Molecular Characterization of Histamine-Producing Psychrotrophic Bacteria Isolated from Red Octopus (Octopus maya) in Refrigerated Storage

The present study aimed at determining the histamine production capacity of Gram (+) and Gram (−) bacteria isolated from Octopus maya, along with identifying the presence of amino acid decarboxylase genes. Of the total 80 psychrotrophic microorganisms, 32 strains were identified as histamine-forming bacteria. The recombinant DNA technique was used for genotypic identification of histidine (hdc), ornithine (odc), and lysine decarboxylases (ldc) genes. Thirty-two strains were able to produce 60–100 ppm in trypticase soy broth with 1.0% l-histidine after 6 h at 20 °C. NR6B showed 98% homology with Hafnia alvei. NR73 represented 18.8% of the total isolates and showed 98% homology with Enterobacter xianfengensis and Enterobacter cloacae. NR6A represented 6% of the total isolates, which were identified as Lactococcus sp. The hdc gen from NR6B showed 100% identity with hdc from Morganella morganii; ldc showed 97.7% identity with ldc from Citrobacter freundii. The Odc gene was detected only in NR73 and showed 100% identity with Enterobacter sp. All the isolated were identified as weak histamine–former. The ingestion of a food containing small amounts of histamine has little effect on humans; however, the formation of biogenic amines is often considered as an indicator of hygienic quality; this emphasizes the importance of improving good management practices and storage.


Introduction
Mexico ranks among the top 10 exporters of octopus to the United States, representing approximately 13% of the market. US octopus imports from Mexico have increased steadily, with almost 18,000 t imported in 2014. The fishery is mainly based on the red octopus Octopus maya, which is endemic to the Yucatan Peninsula, and a unique species among the 150-known species of octopus worldwide [1]. The extraction is done in a small-scale fleet with vessels between 8 and 10 m in length near the coastal areas (2-5-fathom depth) whereby the organisms have a high microbial load on the surface of the skin at the time of capture [2]. These daily trip-boats are not well equipped for cold storage, which renders the octopus a highly perishable product even in subsequent refrigerated storage.
Histamine (C 5 H 9 N 3 ) has been identified as the main causative agent of scombrotoxic fish poisoning (often called "histamine poisoning") which is caused by ingestion of certain species of marine fish that contain high levels of histamine and possibly other biogenic amines (BAs) [3]. The severity of the symptoms can vary, depending on the amount of histamine and other biogenic amines ingested and the individual's sensitivity to specific BAs. Since fish handling practices are critical with regard to BAs production, histamine was defined as a chemical hazard regulated by the US Food and Drug

Amplification of Histidine (Hdc) Ornithine (Odc) and Lysine Decarboxylase (Ldc) Genes
The amplification of the target genes was carried out using oligonucleotides designed by De las Rivas et al. [31]. The hdc gene was amplified using the primer set 106 (5 AAY TCNTTYGAYTTYGARAARGARG3 ) and 107 (5 ATN GGN GAN CCD ATC ATY TTR TGN CC 3 ) that amplifies a region of 534 bp. The odc gene was amplified using PUT1-F (5 TWYMAYGCNGAYA ARACNTAYYYTGT3 ) and PUT1-R (5 ACRCANAGNACNCCNGGNGGRTANGG3 ) amplifying a region of 1440 bp. Finally, the ldc gene was amplified with CAD1-F (5 TTYGAYWCNGCNTGGGTN CCNTAYAC3 ) and CAD1-R (5 CCRTGDATRTCNGTYTCRAANCCNGG3 ) with an amplification region of 1098 bp. The primer set 106/107 failed to amplify the hdc gene from the NR6A strain, which was then amplified by JV16HC (5 AGA TGG TAT TGT TTC TTA TG 3 ) and JV17HC (5 AGA CCA TAC ACC ATA ACC TT 3 ) primer set, specific for Gram-positive bacteria. This primer set generated a PCR product of 367-bp [32].
Amplification was performed in a 50 µL reaction mixture that contained 25 µL of DreamTaq Green PCR Master Mix (2X) (Thermo Scientific, Waltham, MA, USA) 75 pmol of each primer and 2 µg of DNA. The amplification was carried out for 35 cycles (95 • C for 30 s, 58 • C for 30 s and 72 • C for 1 min) in a Thermal Cycler (Major Science, Sea Gull Way Saratoga, CA, USA) with an initial denaturation of 94 • C for 4 min and a final extension at 72 • C for 7 min. The PCR products were separated on a 1.5% agarose gel at 80 V in 1X TAE buffer (400 mM Tris-acetate, 10 mM EDTA, pH = 8. 2-8.4). The amplified fragments containing ethidium bromide (0.5 µg/mL) were visualized in a 3UV-Benchtop transilluminator (UVP, Upland, CA, USA). The product size was confirmed by comparison with a 100-bp molecular marker (Invitrogen, Carlsbad, CA, USA). The amplified products were purified using the Wizard SV Gel and PCR Clean-Up System (Promega Corp., Madison, WI, USA). Finally, the purified products were eluted and solubilized in nuclease-free water, quantified by fluorescence (Quantus Flurorometer E6150, Promega Corp., Madison, WI, USA) and stored at −20 • C for subsequent cloning and sequencing.

Cloning and Sequencing of Decarboxylase Genes
The amplicons from decarboxylase genes were treated with polyethylene glycol, cooled on the ice for 1 h, and pelleted by centrifugation at 15,000× g for 20 min. The pellet was washed with 70% ethanol, dried, and dissolved in Tris-EDTA buffer. The DNA fragments were ligated into a pGEM ® vector using the pGEM ® -T Easy Vector System I ligation kit (Promega Corp., Madison, WI, USA) and transformed into Escherichia coli JM109 competent cells [33]. The ligation reaction was performed at 26 • C for 1 h and 4 • C for 48 h using 150 µL of competent cells. Heat shock was given by incubating the cells at 0 • C for 20 min, 42 • C for 45 s, and at 0 • C for 2 min. Subsequently, 1 mL of Luria-Bertani (LB) broth was added to each reaction and incubated at 37 • C with shaking. After 1 h, the cells were centrifuged at 5000× g for 15 min and re-suspended in 150 µL of LB broth. The transformed colonies were identified on LB agar plates containing ampicillin (100 µg/mL), 0.5 mM IPTG (isopropyl-β-D-thiogalactoside), and X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) (80 µg/mL) at 37 • C. After 18 h, the transformed cells with the insert (white colonies) and without the insert (blue colonies) were selected and cultured on LB agar plates with ampicillin (100 µg/mL) to extract the plasmid DNA.

Identification of Decarboxylase Genes in Transformed Cells
The identification of the cloned decarboxylase gene segments was carried out from the extraction of the plasmid DNA from the transformed cells by the alkaline lysis method [33]. Briefly, 1.5 mL of overnight TSBH culture of transformed cells was centrifuged at 2600× g for 3 min at 4 • C (Eppendorf, 5804R, Westbury, NY, USA). The pellet was then re-suspended in 100 µL of a cold solution containing 50 mM glucose, 25 mM Tris-Cl, 10 mM sodium EDTA at pH 8.0 and 200 µL of a solution consisting of 0.2 N NaOH and 1% SDS. After 3-5 min of cold rest, 150 µL of a cold solution containing 60 mL of 5 M potassium acetate, 11.5 mL of glacial acetic acid, and 28.5 mL of H 2 O were added. The sample was centrifuged at 2600× g for 10 min, and the pellet was washed with phenol/chloroform 1:1 (v/v). After centrifugation, the final pellet was washed with 100% and 70% ethanol, subsequently, and reconstituted with TE buffer containing 20 µg/mL of pancreatic RNAase.
The presence of the insert was verified using universal primers pUC/M13F (5 GTT TTC CCA GTC ACG AC 3 ) and pUC/M13R (5 CAG GAA ACA GCT ATG AC 3 ). The PCR amplification was performed in a 50 µL reaction mixture using 25 µL of the Dream Taq™ Green PCR Master Mix kit (2X), 1 µL of 2 pmol of each primer, and 2 µL of 1 µg of plasmid DNA sample. The amplification was carried out in 30 cycles (94 • C for 60 s, 55 • C for 60 s, and 72 • C for 2 min) with an initial denaturation at 94 • C for 5 min and a final extension at 72 • C for 10 min. The cloned DNA segments were sequenced by an automated DNA sequencer based on Sanger's method (Applied Biosystems 3730xl; Thermo Fisher Scientific, Waltham, MA, USA). The deduced protein sequences were compared with UniProtKB database sequences (Swiss-Prot and TrEMBL). The sequences were aligned with ClustalW using MEGA 7.0 software [28,29]. The evolutionary relationship among the amino acid sequences was inferred by the Neighbor-Joining method [34]. The bootstrap consensus tree inferred from 500 replicates was taken to represent the relationships of the sequences analyzed [35]. The distances were computed using the Poisson's correction method [36].

Determining of the Concentration of Histamine Produced by Bacteria
The concentration of histamine produced by the strains was determined by direct enzyme-linked immunosorbent assays (Multiskan EX, Thermo Scientific, Rockford, IL, USA) using the Veratox ® commercial kit. One colony was suspended in 50 mL of trypticase soy broth supplemented with 1.0% L-histidine (TSBH) and incubated at 20 • C in a shaking incubator at 150 rpm (Thermo Fisher Scientific, Waltham, MA, USA). At every hour, an aliquot of 3 mL was removed from the culture and used to estimate the total concentration of bacteria in soy trypcasein agar (TSA) and cell transmittance using a UV-visible spectrophotometer at 540 nm (GENESYSTM 10S, Thermo Fisher Scientific, Waltham, MA, USA). The remnant sample was centrifuged at 2600× g 15 min and the supernatant was used to quantify the histamine concentration.

Molecular Identification of Histamine Producing Strains
Among the total 80 bacterial isolates from the muscle tissues, 32 strains were identified as histamine-producing by PCR. Table 1 shows the identity of the HFB determined by comparison of the sequences with NCBI-16S rDNA database. NR6B strain, which represented 15% of the total isolates, showed 98% homology with Hafnia alvei (NR_112985.1). NR73 represented 18.8% of the total isolates and 15 of them had 98% homology with Enterobacter xiangfangensis (NR_126208.1) and Enterobacter cloacae (NR_118568.1). Finally, the strain NR6A which represented 6% of the total isolates was identifying as Lactococcus sp.

Phenotypic Characterization of Histamine Producing Strains
Of the 80 strains tested, 38 were identified as histamine forming by the Niven's agar method but only 31 strains were positive by the molecular PCR-identification. With the exception of a Lactococcus sp. most of the histamine producing strains were Gram-negative. Two strains, viz., NR6B and NR73, from the Enterobacteriacea group, were characterized by catalyzing the decomposition of hydrogen peroxide to oxygen as well as the lack of cytochrome oxidase activity. Both strains were negative for the production of sodium thiosulfate, sodium pyruvate, and indole, and positive for glucose fermentation and nitrate reduction.
On the other hand, an immobile strain NR6A strain was catalase and oxidase negative and homofermentative, degrading glucose via glycolysis to produce lactic acid. Table 2 shows the complete phenotypic characterization of the HFB from red octopus. Table 2. Phenotypic identification of histamine-forming bacteria (HFB) isolated from the red octopus (Octopus maya).

Genotypic Identification of Histidine, Ornithine, and Lysine Decarboxylase Genes
The detection of the decarboxylating genes by PCR confirmed the ability of the isolates to decarboxylate BAs and produce histamine, cadaverine, and putrescine. The 106/107 primer set (hdc gene) generated a typical PCR product of 534-bp for H. alvei NR6B and for Enterobacter NR73 (Figure 1a). This primer set failed to amplify the hdc in Lactococcus NR6A, which was later amplified by the JV16HC/JV17HC primers resulting in a 367-bp PCR product (Figure 1b). The odc gene was only detected in Enterobacter NR73 and generated a 1400-bp product (Figure 1c). The ldc gene was detected in all strains giving an amplicon of 1098 bp (Figure 1d). The results of the sequencing of decarboxylase enzymes are presented in Table 3. gene) generated a typical PCR product of 534-bp for H. alvei NR6B and for Enterobacter NR73 ( Figure  1a). This primer set failed to amplify the hdc in Lactococcus NR6A, which was later amplified by the JV16HC/JV17HC primers resulting in a 367-bp PCR product (Figure 1b). The odc gene was only detected in Enterobacter NR73 and generated a 1400-bp product (Figure 1c). The ldc gene was detected in all strains giving an amplicon of 1098 bp (Figure 1d). The results of the sequencing of decarboxylase enzymes are presented in Table 3.   The e-value, match length, and identity percentage for the referred alignment are shown in Table 4. The evolutionary relationship of the taxa based on the amino acid sequences of decarboxylase enzymes and inferred by the Neighbor-Joining method is shown in Figure 2. decarboxylase enzymes and inferred by the Neighbor-Joining method is shown in Figure 2.  The evolutionary history was inferred using the Neighbor-Joining method. The bootstrap consensus tree inferred from 500 replicates was taken to represent the evolutionary history of the taxa analyzed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 replicates) are shown next to the branches.

Histamine Production
The concentration of histamine produced of the isolated strains in TSBH medium is shown in Figure 3. The histamine concentration was increased as a function of growth time. In the first six hours, the concentration of H. alvei NR6B, Lactococcus NR6A, and Enterobacter NR74 attained up to 9.0-10.5 log CFU/mL, but the production of histamine was variable. The highest histamine High-Throughput 2018, 7, 25 9 of 14 concentration was produced by Lactococcus NR6A reaching 100 ppm at 6 h of growth. H. alvei NR6B produced 60 ppm in the same time and Enterobacter NR74 reached less than 40 ppm in 6 h of growth.

Histamine Production
The concentration of histamine produced of the isolated strains in TSBH medium is shown in Figure 3. The histamine concentration was increased as a function of growth time. In the first six hours, the concentration of H. alvei NR6B, Lactococcus NR6A, and Enterobacter NR74 attained up to 9.0-10.5 log CFU/mL, but the production of histamine was variable. The highest histamine concentration was produced by Lactococcus NR6A reaching 100 ppm at 6 h of growth. H. alvei NR6B produced 60 ppm in the same time and Enterobacter NR74 reached less than 40 ppm in 6 h of growth.

Discussion
Three bacterial strains isolated from the refrigerated octopus' muscle were identified as BAforming and were identified as Hafnia, Enterobacter sp. and Lactococcus sp. Hafnia was identified up to species level as Hafnia alvei (DNA group 1).
There are only a few studies on the bacteria related to the production of BAs in octopus. Lougovois et al. [37] studied the spoilage potential of fresh musky octopus (Eledone moschata) by

Discussion
Three bacterial strains isolated from the refrigerated octopus' muscle were identified as BA-forming and were identified as Hafnia, Enterobacter sp. and Lactococcus sp. Hafnia was identified up to species level as Hafnia alvei (DNA group 1).
There are only a few studies on the bacteria related to the production of BAs in octopus. Lougovois et al. [37] studied the spoilage potential of fresh musky octopus (Eledone moschata) by evaluating the changes in the biochemical properties, microbial growth, and sensory quality of the mantle and tentacles over a period of 18 days, and reported that the members of the Enterobacteriaceae were present only in small numbers. Other authors reported Pseudomonas spp., Alteromonas putrefaciens, Flavobacterium spp., Shewanella spp., Acinetobacter spp., and Photobacerium as prevalent bacterial species found in fish during cold storage; however, these genera have rarely been confirmed as HFB [26,38]. It is true that mesophilic Enterobacteriaceae exist as a minor bacterial flora in seafood stored at 0 • C, but psychrotolerant Enterobacteriaceae are capable of growing at refrigeration temperature and associate with the presence of BAs [26]. In the present study, 15 Enterobacter strains codified as NR73 were isolated from the muscle of red octopus and identified as HFB. Other Enterobacter species, e.g., Enterobacter intermedium, have previously been isolated from Bluefin tuna (Thunnus thynnus) at 8 • C confirming not only their adaptability to low temperatures but also the ability to produce histamine [39]. Identification at the species level in the genus Enterobacter is difficult on the basis of 16S rDNA, as the genus is polyphyletic [40]. The strain Enterobacter NR73 showed 98% identity with Enterobacter xiangfangensis sp. nov, and Enterobacter cloacae indicating a close relationship; however, the concatenated partial analysis based on the sequence of rpoB, atpD, gyrB, and infB genes should be analyzed in future studies to confirm a species [41].
Hafnia alvei was isolated from fresh and temperature-abused albacore (Thunnus alalunga) and found to be HFB by Kim et al. [16]. The genus Hafnia originally contained a single species, H. alvei, which was later described to include another species H. paralvei and some of the strains previously designated as the now obsolete Obesumbacterium proteus. H. alvei is genetically heterogeneous and consists of at least two DNA hybridization groups [42,43]. Recently, hybridization studies have distinguished the original H. alvei into H. alvei sensu stricto (DNA group 1) and H. paralvei, formerly known as H. alvei hybridization group 2 [44]. Based on the results of the biochemical tests carried out by Abbot et al. [45] for the identification of Hafnia species, the NC6B strain isolated in the present study was identified as Hafnia alvei (DNA group 1). H. alvei sensu stricto is typically malonate, salicin, and β-glucosidase positive, and D-arabinose negative, while H. paralvei has an opposite pattern. In our study, salicine, esculine, malonate, and β-glucosidase were typically positive and D-arabinose was negative.
Even knowing that decarboxylase is considered as species-specific or sometimes even strain specific activity, in the present study, Hafnia alvei NR6B and Lactococcus NR6A were identified as potential histamine and cadaverine producers, and Enterobacter NR73 as histamine, cadaverine, and putrescine producer. In fact, several studies have reported that the genera isolated from different foodstuffs show lysine and ornithine decarboxylase activity [46]. The ldc gene was found in the Enterbocateriaceae strains (Hafnia NR6B and Enterobacter NR73) amplifying a specific region of 1098-bp, but this region was also found in Streptococcaceae (Lactococcus NR6A). The primer set, CAD1-F, and CAD1-R, used in this study, were previously designed for some representative species of Enterobacteriaceae such as Shigella flexneri, Shigella sonnei, and Escherichia coli [31]. In the present study, this oligonucleotide set also amplified the ldc region from Gram-positive Lactococcus NR6A, which showed 97.7% identity with ldc from Citrobacter freundii (A0A0J1MZ87). The non-amplication of the hdc gene from Lactococcus NR6A with the primer set 106/107 was expected since the oligonucleotides were previously developed considering the two distinct classes of histidine decarboxylase enzymes; the homometric pyridoxal 5-phosphate (PLP)-dependent enzyme from Gram-negative bacteria [47], and the heterometric enzyme that contains an essential pyruvoyl group from Gram-positive bacteria [32,48].
It is difficult to predict the presence of cadaverine and putrescine in the octopus on the basis of gene expression. De Filippis et al. [49] mentioned that the expression of the decarboxylase genes, and the consequent production of cadaverine and putrescine in some microorganisms of the family Enterobacteriaceae, is influenced by the temperature, where a low temperature is associated with gene downregulation. In fact, the strains isolated from octopus, in particular, H. alvei NC6B and Enterobacter NC73 were found to be weak HFB compared with those reported by previous authors. For example, Raoultella planticola and M. morganii isolated from fish were identified as prolific histamine former with an ability to produce 1000 to 4000 ppm in 24 h at 37 • C in a TSBH medium [16,50,51].
The time and temperature of storage play a significant role in histamine accumulation. Torrido et al. [26] measured the histamine production of mesophilic and psychrotrophic isolates and concluded that, unlike the mesophilic, more than half of the 118 psychrophilic isolates produced no more than 500 ppm histamine. H. alvei, in particular, produced >88.7 ppm in fish infusion broth at 30 • C [52] but was identified as weak histamine former (42.1 ppm) in TSBH at 15 • C [53]. Kim et al. [38] also mentioned that only a few isolates of H. alvei and E. cloacae can produce >1000 ppm histamine in culture broth and the most of them are weak histamine formers. In our study, the histamine concentrations produced by the psychrotrophic Enterobacter NC73 and H. alvei NC6B after 6 h of incubation in TSBH were 40 and 60 ppm, respectively. This production classifies them as weak histamine forming strains.
Very little is known about the Gram-positive bacteria isolated from seafood and their role as HFB. The genus Tetragenococcus, created after reclassification of the halophilic lactic acid bacterium Pediococcus halophilus as T. halophilus is a Gram-positive microorganism isolated from salted and fermented fish products and was identified as an HFB [23,54]. A study reported the histamine production of 10.67 ppm from several strains of the Lactobacillus genus after 10 days at 25 • C [55]. This is relatively low compared to the 100 ppm produced for Lactococcus NC6A after 6 h in TSBH. The presence of the genus Lactoccoccus in a refrigerated octopus is unusual and possibly its presence is related to cross-contamination.

Conclusions
Although it is generally believed that histamine fish poisoning is caused by mesophilic histamine-producing bacteria that are active when temperature abuse occurs, this study showed that psychrophilic HFB are also prevalent in refrigerated octopus. To the best of our knowledge, this is the first report to demonstrate the occurrence of HFB in octopus; this emphasizes the importance of improving good management practices and storage. Although some of the enteric bacteria are naturally present in octopus's tissues, most seem to arise from handling during harvesting and storage [24].
The isolated strains were found to produce a low level of histamine in the culture broth. The ingestion of food containing small amounts of histamine has little effect on humans; however, the formation of BAs is important not only from the standpoint of their toxicity, their presence is often being used as an indicator of hygienic quality and possible fecal contamination. Rapid chilling and refrigeration are mandatory to prevent microbial growth and consequent histamine formation. These are the main suggested control measures for seafood. In this sense, this study could assist in identifying better critical control points to minimize contamination of BAs-forming microorganisms in red octopus.