Quality Attributes, Chemical and Microbial Safety of Street-Vended Smoked West Africa Ilisha (Ilisha africana) from Major Markets in Ibadan, Oyo State, Nigeria

This study was carried out to assess the quality and safety of street-vended smoked West African Ilisha fish from major markets in Ibadan, Oyo State, Nigeria. Fifty smoked fish samples were collected from ten major markets in Ibadan, Oyo State, Nigeria and the samples were analyzed in the laboratory for proximate, quality and rancidity indices (PV, TBA, TVB-N, TMA and FFA), polycyclic aromatic hydrocarbon (PAH), heavy metals and microbiological (coliform, fungi, Listeria monocytogenes, Salmonella paratyphi and Staphylococcus aureus) analyses. The results revealed that the mean moisture, protein, fat, crude fibre, ash and carbohydrate contents (%) of smoked fish samples were 11.35-15.18, 52.13-60.22, 11.56-16.13, 1.74-2.52, 1.12-1.54 and 9.53-16.25 respectively. The values of PV, TBA, TVB-N, TMA and FFA of smoked fish samples were 8.96-9.18 meq.peroxide/kg, 1.13-1.62, 1.04-1.16 mg Mol/kg, 17.29-19.36 mgN/kg and 2.15-2.68 mgN/kg respectively. The results also revealed the presence of sixteen (16) PAHs compounds in the smoke fish samples. However, ten of the sixteen PAHs compounds were above the 5.0 μg /kg B(α)P maximum permissible levels set by EU Regulation 1881/2006. The concentrations of Hg, Pb, Cd, and Cr in the smoked fish samples showed that quantities detected were generally below the maximum permissible levels and hence pose no risk to consumers. The results of microbiological analysis indicated the presence of Staphyloccoccus aureus in the smoked fish samples which may be due to post-processing handling. In conclusion, the study revealed that consumption of street-vended smoked fish may expose consumers to chemical and microbiological risks.


Introduction
Fish and fishery products constitute an important food component for a large section of world population, more so in developing countries, where fish forms a cheap source of protein (Amusan et al., 2010;Odu and Imaku, 2013). In the last two decades there has been an increase in awareness about the nutritional and health benefits of fish consumption (Amusan et al., 2010, Odu andImaku, 2013). The low fat content of some fish and the presence of polyunsaturated fatty acids in red meat fishes which are known to reduce the risks of coronary heart diseases, have increased the dietary and health significance of seafood consumption (Amusan et al., 2010;Odu and Imaku, 2013). The Food and Agriculture Organization (1994) asserted that fish contributes about 60% of the world's supply of protein and that 60% of the developing world derives more than 30% of their annual protein from fish (Amusan et al., 2010). However, in Nigeria, fish constitute 40% of the animal protein intake (Olatunde, 1998;Amusan, et al., 2010). They are prone to contamination at various stages of handling and processing and the quality is a major concern to food processors and public health authorities (Oramadike, et al., 2010;Amusan, et al., 2010;Odu and Imaku, 2013). The smoked fish has become so popular in Nigerian dishes and there is need for corresponding concern for safety issues in smoked fish consumption (Riches, 2012;Adeyeye et al., 2015). Although several researchers have worked on quality and safety of smoked fish, there is drought of data on the quality and safety of street-vended smoked fish in Nigeria. Da Silva et al., (2008) examined the microbial safety and quality of smoked blue catfish (Ictalurus furcatus) steaks treated with antimicrobials and antioxidants during 6 weeks ambient storage. Fafioye et al., (2002) studied the fungal infestation of five traditionally smoked dried freshwater fish in Ago-Iwoye, Nigeria and isolated and identified eleven different fungal species of which Aspergillus flavus was the most frequently encountered fungi on the fish species. Adebayo-Tayo et al., (2008) reported the presence of aflatoxin and other metabolites in smoked fish due to Aspergilus flavus in smoked fish sold in Uyo, Akwa Ibom State, Nigeria and confirmed that consumers could have been at risk of aflatoxin poison.
Certain heavy metals such as lead, cadmium, mercury and chromium have been detected in smoked fish and have also been recognized to be potentially toxic within specific limiting values (Smirjakova et al., 2005). Cadmium is spread by air and water (sewage sludge) far over sea and land, but especially in the vicinity of heavy industrial plants. Cadmium is today regarded as the most serious contaminant of the modern age. It is absorbed by many plants and sea creatures and, because of its toxicity, presents a major problem for fish. Cadmium, like lead, is a cumulative poison, that is, the danger lies primarily in the regular consumption of foodstuffs with low contamination (Zeleznik, 1994).
There are also strong pressures on chemical safety for smoked, barbecued and fried fish products from the EU institutions. Thus the Codex Alimentarius Commission on contaminants in food, at its 29 th session from 16 to 20 April 2007 established a reflection on reducing levels of Polycyclic Aromatic Hydrocarbons (PAHs) in food dried and smoked (EC, 2007). In addition, the EU Regulation 1881/2006 requires a formal setting new stricter rule on the content of PAH in smoked products. The presence of PAHs, especially benzo[a]pyrene, in smoked fish has previously been reported (Simko et al., 2002) but little information is available concerning the influence of the smoking and frying processes.
This study was therefore carried out to assess the quality, presence and concentration of PAHs and heavy metals as well as microbial hazards in smoked fish.

Sample collection
A total of 50 samples of smoked West African Ilisha fish were collected from ten major markets in Ibadan, Oyo State, Nigeria by purposive sampling method. The smoked fish samples were taken to the Federal University of Agric ulture, Abeokuta, Nigeria for laboratory analysis.

Proximate analysis
The proximate composition of all the samples were carried out in triplicates according to the standard method AOAC (2000).

Physico-chemical analysis
Kent pH meter (model 7020, Kent Ind. Measurement Ltd., Surrey, U. K) equipped with a glass electrode was used to measure the pH of the flesh in triplicates, employing 10 g of smoked fish homogenized in 10 ml of distilled water. The rancidity (quality) indices of all the samples were carried out in triplicates according to the standard method AOAC (2000). All chemicals used in this study were of the analytical grade unless stated otherwise.

Determination of PAHs by Gas Chromatography (GC)
To start the cold extraction 2 g of each sample was weighed into a clean extraction container and 10 ml of dichloromethane was added, thoroughly mixed and allowed to settle. The mixture was carefully filtered into clean solvent and rinsed into extraction bottle, using filter paper fitted into Buchner funnel. The extract was concentrated to 2 ml and then transferred for clean-up and separation (this involved further purification of the extract prior to gas chromatographic analysis). To achieve this, 1 cm of moderately packed glass wool was placed at the bottom of 10 mm ID × 250 mm long chromatographic column. Slurry of 2 g activated silica in 10 ml dichloromethane was prepared and placed into the chromatographic column. To the top of the column 0.5 cm of sodium sulfite was added. The column was rinsed with additional 10 ml of dichloromethane. The column was pre-eluted with 20 ml of dichloromethane and this was allowed to flow International Journal of Aquaculture, 2015, Vol.5, No.32 1-8 http://ija.biopublisher.ca 3 through the column at a rate of about 2minutes until the liquid in the column was just above the sulfite layer. Immediately, 1 ml of the extracted sample was transferred into the column. The extraction bottle was rinsed with 1 ml of dichloromethane and added to the column as well. The stop-cork of the column was open and the eluant was collected with a 10ml graduated cylinder. Prior to exposure of the sodium sulfite layer to air, dichloromethane was added to the column in 1-2ml increments. Accurately measured 10ml of the eluant was collected and labeled aliphatic. The concentrated aliphatic fractions were transferred into labeled glass vials with rubber crimps caps for GC analysis. 1μL of the concentrated sample was injected by means of syringe through a rubber septum into the column. Separation occurs as the vapour constituents' partition between the gas and liquid phases. The sample components were automatically detected as they emerge from the column (at a constant flow rate) by the Flame Ionization Detection (FID) detector whose response was dependent upon the composition of the vapor.

Heavy metal analysis
Heavy metal, such as Cu, Cd, Hg and (Pb) in smoked fish samples were determined by AOAC (2000) method using atomic absorption spectrophotometer. All chemicals used in this study were of the analytical grade unless stated otherwise.

Microbiological studies
The presence of pathogens in smoked fish samples was investigated in the microbiology laboratory. These include: Listeria monocytogenes, Salmonella paratyphi, Escherichia coli, Staphylococcus aureus and Fungal count. The microbiological procedures recommended in the International Commission on Microbiological Specification for Foods (ICMSF, 1986) were applied. Culture media were those of Oxoid, Biolife and Difco. For each sample, 25g were weighed out and transferred to a sterile blender with 225ml of 0.1% peptone and mixed thoroughly for 2minutes to prepare fish homogenate. The samples were then analysed as follows.

Total viable bacterial counts
Appropriate dilutions of the fish homogenate were prepared and inoculated onto sterile Petri dishes. Plate count agar (Oxoid) media were then poured. Plates were incubated at 37°C for 48hours and colonies were then counted and reported as total colony count/ml. A second set of plates was incubated at 37°C for 48hours in a carbon dioxide incubator or under anaerobic conditions using a gas pack anaerobic jar. Colonies were then counted and reported as anaerobic total bacterial count. In case of spore formers count, the food homogenate was boiled first at 75-80°C and then rapidly cooled. Appropriate serial dilutions were prepared and inoculated onto the surface of sterile and dried plate count agar media. These were incubated finally at 37°C for 48hours.

Detection of Escherichia coli
One ml of each of the decimal dilutions of the smoked fish homogenate was plated on poured Eosine Methylene Blue Agar (Oxoid) and then incubated at 37°C for 24hours. Counts were calculated from the number of growth on the plates. The colonies with green metallic sheen were counted as Escherichia coli.

Detection of Staphylococcus aureus
A sample of 0.1 ml of the smoked fish homogenate and dilutions was inoculated on Baird-Parker (Difco) agar plates and incubated at 37°C for 48hours. Colonies appearing to be black and shiny with narrow white margins and surrounded by clear zones were identified by coagulase test reactions. The coagulase test was carried out by first inoculating typ ical colonies in brain heart infusion broth (Difco) and incubating at 37°C for 24hours. From the resulting cultures, 0.1ml was then added to 0.3 ml of rabbit plasma in sterile tubes and incubated at 37°C for 4hours. The formation of a distinct clot was evidence of coagulase activity.

Detection of Salmonella paratyphi.
Samples of smoked fish homogenate and dilutions were inoculated in Salmonella-shigella agar (Oxoid) and incubated at 37°C for 24hours. For identification, 2-3 suspected colonies were inoculated into tryptone broth for indole test, triple sugar iron agar slant (Oxoid), urea broth and lysine iron agar. These were incubated at 37°C for 24hours. Salmonella species is indole negative, on triple sugar iron it produces acid (yellow) and alkaline (red) with or without gas and hydrogen sulfide, is urea negative, and on lysine iron agar shows an alkaline (purple) reaction throughout International Journal of Aquaculture, 2015, Vol.5, No.32 1-8 http://ija.biopublisher.ca 4 the medium. Serological tests were then carried out.

Listeria monocytogenes.
A sample of 0.1 ml of the smoked fish homogenate and dilutions was inoculated on Brilliant Listeria Agar (Oxoid) plates and incubated at 37°C for 24hours. Colonies appearing were counted and reported as Listeria monocytogenes.

Enumeration of fungi
Appropriate dilutions of Sabouraud dextrose agar plates (Oxoid) were poured over 1 ml of the fried fish homogenate and dilutions. Plates were incubated at 25°C for 3 days and then colonies were counted and reported as fungal count/ml.

Data analysis
The data obtained were subjected to descriptive statistics using IBM SPSS version 21.0 software. One way analysis of variance (ANOVA) was performed followed by Duncan's Multiple Range Test (p<0.05) to find the difference between means. Significant level was set at P<0.05.
The results of the concentrations of Hg, Pb, Cd, and Cr in the smoked fish samples are presented in Table 4. The concentrations (μg/g) of Hg, Pb, Cd, and Cr in the smoked fish samples were 0.0011-0.0013, 0.0010-0.0017, 0.0014-0.0021, and 0.0912-0.0972 respectively.
The results of the microbiological study are presented in Table 5. It was also found that TVC of smoked fish samples were in the range of 1.0 x 10 4 -6.0 x 10 4 cfu/g. Staphylococcal count of smoked fish samples were in the range of 11.0 x 10 2 -57.0 x 10 2 cfu/g. S. paratyphi and E. coli were not detected in the smoked fish samples.

Discussion
The smoked fish samples have high amount of protein.
This can help in reducing the problem of protein energy deficiency in the diets of Nigerians especially the children.
The peroxide values of the smoked fish were below the recommended value of between 20 and 40mgeq.peroxide/kg for rancid taste to begin (da Silva et al., (2008)). The FFA values obtained were very low and this suggests that the level of fat decomposition in the fried and smoked fish samples is low. The values for FFA obtained are below the threshold for rancidity detection in smoked fish. The TBA is used to assess the degree of fish spoilage especially in fatty fish. The TBA test measures a secondary product of lipid oxidation, malonaldehyde (da Silva, 2002). The TBA values of smoked fish samples were also low. The TBA values of (1.03 -1.11 mg TBA/kg) obtained for smoked fish is within the range specified by USFDA for smoked fish (da Silva et al., 2008). TVB-N is related to protein breakdown and is an index of fish spoilage (da Silva, 2002). The legislative standard for TVB-N include: 20mgN/100g for fresh fish, 30 mgN/100g stale fish and 40 mgN/100g for fish that is unfit for human consumption but can be used for animal feed (FAO, 1992;da Silva, 2002). In this study, TVB-N of smoked fish samples is within the range of legislative standard for TVB-N which is 20mgN/100g for fresh fish. This suggests that the level of protein decomposition or breakdown in all the samples is low. The TMA of 1.82 -2.95mgN/kg for smoked fish samples is within the range of < 3mgN/100g for fresh fish, >8mgN/100g for spoiled fish and > 5mgN/100g for doubtful quality specified (da Silva et al., 2008).
Although the result of PAH analysis revealed the presence of sixteen (16) PAHs compounds analysed in the smoked fish samples. Ten of the sixteen (16) PAHs compounds in smoked fish samples have  -Tautua et al., 2013). This may pose danger to the consumer of these products.
The four heavy metals investigated in all the smoked fish samples showed that quantities detected are generally below the maximum permissible levels set by World Health Organization for fried and smoked fish products (Brain and Allen, 1993) for Pb (0.3 ppm); Cd (0.2 ppm), Hg (0.2ppm) and Cr (0.5ppm) and hence    ICMSF (1986) for fish and fishery products, the maximum recommended bacterial counts for good quality products (m) is 5 x 10 5 (5.7 log10 CFU/g). The Staphylococcal count values obtained for the smoked fish samples were below the specified recommended value for all fish samples. S. aureus was isolated from smoked fish samples and this can be attributed to post processing contamination. The S. aureus isolated from smoked fish samples was within the safety level which is equal to or greater than 10 4 /g recommended by FDA, (2001). S. paratyphi was not detected in smoked fish samples and this conforms with the specified microbiological limits recommended by ICMSF (1986) for S. paratyphi count for fish and fishery products which is the presence of the organism, that is zero tolerance. In all cases, the absence of S. paratyphi and E. coli suggests Good Manufacturing Practices (GMP) and non faecal contamination of the products.

Conclusion
Based on this study smoked fish has high protein International Journal of Aquaculture, 2015, Vol.5, No.32 1-8 http://ija.biopublisher.ca 8 content. The rancidity indices were also below the rancidity threshold. However, the presence of