DNA from 32 2,4-dichlorophenoxyacetic acid (2,4-D)-degrading bacteria from diverse locations was probed with the first three genes of the well-known 2,4-D degradation pathway found in Alcaligenes eutrophus JMP134(pJP4). The majority of strains did not show high levels of homology to the first three genes of the 2,4-D degradation pathway, tfdA,-B, and-C. Most strains showed combinations of tfdA-, Band nd C-like elements that exhibited various degrees of homology to the gene probes. Strains having the same genomic fingerprints (as determined by repetitive extragenic palindromic PCR) exhibited the same hybridization pattern regardless of the geographic origin of the strain, with the exception of a strain isolated from Puerto Rico. This strain had the same genomic fingerprint as that of numerous other strains in the collection but differed in its hybridization against the tfdA gene probe. Members of the ␤ subdivision of the Proteobacteria class, specifically Alcaligenes, Burkholderia, and Rhodoferax species, carried DNA fragments with 60% or more sequence similarity to tfdA of pJP4, and most carried fragments showing at least 60% homology to tfdB. However, many strains did not hybridize with tfdC, although they exhibited chlorocatechol dioxygenase activity. Members of the ␣ subdivision of the Proteobacteria class, mostly of the genus Sphingomonas, did not hybridize to either tfdA or tfdC, but some hybridized at low stringency to tfdB. The data suggest that extensive interspecies transfer of a variety of homologous degradative genes has been involved in the evolution of 2,4-D-degrading bacteria.

Muconate cycloisomerase (EC 5.5.1.1) and chloromuconate cycloisomerase (EC 5.5.1.7) were purified from extracts of Rhodococcus erythropolis 1CP cells grown with benzoate or 4-chlorophenol, respectively. Both enzymes discriminated between the two possible directions of 2-chloro-cis,cis-muconate cycloisomerization and converted this substrate to 5-chloromuconolactone as the only product. In contrast to chloromuconate cycloisomerases of gram-negative bacteria, the corresponding R. erythropolis enzyme is unable to catalyze elimination of chloride from (؉)-5-chloromuconolactone. Moreover, in being unable to convert (؉)-2-chloromuconolactone, the two cycloisomerases of R. erythropolis 1CP differ significantly from the known muconate and chloromuconate cycloisomerases of gram-negative strains. The catalytic properties indicate that efficient cycloisomerization of 3-chloro-and 2,4-dichloro-cis,cis-muconate might have evolved independently among gram-positive and gram-negative bacteria.Many chloroaromatic compounds are degraded by bacteria via chlorocatechols as central intermediates. Further catabolism involves ortho-cleavage of the chlorocatechols to chlorosubstituted cis,cis-muconates as well as cycloisomerization and dechlorination of the latter, yielding dienelactones (4-carboxymethylenebut-2-en-4-olides) which are hydrolyzed and finally funneled into the ubiquitous 3-oxoadipate pathway (Fig. 1). Despite much of the early work having been done with an Arthrobacter sp. (3,8,31) and despite many reports of transformation of halogenated aromatic compounds by gram-positive bacteria (recently reviewed in reference 35), the enzymology and genetics of the modified ortho-cleavage pathway outlined above have been elucidated almost exclusively in gram-negative strains. They usually contain separate sets of enzymes for catechol and chlorocatechol conversion, which differ from each other with respect to the affinities and turnover rates for chlorosubstituted catechols or the metabolites formed from them (6,21,25).The gram-positive strain Rhodococcus erythropolis 1CP has previously been reported to utilize 4-chlorophenol and 2,4-dichlorophenol as sole sources of carbon and energy (10). After some adaptation, it also grows slowly with 3-chlorophenol but not with 2-chlorophenol. Like many gram-negative strains, R. erythropolis 1CP possesses separate catechol and chlorocatechol catabolic enzymes (14, 16). The substrate preferences of the chlorocatechol 1,2-dioxygenase (15) and of the dienelactone hydrolase (16) of R. erythropolis 1CP suggest that, corresponding to the growth substrates, only a 4-chlorocatechol branch and a 3,5-dichlorocatechol branch are functional in strain 1CP, but there is no 3-chlorocatechol branch (Fig. 1). In this paper, we show that the substrate preference of the R. erythropolis 1CP chloromuconate cycloisomerase fits well with those of the dioxygenase and of the hydrolase. Moreover, 2-chloro-cis,cis-muconate was found to be converted to only one product, 5-chloromuconolactone, by both the muconate and the ...

Chlorocatechol 1,2-dioxygenase from Rhodococcus erythropolis 1 CP was purified to homogeneity. In contrast to chlorocatechol 1,2-dioxygenase from Gram-negative strains which have a very broad substrate tolerance, the Rhodococcus enzyme was relatively more specific and had a distinct preference for 4-substituted catechols. Protein and metal analysis indicate an unusual stoichiometry of one atom each of iron and manganese/64-kDa homodimer. The N-terminal amino acid sequence (27 residues) of the Rhodococcus chlorocatechol 1,2-dioxygenase was determined and exhibited 15 -22% identity to the published sequences of catechol 1,2-dioxygenases and other chlorocatechol 1,2-dioxygenases. Antiserum was raised in rabbits and antibodies against Rhodococcus chlorocatechol 1,2-dioxygenase were affinity purified. Dot-blot analysis revealed a very weak reaction between the antibodies and partially purified chlorocatechol 1,2-dioxygenases from Alcaligenes eutrophus JMPl34 and Pseudomonas putida 87. No reaction between these antibodies and above enzymes was observed using Western blotting. Kinetic and immunochemical data as well as comparison of subunit molecular mass and suggest that the Rhodococcus enzyme differs significantly from the known highly similar chlorocatechol 1,2-dioxygenases of Gram-negative strains and seems to be only distantly related to them.The ability to catalyze transformation of different monochloroaromatic compounds (chlorophenols, chloroanilines and chlorobenzoates) to the corresponding chlorinated catechols has been shown to be quite common among the representatives of the nocardiafoxm genus Rhodococcus (Golovlev and Eroshina, 1982;

Three 2,4-dichlorophenoxyacetic acid (2,4-D)-degrading bacterial isolates were obtained from the highly saline and alkaline Alkali Lake site in southwestern Oregon contaminated with 2,4-D production wastes. While similar in most respects, the three isolates differed significantly in 2,4-D degradation rates, with the most active strain, 1-18, demonstrating an ability to degrade up to 3000 mg 2,4-D I-1 in 3 d. This strain was well adapted to the extreme environment from which it was isolated, growing optimally on 2,4-D at pH 8.4-9.4 and at sodium ion concentrations of 0.6-1.0 M. According to its optimum salt concentration and pH for growth, this isolate was a moderately halophilic, alkaliphilic bacterium. The 16S RNA gene sequence (303 nt) was identical for all three isolates and most closely resembled those of the moderately halophilic eubacteria of the family Halomonadaceae (91% identity). Biochemical and genetic examination revealed strain 1-18 utilizes the same 2,4-D degradation pathway as most of the 2,4-D-degrading bacteria from non-extreme environments. Hybridization data and comparison of the partial sequences of the tfdA gene from the Alkali Lake isolates with those of bacteria from non-extreme environments suggested a common genetic origin of the 2,4-D degradation pathway in the two groups of micro-organisms.

Burkholderia xenovorans strain LB400, which possesses the biphenyl pathway, was engineered to contain the oxygenolytic ortho dehalogenation (ohb) operon, allowing it to grow on 2-chlorobenzoate and to completely mineralize 2-chlorobiphenyl. A two-stage anaerobic/aerobic biotreatment process for Aroclor 1242-contaminated sediment was simulated, and the degradation activities and genetic stabilities of LB400(ohb) and the previously constructed strain RHA1(fcb), capable of growth on 4-chlorobenzoate, were monitored during the aerobic phase. The population dynamics of both strains were also followed by selective plating and real-time PCR, with comparable results; populations of both recombinants increased in the contaminated sediment. Inoculation at different cell densities (10 4 or 10 6 cells g ؊1 sediment) did not affect the extent of polychlorinated biphenyl (PCB) biodegradation. After 30 days, PCB removal rates for high and low inoculation densities were 57% and 54%, respectively, during the aerobic phase.