JOHANNES V. SWINNEN, DAVID R. JOSEPH, AND MARCO CONTI*. The Laboratories for Reproductive Biology, Departments of Pediatrics and Physiology, ...
Proc. Nati. Acad. Sci. USA Vol. 86, pp. 8197-8201, November 1989 Biochemistry
The mRNA encoding a high-affinity cAMP phosphodiesterase is regulated by hormones and cAMP (follicle-stimulating hormone/desensitization/Sertoli cell/glioma cell)
JOHANNES V. SWINNEN, DAVID R. JOSEPH, AND MARCO CONTI* The Laboratories for Reproductive Biology, Departments of Pediatrics and Physiology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599
Communicated by G. D. Aurbach, May 30, 1989 (received for review February 6, 1989)
a role in regulating cell responsiveness to various stimuli
To elucidate the mechanisms by which horABSTRACT mones regulate cAMP phosphodiesterases (PDEs), a group of cDNA clones that had been isolated from a rat Sertoli cell library were characterized. These cDNAs are derived from a single gene (ratPDE3). The deduced amino acid sequence of the ratPDE3 cDNA corresponds to a 66,200-Da protein homologous to other testicular PDEs, to the Drosophila melanogaster dunce-encoded cAMP PDE, and to bovine and yeast PDEs. Expression of ratPDE3 in eukaryotic and prokaryotic cells leads to the appearance of a cAMP PDE with properties identical to the cAMP PDE purified from Sertoli cells. Although of different size, transcripts corresponding to ratPDE3 were present in all organs studied. In the immature Sertoli cell in culture, the level of mRNA transcripts of ratPDE3 was increased more than 100-fold by follicle-stimulating hormone or N6,02'-dibutyryladenosine 3',5'-cyclic monophosphate treatment. Stimulation of ratPDE3 mRNA by N',02'-dibutyryladenosine 3',5'-cyclic monophosphate was also observed in a C6 glioma cell line. These data demonstrate that cAMP regulates the expression of one of its own degrading enzymes by an intracellular feedback mechanism that involves changes in mRNA levels.
(6-15). Recently, cDNA clones coding for a yeast (16) and a Drosophila PDE (5) with characteristics similar to cAMP PDEs have been isolated. This has paved the way to the isolation of mammalian cDNAs coding for PDEs (3, 17, 18). By screening rat testicular libraries with the Drosophila dunce clone (5), this laboratory has isolated cDNAs that code for four different but homologous PDEs (3). Here we report the complete structure and characterization of one of these PDEs.t We also demonstrate that mRNA levels for this PDE are regulated by hormones and cAMP in the Sertoli cell and in a glioma cell line.
MATERIALS AND METHODS Cell Cultures. Sertoli cell cultures were prepared from 15-day-old Sprague-Dawley rats as described (19) and cultured for 4 days in medium without serum. C6 glioma cells were cultured in Ham's F10 medium with 15% (vol/vol) horse serum. The glioma cells were washed with fresh medium without serum and incubated in this medium for 17 hr. After this preincubation, cell cultures were treated for 5-24 hr with 0.5 mM N6,02'-dibutyryladenosine 3',5'-cyclic monophosphate (Bt2cAMP) or follicle-stimulating hormone (FSH) (NIADDK ov-FSH S16) (100 ng/ml). Control cultures received unsupplemented medium. At the end of the treatments, the medium was removed and the cells were lysed for RNA isolation (20). Sertoli Cell cDNA Library, Screening, and DNA Sequencing. The construction of the Sertoli cell cDNA library, the screening with one of our previously isolated ratPDE cDNA clones, and the DNA sequencing strategy have been described elsewhere (3). Sequences were analyzed by using routines available with the Microgenie (Beckman Instruments) software package. RNA Analysis. Poly(A)+ RNA prepared from tissues and cultured cells (20, 21) was glyoxylated and fractionated by electrophoresis through a 1% agarose gel, transferred to Biotrans nylon membranes (ICN), and hybridized with random-primed 32P-labeled cDNAs (22, 23). Transfection of COS Cells. Monkey kidney cells (COS-7) were grown in high-glucose Dulbecco's modified Eagle's medium with 5% (vol/vol) fetal calf serum to approximately
The biochemical characterization of cyclic nucleotide phosphodiesterases (PDEs) has led to the conclusion that cyclic nucleotide hydrolysis in the cell is catalyzed by a large and heterogeneous family of enzymes. These PDEs can be classified into four major groups on the basis of their affinity for cAMP and cGMP, their intracellular compartmentalization, and their regulation by hormones, neurotransmitters, and intracellular second messengers (1, 2). The least characterized are the type IV PDEs,t a class of enzymes hydrolyzing cAMP with high affinity, but heterogeneous in terms of kinetic behavior and sensitivity to cGMP and to phosphodiesterase inhibitors (1). Purification of these PDE forms has been hampered by this heterogeneity, by their relative low abundance, and their instability. For these reasons, little is known about their molecular structure and the molecular mechanisms that cause their activation. That this class of PDEs has an important role in intracellular cAMP degradation has been emphasized by studies using a new generation of PDE inhibitors specific for type IV PDEs. These inhibitors have been shown to be potent positive inotropic agents, vasodilators, and neurostimulators (4). Furthermore, a cAMP PDE has been characterized in Drosophila melanogaster, and mutations in the gene (dunce) coding for this PDE disrupt learning and fertility (5). Several reports have shown an activation of a cAMP PDE by hormones or neurotransmitters that increase the cAMP level in fibroblasts (6, 7), in a C6 glioma cell line (8), and in Sertoli (9, 10) and granulosa cells (11). This activation is thought to play
Abbreviations: PDE, cyclic nucleotide phosphodiesterase; FSH, follicle-stimulating hormone; Bt2cAMP, N6,02 -dibutyryl adenosine 3',5'-cyclic monophosphate. *To whom reprint requests should be addressed. tThe type IV group of enzymes is also referred to as low Km PDEs, cAMP PDEs, and peak III PDEs. Following suggested guidelines (2), throughout the paper we have used the term cAMP PDE. We have introduced the terms ratPDEl-ratPDE4 to refer to PDEencoding genes of the rat. The presence of these four genes is inferred by our molecular cloning data (3). fThe sequence reported in this paper has been deposited in the GenBank data base (accession no. M25349).
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 8197
8198
Proc. Natl. Acad. Sci. USA 86
Biochemistry: Swinnen et al.
clopentyloxy)-4-methoxyphenyl]-2-pyrrolidone; Berlex, NJ}, and Cilostamide {N-cyclohexy-4-[(1,2-dihydro-2-oxo-6-quinolyl)oxy]-N-methylbutyramide; provided by H. Hidaka, Nagoya University School of Medicine, Showa-Ku, Nagoya 466, Japan} for 5 min. Inhibitors were dissolved in dimethyl sulfoxide. Control tubes contained enzyme and a final dimethyl sulfoxide concentration of 2.5%.
40% confluency and transfected by using DEAE-dextran (24) with 10 ,tg of the pCMV5 expression plasmid (25) containing the ratPDE3 clone in the sense or antisense orientation. Forty-eight hours after the transfection, the medium was removed, and the cells were harvested and homogenized in a buffer containing 20 mM Tris HCl (pH 8.0), 10 mM NaF, 1 mM EDTA, 0.2 mM EGTA, 50 mM benzamidine, leupeptin at 0.5 ,ug/ml, pepstatin at 0.7 ,ug/ml, and 2 mM phenylmethylsulfonyl fluoride. PDE Assay. The PDE assays were performed according to the method of Thompson and Appleman (26) as previously detailed (9, 10). Assay conditions (optimal to measure mammalian high-affinity cAMP PDE activity but unfavorable for the bacterial PDE activity) included 50 mM Tris HCl (pH 8), 10 mM MgCl2, 5 mM 2-mercaptoethanol, and 1 tLM [3H]cAMP (0.1 tCi per tube; 1 Ci = 37 GBq). Incubations were carried out for 3-10 min at 34°C, and separation of the reaction products was as previously described (26). For the determination of the Km of the enzymes (27), approximately 20 ,tg of bacterial extract was incubated with 0.1 ,uCi of [3H]cAMP and 0.1-50 ,M unlabeled cAMP. Activity was measured at three or four different time intervals and expressed as the rate of cAMP hydrolysis per minute. For the inhibitor studies, enzyme was incubated in the presence of 10-8_10-5 M Ro 20-1724 [4-(3-butoxy-4-methoxybenzyl)-2imidazolidinone; Hoffmann-LaRoche], Rolipram {4-[3-(cy-
RESULTS Molecular Cloning of a Rat Sertoli Cell cAMP PDE. On the basis of the finding that FSH and Bt2cAMP increase the activity of a cAMP PDE in the immature rat Sertoli cell (9-11) and that RNA synthesis is required for this activation (10), it was assumed that Sertoli cell stimulation would increase the probability of retrieving clones related to the cAMP PDE. As previously reported, two groups of cDNA clones were isolated (ratPDE3 and ratPDE4) from a cDNA library derived from 15-day-old Sprague-Dawley rat Sertoli cells that, on the fourth day of culture, had been treated with 0.5 mM Bt2cAMP for 24 hr (3). Here, we report the complete sequence and the characterization of the clones corresponding to ratPDE3. Two independent cDNA clones (ratPDE3.1 and ratPDE3.2) of 1.9 and 2.4 kilobases (kb) were sequenced on both strands and were shown to be identical in the overlapping region, except for an internal deletion of 86 nucleotides at the 120
60 180
1
(1989)
.
.t
240
.
AGCCCCCCGCGCGCCGCCGGGCTCTGCMATATGMAGGAGCAGCCCTCATGTGCGGGCACCGGGCATCCGAGCATGGCGGGGTACGGCAGGATGGCCCCCTTTGMACTCGCTGGTGGTCCG
MetLysGLuGLnProSerCysAlaGlyThrG LyH isProSerMetAlaGLyTyrGlyArgMetA LaProPheG LuLeuAlaGLyGLyPro 300
.
.
.
360
.
GTGAAGCGCTTAAGACTGAGTCCCCCTTTCCCTGTCTCTTCGCAGAGGAGGCCTACCAGAAACTGGCCAGCGAGACCCTGGAGGAGCTGGACTGGTGTCTGGACCAGCTAGAGACCCTG 31 Va lLysArgLeuArgThrGluterProPheProCysLeuiPheAlaGluG luA laTyrG lnLysLeuA laSerG luThrLeuG luG luLeuAspTrpCysLeuAspG lnLeuGluThrLeu
CAGACCAGGCACt
420
.
.
.
.
.
480
CCGTCAGTGAGATGGCCTLCCeCMGTTeCAAGAGGATGCTTMTCGGGAGCTCACCCACCTCTCTGAAATGAGTCGGTCTGGCMCCAGGTGTCGGAGTACATATCA 71 G lnTh rArgH isSerVa lSerG luMetA laSerAsnLysPheLysArgMetLeuAsnArgG luLeuTh rH isLeuSerG luMetSerArgSerG lyAsnG lnVa lSerG luTyr I leSer 600 540 AACACGTTCTTAGATAAGCAACATGAAGTAGAAATTCCCTCTCCGACTCAGAAGGAAAAAGAGAAGAAGAAAAG* CGATGTCACAGATCAGTGGGGTCAAGAAGTTGATGCACAGCTCC 111 AsnThrPheLeuAspLysGLnHisGluVaLGul IleProSerProThrGLnLysGLuLysGLuLysLysLysAr roMtSerGlnI LeSerGLyVaLLysLysLeuMetHisSerSer . . . 660 720 AGCCTGACCMATTCCTGCATTCCMAGATTTGGGGTTAAMCAGAGCAGGMAGATGTCCTGGCCMAGGMCETAGMAGACGTGMACMGTGGGGCCTCCACGTTTTCCGMTiAGCGGAGCTG 151 SerLeuThrAsnSerCysI LeProArgPheGLyValLysThrGluGlnGluAspValLeuAlaLysGluLeuGluAspValAsnLysTrpGLyLeuHisVaLPheArgIl eAlaGLuLeu . . . . . 780 840 T CT GGCMACCGGCC-TCT GACTGTTAT CAT GCACACCATT TTT CAGGMACGAGAT T TGTTMAMACGTT TAAMATCCCAGT GGACACTT TGAT TACGTATCTTATGACTCTAGMAGACCAT 191 SerGLyAsnArgProLeuThrVa lIleMetHisThrIl ePheGLnGLuArgAspLeuLeuLysThrPheLysl leProValAspThrLeul LeThrTyrLeuMetThIrLeuGluAspHis . . 900 960 TACCATGCTGACGTGGCCTATCACMACMCATCCATGCTGCAGATGTCGT'CCAGTCMACTCATGTGCTGCTCTCTACACCCGCTTTGGAGGCTGTTTTCACTGACTTGGAGATTCTCGCG 231 TyrHisALaAspVaLAlaTyrHisAsnAsnl leHisAlaAlaAspVaLVaLGLnSerThrHisVaLLeuLeuSerThrProAlaLeuGluAlaValPheThrAspLeuGLuI leLeuAla 1020 1080 GCCAT TTTT GCCAGTGCMATACATGATGT GGATCAT CCTGGT GT GTCAAATCMT T TCT GATCMATACMMACT CGGAACT TGCCTT GAT GTACAACACTCCTCCGT CTTAGAGMATCAT 271 ALa ILePheAlaSerALal leHisAspVaLAspHisProGlyValSerAsnGLnPheLeul LeAsnThrAsnSerluiLeuALaLeuMetTyrAsspSerSerVaLLeuGL uAsnHis . . . 1200 1140 CATT TGGCTGTGGGCTTTMAGTTGCTCCAGAGMGMMACTGTGACATTTT'CCAGMATCTGACCMMMAGCAAAGACAATCTTTAAGGAAAATGGCCATTGACATTGTACTAGCGACAGAC
311 HisLeuALaVaLGLyPheLysLeuLeuGLnGLuGLuAsnCysAspl lePheGLnAsnLeuThrLysLysGLnArgGLnSerLeuArgLysMetALal LeAspl leVaLLeuALaThrAsp 351
. . . . . 1260 1320 AT GTCAAAGCACATGMAT CTGCT GGCT GATCTGMAAACAATGGTT GMMACGAAGAAGGT GACGAGCT CT GGCGT CCTCCT CCTTGATAACTAT TCTGACAGGATCCAGGTCCTCCAGMAT
MetSerLysHiesMetAsnLeuLeuAlaspLeuLysThrMetValGluThrLysLysValThrSerSerGlyValLeuLeuLeuAspAsnTyrSerAspArgleGLnVaLLeuGlnAsn 1380
.
.
.
.
.
1440
1620
.
.
.
.
.
1680
ATGGTGCACTGTGCAGACCTGAGCMACCCCACAAAGCCACTCCA6CTCTACCGCCAGTGGACGGACCGGATAATGGAGGAGTTCT TCCGTCAGGGGGACCGGGAGCGTGAiGCGTGGCATG 391 MetValHisCysALaAspLeuSerAsnProThrLysProLeuGlnLeuTyrArgGlnTrpThrAspArgIl eMetGluGLuPhePheArgGLnGlyAspArgGLuArgGluArgGtlyMet 1500 1560 GAGATMAGTCCCATGTGTGAiCAAGCACAACGCCTCTGTGGAAAAATCAC'AGGTGGGCTTCATAGACTACATT GTTCATCCACTCTGGGAGACATGGGCAGACCTCGTACATCCCGACGCC 431 GLuI LeSerProMetCysAspLysHisAsnALaSerVaLGLuLysSerGLnVaLGlyPhelLeAspTyr ILeValHisProLeuTrpGLuThrTrpAtlaAspLeuVaLHisProAspALa
CAGGACATTTTGGACACTTTGGAGGACAATCGTGAGTGGTACCAGAGCACAATCCCCCAGAGCCCCTCCCC GCACCTGATGACCAAGAGGACGGCCGTCAGGGACAGACTGAAAAATTC 471 GlnAspi LeLeuAspThrLeuGLuAspAsnArgGLuTrpTyrGlnSerThrI LeProGLnSerProSerPrcA LaProAspAspGlnGluAspGlyArgGlnGlyGlnThrGluLysPhe . . . 1800 1740 CAGTTCGMCETAACCTTAGAGGMAGATGGCGAGTCAGACACTGMMAGGAiCAGTGGMAGTCMAGTGGAGGMGACACTAGCTGCAGTGACTCTMAGACTCTGTGCACCCMAGACTCAGAG 511 G lnPheG luLeuThrLeuG luG luAspG lyG luSerAspThrG luLysAspSerG lySerG lnVa lG luGl uAspThrSerCysSerAspSerLysThrLeuCysThrG lnAspSerG lu . . . 1860 1920
TCCACTGAuITTCCCCTTGLCGAGCAGGTsGGAGAGGAGaCTGTAGCAGAAGAGGAAAGCCAGCCCCACTGGCGTCGCAGACGATTGTTGTCpTGACpCsCAGTaTAGGCTGT
551 SerThrGluIl~eProLeuAspGluGlnValGluGluGluAlaValAlaGluGluGluSerGlnProGlnThrGlyValAlaAspAspCysCysProAspThr
FIG. 1. Nucleotide sequence and deduced amino acid sequence of ratPDE3. The DNA sequence presented is a composite of the nucleotide sequences of two individual clones (ratPDE3.1 and ratPDE3.2). Nucleotide residues are numbered at the top of each line starting with the first nucleotide of clone ratPDE3.1. Amino acid residue numbers starting from the first methionine following the in-frame termination codon TAA (boxed) are indicated at the left. The position of the deletion present in clone ratPDE3.2 is marked by vertical arrows. Horizontal arrows delimit a region of nucleotide and amino acid sequence homology with other PDE sequences (3, 5, 16, 17, 18, 28). The boxed amino acid sequence corresponds to a sequence also present in the cAMP binding region of the R1la regulatory subunit of the cAMP-dependent protein kinase (29, 30).
5' end in clone ratPDE3.2 (Fig. 1). Whether this deletion is due to alternative splicing and therefore physiologically relevant is unknown at present. An open reading frame of 1752 nucleotides was identified starting with the ATG in position 151 of clone ratPDE3.1. This ATG is likely the translation start site because the configuration around the ATG fits the consensus Kozak sequence (31) and an in-frame termination codon is present 39 bases upstream of the ATG (Fig. 1). Within this open reading frame, a nucleotide sequence of 1077 bases was found to share similarity with the nucleotide sequence of the dunce PDE from D. melanogaster (5) and other rat testicular PDEs (3). Characteristics of the Protein Encoded by ratPDE3. The conceptual translate of clone ratPDE3.1 consists of 584 amino acids (Fig. 1) corresponding to a molecular mass of 66,249 Da. More than 17% of the residues are acidic and 9% are basic. Because of this composition, the protein is negatively charged with a calculated pI of 5.8. Analysis of the hydrophobicity of the deduced amino acid sequence according to Kyte and Doolittle (32) showed that several hydrophobic regions were present and were clustered between residues 200 and 450 of the deduced amino acid sequence, with the carboxyl and amino terminus of the protein being mostly hydrophilic. As previously reported (3), by comparing the amino acid sequence with sequences of PDEs of various species, a region of 350 amino acids (Fig. 1) bears considerable homology with the D. melanogaster dunce PDE (71.7%) (5), with a PDE isolated from bovine brain (41%) (28), and, to a lesser extent, with a yeast PDE (35.3% homology between residues 224 and 356) (16). The amino- and the carboxyl-terminal regions are, however, different from the other published sequences. Some similarity was observed with the amino-terminal region of clone DPD (18), which corresponds to ratPDE4 (3). The sequence between amino acid residues 296 and 302 was identical to the amino acid sequence of the cAMP binding domain of the RIla regulatory subunit of the cAMP-dependent protein kinase (29, 30) (Fig. 1), suggesting, as proposed by others (5), that this region of the PDE sequence is involved in substrate binding. Catalytic Properties of the Protein Encoded by ratPDE3. To demonstrate that cDNAs corresponding to ratPDE3 encode a protein with phosphodiesterase activity, the ratPDE3 cDNA was expressed in eukaryotic and prokaryotic cells. Transfection of COS-7 cells with a transient expression vector (25) containing the ratPDE3.1 cDNA led to a 5- to 6-fold increase in high-affinity cAMP PDE activity, whereas no change in PDE activity was observed when the cDNA was inserted in the vector in the opposite orientation [shamtransfected COS-7, 90.5 ± 8.0 (mean ± SEM); sense transfected, 402.7 ± 66.0; antisense transfected, 77.6 ± 7.5 pmol/min per mg of protein; n = 3]. It was also found that Escherichia coli DH5a bacteria transformed with the pGEM-3Z plasmid containing the entire coding region of ratPDE3 (construct pGEM-ratPDE3) expressed high levels of PDE activity (Fig. 2A). Thus, assay of cAMP hydrolyzing activity under conditions optimal for the Sertoli cell cAMP PDE showed that extracts from DH5a wild-type bacteria or bacteria carrying the pGEM-3Z plasmid without an insert have low PDE activity (on the order of 10-20 pmol/min per mg of protein) (Fig. 2A). In contrast, extracts from bacteria transformed with the pGEM-ratPDE3 construct had an activity on the order of 400 pmol/min per mg of protein (Fig. 2A). Furthermore, bacteria transformed with pGEM-3Z containing a truncated PDE clone lacking the putative catalytic domain showed levels of PDE activity indistinguishable from DH5a or pGEM-3Z-containing bacteria (Fig. 2A). Bacterial transformation with constructs containing ratPDE3.2 also produced a large increase in PDE activity (data not shown). Similar results were obtained when
8199
Proc. Natl. Acad. Sci. USA 86 (1989)
Biochemistry: Swinnen et al.
A
.5c
(6)
2 400
0L
E .E 300 E 0 S
o 200 E CL
,~ 100
(3)
(10)
(4)
-# rzLEL _
0
LaJ 0C0
QL
0
0)
r
0 0
to
FI
B 0
3 0
N
.-5 0
En
2
NE0Q 1
-1.0 0.0
2.0
1.0
3.0
4.0
5.0
1 / cAMP concentration
6.0
7.0
8.0
(pM)
C 120 -4-,
0
0X
E E x
E SD
o
100 80
60
--
Ao4.-
A~~~~~~~~~, [ 80~~~~ 0~
"
40
20 0
S.......... K.................. 0.01
0.1
1.0
10
Inhibitor conc. (jiM)
FIG. 2. Characterization of the PDE activity of the protein encoded by ratPDE3. (A) PDE activity in pmol/min per mg of protein in extracts of bacteria containing different plasmid constructs. The number of individual experiments is indicated in parentheses. DH5a, E. coli DH5a without a plasmid; pGEM, DH5a with the nonrecombinant pGEM-3Z plasmid; Full length, DH5a with a pGEM-3Z vector containing the complete cDNA of ratPDE3 inserted in the EcoRl site; Truncated, DH5a with pGEM-3Z containing a PDE clone that lacks the putative catalytic site. (B) Lineweaver-Burk analysis of the PDE activity encoded by ratPDE3. Extracts of DH5a containing the pGEM-ratPDE3 construct were assayed for PDE activity with cAMP concentrations ranging from 0.1 to 5 ALM. A representative experiment of the three performed is reported. The estimated Km was 2.0 ± 0.5 uM (n = 3), whereas the Km estimated for DH5a transformed with pGEM-3Z without an insert exceeded 100 tLM (data not shown). (C) Effects of PDE inhibitors on the activity of extracts of DH5a containing the pGEM-ratPDE3 construct. A, Cilostamide; o, Ro 20-1724; *, Rolipram. The activity present in nontransformed bacteria was not inhibited by these compounds.
8200
Biochemistry: Swinnen et al.
Proc. Natl. Acad. Sci. USA 86
the PDE cDNAs were inserted into an expression vector (pRC23) based on a A promoter (33) (data not shown). Further characterization of the ratPDE3-encoded activity expressed in bacteria showed that cAMP was hydrolyzed with high affinity with an estimated Km of 2.0 ± 0.5 1LM cAMP (mean ± SEM; n = 3) (Fig. 2B). Conversely, in agreement with what was previously reported for the E. coli PDE (34, 35), bacteria transformed with pGEM-3Z without an insert or with a truncated PDE clone hydrolyzed cAMP with a Km higher than 0.1 mM (data not shown). Finally, unlike the bacterial PDE activity, the activity present in E. coli extracts carrying the pGEM-ratPDE3 construct was inhibited by Rolipram (IC50 = 0.56 + 0.1 gM; mean ± SEM), Ro 20-1724 (IC50 = 1.67 tkM ± 0.5), and to a lesser extent by Cilostamide (IC50 > 10 uM) (Fig. 2C). The order of potency of these compounds for the inhibition of the recombinant PDE in bacterial extracts was identical to that shown for the inhibition of the Sertoli cell FSH-stimulated PDE (M.C., unpublished data). Like the Sertoli cell PDE, the recombinant PDE was not inhibited by cGMP (data not shown). Tissue Distribution and Regulation of the mRNA Coding for ratPDE3. Transcripts corresponding to ratPDE3 were found in all rat organs surveyed (Fig. 3). Northern blot analysis of poly(A)+ RNA derived from liver and testis showed only a predominant transcript of 6.8 kb, whereas brain, heart, and kidney showed multiple transcripts ranging between 7.4 and 5.8 kb (Fig. 3). Immature Sertoli cells in culture, a cell system responsive to FSH, expressed only low levels of RNA transcripts corresponding to ratPDE3 (Fig. 4 A and B). Conversely, after treatment of the cells with 0.5 mM Bt2cAMP, hybridization to the 6.8- and 3.2-kb transcripts was detected (Fig. 4A). By densitometry of the autoradiograms, it was shown that more than a 100-fold increase in the signal was induced by Bt2cAMP. Similar results were obtained when cells were stimulated with FSH at 100 ng/ml for 20 hr (Fig. 4B). To assess whether the cAMP induction of ratPDE3 is a phenomenon limited to the Sertoli cell or whether it also occurs in other cells, experiments were repeated with C6 glioma cells, a cell line that has been shown to respond to cAMP and to p-adrenergic agonists with an increase in PDE activity (8, 36). C6 glioma cells were treated for 5 hr with 0.5 mM Bt2cAMP, and the poly(A)+ RNA was probed with ratPDE3 cDNA. In unstimulated cells, ratPDE3 hybridized to transcripts of 6.8 and 4.6 kb (Fig. 4C). In a manner similar to what was observed with Sertoli cells, the Bt2cAMP treatment produced a 10-fold increase in the major 6.8-kb tran-
script (Fig. 4C). -
._
& _
Q
_
kb
6.6-
IN
4.4-4
ato
-
-.b
FIG. 3. Organ distribution of ratPDE3 transcripts. A Northern blot containing 5 jug of poly(A)+ RNA from various tissues of 90-day-old male
Sprague-Dawley rats
was
hybridized
with the 32P-labeled ratPDE3 cDNA. The hybridizing mRNA species were visualized by autoradiography. Size markers were 32P-labeled Hindll-digested A DNA and Hae Ill-digested cX174 DNA fragments.
(1989)
DISCUSSION Hormonal activation of low Km PDEs and the consequent increase in cyclic nucleotide degradation have been observed in many systems, but the biochemical steps that lead to these changes in PDE activity are incompletely understood. In platelets (37), adipocytes (38), and cells responding to insulin (39-41), activation of a cAMP PDE has been shown to be a rapid phenomenon and is probably mediated by protein phosphorylation. In other systems, including Sertoli cells (10), C6 glioma cells (8, 36), and fibroblasts (6, 7), the reported hormonal stimulation of the cAMP PDE develops over a period of 3-24 hr, and the activation requires cAMPdependent protein kinase activity (42) and ongoing RNA and protein synthesis (8, 10). The present study reports the complete primary sequence of one of these cAMP PDEs. The mRNA coding for this PDE is regulated by the cAMP analog Bt2cAMP both in Sertoli cells and in a C6 glioma cell line, indicating that the long-term activation of the cAMP PDE by hormones or neurotransmitters is mediated by the regulation of its mRNA levels in the target cell. Thus, cAMP through an intracellular feedback loop regulates the expression of its own degrading enzymes. Since the primary sequence of the FSH-stimulated cAMP PDE is not known, formal proof of the identity of ratPDE3 with the FSH-stimulated cAMP PDE is not available; however, the following findings support the conclusions that the clones that we have isolated code for the FSH-stimulated cAMP PDE. The homology of the deduced ratPDE3 sequence with the sequences of the Drosophila dunce cAMP PDE (5) and PDEs from bovine (28) and yeast (16), together with the finding that a high-affinity cAMP PDE activity appears with expression of ratPDE3 in both prokaryotic and eukaryotic cells, indicate that ratPDE3 encodes a PDE rather than a PDE activator. By analyzing the deduced sequence, the protein encoded by ratPDE3 is an acidic protein with a calculated pI of 5.8 and a molecular mass of 66,200 Da. Similarly, the cAMP PDE purified from Sertoli cells behaved as an acidic protein with a pI of 5.6, and a 67- to 68-kDa polypeptide copurifies with the cAMP hydrolytic activity (M.C., unpublished results). Furthermore, the PDE that appears in bacteria transfected with the pGEM-ratPDE3 construct and the cAMP PDE isolated from Sertoli cells show very similar affinities for cAMP and similar sensitivities to various phosphodiesterase inhibitors. Finally, transcripts corresponding to ratPDE3 are at the limit of detection in the Sertoli cell when only traces of the cAMP PDE are found in the soluble fraction of the cell. On the contrary, large amounts of ratPDE3 transcripts are induced by Bt2cAMP and FSH concomitantly with the appearance of the cAMP PDE activity in the cell (10). Thus all data thus far collected point to identity between the cloned and the purified PDE. Activation of a cAMP PDE by hormones that increase the cAMP levels has been observed in several in vitro systems. Our finding that transcripts corresponding to ratPDE3 are present in all the organs that we have analyzed supports the hypothesis that this enzyme or similar enzymes encoded by the same gene are expressed in most cells in vivo. Furthermore, the fact that Bt2cAMP increases the mRNA levels of this PDE in Sertoli and neural glioma cells again points to a ubiquitous regulation. These data also support the hypothesis that the increase in cAMP PDE activity depends on an increase in the mRNA level and on an increase in PDE synthesis. We cannot exclude at present the possibility that other PDE genes are activated under these conditions in Sertoli and glioma cells or that posttranslational modification of the protein, like phosphorylation, might be a further step required to produce the observed increase in cAMP hydrolytic activity.
Biochemistry: Swinnen et al. A B3t2cAMP l_-
Bt2cAMP
r- -
|
kb
B
FSH
i
r-
+1
FSH 1-
Bt2cAMP X_ J-T+I
8201
Bt2cAMP X-+1
kb :A
.,e
.FIG. 4. Regulation of ratPDE3 transcripts by Bt2cAMP (A) and FSH (B) in the rat Sertoli cell 6.6-_ t t * and by Bt2cAMP in the C6 glioma cell (C). Sertoli 44w s| ~~cell cultures from 15-day-old Sprague-Dawley rats were grown in medium without serum (19). _ SF Cells were incubated in the absence or in the presence of FSH at 100 ng/ml for 20 hr or 0.5 mM Z Bt2cAMP for 24 hr. After preincubation with Q5--.*V 'I r fresh medium without serum for 17 hr, C6 glioma cells were incubated for additional 5 hr with or v without Bt2cAMP. Northern blots with 5 ,ug of poly(A)+ RNA in each lane were hybridized with ratPDE3 cDNA as probe and autoradiographed. After removal of the probe, the filter was hybridized with actin cDNA (A and B) or ubiquitin cDNA (C) to assess whether similar amounts of RNA were present in all lanes. A, 32P-labeled rat PDE 3 HindIll-digested A DNA fragments. ubiquitin
29A94-_~S -
6.64.4-
6.6
2.3-
4.4-
_
2.0
.. .. n
rcstPDE3 rat PDE 3
C
+I
-
kb
9.4'.-
Proc. Natl. Acad. Sci. USA 86 (1989)
Actin
Actin
The cAMP-mediated expression of a cAMP PDE and the consequent change in the pattern of cAMP degradation might be important in long-term adaptation to different stimuli. It
has been proposed that, together with adenylate cyclase desensitization, an increase in PDE activity is responsible for the onset of long-term desensitization (12). In agreement with this, we have shown that in the Sertoli cell the homologous and heterologous long-term desensitization can be partially or completely reversed by cAMP PDE inhibitors (15). The present data on the cAMP-dependent induction of PDE mRNA further support this hypothesis. This latter observation also implies that an altered pattern of gene expression might be one important facet of the desensitized state. In summary, the elucidation of the primary structure of a Rolipram-sensitive, Ro 20-1724-sensitive cAMP PDE provides tools useful in understanding the structure-function relationship of this class of molecules. The notion that cAMP is regulating the expression of its own degrading enzymes represents a step forward in clarifying the exact role of this PDE in cellular adaptation. In view of the similarity between the Drosophila dunce PDE and this PDE, it is possible that cAMP regulation of this enzyme also plays a role in memory formation. The authors are grateful to Dr. R. Crowl for providing the pRC23 expression plasmid. We also thank Susan Hall for making available to us the RNA from FSH-treated Sertoli cells; Kallen Tsikalas for technical assistance in performing PDE assays; and Frank French, Judson Van Wyk, and Kathleen Horner for their critical reading of this manuscript. This work was supported by National institutes of Health Grant HD20788. The Laboratories for Reproductive Biology are supported by Grant P30-HD18968. 1. Beavo, J. A. (1988) in Advances in Second Messenger and Phosphoprotein Research, eds. Greengard, P. & Robison, G. A. (Raven, New York), Vol. 22, pp. 1-38. 2. Strada, S. J. & Thompson, W. J. (1984) Adv. Cyclic Nucleotide Protein Phosphorylation Res. 16, VI. 3. Swinnen, J. V., Joseph, D. R. & Conti, M. (1989) Proc. Natl. Acad. Sci. USA 86, 5325-5329. 4. Weishaar, R. E., Cain, M. H. & Bristol, J. A. (1985) J. Med. Chem. 28, 537-545. 5. Chen, C. N., Denome, S. & Davis, R. L. (1986) Proc. Nail. Acad. Sci. USA 83, 9313-9317. 6. D'Armiento, M., Johnson, G. S. & Pastan, I. (1972) Proc. Natl. Acad. Sci. USA 69, 459-462. 7. Manganiello, V. & Vaughan, M. (1972) Proc. Natl. Acad. Sci. USA 69, 269-273. 8. Onali, P., Schwartz, J. P., Hanbauer, I. & Costa, E. (1981) Biochim. Biophys. Acta 675, 285-292.
9. Conti, M., Geremia, R., Adamo, S. & Stefanini, M. (1981) Biochem. Biophys. Res. Commun. 98, 1044-1050. 10. Conti, M., Toscano, M. L., Petrelli, L., Geremia, R. & Stefanini, M. (1982) Endocrinology 110, 1189-11%. 11. Conti, M., Kasson, B. G. & Hsueh, A. J. W. (1984) Endocrinology 114, 2361-2368. 12. Harden, T. K. (1983) Pharmacol. Rev. 35, 5-32. 13. Schmidtke, J., Meyer, H. & Epplen, J. T. (1980) Acta Embryol. Exp. 95, 404-413. 14. Geremia, R., Rossi, P., Pezzotti, R. & Conti, M. (1982) Mol. Cell. Endocrinol. 28, 37-53. 15. Conti, M., Monaco, L., Geremia, R. & Stefanini, M. (1986) Endocrinology 118, 901-908. 16. Sass, P., Field, J., Nikawa, J., Toda, T. & Wigler, M. (1986) Proc. Nail. Acad. Sci. USA 83, 9303-9307. 17. Davis, R. L., Takayasu, H., Eberwine, M. & Myres, J. (1989) Proc. Nall. Acad. Sci. USA 86, 3604-3608. 18. Colicelli, J., Birchmeier, C., Michaeli, T., O'Neill, K., Riggs, M. & Wigler, M. (1989) Proc. Natl. Acad. Sci. USA 86, 3599-3603. 19. Monaco, L. & Conti, M. (1986) Biol. Reprod. 35, 258-266. 20. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. & Rutter, W. J. (1979) Biochemistry 18, 5294-5299. 21. Aviv, H. & Leder, P. (1972) Proc. Nald. Acad. Sci. USA 69, 1408-1412. 22. Thomas, P. S. (1980) Proc. Natl. Acad. Sci. USA 77, 5201-5205. 23. Feinberg, A. P. & Vogelstein, B. (1984) Anal. Biochem. 137, 266-267. 24. Lopata, M. A., Cleveland, D. W. & Sollner-Webb, B. (1984) Nucleic Acids Res. 12, 5707-5717. 25. Andersson, S., Davis, D. N., Dahlback, H., Jornvall, H. & Russel, D. W. (1989) J. Biol. Chem. 264, 8222-8229. 26. Thompson, W. J. & Appleman, M. M. (1971) Biochemistry 10, 311-316. 27. Lineweaver, H. & Burk, D. (1934) J. Am. Chem. Soc. 56, 658-666. 28. Charbonneau, H., Beier, N., Walsh, K. A. & Beavo, J. A. (1986) Proc. Natil. Acad. Sci. USA 83, 9308-9312. 29. Scott, J. D., Glaccum, M. B., Zoller, M. J., Uhler, M. D., Helfman, D. M., McKnight, G. S. & Krebs, E. G. (1987) Proc. Natl. Acad. Sci. USA 84, 5192-5196. 30. Bubis, J., Neitzel, J. J., Saraswat, L. D. & Taylor, S. S. (1988) J. Biol. Chem. 263, 9668-9673. 31. Kozak, M. (1986) Cell 44, 283-292. 32. Kyte, J. & Doolittle, R. F. (1982) J. Mol. Biol. 157, 105-132. 33. Crowl, R., Seamans, C., Lomedico, P. & McAndrew, S. (1985) Gene 38, 31-38. 34. Nielsen, L. D., Monard, D. & Rickenberg, H. V. (1973)J. Bacteriol. 116, 857-866. 35. Peterkofsky, A. (1976) Adv. Cyclic Nucleotide Res. 7, 1-48. 36. Schwartz, J. P. & Passonneau, J. V. (1974) Proc. Natl. Acad. Sci. USA 71, 3844-3848. 37. Macphee, C. H., Reifsnyder, D. H., Moore, T. A., Lerea, K. M. & Beavo, J. A. (1988) J. Biol. Chem. 263, 10353-10358. 38. Gettys, T. W., Vine, A. J., Simonds, M. F. & Corbin, J. D. (1988) J. Biol. Chem. 263, 10359-10363. 39. Loten, E. G., Assimacopoulos-Jeannet, F. D., Exton, J. H. & Park, C. R. (1978) J. Biol. Chem. 253, 746-757. 40. Makino, H. & Kono, T. (1980) J. Biol. Chem. 255, 7850-7854. 41. Manganiello, V. C., Yamamoto, T., Elks, M., Lin, M. C. & Vaughan, M. (1984) Adv. Cyclic Nucleotide Protein Phosphorylation Res. 16, 291-302. 42. Bourne, H. R., Tomkins, G. M. & Dion, S. (1973) Science 181, 952-954.
