Resuscitation of Viable but Non-Culturable Vibrio Parahaemolyticus in a Minimum Salt Medium

FEMS Microbiology Letters 233 (2004) 269–275 www.fems-microbiology.org

Resuscitation of viable but non-culturable Vibrio parahaemolyticus in a minimum salt medium

Hin-chung Wong *, Peily Wang, Shau-Yan Chen, Shen-Wen Chiu

Department of Microbiology, Soochow University, Taipei 111, Taiwan, ROC

Received 11 June 2003; received in revised form 12 February 2004; accepted 22 February 2004

First published online 3 March 2004

Abstract

Vibrio parahaemolyticus is food-borne pathogen prevalent in Asian countries. This work analyzes factors that inﬂuence the resuscitation of the viable but nonculturable (VBNC) state in this bacterium. The MMS-0.5% NaCl medium alone limited cell multiplication, and in this medium, resuscitation was successful when the temperature was upshifted to 25 °C but not 37 °C. Chloramphenicol inhibition experiments revealed that protein synthesis in the ﬁrst 24 h of temperature upshift was critical in de- termining the success of the three-day resuscitation period. The VBNC state induction period and the age of the VBNC cells for successful resuscitation were strain-dependent. Results of this work facilitate further physiological and pathological study of the VBNC state in this pathogen. Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Vibrio parahaemolyticus; Viable but non-culturable; Resuscitation; Temperature upshift

1. Introduction during the winter. VBNC V. parahaemolyticus has been developed by starving cells at low temperature Vibrio parahaemolyticus, a halophilic Gram negative [5], but has not been well characterized. They also bacterium, causes acute gastroenteritis in humans. reported that the resuscitation of VBNC cells was the This bacterium is a prevalent food-borne pathogen in results of cell regrowth of a limited amount of cul- Japan, Taiwan, and other coastal countries [1,2]. Most turable cells [5]. Mizunoe et al. [6] reported on the clinical isolates are hemolytic on Wagatsuma agar recovery of nonculturable V. parahaemolyticus cells in (Kanagawa-positive, KP+) and produce a major vir- agar medium containing catalase or pyruvate. How- ulence factor, thermostable direct hemolysin (TDH) ever, the cells were probably injured to become non- [1]. culturable by the low salinity used in the study, and V. parahaemolyticus inhabits seawater, but is seldom later recovered by the presence of these protective isolated when the temperature of the seawater is under agents instead of resuscitation of VBNC cells. In this 13–15 °C [3]. Like other vibrios that inhabit similar report, the nonculturable bacterial cells were still rod marine environments [4], V. parahaemolyticus may be shaped with blebs instead of the usual coccoid VBNC present in a viable but nonculturable (VBNC) state cells [6]. This investigation analyzes the resuscitation process of VBNC V. parahaemolyticus triggered by temperature upshift in a minimum salt medium which alone limited * Correspondence author. Tel.: +886-2-28819471x6852; fax: +886-2- multiplication of cells. The induction and resuscitation 28831193. of various strains of V. parahaemolyticus were also E-mail address: [email protected] (H.-c. Wong). compared.

0378-1097/$22.00 Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsle.2004.02.015 270 H. Wong et al. / FEMS Microbiology Letters 233 (2004) 269–275

2. Materials and methods and 1.21 g Tris buffer (pH 7.8) in one liter of deionized water to prevent the carry-over of nutrients [7]. The 2.1. Cultures and media bacterial cells were suspended in the 250-ml Erlenmeyer flask that contained 100 ml MMS-0.5% NaCl medium at V. parahaemolyticus ST550, a serotype O4:K13 and a concentration of about 107 cells mlÀ1, and incubated KP+ clinical strain isolated in Thailand, was used in this at 4 °C in a static state, to induce the VBNC state. study. A total of 20 clinical and 4 environmental strains Bacteria were counted every seven days. were used for comparison (Table 1). These strains were stocked in culture broth with 10% (v/v) glycerol at 2.3. Resuscitation )85 °C, and cultured in Tryptic Soy Broth Medium (TSB, Difco Laboratories, Detroit, Mich.)-3% (w/v) NaCl or The temperature upshift method was used to resus- Tryptic Soy Agar (TSA, Difco)-3% NaCl at 25 °C. citate VBNC V. parahaemolyticus [5]. The temperature of the VBNC culture with under one culturable cell per 2.2. Induction of VBNC state milliliter was changed from 4 to 25 or 37 °C and incubated for one to three days. Viable and culturable cells Bacteria were cultured in 50 ml TSB-3% NaCl me- were counted. Nalidixic acid at 5, 10, 20, 50, 100, or 200 dium in a 250-ml Erlenmeyer flask, and incubated at lgmlÀ1 was added to assay inhibition of bacterial cell 25 °C with shaking at 110 rpm for 4.5 h to reach the multiplication during the temperature upshift [5]. The mid-exponential phase. Bacterial cells were harvested by VBNC culture was also diluted 10-fold to assay the ef- centrifugation at 6000g for 15 min, washed twice in fect of dilution, and diluted 5-fold into TSB-3% NaCl or modified Morita mineral salt solution (MMS-0.5% MMS-0.5% NaCl media, to assay the effect of nutrients NaCl) which consisted of 5 g NaCl, 0.8 g KCl, 5.6 g on resuscitation. Chloramphenicol (100 lgmlÀ1) was MgCl2 Á 6H2O, 5.6 g, 7.6 g MgSO4 Á 7H2O, 0.9 mg added to the VBNC culture to determine the importance FeSO4 Á 7H2O, 1.54 g CaCl2 Á 2H2O, 0.1 g Na2HPO4 of protein synthesis in the resuscitation [8].

Table 1 Induction and resuscitation of VBNC state by diﬀerent strains of Vibrio parahaemolyticus Strain no. Date of isolation Serotype KP VBNC by (day)a Resuscitable period (day)b Clinical ST550 1983 O4:K13 + 35–49 14 1264 1998 O5:K68 ND 28–35 14 1265 1998 O5:K68 ND 28–35 <7 1292 1999 O2:K3 + 35–42 <7 1079 1997 O3:K6 + 35–42 7 1089 1997 O3:K6 + 42–49 7 1091 1997 O3:K6 + 42–49 <7 1092 1997 O3:K6 + 35–42 14 1099 1997 O3:K6 + 42–49 14 1109 1997 O3:K6 + 42–49 7 1119 1997 O3:K6 + 35–42 7 1129 1997 O3:K6 + 35–42 7 1132 1997 O3:K6 + 14–28 7 1137 1997 O3:K6 + 28–35 <7 1147 1997 O3:K6 + 28–35 <7 1188 UN O3:K6 + 14–28 7 1288 1996 O3:K6 + 35–42 <7 1289 1996 O3:K6 + 28–35 7 1290 1996 O3:K6 + 28–35 7 1291 1996 O3:K6 + 35–42 <7 Environmental 1256 1999 ND ND 14–28 7 1257 1999 ND ND 28–35 <7 1258 1999 ND ND 28–35 <7 1266 1998 ND ND 28–35 7 All of the strains except no. ST550 (Thailand), 1264 and 1265 (Singapore) were isolated from clinical or seawater samples from diﬀerent locations in Taiwan. Induction of VBNC state and resuscitation of these strains were determined. a Time in day required to induce VBNC state. b Time in day for the cells in VBNC state successfully resuscitated by temperature upshift method; ND, not determined; KP, Kanagawa phe- nomenon; UN, unknown. H. Wong et al. / FEMS Microbiology Letters 233 (2004) 269–275 271

2.4. Enumeration of bacteria 9

8 The culturability of bacteria was determined by the plate count method. Decimal dilutions were obtained in 7 -1 MMS-0.5% NaCl at 4 °C and plated on TSA-3% NaCl. 6 The plates were incubated at 37 °C for 16 h and the 5 number of colonies counted. Duplicate experiments were performed and the data were obtained as means of 4 three determinations in each experiment. 3

The total number of viable cells was counted using Log cell number ml 2 the LIVE/DEAD BacLight Bacterial Viability Kit (Molecular Probe Inc., Eugene, OR, USA) following the 1 procedures of Mizunoe et al. [6]. This kit has been used 0 to determine the number of total and viable cells of 0 1020304050 several other bacteria [9–11]. Time, day Effect of catalase on the culturability was assayed Fig. 1. Induction and resuscitation of VBNC Vibrio parahaemolyticus according to the method of Mizunoe et al. [6]. Plate in MMS-0.5% NaCl medium. The cells were cultured at 4 or 25 °C. counts were performed using agar plates with or without The 4 °C culture was shifted to 25 °C (arrow) when all cells entered the the supplementation of filter-sterilized Aspergillus niger VBNC state. Viable cell counts and plate counts were determined. Plate counts are given in cfu mlÀ1. Vertical bars designate the standard catalase (Sigma Co., St. Louis, MO, USA), 2000 U per errors. d, plate count of the 25 °C culture; ,, viable count of the 25 °C plate. culture; j, viable count of the 4 °C culture; ., plate count of the 4 °C The resuscitating bacteria were also enumerated by culture without catalase; s, plate count of the 4 °C culture catalase. the most probable number method (MPN) [5]. Decimal dilutions were obtained in MMS-0.5% NaCl. One milliliter of the diluted samples was inoculated into MPN significantly enhance the culturability during the VBNC tubes containing 9-ml of MMS-0.5% NaCl and incu- induction period (Fig. 1). bated at 25 °C in a static state for three days. Equal volume of TSB-3% NaCl was added to the MPN tubes 3.2. Resuscitation of VBNC cells and incubated for one more day. Resuscitation of the VBNC cells was performed by 2.5. Determination of minimum inhibition concentration temperature upshift treatment to 25 °C in the MMS- 0.5% NaCl medium. In this minimal medium, prolifer- An aliquot of 5 llofV. parahaemolyticus ST550 ation of cells of V. parahaemolyticus was not possible culture was inoculated into 0.1 ml of TSB-3% NaCl in due to low salinity and nutrient limitation (Fig. 1). microplate wells supplemented with different concen- When the temperature was increased to 25 °C in the trations of nalidixic acid (5–1000 lgmlÀ1). The micro- resuscitation of VBNC cells, the number of viable cells plate was incubated at 37 °C for 16 h and bacterial remained unchanged, while the culturability resumed growth was determined by a microplate reader (MRXII, rapidly, reaching a maximum level in two days, which Dynex Technologies, Inc., Chantilly, VA, USA). value was near to the initial cell level. No significant increment was observed on the third day (Figs. 1 and 2). When the VBNC cells were treated at 37 °C, the viable 3. Results count remained high and constant, while no culturable cell was recovered on the TSA-3% NaCl medium (data 3.1. Induction of VBNC state not shown). The resuscitated cells were also enumerated by a When the bacterial cells at exponential phase were MPN method. The VBNC culture was diluted into incubated at 25 °C in MMS-0.5% NaCl medium (pH MPN tubes containing MMS-0.5% NaCl medium and 7.8) in a static state, the viable and culturability counts incubated at 25 °C for three days. A level of 108 cells per were similar and only dropped by about one log cell ml was observed close to the initial inoculation level number per milliliter at the end of the incubation period (Fig. 1). (Fig. 1). When the bacterial cells were incubated at 4 °C, Nalidixic acid inhibited the multiplication of cells culturability counts declined while the viable counts and thereby prevented the regrowth of culturable cells were at high level. Zero culturability was reached after during the temperature upshift [5]. The minimum inhi- incubating at low temperature for about 35–49 days in bition concentration (MIC) of nalidixic acid against repeated experiments (Fig. 1). Plate counts with the V. parahaemolyticus was determined to be less than supplementation of 2000 U catalase per plate did not 5 lgmlÀ1. Nalidixic acid (5–200 lgmlÀ1) was added to 272 H. Wong et al. / FEMS Microbiology Letters 233 (2004) 269–275

7 12 11 6 10

-1 5 9

-1 8 4 7 3 6

Log cell number ml 2 5 4

1 Log cell number ml 3 0 01232 Time, day 1

0 Fig. 2. Resuscitation of VBNC Vibrio parahaemolyticus cells by tem- 0 10203040 perature upshift treatment in diluted sample. When all cells entered the Time, h VBNC state, the cells suspended at the original density (d, s)orina 10-fold diluted sample (., ,) were incubated at 25 °C for three days. Fig. 3. Resuscitation of VBNC Vibrio parahaemolyticus cells in nutri- Viable cell counts (open symbols) and plate counts (solid symbols) ent-rich medium. When all cells entered the VBNC state, the cell were determined. Plate counts are given in cfu mlÀ1. suspension was 5-fold diluted in MMS-0.5% NaCl (d, s) or TSB-3% NaCl (., ,) and incubated at 25 °C for 48 h. Viable cell counts (open symbols) and plate counts (solid symbols) were determined. Plate the resusctitation medium, MMS-0.5% NaCl, before the counts are given in cfu mlÀ1. temperature upshift treatment. However, the rate of culturable cell recovery and the maximum number of recovered cells were the same regardless of the presence 8 of nalidixic acid (data not shown). If all VBNC cells were resuscitated, the plate count in 7 the 10-fold diluted sample would be expected to recover 6 to 10% of the level in the undiluted sample. The VBNC -1 cultures were diluted 10-fold and then subjected to 5 temperature upshift. Plate counts reached a maximum 4 of 106 cells mlÀ1 in two days and remained constant thereafter (Fig. 2). 3

Nutrition in the resuscitation medium may influence Log cell number ml the resuscitation of VBNC cells. The VBNC V. para- 2 haemolyticus cells were harvested and resuspended in 1 TSB-3% NaCl or MMS-3% NaCl and incubated at 0 25 °C for 48 h. In the MMS-3% NaCl medium, the 01234 number of culturable cells increased immediately and Time, day reached the original cell density, remaining constant thereafter (Figs. 2 and 3). In the TSB-3% NaCl medium Fig. 4. Inhibition of resuscitation of VBNC Vibrio parahaemolyticus cells by chloramphenicol. When all cells entered the VBNC state, they broth, a lag period of 12 h was observed and then the were moved to MMS-0.5% NaCl and incubated at 25 °C for three 10 culturability and the viability increased, reaching 10 days. Chloramphenicol was not added (d), or was added at 100 lg cells mlÀ1, far in excess of the initial cell density (Fig. 3). mlÀ1 to the VBNC cells on the zeroth (s), first (.), second (,) or third When chloramphenicol was added at the beginning or (j) day during the temperature upshift. Plate counts were determined À1 the first day of the temperature upshift process, no and given in cfu ml . culturable cells were recovered. During the second day or third day of the temperature upshift, adding chlor- bility was examined by plate counts on TSA-3% NaCl amphenicol only reduced the culturable cells by about after 0, 14, 28, 35, 42, and 49 days to induce the VBNC one log level or less cells mlÀ1 (Fig. 4). state. After 35, 42, 49, 56 and 63 days, the cultures underwent temperature upshift treatment, and the 3.3. Induction and resuscitation of VBNC state in number of culturable cells was determined. The length different strains of time for the induction of the VBNC state (14–49 days) was strain-dependent. After certain period of The cultures were separately suspended in MMS- time in VBNC state, the cells could not be resuscitated 0.5% NaCl medium and incubated at 4 °C. Cultura- by the temperature upshift treatment. The VBNC cells H. Wong et al. / FEMS Microbiology Letters 233 (2004) 269–275 273 of strains ST550, 1264, 1092 and 1099 could be re- experiment. The level of nalidixic acid added (5–200 covered in approximately two weeks. Several strains lgmlÀ1) in this study was above the MIC. However, like 1265, 1292, 1137, and others could be recovered in addition of nalidixic acid at these concentrations did less than a week (Table 1). The categories of strains, not affect the number of resuscitated cells. It may be including the clinical, environmental, or serotypical, due to the existing of extremely low number of did not appear to differ in the induction and resusci- growing cells in the VBNC culture, or the multipli- tation of VBNC state. Even in most of the O3:K6 cation of bacterial cells already held by the low sa- strains with very similar genotypes [12], the induction linity medium. Therefore, these results together and resuscitation of the VBNC state also depended on support the hypothesis that resuscitation rather than strain (Table 1). regrowth of VBNC cells of V. parahaemolyticus occurs during the temperature upshift in the MMS-0.5% NaCl medium. 4. Discussion In V. vulnificus, the resuscitation is almost completely inhibited by a adding protein synthesis inhibitor, The VBNC cell is a metabolically active one although chloramphenicol, during the first 24 h of temperature cells are incapable of undergoing the sustained cellular upshift, and is slightly affected when the inhibitor is division required to form a colony on regular agar me- added later [8]. Adding chloramphenicol at the begin- dia. This state has been demonstrated in numerous ning of or after 24 h of temperature upshift completely bacteria [13], although the existence of the VBNC state suppressed the resuscitation of nonculturable V. para- remains somewhat controversial. Application of cata- haemolyticus, but only slightly reduced the culturable lase to the plate count medium did not significantly in- cell density, when this inhibitor was added after 24 h crease the culturable cells of V. parahaemolyticus (Fig. 4). The temperature upshift may trigger the ex- (Fig. 1). Additionally, results of the MPN and other pression of certain genes during resuscitation [16]. The resuscitation experiments suggested that the VBNC state expression of particular genes during the first 24 h of was induced in this species in this study and this state temperature upshift may be critical and generate enough was different from the peroxide sensitive nonculturable components for the recovery of the nonculturable cells cells which could be recovered by adding catalase or on the second or third day of the temperature upshift pyruvate to the medium as described by Bogosian et al. process [17]. [7] for V. vulnificus. During the early stage of resuscitation, proteins or The VBNC state is typically resuscitated by temper- enzymes for building peptidoglycan or other cellular ature upshift treatment [8,14]. However, strong argu- components may be synthesized [8]. Other protective ments concerning the recovery of VBNC cells as the agents, including catalase and superoxide dismutase, results of real resuscitation or just the regrowth of a few may also be synthesized and protect the recovering culturable cells apply during the temperature upshift cells from the adverse effect of increased metabolic treatment. Ravel et al. [15] reported that the recovery of activities [18,19]. According to the authorsÕ unpub- VBNC cells of V. cholerae after a temperature upshift to lished study, a significant quantity of superoxide dis- 30 °C follows the regrowth of surviving cells, but not the mutase is present in the exponential phase cells, resuscitation of all VBNC cells. Jiang and Chai [5] but the quantity becomes undetectable in the VBNC showed that the recovery of VBNC V. parahaemolyticus V. parahaemolyticus. In the present work, the tem- cells in a MMS-2.6% NaCl medium involves the multi- perature upshift to 37 °C failed to recover any VBNC plication of cells during the temperature upshift period. V. parahaemolyticus cells, although this temperature However, in this study, we used MMS-0.5% NaCl me- was optimal for bacterial growth as well as inducing dium and this medium alone did not allow multiplica- VBNC cells (Fig. 3). Metabolic activities may have tion of growing cells of this halophilic bacterium been initiated when the VBNC cells are temperature (Fig. 1). The MPN result also demonstrated that the upshifted to either 25 or 37 °C. However, more per- number of resuscitated cells observed was not likely to oxide or other harmful free radicals may be generated be due to the presence of growing cells in the culture as at 37 °C than at 25 °C. The presence of these harmful these would be below the level of detection. In the di- radicals may be fatal or cause sublethal injury to the lution experiments, the density of culturable cells should sensitive resuscitating cells and inhibits the recovery of increase if multiplication occurs [15], however, in our colony-formation ability at 37 °C. study, the cells density only restored to those of the Resuscitation in a rich medium may be harmful. original or diluted samples and remained constant Incubating VBNC V. vulnificus in a rich heart infusion (Fig. 2). medium at room temperature has been proven not to Nalidixic acid inhibits proliferation of bacterial recover the culturability for 24 h [14]. In this study, cells, so addition of this antibiotic could rule out the the nutrients in TSB-3% NaCl medium were also effect of multiplication of cells in the resuscitation responsible for the 12 h delay in the recovery of 274 H. Wong et al. / FEMS Microbiology Letters 233 (2004) 269–275 the culturability of the VBNC V. parahaemolyticus research under Contract Nos. NSC 89-2313-B-031-001 (Fig. 3). The viability and the culturability increased and NSC90-2313-B-031-004. to around 1010 cells mlÀ1 after the temperature upshift (Fig. 3), a thousand times more than those of the initial population. Therefore, a few of the culturable References cells were probably multiplied and the final population may have been a mixture of regrown and resuscitated [1] Pan, T.-M., Wang, T.-K., Lee, C.-L., Chien, S.-W. and Horng, cells, since VBNC cells, indistinguishable from those C.-B. (1997) Food-borne disease outbreaks due to bacteria in Taiwan, 1986 to 1995. J. Clin. Microbiol. 35, 1260– regrown cells, may have been resuscitated after a de- 1262. lay. When the old VBNC cells were suspended in this [2] Takeda, Y. (1988) Thermostable direct hemolysin of Vibrio rich medium and shifted to 25 °C for resuscitation, parahaemolyticus. Meth. Enzymol. 165, 189–193. the microscopic viable count revealed that these cells [3] Kaneko, T. and Colwell, R.R. (1974) Distribution of Vibrio had multiplied. However, these viable cells failed to parahaemolyticus and related organisms in the Atlantic Ocean off South Carolina and Georgia. Appl. Microbiol. 28, 1009– form colonies on agar medium (data not shown). The 1017. reason for the nutrientsÕ inhibition on the recovery of [4] Oliver, J.D., Hite, F., McDougald, D., Andon, N.L. and culturability of VBNC cells is unclear, but may be Simpson, L.M. (1995) Entry into, and resuscitation from, the attributable to the presence of harmful radicals in the viable but nonculturable state by Vibrio vulnificus in an medium. During the preparation of rich medium, a estuarine environment. Appl. Environ. Microbiol. 61, 2624– 2630. certain amount of peroxide or harmful free radicals [5] Jiang, X.P. and Chai, T.J. (1996) Survival of Vibrio parahaemo- are generated and suppress the growth of Campylo- lyticus at low temperatures under starvation conditions and bacter [20] and other sensitive resuscitating VBNC subsequent resuscitation of viable, nonculturable cells. Appl. cells. They also constrain the recovery of injured cells Environ. Microbiol. 62, 1300–1305. [6]. [6] Mizunoe, Y., Wai, S.N., Ishikawa, T., Takade, A. and Yoshida, S. (2000) Resuscitation of viable but nonculturable cells of Vibrio Under the optimal conditions reported in here, the parahaemolyticus induced at low temperature under starvation. VBNC state of 24 V. parahaemolyticus strains was FEMS Microbiol. Lett. 186, 115–120. induced over a period of 14–49 days, depending on [7] Bogosian, G., Aardema, N.D., Bourneuf, E.V., Morris, P.J. and the strain. VBNC induction for the clinical and envi- OÕNeil, J.P. (2000) Recovery of hydrogen peroxide-sensitive ronmental strains was unrelated to the source or vir- culturable cells of Vibrio vulnificus gives the appearance of resuscitation from a viable but nonculturable state. J. Bacteriol. ulence of these strains (Table 1). The O3:K6 strains 182, 5070–5075. considered are the first pandemic strains of this [8] Novitsky, J.A. and Morita, R.Y. (1976) Morphological charac- pathogen that spread rapidly throughout many Asian terization of small cells resulting from nutrient starvation of a countries since 1996. These O3:K6 strains have a very psychrophilic marine vibrio. Appl. Environ. Microbiol. 32, 617– close clonality, as determined by pulsed-field gel elec- 622. [9] Boulos, L., Prevost, M., Barbeau, B., Coallier, J. and trophoresis or other molecular methods [12,21]. Desjardins, R. (1999) LIVE/DEADâ BacLightTM: application However, the induction time of the VBNC state was of a new rapid staining method for direct enumeration of also strain-dependent (Table 1). These strains may viable and total bacteria in drinking water. J. Microbiol. Meth. respond differently to the low and high temperature 37, 77–86. cycles of the natural marine environment [4]. Jiang [10] Decamp, O., Rajendran, N., Nakano, H. and Nair, G.B. (1997) Estimation of the viability of Vibrio cholerae 0139 by assessing cell and Chai [5] reported that a non-pathological membrane integrity. Microbios 92, 83–89. Kanagawa negative strain lost culturability more [11] Nilsson, L., Oliver, J.D. and Kjelleberg, S. (1991) Resuscitation of slowly than a Kanagawa positive strain. Their results Vibrio vulnificus from the viable but nonculturable state. J. concern only the characteristics of specific strains but Bacteriol. 173, 5054–5059. not the general properties of different categories of V. [12] Wong, H.C., Liu, S.H., Wang, T.K., Lee, C.L., Chiou, C.S., Liu, D.P., Nishibuchi, M. and Lee, B.K. (2000) Characteristics of parahaemolyticus strains. Vibrio parahaemolyticus O3:K6 from Asia. Appl. Environ. In conclusion, recovery of cells by temperature up- Microbiol. 66, 3981–3986. shift in the MMS-0.5% NaCl medium which alone lim- [13] Colwell, R.R. (2000) Viable but nonculturable bacteria: a survival ited cell multiplication allowed only resuscitation of strategy. J. Infect. Chemother. 6, 121–125. VBNC cells. The induction period and the age of the [14] Whitesides, M.D. and Oliver, J.D. (1997) Resuscitation of Vibrio vulnificus from the viable but nonculturable state. Appl. Environ. VBNC cells for successful resuscitation were strain- Microbiol. 63, 1002–1005. dependent. [15] Ravel, J., Knight, I.T., Monahan, C.E., Hill, R.T. and Colwell, R.R. (1995) Temperature-induced recovery of Vibrio cholerae from the viable but nonculturable state: growth or resuscitation? Acknowledgements Microbiology 141, 377–383. [16] Rice, S.A., McDougald, D. and Kjelleberg, S. (2000) Vibrio vulnificus: a physiological and genetic approach to the viable but The authors thank the National Science Council of nonculturable response. J. Infect. Chemother. 6, 115– the Republic of China for financially supporting this 120. H. Wong et al. / FEMS Microbiology Letters 233 (2004) 269–275 275

[17] Albertson, N.H., Nystroem, T. and Kjelleberg, S. (1990) Macro- [20] Field, L.H., Headley, V.L., Payne, S.M. and Berry, L.J. (1986) molecular synthesis during recovery of the marine Vibrio sp. S14 Inﬂuence of iron on growth, morphology, outer membrane from starvation. J. Gen. Microbiol. 136, 2201–2207. protein composition, and synthesis of siderophores in Campylo- [18] Daily, O.P., Debell, R.M. and Joseph, S.W. (1978) Superoxide bacter jejuni. Infect. Immun. 54, 126–132. dismutase and catalase levels in halophilic vibrios. J. Bacteriol. [21] Okuda, J., Ishibashi, M., Hayakawa, E., Nishino, T., Takeda, Y., 134, 375–380. Mukhopadhyay, A.K., Garg, S., Bhattacharya, S.K., Nair, G.B. [19] Visick, K.L. and Ruby, E.G. (1998) The periplasmic, group III and Nishibuchi, M. (1997) Emergence of a unique O3:K6 clone of catalase of Vibrio ﬁscheri is required for normal symbiotic Vibrio parahaemolyticus in Calcutta, India, and isolation of strains competence and is induced both by oxidative stress and by from the same clonal group from Southeast Asian travelers approach to stationary phase. J. Bacteriol. 180, 2087–2092. arriving in Japan. J. Clin. Microbiol. 35, 3150–3155.