Persistence of Vibrio parahaemolyticus in the Pacific Oyster, Crassostrea gigas, Is a Multifactorial Process Involving Pili and Flagella but Not Type III Secretion Systems or Phase Variation

Abstract

Vibrio parahaemolyticus can resist oyster depuration, suggesting that it possesses specific factors for persistence. We show that type I pili, type IV pili, and both flagellar systems contribute to V. parahaemolyticus persistence in Pacific oysters whereas type III secretion systems and phase variation do not.

TEXT

The genus Vibrio consists of a group of bacteria that naturally inhabit aquatic environments worldwide. The human pathogen Vibrio parahaemolyticus is commonly found associated with shellfish, particularly oysters (1, 2). Depuration is a controlled process in which shellfish are placed into clean seawater to reduce the bacterial contaminants in their tissues. However, depuration is not very effective at reducing the numbers of V. parahaemolyticus cells in oysters (3), and little is known about the association of V. parahaemolyticus with oysters.

Both type I and type IV pili are important for bacterial attachment to a variety of surfaces (4–6). V. parahaemolyticus encodes a homologue of type I pili that is most similar to the CsuA/B operon from Acinetobacter baumannii (7) and homologues of two well-studied type IV pili from vibrios, the mannose-sensitive hemagglutinin (MSHA) and the chitin-regulated pilus (PilA) (http://img.jgi.doe.gov/) (8, 9). Vibrio vulnificus uses PilA for persistence in Crassostrea virginica (10, 11). Both type IV pili are involved in biofilm formation in V. parahaemolyticus (12). Flagella are often critical during early stages of bacterial colonization of a surface (13, 14). V. parahaemolyticus possesses a single, polar flagellum used for movement in liquid (reviewed in reference 14) as well as peritrichous (lateral) flagella for surface movement (15–17 and reviewed in reference 18). Many bacterial pathogens use type III secretion systems (T3SSs) for survival in the host by injecting virulence factors directly into the host. V. parahaemolyticus encodes two T3SSs that exhibit different phenotypes during disease (19–21). V. parahaemolyticus also exhibits “phase variation” and can switch from an opaque (OP) to a translucent (TR) phenotype based on polysaccharide production (22). In this study, we examined the role of V. parahaemolyticus pili, flagella, phase variation, and T3SSs in persistence in the Pacific oyster, Crassostrea gigas.

Oysters were collected from Oregon Oyster Farm (Newport, OR) and exposed to ∼10⁵ CFU/ml of V. parahaemolyticus for 16 to 18 h at room temperature (∼20°C) with a recirculating pump. Algae were added to the tank to facilitate uptake according to the manufacturer's recommendation (Phytoplex; Kent Marine). Depuration was conducted at 19 to 20°C for 48 to 72 h. At each time point, animals from each exposure tank were weighed, homogenized, and diluted with phosphate-buffered saline (PBS) for enumeration on tryptic soy agar (TSA) supplemented with 1.5% NaCl and either 100 μg/ml streptomycin, 60 μg/ml chloramphenicol, 10 μg/ml phosphomycin, or 50 μg/ml kanamycin, depending on the strain (see Table S1 in the supplemental material), to eliminate naturally occurring background flora. The data were log transformed prior to data analysis. All strains are compared with their respective parental strain using a one-way analysis of variance (ANOVA) and Tukey's honestly significant difference (HSD) post hoc comparison (P < 0.05). Different letters indicate statistically significant differences between strains for each time point.

To test the role of pili in colonizing the Pacific oyster, V. parahaemolyticus ATCC 17802-defined gene deletions were created in the type I pilin gene csuB, the type IV pilin genes mshA and pilA, a double gene deletion in mshA and pilA, and a single gene deletion in the type IV pilin peptidase pilD (see Table S1 in the supplemental material). The ΔmshA mutant was not defective in uptake, but during depuration, there was significantly less recovered (log CFU/g) after 12 h and 48 h of depuration (Table 1). Unlike the ΔmshA mutant, the ΔpilA, ΔmshA/ΔpilA, and ΔpilD mutants had defects in uptake after the bacterial exposure (t = 0), although no difference was observed for the amount of bacteria added to each tank (data not shown). During depuration, no difference was observed for the ΔpilA mutant (log CFU/g) until 48 h of depuration. The ΔmshA/ΔpilA and ΔpilD mutants were recovered significantly less (log CFU/g) after 12 h and 30 h of depuration. By 48 h of depuration, there was no difference in the ΔmshA/ΔpilA mutant (log CFU/g) recovery but significantly less ΔpilD mutant (log CFU/g) was recovered. As for the type I pilus mutant ΔcsuB, there were no defects in uptake or after 4 h and 24 h of depuration but there were significantly fewer bacteria recovered over the 48 h of depuration (log most probable number [MPN]/g) (Table 2). From these data, it appears that V. parahaemolyticus uses both type I and type IV pili to persist in the Pacific oyster during the depuration process. To verify that these gene deletions did not have a polar effect on neighboring genes, biofilm formation was performed. Type IV pilus mutants had biofilm defects similar to those of previous reports (12), and the type I pilus mutant also had similar defects in biofilm formation; complementation restored these mutants to wild-type levels (data not shown). These depuration results are consistent with what was observed in V. vulnificus, where loss of pilA and pilD resulted in a decrease in bacteria recovered from the tissues of C. virginica compared with that recovered from the wild type (10). It has been previously published that loss of pilD in V. vulnificus also results in loss of type II secretion (23). Although loss of type II secretion was not examined in this study, it cannot be ruled out as playing a possible role in persistence in the oyster during depuration.

Table 1.

Comparison of V. parahaemolyticus wild type and type IV pilus mutants in Pacific oysters during depuration

Strain	Indicated value at each time pointa
	0 h (log mean CFU/g)	12 h		30 h		48 h
		Log mean CFU/g	Log reduction	Log mean CFU/g	Log reduction	Log mean CFU/g	Log reduction
ATCC 17802	5.08 ± 0.10 A	3.82 ± 0.18 A	1.26 ± 0.15 A	2.50 ± 0.22 A	2.58 ± 0.19 A	1.57 ± 0.29 A	3.62 ± 0.27 A
VPAA9 ΔmshA	4.66 ± 0.14 AB	2.72 ± 0.21 B	1.94 ± 0.20 A	1.41 ± 0.29 AB	3.24 ± 0.19 A	0.48 ± 0.23 B	4.50 ± 0.29 A
VPAA27 ΔpilA	4.43 ± 0.14 B	2.98 ± 0.14 AB	1.44 ± 0.17 AB	1.55 ± 0.22 AB	2.88 ± 0.20 AB	0.40 ± 0.19 B	4.16 ± 0.24 A
VPAA28 ΔmshA/ΔpilA	4.43 ± 0.12 B	2.14 ± 0.28 BC	2.3 ± 0.26 BC	1.29 ± 0.28 B	3.15 ± 0.35 AB	0.56 ± 0.20 AB	3.87 ± 0.28 A
VPAA26 ΔpilD	4.56 ± 0.13 B	1.73 ± 0.27 C	2.80 ± 0.21 C	0.70 ± 0.18 B	3.88 ± 0.20 B	0.30 ± 0.15 B	4.50 ± 0.28 A

^aThe log mean or log reduction and standard error are shown. Different letters indicate statistical significance for each time point according to the one-way ANOVA (treatment) and post hoc comparisons using Tukey's HSD test (P < 0.05). These data represent four independent experiments with 12 animals analyzed at each time point.

Table 2.

Comparison of the V. parahaemolyticus wild type and type I pilus mutant in Pacific oysters during depuration

Strain	Indicated value at each time pointa
	0 h (log mean MPN/g)	4 h		24 h		48 h
		Log mean MPN/g	Log reduction	Log mean MPN/g	Log reduction	Log mean MPN/g	Log reduction
ATCC 17802	4.31 ± 0.21 A	3.86 ± 0.25 A	0.44 ± 0.17 A	3.04 ± 0.18 A	1.27 ± 0.14 A	1.88 ± 0.15 A	2.43 ± 0.06 A
VPAA8 ΔcsuB	4.27 ± 0.22 A	3.51 ± 0.10 A	0.76 ± 0.13 A	2.41 ± 0.23 A	1.86 ± 0.22 B	1.08 ± 0.19 B	3.19 ± 0.11 B

^aThe log mean or log reduction and standard error are shown. Different letters indicate statistical significance for each time point according to the one-way ANOVA (treatment) and post hoc comparisons using Tukey's HSD test (P < 0.05). These data represent two independent experiments with 10 animals analyzed at each time point.

To examine the role of flagellar systems, polar (ΔflaM LM5392), lateral (ΔlafK LM7789), and completely nonmotile (ΔflaK/ΔlafK LM7901) flagellar system mutants (see Table S1 in the supplemental material) were tested for persistence in the Pacific oyster during depuration. Only ΔflaM was defective in uptake compared to its isogenic parent LM5431 TR (Table 3). Examining these mutants over 48 h of depuration, all flagellar system mutants, ΔflaM, ΔlafK, and ΔflaK/ΔlafK, were recovered significantly less (log CFU/g) than was LM5431 at 24 h and 48 h. These data suggest that both polar and lateral flagellar systems are important for V. parahaemolyticus persistence in the Pacific oyster.

Table 3.

Comparison of V. parahaemolyticus wild type and flagellar mutants in Pacific oysters during depuration

Strain	Indicated value at each time pointa
	0 h (log mean CFU/g)	24 h		48 h		72 h
		Log mean CFU/g	Log reduction	Log mean CFU/g	Log reduction	Log mean CFU/g	Log reduction
LM5432 OP	4.62 ± 0.21 AB	4.03 ± 0.21 A	0.66 ± 0.14 A	3.64 ± 0.16 A	1.05 ± 0.13 A	2.99 ± 0.49 A	1.71 ± 0.40 A
LM5431 TR	4.83 ± 0.20 A	4.21 ± 0.19 A	0.62 ± 0.11 A	3.53 ± 0.11 A	1.30 ± 0.14 A	3.07 ± 0.46 AB	1.96 ± 0.36 A
LM5392 ΔflaM	3.54 ± 0.41 B	2.12 ± 0.27 B	1.60 ± 0.27 B	1.14 ± 0.34 B	2.58 ± 0.31 B	1.26 ± 0.38 B	2.46 ± 0.31 A
LM7789 ΔlafK	4.33 ± 0.34 AB	2.64 ± 0.37 B	1.69 ± 0.21 B	1.88 ± 0.47 B	2.45 ± 0.32 B	1.77 ± 0.46 AB	2.55 ± 0.31 AB
LM7901 ΔflaK/ΔlafK	4.73 ± 0.30 AB	2.57 ± 0.43 B	2.16 ± 0.28 B	1.41 ± 0.33 B	3.32 ± 0.27 B	1.32 ± 0.30 B	3.42 ± 0.29 B

^aThe log mean or log reduction and standard error are shown. Different letters indicate statistical significance for each time point according to the one-way ANOVA (treatment) and post hoc comparisons using Tukey's HSD test (P < 0.05). These data represent four replicate experiments with 15 animals analyzed for each time point.

Oysters possess an innate immune system and have macrophage-like hemocytes that are responsible for bacterial killing (24). Therefore, to persist in the oyster, bacteria likely have to overcome their immune defenses (25). When the T3SS double mutant was tested (LM7341), there was no significant different between it and the wild type (TR) (data not shown), suggesting that V. parahaemolyticus does not utilize either T3SS for persistence in the Pacific oyster. V. parahaemolyticus may use other unidentified factors or tactics to circumvent the oyster immune system for survival in the oyster host.

One possibility for oyster immune system evasion is phase variation in colony morphology from the opaque to the translucent phenotype, a result of altering surface polysaccharide production (26). In our analyses, we compared strains LM5431 (TR) and LM5432 (OP) to determine if phase variation affected persistence in the oyster. Interestingly, the OP and TR strains were not statistically different during the course of depuration (Table 3), suggesting that phase variation is not involved in persistence in the Pacific oyster. These findings are similar to what was observed in V. vulnificus, where a translucent variant, capable of reverting back to the opaque phenotype, showed no difference in recoverability compared with the opaque wild type in whole oyster homogenates of C. virginica (11).

In conclusion, V. parahaemolyticus persistence in the Pacific oyster is a complex process involving two types of pili, type I and type IV, as well as polar and lateral flagellar systems but not T3SSs or phase variation. This study has provided a glimpse into the intricate Vibrio-oyster interplay. Future studies will expand on this knowledge, and it will be interesting to explore the effects of chitinous particles and/or mannose on improving depuration efficiency of V. parahaemolyticus. Also, exploring other parameters of depuration that may affect flagellar production, i.e., salinity and pH combinations, may also improve depuration conditions, ultimately leading to a highly efficient and cost-effective postharvest processing technique.