Department of Microbiology, Groningen Biomolecular Sciences and Biotechnology Institute (GBB), University of Groningen, PO Box 14, 9750 AA Haren, The Netherlands1
Author for correspondence: Lubbert Dijkhuizen. Tel: +31 50 3632153. Fax: +31 50 3632154. e-mail: L.Dijkhuizen{at}biol.rug.nl
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Keywords: 3-ketosteroid dehydrogenase, isoenzymes, unmarked gene deletion, steroid
Abbreviations: AD, 4-androstene-3,17-dione; ADD, 1,4-androstadiene-3,17-dione; DCPIP, 2,6-dichlorophenolindophenol; KSH, 3-ketosteroid 9-hydroxylase; KSTD, 3-ketosteroid
1-dehydrogenase; 9OHAD, 9
-hydroxy-4-androstene-3,17-dione; PMS, phenazine methosulphate
b The GenBank accession number for the sequence reported in this paper is AY078169
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
General molecular techniques.
DNA modifying enzymes were purchased from Roche, New England Biolabs or Amersham Pharmacia Biotech and were used as described by the manufacturers. Isolation of DNA restriction fragments from agarose gels was done by using the Qiagen gel extraction kit according to the manufacturers protocol. All DNA manipulations were done according to standard protocols. Rhodococcus plasmid DNA was isolated using the Qiagen spin-prep kit with a slight modification: Rhodococcus cells were grown for 23 h in the presence of ampicillin (600 µg ml-1) prior to lysozyme (2 mg ml-1) treatment. PCR was performed under standard conditions using Taq polymerase (Roche) unless stated otherwise: 5 cycles of 1 min at 95 °C, 1·5 min at 60 °C and 1·5 min at 72 °C, followed by 25 cycles of 1 min at 95 °C, 1·5 min at 55 °C and 1·5 min at 72 °C.
Functional complementation experiments.
Electrocompetent Rhodococcus cells of mutant strain RG1-UV29 were obtained as described previously (van der Geize et al., 2000 ). A gene library (mean insert size 6 kb) of R. erythropolis strain RG1 was constructed by cloning Sau3A-digested chromosomal DNA into pRESQ (van der Geize et al., 2002
). Colonies of strain RG1-UV29 obtained after introducing this gene library by electrotransformation were replica-plated onto mineral agar medium containing AD (0·5 g l-1) as sole carbon and energy source. Functional complementation was scored after 3 days incubation at 30 °C. Colonies growing on AD mineral agar medium were cultivated in LBP medium for isolation of plasmid DNA that was subsequently reintroduced into strain RG1-UV29 to check for genuine complementation.
Nucleotide sequencing.
Nucleotide sequencing analysis was done on an ALF-Express sequencing robot using dye-primers in the cycle sequencing method (Murray, 1989 ) with the thermosequenase kit RPN 2538 from Amersham Pharmacia Biotech. Protein and nucleotide sequence comparisons were performed using the facilities of the BLAST server (Altschul et al., 1990
) at NCBI (National Library of Medicine, Washington, DC, USA).
Expression of kstD2 in E. coli BL21(DE3) and preparation of cell-free extracts.
The kstD2 gene was isolated from chromosomal DNA of parent strain SQ1 and mutant strains RG1-UV29, RG1-UV45, RG1-UV50 and RG1-UV51 by PCR, using kstD2 forward primer 5'-GCGCATATGGCTAAGAATCAGGCACCC-3' and reverse primer 5'-GCGGGATCCCTACTTCTCTGCTGCGTGATG-3'. For KSTD2 expression by the T7 RNA polymerase pET3 expression system (Novagen), NdeI (forward primer) and BamHI (reverse primer) restriction sites were introduced (underlined) and used to clone kstD2 into NdeI/BamHI-digested pET3b (Novagen). The resulting plasmid, pKSD114, was introduced into E. coli BL21(DE3) (Studier & Moffatt, 1986 ). For expression of soluble active KSTD2, E. coli BL21(DE3)(pKSD114) was grown for 40 h at 25 °C in LB broth (25 ml) supplemented with 0·5 M sorbitol. Cell pellets (7300 g, 30 min, 4 °C) were washed with phosphate buffer (g l-1: KH2PO4, 2·72; K2HPO4, 3·48; MgSO4.7H2O, 2·46; pH 7·2). Cells were suspended in 5 ml phosphate buffer and disrupted in a French pressure cell at 140 MPa. Cell extracts were centrifuged (40000 g, 20 min, 4 °C) to remove cell debris. The supernatants (510 mg protein ml-1) were used in KSTD activity assays.
Preparation of cell-free extracts of E. coli DH5 containing the KSTD1 isoenzyme was done as described previously (van der Geize et al., 2000
).
Cloning of mutant kstD2 genes.
Mutant kstD2 genes were isolated from four UV mutants of strain RG1 (strains RG1-UV29, RG1-UV45, RG1-UV50 and RG1-UV51) by PCR using Pwo polymerase (Roche) and kstD2 expression primers. The obtained PCR products were cloned into pBlueScript (II) KS for nucleotide sequencing.
Gene deletion mutagenesis of kstD2.
Unmarked kstD2 gene deletions were carried out as described previously (van der Geize et al., 2001 ) using plasmid pKSD201 bearing the sacB counter-selection system. Southern analysis of kstD2 gene deletion mutants was done using a digoxigenin (DIG)-labelled kstD2 probe obtained by PCR (Boehringer PCR DIG probe synthesis kit 1636090) using primers developed for heterologous kstD2 expression in E. coli.
Preparation of cell-free extracts of R. erythropolis strains.
Rhodococcus strains were grown in mineral medium (100 ml) supplemented with 10 mM glucose. Cultures were induced with AD (0·5 g l-1) for 10 h. Following centrifugation (7300 g, 30 min, 4 °C), pellets of non-induced and induced cultures were washed with phosphate buffer, resuspended in phosphate buffer (1 ml) and disrupted by two passages through a French pressure cell at 140 MPa. Supernatants obtained after centrifugation (40000 g, 20 min, 4 °C) were used for KSTD enzyme activity assays.
KSTD enzyme activity assay.
Enzyme activities of KSTD1 or KSTD2 were measured in triplicate spectrophotometrically at 30 °C using phenazine methosulphate (PMS) and 2,6-dichlorophenolindophenol (DCPIP). The reaction mixture (1 ml) consisted of 50 mM Tris/HCl, pH 7·0, 1·5 mM PMS, 40 µM DCPIP, cell-free extract and 500 µM steroid in methanol (2%). Activities are expressed as mean values±SD in mU (mg protein)-1; 1 mU is defined as the reduction of 1 nmol DCPIP min-1 (600=18·7x103 cm-1 M-1).
Steroid bioconversion and steroid analysis.
Steroid bioconversion was done (in duplicate) in shake flasks with Rhodococcus cultures grown in 100 ml YG15 (g l-1: yeast extract, 15; glucose, 15) medium at 28 °C (200 r.p.m. on a rotary shaker). After growth overnight till late exponential phase (OD600 of 59), AD or 9OHAD (5 g l-1 suspended in 0·1%, v/v, Tween 80) was added and bioconversion was followed for several days. Steroids were extracted from the medium (1 ml) with 3 ml methylene-chloride (GC) or 3 ml methylene-chloride/methanol (9:1) extraction (TLC). For HPLC analysis, samples were diluted five times with methanol/water (70:30) and filtered (0·45 µm). Steroids were analysed by HPLC [column, reversed phase Lichrosorb 10RP18, SS 250x3 mm (Varian Chrompack International)], UV254 detection [liquid phase, methanol/water (60:40), 35 °C], GC [column, J&W DB-5MS 30 mx0·25 mm i.d.x0·25 µm (Alltech): pre-column, 2 mx0·25 mm i.d. (InterSciences): FID-40 detection at 300 °C] or TLC [Kieselgel 60 F254 10x20 cm (Merck) in toluene/ethylacetate 1:1]. Retention times for HPLC (flow rate 0·8 ml min-1) were tR,9OHAD=4·2 min, tR,ADD=7·0 min and tR,AD=10·5 min, and for GC (carrier N2, 100 kPa) tR,AD=10·8 min and tR,ADD=11·1 min. Retention factors for TLC were RF,9OHAD=0·23, RF,ADD=0·44 and RF,9AD=0·56. Substrates used (AD, 9OHAD, ADD) were supplied by Diosynth.
In vitro conversion of AD by KSTD2.
A reaction mixture (1 ml) consisting of Tris/HCl, pH 7·0 (50 mM), PMS (150 µM), AD (700 µM) and cell-free extract of E. coli BL21(DE3)(pKSD114) (approx. 25 µg protein) was incubated overnight at 30 °C. Steroid analysis was done by HPLC as described above.
![]() |
RESULTS AND DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
An insert of approximately 6·5 kb was identified in pKSD101 and subjected to restriction mapping analysis. A 3·6 kb EcoRI DNA fragment of pKSD101 was cloned into pBlueScript (II) KS (pKSD105; Fig. 2) and into RhodococcusE. coli shuttle vector pRESQ (pKSD106) for complementation. Plasmid pKSD106 was still able to restore the strain RG1-UV29 phenotype, whereas a 2·4 kb Asp718 DNA fragment (located on this 3·6 kb fragment) in pRESQ could not. Nucleotide sequence analysis of the 3·6 kb EcoRI fragment revealed the presence of a large ORF (ORF1) of 1698 nt (GC content 63·9 mol%), encoding a putative protein of 565 aa (calculated mass 60·2 kDa). The deduced amino acid sequence of ORF1 showed clear similarity with KSTD1 (35% identity), encoded by kstD in R. erythropolis SQ1, and other known KSTD amino acid sequences present in databases (Fig. 3
) and was thus designated kstD2. Highest similarity was found with the ORF3-encoded protein (63% identity) from A. simplex (Dziadek et al., 1998
) (GenBank accession no. Z93338). For alignment purposes, the ORF3 nucleotide sequence was edited to obtain a deduced amino acid sequence matching the KSTD consensus, since ORF3 obviously contained several frame-shift errors. The alignment strongly suggests that A. simplex also contains at least two KSTD isoenzymes, encoded by ksdD (Molnár et al., 1995
) and ORF3 (suggested name ksdD2), respectively. Contrary to kstD2 in R. erythropolis SQ1, ORF3 was found in close proximity to ksdD and ksdI, encoding KSTD1 and steroid 4,5-isomerase, respectively (Dziadek et al., 1998
; Molnár et al., 1995
). Analysis of nucleotide sequences directly upstream and downstream of the kstD2 gene showed no direct evidence for the presence of additional (co-transcribed) nucleotide sequences involved in steroid metabolism. No overlap in nucleotide sequences upstream or downstream of the kstD2 gene was found with sequences flanking kstD (van der Geize et al., 2000
).
|
|
Expression of kstD2 in E. coli
Attempts to express kstD2 from its own promoter in E. coli were unsuccessful. Alternatively, kstD2 was cloned into the pET3b expression vector (pKSD114) for expression via the T7 RNA polymerase expression system in E. coli BL21(DE3). Highly active soluble KSTD2 enzyme was obtained in extracts of E. coli cells grown at 25 °C. Incubation of AD with cell-free extracts (25 µg protein) of E. coli BL21(DE3)(pKSD114) resulted in its conversion into ADD, confirming that in vivo KSTD2 indeed performs the 3-ketosteroid 1-dehydrogenation reaction. ADD formation was not observed in controls with cell-free extracts without AD. Contrary to KSTD1, KSTD2 does not show any activity in the presence of nitro blue tetrazolium. KSTD2 thus could not be visualized by activity staining on native PAGE (van der Geize et al., 2000
). In vitro activity of KSTD2 was detected, however, in the presence of DCPIP and activity was further enhanced by addition of PMS. Optimal KSTD2 activity (6±0·8 U mg-1) was found at pH 78 at 30 °C using AD and the DCPIP/PMS system. A fivefold higher KSTD2 activity was observed with AD compared to 9OHAD. Substrate affinity of KSTD2 for AD (Km,app=17 µM) was 10-fold higher than for 9OHAD (Km,app=165 µM). In contrast, KSTD1 displays similar activities and affinities with AD (Km,app=82 µM) and 9OHAD (Km,app=87 µM).
Construction of single and double gene deletion mutants
For kstD2 unmarked gene deletion mutagenesis, plasmid pKSD201, with a kstD2 remnant (kstD2) of 605 bp, was obtained after deletion of a 1093 bp MluI internal fragment of kstD2 (Fig. 2
). Plasmid pKSD201 was introduced into E. coli S17-1 and mobilized by conjugation to both parent strain R. erythropolis SQ1 and kstD mutant strain RG1. Using the sacB counter-selection system we were able to isolate mutant strains RG7 (kstD2 mutant) and RG8 (kstD kstD2 mutant). Southern analysis, using DIG-labelled kstD2 gene as probe, on Asp718-digested chromosomal DNA of parent strain SQ1, strain RG7 and strain RG8, confirmed deletion of kstD2 in the latter two strains: a 2·4 kb Asp718 DNA fragment of strain SQ1 hybridizing with the kstD2 probe was reduced to a 1·3 kb DNA fragment in both mutant strains (data not shown). PCR performed on chromosomal DNA of strains RG7 and RG8 with the kstD2 primers resulted in PCR products with sizes of 0·6 kb, further demonstrating kstD2 gene deletion.
Characterization of mutant strains RG1, RG7 and RG8
Growth on AD, ADD or 9OHAD mineral agar medium was unaffected in both single gene deletion mutant strains RG1 and RG7. Growth on steroid substrates AD and 9OHAD was blocked in the kstD kstD2 double gene deletion strain RG8, while growth on ADD was not affected, confirming inactivation of the steroid 1-dehydrogenation reaction. Thus, both KSTD1 and KSTD2 are involved in steroid degradation by R. erythropolis SQ1 and the presence of only one of these enzymes is sufficient to allow degradation of the steroid skeleton to proceed. KSTD enzyme activities could not be detected in extracts of cells of parent strain SQ1 and mutant strains RG1, RG7 and RG8 grown on glucose without steroid induction. Upon steroid (AD) induction, KSTD enzyme activities (using AD as substrate) were found in cell-free extracts of parent strain SQ1 (280±40 mU mg-1), strain RG1 (80±5 mU mg-1) and strain RG7 (285±30 mU mg-1), indicating that both KSTD1 and KSTD2 are inducible proteins. No KSTD activity was detected in cell-free extracts of the double gene deletion mutant strain RG8.
Bioconversion of AD
Incubations of parent strain SQ1 and three mutants, i.e. RG1, RG7 and RG8, with 5 g AD l-1 (17·5 mM) or 5 g 9OHAD l-1 (16·6 mM) were performed (Fig. 4). Comparable rates of AD degradation (via KSTD and/or KSH activities; Fig. 1
) were observed for all strains (1·01·7 mmol AD l-1 h-1). No 9OHAD accumulation from AD was observed with strains SQ1 and RG1 (Fig. 4A
). Strain RG7 initially accumulated 9OHAD (1·8±0·1 mmol 9OHAD formed l-1 h-1), stoichiometrically derived from AD (1·7±0·1 mmol AD degraded l-1 h-1), suggesting low, or even absence, of KSTD activity during the early phase of biotransformation. With strain RG7, maximal 9OHAD accumulation was found within 20 h after addition of AD; 9OHAD was subsequently slowly degraded again (0·23±0·03 mmol 9OHAD l-1 h-1). Similar rates of 9OHAD degradation (0·27±0·04 mmol 9OHAD l-1h-1) were found during incubation of strain RG7 with 9OHAD. Much higher rates of 9OHAD degradation were observed for parent strain SQ1 and kstD mutant strain RG1 (1·3±0·2 mmol 9OHAD l-1 h-1). These experiments show that 9OHAD degradation proceeds differently in an environment with only KSTD2 present, compared to a situation in which KSTD1 is the only available KSTD enzyme. With solely KSTD2 present (strain RG1), steroid degradation of 9OHAD proceeds at the same rate as in parent strain SQ1. In contrast, the absence of KSTD2 (strain RG7) results in significant temporary accumulation of 9OHAD during AD bioconversion, and in low rates of 9OHAD degradation. In vivo, KSTD1 thus is considerably less efficient in
1-dehydrogenation of 9OHAD than KSTD2. Incubation of strain RG8 with AD resulted in high accumulation levels of 9OHAD (75% of AD converted into 9OHAD) within 24 h after addition of AD. The 9OHAD formed was not degraded during further incubation. Thus, inactivation of both kstD and kstD2 is necessary to inactivate KSTD activity in R. erythropolis SQ1 and to block 9OHAD degradation.
|
Heterologous expression of mutant KSTD2 protein encoded by kstD2 (Thr503Ile) or kstD2 (Ser325Phe) in E. coli Bl21(DE3) resulted in full-length protein on SDS-PAGE having no detectable KSTD enzyme activity.
Conclusions
R. erythropolis SQ1 employs two KSTD isoenzymes that in vivo may have different functional roles in AD degradation. AD degradation may occur via either ADD or 9OHAD (Fig. 1). Considering the observed mild toxicity of ADD to R. erythropolis SQ1 [growth on agar medium is considerably slower in the presence of higher concentrations (0·5 g l-1) of ADD], we speculate that the presence of two KSTD enzymes may serve to prevent accumulation of relatively high cellular concentrations of ADD. A regulatory system thus is expected that differentially regulates KSTD2 and KSTD1 cellular activity levels in response to ADD levels. kstD is preceded by a putative regulatory gene (kstR) that may fulfil just this role (van der Geize et al., 2000
). Regulation of KSTD1 and KSTD2 activities and/or expression levels, and the role of kstR herein will be at the focus of our future research.
Complete inactivation of steroid 1-dehydrogenation by gene deletion of kstD and kstD2 in R. erythropolis SQ1 (strain RG8) has resulted in a metabolic block at the level of steroid degradation and a stoichiometrical accumulation of 9OHAD from AD. Such mutants are of industrial interest. Accumulation of steroid intermediates (e.g. AD, 9OHAD, ADD) from sterol substrates may produce bioactive (precursor) molecules for medical use.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Choi, K.-P., Molnár, I., Yamashita, M. & Murooka, Y. (1995). Purification and characterization of the 3-ketosteroid 1-dehydrogenase of Arthrobacter simplex produced in Streptomyces lividans. J Biochem 117, 1043-1049.[Abstract]
Drobnic, K., Kriaj, L., Guben
ek, F. & Komel, R. (1993). Improved purification of steroid 1:2-dehydrogenase from Nocardia opaca and partial characterization of its cloned gene sequence. Biochem Biophys Res Commun 190, 509-515.[Medline]
Dziadek, J., Yamashita, M. & Murooka, Y. (1998). Cloning, sequencing and characterization of a downstream region of ksdDI operon of Arthrobacter simplex. Acta Microbiol Pol 47, 345-353.[Medline]
Fujii, C., Morii, S., Kadode, M., Sawamoto, S., Iwami, M. & Itagaki, E. (1999). Essential tyrosine residues in 3-ketosteroid 1-dehydrogenase from Rhodococcus rhodochrous. J Biochem 126, 662-667.[Abstract]
Itagaki, E., Wakabayashi, T. & Hatta, T. (1990). Purification and characterization of 3-ketosteroid-1-dehydrogenase from Nocardia corallina. Biochim Biophys Acta 1038, 60-67.[Medline]
Kaufmann, G., Thole, H., Kraft, R. & Atrat, P. (1992). Steroid-1-dehydrogenase of Rhodococcus erythropolis: purification and N-terminal amino acid sequence. J Steroid Biochem Mol Biol 43, 297-301.[Medline]
Molnár, I., Choi, K.-P., Yamashita, M. & Murooka, Y. (1995). Molecular cloning, expression in Streptomyces lividans, and analysis of a gene cluster from Arthrobacter simplex encoding 3-ketosteroid-1-dehydrogenase, 3-ketosteroid-
5-isomerase and a hypothetical regulatory protein. Mol Microbiol 15, 895-905.[Medline]
Morii, S., Fujii, C., Miyoshi, T., Iwami, M. & Itagaki, E. (1998). 3-Ketosteroid 1-dehydrogenase of Rhodococcus rhodochrous: sequencing of the genomic DNA and hyperexpression, purification, and characterization of the recombinant enzyme. J Biochem 124, 1026-1032.[Abstract]
Murray, V. (1989). Improved double-stranded DNA sequencing using the linear polymerase chain reaction. Nucleic Acids Res 17, 8889.[Medline]
Plesiat, P., Grandguillot, M., Harayama, S., Vragar, S. & Michel-Briand, Y. (1991). Cloning, sequencing and expression of the Pseudomonas testosteroni gene encoding 3-oxosteroid 1-dehydrogenase. J Bacteriol 173, 7219-7227.[Medline]
Quan, S. & Dabbs, E. R. (1993). Nocardioform arsenic resistance plasmid characterization and improved Rhodococcus cloning vectors. Plasmid 29, 74-79.[Medline]
Schäfer, A., Tauch, A., Jäger, W., Kalinowski, J., Thierbach, G. & Pühler, A. (1994). Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 69-73.[Medline]
Sih, C. J. & Bennet, R. E. (1962). Steroid 1-dehydrogenase of Nocardia restrictus. Biochim Biophys Acta 56, 587-592.
Simon, R., Priefer, U. & Pühler, A. (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Bio/Technology 1, 784-791.
Studier, F. W. & Moffatt, B. A. (1986). Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J Mol Biol 189, 113-130.[Medline]
Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighing, position-specific gap penalties and weight matrix. Nucleic Acids Res 22, 4673-4680.[Abstract]
Van der Geize, R., Hessels, G. I., Van Gerwen, R., Vrijbloed, J. W., Van der Meijden, P. & Dijkhuizen, L. (2000). Targeted disruption of the kstD gene encoding a 3-ketosteroid 1-dehydrogenase isoenzyme of Rhodococcus erythropolis strain SQ1. Appl Environ Microbiol 66, 2029-2036.
Van der Geize, R., Hessels, G. I., Van Gerwen, R., Van der Meijden, P. & Dijkhuizen, L. (2001). Unmarked gene deletion mutagenesis of kstD, encoding 3-ketosteroid 1-dehydrogenase, in Rhodococcus erythropolis SQ1 using sacB as counter-selectable marker. FEMS Microbiol Lett 205, 197-202.[Medline]
Van der Geize, R., Hessels, G. I., Van Gerwen, R., Van der Meijden, P. & Dijkhuizen, L. (2002). Molecular and functional characterization of kshA and kshB, encoding two components of 3-ketosteroid 9-hydroxylase, a class IA monooxygenase, in Rhodococcus erythropolis strain SQ1. Mol Microbiol 45, 1007-1018.[Medline]
Wagner, M., Atrat, P. G., Wagner, B., Hanemann, V. & Clark-Curtiss, J. E. (1992). Overexpression of a Rhodococcus erythropolis protein in Escherichia coli with immunological identity to the Rhodococcus steroid 1-dehydrogenase. Immunoelectron microscopic localization and electrophoretic studies. J Basic Microbiol 32, 269-277.[Medline]
Wovcha, M. G., Brooks, K. E. & Kominek, L. A. (1979). Evidence for two steroid 1,2-dehydrogenase activities in Mycobacterium fortuitum. Biochim Biophys Acta 574, 471-479.[Medline]
Received 21 March 2002;
revised 14 June 2002;
accepted 28 June 2002.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |