Shelf-life extension of Atlantic salmon fillets

This research was to extend the shelf-life of fresh Atlantic salmon (Salmo salar) fillets sold in supermarkets across Australia. Improving the shelf-life of salmon fillets from a current commercial limit of 14 days by 2 ‚- 4 days will enable producers, retailers and consumers to have more options for logistics, storage and consumption. The research focussed on combining the effects of two hurdle approaches: (1) novel sanitation methods to reduce the microbial load during the two to five days pre-processing stage for relaxation of rigor mortis prior to packaging; (2) optimisation of packaging conditions to inhibit the microbial growth of any pathogens and principal spoilage species. The novel sanitation method was the application of neutral electrolysed oxidising water (NEW) in the early stages of fish processing after harvest to reduce the natural indigenous bacteria of the fish surfaces and any environmental flora that may become a source of contamination during processing and packaging. The physicochemical stability of NEW was first determined under different storage conditions, and in the presence of organic substances and in natural seawater to understand the practical limits for industrial application. It was found that NEW chlorine equivalents in concentrate stored at room temperatures decreased by 52 % within 28 days of production. Therefore, concentrate should be used within this time to have sufficient oxidising function for microbial efficacy. When applied to fish products during storage in ice-water oxidising power rapidly decreased by over 70% in 30 minutes, due to reaction with organic matter from the fish. This required renewal of the NEW to maintain redox oxidising activity greater than 700 mV. This replenishment was therefore generated by slow release melting of neutral electrolysed oxidising ice (NEI) to maintain the antimicrobial activity for an extended period of up to 5 days during storage. A preliminary small-scale trial investigated the efficacy of using NEW and slow melting NEI against spoilage and pathogenic organisms. The results revealed that 100 ppm chlorine equivalents of NEI was sufficient to inhibit the growth of spoilage microorganisms on salmon fillets during 10 days of storage at 4 ¬∫C. It was also found that 100 ppm NEI decreased the number of Pseudomonas aeruginosa in salmon fillets portions compared to untreated controls by 1.1 log CFU/cm\\(^2\\) and 2.4 log CFU/cm\\(^2\\) for 50 and 100 ppm NEI, respectively. In addition, 100 ppm NEI inhibited Listeria monocytogenes during storage by decreasing the final population after 24 h by 1 log CFU/cm\\(^2\\). The subsequent trials simulated full scale processing by treating head-on gutted (HOG) Atlantic salmon during pre-processing relaxation of rigor mortis followed by applying antimicrobial hurdles during processing and packaging. When 100 ppm chlorine equivalents of NEW and NEI were used during the pre-processing stage the microbial level after six days of pre-treatment was 2.7 log CFU/cm\\(^2\\). When HOG Atlantic salmon were just washed in NEW prior to filleting, the microbial load on the skin of HOGs treated at 20 ppm decreased by 2.5 log CFU/ cm\\(^2\\) down to 2.6 log CFU/cm\\(^2\\). When treated at the higher concentration of 100 ppm, the microbial load after this sanitation step was below the detection level. Full scale trials mimicking industrial practice were then conducted by combining the improved pre-processing stage using NEW/NEI followed by optimisation of modified atmosphere packaging (MAP). Optimisation of MAP to minimise microbial growth in bulk package fillets was carried out at different levels of gas to product (G/P) ratio of MAP. The combinations of sanitation pre-processing and high G/P ratio were effective for lowering the microbial counts by 1.5 log CFU/g compared to the untreated control after 20 days under controlled cold storage at 0¬∫ C and 4¬∫ C. Shelf-life of fresh chilled salmon fillets was also assessed by consumers from the smell of aroma volatiles. The effects of washing the HOGs in NEW/NEI on the type of microbial community that grows on the fillets during storage, and hence the type of volatile organic compounds (VOCs) that are produced, was also determined. Sensory evaluation, microbial enumeration, microbiome profiling and volatile organic compound analysis was performed on samples at intervals for up to 20 days while simulating commercial conditions of storage. The microbial communities were very dynamic with the Jaccard similarity index being less than 50% between every treatment. Twenty-five volatile organic compounds were found in the aroma headspace and changes in their concentration with time were characterised. Freshness volatiles such as hexanal, heptanal and octanal decreased during 20 days of storage. Fifteen sulphide producing cultures obtained from fillets by isolation on iron agar were identified as Shewanella baltica although no sulphides were found in the VOCs or detected by sensory assessment. This was presumably due to consistently low levels of less than 3.1 log CFU/g of hydrogen sulphide producing bacteria (HSPB) in the fish fillets. Altering the microbiome therefore altered the types of VOCs produced. Treatments that favourably altered the microbial inoculum and resulting microbiome of the fillets are therefore a potential way to extend the perceived freshness of the fish product. A final pre-treatment sanitation approach was also tested as a potential further antimicrobial hurdle step to compensate for high packing density, low G/P ratio used in bulk MAP shipments. This pre-processing variant used a combination of NEW/NEI and soluble gas stabilisation (SGS) applied through raised levels of carbonic acid/bicarbonate during storage prior to filleting. The HOGS were first treated by dipping in Shewanella baltica at 3 log CFU/mL to provide a consistent challenge inoculum and were then held at 4¬∫ C for 3 days in the NEW/NEI and SGS pre-processing stage prior to filleting. The fillets were MAP stored with 0.4:1 G/P ratio in line with existing industry practices. Sensory evaluation, microbial load, microbiome profiling and VOC analysis was performed at intervals up to 20 days simulating commercial conditions of storage. The results showed that a combination of NEW/NEI and SGS inhibited the growth of HSPB on fillets packed in MAP during storage by 1 log CFU/cm2 of the skin compared to fillets that did not receive the SGS pre-processing stage treatment. Twenty-seven VOCs and their changes were identified and characterised in the headspace of the fillets. Freshness compounds such as octen-3-ol, hexanal and nonanal were found on Days 1 and 10, but their relative levels decreased with storage time. Microbial community analysis showed that Photobacterium was the dominant flora in every treatment at Days 17 and 20 of the storage periods. However, no sulphide VOC was again detected in the headspace and it is inferred that the final microbial levels of 6.4 log CFU/g were also too low to affect the sensory quality. In conclusion, the combination of improved sanitation using NEW/NEI during pretreatment and more intensive MAP using SGS could be applied within existing industrial processes to extend the shelf-life of fresh chilled Tasmanian Atlantic salmon fillets by maintaining its freshness for at least an extra two days and potentially up to 6 days more than the current shelf life. The integrated hurdle technology approach will enable the industry to better preserve the consumer perceived quality of bulk packed fresh Atlantic salmon fillets compared to the existing partial antimicrobial wash and MAP treatments alone.