1. Introduction
Vibrio parahaemolyticus is a non-spore-forming [
1], halophilic [
2], and thermophilic [
3] Gram-negative bacterium [
4]. Attributed to this viable-but-nonculturable (VBNC) [
5], infectious agent [
3,
4] and, hence, uneasy to detect, the number of
V. parahaemolyticus-associated seafood-borne vibriosis [
6,
7] is increasing worldwide [
3,
8]. This gastroenteritis-causing pathogen [
9] accounts for ca 45% [
6,
10] and 0.1% [
11] of the total vibriosis and toll of foodborne illnesses (48 million) each year in the US, respectively, and is the leading cause of foodborne infections in China [
12]. Infected patients, including all age groups, may carry various symptoms such as watery diarrhea, abdominal cramping, gastroenteritis, nausea, vomiting, fever, wound/soft tissue infections, bacteremia, low blood pressure, blistering skin lesions, and possibly death [
6,
13]. As such, the US Centers for Disease Control and Prevention (CDC) Cholera and Other
Vibrio Illness Surveillance (COVIS) [
10] and the Foodborne Disease Active Surveillance Network (FoodNet) [
14] systems have been closely monitoring this pathogen since 1996. Additionally, the USDA Safe Minimum Internal Temperature chart [
15] published cooking conditions for seafood, including shellfish, that requires a minimum internal cooking temperature of 63 °C to ensure safe consumption.
Persistent, increasing vibriosis incidents associated with
V. parahaemolyticus are generally linked to the consumption of undercooked seafood contaminated with the pathogen [
16,
17] through their ecological niche [
2] and improperly sanitized environment [
18]. However, increasing
V. parahaemolyticus prevalence has been detected in processed, ready-to-eat foods, suggesting that this pathogen can survive the harsh conditions of the food processes [
9]. Hemolysin-related genes, such as thermostable-direct hemolysin (
tdh), thermostable-direct-related hemolysin (
trh), and thermolabile hemolysin (
tlh) [
19,
20], are the primary virulence determinants of
V. parahaemolyticus. In addition, most clinical strains possess the
trh gene alongside the urease-producing gene (encoded by the
uh gene) [
21].
Unlike the thermal resuscitation technique, which uses extreme heat [
22], VBNC
V. parahaemolyticus resuscitation requires an enrichment-based cultivation technique, such as a one- or two-step enrichment technique, that involves alkaline peptone water (APW) selection alone or combined nutrient nourishment and APW (or drug) selection, respectively, followed by plating on selective agar containing thiosulfate-citrate-bile-salt-sucrose (TCBS) [
23,
24]. Of all the enrichment techniques documented, several groups noted that the heated saline enrichment (42 °C) technique was conducive to
Vibrio growth [
25,
26,
27,
28,
29,
30,
31]. However, both Delmore and Crisley [
32] and Andrew et al. [
33] noted that the heat process using temperatures between 49 and 55 °C was detrimental to this pathogen. Hence, these studies suggest that
V. parahaemolyticus is a dormant-enabled, non-spore former, as spore former resuscitation requires a much higher temperature [
22].
In light of the presence of VBNC
Vibrio species, laboratories have employed antibody-labeled microscopy assays [
13,
34,
35], gene-specific multiplex PCR [
13,
19,
36], and one- and/or two-step enrichment-based cultivation techniques [
37,
38] for recovering this type of
Vibrio. These techniques, however, compromise the feasibility of detecting all other 200 serogroups of cholera-causing
Vibrio [
39], as well as contribute to false-negative results as indigenous contaminants populate [
40], which suggest that an efficient technique is needed for the accurate validation of food products for total pathogenic
Vibrio contaminants, including the VBNC strains.
Collectively, these literature reports suggest the need of the present study to re-evaluate the conventional detection technique and cooking recommendation for V. parahaemolyticus containment in seafood products. To the best of our knowledge, the present work has not been noted in peer-reviewed literatures.
V. parahaemolycus-associated vibriosis that is deemed to be a persistent foodborne issue with the increasing number of cases worldwide [
6,
7] may be attributed, in part, to the presence of evasive
V. parahaemolyticus from a conventional detection technique employed today, alterated cooking conditions, and lack of inspection benefits. The present study devised the conventional technique of APW enrichment [
41], followed by heating selection [
22], defined as a modified, two-step enrichment technique, to resuscitate/recover a group of heat-dependent, VBNC
V. parahaemolyticus that could evade the conventional detection technique. The efficacy of the modified, two-step enrichment technique for detecting VBNC
V. parahaemolyticus strains was simultaneously evaluated with the one-step enrichment technique (i.e., uses APW enrichment alone) [
41]. The biology of this type of
V. parahaemolyticus, which distinguishes it from the generally known, non-spore-like type of this pathogen, was characterized for virulence implications, which provides data for more effective validation for seafood safety. In this study, estuarine, lagoon, and wet-market seafoods were examined to isolate a group of VBNC
V. parahaemolyticus requiring heat-dependent resuscitation and TCBS agar plating. Virulence determination tests in the literature, such as the Kanagawa phenomenon test, virulence genes PCR test, urease test, and antibiotic disk diffusion test, were subsequently applied to strains of
V. parahaemolyticus and confirmed by 16S rDNA bacterial identification.
2. Materials and Methods
Sample collection. A total of eight shellfish samples (
Table 1), including environmental samples (3, crab; 2, oyster) and wet market samples (3, shrimp), were collected from the Joe Patti seafood market (wet market, shrimp collection) around Santa Rosa Island (lagoon, oyster collection) and Gulf Island National beach (estuarine environment, crab collection) of Pensacola, Florida in early Fall (19 September 2017). All packaged samples were kept in an icebox (~1 °C) during transportation back to the laboratory, stored at refrigeration temperature (0–2 °C), and processed within a week. Environmental conditions at sample collection sites were determined to be pH 8.5 and temperature 30.3 °C.
Isolation, culture, and storage conditions. Strains of
V. parahaemolyticus used in this study were isolated in the lab using a two-step enrichment technique described previously [
37] with minor modifications (
Figure 1). Briefly, microorganisms in a 25 g sample were homogenized in 225 g of enrichment broth containing APW (pH 8.6, 5%
wt/v) using a Stomacher
® 400 Circulator lab blender (Seward, Weber Scientific, Hamilton, NJ, USA), incubated at specific times, and bacterial dilutions were inoculated on thiosulfate-citrate-bile-salt-sucrose (TCBS, Difco, Detroit, MI, USA) agar plates or heat treated (i.e., 80 °C, 20 min followed by 4 °C, overnight) prior to the plating, followed by an incubation of 72 h at 35 °C before visible colonies were streaked on Brain Heart Infusion agar (BHI, Difco, Detroit, MI, USA) containing 3% NaCl for purification, with pure cultures being stored in a sterile solution containing BHI and glycerol (10%
v/v) at −70 °C. Presumptive
V. parahaemolyticus isolates were confirmed using 16S rDNA bacterial identification and species-gene-specific PCR (
tlh). The bacterial cultures were prepared by inoculating thawed cultures (1/100) in sterile fresh culture solution containing BHI (pH 7.4) supplemented with 3% NaCl, followed by overnight incubation (~20 h) at 35 °C and a repeated subculture before experimental analyses, as described previously [
18].
Heat challenged viability assay. Sub-cultured, fresh, select strains of V. parahaemolyticus were serially diluted to a final cell concentration of 102–103 cfu/mL in LB broth. Dilution aliquots were incubated at 62 °C for 8 h in a water bath, followed by overnight incubation at 35 °C, and were then examined for growth by visible broth turbidity and viable plate count on agar containing BHI and 3% NaCl.
Hemolysis test. Hemolysis activity was examined using sheep erythrocytes [
42] or human erythrocytes [
16] as described previously with minor modifications. Select strains of
V. parahaemolyticus sub-cultured twice in sterile BHI broth were spotted (5 µL) onto agar plates (pH 7.4) containing BHI, 1.5% human erythrocytes, and 1% NaCl. Inoculated plates were incubated at 35 °C for 24 h prior to hemolysis validation.
DNA extraction, PCR, and agarose gel electrophoresis conditions. Genomic DNA was prepared with a previously documented bead extraction method [
43]. Briefly, pelleted fresh
V. parahaemolyticus cultures (16 h) were resuspended in sterile DI water and spun down twice before being subjecting to bead collision with sterile micro-size beads (5 µm) in 0.1 mL sterile Tris buffer (10 mM, pH 7.4) to release the cytoplasmic components using a pulsing vortex mixer (SPW Industrial, Laguna Hills, CA, USA). Chromosomal DNA was phased out from cell debris using a high-speed microcentrifuge (VWR, Suwanee, GA, USA) and stored at −20 °C. The quality of DNA was analyzed using a UV spectrophotometer (Thermo Scientific, South San Francisco, CA, USA) and PCR using gene-specific primers.
PCR mixtures for the amplification of genes were formulated according to the GoTaq Flexi DNA Polymerase’s instructions (Promega, Madison, WI, USA). Briefly, each reaction mixture contained 5× PCR buffer (Promega), 0.4 μM of gene-specific primers (
Table 2) (IDT, Coralville, IA, USA), 1.5 mM MgCl
2 (Promega), 0.2 mM deoxynucleoside triphosphate mix (Fisher Scientific, Fair Lawn, NJ, USA), and 1.25 U of GoTaq polymerase (Promega). The PCR conditions were PCR step 1: 1 cycle of 5 min DNA denaturation at 95 °C; step 2: 40 cycles of 1 min DNA denaturation at 95 °C, 40 s primer-dependent annealing, 72 °C of DNA extension (gene-dependent incubation time); and followed by PCR step 3: 1 cycle of extended DNA extension for 10 min at 70 °C before infinite holding at 4 °C in a GeneAmp PCR System 9700 Thermal Cycler (Applied Biosystems, Thermo Scientific).
PCR products (5 µL) were examined in a 1.6% agarose gel (formed in 1× Tris-borate-EDTA buffer) pre-stained with a GelStarTM Nucleic Acid Gel Stain (ratio 5 µL stain: 50 mL gel solution, Lonza Walkersville Inc., Walkersville, MD, USA) using a UV transilluminator.
Urease assay. V. parahaemolyticus urease activity was evaluated according to the protocol of a commercial urease testing kit. Briefly, a test reaction containing 1 mL of fresh cell solution of V. parahaemolyticus isolates (<20 h) and one tablet of Urease Test Tablets Key Scientific (Hardy diagnostics, Santa Maria, CA, USA) was mixed with a vortex machine and incubated at 37 °C for a maximum of 24 h, or until a pink color indicating a positive test developed.
Antibiotic disc diffusion test. A comparative antibiotic assay for distinguishing strains of
V. parahaemolyticus was performed using the disc diffusion test with minor modifications [
46]. Briefly, BHI agar plates supplemented with 3% NaCl were inoculated for bacterial lawns with 10
8 cfu/mL fresh
V. parahaemolyticus cultures (~20 h) by using sterile glass rods. The antibiotic discs (BD; Sparks, MD, USA) chloramphenicol (CHL, 30 µg/disc), ciprofloxacin (CIP, 5 µg/disc), erythromycin (ERY, 15 µg/disc), gentamycin (GEN, 10 µg/disc), nalidixic acid (NAL, 30 µg/disc), neomycin (NEO, 30 µg/disc), streptomycin (STR, 10 µg/disc), and tetracycline (TET, 30 µg/disc) were distantly placed on the lawns. The inoculated plates were incubated for 24 h before antibiotic susceptibility diameters (mm) were determined and then scored per the Clinical and Laboratory Standards Institute (CLSI) M45 [
47] and M100-S21(M2) (i.e., found in the antibiotic product user manual) guidelines for
Vibrio species and enterococci/
Escherichia coli, respectively.
4. Discussion
The ability of
Vibrio species to form a viable-but-non-culturable (VBNC) state and become resistant to stresses, such as high salt content, pH, high temperatures [
42,
48], and antibiotics [
49,
50] has recently started to shed light on the persistent cases of vibriosis [
35,
51,
52,
53,
54]. Previous studies separately reported by Mizunoe et al. [
5], Okuda et al. [
21], and Xu et al. [
55] have suggested the presence of genotype and phenotype variabilities of
V. parahaemolyticus, which could result in detection inconsistency/limit of one or many other detection techniques reported to date.
Conventionally,
Vibrio spp. or spore-forming bacteria enrichment techniques use nutrient cultivation followed by selective enrichment [
23,
31,
56] or heat treatment followed by cold incubation [
22]. In this study, however, we combine SPW enrichment, heat-treatment, and cold incubation for resuscitating
V. parahaemolyticus. This novel, combined enrichment and heat-selection technique (i.e., modified, two-step enrichment technique) resuscitated a group of environmental, heat-resistant
V. parahaemolyticus strains (41.7%), in addition to regular
V. parahaemolyticus strains (58.3%). This suggests the presence of a distinctive genotype and phenotype of
V. parahaemolyticus that could remain unculturable by many conventional detection/cultivation techniques. Odeyemi noted that
V. parahaemolyticus is more prevalent in oysters than other kinds of seafood investigated by various groups between 2003 and 2015, as demonstrated in a meta-analysis study [
3]. This phenomenon was consistent with the prevalence findings (in this study), whereby
V. parahaemolyticus could only be retrieved from oysters when analyzed along with other seafood samples. Hence, the observations suggest that oyster may be a
V. parahaemolyticus’s preferred symbiont, including the heat-resistant strains reported for the first time in this study, compared with other kinds of shellfish seafood which are deemed frequent targets of contamination by
V. parahaemolyticus. It is worth noting that this technique has not been documented in prior reports, and that the use of an extended SPW enrichment time (i.e., 72 h) in this study is novel (
Table 3).
Delmore and Crisley [
32] and Andrews et al. [
33] noted that
V. parahaemolyticus is susceptible to temperatures between 49 and 55 °C. Cebrián et al. [
57] and Terano et al. [
22] noted numerous non-
Vibrio species of Gram-negative bacteria that are capable of withstanding pasteurization temperature (i.e., 65 °C). The heat-resistant strains of
V. parahaemolyticus investigated in this study were able to survive extended pasteurization conditions (i.e., 62 °C, 8 h) as opposed to other vegetative cells of spore-forming bacteria [
58] or non-spore forming bacteria [
57]. This could suggest the differential expression of heat-shock proteins, such as DnaI [
42], DnaK [
48], GroEL, and GroeS [
42,
48] chaperones, whose role is implicated in bacterial maintenance of heat-deactivated proteins to revert their structure(s) and function(s) [
59]. Both
V. parahaemolyticus VHT1 and VHT2 strains exhibited different heat-tolerance phenotypes (VHT79, VHT80, VHT81) at post-pasteurization temperature (62 °C), as demonstrated by turbidimetric data (35 °C) (
Figure 2) and followed by viable plate count (>8 log cfu/mL cells) data (
Figure 3); nevertheless, these strains are susceptible to the resuscitation temperature (80 °C) used to recover them (
Table 6). It is worth noting that the heat-resistant colonies of VHT80 and VHT81 derived from VHT2 (i.e., pasteurized VHT2) were different phenotypically and that our findings suggest, for the first time, that
V. parahaemolyticus could tolerate pasteurization temperatures (62–65 °C). The latter strain (VHT81) appeared to have sticky colonies. This could be attributed to the heat-dependent alteration of Gram-negative bacteria’s outer membrane, which leads to non-lethal LPS damage [
60] or homeoviscous adaptation [
61]. In contrast to Delmore and Crisley [
32], both strains survived a much higher incubation temperature than elevated temperatures between 49 and 55 °C.
KP was prepared using sheep erythrocytes as described by Ming et al. [
62], as it is more economical than human erythrocytes. However, agar containing BHI was used for the reason that Wagatsuma agar [
16] is costly, and that the
V. parahaemolyticus strains did not grow well on agar plates containing Luria Bertani. In our study, human erythrocytes were able to demonstrate detectable hemolysis activity, disagreeing with Ming et al. [
62]. The findings suggest that the
V. parahaemolyticus heat-resistant strains VHT1 and VHT2 could possess one or more major hemolysin-producing gene(s) of the virulence strains, such as thermostable direct hemolysin (
tdh) and/or
tdh-related hemolysin (
trh) [
38], and that their derivative strains VHT79 (from VHT1), VHT80 (from VHT2), and VHT81 (VHT2) retained their hemolysis phenotypes post pasteurization (
Figure 4 and
Table 7). It is worth mentioning that all KP
+ strains tested, including VHT1 and VHT2, did not exhibit hemolysis activity when cultured on agar media supplemented with sheep erythrocytes instead of human erythrocytes (
Figure 4 and
Table 7).
Wang et al. [
63] noted that the Kanagawa hemolytic test could render false virulence results for
V. parahaemolyticus and that
tdh-specific PCR analysis could be used to confirm their virulence property. Letchumanan et al. [
38] further noted that the non-virulence factor-associated hemolysin gene thermolabile hemolysin (
tlh) could be employed for detecting total
V. parahaemolyticus. Our PCR data revealed that regular
V. parahaemolyticus strains and heat-resistant
V. parahaemolyticus strains possess
tdh−/
tlh− and
tdh−/
tlh+ hemolysin genotypes, respectively (
Table 7). It is worth noting that this is the first case of environmental
V. parahaemolyticus with
tlh−. Additionally, the data demonstrated that these KP
+ V. parahaemolyticus strains (i.e., VHT79, VHT80, VHT81) possess
tdh−/
tlh+ virulence factors (
Table 7) and that this hemolysin gene is phenotypically (
Figure 4) and genotypically (
Figure 6) stable to pasteurization conditions.
Xu et al. [
55] and others [
64,
65,
66,
67,
68] noted the presence of urease activity in
V. parahaemolyticus and that it positively correlates/co-exists with the presence of the thermostable direct hemolysin-related hemolysin (
trh) gene [
21,
64,
65,
68,
69,
70]. However, the other groups reported that urease activity is usually found in non-clinical strains of
V. parahaemolyticus [
65,
66,
67,
68], which is in opposition to Xu et al. [
55]. When urease activity was analyzed for regular (VHT17, 18) and heat-resistant (VHT1, 2, 14, 15, 16) strains of environmentally isolated
V. parahaemolyticus in this study, both groups exhibited positive urease reaction (
Table 7).
Park et al. [
65] also noted that the urease and
trh gene cluster does not positively correlate with the Kanagawa phenomenon (KP). However, the present study revealed that select strains, the heat-resistant, urease
+ V. parahaemolyticus VHT1 and VHT2, were KP
+ (
Table 7). Letchumanan et al. [
38] and Chung et al. [
71] noted other
V. parahaemolyticus hemolysins, such as the thermolabile hemolysin (
tlh) and thermostable direct hemolysin (
tdh), respectively, that could contribute to KP
+. Hemolysin gene-specific PCR analyses revealed that the
V. parahaemolyticus strains VHT1 and VHT2 were
tdh− but
tlh+ (
Table 7). It is worth noting that this is the first report of the discovery of urease
+, KP
+ environmental
V. parahaemolyticus [
65].
The different antibiotics used in this study were adapted in part from previous documented work [
18,
46] and the availability of the antibiotics in the lab. Elexson et al. [
72] and Letchumanan et al. [
18] noted penicillin/ampicillin-resistant strains of
V. parahaemolyticus from environmental and cultured seafood, respectively. When the antibiotic susceptibility test was compared between the heat-resistant
V. parahaemolyticus VHT1 and VHT2, both demonstrated high levels of resistance to penicillin.
The availability of heat-resistant
V. parahaemolyticus in seafoods suggest the potential cause of vibriosis food outbreak by a heat-resistant group of
V. parahaemolyticus due to their ability to evade the conventional protocol of cooking and testing [
15] and become a food safety threat and vibriosis cause. Our findings suggest, for the first time, that
V. parahaemolyticus could survive food processing and cooking criteria [
15], which could be an overlooked root cause of the persistent vibriosis in the US, and that they could be activated/recovered using the modified, two-step enrichment of combined nourishment (APW, 48 h), heating selection (80 °C, 20 min), and cooling steps. As previously noted, the prevalence of the
V. parahaemolyticus variant [
5,
21,
55], such as heat-resistant strains of environmental
V. parahaemolyticus, could be attributed to specific gene transfers within the bacterial complex/mixture community, that could consist of
Vibrio species and heat-resistant symbionts [
73], and induced by global warming [
74]. The results revealing positive KP activity and
tlh amplicons suggest the presence of environmental
V. parahaemolyticus pathogens [
19,
20], and that these strains (i.e., VHT1 and VHT2 derived strains, VHT79, VHT80, and VHT81) (
Table 5 and
Table 7) possess heat-stable hemolysin gene(s), such as
tlh, and that it encodes a heat-stress protecting product, possibly suggesting a positive correlation between
V. parahaemolyticus heat resistance (i.e., 62 °C, 8 h) and persistent vibriosis cases, as noted by Ueda et al. [
75], whose work determined a positive correlation between thermolabile protein expression and thermotolerance in a member of Gram-negative bacteria. The PCR results confirm that this pathogen could use
tdh-independent hemolysin(s), such as
trh [
38,
71],
tlh [
2,
38], VP3048 putative hemolysin [
48], and/or others [
71] for exhibiting KP activity (
Table 7), and that urease-positive
V. parahaemolyticus strains do not simultaneously possess a
tdh-encoding gene, as noted by Wang et al. [
63]. Additionally, the data suggest that
tdh and
tlh hemolysins are not essential in environmental
V. parahaemolyticus metabolism, as these genes were absent in a subset (
tlh gene) or all (
tdh gene) of our culture isolates (
Table 6). Further work (i.e., transcriptomic analyses) exploring the transcriptomic response of this group of
V. parahaemolyticus strains to extreme temperature (i.e., 62 °C) is worthwhile for fostering a seafood processing/cooking and safety regimen. The results demonstrating positive urease activity suggest that both urease-positive heat-resistant and regular
V. parahaemolyticus (
Table 6) could possess a
trh gene. Several groups reported that bacteria could produce urease when exposed to nitrogen limitation [
76], acidic pH [
77], or urea availability [
78,
79]. Berutti et al. [
79] noted the increasing level of nitrogen content in environmental water since 2008, which suggests the adoption of urease in environmental
V. parahaemolyticus. Thus, this indicates that environmental
V. parahaemolyticus could readily produce urease in response to the urea availability supplemented in the urease assay. The results demonstrating high levels of resistance to penicillin could be attributed to the outer membrane of the Gram-negative bacteria, which interferes with antibiotic drugs’ entry and antimicrobial efficacy [
46,
80], as well as to the overuse of antibiotics in aquaculture that leads to bacterial resistance [
81,
82]. VHT2 was relatively more susceptible to ciprofloxacin and nalidixic acid (
Figure 7) than VHT1, suggesting an antibiotic-resistance diversity of heat-resistant, environmental
V. parahaemolyticus. Collectively, the results suggest that the
V. parahaemolyticus strains of VHT1 and VHT2 (heat-resistant strains) possess
tlh,
trh (presumptive), urease, and KP activities.