Rat Seminal Vesicle FAD-dependent Sulfhydryl Oxidase

13 downloads 0 Views 496KB Size Report
Rat Seminal Vesicle FAD-dependent Sulfhydryl Oxidase. BIOCHEMICAL CHARACTERIZATION AND MOLECULAR CLONING OF A MEMBER OF THE NEW.
THE JOURNAL OF BIOLOGICAL CHEMISTRY © 2001 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 276, No. 17, Issue of April 27, pp. 13830 –13837, 2001 Printed in U.S.A.

Rat Seminal Vesicle FAD-dependent Sulfhydryl Oxidase BIOCHEMICAL CHARACTERIZATION AND MOLECULAR CLONING OF A MEMBER OF THE NEW SULFHYDRYL OXIDASE/QUIESCIN Q6 GENE FAMILY* Received for publication, June 2, 1999, and in revised form, January 18, 2001 Published, JBC Papers in Press, January 22, 2001, DOI 10.1074/jbc.M010933200

Be´atrice Benayoun‡, Annick Esnard-Fe`ve‡, Sandrine Castella, Yves Courty, and Fre´de´ric Esnard§ From Equipe “Prote´ases et Vectorisation,” INSERM EMI-U-0010 Universite´ François Rabelais, Faculte´ de Me´decine, 2 bis Boulevard Tonnelle´, 37032 Tours cedex, France

Rat FAD-dependent sulfhydryl oxidase was purified; partial sequencing indicated that it was homologous to human quiescin Q6. A cDNA (GenBank™ accession no. AF285078) was cloned from rat seminal vesicles, and active recombinant sulfhydryl oxidase was expressed in Chinese hamster ovary epithelial cells. This 2472-nucleotide cDNA has an open reading frame of 1710 base pairs, encoding a protein of 570 amino acids including a 32-amino acid leader sequence and two potential sites for N-glycosylation. One of them is used and the 64,000 Mr purified protein was transformed to 61,000 by the action of endoglycosidase F. Northern blotting and reverse transcription-polymerase chain reaction analyses showed that there were small amounts of sulfhydryl oxidase in the rat testis, prostate, lung, heart, kidney, spleen, and liver, and that the gene was highly expressed in seminal vesicles and epididymis. Rat sulfhydryl oxidase cDNA corresponds to the human cell growth inhibiting factor cDNA, which could be a differently spliced form of quiescin Q6. Comparing sulfhydryl oxidase sequences with those of human quiescin Q6 and mammalian and Caenorhabditis elegans quiescin Q6-related genes established the existence of a new family of FAD-dependent sulfhydryl oxidase/quiescin Q6-related genes containing protein-disulfide isomerase-type thioredoxin and yeast ERV1 domains.

Sulfhydryl oxidases are enzymes that catalyze the reaction 2R-SH ⫹ O2 3 R-S-S-R ⫹ H2O2. There are at least three families of these enzymes, each depending on different cofactors for their catalytic activity. They are iron-dependent sulfhydryl oxidases (1– 4), copper-containing enzymes (5– 8), and FAD-dependent enzymes. The last of these have been found in the male rat genital tract (9 –11), in fungi (Aspergillus) (12), and in chicken egg white (13). Sulfhydryl oxidase activity in the mammalian male genital tract was first described in hamster epididymal fluid (9) and in * This work was supported by a grant from the Ligue Nationale Contre le Cancer (Comite´ du Cher). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/ EBI Data Bank with accession number(s) AF285078. ‡ These authors contributed equally to this work (A. E.-F. for the sulfhydryl oxidase purification and characterization and B. B. for the cloning, sequencing, and expression of sulfhydryl oxidase cDNA). § To whom correspondence should be addressed: Laboratoire “Enzymologie et Chimie des Prote´ines,” INSERM EMI-U0010, Faculte´ de Me´decine, 2 bis Blvd. Tonnelle´, 37032 Tours cedex, France. Tel.: 33-247-36-62-06; Fax: 33-2-47-36-60-46; E-mail: [email protected].

seminal vesicle fluids from rats and hamsters (9, 10). Rat sulfhydryl oxidase was purified from seminal vesicles. It is a monomeric enzyme with an apparent Mr of 66,000, a pHi of 7.45, and 1 mol of noncovalently bound FAD/mol of enzyme (14). The rat enzyme accepts a variety of small sulfhydryl substrates including glutathione, cysteine, dithiothreitol (DTT)1, and 2-mercaptoethanol and can also markedly enhance the rate of renaturation of fully reduced ribonuclease (14). Several possible functions have been proposed for sulfhydryl oxidase in the rat male genital tract. These include the generation of disulfide bonds in the proteins of seminal plasma or spermatozoa, the preservation of spermatozoan membrane integrity, antimicrobial activity (through the release of H2O2), and the protection of spermatozoa against the harmful effects of thiol after ejaculation (14). It has been shown recently that flavin-dependent sulfhydryl oxidase from chicken egg white contains one redox-active cystine bridge, and accepts a total of 4 electrons per active site (15). This oxidase has a high catalytic activity toward reduced peptides and proteins including insulin A and B chains, lysozyme, ovalbumin, riboflavin-binding protein, and RNase (16). Flavin-dependent chicken egg white sulfhydryl oxidase and protein disulfide isomerase can also cooperate in vitro in the generation and rearrangement of native disulfide pairings (16). Neither protein nor nucleotide sequences were available for FAD-dependent sulfhydryl oxidases until the partial sequencing of chicken egg white sulfhydryl oxidase (17) (published after the original submission of this work) and rat sulfhydryl oxidase (this work). These sequences were similar to those of four mammalian cDNAs. The first of these, encoding human bone-derived growth factor-1 (BPGF-1; GenBank™ accession no. L42379) was cloned from an osteosarcoma cell line. The second, GEC-3 (GenBank™ accession no. U82982), was the product of a gene whose expression is hormone-dependent in the uterine tissue of the guinea pig (Cavia porcellus). The third, encoding the cell growth inhibiting factor (CGIF; GenBank™ accession no. E12644) was cloned from human lung fibroblast. The last is the product of quiescin Q6 (GenBank™ accession no.

1 The abbreviations used are: DTT, dithiothreitol; BPGF-1, human bone-derived growth factor-1; CGIF, cell growth inhibiting factor; DTNB, 5,5⬘-dithiobis-2-nitrobenzoic acid; E-64, [L-trans-epoxysuccinylleucylamido(4-guanidino)butane]; ERV-1, essential for respiration and viability-1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GEC-3, guinea pig endometrial cell-3; hQ6, human quiescin Q6; MALDI-TOF, matrix-assisted laser desorption/ionization time of flight; PDI, protein-disulfide isomerase; PAGE, polyacrylamide gel electrophoresis; SOx, sulfhydryl oxidase; Z-Phe-Arg-NH-Mec, carbobenzoxy-Lphenylalanyl-L-arginine-4-methylcoumarinyl-7-amide; Mec, 4-methyl7-coumarylamine; DMEM, Dulbecco’s modified Eagle’s medium; RT, reverse transcription; PCR, polymerase chain reaction; bp, base pair(s).

13830

This paper is available on line at http://www.jbc.org

Cloning of Rat Seminal Vesicle FAD-dependent SOx U972760) (18), a gene that is specifically expressed, together with collagen ␣ chains, decorin, and complement C1r, as cultured human lung fibroblasts begin to leave the proliferative cycle and enter quiescence (19). The quiescin Q6 gene has evolved by the fusion of two ancient genes, those for thioredoxin and ERV1 (18). The protothioredoxin gene first evolved into thioredoxin and protein-disulfide isomerase (PDI) genes; then the quiescin Q6 form diverged from the PDI form (18). We have now cloned and sequenced rat seminal vesicle FADdependent sulfhydryl oxidase cDNA. We believe that this enzyme, together with quiescin Q6 and quiescin Q6-related genes, constitutes a new family of FAD-dependent sulfhydryl oxidaserelated proteins. They contain protein-disulfide isomerase/ thioredoxin and ERV1 domains, which may assist in the folding of a number of newly synthesized or secreted proteins. EXPERIMENTAL PROCEDURES

Materials—DTT, 5–5⬘ dithiobis-2-nitrobenzoic acid (DTNB), glutathione and papain were purchased from Roche Molecular Biochemicals (Meylan, France). The papain (EC 3.4.22.2) was repurified on a Mono-S column (Amersham Pharmacia Biotech) to remove the glycyl-endopeptidase (EC 3.4.22.25) which usually contaminates the commercial enzyme (20). Z-Phe-Arg-NH-Mec (Bachem, Bubendorf, Switzerland) was prepared as a 10 mM stock solution in 10% dimethylformamide. E-64 was obtained from the Peptide Research Institute (Osaka, Japan). H2O2 was supplied by Sigma. All other reagents were of analytical grade. Catalase was prepared from frozen rat liver. Purification and Sequencing of Sulfhydryl Oxidase from Rat Seminal Vesicle Fluid—Seminal vesicle fluid from Wistar rats was collected directly (21). It was diluted in four volumes of 0.01 M potassium phosphate buffer, pH 6.8, and homogenized in an Ultraturrax. The fraction of the centrifuged homogenate precipitated by 40 –70% (NH4)2SO4 was redissolved and dialyzed against 0.05 M Tris-HCl, pH 8.0, 0.15 M NaCl and then chromatographed on carboxymethylated papain-Sepharose to eliminate rat cystatin C, a known inhibitor of lysosomal cysteine proteinase inhibitors (22). Rat cystatin C was detected immunochemically in the proteins eluted with phosphate buffer pH 12.0. Most of the papain inhibitory activity was in the flow-through of the column. The unbound material was equilibrated with 1.5 M (NH4)2SO4 in 0.01 M potassium phosphate buffer, pH 6.8, and loaded on to a phenyl-Sepharose column (1.1 ⫻ 7.0 cm) (Amersham Pharmacia Biotech). Proteins were eluted with a gradient of (NH4)2SO4 decreasing from 1.5 to 0.0 M. The column was then washed with 30% ethylene glycol and the inhibitory activity eluted with a gradient of urea (0.0 – 6.0 M) in 0.01 M potassium phosphate buffer, pH 6.8. The inhibiting fractions were dialyzed against 0.01 M potassium phosphate buffer, pH 6.0, and loaded onto a hydroxyapatite column (2.1 ⫻ 7 cm) (Bio-Rad). Proteins were eluted by a potassium phosphate, pH 6.0, gradient from 0.01 to 0.20 M. The purity of the fractions was checked by reverse-phase chromatography using an Aquapore BU300 (C4) cartridge (2.1 ⫻ 30 mm; Applied Biosystems) and an acetonitrile gradient (0 – 60%) in 0.075% trifluoroacetic acid. The molecular mass of the eluted peak was determined by MALDI-TOF mass spectrometry using a Bru¨cker reflex mass spectrometer. The amino acid sequence was determined on an Applied Biosystems 477A pulsed liquid sequencer using the chemicals and programs recommended by the manufacturer. Phenylthiohydantoin derivatives were identified using an on-line Applied Biosystems model 120A analyzer. Rat sulfhydryl oxidase cleavage was carried out on the reduced and pyridylethylated enzyme (23) after purifying by reverse-phase chromatography on an Aquapore BU300 (C4) cartridge using an acetonitrile gradient (0 – 60%) in 0.075% trifluoroacetic acid. CNBr cleavage was performed as previously described (24). The resulting material was either analyzed directly by reverse-phase chromatography on Aquapore BU300 or submitted to endoproteinase Lys-C, used according to the manufacturer’s directions (Roche Molecular Biochemicals), followed by reverse-phase chromatography. Deglycosylation of Rat Sulfhydryl Oxidase—Rat sulfhydryl oxidase was deglycosylated using endoglycosidase F/N-glycosidase F (Sigma) at concentrations of 0.08 –2.5 units/ml. Fractions were incubated overnight in 20 mM phosphate buffer, pH 6.5, 10 mM EDTA, 1% Triton X-100, 0.2% SDS in the presence of 1% 2-mercaptoethanol and then analyzed by SDS-PAGE. Papain Inhibiting Activity—Purified papain was titrated against E-64 (25). Papain (0.01 ␮M range) and sulfhydryl oxidase (nanomolar range) were incubated for 30 min at 30 °C in 0.1 M phosphate buffer, pH

13831

6.8, 1 mM EDTA, 0.1% Brij, and 2 mM DTT in the presence or not of catalase (about 50 nM). The residual enzyme activity was measured with Z-Phe-Arg-NH-Mec (5 ␮M final) on a Dynatech Microfluor reader or a F-2000 Hitachi spectrofluorimeter. Sulfhydryl Oxidase Assays—Dithiotreitol (2 mM) was oxidized in 0.1 M Tris-HCl buffer, pH 8.0, at 25 °C. Aliquots (7 ␮l) of the incubation mixture were removed and added to 650 ␮l of 0.54 mM DTNB in the same buffer, and the absorbance at 412 nm was measured (26). Total RNA Extraction, Poly(A)⫹ RNA Purification, and Northern Blot—Total RNA was extracted from Wistar rat tissues using the TriReagent method (Euromedex, France) following the manufacturer’s instructions. Poly(A)⫹ RNA was isolated using oligo(dT) (27). For Northern blot analysis, total RNA (20 ␮g) or poly(A)⫹ RNA from rat seminal vesicles (1 ␮g) was separated on 1% agarose gel, transferred to a nylon membrane, and prehybridized for 1 h at 68 °C with the QuickHyb hybridization solution (Stratagene). A 300-bp probe was generated from the 3⬘-untranslated region of rat SOx cDNA by PCR using the forward primer (5⬘-AACATCGTCAGAGAC-3⬘) and the reverse primer (5⬘-AGCTGGGTAGGCCAGAGAA-3⬘). One hundred nanograms of this probe was [␣-32P]dCTP random primer-labeled and added to prehybridization buffer. Hybridization was performed overnight at the same temperature. Membranes were washed at 68 °C for 20 min in 2⫻ SSC, 0.1% SDS and twice for 30 min at 68 °C in 0.1⫻ SSC, 0.5% SDS. They were then exposed to Kodak AR x-ray film at ⫺70 °C using intensifying screens for 4 h to 7 days. RNA markers (Promega) were used to determine the size of SOx transcript. Reverse Transcription-Polymerase Chain Reaction (RT-PCR) Analysis—The first cDNA strand was synthesized from 5 ␮g of total RNA prepared from various rat tissues using Moloney murine leukemia virus reverse transcriptase (Promega) and oligo(dT)17 primer following the manufacturer’s instructions. At the end of reverse transcription, the reaction mixture (20 ␮l) was diluted to 1000 ␮l with 10 mM Tris, pH 7.5, 1 mM EDTA and 1 ␮l of this solution was used to amplify a 400-bp SOx fragment or a 800-bp GAPDH fragment (internal standard). The sequences of the 5⬘ and 3⬘ primers for SOx were 5⬘-ACTTGAGCGAGGTGGACAGTCAAG-3⬘ and 5⬘-AGCACAGGCACTCGGGAA-3⬘; for GAPDH they were 5⬘-AGTTCAACGGCACAGTCAAGGCTGAGAAT-3⬘ and 5⬘-GAGGGGCCTCCACAGTCTTCTGAGTGGC-3⬘. Amplification was performed in a Progene thermocycler (Techne) using Platinum® Pfx DNA polymerase (Life Technologies, Inc.). After one cycle at 94 °C for 2 min, 62 °C for 2 min, and 68 °C for 40 min for second strand synthesis, the thermocycling parameters were 30 cycles of amplification at 96 °C for 10 s, 59 °C for 30 s, and 68 °C for 90 s. Ten microliters (1/5 of PCR products) were analyzed on 1% agarose gel. Expression of SOx mRNA was quantified by densitometry and corrected using GAPDH as internal standard. cDNA Cloning of Rat SOx, Sequencing, and Plasmid Construction— The 5⬘ end of the rat SOx cDNA was generated as follows. One microgram of poly(A)⫹ mRNA from rat seminal vesicles was first reverse transcribed using a reverse primer 5⬘-GCGAAGAACTCCACCGCCCAG-3⬘. This primer was synthesized on a conserved 5⬘ region of human quiescin Q6, human BPGF-1, and guinea pig GEC-3 cDNAs. After addition of a poly(C) tail to the first cDNA strand with terminal deoxynucleotide transferase (28), cDNA amplification was performed with the same reverse primer and an oligo(dG) primer, using the Expand Long Template PCR system (Roche Molecular Biochemicals). The thermocycling parameters were: one cycle at each of 95 °C for 5 min, 55 °C for 2 min, and 68 °C for 40 min; and 20 cycles at 94 °C for 10 s, 63 °C for 30 s, and 68 °C for 1 min. The 300-bp amplified product was purified by agarose gel electrophoresis and cloned into a pGEM®T vector (Promega). DNA constructions were transformed into XL2-Blue Epicurian Coli ultracompetent cells (Stratagene). The two cDNA strands were sequenced with a Thermosequenase™ II dye terminator cycle sequencing kit (Amersham Pharmacia Biotech) in an ABI PRISM 377 DNA sequencer (PerkinElmer Life Sciences). Based on the sequence of this 5⬘ end fragment, a specific 5⬘ primer, 5⬘-ACTTGAGCGAGGTGGACAGTCAAG-3⬘, was synthesized to generate the fulllength rat sulfhydryl oxidase cDNA. The reverse transcription reaction was performed from 1 ␮g of rat seminal vesicle poly(A)⫹ RNA with an oligo(dT)17 primer for 1 h at 42 °C. Amplification was performed as described in Ref. 28 with 1/1000 of the reverse transcription product, in the presence of an oligo(dT) primer and the specific 5⬘-primer. Cycling parameters were one cycle at 95 °C for 5 min, 58 °C for 2 min and 68 °C for 40 min; 30 cycles at 94 °C for 10 s, 60 °C for 30 s and 68 °C for 4 min; and one final cycle with an elongation step at 68 °C for 15 min. The 2.5-kilobase pair product corresponding to the full-length rat SOx cDNA was then cloned in pCR3.1 T/A cloning vector (Invitrogen) to generate pCR3.1-rSOx plasmid; it was sequenced with forward and reverse

13832

Cloning of Rat Seminal Vesicle FAD-dependent SOx

rSOx-specific primers. The pcDNA/3.1/LacZ/V5-His-TOPO (pcDNA-3LacZ) came from the Eukaryotic TOPO TA cloning kit (Invitrogen). Cell Culture and Transfections—The Hamster ovary epithelial cell strain CHO was maintained in 10% fetal calf serum (Eurobio, France) in Dulbecco’s modified Eagle’s medium (DMEM) (Eurobio) supplemented with 100 ␮g/ml each penicillin and streptomycin (Eurobio) and 2 mM L-glutamine (Eurobio). Transfection was performed using the electroporation method. Briefly, subconfluent CHO cells were trypsinized and washed twice in DMEM supplemented with 10% fetal calf serum. Cells (2.4 106 cells in 400 ␮l of culture medium) with 10 ␮g of pcDNA-3-LacZ or pCR3.1-rSOx plasmid were then subject to electroporation at 150 V for 25 ms in a ECM399 electroporator (BTX, San Diego, CA). They were then plated in 75-cm2 flasks. Forty-eight hours after transfection, the cells were passaged and the medium was supplemented with 1 mg/ml G418. After 2 weeks in this selective medium, resistant colonies were selected, passaged for 48 h in DMEM supplemented with 2% fetal calf serum, and analyzed for expression and enzymatic activity of SOx recombinant protein or for activity of ␤-galactosidase on lacZ-transfected cells. Immunoprecipitation, Western Blot, and Activity of Recombinant Enzyme—One hundred microliters of stably transfected CHO 10⫻ concentrated supernatant were incubated in 500 ␮l of immunoprecipitation buffer (20 mM Tris, pH 7.5, 0.25 M NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.5% SDS, 1 ␮g/ml leupeptin, 1 ␮g/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride) with 5 ␮l of pre-immune or immune rat seminal vesicle SOx serum. After gentle shaking at 4 °C overnight, 50 ␮l of 50% (v/v) Protein A-Sepharose (Amersham Pharmacia Biotech) was added for an additional 2 h. Protein A-Sepharose was then washed four times in immunoprecipitation buffer, resuspended in Laemmli sample buffer, and heated for 3 min at 100 °C (29). Immunoprecipitation was analyzed by Western blot with immune rat seminal vesicle sulfhydryl oxidase serum after electrophoresis on a 12% polyacrylamide gel and transfer to nitrocellulose membranes. Immunoblots were revealed by chemiluminescence with the detection system RPN 2106 (Amersham Pharmacia Biotech). A concentrated supernatant (10⫻) of CHO transfected cells was used to measure recombinant sulfhydryl oxidase activity, as described previously. Activity of 75 ng of native protein measured in 10⫻ concentrated DMEM supplemented with fetal calf serum was used as a control. RESULTS

Purification and Sequencing of Rat FAD-dependent Seminal Vesicle Fluid Sulfhydryl Oxidase—While purifying the specific inhibitors of lysosomal cysteine proteinases from seminal vesicle fluid, using papain as target enzyme, we found a hitherto unreported inhibitory activity that was not bound to carboxymethylated papain Sepharose (data not shown). This was unusual behavior for any known specific inhibitor of the cystatin superfamily (30). The unbound papain inhibitory activity was fractionated by hydrophobic chromatography. The activity was not eluted by either a decreasing gradient of ammonium sulfate or by 30% ethylene glycol, but it was eluted by an increasing gradient of urea without detectable loss of activity. The inhibitory activity was further fractionated on hydroxyapatite gel using a phosphate buffer gradient. The activity eluted with the yellow retained fractions. One peak was at 39 min by reversephase chromatography on Aquapore BU-300 but there was no inhibitory activity in the eluted peak. It corresponds to a protein with a Mr of 64,000 by SDS-PAGE and of 64,624 by mass spectrometry (Fig. 1). These molecular properties were similar to those of the previously described rat sulfhydryl oxidase (EC 1.8.3.2), an androgen-independent protein in seminal vesicle fluid that catalyzes the oxidation of sulfhydryl groups to disulfides with the reduction of oxygen to hydrogen peroxide (11, 14). Since papain activity must be analyzed under reducing conditions, usually in 2 millimolar DTT, the free thiol content of the reaction mixtures in the presence of inhibitory fractions was measured using DTNB. DTT progressively disappeared from the medium as a function of time until it had all been consumed. The parallel production of H2O2 was demonstrated by the appearance of luminescence in the presence of luminol and peroxidase (data not shown), and this luminescence was

FIG. 1. Molecular mass determination of purified and deglycosylated rat seminal vesicle FAD-dependent sulfhydryl oxidase. A, SDS-PAGE analysis of purified sulfhydryl oxidase. Mr St, molecular weight markers; SOx, purified sulfhydryl oxidase (1 ␮g); endo F⫺, purified sulfhydryl oxidase (2 ␮g) in deglycosylation buffer for 16 h at 37 °C; endo F⫹, purified sulfhydryl oxidase (2 ␮g) in presence of 0.4 unit/ml endoglycosidase F/N-glycosidase F in deglycosylation buffer for 16 h at 37 °C. B, MALDI-TOF mass spectrometry of rat seminal vesicle fluid sulfhydryl oxidase, after reverse-phase high pressure liquid chromatography on Aquapore BU300 of hydroxyapatite fractions, using a Bru¨cker reflex mass spectrometer (RI, relative intensity).

totally abolished by incubation with rat liver catalase. The indirect inhibition of papain by sulfhydryl oxidase was partially reversible, because only a fraction of the inactivated papain could be reactivated when placed under reducing conditions (2 mM DTT) (Fig. 2). This inactivation is explained by the direct effect of H2O2 on protease, because papain remains active when sulfhydryl oxidase acts in the presence of catalase (Fig. 2). Direct sequencing of the unreduced or of the pyridylethylated rat sulfhydryl oxidase gave a sequence of 25 amino acids (ARLSVLYS(F/S)ADPLTLLDADTVRGAV). The pyridylethylated protein was therefore submitted to CNBr cleavage and reverse-phase chromatography (Aquapore BU 300). One narrow peak, eluting at 30.4 min, gave the sequence XXVGSPNAAVLXLXI. Finally, endoproteinase Lys-C cleavage was performed directly on the CNBr cleavage product. A reverse-phase chromatography peak at 33.5 min gave the sequence RLIDALESHRDTWPPACPXL. These amino-terminal and internal sequences of the rat sulfhydryl oxidase did not correspond to any protein sequence in data banks and were therefore the first elements of sequence of FAD-dependent sulfhydryl oxidase. However, they were similar to the amino acid sequences derived from two human cDNAs, BPGF-1 and quiescin Q6 (hQ6) and to the amino acid sequence derived from the guinea pig (C. porcellus) GEC-3 gene product. Bone-derived growth factor-1 was cloned from an osteosarcoma cell line. Quiescin Q6 is a gene that is expressed in confluent cultures of human lung fibroblasts (19). The expression of the GEC-3 gene is hormonedependent in the uterine tissue of guinea pig. We therefore decided to clone the rat sulfhydryl oxidase cDNA Molecular Cloning of Rat Seminal Vesicle FAD-dependent Sulfhydryl Oxidase—A two-step cloning strategy was devel-

Cloning of Rat Seminal Vesicle FAD-dependent SOx

13833

FIG. 2. Effect of rat sulfhydryl oxidase on papain activity. a, activity of papain incubated with sulfhydryl oxidase. Papain (0.01 ␮M) was incubated with sulfhydryl oxidase in 0.1 M phosphate buffer, pH 6.8, 1 mM EDTA, 0.1% Brij 35, and 2 mM DTT at 37 °C. Activity of the papain (0.25 nM final) was analyzed at different times with 5 ␮M Z-Phe-Arg-NH-Mec. Fluorescence is expressed as arbitrary units (AU). b, activity of papain incubated with sulfhydryl oxidase in presence of catalase. c, concentration of H2O2 produced by the action of sulfhydryl oxidase on DTT. H2O2 concentration was deduced from the free sulfhydryl concentration determined using DTNB. The H2O2 produced was calculated on the basis of 1/2 H2O2 per R-SH consumed by sulfhydryl oxidase.

oped to obtain the complete cDNA of sulfhydryl oxidase from seminal vesicles using a primer based on the homology between the amino-terminal sequence of the rat enzyme and the deduced open reading frame of human quiescin Q6, guinea pig GEC3, and human BPGF-1 cDNA (as described under “Experimental Procedures”). The rat 2472-bp sulfhydryl oxidase cDNA contains a polyadenylation signal after the stop codon (Fig. 3) showing that it codes for a full-length sulfhydryl oxidase. This cDNA has an open reading frame of 1710 bp coding for 570 amino acids. The deduced protein contains a 32-amino acid putative signal peptide, which leads to a mature protein with a calculated Mr of 60,035, a value distinct from that of the native enzyme, which is 64,624 by mass spectrometry (Fig. 1). Compared with the deduced protein sequence, the higher mass of the native form is probably due to the carbohydrate content of the mature protein, which is glycosylated, as seen in Fig. 1. However, only one of these two putative N-glycosylation sites seems to be used in vivo, as observed by the modification of SOx electrophoretic mobility on SDS-PAGE after incubation with increasing concentrations of endoglycosidase F (data not shown). The deduced amino acid sequence of rat sulfhydryl oxidase contains all the amino acid sequences of the three peptide fragments obtained from the native seminal vesicle enzyme (Fig. 3). Moreover, the NH2 amino acid sequence of native protein, which started by the ARLSV sequence, confirms the presence of a 32-amino acid cleavable signal peptide, which is released in the mature secreted sulfhydryl oxidase protein. Rat FAD-dependent Sulfhydryl Oxidase Defined a New Gene Family Including Human Quiescin Q6 —A search in the GenBank™ data base showed a strong homology between nucleotide sequences of rat sulfhydryl oxidase cDNA and several cDNA sequences including human quiescin Q6, human cell growth inhibiting factor, human bone-derived growth factor-1, and guinea pig GEC-3. Alignment of sulfhydryl oxidase, hQ6, and GEC-3 cDNAs nucleotide sequences reveals 80% identity in their coding regions (data not shown). This is confirmed by amino acid sequence alignment (Fig. 4), where rat sulfhydryl oxidase is 74.6% identical to human quiescin Q6 and 67.5% identical to GEC-3. On the other hand, the 3⬘ nucleotide noncoding sequences are divergent. Human quiescin Q6 has a 728-bp longer 3⬘ end than rat sulfhydryl oxidase, human CGIF,

FIG. 3. Nucleotide and deduced amino acid sequences of rat sulfhydryl oxidase. The rat sulfhydryl oxidase open reading frame contains 1710 bp between ATG (bp 50 –52) and TGA (1760 –1762) coding for a 570-amino acid protein. The predicted amino acid sequence of rSOx contains a signal peptide (amino acids 1–32, according to rat sulfhydryl oxidase numbering). Arrow indicates the first residue of mature purified enzyme as determined by NH2-terminal sequencing. Sequences corresponding to the amino-terminal and internal amino acid sequences are underlined. The poly(A) signal (bp 2418 –2423) is underlined too. Potential N-glycosylation sites are boxed.

and guinea pig GEC-3 cDNAs (Fig. 5). However, there is a great similarity in the 3⬘ end of rat SOx and other cDNAs with conservation in the sequence surrounding the polyadenylation motif (Fig. 5). Rat sulfhydryl oxidase and human Q6 have the same open reading frame starting with a putative signal sequence. Two conserved putative N-glycosylation motifs are also found in these NH2 regions (Fig. 4). Rat SOx, hQ6, and guinea pig GEC-3 contain a thioredoxin domain (amino acids 38 –127) characterized by a WCGHC PDI-type motif inserted in the well conserved SAWAVEFFASWCGHCIAFAPTWK sequence and an ERV1 domain (amino acids 384 –515) containing the CRDCA motif in the conserved FFGCRDCANHFEQM sequence (Fig. 4) (17, 18). Three genes related to human quiescin Q6, CeQ6r1 (GenBank™ accession no. Z69637), CeQ6r2 (GenBank™ accession no. U80848), and CeQ6r3 (GenBank™ accession no. U39646) from Caenorhabditis elegans containing these PDI-like and ERV1 domains are also members of this family (18). The homologies found among these deduced protein se-

13834

Cloning of Rat Seminal Vesicle FAD-dependent SOx

FIG. 4. Alignment of rat sulfhydryl oxidase amino acid sequence with human quiescin Q6, human BPGF-1, and guinea pig GEC-3 amino acid sequences. The residues conserved in all the four proteins are designated by an asterisk. The amino acids conserved among three of the sequences are shaded gray. The CGHC PDI-type thioredoxin motif and the CRDC redox active disulfide bridge are underlined. The two N-glycosylation motifs are boxed. The sequences were aligned using the Clustal W (1.5) multiple sequence alignment program.

FIG. 5. Comparison of the structure of sulfhydryl oxidase cDNA with human quiescin Q6, human cell growth inhibiting factor, and guinea pig GEC-3 cDNAs. Graphical representation of alignment between rat SOx, human hQ6, human CGIF, and guinea pig GEC-3 cDNAs using BLASTN 2.0.14 protein data base search program. Black areas correspond to more than 75% identity and dashed areas to nonhomologous sequences. Polyadenylation signal is absent from GEC-3 cDNA.

quences associated with the sulfhydryl oxidase activity of rat protein favor the definition of a new sulfhydryl oxidase/quiescin Q6 gene family characterized by the original juxtaposition of thioredoxin and ERV1 domains. Expression and Enzymatic Activity of Recombinant Rat Sulfhydryl Oxidase in CHO Cells—The rat sulfhydryl oxidase cDNA was stably expressed in CHO epithelial cells. This cell line, which has no endogenous sulfhydryl oxidase secretion (Fig. 6A), was transfected with the pCR3.1-rSOx plasmid. After 2 weeks in G418 selective medium, resistant colonies were selected and serially passaged. Four stably transfected colonies were assayed for recombinant sulfhydryl oxidase expression. Recombinant protein was immunoprecipitated from transfected cell supernatants or whole cell extracts with immune rat sulfhydryl oxidase serum and subjected to Western blot analysis. Immunoprecipitation revealed a Mr 64,000 band in the supernatant of rSOx-transfected CHO cells, which was lacking

in the supernatant of nontransfected cells and pcDNA3.1-LacZtransfected cells (Fig. 6A). The electrophoretic mobility of the recombinant protein was identical to that of native enzyme from seminal vesicles. Finally, any expression of rat recombinant rat sulfhydryl oxidase protein was observed in immunoprecipitation from transfected whole cell extracts (data not shown); this confirms that sulfhydryl oxidase protein is mainly secreted. Oxidative activity of recombinant protein in the supernatant of stably transfected CHO cells was measured using DTT as a substrate (Fig. 6C). Sulfhydryl oxidase activities correlated with the expression level of recombinant protein, as shown in the Western blot experiment (Fig. 6A). Activity due to recombinant protein in clone 4 supernatant corresponded to the activity of 75 ng of pure native enzyme measured in the same experimental conditions. These data show that recombinant sulfhydryl oxidase secreted by stably transfected CHO cells

Cloning of Rat Seminal Vesicle FAD-dependent SOx

FIG. 6. Expression and enzymatic activity of recombinant SOx protein in stably transfected CHO cells. A, Western blot analysis of recombinant SOx using antibodies raised against rat SOx. Purified native enzyme (15 ng) was used as a control of electrophoretic mobility (lane 1). Lanes 2–7 correspond to immunoprecipitated products (see “Experimental Procedures”) from 100 ␮l of 10⫻ concentrated supernatant of nontransfected CHO cells (lane 2), pcDNA-3-LacZ-transfected cells (lane 3), and pCR3.1-SOx-transfected cell clones 1– 4 (lanes 4 –7 respectively). B, recombinant protein from clone 3 was immunoprecipitated using preimmune serum and Western blot analysis was performed with immune serum as a specificity control. C, Enzymatic activity of recombinant SOx. Initial velocity of sulfhydryl oxidase reaction is expressed as the relative rate of DTT oxidation using the activity of 75 ng of native protein measured in 10⫻ concentrated DMEM supplemented with fetal calf serum as a reference. Samples 1–7 are the same as those in A (conditions of reaction are detailed under “Experimental Procedures”). Each value represents the mean of three distinct determinations ⫾ S.D.

displays enzymatic activity similar to that of rat seminal vesicle fluid enzyme. Rat Sulfhydryl Oxidase mRNA Is Highly Expressed in the Male Reproductive Tract—To study the expression pattern of sulfhydryl oxidase mRNA, a Northern blot experiment was performed on total RNA derived from different rat tissues. As shown in Fig. 7A, the size of the messenger was around 2600 nucleotides, a value in agreement with the 2472 bp of the cloned cDNA. Only 4 h of autoradiography were necessary to detect a strong signal in seminal vesicles and a weaker signal in epididymis, showing a very high and high level of expression, respectively, in those tissues. Exposure for a week is needed to detect presence of sulfhydryl oxidase mRNA in other tissues (data not shown). To confirm the Northern blot analysis and to create a more sensitive detection, the RT-PCR experiment was conducted using specific rat sulfhydryl oxidase and GAPDH (internal control) primers on total RNA isolated from different rat tissues (Fig. 7, B and C). These data showed a predominant sulfhydryl oxidase mRNA expression in seminal vesicles and epididymis but also a lower basal expression in prostate, kidney, testis, heart, liver, spleen, and lung. DISCUSSION

FAD-dependent sulfhydryl oxidases have previously been identified and characterized by their enzymatic activities and their biochemical properties in several tissues and species including rat seminal vesicles (9, 11, 14), hamster seminal vesicles (10), chicken egg white (13), and human platelets (31). Although sulfhydryl oxidases were thought to be members of the pyridine nucleotide disulfide oxidoreductases (32), no data on the nucleotide or amino acid sequences of these molecules were available in the literature. Purification and partial sequencing of rat FAD-dependent seminal vesicle fluid sulfhydryl

13835

FIG. 7. Expression of sulfhydryl oxidase in rat tissues. A, Northern blot analysis of tissue expression of SOx. SOx mRNA in rat tissues was analyzed using a probe corresponding to the 3⬘-untranslated region of SOx. One microgram of seminal vesicles poly(A)⫹ RNA and 20 ␮g of total RNA of other tissues were loaded per lane (exposition 4 h). B, RT-PCR analysis of SOx in tissues. RT-PCR was performed on 5 ␮g of total RNA from different rat tissues using primers specific for rat SOx and GAPDH (internal standard); 10 ␮l of RT-PCR product were separated on 1% agarose gel. For clarity, and quantification, images from the ethidium bromide stained agarose gels are inverted. C, quantification of SOx mRNA in rat tissues. The amounts of SOx mRNA expression level were corrected using GAPDH as internal standard and are represented as the means of values expressed as percentage of the maximum ratio (SOx/GAPDH in seminal vesicles) considered as 100%.

oxidase and molecular cloning of the corresponding cDNA, indicated a great similarity at the nucleotide and the amino acid levels with human quiescin Q6 and guinea pig GEC-3 cDNA. All three deduced protein sequences possessed a putative peptide signal, two conserved potential N-glycosylation sites, and putative thioredoxin and ERV1 domains. Such structural homologies among these different molecules argue strongly in favor of a new gene family and confirm the preliminary work of Hoober et al. (17) on egg white sulfhydryl oxidase and quiescin Q6. This family also includes CeQ6-related sequences of C. elegans and Drosophila (18). Because the quiescin behavior (i.e. induction of gene expression that takes place when cells enter the quiescent phase) seems thus far to be restricted to human quiescin Q6 cDNA (19, 33), this family should be named the sulfhydryl oxidase/quiescin Q6 family until the enzymatic activity of its members, and their behavior during the proliferative cycle, has been fully determined. Two quiescin Q6 transcripts of 2500 and 3200 nucleotides have simultaneously been found in the WI38 human lung fibroblast cell line (where only the longer form has been cloned) (18) and are also present in several of the cell lines or tissues analyzed (33). Furthermore, two human quiescin Q6-related cDNAs, bone-derived growth factor-1 and cell growth inhibiting factor, have also been independently cloned; they show 98% identity with human quiescin Q6, although a few differences cause modifications in their putative open reading frames. These differences could be due to PCR or sequencing errors as well as to real gene modifications. The 3228-nucleotide BPGF-1 cDNA was cloned from an osteosarcoma cell line and seems to correspond to quiescin Q6 cDNA (3298 nucleotides). The 2500nucleotide CGIF cDNA was isolated from another lung fibroblast cell line (MRC-5), and but it lacks a 728-nucleotide segment (1879 –2606 of quiescin Q6) in its 3⬘-noncoding sequence. CGIF cDNA could therefore represent an alternative transcript of quiescin Q6. Rat sulfhydryl oxidase cDNA (2472 nucleotides), which lacks the same 3⬘-noncoding 728-nucleotide segment, appears therefore to be the homologue of human cell growth inhibiting factor cDNA. Analysis of the expression pattern of rat sulfhydryl oxidase

13836

Cloning of Rat Seminal Vesicle FAD-dependent SOx

by Northern blotting indicated that the gene is actively expressed in the epididymis and seminal vesicles. This was confirmed by RT-PCR and agree with the distribution of sulfhydryl oxidase activity in the rat and guinea pig tissues, showing low activity in testis, more activity in the epididymis, and the greatest activity in seminal vesicles, but no activity in muscle, brain, kidney, liver, or lung tissue (9, 10). On the other hand, RT-PCR and Northern blot revealed low expression of the rat sulfhydryl oxidase gene in all the tissues analyzed. This corresponds to the wide variety of human tissues expressing human quiescin Q6, including the heart, placenta, lung, liver, skeletal muscle, and pancreas (33). Therefore, apart from the high concentration of sulfhydryl oxidase in the male reproductive tract of rodents, sulfhydryl oxidase and quiescin Q6 appear to be ubiquitous. Although this molecule is clearly a secretory protein, sulfhydryl oxidase immunoreactivity has also been found in the matrix of the mitochondria of certain human, rat, and hamster testicular cells at specific stages of functional activation (34 – 36). In rat and hamster, this immunoreactivity appears in pachytene spermatocytes at stage I (34). In photoperiodicallyinduced testicular involution in the Djungarian hamster, immunoreactivity reappeared during recrudescence, when the first spermatogenic wave had reached the pachytene stage (37). In mature human testis, moderate sulfhydryl oxidase immunoreactivity has been found in Leydig cells and spermatogonia and in pachytene spermatocytes (36). Sulfhydryl oxidase also seems to be associated with hypospermatogenesis and impaired fertility (38). Sulfhydryl oxidase appears therefore to be implicated in the functional or the activation state of mitochondria, at least in the cells of the mammalian male reproductive tract. The ERV1 gene is essential for mitochondrial biogenesis and the survival of Saccharomyces cerevisiae cells. ERV1 is found in the cytosol and mitochondria of the yeast cell where it plays an essential role in normal mitochondrial morphology and the stability of these organelles. ERV1p has been recently shown to be a FAD-linked sulfhydryl oxidase (39). Its enzymatic activity is supported by the CXXCA redox disulfide motif located in the COOH-terminal domain of the protein (39). The similarity of this motif to the CRDCA motif present in rat sulfhydryl oxidase, in human quiescin Q6 and guinea pig GEC-3, and to the CQECA motif of chicken egg white sulfhydryl oxidase (17) confirms the functional importance of the ERV1 motif in sulfhydryl oxidase activity and in FAD binding (17). However, a difference in FAD attachment strength could be seen because FAD is removed from ERV1p by 6 M urea, whereas rat sulfhydryl oxidase is stable without any loss of FAD or activity in 6 M urea. A difference in catalytic activity is also found because ERV1p is able to oxidize only protein thiols, whereas rat and chicken sulfhydryl oxidases are able not only to oxidize protein thiols but also low molecular weight thiols including glutathione. Nevertheless, independently of enzyme substrate, a sulfhydryl oxidase activity seems to be essential for cellular processes involving mitochondria in lower and higher eukaryotes. Although rat sulfhydryl oxidase has been shown to enhance markedly the renaturation of fully reduced ribonuclease (14, 16) or the reoxidation of different proteins (16),2 which is a major property of protein-disulfide isomerase (40), its biological role in the secretions as well as in mitochondria is still unknown. The increased expression of quiescin Q6 as cultured lung fibroblasts enter quiescence, together with collagen ␣ chains, decorin and complement C1r (19), and the 2

A. Esnard-Fe`ve and F. Esnard, unpublished observations.

extracellular location of quiescin Q6 in quiescent but not in proliferating cells (33), are indicative of a role for this potential sulfhydryl oxidase in extracellular matrix metabolism as cells begin to leave the proliferation state. These distinct locations of members of sulfhydryl oxidase/quiescin Q6 family could reflect different functions for these proteins. The identification of their biological substrates is therefore of importance. The number of molecules possibly showing sulfhydryl oxidase activity is increasing. It includes the ancient rat, hamster, and chicken egg white sulfhydryl oxidases, and their partial sequencing has led to the identification of other members of this new family (human quiescin Q6 and cell growth inhibiting factor, hamster GEC-3, Q6-related gene products of C. elegans or Drosophila) for which possible sulfhydryl oxidase activity has yet to be demonstrated. Their structure includes a PDI-like thioredoxin and a ERV1 domain. Yeast ERV1p has been recently described as a protein FAD-dependent sulfhydryl oxidase; its homologue, human augmenter of liver regeneration (ALRp) is a mammalian hepatic growth factor that can functionally substitute for ERV1 in yeast (41) and is also probably a sulfhydryl oxidase. Therefore sulfhydryl oxidases are essential molecules in lower and higher eukaryotes implicated in mitochondrial function as well as in the cellular growth regulation, and could be important in pathological states such as cancer development. Acknowledgments—We thank V. Schubnel Courjou for animal handling, M. Brillard-Bourdet for the sequence analysis, and M. Ferrer-Di Martino for the mass spectrometry determination. The English text was edited by Dr. Geoff Watts. REFERENCES 1. Janolino, V. G., and Swaisgood, H. E. (1975) J. Biol. Chem. 250, 2532–2538 2. Isaacs, C. E., Pascal, T., Wright, C. E., and Gaull, G. E. (1984) Pediatr. Res. 18, 532–535 3. Schmelzer, C. H., Swaisgood, H. E., and Horton, H. R. (1982) Biochem. Biophys. Res. Commun. 107, 196 –201 4. Clare, D. A., Pinnix, I. B., Lecce, J. G., and Horton, H. R. (1988) Arch. Biochem. Biophys. 265, 351–361 5. Lash, L. H., Jones, D. P., and Orrenius, S. (1984) Biochim. Biophys. Acta 779, 191–200 6. Lash, L. H., and Jones, D. P. (1986) Arch. Biochem. Biophys. 247, 120 –130 7. Goldsmith, L. A. (1987) Methods Enzymol. 143, 510 –515 8. Takamori, K., Thorpe, J. M., and Goldsmith, L. A. (1980) Biochim. Biophys. Acta 615, 309 –323 9. Chang, T. S., and Morton, B. (1975) Biochem. Biophys. Res. Commun. 66, 309 –315 10. Chang, T. S., and Zirkin, B. R. (1978) Biol. Reprod. 18, 745–748 11. Ostrowski, M. C., Kistler, W. S., and Williams-Ashman, H. G. (1979) Biochem. Biophys. Res. Commun. 87, 171–176 12. de la Motte, R. S., and Wagner, F. W. (1987) Biochemistry 26, 7363–7371 13. Hoober, K. L., Joneja, B., White, H., III, and Thorpe, C. (1996) J. Biol. Chem. 271, 30510 –30516 14. Ostrowski, M. C., and Kistler, W. S. (1980) Biochemistry 19, 2639 –2645 15. Hoober, K. L., and Thorpe, C. (1999) Biochemistry 38, 3211–3217 16. Hoober, K. L., Sheasley, S. L., Gilbert, H. F., and Thorpe, C. (1999) J. Biol. Chem. 274, 22147–22150 17. Hoober, K. L., Glynn, N. M., Burnside, J., Coppock, D. L., and Thorpe, C. (1999) J. Biol. Chem. 274, 31759 –31762 18. Coppock, D. L., Cina-Poppe, D., and Gilleran, S. (1998) Genomics 54, 460 – 468 19. Coppock, D. L., Kopman, C., Scandalis, S., and Gilleran, S. (1993) Cell Growth Differ. 4, 483– 493 20. Buttle, D. J., Kembhavi, A. A., Sharp, S. L., Shute, R. E., Rich, D. H., and Barrett, A. J. (1989) Biochem. J. 261, 469 – 476 21. Esnard, A., Esnard, F., Faucher, D., and Gauthier, F. (1988) FEBS Lett. 236, 475– 478 22. Esnard, A., Esnard, F., and Gauthier, F. (1988) Biol. Chem. Hoppe Seyler 369, 219 –222 23. Hawke, D., and Yuan, P. (1987) Applied Biosystems Bulletin 28, Applied Biosystems, Foster City, CA 24. Esnard, F., Esnard, A., Faucher, D., Capony, J. P., Derancourt, J., Brillard, M., and Gauthier, F. (1990) Biol. Chem. Hoppe Seyler 371, 161–166 25. Barrett, A. J., Kembhavi, A. A., Brown, M. A., Kirschke, H., Knight, C. G., Tamai, M., and Hanada, K. (1982) Biochem. J. 201, 189 –198 26. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70 –77 27. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 28. Frohman, M. A. (1994) PCR Methods Appl. 4, S40 –S48 29. Laemmli, U. K. (1970) Nature 227, 680 – 685 30. Turk, V., and Bode, W. (1991) FEBS Lett. 285, 213–219

Cloning of Rat Seminal Vesicle FAD-dependent SOx 31. Kaul, D., Dhawan, V., and Kaur, M. (1996) Mol. Cell. Biochem. 159, 81– 84 32. Williams, C. H. J. (1990) Chem. Biochem. Flavoenzymes III, 121–211 33. Coppock, D., Kopman, C., Gudas, J., and Cina-Poppe, D. A. (2000) Biochem. Biophys. Res. Commun. 269, 604 – 610 34. Kumari, M., Aumuller, G., Bergmann, M., Meinhardt, A., and Seitz, J. (1990) Histochemistry 94, 365–371 35. Bergmann, M., Oehmen, F., Kumari, M., Aumuller, G., and Seitz, J. (1991) J. Reprod. Fertil. 91, 259 –265

13837

36. Aumuller, G., Bergmann, M., and Seitz, J. (1991) Cell Tissue Res. 266, 23–28 37. Bergmann, M., Kumari, M., Aumuller, G., Hoffmann, K., and Seitz, J. (1990) Int. J. Androl. 13, 488 – 499 38. Bergmann, M., Aumuller, G., Seitz, J., and Nieschlag, E. (1992) Cell Tissue Res. 267, 209 –214 39. Lee, J., Hofhaus, G., and Lisowsky, T. (2000) FEBS Lett. 477, 62– 66 40. Lyles, M. M., and Gilbert, H. F. (1991) Biochemistry 30, 619 – 625 41. Hofhaus, G., Stein, G., Polimeno, L., Francavilla, A., and Lisowsky, T. (1999) Eur. J. Cell Biol. 78, 349 –356