Microbiological quality changes in the intestine of hybrid tilapia (Oreochromis niloticus × Oreochromis aureus) in fresh and frozen storage condition
Ahmed H. Al-Harbi, Fish Culture Project, NRERI, KACST, PO Box 6086, Riyadh 11442, Saudi Arabia (e-mail: email@example.com).
Aims: The goal of this study was to monitor the quantitative and qualitative bacterial flora in the intestine of hybrid tilapia in fresh fish and fish kept in frozen storage conditions for 1 year.
Methods and Results: Quantitative and qualitative analyses of the bacterial flora associated with the intestine of hybrid tilapia (Oreochromis niloticus × Oreochromis aureus) in fresh fish and fish kept in frozen storage conditions for 1 year were carried out. In fresh and frozen fish, aerobic plate count (APC) ranged from 1·6 ± 1·2 × 108 to 1·5 ± 0·9 × 105 CFU g−1 in the intestine of tilapia collected from pond 1, 8·7 ± 2·3 × 107 to 6·5 ± 3·8 × 104 CFU g−1 in the intestine of tilapia from pond 2, and 1·9 ± 2·9 × 108 to 6·2 ± 2·8 × 104 CFU g−1 in the intestine of tilapia from pond 3. APC for all the groups of fish decreased c. 2-log cycles after 1 months frozen storage; thereafter, counts slowly declined during frozen storage for 1 year. Altogether, 16 bacterial genera were identified: Gram-negative rods (67%) dominated. Both in fresh and frozen conditions, four bacterial species viz. Shewanella putrefaciens, Corynebacterium urealyticum, Aeromonas hydrophila and Flavobacterium sp. were always present, with a prevalence of 10% in most cases. Shewanella putrefaciens was the most dominant organism (15% of the total isolates) throughout the studied period. During frozen storage some of the bacteria were not recovered, but most of the bacteria survived after prolonged freezing.
Conclusions: This study describes the aerobic heterotrophic microflora found in the intestine of fresh and frozen tilapia. The unique aspect of this study concerns the data revealing the micro-organisms, which are viable after prolonged freezing. Contamination of edible portions of fish could originate from gastrointestinal sources.
Significance and Impact of the Study: The present results may enhance knowledge in controlling the storage life of fish, and fish product quality. Bacterial activity is by far the most important factor influencing fish quality, so bacterial numbers can be used as an index of quality. Storage of frozen tilapia without evisceration could be avoided.
The duration of bacterial viability in frozen fish tissues or the length of time tissues can remain frozen and still yield viable bacteria upon partial thawing is an area warranting further study. The usefulness of microbiological criteria as objective indicators of the sanitary history of foods has been discussed by many and studied by some, and specific criteria have been proposed for several foods. Fish is one of the most perishable and difficult to handle of all foods. Bacterial activity is by far the most important factor influencing fish quality (Gram and Huss 1996); therefore, it is logical to use bacteria numbers as an index of quality. For high quality fresh fish, the number of bacteria present on the surface varies from 3 to 4 log CFU g−1. On gills, counts are normally one or two orders higher, and intestinal counts can reach 9 log CFU g−1 (Sikorski 1990). Spoilage microflora produce enzymes that cause proteolysis, deamination, decarboxylation resulting in accumulation of unpleasant metabolites and loss of taste substances leading to the sensory rejection of the fish (Shewan and Murray 1979; Connell and Shewan 1980; Huss 1988; Sikorski et al. 1994). Microbiological quality of fresh and frozen fish can be judged using several criteria (Kramer and Liston 1987). Minimum (m) and maximum (M) microbiological limits for aerobic plate counts on fish at 30°C proposed by the ICMSF (1986) are 5 × 105 (5·69 log) and 107 (7 log) CFU g−1 or cm2 respectively. However, some seafood processors specify a 2-class plan for acceptance of frozen fish, where APC should not be higher than 5 log10 CFU g−1 (Elliot 1987).
The frequent contamination of fishery products by pathogenic micro-organisms has been considered a health hazard to consumers (Ingham and Potter 1991). Contamination of these edible portions could originate from gastrointestinal sources. Freezing is one means of preserving fish for longer periods, and frozen fish have become an important commodity both for domestic and export markets in a number of countries of the world. The shelf-life of raw fish depends on the storage conditions, the intrinsic factors, and the qualitative and quantitative composition of the initial microflora, related to the environment where the fish live and are caught, the seasonal period, the fishing method and the early handling (Shewan 1977; Ward and Baj 1988). In Saudi Arabia, one of the largest and perhaps most obvious deficiencies in knowledge concerns the bacteriology of fish. Specially, there is no information on microbiological condition of fish during frozen storage in this country. Storage of frozen tilapia without evisceration is available in the market. By monitoring the bacterial contents of fish organs, the quality of fish can be measured, as these will affect the storage life and quality of the fishery products. This present paper reports on the overall quantitative and qualitative microbiological results for the intestine of fresh fish and frozen storage condition.
Materials and methods
Fish were collected from three artificial earthen ponds, constructed in 1992 and operated by King Abdulaziz City for Science and Technology, KSA. The ponds are completely supplied with groundwater over the year. Only little vegetation exists in the shallow shore areas. There is 15–23 cm of mud in the bottom. Tilapia is the only fish held in these ponds and feed or fertilizers are seldom used. Fish mainly depend on natural food. Although these ponds are not properly managed, many of the public take fish from them, and water from the ponds is also used for irrigation purposes. The area of each pond is 1416 m2 with an average depth of 1·5 m.
Bacteriological sampling and analyses
Three samplings, one in each pond, were performed at a time to catch tilapia by using cast net from April 2001 to March 2002. Immediately after the catch, 12 sample bags from each of the three ponds, a total of 36 bags, each containing nine representative (size, condition, sex, etc.) fish were preserved in a refrigerator at temperature of −20 ± 1°C in walk-in units. For quantitative bacterial analyses samples were tested monthly, but for qualitative analyses at 3-month intervals. One bag of frozen fish was randomly taken from each unit and evaluated for microbiological quality.
Nine frozen fish collected from each of three ponds (three fish in each replication) weighing between 35 and 42 g were used in bacteriological examinations. Fresh fish were only tested one time at the beginning of the storage period. From each of three ponds 15 representative fish (three replication, three × five fish in each pond) were used. They were killed by physical destruction of the brain, and the number of incidental organisms was reduced by washing the fish skin with 70% ethanol before opening the ventral surface with sterile scissors to expose the body cavity. Frozen fish were air thawed (c. 40 min) until it was possible to take out the intestine. Three to five grams of intestine from sampled fish were taken aseptically, mixed, and homogenized in a mortar. To standardize the sampling procedure, intestine was taken from different parts of the whole intestines. Around 2 g of homogenate was suspended in a bottle containing 25 ml of sterile (121°C, 15 min) 0·85% (w/v) NaCl prepared in deionized water. One millilitre of the suspension was serially diluted to 10−8. Volumes (0·1 ml) of the dilutions were spread onto tryptone soya agar plates (TSA; Oxoid, Basingstoke UK), in duplicate.
Total viable count
For total heterotrophic aerobic bacterial counts of tilapia intestine, all the inoculated plates were incubated at 30°C for 48 h and colony-forming units (CFU) counted with a Quebec Darkfield Colony Counter (Leica, Inc., Buffalo, NY, USA) equipped with a guide plate ruled in square centimetres. Readings obtained with ≥30–300 colonies on plate were used to calculate bacterial population results, recorded as CFU ml−1.
Isolation of bacteria
Bacterial strains were isolated from fresh and frozen tilapia intestines. The colonies were divided into different types according to the colony characteristics of shape, size, elevation, structure, surface, edge, colour and opacity, and the number of colonies of each recognizable type was counted. Three to five representatives of each colony type were then streaked on TSA plate repeatedly until pure cultures were obtained. For all populations, around three per cent of primary isolates failed to grow despite repeated attempts on subsequent subculturing. Cultures on TSA slants were kept at 4°C for stock and these were transferred to new slants every 6 weeks.
Identification of bacteria
All the purified isolates were observed for cell shape, motility, flagellation, spores and encapsulation, and Gram staining. The isolates were then subjected to biochemical tests (oxidase, catalase, Hugh and Leifson, amylase, gelatinase, lipase, indole, H2S production, nitrate reduction, etc.) following the criteria described in Bergey's Manual of Determinative Bacteriology (Holt et al. 1994) for identification to genus or species level. The presumptive Vibrio species were tested by their growth on the different concentrations of NaCl, thiosulfate-citrate-bile sucrose (TCBS) agar (Oxoid), and by their sensitivity to vibriostatic agent (0/129) (Oxoid). In parallel, commercial API 20E, API 20 STREP (bioMerieux, Marcy l'Etoile, France) and BIOLOG (BIOLOG, Inc., Hayward, CA, USA) methods were also used.
Table 1 summarizes the quantitative results of APC in tilapia intestine at fresh condition and different storage time for 1 year. During the period of study the APC ranged from 1·6 ± 1·2 × 108 to 1·5 ± 0·9 × 105 CFU g−1 in the intestine of tilapia collected from pond 1, 8·7 ± 2·3 × 107 to 6·5 ± 3·8 × 104 CFU g−1 in the intestine of tilapia collected from pond 2, and 1·9 ± 2·9 × 108 to 6·2 ± 2·8 × 104 CFU g−1 in the intestine of tilapia collected from pond 3. APC for all the groups of fish decreased c. 2-log cycles after 1 months frozen storage; thereafter, counts slowly declined during frozen storage. Each count was the mean value of viable colonies appearing in agar plates made per individual sample. Individual pond results are given in Table 1.
Table 1. Bacterial counts in the intestine of hybrid tilapia in fresh and frozen storage conditions for 1 year
|1||41·2 (1·7)||16·1 (1·5)||1·6 × 108 (1·2)||1·2 × 106 (1·8)||6·2 × 106 (1·9)||6·7 × 106 (3·3)||2·2 × 106 (1·5)||1·1 × 106 (2·3)||1·2 × 106 (3·2)||4·2 × 105 (1·9)||5·6 × 105 (3·5)||2·2 × 105 (1·4)||4·4 × 105 (1·1)||1·5 × 105 (0·9)||2·0 × 105 (2·8)|
|2||37·1 (1·9)||14·6 (1·7)||8·7 × 107 (2·3)||2·3 × 106 (2·7)||1·1 × 106 (2·0)||2·4 × 105 (1·6)||4·9 × 105 (3·4)||6·5 × 105 (2·8)||1·0 × 105 (1·3)||1·1 × 105 (2·8)||3·9 × 105 (1·0)||4·9 × 105 (2·3)||6·1 × 105 (1·9)||6·5 × 104 (3·8)||1·9 × 105 (1·4)|
|3||38·5 (2·1)||15·0 (2·4)||1·9 × 108 (2·9)||3·3 × 106 (3·1)||1·3 × 106 (2·3)||1·1 × 106 (1·9)||2·3 × 106 (2·7)||8·3 × 105 (1·6)||2·3 × 106 (2·5)||7·1 × 105 (3·6)||2·3 × 105 (1·7)||6·2 × 104 (2·8)||1·3 × 105 (3·6)||1·4 × 105 (1·5)||8·3 × 104 (2·9)|
Taxonomic composition of the bacterial flora
Bacterial isolates recovered from the intestine of fresh and frozen tilapia were identified to species level where possible. Table 2 reports the isolation frequency of the recognized bacterial groups from fresh intestine and after different storage times. The results represent the bacterial flora in the intestine of both fresh and frozen tilapia collected from all three ponds because the data obtained from each group were very similar. There were wide variations in distribution pattern and types of bacteria in different sampling times. Bacteria isolated from the samples were predominantly Gram-negative rods (67%). Altogether, 16 bacterial genera were identified from the intestine of tilapia. Salmonella sp. and Vibrio vulnificus were identified presumptively.
Table 2. Bacterial composition and percentage distribution in the intestine of hybrid tilapia in fresh and frozen storage conditions at 3-month intervals for 1 year
|Aeromonas hydrophila||24 (14·55)||14 (11·38)||13 (9·49)||13 (8·61)||19 (9·90)|
|Bacillus sp.||2 (1·21)||4 (3·25)||2 (1·46)||4 (2·65)||2 (1·04)|
|Cellulomonus sp.||5 (3·03)||4 (3·25)||6 (4·38)||8 (5·30)||9 (4·69)|
|Corynebacterium urealyticum||19 (11·52)||15 (12·20)||19 (13·87)||23 (15·23)||30 (15·63)|
|Curtobacterium pusillum||3 (1·82)||–||–||–||–|
|Escherichia coli||15 (9·09)||5 (4·07)||9 (6·57)||9 (5·96)||–|
|Flavobacterium sp.||9 (5·46)||12 (8·94)||12 (8·76)||14 (9·27)||21 (10·94)|
|Micrococcus sp.||14 (8·49)||9 (7·32)||6 (4·38)||5 (3·31)||–|
|Pasteurella sp.||6 (3·64)||6 (4·88)||5 (3·65)||6 (3·97)||10 (5·21)|
|Pseudomonas fluorescens||9 (5·46)||8 (6·50)||2 (1·46)||–||–|
|Salmonella sp.||7 (4·24)||5 (4·07)||10 (7·30)||10 (6·62)||14 (7·29)|
|Serratia liquefaciens||5 (3·03)||5 (4·07)||–||–||–|
|Shewanella putrefaciens||21 (12·73)||13 (10·57)||20 (14·60)||28 (18·54)||37 (19·27)|
|Staphylococcus sp.||4 (2·42)||3 (2·44)||5 (3·65)||4 (2·65)||6 (3·13)|
|Streptococcus sp.||7 (4·24)||6 (4·88)||7 (5·11)||12 (7·94)||15 (7·81)|
|Vibrio vulnificus||7 (4·24)||8 (6·50)||10 (7·30)||8 (5·30)||16 (8·33)|
|Unidentified||8 (4·85)||7 (5·69)||11 (8·03)||7 (4·64)||13 (6·77)|
Four bacterial species viz. She. putrefaciens, Corynebacterium urealyticum, Aeromonas hydrophila and Flavobacterium sp. were present with a prevalence of 10% in most cases. Shewanella putrefaciens was the most dominant organism (15% of the total isolates). Most of the bacteria survived after 12 months frozen storage. Curtobacterium pusillum, Serratia liquefaciens and Pseudomonas fluorescens were not recovered after 3, 6 and 9 months frozen storage, respectively, while Escherichia coli and Micrococcus sp. were not found after 12 months.
The results showed that bacterial load in tilapia intestine varied in fresh and frozen conditions. APC for all the groups of fish decreased c. 2-log cycles after 1 months frozen storage and c. 3 log after 12 months. Al-Harbi and Uddin (2004) reported the seasonal change in intestinal bacterial load of fresh water tilapia in Saudi Arabia as 8·9 × 105 to 1·3 × 109 CFU g−1. During freezing, ice crystals rupture tissue cells and even micro-organisms. The initial killing rate during freezing is rapid but it is followed by a gradual reduction of micro-organisms (Frazier and Westhoff 1990). This research appears closer to our work because of similar observations on the decrease of viable micro-organisms with the frozen storage period of tilapia. Suvanich et al. (2000) reported that the APC of channel catfish frame mince decreased from c. 107–105 CFU g−1 after 2 months frozen storage; thereafter this remained almost unchanged. By monitoring the bacterial load in fish, which will affect the storage life, quality of the fishery products can be measured.
The predominant bacterial flora consisted of Gram-negative rods. Four bacterial species viz. She. putrefaciens, Coryne. urealyticum, Aer. hydrophila and Flavobacterium sp. were present with a prevalence of 10% in most cases, as shown in Table 2. Recovery and retrieval of most of the bacteria from frozen tilapia after prolonged freezing in the present study supports the idea that frozen fish can be a viable alternative to microbiological sampling in instances when fresh healthy or moribund fish diagnostic analysis is unavailable or impractical (Evans et al. 2004). Sugita et al. (1982) found that the gut of tilapia contained a wide variety of bacterial species with the predominant genera being Pseudomonas and Aeromonas species (Enterobacteriaceae). The high prevalence of She. putrefaciens, Coryne. urealyticum, Aer. hydrophila and Flavobacterium sp. throughout the study period suggests that these bacteria may be common bacterial commensals in tilapia intestine. Some strains of She. putrefaciens produced strong and offensive off-odours. Strains of Ps. fluorescens also produced offensive odours, but generally to a lesser extent than She. putrefaciens (Gennary et al. 1999). Although She. putrefaciens count can be useful in quality determination of fish, it is probably not a sufficient parameter for shelf-life prediction. Castro-Escarpulli et al. (2003) isolated Aer. hydrophila from frozen fish. In the present study, isolation of She. putrefaciens, Aer. hydrophila, Flavobacterium sp., V. vulnificus, Streptococcus sp., Salmonella sp., E. coli, Ps. fluorescens and Staphylococcus sp., which are facultative pathogens or agents of food poisoning and spoilage, is of importance. Rashid et al. (1992) recovered V. vulnificus from frozen shrimp. The Vibrios can cause different types of disease and are transmitted through foods and through cross-contamination during handling and processing of foods. Evans et al. (2004) recovered Strep. agalactiae from experimentally infected tilapia after 6 months freezing. The present isolated E. coli was nonpathogenic. The test for E. coli 0157, gave negative results. Al-Harbi (2003) observed significant coliform bacteria, including E. coli, in the same sampled tilapia culture ponds. The presence of pathogenic bacteria should be of concern to fish processors during processing and handling of products. The present report will contribute in controlling the storage life of fish and fish product quality.
The authors are grateful to King Abdulaziz City for Science and Technology (KACST), KSA for research support.