Biogenic Amine Index for Freshness Evaluation in Iced Mediterranean Hake (Merluccius Merluccius)

2433

Journal of Food Protection, Vol. 68, No. 11, 2005, Pages 2433±2438 Copyright ᮊ, International Association for Food Protection

Biogenic Amine Index for Freshness Evaluation in Iced Mediterranean Hake (Merluccius merluccius)

S. BAIXAS-NOGUERAS, S. BOVER-CID, M. T. VECIANA-NOGUEÂ S, A. MARINEÂ -FONT, AND M. C. VIDAL-CAROU*

Departament de NutricioÂ i BromatologiaÐCeRTA, Facultat de FarmaÁcia, Universitat de Barcelona, Avinguda Joan XXIII s/n, E-08028 Barcelona, Spain

MS 05-32: Received 31 January 2005/Accepted 1 June 2005 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/11/2433/1677143/0362-028x-68_11_2433.pdf by guest on 29 September 2021 ABSTRACT

Biogenic amine accumulation was studied during the ice storage of Mediterranean hake. Sensory analysis and counts of Shewanella, Pseudomonas, enterobacteria, psychrotrophic, and mesophilic bacteria provided complementary information on hake spoilage. Putrescine and cadaverine were the main amines accumulated, whereas histamine and tyramine were minor amines but had qualitative interest from the hygienic point of view. Although all biogenic amines were less abundant than in pelagic ®sh, they may also be used as indicators of freshness and/or spoilage in hake. Cadaverine was the amine best correlated with Shewanella, which was the speci®c spoilage organism. Therefore, cadaverine may be regarded as the speci®c spoilage biogenic amine for hake stored at chilling temperatures. However, the biogenic amine index, which considers cadaverine, putrescine, histamine, and tyramine, has several advantages as an indicator of hake quality. Taking into account sensory data, an acceptability limit of the biogenic amine index could be established in 15 to 20 ␮g/g.

Biogenic amines (BAs) are nonvolatile compounds Food and Drug Administration lowered it in 1995 from 100 whose formation requires the presence of decarboxylase- to 50 mg/kg, recommending that not only the histamine positive microorganisms, the availability of free amino ac- level but also other BA contents should be taken into acids, and conditions that allow bacterial growth and decar- count (29). boxylase activity (15, 18). BAs in foods are related to var- In addition to the toxicological implications, BAs are ious health problems, such as histamine intoxication, mi- related to food spoilage, since they are absent or occur at graine, and food intolerance crises. In general, the adverse low levels in fresh ®sh and accumulate as a result of the effects of BAs are due to the high contents of these com- proteolytic and amino acid decarboxylase activity of un- pounds, especially in conjunction with potentiating factors desired contaminant microbial ¯ora (7, 18). Bacterial activ- such as monoamine oxidase inhibiting drugs, alcohol in- ity is the main cause of ®sh spoilage. Nonfermenting gram- take, gastrointestinal diseases, and other food amines (15). negative bacteria have been described as the main microbial The enzyme diamine oxidase, which plays a key role in the group in fresh ®sh, such as Pseudomonas, Shewanella, and human histamine detoxi®cation system, can be genetically enterobacteria. In particular, Shewanella has been reported de®cient in some individuals (33, 35) or inhibited by some (8) as the most proli®c microorganism and the speci®c drugs (34). Thus, undesirable effects, including intestinal spoilage organism. Speci®c relationships between microbial symptoms and headaches, have been described after the groups and amine build-up have been widely studied in consumption of food that contains BAs, especially hista- pelagic ®sh but less is known in lean ®sh and particularly mine, even at low levels. hake. Histamine intoxication, also known as scombroid poi- Mainly histamine but also other BAs have been pro- soning, is the most common health concern related to BAs, posed as markers of ®sh freshness, especially for pelagic particularly histamine, in ®sh (22). It is usually reported species (20, 31, 36). They are promising potential indicators of the degree of hake freshness (3, 5). We aimed to study after the consumption of ®sh belonging to the Scombridae the suitability of BAs, considered either individual or a or Scomberesocidae families. This explains why most stud- combination, as chemical markers of freshness in the over- ies on BAs in ®sh focus on pelagic species, such as tuna, all evaluation of hygienic hake quality. Thus, BA contents mackerel, sardine, and anchovy (7, 14, 17, 19, 20, 32). were monitored throughout the ice storage of Mediterra- However, some authors have also described noticeable nean hake (Merluccius merluccius) in the chilling condi- amounts of BAs in nonpelagic ®sh (21, 24, 26, 30). Al- tions as applied in the merchandising chain. Microbial though several BAs have been associated with ®sh spoilage, counts were also monitored with the aim of studying wheth- a legal limit has only been established for histamine. The er general or speci®c relationships between them and BAs European Union set a maximum average content of 100 could be established for the particular case of hake. mg/kg for fresh and canned products (10), whereas the U.S. MATERIALS AND METHODS * Author for correspondence. Tel: ϩ34-93-402 45 13; Fax: ϩ34-93-403 Samples. Several trials of ice storage were performed during 59 31; E-mail: [email protected]. the following months: May 2000 (trial 1), December 2000 (trial 2434 BAIXAS-NOGUERAS ET AL. J. Food Prot., Vol. 68, No. 11

2), July 2001 (trial 3), and July 2002 (trial 4). Hake samples were ®shed in the Mediterranean coast and stored in ice until being sold. Fish was acquired gutted at a local Barcelona market and delivered directly to the laboratory, always kept in ice. The average weight of individuals was 350 Ϯ 6 g, and their average length was 35 Ϯ 3 cm. Few ®sh were immediately separated and used as time zero samples. All the remaining samples were stored in ¯ake ice (0ЊC) in self-draining boxes, under refrigeration at 4 to 6ЊC, until analysis. The melted ice was replaced daily. Fish samples were taken throughout storage on days 2, 6, 8, 10, and 14. Sensory, microbiological, and chemical determinations were performed in duplicate in all samples.

Sensory analysis. The quality index (QI) method for hake Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/11/2433/1677143/0362-028x-68_11_2433.pdf by guest on 29 September 2021 described by Baixas-Nogueras et al. (2) was applied to assess the quality attributes of raw hake. A panel of 8 trained members scored the general appearance (surface appearance and tissue soft- ness), eyes (clarity of the cornea, shape, and color), and gills (col- or and smell). The QI values were obtained from the quotient between the demerit points scored and the total demerit points. FIGURE 1. Pseudomonas, Shewanella, Enterobacteriaceae, me- Thus, samples were scored from 0 (excellent quality) to 1 (ex- sophilic bacteria, and psychrotrophic bacteria counts of Mediter- tremely spoiled), the limit of acceptability being a QI of 0.60 to ranean hake during storage on ice. 0.65 (2).

Microbiological analysis. To obtain the 10Ϫ1 dilution, the skin was aseptically removed and 10 g of ®sh muscle was blended RESULTS AND DISCUSSION with 90 ml of sterile peptone water for 1 min in a sterile stomacher Sensory evaluation and microbiological counts. The bag using a Seward 400 stomacher (Seward Limited, Thetford, initial value for the QI of raw hake was zero and increased Norfolk, UK). Serial dilutions were obtained by mixing 1-ml al- gradually and similarly in all trials, reaching a ®nal average iquots with 9 ml of peptone water. value of 0.85 (SD ϭ 0.03) after 14 days of ice storage. The Total viable mesophilic and psychrotrophic bacteria were enumerated using spread plates of plate count agar incubated at QI values collected from all trials correlated well with the 2 ϭ Ͻ 32 and 20ЊC, respectively, for 72 h (23). Enterobacteriaceae time of ice storage (r 0.978; P 0.001). In the con- counts were determined by using violet red bile glucose agar with ditions applied, sensory rejection of hake occurred at day a double layer with incubation at 37ЊC for 24 h (23). Pseudo- 10, when samples surpassed the QI value of 0.60 to 0.65 monas on cetrimide agar at 30ЊC for 72 h (23) and random col- established as the limit of acceptability (2). onies were picked for identi®cation with API 20NE. Shewanella, Initial mesophilic and psychrotrophic counts of 3 to 4 as H2S-producing bacteria, was counted from the black colonies log CFU/g increased throughout the ice storage to approx- grown using spread plates of iron agar Lyngby after incubation at imately 7 log CFU/g after 14 days. As Figure 1 shows, 20ЊC for 72 h (12). Black colonies from iron agar Lyngby were psychrotrophic counts were slightly higher than mesophilic picked to be con®rmed as Shewanella using API 20NE. All media counts, but differences were not signi®cant (P Ͼ 0.05). Re- used were obtained from Oxoid (Barcelona, Spain). garding Enterobacteriaceae, levels slowly increased from BA analysis. BAs were analyzed using the high-performance not detected to values close to 4 log CFU/g. Samples liquid chromatographic method described by Veciana-NogueÂs et showed higher mesophilic bacteria counts than the legal al. (30), which is based on the formation of ion pairs between regulated limit of acceptability of 106 CFU/g on day 10, BAs and octanesulfonic acid present in the mobile phase. Their whereas the counts of Enterobacteriaceae surpassed the 3 separation was performed through a C18 reverse-phase column. limit of 10 CFU/g (21) on day 8. Post±column derivatization with o-phthalaldehyde was followed Pseudomonas were present already in fresh hake sam- by spectro¯uorometric detection. ples at approximately 2 log CFU/g. A slight and gradual but signi®cant (P Ͻ 0.001) increase near 5 log CFU/g was Statistical analysis. Data were statistically evaluated using observed during ice storage. Shewanella was not detected the analytical procedures of SPSS 10.0 for Windows software (SPSS Inc., Chicago, Ill.). Nonparametric tests were applied when in samples from zero time, although it showed an early Ͻ data were not normally distributed (signi®cant Shapiro-Wilks signi®cant steep growth (P 0.001) during the 2 ®rst days test). The BA values and bacterial counts between days were com- of storage and reached more than 6 log CFU/g after 14 days pared by the Mann-Whitney U test. Regression analysis was ap- of storage; they were thus the dominant bacterial group plied to correlate changes in BA contents and bacterial counts. A after 1 week of ice storage of Mediterranean hake. Dalgaard receiver operating characteristic curve (9) was applied to evaluate (8) also described the predominance of Shewanella during the suitability of BAs as indexes of hake hygienic quality (from the ice storage of hake. the value of the area under the curve [AUC]) in reference to sensory established limit. The accuracy of the possible BA limits of Biogenic amines. The physiological polyamines sper- acceptability was evaluated from AUC sensitivity and speci®city midine and spermine were always the major amines found values. in fresh samples, the spermine contents being always higher J. Food Prot., Vol. 68, No. 11 BIOGENIC AMINE INDEX TO EVALUATE HAKE FRESHNESS 2435

TABLE 1. Spermine and spermidine content of Mediterranean hake during storage on ice for four trialsa Spermine (␮g/g) Spermidine (␮g/g)

Trial Initial value Final value Initial value Final value

1 4.60 (3.70) 15.00 (2.31) 3.48 (0.48) 10.66 (3.83) 2 55.35 (3.60) 31.42 (5.71) 12.87 (0.95) 11.08 (1.21) 3 12.10 (2.00) 9.50 (1.55) 1.84 (0.33) 2.56 (0.83) 4 11.84 (2.68) 9.48 (1.54) 2.22 (0.89) 2.51 (0.63) a Values are mean (standard deviation). than those of spermidine, as reported for food of animal thine-decarboxylase activity of several bacterial groups re- origin (6). sponsible for ®sh spoilage (4, 28). Statistical correlations Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/11/2433/1677143/0362-028x-68_11_2433.pdf by guest on 29 September 2021 Spermidine and spermine showed a wide range of con- between putrescine and microbial counts indicated that the centrations, depending on the trial (Table 1). Our results most satisfactory correlation was with Pseudomonas (r ϭ did not allow us to draw a de®nitive conclusion about the 0.810, P Ͻ 0.001). As reported by Baixas-Nogueras et al. behavior of the polyamines spermidine and spermine during (4), several Pseudomonas isolates (such as Pseudomonas hake storage. Indeed, data found in the literature also show ¯uorescens, Pseudomonas putida, and Pseudomonas ce- a large variability in contents and pro®les (13, 25). Poly- pacea) from iced Mediterranean hake were able to produce amines are of endogenous origin and not produced by mi- putrescine. croorganisms. Trials 2, 3, and 4 in which polyamine levels Initial cadaverine values were lower than 1 ␮g/g or not did not change or diminished would be in accordance with detected but signi®cantly increased thereafter without dif- this fact but not trial 1, which showed an increase in sperm- ferences among trials, although with a notable variability ine and spermidine levels. Indeed, it has been suggested depending on the sample (Table 2). Cadaverine values dur- that microorganisms can consume polyamines as a nitrogen ing ice storage ®t a linear trend (r2 ϭ 0.635; P Ͻ 0.001). or carbon source (6). However, taking into account the wide Levels of cadaverine were clearly higher than those of pu- range of polyamine contents in hake, they may strongly trescine throughout the storage. However, the levels of pu- depend on the sample rather than the effect of the storage. trescine and cadaverine at the end of the study did not differ In addition, a seasonal variability could be suspected, since signi®cantly. the polyamine amounts were higher in trial 2, which was Moreover, cadaverine was the BA most satisfactorily performed in the winter, than in the other trials. The sea- correlated (P Ͻ 0.002) with the growth of Shewanella pu- sonal differences could be linked to changes in the physi- trefaciens, the speci®c spoilage organism of chilled hake. ological cycle, in accordance with the role of polyamines The earlier and higher cadaverine production compared in growing cells (6, 18). with the other BAs, together with the strong association Putrescine was also detected in all samples at the be- observed with S. putrefaciens development, led to the sug- ginning of storage, although at much lower amounts than gestion that this amine is the speci®c spoilage biogenic polyamines. Putrescine increased during ice storage in all amine in hake stored in ice. The high amount of the pre- trials (Table 2) without signi®cant differences (P Ͼ 0.05). cursor lysine in fresh hake could contribute to the high The best pattern to describe the putrescine accumulation in production of cadaverine as shown elsewhere (24). hake during the ice storage was an exponential curve (r2 The concentration of agmatine was initially very low ϭ 0.738; P Ͻ 0.001). The occurrence of low amounts of or not detected and showed different behavior, depending putrescine in fresh hake samples has also been described on the trial (Figure 2). In trial 2, performed in the winter, elsewhere and is related to its physiological function in liv- agmatine contents did not change throughout ice storage, ing ®sh muscle as the precursor of polyamines, spermidine, whereas in trials 1 and 4, after a stable initial period, ag- and spermine (1). In contrast, the accumulation of this di- matine contents gradually increased. Finally, in trial 3 ag- amine during storage could be associated with the orni- matine achieved 50.3 ␮g/g at day 10 and then decreased.

TABLE 2. Biogenic amines content of Mediterranean hake during storage on icea Biogenic amines (␮g/g) Days in ice Putrescine Cadaverine Histamine Tyramine

0 1.74 (1.18) 0.86 (0.42) 0.12 (0.21) 0.57 (0.06) 2 1.34 (0.69) 3.41 (2.80) 0.88 (0.61) 0.63 (0.05) 6 1.64 (0.91) 5.06 (3.14) 1.15 (0.29) 0.63 (0.11) 8 3.01 (0.77) 7.68 (7.59) 2.28 (1.80) 4.51 (2.76) 10 5.67 (4.00) 16.34 (9.80) 3.14 (1.41) 6.08 (5.54) 14 12.22 (9.40) 20.30 (13.19) 2.20 (0.74) 2.44 (1.44) a Values are mean (standard deviation). 2436 BAIXAS-NOGUERAS ET AL. J. Food Prot., Vol. 68, No. 11

amounts of histamine (higher than 2 ␮g/g) appeared before the rejection point established by sensory analysis. There- fore, despite the lack of toxicological signi®cance, the occurrence of histamine, even at such low levels, could be regarded as qualitatively meaningful from the hygienic point of view, indicating a noticeable decrease of hake freshness. Moreover, since higher amounts of histamine are accumulated during storage at abusive temperature (5), the occurrence of histamine in stored hake could be useful to detect a possible failure of the chilling conditions throughout the merchandising chain. In any case, the maximum acceptable level of histamine should be revised, particularly for hake. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/11/2433/1677143/0362-028x-68_11_2433.pdf by guest on 29 September 2021 The levels of tyramine remained low during the ®rst week of ice storage, showing a similar trend than those of FIGURE 2. Agmatine content (␮g/g) of Mediterranean hake dur- histamine (Table 2). Like histamine, tyramine was a minor ing storage on ice for four trials. amine from a quantitative point of view, but its occurrence in hake should not be undervalued, because it is also related to ®sh spoilage, as also reported in other works (3, 31). This peaking behavior has also been described during the In general, the use of more than a single BA (i.e., a storage of pelagic ®sh (30, 31, 36). The peak pro®le may BA index that consists of a combination of BAs) is rec- be due to the fact that agmatine is an intermediate metab- olite in the putrescine formation pathway from arginine ommended to avoid the limitation of a possible variability (21). Therefore, the use of agmatine does not seem to be in the content of one amine and seems more appropriate as appropriate to evaluate hake freshness, because it did not a quality indicator. Some examples are the sum of cadav- show a regular pattern of change throughout ice storage. erine and putrescine (26), the index of Mietz and Karmas The levels of histamine (Table 2) were relatively low (20), which considers the increases in putrescine, cadaver- (Ͻ3.5 ␮g/g), and in trial 2 they were not detected. The ine, and histamine levels with decreases in spermidine and accumulation of histamine is usually related to spoilage spermine, and the index described by Veciana-NogueÂs et al. processes and to the histaminogenic activity mainly attri- (31) for tuna, which includes putrescine, cadaverine, his- buted to enterobacteria (16). The lack of histamine in trial tamine, and tyramine. 2 is consistent with the observation that enterobacteria did The contents of BAs at the point of rejection by sen- not grow. A statistical correlation (r ϭ 0.779, P Ͻ 0.001) sory and microbial limiting values of hake stored in ice (8 was found between histamine amounts and Enterobacteri- to 10 days) were within the range of 8 to 16 ␮g/g for aceae counts. Histamine levels were lower than those de- cadaverine, 3 to 5 ␮g/g for putrescine and histamine, and scribed in pelagic ®sh, always remaining far below the es- 4to6␮g/g for tyramine in the four trials. The contents of tablished limit of 50 to 100 mg/kg (10, 29). Indeed, al- histamine at the time of rejection in hake were lower than though the toxicological effect of histamine can be en- those reported in tuna (31). However, the levels of the other hanced by diamines, histamine intoxication due to the BAs, such as putrescine, cadaverine, and tyramine, in hake consumption of hake may be rare. Our results support that were similar or slightly higher than those described for tuna histamine does not seem to constitute a serious hazard in (31). This indicates that hake is also a ®sh species that hake, at least when it is stored at chilling temperature. The shows enough BA accumulation during storage to be con- optimum temperature for histidine-decarboxylase activity sidered suitable as a freshness or quality indicator. has been reported to be 20 to 30ЊC (11, 14), and then the The indexes proposed by Mietz and Karmas (20), the ice storage may prevent histaminogenic activity. Moreover, sum of putrescine plus cadaverine plus histamine plus ty- hake contains low amounts of free histidine (24), which ramine, together with the use of the sum of the major may also limit the formation of this amine. Detectable amines putrescine plus cadaverine (26) and cadaverine (36)

TABLE 3. Area under the curve (AUC) and statistical limit of acceptability resulted from the construction of receiver operating characteristic curves of biogenic amine indexes with sensory results and experimental limit of acceptability Statistical limit Experimental limit Reference Biogenic amine indexa AUC of acceptability of acceptability

20 (PU ϩ CA ϩ HI)/(1 ϩ SM ϩ SD) 0.83 0.40 (0.85; 0.91)b 0.45±0.81 31 PU ϩ CA ϩ HI ϩ TY 0.89 18 (0.75; 0.91) 15.98±18.27 26 CA ϩ PU 0.85 8.50 (0.87; 0.85) 10.69±22.01 36 CA 0.83 8 (0.75; 0.87) 7.68±16.34 a PU, putrescine; CA, cadaverine; HI, histamine; SM, spermine; SD, spermidine; TY, tyramine. b Limit of acceptability (sensitivity; speci®city). J. Food Prot., Vol. 68, No. 11 BIOGENIC AMINE INDEX TO EVALUATE HAKE FRESHNESS 2437

appearance depends on many factors, such as handling procedures and spoilage ¯ora, a BA index limit of acceptability may be established in a range of 15 to 20 ␮g/g. However, due to the high variability, further trials are needed to con- ®rm this value and to extrapolate the conclusions to all practical situations concerning chill-stored hake.

ACKNOWLEDGMENTS This research was supported by the European Union project FAIR CT.97.3253 and the ComissioÂ Interdepartamental de Recerca i InnovacioÂ TecnoloÁgica (CIRIT, 2001SGR000132) of the Generalitat de Catalunya.

REFERENCES Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/11/2433/1677143/0362-028x-68_11_2433.pdf by guest on 29 September 2021 1. Ababouch, L., M. E. A®lal, H. Benabdeljelil, and F. F. Busta. 1991. Quantitative changes in bacteria, amino acids and biogenic amines FIGURE 3. Biogenic amine index (BAI) evolution of Mediterra- in sardine (Sardina Pilchardus) stored at ambient temperature (25± nean hake during storage on ice for four trials. 28ЊC) and in ice. Int. J. Food Sci. Technol. 26:297±306. 2. Baixas-Nogueras, S., S. Bover-Cid, M. T. Veciana-NogueÂs,M.L. Nunes, and M. C. Vidal-Carou. 2003. Development of a quality in- values were examined through the building of receiver op- dex method to evaluate freshness in Mediterranean hake (Merluccius erating characteristic curves using the sensory analysis as merluccius). J. Food Sci. 68:1067±1071. a reference parameter. The AUC indicated the suitability of 3. Baixas-Nogueras, S., S. Bover-Cid, M. T. Veciana-NogueÂs, and M. C. Vidal-Carou. 2002. Chemical and sensory changes in Mediter- one variable to distinguish between two states, such as fresh ranean hake (Merluccius merluccius) under refrigeration (6±8 ЊC) and deteriorated hake, whereas the values of sensitivity and and stored in ice J. Agric. Food Chem. 50:6504±6510. speci®city de®ned the best limit of acceptability. Table 3 4. Baixas-Nogueras, S., S. Bover-Cid, M. T. Veciana-NogueÂs, and M. shows the results obtained in each case, together with the C. Vidal-Carou. 2003. Amino acid-decarboxylase activity in bacteria experimental limit of acceptability from the shelf life es- associated with Mediterranean hake spoilage. Eur. Food Res. Tech- nol. 217:164±167. tablished in hake according to sensory and microbiological 5. Baixas-Nogueras, S., S. Bover-Cid, M. C. Vidal-Carou, and M. T. criteria. The limits of acceptability for each index deduced Veciana-NogueÂs. 2001. Volatile and nonvolatile amines in Mediter- from the statistical analysis and experimental data obtained ranean hake as a function of their storage temperature. J. Food Sci. at rejection time were compared. The statistical values ob- 66:83±88. tained from the index of Mietz and Karmas and the sum 6. BardoÂcz, S. 1995. Polyamines in food and their consequences for food quality and human health. Food Sci. Technol. 6:341±346. (putrescine plus cadaverine) were out of the range of the 7. Ben-Gigirey, B., J. M. Vieites Baptista de Sousa, T. G. Villa, and J. experimental data, whereas the statistical limit of accept- Barros-VelaÂzquez. 1998. Changes in biogenic amines and microbi- ability obtained for the index (putrescine plus cadaverine ological analysis in albacore (Thunnus alalunga) muscle during fro- plus histamine plus tyramine) and the cadaverine levels zen storage. J. Food Prot. 5:608±615. were within the experimental range. From these two last 8. Dalgaard, P. 1995. Qualitative and quantitative characterization of spoilage bacteria from packed ®sh. Int. J. Food Microbiol. 26:319± values, the sum (putrescine plus cadaverine plus histamine 333. plus tyramine) showed a more satisfactory value of AUC 9. DeLong, E., D. DeLong, and D. Clarke-Pearson. 1988. Comparing according to the criterion of Swets (27). This index con- the areas under two or more correlated receiver operating character- templates the four amines that are most relevant in hake istic curves: a nonparametric approach. Biometrics 44:837±845. spoilage from the quantitative and qualitative point of view. 10. Directiva CEE. 1991. Directiva de 22 de Julio de 1991 por la que se ®jan las normas aplicables a la produccioÂn y puesta en el mercado Figure 3 shows the evolution of the BA index observed in de los productos pesqueros (91/439/CCE). Diario O®cial de las Co- the four trials performed, which followed a gradual increase munidades Europeas. L268:15±34. during ice storage. The BA index values of trial 2 were 11. Fletcher, G. C., G. Summers, R. V. Winchester, and R. J. Wong. lower than values in other trials, agreeing with the seasonal 1995. Histamine and histidine in New Zealand marine ®sh and shell- differences in amine formation reported herein and proba- ®sh species, particularly kahawai (Arripis trutta). J. Aquat. Food Prod. Technol. 42:53±74. bly related to the lower microbial load of hake in winter. 12. Gram, L., G. Trolle, and H. H. Huss. 1987. Detection of speci®c However, as reported herein, no statistical differences were spoilage bacteria from ®sh stored at low (0ЊC) and high (20ЊC) tem- found in bacterial counts among trials. In addition, differ- peratures. Int. J. Food Microbiol. 4:65±72. ences among trials were especially noticeable only after 10 13. HernaÂndez-Jover, T., M. Izquierdo-Pulido, M. T. Veciana-NogueÂs, A. days of ice storage when hake was clearly sensory rejected. MarineÂ-Font, and M. C. Vidal-Carou. 1997. Biogenic amine and polyamine contents in meat and meat products. J. Agric. Food Chem. Therefore, this variability had a low effect on the BA index 45:1098±2102. value proposed to reject hake. In any case, the limit of 50 14. Koutsoumanis, K., and G. Nychas. 1999. Chemical and sensory ␮g/g reported for tuna (31) was not reached in any trial changes associated with microbial ¯ora of Mediterranean boque even when hake was rejected according to both sensorial (Boops boops) stored aerobically at 0, 3, 7 and 10 degree C. Appl. and microbiological criteria, suggesting that this limit Environ. Microb. 65:698±706. 15. Lehane, L., and J. Olley. 2000. Histamine ®sh poisoning revisited. should be speci®cally reestablished for hake. According to Int. J. Food Microbiol. 58:1±37. the statistical acceptability limits and the experimental data 16. LoÂpez-Sabater, E. I., J. J. RodrõÂguez-Jerez, M. HernaÂndez-Herrero, and taking into account that the variability of ®sh spoilage A. X. Roig-SagueÂs, and M. T. Mora-Ventura. 1995. Sensory quality 2438 BAIXAS-NOGUERAS ET AL. J. Food Prot., Vol. 68, No. 11

and histamine formation during controlled decomposition of tuna 27. Swets, J. A. 1988. Measuring the accuracy of diagnostic systems. (Thunnus thynnus). J. Food Prot. 59:167±174. Science 314:286±293. 17. Malle, P., M. ValleÂ, and S. Bouquelet. 1996. Assay of biogenic 28. Taylor, S. L., and S. S. Summer. 1987. Determination of histamine, amines involved in ®sh decomposition. J. AOAC Int. 79:43±49. putrescine and cadaverine, p. 235±245. In D. E. Kramer and J. Liston 18. MarineÂ-Font, A., M. C. Vidal-Carou, M. Izquierdo-Pulido, M. T. (ed.), Seafood quality determination. Elsevier, New York. Veciana-NogueÂs, and T. HernaÂndez-Jover. 1995. Les amines biogeÁ- 29. U.S. Food and Drug Administration. 1995 Decomposition and his- nes dans les aliments: leur signi®cation, leur analyse. Ann. Fals. Exp. tamine; raw, frozen tuna and mahi-mahi; canned tuna; and related Chim. Toxicol. 88:119±140. species; revisited compliance guide; availability. Fed. Regist. 149: 19. Mendes, R. 1999. Changes in biogenic amines of major Portuguese 39754±39756. blue®sh species during storage at different temperatures. J. Food 30. Veciana-NogueÂs, M. T., T. Hernandez-Jover, A. Marine-Font, and M. Biochem. 23:33±43. C. Vidal-Carou. 1995. Liquid chromatographic method for deter- 20. Mietz, J. L., and E. Karmas. 1981. Biogenic amines as indicators of mination of biogenic amines in ®sh and ®sh products. J. AOAC Int. seafood freshness. Lebensm. Wiss.-Technol. 14:273±275. 78:1045±1049. 21. Ministerio de Sanidad y Consumo. 1991. Orden de 2 de Agosto por 31. Veciana-NogueÂs, M. T., A. MarineÂ-Font, and M. C. Vidal-Carou. el que se aprueban las Normas microbioloÂgicas para los productos 1997. Biogenic amines as hygienic-quality indicators of tuna: rela- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/11/2433/1677143/0362-028x-68_11_2433.pdf by guest on 29 September 2021 de la pesca y de la acuicultura. Boletin O®cial del Estado 195: tionships with microbial counts, ATP related compounds, volatile 27153±27155. amines and organoleptic changes. J. Agric. Food Chem. 45:2036± 22. Morrow, J., G. Margolies, J. Rowland, and J. Roberts. 1991. Evi- 2041. dence that histamine is the causative toxin of scombroid-®sh poi- 32. Veciana-NogueÂs, M. T., A. MarineÂ-Font, and M. C. Vidal-Carou. soning. N. Engl. J. Med. 14:716±720. 1997. Biogenic amines in fresh and canned tuna: effects of canning 23. Pascual Anderson, M. R., and V. CalderoÂn Pascual. 2000. Micro- on biogenic amine contents. J. Agric. Food Chem. 45:4324±4328. biologõÂa alimentaria: metodologõÂa analõÂtica para alimentos y bebidas. 33. Wantke, F., M. GoÈtz, and R. Jarisch. 1994. The red wine provocation DõÂaz de Santos S.A., Madrid. test: intolerance to histamine as a model for food intolerance. Allergy 24. RuõÂz-Capillas, C., and A. Moral. 2001. Production of biogenic Proc. 15:27±32. amines and their potential use as quality control indices for hake 34. Wantke, F., W. Hemmer, M. Focke, W. Stackl, M. GoÈtz, and R. Jar- (Merluccius merluccius) stored in ice. J. Food Sci. 66:1030±1032. isch. 2001. Are adverse effects of sildena®l also caused by inhibition 25. RuõÂz-Capillas, C., and A. Moral. 2002. Effect of controlled and mod- of diamine oxidase? Urol. Int. 67:59±61. i®ed atmospheres on the production of biogenic amines and free 35. Wantke, F., W. Hemmer, T. HaglmuÈller, M. GoÈtz, and R. Jarisch. amino acids during storage of hake. Eur. Food Res. Technol. 214: 1996. Histamine in wine. Int. Arch. Allergy Immunol. 110:397±400. 476±481. 36. Yamanaka, H., K. Shiomi, and T. Kikuchi. 1989. Cadaverine as a 26. Stede, M., and J. Stockemer. 1981. Biogene amine in see®schen. potential index for decomposition of salmonoid ®shes. J. Food Hyg. Lebensm.-Wiss. Technol. 19:283±287. Soc. Jpn. 30:170±174.