Critical role of the plasma membrane for ... - Wiley Online Library

Eur. J. Biochem. 270, 1502–1514 (2003) FEBS 2003

doi:10.1046/j.1432-1033.2003.03516.x

Critical role of the plasma membrane for expression of mammalian mitochondrial side chain cleavage activity in yeast Catherine Duport1,*, Barbara Schoepp1,†, Elise Chatelain1, Roberto Spagnoli2, Bruno Dumas3 and Denis Pompon1 1

Laboratoire d’Ingeńierie des Proteínes Membranaires, CGM du CNRS, Gif sur Yvette, France; 2Lead Discovery Technologies, Aventis Pharma, Romainville, France; 3Functional Genomics, Aventis Pharma, 13 Quai Jules Guesde, F-94403 Vitry sur Seine, France

Engineered yeast cells efficiently convert ergosta-5-eneol to pregnenolone and progesterone provided that endogenous pregnenolone acetylase activity is disrupted and that heterologous sterol D7-reductase, cytochrome P450 side chain cleavage (CYP11A1) and 3b hydroxysteroid dehydrogenase/isomerase (3b-HSD) activities are present. CYP11A1 activity requires the expression of the mammalian NADPH-adrenodoxin reductase (Adrp) and adrenodoxin (Adxp) proteins as electron carriers. Several parameters modulate this artificial metabolic pathway: the effects of steroid products; the availability and delivery of the ergosta5-eneol substrate to cytochrome P450; electron flux and protein localization. CYP11A1, Adxp and Adrp are usually located in contact with inner mitochondrial membranes and are directed to the outside of the mitochondria by the

removal of their respective mitochondrial targeting sequences. CYP11A1 then localizes to the plasma membrane but Adrp and Adxp are detected in the endoplasmic reticulum and cytosol as expected. The electron transfer chain that involves several subcellular compartments may control side chain cleavage activity in yeast. Interestingly, Tgl1p, a potential ester hydrolase, was found to enhance steroid productivity, probably through both the availability and/or the trafficking of the CYP11A1 substrate. Thus, the observation that the highest cellular levels of free ergosta-5-eneol are found in the plasma membrane suggests that the substrate is freely available for pregnenolone synthesis.

The large family of mammalian cytochrome P450 enzymes includes drug metabolizing enzymes and enzymes that mediate individual steps in the biosynthesis of biologically active compounds. Our interest is focused on the cyto-

chrome P450 enzymes that are involved in the synthesis of steroid hormones. These steroids are critical for mammalian life and are involved in such distinct processes as stress response, immunosuppression, ion balance, general metabolite homeostasis and fetal, neonatal and gonadal development [1]. Eukaryotic steroidogenic cells produce a large array of steroids using a limited set of cytochrome P450 enzymes [2]. The biosynthesis of all hormonal steroids begins with the side chain cleavage (SCC) of cholesterol [3] to form pregnenolone, the key precursor of biologically active steroids in all tissues [4,5]. This reaction is catalysed by cytochrome P450scc (also designated CYP11A1 [6]), a mitochondrial protein located on the matrix face of the inner membrane that requires electrons for activity. These electrons are transferred from NADPH through a specific transport chain involving adrenodoxin reductase (Adrp) and adrenodoxin (Adxp) [7]. Adxp is a small soluble iron– sulfur protein localized to the mitochondrial matrix, and Adrp is a larger flavodoxin protein bound to the inner mitochondrial membrane of steroid-producing cells [8]. For many years, pregnenolone formation has been considered to be the rate-limiting step in steroidogenesis [9]. It has been shown that to initiate and sustain steroid production, a constant supply of cholesterol must be available in the cell. Furthermore, there must be a mechanism to ensure the delivery of this substrate to the site where it is cleaved in the inner mitochondrial membrane, where CYP11A1 resides. For example, substrate unavailability is a common cause of congenital lipoid adrenal hyperplasia, a disease characterized by a dramatic decrease in steroid synthesis [10]. In the

Correspondence to B. Dumas, Functional Genomics, Aventis Pharma, 13 Quai Jules Guesde, F-94403 Vitry sur Seine, France. Fax: + 33 1 5893 2625, Tel.: + 33 1 5893 2805, E-mail: [email protected] Abbreviations: ACAT, acyl coenzyme A cholesterol acyltransferase; Adrp, adrenodoxin reductase protein; Adxp, adrenodoxin protein; APAT, acetyl coenzyme A:pregnenolone acetyl transferase; ARE, acyl coenzyme A:cholesteryl acyltransferase-related enzyme; CYP2D6, cytochrome P450 2D6; CYP2E1, cytochrome P450 2E1; CYP11A1, cytochrome P450 steroid side chain cleaving; CYP11B1, cytochrome P450 steroid 11b hydroxylase; D7-Red, sterol D7 reductase; ER, endoplasmic reticulum; 3b-HSD, 3b hydroxysteroid dehydrogenase/isomerase; PM, plasma membrane; PGK1, phosphoglycerate kinase; StAR, steroidogenic acute regulatory protein; SCC, side chain cleavage; TEF1, transcription elongation factor. Enzymes: ACAT, EC 2.3. 2.26; Adrp, EC 1.18.1.2; CYP11A1, EC 1.14.15.6; CYP11B1, EC 1.14.15.4; 3b-HSD, EC 1.1.1.51. *Present address: University of Paris VII and UMR A408, INRA 84914 Avignon Cedex 09, France. Present address: Institut de Biologie Structurale et Microbiologie, 31 Chemin Joseph Aiguier, F-13402 Marseille, France. (Received 20 November 2002, revised 27 January 2003, accepted 11 February 2003)

Keywords: CYP11A1; plasma membrane; ergosta-5-eneol; Tgl1p.

FEBS 2003

CYP11A1 localizes to the yeast plasma membrane (Eur. J. Biochem. 270) 1503

latter case, mutation(s) in the steroidogenic acute regulatory protein (StAR) protein correlate(s) clearly with the absence of pregnenolone synthesis. The well-characterized StAR protein is involved in the rapid transport of cholesterol to the inner mitochondrial membrane [11]. In contrast, the factors and processes responsible for the intracellular supply of cholesterol to the outer mitochondrial membrane are poorly understood. It is known, however, that cholesterol is mobilized from cellular storage sites, such as lipid droplets, in response to trophic hormones [12]. This mobilization requires the enzyme cholesteryl esterase, which mediates the release of free cholesterol from cholesterol esters. In addition, the maintenance of cellular architecture requires a stringent regulation of the concentration of free cholesterol. This is ensured by the enzyme, acyl coenzyme A cholesterol acyltransferase (ACAT), which catalyzes its esterification [13]. The levels of expression of various proteins (Adxp, Adrp, CYP11A1) involved in the reaction can also affect the efficiency of cholesterol side chain cleavage. Indeed, it is apparent that the expression of these three proteins is differentially modulated in hormoneproducing tissues. For example, the corpus luteum and adrenal cortex contain higher concentrations of Adxp and Adrp than does the placenta ([14,15]). In the latter case, the concentration of Adrp limits the production of pregnenolone [15] through a mechanism involving oxidized Adxp, that is in excess in the human placenta [16]. Whether ternary complex formation is required for the optimal flow of electrons from NADPH to CYP11A1 remains controversial. However, recent reports reinforce the idea that there is a complex containing CYP11A1, cytochrome P450 11B1 (CYP11B1), Adxp and Adrp in the mitochondrial membrane of steroid producing cells [17,18]. As reported previously [19], the simultaneous expression of Arabidopsis thaliana sterol D7-reductase (D7-Red), bovine CYP11A1, Adxp, Adrp and human 3b-hydroxysteroid deshydrogenase/isomerase (3b-HSD) in modified Saccharomyces cerevisiae cells allows the self-sufficient biosynthesis of pregnenolone and progesterone (Fig. 1), thus reproducing the properties of steroidogenic tissues of higher eukaryotes. In these recombinant yeast strains, the predominant sterol is ergosta-5-eneol, that replaces ergosterol in membranes and acts as a substrate for CYP11A1. Ergosta-5-eneol differs from ergosterol in that the C7–C8 doublebond is reduced and there is no doublebond at position C22. It also differs from cholesterol in that it has a methyl group at position C24. Ergosta-5-eneol is synthesized [19] and esterified [20] in processes similar to those that control cholesterol accumulation in mammalian cells. Ergosta-5-eneol and cholesterol act similarly as a substrate for CYP11A1 and allow proper folding of CYP11A1 in membrane microdomain. The aim of this study was to determine whether recombinant yeast can be used as a model system to decipher the SCC reaction and its potential regulation during steroidogenesis. To do so, we studied the influence of ergosta-5-eneol availability and electron carrier expression level on the production of pregnenolone and progesterone, and we also determined the localization of the components of the SCC reaction. We found that Adxp, Adrp and CYP11A1 appear to localize to three compartments outside the mitochondrion, without impairing the reaction. This finding has direct implications for the

potential formation of a complex containing CYP11A1, CYP11B1, Adxp and Adrp.

Materials and methods Culture conditions and genetic methods Yeast media, including SG (synthetic medium containing 2% glucose), SL (synthetic medium containing 2% galactose) and YP (complete medium without carbon source) are described [65]. Low-density and high-density cultures were obtained as reported previously [19]. Standard methods were used for transformation [21] and genetic manipulation of S. cerevisiae [22]. pUC-HIS3ADX is an integrative plasmid derived from pUC-HIS3 [23] that carries the TEF1prom::matADX:: PGK1term expression cassette. This expression cassette contains the mature form of the ADX cDNA [24] under the control of the TEF1 promoter and PGK1 terminator [25]. The matADX expression plasmid pTG10917 contains an E. coli replicon with an S. cerevisiae replicon and a URA3 marker. The vector pUC18-HIS3 was linearized at the unique XhoI site in the intergenic region between the S. cerevisiae HIS3 and DDE1 genes and blunt-ended with the Klenow enzyme. A NotI linker was introduced into this linearized vector, giving pUC-HIS3N. The 1235-bp NotI fragment carrying the expression cassette, TEF1prom:: matADX::PGK1term, was isolated from pTG10917 (see above) and subcloned into the NotI site of pUC-HIS3N to obtain pUC-HIS3ADX. pYeDP60 is a 2l replication origin-based expression plasmid that contains the URA3 and ADE2 selectable markers, a galactose inductible GAL10/CYC1 promoter, multiple cloning sites, and the PGK1 terminator [26]. pCD69, a 2l-URA3-ADE2 plasmid expressing TGL1 under the control of the GAL10/CYC1 promoter was constructed as follows. The TGL1 open reading frame was isolated from FY1679 genomic DNA by amplification using oligonucleotides lip1 (5¢-atagacacgcaaacacaaatacaca cactaaattaataatgaccggatcATGTACTTCCCCTTTTTAGG CAGAT-3¢) and lip2 (5¢-cagtagagacatgggagatcccccgcgg aattcgagctcggtacccgggTCATTCTTTATTTAGAGCATC CAGC-3¢). The sequences in lower-case are complementary to the end of the GAL10/CYC1 promoter (lip1) and to the beginning of the PGK1 terminator (lip2). The 1647 bp PCR fragment was transformed into yeast along with BamHI– EcoRI-linearized pYeDP60, permitting cloning by homologous recombination between the plasmid and the PCR fragment and giving pCD69. The pDP10037 (2l-URA3-TRP1), pCD63 (2l-URA3TRP1) and pV13SCC (2l-URA3) plasmids were constructed as reported previously [19]. pDP10037 carries the GAL10/CYC1prom::matADR::PGK1term and GAL10/ CYC1prom::matADX::PGK1term expression cassettes separated by the URA3 marker. pCD63 was obtained from pDP10037 by replacing sequences coding for the mature Adrp (preceded by a methionine codon) by sequences coding for the mature form of cytochrome CYP11A1 (preceded by a methionine codon). pV13SCC expresses the mature form of CYP11A1 (preceded by a methionine codon) driven by the GAL10/CYC1 promoter.

FEBS 2003

1504 C. Duport et al. (Eur. J. Biochem. 270)

Strains Yeast strains used in this study are listed in Table 1. The structure of the CA10 D7 reductase expression locus has been verified by PCR, Southern and direct sequencing analysis of the promoter region, showing that it contains the GAL10/CYC1 promoter instead of the expected PGK1 promoter (Table 1). The APAT-deficient strain, CA14, was generated by disrupting the ATF2 gene of CA10 with the KanMX4 cassette, which confers G418 resistance. Primers 5¢ATF2-Kan 5¢-AGACTTTCAAACGAATAATAACTT CAGCAATAAAAATTGTCCAGGTTAATtccagcgacatg gaggccc-3¢ and 3¢ATF2-Kan: 5¢-TTGTACGAGCTCGG CCGAGCTATACGAAGGCCCGCTACGGCAGTATC GCAcattcacatacgattgacgc-3¢ (nucleotides in lower case are specific to the KanMX4 module) were used for PCR with pFA6-MX4 as a template [27] to produce the KanMX cassette flanked by ATF2 sequences [23]. The strain, CA19 was obtained by introducing the GAL10/CYC1prom::3bHSD::PGK1term cassette into the region between the HIS3 and DDE1 genes of CA14, as previously described [19]. CDR07 (an FY-1679–18B derivative that contains only the GAL10/CYC1prom::D7 reductase::PGK1term expression

cassette) was obtained by sporulation of the diploid resulting from a cross between CA10 (MATa) and FY1679–18B (MATa). TGY120.2 (MATa) was obtained by transformation of FY1679–28C (MATa) with XbaI-linearized pTG10925. This plasmid contains an S. cerevisiae genomic DNA fragment covering the LEU2 and SPL1 locus derived from pFL26CD [19]. A bovine mature Adrp expression cassette (TEF1prom:: matADR::PGK1term) [24] was introduced into the unique NotI site of pFL26CD, that is in the noncoding region between LEU2 and SPL1. Transformed colonies were grown in selective medium, and the expression of mature Adrp was verified by Western blot analysis as described [24]. One clone, TGY120.2 MATa, was selected for further studies. The yeast strain, CA15 was isolated by mating CDR06 (MATa) to TGY120.2 (MATa), that contains the TEF1prom::matADR::PGK1term cassette in the intergenic region between LEU2 and SPL1. The strain, CA17 was generated by integrating a TEF1prom::matADX::PGK1term cassette into the intergenic region between the HIS3 and DDE1 of CA15 with the pUC-HIS3ADX integrative plasmid. The are1::KanMX4 are2::HIS3 double mutant strain CA23 was constructed by crossing CA10 (MATa ARE 1

Table 1. Yeast strains and expression plasmids. Strain or plasmid S. cerevisiae strains FY1679 CDR07 CDS04 CA10

CA15

CA17

CA14

CA19

CA23

TGY 120.2

Relevant genotype or encoded protein (promoter)

Source

MATa/arho+, GAL2, ura3–52, trp1-D63, his3-D200, leu2-D1. MATa, rho+,GAL2, ura3–52, trp1-D63, his3-D200 leu2-D1, ade2::GAL10/CYC1::D7Reductase::PGK1. MATa, rho+, GAL2, ura3–52, trp1-D63, his3-D200, leu2-D1, are1::G418R, are2::HIS3. MATa, rho+, GAL2, ura3–52, trp1-D6, his3-D200, erg5::HYGROR, ade2::GAL10/CYC1::D7Reductase::PGK1, LEU2::GAL10/CYC1::matADR::PGK1. MATa, rho+, GAL2, ura3–52, trp1-D63, his3-D200, erg5::HYGROR, ade2::GAL10/CYC1::D7Reductase::PGK1, LEU2::TEF1::matADR::PGK1 MATa, rho+, GAL2, ura3–52, trp1-D63, erg5::HYGROR, ade2::GAL10/CYC1::D7Reductase::PGK1, LEU2:: TEF1::matADR::PGK1, HIS3::TEF1::matADX::PGK1. MATa, rho+, GAL2, ura3–52, trp1-D63, his3-D200, erg5::HYGROR atf2:: G418R, ade2::GAL10/CYC1::D7Reductase::PGK1, LEU2::GAL10/CYC1::matADR::PGK1. MATa, rho+, GAL2, ura3–52, trp1-D6, his3-D200, erg5::HYGROR, atf2:: G418R, ade2:: GAL10/CYC1::D7Reductase::PGK1, LEU2::GAL10/CYC1::matADR::PGK1, HIS3::GAL10/CYC1::3b-HSD::PGK1. MATa, rho+, GAL2, ura3–52, trp1-D63, his3-D200, erg5:: HYGROR, are1::G418R, are2::HIS3, ade2:: GAL10/CYC1::D7Reductase::PGK1, LEU2::GAL10/CYC1::matADR::PGK1 MATa, rho+, GAL2, ura3–52, trp1-D63, his3-D200, LEU2::TEF1::matADR::PGK1

[64] This study

Plasmids (2micron replicon and URA3 selection marker and GAL10/CYC1 promoter for all the above cDNAs and gene) pV13SCC CYP11A1 pCD63 CYP11A1, matADX pCD69 TGL1 pDP10037 matADX, matADR

[28] [19]

This study

This study

This study

This study

This study

This study

[19] [19] This study [19]

FEBS 2003


ARE2) [19] with CDS04 (MATa are1D are2D) [28] and searching for the appropriate haploid segregants. Subcellular fractionation and Western blot analysis In all subcellular fractionation experiments, recombinant yeast cells were grown in synthetic SL medium to a density of 107 cells per mL. Lipid particles and mitochondrial and endoplasmic reticulum (ER) membranes were prepared according to a published protocol [29]. The plasma membrane (PM) fraction was obtained using electrostatic attachment of spheroplasts on cationic silica beads as described [30]. For differential fractionation and sucrose density centrifugation, cell-free extracts were prepared in a manner similar to that reported above. Spheroplasts were disrupted with a Dounce homogenizer and loaded onto a sucrose gradient after centrifugation at 500 g to eliminate cellular debris. The gradient was continuous from 0.7–1.6 M sucrose in 10 mM Tris/HCl, pH 7.6, 10 mM EDTA and 1 mM dithiothreitol and centrifuged for 4 h at 100 000 g and 4 C in a swinging bucket rotor. 1.5 mL fractions were collected from the bottom of the gradient. The protein contents were determined using a protein assay kit (Pierce Chemical Co.). Immunological analysis of each subcellular fraction was carried out after separation of proteins by 8–12% SDS/ PAGE and transfer onto nitrocellulose membranes (Hybond-C; Amersham Pharmacia Biotech) by standard procedures as described [18, 24]. Filters were probed with antibodies to porin (an outer mitochondrial membrane marker [31]), the 40-kDa microsomal protein [32], 3-PGK (a cytosol marker from Molecular Probes Inc [33]), and Pma1p (an integral membrane-bound H+-ATPase of the PM [34]), to characterize yeast organelles. For detection of recombinant CYP11A1, Adxp, Adrp and 3b-HSD, rabbit polyclonal antibodies obtained from Oxygene (Dallas, Tx, USA) were used. Anti-peptide D7-Red was generated using a synthetic peptide containing amino acids 311–324 of D7-reductase (H2N-Tyr-Asp-Arg-Gln-Arg-Gln-Glu-PheArg-Arg-Thr-Asn-Gly-Lys-COOH) coupled to keyhole limpet hemocyanin by N-maleimidobenzoyl-N-hydrosuccinimide ester cross-linking. The resulting peptide/keyhole limpet hemocyanin conjugate was injected subcutaneously into female New Zealand White Rabbits (Neosystem Laboratoire, Strasbourg, France). Immune complexes were visualized using HRP-conjugated secondary antibodies (Amersham Biosciences, Little Chalfont, UK), followed by chemiluminescence (SuperSignal, Pierce Chemical Co., Rockford, IL USA). Fluorescence and confocal microscopy Yeast cells were fixed in 2% paraformaldehyde, converted to spheroplasts, attached to poly L-lysine coated coverslips and permeabilized as described [31]. Samples were incubated with 1/20 dilutions of anti-CYP11A1, anti-Adrp, antiAdxp, anti-PGK, anti-Porin, anti-Gpa1p [35] (Plasma Membrane marker, Santa Cruz Biotech. Inc) Igs and a 1/5 dilution of anti-Dpm1p (dolichol phosphate mannose synthase, Molecular Probes Inc) [36] in NaCl/Pi containing 1% BSA and 0.1% Tween 20 overnight at 4 C. They were

washed four times in 1 · NaCl/Pi and stained with a 1/150 dilution of CY2TM-conjugated anti-rabbit IgG (Interchim Inc.) and a 1/150 dilution of FITC-conjugated goat antimouse IgG (Santa Cruz Biotech. Inc) for 30 min. Samples were washed and mounted in 95% glycerol containing 0.1% p-phenylenediamine. Observations were made with a confocal microscope (model MRC-1000; Bio-Rad House, Hertfordshire, England; 1-lm optical serial sections) attached to a camera (model Optiphot; Nikon Inc.) equipped with a 60 · plan apochromat objective (NA 1.3; Carl Zeiss Inc.). Images were collected using Bio-Rad image capture software, and projections were generated using confocal assistant software LASER SHARP 2000 (Bio-Rad). Enzymatic activity assays Previously, steryl ester hydrolase assays were performed as reported [37] except that cholesteryl[4-14C]oleate (100 mCiÆmL)1 in toluene, NEN Life Science Products Inc.) was used instead of cholesteryl[1-14C]oleate. Side chain cleavage activity was measured as described [18]. Steroid and sterol analyses Steroids and sterols were extracted from yeast cells and analyzed by gas chromatography as described previously [19]. Total concentration was measured after 100 h galactose induction in three independent assays. Error bars on histograms indicate the standard errors of the means. Statistical significance by paired t-tests was performed using the STATVIEW program. Statistical significance was assumed when P < 0.05.

Results The cellular pool of ergosta-5-eneol is not limiting for CYP11A1 activity. In the strain, CA10/pCD63, the reconstituted SCC system catalyzes the conversion of ergosta-5-eneol to pregnenolone, that accumulates also as the 3-acetyl ester form ([19], and Fig. 1). Pregnenolone esterification was found to compete with progesterone production when mammalian 3b-HSD activity was introduced into this recombinant strain [19]. Cauet and coworkers [38] further showed that in S. cerevisiae, pregnenolone is acetylated by acetyl-coenzyme A: pregnenolone acetyl-transferase (APAT) activity encoded by the ATF2 gene. Therefore, both free pregnenolone production and the efficient coupling of the SCC and the 3b-HSD activities in yeast require the use of strains lacking this acetylating activity. For this reason, the CA10 ATF2 gene was disrupted, yielding the strain, CA14. The strain, CA19 was constructed by integrating a human 3b-HSD expression cassette into the CA14 genome (see Materials and methods and Table 1). As expected, pregnenolone acetate accumulation was not observed in the atf2D strains, CA14 and CA19 transformed with pCD63 compared to the ATF2 control strain CA10/ pCD63 (Fig. 2A). In addition, pregnenolone was almost completely converted into progesterone when 3b-HSD activity was expressed (compare the CA19/pCD63 and


FEBS 2003

Fig. 1. Schematic representation of the connected sterol and steroid pathway in yeast. C, cytosol; ER, endoplamic reticulum; LP, lipid particles; mat, mature form of the proteins; PM, plasma membrane. Steroids are shown in green. Ncp1p, NADPH P450 reductase; Adxp, adrenodoxin; Adrp, adrenodoxin reductase; Are1p, Are2p, Atf2p, alcohol O-acetyltransferase (acetyl pregnenolone acetyl transferase); CYP11A1, P450 side chain cleaving; Erg2p, sterol C8-C7 isomerase; Erg3p, C-5 sterol desaturase; Erg5p, D 22(23) sterol desaturase; Erg6p, S-adenosyl methionine D-24-sterolC-methyl-transferase; 3b-HSD, 3b-hydroxy steroid dehydrogenase.

CA14/pCD63 gas chromatography profiles). However, in these two strains, the accumulation of ergosta-5-eneol, which is both a CYP11A1 substrate and the main sterol of CA10/pCD63 membranes, was greatly reduced, while at least four other sterols accumulated (Fig. 2B). They were identified (by mass spectrometry and relative retention time to cholesterol [39]) as intermediates of the ergosterol biosynthesis pathway. Namely, in CA14/pCD63 membranes, desmosterol (cholesta-5,24-dieneol) becomes the major sterol while ergosta-5-enol, zymosterol (cholesta8,24-dieneol), fecosterol [ergosta-8,24(28)-dieneol], ergosta5,7-dieneol and another sterol (Mr ¼ 384, putatively cholesta-7,24-dieneol) accumulate (Figs 1 and 2B). To a lesser extent, the same phenomenon is observed in CA19/ pCD63 membranes; but in this strain, ergosta-5-eneol remains the main sterol (Fig. 2B). Thus, the production of both pregnenolone and progesterone correlates with the depletion of ergosta-5-eneol and the accumulation of ergosta-5-eneol precursors in atf2D strains. Moreover, the addition of pregnenolone, or to a lesser extent of progesterone, into the culture medium of CA14 (in the absence or presence of CYP11A1) similarly induces the accumulation

of ergosterol biosynthesis intermediates (data not shown). The levels of free ergosta-5-eneol final (stationary phase) were reduced 10- and fivefold for CA14/pCD63 and CA19/ pCD63, respectively when compared to CA10/pCD63 (Fig. 3B). However, the extent of steroid formation did not reflect this dramatic change in CYP11A1 substrate availability and was comparable for the strains CA10/ pCD63, CA14/pCD63, CA19/pCD63, which produce pregnenolone acetate, pregnenolone and progesterone, respectively (Fig. 3A). An almost complete conversion of pregnenolone into progesterone in CA19/pCD63 was associated with a more limited decrease in ergosta-5-eneol content than in the pregnenolone accumulating strain CA14/pCD63. This result is consistent with observation of the absence of inhibition by pregnenolone in the pregnenolone acetylation-competent strain CA10 (Fig. 2). In conclusion, we showed that in atf2D strains the SCC reaction is readily coupled with 3b-HSD activity, permitting the efficient biosynthesis of progesterone. A disruption of pregnenolone acetylase activity causes in turn a dramatic decrease in the cellular production of free ergosta-5-eneol, the CYP11A1 substrate, but has only limited effects on the

FEBS 2003


Fig. 2. The efficient production of pregnenolone and progesterone (A) in the atf2D recombinant strains CA14/pCD63 (C) (Derg5, expressing CYP11A1) and CA19/pCD63 (Derg5, expressing CYP11A1 and 3bHSD) is accompanied by an accumulation of ergosta-5-eneol precursors (B). Gas chromatography (GC) profiles were obtained from cellular lysates prepared from cultures harvested after 100 h of induction with galactose. The sterol extraction procedure allows free sterol to be detected (B). The ATF2 strain CA10/pCD63 was used as a control. Relative retention times to cholesterol under our conditions are shown between brackets. Steroids are: P, pregnenolone (0.598); Pr, progesterone (0.685), PA, pregnenolone acetate (0.714). Atypical sterols detected in CA14/pCD63 and CA19/pCD63 cells are the following: D, desmosterol (1.04); E5, ergosta-5-enel (1.13); E5,7, ergosta-5,7-dieneol (1.16); F, fecosterol (1.11); U, unknown sterol with a MW of 384 which might be cholesta-7,24-dieneol (1.17); Z, zymosterol (1.06).

activity of CYP11A1. This suggests that the availability of the CYP11A1 substrate is not limiting in the experimental conditions used, but it might be limiting for higher levels of CYP11A1 activity or expression. Tgl1p, a putative ester hydrolase, regulates SCC activity In wild-type yeast, more than 90% of the predominant sterol is stored as esters [37]. The balance between the levels of free and esterified sterols is regulated by esterification and hydrolysis and is modified in strains disrupted for the estersynthase genes ARE 1 and ARE 2 ([20,28]), or in which steryl ester hydrolase activity is altered [40]. To further investigate the potential limiting effect of sterol availability on CYP11A1 activity, we cloned and expressed the TGL1 gene, that is predicted to code for a 46-kDa protein with a potential steryl ester hydrolase catalytic domain [41]. Tgl1p function was first evaluated in vitro by monitoring the cholesteryl hydrolase activities of cell-free extracts prepared from the wild-type strain, FY1679/pCD69 and an isogenic are1D are2D double mutant, CDS04/pCD69. Surprisingly, ester hydrolase activity in extracts of the control strain, FY1679/pYeDP60 was higher than in extracts of FY1679/ pCD69, which overexpresses Tgl1p (Fig. 4A). In contrast, although control CDS04/pYeDP60 extracts exhibited the same activity as FY1679/pYeDP60 extracts, CDS04/ pCD69 cell-free extracts exhibited a twofold increase in cholesteryl ester hydrolase activity. These apparently contradictory results can be rationalized if cholesteryl esterase activity can be monitored in vitro only in cellular extracts devoid of endogenous sterol esters, that likely compete with the labeled cholesteryl oleate used as a substrate. Inhibition in FY1679/pCD69 extracts also suggested a strong preference of Tgl1p for yeast sterol esters compared to cholesteryl oleate. The hydrolytic activity of Tgl1p was further analyzed in vivo (Fig 4B,C). Similar twofold, increases in the ratios of free vs. total ergosterol and ergosta-5-eneol were observed in Tgl1p-overexpressing FY-1679 and CA10 cells, when compared to their respective control strains (Fig. 4B). This effect was not observed in the corresponding are1D are2D double mutant strains CDS04 and CA23, consistent with the disruption of ester synthase genes (data not shown). As established for ergosterol in wild-type strains [40], the highest free ergosta-5-eneol to total protein ratio was


FEBS 2003

strain was increased (Fig. 4B and data not shown), whereas pregnenolone production was enhanced in both strains, with similar final concentrations of the steroid (Fig. 5). Therefore, these results reveal two levels of complexity of the CYP11A1 reaction in yeast. On one hand, the content of free ergosta-5-eneol poorly correlates with the extent of pregnenolone production, suggesting that SCC activity is only partially limited by substrate concentration (Fig. 1). On the other, an artificial increase in the level of free ergosta5-eneol improves the yield of steroid (Fig. 5). This latter effect may involve steryl hydrolase activity potentially encoded by TGL1. Moreover, the effects of Tgl1p on CYP11A1 activity is observed both in the presence and absence of ester synthase activity. In recombinant yeast, the concentration of Adxp, but not of Adrp, controls SCC activity

Fig. 3. The final levels of accumulated steroids are independent of Datf2 genetic background (A) and do not correlate with the content of free ergosta-5-eneol (B). Steroid-producing cells were grown as described in Material and methods. Accumulated steroid contents are the sum of pregnenolone and pregnenolone acetate (CA10/pCD63) (Derg5, expressing CYP11A1), pregnenolone (CA14/pCD63) (Fig. 2), pregnenolone and progesterone (CA19/pCD63) (Fig. 2). Statistical significance by paired t-tests was performed using the STATVIEW program. **P < 0.05 when a strain is compared to the two others (B).

observed in the PM of steroid-producing cells (Fig. 4C). In contrast, internal membranes (ER, lipid particles and mitochondrial membranes) exhibited only residual levels of ergosta-5-eneol. Tgl1p overexpression did not modify this subcellular distribution, and the influence of Tgl1p was observed only in the PM fraction. In summary, we showed that Tgl1p containing extracts have a steryl ester hydrolytic activity and that these extracts could mediate the release of free ergosta-5-eneol from esterified forms in steroid-producing cells with the same efficiency as observed for ergosterol in wild-type yeast cells. In addition, Tgl1p overproduction leads to an increase of free ergosta-5-eneol in tested organelles, especially in the PM, which contains the highest concentration of substrate. To determine whether the Tgl1p-dependent increase in free ergosta-5-eneol content could affect the SCC reaction, the concentrations of accumulated pregnenolone were measured in both the wild-type strain CA10/pCD63 and in the are1D are2D double mutant CA23/pCD63, in the absence or presence of the Tgl1p overexpression construct (Fig. 5). As expected, CA23/pCD63, that is devoid of ester synthase activity [42], contained higher amounts of free ergosta-5-eneol than CA10/pCD63 (data not shown). This phenomenon could explain the limited enhancement of CYP11A1 activity observed in the are1D are2D mutant (Fig. 5). When CA10 and CA23 were cotransformed with pCD69, only the free ergosta-5-eneol content of the former

To determine whether electron transfer from NADPH to CYP11A1 via Adrp and Adxp could regulate the extent of synthesis, we built two CA10 derivatives, CA15 and CA17. Like CA10, CA15 carries a unique ADR expression cassette integrated in the LEU2-SPL1 intergenic region but in CA15 the mature ADR ORF is under the control of the constitutive TEF1 promoter instead of the inducible GAL10/CYC1 promoter. CA17 also carries a cassette with the mature version of ADX integrated in the HIS3-DED1 intergenic region. (Table 1). The accumulation of steroids in the strains CA15/pCD63 (that carries a multicopy plasmid bearing CYP11A1 and ADX) and CA17/pV13sccm (that carries a multicopy plasmid for CYP11A1 and has a single integrated copy of ADX) was compared to that in CA10/pCD63. The level of expression of mature Adrp was found to be lower in CA15/pCD63 as compared to CA10/pCD63, as judged by Western blot analysis (data not shown and [43]). In contrast, CA17 exhibited the same level of Adrp as CA15 but a lower level of Adxp as Adxp was expressed from a single genomic copy (data not shown and Table 1). Similar concentrations of pregnenolone (2.9 ± 0.5 mgÆL)1A600 units) were observed for CA10/pCD63 and CA15/pCD63, whereas a 36-fold decrease was observed for CA17/pCD63. These results suggest that CYP11A1 activity depends strongly on the levels of expression of Adxp but not Adrp and therefore that the concentration of Adxp is the major factor controlling pregnenolone synthesis in recombinant yeast. In conclusion, the SCC reaction appears to be regulated similarly in yeast and in adrenal cells; in the latter, Adxp and not Adrp limits the activity of CYP11A1 and hence controls the extent of pregnenolone production [14]. Protein partners involved in the yeast SCC system localize to three distinct subcellular compartments To gain a deeper insight into how the artificial steroid pathway is coordinated with the endogenous ergosta-5-eneol pathway, we determined the subcellular localizations of Adrp, Adxp, D7-Red and 3b-HSD. A total cell extract from the progesterone-producing strain CA19/pCD63 was subjected to sucrose gradient fractionation (Fig. 6). Characterization of each fraction with anti-Pma1p (PM), anti3-PGKp (cytosol), anti-40 kDa protein (ER) and anti-porin (mitochondrial membranes) antisera revealed that Adrp,

FEBS 2003


Fig. 4. Tgl1p has a steryl ester hydrolase activity. Tgl1p activity was illustrated by comparing the steryl esterase cell free extract activities of Tgl1p-overexpressing cells (strains transformed with pCD69, stripped bars) vs. the controls (strains transformed with the vector pYeDP60, gray bars). The CDS04 strain is an Dare1, Dare2 isogenic derivative of FY1679-28 C and was generated by disruption of both ARE 1 and ARE 2 genes, that encode two sterol ester-transferases that catalyze the synthesis of steryl ester in yeast. (A) The cholesteryl esterase activity of Tgl1p is detected only in the sterol esterification deficient strain, CDS04 (Dare1, Dare2). Experiments used cholesteryl[4-14C]oleate as a substrate [37]. Specific activities were measured in whole cell extracts prepared from lysed spheroplasts of FY1679/ pYeDP60, FY1679/pCD69 (expressing Tgl1p), CDS04/pCD69 (see above), and CDS04/pYeDP60 cells. Data are mean values ± SEM from three independent experiments with a maximum deviation of 5%. (B) The free ergosterol and ergosta-5-eneol contents are increased in the Tgl1p-overproducing strains FY1679, CA10 (Derg5) and CA10/pCD63 (Derg5 expressing CYP11A1), respectively. Sterols were extracted from cellular lysates and analyzed by gas chromatography with or without preliminary saponification for detection of total sterols and free sterols, respectively. Data are expressed as ratio of free vs. total sterol. Data are mean values ± SEM obtained from three independent experiments. (C) The subcellular partitioning of free ergosta-5-eneol is not changed by TGL1 overexpression. Whole cell lysates were subjected to fractionation to isolate subcellular organelles, mitochondria, lipid particles and ER as described [29] and PM as described [30]. Data are expressed as the sterol : protein ratio and are mean values from three independent experiments with a maximum deviation of 5%. The ratio obtained in lipid particles were 6.69 · 10)4 and 6.78 · 10)4 mgÆmg protein)1 in CA10/ pYeDP60 and CA10/pCD69, respectively.

D7-Red and 3b-HSD clearly localize to the ER whereas Adxp is a soluble protein, as shown previously [19]. In contrast to the other enzymes, recombinant CYP11A1 was not restricted to a single subcellular compartment but instead was found distributed broadly throughout the gradient. Either this experiment reflects a broad intracellular distribution of the protein, or a cell surface transport of the protein as described [44,45]. To distinguish between these two hypotheses, indirect immunofluorescence studies were performed with polyclonal Igs raised against markers for each of the different compartments (Fig 7. A,D,G, red fluorescence, CY-2TM conjugated secondary Ig). The PM (Fig. 7B), ER (Fig. 7E) and mitochondrial membranes (Fig. 7H) were simultaneously visualized in the same cells with Igs to Gpa1p, Dpm1p and porin, respectively (green fluorescence, FITC-conjugated secondary antibodies). Fluorescence was not detected in control experiments performed in the absence of the primary anti-CYP11A1 and anti-marker Igs (data not shown). Confocal microscopy was used to evaluate the degree of CYP11A1 colocalization with each of these organelle markers, as seen in merged images (Fig 7. C,F,I). CYP11A1 was found to colocalize with the PM marker Gpa1p (Fig. 7C; yellow areas correspond to regions where the red and green signals are superimposed), suggesting that most of the CYP11A1 antigen resides in the PM. There was no overlap between the red and green signals corresponding to the CYP11A1 and porin mitochondrial markers, respectively, but some yellow

signals could be seen in cells labeled with both the CYP11A1 and the ER marker Ig (Fig. 7). However, the best correspondence was observed for the CYP11A1-derived signal and the PM-derived signal. In conclusion, the CYP11A1 antigen appears to be excluded from mitochondria in vivo, and most of the antigen is detected in the plasma membrane, with a minor fraction localizing to the ER. To determine whether the SCC reaction occurs in the PM, where most CYP11A1 is found, or in the ER, we performed an analysis of free ergosta 5-eneol distribution in the steroid-producing strain, CA19/pCD63 and in the control strain, CA19/pDP10037. Figure 8 shows that the level of free sterol in the PM is significantly depleted upon the expression of SCC activity, whereas no decrease is

FEBS 2003


NH2 terminus) is primarily in the PM in yeast, and the SCC system depends on electron transfer from ERlocalized Adrp to cytosolic Adxp and finally to CYP11A1 in the PM.

Discussion

Fig. 5. Pregnenolone biosynthesis is increased in recombinant yeast cells overproducing Tgl1p. Pregnenolone was extracted from cellular lysates prepared from cultures harvested after 100 h of induction by galactose [19]. The Tgl1p-overproducing effect was analyzed in ARE 1 ARE 2 strains (CA10/pCD63 + pCD69) (Derg5, expressing CYP11A1 and Tgl1p) and (CA10/pCD63 + pYeDP60) (Derg5, expressing CYP11A1) and in Dare 1 Dare 2 strains (CA23/pCD63 + pCD69) (Derg5, expressing CYP11A1 and Tgl1p) and CA23/pCD63 + pYeDP60) (Derg5, expressing CYP11A1).

Fig. 6. Subcellular localization of heterologous CYP11A1, Adrp, Adxp, D7-Red, 3b-HSD and organelle membrane after sucrose density fractionation in recombinant yeast. The heterologous Adrp, D7-Red and 3b-HSD proteins cofractionate with the ER marker (40 kDa protein) and Adxp cofractionates with the cytosol marker 3-PGK on a sucrose continuous gradient. No clear subcellular localization in the PM, ER or mitochondrial membranes was observed for CYP11A1. CA19/ pCD63 (Datf2, Derg5, expressing CYP11A1 and 3bHSD) cells were grown as described in Materials and methods, converted to spheroplasts, lysed and fractionated on a continuous sucrose gradient (0.7– 1.6 M) from the bottom (fraction 1) to the top (fraction 25). Fractions were subjected to Western blot analysis with Igs against the following organelle marker proteins: 3-PGK for cytosol, Pma1p for the PM, the 40-kDa protein for the ER and porin for mitochondria. The distribution of heterologous proteins was similarly detected with Igs against CYP11A1, Adrp, Adxp, 3b-HSD and D7-Red.

observed in the ER or mitochondrial fractions. These data are consistent with ergosta-5-eneol bioconversion at the PM level. In conclusion, the mature form of recombinant CYP11A1 (e.g., the protein without the mitochondrial targeting sequence and with an extra methionine at the

The reconstruction in S. cerevisiae of the steroidogenic pathway allows the self-sufficient production of pregnenolone and progesterone, as reported previously [19] (Fig. 1). Like mammals, S. cerevisiae possesses a system that efficiently protects against the toxic accumulation of pregnenolone. In mammals, pregnenolone is present as a biologically active sulfo-conjugate [46], whereas in yeast, APAT activity converts pregnenolone into the corresponding acetate ester [38]. Acetylation or sulfatation at position 3 of pregnenolone prevents further metabolism by 3b-HSD. Yeast strains devoid of APAT activity allow high-yield biosynthesis of free pregnenolone or progesterone but accumulate ergosta-5-eneol biosynthesis intermediates, such as desmosterol, fecosterol and zymosterol (Fig. 1). This phenomenon likely results from the inhibition of S-adenosyl sterol 24-methyl transferase (Erg6p [47]), and to a lesser extent of sterol C8–C7 isomerase (Erg2p [48]). Erg6p permits the transformation of cholesta-derivatives into ergosta-derivatives. Thus, inhibition of Erg6p allows the accumulation of zymosterol that is sequentially transformed by Erg2p and Erg3p (the C-5 desaturase [49]), into cholesta7,24-dieneol and cholesta-5,7,24-trieneol, respectively (Fig. 1). In the presence of D7-reductase, the latter is modified into cholesta-5,24-dieneol (desmosterol) that is detected in the membranes of CA14/pCD63 (Fig. 2B). Accumulation of zymosterol indicates that Erg2p might also be inhibited. A similar effect was observed in mammalian cells, in that progesterone [50] and pregnenolone [51] inhibit cholesterol biosynthesis, resulting in the accumulation of cholesterol precursors. Finally, in yeast strains devoid of APAT activity, there is almost no accumulation of pregnenolone when 3b-HSD activity is present, and the rate of progesterone biosynthesis is only marginally reduced compared to the rate observed for pregnenolone alone. Therefore, an efficient coupling of the two first steps of steroidogenesis is possible, and as reported for mammalian cells [9], the SCC reaction is the rate-determining step in progesterone synthesis in yeast. The yeast SCC system offers the possibility of increasing or decreasing the availability of the endogenous substrate. Ergosterol or related yeast sterols exist in the free form or as esters conjugated to fatty acids. Conversion between free sterols and steryl esters is thus a critical homeostatic determinant for membrane function in yeast as in all eukaryotic cells ([52,53]). The PM is the major subcellular location of free sterol [40]. Steryl esters are synthesized in the ER by the ACAT-related ARE 1 and ARE 2 gene products [20,42], stored in lipid droplets and mobilized by a process involving steryl ester hydrolases in the PM [37]. Our work shows that Tgllp over-expressing cells and extracts exhibit steryl ester hydrolase activities in vivo and in vitro, respectively. As it has been shown previously that Tgl1p had significant homologies to triglyceride lipase, it is rather likely that Tgl1p is a steryl ester hydrolase [41]. In the steroid-producing yeast strain, CA10/pCD63 free

FEBS 2003


Fig. 7. CYP11A1 colocalizes with the endogenous Gpa1p plasma membrane protein. Spheroplasts from CA19/pCD63 (Datf2, Derg5, expressing CYP11A1 and 3b)HSD) were fixed, permeabilized and then incubated with primary polyclonal CYP11A1 Igs. A CY2TM conjugated secondary Ig (red fluorescence in A, D, G) was used to visualize the CYP11A1 protein. PM (B), ER (E) and mitochondrial (H) membranes were detected in the same cells with anti-Gpa1p (guanine nucleotide-binding protein alpha subunit), Dpm1p (dolichol-phosphate mannosyltransferase) and porin monoclonal Igs coupled to FITC secondary Igs (green fluorescence). The merged CY2TM and FITC images are shown in C, F and I.

Fig. 8. Distribution of the CYP11A1 substrate. Whole cell lysates were fractionated to isolate the ER and mitochondrial membranes as described [29] and PM as described [30]. Data are expressed as percent of the total activity measured for the whole cellular extract. Relative to CA10/pDP10037 (gray bars, Derg5), the pregnenolone-producing strain CA10/pCD63 (stripped bars, Derg5 expressing CYP11A1) shows a significant depletion of the levels of free ergosta 5-eneol in the PM fraction and increased overall levels in the ER and mitochondrial fractions.

ergosta-5-eneol represents, as expected, only a minor fraction of the cellular pool of sterols, and this fraction increases when Tgl1p is overexpressed or when both the

ARE 1 and ARE 2 genes are deleted. Thus, the two ester synthases, Arelp and Are2p and the probable steryl ester hydrolase Tgllp, play complementary roles in maintaining free ergosta-5-enol at appropriate levels, reminiscent of the mechanism reported for cholesterol in adrenal cells ([40,54]). In addition to normally cycling between the PM and other cellular organelles, free ergosta-5-eneol also must be available in the yeast cell to serve as a substrate for CYP11A1. While SCC driven formation of pregnenolone likely occurs at the PM, ergosta-5-enol biosynthesis occurs mainly in the ER, consistent with D7-Red localization (Fig. 6). Are1p and Are2p are localized in the ER [42] while Tgl1p could be localized in lipid particles, and activated with the supply of sterol from lipid particles to PM. Therefore, de novo synthesis and transport processes are critical for continued activity. In theory, the sterol pool used for steroidogenesis must be constantly supplied from cellular sterol stores while the membrane structural pool is more static, at least for cells in stationary phase, in which CYP11A1-dependent conversion is active but cell growth has ceased. Thus, free and esterified sterol pools are exchanged rapidly. Indeed, Tgllp overproduction results in an increased level of free sterol (Fig. 4B), and it would also be expected to increase the cycling of sterols by esterification and hydrolysis, resulting in an increase in intracellular sterol trafficking. It is unclear, then, whether the corresponding increase in pregnenolone production results from an increase in free sterol concentration or from an enhancement of sterol trafficking.

FEBS 2003


In conclusion, the recombinant yeast strain described in this report mimics mammalian steroid-producing cells with respect to the coupling of the SCC and 3b-HSD reactions; the inhibitory effect of steroid products on CYP11A1 substrate biosynthesis; the regulation of intracellular free sterol concentration by the ACAT-related enzymes Are1p and Are2p; the steryl ester hydrolase Tgl1p and finally, the modulation of substrate availability by Tgl1p. To determine if the delivery of reducing equivalents to CYP11A1 could contribute to limit SCC activity in yeast, the importance of the concentrations of Adxp and Adrp was evaluated. The recombinant SCC system in yeast is sensitive to the levels of expression of these components, similar to that found for mammalian adrenal cells [14] or for in vitro assay systems [55]. In yeast, as in transfected mammalian cells [56], a decrease in the level of expression of Adxp has dramatic consequences for SCC activity. Alterations in the level of expression of Adrp in yeast have no effect as described above. This contrasts with a recent paper where, in placenta, the Adrp concentration is limiting, giving an alternative regulation of the SCC reaction [57]. It was of interest to characterize the localization of the different proteins of the reaction – namely, CYP11A1, Adxp and Adrp in yeast cells that exhibit mammalian steroidogenic activity. In order to avoid the difficulty of transporting the CYP11A1 substrate to yeast mitochondria, we chose to express CYP11A1, Adxp and Adrp outside the mitochondrion by simply replacing their respective mitochondrial targeting sequences with a methionine residue. Adrp and CYP11A1 were expected to be found in the ER, while Adxp was expected to be in the cytosol. Classical differential centrifugation (data not shown) or sucrose gradient fractionation (Fig. 6) confirmed the localization of Adxp and Adrp, but CYP11A1 could not be definitively localized using these biochemical fractionation techniques. Finally, indirect immunofluorescence with polyclonal antiCYP11A1 antibodies together with confocal microscopy was the only technique that reliably and reproducibly showed that CYP11A1 is mainly localized to the PM. We cannot exclude the possibility that the deletion of the mitochondrial targeting sequence from CYP11A1 reveals a cryptic recognition motif that allows targeting to the PM. For example, the mature forms of the human, bovine and pig CYP11A1 have two positively charged residues at their NH2 termini that could favor targeting to the PM [58]. The presence of CYP11A1 in the yeast PM is rather puzzling considering that mammalian CYP11A1 is a specialized enzyme localized to the inner mitochondrial membrane of adrenal or placenta cells. Moreover, the protein is apparently catalytically competent in the yeast PM, based on the observation of the specific depletion of ergosta-5-eneol in the PM ([19], and this work). Interestingly, both Adrp and Adxp are absent from the PM, as evidenced by Western blot analyses, and their respective ER and cytosolic localizations were confirmed by immunofluorescence and differential centrifugation (data not shown, and [59]). This result indicates that Adxp functions as a soluble transporter of electrons from the ER associated Adrp to the PM-bound CYP11A1. Therefore, our work supports a shuttle mechanism of electron transport in which Adxp dissociates from the ER (where Adrp is located) before delivering electrons to CYP11A1 at the PM. In

conclusion, these findings demonstrate that the cluster model proposed for the mitochondrial system ([17,60]), is not obligatory and that the mitochondrial environment is not absolutely required for CYP11A1 function. The results of this work provide convincing evidence that recombinant CYP11A1 associates functionally with the PM in yeast but do not allow us to exclude the possibility that it has a second association with ER membranes (Fig. 7). Ectopic localization at the PM of a functional microsomal cytochrome P450 protein, has been reported for CYP2D6 expressed heterologously in yeast [30] and endogenously in rat hepatocytes [61]. Dual localization has been also reported for CYP2E1 and other P450 enzymes, and these enzymes can exhibit different substrate specificities in the mitochondrion and in microsomes. Whereas CYP2D6 requires NADPH-P450 reductase as an electron carrier at both sites, CYP2E1 is bifunctional: in the mitochondrion it receives electrons from Adxp and Adrp, and in microsomes it receives electrons from NADPH-P450 reductase. CYP11A1 cannot receive electrons from NADPH-P450 reductase [62] and the existence of alternate electron carriers has not been established. However, cytochrome b5 has been shown to interact specifically with CYP11A1, but this interaction is unproductive in the absence of Adxp and Adrp [63]. In conclusion, we have used steroid-producing S. cerevisiae strains to study the different factors involved in the SCC of ergosta-5-eneol into pregnenolone, including the effect of the steroid products on the yeast sterol synthesis pathway, acetylation of the product, availability of the substrate, electron flow and localization of the different protein partners. We find that yeast mimics mammalian adrenal cells in these respects. The quantity of steroid produced is controlled by the availability and mobility of the substrate together with the concentration of Adxp. Unexpectedly, Adxp, Adrp and CYP11A1 are localized in the cytosol, ER and PM, respectively, without impairing the SCC reaction and its coupling to the ER-associated 3b-HSD activity. Nonetheless, Adxp can shuttle electrons from ADR to CYP11A1 in a productive fashion. The unusual presence of the mitochondrial CYP11A1 in the PM may reflect a possible alternative localization of this enzyme in mammalian cells, suggestive of an alternative way of producing pregnenolone.

Acknowledgements The anti-PmaIp Igs were kindly supplied by R. Hagenauer-Tsapis. We thank J. Loeper and N. Chaverot for their helpful assistance in confocal scanning microscopy. We also thank C. Roubal for continuous support. This work was supported by Aventis Pharma.

References 1. Fannon, S.A., Vidaver, R.M. & Marts, S.A. (2001) An abridged history of sex steroid hormone receptor action. J. Appl. Physiol. 91, 1854–1859. 2. Miller, W.L. (1988) Molecular biology of steroid hormone synthesis. Endocr. Rev. 9, 295–318. 3. Sugano, S., Miura, R. & Morishima, N. (1996) Identification of intermediates in the conversion of cholesterol to pregnenolone with a reconstituted cytochrome p-450scc system: accumulation of

FEBS 2003

4.

5.

6.

7.

8.

9. 10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.


the intermediate modulated by the adrenodoxin level. J. Biochem. (Tokyo). 120, 780–787. Morohashi, K., Fujii-Kuriyama, Y., Okada, Y., Sogawa, K., Hirose, T., Inayama, S. & Omura, T. (1984) Molecular cloning and nucleotide sequence of cDNA for mRNA of mitochondrial cytochrome P-450 (SCC) of bovine adrenal cortex. Proc. Natl Acad. Sci. USA. 81, 4647–4651. Zuber, M.X., Mason, J.I., Simpson, E.R. & Waterman, M.R. (1988) Simultaneous transfection of COS-1 cells with mitochondrial and microsomal steroid hydroxylases: incorporation of a steroidogenic pathway into nonsteroidogenic cells. Proc. Natl Acad. Sci. USA. 85, 699–703. Nelson, D.R., Koymans, L., Kamataki, T., Stegeman, J.J., Feyereisen, R., Waxman, D.J., Waterman, M.R., Gotoh, O., Coon, M.J., Estabrook, R.W., Gunsalus, I.C. & Nebert, D.W. (1996) P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature. Pharmacogenetics. 6, 1–42. Vickery, L.E. (1997) Molecular recognition and electron transfer in mitochondrial steroid hydroxylase systems. Steroids. 62, 124– 127. Grinberg, A.V., Hannemann, F., Schiffler, B., Muller, J., Heinemann, U. & Bernhardt, R. (2000) Adrenodoxin: structure, stability, and electron transfer properties. Proteins. 40, 590–612. Miller, W.L. (1995) Mitochondrial specificity of the early steps in steroidogenesis. J. Steroid Biochem. Mol. Biol. 55, 607–616. Lin, D., Gitelman, S.E., Saenger, P. & Miller, W.L. (1991) Normal genes for the cholesterol side chain cleavage enzyme, P450scc, in congenital lipoid adrenal hyperplasia, J. Clin. Invest. 88, 1955– 1962. Bose, H., Lingappa, V.R. & Miller, W.L. (2002) Rapid regulation of steroidogenesis by mitochondrial protein import. Nature. 417, 87–91. Stocco, D.M. & Clark, B.J. (1996) Regulation of the acute production of steroids in steroidogenic cells. Endocr. Rev. 17, 221–244. Yang, H., Cromley, D., Wang, H., Billheimer, J.T. & Sturley, S.L. (1997) Functional expression of a cDNA to human acyl-coenzyme A: cholesterol acyltransferase in yeast. Species-dependent substrate specificity and inhibitor sensitivity. J. Biol. Chem. 272, 3980– 3985. Hanukoglu, I. & Hanukoglu, Z. (1986) Stoichiometry of mitochondrial cytochromes P-450, adrenodoxin and adrenodoxin reductase in adrenal cortex and corpus luteum. Implications for membrane organization and gene regulation. Eur. J. Biochem. 157, 27–31. Tuckey, R.C. & Sadleir, J. (1999) The concentration of adrenodoxin reductase limits cytochrome p450scc activity in the human placenta. Eur. J. Biochem. 263, 319–325. Tuckey, R.C., McKinley, A.J. & Headlam, M.J. (2001) Oxidized adrenodoxin acts as a competitive inhibitor of cytochrome P450scc in mitochondria from the human placenta. Eur. J. Biochem. 268, 2338–2343. Muller, E.C., Lapko, A., Otto, A., Muller, J.J., Ruckpaul, K. & Heinemann, U. (2001) Covalently crosslinked complexes of bovine adrenodoxin with adrenodoxin reductase and cytochrome P450scc. Mass spectrometry and Edman degradation of complexes of the steroidogenic hydroxylase system. Eur. J. Biochem. 268, 1837–1843. Cauet, G., Balbuena, D., Achstetter, T. & Dumas, B. (2001) CYP11A1 stimulates the hydroxylase activity of CYP11B1 in mitochondria of recombinant yeast in vivo and in vitro. Eur. J. Biochem. 268, 4054–4062. Duport, C., Spagnoli, R., Degryse, E. & Pompon, D. (1998) Selfsufficient biosynthesis of pregnenolone and progesterone in engineered yeast. Nature Biotechnol. 16, 186–189. Yang, H., Bard, M., Bruner, D.A., Gleeson, A., Deckelbaum, R.J., Aljinovic, G., Pohl, T.M., Rothstein, R. & Sturley, S.L.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

(1996) Sterol esterification in yeast: a two-gene process. Science. 272, 1353–1356. Gietz, R.D., Schiestl, R.H., Willems, A.R. & Woods, R.A. (1995) Studies on the transformation of intact yeast cells by the LiAc/ SS-DNA/PEG procedure. Yeast. 11, 355–360. Adams, A., Gottsching, D., Kaiser, C. & Stearns, T. (1997) Methods in Yeast Genetics: a Laboratory Course Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Baudin, A., Ozier-Kalogeropoulos, O., Denouel, A., Lacroute, F. & Cullin, C. (1993) A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae. Nucleic Acids Res. 21, 3329–3330. Dumas, B., Cauet, G., Lacour, T., Degryse, E., Laruelle, L., Ledoux, C., Spagnoli, R. & Achstetter, T. (1996) 11 Beta-hydroxylase activity in recombinant yeast mitochondria. In vivo conversion of 11-deoxycortisol to hydrocortisone. Eur. J. Biochem. 238, 495–504. Degryse, E., Dumas, B., Dietrich, M., Laruelle, L. & Achstetter, T. (1995) In vivo cloning by homologous recombination in yeast using a two- plasmid-based system. Yeast 11, 629–640. Cullin, C. & Pompon, D. (1988) Synthesis of functional mouse cytochromes P-450 P1 and chimeric P-450 P3–1 in the yeast Saccharomyces cerevisiae. Gene 65, 203–217. Wach, A., Brachat, A., Pohlmann, R. & Philippsen, P. (1994) New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae. Yeast 10, 1793–1808. Ness, F., Achstetter, T., Duport, C., Karst, F., Spagnoli, R. & Degryse, E. (1998) Sterol uptake in Saccharomyces cerevisiae heme auxotrophic mutants is affected by ergosterol and oleate but not by palmitoleate or by sterol esterification. J. Bacteriol. 180, 1913– 1919. Zinser, E. & Daum, G. (1995) Isolation and biochemical characterization of organelles from the yeast, Saccharomyces cerevisiae. Yeast 11, 493–536. Loeper, J., Louerat-Oriou, B., Duport, C. & Pompon, D. (1998) Yeast expressed cytochrome P450 2D6 (CYP2D6) exposed on the external face of plasma membrane is functionally competent. Mol. Pharmacol. 54, 8–13. Mihara, K., Blobel, G. & Sato, R. (1982) In vitro synthesis and integration into mitochondria of porin, a major protein of the outer mitochondrial membrane of Saccharomyces cerevisiae. Proc. Natl Acad. Sci. USA. 79, 7102–7106. Daum, G., Bohni, P.C. & Schatz, G. (1982) Import of proteins into mitochondria. Cytochrome b2 and cytochrome c peroxidase are located in the intermembrane space of yeast mitochondria. J. Biol. Chem. 257, 13028–13033. Hermann, G.J., Thatcher, J.W., Mills, J.P., Hales, K.G., Fuller, M.T., Nunnari, J. & Shaw, J.M. (1998) Mitochondrial fusion in yeast requires the transmembrane GTPase Fzo1p. J. Cell Biol. 143, 359–373. Serrano, R., Kielland-Brandt, M.C. & Fink, G.R. (1986) Yeast plasma membrane ATPase is essential for growth and has homology with (Na+ + K+), K+- and Ca2+-ATPases. Nature. 319, 689–693. Hirschman, J.E., De Zutter, G.S., Simonds, W.F. & Jenness, D.D. (1997) The G beta gamma complex of the yeast pheromone response pathway. Subcellular fractionation and protein–protein interactions. J. Biol. Chem. 272, 240–248. Jackson, D.D. & Stevens, T.H. (1997) VMA12 encodes a yeast endoplasmic reticulum protein required for vacuolar H+-ATPase assembly. J. Biol. Chem. 272, 25928–25934. Zinser, E., Paltauf, F. & Daum, G. (1993) Sterol composition of yeast organelle membranes and subcellular distribution of enzymes involved in sterol metabolism. J. Bacteriol. 175, 2853–2858. Cauet, G., Degryse, E., Ledoux, C., Spagnoli, R. & Achstetter, T. (1999) Pregnenolone esterification in Saccharomyces cerevisiae.


39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

A potential detoxification mechanism. Eur. J. Biochem. 261, 317– 324. Gachotte, D., Sen, S., Eckstein, J., Barbuch, R., Krieger, M., Ray, B. & Bard, M. (1999) Characterization of the Saccharomyces cerevisiae ERG27 gene encoding the 3-keto reductase involved in C-4 sterol demethylation. Proc. Natl Acad. Sci. USA. 96, 12655– 12660. Zinser, E., Sperka-Gottlieb, C.D., Fasch, E.V., Kohlwein, S.D., Paltauf, F. & Daum, G. (1991) Phospholipid synthesis and lipid composition of subcellular membranes in the unicellular eukaryote Saccharomyces cerevisiae. J. Bacteriol. 173, 2026–2034. Anderson, R.A. & Sando, G.N. (1991) Cloning and expression of cDNA encoding human lysosomal acid lipase/cholesteryl ester hydrolase. Similarities to gastric and lingual lipases. J. Biol. Chem. 266, 22479–22484. Zweytick, D., Leitner, E. & Kohlwein, S.D., Yu, C., Rothblatt, J. & Daum, G. (2000) Contribution of Are1p and Are2p to steryl ester synthesis in the yeast Saccharomyces cerevisiae. Eur. J. Biochem. 267, 1075–1082. Nacken, V., Achstetter, T. & Degryse, E. (1996) Probing the limits of expression levels by varying promoter strength and plasmid copy number in Saccharomyces cerevisiae. Gene. 175, 253–260. Kolling, R. & Hollenberg, C.P. (1994) The ABC-transporter Ste6 accumulates in the plasma membrane in a ubiquitinated form in endocytosis mutants. EMBO J. 13, 3261–3271. Chen, C., Ingram, M.F., Rosal, P.H. & Graham, T.R. (1999) Role for Drs2p, a P-Type ATPase and potential aminophospholipid translocase, in yeast late golgi function. J. Cell Biol. 147, 1223– 1236. Akwa, Y., Ladurelle, N., Covey, D.F. & Baulieu, E.E. (2001) The synthetic enantiomer of pregnenolone sulfate is very active on memory in rats and mice, even more so than its physiological neurosteroid counterpart: distinct mechanisms? Proc. Natl Acad. Sci. USA. 98, 14033–14037. Venkatramesh, M., Guo, D.A., Harman, J.G. & Nes, W.D. (1996) Sterol specificity of the Saccharomyces cerevisiae ERG6 gene product expressed in Escherichia coli. Lipids 31, 373–377. Moebius, F.F., Bermoser, K., Reiter, R.J., Hanner, M. & Glossmann, H. (1996) Yeast sterol C8–C7 isomerase: identification and characterization of a high-affinity binding site for enzyme inhibitors. Biochemistry. 35, 16871–16878. Lees, N.D., Skaggs, B., Kirsch, D.R. & Bard, M. (1995) Cloning of the late genes in the ergosterol biosynthetic pathway of Saccharomyces cerevisiae – a review. Lipids. 30, 221–226. Metherall, J.E., Waugh, K. & Li, H. (1996) Progesterone inhibits cholesterol biosynthesis in cultured cells. Accumulation of cholesterol precursors. J. Biol. Chem. 271, 2627–2633. Metherall, J.E., Li, H. & Waugh, K. (1996) Role of multidrug resistance P-glycoproteins in cholesterol biosynthesis. J. Biol. Chem. 271, 2634–2640.

FEBS 2003 52. Lewis, T.A., Rodriguez, R.J. & Parks, L.W. (1987) Relationship between intracellular sterol content and sterol esterification and hydrolysis in Saccharomyces cerevisiae. Biochim. Biophys. Acta. 921, 205–212. 53. Sturley, S.L. (2000) Conservation of eukaryotic sterol homeostasis: new insights from studies in budding yeast, Biochim. Biophys. Acta. 1529, 155–163. 54. Yeaman, S.J. (1990) Hormone-sensitive lipase – a multipurpose enzyme in lipid metabolism. Biochim. Biophys. Acta. 1052, 128– 132. 55. Beckert, V. & Bernhardt, R. (1997) Specific aspects of electron transfer from adrenodoxin to cytochromes p450scc and p45011beta. J. Biol. Chem. 272, 4883–4888. 56. Cao, P.R. & Bernhardt, R. (1999) Interaction of CYP11B1 (cytochrome P-45011 beta) with CYP11A1 (cytochrome P-450scc) in COS-1 cells. Eur. J. Biochem. 262, 720–726. 57. Tuckey, R.C. & H.M. (2002) Placental cytochrome P450scc (CYP11A1): comparison of catalytic properties between conditions of limiting and saturating adrenodoxin reductase. J. Steroid Biochem. Mol. Biol. 81, 153–158. 58. Neve, E.P. & Ingelman-Sundberg, M. (2000) Molecular basis for the transport of cytochrome P450 2E1 to the plasma membrane. J. Biol. Chem. 275, 17130–17135. 59. Akiyoshi-Shibata, M., Sakaki, T., Yabusaki, Y., Murakami, H. & Ohkawa, H. (1991) Expression of bovine adrenodoxin and NADPH-adrenodoxin reductase cDNAs in Saccharomyces cerevisiae. DNA Cell Biol. 10, 613–621. 60. Hara, T., Koba, C., Takeshima, M. & Sagara, Y. (2000) Evidence for the cluster model of mitochondrial steroid hydroxylase system derived from dissociation constants of the complex between adrenodoxin reductase and adrenodoxin. Biochem. Biophys. Res. Commun. 276, 210–215. 61. Wu, D. & Cederbaum, A.I. (1992) Presence of functionally active cytochrome P-450IIE1 in the plasma membrane of rat hepatocytes. Hepatology. 15, 515–524. 62. Sakaki, T., Akiyoshi-Shibata, M., Yabusaki, Y. & Ohkawa, H. (1992) Organella-targeted expression of rat liver cytochrome P450c27 in yeast. Genetically engineered alteration of mitochondrial P450 into a microsomal form creates a novel functional electron transport chain. J. Biol. Chem. 267, 16497–16502. 63. Chudaev, M.V., Gilep, A.A. & Usanov, S.A. (2001) Site-directed mutagenesis of cytochrome b5 for studies of its interaction with cytochrome P450. Biochemistry (Mosc). 66, 667–681. 64. Thierry, A., Fairhead, C. & Dujon, B. (1990) The complete sequence of the 8.2 kb segment left of MAT on chromosome III reveals five ORFs, including a gene for a yeast ribokinase. Yeast. 6, 521–534. 65. Lecain, E., Chenivesse, X., Spagnoli, R. & Pompon, D. (1996) Cloning by metabolic interference in yeast and enzymatic characterization of Arabidopsis thaliana sterol delta 7-reductase. J. Biol. Chem. 271, 10866–10873.