|dc.description.abstract||Sushi is a traditional Japanese dish that has become a popular ready-to-eat (RTE) product available in retail stores. Retail sushi is usually offered as complete meals, packed in a plastic lidded tray with a claimed shelf-life of three days at ≤ 4°C. To ensure sensory and microbiological quality, rice acidification and refrigeration are applied as hurdles against microbial growth. Little information is available on the composition and dynamics of the microbiota of these products prepared with raw fish and raw vegetables. They are subjected to contamination during handling, and susceptible to microbial growth during storage. This thesis aimed at establishing knowledge that can contribute to increased quality and safety of RTE sushi to the retail market, with special focus on the pathogenic Aeromonas species.
Screening of the microbiological quality of fresh retail sushi purchased in selected Norwegian supermarkets revealed that 48% of the products could be rated as microbiologically unsatisfactory. Significant differences in microbiological quality (especially H2S-producing bacteria associated with fish spoilage) were found between products purchased in stores and sushi collected directly from the factory in a follow-up of one producer. Poor temperature control after assembly was the most likely reason for quality loss during shelf-life. The main food safety issue associated with the analyzed sushi was the high prevalence of mesophilic Aeromonas spp., found in 71% of the products, and in concentrations > 4 log CFU/g in several boxes. Aeromonas spp. are ubiquitous bacteria in marine environments and have received increasing attention as human pathogens because of their virulence traits and widespread occurrence in food, especially in seafood. Sampling of sushi ingredients before assembly confirmed that Aeromonas spp. were introduced to the product from both raw fish and raw vegetables. Moreover, raw vegetables were identified as a source of bacteria that could contribute to spoilage of the assembled product.
The concentrations and maximum specific growth rates of total aerobic bacteria, lactic acid bacteria (LAB) and H2S-producing bacteria in retail sushi were quantified as a function of five storage temperatures (4 to 20°C) during a five day period. The bacterial communities were analyzed using the culture-independent method PCR-denaturing gradient gel electrophoresis (DGGE). Refrigeration resulted in no growth of H2S-producing bacteria, but this group had the strongest response to temperatures > 4°C, indicated by their temperature coefficients in the square-root model. The growth rate of total aerobic bacteria displayed a two-fold and six-fold increase at 8 and 20°C, respectively, and all bacterial groups proliferated without any lagphase at temperatures > 4°C. The relationship between bacterial growth rates and temperature demonstrated that retail sushi products are highly sensitive to deviations from optimal storage temperature and that storage at ≤ 4°C is an efficient hurdle to inhibit many spoilage bacteria. However, storage temperature alone was not the major determinant for the bacterial community structure. The total number of bacteria was the variable that most successfully explained the differences between the communities, which were dominated by Gram positive bacteria such as genera of LAB and Brochothrix thermosphacta.
The Aeromonas strains (n=118) isolated from retail sushi products were characterized genetically and phenotypically. The strains were identified as A. salmonicida (74%), A. bestiarum (9%), A. dhakensis (5%), A. caviae (5%), A. media (4%), A. hydrophila (2%), and A. piscicola (1%) by phylogenetic analysis based on sequencing of the gyrB gene. All isolates were characterized as potentially pathogenic due to the high prevalence of genes encoding hemolysin (hlyA), aerolysin (aerA), cytotoxic enterotoxin (act), heat-labile cytotonic enterotoxin (alt), and heat-stable cytotonic enterotoxin (ast). β-hemolysis was species dependent and for the non-hemolytic isolates, the lack of hemolysis was possibly linked to the absence of the act gene. The combination of recognized virulence genes and the expressed hemolysis and motility were comparable to clinical Aeromonas strains. These findings prompted further investigation of the growth potential of mesophilic Aeromonas in refrigerated RTE seafood.
A microbiological challenge test using a mesophilic strain of A. salmonicida in a nigari sushi model demonstrated that Aeromonas spp. may represent a microbial hazard in retail sushi (and other salmon products) during refrigerated storage. Refrigeration alone was not sufficient to prevent growth of the inoculated strain on Atlantic salmon during storage at 4°C, whereas acidified rice (nigiri sushi) resulted in a pH-drop in the fish which inhibited growth of the Aeromonas strain at 4°C. However, the effect of lower pH was cancelled at 8°C, representing a mild temperature abuse. Moreover, a screening of the ability to grow in a wide range of pH (pH 3.5 to 10) for a selection of Aeromonas strains showed strain to strain variability in pH-tolerance. Several strains had the ability to grow in the low range (pH 4.5 to 5) when the stress from low temperature was removed. Thus, the combination of low pH and low storage temperatures are prerequisites to prevent growth of Aeromonas to potentially disease-causing levels during the shelf-life of retail sushi products.
In conclusion, this work demonstrated that the quality and food safety of retail sushi products must be founded through high quality ingredients with emphasis on thorough rinsing of raw vegetables and initial fish quality, and that hurdles against microbial growth such as temperature control and rice acidification must be combined to prevent growth of spoilage-associated bacteria as well as potentially pathogenic bacteria in sushi. Mesophilic Aeromonas strains can grow in unpreserved salmon during refrigerated storage and may constitute a food safety risk if other hurdles are not applied. Retailers have a responsibility to preserve the initial product quality by efficient goods reception and proper refrigeration during display in stores. The observed differences in microbiological quality between products purchased in stores compared to those collected from the factory suggest that quality loss can be a result of poor temperature control after product assembly. Shelf-life stability and food safety is thus a result of joint effort in the entire food chain.||nb_NO