sal Genes Determining the Catabolism of Salicylate Esters Are Part of a Supraoperonic Cluster of Catabolic Genes in Acinetobacter sp. Strain ADP1

Abstract

A 5-kbp region upstream of the are-ben-cat genes was cloned from Acinetobacter sp. strain ADP1, extending the supraoperonic cluster of catabolic genes to 30 kbp. Four open reading frames, salA, salR, salE, and salD, were identified from the nucleotide sequence. Reverse transcription-PCR studies suggested that these open reading frames are organized into two convergent transcription units, salAR and salDE. The salE gene, encoding a protein of 239 residues, was ligated into expression vector pET5a. Its product, SalE, was shown to have esterase activity against short-chain alkyl esters of 4-nitrophenol but was also able to hydrolyze ethyl salicylate to ethanol and salicylic acid. A mutant of ADP1 with a Km^r cassette introduced into salE had lost the ability to utilize only ethyl and methyl salicylates of the esters tested as sole carbon sources, and no esterase activity against ethyl salicylate could be detected in cell extracts. SalE was induced during growth on ethyl salicylate but not during growth on salicylate itself. salD encoded a protein of undetermined function with homologies to the Escherichia coli FadL membrane protein, which is involved in facilitating fatty acid transport, and a number of other proteins detected during aromatic catabolism, which may also function in hydrocarbon transport or uptake processes. A Km^r cassette insertion in salD deleteriously affected cell growth and viability. The salA and salR gene products closely resemble two Pseudomonas proteins, NahG and NahR, respectively encoding salicylate hydroxylase and the LysR family regulator of both salicylate and naphthalene catabolism. salA was cloned into pUC18 together with salR and salE, and its gene product showed salicylate-inducible hydroxylase activity against a range of substituted salicylates, with the same relative specific activities as found in wild-type ADP1 grown on salicylate. Mutations involving insertion of Km^r cassettes into salA and salR eliminated expression of salicylate hydroxylase activity and the ability to grow on either salicylate or ethyl salicylate. Studies of mutants with disruptions of genes of the β-ketoadipate pathway with or without an additional salE mutation confirmed that ethyl salicylate and salicylate were channeled into the β-ketoadipate pathway at the level of catechol and thence dissimilated by the cat gene products. SalR appeared to regulate expression of salA but not salE.

Acinetobacter sp. strain ADP1 is capable of utilizing a range of aromatic compounds as sole sources of carbon. These compounds are dissimilated via the β-ketoadipate pathway either through benzoate and catechol or, alternatively, through 4-hydroxybenzoate and 3,4-dihydroxybenzoate (protocatechuate) (15, 22). Although the two branches have three terminal reactions in common, from β-ketoadipate enol lactone to the coenzyme A esters of succinate and acetate, each branch has its own genes and enzymes. The genes for the two branches are located in two supraoperonic clusters, the ben-cat cluster (7, 8, 20) (GenBank accession no. AF009224 and AF150928) and the pob-qui-pca cluster (9, 10) (GenBank accession no. L05770), which both extend for >20 kbp but are separated on the chromosome by approximately 270 kbp (13). Recently we reported that at one end of the ben-cat cluster is a group of four genes, areA, -B, -C, and -R, encoding an esterase, two dehydrogenases, and a regulator protein, that are responsible for the catabolism of alkanoate esters of benzyl alcohol, 2-hydroxybenzyl (salicyl) alcohol, and 4-hydroxybenzyl alcohol to benzoate, salicylate, and 4-hydroxybenzoate, respectively, and thus feeding these substrates into one of the two branches of the β-ketoadipate pathway (17). In this paper, we report a further extension, by about 5 kbp, of the ben-cat supraoperonic cluster beyond the are genes to include previously unreported genes responsible for the catabolism of alkyl salicylates through salicylic acid to catechol and thus channeling new substrates into the catechol branch of the pathway.

MATERIALS AND METHODS

Strains and plasmids.

The plasmids and bacterial strains used in this study are listed in Table 1. Isolates ADPW1 and ADPW38 were spontaneous mutants selected by the procedure outlined in reference 31.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Descriptiona	Reference or source
Acinetobacter strains
ADP1 (BD413)	Wild type	18
ADPW1	catA	31
ADPW38	ben	31
ADP6	pcaG	12
ADPW57	areB::Kmr; obtained by transformation of ADP1 with pADPW26	17
ADPW67	salA::Kmr; obtained by transformation of ADP1 with pADPW44	This study
ADPW70	salE::Kmr; obtained by transformation of ADP1 with pADPW76	This study
ADPW72	salR::Kmr; obtained by transformation of ADP1 with pADPW79	This study
ADPW78	salD::Kmr; obtained by transformation of ADP1 with pADPW86	This study
ADPW86	salE::KmrcatA; obtained by transformation of ADPW1 with pADPW76	This study
ADPW87	salE::Kmrben; obtained by transformation of ADPW38 with pADPW76	This study
ADPW88	salE::KmrpcaG; obtained by transformation of ADP6 with pADPW76	This study
E. coli strains
DH5α	F− φ80dlacZΔM15 Δ(lacZYA-argF)U169 deoR recA1 endA1 hsdR17(rK− mK+) phoA supE44 λ− thi-1 gyrA96 relA1	Gibco BRL
XL1-Blue MRF′	Δ(mcrA)183 Δ(mcrCB-hsdSMR-mrr)173 endA1 supE44 thi-1 recA1 gyrA96 relA1 lac [F′ proAB lacIqZ ΔM15 Tn10 (Tetr)]	Stratagene
BL21(DE3)pLysS	F−ompT hsdSB(rB− mB−) dcm gal (DE3) pLysS Cmr	Promega
Plasmids
pET5a	Apr; T7 expression vector	Promega
pUC18	Apr; cloning vector	32
pUC4K	Apr Kmr; source plasmid for Kmr cassette	28
pUI1637	Apr Kmr; source plasmid for Kmr cassette	11
pADPW32	8.0-kbp SacI fragment cloned from ADPW57 containing Kmr cassette and the entire areA in pUC18	17
pADPW33	3.6-kbp HindIII subclone of pADPW32 in pUC18	17
pADPW34	4.1-kbp SacI-XbaI subclone of pADPW32 in pUC18	This study
pADPW40	NdeI*-EcoRI* fragment containing areA in pET5a	17
pADPW41	1.6-kbp SacI-HindIII subclone of pADPW34 in pUC18	This study
pADPW44	pADPW41 with Kmr cassette from pUI1637 cloned into ClaI site in salA	This study
pADPW49	1.0-kbp EcoRI* fragment containing salE in pUC18	This study
pADPW70	NdeI*-EcoRI* fragment containing salE in pET5a	This study
pADPW76	pADPW49 with Kmr cassette from pUI1637 cloned into ClaI site in salE	This study
pADPW78	1.0-kbp EcoRI*-HindIII* fragment containing part of salR in pUC18	This study
pADPW79	pADPW78 with Kmr cassette from pUI1637 cloned into ClaI site in salR	This study
pADPW82	1.2-kbp EcoRI* fragment containing part of salD in pUC18	This study
pADPW86	pADPW82 with Kmr cassette from pUC4K cloned into NsiI site in salD	This study

^aAsterisks denote restriction sites added by PCR. 

Chemicals and media.

Aromatic substrates were obtained from Sigma-Aldrich Co. Ethyl salicylate was redistilled under reduced pressure to remove small amounts of contaminating ethanol. Luria-Bertani (LB) medium (23) was used for the cultivation of bacteria unless otherwise noted. For growth on defined carbon sources in liquid medium, the substrates were added to minimal salts medium (4) at the following concentrations: ethyl salicylate, sodium salicylate, benzyl acetate, benzoate, and 4-hydroxybenzoate at 2.5 mM and succinate at 10 mM. For growth on solid medium, a single nonvolatile carbon source (succinate, benzoate, salicylate, or 4-hydroxybenzoate) was added to minimal agar at the same concentrations, but volatile compounds (i.e., all of the esters) were presented in small tubes in the lids of inverted petri dishes containing minimal medium. When appropriate, ampicillin was incorporated at 100 μg/ml and kanamycin was added at 50 μg/ml for Escherichia coli and at 10 μg/ml for Acinetobacter.

DNA manipulations.

Standard methods were used for DNA manipulations (23). Total DNA was prepared from Acinetobacter sp. strain ADP1 by the cetyltrimethylammonium bromide method (3). Plasmids carrying insertions of Acinetobacter DNA were isolated from and maintained in E. coli host strain XL1-Blue MRF′ or DH5α (Table 1) unless otherwise noted. Plasmid DNA was prepared from E. coli by the alkaline lysis miniprep method (23) or by using Qiaprep columns (Qiagen). DNA fragments were recovered from agarose gels by using Qiaquick columns (Qiagen). Southern blots were prepared as described by Sambrook et al. (23), and hybridizations were carried out with an ECL direct labeling kit (Amersham) in accordance with the manufacturer's instructions.

PCR amplification.

PCR amplifications were carried out in 50-μl volumes of reaction buffer (New England Biolabs) containing 10 ng of template DNA, 100 pmol of each primer, 2.5 nmol of each deoxynucleoside triphosphate, 300 nmol of MgSO₄, and 1 U of Vent polymerase (New England Biolabs). In some reactions, 200 nmol of MgCl₂ and 1 U of Taq polymerase were used in place of the MgSO₄ and Vent polymerase. The mixtures were subjected to a 4-min hot start at 94°C and then to 30 cycles of 1 min at 94°C, 1 min at 56°C, and 2 min at 74°C.

RT-PCR.

Cells were grown on minimal medium containing either ethyl salicylate, salicylate, or succinate until they reached a density of about 10⁸/ml. Total RNA was prepared from 10 ml of the culture by using RNeasy Mini columns (Qiagen), with elution in 50 μl of water. To remove any contaminating genomic DNA, the RNA was incubated with 1 U of RNase-free DNase (Promega) and 1 U of RNasin (Promega) in 40 mM Tris-HCl (pH 7.9) containing 10 mM NaCl (10 mM), CaCl₂ (10 mM), and MgSO₄ (6 mM) for 30 min at 37°C. The RNA was cleaned by passage through an RNeasy Mini column prior to use in reverse transcription (RT)-PCR. RT-PCR was carried out with an Access RT-PCR kit (Promega). Amplifications were carried out across the salD-salE and salA-salR intergenic regions by using primer pair EDf (5′-AGATTTAGGTATTCAGCAATTCAGGGCAAAAGGTG-3′)-EDr (5′-AAGGCTCAGGCGTAAGCATCTTGTAAGTTTCCTC-3′) and ARf (5′-CCATGGACACGTGCGGTAGAC-3′)-ARr (5′-TTTTTGGTGCATGTGCTCGTAAGT-3′), respectively. PCRs were carried out in 50-μl volumes of reaction buffer (Promega) containing 0.5 μg of total RNA, 50 pmol of each primer, 50 μM (each) deoxynucleoside triphosphate, 1 mM MgSO₄, 5 U of avian myeloblastosis virus reverse transcriptase, and 5 U of Tfl DNA polymerase. After RT at 48°C for 1 h, the reaction mixtures were heated to 94°C for 2 min and subjected to 40 cycles of 30 s at 94°C, 1 min at 55°C, and 2 min at 68°C. Negative-control reactions, designed to ensure that residual genomic DNA was not amplified, were performed in the same way, except that the reverse transcriptase was omitted from the reaction mixtures.

Cloning of Acinetobacter sp. strain ADP1 DNA.

DNA adjacent to areABC was isolated by using the chromosomal drug resistance cassette in areB of strain ADPW57 (17). A plasmid library in E. coli (pUC18) was made from chromosomal DNA of ADPW57, from which plasmid pADPW32 with an 8.0-kbp SacI insert (Fig. 1) was selected by screening for Km^r Ap^r colonies. pADPW34 was constructed as a SacI-to-XbaI subclone of pADPW32 (Fig. 1) so as to have sequence overlap with the previously sequenced plasmid pADPW33 (17). pADPW34 was used as the DNA sequencing template. Sequence alignment confirmed its overlap with pADPW33.

FIG. 1.

Physical map of the salA, salR, salE, and salD genes and their location relative to areABC at the left-hand end (as drawn) of the supraoperonic ben-cat cluster. The inserts of the plasmids produced from cloning genomic DNA into vectors are specified in Table 1. pADPW32 was cloned directly from genomic DNA. All other plasmids were produced by PCR from genomic DNA (denoted by asterisks) or by subcloning from plasmids containing genomic DNA. Sites at the termini of the inserts marked with asterisks were incorporated via PCR primers. The Km^r cassette insertions are not to scale. The abbreviations for the restriction sites are as follows: Bc, BclI; C, ClaI; E, EcoRI; H, HindIII; N, NsiI; Nd, NdeI; S, SacI; and X, XbaI.

Expression of salE in E. coli.

Oligonucleotide primers were designed to produce a PCR fragment of the salE gene with (i) an NdeI site introduced at the putative start site of the reading frame, (ii) a constructed EcoRI site upstream of the NdeI site, and (iii) an EcoRI site downstream of the gene. The PCR fragment generated from pADPW34 was cut with EcoRI and first ligated into EcoRI-cut pUC18 to create pADPW49 (Table 1). The insert was sequenced on one strand to ensure that mutations had not been incorporated during the PCR. A fragment was excised with NdeI and EcoRI, religated into the expression vector pET5a, and transformed into E. coli BL21(DE3)pLysS to produce plasmid pADPW70 (Fig. 1). Since salE contained a NdeI restriction site from bp 8 to 14, the forward primer was designed with a mutation in the ninth base pair (A→G) that destroyed the native NdeI site but did not change the amino acid encoded (Thr). The primers used were 5′-AGGAGAATTCATATGATAACGTATGTACTTGTTC-3′ (forward) and 5′-AGCGAATTCCTCGGATATGGTTGATTCAAAC-3′ (reverse) (the NdeI site is italicized, the EcoRI restriction sites used for cloning into pUC18 are underlined, and the bases which differ from the wild-type sequence are in boldface type). The SalE protein encoded on the expression vector pADPW70 was expressed in E. coli BL21(DE3)pLysS by growth of the bacterium in LB medium to an optical density at 600 nm of 0.6 and subsequent induction for 4 h by addition of 0.4 mM isopropyl-β-d-thiogalactopyranoside (IPTG). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out in a discontinuous gel in a Mini-PROTEAN II electrophoresis cell (Bio-Rad) in accordance with the manufacturer's instructions.

Chromosomal disruption of salA, salR, salE, and salD in Acinetobacter sp. strain ADP1.

As a first step in the construction of gene knockouts, pUC18-derived plasmids carrying part or all of gene salA, salR, salE, or salD disrupted by a Km^r cassette were constructed. For salA, a SacI-HindIII fragment of pADPW34 was cloned to create pADPW41 (Fig. 1). The Km^r cassette of pUI1637 (11) was cloned into a unique ClaI site of pADPW41, creating pADPW44. The salE gene was disrupted by the insertion of the cassette from pUI1637 into the unique ClaI site of pADPW49 (see above) to form pADPW76. Plasmid-borne salR disruption was achieved after first creating pADPW78, which has a PCR-generated EcoRI insert in pUC18. This 1.0-kbp fragment, internal to salR and flanking its ClaI site, was amplified from pADPW34. Primer sequences (with the EcoRI sites underlined and the altered bases in boldface) were as follows: 5′-TGGAATTCATGAACAGATCCGAAAAGAACG-3′ (forward) and 5′-CATGAATTCCCTGAGTATGCCCGGTA-3′ (reverse). The central ClaI site in the pADPW78 was used as the insertion site for the Km^r cassette from pUI1637 (11) to create pADPW79. Disruption of plasmid-borne salD was performed after first creating pADPW82, which has a PCR-generated EcoRI insert in pUC18. This 1.2-kbp fragment, flanking the NsiI site in salD, was amplified from pADPW32. Primer sequences (with the EcoRI sites underlined and the altered bases in boldface) were as follows: 5′-AGGGGGAATTCTGGCAGCAATCACTG-3′ (forward) and 5′-GGGCTGGAATTCCCAAGTACTACCTAT-3′ (reverse). The central NsiI site in pADPW82 was used as the insertion site for a Km^r cassette from pUC4K (28) to create pADPW86. Plasmids pADPW44, pADPW76, pADPW79, and pADPW86 were linearized by digestion with an appropriate restriction enzyme, and each was used to transform ADP1 via natural transformation. Southern hybridization confirmed that in strains ADPW67 (salA::Km^r), ADPW70 (salE::Km^r), ADPW72 (salR::Km^r), and ADPW78 (salD::Km^r) the altered plasmid-borne allele had replaced the corresponding chromosomal wild-type region (data not shown).

Preparation of cell extracts.

Cells were harvested by centrifugation, washed with 100 mM phosphate buffer (pH 7.4), and stored as pellets at −20°C. Cell extracts were prepared by disrupting frozen pellets, suspended in ice-cold 100 mM phosphate buffer (pH 7.4), with a French pressure cell (SLM Instruments, Inc., Urbana, Ill.) and centrifuging the broken cells at 120,000 × g for 30 min at 4°C. The supernatant was stored frozen as 1-ml portions at −20°C.

Transformation of metabolites.

Transformation of ethyl salicylate into salicylate by cell extracts of SalE was monitored spectrophotometrically. The measuring cell contained 100 μM substrate, 100 μM Tris (pH 7.5), and 10 μl of cell extract, while the reference cell contained only buffer and enzyme.

Enzyme assays.

Salicylate hydroxylase (SalA) activity was measured in 3-ml reaction mixtures containing 50 mM Tris (pH 7.5), 100 μM NADH, and 100 μM salicylate. The reaction was initiated by addition of 20 μl of cell extract, and the rate of oxidation of NADH was determined spectrophotometrically at 340 nm (extinction coefficient, 6,220 mol⁻¹ cm⁻¹). Salicylate esterase (SalE) activity was assayed by spectrophotometric monitoring (405 nm) of the hydrolysis of 4-nitrophenyl ester substrates in 1-ml reaction mixtures containing 50 mM phosphate buffer (pH 8) and 2 mM 4-nitrophenyl ester. The 4-nitrophenyl esters with an aliphatic moiety of six or less carbon atoms were dissolved in methanol, and an aliquot was added to the assay mixture such that the final concentration of the ester was 2 mM. The 4-nitrophenyl esters with an aliphatic moiety of eight carbon atoms or longer were first dissolved in 2-propanol at 60°C and then added dropwise to 50 mM Tris-HCl (pH 8.0), prewarmed to 60°C, to a final ester concentration of 2 mM. The assay reactions were initiated by the addition of 10 μl of enzyme. The molar extinction coefficient of 4-nitrophenol was taken as 14,800 mol⁻¹ cm⁻¹. The activity of the esterase with ethyl salicylate as the substrate was determined in a linked assay. The rate of increase of absorbance at 340 nm was measured in 1-ml reaction mixtures containing 100 mM phosphate buffer (pH 8), 2 mM NAD⁺, 100 μM substrate, and 10 U of yeast alcohol dehydrogenase (Sigma-Aldrich Co.). The reaction was initiated by the addition of esterase. A linear response of rate to added esterase verified that the esterase-catalyzed reaction was the rate-limiting step. The assay produced 1 mol of NADH per mol of ethyl salicylate utilized.

Determination of kinetic parameters for salicylate esterase.

To obtain K_m and maximum velocity (V_max) values, initial velocities were measured at several nonsaturating concentrations of each compound. Preliminary experiments determined the approximate value of K_m, and accurate rate determinations were then performed with from 7 to 10 different substrate concentrations spanning the approximate K_m value. Initial velocities were analyzed by direct linear analysis using the program EnzPack, which calculates the most probable values for the kinetic parameters with their 68% confidence limits (30). Each reaction velocity was determined in triplicate with two separate extract preparations. The concentration of the substrate stock solution was accurately determined enzymatically by making the substrate limiting in the assay while other components were in excess, and the change in absorbance at 405 nm, corresponding to the total conversion of added substrate, was determined.

Nucleotide sequencing and sequence analysis.

DNA sequences were determined by primer walking of fragments cloned in pUC18 by MWG-Biotech Ltd. (Ebersberg, Germany). Searches of the GenBank database were carried out with the BLASTN and BLASTP programs from the National Center for Biotechnology Information, Bethesda, Md. (2). Sequence data were aligned and edited by using the Lasergene software package (DNAStar, Inc., Madison, Wis.). Amino acid sequence alignments were performed with the program ClustalW (PAM350 matrix) (27).

Nucleotide sequence accession number.

The DNA sequence obtained in this study has been added to the GenBank database (accession no. AF150928).

RESULTS

Analysis of nucleotide sequences of protein products.

Analysis of the nucleotide sequence downstream of areA revealed the presence of four open reading frames (Table 2). Immediately downstream of areA, but separated by 195 bp and an inverted repeat, which could serve as a termination loop for areA expression (17), is a gene designated here as salD. SalD shows 13 to 17% amino acid sequence similarity to putative proteins encoded by other operons of aromatic catabolism in different bacteria, which include TodX (29), XylN (GenBank accession no. D63341), and TbuX (H.-Y. Kahng, A. M. Byrne, R. H. Olsen, and J. J. Kukor, submitted for publication), which have been suggested to be membrane proteins involved in the transport of hydrocarbons (J. J. Kukor, personal communication). Downstream of salD is a gene, which we have called salE, whose product shows homology to serine esterases of the α/β hydrolase family of enzymes (21, 26) but which is not closely related to the benzyl esterase AreA-encoding gene lying upstream of it (26% similar, 15% identical). Transcribed convergently toward salE is a gene, which we have called salR, apparently encoding a regulatory protein of the LysR family. The closest matches to SalR are two NahR proteins involved in naphthalene catabolism, one from Pseudomonas stutzeri (6) and one from the Pseudomonas putida NAH7 plasmid (25). The latter protein has a dual role as a positive regulator of two functionally related operons, for the conversion of naphthalene to salicylate and for the further conversion of salicylate to central metabolites via catechol and the subsequent extradiol (meta) cleavage pathway (24, 25, 33). The final gene in this cluster is salA, whose putative product's closest relatives have both been named NahG. These enzymes are salicylate hydroxylases, converting the salicylate produced from naphthalene to catechol. The genes encoding both enzymes head the salicylate catabolic operon on the NAH7 plasmid (34) and in P. stutzeri (6), respectively. However, it has been noted that ADP1 does not grow on naphthalene as a sole carbon source. The alignments of both salA and salR with their Pseudomonas homologues strongly indicate that unlike the Pseudomonas genes, they both have a GTG start codon.

TABLE 2.

Acinetobacter sp. strain ADP1 genes and gene products

Gene designation	Putative function of gene product	Size of gene (bp)	% (A+T)	Size of gene product		Most-similar gene products (% amino acid identity/similarity)a [GenBank/Swissprot accession no.]
				Residues	kDa
salD	Putative membrane protein	1,161	58	386	41.8	XylN (P. putida pWW0) (19%/36%) [D63341], TodX (P. putida) (21%/36%) [U18304]
salE	Salicylate esterase	720	54	239	27.0	Putative esterase/lipase 2 (Mycoplasma pneumoniae) (17%/34%) [P75311]
salR	Regulatory protein of LysR family	891	65	296	34.1	NahR (P. stutzeri) (33%/53%) [AF039534]
						NahR (P. putida NAH7) (31%/52%) [M22723]
salA	Salicylate hydroxylase	1,272	54	423	46.9	NahG (P. putida) (48%/65%) [X83926]
						NahG (P. putida NAH7) (47%/65%) [M60055]

^aMeasured over whole protein. 

Insertional inactivation of single genes.

The chromosomal copies of salA, salE, salR, and salD were specifically disrupted, individually, by the insertion of a Km^r cassette into each of the genes in plasmid constructs (pADPW44, pADPW76, pADPW79, and pADPW86, respectively [Fig. 1]). The disrupted genes were introduced into ADP1, using the high frequency of natural transformation of which this strain is capable. Two of the resulting strains, ADPW67 (salA::Km^r) and ADPW72 (salR::Km^r), failed to grow on salicylate, unlike ADP1, which grows vigorously on salicylate overnight. By contrast, ADPW70 (salE::Km^r) grew on salicylate as well as did ADP1. ADPW78 (salD::Km^r) was a very unhealthy strain; it grew slowly compared with ADP1 even on succinate minimal medium and rich (LB) medium and lost viability when maintained on agar after 2 to 3 days. However, despite these limitations it, too, like ADPW70, grew on salicylate. Because SalE showed amino acid sequence homologies with other esterases, we compared the abilities of ADPW70 and ADP1 to grow on a range of esters containing aromatic components both as the alcohol and as the acid moiety, as well as a number of exclusively aliphatic esters. Thirteen esters (n-propyl acetate, benzyl acetate, n-butyl propionate, ethyl propionate, benzyl propionate, ethyl valerate, ethyl butyrate, benzyl butyrate, ethyl caproate, ethyl benzoate, n-propyl benzoate, n-butyl benzoate, and ethyl benzoylacetate) were growth substrates for both strains, and 6 (ethyl acetate, n-butyl acetate, n-butyl butyrate, benzyl benzoate, n-propyl cinnamate, and benzyl salicylate) were substrates for neither strain. However, ethyl salicylate and methyl salicylate supported growth of ADP1 but not ADPW70, suggesting that SalE is a hydrolase specific for these esters. Neither ADPW67, ADPW72, nor ADPW78 was able to grow on either of the two alkyl salicylates.

Expression of cloned salE.

The salE gene was cloned into expression vector pET5a as plasmid pADPW70 with its start codon (ATG) located in the optimal position for expression. SDS-PAGE of the induced E. coli BL21(DE3)pLysS containing pADPW70 revealed a strong protein band with an expected molecular mass of 27 kDa (Fig. 2). The esterase activity in extracts of induced cells against a range of 4-nitrophenyl esters was measured and compared with the activity of the upstream benzyl esterase AreA (17) (Table 3). Whereas AreA shows a broad specificity for the alkanoate side chain, going up to C₁₆, SalE shows a much more restricted range, up to only C₆. The K_m values of SalE for 4-nitrophenyl acetate and 4-nitrophenyl butyrate were 106 and 77 μM, respectively, whereas the relative V_max dropped 10-fold, from 28 to 2.8 μmol/min/mg, for the same two substrates.

FIG. 2.

SDS-PAGE of overexpressed SalE and AreA proteins. The lanes contain lysates from E. coli BL21(DE3)pLysS carrying the following plasmids, induced with IPTG (I) or uninduced (U): lane 2, pADPW70 (I); lane 3, pADPW70 (U); lane 4, pADPW40 (I); lane 5, pADPW40 (U); and lane 6, pET5a. Lane 1 contained molecular mass standards (A, 97.4 kDa; B, 66.2 kDa; C, 45.0 kDa; D, 31.0 kDa; and E, 21.5 kDa). The estimated molecular masses for the overexpressed bands are 27 kDa for SalE (lane 2) and 37 kDa for AreA (lane 4).

TABLE 3.

Relative specific activities of Acinetobacter esterases against 4-nitrophenyl alkanoates

No. of carbons in alkyl chain	Relative specific activitya
	SalEb	AreAc
2	450	1.9
4	100	100
6	0.5	87
8	<0.1	6.5
10	<0.1	1.4
12	<0.1	0.9
14	<0.1	0.2
16	<0.1	<0.1

^aRelative to the activity against 4-nitrophenyl butyrate (C₄) (set at 100). All reaction rates were based on the average of three measurements, none of which varied by >5%. 

^bMeasured in cell extracts of E. coli BL21(pADPW70). The specific activity against 4-nitrophenyl butyrate was 11.2 U/mg of protein. 

^cMeasured in cell extracts of E. coli BL21(pADPW40). The specific activity against 4-nitrophenyl butyrate was 38.0 U/mg of protein. 

We attempted to set up a SalE assay using ethyl salicylate as a substrate by linkage to salicylate hydroxylase SalA overexpressed from a pET5a-derived plasmid. Unfortunately, the SalA construct failed to show activity for as-yet-undiagnosed reasons. However, we did set up an alternative assay with ethyl salicylate as a substrate by linking the assay to yeast alcohol dehydrogenase, acting on the ethanol produced by SalE action. This was successfully carried out and showed that after IPTG induction the E. coli(pADPW70) exhibited high-level hydrolytic activity against ethyl salicylate (Table 4).

TABLE 4.

Specific activities of SalE salicylate esterase in crude extracts of cells grown on different media

Strain	Growth substrate(s)	Specific activitya (U/mg of protein)
ADP1	Succinate	<0.04
ADP1	Salicylate	<0.04
ADP1	Ethyl salicylate	3.44
ADP1	Succinate + ethyl salicylate	0.67
ADPW70	Succinate	<0.04
ADPW70	Salicylate	<0.04
ADPW70	Succinate + ethyl salicylate	<0.04
ADPW72	Succinate	<0.04
ADPW72	Succinate + salicylate	<0.04
ADPW72	Succinate + ethyl salicylate	0.45
E. coli(pADPW70)	LB + IPTG	52.5
E. coli(pADPW70)	LB	<0.04

^aAll reaction rates were based on the average of three measurements, none of which varied by >5%. 

Expression of salicylate hydroxylase.

Using the standard NADH-linked assay procedure, we were able to detect salicylate hydroxylase activity in a number of strains (Table 5). For ADP1, activity was not detectable in succinate-grown cells but was induced by growth on both salicylate and ethyl salicylate, and when grown on succinate in the presence of both salicylate and ethyl salicylate the activity was significantly induced, although at a lower level than in the absence of succinate. For both ADPW67 (salA::Km^r) and ADPW72 (salR::Km^r), no activity was detected when the cells were grown on succinate plus salicylate or, for the latter, succinate in the presence of ethyl salicylate. However, for ADPW70 (salE::Km^r) there was high-level induction when grown on salicylate and a low, but significant, level of induction when grown in the presence of ethyl salicylate, which it is unable to transform.

TABLE 5.

Specific activities of salicylate hydroxylase in crude extracts of cells grown on different carbon sources

Growth mediuma	Specific activityb (U/mg of protein) in extracts of:
	ADP1 (wild type)	ADPW67 (salA::Kmr)	ADPW70 (salE::Kmr)	ADPW72 (salR::Kmr)	DH5α(pADPW34) (pUC18::salARE)	DH5α(pUC18)
Salicylate	0.70	NDc	0.50	ND	ND	ND
Ethyl salicylate	0.30	ND	ND	ND	ND	ND
Salicylate + succinate	0.55	<0.01	ND	<0.01	ND	ND
Ethyl salicylate + succinate	0.11	ND	0.09	<0.01	ND	ND
Succinate	<0.01	<0.01	<0.01	<0.01	ND	ND
LB + salicylate	ND	ND	ND	ND	1.37	<0.01
LB	ND	ND	ND	ND	<0.01	ND

^aAcinetobacter strains were grown in minimal medium containing aromatic substrates at 2 mM and/or succinate at 10 mM as indicated. 

^bAll reaction rates were based on the average of three measurements, none of which varied by >5%. 

^cND, not done. 

Although we were unable to measure salicylate hydroxylase activity encoded by the gene cloned into expression vector pET5a (see above), activity was detected in E. coli DH5α(pADPW34), which carries salA, salR, and salE (Fig. 1). Moreover, the activity was induced only when 2 mM salicylate was added to the LB medium. The specific activity was twofold higher in salicylate-induced E. coli(pADPW34) than in salicylate-grown ADP1. The relative activities exhibited by the strain with this cloned gene against a range of substituted salicylates were compared with those expressed in the wild-type ADP1 and found to be identical, within experimental error, with a broad substrate specificity except against the only available position 3-substituted salicylate (Table 6).

TABLE 6.

Relative activities of SalA salicylate hydroxylases in crude extracts of cells

Assay substrate	Relative activitya in cell extracts of:
	ADP1b	DH5α(pADPW34)c
Salicylate	100	100
3-Methyl salicylate	<0.1	<0.1
4-Methyl salicylate	56	54
5-Methyl salicylate	35	33
4-Chlorosalicylate	45	40
5-Chlorosalicylate	27	22

^aActivity relative to that on salicylate, which is set at 100. All reaction rates were based on the average of three measurements, none of which varied by >5%. 

^bADP1 was grown on minimal medium containing 2 mM salicylate; the specific activity was 0.71 U/mg of protein. 

^cE. coli DH5α(pADPW34) was grown on LB containing 2 mM salicylate; the specific activity was 1.37 U/mg of protein. 

Phenotypes of mutants.

To demonstrate that the aromatic moiety arising from salicylate and ethyl salicylate is channeled down the β-ketoadipate pathway, we checked the phenotypes of mutants blocked both in the sal genes and at three different points in the β-ketoadipate pathway (Table 7). The three β-ketoadipate pathway mutants were ADP6 (pcaG), which does not grow on 4-hydroxybenzoate or any substrate that feeds into the protocatechuate branch of the pathway; ADPW1 (catA), with a functionless catechol 1,2-dioxygenase which blocks the utilization of benzoate and catechol and which accumulates catechol as demonstrated by the brown coloration on agar plates from any substrate that feeds into the catechol branch; and ADPW38 (ben), which has an uncharacterized lesion in benABC and is unable to convert benzoate to catechol. The Km^r cassette insertion mutation in salE was also individually introduced into ADP6, ADPW1, and ADPW38 by natural transformation to produce double mutants, all of which were tested for growth on the appropriate carbon sources (Table 7). Only ADPW1 of the single β-ketoadipate pathway mutants failed to grow on salicylate or ethyl salicylate, with catechol accumulating in both media. The growth phenotypes of the double mutants were consistent with the proposed pathway (Fig. 3) in which the salicylate nucleus is fed into the benzoate branch at the level of catechol.

TABLE 7.

Growth phenotypes of Acinetobacter strains

Growth substrate	Growth of straina (relevant genotype)
	ADP1 (wild type)	ADPW67 (salA)	ADPW70 (salE)	ADPW72 (salR)	ADPW38 (ben)	ADPW1 (catA)	ADP6 (pcaG)	ADPW88 (salE ben)	ADPW86 (salE catA)	ADPW87 (salE pcaG)
Benzoate	+	+	+	+	−	−B	+	−	−B	+
4-Hydroxybenzoate	+	+	+	+	+	+	−	+	+	−
Salicylate	+	−	+	−	+	−B	+	+	−B	+
Ethyl salicylate	+	−	−	−	+	−B	+	−	−	−
Succinate	+	+	+	+	+	+	+	+	+	+

^aGrowth was assessed on agar plates: +, good growth; −, no growth; −B, no growth, with accumulation of brown or black coloration (catechol). 

FIG. 3.

Proposed pathway for the catabolism of alkyl salicylates linked to the ben-cat branch of the β-ketoadipate pathway in Acinetobacter sp. strain ADP1.

RT-PCR analysis of ADP1 transcripts.

To confirm that the sal genes are transcribed during both ethyl salicylate and salicylate catabolism and that the operon structure is as implied by the gene organization (Fig. 1), transcripts from cells grown on both of these substrates and on succinate as a noninducing negative control were examined. Two primer sets, spanning from salD through to salE and from salA through to salR, were constructed (Fig. 4A). The expected RT-PCR product sizes for the salD-salE and salA-salR amplicons were 1,195 and 988 bp, respectively. The PCR products obtained, together with restriction digests chosen to confirm the presence of expected restriction sites, were analyzed by agarose gel electrophoresis. The salAR products obtained from the total RNA of cells grown on both ethyl salicylate and salicylate were of the expected sizes (Fig. 4B). Also, a salED product of the expected size was obtained from the total RNA of cells grown on ethyl salicylate. The presence of restriction sites in the expected positions within the fragments was confirmed by digestion with ClaI (salED) and HindIII (salAR). No products were obtained from total RNA of succinate-grown cells or from reaction mixtures from which the reverse transcriptase had been omitted (data not shown).

FIG. 4.

RT-PCR of sal genes. (A) Positions of the genes relative to the SacI site (at bp 1), the primers used for the RT-PCR, and the HindIII and ClaI restriction sites. (B) Agarose gel electrophoresis of RT-PCR products amplified from ADP1 grown on ethyl salicylate and salicylate. The sizes of molecular size markers (in base pairs) in lanes S (HyperLadder I; Bioline, London, United Kingdom) are indicated on the right. Lanes: 1, salAR, salicylate-grown cells (expected size, 988 bp); 2, salAR, salicylate-grown cells digested with HindIII (582 and 406 bp); 3, salAR, ethyl salicylate-grown cells (expected size, 988 bp); 4, salAR, salicylate-grown cells digested with HindIII (582 and 406 bp); 5, salED, ethyl salicylate-grown cells (expected size, 1195 bp); 6, salED, ethyl salicylate- grown cells digested with ClaI (860 and 335 bp). No detectable products were obtained in control reactions, with each pair of primers, from which reverse transcriptase had been omitted or in reactions carried out on succinate-grown cells (data not shown).

DISCUSSION

Catabolic role of sal genes and proteins.

The two enzymes salicylate esterase (SalE) and salicylate hydroxylase (SalA) reported in this paper are involved in the sequential catabolism of alkyl salicylates via salicylate to catechol. They represent another route into the β-ketoadipate pathway for substrates which are likely to be found as natural products, either directly of plant origin or as microbial breakdown products of plant compounds. In this study, both enzymes were expressed from cloned genes, SalE from the expression vector pET5a, from which it is expressed to a very high specific activity, and SalA from a pUC18 clone carrying salERA. The SalA salicylate hydroxylase activity shows the same relative substrate preferences as does the activity found in wild-type ADP1 grown on salicylate alone. In addition, insertional salE and salA knockout mutants, whose construction was facilitated by the natural transformation of ADP1, show the phenotype expected from the proposed pathway (Fig. 3). The further catabolism of the aromatic moiety into the ben-cat branch of the β-ketoadipate pathway at the level of catechol was also confirmed by the fact that only mutations in catA (for catechol 1,2-dioxygenase), and not those in the ben or the pca genes, eliminated the ability to utilize either the ester or the free salt of salicylate.

It is interesting that the location of these genes is directly adjacent to the areCBA operon, which we have recently described (17), whose role is to channel benzyl alkanoates into the β-ketoadipate pathway by hydrolysis of the esters to benzyl alcohol and two sequential dehydrogenase-catalyzed oxidations of benzyl alcohol to benzoate. The two sets of genes thus appear complementary in that the are genes are responsible for the catabolism of esters in which the alcohol moiety is aromatic whereas the sal genes encode proteins that function in the catabolism of esters in which the acid moiety is aromatic.

Comparisons of Sal proteins.

Examination of the deduced amino acid sequence of SalE in the PROSITE database (16) shows that from residues 68 to 77 (IVLLGHSYGG) it has the signature characteristic of serine lipases, [LIV]-x-[LIVFY]-[LIVMST]-G-[HYWV]-S-x-G-[GSTAC], in which the serine is the active-site nucleophile. Its primary sequence does not align closely (<26% similarity) with other reported Acinetobacter esterases, one from Acinetobacter lwoffii RAG-1 (1) and two (a carboxylesterase [19] and the adjacent benzyl esterase, AreA [17]) from strain ADP1, all of which are longer and have their putative active site serines further from the N terminus than SalE.

The regulator protein SalR is clearly a member of the LysR family of regulator proteins, with the family signature motif containing a helix-turn-helix at residues 17 to 47 (NISKAAEILNLSQPSVTYNLNRLRKHLNNPL) according to the PROSITE database (16). The high degree of similarity of both SalR and SalA to the two Pseudomonas isofunctional proteins, NahR and NahG, from the P. putida NAH7 plasmid (24, 33) and P. stutzeri (6) implies the occurrence of past intergeneric exchange of genes by horizontal transfer and yet, within the context of conservation of amino acid sequence, an equilibrium of DNA composition, in terms of AT/GC ratio, with that of the host. Whereas the four Pseudomonas genes have G+C ratios of between 60 and 65%, the salA and salR genes have a composition more characteristic of the A+T-rich Acinetobacter genome (Table 2).

The open reading frame salD appears to encode a member of a family of proteins of which FadL from E. coli (5) is perhaps the archetype. A number of these proteins encoded by open reading frames within gene clusters associated with aromatic catabolism have been reported, including TodX (29), XylN (GenBank accession no. D63341), TbuX (H.-Y. Kahng, A. M. Byrne, R. H. Olsen, and J. J. Kukor, submitted for publication), and CumH (14), but their function in this context has yet to be definitively determined. The overall level of similarity within the family is low, only 13 to 17%, but there are 23 conserved residues, which are also found in SalD (J. J. Kukor, personal communication). We have created a mutant of SalD, with a Km^r insertion, which is unable to grow on ethyl salicylate, but because this insertion will exert a polar effect on salE expression, this does not prove that SalD is essential for its catabolism. However RT-PCR has shown that salD and salE are cotranscribed during growth on salicylate and ethyl salicylate, so it is reasonable to assume that they have related functions. Definitive proof of this would be obtained by constructing a salD deletion mutant without a concomitant polar effect on salE, and so far we have been unable to make such a mutant. What is clear is that the growth rate of the salD::Km^r mutant is severely reduced, even on noninducing substrates, and its cell viability is also impaired, with plate cultures of ADPW78 dying after 2 to 3 days, whereas cultures of ADPW70 with a salE::Km^r insertion remain as viable as those of wild-type ADP1.

Regulation of sal genes.

SalA activity is induced by growth of ADP1 on salicylate or on ethyl salicylate (Table 5). It is probable that SalR is the protein that regulates expression of SalA, since (i) insertion of a Km^r cassette into salR in strain ADPW72 stops induction of salicylate hydroxylase activity by salicylate (Table 5); (ii) salicylate hydroxylase is also salicylate inducible in E. coli from pADPW34, which carries only salARE (Table 5); and (iii) there are close amino acid sequence homologies between SalR, SalA, and the Pseudomonas homologues NahR and NahG. A further point of comparison with the Pseudomonas genes is that there is also homology with the regions upstream of salA identified by Schell (24) as being the promoter sites at which NahR interacts; upstream of nahAa is sequence TCA-N₆-TGA, and upstream of nahG is sequence TCA-N₃-TGATGA (24) (GenBank accession no. M11863). The sequence upstream of salA is TCA-N₃-TGATGG. Just downstream of this there is also a −35 sequence, TAGGCAATT, which has 5 bases that correspond to those of the −35 sequence identified for nahAa, TGGTGTATT (24) (GenBank accession no. M11863), but the putative −10 region shows no similarity. A major difference between the sal and nah genes is that whereas nahR is transcribed divergently from its adjacent catabolic gene, nahAa, salA, and salR appear to be cotranscribed as a single regulatory unit, as shown by the RT-PCR results. This implies that SalR controls its own expression.

Similarly, RT-PCR suggests that salD and salE are cotranscribed, although there remains a remote possibility, as is also the case with salAR, that the two genes are separately transcribed on two overlapping mRNAs. The induction results show that growth on ethyl salicylate is necessary for induction of SalE activity but that salicylate does not act as the inducer (Table 4). It is also clear that (i) when salR is inactivated in ADPW72 (salR::Km^r), SalE remains inducible by ethyl salicylate; and (ii) there are no obvious potential binding sites upstream of salD similar to nahR promoter sites. This points to the possibility that the regulatory mechanism for salDE differs from that of salAR and implies that there might be an additional regulator gene on the ADP1 chromosome that is involved in the induction of salDE but is not located in the immediate vicinity; we have sequenced about 5 kb further upstream of salA and found no obvious regulator gene present. A further possibility which needs to be tested is that salD and -E are cotranscribed with areCBA under the control of AreR.