TEXT

Top Abstract Text References

Chloroparaffins are produced by chlorination of alkanes and are classified into three groups according to their chain length: short (C₁₀ to C₁₃), medium (C₁₄ to C₁₇), and long (C_>17) chains. They are widely used as plasticizers in polyvinylchloride, as lubricants, in paints, and in fire retardants. Their total world production is about 300,000 metric tons per year (10). In Germany, 5,000 tons of short-chain chloroparaffins were produced in 1994 and were mainly used in the metal industry (13). This widespread use of chloroparaffins has raised concern and interest in the environmental fate of these water-insoluble compounds which are classified as nonbiodegradable (3).

The degradation of haloalkanes can proceed through different pathways. Haloparaffins (C₁₂ to C₁₈) have been reported to be incorporated into fatty acids in bacteria, yeasts, and fungi (9, 17), resulting in their accumulation in the food chain. Another pathway is the oxygenation at the nonhalogenated end of monohalogenated alkanes by an inherent oxygenase with a tight substrate selectivity (5). In this case fluoroalkanes were defluorinated, but no dehalogenation was observed with chloro-, bromo-, or iodoalkanes. Chain length was reported to have minor effects on this oxygenation reaction. In general, - and ,-chlorinated haloalkanes with short carbon chains (C₁ to C₆) are dehalogenated hydrolytically or by a glutathione-dependent mechanism (6, 26). In contrast, - and ,-haloalkanes with longer chains, e.g., 1,9-dichlorononane and 1,10-dichlorodecane (1,10-DCD), have been proposed to be dehalogenated by oxidative mechanisms (1, 18, 19). Studies on the biodegradation of this class of compounds are rare, because haloalkane-degrading microorganisms are not easily found. Omori and Alexander (18) obtained only two bacterial cultures from 500 enriched cultures growing on 1,9-dichlorononane. One of these organisms was a Pseudomonas species that grew aerobically on C₇ and C₉ chloro-, bromo-, and iodoalkanes. Another Pseudomonas species with an n-alkane-inducible enzyme system dehalogenated these compounds cometabolically when grown on n-alkanes (19). The chlorinated analogues did not induce the oxygenolytic enzyme system, which might explain why it is difficult to enrich for chloroalkane degraders.

We monitored various soil samples from pristine and contaminated sites for their potential to release chloride ions from 1,10-DCD over 5 years. This article described the isolation and characterization of a novel Pseudomonas species that was able to grow in suspensions of 1,10-DCD and other ,-dichloroalkanes.

Garden soil samples (5 g [dry weight]) from five different locations in the area of Stuttgart, Germany, were suspended in 50-ml portions of mineral salts medium. Mineral salts medium contained the following (per liter): 5.37 g of Na₂HPO₄ · 12H₂O, 1.36 g of KH₂PO₄, 0.5 g of (NH₄)₂SO₄, 0.2 g of MgSO₄ · 7H₂O, 0.2 ml of vitamin solution, 1 ml of trace element solution, 0.05% (wt/vol) yeast extract, and 10 mmol of 1,10-DCD. The trace element solution contained the following (per liter): 20.0 g of NaOH, 10 g of MgSO₄ · 7H₂O, 4 g of ZnSO₄ · 7H₂O, 1 g of CuSO₄ · 5H₂O, 3.2 g of MnSO₄ · H₂O, 20 g of Fe₂(SO₄) ·  7H₂O, 100 g of Na₂SO₄, 1 g of NaMoO₄ · 2H₂O, 1 ml of H₂SO₄ (concentrated), and 120 g of EDTA. The vitamin solution contained (per liter) 100 mg of vitamin B₁₂, 20.0 mg of pyridoxal, 100 mg of riboflavin, and 100 mg of thiamine.

Cultures were incubated at 30°C with 10 mM 1,10-DCD as the sole carbon and energy source and shaken at 200 rpm. Samples (1 ml) were withdrawn weekly and analyzed for chloride release with the chlorocounter (Marius Instrumenten, Nieuwegein, The Netherlands). Enriched cultures that had released more than 15 mM chloride were transferred (10% inoculum [vol/vol]) to fresh medium containing 10 mM 1,10-DCD. Of the five different soils incubated with 1,10-DCD as the only available carbon and energy source, chloride release was detected from only one garden soil after 1 month. Dechlorinating activity was maintained during six subsequent transfers with 1,10-DCD as the only growth substrate. Aliquots (100 µl) of the culture fluid were then spread onto solid mineral salts medium with 1,10-DCD as the sole source of carbon and energy. Small, flat, opaque colonies formed within 1 week. When streaked on HNB (5.0 g of yeast extract [Merck], 8.0 g of nutrient broth [Difco Laboratories, Detroit, Mich.], 5.0 g of NaCl [all per liter]) agar plates, uniform colonies were obtained. Colony material appeared homogeneous when examined by phase-contrast microscopy. BOX-PCR (targeted to repetitive intergenic sequence elements of Streptoccus) and ERIC-PCR (targeted to enterobacterial repetitive intergenic consensus sequence elements) (21) performed with whole cells grown under different conditions (HNB, tryptic soy agar, or mineral salts medium with 1,10-DCD) gave identical banding patterns, confirming the purity of the isolate, which was designated strain 273.

Identification of the isolate was initially performed with API 20 NE test strips (bioMérieux, Nürtingen, Germany). Further physiological characterization of the isolate was done by the method of Süßmuth et al. (27). The isolate was a gram-negative, motile rod, 0.8 to 1.5 µm long, with more than one polar flagellum. Strain 273 lysed in 3% potassium hydroxide and was aminopeptidase positive. The isolate grew aerobically and under denitrifying conditions. Nitrate was reduced to nitrite and dinitrogen gas. No fermentative growth was observed under anaerobic conditions. Oxidase and catalase were present. The round sticky colonies were white to yellowish and produced a diffusible fluorescent pigment on HNB agar plates. No polyphosphates or polyhydroxyalkanoates were formed as storage products. Arginine dihydrolase was present. Polyamine analysis (2) revealed 9.1 µmol of spermidine, 6.8 µmol of putrescine, and 1.0 µmol of 1,3-diaminopropane per g (dry weight). Strain 273 grew on D-glucose, gluconate, caprate, adipate, malate, citrate, and phenylacetate. Arabinose, mannose, mannitol, N-acetylglucosamine, and maltose did not support growth. Urease, -glucosidase, gelatin protease, and -galactosidase were absent. Growth was not significantly inhibited by NaCl concentrations up to 2%, and growth occurred between pH 6.0 and 7.8. The optimum temperature for growth was 37°C, but the organism also grew at 41°C. No growth was detected at 4°C.

Near-complete (ca. 1,500-bp) small-subunit rRNA genes from strains 273 and B13 were PCR amplified using eubacterium-specific primers and conditions previously described (34). Excess amplification primers were removed prior to sequencing by using commercially available columns per the manufacturer's instructions (Wizard PCR Preps; Promega, Madison, Wis.). Double-stranded sequencing of PCR products was performed by automated, fluorescent cycle sequencing, using modified versions of previously described eubacterium-specific primers targeted to conserved regions of the 16S rRNA gene (12, 31, 32, 33). Sequences from the closest relatives of strain 273 were identified and obtained from the Ribosome Database Project by using the SIMILARITY_RANK and SUBALIGNMENT programs (14), and related sequences in other databases were identified by BLAST analysis. Sequences were then manually aligned based upon both primary and secondary structure by using the ARB editor (www.biol.chemie.tu-muenchen.de/pub/ARB). The sequences used for phylogenetic analysis are as follows (GenBank accession numbers given in parentheses): Pseudomonas sp. strain 273 (AFO30488), Pseudomonas sp. strain B13 (AFO39489), Pseudomonas aeruginosa LMG1242 (Z76651), Pseudomonas alcaligenes LMG1224 (Z76653), Pseudomonas citronellosis DSM50332 (Z76659), Pseudomonas flavescens NCPPB3063 (U01916), Pseudomonas fluorescens DSM50090 (Z76662), Pseudomonas mendocina LMG1223 (Z76664), Pseudomonas nitroreductans IAM1439 (D84006), Pseudomonas oleovorans DSM1045 (Z76665), Pseudomonas putida DSM291 (Z76667), Pseudomonas stutzeri CCUG11256 (U25432), Pseudomonas sp. strain IpA-1 (X96787), and Pseudomonas sp. strain A3 (Y13246). Natural relationships were inferred by neighbor-joining analysis (24). Phylogenetic analyses placed strain 273 in the gamma subgroup of the Proteobacteria and within the genus Pseudomonas. Its closest relatives were a Pseudomonas species (GenBank accession no. Y13246) (98.9% similarity), P. citronellolis (98.8% similarity), and Pseudomonas sp. strain B13 (98.5% similarity). Strain 273 was more distantly related to P. aeruginosa (94.8% similarity) (Fig. 1). This is consistent with the results of the phenotypic analyses described above and with a preliminary identification done by the Deutsche Sammlung von Mikroorganismen und Zellkulturen (Braunschweig, Germany) based upon the organism's fatty acid profile and partial rDNA sequence.

View larger version (30K): [in this window] [in a new window]   FIG. 1.   Phylogenetic tree of Pseudomonas sp. strain 273 and its closest relatives based on 16S RNA sequence analysis.

1,10-DCD formed a separate phase on top of the aqueous phase due to its very low solubility in water (<100 µg liter¹ [3]). Pseudomonas sp. strain 273 started to grow after 24 h, and the organic phase disappeared. When baffled Erlenmeyer flasks were used, 1,10-DCD formed small droplets in the culture medium when shaken at 250 rpm. The same effect was achieved in regular flasks containing stainless steel spirals (1-cm diameter; 1-mm-thick wire) covering the bottom of the growth vessel. Alternatively, 1,10-DCD was added to the culture medium as a 10% emulsion in medium, which had been sonicated until the suspension was milky. When these techniques were applied, growth and chloride release started after only 12 h. During growth of Pseudomonas sp. strain 273 with 1,10-DCD, stoichiometric amounts of chloride were released (Fig. 2). The pH dropped from 7.4 to 4.7 within 48 h when no pH control was installed. The cultures began to foam when the substrate was exhausted, and the pH dropped below 6.2. At pH 4.7, the culture no longer contained viable cells. Therefore, the pH of the cultures was routinely adjusted to pH 7.4 with 4 M potassium hydroxide before the pH had dropped below 6.2, which also prevented foaming. Strain 273 was able to grow on and release stoichiometric amounts of chloride from a variety of chlorinated aliphatic compounds, including 1,12-dichlorododecane, 1,10-DCD, 1-chlorodecane, 10-chlorodecane-1-ol, 1,9- dichlorononane, 1,8-dichlorooctane, 1,6-dichlorohexane, and 1,5-dichloropentane. Growth on 1,6-dichlorohexane and 1,5-dichloropentane was inhibited when these compounds were present at concentrations above 2 mM. In contrast, growth and dechlorination were not influenced by C₈ to C₁₂ ,-dichloroalkanes, even when present at saturating concentrations. Growth was also observed with the following compounds (concentrations tested given in parentheses): n-heptadecane (5 mM), n-tetradecane (1 mM), n-dodecane (5 mM), n-decane (5 mM), phenylnonane (1 mM), n-octane (5 mM), n-hexane (1 mM), n-pentane (1 mM), 1-eicosanol (1 mM), 1-tetradecanol (1 mM), 1-dodecanol (1 mM), 1-decanol (2 mM), 1-octanol (1 mM), 1-hexanol (1 mM), 1,10-decanediol (2 mM), decanal (5 mM), 2-decanone (5 mM), n-decanoic acid (5 mM), 1,10-decanedioic acid (sebacic acid) (5 mM), 1-decene (2 mM), n-decanoic acid dimethyl ester (dimethyl sebacate) (2 mM), 10-decanediamide (sebacamid) (5 mM), acetate (10 mM), glucose (10 mM), and glycerol (10 mM). No growth was observed within 2 weeks with 1 mM isooctane, cyclohexane, cycloheptane, 1,10-diaminodecane, phenol, benzene, toluene, xylene, and 3-chlorobenzoate. Substances that are solid at ambient temperatures, like eicosane, did not support growth. In resting cell experiments with Pseudomonas sp. strain 273 under anaerobic conditions, no chloride release from 1,10-DCD, 1,9-dichlorononane, or 1,8-dichlorooctane was observed, indicating that oxygen was required for the dechlorinating reactions.

View larger version (19K): [in this window] [in a new window]   FIG. 2.   Batch fermentation of strain 273 in a 10-liter bioreactor with mineral salts medium containing 5 mM 1,10-DCD as the sole carbon source. The consumption of 1,10-DCD (circles), the release of chloride (triangles), and the increase in biomass (squares) was monitored over time. pH was kept constant at 7 by automatic titration with NaOH.

For induction experiments, strain 273 was grown for 48 h in LC medium (7.15 g of Na₂HPO₄ · 2H₂O, 1.33 g of KH₂PO₄, 1.0 g of (NH₄)₂SO₄, 0.2 g of MgSO₄ · 7H₂O, 1 ml of trace element solution, and 1 ml of vitamin solution) containing 10 mM decane or 10 mM 1,10-DCD or in HNB medium. The cells were harvested by centrifugation at 5,000 × g for 20 min, washed once with 25 mM phosphate buffer (pH 7.5), and suspended in 50 ml of either 50 mM phosphate buffer, 50 mM phosphate buffer containing 100 µg of chloramphenicol ml¹ (from an ethanol stock solution of 10 mg ml¹), 50 mM phosphate buffer containing 100 µg of kanamycin ml¹, or LC medium to give a final concentration of 4 × 10⁸ cells ml¹. Each flask contained 10 mM 1,10-DCD and was incubated at 28°C and shaken at 200 rpm. Aliquots were taken every hour for to monitor optical density (A₆₀₅) and protein and chloride concentrations.

Cells grown on 1,10-DCD immediately dehalogenated 1,10-DCD in LC medium with or without chloramphenicol or kanamycin. Cells grown in HNB medium did not dechlorinate 1,10-DCD within 12 h, but longer incubation resulted in release of chloride and growth in the LC medium not amended with chloramphenicol or kanamycin. No dehalogenation was observed in the medium containing the antibiotics. Control experiments confirmed that the antibiotic concentration was high enough to inhibit growth. Cells grown with glucose, glycerol, or acetate dechlorinated 1,10-DCD after a lag phase of 12 h. In contrast, cells grown on decane dechlorinated 1,10-DCD without a lag phase in the presence and absence of chloramphenicol or kanamycin, suggesting that decane induces the dechlorinating enzyme system. The dehalogenation rates, however, were lower than those of cells grown on 1,10-DCD. Dechlorination rates were twice as high in cells grown on 1,10-DCD (25.2 µmol of chloride released h¹ mg of protein¹) than in cells grown on decane (11.3 µmol of chloride released h¹ mg of protein¹). These results show that the 1,10-DCD-dechlorinating enzyme system was not constitutive and that the dechlorinating enzyme system was induced by 1,10-DCD and decane. In the presence of 5 mM decane and 5 mM 1,10-DCD, the dechlorination rate decreased by 60% (4.5 µmol of chloride h¹ mg¹ for decane-grown cells and 10.1 µmol of chloride h¹ mg¹ for 1,10-DCD-grown cells). These results suggest that the enzyme system responsible for the dehalogenation of 1,10-DCD is the same enzyme system that initiates the degradation of decane. These findings were supported by oxygen consumption measurements. Table 1 shows oxygen consumption rates of resting cells grown with decane or 1,10-DCD. The rate of oxygen consumption with decane (12.9 µmol of O₂ h¹ mg¹) was about twofold higher than the rates measured with 1,10-DCD (5.3 µmol of O₂ h¹ mg¹) in cells grown in decane and 1,10-DCD. The oxygen consumption rates are considerably lower than the rates we determined for chloride release. This discrepancy may be explained by the fact that the mixing of the insoluble substrates is less efficient in the oxygen measuring chamber and that the reaction is therefore limited by the transport of the insoluble substrates to the cells. Nevertheless, these results prove that the enzyme system induced by decane dehalogenates 1,10-DCD and that the enzyme system induced by 1,10-DCD oxidizes decane. The most likely explanation is that in both cases the same enzyme system is induced. Cells grown on 10% LB medium did not show any increase of oxygen consumption with decane or 1,10-DCD but did with 1-decanol. The presence of chloramphenicol had no effect on oxygen uptake within the 1-h analysis period.

Since Pseudomonas sp. strain 273 grew well on nonchlorinated alkanes, we tested other known alkane degraders for their ability to grow on 1,10-DCD. The substrate ranges for alkane-degrading organisms differ markedly. Rhodococcus erythropolis Y2 grew with C₇ to C₁₈ n-alkanes and monochloroalkanes, and the best substrate for dechlorination in resting cell experiments was 1-chlorotetradecane and 1-chlorohexadecane (1). Scholtz et al. (25) described an Arthrobacter species which was able to grow with C₄ to C₈ -monochlorinated alkanes, and a hydrolytic dehalogenase was purified from this organism. This enzyme dechlorinated 1-chlorodecane but exhibited no activity toward 1,10-DCD. Pseudomonas sp. strain C12B grew with C₉ to C₁₂ n-alkanes, and optimum growth was observed with C₁₁ n-alkanes, but no dehalogenating activity was reported for this strain (11). P. oleovorans GPo1 grew with C₆ to C₁₂ n-alkanes, and optimum growth occurred on C8 and C9 n-alkanes (15, 29). Omori and Alexander (19) studied Pseudomonas sp. strain B which dehalogenated mono- and di-substituted chloroalkanes when grown on n-undecane. In the latter two cases, the corresponding chlorinated alkanes were cometabolically dehalogenated, but none of them supported growth when supplied as the only growth substrate. P. oleovorans GPo1 harbors the OCT plasmid which encodes alkane monooxygenase. Hybridization experiments with the alk genes from this plasmid with total DNA from Pseudomonas sp. strain 273 revealed no homologous bands. P. oleovorans GPo1 grew well with octane, decane, 1-chlorodecane, 1-chloro-10-decanol, 1-decanol, and decanal. However, this organism did not grow with 1,10-DCD. In the presence of octane as the growth substrate, P. oleovorans GPo1 released chloride from 1,8-dichlorooctane and 1,9-dichlorononane but not from 1,10-DCD. P. oleovorans DSM1045^T did not grow with any of the ,-dichloroalkanes. These results confirm the unique ability of Pseudomonas sp. strain 273 to grow on ,-chlorinated alkanes, an ability not shared by other known alkane degraders.

Cell extracts of Pseudomonas sp. strain 273 dechlorinated 3 mM 1,10-DCD within 2 h (Fig. 3). The addition of glutathione (5 mM) had no effect, whereas the divalent metal chelator EDTA had an inhibitory effect on dechlorination. Only low dechlorination activities were measured with the soluble protein fraction or with the particulate fraction alone. Complete dechlorination of the initial amount of 1,10-DCD was achieved only when both the particulate fraction and the soluble fraction were added. Furthermore, for the complete dechlorination of 5 mM 1,10-DCD, the addition of NADH and Fe²⁺ was required after dialysis of the cell extract. The reaction was strictly oxygen dependent, and no dechlorination was observed under anaerobic conditions. The addition of catalase did not affect dehalogenation, indicating that a peroxidase was not involved in the reaction. In the soluble protein fraction obtained from cells grown on 1,10-DCD, aldehyde dehydrogenase activity was measured (7) (0.31 U mg of protein¹ with decylaldehyde as the substrate), but no alcohol dehydrogenase activity (8) (with 10-chlorodecane-1-ol as the substrate) could be detected.

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Enzymatic conversion of 1,10-DCD (circles) and chloride release (triangles) in cell extracts of strain 273 in the presence (solid symbols) or absence (open symbols) of NADH (5 mM) and Fe2+ (0.2 mM). Data were averaged from two independent experiments.

Several bacteria are known to contain monooxygenases with broad substrate specificity, e.g., octane monooxygenase (28), propane monooxygenase (30), and methane monooxygenase (4). Propane monooxygenase in the presence of oxygen dechlorinated 1-chlorobutane, but 1-chlorobutane was not an inducer for either enzyme. This may be the reason why it is difficult to find dechlorinating activities in microcosms when chloroalkanes are used as the only substrate. For example, P. aeruginosa S7B1, isolated and grown with n-hexadecane, aerobically dechlorinated 1,10-DCD (20). However, attempts by these researchers to culture for dechlorinating organisms with chloroalkanes as the only growth substrate failed. These findings suggest that enzymes involved in the initial catabolic step of n-alkane degradation also play a role in chloroalkane dechlorination. However, these enzymes were induced only by the nonchlorinated alkane and not by the corresponding chloroalkane. Similar data were reported for an oxygenase-type dehalogenase from R. erythropolis Y2 (1). Hexadecane-grown resting cells of R. erythropolis Y2 did dechlorinate 1,10-DCD, but no growth occurred on the ,-chlorinated compound. In contrast, dechlorinating activity in Pseudomonas sp. strain 273 was induced by both 1,10-DCD and decane. The -hydroxylating multienzyme system involved in the hydroxylation of n-alkanes has been characterized from octane-grown cells of P. oleovorans (16, 22, 23). This enzyme system is composed of rubredoxin, NADH-rubredoxin reductase, and the nonheme ion -hydroxylase (a monooxygenase). -Hydroxylase has been described as a relatively insoluble and unstable protein that required the presence of ferrous ion and phosholipids for full activity (22, 23). At this stage, it seems likely that the ,-chloroalkane-dechlorinating enzyme system from strain 273 is a monooxygenase with a composition similar to that of the well-known -hydroxylase.

Nucleotide sequence accession numbers. The 16S ribosomal DNA (rDNA) sequences of Pseudomonas sp. strain 273 and Pseudomonas sp. strain B13 were deposited as GenBank accession numbers AFO39488 and AFO39489, respectively.