Abstract
Shewanella putrefaciens has been considered the main spoilage bacteria of low-temperature stored marine seafood. However, psychrotropic Shewanella have been reclassified during recent years, and the purpose of the present study was to determine whether any of the new Shewanella species are important in fish spoilage. More than 500 H2S-producing strains were isolated from iced stored marine fish (cod, plaice, and flounder) caught in the Baltic Sea during winter or summer time. All strains were identified as Shewanella species by phenotypic tests. Different Shewanella species were present on newly caught fish. During the warm summer months the mesophilic human pathogenic S. algae dominated the H2S-producing bacterial population. After iced storage, a shift in the Shewanella species was found, and most of the H2S-producing strains were identified as S. baltica. The 16S rRNA gene sequence analysis confirmed the identification of these two major groups. Several isolates could only be identified to the genus Shewanella level and were separated into two subgroups with low (44%) and high (47%) G+C mol%. The low G+C% group was isolated during winter months, whereas the high G+C% group was isolated on fish caught during summer and only during the first few days of iced storage. Phenotypically, these strains were different from the type strains of S. putrefaciens, S. oneidensis, S. colwelliana, and S. affinis, but the high G+C% group clustered close to S. colwelliana by 16S rRNA gene sequence comparison. The low G+C% group may constitute a new species. S. baltica, and the low G+C% group of Shewanella spp. strains grew well in cod juice at 0°C, but three high G+C Shewanella spp. were unable to grow at 0°C. In conclusion, the spoilage reactions of iced Danish marine fish remain unchanged (i.e., trimethylamine-N-oxide reduction and H2S production); however, the main H2S-producing organism was identified as S. baltica.
In 1985 Shewanella was defined as a new genera (28) named after James Shewan for his work in fisheries microbiology (38). The genus is a member of the Order Alteromonadales, Family Alteromonadceae within the gamma subdivision of the Proteobacteria. Shewanella species are gram-negative, motile rods, which in general are nonfermentative, although the ability to ferment glucose has been reported in a few species (1, 23). Shewanella species produce H2S from both organic and inorganic sources. They are capable of anaerobic respiration using several electron acceptors and most strains reduce trimethylamine-N-oxide (TMAO) (3, 42). Shewanella species can be isolated from a wide range of habitats. Species may be mesophilic (S. algae, S. amazonensis, S. colwelliana, S. oneidensis, and S. putrefaciens), psychrotrophic (S. putrefaciens, S. baltica, S. frigidimarina, S. woodyi, and S. dinitrificans), psychrophilic (S. gelidimarina and S. hanedai), or psychrophilic, as well as barophilic (S. benthica) (2, 24, 29, 42). The species S. putrefaciens plays a prominent role as a spoilage organism of fish and other food products (14). However, S. putrefaciens (10) strains are heterogeneous, and recent taxonomic description using modern molecular methods bifurcated this species; several new species have been described. Initially, mesophilic strains of S. putrefaciens were identified as S. algae (10, 30), S. waksmanii (22), S. affinis (21), and S. aquimarina (43). Later, additional psychrotrophic species such as S. baltica (45), S. oneidensis (42), S. gelidimarina (1), S. frigidimarina (1, 33), S. livingstonensis (2), S. olleyana (39), S. denitrificans (3), and S. profunda (41) were described.
The psychrotrophic nature and the ability of the S. putrefaciens to reduce TMAO to trimethylamine (TMA) explains its importance in spoilage of fish stored at low temperatures where the “fishy” off-odor of spoiling fish is caused by the production of TMA. The bacterium also degrades sulfur-containing amino acids and produces volatile sulfides including H2S (16, 19). Bacteria identified as S. putrefaciens have also been associated with spoilage of chilled-seawater-stored prawns (4), high-pH beef (5), and broiler chicken carcasses (34) and have been isolated in high numbers from ground beef (32). H2S-producing bacteria constitute only a minor fraction of the initial microflora on newly caught fish, but gram-negative, psychrotrophic species become dominant during iced storage, and H2S-producing bacteria will typically grow to levels of 107 × 109 CFU/g (19, 25, 37). The remaining shelf-life of iced fish can be predicted by the number of bacteria capable of reducing TMAO and producing H2S (25).
With the recent changes in Shewanella taxonomy, the objectives of the present study were to determine whether any of the newly named Shewanella species were present on newly caught Danish marine fish and whether any of these appeared more adapted to the selective pressure of prolonged cold storage.
MATERIALS AND METHODS
H2S-producing bacterial examination during storage.
Bacterial examination were carried out on fish caught from the Baltic Sea in August 1995 (one plaice and two cod), January 1996 (two cod), and September 2001 (one plaice, one flounder, and two cod). Water temperatures of the Baltic Seawater were 20, 1, and 17°C, respectively. All of these nine fish (six cod [Gadus morhua], two plaice [Pleuronectes platessa], and one flounder [Platichtys flesus]) were stored at 0°C in ice. Samples were taken from the belly flap area on days 0, 7, 14, and 21 during 1995 and1996 and on days 0, 1, 2, 7, and 14 during 2001. Samples were homogenized and serially diluted in sterile peptone saline and pour plated in iron agar (Oxoid CM964; Oxoid, Ltd., Basingstoke, England) (17). Plates were incubated at 25°C for 3 days and black (H2S-producing) and white colonies formed were counted. Total aerobic counts were determined, combining the counts of black and white colonies. All samplings were carried out in duplicate.
Isolation of bacterial strains.
Approximately 20 black colonies (H2S production) were picked randomly from iron-agar plates from each sampling. The isolates were further purified in veal infusion broth (VIB; catalog no. 0344-17-6; Difco Laboratories, Detroit, MI) and on iron agar at 25°C. The strains were stored at −80°C in broth with 4% (vol/vol) glycerol and 2% (wt/vol) dried skim milk (12) for further characterization.
Identification of bacteria.
All strains were tested at 25°C for the following key characteristics: gram reaction (18), motility and cell shape (phase-contrast microscopy after growth in VIB for 24 h), cytochrome oxidase (BBL DrySlide oxidase, catalog no. 231746; Becton Dickinson, Detroit, MI) (27), catalase reaction (3% H2O2), reduction of TMAO in TMAO medium (17), and production of H2S from thiosulfate (17). Fermentation of glucose was tested in the O-F medium (catalog no. 10282; Merck, Darmstadt, Germany) of Hugh and Leifson (20) at 25°C. Growth at various temperatures (4, 37, and 42°C), tolerance to 6% NaCl (10), and assimilation of several carbohydrates (citrate, gluconate, glucose, lactate, and sucrose) (44) were used to group the isolates as putative Shewanella spp. Briefly, 0.1% (wt/vol) of the tested carbohydrate was added to a sterile mineral basic substrate [10 mM K2HPO4, 10 mM KH2PO4, 2 mM NH4Cl, 2 mM MgSO4(H2O)7, 200 μM C6H8O7FeNH4, 50 μM CaCl2, 154 mM NaCl] supplemented with 1.5% agar. A total of 5 μl of an overnight grown culture was spotted on the plates. Cultures spotted on plates containing no carbohydrate were used as controls of growth due to carryover from the media of the overnight culture. Assimilation of carbohydrate was determined by visual inspection of the plates for growth after 2 and 7 days at 25°C. Type strains of S. putrefaciens, S. baltica, and S. algae were included in each trial and served as control. The genomic guanine-plus-cytosine (G+C mol%) content was determined for a subset of strains (87 strains of 518) by high-pressure liquid chromatography (10). The ability of strains from this subset and some additional strains (in total 98) to degrade gelatin (11), DNA (Difco DNase Test agar w/methyl green, catalog no. 0220-17-5) and ornithine (Difco Decarboxylase base Moeller, catalog no. 289020) was also tested.
Growth of Shewanella in cod juice.
The ability to grow in cod juice at low temperatures was tested on nine strains representing three major groups of Shewanella spp. (three strains each of S. baltica and high-G+C and low-G+C Shewanella spp.). Six strains were precultured in cod juice as reported elsewhere (6) at 0°C. Three strains of high-G+C Shewanella spp. were precultured in nutrient broth at 5°C since growth of these strains at 0°C was very weak when cultured in cod juice. All strains were inoculated in cod juice at 104 × 105 CFU/ml. Flasks were incubated at 0°C, samples were withdrawn twice a week, and growth was estimated by surface plating on iron agar. The experiment was conducted twice, each time with one determination per strain.
DNA extraction and PCR amplification.
DNA was extracted from the overnight grown culture according to standard lysozyme/organic solvent extraction protocols (35). Bacterial small subunit rRNA genes were PCR amplified with eubacterially biased primers B27F (5′-GAGTTTGATCMTGGCTCAG-3′) and B1512R (5′-AAGGAGGTGATCCANCCRCA-3′). PCR conditions were as follows: denaturation for 1 min at 95°C, annealing for 2 min at 55°C, and elongation for 3 min at 72°C for 35 cycles using a thermal cycler (MJ Research, Waltham, MA). The purified amplicons were then fully, bidirectionally sequenced (Davis Sequencing, Davis, CA).
Sequence analysis and phylogenetics.
The phylogenetic relationships of organisms covered in the present study were determined by comparison of individual rRNA gene sequences to those in the public database (http://www.ncbi.nlm.nih.gov/BLAST). A maximum-parsimony evolutionary tree encompassing 500 bootstrapped replicates was constructed by using PAUP software (40a). The rRNA sequences were entered into the GenBank database and were assigned accession numbers AB205566 to AB205581.
DNA-DNA hybridization was studied by microplate hybridization methods (8) with photobiotin labeling and colorimetric detection, with 1,2-phenylenediamine (Sigma) as the substrate and streptavidin-peroxidase conjugate (Boehringer Mannheim) as the colorimetric enzyme (36).
RESULTS
Changes in H2S-producing bacterial population during storage.
Initial counts of H2S-producing black colonies in iron agar varied from 10 CFU/g (winter 1996) to approximately 103 CFU/g (summer 1995). The H2S-producing counts increased on fish stored in ice over the 3-week storage period to 107 to 108 CFU/g irrespective of initial counts and period of collection (Fig. 1). Similarly, total aerobic counts increased from 103 to 104 CFU/g to 108 to 109 CFU/g after 3 weeks of storage in ice (Fig. 1). The initial counts of both H2S-producing and total aerobic bacteria of plaice and cod fish caught during the summer of 2001 were comparable except for the flounder that showed at least 2- to 3-log higher counts than cod or plaice sampled in the same period (Fig. 2). The counts on fish caught in 2001 and stored in ice did not increase to 108 or 109 CFU/g, but the aerobic counts at day 14 were only 104 or 105 CFU/g and H2S-producing microbes were 102 or 104 CFU/g. A log increase in counts was observed when flounder was stored at 0°C (Fig. 2). In 2001, a more frequent sampling was done during the first week; it was clear that H2S counts decreased slightly during the first few days, and a significant increase was not seen until day 14 (Fig. 2). The water temperature reached 21°C in August 1995 and fell to 0 to 1°C in January 1996. In August 2001, the water temperature averaged 17 to 18°C (data not shown).
FIG. 1.
Changes in H2S counts and aerobic counts of five fish (four cod and one plaice) during storage in ice. Fish were sampled in summer 1995 or winter 1996.
FIG. 2.
Changes in H2S counts and aerobic counts of four fish (two cod, one plaice, and one flounder) during storage in ice. Fish were sampled in summer 2001.
Characterization of H2S-producing microbes.
A total of 518 H2S-producing strains were isolated from the nine fish (Table 1). All isolates picked as H2S producing were gram-negative, motile rods with positive oxidase and catalase reactions. They were unable to ferment glucose but reduced TMAO and produced H2S. Based on these traits, the strains were tentatively classified as S. putrefaciens according to the method of Stenstrøm and Molin (40). However, these characteristics are not sufficient to allow for a complete differentiation between S. putrefaciens and S. algae (10) or for differentiation between some of the psychrotrophic shewanellae (44). A simple phenotypic scheme derived from various biochemical tests was used to further distinguish the 518 strains (Table 1). Based on these phenotypic traits, the changes in the abundance of these Shewanella species on fish stored at 0°C, as well as their succession, were delineated (Table 2).
TABLE 1.
Differentiation between Shewanella species based on phenotypic criteria and G+C mol% valuesa
| Species | Phenotypic code | No. of strains | Assimilation of
|
Grow at or in:
|
No. of strains for enzyme assays | Production of:
|
No. of strains for G+C% analysis | Mean G+C mol% ± SDc | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Citrate | Gluconate | Glucose | Lactate | Sucrose | 4°C | 37°C | 42°C | 6% NaCl | Gelatinase | DNase | Ornithine decarboxylase | ||||||
| S. baltica NCTC 10735T1 | A | 1 | + | + | + | + | + | + | − | − | + | 1 | + | + | + | 1 | 46.1 |
| A1 | 345 | + | + | + | + | + | + | − | − | + | 42 | + | + | + | 41 | 46.7 ± 0.5 | |
| A2 | 27 | − | + | + | + | + | + | − | − | + | 5 | + | + | + | 4 | 46.6 ± 0.2 | |
| A3 | 14 | + | + | + | + | − | + | − | − | + | 3 | + | + | + | 2 | 46.8 ± 0.2 | |
| A4 | 8 | + | + | + | + | + | + | + | − | + | 4 | + | + | + | 4 | 46.9 ± 0.3 | |
| S. algae IAM 14159T2 | B | 1 | − | − | + | + | − | − | + | + | + | 1 | + | + | + | 1 | 53.1 |
| B1 | 27 | − | − | − | + | − | − | + | + | + | 10 | + | + | + | 9 | 53.2 ± 0.7 | |
| B2 | 17 | + | − | − | + | − | − | + | + | + | 8 | + | + | + | 8 | 53.0 ± 0.7 | |
| Shewanella spp. | C1 | 30 | − | − | − | + | − | + | + | − | + | 7 | − | − | − | 6 | 47.3 ± 0.2 |
| C2 | 10 | − | − | − | + | − | + | − | − | + | 6 | − | − | − | 4 | 47.3 ± 0.2 | |
| C3 | 25 | − | + | − | − | − | + | − | − | − | 5 | + | + | + | 5 | 44.2 ± 0.6 | |
| C4 | 8 | − | + | − | − | + | + | − | − | − | 4 | + | + | + | 2 | 44.0 ± 0.1 | |
| C5 | 7 | − | + | − | + | − | + | − | − | − | 4 | + | + | + | 3 | 44.1 ± 0.6 | |
| S. putrefaciens ATCC 8071T3 | D | 1 | − | − | − | + | − | + | + | − | − | 1 | + | + | + | 1 | 44.7 |
| S. oneidensis MR1T4 | E | 1 | − | − | − | + | − | + | + | − | − | 1 | + | + | + | 45* | |
| S. colwelliana ATCC 39565T5 | F | 1 | − | − | − | − | − | + | − | − | − | 1 | + | + | + | 46* | |
| S. affinis ATCC BAA-642T6 | G | 1 | − | − | − | − | − | − | − | − | + | 1 | + | + | + | 45* | |
Included are 518 strains isolated as hydrogen sulfide-producing bacteria from iced stored marine fish, as well as the type strains of Shewanella baltica, Shewanella algae, Shewanella putrefaciens, Shewanella oneidensis, and Shewanella colwelliana.
Superscript T1 to T6, type strains of S. baltica, S. algae, S. putrefaciens, S. oneidensis, S. colwelliana, and S. affinis, respectively.
Data for some G+C mol% values have been taken from other studies (*).
TABLE 2.
Succession of different Shewanella species during iced storage of marine fish caught in winter and summer months in Denmarka
| Time | Fish | Storage time (days) | Total no. of H2S strains (n = 518)b | Species
|
|||
|---|---|---|---|---|---|---|---|
| S. baltica (46-47 mol%) (n = 394) | S. algae (52-54 mol%) (n = 44) |
Shewanella spp.
|
|||||
| 46-47 mol% (n = 40) | 44-45 mol% (n = 40) | ||||||
| August 1995 | Plaice | 0 | 15 | 1 | 14 | ||
| 7 | 15 | 15 | |||||
| 14 | 14 | 14 | |||||
| 21 | 15 | 15 | |||||
| Cod | 0 | 14 | 1 | 11 | 2 | ||
| 7 | 15 | 15 | |||||
| 14 | 15 | 15 | |||||
| 21 | 15 | 15 | |||||
| Cod | 0 | 15 | 2 | 13 | |||
| 7 | 14 | 14 | |||||
| 14 | 15 | 15 | |||||
| 21 | 14 | 15 | |||||
| January 1996 | Cod | 0 | 20 | 19 | 1 | ||
| 7 | 19 | 13 | 6 | ||||
| 14 | 20 | 16 | 3 | ||||
| 21 | 20 | 11 | 9 | ||||
| Cod | 0 | 14 | 12 | 1 | 1 | ||
| 7 | 14 | 10 | 4 | ||||
| 14 | 15 | 8 | 7 | ||||
| 21 | 14 | 5 | 9 | ||||
| August 2001 | Plaice | 0 | 5 | 3 | 2 | ||
| 1 | 3 | 3 | |||||
| 2 | 9 | 2 | 7 | ||||
| 7 | 19 | 19 | |||||
| 14 | 16 | 16 | |||||
| Flounder | 0 | 3 | 3 | ||||
| 1 | 2 | 2 | |||||
| 2 | 0 | ||||||
| 7 | 20 | 20 | |||||
| 14 | 17 | 17 | |||||
| Cod | 0 | 13 | 7 | 1 | 5 | ||
| 1 | 6 | 3 | 3 | ||||
| 2 | 1 | 1 | |||||
| 7 | 0 | ||||||
| 14 | 17 | 17 | |||||
| Cod | 0 | 9 | 3 | 6 | |||
| 1 | 13 | 11 | 2 | ||||
| 2 | 16 | 9 | 7 | ||||
| 7 | 19 | 19 | |||||
| 14 | 18 | 18 | |||||
n, total number.
That is, H2S-producing strains.
The majority of the strains (345 of 518) were phenotypically similar to the psychrotrophic S. baltica NCTC 10735T strain and assimilated all of the carbohydrates tested (code A1). Forty-nine strains exhibited very similar phenotypic reactions to the S. baltica type strain except for at least one phenotypic variation. Among these 49 atypical strains, code A2 strains (27 isolates) and code A3 strains (14 isolates) did not assimilate citrate or sucrose, respectively, whereas code A4 strains (8 isolates) grew slowly at 37°C (Table 1). A subset of strains (54 isolates) were positive for production of gelatinase, DNase, and ornithine decarboxylase. Any strain displaying an atypical reaction was retested. The G+C mol% content of code A strains (50 isolates) varied from 46 to 47 G+C mol%, which is also reported for S. baltica. Compiling these results, we identified the 394 strains as S. baltica.
The H2S-producing strains that grew at 42°C but failed to grow at 4°C were tentatively grouped as S. algae. These strains were unable to assimilate gluconate and sucrose as carbon substrate. Among the 44 isolates, code B1 strains (27 isolates) were similar to S. algae IAM 14159T and code B2 strains (17 isolates) assimilated citrate as a sole carbon source. A subset of code B strains were tested for gelatinase activity, DNase, and ornithine decarboxylase production and were all positive for these traits. G+C mol% were between 52 and 54%, and all of these strains were identified as S. algae.
A third group of 80 strains grew well at 4°C but only assimilated one or two of the carbohydrates tested (Table 1, phenotypic trait codes C1 to C5). They were phenotypically different from the type strains of S. putrefaciens (ATCC 8071 T), S. oneidensis (ATCC 700550T), and S. colwelliana (ATCC 39565T). The halotolerant strains (growing in 6% NaCl; codes C1 and C2) did not produce gelatinase, DNase, or ornithine decarboxylase, and the G+C mol% of these strains ranged from 46 to 47%. Code C1 strains were able to grow at 37°C, whereas code C2 strains were not. The 40 strains that did not grow in 6% NaCl (codes C3, C4, and C5) were also able to degrade gelatin, DNA, and ornithine, and the G+C mol% ranged from 44 to 45. The code C strains may belong to a new Shewanella spp., but more detailed taxonomic study is necessary to validate this claim. However, we designate C1 and C2 strains as high-G+C Shewanella spp., and C3, C4, and C5 strains are called low-G+C Shewanella spp.
Phylogenetic analysis based on 16S rRNA gene sequence.
The 1.4-kb nucleotide sequences of 16S rRNA genes (covering base positions 44 to 1471 [E. coli numbering]) were used for phylogenetic analyses. The 16S rRNA gene sequences of 16 selected strains representing all 11 subgroups of different phenotypic traits were compared to the sequences of other Shewanella species and closely related bacteria such as Pseudoalteromonas haloplanktis and Marinospirillum minutulum (Fig. 3). Phylogenetic analyses, based on 16S rRNA sequences, unambiguously demonstrated that the tested strains belonged to the γ-proteobacteria. The 16S rRNA sequences of all known members of the γ-proteobacteria were compared to that of tested strains. Bootstrapping (1,000 replicates) analysis was performed to avoid sampling artifacts. Neighbor-joining, parsimony, and maximum-likelihood analyses were undertaken on this subset of bacteria, using several subdomains of the 16S rRNA. The tested strains shares a close phylogenetic relationship with Shewanella species. The sequence similarities of 16S rRNA genes among the phenotypically coded strains were >99.0%. For example, strains of codes A1, A2, and A3 share >99.0% 16S rRNA sequence similarities among themselves, as well as with sequences of S. baltica NCTC 10735T and S. putrefaciens ATCC 8071T. Likewise, strains of codes B1 and B2 exhibited 100% sequence similarities with S. algae IAM 14159T. Even though strains belonging to high-G+C Shewanella spp. (C1 and C2) were similar to sequences of S. colwelliana ATCC 33888T, the strains of low-G+C Shewanella spp. (C3, C4, and C5) were related to either S. baltica or S. putrefaciens. The strains of halotolerant high-G+C Shewanella spp. and nonhalotolerant low-G+C Shewanella spp. showed <96.0% similarities in their 16S rRNA sequences.
FIG. 3.
Phylogenetic tree based on 16S rRNA gene sequence comparison, showing the taxonomic positions of 16 selected strains representing each subgroup of the different phenotypic traits (Table 1) within the genus Shewanella. The branching pattern was generated by neighbor-joining methods, and bootstrap values were calculated from 1,000 trees. Bar, sequence similarity of 0.01.
DNA-DNA hybridization studies revealed that representative strains belonging to the groups A1 and A2 exhibited >70% DNA reassociation values with S. baltica. However, strains belonging to the group A3 showed only 38% DNA:DNA hybridization values with S. putrefaciens ATCC 8071T and 43% with S. baltica NCTC 10735T. Similarly, the strains of the groups C3 and C4 exhibited 9 to 34% DNA relatedness with several authentic Shewanella species: S. putrefaciens ATCC 8071T, S. frigidimarina ACAM 591T, and S. baltica NCTC 10735T. Furthermore, the strains of the groups A3, C3, and C4 had only 20 to 34% DNA hybridization values between the groups. This finding strongly supports the claim that the isolates of the groups A3, C3, and C4 represent two novel species within the Shewanella genus and will be described in detail elsewhere (M. Satomi, unpublished data). The strains of group C5 are very closely related to groups C3 and C4 (55 to 65% DNA relatedness), and further research is necessary to define their phylogenetic affiliation. In this communication, we discuss the predominance of these strains based on phenotypic groupings.
Changes in Shewanella species during storage of marine fish.
The changes in Shewanella species during storage of marine fish stored at 0°C are presented in Table 2. S. algae was the most commonly identified H2S-producing bacteria on all fish caught in 1995 and on plaice caught in 2001 but, during storage at 0°C, psychrotrophic S. baltica outgrew the mesophilic S. algae. In contrast, S. baltica were present as a prominent member of the H2S-producing bacteria on the fish caught in the colder periods. At all three samplings there was a clear pattern in the growth of certain Shewanella species during storage at 0°C. In August 1995, S. baltica was the only H2S-producing bacteria that proliferated. In January 1996 and August 2001, appreciable numbers of low G+C and high G+C Shewanella spp. grew during iced storage despite the presence of S. baltica on the newly caught fish. In some instances, low-G+C Shewanella spp. accounted ca. 64% (9 of 14 isolates from cod of January 1996; Table 2) of the H2S-producing strains. The H2S-producing bacteria on plaice sampled during August 2001 consisted of either S. algae or high-G+C Shewanella spp., and S. baltica was not isolated. However, S. baltica completely outgrew the other H2S-producing bacteria after 1 week of iced storage Similarly, the flounder of August 2001 exhibited a decline in S. baltica incidence at day 2, but this was the only Shewanella species isolated after 7 and 14 days of storage at 0°C.
Growth patterns in cod juice at 0°C.
Representative strains of S. baltica and Shewanella spp. (both low- and high-G+C isolates) were selected, and their growth pattern in cod juice at 0°C was determined (Fig. 4). The three S. baltica and the three low-G+C Shewanella spp. grew well in cod juice at 0°C and reached densities of 109 CFU/ml in 21 days. The three S. baltica strains had an extended lag phase (7 days) but grew well thereafter. Even though the initial inoculation was 105 CFU/g, the high-G+C Shewanella spp. did not grow at 0°C, and numbers declined slowly over the period of cold storage.
FIG. 4.
Growth of three S. baltica strains and six Shewanella spp. strains in cod juice at 0°C. Shewanella spp. strains belonging to groups C3, C4, and C5 had G+C mol% values of 43.1 to 44.2%, whereas strains belonging to groups C1 and C2 had G+C mol% values of 47.1 to 47.5%. G+C mol% of S. baltica were 46.4 to 46.8%. A1 and A2 and C1 and C2 correspond to the phenotypic trait codes in Table 1.
DISCUSSION
H2S-producing gram-negative bacteria are commonly associated with the spoilage of foods (19). These bacteria usually constitute only a small fraction of the initial flora on newly caught fish but constitutes a significant, sometimes dominant, part of the microbiota during chilled storage (15, 25), and their numbers determine the shelf life of the product (25). All of the 518 H2S-producing strains isolated from the nine fish (Table 1) were tentatively classified as S. putrefaciens according to the method of Stenstrøm and Molin (40). Our study confirmed that H2S producing bacteria grow well during storage of iced marine fish (Fig. 1 and 2), and all strains also produced the fishy compound TMA. These H2S-producing bacteria have hitherto been identified as S. putrefaciens (26, 40); however, the present study demonstrates that the H2S-producing bacteria proliferating during ice stored fish of the Baltic Sea were in fact S. baltica.
Bacterial isolates that have until recently been identified as S. putrefaciens by standard microbiological key characteristics as being gram-negative motile rods with positive oxidase and catalase reactions that produce H2S have strictly respiratory metabolism and are able to reduce several electron-acceptors, including TMAO (40). These traits can, however, only resolve the identity of the isolate to the genus level (42). Based on additional phenotypic characterization, we propose that the Shewanella strains of the Baltic Sea can be further distinguished into its species (Table 1). The criteria chosen were carefully selected from other studies where they have been used to differentiate within the Shewanella species (40, 42, 45). For example, the inability to grow at 4°C and concurrent ability to grow at 42°C distinguishes S. algae from S. baltica and S. putrefaciens.
Stenström and Molin (40) reported phenotypic differences for several strains of S. putrefaciens (31). One of these groups was later described as S. baltica by Ziemke et al. (44). Similarly, in the present study, a phenotypic heterogeneity was observed in the Shewanella species isolated from fish caught in the Baltic Sea. Ziemke et al. (44) concluded that strains identified as S. baltica could use citrate and sucrose as sole sources of carbon and energy and could not grow at 37°C similar to 345 strains in our study. Approximately 12% of strains (49 of 394 isolates) identified as S. baltica did not follow the typical phenotypic pattern of the S. baltica type strain NCTC 10735. Our 16S rRNA gene sequence analysis confirmed the phenotypic grouping established in the present study and identified the 394 strains as S. baltica. The majority of strains tested were positive for the production of ornithine decarboxylase in contrast to the findings of Venkateswaran et al. (42). Gelatinase was produced by all high-G+C Shewanella spp. strains tested, whereas Venkateswaran et al. (42) reported that S. putrefaciens could be separated from other halotolerant strains by being gelatinase-negative. Both these differences could be explained by the fact that our study covered a larger collection of strains (80 isolates) compared to the study by Venkateswaran et al. (42) (10 strains). Also, Shewanella is clearly a heterogeneous genus.
It is possible to systematize any group of isolates on the basis of phenotypic characters; however, the comparative value of a given trait or group of traits over another remains difficult to assess (42). Also, variations in the approaches used for the culture and testing of isolates can produce different results. Hence, we also included G+C content, and 16S rRNA sequence analysis, and it is evident that some of the strains examined in the present study may encompass two novel Shewanella species (code A3 and code C3 to C5 strains). A detailed polyphasic taxonomic approach, as well as DNA-DNA hybridization studies, is necessary to further describe these novel species. Since the objective of our present study was not the description of new Shewanella species, taxonomic characterizations will not be discussed further.
Our study demonstrates that, as reported by several others (10, 42, 45), mesophilic Shewanella strains are often S. algae. This organism commonly occurs in the waters around Denmark during periods of high water temperatures (13) and has been associated with infections in humans (7, 9). The present study demonstrates that S. algae also predominates on fish during the warm summer months. There was a clear correlation between water temperature and the occurrence of S. algae in the Danish waters (13), and was this also reflected in the fish microflora. S. algae were not detected on fish caught in winter but only on fish caught during summer. The significantly higher proportion of S. algae found during the warmer summer of 1995 than in 2001 might be attributed to the seawater temperature, which was at least 3 to 4°C warmer in 1995 than in 2001.
The incidences of Shewanella species on fish caught at different times exhibit different patterns. In 1995, both S. algae and S. baltica coexisted on the newly caught fish. In 1996, due to colder temperature, S. algae strains were not isolated, but S. baltica and psychrotrophic, low-G+C Shewanella spp. were the most abundant H2S-producing bacteria. In the summer of 2001 both S. baltica and high-G+C Shewanella spp. outnumbered S. algae. This higher incidence of high G+C Shewanella spp. in fish requires more detailed study, e.g., monitoring Shewanella spp. occurrence in the surrounding water column etc. The persistence of S. baltica during storage at 0°C is likely due to its faster growth rate at low temperature. The disappearance of high G+C Shewanella spp. (capable of growing at 0°C; Table 1) during storage at low temperature suggested that not all psychrotrophic species grew equally well during prolonged cold storage.
Since S. putrefaciens has been identified as the main spoilage bacterium of iced marine fish (14, 15, 16), we further studied the ability of S. baltica and Shewanella spp. to grow at 0°C. As observed in the prolonged cold storage experiment, members of S. baltica and low G+C Shewanella spp. strains grew well at 0°C when inoculated in the cod juice. The decline in the counts of high G+C Shewanella spp. under the low-temperature conditions further substantiates the notion that not all psychrotrophic species could grow well under low-temperature storage conditions. All species reduced TMAO, produced H2S, and grew well in fish juice at higher temperatures (data not shown). It may also be hypothesized that S. baltica can produce substances that suppress the growth of other bacterium because S. baltica was always cultured from the fish that was stored for >7 days at 0°C. Further research is warranted to elucidate such a phenomenon. In summary, the main H2S-producing organism growing on iced marine fish caught in the Baltic Sea was identified as S. baltica; however, when the water temperature was low, a novel low-G+C Shewanella spp. could also play a role in the spoilage of aerobically stored fish in ice (0°C).
Acknowledgments
We thank Anemone Bundvad and Jette Melchiorsen for excellent technical assistance.
REFERENCES
- 1.Bowman, J. P., S. A. McCammon, D. S. Nichols, J. H. Skerratt, S. M. Rea, P. D. Nichols, and T. A. McMeekin. 1997. Shewanella gelidimarina sp. nov. and Shewanella frigidimarina sp. nov., novel Antarctic species with the ability to produce eicosapentaenoic acid (20:5ω3) and grow anaerobically by dissimilatory Fe(III) reduction. Int. J. Syst. Bacteriol. 47:1040-1047. [DOI] [PubMed] [Google Scholar]
- 2.Bozal, N., M. J. Montes, E. Tudela, F. Jimenez, and J. Guinea. 2002. Shewanella frigidimarina and Shewanella livingstonensis sp. nov. isolated from Antarctic coastal areas. Int. J. Syst. Evol. Microbiol. 52:195-205. [DOI] [PubMed] [Google Scholar]
- 3.Brettar, I., R. Christen, and M. G. Hofle. 2002. Shewanella denitrificans sp. nov., a vigorously denitrifying bacterium isolated from the oxic-anoxic interface of the Gotland Deep in the central Baltic Sea. Int. J. Syst. Evol. Microbiol. 52:2211-2217. [DOI] [PubMed] [Google Scholar]
- 4.Chiniwasagam, H. N., H. A. Bremner, S. J. Thrower, and S. M. Nottingham. 1996. Spoilage pattern of five species of Australian prawns: deterioration is influenced by environment of capture and mode of storage. J. Aquat. Food Prod. Technol. 5:25-50. [Google Scholar]
- 5.Dainty, R. H., R. A. Edwards, C. M. Hibbard, and J. J. Marnewick. 1989. Volatile compounds associated with microbial growth on normal and high pH beef stored at chill temperatures. J. Appl. Bacteriol. 66:281-289. [DOI] [PubMed] [Google Scholar]
- 6.Dalgaard, P., L. Gram, and H. H. Huss. 1993. Spoilage and shelf-life of cod fillets packed in vacuum or modified atmospheres. Int. J. Food Microbiol. 19:283-294. [DOI] [PubMed] [Google Scholar]
- 7.Dominguez, H., B. F. Vogel, L. Gram, S. Hoffmann, and S. Schaebel. 1996. Shewanella algae bacteremia in two patients with lower leg ulcers. Clin. Infect. Dis. 22:1036-1039. [DOI] [PubMed] [Google Scholar]
- 8.Ezaki, T., Y. Hashimoto, and E. Yabuuchi. 1989. Fluorometric deoxyribonucleic acid-deoxyribonucleic acid hybridization in microdilution wells as an alternative to membrane filter hybridization in which radioisotopes are used to determine genetic relatedness among bacterial strains. Int. J. Syst. Bacteriol. 39:224-229. [Google Scholar]
- 9.Fonnesbech Vogel, B., H. M. Holt, P. Gerner-Smidt, A. Bundvad, P. Sogaard, and L. Gram. 2000. Homogeneity of Danish environmental and clinical isolates of Shewanella algae. Appl. Environ. Microbiol. 66:443-448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fonnesbech Vogel, B., K. Jorgensen, H. Christensen, J. E. Olsen, and L. Gram. 1997. Differentiation of Shewanella putrefaciens and Shewanella algae on the basis of whole-cell protein profiles, ribotyping, phenotypic characterization, and 16S rRNA gene sequence analysis. Appl. Environ. Microbiol. 63:2189-2199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Frazier, W. C. 1926. A method for the detection of changes in gelatin due to bacteria. J. Infect. Dis. 39:302-306. [Google Scholar]
- 12.Gibson, L. F., and J. T. Khoury. 1986. Storage and survival of bacteria by ultra-freeze. Lett. Appl. Microbiol. 3:127-129. [Google Scholar]
- 13.Gram, L., A. Bundvad, J. Melchiorsen, C. Johansen, and B. Fonnesbech Vogel. 1999. Occurrence and survival of Shewanella algae in marine waters depending on water temperature and mode of growth. Appl. Environ. Microbiol. 65:3896-3900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gram, L., and P. Dalgaard. 2002. Fish spoilage bacteria: problems and solutions. Curr. Opin. Biotechnol. 13:262-266. [DOI] [PubMed] [Google Scholar]
- 15.Gram, L., and B. Fonnesbech Vogel. 2000. Shewanella, p. 2008-2015. In R. Robinson, C. Batt, and P. Patel (ed.), Encyclopedia of food microbiology. Academic Press, Inc., San Diego, Calif.
- 16.Gram, L., and H. H. Huss. 1996. Microbiological spoilage of fish and fish products. Int. J. Food Microbiol. 33:121-137. [DOI] [PubMed] [Google Scholar]
- 17.Gram, L., G. Trolle, and H. H. Huss. 1987. Detection of specific spoilage bacteria from fish stored at low (0°C) and high (20°C) temperatures. Int. J. Food Microbiol. 4:65-72. [Google Scholar]
- 18.Gregersen, T. 1978. Rapid method for distinction of gram-negative from gram-positive bacteria. Eur. J. Appl. Microbiol. Biotechnol. 5:123-127. [Google Scholar]
- 19.Herbert, R. A., M. S. Hendrie, D. M. Gibson, and J. M. Shewan. 1971. Bacteria active in spoilage of certain seafoods. J. Appl. Bacteriol. 34:41-45. [DOI] [PubMed] [Google Scholar]
- 20.Hugh, R., and E. Leifson. 1953. The taxonomic significance of fermentative versus oxidative metabolism of carbohydrates by various gram-negative bacteria. J. Bacteriol. 66:24-26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ivanova, E. P., O. I. Nedashkovskaya, T. Sawabe, N. V. Zhukova, G. M. Frolova, D. V. Nicolau, V. V. Mikhailov, and J. P. Bowman. 2004. Shewanella affinis sp. nov., isolated from marine invertebrates. Int. J. Syst. Evol. Microbiol. 54:1089-1093. [DOI] [PubMed] [Google Scholar]
- 22.Ivanova, E. P., O. I. Nedashkovskaya, N. V. Zhukova, D. V. Nicolau, R. Christen, and V. V. Mikhailov. 2003. Shewanella waksmanii sp. nov., isolated from a sipuncula (Phascolosoma japonicum). Int. J. Syst. Evol. Microbiol. 53:1471-1477. [DOI] [PubMed] [Google Scholar]
- 23.Ivanova, E. P., T. Sawabe, N. M. Gorshkova, V. I. Svetashev, V. V. Mikhailov, D. V. Nicolau, and R. Christen. 2001. Shewanella japonica sp. nov. Int. J. Syst. Evol. Microbiol. 51:1027-1033. [DOI] [PubMed] [Google Scholar]
- 24.Ivanova, E. P., T. Sawabe, N. V. Zhukova, N. M. Gorshkova, O. I. Nedashkovskaya, K. Hayashi, G. M. Frolova, A. F. Sergeev, K. G. Pavel, V. V. Mikhailov, and D. V. Nicolau. 2003. Occurrence and diversity of mesophilic Shewanella strains isolated from the North-West Pacific Ocean. Syst. Appl. Microbiol. 26: 293-301. [DOI] [PubMed] [Google Scholar]
- 25.Jorgensen, B. R., D. M. Gibson, and H. H. Huss. 1988. Microbiological quality and shelf-life prediction of chilled fish. Int. J. Food Microbiol. 6: 295-307. [DOI] [PubMed] [Google Scholar]
- 26.Jorgensen, B. R., and H. H. Huss. 1989. Growth and activity of Shewanella putrefaciens isolated from spoiling fish. Int. J. Food Microbiol. 9:51-62. [DOI] [PubMed] [Google Scholar]
- 27.Kovacs, N. 1956. Identification of Pseudomonas pyocyanea by the oxidase reaction. Nature 178:703. [DOI] [PubMed] [Google Scholar]
- 28.Macdonell, M. T., and R. R. Colwell. 1985. Phylogeny of the Vibrionaceae and recommendation for two new genera, Listonella and Shewanella. Syst. Appl. Microbiol. 6:171-182. [Google Scholar]
- 29.Nogi, Y., C. Kato, and K. Horikoshi. 1998. Taxonomic studies of deep-sea barophilic Shewanella strains and description of Shewanella violacea sp. nov. Arch. Microbiol. 170:331-338. [DOI] [PubMed] [Google Scholar]
- 30.Nozue, H., T. Hayashi, Y. Hashimoto, T. Ezaki, K. Hamasaki, K. Ohwada, and Y. Terawaki. 1992. Isolation and characterization of Shewanella algae from human clinical specimens and emendation of the description of S. alga Simidu et al., 1990, 335. Int. J. Syst. Bacteriol. 42:628-634. [DOI] [PubMed] [Google Scholar]
- 31.Owen, R. J., R. M. Legros, and S. P. Lapage. 1978. Base composition, size, and sequence similarities of genome deoxyribonucleic acids from clinical isolates of Pseudomonas putrefaciens. J. Gen. Microbiol. 104:127-138. [DOI] [PubMed] [Google Scholar]
- 32.Parker, L. L., and R. E. Levin. 1983. Relative incidence of Alteromonas putrefaciens and Pseudomonas putrefaciens in ground beef. Appl. Environ. Microbiol. 45:796-799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Reid, G. A., and E. H. J. Gordon. 1999. Phylogeny of marine and freshwater Shewanella: reclassification of Shewanella putrefaciens NCIMB 400 as Shewanella frigidimarina. Int. J. Syst. Bacteriol. 49:189-191. [DOI] [PubMed] [Google Scholar]
- 34.Russell, S. M., D. L. Fletcher, and N. A. Cox. 1995. Spoilage bacteria of fresh broiler chicken carcasses. Poult. Sci. 74:2041-2047. [DOI] [PubMed] [Google Scholar]
- 35.Sambrook, J., E. Fritsch, and D. Russel. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, New York, N.Y.
- 36.Satomi, M., B. Kimura, M. Mizoi, T. Sato, and T. Fujii. 1997. Tetragenococcus muriaticus sp. nov., a new moderately halophilic lactic acid bacterium isolated from fermented squid liver sauce. Int. J. Syst. Bacteriol. 47:832-836. [DOI] [PubMed] [Google Scholar]
- 37.Shewan, J. M. 1977. The bacteriology of fresh and spoiling fish and the biochemical changes induced by bacterial action, p. 51-60. In Proceedings of the Conference on Handling, Processing, and Marketing of Tropical Fish. Tropical Products Institute, London, United Kingdom.
- 38.Shewan, J. M., G. Hobbs, and W. Hodgkiss. 1960. A determinative scheme for the identification of certain genera of gram-negative bacteria with special reference to Pseudomonodaceae. J. Appl. Bacteriol. 23:379-390. [Google Scholar]
- 39.Skerratt, J. H., J. P. Bowman, and P. D. Nichols. 2002. Shewanella olleyana sp. nov., a marine species isolated from a temperate estuary which produces high levels of polyunsaturated fatty acids. Int. J. Syst. Evol. Microbiol. 52:2101-2106. [DOI] [PubMed] [Google Scholar]
- 40.Stenstrom, I. M., and G. Molin. 1990. Classification of the spoilage flora of fish, with special reference to Shewanella putrefaciens. J. Appl. Bacteriol. 68:601-618. [DOI] [PubMed] [Google Scholar]
- 40a.Swofford, D. 1990. PAUP: phylogenetic analysis by using parsimony, version 3.0. Illinois Natural History Survey, Champaign, Ill.
- 41.Toffin, L., A. Bidault, P. Pignet, B. J. Tindall, A. Slobodkin, C. Kato, and D. Prieur. 2004. Shewanella profunda sp. nov., isolated from deep marine sediment of the Nankai Trough. Int. J. Syst. Evol. Microbiol. 54:1943-1949. [DOI] [PubMed] [Google Scholar]
- 42.Venkateswaran, K., D. P. Moser, M. E. Dollhopf, D. P. Lies, D. A. Saffarini, B. J. MacGregor, D. B. Ringelberg, D. C. White, M. Nishijima, H. Sano, J. Burghardt, E. Stackebrandt, and K. H. Nealson. 1999. Polyphasic taxonomy of the genus Shewanella and description of Shewanella oneidensis sp. nov. Int. J. Syst. Bacteriol. 49:705-724. [DOI] [PubMed] [Google Scholar]
- 43.Yoon, J. H., S. H. Yeo, I. G. Kim, and T. K. Oh. 2004. Shewanella marisflavi sp. nov. and Shewanella aquimarina sp. nov., slightly halophilic organisms isolated from sea water of the Yellow Sea in Korea. Int. J. Syst. Evol. Microbiol. 54:2347-2352. [DOI] [PubMed] [Google Scholar]
- 44.Ziemke, F., I. Brettar, and M. G. Hofle. 1997. Stability and diversity of the genetic structure of a Shewanella putrefaciens population in the water column of the central Baltic. Aquat. Microb. Ecol. 13:63-74. [Google Scholar]
- 45.Ziemke, F., M. G. Hofle, J. Lalucat, and R. Rossello-Mora. 1998. Reclassification of Shewanella putrefaciens Owen's genomic group II as Shewanella baltica sp. nov. Int. J. Syst. Bacteriol. 48:179-186. [DOI] [PubMed] [Google Scholar]