are highly concentrated through advection, mixing, dispersion, and dilution of sewage [2].Te sampling sites in Small Bay, sheltered from the sea swells and close to the sewage discharge point, sewage pump stations, and stormwater drains may not beneft from this natural process and therefore presented higher contamination levels.Tese natural processes are also evident in the analysis results of the Big Bay and Outer Bay sampling sites where lower contamination levels were observed.Both sites are semiexposed to the sea swells, explaining the relative improvement in water quality.In other words, the possibility of having shellfsh farms far away from sewage discharge points could eliminate the microbial contamination problem.Similarly, Florini et al. [45] reported a decrease in the concentrations of faecal indicator species with an increase in distance from sewage discharge points.Te low concentration results were ascribed to possible dilution and die-of efects.
Contamination of water bodies by wastewater is a fundamental problem worldwide.Te bacteria, parasites, and viruses from animals and humans reach the oceans through runof from roads, agricultural areas, and sewage discharges [46].In addition, heavy rainfall may cause sewage overfows and drain leakages [47].As mentioned, faecal coliform and Escherichia coli are indicators of water quality.Te presence of these organisms is undesirable in areas used for shellfsh farming.
No correlation could be drawn between the total MPN count in water (microorganisms/100 ml) and shellfsh (microorganisms/100 g) samples, physicochemical parameters, as well as between rainfall patterns and MPN counts in water and shellfsh (p > 0.05) (Table 4).As the total MPN count in water samples increased, the total MPN count in shellfsh samples increased (r � 0.997, n � 11, p ≤ 0.001).

Bacterial Species Isolated from Selected Sample Sites.
Salmonella spp., Vibrio cholerae, and Vibrio parahaemolyticus were not detected.Bacterial species identifed included the Enterobacter cloacae complex, Citrobacter freundii, Klebsiella pneumoniae spp.pneumoniae, Aeromonas sobria, Vibrio alginolyticus, and Sphingomonas paucimobilis (Table 5).Tese microorganisms may be grouped into pathogens that are often present in aquatic Aquaculture Research environments (e.g., Klebsiella pneumoniae spp., Aeromonas sobria, Vibrio alginolyticus, and Sphingomonas paucimobilis).Tese microorganisms are pathogens of high priority as some are antimicrobial resistant and may cause illnesses in humans.In addition, pathogens naturally present in human beings and animals (e.g., Citrobacter freundii and Enterobacter cloacae complex) are also high priority pathogens and their presence should not be taken lightly [48].Te Enterobacter cloacae complex has also proved to be abundant in aquatic environments [49].

Conclusion
Te study used conventional culture methods to isolate Salmonella and Vibrio spp.in mussels, oysters, and seawater samples obtained from the Saldanha Bay Harbour.Te most probable number (MPN) analysis technique was used for detecting and enumerating faecal coliforms and E. coli in the obtained samples.Te identifcation of species was conducted using the VITEK 2 automated system, which successfully identifed species such as Enterobacter cloacae complex, Citrobacter freundii, Klebsiella pneumoniae spp.pneumoniae, Aeromonas sobria, Vibrio alginolyticus, and Sphingomonas paucimobilis.Te fndings observed with correlations between MPN counts in seawater, mussels, and oysters, and correlations between physicochemical parameters and rainfall, did not show any signifcant diference.However, total MPN count spikes were observed in mussels, oysters, and seawater, which could be ascribed to the rainfall period and winter season, although the spikes did not have a signifcant impact on the E. coli concentrations, as the concentrations were below the permissible limits.Tis information may be used as a basis to conduct an indepth investigation of sources of pollutants.Further studies need to be conducted on the bacterial species

Table 1 :
Physicochemical parameters of shellfsh production areas and rainfall in Saldanha Bay.

Table 2 :
Faecal coliforms and Escherichia coli in mussels and oysters homogenate.

Table 3 :
Prevalence of faecal coliforms and Escherichia coli in seawater samples.

Table 5 :
Bacterial species isolated from mussels and oysters sampling points.