2023 Volume 29 Issue 6 Pages 559-565
We assessed the bacterial diversity in five Alaska pollock surimi samples using 16S rRNA sequencing to plan the pre-processing of fermenting surimi. The viable bacterial count of the surimi samples was approximately 104-105 cfu/g. Phylogenetic tree analysis revealed that the surimi bacterial profile was dominated by phylum Pseudomonadota, with the dominant genera being Pseudomonas and Psychrobacter. Only one strain of spore-producing bacteria was isolated from the five surimi samples. These results suggest that pasteurization before inoculation of bacterial starters was necessary for stable and hygienic fermentation of Alaska pollock surimi.
Surimi, which is used to produce the traditional Japanese food “kamaboko,” has health benefits due to its high protein content. Fish proteins play important roles in human health (Hosomi et al., 2012) and inhibit lipid accumulation (Mizushige et al., 2010). Additionally, Alaska pollock protein has been shown to cause muscle hypertrophy in rats (Morisawa et al., 2019).
Despite these benefits, seafood consumption in Japan has declined recently. To address this, surimi fermentation using lactic acid bacteria has been conducted as a novel way to consume surimi (Yoshikawa et al., 1994; Shan et al., 2007). These previous studies reported that lactic acid bacteria multiplied and produced proteases during fermentation, resulting in the generation of peptides or gelation of surimi. Thus, the fermentation process is expected to improve the value of surimi.
On the contrary, fermenting raw surimi has the risk of spoilage because bacteria contaminating surimi may multiply during the fermentation process. Microorganisms are a major cause of spoilage in most seafood products (Gram and Dalgaard, 2002), and surimi has been reported to have a total bacterial count > 104 cfu/g (Himelbloom et al., 1991).
To produce fermented surimi safely and stably, an effective heat-treatment process that is implemented before fermentation must be designed. Understanding the bacterial community is crucial for hygienic fermentation of surimi. Thus, the purpose of this study was to characterize the bacterial biodiversity in surimi.
Materials Frozen surimi samples made from Alaska pollock were purchased from a commercial retailer. Prior to analysis, all samples were stored at or below −18 °C.
Lactic acid bacteria Lactobacillus helveticus JCM1004 was used for lactic acid fermentation of surimi. This strain was provided by the RIKEN BRC through the National BioResource Project of the MEXT/AMED, Japan. Before inoculation, lactic acid bacteria were pre-cultured using MRS broth and bacterial cells were collected by centrifugation. Precipitate was washed with 1 % NaCl solution, and prepared to an O.D. at 660 nm of 1.0.
Incubation of raw surimi After thawing, 1 % glucose was added to surimi. For lactic acid fermentation, the starter was inoculated at 1 % (v/w) of surimi. Fifty grams of samples were placed in nylon bags and incubated at 37 °C.
Heat treatment of surimi Fifty grams of the surimi samples were placed in nylon bags and heated at 75 °C for 30 min. After heat treatment, the surimi samples were cooled with water quickly.
Bacterial count and pH measurement Five grams of samples were placed in a stomacher bag (Eiken Chemical Co., Ltd., Tokyo, Japan), diluted with 45 mL of sterilized 1 % NaCl solution, and homogenized for 60 s using the stomacher. The sample solution was then spread on a counting plate with standard methods agar (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) at different dilution gradients, and incubated at 37 °C for 48 h. To identify thermoresistant bacteria, the sample solution was heated at 70 °C for 20 min. Viable bacteria were subsequently counted using the same procedure. MRS agar with 0.5 % CaCO3 was used to detect acid-producing bacteria, and the colonies that showed a clear zone of dissolved CaCO3 were counted. MRS agar was prepared using Difco lactobacilli MRS broth (Becton, Dickinson and Company, Sparks, MD USA) agar powder (15 g/L) and CaCO3 (5 g/L). The pH of sample solutions was measured using LAQUAtwin-pH-33 (Horiba, Ltd., Kyoto, Japan).
DNA extraction Bacterial DNA was directly extracted from a colony using Cica Geneus DNA Extraction Reagent ST (Kanto Chemical Co., Inc., Tokyo, Japan). The DNA extract was stored at −30 °C prior to analysis.
Randomly amplified polymorphic DNA PCR methods Randomly Amplified Polymorphic DNA Polymerase Chain Reaction (RAPD-PCR) was performed using a random primer (5′-CGTGCGGGAA-3′). Each reaction mixture contained 25 μL of PCR buffer, 10 μL of 2 mM dNTPs, 3 μL of 20 μM primer, KOD FX neo (Toyobo Co., Ltd., Osaka, Japan), and 1 μL of DNA template, constituted to 50 μL with water. PCR conditions included an initial denaturation step at 94 °C for 2 min, followed by 40 cycles of denaturation at 98 °C for 10 s, annealing at 37 °C for 30 s, and extension at 68 °C for 1 min. Two microliters of the PCR products were mixed with loading buffer (6X) and Tris-borate-EDTA (TBE) buffer. The mixtures were loaded onto a 2.0 % agarose gel, electrophoresed in TBE buffer, stained with GelRed (Biotium, Inc., Fremont, CA, USA), and photographed with a Fluor-S Multimanager (Bio-Rad Laboratories Inc., Hercules, CA, USA). The RAPD-PCR band patterns were visually evaluated.
Phylogenetic tree analysis The 16S rRNA gene was amplified by PCR using two primers, 27F (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492R (5′-GGYTACCTTGTTACGACTT-3′). Each reaction contained 25 μL of PCR buffer, 10 μL of 2 mM dNTPs, 1.5 μL of 10 μM primers, KOD FX neo (Toyobo Co., Ltd.), and 1 μL of DNA template, constituted to 50 μL with water. PCR conditions included an initial denaturation step at 94 °C for 2 min, followed by 35 cycles of denaturation at 98 °C for 10 s, annealing at 54 °C for 30 s, and extension at 68 °C for 1 min. Sanger sequencing was used to identify the bacteria. PCR products were purified with a QIAquick PCR Purification Kit (QIAGEN GmbH, Hilden, Germany) and sequenced by a commercial provider using the primer 5′-ACTCCTACGGGAGGCAGCAG-3′ to analyze the V4, V5, and V6 regions (Fasmac Co., Ltd., Midorigaoka Atsugi, Japan). For phylogenetic analysis, 25 representative strains were selected based on previous reports (Table 1). Multiple nucleotide sequence alignments were performed using MAFFT version 7 (Kato et al., 2019). The phylogenetic tree was constructed by the maximum likelihood method using RAxML (Kozlov et al., 2019). The resulting sequences were queried in the NCBI database using BLASTn to identify species with a high degree of homology.
| Isolated bacteria* (genus or species) |
Type species | Accession No. |
|---|---|---|
| Flavobacterium | Flavobacterium aquatile | NR_042495 |
| Pseudomonas | Pseudomonas aeruginosa | NR_026078 |
| Pseudomonas gessardii | AF074384 | |
| Moraxella | Moraxella lacunata | NR_036825 |
| Aeromonas | Aeromonas hydrophila | NR_043638 |
| Aeromonas media | X60410 | |
| Lactobacillus | Lactobacillus delbrueckii | NR_117074 |
| Serratia | Serratia marcescens | M59160.1 |
| Acinetobacter | Acinetobacter calcoaceticus | NR_042387 |
| Acinetobacter johnsonii | MG846022 | |
| Arthrobacter | Arthrobacter globiformis | NR_026187 |
| Corynebacterium | Corynebacterium diphtheriae | X84248 |
| Bacillus subtilis | AB042061 | |
| Bacillus simplex | AJ439078 | |
| Bacillus licheformis | KX785171 | |
| Enterobacter | Enterobacter cloacae | AJ251469 |
| Leuconostoc | Leuconostoc mesenteroides | AB681194 |
| Chryseobacterium | Chryseobacterium gleum | AB680759 |
| Pantoea | Pantoea agglomerans | AJ251466 |
| Vagococcus | Vagococcus fluvialis | Y18098.1 |
| Psychrobacter | Psychrobacter immobilis | AJ309942 |
| Macrococcus | Macrococcus equipercicus | NR_044926 |
| Carnobacterium | Carnobacterium divergens | AB680940 |
| Brochothrix | Brochothrix thermosphacta | HQ890942 |
| Escherichia | Escherichia coli | AB681728 |
The bacterial counts and pH of raw surimi samples incubated with/without inoculation of lactic acid bacteria are presented in Table 2. The bacterial counts of surimi without starter were higher than those of surimi with starter after 24 h incubation, indicating that the bacteria contaminating surimi multiplied vigorously during incubation. The pH levels of the surimi samples were decreased during incubation with or without inoculation. This reduction in pH was associated with acid production by bacteria, because acid-producing bacteria were detected in both of the incubated surimi samples. The effect of inoculation using Lactobacillus helveticus JCM1004 on the viable bacterial counts was slight in the fermented surimi. It is difficult to inhibit the vigorous growth of contaminating bacteria via inoculation with lactic acid bacteria. Thus, the reduction of bacteria in surimi is necessary for stable fermentation and to prevent the risks of spoilage and food poisoning.
| Incubation time hour | pH | Viable bacteria cfu/g | Acid-producing bacteria cfu/g | |
|---|---|---|---|---|
| Before incubation | 0 | 7.29 | 2.90 × 103 | Not analyzed |
| Control | 24 | 4.64 | 1.31 × 109 | 1.90 × 109 |
| (without inoculum) | 48 | 4.69 | 2.77 × 108 | 2.97 × 108 |
| Fermented | 24 | 4.59 | 7.13 × 108 | 7.57 × 108 |
| (with inoculum) | 48 | 4.71 | 3.97 × 108 | 3.43 × 108 |
Data were obtained in triplicate. The surimi sample used in this test was a different lot of A and B in Table 3.
| Sample | Manufacturer | Grade | Country | Place | Bacterial count (cfu/g) | Thermoresistant bacterial count (cfu/g) |
|---|---|---|---|---|---|---|
| A | 1 | A | United States | Factory trawl ship | 7.37 × 104 | 6.67 |
| B | 1 | A | United States | Factory trawl ship | 2.30 × 104 | n.d. |
| C | 1 | B | United States | Factory trawl ship | 1.52 × 105 | 1.33 × 101 |
| D | 2 | B | United States | Shore plant | 7.03 × 104 | 3.33 |
| E | 3 | A | United States | Shore plant | 9.67 × 103 | n.d. |
Each sample was measured in triplicate. Sample B was different lot from A.
n.d., not detected.
To design a bactericidal process using heat treatment, we prepared five different Alaska pollock surimi samples. The surimi samples used in this study and their microbiological characteristics are shown in Table 3. Approximately 104–105 cfu/g of viable bacteria were detected and a few thermoresistant bacteria were also found in each surimi sample (Table 3). These findings are consistent with previous research (Himelbloom et al., 1991). These bacteria, culturable at 37 °C, can grow under conditions of lactic acid fermentation.
To assess bacterial diversity, 20 strains were randomly selected for DNA extraction from the viable bacterial colonies from each surimi sample. Additionally, five thermoresistant bacterial strains were selected, one from A, three from C, and one from D. DNA profiling of the five surimi samples was performed using the RAPD-PCR method (Fig. 1). The number of independent strains identified in each sample was as follows: A, 17/21; B, 12/20; C, 18/23; D, 12/21; and E, 15/20. A total of 74 independent strains were detected from 105 isolates, indicating that the bacteria contaminating the surimi samples were diverse.
DNA profile using randomly amplified polymorphic DNA polymerase chain reaction (RAPD-PCR). A, B, C, D, and E indicate the DNA profiles derived from five different surimi samples. M indicates the 100 bp DNA ladder (Takara Bio Inc., Shiga, Japan). Isolates are represented with numbers. Electrophoresis was carried out using 2.0 % agarose gel in 1 X TBE buffer at 100 V for 30 min.
To identify the bacteria contaminating surimi, a phylogenetic tree of the 16S rRNA sequences from the 74 independent strains and 25 representative strains was constructed (Fig. 2). At the phylum level, the bacterial profile of Alaska pollock surimi was dominated by Pseudomonadota (51/74), Bacillota (12/74), Bacteroidota (6/74), and Actinomycetota (5/74). The dominant genera were Pseudomonas and Psychrobacter, followed by Chryseobacterium, Acinetobacter, and Carnobacterium (Table 4). Pseudomonas and Psychrobacter were isolated from all of the tested surimi samples. These two genera were previously identified to be surimi spoilage bacteria (Wang et al., 2021 and Huang et al., 2022). Other genera were also isolated from some surimi samples, indicating that the microbial diversity differed among the manufacturers, grades, and/or lots.
Phylogenetic tree using maximum likelihood method for V4, V5 and V6 regions of 16S rRNA. Bootstrap values for a total of 100 replicates are shown at the nodes of the tree. The bar shows 10 % sequence divergence.
| Genera | A | B | C | D | E | Total |
|---|---|---|---|---|---|---|
| Pseudomonas | 9 | 2 | 3 | 3 | 13 | 30 |
| Psychrobacter | 6 | 10 | 7 | 1 | 1 | 25 |
| Chryseobacterium | 0 | 0 | 2 | 3 | 1 | 6 |
| Acinetobacter | 3 | 0 | 1 | 9 | 0 | 13 |
| Carnobacterium | 1 | 4 | 3 | 0 | 0 | 8 |
| Bacillus | 0 | 0 | 0 | 1 | 0 | 1 |
| Others | 2 | 4 | 7 | 4 | 5 | 22 |
| Total | 21 | 20 | 23 | 21 | 20 | 105 |
These results suggest that the dominant genera in surimi are non-spore-forming bacteria. Spore-forming bacteria were only isolated from surimi D (D-21). The spore-forming strain was classified into the Bacillus genus in the phylogenetic tree, and 16S rDNA sequence analysis indicated 99.7 % homology with Bacillus racemilacticus (Accession number, D16279). Although four other strains survived the heat treatment, they were identified as non-spore-forming bacteria. Only a minimal number of thermoresistant bacteria were found in surimi made from Alaska pollock.
As the dominant genera isolated from surimi are non-spore-forming bacteria, pasteurization is an adequate bactericidal process for surimi. The manufacturing standard for heat treatment of kamaboko is set at 75 °Ci). In fact, no viable bacteria were detected from the five surimi samples tested in the present study after heat treatment at 75 °C for 30 min except for surimi E. The viable counts of surimi E were also minimal (< 10 cfu/g). The differences between this result and that shown in Table 3 are due to the detection limit of the dilution plate method because of the low thermoresistant bacterial count in the surimi samples. Therefore, kamaboko (the pasteurized surimi) is suitable for preparing materials for fermentation because of its low viable bacterial counts. However, some non-spore-forming bacteria survived pasteurization. Thus, the temperature and time of the pasteurization should be adjusted to the conditions of surimi, for example the size, shape, the bacterial diversity and the chemical composition (including the salt concentration).
In conclusion, the bacterial profile of Alaska pollock surimi was dominated by Pseudomonas and Psychrobacter, which are known surimi spoilage bacteria. Almost all bacteria in the surimi samples were non-spore-forming, even though bacterial diversity differed among the surimi samples. Thus, pasteurization equivalent to the heat treatment standard for kamaboko before bacterial inoculation is necessary for hygienic fermentation of surimi. However, gel formation of surimi is induced during the heating process (Nguyen et al., 2020). We predict that differences in the structural strength between raw surimi and kamaboko affect lactic acid fermentation. Therefore, further study is needed to design a suitable manufacturing process for lactic acid fermentation of surimi.
Acknowledgements We thank Dr. Fukunaga, Mr. Yoshimizu, and Mr. Sato for their kind advice on this study.
Conflict of interest There are no conflicts of interest to declare.
