Oxidation of Methyl tert-Butyl Ether by Alkane Hydroxylase in Dicyclopropylketone-Induced and n-Octane-Grown Pseudomonas putida GPo1

Abstract

The alkane hydroxylase enzyme system in Pseudomonas putida GPo1 has previously been reported to be unreactive toward the gasoline oxygenate methyl tert-butyl ether (MTBE). We have reexamined this finding by using cells of strain GPo1 grown in rich medium containing dicyclopropylketone (DCPK), a potent gratuitous inducer of alkane hydroxylase activity. Cells grown with DCPK oxidized MTBE and generated stoichiometric quantities of tert-butyl alcohol (TBA). Cells grown in the presence of DCPK also oxidized tert-amyl methyl ether but did not appear to oxidize either TBA, ethyl tert-butyl ether, or tert-amyl alcohol. Evidence linking MTBE oxidation to alkane hydroxylase activity was obtained through several approaches. First, no TBA production from MTBE was observed with cells of strain GPo1 grown on rich medium without DCPK. Second, no TBA production from MTBE was observed in DCPK-treated cells of P. putida GPo12, a strain that lacks the alkane-hydroxylase-encoding OCT plasmid. Third, all n-alkanes that support the growth of strain GPo1 inhibited MTBE oxidation by DCPK-treated cells. Fourth, two non-growth-supporting n-alkanes (propane and n-butane) inhibited MTBE oxidation in a saturable, concentration-dependent process. Fifth, 1,7-octadiyne, a putative mechanism-based inactivator of alkane hydroxylase, fully inhibited TBA production from MTBE. Sixth, MTBE-oxidizing activity was also observed in n-octane-grown cells. Kinetic studies with strain GPo1 grown on n-octane or rich medium with DCPK suggest that MTBE-oxidizing activity may have previously gone undetected in n-octane-grown cells because of the unusually high K_s value (20 to 40 mM) for MTBE.

Methyl tert-butyl ether (MTBE) is widely used as a gasoline oxygenate to reduce automobile emissions of carbon monoxide and compounds involved in photochemical smog formation. Extensive use of MTBE in the United States over the last 25 years has led to widespread contamination of groundwater resources, mainly as a result of leaking underground fuel storage tanks (13, 27, 31). There is considerable interest in MTBE biodegradation, as this may contribute significantly to natural attenuation processes for MTBE and may also provide effective approach for engineered remediation of MTBE contamination.

Biodegradation of MTBE is known to occur under both anaerobic (2, 5, 26, 38) and aerobic (4, 6, 9-12, 16, 17, 22, 24, 25, 30) conditions. Little is known about the organisms, pathways, or intermediates involved in anaerobic MTBE oxidation. Under aerobic conditions, MTBE biodegradation occurs either through growth-related metabolism (9, 11, 22) or through cometabolic processes catalyzed by organisms previously grown on various hydrocarbons (4, 6, 10, 12, 17, 24, 25, 30). Growth-related metabolism of MTBE by pure bacterial cultures can lead to substantial mineralization of this compound. In contrast, cometabolism of MTBE by pure cultures can lead to the substoichiometric accumulation of a variety of partially oxidized intermediates including tert-butyl alcohol (TBA), tert-butyl formate (TBF), 2-methyl-1,2-propanediol, 2-hydroxyisobutyric acid, acetone, formaldehyde, and formate (24, 30).

An important early study describing aerobic cometabolic MTBE degradation showed that this activity is widespread in propane-oxidizing bacteria but is inconsistently observed in bacterial strains grown on selected substrates known to lead to the expression of several well-characterized, nonspecific oxygenases (30). For instance, camphor-grown cells of Pseudomonas putida (ATCC 17453) oxidized low MTBE concentrations (20 ppm [∼200 μM MTBE in solution]). As no MTBE-oxidizing activity was observed in cells grown on rich medium, this MTBE-degrading activity was attributed to the camphor-oxidizing cytochrome P-450_cam found in this organism (30). By the same approach, no MTBE-degrading activity was observed with either toluene-grown Pseudomonas mendocina KR-1 or n-octane-grown P. putida GPo1. From these observations it was concluded that neither toluene toluene-4-monooxygenase (P. mendocina KR-1) nor alkane hydroxylase (P. putida GPo1) is reactive toward MTBE (30). The apparent lack of reactivity of the latter enzyme toward MTBE has remained puzzling for several reasons. First, alkane hydroxylase readily oxidizes a variety of other methoxylated substrates (32), including branched alkyl methyl ethers (14, 15), and it is unclear what structural features might prevent MTBE from being a substrate for this enzyme. Second, several later studies have shown that cometabolic MTBE oxidation is catalyzed by a variety of n-alkane-oxidizing microorganisms (6, 10, 17, 24, 25, 30). The n-alkane growth substrates for these organisms include n-pentane and some of the other shorter-chain (C_≤10) n-alkanes that are abundant in gasoline, the principal source of most MTBE found in the environment. Although none of the monooxygenases responsible for MTBE oxidation in these alkane-oxidizing organisms have been unequivocally identified, alkane hydroxylases and their close homologs appear to be widely distributed, especially among gram-negative organisms that utilize shorter-chain (C₅ to C₁₀) n-alkanes as growth substrates (1, 33, 34, 36).

In this study we have reexamined the potential for MTBE oxidation by P. putida GPo1 with cells grown in the presence of dicyclopropylketone (DCPK). This compound is a potent gratuitous inducer of OCT plasmid-encoded alk genes required for alkane hydroxylase activity and n-alkane metabolism (8, 32). This inducer has been widely used in biocatalytic (32) and physiological (8, 28, 29) studies with this organism when cells with reproducibly high levels of alkane hydroxylase activity are required. The results of this study demonstrate that DCPK-treated cells readily oxidize MTBE and provide several strong lines of evidence linking MTBE oxidation to alkane hydroxylase activity. We have also examined the kinetics of MTBE oxidation and the effects of various growth conditions on the MTBE-oxidizing activity of n-octane-grown cells. Our results suggest that MTBE oxidation by this organism may have previously gone unnoticed because of the unusually high K_s value exhibited by alkane hydroxylase for MTBE.

MATERIALS AND METHODS

Materials.

The bacterial strains used in this study were P. putida GPo1 (ATCC 29347) and P. putida GPo12, a derivative of strain GPo1 cured of the alkane hydroxylase-encoding OCT plasmid (3) (kindly supplied by J. van Bielen, ETH Hönggerberg, Zürich, Switzerland). n-Butane (99% purity), n-decane (99% purity), n-dodecane (99% purity), DCPK (95% purity), ethyl tert-butyl ether (ETBE; 99% purity), formaldehyde (37% aqueous solution), n-heptane (99% purity), n-hexane (>99% purity), MTBE (99.8% purity), n-nonane (99% purity), n-octane (99% purity), tert-amyl methyl ether (TAME; 97% purity), tert-amyl alcohol (TAA; 99% purity), and TBA (99.3% purity) were obtained from Sigma Aldrich Chemical Co. (Milwaukee, Wis.). Acetylacetone (reagent grade) and n-pentane (99.5% purity) were obtained from Fisher Scientific (Pittsburgh, Pa.). Methane and ethane (chemically pure [CP] grade) were supplied by National Specialty Products, (Durham, N.C.). Propane (instrument grade) was supplied by Air Products and Chemicals Inc. (Allentown, Pa.). All other gases for gas chromatography (air, H₂, and N₂) were obtained from local vendors.

Cell growth.

Cells for all of the experiments in this study were grown in batch culture in glass bottles (600 ml) sealed with screw caps fitted with butyl rubber septa (Wheaton Scientific, Millville, N.J.). Most of the cells were grown in 100 ml of Plate Count Medium (PCM; Difco Laboratories, Sparks, Md.) that contained (per liter) dextrose (1.0 g), pancreatic digest of casein (5.0 g), and yeast extract (2.5 g). When required, DCPK was added as an undiluted compound to an initial concentration of 1 mM. The cultures were inoculated (initial optical density at 600 nm [OD₆₀₀], <0.005) with a suspension of cells obtained from axenic cultures previously grown on PCM agar plates. The culture vials were incubated at 30°C in the dark for 24 h in an Innova 4900 (New Brunswick Scientific Co., Inc., Edison, N.J.) environmental shaker operated at 150 rpm. Culture growth was determined by measuring OD₆₀₀ with a Shimadzu 1601 UV/Vis spectrophotometer (Shimadzu, Kyoto, Japan). In every experiment, a sample (50 μl) of each cell culture was streaked onto PCM agar plates to subsequently confirm the purity of the culture.

In experiments with n-octane-grown cells, the cells were grown as described above with various mineral salts media and n-octane (0.05% [vol/vol] liquid phase) as the sole carbon and energy source. The mineral salts media used included the medium recommended by the American Type Culture Collection (ATCC) for cultivation of P. putida GPo1 (ATCC medium 910). This medium contained (per liter) (NH₄)₂HPO₄ (10.0 g), K₂HPO₄ (5.0 g), and Na₂SO₄ (0.5 g). The medium was made with either tap water (as recommended by the ATCC), deionized water, or deionized water containing FeSO₄ (10 μM) added from a stock solution containing Na₂EDTA (50 mM) and FeSO₄ (30 mM). The cells were also grown in another mineral salts medium (G4) that contained (per liter) NH₄NO₃ (0.5 g), MgSO₄ · 7H₂O (0.2 g), CaCl₂ · 2H₂O (0.05 g), Na₂EDTA (0.01 g), FeCl₃ (0.005 g), and a trace element solution (0.01%, vol/vol). This solution contained (per liter); H₃BO₃ (143 mg), MgSO₄ · 7H₂O (102 mg), ZnSO₄ · 7H₂O (32 mg), CoCl₂ · 4H₂O (10 mg), CuSO₄ · 5H₂O (8 mg), and Na₂MoO₄ · 2H₂O (5 mg). The medium was also buffered (pH 7.0) by the addition of 1 M KH₂PO₄-K₂HPO₄ (0.05%, vol/vol). In all cases the cells were grown for 4 days at 30°C in the environmental shaker operated at 150 rpm. In every experiment, a sample (50 μl) of each cell culture was streaked onto PCM agar plates to subsequently confirm the purity of the culture.

Cell harvesting.

In all cases the cells were harvested from their culture medium by centrifugation (10,000 × g, 10 min) and the resulting cell pellet was resuspended in buffer (10 ml of 50 mM sodium phosphate, pH 7.0). The washed cells were sedimented again by centrifugation (as described above), and the resulting cell pellet was finally resuspended with buffer (∼1 ml) to a final protein concentration of 25 to 50 mg of total cell protein ml⁻¹ for cells grown on PCM and 5 to 12 mg of total cell protein ml⁻¹ for cells grown on n-octane. In all cases the cells were stored at 4°C and used within 4 h.

Reaction conditions.

Unless otherwise stated, reactions following the degradation of MTBE and TBA were conducted in glass serum vials (10 ml). The reaction vials were prepared by adding buffer (50 mM sodium phosphate, pH 7.0) (∼900 μl), after which the vials were sealed with butyl rubber stoppers and aluminum crimp seals (Wheaton Scientific). When required, liquid n-alkanes (C₅ to C₁₂), 1,7-octadiyne, ether oxygenates (MTBE, ETBE, and TAME), and tertiary alcohols (TBA and TAME) were added to the sealed vials as undiluted compounds or from aqueous stock solutions with microsyringes. Gaseous n-alkanes (methane, ethane, propane, and butane) were added to the sealed vials with microsyringes, and the excess pressure was released by briefly inserting a syringe needle (2.5 cm, 22 gauge) into the stopper. The final concentration of gaseous n-alkane in the reaction vial was then determined by analysis of the gas phase by gas chromatography (see below). The aqueous phase concentration of each gaseous n-alkane was then calculated with aqueous solubility data for each gas assuming a total gas (air plus n-alkane) pressure of 1 atm. In all experiments the reaction vials were prepared immediately before use and then incubated for 5 min in a shaking water bath (30°C and 150 rpm) to allow equilibration of organic compounds between the gas and liquid phases. The reactions were initiated by addition of an aliquot (100 μl) of a concentrated cell suspension to give a final reaction volume of 1 ml and a total cell protein content of up to 5 mg. The reaction vials were then returned to the shaking water bath and sampled for analysis by gas chromatography in accordance with the demands of each experiment.

Formaldehyde was quantified colorimetrically by the method described by Nash (23). To quantify changes in formaldehyde concentration, samples (500 μl) were removed from the reaction vial and centrifuged in an Eppendorf model 5415C microcentrifuge (14,000 rpm, 2 min) to sediment the cells. Three samples (100 μl each) were taken from the supernatant and diluted to a final volume of 1 ml with water. The diluted sample was then mixed with the test reagent (23) and incubated in the dark at 37°C for 1 h. The concentration of formaldehyde was determined spectrophotometrically (412 nm) with calibration plots developed with standard solutions of formaldehyde (0 to 2 mM). Neither MTBE nor TBA interfered with this assay at concentrations of up to 1 mM.

Determination of kinetic constants.

Kinetic constants (V_max and K_s) were determined for MTBE oxidation with the small-scale reactions described above. The reactions were conducted with cells incubated in a range of initial MTBE concentrations (0 to ∼50 mM dissolved MTBE), and the final concentration of TBA generated after 30 min was determined by gas chromatography (see below). The rate of MTBE oxidation was derived from the final TBA concentration assuming that the rate of oxidation of MTBE remained constant for 30 min (see Fig. 3) and that no further oxidation of TBA occurred during the reaction period (see Fig. 1 and Table 1). The kinetic constants were derived by computer fitting the data by nonlinear regression to a single substrate-binding model {y = V_max ·  [x/(K_s + x)]} with GraphPad Prism version 3.0a for Macintosh (GraphPad Software, San Diego, Calif.).

FIG. 3.

Inhibition of MTBE oxidation by DCPK-treated cells of P. putida GPo1 by gaseous n-alkanes. Cells (4.4 mg of total protein) of strain GPo1 grown on rich medium with DCPK were incubated in glass serum vials (10 ml) in the presence of MTBE (42 μmol) as described in Materials and Methods. Shown is the time course for TBA production from MTBE for cells incubated with MTBE alone (▪) or with MTBE and methane (▵), ethane (▴), propane (○), or n-butane (•). Each gaseous n-alkane was added to an initial gas phase concentration of 50% (vol/vol). (Inset) Cells (3.2 mg of total protein) of strain GPo1 grown on rich medium with DCPK were incubated for 30 min in glass serum vials (10 ml) in the presence of MTBE (42 μmol) and various gas phase concentrations of propane and n-butane as described in Materials and Methods. Shown are the effects of various dissolved concentrations of propane (•) and n-butane (○) on the rate of TBA production. Also shown is the resulting curve when the results for each culture were computer fitted to a single substrate-binding model {y = V_max · [x/(K_s + x)]}. The error bars represent the range of values for two different cultures of cells grown on each substrate.

FIG. 1.

Oxidation of MTBE by DCPK-treated cells of P. putida GPo1. Washed cells (∼18 mg of total protein) of strain GPo1 grown on PCM with DCPK were incubated in phosphate buffer (25 ml) in a glass serum vial (160 ml) sealed with a butyl rubber stopper and an aluminum crimp seal. The vials contained either MTBE (28 μmol; ∼900 μM MTBE in solution) or TBA (28 μmol; ∼1.1 mM TBA in solution). At the indicated times, samples (2 μl) were removed from the reaction medium and analyzed for MTBE consumption and TBA production by gas chromatography and additional samples (500 μl) were also removed for quantification of formaldehyde as described in Materials and Methods. Shown is the time course for MTBE consumption (▪), TBA production (□), and formaldehyde production (•) for cells incubated with MTBE. Also shown is the percentage of MTBE consumed that was detected as TBA for cells incubated with MTBE (○), the residual TBA in reaction mixtures in which cells were incubated with TBA alone (⧫), and the residual TBA in a reaction vial with TBA alone (▴). The data are plotted as the average of two separate cultures. The error bars show the range of the individual data points.

TABLE 1.

Specific rates of ether and alcohol oxidation

Strain	Growth conditions	Sp act (nmol min−1 mg of total protein−1)
		TBA from MTBE	TBA from ETBE	TAA from TAME	TBA consumption	TAA consumption
GPo1	PCM	≤0.01	0.08 (0.15)a	0.18 (0.52)	0.46 (1.7)	≤0.01
GPo1	PCM + DCPK	7.73 (1.8)	0.12 (0.12)	3.13 (0.41)	0.46 (0.41)	≤0.01
GPo12	PCM + DCPK	≤0.01	0.17 (0.11)	0.21 (0.22)	≤0.01	≤0.01

^aAll values in parentheses represent the standard deviation from the reported mean for assays conducted with three separate cultures.

Analytical methods.

In all experiments the concentrations of MTBE and TBA were determined by gas chromatography. Aqueous samples (2 μl) were taken directly from the sealed reaction vials and immediately injected into a Shimadzu GC-8A gas chromatograph fitted with a flame ionization detector and a stainless steel column (0.3 by 183 cm) filled with Porapak Q (60 to 80 mesh; Waters Associates, Framingham, Mass.). The analysis was conducted with a column temperature of 160°C, an injection port temperature of 200°C, and a detector temperature of 220°C. Nitrogen was used as the carrier gas at a flow rate of 15 ml min⁻¹. The gas chromatograph was interfaced with a Hewlett Packard HP3395 (Hewlett Packard, Palo Alto, Calif.) integrator for data collection. Cell protein concentrations were determined with the biuret assay (7) after solubilization of cell material for 30 min at 65°C in 3 N NaOH and sedimentation of insoluble material by centrifugation in an Eppendorf model 5415C microcentrifuge (14,000 rpm, 5 min). Bovine serum albumin was used as the standard. The concentration of MTBE in saturated aqueous solution at room temperature (23°C) was taken as 0.544 M (10). The dimensionless Henry constant for MTBE at 30°C was taken as 0.0255 (21). The aqueous solubilities of propane and n-butane at 1 atm and 30°C were taken as 1.44 and 1.01 mM, respectively (37).

RESULTS

We initially examined whether cells of strain GPo1 grown on PCM in the presence of DCPK (PCM with DCPK) could oxidize MTBE at concentrations (∼1 mM) comparable to those we have used in previous studies of MTBE oxidation by n-alkane-grown bacteria (24, 25). The DCPK-treated cells consumed MTBE without a lag phase, and TBA was the only product of this reaction detected (Fig. 1). The initial rate of MTBE oxidation (0 to 2 h) was estimated at 2.7 nmol min⁻¹ mg of total protein⁻¹ under these conditions. The accumulation of TBA was close to stoichiometric with respect to MTBE consumption (≥92%) throughout the reaction time course, and there was minimal (0.6%: 0.17 μmol) consumption of TBA by cells incubated with TBA alone (Fig. 1).

Previous studies of MTBE oxidation have demonstrated that in addition to TBA other products, including formaldehyde and TBF, can be observed as immediate products of MTBE oxidation (24, 25, 30). We did not observe accumulation of either TBF or formaldehyde in the experiment described in Fig. 1. In separate experiments, we observed complete consumption of formaldehyde when cells grown on PCM with DCPK were incubated with formaldehyde (1 mM) for 1 h in the absence of MTBE. In contrast, ≥95% of the added formaldehyde was recovered from abiotic control incubations over the same time period (data not shown). We also observed that losses of TBF from short-term (1-h) incubations conducted with the same cells incubated with TBF (1 mM) alone could not be differentiated from the small (≤5%) losses observed in abiotic control incubations (data not shown).

Several organisms that have been shown to oxidize MTBE also oxidize TBA (10, 11, 17, 24, 30). In the case of Mycobacterium vaccae JOB5 this activity has been attributed to the same enzyme responsible for initiating MTBE oxidation (24). Our present results (Fig. 1) obtained with strain GPo1 suggest that MTBE oxidation did not appear to involve concurrent oxidation of TBA. However, TBA-oxidizing activity may have been obscured by several factors, including competitive interactions with the residual MTBE in the reaction mixture, as well as a kinetic limitation caused by low TBA concentrations. We subsequently examined the reactivity of DCPK-treated cells toward both MTBE and TBA alone with higher concentrations of these compounds (10 μmol) in short-term (30-min) incubations conducted in smaller glass serum vials (10 ml). Similar reactions were also conducted with ETBE, TAME, and TAA to determine whether the cells were also reactive toward other ether oxygenates (ETBA and TAME) and other dealkylated alcohol products (TAA). Higher specific rates of MTBE oxidation (∼7.7 nmol min⁻¹ mg of total protein⁻¹) were observed under these conditions, and the cells also slowly oxidized TAME to TAA (Table 1). In contrast, equivocal evidence (low rates with equivalent standard deviations) for ETBE, TBA, and TAA oxidation was obtained under these conditions. We also examined whether similar results were obtained with either cells of strain GPo1 grown on PCM without DCPK or with cells of strain GPo12 grown on PCM with DCPK. Strain GPo12 is a derivative of GPo1 cured of the alkane hydroxylase-encoding OCT plasmid. Our results show that neither cells of strain GPo1 grown without DCPK nor cells of strain GPo12 grown on PCM with DCPK oxidized MTBE, TBA, or TAA at detectable rates. Equivocal results were obtained for ETBE and TAME oxidation with these cells (Table 1).

The lack of MTBE-oxidizing activity in DCPK-treated cells of strain GPo12 provides strong evidence for the role of alkane hydroxylase activity in MTBE oxidation. However, these experiments did not eliminate the possibility that another undefined enzyme system that responds to DCPK induction was responsible for this activity. To further test the role of alkane hydroxylase in MTBE oxidation we examined whether n-alkane substrates for this enzyme inhibited MTBE oxidation by cells of strain GPo1 grown on PCM with DCPK. All of the n-alkanes tested (C₅ to C₁₂) inhibited MTBE oxidation to various degrees when present at a fixed initial concentration (0.1% [vol/vol] liquid phase) (Fig. 2). The maximal level of inhibition was caused by n-pentane, while the least effective inhibitor was n-dodecane. We also tested 1,7-octadiyne as an inhibitor of MTBE oxidation. This compound completely (≥99%) inhibited TBA production.

FIG. 2.

Inhibition of MTBE oxidation by DCPK-treated cells of P. putida GPo1. Cells (3.8 to 5 mg of total protein) of strain GPo1 grown on rich medium with DCPK were incubated in glass serum vials (10 ml) in the presence of MTBE (5.5 μmol) for 30 min as described in Materials and Methods. Shown is the average inhibition of TBA production in the presence of each tested n-alkane or 1,7-octadiyne. Each of the n-alkanes and 1,7-octadiyne were added to an initial concentration of 0.1% (vol/vol) of the reaction medium (1 ml). The error bars indicate the standard error of values obtained from a minimum of four replicate incubations conducted with different cell cultures.

To examine whether the effects of n-alkanes described in Fig. 2 were due to competitive interaction between n-alkanes and MTBE, we examined the effects of various concentrations of individual n-alkanes on the MTBE-oxidizing activity of cells of strain GPo1 grown on PCM with DCPK. Rather than using the poorly soluble liquid n-alkanes, we used more water-soluble gaseous n-alkanes as potential inhibitors. We have recently described this approach in a study of MTBE oxidation by P. mendocina KR-1 (25). Like P. mendocina KR-1 (25), neither methane nor ethane had any detectable effect on the rate (27.2 nmol min⁻¹ mg of total protein⁻¹) of TBA production from MTBE (∼35 mM in solution) (Fig. 3). In contrast, both propane and n-butane inhibited TBA production by 52 and 83%, respectively, when added at equimolar gas phase concentrations. A further analysis (Fig. 3 inset) of the inhibitory effects of various gas phase concentrations of propane and n-butane showed that these gases had a saturable inhibitory effect on TBA production. The data fitted a single-substrate-binding model well (r², ≥0.98), and apparent K_i (K_i^app) values of 192 and 37 μM were obtained for propane and n-butane, respectively.

Although our results to this point strongly suggest that MTBE oxidation in DCPK-treated cells was due to alkane hydroxylase activity, none of our experiments reveal why this activity had previously gone undetected. In our subsequent experiments we considered two possible explanations for this. First, we noted that the original report describing the lack of reactivity of strain GPo1 to MTBE did not provide experimental details of how cells were grown on n-alkanes and what media were used (30). We therefore examined the MTBE-oxidizing activity of cells grown on n-octane (0.05%, vol/vol) in a variety of mineral salts media. n-Octane was used as a growth substrate as this n-alkane is the most frequently used substrate in studies of n-alkane-dependent growth of this organism. The media included the mineral salts medium recommended by the ATCC (ATCC medium 910) made with tap water. The same medium was also made with deionized water, as well as with deionized water supplemented with Fe to overcome the reported increased demand for this metal in n-alkane-grown cells (29). We also used another mineral salts medium (G4) that contained a wider variety of micronutrients than ATCC medium 910. The cells were grown for 4 days in each medium containing n-octane and then harvested and tested for MTBE-oxidizing activity in the presence of MTBE (∼35 mM in solution). The results (Table 2) show that, like DCPK-treated cells, n-octane-grown cells also oxidized MTBE to TBA. However, compared to the rate for DCPK-treated cells incubated with the equivalent MTBE concentration (Fig. 3, 27 nmol min⁻¹ mg of total protein⁻¹), the specific rates of TBA production by n-octane-grown cells were considerably lower and varied by more than threefold (3.6 to 12 nmol min⁻¹ mg of total protein⁻¹), depending on the growth medium used during cell growth. The highest specific rate of MTBE oxidation was obtained with cells grown on n-octane-G4 medium.

TABLE 2.

MTBE-oxidizing activity after growth on various media

Growth medium	Final OD600a	MTBE-oxidizing activityb
ATCC 910 (deionized water + Fe)	0.36 (0.067)c	5.64 (0.06)
ATCC 910 (deionized water)	0.41 (0.024)	3.56 (0.41)
ATCC 910 (tap water)	0.39 (0.212)	6.54 (0.59)
G4 (deionized water)	0.63 (0.058)	11.94 (1.05)

^aAll cultures were grown for 4 days.

^bUnits are nanomoles of TBA generated per minute per milligram total protein.

^cAll values in parentheses are the range of values around the reported mean for three separate cultures.

The second potential explanation we considered was that MTBE-oxidizing activity had previously gone unobserved because of a kinetic limitation. The kinetic constants K_s and V_max for MTBE oxidation by strain GPo1 grown on PCM with DCPK or n-octane-G4 medium were determined from rates of TBA production from a range of MTBE concentrations (0 to ∼50 mM) (Fig. 4). In each case the rate of MTBE oxidation to TBA was saturable and fitted a single-substrate-binding model well (r², ≥0.98). V_max estimates of 38.9 (standard error [SE] = 3.1) and 19.7 (SE = 1.67) nmol min⁻¹ mg of total protein⁻¹ and K_s estimates of 18.5 (SE = 3.4) and 39.3 (SE = 5.8) mM were derived from these plots for MTBE oxidation by cells grown on PCM with DCPK and n-octane-G4-medium, respectively.

FIG. 4.

Kinetics of MTBE oxidation by DCPK-treated and n-octane-grown cells of P. putida GPo1. Cells of strain GPo1 were grown on PCM with DCPK and n-octane (0.05%, vol/vol) in G4 mineral salts medium as described in Materials and Methods. Cells (1.1 to 2.1 mg of total protein) were incubated with a range of dissolved MTBE concentrations (0 to ∼50 mM) for 30 min, and the amount of TBA generated was detected by gas chromatography as described in Materials and Methods. Reaction rates for DCPK-treated cells (▪) and n-octane-grown cells (▾) were calculated from the amount of TBA detected after 30 min, assuming that no oxidation of TBA occurred during the reaction. Shown is the resulting curve when the results for each culture were computer fitted to a single substrate-binding model {y = V_max · [x/(K_s + x)]}. The error bars represent the range of values for two different cultures of cells grown on each substrate.

DISCUSSION

In this study we have used the strong inducing effects of DCPK to ensure that cells of strain GPo1 expressed high levels of alkane hydroxylase activity. Although we have not attempted to establish optimal conditions for obtaining cells with maximal MTBE-oxidizing activities, our results clearly show that DCPK-treated cells of strain GPo1 can oxidize MTBE to TBA (Fig. 1 and Table 1). Our conclusion that alkane hydroxylase is responsible for this reaction is supported by our observation that MTBE oxidation generated TBA, a product compatible with a monooxygenase reaction. Furthermore, MTBE oxidation did not occur with cells of strain GPo1 grown on PCM without DCPK (Table 1) or with DCPK-treated cells of strain GPo12 that lack the alkane hydroxylase-encoding OCT plasmid (Table 1). Oxidation of MTBE in cells of strain GPo1 grown on PCM with DCPK was also inhibited by n-alkane substrates for alkane hydroxylase (Fig. 2 and 3) and by 1,7-octadiyne (Fig. 2), a putative mechanism-based inactivator of this enzyme (14, 20). Our conclusion that MTBE oxidation is catalyzed by alkane hydroxylase is also complemented, but not supplanted, by the same activity in n-octane-grown cells under appropriate culture (Table 2) and experimental (Fig. 4) conditions. Further discussion of these lines of evidence and the broader significance of alkane hydroxylase-catalyzed MTBE oxidation is presented below.

Products of MTBE oxidation.

Microbial oxidation of MTBE under aerobic conditions appears to consistently involve monooxygenase-catalyzed dealkylation reactions that are thought to proceed via an unstable hemiacetal intermediate (24). Formaldehyde and TBA are the expected products derived from the subsequent chemical decomposition of this hemiacetal, whereas TBF has been proposed as the biotic product of an alcohol dehydrogenase-catalyzed oxidation of this transient intermediate (24). We did not observe TBF accumulation during MTBE oxidation, and DCPK-treated cells did not consume this compound. We therefore conclude that, unlike other n-alkane grown, MTBE-oxidizing organisms, such as M. vaccae JOB5, TBF is not an obligatory intermediate in the pathway of MTBE degradation in strain GPo1. In contrast, we did observe rapid consumption of formaldehyde. This suggests that formaldehyde production during MTBE oxidation can be obscured by rapid concurrent consumption of this compound. Taken together, these observations suggest that MTBE oxidation by DCPK-treated cells of strain GPo1 involves a conventional monooxygenase-catalyzed O-dealkylation reaction and that the alcohol dehydrogenase activity in both DCPK-treated and n-octane grown cells is unreactive to the putative hemiacetal generated from MTBE.

It is notable that unlike M. vaccae JOB5 and other propane-oxidizing bacteria (24), the activity of alkane hydroxylase is apparently restricted to methoxy groups in MTBE and TAME (Table 1) and does not appear to include the ethoxy group in ETBE or tertiary methyl groups (Table 1). In vivo and in vitro studies (32) have shown that alkane hydroxylase does not oxidize multiple-branched alkanes such as 2,2- and 2,3-dimethylbutane. This lack of activity toward multiple-branched alkanes may explain why neither TAA nor TBA (Fig. 1 and Table 1) appears to be oxidized by this enzyme.

Inhibitors of and substrates for alkane hydroxylase.

Important evidence supporting the role of alkane hydroxylase in MTBE oxidation by DCPK-treated cells is the effect of alkane hydroxylase inhibitors and substrates on this activity (Fig. 2 and 3). For example, 1,7-octadiyne is believed to be a mechanism-based inactivator of alkane hydroxylase (14, 20) and the complete inhibition of TBA production observed with this compound is compatible with this inhibition mechanism. In contrast, the lower and variable inhibition caused by n-alkanes is compatible with competitive interactions at the active site of alkane hydroxylase. In this case the level of inhibition would be dictated by both the enzyme affinity for each substrate and the aqueous concentration of each substrate. While we have no data for the first of these issues, the inhibition pattern does closely follow the aqueous solubility of n-alkanes. For example, the most potent inhibition was caused by n-pentane (saturated solution = 571 μM at 25°C) (18) while n-dodecane (saturated solution = 0.02 μM at 25°C) (18) was the least effective compound tested. An alternative explanation for the ordered inhibitory effects of n-alkanes on MTBE oxidation is that these compounds all exert a toxic effect on DCPK-treated cells and that the degree of toxicity is again directly related to the aqueous solubility of the n-alkanes tested. Again, we have not addressed the possibility of potential toxic effects in this study, However, an interpretation based on toxicity seems unlikely when considered in the light of our findings with various concentrations of n-butane and propane (Fig. 3). These results show that the inhibitory effect of these gases follows the behavior expected of a competitive inhibitor and is proportional to their dissolved concentration at low concentrations and saturable at higher concentrations.

Kinetics of MTBE oxidation.

An important feature characterized in this study is the high K_s value for MTBE (Fig. 4), and we believe that this is likely to be the main reason why this activity was not previously observed in n-alkane-grown cells of strain GPo1 (30). With the extrapolated V_max and K_s values for our most active cells grown on n-octane-G4 medium (Fig. 4), the Henri-Michaelis-Menten Michaelis equation {v = ([S]/K_s + [S]) · V_max} predicts that the expected reaction velocity (v) under the conditions previously tested (∼200 μM MTBE in solution) (30) would be ≤0.1 nmol min⁻¹ mg of total protein⁻¹. Our evidence (Table 2) also indicates that this rate could be more than threefold lower with cells grown under different conditions (Table 2). Although a rate of MTBE oxidation of 0.4 nmol min⁻¹ mg of total protein⁻¹ was previously reported for camphor-grown cells of P. putida (30), rates an order of magnitude lower may be below the detection limit of the analytical approach used.

The K_s values for MTBE that we have determined for strain GPo1 appear to be dependent on whether alkane hydroxylase activity is induced by DCPK or alkanes (Fig. 4). The cause of this variability is unknown and was not further explored in this study. However, irrespective of this variability, the K_s values we have determined in this study are unusually high compared to those of other alkane-oxidizing bacteria that can cometabolically oxidize MTBE. For example, butane-oxidizing strains have K_s values for MTBE as low as ∼25 μM (17) and propane-grown M. vaccae JOB5 has a reported K_s value of 1.36 mM (24). It is also important to note that both of these previous studies involved organisms that, unlike strain GPo1, can further oxidize TBA and appear to oxidize both MTBE and TBA through the activity of the same enzyme. The K_s values for MTBE in these organisms may therefore be overestimates due to concurrent TBA production and oxidation. As alkane hydroxylase in strain GPo1 does not appear to oxidize TBA, the differences between the true K_s values of this organism and gaseous n-alkane-oxidizing bacteria such as M. vaccae for MTBE may be even larger than suggested by the present data.

Similarities to MTBE oxidation by P. mendocina KR-1.

Our present results show many similarities to those we have recently reported for MTBE oxidation by P. mendocina KR-1 (25). These similarities include (i) the overlapping n-alkane growth substrate range of these organisms, (ii) the production of TBA but not TBF during MTBE oxidation, (iii) the apparent lack of further TBA oxidation, and (iv) an unusually high K_s value for MTBE. Additional similarities can also be seen in the inhibitory effects of non-growth-supporting gaseous n-alkanes on MTBE oxidation (Fig. 3, inset). Neither methane nor ethane is oxidized by n-alkane-grown P. mendocina KR-1, and neither gas inhibits MTBE oxidation. In contrast, both propane (K_i = 53 μM) and n-butane (K_i = 16 μM) inhibit MTBE oxidation and both gases are co-oxidized by n-alkane-grown cells of this organism. Like P. mendocina KR-1, gaseous n-alkanes (C₁ to C₄) are not recognized growth substrates for strain GPo1. Despite this, both propane and n-butane strongly inhibited MTBE oxidation by DCPK-treated cells of this organism (Fig. 3). Although we have not presented direct evidence for propane and n-butane oxidation by DCPK-treated cells, it is likely that these gases inhibit MTBE oxidation by acting as alkane hydroxylase substrates. This conclusion is supported by the previous observations that n-octane-grown cells of strain GPo1 hydroxylate both propylene and 1-butene (19) and that butane oxidation is supported by a P. aeruginosa strain (strain 473) that grows on n-octane (35) and has an alkane hydroxylase system closely related to that found in strain GPo1 (34). Assuming that the inhibitory effects of propane and n-butane are the result of these gases acting as competitive alkane hydroxylase substrates, the K_i^app values derived (Fig. 3, inset) for propane (191 μM) and n-butane (37 μM) can be converted to true K_i values as follows: K_i^true = K_i^app/1 + [MTBE]/K_s(MTBE). On the basis of the K_s value for MTBE for cells of strain GPo1 grown on PCM with DCPK (18.5 mM, Fig. 4), the true K_s values for propane and n-butane are 66 and 13 μM, respectively. Bearing in mind the other similarities discussed above, the almost identical K_i values for propane and n-butane in these two organisms provide strong circumstantial evidence that the n-alkane-oxidizing enzyme systems involved are closely related. This possibility is being explored in this laboratory, as is the potential role of alkane hydroxylase in the biodegradation of MTBE in gasoline-impacted environments.