| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Cell Biology and Physiology, University of Pittsburgh, Pittsburgh, Pennsylvania 15261
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cyclic adenosine monophosphate, gene regulation, Sertoli cells, signal transduction, spermatogenesis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CREB and CREM are members of the bZIP family of transcription factors that contain within their carboxy-terminal region a basic domain responsible for DNA recognition located adjacent to a leucine zipper required for dimerization [6]. These transcription factors bind as dimers to cAMP response elements (CREs) within gene promoters, and when phosphorylated on a specific serine, the proteins interact with the CREB binding protein (CBP) coactivator [7, 8]. The addition of CBP contributes to the induction of gene transcription by facilitating the recruitment of general transcription factors and RNA polymerase to the promoter.
Whereas the CREM gene promoter is not thought to be regulated [9], the CREB gene has been shown to be positively autoregulated by the binding of CREB protein to CREs within the CREB promoter [10–12]. The CREB promoter can also be induced by NF-
B transcription factors through 2 B enhancer motifs in the CREB promoter [13]. In addition, an Sp1 transcription factor binding motif is required for basal and cAMP-inducible CREB promoter activity [10], although the direct effects of Sp1 expression on CREB promoter activity were not tested previously.
Alternative exon splicing is responsible for the generation of numerous isoforms of CREB and CREM in the testis [2]. Thus far, only the 2 ubiquitous forms of CREB, the full length 341-amino acid protein and the 327-amino acid exon D-deleted isoforms of CREB have been detected in Sertoli cells [14]. Both CREB341 and CREB327 activate gene transcription. The only CREM protein detected in Sertoli cells is a repressor isoform called inducible cAMP early repressor (ICER) [9]. ICER has been shown to compete with CREB for binding CREs within the CREB promoter. Repression of CREB promoter activity by ICER is thought to limit FSH induction of the CREB gene in Sertoli cells to stages II–VI of the cycle of the seminiferous epithelium [15].
In germ cells, CREM mRNA is not expressed until the pachytene spermatocyte stage of development, when mRNAs encoding various repressor isoforms are produced [16, 17]. After the completion of meiosis, an activator form of CREM (CREM-
) is expressed in round spermatids [17] followed by the production of an inhibitory isoform of CREM lacking transactivation domains (CREM
C-G) coincident with the termination of transcription activity in elongated spermatids [18]. A number of alternatively spliced CREB mRNAs encoding repressors have been described in germ cells. Pachytene spermatocytes in stages I–VI of the cycle of the seminiferous epithelium express an alternatively spliced CREB transcript called CREB-W [14, 19]. CREB-W contains an extra exon (W) that encodes a stop codon, resulting in a carboxy-terminal truncated protein that cannot bind DNA or dimerize with other bZIP proteins [14]. As a result of the premature termination of CREB-W protein translation, initiation codons downstream of exon W are used to produce 16- and 8-kDa inhibitor CREB proteins (I-CREBs) that repress CREB-mediated transcription [20, 21]. The I-CREB proteins are expressed in pachytene spermatocytes during stages V–XIV. A second alternatively spliced CREB mRNA (CREB-Y) containing an extra exon that causes premature termination of CREB has also been detected in pachytene spermatocytes during stages VI–XII [14, 19], but neither the predicted 6-kDa truncated protein nor I-CREB-like proteins have been identified to result from this transcript.
Although previous studies have identified repressor forms of CREB and various CREM proteins expressed in germ cells from the midpachytene stage onward, activator forms of CREB in germ cells have not been characterized. Furthermore, the regulation of CREB gene expression and the potential requirement for stimulation of CRE-mediated transcription in early germ cells is not understood. As has been demonstrated for CREM in spermatid germ cells, CREB may be an important inducer of gene expression in earlier germ cells. Therefore, the expression of the activator form of CREB in germ cells was investigated. In this study, we demonstrated that in testis seminiferous tubules the activator form of CREB is limited to those cells containing Sp1, including Sertoli cells and primary spermatocytes from the preleptotene through midpachytene stages of development. We also demonstrated that Sp1 activates the CREB promoter. In further characterization of the CREB promoter, we identified an additional regulatory element required for CREB promoter activity located between the Sp1 and CREB binding motifs. The newly characterized regulatory element specifically bound protein(s) present in Sertoli cells and germ cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adult rat testes were decapsulated and digested with collagenase (0.5 mg/ml; 33°C, 12 min) in enriched Krebs-Ringer bicarbonate (EKRB) medium prepared as described previously [22]. The resultant seminiferous tubules were washed 3 times in EKRB and digested in trypsin (0.5 mg/ml) and DNase I (1 µg/ml) until a single-cell suspension was obtained. Cells were pelleted in Dulbecco modified Eagle medium (DMEM) with 10% fetal bovine serum and were fractionated by unit gravity sedimentation through a 2–4% BSA gradient [22]. Collected cell fractions were analyzed by differential interference contrast light microscopy, and fractions (80–90% pure) containing similar cell types were pooled. To prepare partially purified germ cells from 17-day-old rats, a single-cell suspension was prepared from testis as described above. Somatic cells and aggregated cells were pelleted (25 x g, 5 min), and the remaining germ cell enriched fraction was collected by centrifugation (600 x g, 5 min). Animals used in these studies were maintained and killed according to the principles and procedures described in the NIH Guide for the Care and Use of Laboratory Animals.
Cell Culture, Extract Preparation, and Immunoblotting
Primary Sertoli cells were maintained in serum-free medium as described previously [10]. JEG-3 human choriocarcinoma and GC-1 spg mouse spermatogonia-spermatocyte cell lines were cultured in DMEM plus 10% fetal bovine serum. Cells were disrupted in ELB+ buffer (250 mM NaCl, 0.1% NP40, 50 mM Hepes, pH 7.0, 5 mM EDTA, 0.5 mM dithiothreitol) and a protease inhibitor cocktail to prepare whole-cell extracts [23]. Nuclear extracts were prepared by detergent lysis [24] as previously described [25]. For immunoblotting, whole-cell extracts were fractionated by SDS-PAGE, transferred to polyvinylidene fluoride membranes, and probed with R1090 antisera directed against CREB amino acids 92–124 [14]. Western immunoblot analysis was performed using the enhanced chemiluminescence system.
Immunohistochemistry
Immunostaining was performed on paraffin-embedded sections (5 µm) from adult rat testis fixed in Bouin fixative. Testis sections were deparaffinized in xylene, rehydrated, and then permeabilized for 1 min in cold 100% methanol. The sections were microwaved on high power for 20 min in either citrate buffer (10 mM citrate, 30 mM NaCl, pH 5.5) for Sp1 immunostaining or glycine buffer (50 mM glycine, 0.01% EDTA, pH 3.0) twice for 5 min with a 5-min intermediate incubation for CREB. The sections were left undisturbed for 20 min, washed twice in PBS, and blocked with normal goat serum, 0.5% BSA, and 0.15% glycine. The tissue sections were then incubated for 24–48 h with rabbit polyclonal antisera against Sp1 amino acids 436–454 (Sp1 PEP2; Santa Cruz Biotechnology, Santa Cruz, CA) or against CREB amino acids 5–24 (CREBNT; Upstate Biotechnology, Lake Placid, NY). Each antiserum was incubated in the presence of either the PEP2 or CREBNT peptides (200 ng/ml). Anti-rabbit biotinylated secondary antibody (Vectastain Elite ABC Kit; Vector Laboratories, Burlington, CA) was added, and immunocomplexes detected using an AEC solution (0.02% 3-amino-9-ethylcarbazole, 5% N,N-dimethyl formamide, 0.015% H2O2, 0.1 M sodium acetate, pH 5.0) as the colorimetric reagent. Slides were washed and counterstained with hematoxylin. A charge coupled device video camera was used to capture images of stained tissue sections.
Transient Transfections, Plasmid Constructs, and Luciferase Assays
Drosophila SL2 (embryonic, epithelial origin) cells in 25-cm2 flasks were transfected using Lipofectin (Life Technologies, Gaithersburg, MD) according to the manufacturer's instructions for 1 h with 1 µg of -1264 CREBLUC [13] reporter plasmid and 1 µg of empty pPacO expression vector or pPacO-encoding Sp1 (pPacOSp1) [26]. Whole-cell lysates were prepared 48 h after transfection, and luciferase assays were performed as described previously [25]. Primary Sertoli cells isolated from 16-day-old rats as described previously [10] and JEG-3 cells were transfected with -1264 CREBCAT vectors and derivatives (5 µg/60-mm2 dish). Sertoli cells were transfected using FuGENE reagent (Roche Molecular Biochemicals, Indianapolis, IN) [13], and JEG-3 cells were transfected using the calcium phosphate method [10]. Luciferase (LUC) and chloramphenicol acetyltransferase (CAT) activities were determined 48 h posttransfection and normalized for protein concentration as determined by a Bradford assay (Bio-Rad Laboratories, Richmond, CA).
Glutathione S-Transferase Pull-Down Assays
Bacterial proteins produced in Escherichia coli BL21 cells transformed with pGEX1Sp1FLU (a gift from R. Tjian) or pGEXCREB were used for glutathione S-transferase (GST) pull-down assays as previously described [27]. The 35S-labeled in vitro-translated protein was prepared in a rabbit coupled transcription translation system (Promega, Madison, WI) using either CREB or CREB-W cDNAs in the pCRII vector (Invitrogen, Carlsbad, CA) as a substrate for Sp6 polymerase. For the pull-down assays, 50 µl of a 100-µl in vitro translation reaction was loaded onto glutathione sepharose 4B columns (Pharmacia, Piscataway, NJ). To measure the relative levels of CREB and CREB-W in vitro translation products, 5 µl of the reaction was fractionated in parallel with the column-eluted proteins on SDS-polyacrylamide gels and visualized by fluorography.
Site-Directed Mutagenesis
Polymerase chain reaction (PCR)-based mutagenesis was used to generate a HindIII restriction site by incorporating a GG-to-TT mutation at -133 and -132 of the CREB promoter. The -1264 CREBCAT vector [10] was used as the substrate in PCRs. The -278 to -129 region of the CREB promoter was amplified using the 5' primer 5'-GGGCCCGAATTCTCGAGCTGCTCCGGGGC-3' and the 3' primer 5'-ACTAAGCTTCTCCCAGCCGCC-3'. The CREB promoter extending downstream of -145-including vector sequences was amplified using the 5' primer 5'-GCTGGGAGAAGCTTAGTGTTGGTCA-3' and the 3' primer 5'-GGTTATAGGTACATTGAGC-3' derived from the CAT gene. The underlined sequences denote HindIII restriction sites and the mutated bases are shown in bold. The 2 amplified DNA fragments were digested with HindIII and ligated together, and the resulting mutated CREB promoter fragment was amplified by PCR. High fidelity pfu polymerase was used for all PCRs. The PCR fragment was digested with XhoI, and the resulting -278 to -51 mutated fragment was used to replace the -278 to -51 region of XhoI-digested -1264 CREBCAT MCS mt to produce the -1264 CREBCAT GG-to-TT plasmid. The -1264 CREBCAT MCS mt construct was derived from -1264 CREBCAT by using site-directed mutagenesis to replace the HindIII site of the polylinker with an SacI site. The -1264 CREB CAT TT-to-GG vector was digested with HindIII, and the annealed oligonucleotides 5'-AGCTCTGCAGATATC-3' and 5'-AGCTGATATCTGGAC-3' or 5'-AGCTTCTGCAGGCGGCCGCT-3' and 5'-AGCTAGCGGCCGCCTGCAGA-3' having HindIII compatible ends were inserted to produce the 21- and 15-base pair (bp) inserts, respectively. All constructs were confirmed by DNA sequencing.
Electrophoretic Mobility Shift Assays and UV Crosslinking
Probes encompassing the -144 to -119 region of the CREB promoter were prepared by annealing wild-type 5'-CTGGGAGAAGCGGAGTGTTGGTGAGT-3' or mutant coding strand templates containing either 2 mismatches (5'-CTGGGAGAAGCTTAGTGTTGGTGAGT-3') or 4 mismatches (5'-CTGGGAGAACGTTAGTGTTGGTGAGT-3') to a complimentary 10-base primer (5'-ACTCACCAAC-3') followed by filling in the overhang with the Klenow fragment of DNA polymerase in the presence of [
-32P]dATP, dCTP, and dGTP and equimolar amounts of dTTP and 5-bromo-2'-deoxyuridine 5'-trisphosphate. For DNA-binding studies, 1–5 µg of nuclear protein extract was incubated with the 32P-labeled probes as previously described [25]. For electrophoretic mobility shift assays (EMSAs), the DNA-protein complexes were resolved via nondenaturing PAGE. DNA-protein binding was analyzed by autoradiography. For ultraviolet (UV) crosslinking, the DNA-protein binding reactions were subjected to UV irradiation (302 nm) for 15 min and fractionated by SDS-PAGE as previously described [25]. The crosslinked DNA-protein complexes were identified by autoradiography.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To survey CREB protein expression during germ cell development, testis fractions enriched for germ cells at specific stages of development were prepared by density gradient sedimentation. Western analysis of whole cell extracts from the pachytene spermatocyte, round spermatid, elongated spermatid, and residual body fractions and from primary Sertoli cell cultures isolated from 16-day-old rats confirmed that CREB was present in primary Sertoli cells (Fig. 1). CREB was also detected in the enriched spermatocyte fraction, although at lower levels than that of Sertoli cells. CREB expression was below detectable levels in round and elongated spermatids and in residual bodies. It is not likely that the immunoreactive CREB band present in spermatocytes was due to cross-reactivity with the closely related CREM protein [16] because CREM is not expressed until the spermatid stage of development [17, 28]. A 29-kDa protein corresponding to the carboxy-terminal truncated CREB-W isoform was also detected in pachytene spermatocytes and round spermatids, in agreement with earlier reports [14, 20].
|
Sp1 and CREB Are Expressed in Sertoli Cells and Germ Cells Through the Midpachytene Stage of Spermatogenesis
Because CREB promoter activity is reduced by the mutation of a Sp1 motif in the CREB promoter [10], immunocytochemistry studies were performed using testis tissue to determine whether CREB expression was limited to cells in which Sp1 was present. In the seminiferous tubules, both CREB and Sp1 immunostaining was present only in the nuclei of Sertoli cells and early germ cells from preleptotene to midpachytene spermatocytes (Fig. 2). Specifically, Sp1 and CREB immunostaining was evident in the nuclei of interstitial (Leydig) cells. Sp1 staining remained fairly constant in Sertoli cells throughout the cycle of the seminiferous epithelium. As demonstrated previously [14], CREB immunoreactivity of Sertoli cell nuclei was highest in stages III–VII, with little or no CREB detectable in stages IX–XIV. In germ cells, Sp1 and CREB immunostaining was most intense for preleptotene, leptotene, and zygotene spermatocytes during stages VI–XIII. Sp1 and CREB were also detectable in the nuclei of pachytene spermatocytes during stages XIV–III, with levels diminishing to near background by stage VIII.
|
Sp1 Stimulates CREB Promoter Activity
The coincident expression of CREB and Sp1 in early germ cells followed by the simultaneous loss of expression during the pachytene spermatocyte stage raised the possibility that Sp1 is required for CREB expression in germ cells. Further support for the potential regulation of CREB expression by Sp1 was provided by an earlier study demonstrating that an Sp1 binding motif is required for basal and cAMP-inducible CREB promoter activity [10]. Because the presence or absence of Sp1 in germ cells may determine whether CREB is expressed, Sp1 regulation of the CREB gene promoter was studied using transient transfection assays. Sp1 overexpression and antisense studies have proven difficult to conduct in mammalian cell lines because of high levels of endogenous Sp1; therefore, Drosophila SL2 cells were used to study Sp1 actions [26]. The SL2 cell line is a common model used to study transcription factor actions because it contains general transcription factors that are similar to those in mammalian cells, but the Drosophila cells lack many sequence-specific mammalian transcription factors, including Sp1. SL2 cells transfected with the -1264 CREBLUC vector containing the full-length CREB promoter linked to the luciferase reporter gene displayed relatively low basal transcription activity (Fig. 3). Addition of an expression vector containing the Drosophila actin promoter alone did not alter CREB promoter activity; however, cotransfection of the same vector having the Sp1 gene driven by the actin promoter resulted in a 72-fold induction of CREB promoter activity. These data demonstrate that Sp1 is a potent inducer of CREB promoter activity.
|
GST Pull-Down Assays Do Not Detect Interactions Between CREB and Sp1
The Sp1 motif within the CREB promoter, previously found to be required for basal and cAMP inducible expression, is located at position -148, exactly 3 helical turns (32 bp) upstream of a CRE motif that is in turn located 2 helical turns (21 bp) upstream of a second CRE motif [10]. Because the spacing of the transcription factor binding sites predicts that the binding factors are all located on the same side of the DNA helix, the possibility was investigated that the Sp1 and CREB proteins binding to these sites physically interact. Furthermore, it was thought that Sp1-CREB interactions may be significant for the regulation of the CREB promoter because Sp1 may facilitate CREB binding to the imperfect CREs or the 2 proteins may contribute to a more complex structure that facilitates the recruitment of general transcription factors to the CREB promoter. To determine whether Sp1 and CREB could directly interact, GST pull-down assays were performed. For these studies, chimeric proteins containing CREB or Sp1 linked to GST were produced in bacteria. Cell extracts were tested by Western blot analysis and DNA binding assays to confirm the production of the proteins (data not shown). 35S-Labeled CREB and the 29-kDa carboxy-CREB-W isoform lacking the bZIP dimerization domain [14] were produced by coupled in vitro transcription/translation and incubated separately with column-immobilized GST, GST-CREB, or GST-Sp1 proteins. After washing, proteins retained on the columns were eluted and fractionated by SDS-PAGE. As expected, in vitro-translated CREB was able to form dimers with bacterially expressed GST-CREB, but the CREB-W isoform lacking the leucine zipper dimerization domain was not able to interact with GST-CREB (Fig. 4). Neither CREB nor CREB-W was bound by GST-Sp1. These data suggest that CREB and Sp1 do not form high affinity interactions and that these proteins act independently to support transcription from the CREB promoter.
|
A Region Including -133 and -132 Specifically Binds Nuclear Proteins and Is Required for Full CREB Promoter Activity
Additional studies originally designed to determine whether the induction of the CREB promoter depended on Sp1 and CREB maintaining their relative positions on the DNA helix unexpectedly identified a new element regulating CREB promoter activity. A HindIII restriction site created by mutating GG to TT at positions -133 and -132 of the CREB promoter was used to insert random 15- or 21-bp (1.5 or 2 helical turns) DNA sequences between the Sp1-1 and CRE-2 motifs. Because of the low transfection efficiency of primary germ cells, the activity of various CREB promoter-CAT chimeras was first tested in human choriocarcinoma JEG-3 cells. In addition to the wild-type promoter (-1264 CREB), an intermediate product of the site-directed mutagenesis process having 2 bases altered in the vector polylinker (-1264 CREBmcs mt), another construct having the HindII site introduced into the CREB promoter (-1264 CREB GG to TT), and constructs containing the 1.5- and 2-helical turn insertions (15-bp insert and 21-bp insert, respectively) were transfected into the JEG-3 cells. Both the 1.5- and 2-helical turn insertions severely limited CREB promoter activity (74% and 95%, respectively); however, the GG-to-TT mutation similarly reduced CREB promoter activity (83%) (Fig. 5A).
|
To determine whether sequences between the Sp1 and CREB binding sites were also required for expression in testis cells, transfection studies were extended to primary rat Sertoli cells. Because the -1264 CREB and -1264 CREBmcs mt constructs had nearly identical activities in JEG-3 cells, only the -1264 CREBmcs mt plasmid was used for comparison in Sertoli cells. As in JEG-3 cells, mutation of sequences between the Sp1 and CRE motifs reduced CREB promoter activity by at least 50% compared with -1264 CREBmcs mt. Specifically, the insertion of 1.5 and 2 helical turns decreased promoter activity by 52% and 63%, respectively (Fig. 5B). The GG-to-TT mutation reduced promoter activity by 67%. The JEG-3 and Sertoli cell transfection data suggest that a specific DNA sequence including positions -133 and -132 between the Sp1 and CRE motifs is required for basal CREB promoter activity.
After the discovery that the region between the Sp1 and CRE motifs was required for full CREB promoter activity, studies were performed to determine whether transcription factors bound to this region. EMSAs were performed using nuclear extracts from immature primary rat Sertoli cells isolated from 15-day-old rats or from 17-day-old rat germ cells. The extracts were incubated with either a wild-type probe containing the region between the Sp1 and CRE motifs (positions -144 to -119) or a similar probe containing the GG-to-TT mutation at positions -133 and -132. The DNA-protein binding studies demonstrated that at least 2 complexes (C1 and C2) were formed with the wild-type probe but that complex formation was dramatically reduced using the mutant probe (Fig. 5C). These studies indicate that a protein or proteins specifically bind to the CREB promoter and that the binding site includes positions -133 and -132 that are required for full CREB promoter activity.
As a first step toward identification of the proteins present in the C1 and C2 complexes, we sought to determine the relative molecular weights of the protein(s) binding to the region. A DNA-protein crosslinking approach was used in which a photoreactive oligonucleotide probe encompasing CREB promoter positions -144 to -119 was incubated with Sertoli cell nuclear proteins followed by exposure of the binding complex to UV light. The resulting crosslinked DNA-protein adducts were then fractionated by SDS-PAGE and identified by autoradiography. Lower molecular weight complexes of 29 and 32 kDa and higher molecular weight complexes ranging from 50 to 78 kDa were formed with the wild-type probe (Fig. 5D). The proteins present in the complexes are predicted to be slightly smaller than suggested by the relative mobilities of the DNA-protein adducts because the probe contributes approximately 5 kDa to the complex [29]. The DNA-protein interactions were specific; the lower molecular weight complexes were abolished, and the higher molecular weight complexes were greatly reduced by employing a mutant probe altered at positions corresponding to -135 to -132 of the CREB promoter.
Because FSH activates CREB in Sertoli cells [10] and has the potential to stimulate Sp1 actions [30], FSH induction of DNA-protein complex formation in the region between the binding sites for Sp1 and CREB was also investigated. Stimulation of Sertoli cells for 6 or 24 h with ovine FSH (100 ng/ml) did not alter the levels of any DNA-protein complex, suggesting that proteins binding to this region are not induced by FSH.
Further crosslinking studies were conducted to determine whether germ cells contain similar proteins binding to the CREB promoter region. Wild-type and mutant probes were incubated with nuclear extracts isolated from the GC-1 spg cell line that exhibits characteristics of germ cells between the B spermatogonia and primary spermatocyte stages of development [31]. The pattern of DNA-protein complex formation for GC-1 spg cells was similar to that observed using Sertoli cell extracts except that an additional complex of 38 kDa was observed and some of the 50- to 72-kDa complexes were absent. The specificity of the GC-1 spg nuclear protein-DNA interactions was again indicated by the fact that complex formation was markedly reduced using the mutant probe. These data suggest that Sertoli and GC-1 spg cells express many of the same proteins binding to the -133 to -132 region of the CREB promoter.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The limitation of CREB and Sp1 protein expression exclusively to early germ cells differs from the findings of earlier reports. As part of an earlier survey of numerous tissues, Sp1 immunostaining was identified in the testis but reportedly only in round spermatids [34]. However, the inability to detect Sp1 in the somatic cells of the testis in the earlier study is unusual and raises the possibility that the immunostaining that was detected may have been due to Sp1 antisera bound nonspecifically to the components of the spermatid acrosome. In spermatids, only a 2.4-kb Sp1 mRNA is produced, which is predicted to encode an Sp1 protein that differs from the somatic form by the absence of 7 amino-terminal amino acids [33]. Because the antiserum used in this study was directed against amino acids 436–454 of Sp1, the Sp1 protein predicted to be produced by the 2.4-kb transcript probably would have been identified if produced. CREB expression has also been described in spermatids [14]; however, the antiserum used for these early studies, was later found to cross-react with CREM after the CREB-related protein was identified [16]. Therefore, the immunoreactivity detected in spermatids probably was due to the presence of CREM.
Although Sp1 regulation of CREB gene expression was not directly studied in germ cells, the results of the SL2 cell transfection studies in which Sp1 induced CREB promoter activity 70-fold suggests that Sp1 is a potent activator of CREB in mammalian cells, including germ cells. This notion is supported by results from a previous study in which mutation of the Sp1-1 motif in the CREB promoter resulted in a 50% decrease in basal activity in Sertoli cells [10]. The loss of Sp1 and Sp1 stimulation of the CREB promoter after the midpachytene stage of germ cell development would be predicted to decrease CREB promoter activity by at least the same extent as the mutation of the Sp1 binding site. However, further promoter and gene expression studies employing spermatocytes will be required to confirm these predictions.
The finding that Sp1 activates CREB promoter activity and the known positioning of the binding sites for Sp1 and CREB on the same sides of the DNA helix raised the possibility that Sp1 and CREB interact. The results of GST pull-down assays suggest that Sp1 and CREB do not interact directly; however, the 2 transcription factors are both capable of binding with the TAFII110 general transcription factor [35, 36]. The presence of the 2 transcription factors may increase the recruitment of TAFII110 and associated TATA binding protein (TBP) to the CREB promoter. CREB and Sp1 also bind to different regions of the CBP [7, 8, 37] and may act in concert to recruit this coactivator to the CREB promoter. Sp1 is also capable of binding to recognition sites on DNA that are constituted into nucleosomes [38] and may act to recruit coactivators to the CREB promoter that are needed to open the chromatin and facilitate the access of CREB or other transcription factors. The opening of the chromatin after Sp1 binding may be important to facilitate CREB binding. The CREs present within the CREB promoter match the consensus CRE at 5 of 8 positions and may not be ideal binding sites for CREB. FSH, through the activation of cAMP-dependent protein kinase A (PKA) is a major regulator of CREB activity in Sertoli cells. PKA also induces Sp1 DNA binding and transactivation properties [30], providing for the possible stimulation of the CREB promoter by a coordinated Sp1 and CREB response to FSH.
Mutagenesis of positions -133 and -132 of the CREB promoter between Sp1 and CRE binding sites reduced promoter activity by more than 70%. This newly identified regulatory region of the CREB promoter specifically bound proteins from Sertoli cells and germ cells. The protein(s) binding to the region may provide a link between the flanking CREB and Sp1 proteins or may act independently to induce transcription. The EMSA DNA-protein binding studies identified 2 complexes, C1 and C2, that were formed specifically by interactions that required the GG pair of nucleotides at positions -133 and -132. Crosslinking studies identified complexes having 2 ranges of mobilities (50–78 kDa and 29–32 kDa) that may correspond to complexes C1 and C2. Related proteins or isoforms of the same protein may form the various complexes. Similar results were seen with a total of 5 Sertoli cell and GC-1 spg nuclear extract preparations (data not shown), suggesting that the multiple complexes are not the result of protein degradation. A search of the TFSEARCH and TESS databases using the -144 to -119 region of the CREB promoter did not reveal any consensus transcription factor binding sites that included positions -133 and -132. This finding raises the possibility that the proteins binding to this region of the promoter have yet to be identified. Studies are ongoing to determine the exact DNA binding motifs responsible for protein binding, and efforts are being made to clone the potentially important transcription factors.
Specific regulators of CRE-mediated transcription appear to be produced for the various stages of spermatogenesis. We demonstrated that in early spermatocytes the activator form of CREB is expressed, whereas previous studies have shown that in later stage spermatocytes only smaller I-CREB repressor forms of CREB are present [2]. After the completion of meiosis, the round spermatids predominately express activator forms of the related CREM transcription factor [17, 39], followed by the production of a repressor CREM isoform, CREMC-G, in elongating spermatids before transcription ceases [18]. These data suggest that CRE-mediated transcription is temporally controlled by a series of related proteins in successive stages of germ cell development.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by NIH grant R29-HD34913 (to W.H.W.). Preliminary results of this study were presented at the 81st Annual Meeting of the Endocrine Society, 12–15 June 1999, San Diego, CA. 
2 Correspondence: William H. Walker, Department of Cell Biology and Physiology, University of Pittsburgh, 820 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261. FAX: 412 648 8330; walkerw+{at}pitt.edu
Accepted: October 12, 2001.
Received: April 19, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B induces cAMP-response element-binding protein gene transcription in Sertoli cells. J Biol Chem 1999;; 274:35607-35613B in mammalian testis. Mol Endocrinol 1998; 12:1696-1707B: A family of inducible and differentially expressed enhancer-binding proteins in human T cells. Proc Natl Acad Sci U S A 1990; 87:10028-10032
isoforms expressed by two newly identified cAMP-responsive promoters active in the testis. Endocrinology 2000; 141:3923-3930This article has been cited by other articles:
|
|
 |
|
  K. Thomas, D.-Y. Sung, J. Yang, K. Johnson, W. Thompson, C. Millette, J. McCarrey, A. Breitberg, R. Gibbs, and W. Walker Identification, Characterization, and Functional Analysis of Sp1 Transcript Variants Expressed in Germ Cells During Mouse Spermatogenesis Biol Reprod, April 1, 2005; 72(4): 898 - 907. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Matlhagela, M. Borsick, T. Rajkhowa, and M. Taub Identification of a Prostaglandin-responsive Element in the Na,K-ATPase {beta}1 Promoter That Is Regulated by cAMP and Ca2+: EVIDENCE FOR AN INTERACTIVE ROLE OF cAMP REGULATORY ELEMENT-BINDING PROTEIN AND Sp1 J. Biol. Chem., January 7, 2005; 280(1): 334 - 346. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bourguiba, S. Chater, C. Delalande, M. Benahmed, and S. Carreau Regulation of Aromatase Gene Expression in Purified Germ Cells of Adult Male Rats: Effects of Transforming Growth Factor {beta}, Tumor Necrosis Factor {alpha}, and Cyclic Adenosine 3',5'-Monosphosphate Biol Reprod, August 1, 2003; 69(2): 592 - 601. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
