- Email: [email protected]
Viability of the Nonculturable Vibrio cholerae O1 and O139
System. Appl. Microbiol. 24, 331–341 (2001) © Urban & Fischer Verlag http://www.urbanfischer.de/journals/sam
Viability of the Nonculturable Vibrio cholerae O1 and O139 SITTHIPAN CHAIYANAN1, SAIPIN CHAIYANAN1, ANWARUL HUQ1,3, TIMOTHY MAUGEL2, and RITA R. COLWELL1,3* 1
Center of Marine Biotechnology, University of Maryland Biotechnology Institute, Baltimore, USA Department of Zoology, University of Maryland, 3 Department of Cell and Molecular Genetics, University of Maryland, 2
Received April 23, 2001
Summary Vibrio cholerae is capable of transforming into a viable but nonculturable (VBNC) state, and, in doing so, undergoes alteration in cell morphology. In the study reported here, Vibrio cholerae O1 and O139 cells were maintained in laboratory microcosms prepared with 1% Instant Ocean and incubated at 4 °C, i.e., conditions which induce the VBNC state. Cells were fixed at different stages during entry into the VBNC state and, when no growth was detectable on solid or in liquid media, the ultrastructure of these cells was examined, using both transmission and scanning electron microscopy. As shown in earlier studies, the cells became smaller in size and changed from rod to ovoid or coccoid morphology, with the central region of the cells becoming compressed and surrounded by denser cytoplasm. Because the coccoid morphology, indicative of the VBNC state is common for Vibrio cholerae in the natural environment, as well as in starved cells (BAKER et al., 1983; HOOD et al., 1986) viability of the coccoid, viable but nonculturable cell was investigated. The percentage of coccoid (VBNC) cells showing metabolic activity and retention of membrane integrity was monitored using direct fluorescence staining (LIVE/DEAD BacLight Bacterial Viability kit), with 75 to 90% of the viable but nonculturable coccoid cells found to be metabolically active by this test. Furthermore, the proportion of actively respiring cells, using the redox dye, 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC), relative to total cells, the latter determined by DAPI staining, ranged from 10 to 50%. VBNC coccoid cells retained the antigenic determinants of Vibrio cholerae O1 and O139, respectively, evidenced by positive reaction with monoclonal fluorescent antibody. Viability was further established by susceptibility of the VBNC cells to chlorine, copper sulfate, zinc sulfate, and formaldehyde. Since retention of cell membrane integrity is a determining characteristic of viable cells, DNA was extracted from VBNC cells in microcosms maintained for two months and for one year. Conservation of cholera toxin and toxin-associated genes, ctxA, toxR, tcpA, and zot in chromosomal DNA of VBNC cells was demonstrated using PCR and employing specific primers. It is concluded that not only do VBNC V. cholerae 01 and 0139 retain viability up to one year, but genes associated with pathogenicity are retained, along with chromosomal integrity. Key words: Viable but nonculturable Vibrio cholerae – Vibrio cholerae morphology – Coccoid Vibrios
Introduction Serotypes O1 and O139 of Vibrio cholerae, the causative agent of epidemic cholera, not only produce cholera toxin, but are native to the aquatic environment (COLWELL et al., 1985; FARUQUE et al., 1998). V. cholerae O1 possesses an outer membrane lipopolysaccharide that defines the O1-specific antigen, whereas V. cholerae O139 lacks a portion of the O1 antigen (MANNING et al., 1994) and, therefore, does not agglutinate with O1 antibody. Furthermore, V. cholerae O139 produces a polysaccharide capsule.
The ability to enter the viable but nonculturable (VBNC) state has been described for several enteric pathogens, notably Vibrio cholerae (XU et al., 1982), Vibrio vulnificus (OLIVER and WANUCHA, 1989), Salmonella enteritidis (ROSZAK et al., 1984), enterotoxigenic Escherichia coli (FLINT, 1987), Helicobacter pylori and Campylobacter jejuni (ROLLINS and COLWELL, 1986). VBNC cells cannot be detected by standard culture methods and the VBNC state has been concluded to be a survival strategy in response to environmental stress, with 0723-2020/01/24/03-331 $ 15.00/0
332
S. CHAIYANAN et al.
VBNC bacteria retaining metabolic activity (NILSSON et al., 1991; ROLLINS and COLWELL, 1986; ROSZAK et al., 1984; and XU et al., 1982) and virulence (COLWELL et al., 1990: COLWELL et al., 1996: MCKAY, 1992; and ROLLINS and COLWELL, 1986). In this state, cells change from a rod to ovoid morphology and, in contrast to starved cells, are unable to grow on standard laboratory media (GRIMES et al., 1986). COLWELL et al. (1985) suggested that V. cholerae O1 possesses the ability to enter a state of, or one approximating, dormancy in response to nutrient deprivation, elevated salinity, and/or reduced temperature. The response to nutrient deprivation by bacteria appears to be one of many examples of a common strategy of bacteria for survival in nutrient-poor environments. NOVITSKY and MORITA (1976, 1977, 1978) demonstrated that cultures of the marine psychrophilic Vibrio sp. ANT300 responded to starvation in either natural or artificial seawater by increasing in cell number (reductive division) and producing cells significantly decreased in volume and altered in morphology from the typical bacillus, i.e., from rod shape to a coccoid morphology. The morphological changes that occur within the cell, when the cells “round up” (BAKER et al., 1983; GUETIN et al., 1979; and KENNEDY et al., 1970) and decrease in volume from 15 fold (GUETIN et al., 1979) to 300 fold (MISHISTINA and KAMENEVA, 1981), have been reported to be typical physiological responses of some bacteria upon exposure to organic nutrient-free conditions. STEVENSON (1978), in reviewing the occurrence of dormancy in bacterial cells, maintained that the small coccoid forms of bacteria, normally observed in natural systems by direct microscopy, represent exogenously dormant forms in delayed development, responding to unfavorable chemical or physical conditions in the environment. Such changes are often described as part of a strategy to minimize cell maintenance requirements. Although nonculturable cells can readily be produced in laboratory microcosms and detected in environmental samples, the methods by which they are determined to be viable are not well established. Unfortunately, assessment of the viability of bacterial cells is very important for public health microbiology and environmental and industrial applications. Starved cells of Vibrio cholerae both in microcosms and in the natural environment are coccoidal, and the coccoid morphology is concluded to be common for Vibrio cholerae in nature. The hypothesis that the coccoid form is typical of VBNC cells as well has not yet been fully proven. Therefore, characteristics related to viability of the nonculturable, coccoid form of V. cholerae serotypes O1 and O139 were investigated. Metabolic activity, retention of membrane integrity, and susceptibility to disinfectants were tests employed to demonstrate viability of the coccoid VBNC cells. In addition, the persistence of genes associated with toxin production and secretion (ctxA, toxR, tcpA, and zot) in coccoid cells of both V. cholerae O1 and O139 in long-term microcosms (up to one year) was examined using PCR. Demonstrating the presence of these genes in the genome of the small, coccoid VBNC cells provides yet another indication of potential pathogenicity.
Materials and Methods Bacterial strains employed and construction of microcosms V. cholerae O1 strain ATCC 14035 and V. cholerae O139 strain NT330 were used in this study. Single colonies of both strains of Vibrio cholerae were picked from LB (Luria - Bertani) agar and transferred to 100 ml of LB broth and incubated for 18 h at 35 °C. Bacterial cells in logarithmic phase of growth were harvested by inoculating 1 liter of LB broth, incubating the inoculated LB broth for 6 h at 35 °C, and centrifuging at 5000 g for 20 min at 4 °C. After centrifugation, the cells were washed three times and resuspended in phosphate buffered saline (PBS 0.01M, pH 7.3). The washed cells were resuspended in 1 liter of PBS and in autoclaved and filter-sterilized (0.22 µm) artificial sea water [Instant OceanTM (IO) Aquarium system, Mentor, OH] at a salinity of 1%. To induce the VBNC state, flasks containing the cell suspensions were maintained at 4 °C for six months or until the bacterial cells in the microcosms did not yield growth when transferred to solid or liquid media. Preparation of coccoid VBNC cells from microcosms Sequential filtration through different pore-size filter membranes (0.8, 0.4, and 0.2 µm) allowed selection of coccoid VBNC cells from populations of mixed cell morphologies. The 0.2-0.4 µm cells were collected and confirmed to be 100% coccoid in morphology by both light and electron microscopy. Formation of colonies on both nonselective (LB) and selective (TCBS) media was tested and shown not to occur. Bacterial enumeration Total numbers of bacterial cells were determined by acridine orange direct counting (AODC). Culturable cells were enumerated by spread plating onto nonselective media, LB agar, and selective media, including TCBS agar. Detection and enumeration of specific serogroups of viable but nonculturable (VBNC) bacterial cells were achieved by direct fluorescent monoclonal antibody assay (New Horizons Diagnostics, Inc., MD) (HASAN et al., 1994). Direct viable counts were performed, employing three different techniques: LIVE/DEAD BacLight Viability Staining (Molecular Probes, OR), 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC) redox dye staining (Polysciences Inc., PA), and direct fluorescent antibody - direct viable count (DFA-DVC). At least three replicates of total and viable bacterial counts were carried out for each method and sampling time. AODC staining The acridine orange direct count was used for total bacterial enumeration (HOBBIE et al., 1977). One ml of cell suspension was stained with 0.1% acridine orange (ICN Biochemicals, OH) in 0.2 M acetate buffer at pH 4.5 (Sigma Diagnostics, MO). After incubation at room temperature in the dark for 5 min, 0.1 ml of 2% (vol/vol) formaldehyde was added to 1 ml of the sample to stop the reaction and preserve the sample. The stained bacterial cells were captured on 47-mm, black polycarbonate membrane filters of 0.2 µm-pore size (Poretics Co., CA) and placed on a Millipore glass filter funnel unit (Millipore). The membrane was removed after filtration and mounted on a glass slide, with low-fluorescence immersion oil employed. A minimum of 10 random fields (≥ 400 cells were counted with a Olympus VANOX-T microscope attachment for reflected and transmitted light fluorescence (Model AH 2, Olympus, Tokyo, Japan). The total bacterial concentration per ml of bacterial suspension was calculated as follows: Number of bacterial cells =
M × CF × DF V
Viability of VBNC Vibrio cholerae Where M is the microscopic field count, representing the average number of bacteria per field. CF is the membrane conversion factor, and DF is the dilution factor. CF is the ratio of drainage area (873.09 mm2) to microscopic field area (0.0154 mm2). Direct Fluorescence Antibody staining (DFA) Five µl of bacterial cell suspension was uniformly smeared on a precleaned glass slide. After air drying, the smear was fixed with absolute ethanol, followed by addition of anti-O1 or antiO139 Mab-FITC conjugate (New Horizons Diagnostics Corp., MD) for 10 min. The slide was washed with PBS, air dried, and mounted with fluorescent mounting medium before observation of the bacterial cells under an Olympus epifluorescent microscope (Tokyo, Japan). Direct Fluorescence Antibody staining – Direct Viable Count (DFA-DVC) The direct viable counting (DVC) procedure of KOGURE et al. (1979) was modified by combining with specific monoclonal antibody staining. Yeast extract (0.1% wt/vol) and nalidixic acid (10 µg/ml) were added to the samples, which were incubated overnight at 30 °C in the dark with shaking. Cells were stained by DFA. Viable cells respond to this treatment by becoming elongated in the presence of the yeast extract and nalidixic acid. The degree of elongation was observed by comparison with nontreated DFA-stained cells. LIVE/DEAD BacLight Viability staining Aliquots (1 ml) of an appropriate 10-fold dilution of a bacterial suspension were stained with a 3 µl mixture (1:1) of SYTO9 and propidium iodide (PI) nucleic acid stain and incubated in the dark for 25 min at room temperature (Molecular Probes, OR). The stained bacteria were added to 20 ml of 0.1 M phosphate buffer (pH 7.2), filtered through a 47 mm diameter, 0.2 µm-pore size black polycarbonate membrane filter (Poretics Co., CA.) using vacuum suction and then washed twice with 20 ml of deionized water to remove unbound stain. The filter was removed and mounted on a glass slide and low-fluorescence immersion oil was added. The LIVE/DEAD BacLight stain mixture distinguishes live bacterial cells from dead by membrane integrity. Ideally, healthy living bacteria with an intact cytoplasmic membrane stain with a green fluorescence, and dead or injured cells with a compromised membrane stain fluorescent red (TERZIEVA et al., 1996). The viable bacterial count was obtained by employing the same equation as was used for the AODC technique, in which the microscopic field count, M, of green fluorescent cells represents the viable cell count. CTC staining 5-cyano-2,3-ditolyl tetrazolium chloride was employed for direct epifluorescent microscopic enumeration of actively respiring bacteria. Cell suspensions were amended with CTC and nutrient, and counterstained with the DNA-binding fluorochrome, 4’,6-diamidino-2-phenylindole (DAPI), in accordance with modification of the procedure described by RODRIGUEZ et al., 1992). One ml cell suspension was amended with 3.0 ml of 1% yeast extract and 1.0 ml of CTC stock solution (0.5%). The mixture was incubated for 2 h in the dark at 35 °C. After incubation, the sample was counterstained with 1.0 µg of DAPI per ml for 5 min. Stained bacteria were filtered through a 0.2 µm pore size black polycarbonate membrane filter and air dried. The nonrespiring bacteria appeared green when stained with DAPI, whereas the respiring cells were green but contained intracellular crystals of red CTC-formazan. The number of actively respiring cells was determined by the same equation as in the BacLight staining method.
333
Scanning electron microscopy (SEM) Vibrio cholerae O1 and O139 in microcosms were examined at different time intervals (0, 1, 2, and 6 months). Cells were harvested by centrifugation and the bacterial cell pellets were washed three times with sterile Millonig buffer (374 mM/kg), pH 7.4 (HAYET, 1989). Bacterial cells were fixed with 2% (v/v) glutaraldehyde in Millonig buffer at room temperature for one hour. The samples were filtered through 13 mm diameter, 0.8 mm pore size nucleopore polycarbonate filters, post-fixed in 1% osmium tetroxide at room temperature for one hour, dehydrated in a graded ethanol series (75%, 95%, and 100%), critical point dried in CO2 in a DCPI-CPD apparatus (Morristown, NJ), and coated with palladium gold (AuPd) in a Denton DV503 vacuum evaporator coating apparatus (Amray Inc, MA). The coated samples were examined in an Amray 1802D scanning electron microscope (Amray Inc, MA). Transmission electron microscopy (TEM) Bacterial cells were harvested from microcosms of both strains at 0, 1, 2, and 6 months. The cells were immediately fixed with 2% glutaraldehyde in cacodylate buffer for one hour at room temperature. Post-fixation was done over 30 min in 1% osmium tetroxide in cacodylate buffer, followed by washing for 10 min with double distilled water three times to remove osmium. Fixation in 2% aqueous uranyl acetate was for one hour. Samples were then dehydrated in a graded series (35%, 50%, 75%, 95%, and three changes of 100% ethanol solution, 10 min for each change). After dehydration, the samples were embedded in Spurr’s resin by a rapid embedding technique. Polymerization was done at 70 °C for at least eight hours. Thin sections were cut and placed on 75/300-mesh Formvar-carbon grids, using a Reichert Ultracut E Ultratome microtome with a diamond knife (Leica, Vienna, Austria). Post-staining consisted of 0.2% lead citrate and 2% aqueous uranylacetate at pH 3.6 and 7.4. Sections were examined under a Zeiss EM10CA transmission electron microscope (LEO Electron Microscopy, NY) (AMY et al., 1983; BAKER et al., 1983). Susceptibility to disinfectants Susceptibility to disinfectants was also used to demonstrate viability of the coccoid VBNC cells. Inactivation of respiration by VBNC V. cholerae cells was determined employing four disinfectants: sodium hypochlorite; copper sulfate; zinc sulfate; and formaldehyde. Stock solutions of the disinfectants were prepared as follows. Sodium hypochlorite solution (5%) was obtained from the J. T. Baker Chemical Co. (Phillipsburg, NJ). Copper and zinc solutions were prepared in 20 mM phosphate buffer (pH 7.0) with cupric sulfate and zinc sulfate (Fisher Scientific Co., Fair Lawn. NJ), respectively. Formaldehyde (2.0% wt/vol) was prepared in filter-sterilized 1% IO. VBNC samples were treated with various concentrations of disinfectant at 4 °C for 24 h. Residual disinfectant was removed by three sequential washes using filtered 1% IO (5 min each). One ml of treated cell sample was incubated for two hours with three ml of 1% yeast extract and one ml of CTC stock solution (0.5). After incubation, the sample was counterstained with 1.0 µg of DAPI per ml for 5 min. Stained bacteria were filtered through a 0.2 µm pore size black polycarbonate nucleopore membrane (Poretics Corp., Livermore, CA) and air dried. Respiring cells contained intracellular crystals of red CTC-formazan, observed by epifluorescence microscopy. Chromosomal DNA extraction Active growing cells (6 h) of V. cholerae O1 and O139 from LB broth were harvested by centrifugation at 5,000g for 10 min and resuspended in 9.5 ml of TE buffer (10 mM Tris/HCl, 1 mM EDTA, pH 8.0) amended with 250 µl 10% SDS solution and 25 µl freshly prepared proteinase K (20 mg/ml) (Sigma),
334
S. CHAIYANAN et al.
Table 1. Primer sequences used for amplification of toxin-associated genes. Gene
Primer
Oligonucleotide sequence (5’___3’)
Fragment size (bp)
Reference
ctxA
94F 614R
CGGGCAGATTCTAGACCTCCTG CGATGATCTTGGAGCATTCCCAC
564
SHIRAI et al., 1991
toxR
101F 837R
CCTTCGATCCCCTAAGCAATAC AGGGTTAGCAACGATGCGTAAG
779
HERRINGTON et al., 1988
zot
225F 1129R
TCGCTTAACGATGGCGCGTTTT AACCCCGTTTCACTTCTACCCA
947
CHOWDJURY et al., 1994
tcpA
72F 647R* 477R#
CACGATAAGAAAACCGGTCAAGAG TTACCAAATGCAACGCCGAATG CGAAAGCACCTTCTTTCACGTTG
575/405
DIRITA, 1991
* specific for V. cholerae O1, # specific for V. cholerae O139
Fig. 1. Transmission electron micrographs of viable cells of Vibrio cholerae O1 in 1% Instant Ocean. A, transverse and cross-sections of actively-growing cells; B, one month (cells oval, with condensed cytoplasm); C, two months (cells without the distinct three-layered integrity of the outer membrane and cell membrane); and D, six months (coccoid, with a large periplasmic space).
Viability of VBNC Vibrio cholerae mixed gently and incubated for one hour at 37 °C. After incubation, 1.2 ml 5M NaCl was added, mixed thoroughly with 750 µl pre-warmed (65 °C) CTAB/NaCl [10% cetyltrimethylammonium bromide-(CTAB) in 0.7 M NaCl and incubated at 65 °C for 20 min. To the nucleic acid in the aqueous phase was added an equal vol. chloroform-isoamyl-alcohol (24:1 v/v) and the mixture was centrifuged at 8000 rpm for 20 min at 22 °C. The aqueous phase was treated again with an equal vol. chloroform. The nucleic acid solution was precipitated with 0.6 vol. 2-propanol, washed with 70% (v/v) ethanol, and dried under vacuum. The nucleic acid pellet was resuspended in 3 ml modified TE buffer (10 mM Tris/HCl. 0.1 mM EDTA, pH 8.0) and treated with RNase A (10 mg/ml) at 37 °C for one hour. The tubes were kept on ice and amended with 60 µl 3 M sodium-acetate (pH 5.0) and the DNA precipitated with 2 vol. cold (–20 °C) 100% ethanol. Spooled DNA was washed with cold (–20 °C) 70% ethanol and dried under vacuum. The DNA pellet was dissolved in 1 to 2 ml modified TE buffer and stored at 4 °C. Chromosomal DNA extraction of VBNC cells VBNC cells from two-month and one- year microcosms were selectively collected for 0.22–0.45 µm cell size by using 0.45 µm pore-size filter membranes to exclude cells that were larger than
Fig. 2. Transmission electron micrographs of viable cells of Vibrio cholerae O139 in 1% Instant Ocean. A, actively-growing cells; B, one month (cells oval, with condensed cytoplasm); C, two months (cells with distinctive periplasmic space); and D, six months (coccoid, with enlarged periplasmic space).
335
0.45 µm and filtering again through 0.22 µm pore-size filter membranes. The cells collected on the 0.22 µm membrane were centrifuged and washed with 1% IO at least twice or until there was no UV absorption in the supernatant indicating no nucleic acid outside the cell pellets. Chromosomal DNA extraction was performed as for the active growing cells, described above. PCR Amplification The primers, as shown in Table 1, were synthesized and used to amplify the toxin-associated genes. Approximately 50 ng of DNA was subjected to PCR amplification in a total volume of 50 µl containing primers (each at a concentration of 0.4 mM), a mixture of deoxynucleoside triphosphates (each at a concentration of 200 mM), Taq polymerase, and buffer (Promega, Madison, Wis.). A DNA thermal cycler (PTC-200; MJ Research) used for thermal amplification was programmed for the following: (1) an initial extensive denaturation step, consisting of treatment at 94 °C for 2 min; (2) 30 reaction cycles, with each cycle consisting of treatment at 94 °C for one min., 60 °C for one min., and 72 °C for one min; and (3) a final extension step, consisting of treatment at 72 °C for 10 min. The PCR products were separated by 1% agarose gel electrophoresis, stained with ethidium bromide, and visualized with UV light.
336
S. CHAIYANAN et al.
Results Morphological changes associated with entry into the viable but nonculturable state. VBNC cells of Vibrio cholerae O1 and O139 exhibited several characteristic morphological changes when observed by TEM (Figs. 1–2). Compared to electron micrographs of actively growing cells, there were no granules or inclusion bodies in the VBNC cells. The distinct three-layered integrity of the outer membrane and cell membrane of the VBNC was also lost, but remnants of the structures remained. The nuclear region (electron clear area) was compressed into the center of the cell, surrounded by a denser cytoplasm. A large periplasmic space was a significant aspect of the morphological change associated with entry into the VBNC state. The cell wall formed an extended or convoluted structure, which pulled away from the cell membrane. Ribosomal structure, however, was conserved, with no apparent morphological change. Based on the
electronmicrographs, it was concluded that the morphological changes occurring in V. cholerae O1 differed from those of O139, in that V. cholerae O1 exhibited significantly less convolution of the cell wall, while V. cholerae O139 showed distinguishable distortion of the cell wall. However, the two serotypes shared certain structural changes, namely total size reduction, intracellular integrity, disappearance of granules, and compacted nuclear regions. Based on SEM micrographs, cells of Vibrio cholerae O1 and O139 in logarithmic growth phase were not significantly different in morphology (Figs. 3A and 4A). However, SEM of both Vibrio cholerae O1 (Fig. 3, B, C, and D) and O139 (Fig. 4, B, C, and D) showed a few morphological differences. Both vibrios gradually changed in cell shape from curved rods to coccoid cells. At about two months, Vibrio cholerae O1 cells became shorter and fatter, whereas Vibrio cholerae O139 cells became shorter and thinner than actively growing cells. At six months, when all cells were VBNC, both Vibrio cholerae O1 and O139 were coccoid in morphology.
Fig. 3. Scanning electron micrographs of Vibrio cholerae O1 entering into the VBNC state. A, actively growing cells; B, one month (short rods); C, two months (oval, with decreased volume); and D, six months (coccoid cells with a greater than 50% decrease in volume).
Viability of VBNC Vibrio cholerae
Viability and antigenicity of coccoid VBNC cells from microcosms The cells of both serotypes in the microcosms decreased in size during exposure to nutrient depletion. In two-month microcosms, more than 90% of the cells were coccoid and their size was ca. 0.4 µm. In one-year microcosms, all cells were coccoid and between 0.2–0.4 µm in size. Filtration of cell suspensions from each microcosm through 0.8 µm-pore-size filter membranes prior to collection of the cells on selected pore-size membranes was done to remove cell debris and clumps of dead cells. VBNC cells were concentrated up to 10-fold by molecular sieve filtration centrifugation. Comparative cell numbers of V. cholerae O1 obtained by acridine orange total count, direct fluorescent monoclonal antibody staining, and different viable staining techniques are shown in Table 2. Viable counts were obtained as relative percent of the total cell count for each cell sample. Results were the same for both serotypes and demonstrated that populations of coccoid VBNC during long-term incubation changed in number of metabolically active cells. Cells that were VBNC for two months showed a large number of cells with intact cell membranes and active respiring systems. However, after one year, the number of actively respiring VBNC cells, i.e., cells responding to the tests employed, decreased. Coccoid VBNC cells responsive to
Fig. 4. Scanning electron micrographs of Vibrio cholerae O139 entering into the VBNC state. A, actively growing cells; B, one month (short rods); C, two months (oval, with decreased volume); and D, six months (coccoid cells with a, greater than 50% decrease in volume).
337
yeast extract and nalidixic acid were clearly demonstrable in two-month microcosms, while only a few percent of responsive cells could be detected in one-year microcosm samples, employing the standard DVC staining procedure. From Table 2, it can be seen that V. cholerae O1 and O139 cells retained their antigenicity, even in longterm microcosms. The number of fluorescent monoclonal antibody-staining cells was not significantly different from that of the acridine orange staining cells. Viability of the VBNC coccoid cells, determined by staining with the redox dye, 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC), is shown in Fig. 5. Susceptibility of coccoid VBNC cells to disinfectants Coccoid VBNC cells in the size range of 0.2–0.4 µm at a concentration of 108 cells/ml, as well as actively growing Vibrio cells, were exposed to various concentrations of sodium hypochlorite, copper sulfate, zinc sulfate, and formaldehyde for 24 h at 4 °C. Enumeration of respiring cells was performed by the CTC assay. Inhibition of respiration of the VBNC cells confirmed the viability of these coccoid, VBNC cells. The percent decrease in respiring cells was calculated, relative to control (untreated) cell samples. Table 3 shows the concentration of disinfectant that caused cessation of respiration of cells of the vibrios, both control and VBNC cells. The effective concentration
338
S. CHAIYANAN et al.
Table 2. Vibrio cholerae O1 and O139 enumerated by acridine orange total count (AO), direct fluorescent monoclonal antibody staining (DFA), BacLight viable staining, and 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC) redox dye staining. Method
Relative percentage of V. cholerae cell count, with reference to control* Actively growing cells
Two-month microcosm VBNC cells
One-year microcosm VBNC cells
O1
O139
O1
O139
O1
O139
Total count (AO)
100
100
100
100
100
100
DFA
98.9 ± 0.20
99.2 ± 0.70
99.7 ± 0.10
98.6 ± 0.66
98.2 ± 0.70
98.8 ± 0.26
DFA-DVC
90.2 ± 3.08
89.1 ± 1.59
80.1 ± 1.99
72.0 ± 2.36
5.5 ± 1.32
8.2 ± 1.00
BacLight
92.0 ± 1.44
90.3 ± 0.72
90.2 ± 0.85
75.1 ± 2.50
50.4 ± 2.43
52.0 ± 4.42
CTC
89.2 ± 1.15
87.4 ± 2.36
52.0 ± 1.80
49.6 ± 2.36
10.4 ± 0.82
12.2 ± 2.43
* Total cell counts were in the range of 108 cells/ml. Data given are mean and standard error of three determinations. Table 3. Efficacy of disinfectants in inhibiting actively growing cells and viable but non-culturable cells of Vibrio cholerae O1 and Vibrio cholerae O139. Cell state
% Decrease in viable counta Chlorine (µmg/l) O1 1
Actively growing Cells
5 10
CuSO4 (µmg/l) O139
1
O1
5 10
1
5
O139 10
100 100 100 100 100 100 100 100 100
VBNC 2 month microcosm
25 64 90
VBNC 1 year microcosm
6 65 70
30 40 80
8 60
ZnSO4 (µmg/l)
68
25 50 75
4 45 48
1
O1
5 10
1
5 10
100 100 100 89 100 100
32 54
Formaldehyde (µg/l) O139 1
5 10
O1
O139
1.11 1.85 3.7 1.11 1.85 3.7
88 100 100 100
100 100
100 100 100
81
4 49 68
20 52 63
88
89 92
92 100 100
5 42 55
3 42 56
4 40 61
87
89 96
95 100 100
a Cells were exposed for 24 h at 4 °C in 1% IO. Concentrations of disinfectants, as given, are final concentrations. Data shown are averages of duplicate samples.
Viability of VBNC Vibrio cholerae
of zinc sulfate was less than 5 mg/l. Percent decrease in the number of respiring cells, in response to disinfectants, at two months was greater than for one-year microcosms. Formaldehyde showed strong inhibition of respiration of both actively growing and VBNC cells. The response of VBNC cells to the concentrations of sodium hypochlorite, copper sulfate, and zinc sulfate used was not a first order reaction. PCR amplification of toxin-associated genes in VBNC coccoid cells Using chromosomal DNA extracted from actively growing cells as a control, the PCR products of ctxA, toxR, tcpA, and zot genes were detected in coccoid VBNC cells of two-month and one-year microcosms for both serotypes. The persistence of these toxin-associated genes in long-term VBNC cells show that these genes remain detectable in the intact VBNC cells for up to one year, or longer.
Discussion Micrographs prepared using both TEM and SEM revealed changes in cell structure, both internally and externally (Figs. 1–4). However, since cells maintaining an intact cell structure showed no significant morphological differences from the controls, the VBNC cells were considered to be viable cells. During entry into the VBNC state, Vibrio cholerae O1 and O139 decrease in size and become coccoid in morphology, with loss of outer cell wall rigidity. The cytoplasm of the cells condenses, resulting in a significant space between the cytoplasmic membrane and the cell wall. Condensation of the cytoplasm may be related to a dehydration that causes enzymes remaining in the cytoplasm to become inactive, as occurs during formation of endospores (FREESE and HEINZE, 1984). Intracellular integrity decreased in both V. cholerae O1 and O139, accompanied by disappearance of granules and compacting of the nuclear regions. The cell wall of Vibrio cholerae O1 was thicker than that of Vibrio cholerae O139, which was not only thinner but relatively distorted when viewed by transmission electron microscopy. Vibrio cholerae, being an aquatic Gramnegative bacterium, undergoes macromolecular changes when exposed to conditions adverse to growth and multiplication. These changes include construction of a protective layer that is not unlike an exosporium, spore coat, or a cortex, as in spore formation, but with much less definition. In VBNC Vibrio cells, ribosomal structure is conserved, although the number of ribosomes is much reduced. Preservation of ribosomes also occurs in endospores, where it has been suggested that, because protein synthesis is energy-consuming, morphological
339
changes comprise part of the strategy for survival (FREESE and HEINZE, 1984). In the experiments reported here, nonculturable cells of Vibrio cholerae O1 and O139 were further examined to elucidate the association between coccoid morphology and the viable but nonculturable (VBNC) state. Viable counts measured by BacLight staining were significantly higher than DVC and CTC counts (direct viable counts of cells responding to addition of yeast extract), suggesting that nonsubstrate-responsive cells may also be viable, despite lack of response to nalidixic acid and nutrient addition. DVC (elongation in the presence of DNA gyrase inhibitors) and CTC (cellular accumulation of respiring product, insoluble fluorescent crystal of CTC) assays, using 1% yeast extract to prevent nutrient shock of oligotrophic adapted cells, were also done. A longer incubation time for the VBNC cells (24 h) than for actively growing cells (2 h) was employed to allow for substrate uptake and respiration. Prolonged incubation in microcosms was found to affect the viability response of the VBNC cells. However, a minimum of 10% of the VBNC cells maintained in microcosms for one year remained viable i.e., responsive to the test methods used in this study. Environmental factors play a role in prolonging viability of VBNC cells, as well as conditioning cells before they enter the VBNC state. VBNC cells of V. cholerae were very small in size, yet retained antigenicity (Fig. 5). Studies were also done to determine whether respiration of the VBNC V. cholerae could be inhibited as were controls, i.e., actively growing cells. Four disinfectants were used (sodium hypochlorite, copper sulfate, zinc sulfate, and formaldehyde). Viability of VBNC cells was interpreted from the degree of susceptibility to disinfectant. Interestingly, VBNC cells of both serotypes showed greater resistance to all treatments, compared to exponentially growing cultures, a finding that is in agreement with results of studies of bacteria subjected to nutrient starvation, i.e., starved cells (not VBNC) become more resistant to subsequent stresses, such as pressure (BERLIN et al., 1999), heat (GIARD et al., 1996; HARTE et al., 1994; HARTKE et al., 1998), ethanol (HARTE et al., 1994); acid (GIARD et al., 1996; HARTKE et al., 1998); UV light (GIARD et al., 1996); and sodium hypochlorite (LAPLACE et al., 1997). We conclude that the coccoid VBNC cells are more resistant to disinfectants than actively growing cells, with these cells maintaining respiration, as well as cellular integrity, at concentrations of disinfectant inhibitory to actively growing cells. Furthermore, inhibition of respiration of the VBNC cells by disinfectants at higher concentrations, provides evidence of continuing metabolic function in VBNC cells. Since retention of cell membrane integrity is a significant characteristic of viable cells, chromosomal DNA was extracted from the intact VBNC cells maintained for two months and one year in microcosms and conservation of
Fig. 5. Vibrio cholerae O1 from 1% Instant Ocean microcosm. A, Direct fluorescent monoclonal antibody staining and B, 5-cyano2, 3-ditolyl tetrazolium chloride redox dye staining.
340
S. CHAIYANAN et al.
the cholera toxin and toxin-associated genes in VBNC cells in long-term microcosms was demonstrated, using PCR primers for the ctxA, toxR, tcpA, and zot genes. Earlier studies showing that VBNC cells can cause accumulation of fluid in rabbit ligated ileal loops and also can cause diarrhoea in human volunteers (COLWELL et al., 1985) are corroborated at the molecular level by persistence of the genes relevant to pathogenicity in VBNC cells of V. cholerae 01 and 0139. From all of the evidence gathered in this study, it is concluded that VBNC cells retain viability as well as the potential for pathogenicity for significant periods of time, i.e., one year or longer. Acknowledgments The authors gratefully acknowledge support of NIH Grant No. 1RO1A139129-01 and NIH-NINR Grant No. RO1 NR04527-01A1.
References AMY, P. S., PAWLING, C., MORITA, R. Y.: Recovery from nutrient starvation by a marine Vibrio spp. Appl. Environ. Microbiol. 45, 1685–1690 (1983). BAKER, R. M., SINGLETON, E L., HOOD, M. A.: Effects of nutrient deprivation on Vibrio cholerae. Appl. Environ. Microbiol. 46, 930–940 (1983). BERLIN, D. L., HERSON, D. S., HICKS, D. T., HOOVER, D. G.: Response of pathogenic Vibrio species to high hydrostatic pressure. App. Environ. Microbiol. 65, 2776–2780 (1999). CHOWDHURY, M. A. R., HILL, R. T., COLWELL, R. R.: A gene for the enterotoxin zonula occludens toxin is present in Vibrio mimicus and Vibrio cholerae O139. FEMS Microbiology Letters 119, 377–380 (1994). COLWELL, R. R., BRAYTON, P. R., GRIMES, D. J., ROSZAK, D. B., HUG, S. A., PALMER, L. M.: Viable but non-culturable Vibrio cholerae and related pathogens in the environment: implications for release of genetically engineered microorganisms. Bio/Technology 3, 817–820 (1985). COLWELL, R. R., TAMPLIN, M. L., BRAYTON, P. R., GAUZENS, A. L., TALL, B. D., HERRINGTON, D., LEVINE, M. M., HALL, S., HUQ, A., SACK, D. A.: Environmental aspects of Vibrio cholerae in transmission of cholera, pp. 327–343. In: Advances on cholera and related diarrheas, Vol. 7 (R. B. SACK, Y. ZINNAKA, eds.) KTK Scientific, Tokyo, Japan 1990. COLWELL, R. R., BREYTON, P., HERRINGTON, D., TALL, B., HUQ, A., LEVINE, M. M.: Viable but non-culturable Vibrio cholerae O1 revert to a cultivable state in the human intestine. World J. Microbiol. Biotechnol. 12, 28–31 (1996). DIRITA, V. J., PARSOT, C., JANDER, G., MEKALANOS, J. J.: Regulatory cascade controls virulence in Vibrio cholerae. Proc. Natl. Acad. Sci. 88, 5403–5407 (1991). FARUQUE, S. M., ALBERT, M. J., MEKALANOS, J. J.: Epidemiology, Genetics, and Ecology of Toxigenic Vibrio cholerae. Micro. Mol. Bio. Rev. 62, 1301–1314 (1998). FLINT, K. P.: The long-term survival of Escherichia coli in river water. J. Appl. Bacteriol. 63, 261–270 (1987). FREESE, E., HEINZE, J.: Metabolic and genetic control of bacterial sporulation, pp. 101–172. In: The bacterial spore (A. HURST, G. W. GOULD, eds.), Academic Press, Inc., New York 1984. GIARD, J. C., HARTKE, A., FLAHAUT, S., BENACHOUR, A., BOUTIBONNES, P., AUFFRAY, Y.: Starvation-induced multiresistance in Enterococcus faecalis JH2-2. Curr. Microbiol. 32, 264–271 (1996).
GRIMES, D. J., ATWELL, R. W., BRAYTON, P. R., PALMER, L. M., ROLLINS, D. M., ROSZAK, D. B., SINGLETON, F. L., TAMPLIN, M. L., COLWELL, R. R.: The fate of enteric pathogenic bacteria in estuarine and marine environments. Microbiol. Sci. 3, 324–329 (1986). GUETIN, A. M., MISHISTINA, I. E., ANDREEV, L. V., BOBYK, M. A., LAMBINA, V. A.: Some problems of the ecology and taxonomy of marine microvibrios. Biol. Bull. Acad. Sci. USSR. 5, 336–340 (1979). HARTKE, A., BOUCHE, S., GANSEL, X., BOUTIBONNES, P., AUFFRAY, Y.: Starvation-induced stress resistance in Lactococcus lactis subsp. lactis IL 1403. Appl. Environ. Microbiol. 60, 3474–3478 (1994). HARTKE, A. J. GIARD, LAPLACE, J., AUFFRAY, Y.: Survival of Enterococcus faecalis in an oligotrophic microcosm: changes in morphology, development of general stress resistance, and analysis of protein synthesis. Appl. Environ. Microbiol. 64, 4238–4245 (1998). HASAN, J. A. K., A. HUQ, M. L. TAMPLIN, R. SIEBELING, R. R. COLWELL. A Novel Kit for Rapid Detection of V. cholerae 01. J. Clin. Microbiol. 32, 249–252 (1994). HAYET, M. A. (ed.): Principle and techniques of electron microscopy: biological applications. Macmillan Press, London 1989. HERRINGTON, D. A., HALL, R. H., LOSONSKY, G., MEKALANOS, J. J., TAYLOR, R. K., LEVINE, M. M. Toxin, toxin-coregulated pili, and the toxR regulon are essential for Vibrio cholerae pathogenesis in humans. J. Exp. Med. 168, 1487–1492 (1988). HOBBIE, J. E., DALEY, R. J., JASPER, S.: Use of nucleopore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33, 1225–1228 (1977). HOOD, M. A., GUCKERT, J. B., WHITE, D. C., DECK, F.: Effect of nutrient deprivation on lipid, carbohydrate, DNA, RNA, and protein levels in Vibrio cholerae. Appl. Environ. Microbiol. 52, 788–793 (1986). KENNEDY, S. F., COLWELL, R. R., CHAPMAN, G. B.: Ultrastructure of a psychrophilic marine vibrio. Can. J. Microbiol. 16, 1027–1032 (1970). KOGURE, K., SIMIDU, U., TAGA, N.: A tentative direct microscopic method for counting living marine bacteria. Can. J. Microbiol. 25, 415–420 (1979). LAPLACE, J. M., THUAULT, M., HARTKE, A., BOUTIBONNES, P., AUFFRAY, Y.: Sodium hypochlorite stress in Enterococcus faecalis: influence of antecedent growth conditions and induced proteins. Curr. Microbiol. 34, 284–289 (1997). MANNING, P. A., STROEHER, U. H., MORONA, R.: Molecular basis for O-antigen biosynthesis in Vibrio cholerae Ol: Ogawa-Inaba switching, pp. 77–94. In: Vibrio cholerae and cholera: molecular to global perspectives (I. K. WACHSMUTH, P. BLAKE, O. OLSVIK, eds.), ASM Press, Washington, DC 1994. MCKAY, A. M.: Viable but non-culturable forms of potentially pathogenic bacteria in water. Lett. Appl. Microbiol. 14, 129–135 (1992). MISHISTINA, I. E., KAMENEVA, T. G.: Bacterial cells of minimal size in the Barents Sea during the polar night. Microbiology. 50, 360–363 (1981). NILSSON, L., OLIVER, J. D., KJELLEBERG S.: Resuscitation of Vibrio vulnificus from the viable but nonculturable state. J. Bact. 173, 5054–5059 (1991). NOVITSKY J. S., MORITA, R. Y.: Morphological characterization of small cells resulting from nutrient starvation of a psychrophilic marine vibrio. Appl. Environ. Microbiol. 32, 617–622 (1976).
Viability of VBNC Vibrio cholerae NOVITSKY J. S., MORITA, R. Y.: Survival of a psychrophilic marine vibrio under long-term nutrient starvation. Appl. Environ. Microbiol. 33, 635–641 (1977). NOVITSKY J. S., MORITA, R. Y.: Possible strategy for the survival of marine bacteria under starvation conditions. Mar. Biol. 48, 289–295 (1978). OLIVER, J. D., WANUCHA, D.: Survival of Vibrio vulnificus at reduced temperatures and elevated nutrient. J. Food. Saf. 10, 79–86 (1989). RODRIGUEZ, G. G., PHIPPS, D., ISHIGURO, K., RIDGWAY, H. F.: Use of a fluorescent redox probe for direct visualization of actively respiring bacteria. Appl. Environ. Microbiol. 58, 1801–1808 (1992). ROLLINS, D. M., COLWELL, R. R.: Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Appl. Environ. Microbiol. 52, 531–538 (1986). ROSZAK, D. B., GRIMES, D. J., COLWELL, R. R.: Viable but nonrecoverable stage of Salmonella enteritidis in aquatic systems. Can. J. Microbiol. 30, 334–338 (1984). SHIRAI, H., NISHIBUCHI, M., RAMAMURTHY, T., BHATTACHARYA, S. K., PALL, S. C., TAKEDA, Y.: Polymerase chain reaction for de-
341
tection of the cholera enterotoxin operon of Vibrio cholerae. J. Clin. Microbiol. 29, 2517–2521 (1991). STEVENSON, L. H.: A case for bacterial dormancy in aquatic systems. Microb. Ecol. 4, 127133 (1978). TERZIEVA S., DONNELLY, J., ULEVICIUS, V., GRINSHPUN, S. A., WILLEKE, K., STELMA, G. N., BRENNER, K. P.: Comparison of methods for detection and enumeration of airborne microorganisms collected by liquid impingement. Appl. Environ. Microbiol. 62, 2264–2272 (1996). XU, H. S., ROBERT, N., SINGLETON, F. L., AATTWELL, R. W., GRIMES, D. J., COLWELL, R. R.: Survival and viability of nonculturable Escherichia coli and Vibrio cholerae in the estuarine and marine environment. Microb. Ecol. 8, 313–323 (1982). Corresponding author: RITA C. COLWELL, Center of Marine Biotechnology, University of Maryland Biotechnology Institute, 701 E. Pratt Street, Baltimore, MD 21202, USA. Tel.: ++1-703-2928000; Fax: ++1-703-2929232; e-mail: [email protected]
Viability of the Nonculturable Vibrio cholerae O1 and O139 SITTHIPAN CHAIYANAN1, SAIPIN CHAIYANAN1, ANWARUL HUQ1,3, TIMOTHY MAUGEL2, and RITA R. COLWELL1,3* 1
Center of Marine Biotechnology, University of Maryland Biotechnology Institute, Baltimore, USA Department of Zoology, University of Maryland, 3 Department of Cell and Molecular Genetics, University of Maryland, 2
Received April 23, 2001
Summary Vibrio cholerae is capable of transforming into a viable but nonculturable (VBNC) state, and, in doing so, undergoes alteration in cell morphology. In the study reported here, Vibrio cholerae O1 and O139 cells were maintained in laboratory microcosms prepared with 1% Instant Ocean and incubated at 4 °C, i.e., conditions which induce the VBNC state. Cells were fixed at different stages during entry into the VBNC state and, when no growth was detectable on solid or in liquid media, the ultrastructure of these cells was examined, using both transmission and scanning electron microscopy. As shown in earlier studies, the cells became smaller in size and changed from rod to ovoid or coccoid morphology, with the central region of the cells becoming compressed and surrounded by denser cytoplasm. Because the coccoid morphology, indicative of the VBNC state is common for Vibrio cholerae in the natural environment, as well as in starved cells (BAKER et al., 1983; HOOD et al., 1986) viability of the coccoid, viable but nonculturable cell was investigated. The percentage of coccoid (VBNC) cells showing metabolic activity and retention of membrane integrity was monitored using direct fluorescence staining (LIVE/DEAD BacLight Bacterial Viability kit), with 75 to 90% of the viable but nonculturable coccoid cells found to be metabolically active by this test. Furthermore, the proportion of actively respiring cells, using the redox dye, 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC), relative to total cells, the latter determined by DAPI staining, ranged from 10 to 50%. VBNC coccoid cells retained the antigenic determinants of Vibrio cholerae O1 and O139, respectively, evidenced by positive reaction with monoclonal fluorescent antibody. Viability was further established by susceptibility of the VBNC cells to chlorine, copper sulfate, zinc sulfate, and formaldehyde. Since retention of cell membrane integrity is a determining characteristic of viable cells, DNA was extracted from VBNC cells in microcosms maintained for two months and for one year. Conservation of cholera toxin and toxin-associated genes, ctxA, toxR, tcpA, and zot in chromosomal DNA of VBNC cells was demonstrated using PCR and employing specific primers. It is concluded that not only do VBNC V. cholerae 01 and 0139 retain viability up to one year, but genes associated with pathogenicity are retained, along with chromosomal integrity. Key words: Viable but nonculturable Vibrio cholerae – Vibrio cholerae morphology – Coccoid Vibrios
Introduction Serotypes O1 and O139 of Vibrio cholerae, the causative agent of epidemic cholera, not only produce cholera toxin, but are native to the aquatic environment (COLWELL et al., 1985; FARUQUE et al., 1998). V. cholerae O1 possesses an outer membrane lipopolysaccharide that defines the O1-specific antigen, whereas V. cholerae O139 lacks a portion of the O1 antigen (MANNING et al., 1994) and, therefore, does not agglutinate with O1 antibody. Furthermore, V. cholerae O139 produces a polysaccharide capsule.
The ability to enter the viable but nonculturable (VBNC) state has been described for several enteric pathogens, notably Vibrio cholerae (XU et al., 1982), Vibrio vulnificus (OLIVER and WANUCHA, 1989), Salmonella enteritidis (ROSZAK et al., 1984), enterotoxigenic Escherichia coli (FLINT, 1987), Helicobacter pylori and Campylobacter jejuni (ROLLINS and COLWELL, 1986). VBNC cells cannot be detected by standard culture methods and the VBNC state has been concluded to be a survival strategy in response to environmental stress, with 0723-2020/01/24/03-331 $ 15.00/0
332
S. CHAIYANAN et al.
VBNC bacteria retaining metabolic activity (NILSSON et al., 1991; ROLLINS and COLWELL, 1986; ROSZAK et al., 1984; and XU et al., 1982) and virulence (COLWELL et al., 1990: COLWELL et al., 1996: MCKAY, 1992; and ROLLINS and COLWELL, 1986). In this state, cells change from a rod to ovoid morphology and, in contrast to starved cells, are unable to grow on standard laboratory media (GRIMES et al., 1986). COLWELL et al. (1985) suggested that V. cholerae O1 possesses the ability to enter a state of, or one approximating, dormancy in response to nutrient deprivation, elevated salinity, and/or reduced temperature. The response to nutrient deprivation by bacteria appears to be one of many examples of a common strategy of bacteria for survival in nutrient-poor environments. NOVITSKY and MORITA (1976, 1977, 1978) demonstrated that cultures of the marine psychrophilic Vibrio sp. ANT300 responded to starvation in either natural or artificial seawater by increasing in cell number (reductive division) and producing cells significantly decreased in volume and altered in morphology from the typical bacillus, i.e., from rod shape to a coccoid morphology. The morphological changes that occur within the cell, when the cells “round up” (BAKER et al., 1983; GUETIN et al., 1979; and KENNEDY et al., 1970) and decrease in volume from 15 fold (GUETIN et al., 1979) to 300 fold (MISHISTINA and KAMENEVA, 1981), have been reported to be typical physiological responses of some bacteria upon exposure to organic nutrient-free conditions. STEVENSON (1978), in reviewing the occurrence of dormancy in bacterial cells, maintained that the small coccoid forms of bacteria, normally observed in natural systems by direct microscopy, represent exogenously dormant forms in delayed development, responding to unfavorable chemical or physical conditions in the environment. Such changes are often described as part of a strategy to minimize cell maintenance requirements. Although nonculturable cells can readily be produced in laboratory microcosms and detected in environmental samples, the methods by which they are determined to be viable are not well established. Unfortunately, assessment of the viability of bacterial cells is very important for public health microbiology and environmental and industrial applications. Starved cells of Vibrio cholerae both in microcosms and in the natural environment are coccoidal, and the coccoid morphology is concluded to be common for Vibrio cholerae in nature. The hypothesis that the coccoid form is typical of VBNC cells as well has not yet been fully proven. Therefore, characteristics related to viability of the nonculturable, coccoid form of V. cholerae serotypes O1 and O139 were investigated. Metabolic activity, retention of membrane integrity, and susceptibility to disinfectants were tests employed to demonstrate viability of the coccoid VBNC cells. In addition, the persistence of genes associated with toxin production and secretion (ctxA, toxR, tcpA, and zot) in coccoid cells of both V. cholerae O1 and O139 in long-term microcosms (up to one year) was examined using PCR. Demonstrating the presence of these genes in the genome of the small, coccoid VBNC cells provides yet another indication of potential pathogenicity.
Materials and Methods Bacterial strains employed and construction of microcosms V. cholerae O1 strain ATCC 14035 and V. cholerae O139 strain NT330 were used in this study. Single colonies of both strains of Vibrio cholerae were picked from LB (Luria - Bertani) agar and transferred to 100 ml of LB broth and incubated for 18 h at 35 °C. Bacterial cells in logarithmic phase of growth were harvested by inoculating 1 liter of LB broth, incubating the inoculated LB broth for 6 h at 35 °C, and centrifuging at 5000 g for 20 min at 4 °C. After centrifugation, the cells were washed three times and resuspended in phosphate buffered saline (PBS 0.01M, pH 7.3). The washed cells were resuspended in 1 liter of PBS and in autoclaved and filter-sterilized (0.22 µm) artificial sea water [Instant OceanTM (IO) Aquarium system, Mentor, OH] at a salinity of 1%. To induce the VBNC state, flasks containing the cell suspensions were maintained at 4 °C for six months or until the bacterial cells in the microcosms did not yield growth when transferred to solid or liquid media. Preparation of coccoid VBNC cells from microcosms Sequential filtration through different pore-size filter membranes (0.8, 0.4, and 0.2 µm) allowed selection of coccoid VBNC cells from populations of mixed cell morphologies. The 0.2-0.4 µm cells were collected and confirmed to be 100% coccoid in morphology by both light and electron microscopy. Formation of colonies on both nonselective (LB) and selective (TCBS) media was tested and shown not to occur. Bacterial enumeration Total numbers of bacterial cells were determined by acridine orange direct counting (AODC). Culturable cells were enumerated by spread plating onto nonselective media, LB agar, and selective media, including TCBS agar. Detection and enumeration of specific serogroups of viable but nonculturable (VBNC) bacterial cells were achieved by direct fluorescent monoclonal antibody assay (New Horizons Diagnostics, Inc., MD) (HASAN et al., 1994). Direct viable counts were performed, employing three different techniques: LIVE/DEAD BacLight Viability Staining (Molecular Probes, OR), 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC) redox dye staining (Polysciences Inc., PA), and direct fluorescent antibody - direct viable count (DFA-DVC). At least three replicates of total and viable bacterial counts were carried out for each method and sampling time. AODC staining The acridine orange direct count was used for total bacterial enumeration (HOBBIE et al., 1977). One ml of cell suspension was stained with 0.1% acridine orange (ICN Biochemicals, OH) in 0.2 M acetate buffer at pH 4.5 (Sigma Diagnostics, MO). After incubation at room temperature in the dark for 5 min, 0.1 ml of 2% (vol/vol) formaldehyde was added to 1 ml of the sample to stop the reaction and preserve the sample. The stained bacterial cells were captured on 47-mm, black polycarbonate membrane filters of 0.2 µm-pore size (Poretics Co., CA) and placed on a Millipore glass filter funnel unit (Millipore). The membrane was removed after filtration and mounted on a glass slide, with low-fluorescence immersion oil employed. A minimum of 10 random fields (≥ 400 cells were counted with a Olympus VANOX-T microscope attachment for reflected and transmitted light fluorescence (Model AH 2, Olympus, Tokyo, Japan). The total bacterial concentration per ml of bacterial suspension was calculated as follows: Number of bacterial cells =
M × CF × DF V
Viability of VBNC Vibrio cholerae Where M is the microscopic field count, representing the average number of bacteria per field. CF is the membrane conversion factor, and DF is the dilution factor. CF is the ratio of drainage area (873.09 mm2) to microscopic field area (0.0154 mm2). Direct Fluorescence Antibody staining (DFA) Five µl of bacterial cell suspension was uniformly smeared on a precleaned glass slide. After air drying, the smear was fixed with absolute ethanol, followed by addition of anti-O1 or antiO139 Mab-FITC conjugate (New Horizons Diagnostics Corp., MD) for 10 min. The slide was washed with PBS, air dried, and mounted with fluorescent mounting medium before observation of the bacterial cells under an Olympus epifluorescent microscope (Tokyo, Japan). Direct Fluorescence Antibody staining – Direct Viable Count (DFA-DVC) The direct viable counting (DVC) procedure of KOGURE et al. (1979) was modified by combining with specific monoclonal antibody staining. Yeast extract (0.1% wt/vol) and nalidixic acid (10 µg/ml) were added to the samples, which were incubated overnight at 30 °C in the dark with shaking. Cells were stained by DFA. Viable cells respond to this treatment by becoming elongated in the presence of the yeast extract and nalidixic acid. The degree of elongation was observed by comparison with nontreated DFA-stained cells. LIVE/DEAD BacLight Viability staining Aliquots (1 ml) of an appropriate 10-fold dilution of a bacterial suspension were stained with a 3 µl mixture (1:1) of SYTO9 and propidium iodide (PI) nucleic acid stain and incubated in the dark for 25 min at room temperature (Molecular Probes, OR). The stained bacteria were added to 20 ml of 0.1 M phosphate buffer (pH 7.2), filtered through a 47 mm diameter, 0.2 µm-pore size black polycarbonate membrane filter (Poretics Co., CA.) using vacuum suction and then washed twice with 20 ml of deionized water to remove unbound stain. The filter was removed and mounted on a glass slide and low-fluorescence immersion oil was added. The LIVE/DEAD BacLight stain mixture distinguishes live bacterial cells from dead by membrane integrity. Ideally, healthy living bacteria with an intact cytoplasmic membrane stain with a green fluorescence, and dead or injured cells with a compromised membrane stain fluorescent red (TERZIEVA et al., 1996). The viable bacterial count was obtained by employing the same equation as was used for the AODC technique, in which the microscopic field count, M, of green fluorescent cells represents the viable cell count. CTC staining 5-cyano-2,3-ditolyl tetrazolium chloride was employed for direct epifluorescent microscopic enumeration of actively respiring bacteria. Cell suspensions were amended with CTC and nutrient, and counterstained with the DNA-binding fluorochrome, 4’,6-diamidino-2-phenylindole (DAPI), in accordance with modification of the procedure described by RODRIGUEZ et al., 1992). One ml cell suspension was amended with 3.0 ml of 1% yeast extract and 1.0 ml of CTC stock solution (0.5%). The mixture was incubated for 2 h in the dark at 35 °C. After incubation, the sample was counterstained with 1.0 µg of DAPI per ml for 5 min. Stained bacteria were filtered through a 0.2 µm pore size black polycarbonate membrane filter and air dried. The nonrespiring bacteria appeared green when stained with DAPI, whereas the respiring cells were green but contained intracellular crystals of red CTC-formazan. The number of actively respiring cells was determined by the same equation as in the BacLight staining method.
333
Scanning electron microscopy (SEM) Vibrio cholerae O1 and O139 in microcosms were examined at different time intervals (0, 1, 2, and 6 months). Cells were harvested by centrifugation and the bacterial cell pellets were washed three times with sterile Millonig buffer (374 mM/kg), pH 7.4 (HAYET, 1989). Bacterial cells were fixed with 2% (v/v) glutaraldehyde in Millonig buffer at room temperature for one hour. The samples were filtered through 13 mm diameter, 0.8 mm pore size nucleopore polycarbonate filters, post-fixed in 1% osmium tetroxide at room temperature for one hour, dehydrated in a graded ethanol series (75%, 95%, and 100%), critical point dried in CO2 in a DCPI-CPD apparatus (Morristown, NJ), and coated with palladium gold (AuPd) in a Denton DV503 vacuum evaporator coating apparatus (Amray Inc, MA). The coated samples were examined in an Amray 1802D scanning electron microscope (Amray Inc, MA). Transmission electron microscopy (TEM) Bacterial cells were harvested from microcosms of both strains at 0, 1, 2, and 6 months. The cells were immediately fixed with 2% glutaraldehyde in cacodylate buffer for one hour at room temperature. Post-fixation was done over 30 min in 1% osmium tetroxide in cacodylate buffer, followed by washing for 10 min with double distilled water three times to remove osmium. Fixation in 2% aqueous uranyl acetate was for one hour. Samples were then dehydrated in a graded series (35%, 50%, 75%, 95%, and three changes of 100% ethanol solution, 10 min for each change). After dehydration, the samples were embedded in Spurr’s resin by a rapid embedding technique. Polymerization was done at 70 °C for at least eight hours. Thin sections were cut and placed on 75/300-mesh Formvar-carbon grids, using a Reichert Ultracut E Ultratome microtome with a diamond knife (Leica, Vienna, Austria). Post-staining consisted of 0.2% lead citrate and 2% aqueous uranylacetate at pH 3.6 and 7.4. Sections were examined under a Zeiss EM10CA transmission electron microscope (LEO Electron Microscopy, NY) (AMY et al., 1983; BAKER et al., 1983). Susceptibility to disinfectants Susceptibility to disinfectants was also used to demonstrate viability of the coccoid VBNC cells. Inactivation of respiration by VBNC V. cholerae cells was determined employing four disinfectants: sodium hypochlorite; copper sulfate; zinc sulfate; and formaldehyde. Stock solutions of the disinfectants were prepared as follows. Sodium hypochlorite solution (5%) was obtained from the J. T. Baker Chemical Co. (Phillipsburg, NJ). Copper and zinc solutions were prepared in 20 mM phosphate buffer (pH 7.0) with cupric sulfate and zinc sulfate (Fisher Scientific Co., Fair Lawn. NJ), respectively. Formaldehyde (2.0% wt/vol) was prepared in filter-sterilized 1% IO. VBNC samples were treated with various concentrations of disinfectant at 4 °C for 24 h. Residual disinfectant was removed by three sequential washes using filtered 1% IO (5 min each). One ml of treated cell sample was incubated for two hours with three ml of 1% yeast extract and one ml of CTC stock solution (0.5). After incubation, the sample was counterstained with 1.0 µg of DAPI per ml for 5 min. Stained bacteria were filtered through a 0.2 µm pore size black polycarbonate nucleopore membrane (Poretics Corp., Livermore, CA) and air dried. Respiring cells contained intracellular crystals of red CTC-formazan, observed by epifluorescence microscopy. Chromosomal DNA extraction Active growing cells (6 h) of V. cholerae O1 and O139 from LB broth were harvested by centrifugation at 5,000g for 10 min and resuspended in 9.5 ml of TE buffer (10 mM Tris/HCl, 1 mM EDTA, pH 8.0) amended with 250 µl 10% SDS solution and 25 µl freshly prepared proteinase K (20 mg/ml) (Sigma),
334
S. CHAIYANAN et al.
Table 1. Primer sequences used for amplification of toxin-associated genes. Gene
Primer
Oligonucleotide sequence (5’___3’)
Fragment size (bp)
Reference
ctxA
94F 614R
CGGGCAGATTCTAGACCTCCTG CGATGATCTTGGAGCATTCCCAC
564
SHIRAI et al., 1991
toxR
101F 837R
CCTTCGATCCCCTAAGCAATAC AGGGTTAGCAACGATGCGTAAG
779
HERRINGTON et al., 1988
zot
225F 1129R
TCGCTTAACGATGGCGCGTTTT AACCCCGTTTCACTTCTACCCA
947
CHOWDJURY et al., 1994
tcpA
72F 647R* 477R#
CACGATAAGAAAACCGGTCAAGAG TTACCAAATGCAACGCCGAATG CGAAAGCACCTTCTTTCACGTTG
575/405
DIRITA, 1991
* specific for V. cholerae O1, # specific for V. cholerae O139
Fig. 1. Transmission electron micrographs of viable cells of Vibrio cholerae O1 in 1% Instant Ocean. A, transverse and cross-sections of actively-growing cells; B, one month (cells oval, with condensed cytoplasm); C, two months (cells without the distinct three-layered integrity of the outer membrane and cell membrane); and D, six months (coccoid, with a large periplasmic space).
Viability of VBNC Vibrio cholerae mixed gently and incubated for one hour at 37 °C. After incubation, 1.2 ml 5M NaCl was added, mixed thoroughly with 750 µl pre-warmed (65 °C) CTAB/NaCl [10% cetyltrimethylammonium bromide-(CTAB) in 0.7 M NaCl and incubated at 65 °C for 20 min. To the nucleic acid in the aqueous phase was added an equal vol. chloroform-isoamyl-alcohol (24:1 v/v) and the mixture was centrifuged at 8000 rpm for 20 min at 22 °C. The aqueous phase was treated again with an equal vol. chloroform. The nucleic acid solution was precipitated with 0.6 vol. 2-propanol, washed with 70% (v/v) ethanol, and dried under vacuum. The nucleic acid pellet was resuspended in 3 ml modified TE buffer (10 mM Tris/HCl. 0.1 mM EDTA, pH 8.0) and treated with RNase A (10 mg/ml) at 37 °C for one hour. The tubes were kept on ice and amended with 60 µl 3 M sodium-acetate (pH 5.0) and the DNA precipitated with 2 vol. cold (–20 °C) 100% ethanol. Spooled DNA was washed with cold (–20 °C) 70% ethanol and dried under vacuum. The DNA pellet was dissolved in 1 to 2 ml modified TE buffer and stored at 4 °C. Chromosomal DNA extraction of VBNC cells VBNC cells from two-month and one- year microcosms were selectively collected for 0.22–0.45 µm cell size by using 0.45 µm pore-size filter membranes to exclude cells that were larger than
Fig. 2. Transmission electron micrographs of viable cells of Vibrio cholerae O139 in 1% Instant Ocean. A, actively-growing cells; B, one month (cells oval, with condensed cytoplasm); C, two months (cells with distinctive periplasmic space); and D, six months (coccoid, with enlarged periplasmic space).
335
0.45 µm and filtering again through 0.22 µm pore-size filter membranes. The cells collected on the 0.22 µm membrane were centrifuged and washed with 1% IO at least twice or until there was no UV absorption in the supernatant indicating no nucleic acid outside the cell pellets. Chromosomal DNA extraction was performed as for the active growing cells, described above. PCR Amplification The primers, as shown in Table 1, were synthesized and used to amplify the toxin-associated genes. Approximately 50 ng of DNA was subjected to PCR amplification in a total volume of 50 µl containing primers (each at a concentration of 0.4 mM), a mixture of deoxynucleoside triphosphates (each at a concentration of 200 mM), Taq polymerase, and buffer (Promega, Madison, Wis.). A DNA thermal cycler (PTC-200; MJ Research) used for thermal amplification was programmed for the following: (1) an initial extensive denaturation step, consisting of treatment at 94 °C for 2 min; (2) 30 reaction cycles, with each cycle consisting of treatment at 94 °C for one min., 60 °C for one min., and 72 °C for one min; and (3) a final extension step, consisting of treatment at 72 °C for 10 min. The PCR products were separated by 1% agarose gel electrophoresis, stained with ethidium bromide, and visualized with UV light.
336
S. CHAIYANAN et al.
Results Morphological changes associated with entry into the viable but nonculturable state. VBNC cells of Vibrio cholerae O1 and O139 exhibited several characteristic morphological changes when observed by TEM (Figs. 1–2). Compared to electron micrographs of actively growing cells, there were no granules or inclusion bodies in the VBNC cells. The distinct three-layered integrity of the outer membrane and cell membrane of the VBNC was also lost, but remnants of the structures remained. The nuclear region (electron clear area) was compressed into the center of the cell, surrounded by a denser cytoplasm. A large periplasmic space was a significant aspect of the morphological change associated with entry into the VBNC state. The cell wall formed an extended or convoluted structure, which pulled away from the cell membrane. Ribosomal structure, however, was conserved, with no apparent morphological change. Based on the
electronmicrographs, it was concluded that the morphological changes occurring in V. cholerae O1 differed from those of O139, in that V. cholerae O1 exhibited significantly less convolution of the cell wall, while V. cholerae O139 showed distinguishable distortion of the cell wall. However, the two serotypes shared certain structural changes, namely total size reduction, intracellular integrity, disappearance of granules, and compacted nuclear regions. Based on SEM micrographs, cells of Vibrio cholerae O1 and O139 in logarithmic growth phase were not significantly different in morphology (Figs. 3A and 4A). However, SEM of both Vibrio cholerae O1 (Fig. 3, B, C, and D) and O139 (Fig. 4, B, C, and D) showed a few morphological differences. Both vibrios gradually changed in cell shape from curved rods to coccoid cells. At about two months, Vibrio cholerae O1 cells became shorter and fatter, whereas Vibrio cholerae O139 cells became shorter and thinner than actively growing cells. At six months, when all cells were VBNC, both Vibrio cholerae O1 and O139 were coccoid in morphology.
Fig. 3. Scanning electron micrographs of Vibrio cholerae O1 entering into the VBNC state. A, actively growing cells; B, one month (short rods); C, two months (oval, with decreased volume); and D, six months (coccoid cells with a greater than 50% decrease in volume).
Viability of VBNC Vibrio cholerae
Viability and antigenicity of coccoid VBNC cells from microcosms The cells of both serotypes in the microcosms decreased in size during exposure to nutrient depletion. In two-month microcosms, more than 90% of the cells were coccoid and their size was ca. 0.4 µm. In one-year microcosms, all cells were coccoid and between 0.2–0.4 µm in size. Filtration of cell suspensions from each microcosm through 0.8 µm-pore-size filter membranes prior to collection of the cells on selected pore-size membranes was done to remove cell debris and clumps of dead cells. VBNC cells were concentrated up to 10-fold by molecular sieve filtration centrifugation. Comparative cell numbers of V. cholerae O1 obtained by acridine orange total count, direct fluorescent monoclonal antibody staining, and different viable staining techniques are shown in Table 2. Viable counts were obtained as relative percent of the total cell count for each cell sample. Results were the same for both serotypes and demonstrated that populations of coccoid VBNC during long-term incubation changed in number of metabolically active cells. Cells that were VBNC for two months showed a large number of cells with intact cell membranes and active respiring systems. However, after one year, the number of actively respiring VBNC cells, i.e., cells responding to the tests employed, decreased. Coccoid VBNC cells responsive to
Fig. 4. Scanning electron micrographs of Vibrio cholerae O139 entering into the VBNC state. A, actively growing cells; B, one month (short rods); C, two months (oval, with decreased volume); and D, six months (coccoid cells with a, greater than 50% decrease in volume).
337
yeast extract and nalidixic acid were clearly demonstrable in two-month microcosms, while only a few percent of responsive cells could be detected in one-year microcosm samples, employing the standard DVC staining procedure. From Table 2, it can be seen that V. cholerae O1 and O139 cells retained their antigenicity, even in longterm microcosms. The number of fluorescent monoclonal antibody-staining cells was not significantly different from that of the acridine orange staining cells. Viability of the VBNC coccoid cells, determined by staining with the redox dye, 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC), is shown in Fig. 5. Susceptibility of coccoid VBNC cells to disinfectants Coccoid VBNC cells in the size range of 0.2–0.4 µm at a concentration of 108 cells/ml, as well as actively growing Vibrio cells, were exposed to various concentrations of sodium hypochlorite, copper sulfate, zinc sulfate, and formaldehyde for 24 h at 4 °C. Enumeration of respiring cells was performed by the CTC assay. Inhibition of respiration of the VBNC cells confirmed the viability of these coccoid, VBNC cells. The percent decrease in respiring cells was calculated, relative to control (untreated) cell samples. Table 3 shows the concentration of disinfectant that caused cessation of respiration of cells of the vibrios, both control and VBNC cells. The effective concentration
338
S. CHAIYANAN et al.
Table 2. Vibrio cholerae O1 and O139 enumerated by acridine orange total count (AO), direct fluorescent monoclonal antibody staining (DFA), BacLight viable staining, and 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC) redox dye staining. Method
Relative percentage of V. cholerae cell count, with reference to control* Actively growing cells
Two-month microcosm VBNC cells
One-year microcosm VBNC cells
O1
O139
O1
O139
O1
O139
Total count (AO)
100
100
100
100
100
100
DFA
98.9 ± 0.20
99.2 ± 0.70
99.7 ± 0.10
98.6 ± 0.66
98.2 ± 0.70
98.8 ± 0.26
DFA-DVC
90.2 ± 3.08
89.1 ± 1.59
80.1 ± 1.99
72.0 ± 2.36
5.5 ± 1.32
8.2 ± 1.00
BacLight
92.0 ± 1.44
90.3 ± 0.72
90.2 ± 0.85
75.1 ± 2.50
50.4 ± 2.43
52.0 ± 4.42
CTC
89.2 ± 1.15
87.4 ± 2.36
52.0 ± 1.80
49.6 ± 2.36
10.4 ± 0.82
12.2 ± 2.43
* Total cell counts were in the range of 108 cells/ml. Data given are mean and standard error of three determinations. Table 3. Efficacy of disinfectants in inhibiting actively growing cells and viable but non-culturable cells of Vibrio cholerae O1 and Vibrio cholerae O139. Cell state
% Decrease in viable counta Chlorine (µmg/l) O1 1
Actively growing Cells
5 10
CuSO4 (µmg/l) O139
1
O1
5 10
1
5
O139 10
100 100 100 100 100 100 100 100 100
VBNC 2 month microcosm
25 64 90
VBNC 1 year microcosm
6 65 70
30 40 80
8 60
ZnSO4 (µmg/l)
68
25 50 75
4 45 48
1
O1
5 10
1
5 10
100 100 100 89 100 100
32 54
Formaldehyde (µg/l) O139 1
5 10
O1
O139
1.11 1.85 3.7 1.11 1.85 3.7
88 100 100 100
100 100
100 100 100
81
4 49 68
20 52 63
88
89 92
92 100 100
5 42 55
3 42 56
4 40 61
87
89 96
95 100 100
a Cells were exposed for 24 h at 4 °C in 1% IO. Concentrations of disinfectants, as given, are final concentrations. Data shown are averages of duplicate samples.
Viability of VBNC Vibrio cholerae
of zinc sulfate was less than 5 mg/l. Percent decrease in the number of respiring cells, in response to disinfectants, at two months was greater than for one-year microcosms. Formaldehyde showed strong inhibition of respiration of both actively growing and VBNC cells. The response of VBNC cells to the concentrations of sodium hypochlorite, copper sulfate, and zinc sulfate used was not a first order reaction. PCR amplification of toxin-associated genes in VBNC coccoid cells Using chromosomal DNA extracted from actively growing cells as a control, the PCR products of ctxA, toxR, tcpA, and zot genes were detected in coccoid VBNC cells of two-month and one-year microcosms for both serotypes. The persistence of these toxin-associated genes in long-term VBNC cells show that these genes remain detectable in the intact VBNC cells for up to one year, or longer.
Discussion Micrographs prepared using both TEM and SEM revealed changes in cell structure, both internally and externally (Figs. 1–4). However, since cells maintaining an intact cell structure showed no significant morphological differences from the controls, the VBNC cells were considered to be viable cells. During entry into the VBNC state, Vibrio cholerae O1 and O139 decrease in size and become coccoid in morphology, with loss of outer cell wall rigidity. The cytoplasm of the cells condenses, resulting in a significant space between the cytoplasmic membrane and the cell wall. Condensation of the cytoplasm may be related to a dehydration that causes enzymes remaining in the cytoplasm to become inactive, as occurs during formation of endospores (FREESE and HEINZE, 1984). Intracellular integrity decreased in both V. cholerae O1 and O139, accompanied by disappearance of granules and compacting of the nuclear regions. The cell wall of Vibrio cholerae O1 was thicker than that of Vibrio cholerae O139, which was not only thinner but relatively distorted when viewed by transmission electron microscopy. Vibrio cholerae, being an aquatic Gramnegative bacterium, undergoes macromolecular changes when exposed to conditions adverse to growth and multiplication. These changes include construction of a protective layer that is not unlike an exosporium, spore coat, or a cortex, as in spore formation, but with much less definition. In VBNC Vibrio cells, ribosomal structure is conserved, although the number of ribosomes is much reduced. Preservation of ribosomes also occurs in endospores, where it has been suggested that, because protein synthesis is energy-consuming, morphological
339
changes comprise part of the strategy for survival (FREESE and HEINZE, 1984). In the experiments reported here, nonculturable cells of Vibrio cholerae O1 and O139 were further examined to elucidate the association between coccoid morphology and the viable but nonculturable (VBNC) state. Viable counts measured by BacLight staining were significantly higher than DVC and CTC counts (direct viable counts of cells responding to addition of yeast extract), suggesting that nonsubstrate-responsive cells may also be viable, despite lack of response to nalidixic acid and nutrient addition. DVC (elongation in the presence of DNA gyrase inhibitors) and CTC (cellular accumulation of respiring product, insoluble fluorescent crystal of CTC) assays, using 1% yeast extract to prevent nutrient shock of oligotrophic adapted cells, were also done. A longer incubation time for the VBNC cells (24 h) than for actively growing cells (2 h) was employed to allow for substrate uptake and respiration. Prolonged incubation in microcosms was found to affect the viability response of the VBNC cells. However, a minimum of 10% of the VBNC cells maintained in microcosms for one year remained viable i.e., responsive to the test methods used in this study. Environmental factors play a role in prolonging viability of VBNC cells, as well as conditioning cells before they enter the VBNC state. VBNC cells of V. cholerae were very small in size, yet retained antigenicity (Fig. 5). Studies were also done to determine whether respiration of the VBNC V. cholerae could be inhibited as were controls, i.e., actively growing cells. Four disinfectants were used (sodium hypochlorite, copper sulfate, zinc sulfate, and formaldehyde). Viability of VBNC cells was interpreted from the degree of susceptibility to disinfectant. Interestingly, VBNC cells of both serotypes showed greater resistance to all treatments, compared to exponentially growing cultures, a finding that is in agreement with results of studies of bacteria subjected to nutrient starvation, i.e., starved cells (not VBNC) become more resistant to subsequent stresses, such as pressure (BERLIN et al., 1999), heat (GIARD et al., 1996; HARTE et al., 1994; HARTKE et al., 1998), ethanol (HARTE et al., 1994); acid (GIARD et al., 1996; HARTKE et al., 1998); UV light (GIARD et al., 1996); and sodium hypochlorite (LAPLACE et al., 1997). We conclude that the coccoid VBNC cells are more resistant to disinfectants than actively growing cells, with these cells maintaining respiration, as well as cellular integrity, at concentrations of disinfectant inhibitory to actively growing cells. Furthermore, inhibition of respiration of the VBNC cells by disinfectants at higher concentrations, provides evidence of continuing metabolic function in VBNC cells. Since retention of cell membrane integrity is a significant characteristic of viable cells, chromosomal DNA was extracted from the intact VBNC cells maintained for two months and one year in microcosms and conservation of
Fig. 5. Vibrio cholerae O1 from 1% Instant Ocean microcosm. A, Direct fluorescent monoclonal antibody staining and B, 5-cyano2, 3-ditolyl tetrazolium chloride redox dye staining.
340
S. CHAIYANAN et al.
the cholera toxin and toxin-associated genes in VBNC cells in long-term microcosms was demonstrated, using PCR primers for the ctxA, toxR, tcpA, and zot genes. Earlier studies showing that VBNC cells can cause accumulation of fluid in rabbit ligated ileal loops and also can cause diarrhoea in human volunteers (COLWELL et al., 1985) are corroborated at the molecular level by persistence of the genes relevant to pathogenicity in VBNC cells of V. cholerae 01 and 0139. From all of the evidence gathered in this study, it is concluded that VBNC cells retain viability as well as the potential for pathogenicity for significant periods of time, i.e., one year or longer. Acknowledgments The authors gratefully acknowledge support of NIH Grant No. 1RO1A139129-01 and NIH-NINR Grant No. RO1 NR04527-01A1.
References AMY, P. S., PAWLING, C., MORITA, R. Y.: Recovery from nutrient starvation by a marine Vibrio spp. Appl. Environ. Microbiol. 45, 1685–1690 (1983). BAKER, R. M., SINGLETON, E L., HOOD, M. A.: Effects of nutrient deprivation on Vibrio cholerae. Appl. Environ. Microbiol. 46, 930–940 (1983). BERLIN, D. L., HERSON, D. S., HICKS, D. T., HOOVER, D. G.: Response of pathogenic Vibrio species to high hydrostatic pressure. App. Environ. Microbiol. 65, 2776–2780 (1999). CHOWDHURY, M. A. R., HILL, R. T., COLWELL, R. R.: A gene for the enterotoxin zonula occludens toxin is present in Vibrio mimicus and Vibrio cholerae O139. FEMS Microbiology Letters 119, 377–380 (1994). COLWELL, R. R., BRAYTON, P. R., GRIMES, D. J., ROSZAK, D. B., HUG, S. A., PALMER, L. M.: Viable but non-culturable Vibrio cholerae and related pathogens in the environment: implications for release of genetically engineered microorganisms. Bio/Technology 3, 817–820 (1985). COLWELL, R. R., TAMPLIN, M. L., BRAYTON, P. R., GAUZENS, A. L., TALL, B. D., HERRINGTON, D., LEVINE, M. M., HALL, S., HUQ, A., SACK, D. A.: Environmental aspects of Vibrio cholerae in transmission of cholera, pp. 327–343. In: Advances on cholera and related diarrheas, Vol. 7 (R. B. SACK, Y. ZINNAKA, eds.) KTK Scientific, Tokyo, Japan 1990. COLWELL, R. R., BREYTON, P., HERRINGTON, D., TALL, B., HUQ, A., LEVINE, M. M.: Viable but non-culturable Vibrio cholerae O1 revert to a cultivable state in the human intestine. World J. Microbiol. Biotechnol. 12, 28–31 (1996). DIRITA, V. J., PARSOT, C., JANDER, G., MEKALANOS, J. J.: Regulatory cascade controls virulence in Vibrio cholerae. Proc. Natl. Acad. Sci. 88, 5403–5407 (1991). FARUQUE, S. M., ALBERT, M. J., MEKALANOS, J. J.: Epidemiology, Genetics, and Ecology of Toxigenic Vibrio cholerae. Micro. Mol. Bio. Rev. 62, 1301–1314 (1998). FLINT, K. P.: The long-term survival of Escherichia coli in river water. J. Appl. Bacteriol. 63, 261–270 (1987). FREESE, E., HEINZE, J.: Metabolic and genetic control of bacterial sporulation, pp. 101–172. In: The bacterial spore (A. HURST, G. W. GOULD, eds.), Academic Press, Inc., New York 1984. GIARD, J. C., HARTKE, A., FLAHAUT, S., BENACHOUR, A., BOUTIBONNES, P., AUFFRAY, Y.: Starvation-induced multiresistance in Enterococcus faecalis JH2-2. Curr. Microbiol. 32, 264–271 (1996).
GRIMES, D. J., ATWELL, R. W., BRAYTON, P. R., PALMER, L. M., ROLLINS, D. M., ROSZAK, D. B., SINGLETON, F. L., TAMPLIN, M. L., COLWELL, R. R.: The fate of enteric pathogenic bacteria in estuarine and marine environments. Microbiol. Sci. 3, 324–329 (1986). GUETIN, A. M., MISHISTINA, I. E., ANDREEV, L. V., BOBYK, M. A., LAMBINA, V. A.: Some problems of the ecology and taxonomy of marine microvibrios. Biol. Bull. Acad. Sci. USSR. 5, 336–340 (1979). HARTKE, A., BOUCHE, S., GANSEL, X., BOUTIBONNES, P., AUFFRAY, Y.: Starvation-induced stress resistance in Lactococcus lactis subsp. lactis IL 1403. Appl. Environ. Microbiol. 60, 3474–3478 (1994). HARTKE, A. J. GIARD, LAPLACE, J., AUFFRAY, Y.: Survival of Enterococcus faecalis in an oligotrophic microcosm: changes in morphology, development of general stress resistance, and analysis of protein synthesis. Appl. Environ. Microbiol. 64, 4238–4245 (1998). HASAN, J. A. K., A. HUQ, M. L. TAMPLIN, R. SIEBELING, R. R. COLWELL. A Novel Kit for Rapid Detection of V. cholerae 01. J. Clin. Microbiol. 32, 249–252 (1994). HAYET, M. A. (ed.): Principle and techniques of electron microscopy: biological applications. Macmillan Press, London 1989. HERRINGTON, D. A., HALL, R. H., LOSONSKY, G., MEKALANOS, J. J., TAYLOR, R. K., LEVINE, M. M. Toxin, toxin-coregulated pili, and the toxR regulon are essential for Vibrio cholerae pathogenesis in humans. J. Exp. Med. 168, 1487–1492 (1988). HOBBIE, J. E., DALEY, R. J., JASPER, S.: Use of nucleopore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33, 1225–1228 (1977). HOOD, M. A., GUCKERT, J. B., WHITE, D. C., DECK, F.: Effect of nutrient deprivation on lipid, carbohydrate, DNA, RNA, and protein levels in Vibrio cholerae. Appl. Environ. Microbiol. 52, 788–793 (1986). KENNEDY, S. F., COLWELL, R. R., CHAPMAN, G. B.: Ultrastructure of a psychrophilic marine vibrio. Can. J. Microbiol. 16, 1027–1032 (1970). KOGURE, K., SIMIDU, U., TAGA, N.: A tentative direct microscopic method for counting living marine bacteria. Can. J. Microbiol. 25, 415–420 (1979). LAPLACE, J. M., THUAULT, M., HARTKE, A., BOUTIBONNES, P., AUFFRAY, Y.: Sodium hypochlorite stress in Enterococcus faecalis: influence of antecedent growth conditions and induced proteins. Curr. Microbiol. 34, 284–289 (1997). MANNING, P. A., STROEHER, U. H., MORONA, R.: Molecular basis for O-antigen biosynthesis in Vibrio cholerae Ol: Ogawa-Inaba switching, pp. 77–94. In: Vibrio cholerae and cholera: molecular to global perspectives (I. K. WACHSMUTH, P. BLAKE, O. OLSVIK, eds.), ASM Press, Washington, DC 1994. MCKAY, A. M.: Viable but non-culturable forms of potentially pathogenic bacteria in water. Lett. Appl. Microbiol. 14, 129–135 (1992). MISHISTINA, I. E., KAMENEVA, T. G.: Bacterial cells of minimal size in the Barents Sea during the polar night. Microbiology. 50, 360–363 (1981). NILSSON, L., OLIVER, J. D., KJELLEBERG S.: Resuscitation of Vibrio vulnificus from the viable but nonculturable state. J. Bact. 173, 5054–5059 (1991). NOVITSKY J. S., MORITA, R. Y.: Morphological characterization of small cells resulting from nutrient starvation of a psychrophilic marine vibrio. Appl. Environ. Microbiol. 32, 617–622 (1976).
Viability of VBNC Vibrio cholerae NOVITSKY J. S., MORITA, R. Y.: Survival of a psychrophilic marine vibrio under long-term nutrient starvation. Appl. Environ. Microbiol. 33, 635–641 (1977). NOVITSKY J. S., MORITA, R. Y.: Possible strategy for the survival of marine bacteria under starvation conditions. Mar. Biol. 48, 289–295 (1978). OLIVER, J. D., WANUCHA, D.: Survival of Vibrio vulnificus at reduced temperatures and elevated nutrient. J. Food. Saf. 10, 79–86 (1989). RODRIGUEZ, G. G., PHIPPS, D., ISHIGURO, K., RIDGWAY, H. F.: Use of a fluorescent redox probe for direct visualization of actively respiring bacteria. Appl. Environ. Microbiol. 58, 1801–1808 (1992). ROLLINS, D. M., COLWELL, R. R.: Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Appl. Environ. Microbiol. 52, 531–538 (1986). ROSZAK, D. B., GRIMES, D. J., COLWELL, R. R.: Viable but nonrecoverable stage of Salmonella enteritidis in aquatic systems. Can. J. Microbiol. 30, 334–338 (1984). SHIRAI, H., NISHIBUCHI, M., RAMAMURTHY, T., BHATTACHARYA, S. K., PALL, S. C., TAKEDA, Y.: Polymerase chain reaction for de-
341
tection of the cholera enterotoxin operon of Vibrio cholerae. J. Clin. Microbiol. 29, 2517–2521 (1991). STEVENSON, L. H.: A case for bacterial dormancy in aquatic systems. Microb. Ecol. 4, 127133 (1978). TERZIEVA S., DONNELLY, J., ULEVICIUS, V., GRINSHPUN, S. A., WILLEKE, K., STELMA, G. N., BRENNER, K. P.: Comparison of methods for detection and enumeration of airborne microorganisms collected by liquid impingement. Appl. Environ. Microbiol. 62, 2264–2272 (1996). XU, H. S., ROBERT, N., SINGLETON, F. L., AATTWELL, R. W., GRIMES, D. J., COLWELL, R. R.: Survival and viability of nonculturable Escherichia coli and Vibrio cholerae in the estuarine and marine environment. Microb. Ecol. 8, 313–323 (1982). Corresponding author: RITA C. COLWELL, Center of Marine Biotechnology, University of Maryland Biotechnology Institute, 701 E. Pratt Street, Baltimore, MD 21202, USA. Tel.: ++1-703-2928000; Fax: ++1-703-2929232; e-mail: [email protected]