Identifcation of a thiolase gene essential for ß-oxidation of the acyl side chain of the steroid compound cholate in Pseudomonas sp. strain Chol1

(1)

Identification of a thiolase gene essential for p-oxidation of the

acyl side chain of the steroid compound cholate in Pseudomonassp.

strain Chol1

Antoinette Birkenmaier\ Heiko M. M611er²& Bodo Philipp1

'Universitat Konstanz. Fachbereich Biologie. Mikrobielle Okologie, Konstanz. Germany; and 2Universitat Konstanz, Fachbereich Chemie.

Konstanz, Germany

Correspondence: Bodo Philipp, Universitat Konstanz, Fachbereich Biologie, Mikrobielle Okologie. Fach M654, 78457 Konstanz, Germany. Tel.: +497531 884541;

fax: +497531 884047;

email: bodo.philipp@unikonstanz.de

DOl: 1 0.1111/j.1574 6968.2011.02250.x

Editor: Dieter Jahn

Keywords

steroid degradation; bile salts; cholate;

P ketothiolase.

Introduction

Abstract

Bile salts such as cholate are steroid compounds occurring ubiquitously in the environment through excretion by animals. Cholate degradation by Pseudomonas sp. strain Choll is initiated by A ring oxidation and ~ oxidation of the acyl side chain. A transposon mutant of strain Choll was isolated that could not grow with cholate, but transformed it into several steroid compounds accumulating in culture supernatants. The main product was identified as (22E) 7cx,12cx dihy droxy 3 oxochola 1,4,22 triene 24 oate (DHOCTO). A further compound was identified as 7cx,12cx,22 trihydroxy 3 oxochola 1,4 diene 24 oate (THOCDO). The structures of DHOCTO and THOCDO indicate that they are intermediates of the ~ oxidation of the acyl side chain. The interrupted gene was named skt and had similarities to the 3 keto acyl CoA thiolase domain of the eukaryotic sterol carrier protein SCP x. An skt mutant grew with intermediates of cholate degrada tion, from which the acyl side chain had been partly or completely removed.

Growth with cholate was restored by an intact skt copy on a plasmid. These results strongly suggest that skt encodes a ~ ketothiolase responsible for the cleavage of acetyl CoA from the acyl side chain of cholate. Sequence comparisons revealed that other steroid degrading bacteria such as Coma monas testosteroni contain genes encoding proteins very similar to Skt, suggesting a widespread role of this enzyme in bacterial steroid degradation.

Steroids are ubiquitous natural compounds with diverse functions for eukaryotic organisms. They act as membrane constituents (e.g. cholesterol, sitosterol, ergosterol) and as hormones (e.g. testosterone, estradiol, ecdyson). Bile salts (e.g. cholate, deoxycholate) are surface active steroid com pounds with important functions for the digestion of lipophilic nutrients in vertebrates (Hofmann & Mysels, 1988; Russell, 2003). Recently, cholate has been identified as a plant elicitor, thereby adding a completely new function to this bile salt (Shimizu et aI., 2008). Steroids enter the environment via decay of and excretion from eukaryotic organisms. Bile salts are mainly released by fecal excretion;

in humans, this excretion is in the range of 300 600 mg

per day and person (Ridlon et aI., 2006). In bacteria, steroids occur only as a rare exception, but many bacteria are capable of transforming and degrading steroid com pounds (for recent reviews, see Horinouchi et aI., 201Oa;

Philipp, 2011). As steroids are ubiquitous and abundant in the environment, bacterial steroid degradation is an impor tant part of the CO2 releasing site of the global carbon cycle.

Bacterial degradation is particularly important for the degradation of natural and synthetic steroid hormones, which can influence the fertility of animals as endocrine disruptors (Carson et aI., 2008; Combalbert & Hernandez Raquet, 2010). Furthermore, bacterial transformation of steroids is an essential part of the production of steroid drugs in biotechnology (Bortolini et al., 1997; Mahato &

Garai, 1997).

Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-179447

(2)

Despite the ecological and biotechnological importance of bacterial steroid metabolism, the knowledge of this process is scarce compared with the bacterial metabolism of for example aromatic compounds. Only recently has interest in bacterial steroid degradation increased considerably since it was found that Mycobacterium tuberculosis utilizes host cholesterol during infection (Pandey & Sassetti, 2008; Hu et aI., 2010).

We study bacterial steroid degradation using the bile salt cholate (compound I in Fig. 1) as a model compound and Pseudomonas sp. strain Choll as a model organism. Strain Choll initiates cholate degradation by oxidation of the A ring and

P

oxidation of the acyl side chain (Fig. 1). By these reactions, cholate is converted into 7,12 dihydroxy andros ta 1,4 diene 9,17 dione (DHADD, VIII) and its subsequent degradation product 3,7,12 trihydroxy 9,10 secoandrosta 1,3,5(lO)triene 9,17 dione (THSATD, IX; Philipp et aI., 2006). THSATD is then degraded to CO2 via the so called 9,10 seco pathway (Philipp, 2011).

o

We have studied

p

oxidation' of the acyl side chain of cholate by characterization of the transposon mutant strain R1, which is interrupted in a gene (acad) encoding an acyl CoA dehydrogenase (Birkenmaier et aI., 2007). This defect causes cholate degradation to stop at the intermediate 71X,121X dihydroxy 3 oxopregna 1,4 diene 20 carboxylate (DHOPDC, XIII), which has a C3 acyl side chain, indicating the removal of an acetyl residue from the Cs acyl side chain of cholate. A prerequisite for

P

oxidation of carboxylic acids is the forma tion of CoA esters. In agreement with this, we have detected the enzymatic formation of cholyl CoA and four further CoA esters of degradation intermediates of cholate, including L'1¹,4

3 ketocholyl CoA (II) and DHOPDC CoA (VI), in cell ex tracts of strain Choll (Birkenmaier et aI., 2007). The reaction steps preceding and following the formation of DHOPDC CoA have, to our knowledge, not been detected so far. For elucidating

P

oxidation of the acyl side chain of cholate further, we continued our screening of transposon mutants that showed an altered growth with cholate.

S-CoA

o

acad

OH 0

VIII:71~

ot::CC

+ Proplonyl-CoA

:" d=--

^VI

1 o~.

⁰

~

S·CoA

A-ring oxidation and CoA activation

'~ (V) ¹

o

skt

S·CoA

Further degradation to CO₂

Fig. 1. Section of the proposed pathway of cholate (compound I) degradation in Pseudomonas sp. strain Chol1. The following compounds have been identified: II. 1l',4 3 ketocholyl CoA; VI, CoA ester of 7!J., 12!J. dihydroxy 3 oxopr~gna 1,4 diene 20 carboxylate (DHOPDC); VIII, 7,12 dihydroxy androsta 1,4 diene 9,17 dione (DHADD); IX, 3,7,12 trihydroxy 9,10 secoandrosta 1,3,5(1 O)triene 9,17 dione (THSATD); X, 1l',4 3 ketocholate (P3 in Fig. 4); XI, (22E)7!J., 12!J. dihydroxy 3 oxochola 1 ,4,22E triene 24 oate (DHOCTO; P1 in Fig. 4); XII, 7!J.,12!J.,22 trihydroxy 3 oxochola 1,4 diene 24 oate (THOCDO; P2 in Fig. 4); XIII, DHOPDC. The CoA esters III, IV, V and VII (in brackets) have not been detected as yet.

(3)

Materials and methods

Bacteria, media and growth experiments Pseudomonas sp. strain ChoU and mutant strains derived from it were grown in the phosphate buffered mineral medium MMChol as described previously (Philipp et aI., 2006). The transposon mutant strain G 12 and strain ChoU KO[skt] (with and without the plasmid pBBR1MCS 5) were grown in the presence of kanamycin (10 ~lg mL -1) and gentamycin (20 Ilg mL -1), respectively. Growth experiments were carried out as described previously (Philipp et aI., 2006;

Birkenmaier et aI., 2007).

Transposon mutagenesis

Pseudomonas sp. strain ChoU was subjected to random transposon mutagenesis by insertion of the transposon mini Tn5 Km1 and screened for transposon mutants show ing altered growth with cholate as described previously (Birkenmaier et aI., 2007). Transposon insertions were identified by screening a gene library of strain G12 in Escherichia coli strain JM109 for kanamycin resistant clones as described previously (Birkenmaier et aI., 2007).

Construction and complementation of the mutant strain Chol1-KO[skt]

For the construction of the mutant strain ChoU KO[skt]

genomic DNA of strain Chol1 was purified as described previously (Jagmann et aI., 2010) and used as a template to amplify an internal fragment of skt using the primers KOskt F1 (5' CGATGGGGCCGGACGAAGAC 3') and KOskt R1 (5' TGCCGCGCCAGGTGAGGTC 3') by PCR. The ampli con was ligated into the vector pMBL T/A (Genaxxon). The resulting vector was digested with SpeI and PstI, and the internal skt fragment was ligated into the SpeIlPstI digested and dephosphorylated suicide vector pKnockout G (Wind gas sen et aI., 2000). The' resulting vector was transformed into E. coli strain S 17 1 and conjugated into strain Choll by biparental mating as described previously (Jagmann et aI., 2010). Insertional mutants were selected on MMChol agar plates (Philipp et al., 2006) containing 12 mM Na2 succi nate, 2 mM Na cholate and 20 ~lg mL ^-¹gentamycin. Vector insertion was verified by PCR using the vectors PKO G (5' GCGCGTTGGCCGATTCATTA 3') and KOskt R1. For complementation of strain Choll KO[skt], the skt gene was amplified from genomic DNA of strain Choll using the primers SktFl (5' CCCCGGCTGGCACCTTTGAACC 3') and SktRl (5' CGGCGCGGAAATCTCGGTCATCAC 3').

The amplicon was further processed using the TA cloning Kit (Invitrogen) as described previously (Birkenmaier et aI., 2007). The skt gene was excised from the cloning vector by digestion with HindIIlXhoI and ligated into vector

pBBRMCS 5 (Kovach et al.; 1995) digested with the same enzyme combination. The resulting vector pBBRIMCS 5[skt] was transformed into E. coli strain DH5()( and conjugated into strain Choll KO[skt] by triparental mating as described previously (Birkenmaier et aI., 2007).

HPLC analysis

Analysis was performed with an HPLC system described previously (Jagmann et al., 2010) using K Na phosphate buffer (10 mM, pH 7.1) and acetonitrile as eluents A and B, respectively. A gradient was applied, starting with 20% B (0 2 min), increasing to 50% B (2 16 min) and returning to 20% B within 1 min, followed by an equilibration of 4 min.

Preparation of steroid compounds

Steroid compounds were purified from culture supernatants by organic extraction and preparative HPLC analysis as described previously for DHOPDC (Birkenmaier et aI., 2007). DHADD and THSATD containing supernatants for growth experiments were prep~red as described previously (Philipp et aI., 2006).

Liquid chromatography-coupled to MS (LC-MS) MS analysis was performed on an LTQ Orbitrap Discovery LC MS/MS (Thermo Scientific) using nano electrospray in the positive ion mode. Chromatographic separation was performed using a nano HPLC system (Eksigent) equipped with a C18 column (Hypersil Gold C18, Thermo Scientific, particle size: 51lm; length: 100 mm; ID: 0.075 mm) using 0.1 % formic acid in water and 0.1 % formic acid in acetoni tril as eluents. The mass spectrometer was operated in the data dependent mode to automatically switch between orbitrap MS and ion trap MS/MS (MS2) acquisition. Sur vey full scan MS spectra (from mlz 350 1800) were acquired in the orbitrap with resolution R = 30 000 at mlz 400 (after accumulation to a target value of 1000000 charges in the linear ion trap). The five most intense ions were sequentially isolated and fragmented in the linear ion trap using colli sionally induced dissociation at a target value of 100000 charges. For accurate mass measurements, the lock mass option was enabled in the MS mode and the polydimethyl cyclosiloxane ions generated in the electrospray process from ambient air [protonated (Si(CH3lzO))6; mlz= 445.12003] were used for internal recalibration in real time.

Target ions already selected for MS/MS were dynamically excluded for 30 s. General MS conditions were: electrospray voltage, 2.3 kV; no sheath and auxiliary gas flow; ion transfer tube temperature: 110 DC; collision gas pressure: 1.3 mT; and normalized collision energy: 35% for MS2. The ion selection threshold was 500 counts for MS2.

(4)

Nuclear magnetic resonance (NMR) spectroscopy

NMR measurements were carried out with HPLC purified DHOCTO and THOCDO dissolved in DzO to a final concentration of c. 100 500 11M. All NMR spectra were recorded at 300 K on a Bruker AVANCE III 600 MHz spectrometer equipped with a 5 mm TCI H/C/N cryoprobe with an actively shielded Z gradient. The proton 1D spectra were recorded with a spectral width of 16 p.p.m. and 32k complex points. Residual HDO was suppressed by presa turation during the recycle delay of 2 s. Homonuclear 2D COSY, TOCSYand NOESYexperiments were recorded with 4k complex points in the detected and 256 complex points in the indirect dimension. TOCSY spin lock was achieved with MLEV17 at a 10 kHz field strength and a duration of 80 ms. The HSQC and HMBC spectra were recorded with 1k and 4k complex points, respectively, in the proton dimen sion and 256 complex points in the l3C dimension. The spectral width in the carbon dimension was 170 p.p.m. and 180 p.p.m., respectively. All spectra were processed and analyzed using Bruker's TOPSPIN (v3.0) software. Usually, zero filling was applied to double the number of real points in each dimension. Chemical shifts were referenced to the HDO resonance at 4.7 p.p.m. Chemical shift assignments for 13C were determined indirectly from HSQC and HMBC spectra.

Results and discussion

Characterization of the transposon mutant strain G12

Pseudomonas sp. strain Choll was subjected to random transposon mutagenesis by insertion of the transposon mini Tn5 Km1 and screened for transposon mutants show ing altered growth with cholate as described previously (Birkenmaier et aI., 2007). One mutant, strain G12, was analyzed further. Strain G12 could not grow with cholate as the sole substrate, but it could grow with succinate in the presence of cholate. HPLC analysis of supernatants from these cultures revealed that strain G 12 did not transform cholate at all. We then checked whether strain G 12 could grow with intermediates of cholate degradation. With supernatants containing DHADD (VIII), strain G12 could grow after a long lag phase. Notably, cells of strain G12 induced for growth with DHADD were also induced for cholate transformation during growth with succinate in the presence of cholate. HPLC analysis revealed that cholate was transformed into several compounds with an absorption maximum at 244 nm, which is indicative of steroids with a 3 keto 1,4 diene structure of the A ring (Philipp et ai., 2006).

Identification and inactivation of the skt gene In the next step, we identified the gene in strain G 12, in which the mini Tn5 Km1 had been inserted. The transpo son was inserted into an ORF of 1212 bp at bp 333. The predicted protein had 403 amino acids and showed high identity to nonspecific lipid transfer proteins from various bacteria. Among these were two bacteria, for which growth with cholate had been demonstrated, namely Pseudoalter omonas haloplanktis strain TAC125 (Birkenmaier et aI., 2007) and Comamonas testosteroni strain KF 1 (Rosch et aI., 2008). The nonspecific lipid transfer proteins from strains TAC125 and KF 1 showed 80% and 68% identity, respectively, to the gene product from strain Choll (Fig. 2).

This gene was named skt (for steroid

P

ketothiolase) for reasons that will be described below. To investigate the function of skt for cholate degradation further, we decided to construct a defined mutant of this gene by subjecting strain Choll to insertional mutagenesis with the suicide vector pKnockoutG. The resulting strain Choll KO[skt]

could not grow with cholate; growth with cholate was restored when an intact copy of skt was provided in trans on the vector pBBR1MCS 5 (Fig. 3a). This complementa tion clearly showed that the phenotype of this mutant was caused by the inactivation of skt. Strain Choll KO[skt]

could grow with succinate in the presence of cholate (Fig. 3b). HPLC analysis of supernatants from these cultures showed that strain Choll KO[skt] transformed cholate into several products with an absorption maximum at 244 nm (Fig. 4), which were identical to the aforementioned pro ducts in culture supernatants of the transposon mutant strain G12. Notably, in contrast to strain G12, strain Choll KO[skt] performed this conversion without prior induction through growth with DHADD. The reason for this differ ence between strains G12 and Choll KO[skt] is not known.

Identification of the steroid compounds P1, P2 and P3

Among the accumulating products, one peak, PI, was dominant (Fig. 4). This compound had a second UV absorption maximum around 210 nm in addition to the maximum at 244 nm. A further compound with a UV absorption maximum at 244 nm eluted very close to PI, thereby causing a shoulder tailing off from the PI peak. As a better separation of these two compounds could not be achieved, it is likely that they have a very similar structure. A relatively small peak, P2, eluted several minutes earlier than all other products. This compound occurred in low amounts and was relatively unstable. Compounds PI and P2 were purified and analyzed by NMR and MS.

As sample PI contained a slight amount of impurities from the compound eluting very close to it and as sample P2 had a relatively low concentration, the de novo chemical shift

(5)

Fig. 2. Alignment of the amino acid sequences of the putative ~ ketothiolase Skt in Pseudomonas sp. strain Choll, of the putative ~ ketothiolases encoded by the genes PSHAa0897 in Pseudoalter omonas ha/oplanktis TAC 125 (80% identity) and PD3654 in Comamonas testoseroni strain KF 1 (68% identity) and of the human SCP X protein (GenBank accession number AAA03557.1).

Alignment was performed using the software CLUSTAlW2. Residues identical in all three bacterial proteins are shaded. Residues identical in all four proteins are indicated by"'; conserved and semi conserved substitutions are indicated by':' and '.', respectively. The GenBank nucleotide se quence accession number for skt is HM776520.

Choll TAC125 KF 1 SCP )(

····PE;kY'ATt~Rf-iE'Hi..EQVA\:iisAARALDpEGLKAEDLD@i,i 53 EEK;{QTl\J:\QMFHLEQVADDANQALLDAGLQKSDLDG:(l 53 MGLQgKAI1LZ@V !jQ~:iATAg~JJEQV!iDLTLQI1~EMGMELsEvg§J;J 53 MSSSPWEPATLRRVFVVGVGMTKFVKPGAENSRDYPDLAEEAGKKALADAQIPYSAVDQA 60

:~* ** : VING,PfllfflEA.S, ,Vf

m,

N'll\A. )§.,.YLqLAV~F'AF,VVD~GqcTGMGMV~!W'AAtl'i.LGL," Co

VING,PflFHEl\.SYFVP~l\.l\.~ifLqI EVNFl\.?VVDLGciCTA VAQ I\'lMAAA~ 8):.GL

8

ITSAmf.lil'il'.sCIfVP~q;X)::gVRL~fl\.?VJLQ1&"ASSVAMVW~;l;S;!oG.L. 9

CVG YVFGDSTCGQRAIYHSLGMTGIPIINVNNNCATGSTALFMARQLIQGGVAECVLA 113 113 113 118

VMtA~AP}1GPDEQ PTAMMRAMRFGGH$"'.ilNYp FQIiPYGHM 169

*f'A~AP,MAPNEq PAWMAQ/:IMRYGGHSToAFpAPE DLPYI:iH~G:Q ¹⁶⁹

MLgs\3!:1AP,ISEHDDHIKEVMKASRFGGHS'r:RFp,\PEI'. DLPYGI;\r-jAQ D!}Q 170 LGFEKMSKG SLGIKFSDRTIPTDKHVDLLINKYGLSAHPVAPQMFGYAGK 168

RYiI\f\RYG;{DPV~ ~iiQI'.NPE!iIFCRQPL'I11 DDV!,N:S;Rj1iJAPPLH. 229 RY!~QYGYDQA~ ~!\QYHKDl'iIFK~KEL'I\II'.QVIJ!,SK~VADPLI:L 229

~~GhVHGKQAA~ ~ACHNPD8MFYBQPI~VDDVMNs'RMY ~ 230 EHMEKYGTKIEHFAKIGWKNHKHSVNNPYSQFQD EYSLDEVMASKEVFDFLTILQCCPT 227

.. . *. * . * * * .* . .. . . .

VASl~A!'1V1V'ASKELMARARN~GA WTGFOJ§rLG Y+<SPSy!S Pit:i*EtI)VG PMRS 283 vsp.qA;flv:rV'ASKEVAARAKK,~PAFIT13FfiE;RLSFf'S~SY:AQPI1TV7'PVA Ei\AKR 283 AApgG)\MI'{TRADRARTTRHSPVSIVdCGJ:1'HVSS~.sgT:(MAQ.I1LQ!l'PIG P~SAK 284 SDGAAAAILASEAFVQKYGLQSKAVEILI'.QEMMTDLPSSFEEKSIIKMVGFDMSKEAARK 287

. *. * *. .. ..

343 343 344 347

,GGQ1.SFGQAGAt;,G(J HQrAGRAGDReDKSCQNVFVT 386

pGQLSFGQAAAi\'Gg 'HQISNsNDQRQrJKRCbNVFVS 386 9QQi:!$E;qqSGTl>§g ~QJJQGR~GD~Q.*RNQLAYVS 387 GGLISKGHPLGATGLAQCAELCWQLRGEAGKRQVPGAKVALQHNLGIGGAWVTLYKMGF 407

PEAASSFRTHQIEAVPTSSASDGFKANLVFKEIEKKLEEEGEQFVKKIGGIFAFKVKDGP 467

QGl\L:r:i!Q't( 402 QG1\r.:n;QQ 402

QGAr.:l~R.Q 403 GGKEATWVVDVKNGKGSVLPNSDKKADCTITMADSDFLALMTGKMNPQSAFFQGKLKITG 527

A 403

G 403

A 404

NMGLAMKLQNLQLQPGNAKL 547

assignment was difficult. However, the NMR spectra of both compounds showed high similarities in their ~ 1,4 3 keto cholate framework such that the assignment of the four steroid rings was facilitated by comparison with the chemi cal shift assignment of DHOPDC (Birkenmaier et ai., 2007) (Table 1). Compound PI contains an additional unsatura tion, whereas compound P2 contains an additional hydroxyl group. Both modifications do not affect the pattern of chemical shifts of the four steroid rings. The attachment of the hydroxyl group ofP2 at C22 could be identified from the characteristic HSQC crosspeak at 4.09170.5 p.p.m. and cor relations, from COSY, TOCSY and HMBC, into the side

chain and ring D. Compound PI exhibits an additional C C double bond with chemical shifts of 5.82/118.3 p.p.ill. and 6.931157.4 p.p.m., respectively. The location of this olefinic group could be established again from its correlations within the side chain and to the D ring. According to the scalar coupling of 15 Hz between the olefinic protons, the double bond has an E configuration. The absolute configuration at C20 (PI, P2) and C22 (P2) could not be determined because of insufficient amount of sample. The stereospecific assign ments at C6, C7, CII, C12, Cl5 and Cl6 were carried out according to their similarity of chemical shifts as compared with DHOPDC (Birkenmaier et ai., 2007).

(6)

1.5, -- - - -- , 1.25

E c:

~ 0.75

o

o ^0.5

0.25 (a)

O~~~~~~uu~u.~

E

c:

&

Q)

u c:

1l g

£J

'"

^Q)_>

~ Qi

a:

o

10

.k

2 4

20 Time (h)

30

P2

f ' .

6 8

Time (min) 40

1.5 1.25

(b)

E c:

~ 0.75

0

•

0 _0.5

.;!

0.25

0

•••

0 10

P1

P3

10 12 14

Fig. 4. HPLC chromatogram of a culture supernatant of Pseudomonas sp. strain Choll KO[skt] grown with succinate in the presence of cholate.

The products Pl, P2 and P3 were identified as DHOCTO (XI in Fig. 1), THOCDO (XII) and ,..1.43 ketocholate (X), respectively. The same pro ducts accumulated in the culture supernatants of the transposon mutant strain G 12 grown with succinate in the presence of cholate.

According to these NMR spectroscopic data, Pi was identified as (22E) 7ex,12ex dihydroxy 3 oxochola 1,4,22 triene 24 oate (DHOCTO, XI) and P2 was identified as 7ex, 12ex,22 trihydroxy 3 oxochola 1,4 diene 24 oate (THOCDO, XII).

Analysis by LC MS revealed ions [M

+

HJ + with mlz 401.23 and mlz 419.24 for Pi and P2, respectively. These spectroscopic data are in agreement with a molecular mass of 400Da for DHOCTO (C21H320S) and 418Da for THOCDO (C24H3106), and thus confirm the results of NMR spectroscopy. While an analogue of DHOCTO had been detected in an earlier study on deoxycholate degrada tion by another Pseudomonas sp. (Leppik, 1983), a structure similar to THOCDO has, to our knowledge, not been described in any study on bacterial degradation of bile salts.

Within the shoulder tailing off from the DHOCTO peak (Fig. 4), LC MS analysis revealed an ion [M

+

HJ + with mlz

•••

H!

_{Fig. 3.}Growth of Pseudomonas sp. strains Choll (e), Choll KO[skt] (_) and Choll KO[skt]

pBBR1[skt] (A) with (a) 2 mM cholate and (b) 20 30 40 2 mM cholate and 12 mM succinate as sources of Time (h) carbon and energy. Error bars indicate SD (n = 3).

Table 1. IH and 13C chemical shifts of DHOCTO (XI in Fig. 1 and Pl in Fig. 4) and THOCDO (XII in Fig. 1 and P2 in Fig. 4)

DHOCTO THOCDO

Atom no. BC (p.p.m.) BH (p.p.m.) BC (p.p.m.) oH (p.p.m.)

1 160.15 7.40 160.2 7.34

2 125.6 6.30 125.5 6.23

3 188.9 188.8

4 125.1 6.174 124.8 6.12

5 171.4 171.5

6- 40.0 2.83 (P)/2.53 (ex) 40.2 2.77 (P)/2.46 (ex)

7t 69.7 4.13 69.8 4.07

8 38.7 1.89 38.9 1.82

9 38.5 1.73 38.4 1.67

10 44.1 43.9

11- 29.4 2.07 (P)/l.85 (ex) 29.5 1.78 (P)I1.99 (ex)

12' 72.2 4.09 72.3 4.02

13 41.7 46.0

14 41.8 1.68 41.8 1.64

15~ 22.5 1.67 (P)/l .25 (ex) 22.4 1.64 (P)/l .19 (ex) 16~ 26.4 1.67 (P)/l .28 (ex) 26.2 1.86 (P)/l.31 (ex)

17 45.7 1.79 43.6 1.85

18 11.9 0.82 11.5 0.72

19 17.1 1.25 17.1 1.18

20 38.9 2.37 39.6 1.41

21 17.3 1.09 9.9 0.87

22~,1I 157.4 6.93 70.5 4.09

23 118.3 5.82 41.8 2.42/2.30

24 171.2 178.9

-The stereochemical assignment was made with reference to DHOPDC.

tThe 7 OH group is in the ex configuration. This stereochemical assign ment was made with reference to DHOPDC.

'The 12 OH group is in the ex configuration. This stereochemical assign ment was made with reference to DHOPDC.

'Because of signal overlap, the assignment of C 15 and C 16 as well as the stereochemical assignments were made with reference to DHOPDC.

'IThe C22 C23 double bond of DHOCTO is in the E configuration ('h122 ^H23= 15 Hz).

liThe absolute configuration of the stereogenic center C22 in THOCDO has not been determined.

As indicated in the footnotes, most of the stereochemical assignments were made with reference to DHOPDC (Birkenmaier et al., 2007).

(7)

403.24, which could be the monoene derivative of DHOCTO.

With the transposon mutant strain Rl, we also observed the transient accumulation of the monoene derivative ofDHOPDC in the culture supernatants (Birkenmaier et aI., 2007).

The compound P3, which formed the second largest peak in HPLC analysis, coeluted with and had a UV spectrum very similar to that of the previously identified ~ 1,4 3 ketocholate (X) from culture supernatants of the transposon mutant strain Rl (Birkenmaier et al., 2007).

Characterization of strain ChoI1-KO[skt] and the sktgene

To characterize the mutant strain Choll KO[skt] further, it was tested for growth with intermediates of cholate degra dation. Strain Choll KO[skt] could grow with DHADD (VIII) and THSATD (IX). Importantly, strain Choll KO[skt] could also grow with DHOPDC (XIII) that was provided with filter sterilized supernatants of a culture of the acad mutant strain Rl that had been grown with succinate in the presence of cholate as described previously (Birkenmaier et aI., 2007).

The growth of strain Choll KO[skt] with DHOPDC clearly showed that the skt gene must be responsible for a reaction step preceding the formation of DHOPDC. The accumulation of DHOCTO and THOCDO supports this conclusion because both compounds could have arisen from hydrolyzed CoA esters III and IV that are presumptive intermediates of

P

oxidation of the acyl side chain of cholate (Fig.

O.

Thus, the accumulation of DHOCTO and THOC DO indicates that at least the first two steps of

P

^oxidation

starting from ~ ^1,43 ketocholyl CoA (II) could be catalyzed in the skt mutant. This narrowed the probable function of the skt encoded protein down to being either a 3 hydroxy acyl CoA dehydrogenase or a

P

ketothiolase.

A closer analysis of the predicted protein reveals that Skt and its orthologs in other cholate degrading bacteria (Fig. 2) have similarities to the

P

ketothiolase domain of eukaryotic sterol carrier protein SCP x (Stolowich et al., 2002). SCP x, which is also referred to as a nonspecific lipid transfer protein, is a fusion protein with a smaller C terminal and a larger N terminal domain. While the C terminal domain (also called the SCP 2 domain) is responsible for intracel lular targeting and the uptake of sterols, the N terminal domain has 3 ketoacyl CoA thiolase activity for branched chain acyl CoA esters. Interestingly, SCP x is also responsi ble for the final step of cholate biosynthesis in mammals (Kannenberg et aI., 1999; Russell, 2003). There, it catalyzes the thiolytic release of propionyl CoA from 3cx,7cd2cx trihydroxy 5

P

cholest 24 one CoA. Phylogenetic analyses of eukaryotic SCP x thiolase domains reveal that they are related to putative thiolases encoded in proteobacterial genomes (Pereto et aI., 2005).

Based on the phenotype of the skt mutant strains G12 and Choll KO[skt] and on the similarities to the SCP x thiolase domain, we conclude that the gene skt encodes a

P

ketothiolase that catalyzes the thiolytic release of acetyl CoA from the CoA ester of the so far presumptive 7,12 dihydroxy 3,22 dioxo 1,4 diene 24 oate (V). The reaction products would be DHOPDC CoA (VI), which has been detected in cell extracts of strain Choll previously (Birken maier et aI., 2007), and acetyl CoA. As the gene product of skt and its orthologs in the other cholate degrading bacteria mainly show similarities to the SCP x thiolase domain only and not to the SCP 2 domain of SCP x, the annotation of these putative proteins as nonspecific lipid transfer proteins is misleading. However, Skt and its orthologs have a highly conserved motif at their C terminus that is very similar to two short motifs within the sterol binding SCP 2 domain of the human SCP x (Fig. 2), suggesting that this region of the bacterial proteins might be involved in interacting with the steroid skeleton of cholate.

Regarding the function ofSkt, it appeared surprising that DHOCTO was the major accumulating product because one would rather expect 7,12 dihydroxy 3,22 dioxo 1,4 diene 24 oate (DHDODO), the presumptive hydrolysis product of CoA ester V, to accumulate as a dead end metabolite.

DHDODO is a

P

ketoacid, which is prone to spontaneous decarboxylation. However, we did not detect DHDODO or a presumptive decarboxylation product in our analyses.

Thus, the fact that DHOCTO was the major accumulating compound suggests that blocking

P

^oxidation^at^{the last}

step causes a negative .feedback inhibition on the previous enzymatic steps. As a consequence, the CoA esters of DHOCTO and THOCDO are hydrolyzed and the free bile salts are released. In our earlier study on the transposon mutant strain Rl, we had never detected DHOCTO or THOCDO in culture supernatants (Birkenmaier et aI., 2007). This indicates that the conversion of ~ 1,4 3 ketocho Iyl CoA (II) to DHOPDC CoA (VI) may proceed in a tightly controlled canalized process without a significant release of degradation intermediates. In agreement with this hypothesis, it is also believed that

P

oxidation of fatty acids occurs by substrate channelling in multienzym~ complexes (Kunau et al., 1995; Pereto et aI., 2005).

Our study is a further step towards the verification of the pathway for the

P

oxidation of the acyl side chain of cholate by strain Choll. To elucidate this reaction sequence further, biochemical investigations regarding the formation and metabolism of the respective CoA esters of DHOCTO and THOCDO are under way in our laboratory. We have now identified two genes, acad and skt, that encode proteins required for this part of cholate degradation. As genes encoding proteins very similar to Skt and ACAD of strain Choll occur in different strains of C. testosteroni (Horinou chi et aI., 2010b) and in P. haloplanktis strain TACI25, it is

(8)

likely that the same pathway for steroid degradation prevails in these organisms as well. Recently, the thiolase FadA5 from M. tuberculosis H37Rv has been shown to be involved in the degradation of the side chain of cholesterol (Nesbitt et aI., 2010). According to the Conserved Domain Database (CCD; Marchler Bauer et aI., 2009), FadA5 and Skt fall into different subfamilies of the thiolase superfamily (subfamily cd00751 for FadA5 and subfamily cd0829 for Skt), indicat ing that Fad5A might be involved in a different step of steroid side chain oxidation.

Acknowledgements

The authors thank Anke Friemel for excellent assistance with NMR analysis and Andreas Marquardt for performing LC MS analysis. The authors acknowledge Kathrin Happle and Antje Wiese for technical assistance and Bernhard Schink for continuous support. This work was funded by grants from the Deutsche Forschungsgemeinschaft (DFG;

PH7113 1; TP B9 in SFB454) and the University of Konstanz (AFF project 58/03) to B.P.

References

Birkenmaier A, Holert J, Erdbrink H, Moller HM, Friemel A, Schoenenberger R, Suter MJ, Klebensberger J & Philipp B (2007) 'Biochemical and genetic investigation of initial reactions in aerobic degradation of the bile acid cholate in Pseudomonas sp. strain Choli. J Bacteriol189: 7165 7173.

Bortolini 0, Medici A & Poli S (1997) Biotransformations on steroid nucleus of bile acids. Steroids 62: 564 577.

Carson JD, Jenkins RL, Wilson EM, Howell W & Moore R (2008) Naturally occurring progesterone in loblolly pine (Pinus taeda

L.): a major steroid precursor of environmental androgens.

Environ Toxicol Chern 27: 1273 1278.

Combalbert S & Hernandez Raquet G (2010) Occurrence, fate, and biodegradation of estrogens in sewage and manure. Appl Microbiol Biot 86: 1671 1692.

Hofmann AF & Mysels KJ (1988) Bile salts as biological surfactants. Colloids surface 30: 145 173.

Horinouchi M, Hayashi T & Kudo T (20IOa) Steroid degradation in Comamonas testosteroni. J Steroid Biochem DOl: 10.10161 j.jsbmb.20 10.1 0.008.

Horinouchi M, Kurita T, Hayashi T & Kudo T (2010b) Steroid degradation genes in Comamonas testosteroni TA441: isolation of genes encoding ad4(5) isomerase and 3()( and 3~

dehydrogenases and evidence for a 100kb steroid degradation gene hot spot. J Steroid Biochem 122: 253 263.

Hu Y,'van der Geize R, Besra GS, Gurcha SS, Liu A, Rohde M,

Sin~h M & Coates A (2010) 3 Ketosteroid 9alpha hydroxylase is an essential factor in the pathogenesis of Mycobacterium tuberculosis. Mol Microbiol75: 107 121.

Jagmann N, Brachvogel HP & Philipp B (2010) Parasitic growth of Pseudomonas aeruginosa in co culture with the chitinolytic

bacterium Aeromonas hydrophila. Environ Microbial 12:

1787 1802.

Kannenberg F, Ellinghaus P, Assmann G & SeedorfU (1999) Aberrant oxidation of the cholesterol side chain in bile acid synthesis of sterol carrier protein 2/sterol carrier protein x knockout mice. J Bioi Chern 274: 35455 35460.

Kovach ME, Elzer PH, Hill SD, Robertson GT, Farris MA, Roop

RM & Peterson KM (1995) Four new derivatives of the broad

host range cloning vector pBBRIMCS, carrying different antibiotic resistance cassettes. Gene 166: 175 176.

Kunau WH, Dommes V & Schulz H (1995) ~ Oxidation offatty acids in mitochondria, peroxisomes, and bacteria: a century of continued progress. Prog Lipid Res 34: 267 342.

Leppik RA (1983) Deoxycholic acid degradation by a Pseudomonas sp. acidic intermediates with A ring unsaturation. Biochem J 210: 829 836.

Mahato SB & Garai S (1997) Advances in microbial steroid biotransformation. Steroids 62: 332 345.

Marchler Bauer A, Anderson JB, Chitsaz F et al. (2009) CDD:

specific functional annotation with the Conserved Domain Database. Nucleic Acids Res 37: D205 D210.

Nesbitt NM, Yang X, Fontan P, Kolesnikova l, Smith 1, Sampson

NS & Dubnau E (2010) A thiolase of Mycobacterium

tuberculosis is required for virulence and production of androstenedione and androstadienedione from cholesterol.

Infect Immun 78: 275 282.

Pandey AK & Sassetti CM (2008) Mycobacterial persistence requires the utilization of host cholesterol. P Natl Acad Sci USA 105: 4376 4380.

Peret6 J, Lopez Garcia P & Moreira D (2005) Phylogenetic analysis of eukaryotic thiolases suggests multiple proteobacterial origins. J Mol Evo161: 65 74.

Philipp B (2011) Bacterial degradation of bile salts. App/

Microbiol Biot 89: 903 915.

Philipp B, Erdbrink H, Suter MJ & Schink B (2006) Degradation of and sensitivity to cholate in Pseudomonas sp. strain Choli. Arch Microbiol185: 192 201.

Ridlon JM, Kang DJ & Hylemon PB (2006) Bile salt

transformations by human intestinal bacteria. J Lipid Res 47:

241 259.

Rosch V, Denger K, Schleheck D, Smits THM & Cook AM (2008) Different bacterial strategies to degrade taurocholate. Arch Microbiol190: 11 18.

Russell DW (2003) The enzymes, regulation, and genetics of bile acid synthesis. Annu Rev Biochem 72: 137 174.

Shimizu T, Jikumaru Y, Okada A et al. (2008) Effects of a bile acid elicitor, cholic acid, on the biosynthesis of diterpenoid phytoalexins in suspension cultured rice. Phytochemistry 69:

973 981.

Stolowich NJ, Petrescu AD, Huang H, Martin GG, Scott AI &

Schroeder F (2002) Sterol carrier protein 2: structure reveals function. Cell Mol Life Sci 59: 193 212.

Windgassen M, Urban A & Jaeger KE (2000) Rapid gene inactivation in Pseudomonas aeruginosa. FEMS Microbial Lett 193: 20 I 205.