Use of citric and lactic acids in ice to enhance quality of two fish species during on-board chilled storage
Bibiana García-Soto1, Inmaculada Concepción Fernández-No2, Jorge Barros-Velázquez2, and Santiago P.Aubourg3,*
1 Cooperativa de Armadores de Pesca del Puerto de Vigo (ARVI), Vigo (Spain)
2 Department of Analytical Chemistry, Nutrition and Food Science, School of Veterinary Sciences, University of Santiago de Compostela, Lugo (Spain)
3 Department of Food Science and Technology, Marine Research Institute (CSIC), Vigo (Spain)
* Corresponding author. Prof. Santiago P. Aubourg; C/ Eduardo Cabello, 6. 36208-Vigo (Spain); Phone: +34 986231930; Fax: +34 986292762; e-mail:
ABSTRACT
This work focused on the on-board chilled storage of European hake (Merlucciusmerluccius) and megrim (Lepidorhombus whiffiagonis). To enhance fish quality, an aqueous solution including citric (1.25 g l-1) and lactic (0.50 g l-1) acids was prepared, frozen, ground and employed as icing medium. Its effect on sensory, microbiological and chemical changes was monitored after 9, 12 and 15 days of on-board storage. Lower (p<0.05) bacterial growth was detected according to microbiological (aerobe, anaerobe, psychrotrophe, proteolytic, and Enterobacteriaceae counts) and chemical (trimethylamine content) assessments. An inhibitory effect (p<0.05) on autolysis development (K value assessment) in hake was also detected. Finally, an enhancement of sensory scores (eyes, external odour and gills) in both species was obtained. Results described allow to conclude that on-board employment of such acid-mixture icing system can provide a profitable strategy to obtain higher quality and safe products while unloading.
Keywords: Merluccius merluccius; Lepidorhombus whiffiagonis; on-board chilling; citric acid; lactic acid; shelf life.
Running Title: On-board storage of commercial fish
1. INTRODUCTION
Deterioration of marine species begins immediately upon capture or harvest, and the degree to which it continues depends directly on storage conditions. Flake ice has been the most employed method to cool and store fish products and partially inhibit detrimental effects on the commercial value. However, significant deterioration of sensory quality and nutritional value has been detected in chilled fish as a result of microbial and biochemical degradation mechanisms (Whittle et al., 1990; Beaufort et al., 2009).To retard fish damage as long as possible and accordingly extend shelf life, a wide number of preservative strategies to be combined with flake ice chilling have been tested satisfactorily such as previous chemical treatment (Manju et al., 2007), employment of preservative packaging (Ruiz-Capillas et al., 2001) and presence of preservative compounds (ozone) (Pastoriza et al., 2008) or plant extracts (thyme hydrosol, rosemary extract) (Oral et al., 2008; Özyurt et al., 2012) in the icing medium.
Among previous chemical treatments used during chilling storage, natural low molecular weight organic acids and their sodium salts represent a relevant choice because of their easy availability, low commercial cost and wide range of permitted concentrations for use. Thus, citric acid (CA) is widely known for its role as a chelator and an acidulant in biological systems; its presence has resulted in a profitable effect on fish fillet (Badii and Howell, 2002; Kilinc et al., 2009) and whole fish (Aubourg et al., 2004) quality. Further, lactic acid (LA) has been reported to be effective in preserving and extending shelf-life for fish fillets (Kim et al., 1995; Metin et al., 2001) and coated fish (Gogus et al., 2006).
Unlike other muscle food, fish are usually harvested in remote locations. Among them, the Grand Sole North Atlantic fishing bank has been exploited by a wide number of European countries. Due to the fast post-mortem deterioration of fish species, most problems are encountered because the time elapsed between catching in these locations, and arrival at the ultimate destination can reach a 14-17-day period. Consequently, the threat of having fish condemned, withdrawn from sale, or sold at low prices at harbour, may limit the length of the voyage (Aubourg et al., 2006; Barros-Velázquez et al., 2008). As a result, substantial efforts are needed for the optimisation of the refrigeration systems employed on-board to meet the increasing consumer demand for high quality and safe, fresh products.
The present work is focused on the on-board storage and commercialisation of two abundant fish species (European hake, Merlucciusmerluccius; megrim, Lepidorhombus whiffiagonis.) from the Grand Sole bank. Its basic objective was the quality enhancement of fish captured during the first period of the trawler trip. Previous research carried out at laboratory level showed that quality loss could be inhibited in both species when applying ice prepared from an aqueous solution of CA and LA (García-Soto et al., 2013a, 2013b). With the aim of attaining a quality enhancement, an aqueous solution including both acids was prepared and employed on-board as an icing medium. Its effect on sensory, chemical and microbiological changes was monitored at different catching times during a trawler trip.
2. MATERIAL AND METHODS
2.1. Icing systems
An aqueous solution containing 1.25 g l-1 of CA and 0.50 g l-1 of LA was prepared, packed in polythene bags and kept frozen at –20ºC until use. Traditional ice was prepared starting only from tap water that was packed and kept frozen in the same way as the ice including both acids. Before addition to individual fishes, the two ices were ground to obtain common flakes. Organic acids encountered in the present research are regarded as safe (GRAS) for use in foods according to European and American administrations (Madrid et al., 1994; Giese, 1996).
Previous research was conducted onshore to assess a convenient concentration of CA and LA to prepare the ice (García-Soto et al., 2013a, 2013b). Thus, solutions combining the two acids in the 0.05-2.50 g l-1 concentration range were preliminary tested. According the evaluation of sensory, microbiological and chemical indices related to quality loss, the above-mentioned combination of both acids (1.25 g l-1 of CA and 0.50 g l-1) of LA was chosen.
2.2. Fish material, processing and sampling
European hake (Merluccius merluccius; length 32-35 cm, weight 180-210 g), and megrim (Lepidorhombus whiffiagonis; length 20-23 cm, weight 95-120 g) were captured in the Grand Sole North Atlantic fishing bank throughout a single trip (May-June 2012). All fish were gutted immediately after catching, but none were beheaded. For each fish species, individuals were distributed on-board into acid (treated batch, T) or traditional (control batch, C) icing treatments. Individuals were surrounded by ice at a fish/ice ratio of 1/1 (w/w) and stored on-board in a refrigerated room at 0-1ºC. Each fish species was captured at three different times of the trip. At each sampling time, and for both C and T batches, individuals were separated into three groups (three individuals per group) in order to be analysed separately (n=3).
Once the fishing boat arrived at Vigo harbour, fish specimens were transported to the laboratory to be analysed. Consequently, sensory, microbiological and chemical analyses were performed after 9, 12 and 15 days of on-board chilled storage from catching time. Sensory analysis was conducted on the whole fish, while microbiological and chemical analyses were done on the white muscle.
2.3. Sensory analysis
Sensory analysis was conducted by a sensory panel consisting of five experienced judges (three male and two female with an age in the 30-55 yr range), according to traditional guidelines concerning fresh and refrigerated fish adapted to the species under study (Council Regulation, 1996). Before starting the present experiment, the panel was trained on chilled hake and megrim. In this training, evaluation of chilled specimens belonging to the two species and corresponding to different chilling times (from starting material until the time the fish was no more acceptable) and quality degrees were tested. These preliminary chilled experiments were carried out several times, with a marked attention to the evolution of each of the different sensory descriptors; special emphasis was given to descriptors that were found as limiting factors of acceptability (namely, eyes, gills and external and muscle odours).
According to the mentioned Council Regulation (1996) procedure, four categories were ranked: highest quality (E), good quality (A), fair quality (B), and unacceptable quality (C). Sensory assessment of the fish included the following descriptors: skin and mucus development, eyes, external odour, gills appearance and odour, consistency, flesh odour (raw and cooked) and flesh taste (cooked). At each sampling time, the fish were presented to panellists and were scored individually by panellists; each descriptor of each sample was scored a single time by each member of the panel. The panel members shared samples tested.
2.4. Microbiological analyses
Samples of 10 g of fish muscle were dissected aseptically from chilled fish specimens, mixed with 90 ml of 1 ml l-1 peptone in water (Merck, Darmstadt, Germany), and homogenized in sterilized stomacher bags (AES, Combourg, France) as previously described (Ben-Gigirey et al., 1998).In all cases, serial dilutions from the microbial extracts were prepared in 1 ml l-1 peptone in water.
The number of aerobic mesophiles was determined by surface inoculation on plate count agar (PCA; Oxoid Ltd., London, UK) after incubation at 30ºC for 48 h. The anaerobe counts were also determined in PCA at 30ºC, except that an anaerobic atmosphere kit (Oxoid) was placed together with the plates inside the anaerobiosis jar. Psychrotrophs were also investigated in PCA, but incubation was performed at 7-8ºC for 7 d. Enterobacteriaceae were investigated by pour plating using Violet Red Bile Agar (VRBA) (Merck, Darmstadt, Germany) after incubation at 37ºC for 24 h. Microorganisms exhibiting a proteolytic phenotype were investigated in casein-agar medium after incubation at 30ºC for 48 h, as previously describedby Ben-Gigirey et al. (2000).
In all cases, bacterial counts were transformed into log CFU g-1 muscle before undergoing statistical analysis. All analyses were performed in triplicate.
2.5. Chemical analyses
Total volatile base-nitrogen (TVB-N) values were measured as previously reported (Aubourg et al., 2006). Briefly, fish muscle (10 g) was extracted with 60 g l-1 perchloric acid in water (30 ml) and brought up to 50 ml. An aliquot of the acid extracts was rendered alkaline to pH 13 with 200 g l-1 aqueous NaOH and then steam-distilled. Finally, the TVB-N content was determined by titration of the distillate with 10 mM HCl. Results were expressed as mg TVB-N kg-1 muscle.
Trimethylamine-nitrogen (TMA-N) values were determined by the picrate method, as previously described by Tozawa et al. (1971).This involved the preparation of a 50 g l-1 aqueous trichloroacetic acid extract of fish muscle (10 g/25 ml). Results were expressed as mg TMA-N kg-1 muscle.
Nucleotides were obtained by extraction with 60 g l-1 aqueous perchloric acid solution and analysed by HPLC according to Aubourg et al. (2005). Standard curves for adenosine 5’-triphosphate (ATP) and each compound involved in its degradation pathway, adenosine 5’-diphosphate (ADP), adenosine 5’-monophosphate (AMP), inosine 5’-monophosphate (IMP), inosine (INO) and hypoxanthine (HX), were constructed in the 0-1 mM range. Results obtained for each degradation compound were calculated as mmol kg-1 muscle. The K value was determined according to the following concentration ratio: K value (%) = 100 x (INO + HX) / (ATP + ADP + AMP + IMP + INO + HX).
2.6. Statistical analysis
Data obtained from the different microbial and chemical analyses were subjected to the ANOVA method to explore differences in two ways: icing condition effect and chilling time effect. For these analyses, the PASW Statistics 18 software for Windows (SPSS Inc., Chicago, IL, USA) was employed. The comparison of means was performed using the least-squares difference (LSD) method. Differences between batches were considered significant for a confidence interval at the 95% level (p<0.05) in all cases.
3. RESULTS AND DISCUSSION
3.1. Microbiological analyses
The evolution of aerobic mesophiles in hake and megrim muscle throughout the storage time under study is shown in Figure 1. Statistically significant (p<0.05) lower mesophile counts were determined in hake and megrim stored in the treated batch (T), compared with the control batch (C). Remarkably, the average differences after 15 d of storage for hake and megrim between C and T batches were 0.63 log CFU g-1in both cases. With respect to the anaerobes, the T batch also showed lower counts than the C batch both in hake and megrim (Figure 2). Statistically significant (p<0.05) differences between T and C batches were determined in hake. Remarkably, differences up to 1.98 log CFU g-1in the anaerobe counts between T and C hake batches were observed on day 9, while such differences decreased to 0.95 log CFU g-1 units on day 15. With respect to megrim, the T batch exhibited a better control of anaerobes at all storage times, although the differences with respect to the C batch were only statistically significant (p<0.05) on day 9.
The development of psychrotrophs exhibited statistically significant (p<0.05)differences between T and C batches in the case of hake on day 9 (Figure 3). Moreover, the psychrotroph counts were lower in the hake T batch at all storage times compared with the C batch. In the case of megrim, statistically significant (p<0.05) differenceswere also observed on days 12 and 15 (Figure 3). With respect to proteolytic bacteria, statistically significant (p<0.05) differences were observed between T and C batches only in the case of hake after 15 d of storage (Table 1). In the case of megrim, no statistically significant (p>0.05) differencesbetween batches were observed at any storage time (Table 1). However, the megrim T batch exhibited lower counts of proteolytic bacteria than the C batch, with the highest differences between batches being determined on day 12 (0.43 log CFU g-1 units). The role of proteolytic bacteria in the spoilage of fish muscle has been reported (Rodríguez et al., 2003). Accordingly, the inhibition of this microbial group in hake muscle by the acids included in the icing system is a remarkable result in terms of fish quality.
Finally, the Enterobacteriaceae counts revealed statistically significant (p<0.05) differences between hake T and C batches at all storage times (Table 1); the highest difference between both batches (0.97 log CFU g-1 units) was observed on day 9. With respect to megrim muscle, lower counts of Enterobacteriaceae were determined in the T batch at all storage times (Table 1), although such differences were not statistically significant (p>0.05).
The results of the microbiological analysis indicated that the T batch, corresponding to the icing system that included 1.25 g l-1 of CA and 0.50 g l-1 of LA, significantly (p<0.05)slowed down the growth of all microbial parameters analysed in the case of hake, and this led to an improved microbial quality of this fish species even after 15 d of storage. Indeed, the aerobic mesophiles, anaerobes, proteolytic bacteria and Enterobacteriaceae did not reach 7, 5, 4 and 5 log CFU g-1 units in the hake T batch, respectively, while the C batch was above such numbers. In the case of megrim, a similar significant (p0.05)beneficial effect in the presence of both acids in the icing system was observed for aerobic mesophiles and psychrotrophs. Therefore, and according to the results of the microbial analysis, the presence of organic acids in the icing system evaluated in this study significantly slowed down the growth of certain microbial groups in fish muscle, this result being especially relevant for hake and to a lesser extent, for megrim.
This microbial activity inhibition agrees to previous research carried out onshore, where both species were kept under the same icing conditions (García-Soto et al., 2013a, 2013b); in such experiments, 0-13-day storage periods were analysed for both species. Likewise, other authors reported the effectiveness on microbial activity inhibition by including preservative compounds and natural compounds in the icing system. These account for ozone during on-board chilled storage of megrim (Pastoriza et al., 2008),rosemary extract during chilling storage of sardine (Sardinella aurita) (Özyurt et al., 2012) and wild-thyme hydrosol in chilled Transcaucasian barb (Capoeta capoeta capoeta) (Oral et al., 2008).
Previous studies performed with ice slurries prepared from marine water indicated that when the microscopic ice crystals melt, the salt solution exerts a washing effect of the fish surface, which reduces the microbial load of the fish surface, thus reducing microbial diffusion through the skin towards the muscle (Campos et al., 2005; Aubourg et al., 2006). Likewise, the melting of the ice flakes prepared with CA and LA in the present study may exert a similar washing effect, which would also prevent the formation of biofilms in the fish surface, thus limiting fish spoilage.
3.2. Chemical analyses
Volatile amine formation was observed by measuring the TVB-N and TMA-N contents (Table 2). In all types of fish samples, increasing values for both parameters could be observed by increasing the chilling time (p<0.05). In the storage period analysed, this increase was found more relevant for trimethylamine (TMA) formation.
TVB-N values did not differ when acid was present in the ice (p>0.05). Our results indicate that none of the batches analysed reached the legal limit of 300-350 mg kg-1 that was set for this index (Directive 95/149/EEC) (Baixas-Nogueras et al., 2003). Concerning the TMA assessment, lower mean values were obtained in fish corresponding to the acid-icing treatment; differences were found significant at days 12 and 15 for hake and at day 12 for megrim. It is worth pointing out that hake stored under traditional ice surpassed the legal limit established for this species (5 mg kg-1) (Baixas-Nogueras et al., 2003), while megrim samples did not attain its legal limit (12 mg kg-1; Directive 91/493/EEC) (Aubourg et al., 2006).
Volatile amine compounds are produced partially by means of endogenous enzyme activity but mostly as a result of microbial development (Whittle et al., 1990). In the present case, agreement was found between microbiological (aerobes, anaerobes, psychrotrophs, proteolytics and Enterobacteriaceae) and chemical (TVB-N and TMA-N) parameters related to microbial activity development during the chilled storage in both fish species.
Previous research shows the inhibitory effect on amine formation as a result of including a preservative component in the icing system. Thus, the presence of wild-thyme tyrosol in the icing medium led to a marked decrease in TVB-N content formation during chilled storage of Transcaucasian barb (Oral et al., 2008).However, the inclusion of a rosemary extract in the icing medium did not lead to a TVB-N content decrease, but produced a lower formation of biogenic amines during the chilled storage of sardine (Özyurt et al., 2012). Oregano and rosemary extracts were also included successfully in the icing medium during Chilean jack mackerel (Trachurus murphyi) chilled storage (Quitral et al., 2009); thus, a lower TVB-N content was detected in fish stored under plant extract icing. A lower amine formation (TVB-N and TMA-N) was also obtained in hake and megrim by applying the present icing conditions during a 0-13-day storage period (García-Soto et al., 2013a, 2013b).