ANewBacterialSteroidDegradationGeneCluster

A PPLIED AND E NVIRONMENTAL M ICROBIOLOGY,Aug.2003,p.4421–4430Vol.69,No.8 0099-2240/03/$08.00?0DOI:10.1128/AEM.69.8.4421–4430.2003

Copyright?2003,American Society for Microbiology.All Rights Reserved.

A New Bacterial Steroid Degradation Gene Cluster in Comamonas

testosteroni TA441Which Consists of Aromatic-Compound

Degradation Genes for Seco-Steroids and3-Ketosteroid

Dehydrogenase Genes

Masae Horinouchi,1*Toshiaki Hayashi,1Takako Yamamoto,1and Toshiaki Kudo1,2,3 RIKEN1and JST,2Wako-shi,Saitama351-0198,and Science of Biological Supermolecular Systems,Graduate School of Integrated Science,Yokohama City University,Suehiro,Tsurumi-ku,Yokohama230-0045,3Japan

Received2December2002/Accepted27May2003

In Comamonas testosteroni TA441,testosterone is degraded via aromatization of the A ring,which is cleaved

by the meta-cleavage enzyme TesB,and further degraded by TesD,the hydrolase for the product of TesB.

TesEFG,encoded downstream of TesD,are probably hydratase,aldolase,and dehydrogenase for degradation

of2-oxohex-4-enoicacid,one of the products of TesD.Here we present a new and unique steroid degradation

gene cluster in TA441,which consists of ORF18,ORF17,tesI,tesH,ORF11,ORF12,and tesDEFG.TesH and

TesI are3-ketosteroid-?1-dehydrogenase and3-ketosteroid-?4(5?)-dehydrogenase,respectively,which work in

the early steps of steroid degradation.ORF17probably encodes the reductase component of9?-hydroxylase for

1,4-androstadiene-3,17-dione,which is the product of TesH in testosterone degradation.Gene disruption

experiments showed that these genes are necessary for steroid degradation and do not have any isozymes in

TA441.By Northern blot analysis,these genes were shown to be induced when TA441was incubated with

steroids(testosterone and cholic acid)but not with aromatic compounds[phenol,biphenyl,and3-(3-hydroxy-

phenyl)propionic acid],indicating that these genes function exclusively in steroid degradation.

Comamonas testosteroni is able to utilize certain C19and C21 steroids,as well as a number of aromatic compounds,as sole carbon and energy sources via a complex metabolic pathway involving many steroid-inducible enzymes.The mechanism by which testosterone is degraded in Nocardia restrictus and C. testosteroni was proposed in the1960s(4,8,14).However,the genes involved in this degradation pathway in C.testosteroni have yet to be identi?ed,except for the initial17?-dehydroge-nation(1,3,7)and the following?1-dehydrogenation(13).In previous studies,we identi?ed the testosterone degradation genes tesB and tesDEFG among some open reading frames (ORFs)which are involved in testosterone degradation in C.testosteroni TA441(9,10).tesB encodes the meta-cleavage enzyme for3,4-dihydroxy-9,10-secoandrosta-1,3,5(10)-triene-9,17-dione(3,4-DHSA)(10),and tesD encodes the hydrolase for4,5-9,10-diseco-3-hydroxy-5,9,17-trioxoandrosta-1(10),2-dien-4-oic acid(4,9-DSHA)(9).TesEFG are probably the enzymes for degradation of2-oxohex-4-enoic acid,which is one of the two products of the hydrolysis of4,9-DSHA by TesD.

Studies on these genes have shown the probable testoster-one degradation pathway of C.testosteroni TA441to be that presented in Fig.1.The process is initiated by dehydrogenation of the17?-hydroxyl group on testosterone to4-androstene-3,17-di-one(4-AD)(Fig.1,reaction1),which then undergoes?1-dehy-drogenation to1,4-androstadiene-3,17-dione(ADD)(reaction2), followed by9?-hydroxylation to produce3-hydroxy-9,10-secoan-drosta-1,3,5(10)-triene-9,17-dione(3-HSA)(reaction3and the following spontaneous cleavage).The C-4of3-HSA is hydroxy-lated to yield3,4-DHSA(Fig.1,reaction4),which is cleaved between C-4and C-5via a meta-cleavage reaction by TesB(re-action5)(10).The product of the meta-cleavage reaction of 3,4-DHSA,4,9-DSHA,is degraded into9,17-dioxo-1,2,3,4,10,19-hexanorandrostan-5-oic acid and2-hydroxyhexa-2,4-dienoic acid by TesD(reaction6)(9).One of the products,probably2-hy-droxyhexa-2,4-dienoic acid,is degraded by TesEFG.

TesB,-D,-E,-F,and-G show homology to the correspond-ing enzymes in bacterial aromatic compound degradation,such as BphC,-D,-H,-I,and-J,respectively,in biphenyl degrada-tion.However,the homologies between TesB and BphC and between TesD and BphD are about40%at most.In the up-stream gene region of TesD,ORF11and ORF12were identi-?ed.ORF11is oriented in the direction opposite that of ORF12in relation to tesG.Both of the ORFs are indispens-able,are induced in TA441grown on testosterone,and are suggested to encode oxygenases.The region from ORF11to tesG was not included in the gene region of approximately20 kb containing tesB.

In the present study,we report a new steroid degradation gene cluster consisting of ORF18,ORF17,tesI,tesH,ORF11,ORF12, and tesDEFG.3-Ketosteroid-?1-dehydrogenase(TesH)and3-ke-tosteroid-?4(5?)-dehydrogenase(TesI)are encoded in sequence, followed by two ORFs,ORF17and ORF18,which are indispens-able and are induced in TA441cells grown on steroids.ORF17 probably encodes the reductase component of9?-hydroxylase for ADD,which is the product of the reaction catalyzed by TesH in testosterone degradation.In contrast to the3-ketosteroid-?1-dehydrogenases in Rhodococcus erythropolis SQ1(KstD1and KstD2)(17,18),TesH is a unique?1-dehydrogenase for4-AD in TA441.The enzymes encoded by ORF11to ORF18and by ORF12to tesG are induced when TA441is grown with steroids

*Corresponding author.Mailing address:RIKEN,2-1Hirosawa, Wako-shi,Saitama351-0198,Japan.Phone:81484679545.Fax:8148 4624672.E-mail:masae@postman.riken.go.jp.

4421

FIG.1.Proposed pathway of testosterone degradation in C.testosteroni TA441.

4422HORINOUCHI ET AL.A PPL .E NVIRON .M ICROBIOL .

but not when it is grown with aromatic compounds [phenol,bi-phenyl,and 3-(3-hydroxyphenyl)propionic acid],indicating that the enzymes encoded by this gene cluster are used exclusively for steroid degradation in TA441.

MATERIALS AND METHODS

Culture conditions.C.testosteroni TA441and mutant strains were grown at 30°C in either Luria-Bertani (LB)medium,C medium (10),or a mixed LB-plus-C medium (a mixture of equal volumes of LB and C media)with suitable carbon sources when necessary.Steroids and aromatic compounds,except for 3-(3-hydroxyphenyl)propionic acid,were added in a ?ltered dimethyl sulfoxide solution at a ?nal concentration of 0.1%(wt/vol).3-(3-Hydroxyphenyl)propionic acid was added as a CH 3CN solution at a ?nal concentration of 0.1%(wt/vol).Construction of plasmids and mutant strains for gene disruption experi-ments.The tesH gene was disrupted by insertion of a kanamycin resistance gene into the Not I site in tesH .The resultant plasmid,pTesH-Km r ,was used for inactivation of the tesH gene in TA441by homologous recombination.The plasmid was introduced into TA441by electroporation,and a Km r and carben-icillin-sensitive TA441mutant was selected.Insertion of the Km r gene in tesH was con ?rmed by Southern hybridization using the Km r gene and tesH as probes.Gene disruption of tesI ,ORF17,and ORF18was performed in the same manner.The restriction site at which the Km r gene was inserted is indicated in Fig.2.Growth of TA441and mutant strains on testosterone.Growth was monitored by CFU counting,because the culture to which testosterone was added in a dimethyl sulfoxide solution was turbid,which prevented measurement of cell absorbance.Mutants were grown with 0.1%(wt/vol)testosterone or andros-terone,and growth was monitored by counting colonies that appeared on LB plates on which appropriate dilutions of the culture had been spread and incu-bated at 30°C,as previously described (10).

Three-dimensional HPLC.A volume of methanol equivalent to twice the vol-ume of culture was added to the culture,which was then centrifuged.The super-natant was directly injected into a high-pressure liquid chromatography (HPLC)system.A Waters 2690HPLC was utilized with an Inertsil ODS-3column (2.1by 150mm),and elution was carried out using a linear gradient from 20%solution A (CH 3CN –CH 3OH –tri ?uoroacetic acid [95:5:0.05])and 80%solution B (CH 3CN –H 2O –tri ?uoroacetic acid [95:5:0.05])to 65%solution A and 35%solution B over 10min;the latter proportion was maintained for 3min,and then the solvent was adjusted to 20%solution A for 5min.The ?ow rate was 0.21ml/min.

Isolation of the intermediate product X4accumulated by the TesH disruption mutant.For isolation of the intermediate product accumulated by the TesH disruption mutant (the compound X4),a 1,200-ml culture of the tesH disruption mutant incubated in C medium with 0.1%testosterone was twice extracted with the same volume of ethyl acetate.The ethyl acetate layer was concentrated in vacuo,and the residue was dissolved in a small amount of methanol and loaded onto a Waters 600HPLC (Nihon Waters,Tokyo,Japan)with an Inertsil ODS-3

column (20by 250mm;GL Science)and a solvent composed of CH 3CN and H 2O (1:1),with a ?ow rate of 1ml/min,at 40°C.X4was detected at 245nm.The fraction containing X4was collected and studied further.

General experimental procedures.For gas chromatography-mass spectrom-etry (GC-MS),an API 2000LC/GC/MS spectrometer (Perkin-Elmer,Inc.,Boston,Mass.)was used.UV spectra were recorded with an Ultrospec 3300(Amersham Pharmacia Biotech,Piscataway,N.J.).Nuclear magnetic resonance (NMR)(1D,1H,and 13C,DEPT,PFG-DQFCOSY,PFG-HMQC,PFG-HMBC)spectra were taken on a JNM-ECP500spectrometer (JEOL,Tokyo,Japan)in a CDCl 3solution with tetramethylsilane at 0ppm as an internal standard for 1H and 13C chemical shifts.

Northern blot analysis.For the data presented in Fig.8,TA441cells incubated in LB medium for about 12h were collected and added to newly prepared LB medium containing 0.1%testosterone and succinate.Total RNA was puri ?ed every 2h after the start of the incubation and analyzed by Northern hybridization with suitable probes.Succinate was used as a negative control.For the data presented in Fig.9,TA441cells incubated in LB medium for about 12h were collected and added to newly prepared LB medium containing 0.1%of each substrate [testosterone,cholic acid,phenol,biphenyl,3-(3-hydroxyphenyl)pro-pionic acid,and succinate].Total RNA was puri ?ed at 4h after the start of the incubation and analyzed by Northern hybridization with suitable

probes.

FIG.2.Partial restriction map of the DNA fragment containing the steroid degradation genes of C.testosteroni https://www.sodocs.net/doc/c512677189.html,rge open arrows,genes and putative ORFs;small open arrowheads above large arrows,putative promoters.Putative terminators are located just downstream of tesG and ORF18.Small solid arrowheads above genes and ORFs indicate restriction sites where the Km r gene was inserted.DNA segments used as probes are shown just below the ORFs.Deletion plasmids and remaining gene segments are also shown below the restriction map.Abbreviations:Bm,Bam HI;Bg,Bgl II;EI,Eco RI;EV,Eco RV;H,Hin dIII;K,Kpn I;Mn,Mun I;Nr,Nru I;Nt,Not I;Ps,Pst I;Sc,Sac I;Sl,Sal I;Sm,Sma I;Sn,Sna BI;Xb,Xba I;Xh,Xho

I.

FIG.3.Phylogenetic tree for TesH and ?1DHs.The branching pattern was generated by the neighbor-joining method (CLUSTAL X).The putative amino acid sequence of ORF18was used as an outgroup.KsdD(IFO12069),?1DH of A.simplex IFO12069(accession number S61889);KsdD(IFO3338),?1DH of R.rhodochrous IFO3338(acces-sion number AB007847);KstD1and KstD2,?1DHs of R.erythropolis SQ1(accession numbers AF096929and AY078169,respectively);?1DH(ATCC17410),?1DH of C.testosteroni ATCC 17410(accession number M68488).

V OL .69,2003STEROID DEGRADATION GENE CLUSTER IN C .TESTOSTERONI 4423

Nucleotide sequence accession number.The nucleotide sequence data re-ported in this paper will appear in the DDBJ/EMBL/GenBank nucleotide se-quence databases with accession number AB076368.

RESULTS

Isolation and sequencing of the gene segment downstream of ORF11.The two DNA segments of C.testosteroni TA441containing genes necessary for testosterone degradation have been reported previously.One contains tesB ,the gene encod-

ing the meta-cleavage enzyme which converts 3,4-DHSA to 4,9-DSHA (10).In another gene segment,tesD ,a gene encod-ing the hydrolase for 4,9-DSHA,and tesEFG ,genes encoding the enzymes for the following reaction (Fig.2)(9),occur in sequence.ORF11and ORF12,showing homology to some oxygenases,are located upstream of tesD ,and ORF11is ori-ented in the direction opposite that of ORF12in relation to tesG.Downstream of tesG ,a putative terminator was found,which indicates that this is one end of this gene

cluster.

FIG.4.Alignment of TesH and the ?1DHs.Asterisks indicate amino acids identical among all the proteins;colons and periods indicate amino acids of high and low similarity,respectively.Regions in which the amino acid sequences of TesH and the ?1DH of ATCC 17410differ greatly are underlined.In these regions,number signs indicate amino acids identical among all the proteins except the ?1DH of ATCC 17410,and plus signs indicate amino acids identical among all the proteins except TesH.Abbreviations are the same as for Fig.3.

4424HORINOUCHI ET AL.A PPL .E NVIRON .M ICROBIOL .

To obtain the gene segment downstream of ORF11,several pUC19-derived plasmids were constructed from pC26and pC29,SuperCosI-derived plasmids carrying the DNA of TA441described previously (9)(Fig.2).Analysis of a 17-kb region downstream of ORF11revealed nine ORFs (tesH ,tesI ,and ORFs 17,18,50,51,52,53,and 54).Of the 17-kb region se-quenced,about 10kb comprising the region from tesH to ORF52is shown in Fig.2,because the following experiments showed that ORFs 50to 54are not involved in testosterone degradation in TA441.Two ORFs just downstream of ORF11are named tesH and tesI ,because they encode enzymes with high homologies to 3-ketosteroid-?1-dehydrogenase (?1DH)(83.0%identity)and 3-ketosteroid-?4(5?)-dehydrogenase [?4(5?)DH](75.8%identity),respectively,in C.testosteroni ATCC 17410(6,13).TesH also showed identities of about 33,34,38,and 33%to the following ?1DHs,respectively:Arthro-bacter simplex IFO12069KsdD (11),Rhodococcus rhodochrous IFO3338KsdD (12),and R.erythropolis SQ1KstD1and KstD2(identity between KstD1and KstD2,35%)(17,18).The phylo-genetic tree and the alignment of these ?1DHs are shown in Fig.3and 4,respectively (KstD2is not included in the alignment).The alignment revealed three regions in which the amino acid sequences of TesH and the ?1DH of ATCC 17410differ greatly (Fig.4).These differences are the results of a frame-shift caused by the lack of several nucleotides in the gene sequence of the ?1DH of ATCC 17410relative to that of tesH.A search of TesI homology revealed signi ?cant homology only with the ?4(5?)DH of ATCC 17410(75.0%identity).Align-ment of TesI and the ?4(5?)DH of ATCC 17410also revealed three signi ?cantly different regions as a result of a frameshift.Downstream of tesI ,two ORFs,ORF17and ORF18,fol-lowed by a putative terminator,were found.The putative amino acid sequence of ORF17showed homology with several kinds of ferredoxin reductases and possessed three conserved motifs of ferredoxin reductase:(i)the consensus sequence (RXFS),which is considered to be the binding region of a ?avin adenine dinucleotide-isoalloxazine ring,(ii)the G-rich region,which is considered to be the binding region of an NAD(P)-ribose,and (iii)the consensus sequence (CXXXX CXXC-any 29amino acids-C)of the chloroplast type Fe-S cluster (Fig.5).The putative amino acid sequence of ORF17showed the highest homology,about 46.0%,with KshB,the ferredoxin reductase component of R.erythropolis SQ13-ke-tosteroid-9?-hydroxylase (KSH)(16).All of this evidence in-dicates that ORF17encodes the ferredoxin reductase compo-nent of 9?-hydroxylase for ADD in testosterone degradation in C.testosteroni TA441.However,none of the putative proteins that we cloned showed signi ?cant homology to KshA,the ter-minal oxygenase component of R.erythropolis SQ1KSH (16).ORF18showed the maximum homology,about 45%,to several putative acid-coenzyme A (CoA)ligases,about 30%homology to long-chain fatty acid –CoA ligases,and lower homologies to other CoA ligases.

Downstream of ORF18,several putative ORFs (ORF50to ORF54)were found.Because the growth of the gene disrup-tion mutants of these ORFs on testosterone was almost equal to that of TA441(data not shown),these ORFs are not con-sidered to be involved in testosterone degradation.

Growth of tesH ,tesI ,ORF17,and ORF18disruption mu-tants on testosterone.tesH ,tesI ,ORF17,and ORF18in TA441were individually disrupted with a Km r gene by homologous recombination,and the resultant mutant strains were desig-nated TesH ?,TesI ?,ORF17?,and ORF18?,respectively.Figure 6a shows a comparison of the growth of each mutant strain and TA441on testosterone as the sole carbon source.The growth of each strain until around 6h after the start of the incubation was probably caused by the nutrients carried by the cells from preculture on LB medium.After that,growth of the TesH ?,ORF17?,and ORF18?mutants on testosterone as the sole carbon source was limited,indicating that these

genes

FIG.5.Alignment of the putative amino acid encoded by ORF17with R.erythropolis SQ1KshB.KshB is the ferredoxin reductase component of the two-component class IA monooxygenase KSH.Asterisks indicate amino acids identical in the two proteins;colons and periods indicate amino acids of high and low similarity,respectively.Three conserved motifs of ferredoxin reductase are boxed.

V OL .69,2003STEROID DEGRADATION GENE CLUSTER IN C .TESTOSTERONI 4425

are necessary for testosterone degradation in TA441.The growth of the TesI ?mutant was almost the same as that of TA441.This result is consistent with the function of TesI as ?4(5?)DH;because testosterone has a double bond at the ?4(5?)position,?4(5?)DH is not necessary for testosterone degradation.To con ?rm the function of TesI as ?4(5?)DH,we also examined the growth of these mutants and TA441on androsterone,whose ?4(5?)bond is single (Fig.6b).Only TA441showed signi ?cant growth on androsterone;the TesI ?mutant,as well as the other mutants,showed limited growth.HPLC analysis of the culture media of the mutants.The TesH ?,TesI ?,ORF17?,and ORF18?gene disruption mu-tants were incubated in a mixed LB-plus-C medium with tes-tosterone,and after about 40h the culture media were ana-lyzed by HPLC.The mixed LB-plus-C medium is a mixture of equal volumes of LB and C medium and was used because TA441mutants accumulated larger amounts of intermediate compounds in this medium than in C medium (9).Figure 7shows a three-dimensional HPLC chart of the culture of each mutant.The chart of the culture of the TesD disruption mu-tant,characterized in our previous study (9),is presented as authentic (Fig.7a).In the culture of the TesH ?mutant,4-AD and an intermediate compound named X4accumulated (Fig.7b).Compound X4was not detected in the culture of any other mutants.The identi ?cation of X4is described in the next section.The TesI ?mutant did not accumulate a signi ?

cant

FIG.6.Growth of TA441and gene disruption mutants on testosterone (a)or androsterone (b)as the sole carbon source.The genes were disrupted by insertion of a Km r gene.Each strain was grown in LB medium,washed twice,and inoculated into 10ml of C medium supplemented with 0.1%(wt/vol)testosterone or androsterone.Symbols:s ,TA441;?,TesH disruption mutant;?,TesI disruption mutant;F ,ORF17disruption mutant;E ,ORF18disruption mutant.Growth is expressed in CFU.Data are averages of more than three experimental determinations.

4426HORINOUCHI ET AL.A PPL .E NVIRON .M ICROBIOL .

amount of any intermediate (Fig.7c),a result usually observed in the culture of TA441or a well-growing mutant strain.The ORF17?and ORF18?mutants accumulated 4-AD,ADD,and 17?-hydroxylated ADD (Fig.7d and e).These intermediate compounds are detected with most gene disruption mutants of steroid degradation genes (9,10;also unpublished data).Fig-ure 7f and g show the culture of the TesH ?and TesI ?mutants,respectively,incubated with androsterone.In the culture of the TesH ?mutant incubated with androsterone,4-AD and X4accumulated,the same result as for incubation with testoster-one.In the culture of the TesI ?mutant,two intermediate compounds (XA1and XA2)accumulated;these were not de-tected in the culture of any other mutant.

Puri?cation and identi?cation of compound X4.For isola-tion of X4,the TesH ?mutant was incubated with testosterone,and the culture was extracted with ethyl acetate.The ethyl

acetate layer was concentrated,and the residue was dissolved in a small amount of methanol.The solution was loaded onto the HPLC column,and the fraction containing X4was col-lected from the eluent.The puri ?ed X4gave a molecular ion at m/z 325(M ?Na ?)and m/z 303(M ?H ?)by direct MS analysis,indicating its molecular formula to be C 19H 26O https://www.sodocs.net/doc/c512677189.html,pound X4showed maximum absorbance at 241.5nm (ε?16,000)in methanol solution,which agreed well with the value obtained by Dodson and Muir (5)for 9?-hydroxyandrost-4-ene-3,17-dione.Because complete NMR data were not avail-able from previous studies of 9?-hydroxyandrost-4-ene-3,17-dione,complete NMR assignments were performed (Table 1).The 13C NMR spectrum con ?rmed the presence of 19carbons including 2methyls at 12.7and 19.9ppm and 2carbonyl car-bons at 198.8and 219.9ppm.One hydroxy signal was observed at 76.7ppm.The 1H NMR spectrum con ?rmed two methyl protons at 0.93(s)and 1.35(s)ppm and one ole ?nic methine proton at 5.90(d)ppm.These characteristic signals of 1H NMR spectra are consistent with the partial 1H NMR spectral data of 9?-hydroxyandrost-4-ene-3,17-dione reported by Bell et al.(2),in which 9?-hydroxyandrost-4-ene-3,17-dione was prepared by the conversion of 4-AD by the fungus Diaporthe celastrina.Because the measured value of compound X4correlated well with the MS,UV spectrum,and partial NMR analysis data reported in previous studies,X4was identi ?ed as 9?-hy-droxyandrost-4-ene-3,17-dione.

Induction of the steroid degradation genes in TA441.Total RNA of TA441incubated in LB medium with testosterone or succinate (negative control)was puri ?ed every 2h for 8h and subjected to Northern blot analysis.LB medium,which has been shown not to inhibit the induction of testosterone degra-dation genes in TA441(9),was used to obtain almost the same volume of cells from cultures of all the compounds tested.Northern blot analysis showed that tesH and ORF18were induced in testosterone-grown cells (Fig.8).Induction

of

FIG.7.Intermediate compounds accumulating in the cultures of the gene disruption mutants of TA441incubated with testosterone (a through e)or androsterone (f and g).Data are represented in three-dimensional HPLC charts (the vertical axis indicates wavelength,the horizontal axis indicates reaction time,and the UV absorbance of each compound is represented in contour).(a)HPLC chart of the culture of the TesD disruption mutant grown on the mixed LB-plus-C medium with testosterone (presented as authentic).(b through g)HPLC charts showing cultures of the gene disruption mutants of tesH (b),tesI (c),ORF17(d),and ORF18(e)incubated with testosterone and of the gene disruption mutants of tesH (e)and tesI (g)incubated with andro-sterone.Ts,testosterone;17OH-ADD,17-hydroxy-1,4-androstadiene-3-one.

TABLE 1.NMR data for X4

Carbon no.

13

C (ppm)

1

H NMR (?[ppm])a

128.6 2.43m,1.75m 233.9 2.42m 3198.84127.2 5.90d (1.8)

5167.7631.5 2.51ddd (15.1,6.4,1.8),2.36ddd (15.1,5.1,1.8)724.2 1.75m,1.54m

837.2 2.11ddd (12.0,11.5,3.7)

976.71044.51126.2 1.78ddd (14.2,13.3,4.1),1.65ddd (14.2,4.1,2.3)

1227.0 1.69ddd (13.3,4.1,2.3),1.56m 1347.21444.3 1.86ddd (12.0,6.0,6.0)1521.6 1.91m,1.56m

1635.7 2.48br.dd (18.8,8.7),2.13ddd (18.8,8.7,8.7)17219.91812.70.93s 19

19.9

1.35s

a

Coupling constants (J values in hertz)are given in parentheses.

V OL .69,2003STEROID DEGRADATION GENE CLUSTER IN C .TESTOSTERONI 4427

ORF11,ORF12,and tesB in TA441cells incubated with ste-roids (testosterone and cholic acid)or with aromatic com-pounds [phenol,biphenyl,and 3-(3-hydroxyphenyl)propionic acid]at 4h after the start of the incubation is shown in Fig.9.Cholic acid was tested because TA441shows good growth on it.Induction of ORF11,ORF12,and tesB was examined be-cause these genes are located at the very beginning in each sequence of genes (putative operon).ORF11,ORF12,and tesB were induced during growth on steroids but not during growth on aromatic compounds.The mhpB gene,which en-codes the meta-cleavage enzyme for 3-(3-hydroxyphenyl)pro-pionic acid degradation,was used as a control.In contrast to ORF11,ORF12,and tesB ,mhpB was induced in cells incu-bated with 3-(3-hydroxyphenyl)propionic acid but not with ste-roids.

DISCUSSION

In this paper,we reported a bacterial steroid degradation gene cluster with a unique combination of genes that has not been reported before.It consists of ORF18,ORF17,tesI ,tesH ,ORF11,ORF12,and tesDEFG .TesDEFG,reported in our previous study (9),are the hydrolase for 4,9-DSHA (TesD)and enzymes for the subsequent reaction (TesEFG).The present study revealed the presence of a ?1DH gene and a ?4(5?)DH gene,tesH and tesI ,followed by two ORFs,ORF17

and ORF18,in the region upstream of tesD .All these genes are involved and induced in testosterone degradation in TA441.Involvement of TesH and TesI in androsterone degradation,as well as induction of isolated testosterone degradation genes in TA441cells incubated with cholic acid,indicates that these genes are targeted not only to testosterone but also to various kinds of steroids.Putative terminator sequences were found just next to the end of ORF18and just next to the end of tesG ,and ORFs located downstream of ORF18were not necessary for steroid degradation in TA441,indicating that this cluster consists of ORF18through tesG .Genes tesH ,tesI ,and proba-bly ORF17encode enzymes for steroids with an androstane structure,and ORF11,ORF12,and tesDEFG encode enzymes for the aromatized A ring of seco-steroids.(The roles of ORF11and ORF12will be described in a subsequent paper.)Gene disruption experiments and analysis of the culture media of the mutants showed that all these enzymes are involved in steroid degradation.The induction of these genes by steroids (testosterone and cholic acid)but not by aromatic compounds [phenol,biphenyl,and 3-(3-hydroxyphenyl)propionic acid]was con ?rmed by Northern blot analysis.These results clearly

show

FIG.8.Induction of tesH and ORF18in TA441incubated with testosterone.Total RNA was puri ?ed every 2h and analyzed by North-ern hybridization with suitable probes (indicated in Fig.2).Succinate was used as a negative

control.

FIG.9.Induction of ORF11,ORF12,and tesB in TA441cells in-cubated with steroids or aromatic compounds.Cells were collected 4h after the start of incubation and were subjected to Northern blot analysis with suitable probes (ORF11,ORF12,tesB ,and mhpB ).mhpB is the gene encoding the meta-cleavage enzyme of TA441in 3-(3-hydroxyphenyl)propionic acid degradation and was used as a control.Substrates tested were testosterone (Ts),cholic acid (CA),biphenyl (Bp),phenol (Ph),3-(3-hydroxyphenyl)propionic acid (PP),and suc-cinate (Sc).

4428HORINOUCHI ET AL.A PPL .E NVIRON .M ICROBIOL .

that tesDEFG developed exclusively for steroid degradation in TA441in spite of their similarity to aromatic compound deg-radation genes.From the evolutionary viewpoint,the relation-ship between the tes genes and aromatic compound degrada-tion genes is particularly interesting.

TesH showed about 83and 38%identity with C.testosteroni ATCC 17410?1DH (13)and R.erythropolis SQ1KstD1(17,18),respectively.R.erythropolis SQ1has two ?1DHs,KstD1and KstD2,and the kstD1disruption mutant of SQ1still pos-sessed the ability to grow on testosterone.The tesH disruption mutant of TA441did not show signi ?cant growth on testoster-one and accumulated 4-AD and 9?-hydroxyandrost-4-ene-3,17-dione,indicating that TesH is a unique ?1DH in TA441.4-AD and ADD were detected in the cultures of most of the mutants incubated with testosterone and in the culture of TA441in the early phase of growth on testosterone,while 9?-hydroxyandrost-4-ene-3,17-dione was detected only in the culture of the tesH disruption mutant.These results indicate that in TA441,?1-dehydrogenation of 4-AD to ADD is faster and more signi ?cant than 9?-hydroxylation of 4-AD.Figure 10shows proposed testosterone and androsterone conversion in TA441.

TesI,encoded just next to TesH,showed 75.8%identity with C.testosteroni ATCC 17410?4(5?)DH,and the tesI disruption mutant grew quite well on testosterone [in which the ?4(5?)bond is double]but failed to show signi ?cant growth on an-drosterone [in which the ?4(5?)bond is single],indicating that TesI is a unique ?4(5?)DH for steroid degradation in TA441.In the culture of the tesI disruption mutant incubated with androsterone,two intermediate compounds (XA1and XA2),which were not detected in the cultures of other mutants,were detected by HPLC.By comparison with peaks detected in the tesH disruption mutant incubated with androsterone,these compounds are believed to be 9?-hydroxyandrost-1-ene-3,17-dione and androst-1-ene-3,17-dione.

The putative amino acid sequence of ORF17possessed con-served motifs of a ferredoxin reductase component of hydroxy-lases and showed similarity to ferredoxin reductases;the high-est identity,about 48%,was to R.erythropolis SQ1KshB (16).Since KshB is the ferredoxin reductase component of SQ1KSH,there is a high possibility that ORF17encodes a ferre-doxin reductase component of KSH.This is supported by the result of HPLC analysis of the culture of the ORF17disruption mutant incubated with testosterone:most of the testosterone added was converted to ADD.KshA is the terminal oxygenase of SQ1KSH,though none of the putative proteins that were identi ?ed in TA441as involved in steroid degradation showed signi ?cant homology to KshA.Examination of the function of ORF17and a search for the probable terminal oxygenase com-ponent are now under way.ORF18is also necessary and in-duced in testosterone degradation and may encode an acid-CoA ligase.CoA-ligase adds CoA to a carboxyl group,but its role in steroid degradation is unclear.

Another gene cluster isolated from TA441contains tesB ,encoding a meta-cleavage enzyme for 3,4-DHSA,followed by three putative ORFs,ORF1to ORF3,all of which are neces-sary for testosterone degradation in TA441.Recently,a tes-tosterone-inducible biphenyl-2,3-diol-1,2-dioxygenase of

C.

FIG.10.Conversion of testosterone and androsterone in C.testosteroni TA441.Solid arrowheads indicate the probable main degradation pathway.Open arrowheads indicate minor reactions.

V OL .69,2003STEROID DEGRADATION GENE CLUSTER IN C .TESTOSTERONI 4429

testosteroni ATCC11996,TIP1,which is almost identical to TesB(only two amino acids are different),was reported(15). The tip1gene is followed by ORF1,tip6,ORF2,and ORF3, which are almost identical to ORFs1to3in TA441.(The ORF2of ATCC11996presented in the report of Skowasch et al.is shorter than ORF2in TA441.This is probably the result of an error in DNA sequence determination,which resulted in the division of an ORF,which corresponds to ORF2in TA441, into tip6and ORF2,because the estimated molecular weight of TIP6in their sodium dodecyl sulfate-polyacrylamide gel elec-trophoresis experiment was not consistent with the molecular weight of the putative protein encoded by tip6but was consis-tent with ORF2of TA441.)Extradiol dioxygenase,another name of the meta-cleavage enzyme,is well known for its wide substrate speci?city,so the activity on2,3-dihydroxy biphenyl and the approximately40%homology to BphC’s are not con-sidered to be enough evidence to conclude that it is biphenyl-2,3-diol-1,2-dioxygenase.To identify it as biphenyl-2,3-diol-1,2-dioxygenase,it is necessary to examine the growth of the gene disruption mutant on biphenyl and its induction in biphenyl-grown cells.TA441does not grow on biphenyl.Northern blot analysis showed that tesB is induced in TA441incubated with steroids(testosterone and cholic acid)but not with aromatic compounds[phenol,biphenyl,and3-(3-hydroxyphenyl)propi-onic acid].These results showed that TesB is an enzyme only for steroid degradation.The gene cluster containing tesB must be another steroid degradation gene cluster in TA441.Analysis of genes in this cluster is also necessary in order to clarify the entire degradation pathway of steroids in C.testosteroni.

ACKNOWLEDGMENTS

We thank Hiroyuki Koshino for practical advice on identi?cation of puri?ed intermediate compounds.We thank Y.Ichikawa and R.Na-kazawa(Biodesign DNA Sequence Facility,RIKEN)for determina-tion of the nucleotide sequence.

This work was partly supported by a grant from the Eco Molecular Sciences Research Program of RIKEN.M.H.was supported by a grant from the Special Postdoctoral Researchers Program of RIKEN.

REFERENCES

1.Abalain,J.H.,S.Di Stefano,Y.Amet,E.Quemener,M.L.Abalain-Colloc,

and H.H.Floch.1993.Cloning,DNA sequencing,and expression of(3–

17)?-hydroxysteroid dehydrogenase from Pseudomonas testosteroni.J.Ste-

roid Biochem.Mol.Biol.44:133–139.

2.Bell,A.M.,A.D.Boul,E.R.Jones,G.D.Meakins,J.O.Miners,and A.L.

Wilkins.1975.Microbiological hydroxylation.Part XVII.Introduction of

16?-,9?-,and3?-hydroxy-groups into dioxygenated5?-androstanes by the fungus Diaporthe celastrina.J.Chem.Soc.14:1364–1366.

3.Cabrera,J.E.,J.L.Pruneda Paz,and S.Genti-Raimondi.2000.Steroid-

inducible transcription of the3?/17?-hydroxysteroid dehydrogenase gene (3?/17?-hsd)in Comamonas testosteroni.J.Steroid Biochem.Mol.Biol.

73:147–152.

4.Coulter,A.W.,and P.Talalay.1968.Studies on the microbial degradation of

steroid ring A.J.Biol.Chem.243:3238–3247.

5.Dodson,R.M.,and R.D.Muir.1958.Microbiological transformations.II.

Microbiological aromatization of steroids.J.Am.Chem.Soc.80:5004–5005.

6.Florin,C.,T.Kohler,M.Grandguillot,and https://www.sodocs.net/doc/c512677189.html,amonas

testosteroni3-ketosteroid-?4(5?)-dehydrogenase:gene and protein charac-terization.J.Bacteriol.178:3322–3330.

7.Genti-Raimondi,S.,M.E.Tolmasky,L.C.Patrito,A.Flury,and L.A.Actis.

1991.Molecular cloning and expression of the?-hydroxysteroid dehydroge-nase gene from Pseudomonas testosteroni.Gene105:43–49.

8.Gibson,D.T.,K.C.Wang,C.J.Sih,and H.Whitlock,Jr.1966.Mechanisms

of steroid oxidation by microorganisms.IX.On the mechanism of ring A cleavage in the degradation of9,10-seco steroids by microorganisms.J.Biol.

Chem.241:551–559.

9.Horinouchi,M.,T.Hayashi,H.Koshino,T.Yamamoto,and T.Kudo.2003.

Gene encoding the hydrolase for the product of the meta-cleavage reaction in testosterone degradation by Comamonas testosteroni.Appl.Environ.Mi-crobiol.69:2139–2152.

10.Horinouchi,M.,T.Yamamoto,K.Taguchi,H.Arai,and T.Kudo.2001.

Meta-cleavage enzyme gene tesB is necessary for testosterone degradation in Comamonas testosteroni TA441.Microbiology147:3367–3375.

11.Molnar,I.,K.P.Choi,M.Yamashita,and Y.Murooka.1995.Molecular

cloning,expression in Streptomyces lividans,and analysis of a gene cluster from Arthrobacter simplex encoding3-ketosteroid-?1-dehydrogenase,3-ke-tosteroid-?5-isomerase and a hypothetical regulatory protein.Mol.Micro-biol.15:895–905.

12.Morii,S.,C.Fujii,T.Miyoshi,M.Iwami,and E.Itagaki.1998.3-Ketoster-

oid-?1-dehydrogenase of Rhodococcus rhodochrous:sequencing of the genomic DNA and hyperexpression,puri?cation,and characterization of the recombinant enzyme.J.Biochem.124:1026–1032.

13.Plesiat,P.,M.Grandguillot,S.Harayama,S.Vragar,and Y.Michel-Briand.

1991.Cloning,sequencing,and expression of the Pseudomonas testosteroni gene encoding3-oxosteroid?1-dehydrogenase.J.Bacteriol.173:7219–7227.

14.Sih,C.J.,S.S.Lee,Y.Y.Tsong,and K.C.Wang.1966.Mechanisms of

steroid oxidation by microorganisms:3,4-dihydroxy-9,10-secoandrosta-1,3,5(10)-triene-9,17-dione,an intermediate in the microbiological degrada-tion of ring A of androst-4-ene-3,17-dione.J.Biol.Chem.241:540–550. 15.Skowasch,D.,E.Mo¨bus,and E.Maser.2002.Identi?cation of a novel

Comamonas testosteroni gene encoding a steroid-inducible extradiol https://www.sodocs.net/doc/c512677189.html,mun.294:560–566.

16.van der Geize,R.,G.I.Hessels,R.van Gerwen,P.van der Meijden,and L.

Dijkhuizen.2002.Molecular and functional characterization of kshA and kshB,encoding two components of3-ketosteroid9?-hydroxylase,a class IA monooxygenase,in Rhodococcus erythropolis strain SQ1.Mol.Microbiol.

45:1007–1018.

17.van der Geize,R.,G.I.Hessels,R.van Gerwen,P.van der Meijden,and L.

Dijkhuizen.2001.Unmarked gene deletion mutagenesis of kstD,encoding 3-ketosteroid-?1-dehydrogenase,in Rhodococcus erythropolis SQ1using sacB as counter-selectable marker.FEMS Microbiol.Lett.205:197–202. 18.van der Geize,R.,G.I.Hessels,R.van Gerwen,J.W.Vrijbloed,P.van Der

Meijden,and L.Dijkhuizen.2000.Targeted disruption of the kstD gene encoding a3-ketosteroid-?1-dehydrogenase isoenzyme of Rhodococcus erythropolis strain SQ1.Appl.Environ.Microbiol.66:2029–2036.

4430HORINOUCHI ET AL.A PPL.E NVIRON.M ICROBIOL.