Transcriptomic Analysis Reveals a Bifurcated Terephthalate Degradation Pathway in Rhodococcus sp. Strain RHA1

Abstract

Phthalate isomers and their esters are important pollutants whose biodegradation is not well understood. Rhodococcus sp. strain RHA1 is notable for its ability to degrade a wide range of aromatic compounds. RHA1 was previously shown to degrade phthalate (PTH) and to have genes putatively encoding terephthalate (TPA) degradation. Transcriptomic analysis of 8,213 genes indicated that 150 were up-regulated during growth on PTH and that 521 were up-regulated during growth on TPA. Distinct ring cleavage dioxygenase systems were differentially expressed during growth on PTH and TPA. Genes encoding the protocatechuate (PCA) pathway were induced on both substrates, while genes encoding the catechol branch of the PCA pathway were additionally induced only on TPA. Accordingly, protocatechuate-3,4-dioxygenase activity was induced in cells grown on both substrates, while catechol-1,2-dioxygenase activity was induced only in cells grown on TPA. Knockout analysis indicated that pcaL, encoding 3-oxoadipate enol-lactone hydrolase and 4-carboxymuconolactone decarboxylase, was required for growth on both substrates but that pcaB, encoding β-carboxy-cis,cis-muconate lactonizing enzyme, was required for growth on PTH only. These results indicate that PTH is degraded solely via the PCA pathway, whereas TPA is degraded via a bifurcated pathway that additionally includes the catechol branch of the PCA pathway.

Phthalate esters are synthetic compounds used predominantly as additives in plastic to improve the mechanical properties of plastic resin, particularly its flexibility. In order to provide the required flexibility, phthalate ester plasticizers are not bound covalently to plastic resins and are able to migrate into the environment (14, 16). After decades of global industrial use as plasticizers, phthalate esters are now recognized as ubiquitous environmental pollutants detected in every environment in which they have been sought, with the highest concentrations detected adjacent to phthalate ester production or processing facilities (8, 42). There are now serious concerns about the impacts of these compounds on human health and the environment due to wide-ranging adverse effects on animal cells (8, 10, 14, 21, 23, 24, 26, 31).

Phthalate esters are generally considered biodegradable. However, our understanding of these biodegradation processes is extremely limited and largely based on observations and experiments with undefined biological systems (4, 5, 18). Limited studies suggest that the primary mechanism of phthalate ester biodegradation involves hydrolysis (4, 35) to the corresponding parent compound isomers, i.e., phthalate (PTH), terephthalate (TPA), and isophthalate.

Mechanisms for biodegradation of the isomers of PTH have been partially characterized. In particular, several gram-negative bacteria were found to degrade PTH via cis-4,5-dihydroxyphthalate to protocatechuate (1, 6, 32). On the other hand, gram-positive bacteria were found to degrade PTH via 3,4-dihydroxyphthalate to protocatechuate (11, 12, 20, 33). In most cases, protocatechuate from PTH is presumed to be degraded via the β-ketoadipate pathway, involving ortho-cleavage. But in Arthrobacter keyseri 12B, protocatechuate is degraded via both ortho- and meta-cleavage (11). The remaining isomers of PTH, TPA and isophthalate, have been investigated much less thoroughly but were similarly found to be degraded by bacteria, via dihydroxylation (7, 36, 37, 39, 48). For these substrates, protocatechuate also appears to be a common intermediate. Evidence suggests that Rhodococcus sp. strain DK17 degrades protocatechuate from TPA via ortho-cleavage (7) but that Comamonas sp. strain E6 and Comamonas testosteroni T-2 do so via meta-cleavage (36, 37).

Rhodococcus sp. strain RHA1 was originally isolated from γ-hexachlorocyclohexane-contaminated soil and is one of the best-characterized polychlorinated biphenyl degraders (13, 38). The sequence of its 9.7-Mb genome (29) further indicates the ability of RHA1 to degrade a broad range of aromatic compounds. Working in the laboratory of M. Fukuda, W. Kitagawa (25) showed that RHA1 can grow on PTH and TPA and identified two identical, functional copies of the pad genes, encoding PTH degradation proteins, on two large, linear plasmids, pRHL1 and pRHL2. The pad genes encode a ring-hydroxylating 3,4-dioxygenase system (PadAaAbAcAd), a dihydrodiol dehydrogenase (PadB), and a decarboxylase (PadC) as well as a regulatory protein (PadR). Patrauchan et al. (33) found that the pad genes were contained in two nearly identical (>99% nucleotide identity) 32.5-kb regions on the plasmids, which also included pat genes encoding an ABC transport system (PatABCD) and an ester hydrolase (PatE), both of which are putatively involved in PTH or TPA degradation, as well as tpa genes putatively encoding TPA uptake and degradation proteins (Fig. 1). The tpa genes encode a transport protein (TpaK), the large and small subunits plus the reductase of a ring-hydroxylating 1,2-dioxygenase system (TpaAaAbB), a dihydrodiol dehydrogenase (TpaC), and a regulatory protein (TpaR). Interestingly, the nearly identical regions of pRHL1 and pRHL2 are also nearly identical (>99% nucleotide identity) to regions that are duplicated in pDK2 and pDK3 in Rhodococcus sp. strain DK17 (7), which have also been shown to encode PTH and TPA degradation proteins. The tpa genes of Comamonas sp. strain E6 are also duplicated (36). With the exception of the transport proteins, the proteins encoded by the tpa genes of E6 and RHA1 are homologous, with the encoded enzymes sharing 42% to 68% amino acid identity and the regulatory proteins sharing 37% amino acid identity. The order of the tpa genes is different in RHA1 and E6. Proteomic analysis of RHA1 confirmed the role of pad gene products in PTH degradation, and gene knockout analysis showed that further degradation was via the protocatechuate branch of the β-ketoadipate pathway (33).

FIG. 1.

Proposed degradation pathways for phthalate and terephthalate by Rhodococcus sp. strain RHA1 (A) and organization of phthalate and terephthalate degradation gene cluster on pRHL1 and pRHL2 (B). RHA1 has identical versions of this gene cluster on plasmids pRHL1 and pRHL2. DDP, cis-3,4-dihydroxy-3,4-dihydrophthalate; DHP, 3,4-dihydroxyphthalate; DDT, cis-1,2-dihydroxy-1,2-dihydroterephthalate.

In this study, we address the lack of knowledge of TPA biodegradation, particularly for gram-positive bacteria. Transcriptomes of RHA1 grown on PTH and TPA were compared, providing a more comprehensive understanding than was previously possible of the genes and proteins involved in the degradative pathways. The lower (ring cleavage) pathways were distinct for the two compounds. Unexpectedly, a bifurcated pathway for TPA degradation was demonstrated with enzyme assays and gene knockout analysis.

MATERIALS AND METHODS

Strains, plasmids, and cultures.

The strains and plasmids used in this study are listed in Table 1. Rhodococcus sp. strain RHA1 was grown at 30°C in W minimal salt medium (38) containing 20 mM of each carbon source (phthalate or terephthalate) or in Luria-Bertani (LB) medium for genetic manipulation. Cultures were incubated at 30°C with shaking. Phthalate and terephthalate were >99% pure and were purchased from Sigma-Aldrich (St. Louis, MO).

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Relevant characteristics	Reference or source
Strains
Rhodococcus sp. strains
RHA1	Wild type; Cmr	38
RDO5	Derivative of RHA1 cured of pRHL2; biphenyl− Kmr	40
Rha1_005	Mutant derivative of RHA1 with pcaL replaced with Aprr gene	33
Rha1_014	Mutant derivative of RHA1 with pcaB replaced with Aprr gene	This study
E. coli strains
BW25113	K-12 derivative; ΔaraBAD ΔrhaBAD Ampr	19
DH10B	Host for pUZ8002 and mutagenized fosmid	43
BL21(DE3)	hadS gal (λcIts857 indI sam-7 nin-5 lacUV5-T7 gene 1)	19
Plasmids
pKD46	λ-Red (gam bet exo) araC rep101(Ts) Ampr	9
pIJ773	aac(3)1V oriT, source of Aprr cassette	19
pUC-HY	Source of Hygr gene, Ampr	27
pUZ8002	tra neo RP4, Kmr	19
pET21(+)	Expression vector, Ampr T7 promoter	Novagen
RF00129K16	Fosmid clone carrying pcaB, Cmr	This study
RFPCB	RF00129K16 with Cmr replaced with Hygr	This study
RFHPB	RFCPB with Apar replaced with pcaB, Hygr Apar	This study

Transcriptomic analyses.

For transcriptomic analysis, 500-ml cultures were grown to mid-log phase, i.e., an optical density at 600 nm of 2.0 for phthalate or terephthalate and 1.0 for pyruvate, and growth was stopped with 50 ml of 10% phenol, pH 5.0, in ethanol (2). Cells were harvested by centrifugation for 10 min at 7,400 × g at 4°C. The cell pellet was suspended in 0.5 ml Tris-EDTA (TE) buffer plus 1.0 ml RNA Protect (QIAGEN) and incubated for 5 min at room temperature. Cells were centrifuged again as described above, and the cell pellets were stored at −80°C.

Microarray analyses were performed as described previously (15). Briefly, total RNA isolation involved vortexing with glass beads, incubation with hot phenol plus sodium dodecyl sulfate, precipitation of debris with acetate, incubation with phenol plus chloroform, precipitation of nucleic acids with acetate plus isopropanol, DNase treatment, and purification with an RNeasy mini column (QIAGEN). Aliquots of total RNA were stored at −80°C until use. Cy3- and Cy5-labeled cDNAs were prepared by indirect labeling. Equal amounts of Cy3- and Cy5-labeled cDNAs were used for each microarray hybridization. The microarray contained duplicate 70-mer oligonucleotide probes for 8,213 RHA1 genes. The probes were designed and synthesized by Operon Biotechnologies, Inc. (Huntsville, AL), on the basis of the RHA1 genome sequence (29). For each substrate, three independent cultures were used. Differential display hybridizations compared cDNAs from PTH or TPA cultures with those from pyruvate control cultures. One array was used for each pair of cultures compared. Dye swapping was employed to minimize bias (47). The slides were scanned with a GenePix 4000B scanner (Axon Instruments). The spot intensities were quantified using ImaGene 5.6 (Biodiscovery Inc.).

Microarray data were analyzed using the GeneSpring (Silicon Genetics) software package. Only data points that were flagged as present or marginal were included in the analysis. The data were normalized using intensity-dependent Lowess normalization, with 50% of the data used for smoothing. The average normalized expression ratio (treatment/control) was calculated for each gene. Ratios were considered significant if they were >2.0 or <0.5 and Student's t test indicated a P value of <0.05.

Enzyme activity.

Protocatechuate-3,4-dioxygenase activity was measured spectrophotometrically, as described by Stanier and Ingraham (41). The assay mixture contained 50 mM Tris-HCl (pH 8.5), 10 μg of crude cell extract, and 160 μM protocatechuate in a total volume of 1 ml. The reaction was initiated by the addition of substrate, and the decrease in absorbance at 290 nm at 30°C was recorded spectrophotometrically using a Varian Cary 1E spectrophotometer. One unit of enzyme activity was defined as the amount of enzyme that oxidized protocatechuate at an initial rate of 1 μmol per min. Reaction rates were calculated using an extinction coefficient of 2.3 mM⁻¹ cm⁻¹ for the conversion of protocatechuate to β-carboxy-cis,cis-muconate. Specific activity was expressed in units per milligram of protein. Catechol-1,2-dioxygenase activity was measured by the increase in absorbance at 260 nm (3), using the method described above, with the exception of catechol as the substrate instead of protocatechuate. The activity of protocatechuate 4,5-dioxygenase was similarly determined by measuring the increase in absorbance at 410 nm.

Knockout mutagenesis.

The pcaB knockout mutant was constructed using fosmid RF00129K16 (Table 1) and the λRed system, with a slight modification (33). The pcaB gene was replaced using primers PATEfor1 (CCCTGTAATCGGTCACAACGCAGAAAGGTGTCGCGCGTGATTCCGGGG ATCCGTCGACC) and PATErev1 (TTCGGGGTCCGGAGATTTCGTGTGCGAGTGCGACTGTCATGTAGGCTGGAGCTGCTTC).

Nucleotide sequence accession numbers.

The RHA1 genome was submitted to NCBI under accession numbers NC_8268, NC_8269, NC_8270, and NC_8271. Additional data, including files for whole-genome visualization (in Artemis and GBrowse formats), are available at http://www.rhodococcus.ca/.

Microarray accession number.

Details of the microarray design, transcriptomic experimental design, and transcriptomic data have been deposited in the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO Series accession number GSE6685.

RESULTS AND DISCUSSION

Gene expression.

We confirmed that Rhodococcus sp. strain RHA1 can grow on PTH and TPA as sole organic substrates and found that it was unable to grow on the third isomer, isophthalate. Accordingly, the annotated RHA1 genome (29) does not contain isophthalate degradation pathway genes, and a BLAST search revealed no homologs of the gene encoding the isophthalate dioxygenase large subunit (AAX18934). During growth of RHA1 on PTH, the expression of 282 genes differed significantly from their expression in the pyruvate control, among which 203 were up-regulated and 79 were down-regulated. On TPA, the expression of 814 genes differed significantly from their expression in the pyruvate control, among which 574 were up-regulated and 240 were down-regulated. Mainly distinct genes responded to PTH versus TPA, with only 53 being up-regulated and 63 being down-regulated on both substrates. Many of the genes up-regulated on either substrate have functions likely related to PTH or TPA catabolism (Table 2).

TABLE 2.

Selected genes with significant ratios for changes in expression on at least one substrate^a

Functional category	Open reading frame nameb	Gene name	Gene product	Expression ratio		Signal intensity
				PTH	TPA	PYR	PTH	TPA
Phthalate degradation pathway gene cluster	ro08168/ro10207	padR	Transcriptional regulator, IclR family	1.69	−1.08	2,520	4,420	2,180
	ro08167/ro10208	padAa	Phthalate 3,4-dioxygenase alpha subunit	185	2.62	111	32,800	78
	ro08166/ro10209	padAb	Phthalate 3,4-dioxygenase beta subunit	211	2.48	107	32,900	156
	ro08164/ro10211	padB	Phthalate 3,4-dihydrodiol dehydrogenase	52.6	−1.23	222	15,500	123
	ro08163/ro10212	padAc	Phthalate 3,4-dioxygenase ferredoxin subunit	117	2.01	81	13,900	94
	ro08162/ro10213	padAd	Phthalate 3,4-dioxygenase ferredoxin reductase subunit	84.5	2.01	139	16,600	171
	ro08161/ro10214	padC	3,4-Dihydroxyphthalate decarboxylase	19.9	1.01	164	4,000	130
ABC transporter	ro08173/ro10202	patD	Probable ABC transporter, substrate binding component	47.3	4.16	54	3,500	122
	ro08172/ro10203	patA	Probable ABC transporter, ATP-binding component	167	9.54	129	26,500	935
	ro08171/ro10204	patB	Probable ABC transporter, permease component 1	19.5	1.76	551	11,900	914
	ro08170/ro10205	patC	Probable ABC transporter, permease component 2	18.5	3.42	264	7,810	316
Putative ester hydrolase	ro08169/ro10206	patE	Putative phthalate ester hydrolase	11.9	1.74	294	4,750	331
Terephthalate degradation pathway gene cluster	ro08180/ro10195	tpaK	Probable terephthalate transporter, MFS superfamily	5.49	120	185	717	28,100
	ro08179/ro10196	tpaB	Terephthalate 1,2-dioxygenase ferredoxin reductase subunit	4.87	150	82	261	13,600
	ro08176/ro10199	tpaAa	Terephthalate 1,2-dioxygenase alpha subunit	20.5	300	116	1,830	41,200
	ro08175/ro10200	tpaR	Transcriptional regulator, IclR family	2.2	5.25	306	790	1,320
Protocatechuate degradation pathway genes	ro01333	pcaJ	Probable 3-oxoacid coenzyme A transferase beta subunit	9.1	3.58	72	685	242
	ro01334	pcaI	Probable 3-oxoacid coenzyme A transferase alpha subunit	5.92	2.81	92	579	246
	ro01336	pcaG	Protocatechuate dioxygenase alpha subunit	3.33	1.58	416	1,530	599
	ro01338	pcaL	3-Oxoadipate enol-lactone hydrolase/ 4-carboxymuconolactone decarboxylase	4.19	2.78	145	634	388
Catechol degradation pathway genes	ro02371	catC	Muconolactone delta-isomerase	1.54	39.5	24	44	767
	ro02372	catB	Muconate cycloisomerase	1.72	33.7	56	109	1,550
	ro02373	catA	Catechol 1,2-dioxygenase	1.55	18.8	246	425	4,080
	ro02374	catR	Transcriptional regulator, IclR family	1.27	1.15	5,410	7,530	5,350
Other genes, including	ro01857	tcpC	6-Chlorohydroxyquinol-1,2-dioxygenase	7.06	1.22	68	636	55
carboxylase and aromatic	ro02782		Carboxylase	1.38	38.9	184	283	6,510
compound degradation	ro08054	bphB	cis-2,3-Dihydrobiphenyl-2,3-diol dehydrogenase	−1.85	−3.33	470	256	123
genes	ro08055	bphC1	2,3-Dihydroxybiphenyl 1,2-dioxygenase	−2.11	−4.61	487	264	78
	ro08057	bphAd	Biphenyl 2,3-dioxygenase, reductase	−1.71	−3.11	234	169	53
	ro08058	bphAc	Biphenyl 2,3-dioxygenase, ferredoxin component	−1.65	−2.71	454	306	121
	ro08059	bphAb	Biphenyl 2,3-dioxygenase beta subunit	−1.6	−3.19	519	386	107
	ro08060	bphAa	Biphenyl 2,3-dioxygenase alpha subunit	−1.68	−3.72	1,030	634	240
	ro08051/ro10121	bphT1/bphT2	Response regulator, two-component system	1.26	18.8	980	1,610	12,100
	ro08052	bphS1	Sensor kinase, two-component system	1.38	1.22	412	609	459
	ro10122	bphS2	Sensor kinase, two-component system	−1.12	−1.17	2,420	2,500	1,810
	ro10148	bphS3	Sensor kinase, two-component system	1.45	3.28	66	116	176

^aRatios were considered significant if they were >2.0 or <0.5 and Student's t test indicated that the P value was <0.05.

^bTwo open reading frame names indicate identical or nearly identical genes whose expression could not be distinguished by the microarray.

As expected, a putative operon encoding the initial steps of PTH degradation, i.e., padAaAbBAcAdC (Fig. 1), was highly up-regulated, up to 200-fold, on PTH (Table 2). These data are consistent with a previous proteomic analysis (33) in which all of the Pad proteins were highly abundant in PTH-grown cells. The pad genes were not up-regulated during growth on TPA. Another putative operon, tpaAaAbCBK, was highly up-regulated, up to 120- to 300-fold, on TPA and was much less highly up-regulated on PTH (5- to 20-fold). This may reflect a lower specificity of the regulatory system for the tpa genes than for the pad genes. The microarray lacked probes for tpaAb and tpaC, but their up-regulation was inferred by their location in the putative operon. These results confirm the expected role of the tpaAaAbBC genes, encoding the initial steps in TPA degradation, and strongly suggest that tpaK encodes a terephthalate transporter used for uptake. Genes for an ABC transporter (patABCD) and a putative ester hydrolase (patE), located in a single putative operon between the pad and tpa genes, were highly up-regulated on PTH but not on TPA. Our microarray could not distinguish between expression of the two identical copies of the pad, tpa, and pat genes. However, since these copies are on nearly identical 32.5-kb regions of pRHL1 and pRHL2, including up- and downstream regions likely to be involved in the regulation of these genes, it is very likely that duplicate genes are expressed at very similar levels. A separate investigation is under way to examine the functions of the ABC transporter and ester hydrolase.

The pca genes, encoding the complete protocatechuate branch of the β-ketoadipate pathway (Fig. 1), were up-regulated on both PTH and TPA (Table 2). This confirms the previous finding that PTH is degraded via this pathway (33) and indicates, as expected, that TPA is also degraded via this pathway. Surprisingly, the cat genes, encoding the catechol ortho-cleavage branch of the β-ketoadipate pathway, were highly up-regulated (up to 40-fold increase) on TPA but not on PTH. This suggests that TPA, but not PTH, is degraded via a bifurcated pathway, with both protocatechuate and catechol intermediates and with the pathways converging at the common intermediate, β-ketoadipate enol-lactone.

We did not find genes encoding the protocatechuate 4,5-cleavage (meta-cleavage) pathway in the RHA1 genome, nor did we detect protocatechuate 4,5-dioxygenase activity in a crude extract of RHA1 grown on phthalate (not shown). These results indicate that RHA1 degrades protocatechuate derived from PTH via only the β-ketoadipate (ortho-cleavage) pathway, not via both ortho- and meta-cleavage pathways, as is the case for Arthrobacter keyseri 12B (11).

Enzyme activity of ring cleavage dioxygenases.

To further test whether TPA is degraded by RHA1 via both protocatechuate and catechol, protocatechuate 3,4-dioxygenase (PcaHG) and catechol 1,2-dioxygenase (CatA) activities were measured in crude cell extracts. Compared to cells grown on pyruvate, cells grown on PTH had much higher PcaHG activity but only slightly higher CatA activity (Fig. 2). On the other hand, compared to cells grown on pyruvate, cells grown on TPA had substantially higher activities of both PcaHG and CatA. These trends in enzyme activities are consistent with measured expression levels of the corresponding genes and support the hypothesis that TPA is degraded via a bifurcated pathway (Fig. 1).

FIG. 2.

Catechol 1,2- and protocatechuate 3,4-dioxygenase activities in cells grown on pyruvate, phthalate, and terephthalate. Gray bars show protocatechuate 3,4-dioxygenase (PcaHG) activity, and white bars show catechol 1,2-dioxygenase (CatA) activity. Error bars indicate standard deviations for three independent cultures.

The pathway for transformation of TPA to catechol remains uncertain. The simplest route would be decarboxylation of protocatechuate to catechol (Fig. 1), as previously reported for Klebsiella aerogenes (17). A protocatechuate decarboxylase from Clostridium hydroxybenzoicum was characterized, including determination of its N-terminal amino acid sequence (22). However, there were no RHA1 proteins found with substantial similarity to that amino acid sequence. A highly up-regulated gene in TPA-grown RHA1 cells, ro02782, is annotated as a gene encoding a carboxylase but also has similarity to genes encoding decarboxylases (Table 2). The product of this gene might decarboxylate protocatechuate to catechol. The genomic context of this gene gives no further indication of its function or its regulation.

Characterization of pcaL and pcaB knockout mutants.

To confirm the use of the lower β-ketoadipate pathway during growth on both PTH and TPA (Fig. 1), a pcaL knockout mutant (RHA1_005) was tested for growth on each compound. As previously reported (33), RHA1_005 was unable to grow on PTH. RHA1_005 was also unable to grow on TPA, confirming that both compounds are degraded via the lower β-ketoadipate pathway, with β-ketoadipate enol-lactone as a common intermediate (Fig. 1). This further eliminates the possibility that either compound is degraded via meta-cleavage of the protocatechuate intermediate.

To confirm the use of the catechol branch of the β-ketoadipate pathway during growth on TPA but not on PTH, a pcaB knockout mutant (RHA1_014) was constructed and tested for growth on each compound. RHA1_014 was unable to grow on PTH but grew on TPA at a similar rate to that of the wild type (0.052 versus 0.058 doublings/hour, respectively). This confirms that PTH is degraded solely via the protocatechuate branch of the β-ketoadipate pathway but indicates that TPA can be degraded via another pathway. Together, the transcriptomic analysis, enzyme assays, and gene knockout analyses indicate that TPA is simultaneously degraded via both the protocatechuate and catechol branches of the β-ketoadipate pathway. This is the first report of this bifurcated pathway. Evidence that the protocatechuate intermediate of TPH is degraded via ortho-cleavage in Rhodococcus sp. strain DK17 (7) and by meta-cleavage in Comamonas testosteroni T-2 (37) and Comamonas sp. strain E6 (36) does not rule out the possibility of a bifurcated pathway in these or other organisms. Other studies of TPA degradation (39, 48) do not indicate how the protocatechuate intermediate is degraded. Thus, it is currently unclear how broadly representative is the pathway for TPA degradation by RHA1.

Gene regulation.

The distinct expression patterns of the pad and tpa gene clusters indicate that they are independently regulated. Each cluster contains genes encoding putative regulatory proteins, namely, padR and tpaR, respectively. Both genes encode regulatory proteins of the IclR family, based on the presence of a conserved signature region (30). These two regulators have helix-turn-helix domains, which are 42% similar to one another. Each of these regulatory genes is transcribed divergently with the corresponding catabolic gene. Both regulatory genes appeared to be expressed at constitutively high levels on all substrates on the basis of signal intensity (Table 2). The tpaR gene was up-regulated on TPA. The padR gene did not meet our criterion of a twofold difference in expression, but based on a P value of <0.05, it was up-regulated on PTH. Their genomic contexts and expression patterns suggest that padR and tpaR encode regulators for their respective operons, which is consistent with the case for IclR-type regulatory proteins for other aromatic catabolism pathways (28, 34, 46).

Additional regulatory genes were up-regulated on TPA (Table 2), including one or both of bphT1 and bphT2 (too similar to be distinguished by microarray probes), which encode response regulators, and bphS3, which encodes a sensor kinase. The former two genes are linked to genes encoding sensor kinases, i.e., bphS1 and bphS2, which both appeared to be expressed constitutively at high levels. The two-component system encoded by bphS1T1 is known to play a role in the regulation of bph genes involved in biphenyl degradation (44, 45) and of other genes, broadly dispersed throughout the genome and sharing an upstream regulatory motif. Interestingly, the bphA1A2A3A4B1C genes were down-regulated on both PTH and TPA. One or more of the above two-component systems may help to account for the up- and down-regulation of a relatively large number of genes (574) on TPA.

Interestingly, RDO5, an RHA1 derivative lacking pRHL2, grows like the wild type on PTH but does not grow on TPA (data not shown). Since the tpa catabolic genes on pRHL2 are duplicated on pRHL1, some other factor(s) associated with pRHL2 must be essential for growth on TPA. Among the genes carried on pRHL2 are bphS2T2 and bphS3. Thus, it is possible that one or more of these genes and the associated two-component system(s) are essential for the induction of genes involved in TPA catabolism. Further work is required to test this hypothesis and to elucidate the regulation of catabolism of both PTH and TPA.