Unique Kinetic Properties of Phenol-Degrading Variovorax Strains Responsible for Efficient Trichloroethylene Degradation in a Chemostat Enrichment Culture

Abstract

A chemostat enrichment of soil bacteria growing on phenol as the sole carbon source has been shown to exhibit quite high trichloroethylene (TCE)-degrading activities (H. Futamata, S. Harayama, and K. Watanabe, Appl. Environ. Microbiol. 67:4671-4677, 2001). To identify the bacterial populations responsible for the high TCE-degrading activity, a multidisciplinary survey of the chemostat enrichment was conducted by employing molecular-ecological and culture-dependent approaches. Three chemostat enrichment cultures were newly developed under different phenol-loading conditions (0.25, 0.75, and 1.25 g liter⁻¹ day⁻¹) in this study, and the TCE-degrading activities of the enrichments were measured. Among them, the enrichment at 0.75 g liter⁻¹ day⁻¹ (enrichment 0.75) expressed the highest activity. Denaturing gradient gel electrophoresis of PCR-amplified 16S rRNA gene fragments detected a Variovorax ribotype as the strongest band in enrichment 0.75; however, it was not a major ribotype in the other samples. Bacteria were isolated from enrichment 0.75 by direct plating, and their 16S rRNA genes and genes encoding the largest subunit of phenol hydroxylase (LmPHs) were analyzed. Among the bacteria isolated, several strains were affiliated with the genus Variovorax and were shown to have high-affinity-type LmPHs. The LmPH of the Variovorax strains was also detected as the major genotype in enrichment 0.75. Kinetic analyses of phenol and TCE degradation revealed, however, that these strains exhibited quite low affinity for phenol compared to other phenol-degrading bacteria, while they showed quite high specific TCE-degrading activities and relatively high affinity for TCE. Owing to these unique kinetic traits, the Variovorax strains can obviate competitive inhibition of TCE degradation by the primary substrate of the catabolic enzyme (i.e., phenol), contributing to the high TCE-degrading activity of the chemostat enrichments. On the basis of physiological information, mechanisms accounting for the way the Variovorax population overgrew the chemostat enrichment are discussed.

Trichloroethylene (TCE) is a suspected carcinogen and is one of the most commonly detected volatile organic contaminants in soil and groundwater. Indeed, TCE is considered to be a potentially serious threat to drinking water sources (26). Considering this, research that has focused on both anaerobic reductive dechlorination and aerobic degradation via cometabolism has been under development (25, 33). A number of laboratory studies have demonstrated that aliphatic and aromatic hydrocarbon-degrading bacteria, such as those that degrade phenol, toluene, and methane, are capable of aerobic cometabolic transformation of TCE to readily degradable oxygenated compounds (6, 32, 36). Field trials using these bacteria and/or the inducing substrates (e.g., phenol, toluene, and methane) for TCE bioremediation have also been reported previously (1-3, 12, 18, 19, 22, 27, 32, 35). However, since the enzymes involved in TCE transformation preferentially catalyze the genuine substrates (i.e., the inducers for TCE-degrading activity) rather than TCE, TCE is not always efficiently degraded in the presence of inducers such as phenol (7, 14, 21).

We have been studying phenol-degrading bacteria to develop efficient TCE bioremediation strategies (13-16), because previous studies have shown that phenol is the most potent inducer of bacterial TCE-degrading activity (20). It has been shown that phenol-loading methods (e.g., batchwise or continuous supply) largely affected the TCE-degrading activity of enrichment cultures developed from a TCE-contaminated soil (15, 16). The most potent enrichment was developed in a chemostat culture (i.e., continuously phenol-fed enrichment) that showed TCE-degrading activity as high as the activity expressed by a pure culture of Methylosinus trichosporium OB3b (15). In parallel, we have isolated many phenol-degrading bacteria and analyzed their kinetic properties for phenol and TCE degradation and their phenol-hydroxylase genes (summarized in reference 37). Taking all these data together, it has been suggested that phenol-degrading bacteria having high-affinity-type phenol hydroxylases were present in the chemostat enrichment (15), although bacterial species responsible for the high TCE-degrading activity in the chemostat enrichment have not yet been identified.

The primary aim of the present study was to identify bacterial species responsible for the high TCE-degrading activity in the chemostat enrichment. Identification of the major players has been suggested to provide pivotal information as to the utilization and manipulation of microbial communities (37, 42). In addition, this study included detailed kinetic analyses of the major players (i.e., Variovorax strains) and showed that they exhibited unique kinetic properties. On the basis of the information obtained in this study, we discuss how these Variovorax strains became the major players under conditions of chemostat enrichment when fed phenol as the sole carbon source.

MATERIALS AND METHODS

Construction of enrichment cultures.

A total of 1.5 liters of BSM medium (15) containing 0.2 mM of phenol as the sole carbon source (in a chemostat reactor [2 liters in capacity]) was inoculated with 30 g (wet weight) of aquifer soil obtained from a depth of −10 m (Kururi, Chiba prefecture, Japan) (17). After the initial phenol was completely degraded, BSM medium containing phenol (500, 1,500, or 2,500 mg liter⁻¹) was continuously supplied at a flow rate of 31.5 ml h⁻¹ (i.e., the hydraulic residence time was 2.0 days). The culture volume was maintained at 1.5 liters. The culture was stirred at 150 rpm, and the temperature and pH were kept at 25°C and 7.0, respectively. Air was filtered through 0.2-μm-pore-size membrane filters (Millipore) and supplied to the culture at 1.5 liters min⁻¹. The concentration of phenol in the culture was measured by a colorimetric assay with a Phenol Test Wako kit (Wako Pure Chemicals) (15) or by high-pressure liquid chromatography (14). When phenol was detected at more than 0.5 mg liter⁻¹, the addition of phenol was stopped until phenol was no longer detected. Total direct counts of microbial cells were measured by using fluorescence microscopy after cells were stained with 4′,6′-diamidino-2-phenylindole (DAPI; Sigma-Aldrich, Chuo-ku, Japan).

DGGE.

Bacterial community structures were analyzed by denaturing gradient gel electrophoresis (DGGE) targeting 16S rRNA genes. Variable region V3 of bacterial 16S rRNA genes (corresponding to positions 341 to 534 in the Escherichia coli sequence) was amplified using primers P2 and P3 (containing a 40-bp GC clamp (28) and a Takara thermal cycler (Takara Shuzo) as described previously (39). A Dcode DGGE system (Bio-Rad) was used for electrophoresis as recommended by the manufacturer. A total of 10 μl of a PCR-amplified mixture was subjected to electrophoresis in a 10% (wt/vol) polyacrylamide gel at 200 V for 3.5 h. Gel gradients used for separation, which were applied in parallel to the electrophoresis direction, were 30 to 50%. After electrophoresis, the gel was stained with SYBR Green I (FMC Bioproducts) for 30 min as recommended by the manufacturer. The nucleotide sequences of DGGE bands were determined as described previously (39).

Quantitative PCR (q-PCR). (i) Group-specific monitoring of phenol-hydroxylase gene fragments.

Fragments of genes encoding the largest subunit of multicomponent phenol hydroxylase (LmPHs) were amplified by PCR using specific sets of primers as described previously (15). The primer set pheUf and pheLr (hereafter described as pheUf/pheLr) was used to amplify LmPHs of low-K_s groups (high-activity TCE degraders), and the primer set pheUf/pheHr was used to amplify LmPHs of high-K_s groups (low-activity TCE degraders) (14, 15). The nucleotide sequences of these primers, compositions of PCR mixtures, and amplification conditions were described previously (15). Competitor fragments used were constructed in our previous study (15). Gel electrophoresis, quantification of band intensities, and estimation of copy numbers were conducted as previously described (15).

(ii) Monitoring of LmPHs of the Variovorax strains.

PCR primers were designed for specifically detecting LmPHs of Variovorax strains by comparing LmPH sequences of 35 phenol-degrading bacteria (including Variovorax strains) (15); primers designed were a forward primer (pheVf [5′-CGGCTATGCCATGGCCGGCGCTTC-3′], corresponding to Pseudomonas putida CF600 LmPH positions 342 to 365) and a reverse primer (pheVr [5′-TTGTCCGGGTCCTGCTCGAGG-3′], corresponding to positions 744 to 764). Competitor fragments were produced by using a competitive DNA construction kit (Takara Shuzo). The sizes of the target fragment and competitor fragment were 422 and 245 bp, respectively. The conditions of amplification were as stated above (15). PCR conditions were as follows: 10 min of activation at 94°C, followed by 5 cycles consisting of 1 min at 94°C, 1 min at 60°C, and 1 min at 72°C. The annealing temperature was decreased by 1°C in the next 5 cycles, followed by 25 cycles in which the annealing temperature was 58°C. Finally, 10 min of extension at 72°C was conducted.

Isolation of bacteria.

Three types of media which were supplemented with phenol at 0.5 mM were used to isolate soil bacteria: MP medium (39) (MP50phe), 1/10-strength Trypticase soy broth (TSB) medium (1/10TSB50phe; Difco, Detroit, Mich.), and TSB medium (TSB50phe). The diluted soil suspension or the diluted culture suspension was directly spread onto agar plates containing each medium. After the plates were incubated at 25°C for 7 to 14 days, colonies formed were picked and spread on 1/10TSB50phe. This purification procedure was repeated several times.

Sequencing of 16S rRNA gene.

Purified bacteria were incubated in the 1/10TSB50phe for 2 days at 25°C, and their DNA was then extracted from a bacterial culture by a method described previously (15). The quality and quantity of the extracted DNA were checked by measuring the absorbance at 260 and 280 nm. Almost-full-length fragments of 16S rRNA gene were amplified using the following primers: 5′-AGAGTTTGATCCTGGCTCAG-3′ (E. coli 16S rRNA gene positions 8 to 27) (3) and 5′-AAGGAGGTGATCCAGCC-3′ (positions 1525 to 1542). Amplification was performed with a Progene thermal cycler (Techne) by using a 50-μl mixture containing 2.5 U of Taq DNA polymerase (Amplitaq Gold; Applied Biosystems), 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, 0.001% (wt/vol) gelatin, each deoxynucleoside triphosphate at a concentration of 200 μM, 100 pmol of each primer, and 50 ng of templated DNA. The PCR conditions used were as follows: 10 min of activation of the polymerase at 94°C, followed by 30 cycles consisting of 1 min at 94°C, 1 min at 52°C, and 1 min at 72°C, and, finally, 10 min of extension at 72°C. The PCR products were electrophoresed through a 1.5% (wt/vol) agarose gel with Tris-borate-EDTA buffer (30) and then purified with a MicroSpin S-400 HR column (Amersham Pharmacia Biotech Inc.). The nucleotide sequences of the PCR products were determined by using a Dye terminator cycle DNA sequencing kit (Perkin-Elmer) as described by Edwards et al. (5). The products of the sequencing reactions were analyzed with a model 377 DNA sequencer (Perkin-Elmer).

Determination of TCE-degrading activity.

The TCE-degrading activities of the enrichments and isolates were measured at the initial TCE concentration of 0.5 mg liter⁻¹ by a bottle incubation assay as previously described (14). TCE-degrading activities were calculated as the pseudo-first-order degradation rate constant (k₁) according to previous studies (14, 15, 17, 34, 35). All isolate cultures to be analyzed were prepared by chemostat enrichment growing on phenol at a loading rate of 0.75 g per liter per day. Samples of cultures were obtained after the parameters (pH, phenol concentration, dissolved oxygen concentration, and optical density at 600 nm) became stable.

Determination of phenol- and catechol-oxygenating activities.

The chemostat culture used for measuring the TCE-degrading activity was also used to determine the phenol- and catechol-oxygenating activity of the culture. The phenol- and catechol-oxygenating activities (phenol- and catechol-oxygen consumption rates) were measured at various substrate concentrations by using an oxygen electrode (DO Meter B-505; Iijima Electronics Co.) after the respiratory oxygen consumption was suppressed by adding potassium cyanide (38). One unit of activity was equivalent to 1 mmol of oxygen consumed per min, while the specific activity was defined as the activity per gram of dried cells.

Kinetic analysis.

Kinetic parameters were calculated according to Michaelis-Menten's equation or Haldane's equation (11, 39). The activities at more than 10 different substrate concentrations were used to calculate these kinetic parameters with JMP statistical visualization software (SAS Institute Inc.). The apparent kinetic constants, K_S (affinity constant), K_SI (inhibition constant), and V_max (theoretical maximum activity) were determined by the nonlinear regression method as described previously (38, 40). Following Folsom et al. (11), the term K_s was employed instead of K_m, because the activity was measured by using intact cells rather than purified enzymes.

Statistics.

Data were statistically analyzed by the Student t test. A value of P = 0.05 was considered significant.

Nucleotide sequence accession numbers.

The nucleotide sequences reported in this paper have been deposited in the GSDB, DDBJ, and EMBL nucleotide sequence databases under accession numbers AB167174 to AB167268.

RESULTS

Performance of chemostat enrichments.

To investigate the effects of phenol-loading rates on TCE-degrading activity of the chemostat enrichment, phenol was continuously supplied at the loading rates of 0.25, 0.75, and 1.25 g liter⁻¹ day⁻¹, and the k₁ values (TCE-degrading activity) of the enrichments were determined (Fig. 1). Under these operational conditions, the culture parameters became stable after day 20. Among these enrichment cultures, the enrichment at 0.75 g liter⁻¹ day⁻¹ (enrichment 0.75) exhibited the highest k₁ value (the mean ± standard deviation of the k₁ value from day 20 to 36 [n = 5] was 51.1 ± 4.0 liter g⁻¹ h⁻¹), and the second highest k₁ value was observed in enrichment 0.25 (33.8 ± 3.1 liter g⁻¹ h⁻¹). Enrichment 1.25 exhibited a low k₁ value (10.2 ± 1.7 liter g⁻¹ h⁻¹). During the enrichment culture procedure, phenol was occasionally detected at more than 1.0 mg liter⁻¹ in enrichment 1.25, while no phenol was detected in the other enrichment cultures.

FIG. 1.

Changes in TCE-degrading activities (k₁ values) expressed by the enrichment cultures. The enrichments were constructed by batch culture with 0.2 mM phenol for 3 days, after which they were enriched by chemostat culture with the following phenol loading rates: 0.25 g liter⁻¹ day⁻¹ (□), 0.75 g liter⁻¹ day⁻¹ (▪), and 1.25 g liter⁻¹ day⁻¹ (•). The means of three determinations are shown, and the error bars indicate standard deviations.

DGGE.

DGGE analysis of PCR-amplified partial 16S rRNA genes was conducted to compare the bacterial populations present in these three enrichment cultures (Fig. 2). Major bands were excised from the DGGE gels, and their nucleotide sequences were determined and used to deduce the related organisms, although the sequenced DNA lengths were very short (only 120 to 160 bp) for phylogenetic analysis. The analysis indicated that a band representing a Variovorax population (band S8; identity to Variovorax sp. HAB-24 [accession no. AB051691] was 100%) appeared solely in enrichment 0.75. In enrichment 0.25, a Variovorax population (band S13) appeared as one of the several major bands, while it was hardly seen in enrichment 1.25 (band S17). The other bacterial populations detected in the DGGE analysis included strains of Pseudomonas (bands S1, S2, S3, S9, and S15), Acinetobacter (bands S4 and S10), Acidovorax (bands S5, S11, and S14), Chryseobacterium (band S16), Ralstonia (band S18), and Azoarcus (band S20) and uncultured or unidentified bacterial strains (bands S6, S7, and S19). The identity of almost all DGGE bands analyzed was above 98%. The results suggest that the Variovorax population preferentially occurred in enrichment 0.75, which was also suggested by the PCR-DGGE analysis targeting LmPH (data not shown).

FIG. 2.

DGGE profiles of partial 16S rRNA gene fragments during enrichment at 0.75 g liter⁻¹ day⁻¹ (enrichment 0.75). Bands marked with arrows were excised and sequenced (described in section concerning DGGE in Results). Numbers noted above the photograph indicate sampling days.

Isolation of bacteria.

On the basis of the results of the DGGE analysis described above, we hypothesized that the Variovorax population was responsible for the high TCE-degrading activity in the chemostat enrichment. To test this hypothesis, we attempted to isolate bacterial strains that corresponded to the major DGGE bands from enrichment 0.75 at different time points with the intention of determining their role in the TCE-phenol biodegradation process. Colonies formed on the agar plates were randomly picked (74 colonies in total), purified, and subjected to 16S rRNA gene analysis. The isolates were then classified on the basis of their 16S rRNA gene sequences (Table 1). Table 1 shows that the isolated strains were affiliated with members of the Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, Actinobacteria, and Firmicutes, while several strains were unable to be affiliated with a known phylum. Some strains had 16S rRNA gene sequences identical to those represented by the major DGGE bands in Fig. 2. In particular, the Variovorax strains corresponding to band S8 were isolated from the enrichment on days 10 and 36; the most abundant isolate from the day 36 enrichment was the Variovorax strain. This tendency of bacterial isolation efficiency corresponded to the DGGE analysis of bacterial populations in the chemostat enrichment (Fig. 2).

TABLE 1.

List of bacterial strains isolated from the enrichment^a

Day	Bacterial group	No. of isolates	Closest relative (% similarity)	Identical DGGE band
0	Alphaproteobacteria	1	Sinorhizobium sp. (99)
	Gammaproteobacteria	1	Pseudomonas syringae pv. phaseolicola (99)
		1	Pseudomonas borealis (99)
	Actinobacteria	5	Arthrobacter polychromogenes (98)
		1	Arthrobacter sp. strain SMCC ZAT056 (98)
		1	Actinomyces sp. strain (92)
		1	Cellulomonas variformis strain MX5 (99)
		2	Nocardioides jensenii (97-98)
	Unidentified	2	Uncultured eubacterium (98)
3	Betaproteobacteria	1	Acidovorax sp. strain UFZ-B517 (98)	S5
		1	Ralstonia eutropha KT1 (96)
		1	Ralstonia sp. strain JMP134 (98)
	Gammaproteobacteria	1	Pseudomonas beteli (99)
		1	Pseudomonas sp. strain PsA (99)
		2	Stenotrophomonas maltophilia (99)
	Actinobacteria	1	Arthrobacter oxydans (99)
		1	Arthrobacter polychromogenes (99)
10	Alphaproteobacteria	2	Brevundimonas diminuta (98)
		1	Chelatobacter heintzii (99)
		1	Sinorhizobium sp. strain BK10 (99)
	Betaproteobacteria	2	Ralstonia eutropha strain KT1 (99)
		1	Ralstonia sp. strain KN1 (99)
		1	Variovorax sp. strain WFF52 (99)
		1	Variovorax sp. strain 6C-13 (99)	S8
	Gammaproteobacteria	15	Acinetobacter sp. strain OM-E81 (99)	S4
		1	Pseudomonas sp. strain IpA-2 (99)
	Firmicutes	4	Bacillus fusiformis strain D (99)
	Unidentified	1	Uncultured eubactrium WD2115 (99)
36	Alphaproteobacteria	4	Rhizobium sp. strain H-4 (97)
	Betaproteobacteria	3	Ralstonia sp. strain KN1 (99)
		11	Variovorax sp. strain HAB-24 (99)	S8
	Gammaproteobacteria	2	Acinetobacter sp. strain OM-E81(99)	S4

^aThe enrichment constructed at phenol loading rate of 0.75g L⁻¹ day⁻¹ was used.

Analyses of the LmPH fragments of Variovorax strains.

LmPH fragments of the Variovorax strains isolated in the present study were amplified by PCR, and their nucleotide sequences were determined. The sequences were compared to those of LmPH fragments determined in our previous study (15). It was found that the LmPH sequences of the Variovorax strains determined in the present study were almost identical to each other and were closely related to the LmPH sequences of the Variovorax strains isolated in our previous study (including strains HAB-24 and HAB-30) (data not shown).

To specifically monitor the Variovorax population in the course of the enrichment culture and to demonstrate that it was actually dominant in enrichment 0.75, PCR primers were designed by comparing the LmPH sequences of the Variovorax strains and those of other phenol-degrading bacteria (15, 38) (Fig. 3A). Analysis of isolated bacteria by use of the designed primers (pheVf and pheVr) showed that PCR was positive only for the Variovorax strains (data not shown). A q-PCR assay was developed using these primers and used to analyze the population dynamics of the Variovorax strains in the course of enrichment. The population dynamics of the Variovorax strains were quantitatively compared to population dynamics of different types of phenol-degrading bacteria; we have previously developed q-PCR assays that allowed group-specific monitoring of different types of LmPHs, namely, LmPHs showing high affinity to phenol (low-K_s type) and those showing low affinity (high-K_s type) (15). It was found that bacteria with low-K_s-type LmPHs more efficiently degrade TCE than those with high-K_s-type LmPHs (14, 15). Sequence analysis indicated that the LmPHs of the Variovorax strains were affiliated with the low-K_s-type LmPH (data not shown). Figure 3B shows the results of q-PCR assays for enrichment 0.75. It can be seen that the initial batch cultivation resulted in the rapid growth of bacteria with high-K_s-type LmPHs, while the dominant genotype was shifted from the high-K_s type to the low-K_s type during the subsequent chemostat cultivation. After the culture parameters became stable (day 20 afterward), the Variovorax LmPH occupied more than 80% of the total LmPH copies (sum of high-K_s type and low-K_s type).

FIG. 3.

The q-PCR for monitoring LmPHs of Variovorax strains. (A) Signature amino acid residues for LmPH of Variovorax strains. Regions used for designing the Variovorax-specific primers are underlined. Numbers above the sequence correspond to the numbering in the DmpN sequence (29). (B) Changes in phenol-degrading populations in enrichment 0.75 as analyzed by q-PCR. Variovorax LmPH (•), high-K_S type (▵), low-K_S type (▴), total direct count (○), and k₁ value (▪) are shown. The means of three determinations are shown, and error bars indicate standard deviations.

Kinetic properties of the enrichment and isolates.

Kinetic analyses of the TCE-, phenol-, and catechol-degrading activities of the chemostat enrichment (i.e., enrichment 0.75) and isolated bacteria were conducted to compare their kinetic parameters (Table 2). During the course of enrichment, the K_s value for TCE decreased from 300 ± 3.5 μM (after the batch culture) to 5.3 ± 0.7 μM (during the chemostat phase). The K_SI value for phenol dramatically decreased from 3,500 ± 690 μM to 300 ± 130 μM. The substrate-degrading efficacy (i.e., V_max/K_s value) has been recognized as an important index for the evaluation of enzymatic reactions (10). The V_max/K_s values for TCE greatly increased from 0.002 to 0.340, while the increase of the V_max/K_s values for phenol (from 9.7 to 28.5) was relatively small. This study also analyzed the kinetics of catechol degradation, since catechol is also considered to affect bacterial population structure due to its acute toxicity (31). The K_s value of enrichment 0.75 for catechol decreased from 38.5 ± 8.3 μM to 7.0 ± 0.6 μM, while the K_SI value increased from ∼5,000 to 57,000 ± 25,000 μM. The V_max/K_s values for catechol increased from 5.2 to 77 ± 15 μM, indicating that catechol degradation became quite efficient during the chemostat enrichment on phenol.

TABLE 2.

Comparison of kinetic parameters for TCE, phenol, and catechol of enrichment 0.75 and isolates^a

Reactor and day or isolate	Kinetic parameters for TCE					Kinetic parameters for phenol				Kinetic parameters for catechol
	Ks (μM)	Vmax (μmol min−1 g−1 of dry cells)	Kst (μM)	Vmax/Ks (liters min−1 g−1 of dry cells)	Kl (liters g−1 h−1)	Ks (μM)	Vmax (μmol min−1 g−1 of dry cells)	Kst (μM)	Vmax/Ks (liters min−1 g−1 of dry cells)	Ks (μM)	Vmax (μmol min−1 g−1 of dry cells)	Kst (μM)	VmaxKs
Reactor
3	300 ± 35	1.4 ± 0.5	>1,300	0.002	5.6 ± 0.1	3.2 ± 0.8	31 ± 1.5	3,500 ± 690	9.7	38.5 ± 8.3	200 ± 8.0	>5,000	5.2
10	5.1 ± 3.4	0.9 ± 0.06	7,800 ± 190	0.112	31 ± 4.2	2.7 ± 1.0	46 ± 15	1,900 ± 1,500	16.9	7.5 ± 2.2	307 ± 22	3,800 ± 1,270	41
14	6.4 ± 1.0	1.3 ± 0.04	>1,300	0.203	45 ± 1.7	4.5 ± 0.9	25 ± 1.4	830 ± 160	5.6	6.8 ± 1.8	480 ± 33	8,500 ± 1,860	71
21	5.1 ± 1.4	1.4 ± 0.08	>1,300	0.274	49 ± 4.6	1.9 ± 1.4	58 ± 22	400 ± 350	30.5	6.2 ± 1.8	560 ± 33	13,500 ± 5,860	90
36	4.7 ± 2.5	1.6 ± 0.13	>1,300	0.340	56 ± 4.5	3.7 ± 1.5	98 ± 18	300 ± 130	26.5	7.3 ± 1.8	500 ± 30	57,000 ± 25,000	69
Isolate
Acinetobactor sp. strain c1b	71 ± 43	0.75 ± 0.2	>1,300	0.010	10 ± 0.5	1.1 ± 0.4	31 ± 1.4	4,800 ± 1,620	28.2	7.1 ± 0.9	273 ± 8.1	>12,500	38.5
Acinetobactor sp. strain c26b	280 ± 100	0.25 ± 0.03	>1,300	0.001	0.4 ± 0.02	2.4 ± 1.8	16 ± 2.4	80 ± 30	6.5	4.6 ± 1.4	63 ± 2.4	>10,000	13.8
Variovorax sp. strain c24b	3.5 ± 1.5	1.2 ± 0.1	2,540 ± 1,270	0.346	51.4 ± 9.8	8.2 ± 1.2	93 ± 6.5	220 ± 45	11.3	31 ± 4.4	620 ± 26	12,000 ± 2,000	20.1
Variovorax sp. strain YN07b	2.9 ± 2.2	0.7 ± 0.2	415 ± 200	0.252	88 ± 7.8	12.4 ± 1.4	66 ± 2.5	1,200 ± 120	5.4	36 ± 14	890 ± 110	22,000 ± 2,800	24.8
Variovorax sp. HAB-24c	6.7 ± 1.2	1.4 ± 0.1	2,300 ± 1,300	0.213	98 ± 0.3	7.6 ± 1.5	107 ± 12	180 ± 47	14.1	8.0 ± 2.4	330 ± 46	14,600 ± 5,200	42.3
Variovorax sp. HAB-30c	6.2 ± 1.0	1.7 ± 0.1	1,230 ± 260	0.270	199 ± 6.7	5.8 ± 1.0	160 ± 11	200 ± 40	27.6	7.4 ± 4.5	320 ± 35	42,000 ± 5,750	42.7

^aThe reactor consortium was constructed with phenol at a loading rate of 0.75 g of phenol per liter per day. Data are means ± standard deviations (n = 3).

^bThese bacteria were isolated in this study. Strains c26 and YN07 were isolated on day 10, and strains c1 and c24 were isolated on day 36. The similarity of 16S rRNA genes between strain c1 and c26 is 99.5%.

^cThese bacteria were reported by Futamata et al. (15).

The kinetics of some isolated bacteria that corresponded to the major DGGE bands in enrichment 0.75 (Fig. 2) were analyzed (Table 2). Among them, Acinetobacter sp. strain c26 and Variovorax strain YN07 were isolated on day 10, while three strains (Acinetobacter c1 strains and Variovorax strain c24) were isolated on day 36. In addition, two Variovorax strains (strains HAB-24 and HAB-30) that had been isolated in our previous study (15) from a chemostat enrichment of the same aquifer soil were also analyzed, because their 16S rRNA gene sequences were also identical to that of the major DGGE band S8 (Fig. 2). Even though the two Acinetobacter strains had almost identical 16S rRNA genes (the similarity of the 16S rRNA genes of strain c1 and c26 is 99.5%), their kinetics were different in many ways. Some kinetic parameters of strain c26 were similar to those of enrichment on the early days. The four Variovorax strains exhibited similar types of kinetics, which were also similar to those of the enrichment on the later days (e.g., day 36); on these days, band S8 representing the genus Variovorax was detected as the dominant DGGE band (Fig. 2). These data thus support the idea that the Variovorax population played the dominant role in enrichment 0.75.

DISCUSSION

In the present study we identified the bacterial population that was responsible for the high TCE-degrading activity expressed by a chemostat enrichment of soil bacteria and characterized its kinetic properties for phenol, catechol, and TCE degradation. Several different lines of evidence, including those obtained by molecular ecological analyses targeting 16S rRNA genes and LmPH and kinetic analyses of enrichment and isolated bacteria, indicate that the Variovorax population possessing low-K_s-type LmPH played the major role in TCE degradation.

Previously, we showed that phenol-degrading bacteria could be classified into three types on the basis of their kinetic properties for phenol and TCE degradation, i.e., low-K_s, moderate-K_s, and high-K_s types (14). A positive correlation between affinities for phenol and TCE is found, i.e., low-K_s bacteria exhibit high affinities for both phenol and TCE whereas high-K_s bacteria exhibit low affinities for both phenol and TCE (14). Low-K_s-type phenol-degrading bacteria degrade TCE more efficiently than other types (14, 15). Unexpectedly, however, the K_s values for phenol of the Variovorax strains were found to be very high compared to those of other phenol-degrading bacteria, while the strains exhibited very low K_s values for TCE (Fig. 4). This reverse correlation between kinetics for phenol and TCE was found in the present study for the first time and placed the Variovorax strains at a unique position in a kinetic matrix for phenol and TCE. We suggest that these unique kinetics of the Variovorax strains might be very advantageous for cometabolic TCE bioremediation, since this property can minimize the major limitation of TCE bioremediation, i.e., competitive inhibition of TCE degradation by the genuine substrate for the catabolic enzymes, i.e., phenol (14). The high TCE-degrading activity of enrichment 0.75 could thus be attributed to the overgrowth of Variovorax strains (Fig. 3).

FIG. 4.

A kinetic matrix for the functional grouping of phenol-degrading bacteria constructed on the basis of their affinities to TCE and phenol. Filled squares, Variovorax strains; gray-shaded squares, Acinetobacter strains; open circles, phenol-degrading bacteria reported previously (14).

Analyses of phenol-hydroxylase genes have shown that phylogenetic classification of LmPH corresponds to the kinetic classification of phenol-degrading bacteria, allowing us to develop molecular monitoring tools for different types of phenol-degrading bacteria in microbial communities. From these results, we postulated that the overall genetic features of the largest subunit of phenol hydroxylase are correlated to their kinetic properties (41). The present study revealed, however, that the Variovorax strains possessing low-K_s-type LmPH exhibited very low affinity for phenol (high K_s values), while they showed high affinity for TCE. We anticipate that substitutions of specific amino acids in the low-K_s-type LmPHs resulted in this unique kinetic property. Several specific amino acid residues in the LmPHs of the Variovorax strains are presented in Fig. 3A. Cloning of whole phenol-hydroxylase genes and site-directed mutagenesis analyses will contribute to our understanding of which amino acid residues are most vital for phenol and TCE degradation.

How could the Variovorax population become dominant in the chemostat enrichment despite its low affinity for phenol? Understanding the mechanisms of bacterial community succession may provide valuable information with respect to the control and utilization of microbial communities (37), and some scientists have discussed this point for a variety of microbial communities (4, 8, 9, 23, 24). In our case, it is clear that the ability to degrade phenol (the primary and limiting substrate) is not the sole determinant. Considering this, one interesting finding is that the kinetic index (V_max/K_s, useful to assess catabolic efficiency) for catechol was largely increased during the chemostat enrichment; the increase was larger than the increase in the index for phenol (Table 2), and this was contrary to our expectations, because phenol was provided as the sole carbon and energy source. Following this pattern of increasing catechol-degrading efficiency, the Acinetobacter strain isolated on day 36 (strain c1) showed a higher V_max/K_s value for catechol than Acinetobacter strain c26 isolated on day 10. The pattern of increasing catechol-degrading efficacy was similarly observed in Variovorax strains; Variovorax strain c24 isolated on day 36 and its closely related strains (Variovorax sp. HAB-24 and HAB-30) showed a V_max/K_s value for catechol similar to or higher than that of Variovorax strain YN07 isolated on day 10. Effects of catechol-degrading activity on the chemostat community will be further addressed by using defined mixed cultures of kinetically characterized phenol-degrading bacteria.

In conclusion, the Variovorax strains have been shown to be useful bacteria for TCE bioremediation. Understanding the mechanisms with respect to how they are enriched from natural sources and how they degrade TCE at the molecular level will facilitate our progress in establishing a basis for efficient TCE bioremediation. In addition, more detailed analyses of the chemostat enrichment system used in this study will provide useful information for understanding the mechanisms of succession in microbial communities.