APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1978, p. 512-516

0099-2240/78/0035-0512$02.O0/0 CopyrigL.t i) 1978 American Society for Microbiology

Vol. 35, No. 3 Printed in U.S.A.

Bacterial and Spontaneous Dehalogenation of Organic Compounds T. OMORIt AND M. ALEXANDER* Laboratory of Soil Microbiology, Department of Agronomy, Cornell University, Ithaca, New YorR 14853

Received for publication 6 October 1977

Only 3 of more than 500 soil enrichments contained organisms able to use 1,9dichlorononane as a sole carbon source. One isolate, a strain of Pseudomonas, grew on the compound and released much of the halogen as chloride. Resting cells dehalogenated 1,9-dichlorononane aerobically but not anaerobically. Pseudomonas sp. grew on and resting cells dehalogenated 1,6-dichlorohexane, 1,5-dichloroheptane, 2-bromoheptanoate, and 1-chloro-, 1-bromo-, and 1-iodoheptane, but the bacterium cometabolized but did not grow on 3-chloropropionate. pMethylbenzyl alcohol, chloride, and p-methylbenzoate were formed when resting cells were incubated with a-chloro-p-xylene; the first two products were also formed in the absence of the bacteria. Similarly, o- and m-methylbenzyl alcohols were generated from the corresponding chlorinated xylenes in the presence or absence of Pseudomonas sp. The formation of m- and p-chlorobenzoic acid from m- and p-chlorobenzyl chloride proceeded only in the presence of the cells, but pchlorobenzyl alcohol was generated from p-chlorobenzyl chloride even in the absence of the bacterium. These results are discussed in terms of possible mechanisms of dehalogenation.

Low-molecular-weight petroleum hydrocarbons are used on a large scale for the synthesis of organic chemicals. Once certain substituents are added to the original hydrocarbons, which are characteristically susceptible to microbial attack, the organic molecules frequently become resistant to biodegradation. Sometimes large quantities are discharged into waters and soils, and occasionally deleterious effects arise because of the resistance of these compounds to appreciable enzymatic modification. Although there are many such chemicals, few generalizations exist to explain why these molecules are not subject to rapid degradation or are not attacked at all. Several studies have been concerned with the metabolism of water-soluble a- and ,B-halogenated fatty acids, and information exists on some of the microorganisms involved and the mechanism of enzymatic cleavage of the carbon-halogen bonds (2, 5, 7). A few investigators have focused attention on halogenated aromatic compounds, but the interest in these instances was directed largely to products and pathways of decomposition (6, 8, 11). Nevertheless, generalizations are lacking on the patterns of microbial dehalogenation as related to chemical and physical properties of organic molecules. t Present address: Department of Agricultural Chemistry, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo,

Japan. 512

Because of the lack of information on whether a single enzyme may act on a natural product and a structurally similar but synthetic chemical, a study was undertaken to establish the mechanisms of removal of substituents that might affect biodegradation. For this purpose, halogenated compounds were used because such chemicals are widely used, some are generated as water is chlorinated to destroy pathogens, and a few are known to be environmentally objectionable. MATERIALS AND METHODS Isolation and growth. Bacteria capable of utilizing 1,9-dichlorononane and 1-chlorooctane as their carbon source were obtained from soil by enrichment culture with 1,9-dichlorononane as the organic substrate. The 1,9-dichlorononane-utilizing isolate was

grown in a medium containing 2.0 g of 1,9-dichlorononane or 30 g of glycerol, 1.5 g of KH2PO4, 1.5 g of Na2HPO4, 4.0 g of NH4NO3, 10 mg of MgSO4, 5 mg of FeSO4 7H20, 10 mg of CaSO4 2H20, and 5 mg of yeast extract per liter of deionized water. The same salt solution was used for the enrichment cultures. The final pH was adjusted to 7.0. Incubations were carried out at 30°C on a rotary shaker operating at 160 rpm. To measure bacterial growth, the medium was centrifuged to separate 1,9-dichlorononane from the cells, and after the cells were suspended in a volume of phosphate buffer equal to the original volume, the optical density was determined at 460 nm. Preparation of resting celis. One liter of a liquid culture of Pseudomonas sp. grown on 3% glycerol for

VOL. 35, 1978

DICHLORONONANE DEHALOGENATION

48 h was harvested by centrifugation, and the cells were washed once with 0.067 M phosphate buffer, pH 7.0. The cell paste was then suspended in 300 ml of the same buffer, and 50 ml of the cell suspension was incubated with about 20 mg of substrate for 24 to 48 h on a rotary shaker operating at 300C. The reaction mixtures were centrifuged, and the supernatant fluid was used for assay of chloride or for other chemical procedures. Chemical analysis. After the cells were removed from the reaction mixtures, the supernatant fluids were extracted with equal volumes of ether. The ether phases were dried over Na2SO4 and evaporated in vacuo. The crude crystals were recrystallized from ether-ethanol, and the resulting crystals were subjected to mass spectrometry, gas chromatographymass spectrometry, and infrared analysis. Combined gas chromatography-mass spectrometry was performed with a quadrupole mass spectrometer (Finnigan Instruments Corp., Sunnyvale, Calif., model 3300) coupled to a gas chromatograph (Finnigan model 9500). The operating temperature of the gas chromatograph was either 100 or 1400C for the column. A stainless-steel column (180 cm by 3 mm) was packed with 5% SE30 on 80/100 Chromosorb 6W (H.P.). The carrier gas, N2, was passed through the column at a flow rate of 40 ml/min. Chloride release was determined by the technique of Bergmann and Sanik (1). Chemicals. n-Undecane, 1,9-dichlorononane, 1,9dibromononane, 1,6-dichlorohexane, 1,5-dichloropentane, 1-chlorooctane, and halogenated heptanes, xylenes, and benzyl chlorides were obtained from Aldrich Chemical Co., Milwaukee, Wis.; 3-chloropropionic acid, 6-bromohexanoic acid, iodoacetic acid, 3-iodopropionic acid, and chloroacetic acid were provided by Eastman Organic Chemicals, Rochester, N.Y.; and 2bromohexanoic acid and 7-bromoheptanoic acid were from K & K Laboratories, Plainveiw, N.Y.

RESULTS More than 500 enrichment cultures were established to obtain isolates able to grow on 1,9dichlorononane and 1-chlorooctane, but only three bacteria were obtained. Two of these could grow on 1,9-dichlorononane, and one used 1chlorooctane as a sole carbon source. One of the 1,9-dichlorononane-utilizing organisms was studied further. This soil bacterium was identified as a strain of Pseudomonas on the basis of its morphological and biochemical characteristics. It is a short, gram-negative, motile rod with two polar flagella. The organism hydrolyzed gelatin and gave positive reactions for catalase and cytochrome oxidase. Indole production, the methyl red test, and nitrate reduction were negative. H2S was not produced on triple sugar iron agar. Acids were formed from glucose and sucrose but not from lactose. During glucose fermentation, gas was not produced. The bacterium grew rapidly in media containing 1,9-dichlorononane or glycerol as the carbon source. The dehalogenating activity of Pseudomonas

513

sp. during its growth on 1,9-dichlorononane was investigated, using a culture grown in a 2,000-ml flask containing 500 ml of medium. As shown in Fig. 1, chlorine release was parallel to the increase in bacterial density, and dehalogenation ceased when growth stopped. At that time, about 40% of the chlorine was cleaved. The pH fell as halide was removed from the hydrocarbon, and this increase in acidity may have been responsible for the cessation in growth and dehalogenation. Therefore, CaCO3 was added to the liquid medium to maintain the pH values, and 1.0 g of 1,9-dichlorononane was also added. Under such circumstances, dechlorination was maintained for much longer periods (Fig. 2). The amount of chloride released at 72 and 120 h was 600 and 1,000 itg/ml, respectively. The substrates dehalogenated by Pseudomonas sp. were then investigated by two methods. (i) The cells were tested for growth on agar slants (containing the inorganic salts) on which a drop of a chlorinated compound was placed (so the organisms were growing on the liquid or the vapor); (ii) dehalogenation by resting cells derived from glycerol-grown cultures was tested. Eight of the chemicals supported abundant growth as measured visually (Table 1). All but one of the 16 chemicals were dehalogenated by resting cells, and halogen was even cleaved from compounds that would not support growth; the

E

11-1.

8 cpb

=k

HOURS

FIG. 1. Growth of Pseudomonas sp. on 1,9-dichlorononane and the liberation of chloride.

APPL. ENVIRON. MICROBIOL.

OMORI AND ALEXANDER

514

w-halogenated fatty acids, and aromatic compounds with halogenated alkyl side chains were attacked by the resting cells with the release of the free halide. Chloride was not released, however, from 1,9-dichlorononane under anaerobic conditions; the anaerobic breakdown of only this compound was tested. A study was conducted of halogen cleavage by resting cells of Pseudomonas-sp. grown on glycerol for 24 h. The test compounds were various halogenated alkanes and fatty acids. The extent of dehalogenation in 24 h of the alkanes was governed by the identity of the halogen substituent (Table 2). Also, the w-substituted fatty acids were readily dehalogenated. It is of interest that the extent of microbial cleavage of the halogen substituents in 24 h is not correlated with the carbon-halogen bond energy (9). The products formed in the presence of the resting cells of Pseudomonas sp. were investigated with mass spectrometry, gas chromatography-mass spectrometry, and infrared analysis. The cells were grown on glycerol, and the resting cells were incubated for 48 h with 30 mg of each FIG. 2. Dehalogenation of 1,9-dichlorononane by substrate. Under these conditions, a-chloro-pgrowing cultures of Pseudomonas sp. in the presence xylene was found to be converted to p-methylof CaCO3. benzyl alcohol, the identity of which was determined by infrared spectrometry and combined TABLE 1. Growth on halogenated compounds and gas chromatography-mass spectrometry. The intheir dehalogenation by resting cell suspensions of frared spectrum of the product showed absorpPseudomonas sp. tion at 3,400 and 800 cm-'. Combined gas chroChloride matography-mass spectrometry showed a moreleased Growth agar on (ug/ml per lecular ion peak at m/e 122 and fragmentation Substrate peaks at m/e 107, 93, 91, 79, and 77. These 48 h) peaks were identical to data obfragmentation 303 1,9-Dichlorononane ............. + authentic with p-methylbenzyl alcohol. tained 6 + .............. 1,6-Dichlorohexane was anfrom a-chloro-p-xylene product Another 4 1,5-Dichloropentane ............. + alyzed by gas-liquid chromatography, using 10% 104 1,9-Dibromononane ............. + 30 diethylene glycol adipate and 2% phosphoric 1-Chloroheptane ................ + 35 acid on 80/100 Chromosorb W at 180°C. This 1-Bromoheptane ................ + 460 1-lodoheptane ................. + compound showed the same retention time as 10 2-Bromoheptanoic acid .......... + that of authentic p-methylbenzoic acid. Simi258 7-Bromoheptanoic acid ........ ±..a larly, a-chloro-m-xylene was observed to be 220 3-Chloropropionic acid .......... transformed to m-methylbenzyl alcohol, the 0 Trichloroacetic acid ............. identity of which was demonstrated by gas chro+b a-Chloro-p-xylene ................ + a-Chloro-m-xylene .............. TABLE 2. Halogen release from test compounds by + a-Chloro-o-xylene ............... Pseudomonas sp. + p-Chlorobenzyl chloride ......... I

ig

+ m-Chlorobenzyl chloride ......... aSlight growth. b Chloride was released, but 'the amount was not determined.

latter thus appear to be subject to cometabolism. Not only were chlorinated compounds dehalogenated, but also bromide and iodide were released from alkanes and fatty acids. Furthermore, mono- and dihalogenated alkanes, a- and

Substrate

released Halide per 24 h) (pg/ml

260 1,9-Dichlorononane ............... 100 1,9-Dibromononane ............... 1-Chloroheptane .......... ..... 20 1-Bromoheptane .......... ..... 30 340 ....... 1-lodoheptane ........

10 2-Bromoheptanoic acid ............... 7-Bromoheptanoic acid ............... 250 3-Chloropropionic acid ............... 220 3-Iodopropionic acid ............... 200

VOL. 35, 1978

DICHLORONONANE DEHALOGENATION

matography-mass spectrometry. A molecular ion peak was evident at m/e 122, and fragmentation peaks were at m/e 107, 93, 81, 79, and 77; identical peaks were found with authentic mmethylbenzyl alcohol. The mass spectrum of the product from a-chloro-o-xylene showed a molecular ion peak at m/e 122 and prominent fragmentation peaks at m/e 104, 93, 81, 79, and 77; these peaks were identical to those published for authentic o-methylbenzyl alcohol (10). In addition, p-chlorobenzyl chloride was converted top-chlorobenzyl alcohol, the latter being identified by combined gas chromatographymass spectrometry. The mass spectrum exhibited a molecular ion peak at m/e 142 and intensive fragmentation peaks at m/e 107, 79, and 77. The p-chlorobenzoic acid formed from p-chlorobenzyl chloride was identified by infrared analysis and gas chromatography-mass spectrometry. The mass spectrum of the product showed a molecular ion peak at m/e 156 and prominent fragmentation peaks at 141, 139, 104, and 101. The infrared spectrum and gas chromatographic and mass spectral characteristics of this compound were identical to those of authentic p-chlorobenzoic acid (10). Finally, mass spectrometry and infrared spectrometry showed that m-chlorobenzoic acid was generated from m-chlorobenzyl chloride. The infrared spectrum of the product from m-chlorobenzyl chloride was identical to that of authentic m-chlorobenzoic acid, and the mass spectrum showed a molecular ion peak at m/e 156 and fragmentation peaks at m/e 141, 139, 103, and 101, in agreement with published data for this compound (10). p-Methylbenzyl alcohol, m-methylbenzyl alcohol, o-methylbenzyl alcohol, andp-chlorobenzyl alcohol, which were formed from a-chloro-pxylene, a-chloro-m-xylene, a-chloro-o-xylene, and p-chlorobenzyl chloride, respectively, were also generated nonenzymatically, i.e., in controls incubated under the same conditions but without resting cells. Hence, it was not possible to determine whether these compounds were made by biological or nonenzymatic reactions. However, p-methylbenzoic acid, p-chlorobenzoic acid, and m-chlorobenzoic acid, which were produced from a-chloro-p-xylene, p-chlorobenzyl chloride, and m-chlorobenzyl chloride, respectively, were formed only in the presence of resting cells. These products and their precursors are shown in Fig. 3. Several other halogenated chemicals were found to be susceptible to spontaneous dehalogenation. In these tests, various compounds were incubated at 30°C in 250-ml flasks containing 30 mg of the organic compound in 50 ml of phos-

515

phate buffer. The data provide clear evidence of nonbiological dehalogenation of both aromatic and aliphatic molecules (Table 3).

DISCUSSION It has been reported that a-substituted aliphatic acids are generally more prone to dehalogenation than are,B-substituted compounds (5, 7). Moreover, a strain of Paracoccus denitrificans was found to be specific for the dehalogenation of ,8-substituted fatty acids (2). By contrast, the results of this investigation provide evidence that Pseudomonas sp. actively removed halide from w-halogenated alkanoic acids and alkanes having halogens on one or both terminal carbon atoms. The present report also shows a more broad substrate specificity for a dehalogenating isolate than has been previously reported (2, 5, 7). SUBSTRATE

PRODUCT

|H37@COOH cS CH3

CH20H

CH2CI

Ck~~CH2CI

CC@OOH OH

clK

Cl

FIG. 3. Products formed from chlorinated organic compounds incubated in the presence of resting cells of Pseudomonas sp.

TABLE 3. Spontaneous dehalogenation in phosphate buffer Halideper release Compound 48 h)

(pug/ml

o-, m-, p-Chlorobenzyl chloride ....

390

a-Chloro-p-xylene .......... ........... 200 2-Bromoheptanoic acid ................ 50 7-Bromoheptanoic acid ................