INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Aromatic-hydrocarbon-degrading bacteria often display remarkable metabolic versatility due to the broad range of substrates of the catabolic enzymes. This low substrate specificity allows bacteria to degrade related molecules through the same catabolic pathway. This is the case for toluene, m-xylene, and p-xylene, which are normally metabolized through the progressive oxidation of a methyl group leading to the formation of a methylbenzoic acid (TOL upper pathway), which is further converted to a (methyl)catechol (9). However, catabolic enzymes have limitations with regard to their substrate range. For instance, o-xylene cannot be degraded through the TOL upper pathway, as ortho-substituted toluene derivatives cannot act as substrates for xylene monooxygenase, the first enzyme of this route (41). A few exceptionsin which o-xylene is converted to 2-methylbenzyl alcohol and then to 2-methylbenzoateare known, but there is no clear evidence that this pathway supports growth (2, 8).

Further metabolic versatility can be achieved through the acquisition of catabolic transposons carrying the genetic information for novel pathways (35, 42). Also, insertion sequences (ISs) can contribute to bacterial metabolic versatility, causing genomic rearrangements that may lead to the expression of genes that are otherwise silent (35).

Pseudomonas stutzeri OX1 is able to grow on toluene or o-xylene as the sole carbon and energy source (3). The first step of the catabolism of these compounds consists of the hydroxylation of the aromatic ring (TOU pathway, for toluene/o-xylene utilization) catalyzed by the enzymatic complex of the toluene/o-xylene monooxygenase (6, 7). m-Xylene and p-xylene are not used for growth by P. stutzeri OX1, but they are cometabolized through toluene/o-xylene monooxygenase to intermediates unproductive for growth (4, 6) (Fig. 1). Nevertheless, from P. stutzeri OX1 cultures, we isolated spontaneous mutants which had acquired the ability to grow on m-xylene and p-xylene through the TOL pathway but which lost the ability to utilize the ortho isomer (4) and, successively, revertants endowed with both catabolic pathways and able to grow on all three isomers of xylene were isolated (11) (Fig. 1).

View larger version (19K): [in this window] [in a new window]   FIG. 1.   Transformations of ortho-, meta-, and para-xylene in P. stutzeri OX1, M1, and R1 via the TOL or the TOU pathway. Solid, dashed, or dotted lines that end with vertical lines rather than arrowheads indicate that the transformation does not occur. P. stutzeri OX1 and R1 grow on o-xylene, and the first catabolic step consists of the monooxygenation of the aromatic nucleus. Both strains cometabolize m-xylene and p-xylene to give the corresponding dimethylphenols, which are not used for further growth. P. stutzeri M1 and R1 grow on m-xylene and p-xylene, whose catabolism proceeds through the progressive oxidation of a methyl group. P. stutzeri M1 is unable to grow on o-xylene and does not produce dimethylphenols from any xylene isomer. None of the three strains is able to oxidize a methyl group of o-xylene. See the text for further details and references.

In this study, we set out to determine the events which switch these catabolic pathways on and off.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and general procedures. The bacterial strains and plasmids used in this work are listed in Table 1. Pseudomonas cultures were routinely grown at 30°C, and Escherichia coli cultures were grown at 37°C in Luria-Bertani medium. For selective medium, ampicillin was added at 100 µg/ml and tetracycline was added at 25 µg/ml. Induction of the lac promoter of plasmids based on pUC19 was performed by the addition of isopropyl--D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM. E. coli cells were transformed with plasmid DNAs by electroporation (12). Plasmid preparations and all the DNA manipulations were carried out in E. coli JM109 by following standard procedures (32).

Southern analyses. Genomic DNA from Pseudomonas cultures was extracted by the cetyltrimethylammonium bromide method and purified by CsCl gradient (32). After digestion, DNA fragments were transferred to Hybond-N nylon membranes (Amersham). The probe was labeled with [-³²P]dATP by using a random-primed DNA labeling kit (Boehringer Mannheim). Hybridization was performed at 42°C, and washing was performed at 65°C.

Cloning strategy. To clone regions of interest from chromosomal DNA, mini-libraries were constructed as follows. After DNA digestion with the suitable enzyme(s), restriction fragments were separated by standard agarose gel electrophoresis. Fragments of the desired length (±200 bp) were identified by using suitable DNA markers, slices were cut from the agarose gel, and the DNA was purified. To verify that the desired fragment was included, 300 ng of each fraction was transferred to a nylon membrane, and a Southern analysis was performed with the appropriate probe. Approximately 1.5 µg of DNA from the appropriate fraction was then used for the ligation with the dephosphorylated cloning vector. The mini-libraries obtained were screened by colony hybridization (32).

Enzyme assays. Xylene monooxygenase activity, responsible for the hydroxylation of a methyl group, was assayed in E. coli DH5 cells. Cultures (100 ml) were grown in M9-salts medium (21) to an optical density at 600 nm (OD₆₀₀) of 0.4 before IPTG induction. The cultures were further incubated to reach an OD₆₀₀ of approximately 0.7. The cells were harvested, washed in phosphate buffer (pH 7), and resuspended in the same buffer to obtain an OD₆₀₀ of 2. Next, 20 mM glucose and 3 µl of p-xylene/ml were added, and the suspensions, in tightly closed containers, were incubated at 37°C. At 30-min intervals, samples were collected. The cells in these samples were eliminated by filtration, and the supernatants were analyzed by reverse-phase high-pressure liquid chromatography using a NovaPack C18 cartridge column eluted with acetonitrile-water at a concentration of 50:50 (flow rate, 1 ml/min). The detector was set to 254 nm, and the amount of 4-methylbenzyl alcohol produced was determined by using a calibration curve.

DNA amplification, nucleotide sequencing, and sequence analysis. The DNA sequence of the P. stutzeri R1 xylWC and touA regions were determined directly from the PCR amplification products. Amplification of the xylWC region was performed with the primers 5'-GGGGGCGGATGAATGCAT-3' and 5'-CCGGCCTCCCATAAGCCC-3'. The touA region was amplified with the primers 5'-GTGGGAAGGGCTACGTCTGA-3' and 5'-CCAATCCCATTTCCTGCTCC-3'. Plasmid templates for DNA sequencing were isolated with purification kits purchased from Macherey-Nagel-Düren or Qiagen. Nucleotide sequences were determined directly from pFC1965, pFC4037, or their derivatives with the Deaza G/A^T7Sequencing Mixes kit, used according to the supplier's instructions (Pharmacia Biotech), and with [-³⁵S]dATP and T7, SP6, or specific synthetic primers. The sequence was analyzed with the Genetics Computer Group (Madison, Wis.). software package (10). The National Center for Biotechnology Information BLASTP program (1) was used to search a nonredundant peptide sequence database.

Nucleotide sequence accession number. The nucleotide sequence of ISPs1 has been submitted to the EMBL data bank under accession no. AJ012352.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Southern analyses of P. stutzeri OX1, M1, and R1 genomes. Any differences among OX1, M1, and R1 genomes with regard to o-xylene or m-xylene and p-xylene catabolic genes were investigated by Southern analyses. The genes involved in the initial steps of o-xylene catabolism (tou) were isolated from a P. stutzeri OX1 genomic library in the pFB3401 cosmid (6). As a probe for the genes involved in m-xylene and p-xylene catabolism, a fragment containing the pWW0 upper pathway genes (xyl) cloned in pED3306 from P. putida PaW1 (28), with which the P. stutzeri OX1 genome was previously demonstrated to have homology (4), was used.

View larger version (47K): [in this window] [in a new window]   FIG. 2.   Southern hybridizations of the P. stutzeri OX1 (lane 2), M1 (lane 3), and R1 (lane 4) genomic DNAs digested with the indicated restriction enzymes and probed with the tou genes (pFB3401) (A) and the pWW0 xyl upper pathway genes (pED3306) (B). Lane 1, pFB3401.

View larger version (21K): [in this window] [in a new window]   FIG. 3.   Schematic representation of tou and xyl gene arrangement in P. stutzeri OX1 (A and B) and M1 (C and D) and of the reactions they are involved in. The bars represent the physical maps of the inserts of pBZ2035 (A), pFC1965 (C), and pFC3095 (D). In pFC3095 and in P. stutzeri OX1 (B), xyl genes were mapped by Southern hybridization. The 3.7-kb EcoRI-SalI DNA fragment from the P. stutzeri OX1 xyl region illustrated in panel B was cloned in pFC4037. Only relevant restriction sites are shown. In panels B and C, the black boxes indicate ISPs1; IRL is close to the AvaI site. In the sequences shown below, the 8-bp direct duplications of the target DNA are underlined (only the coding strands are shown) and 7 bp of the 24-bp ISPs1 inverted repeats are shown in boldface (complementary strand) in panel C. Abbreviations: A, AvaI; B, BstEII; D, DraI; E, EcoRI; H, HindIII; P, PvuII; S, SalI.

Cloning of the upper pathway genes for m-xylene and p-xylene catabolism from P. stutzeri M1. The region homologous to pWW0 xyl genes was cloned from the M1 genome to demonstrate that it actually coded for activities involved in m-xylene and p-xylene catabolism. A map of this region was also needed to precisely locate the additional 3-kb sequence detected in the corresponding OX1 region. A mini-library of approximately 9- to 10-kb EcoRI fragments was obtained from M1 and screened by using the pWW0 xylAM genes as probes, and plasmid pFC3095 was selected (see Materials and Methods for the cloning strategy). Southern hybridization of pFC3095 with pWW0 xylAM, xylBN, and xylC genes (results not shown) made it possible to draw the map depicted in Fig. 3D and revealed a highly conserved gene order with respect to the pWW0 upper operon gene organization (19). By analogy with the pWW0 upper pathway genes, it can be supposed that the cloned genes are transcribed from left to right with respect to the map shown in Fig. 3D. The sequence (not shown) of approximately 800 bp at the left end of the map reported in Fig. 3D revealed 96% overall identity with the pWW0 upper pathway genes. In particular, sequences homologous to the 3' end of the xylW gene of unknown function (38) and to the 5' end of xylC coding for benzaldehyde dehydrogenase (19) were easily recognizable.

1

1

1

1

1

1

Cloning and comparison of the P. stutzeri OX1 and M1 IS. Though hybridization experiments could give only preliminary data about the two unidentified sequences detected in M1 and OX1 genomes, it was observed that both elements showed similar lengths and restriction patterns. In fact, in Southern hybridizations, in both the M1 and OX1 insertion elements no restriction sites were detectable for EcoRI and SalI (Fig. 2A and B) or for BglII, XhoI, and DraI (Southern hybridizations not shown), while both contained at least one HindIII cleavage site (Fig. 2A and B). Based on these data, it was possible to speculate that a similarity between the genetic elements found in the OX1 and M1 genomes existed. To verify this hypothesis, the two elements were cloned from the M1 and OX1 genomes.

View larger version (84K): [in this window] [in a new window]   FIG. 4.   Southern hybridization with the AvaI-BstEII 2-kb internal probe derived from ISPs1 of the P. stutzeri OX1 (lane 1), M1 (lane 2), and R1 (lane 3) genomic DNAs digested with the indicated restriction enzymes.

Sequence analysis of the ISs cloned from P. stutzeri M1 and OX1. A schematic representation of the insertion element cloned from the M1 strain is depicted in Fig. 5. Its nucleotide sequence was determined from a set of subclones obtained from pFC1965. The IS, designated ISPs1, was delimited by 24-bp inverted repeats (IR), which showed a 4-bp mismatch. External to the IS, the direct duplication of an 8-bp AT-rich region belonging to touA (7) was found (Fig. 3C). Based on these features, and despite its atypical length, ISPs1 seems to belong to the ISL3 family, a sparsely distributed class of insertion elements (26).

View larger version (13K): [in this window] [in a new window]   FIG. 5.   Schematic representation and restriction map of ISPs1. The black boxes represent the IRs (not shown to scale). Nucleotide sequences of IRL and IRR are shown at bottom. Mismatches are marked with an asterisk. The arrows indicate the directions of transcription of ORF1 and ORF2. Abbreviations: A, AvaI; B, BstEII; H, HindIII; K, KpnI; P, PvuII; X, XbaI.

Sequence analysis of the tou and xyl regions of P. stutzeri R1. Several transposons are known to be capable of excising precisely and thereby regenerating the target DNA to its original sequence and function (23).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial degradation of methylbenzenes has been extensively investigated. A model for the degradation of m-xylene and p-xylene is represented by the TOL pathway encoded by the xyl genes (9). Less information, especially at the genetic level, is available with regard to o-xylene degradation; however, the direct oxygenation of the aromatic ring seems to be required for the catabolism of this compound (6, 15, 33). Toluene can be degraded via both pathways.

Although TOL strains are widespread in the environment, laboratory competition experiments showed that, under some conditions, the strains which degrade toluene through the TOL pathway are disadvantaged with respect to those which attack this molecule through the direct oxygenation of the aromatic nucleus, regardless of whether it is a mono- or a dioxygenation (13). It is thus possible that, under particular conditions, bacterial strains endowed with the ring oxidation pathway for methylbenzene degradation are competitive, as is also suggested by the number of isolates displaying this metabolic feature (16, 20, 31, 34, 44). Furthermore, depending on the substrate range of the oxygenases involved, the direct oxygenation of the aromatic ring may allow bacteria to use o-xylene (6, 15, 33).

At the phenotypic level, P. stutzeri OX1 shows no analogy with typical TOL strains, but it carries all the genetic information required for the catabolism of m-xylene and p-xylene. In fact, the spontaneous mutant P. stutzeri M1 displays a phenotype undistinguishable from that of several environmental isolates selected for growth on m-xylene and p-xylene and here we demonstrate that even at the genetic level P. stutzeri M1 can be considered a TOL strain, carrying genes coding for activities involved in m-xylene and p-xylene catabolism, highly homologous to and arranged as for the archetypal pWW0 xyl upper pathway genes. Even if the xyl genes are not expressed in the wild-type strain P. stutzeri OX1, their maintenance may confer a selective advantage to the bacterial population. From this perspective, xyl genes represent a reservoir of genetic information and, following changes in environmental conditions, may ensure the survival of the population members in which they are expressed.

In contaminated environments, methylbenzenes are usually present as a mixture. The ability to degrade all the three isomers of xylene may thus represent a selective advantage for bacteria colonizing polluted environments. Although bacterial strains endowed with multiple pathways for toluene degradation are not unknown (20), to the best of our knowledge, only one report of bacterial strains able to grow on the three isomers of xylene has appeared (11). The studies of methylbenzene degradation carried out before now suggest that the three xylene isomers cannot usually be degraded through a unique pathway. This is likely due to the biochemical features of the catabolic enzymes involved. These either have limitations in their range of substrates or lead to the formation of dead-end intermediates (4, 11, 15, 41). Further, bacterial strains such as P. stutzeri R1 (11) that are endowed with both the TOL pathway for m-xylene and p-xylene degradation and the aromatic ring oxidation pathway for o-xylene catabolism do not seem to be widespread. This does not appear to be due to difficulties in recruiting genetic information: several catabolic genes, and among them the xyl genes themselves, are known to reside on transposons which contribute to their diffusion (42). Rather, interference between the two pathways may be responsible for the scarce diffusion of strains endowed with both catabolic routes. In fact, although P. stutzeri R1 is able to grow on m-xylene and p-xylene, it also transforms small amounts of these two isomers into the corresponding dimethylphenols (Fig. 1), which are not used for growth and are toxic to the cells (11). Toluene/o-xylene monooxygenase, which seems to be inducible by each xylene isomer (11), is responsible for this cometabolism (6). Such a strain would be obviously disadvantaged in an environment in which a mixture of the three xylene isomers is present.

At the population level, such interference can be bypassed by switching one of the two catabolic pathways on or off. In P. stutzeri OX1 and its derivatives, the metabolic versatility is brought about by genome rearrangements mediated by a transposable element. ISPs1 can transpose into and precisely excise out of catabolic genes, causing the inactivation or the activation of the corresponding catabolic functions. Consistent with its phenotype, no IS was detected in either the tou or xyl gene cluster in P. stutzeri R1. Although rough, this mechanism contributes to genome plasticity and may provide P. stutzeri with enough variability to allow some members of the population to cope with challenging environmental conditions.

The genomic shock theory proposes that environmental stress stimulates transposon-mediated genomic rearrangements (27). In this way, greater genetic variability would be created in a short time and would subsequently undergo environmental selection. The hypothesis that transposition events are triggered by environmental stress is tantalizing, and further experiments are underway to verify if, in P. stutzeri, transposition frequency is increased under stressful conditions.