Journal of Applied Microbiology 2002, 92, 941–950 Sensory, chemical and bacteriological changes during storage of iced squid (Todaropsis eblanae) T. Paarup1, J.A. Sanchez1, A. Moral1, H. Christensen2, M. Bisgaard2 and L. Gram3 1 Instituto del Frı´o (CSIC), Department of Science and Technology of Meat and Meat Products and Fish and Fishery Products, Ciudad Universitaria, Madrid, Spain, 2Department of Veterinary Microbiology, Royal Veterinary and Agricultural University, Frederiksberg, and 3Danish Institute for Fisheries Research, Department of Seafood Research, c/o Technical University of Denmark, Lyngby, Denmark 2001/299: received 8 October 2001, revised 29 November 2001 and accepted 19 December 2001 T . P A A R U P , J . A . S A N C H E Z , A . M O R A L , H . C H R I S T E N S E N , M . B I S G A A R D A N D L . G R A M . 2002. Aims: To relate sensory shelf-life of iced whole and gutted squid to bacterial growth and chemical changes. Methods and Results: Cooked mantles from whole and gutted individuals were rejected after 10 and 12 days of storage, respectively, due to ammoniacal off-odours. Rate of production of both ammonia and trimethylamine was highest in the whole lot. Agmatine, which was only present in trace amounts in freshly-caught squid, increased rapidly in both lots. The main microflora at the time of sensory rejection of iced whole squid included Gram-negative, motile and non-fermentative rods, which were psychrophilic and had a requirement for NaCl. 16S rDNA sequence analyses identified the strains as belonging to the genus Pseudoalteromonas. Shewanella putrefaciens, Pseudoalteromonas sp. and Pseudomonas sp. dominated in spoiled gutted squid. Identification of strains from the stomach and digestive gland of recently-captured squid showed that the main flora consisted of Photobacterium phosphoreum. Conclusions: Spoilage of iced squid is likely to result from a combination of autolytic and bacterial changes. Agmatine seems to be an excellent freshness indicator. Photobacterium phosphoreum may contribute to spoilage through activity in the digestive gland, followed by diffusion of volatile compounds and amines to the mantle. Significance and Impact of the Study: Due to the psychrophilic nature of P. phosphoreum and Pseudoalteromonas sp., spread-plating and low temperature incubation are recommended for bacteriological evaluation of iced squid. INTRODUCTION Decreased fish stocks, probably related to over-fishing, has led to increasing exploitation of under-utilized marine resources such as cephalopods, of which the order Teuthoidea (squid) is quantitatively the most important. In 1996, cephalopods made up 3Æ2% of the total catches compared with 2% in 1966 (FAO 1998), corresponding to approximately 3 and 0Æ85 million tonnes, respectively. The most valuable species belong to the orders Octopoda (octopus) and Sepioidea (cuttlefish), and to the family Correspondence to: Torsten Paarup, Instituto del Frı´o (CSIC), Ciudad Universitaria s/n, E-28040 Madrid, Spain (e-mail: tpaarup@if.csic.es). ª 2002 The Society for Applied Microbiology Loliginidae (‘inshore squid’) within the suborder Myopsida of Teuthoidea. However, the so-called oceanic squids are interesting from a commercial point of view because of their abundance. The species that belong to the family Ommastrephidae are considered most important, despite their relatively hard and elastic texture. Such species are mainly landed frozen as they are often captured by long-distance fishing. Nevertheless, other oceanic squid families have received practically no commercial attention, and it has been estimated that only 10% of the detected stocks are captured (Guerra 1996). The main reasons for this are probably the great dispersion of these oceanic species in the sea, lack of knowledge on technological characteristics, existence of species with an extremely high ammonia content (Iida et al. 1992), and a short iced shelf-life. 942 T . P A A R U P ET AL. Squid consumption is limited in large parts of the world, especially in North America and northern Europe, where this commodity is mainly commercialized as frozen (breaded) rings destined for frying (‘a la romana’). On the other hand, considerable amounts of squid are consumed in east and south-east Asia, where a wide range of squid products are commercialized, and in the Mediterranean countries, where squid is often commercialized ungutted in ice for household preparation of various dishes. The microbiology of fresh and spoiling fish has been extensively studied, as reviewed by Liston (1980), Hobbs and Hodgkiss (1982) and Gram and Huss (1996), but qualitative microbiological data on cephalopods are scarce as autolytic changes are believed to be the main reason for deteriorating sensory quality (LeBlanc and Gill 1984). Most studies on fresh chilled squid have been conducted on isolated sensory characteristics (Ohmori et al. 1975; Learson and Ampola 1977; Botta et al. 1979; Ke et al. 1979). Such studies have mainly focused on skin colour, which is a delicate quality parameter due the presence of chromatophores that easily break post mortem when exposed to direct contact with ice, leading to discoloration. Combined sensory and biochemical studies have been conducted by Ke et al. (1984, 1991), who elaborated a useful grading procedure for fresh Illex illecebrosus based on its content of total volatile bases, trimethylamine and free fatty acids, previously correlated with organoleptic results, and Yamanaka et al. (1987), who related the production of biogenic amines to the sensory quality of Todarodes pacificus. Only a few studies have covered sensory, chemical and microbiological changes (Slabyj and True 1981; Licciardello et al. 1985; Ohashi et al. 1991; Morales 1997), and microbiological studies have mainly focused on total aerobic counts. A better understanding of the spoilage mechanism could lead to measures for increasing shelf-life, which could stimulate the marketability of fresh chilled squid. In this work, the sensory, chemical and bacteriological changes of the squid mantle were studied during iced storage of whole and gutted individuals. The microbiological load and composition of the stomach contents, the digestive gland and the mantle of recently-caught squid were also studied. M A T E R I A LS A N D M E T H O D S Storage experiment and conditions The squid species Todaropsis eblanae was used in the present study. It was caught along the coast of Galicia (Spain) in winter and brought to the Instituto del Frı́o in ice within 48 h of its capture. Two lots that originated from the same catch were prepared. An eviscerated lot was prepared manually by squeezing out the viscera, head and arms, and briefly washing the inner and outer surfaces of the mantle with tap water; the whole lot was not subjected to any manipulation. All lots were stored in ice. Samples were analysed immediately upon arrival at the laboratory and after 4, 7 and 11 days of storage. Thus, sampling points reflected storage times of 2, 6, 9 and 13 days after catch. The squid-to-ice ratio was approximately 2:1 (crushed freshwater ice) and new ice was added at regular intervals to maintain the ratio. The individuals were separated from the ice with rough paper, which avoided direct contact between the individuals and the ice but permitted the passage of melt-water, as this storage method allows for effective chilling and protects chromatophores from physical damage. The storage temperature was measured at 3 h intervals by thermocouples inserted in the mantle cavity of both lots and in the cold chamber. Two mantles from each lot were subjected to bacteriological and chemical analyses, and three to sensory analysis, at each sampling point. In addition, pooled stomach contents from 10 individuals were subjected to bacteriological analyses, while duplicate samples of the digestive gland were analysed by chemical and microbiological methods at the time of arrival of the squid (approximately 48 h from the time of capture). Sensory analyses Sensory analyses were conducted, by six to eight panellists, on mantles which had previously been skinned, washed briefly in tap water, placed in plastic bags and cooked in boiling water for 30 min after sealing. The parameters odour of cook water, and flavour and texture of the samples, were awarded points on a scale of 0–10, with 10 expressing optimum quality and 4 the rejection limit. However, as the scale must be considered bipolar in the case of texture, the value of 5 expressed what the panellists perceived as normal squid texture, higher and lower values expressing a harder and softer texture, respectively; the rejection limits were 8 and 2, respectively. Mantles that had been frozen at the time of arrival of the squid were included as controls. Chemical analyses Trimethylamine (TMA) and trimethylamine oxide (TMAO) were determined by the methods of AOAC (1995) and Bystedt et al. (1959), respectively. Ammonia (NH3) and urea were determined using a commercial kit (Boehringer Mannheim, Mannheim, Germany). pH was measured using a Radiometer PHM 93 reference pH meter (Radiometer, Copenhagen, Denmark). Analyses of biogenic amines were carried out by mixing 25 g mantle muscle with 50 ml 7Æ5% trichloroacetic acid solution (Panreac). Samples were homogenized, centrifuged for 10 min at 25 000 g, and then passed through a 0Æ22 lm MFS nylon filter (Advantec MSF, Inc., Pleasanton, CA, USA). The concentration of ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 CHANGES DURING STORAGE OF ICED SQUID biogenic amines was determined in a Perkin-Elmer model 1022 HPLC (Perkin Elmer, Norwalk, CT, USA) fitted with a Pickering ALKIONTM ion-exchange column (4 · 150 mm, Pickering Labs, Inc., Mountain View, CA, USA) and a Pickering PCX 3100 post-column system, using o-phthalaldehyde (OPA, Pickering Labs) as the derivatizing reagent (Tracy et al. 1995). Mobile phases used were those commercially provided by Pickering Labs, with the Polyamine Application Kit (0052–0040) for biogenic amines analysis: K-600 potassium phosphate buffer, K-563 potassium phosphate buffer and K-130 potassium column regenerant. A gradient elution method (Pickering Labs., with slight adaptations (unpublished results) for determining agmatine and 2-phenyl-ethyl-amine) was programmed with flows of 0Æ8 ml min–1 for the mobile phases and 0Æ3 ml min–1 for the derivatizing reagent. The fluorescence detector (PerkinElmer) wavelengths were 330 nm (excitation) and 465 nm (emission). This procedure was used to determine six biogenic amines usually tested in foods (Tapia-Salazar et al. 2000), plus agmatine and 2-phenylethylamine which were quantified by comparison with the following external standards: agmatine sulphate salt (Sigma), cadaverine dihydrochloride (Sigma), histamine dihydrochloride (Sigma), 2-phenylethylamine hydrochloride (Sigma), putrescine dihydrochloride (Sigma), spermidine trihydrochloride (Sigma), spermine tetrahydrochloride and tyramine hydrochloride (Sigma). Bacteriological analyses From each mantle and digestive gland, 10 g samples were homogenized with 90 ml peptone water (0Æ1%) in a stomacher (400 lab blender, Seward, London, UK) and appropriate decimal dilutions were prepared. The stomach contents were pooled from 10 individuals in a sterile Petri dish. After manual mixing, 10 g were taken and diluted as described above. The following non-selective, indicative or selective media were used to monitor changes in the bacterial flora during storage. Spread-plated chilled Iron Agar Lyngby (ADSA Micro) supplemented with 0Æ5% NaCl (Panreac) and incubated at 15C for 5 days (Iron Agar with salt (IAS)) was used for aerobic plate counts. This medium was also used for counting luminous bacteria by observing the plates in a dark room. Pour-plated, double layered Iron Agar, incubated at 20C for 3 days, was used for enumeration of H2S-producing bacteria. Pseudomonas spp. were enumerated on Cetrimide-Fucidin-Cephaloridine (CFC) agar (Oxoid), incubated at 20C for 48 h. Isolation and identification Composition of the microflora was determined by isolating and identifying 20% of colonies from IAS (15C) plates. 943 Colonies were randomly isolated from plates with 30–250 colonies taking all colonies from a section of the plate. Prior to identification, colonies were subcultured in Veal Infusion Broth (0344–17–6, Difco, Detroit, Michigan, USA), supplemented with 0Æ5% NaCl (VIBS) and plated on IAS and incubated at 15C for 18–24 h or until visible growth was observed. Morphology and motility studies were carried out on pure cultures in VIBS grown at 15C by phase contrast microscopy. Gram-reaction was tested by the KOH-method (Gregersen 1978), cytochrome oxidase by the method of Kovacs (1956) and catalase on a sterile Petri dish using a drop of 3% H2O2. Unless otherwise stated, the following tests were carried out at 15C and reactions monitored for up to 7 days. Oxidative or fermentative metabolism of glucose was performed in the medium of Hugh and Leifson (H & L) (Hugh and Leifson 1953) supplemented with 0Æ5% NaCl. Cultures that showed neither fermentative nor oxidative reactions were tested in H & L, supplemented with half-strength artificial sea water (ASW, MacLeod 1968), to favour acidification by marine bacteria, and an extra open tube, in which peptone had been replaced with ((NH)4)2SO4, was added to permit acidification by Pseudomonas sp. TMAO reduction and H2S production were tested by the method of Gram et al. (1987). Some isolates were tested for resistance to vibriostaticum (150 lg, 0/129 discs, Oxoid) on IAS, ability to grow without NaCl in VIBS and ability to grow in VIBS at 1C during 21 days of incubation. Based on the reactions shown in Table 1, the strains were tentatively identified according to Barrow and Feltham (1993). A subset of strains was also tested for arginine deamination (Møller 1955) and hydrolysis of gelatine (Anon 1965) and urea (Christensen 1946). The arginine test was performed in media prepared with half-strength ASW, while the NaCl concentration of the gelatine and urea media was adjusted to 1% (w/v). Incubation was carried out at 15C for up to 7 days except for gelatine, which was incubated for up to 21 days. Determination of G + C mol% and 16S rDNA gene sequence analysis G + C mol% and 16S rDNA gene sequence were determined for 11 and five strains, respectively. The strains were isolated from spoiled squid. Chromosomal DNA was extracted and purified as described by Fonnesbech Vogel et al. (1997). Briefly, a few freshly-grown bacterial colonies were suspended in phosphate-buffered saline (PBS) and treated with RNase. After lysing with sodium dodecyl sulphate (SDS), DNA was precipitated with isopropanol and washed twice with icecold ethanol. After drying, DNA was dissolved in Millipore ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 + nd nd + nd nd nd + Nd§ nd + nd nd nd Photobacterium phosphoreum Shewanella putrefaciens Pseudomonas sp. Pseudoalteromonas sp.– Moraxella or Psychrobacter sp. Vibrio sp.** Aeromonas sp. quality water. G + C mol% was determined according to Fonnesbech Vogel et al. (1997) following the procedure described by Mesbah et al. (1989). A strain of Shewanella putrefaciens (ATCC 8071) with a G + C% of 45% was included as a control. PCR amplification was performed as described by Fonnesbech Vogel et al. (1997). Oligonucleotides for both PCR amplification and sequencing were synthesized according to sequences and 16S rRNA positions given in Dewhirst et al. (1989) and Paster and Dewhirst (1988). PCR-amplified fragments were purified on Microspin columns (Pharmacia Biotech, Piscataway, NJ, USA) and the cycle sequenced (Thermo sequenase fluorescent-labelled primer cycle sequencing kit, Amersham, Little Chalfont, UK) on the A.L.F. Sequencer (Pharmacia Biotech) using fluoresceinlabelled primers. Searches for 16S rRNA sequences were performed by fastA and Blast using the Wisconsin Sequence Analysis Package (Genetics Computer Group, Madison, USA). The nucleotide sequences described in this report have been deposited with Genbank under the accession number AF293975. RESULTS ) + ) ) ) ± ± Growth at 1C H2 S production NaCl requirement Identification 944 T . P A A R U P ET AL. *Coccobacilli. Large round cells. Rods. §Not determined. –16S rDNA gene sequence analysis carried out on five strains and G + C mol% on 11 strains. **Sensitive to vibriostaticum (O/129). Resistant to vibriostaticum (O/129). ± + ) ) ) ± ± F –/ox. Ox. – – F F Cb* r r r cb/r r r - ± + + + ) ± + ) + + + + + + + + + + + + + TMAO reduction H&L Catalase Oxidase Motility, 15C Morphology Gram reaction Table 1 Tentative identification of strains isolated from squid (Todaropsis eblanae) stored in ice Storage conditions The average temperatures of the whole and gutted individuals were 0Æ3 ± 0Æ1C and –0Æ1 ± 0Æ1C, respectively; the temperature of the cold chamber in which the storage experiment was conducted was of 2Æ7 ± 0Æ4C. Shelf-life, sensory and chemical characteristics of spoiled squid The evaluation of the sensory parameters, flavour and smell of cook water, by the taste panellists showed shelf-lives of 10 and 12 days for the whole and gutted lot, respectively (Table 2). As sensory assessment was carried out on days 9 and 13, the shelf-life limits were obtained by interpolation and thus, in the case of flavour, whole squid was awarded scores of 5 and 0Æ8 on days 9 and 13, respectively, and gutted squid, 5Æ5 and 3Æ3, respectively. Linear regression between these points showed that a score of 4 (rejection) in whole and gutted individuals was obtained on days 10 and 12, respectively, assuming that spoilage was close to linear between days 9 and 13. Most panellists complained of ammoniacal off-odours and off-flavours in both lots, and soft texture in the whole lot on day 13. Interpolated NH3-N concentrations by the end of shelf-life of whole and gutted squid were 33 and 24 mg 100 g–1, respectively, while TMAN constituted 17 and 12 mg 100 g–1. pH values were 7Æ2 and 7Æ3, respectively. It was also assumed in this case that spoilage was close to linear between days 9 and 13. ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 CHANGES DURING STORAGE OF ICED SQUID Chemical changes during storage 60 (a) mg 100 g–1 50 40 30 20 10 0 2 6 9 13 9 13 Days in ice 60 (b) 50 mg 100 g–1 The initial concentration of TMAO-N in the mantle was 20 mg 100 g–1. The initial pH value was 6Æ8, which increased gradually to 7Æ5 and 7Æ4 in the whole and gutted lot, respectively (results not shown). NH3 and TMA production increased from day 2 in whole squid, and the rate of production of these compounds increased in both lots from day 9 to day 13 (Fig. 1). The production of NH3 and TMA was lower in gutted squid during the whole storage period, but the evolution pattern was similar to the whole lot. On day 10, concentrations of NH3 and TMA in the gutted lot constituted 61 and 59%, respectively, of the concentrations detected in the whole lot. The concentration of urea remained unchanged at approximately 10 mg 100 g–1 in both lots throughout storage. The biogenic amine agmatine, which was only present in trace amounts in fresh squid, increased rapidly in both lots during storage, from less than 1 to 120 mg 100 g–1 in 9 days (Fig. 2). Other biogenic amines, cadaverine, tyramine and putrescine, were detected in whole squid on day 13 (Fig. 2). Histamine and spermine never exceeded 2 and 5 mg 100 g–1, respectively, and only trace amounts of phenyl-ethyl-amine and spermidine were found during the whole storage period for both lots (data not shown). 945 40 30 20 10 0 2 6 Days in ice Bacteriological changes Initially, gutting and washing led to a reduction of 1–1Æ5 log units of the counts of whole squid, with the exception of the luminous bacteria where the initial concentration was 103 cfu g–1 (Fig. 3). Luminous bacteria increased to 105 cfu g–1 in both whole and gutted squid after 13 days of iced storage. On that day, aerobic plate counts and counts of H2S producers in the gutted lot were 5 · 107 and 1 · 107 cfu g–1, respectively. Levels in whole squid were lower at 107 and 106 cfu g–1. Pseudomonas counts were, in general, lower than numbers of H2S producers in both lots. Identification of bacterial flora at initial and final sampling points A heterogeneous group of Gram-negative bacteria was isolated from fresh squid (Table 3). With the exception of the dominating flora subsequently identified as Pseudoalteromonas sp., all isolates were allocated based on Table 2 Shelf life, sensory and chemical characteristics at the time of sensory rejection of squid stored in ice Fig. 1 Evolution of NH3-N (s) and TMA-N (n) in (a) whole and (b) gutted squid stored in ice the simple scheme of Table 1. The dominating flora included motile, Gram-negative, non-fermentative rods with positive oxidase and catalase reactions. These were neither non-alkali-producing Pseudomonas nor Alcaligenes spp., as they had a requirement for NaCl. The G + C mol% varied between 39Æ8 and 41Æ2% for 11 strains tested. 16S rDNA gene sequence analysis identified these strains as Pseudoalteromonas (see below). As part of the qualitative characterization of the isolates from spoiled squid, originating from this experiment and two unpublished experiments, 96% of 86 presumptive Pseudoalteromonas sp. isolated from spoiled squid liquefied gelatine and only 3% metabolized arginine. These reactions, together with a negative nitrate reaction and strict aerobic growth, are in agreement with the description of the genus Characteristics at end of shelf-life Lot Shelf-life (days) Sensory characteristics NH3-N 100g)1 TMA-N 100g)1 pH Whole Gutted 10 12 33 24 17 12 ammonia, soft texture ammonia ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 7Æ2 7Æ3 946 T . P A A R U P ET AL. 180 9 (a) 8 140 7 120 6 Log cfu g –1 mg 100 g–1 160 100 80 60 5 4 3 40 2 20 1 0 2 6 9 (a) 0 13 2 6 Days in ice 180 160 13 9 13 Days in ice 9 (b) 8 140 (b) 7 Log cfu g –1 mg 100 g–1 9 120 100 80 60 6 5 4 3 40 2 20 1 0 2 6 9 13 0 2 Days in ice Days in ice Fig. 2 Evolution of the biogenic amines agmatine (s), cadaverine (n), putrescine (h) and tyramine (d) in (a) whole and (b) gutted squid (Gauthier et al. 1995). Moreover, 92% produced NH3 from urea (Paarup, unpublished data). During iced storage of whole squid, these Pseudoalteromonas and strains identified as Photobacterium phosphoreum became the dominant bacteria, accounting for 85% of the flora. In gutted squid, a more heterogeneous flora appeared, containing equal numbers of Shewanella putrefaciens, Pseudomonas sp. and Pseudoalteromonas (Table 3). Neither Fig. 3 Microbiological changes during storage of (a) whole and (b) gutted squid stored in ice. Aerobic plate count, IAS 15C (s), H2Sproducing bacteria (n), Pseudomonas (h) and luminous bacteria (d). All counts are from selective or indicative media Pseudoalteromonas sp. nor P. phosphoreum were detected by isolation of colonies from Iron agar incubated at 20C; they were isolated only from spread plates, which indicates that they do not survive the contact with the warm agar. Moraxella/Psychrobacter, Vibrio and Aeromonas sp. were all detected in lower percentages at the end of the storage Number of strains (%) Genus or species Whole, initial Whole, day 13 Gutted, day 13 Photobacterium phosphoreum Shewanella putrefaciens Pseudomonas sp. Pseudoalteromonas sp. Moraxella sp., Psychrobacter sp. Vibrio sp. Aeromonas sp. 1 5 4 11 2 4 8 7 5 2 33 0 0 0 0 19 12 15 0 4 1 Total 35 (100) (3) (14) (11) (31) (6) (11) (23) 6 (15) (11) (4) (70) (0) (0) (0) 47 (100) (0) (37) (24) (29) (0) (8) (2) Table 3 Composition of the microflora of whole and gutted squid stored in ice, initially and in samples rejected by sensory analyses. Percentage values are based on random isolation from Iron Agar (with salt) plates and subsequent identification 51 (100) ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 CHANGES DURING STORAGE OF ICED SQUID period of gutted squid, being absent in the iced whole animal. Identification of Pseudoalteromonas by 16S rDNA gene sequence analysis Sequence was obtained for the region 28–1492 (Escherichia coli position) of the 16S rRNA gene. The sequences were 1456 bases in length. The five strains sequenced were identical. The search in GenBank and EMBL databases showed 99Æ7% similarity to Pseudoalteromonas [gracilis] (gracilis is not a valid species name), with GenBank accession number AF038846. High similarity to other strains of Pseudoalteromonas of between 98 and 99% was also found; 97–98% similarity was found to Pseudoalteromonas antartica strains, and between 95 and 96% similarity was found to other strains listed as Pseudoalteromonas sp. and to Pseudoalteromonas citrea and Pseudoalteromonas aurantia. Bacteriological characteristics of the stomach contents and bacteriological and chemical characteristics of the digestive gland of recently-caught squid (48 h) The mean weight of the digestive gland was 30 ± 9 g, which made up 13% of the total weight of the whole individuals. Values of pH (6Æ8) and urea (7Æ4 mg 100 g–1) did not differ from those of the mantle, but ammonia and TMA concentrations were about five and 18 times higher, respectively. Also, its content of biogenic amines was much higher than in the mantle (results not shown). The aerobic plate count of the stomach contents 48 h after capture was 106 cfu g–1. Photobacterium phosphoreum dominated the bacterial flora, followed by strains identified as Pseudoalteromonas and Pseudomonas sp. The microbial load of the digestive gland was about one log unit lower than the stomach contents, and all isolated colonies were identified as P. phosphoreum (data not shown). DISCUSSION Iced squid spoiled rapidly in less than two weeks. Gutting of squid allowed shelf-life to be extended by approximately two days, which is not in agreement with Park and Hur (1990), the only work on the subject as far as is known. These authors reported a shelf-life of Loligo vulgaris of 10–12 days, independent of gutting. In contrast, the extension is similar to the range of one to six days reported for various fish species (Huss and Asenjo 1976). The presence of ammoniacal off-odours is in accordance with other reports on sensory characteristics of spoiled squid (Botta et al. 1979; Ke et al. 1984). In addition, NH3 has been shown to be an excellent indicator of squid quality (Illex illecebrosus) according to LeBlanc and Gill (1984), who 947 attributed production to autolysis due to a linear increase during the whole storage period, using the same method as in the present work. The rapid onset of NH3 production at low bacterial cell densities indicates that autolysis was causing the production during the first nine days of storage in the present work. The change in rate of production after nine days of storage, particularly in whole individuals, indicates a bacterial contribution. Similarly, Licciardello et al. (1985) observed a final exponential increase of NH3 in whole iced Loligo pealei. These authors reported a concentration of 29 mg NH3-N 100 g–1 in samples of marginal sensory quality, which corresponds with the level in the present work. Although ultimate evidence would require monitoring of the digestive gland during the whole storage period, the high initial content of NH3 makes diffusion from gland to mantle during storage probable. Thus, LeBlanc and Gill (1984) detected NH3 concentrations twice as high in parts of the mantle that had been discoloured by direct contact with the gland. Also, diffusion of autolytic exoenzymes from gland to mantle had been suggested earlier by Moral et al. (1998) based on higher levels of free amino acids in the mantles of whole individuals than in gutted ones during 13 days of storage. Diffusion may also explain the slightly higher level of TMA found in the whole lot compared with the gutted lot. The levels reported in the present study are similar to the 13 mg-N 100 g–1 reported by Licciardello et al. (1985). The overall bacterial numbers are too low to account for the TMA production (Gram and Huss 1996), but the existence of a niche for bacteria that produce enzymes diffusing into the flesh cannot be ruled out. The very early onset of agmatine production indicates an autolytic mechanism, but the fact that it is produced by decarboxylation of arginine suggests the involvement of bacterial decarboxylase activity. This amine seems to be an excellent freshness index in squid, which is in agreement with Yamanaka et al. (1987). The appearance of cadaverine, putrescine and tyramine between days nine and 13 supports the view that bacterial activity contributes to spoilage at the late storage stage. Cadaverine and putrescine have both been suggested as spoilage indicators of squid (Takagi et al. 1971; Yamanaka et al. 1987). The possible role of agmatine in intoxication by biogenic amines is not known but according to Halász et al. (1994), it may act as a histamine poisoning potentiator alongside cadaverine and putrescine. However, the low levels of histamine detected during the present experiment make such intoxication problems in squid irrelevant. Pseudoalteromonas is a marine genus (Baumann et al. 1984; Gauthier et al. 1995) and as far as is known, this is the first report linking the organism to spoilage of marine seafood. This genus was created by Gauthier et al. (1995) who included all Alteromonas species with the exception of ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 948 T . P A A R U P ET AL. A. macleodii. van Spreekens (1977) isolated some shrimpspoiling strains with typical ‘Alteromonas’ reactions, which may have been Pseudoalteromonas. Pseudoalteromonas spp. are antagonistic and produce bacteriostatic and bacteriolytic compounds (Nair and Simidu 1987; Holmström and Kjelleberg 1999) as well as iron-binding siderophores (Reid et al. 1993). Such properties may have been involved in its dominance in whole squid, but they do not explain the complex microbiological picture in gutted individuals. Urease activity in Pseudoalteromonas sp. may contribute to late ammonia accumulation in both whole and gutted squid where similar numbers were detected. The predominance of P. phosphoreum in the digestive gland and stomach is probably the reason for the presence of this bacterium in spoiled whole squid. Even higher numbers (107–108) of this bacterium have been described in the gut contents of cod (van Spreekens 1974; Dalgaard et al. 1993). This micro-organism, which produces large amounts of TMA, is the specific spoilage organism of modifiedatmosphere-packed cod fillets (Dalgaard et al. 1993; Dalgaard 1995). Several studies have shown that 107 cfu g)1 are required to produce detectable levels of TMA and therefore, the final level of 106 cfu g–1 of this bacterium detected in the mantle of the whole lot is not likely to have contributed to the production of this compound. It is possible that growth in the digestive gland, where initial levels were higher, caused TMA production which then, by diffusion, entered the mantle. Photobacterium phosphoreum may also be involved in agmatine production, since preliminary results obtained in a model system have shown that it exhibits strong arginine decarboxylase activity (Paarup, unpublished data). Also, the precursor arginine is abundant in the free state in squid (Suyama and Kobayashi 1980). Many squid species possess photophores (light organs) of the symbiotic type based on emission of light from P. phosphoreum (Guerra 1992), and such species may show an accelerated spoilage pattern. The absence of P. phosphoreum in pour-plated iron agar is in accordance with Dalgaard et al. (1997). Shewanella putrefaciens is unlikely to play any important role in the spoilage of the whole lot, with a final level of approximately 106 cfu g–1, since concentrations of 108–109 are required to produce off-odours in cod juice (Jørgensen and Huss 1989). Similarly, Pseudomonas spp. are of little importance in the spoilage of the whole lot, and a final concentration of 107 cfu g–1 in the gutted lot is very low compared with the 108–109 cfu g–1 required to spoil chilled fish (Gram et al. 1989). In conclusion, the gutting of T. eblanae extends shelf-life by approximately two days and reduces production of ammonia and TMA. Agmatine production occurs during the early storage stages. For the same reasons, this amine seems to be an excellent freshness indicator. Plate count methods based on pour-plating (i.e. incorporating the sample in melted, 45C warm agar) are inappropriate for enumeration of the microflora of squid since psychrophilic, heat-sensitive bacteria, which constitute a substantial part of the spoilage flora, are lost by such methods. Pseudoalteromonas sp. is constantly present in spoiled T. eblanae. Growth kinetics and spoilage potential/activity of the quantitatively most important micro-organisms isolated from spoiled squid are currently being investigated in a model system. ACKNOWLEDGEMENTS This work was carried out under EU contract FAIR no. GT95 1836 entitled ‘The elaboration of new squid products based on initial tenderization’. REFERENCES Anon. (1965) Standard Methods for the Examination of Water and Waste Water 12th edn. pp. 596–597. New York: American Public Health Association, American Waterworks, Water Pollution Control Federation, APHA Inc. AOAC (1995) Trimethylamine nitrogen in seafood. In Official Methods of Analysis of AOAC International 16th edn, Vol. II. p. 7. Virginia, USA: Association of Official Analytical Chemists. Barrow, G.I. and Feltham, R.K.A. (1993) Cowan and Steel’s Manual for the Identification of Medical Bacteria. Cambridge, UK: Cambridge University Press. Baumann, P., Gauthier, M.J. and Baumann, L. (1984) Genus Alteromonas Baumann, Baumann, Mandel and Allen 1972. In Bergey’s Manual of Systematic Bacteriology Vol. 1, ed. Krieg, N.R. and Holt, J.G. pp. 343–352. Baltimore: Williams & Wilkins. Botta, J.R., Downey, A.P., Lauder, J.T. and Noonan, P.B. (1979) Preservation of Raw Whole Short-finned Squid (Illex Illecebrosus) During the Period from Catching to Processing: Skin Color of Raw Squid and Sensory Quality of the Subsequently Cooked Squid. Fisheries & Marine Service Technical Report no. 855, Department of Fisheries and Oceans, Canada. Bystedt, J., Swenne, L. and Aas, H.W. (1959) Determination of trimethylamine oxide in fish muscle. Journal of the Science of Food and Agriculture 10, 301–304. Christensen, W.B. (1946) Urea decomposition as a means of differentiating Proteus and paracolon cultures from each other and from Salmonella and Shigella. Journal of Bacteriology 52, 461. Dalgaard, P. (1995) Qualitative and quantitative characteristics of spoilage bacteria from packed fish. International Journal of Food Microbiology 26, 319–333. Dalgaard, P., Gram, L. and Huss, H.H. (1993) Spoilage and shelf life of cod fillets packed in vacuum or modified atmospheres. International Journal of Food Microbiology 19, 283–294. Dalgaard, P., Mejlholm, O., Christiansen, T.J. and Huss, H.H. (1997) Importance of Photobacterium phosphoreum in relation to spoilage of modified atmosphere-packed fish products. Letters in Applied Microbiology 24, 373–378. ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 CHANGES DURING STORAGE OF ICED SQUID Dewhirst, F.E., Paster, B.J. and Bright, P.L. (1989) Chromobacterium, Eikenella, Kingella, Neisseria, Simonsiella and Vitreoscilla species comprise a major branch of the beta group Proteobacteria by 16S ribosomal ribonucleic acid sequence comparison: transfer of Eikenella and Simonsiella to the family Neisseriaceae (emend.). International Journal of Systematic Bacteriology 38, 258–266. FAO (1998) FAOSTAT. http://www.fao.org. Fonnesbech Vogel, B., Jørgensen, K., Christensen, H., Olsen, J.E. and Gram, L. (1997) Differentiation of Shewanella putrefaciens and Shewanella alga using ribotyping, whole-cell protein profiles, phenotypic characterization and 16S rRNA sequence analysis. Applied and Environmental Microbiology 63, 2189–2199. Gauthier, G., Gauthier, M. and Christen, R. (1995) Phylogenetic analysis of the genera Alteromonas, Shewanella and Moritella using genes coding for small-subunit rRNA sequences and division of the genus Alteromonas into two genera, Alteromonas (emended) and Pseudoalteromonas gen. Nov., and proposal of twelve new species combinations. International Journal of Systematic Bacteriology 45, 755–761. Gram, L. and Huss, H.H. (1996) Microbiological spoilage of fish and fish products. International Journal of Food Microbiology 33, 121– 137. Gram, L., Oundo, O. and Bon, J. (1989) Storage life of Nile perch (Lates niloticus) dependent on storage temperature and initial bacterial load. Tropical Science 29, 221–236. Gram, L., Trolle, G. and Huss, H.H. (1987) Detection of specific spoilage bacteria from fish stored at low (0C) and high (20C) temperatures. International Journal of Food Microbiology 4, 65–72. Gregersen, T. (1978) Rapid method for distinction of Gram-negative from Gram-positive bacteria. European Journal of Applied Microbiology and Biotechnology 5, 123–127. Guerra, A. (1992) Mollusca, Cephalopoda. In Fauna Ibe´rica Vol. 1, ed. Ramos, M.A. Madrid, Spain: Museo Nacional de Ciencias Naturales, CSIC. Guerra, A. (1996) Explotación mundial de cefalópodos. In II Jornadas Internacionales Sobre Utilizacio´n de Cefalo´podos: Aspectos Cientı´ficos y Tecnolo´gicos, Instituto Del Frı́o, Madrid. Halász, A., Baráth, À., Simon-Sarkadi, L. and Holzapfel, W. (1994) Biogenic amines and their production by microorganisms in food. Trends in Food Science and Technology 5, 42–49. Hobbs, G. and Hodgkiss, W. (1982) The bacteriology of fish handling and processing. In Developments in Food Microbiology ed. Davis, R. pp. 71–117. London: Applied Science Publishers. Holmström, C. and Kjelleberg, S. (1999) Marine Pseudoalteromonas species are associated with higher organisms and produce biologically active extracellular agents. FEMS Microbiology and Ecology 30, 285–293. Hugh, R. and Leifson, E. (1953) The taxonomic significance of fermentative versus oxidative Gram-negative bacteria. Journal of Bacteriology 66, 24–26. Huss, H.H. and Asenjo, I. (1976) Storage life of gutted an ungutted white fish. In Annual Report, Technological Laboratory, Danish Ministry of Fisheries, Technical University, Lyngby, Denmark. pp. 32–47. Iida, H., Nakamura, K. and Tokunaga, T. (1992) Non-protein nitrogenous compounds in muscle extract of oceanic cephalopods. Nippon Suisan Gakkaishi 58, 2383–2390. 949 Jørgensen, B.R. and Huss, H.H. (1989) Growth and activity of Shewanella putrefaciens isolated from spoiling fish. International Journal of Food Microbiology 9, 51–62. Ke, P.J., Burns, B.G. and Woyewoda, A.D. (1984) Recommended procedures and guidelines for quality evaluation of Atlantic short-fin squid (Illex illecebrosus). Lebensmittel-Wissenschaft und -Technologie 17, 276–281. Ke, P.J., Fierheller, M. and Lemon, D.W. (1991) Studies on processing technology for Atlantic short-fin squid (Illex illecebrosus). Lebensmittel-Wissenschaft und -Technologie 24, 328–333. Ke, P.J., Woyewoda, A. and Fierheller, M. (1979) Handling methods and quality evaluation of fresh Canadian atlantic squid (Illex illecebrosus). Fisheries & Marine Service Technical Report no. 898. Canada: Department of Fisheries and Oceans. Kovacs, N. (1956) Identification of Pseudomonas pyocyanae by the oxidase reaction. Nature 178, 703. Learson, R.J. and Ampola, V.G. (1977) Care and maintenance of squid quality. Marine Fisheries Review 39, 15–16. LeBlanc, R.J. and Gill, T.A. (1984) Ammonia as an objective quality index in squid. Canadian Institute of Food Science and Technology Journal 17, 195–201. Licciardello, J.J., Ravesi, E.M., Gerow, S.M. and D’entremont, D. (1985) Storage characteristics of iced whole Loligo squid. In Proceedings of the International Institute of Refrigeration, Conference on Storage Lives of Chilled and Frozen Fish and Fish Food, Aberdeen, UK, pp. 249–257. Liston, J. (1980) Microbiology in fishery science. In Advances in Fishery Science and Technology ed. Connell, J.J. pp. 138–157. Farnham, UK: Fishing News Books Ltd. MacLeod, R.A. (1968) On the role of inorganic ions in the physiology of marine bacteria. Advances in Microbiology of the Sea 1, 95–126. Mesbah, M., Premachandran, U. and Whitman, W.B. (1989) Precise measurement of G+C content of deoxyribonucleic acid by highperformance liquid chromatography. International Journal of Systematic Bacteriology 39, 159–167. Møller, V. (1955) Simplified tests for some amino acid decarboxylases and for the arginine dihydrolase system. Acta Pathologica et Microbiologica Scandinavica 36, 158. Moral, A., Paarup, T. and Sanchez, J.A. (1998) Evolution of free amino acids in whole and eviscerated squid stored in ice. Poster presented at the 28th WEFTA meeting, October 4–7, Tromsø, Norway. Morales, J. (1997) Caracterización bioquı́mica de Illex coindetii, Todaropsis eblanae y Eledone cirrhosa y comportamiento durante la conservación en estado refrigerado, bajo distintas concentraciones de CO2 y O2, y congelado. Aspectos sobre identificación y modificaciones del colágeno. PhD Thesis, Complutense University, Faculty of Pharmacy, Madrid, Spain. Nair, S. and Simidu, U. (1987) Distribution and significance of heterotrophic marine bacteria with antibacterial activity. Applied and Environmental Microbiology 53, 2957–2962. Ohashi, E., Okamoto, M., Ozawa, A. and Fujita, T. (1991) Characterization of common squid using several freshness indicators. Journal of Food Science 56, 161–163, 174. Ohmori, H., Hori, T. and Nakamura, K. (1975) Preservation of skin colour of squid in refrigerated seawater. Refrigeration 50, 435–438. ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950 950 T . P A A R U P ET AL. Park, H.-Y. and Hur, J.-W. (1990) A study on the processing aptitude and storage of common-European squid (Loligo vulgaris). 1. Changes of freshness during storage. Journal of Korean Society of Food and Nutrition 19, 168–174. Paster, B.J. and Dewhirst, F.E. (1988) Phylogeny of Campylobacters, Wolinellas, Bacteriodes gracilis and Bacteriodes ureolytica by 16S ribosomal ribonucleic acid sequencing. International Journal of Systematic Bacteriology 38, 56–62. Reid, R.T., Live, D.H., Faulkner, D.J. and Butler, A. (1993) A siderophore from a marine bacterium with an exceptional ferric ion affinity constant. Nature 366, 455–458. Slabyj, B.M. and True, R.H. (1981) Pre-process holding of squid (Illex illecebrosus) and quality of canned mantles. Journal of Food Protection 44, 109–111, 114. Suyama, M. and Kobayashi, H. (1980) Free amino acids and quaternary ammonium bases in mantle muscles of squid. Bulletin of the Japanese Society of Fisheries Science 46, 1261–1264. Takagi, M., Iida, A. and Oka, S. (1971) On the formation of unvolatile amines during putrefaction in the muscle of squid and octopus. Bulletin of the Japanese Society of Fisheries Science 37, 1079–1083. Tapia-Salazar, M., Smith, T.K. and Harris, A. (2000) High performance liquid chromatographic method for detection of biogenic amines in feedstuffs, complete feeds and animal tissues. Journal of Agricultural and Food Chemistry 48, 1708–1712. Tracy, M.L., Pickering, M.V. and Verhulst, T. (1995) Cation exchange analysis of foods and beverages for biogenic amines. Food Testing and Analysis 1, 48–50. van Spreekens, K.J.A. (1974) The suitability of a modification of Long and Hammer’s medium for the enumeration of more fastidious bacteria from fresh fishery products. Archiv für Lebensmittel Hygiene 25, 213–219. van Spreekens, K.J.A. (1977) Characterization of some fish and shrimp spoiling bacteria. Antonie van Leeuwenhoek 43, 283–303. Yamanaka, H., Shiomi, K. and Kikuchi, T. (1987) Agmatine as a potential index for freshness of common squid (Todarodes pacificus). Journal of Food Science 52, 936–938. ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 92, 941–950