Molecular Human Reproduction, Vol. 8, No. 11, 965-976,
November 2002
© 2002 European Society of Human Reproduction and Embryology
Testis and spermatogenesis |
Novel leader exons of the cyclic adenosine 3',5'-monophosphate response element modulator (CREM) gene, transcribed from promoters P3 and P4, are highly testis-specific in primates*
1 IHF Institute for Hormone and Fertility Research, University of Hamburg, D-22529 Hamburg and 2 Institute of Reproductive Medicine of the University of Münster, D-48149 Münster, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Testicular expression of CREM is essential for spermatogenesis in the mouse. From a monkey testis cDNA library we isolated a CREM transcript isoform with a novel 5' exon
2 which provides at its 3'-end an in-frame ATG to the downstream reading frame. 5'-RACE on human testis cDNA indicated that exon 2 is 
113 bp in size. Moreover, a second novel leader exon, 1, of 289 bp was identified and encodes a putative open reading frame of 26 amino acids. In-vitro translation and cellular expression of CREM-1 and CREM-2 splice variants cloned from human testis yielded not only full length proteins but also shorter repressor products resulting from downstream translation initiation. Upon co-transfection, products of CREM-2 cDNA repressed protein kinase A-induced activation of a CRE-driven reporter construct. RT–PCR analysis of primate tissues for CREM-2 transcripts showed abundant expression in the testis and very low levels or absence from all other tissues tested. CREM-1 mRNA was exclusively expressed in the testis. Promoters P3 and P4, flanking exons 1 and 2, were cloned and found to be non-responsive to protein kinase A in transfection assays. Furthermore, we show differential activation of P1, P3 and P4 during mouse postnatal testicular development, suggesting cell- and stage-specific regulatory mechanisms for these CREM promoters. alternative promoter/CREM/isoform/primate spermatogenesis/testis
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Spermiogenesis encompasses the post-meiotic cellular differentiation of male germ cells from spherical immotile cells to highly motile sperm (Russell et al., 1990
). This includes complex reorganization of the cytoskeleton and chromatin as well as the establishment of the acrosome, a derivative of the Golgi apparatus, containing proteolytic enzymes. Products of the CREM gene are considered to be key factors for rodent spermiogenesis. Deletion of the CREM gene leads to infertility in the male mouse (Blendy et al., 1996; Nantel et al., 1996). Histological analysis of CREM-deficient mice has revealed a highly specific arrest at early post-meiotic germ cell differentiation. This observation matches with the onset of highest CREM activator protein expression in the rodent testis (Delmas et al., 1993; Behr and Weinbauer, 1999). Moreover, several CREM target genes with high expression in spermatids have been identified (Behr and Weinbauer, 2001). When related to spermatid development, the CREM protein expression pattern is very similar between the rodent and the primate testis (Behr and Weinbauer, 1999). CREM expression attains peak levels in those stages associated with the formation of the acrosome (Russell et al., 1990), and vanishes at the initiation of nuclear elongation. CREM expression has also been investigated in patients with spermatid maturation arrest by immunohistochemistry and in-situ hybridization. The findings revealed reduced or absent CREM expression in patients with blocked spermatid maturation (Weinbauer et al., 1998; Steger et al., 1999), and suggest that CREM also functions as a key regulator of spermiogenesis in the primate testis. However, primate CREM gene regulation and function has been much less intensively investigated than in the rodent. In order to gain better insight into the expression and role of CREM in the primate testis, we constructed a cynomolgus monkey testis cDNA library and screened for novel CREM transcripts.
The CREM gene gives rise to a plethora of products due to mechanisms of alternative promoter usage, alternative splicing and alternative translation initiation (de Groot and Sassone-Corsi, 1993; Lamas et al., 1996; Walker and Habener, 1996; Sanborn, 2000; Gellersen et al., 1997; Behr and Weinbauer, 2000). When transcription is initiated at promoter 1 (P1), the translational start codon is located in exon B which can be spliced to various combinations of downstream exons including exon 
(resulting in an early stop codon), C (the first Q-rich transactivation domain, 
1), E plus F (the kinase-inducible domain, KID), G (the second transactivation domain, 2), the small 
-exon, H (the basic region, BR), and the entire exon I (Ia + Ib) (resulting in translation of the first DNA-binding domain, DBD I, followed by a stop codon), or the 3'-portion of exon I (Ib) (resulting in translation of DBD II). A cAMP-inducible intronic downstream promoter, P2, gives rise to the transcriptional repressor ICER (inducible cAMP early repressor) lacking phosphorylation and transactivation domains (Masquilier et al., 1993; Molina et al., 1993; Fujimoto et al., 1994; Bodor et al., 1996; Gellersen et al., 1997; Müller et al., 1998). Recently, additional exons 1 and 2, transcribed from novel promoters P3 and P4 with activity in the testis, have been identified in the rat CREM gene (Daniel et al., 2000).
Here we report the isolation of CREM exons 1 and 2 from the primate testis and demonstrate highly testis-specific activity of promoters P3 and P4 in primates. Characterization of the CREM-1 and CREM-2 transcripts and their protein products suggests a specific role during the stage-specific establishment of spermatogenesis.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Screening of a monkey testis cDNA library
Construction and screening of a cynomolgus monkey (Macaca fascicularis) testis cDNA library was performed as described earlier (Behr et al., 2000
). The oligo-dT primed cDNA synthesis kit (Stratagene, Heidelberg, Germany) was used, employing the Uni-ZAP XR unidirectional phagemid vector. Approximately 1x106 recombinant phage clones were screened for the presence of CREM transcripts using a [32P]dCTP-labelled human CREM cDNA fragment from the 3'-untranslated region (positions 1214–1522; Genbank HUMCREM1A) (Fujimoto et al., 1994). Plating of recombinant phages and generation of replica filters were performed according to the manufacturer’s protocol. The filters were prehybridized at 60°C for 2–4 h. For hybridization, labelled cDNA was added to a final concentration of 1x106 c.p.m./ml of hybridization solution. Hybridization was performed at 60°C for 16 h. Autoradiographic exposure of the filters was carried out at –80°C for 2 days. Positive clones were purified by plaque isolation and a subsequent secondary screening procedure. Inserts were excised according to the manufacturer’s protocol. The resulting phagemid contained the cDNA inserted at the EcoRI site of pBluescript SK(–). Sequencing of the cDNA was carried out on both strands using the LICOR sequencer (MWG Biotech, Ebersberg, Germany).
PCR primers and probes and RT–PCR analysis
Total RNA was isolated from human testis or a range of cynomolgus monkey tissues by the Ultraspec method (Biotecx, Houston, TX, USA). Oligo(dT)-primed first strand cDNA synthesis was performed with MMLV reverse transcriptase (Promega, Madison, WI, USA) or SuperScript RNase H-reverse transcriptase (Life Technologies, Karlsruhe, Germany), and subsequent PCR was performed with Taq polymerase (Promega). Products were separated on 1.5% agarose gels, denatured in 0.4 mol/l NaOH, blotted onto nylon membranes (Hybond-N; Amersham Pharmacia Biotech, Freiburg, Germany) and hybridized with internal DIG-labelled oligonucleotide probes. Visualization was performed with the DIG detection kit (Roche Molecular Biochemicals, Mannheim, Germany). Sequences of human CREM-specific oligonucleotides used as primers or probes are given in Table I. Primers for amplification of monkey GAPDH cDNA were: 5'-CCAGGGCTGCTTTTAACTCTG-3' (sense); 5'-GCAGGGATGATGTTCTGGAGA-3' (antisense), yielding a product of 571bp.
|
Transcripts from mouse testis were analysed by radioactive RT–PCR. Total testicular RNA from different postnatal time points between days 3 and 70 was isolated using TriReagent LS (Molecular Research Center, Cincinnati, OH, USA) according to the manufacturer’s instructions. RT–PCR analysis was performed essentially as described (Wilson and Melton, 1994
) using random hexamer primers and SuperScript RNase H-reverse transcriptase. Radioactively labelled PCR products were separated on a 6% polyacrylamide gel. The gel was transferred onto Whatman paper and dried. Signals were detected using a PhosphorImager (Molecular Dynamics). PCR was performed in a buffer containing 1.5 mmol/l MgCl2 and the conditions used were 1 cycle at 94°C for 3 min, 22 cycles at 94°C for 45 s, 54°C for 45 s, and 72°C for 1 min, followed by 1 cycle at 72°C for 5 min. After 22 PCR cycles, the reaction was proven to be in the exponential phase of product amplification by serial dilution of cDNA from an adult testis. The exon B-specific forward primer (#3350) and the antisense primer to exon H (#3336) used in this experiment are listed in Table I
; mouse-specific sense primers to exons 1 and 2 were: 5'-GTGGATGTGGTGGCATCAGCA-3' and 5'-GACAGTTCCAGGACAGTGAC-3'.
Cloning of human testicular CREM transcripts
The 5'-ends of CREM transcripts from human testis were isolated using a 5'-RACE kit (Roche Molecular Biochemicals, Mannheim, Germany). CREM-specific first strand cDNA synthesis was primed with #3047 in exon H, 5'-RACE outer and inner primers were located in exon E (#3285 and #3060 respectively). Nested PCR was performed with the 5'-anchor primer (5'-GACCACGCGTATCGATGTCGAC-3') with either Taq or Pfu polymerase (Promega); products were cloned into the pGEM-T (Promega) or pCR-Blunt (Invitrogen) vectors respectively, and sequenced on both strands.
Full length cDNA with exon 2 as the leader exon were amplified by RT–PCR from human testis RNA using primers #3261 (in exon 2) and #3264 (adding a FLAG epitope and a NotI site to the 3'-end of the DBD II) and Taq polymerase. Products were cloned into pGEM-T Easy (Promega); and sequenced. Inserts encompassing exons 2, E, F, H and Ib and either containing or lacking exon G were excised with EcoRI/NotI and subcloned into pcDNA3.1(+) (Invitrogen, Karlsruhe, Germany) to yield eukaryotic expression vectors pcDNA/CREM-2-2-ß and pcDNA/CREM-2-ß respectively.
Full length cDNA with exon 1 as the leader exon were amplified by RT–PCR from human testis using primers #3311 (in exon 1) and #3048 (in the 3'-UTR of exon Ib) and Taq polymerase, cloned into pGEM-T Easy and sequenced. Inserts encompassing exons 1, E, F, H and Ib and either containing or lacking exon G were amplified by PCR using primers #3311 and #3264 (see above) and Pfu polymerase, cut at the 3'-NotI site added by #3264, and inserted into the EcoRV/NotI sites of pcDNA3.1(+) to yield pcDNA/CREM-1-2-ß and pcDNA/CREM-1-ß respectively.
Generation of promoter constructs
A 2.9 kb DNA fragment flanking exon 1 was amplified from human genomic DNA (Roche Molecular Biochemicals) using Pfu polymerase and primers #3320 and #3321 (extending to position –173 relative to the start ATG in exon 1). The product was digested with SstI (position –2891 relative to the start ATG in exon 1) and cloned into the SstI/SmaI sites of the luciferase reporter vector pGL3-Basic (Promega) to yield the P3 promoter construct CREM-P3-2.9/luc3. This construct was digested with Ecl136II at the 5'-SstI site and with Bpu1102I at –389 relative to the start ATG, the released fragment discarded, the overhang filled in with Klenow enzyme, and the ends re-ligated to yield the truncated promoter construct CREM-P3-0.4/luc3. Similarly, a further truncation to yield CREM-P3-0.3/luc3 was generated by Ecl136II/EcoRI collapse on CREM-P3-2.9/luc3 using the EcoRI site at position –290.
Sequence flanking exon 2 was isolated with the Genome Walker Kit (BD Clontech, Heidelberg, Germany). Oligonucleotides #3263 and #3278 (antisense to exon 2) were used as nested gene-specific primers GSP1 and GSP2 on the DNA libraries DL-EcoRV and DL-DraI with Expand polymerase (Roche Molecular Biochemicals). Products were cloned into pCR-XL-TOPO (Invitrogen) and sequenced at the ends. A 3.2 kb insert from one of the DL-EcoRV clones, extending from the genomic EcoRV site at position –3255 relative to the ATG in exon 2 to position –53, was excised with SmaI (in the Genome Walker adapter sequence at the 5'-end) and EcoRV (in the 3'-polylinker of pCR-XL-TOPO) and ligated into the SmaI site of pGL3-Basic to yield P4 promoter construct CREM-P4-3.2/luc3. From one of the DL-DraI clones, a 0.6 kb fragment extending from the genomic XbaI site (position –614) to –53 was isolated by XbaI digestion, followed by polishing, and EcoRV digestion in the 3'-polylinker of pCR-XL-TOPO. The fragment was inserted into the SmaI site of pGL3-Basic to yield CREM-P4-0.6/luc3.
Cell culture and transient transfections
The human uterine sarcoma cell line SKUT-1B (HTB 115, American Type Culture Collection, Rockville, MD, USA) was maintained in DMEM/Ham’s F-12 (1 + 1) with 10% FCS, 50 IU/ml penicillin and 50 µg/ml streptomycin. Transfections were performed by the calcium phosphate precipitation method (Profection; Promega). Cells were plated in 24-well plates (0.7x105 cells/well) and received 1 µg of reporter plasmid and 0.05 µg of each expression vector including the ß-galactosidase expression vector pCMV/LacZ (kindly provided by Dr G.E.DiMattia, London Regional Cancer Centre, Ontario). When promoter constructs of various lengths were compared, equimolar amounts were applied, and the total amount of DNA was kept constant by the addition of irrelevant plasmid DNA. The DNA precipitate was removed from the cells 16 h later and replaced by fresh medium. Cell extracts were harvested after an additional 24 h for chemiluminescent luciferase assay (Promega) and ß-galactosidase assay (Galacto-Light; Tropix, Bedford, MA, USA). Transfections were performed in triplicates on several independent cultures. Expression vectors for the catalytic subunit of PKA, pRSV-Cß, and an inactive mutant thereof, pRSV-Cßmut, were kindly provided by Dr Richard Maurer (Oregon Health Sciences University, Portland, OR, USA) (Maurer, 1989). The cAMP-responsive reporter construct pCRE/-36rPRL/luc3 has been described previously (Gellersen et al., 1997).
Protein analysis
CREM/FLAG constructs in pcDNA3.1 were subjected to in-vitro transcription/translation using the TNT T7 coupled reticulocyte lysate system (Promega) as described previously (Gellersen et al., 1997). Immunoprecipitation with FLAG antiserum (D-8; Santa Cruz Biotechnology, Santa Cruz, CA, USA) and sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) analysis have been described (Gellersen et al., 1997). For Western blot analysis of cellular proteins, COS-7 cells were transfected with CREM/FLAG constructs in pcDNA3.1 and harvested 40 h later directly in SDS sample loading buffer. Proteins were resolved by SDS–PAGE and transferred to polyvinylidene fluoride membrane (Millipore, Eschborn, Germany). Immunodetection was performed by enhanced chemiluminescence (Super-Signal; Pierce, Bonn, Germany) with antiserum against human CREM-1 (1:500; Santa Cruz) or monoclonal FLAG antibody M2 (1:1000; Sigma).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Isolation of novel CREM isoforms from primate testis
Screening of a cDNA library from cynomolgus monkey (Macaca fascicularis) testis with a human CREM cDNA fragment located in the 3'-untranslated region of exon Ib yielded five clones. All clones contained exons E, F, G, H, and Ib, encoding the kinase-inducible domain (KID), the second transactivation domain (
2), the basic region (BR) and the second of two alternative DNA binding domains (DBD II) (see also Figure 1A for modular structure of CREM). The 5'-end of three of the clones consisted of an exon B sequence. These clones therefore represented a transactivator isoform (CREM-2-ß) previously isolated from other species (de Groot and Sassone-Corsi, 1993; Laoide et al., 1993; Walker et al., 1994; Behr et al., 2000). However, two of the clones displayed up to 98 bp of novel sequence which was spliced to exon E. This prompted us to perform 5'-RACE on human testicular cDNA in searching for the corresponding human sequence. Of ten 5'-RACE clones isolated using antisense primers anchored in exon E, two contained either exon A0 or B spliced to exon E. Four clones comprised novel 5' sequence corresponding to that isolated from monkey testis (Figure 1A, exon 2) and four clones contained unrelated novel sequence (Figure 1A, exon 1). During the course of our work, isolation of exons 1 and 2 from rat testis was reported (Daniel et al., 2000). An alignment of human and rat exons 1, and of human, monkey and rat exons 2, is shown in Figure 1B. Exon 1 contributes an open reading frame (ORF) of 26 amino acids when spliced to exon E. While the protein coding regions are highly homologous between human and rat, human exon 1 extends 150 bp further upstream than rat exon 1. The 5'-end of rat exon 1 has been deduced from the longest mouse EST clone (Daniel et al., 2000); alignment with the 5'-flanking rat genomic DNA reveals high homology to the human exon 1 5'-UTR. Human and monkey exons 2 are at least 79 and 64 bp longer than the corresponding rat exon respectively; in all species the 3'-end of exon 2 provides a methionine codon which is in-frame with the downstream exon E.
|
Structural organization of the human CREM gene and analysis of CREM
splice variantsTranscription of the CREM gene can be driven by promoter P1 or the alternative cAMP-inducible intronic promoter P2 which gives rise to the transcriptional repressor ICER. The presence of exons
1 and 2 suggests transcriptional initiation at novel promoters P3 and P4. A complete map of the human CREM gene was assembled from database entries (Figure 2). The gene spans ~85 kb on chromosome 10. The 5'-UTR of transcripts initiated at P1 is encoded on three exons (A0, A, and B). Exon 1 is located 15 kb downstream of exon and is separated by a 7.8 kb intron from exon 2. The intron between exon 2 and exon C is very short both in the human and the rat gene (161 and 145 bp respectively). The precise 5' boundary of exon C is ambiguous; the 14 bp at the 5'-end of exon C (Masquilier et al., 1993) deviate from the working draft sequence of human chromosome 10. In contrast to rodent CREM mRNA, the 1 is seldom found in human CREM transcripts; splicing of exon C appears to be a rare event. This was confirmed by RT–PCR analysis of splice variants in human testis as described below.
|
The cDNA from two individual human testes was subjected to PCR to amplify transcripts spanning from either exon
1 or 2 to the 3'-UTR in exon Ib (Figure 3). Multiple products were obtained which were analysed by Southern hybridization using exon-specific probes. Apparently all products contain exons E, F and H. The largest of the more abundant PCR products amplified with primers anchored in exons 1 or 2 (sense) and the 3'-UTR in exon Ib (antisense) most likely include exons 1 or 2 respectively, E, F, G (± ), H and Ib, and lack C and Ia. They are so abundant that they are visible by ethidium bromide staining. The even larger products detected by hybridization with a probe to exon C are likely to contain exons 1 or 2 respectively, C, E, F, G (± ), H and Ib. They are rare, only become visible after extended exposure of the Southern blot and are not detectable by ethidium bromide staining or upon the shorter exposure times used for the Southern hybridizations with probes to exons E, F, G, or H. The 1.2 kb product in the lower right hand panel of Figure 3 represents a rare splice variant including exon Ia which adds 400 bp and again is only seen after prolonged exposure. The 2 domain (exon G) is more frequently present in exon 1 than in exon 2 transcripts.
|
Tissue-specific expression of CREM
transcriptsWe next investigated the tissue-specificity of CREM transcripts initiated at promoters P3 and P4, using 17 tissues of the male cynomolgus monkey plus uterus and ovary (Figure 4
). PCR primers anchored in exon 1 or 2 were paired with an antisense primer to exon H. Exon 1 was exclusively found in testis. Exon 2 was abundantly expressed in testis, and weakly in prostate, seminal vesicle, and uterus. The transcript isoforms most likely corresponding to the different sizes of the PCR products are given in the legend to Figure 4. For comparison, transcripts including exon B, derived from promoter P1, were amplified and found to be ubiquitously expressed. Interestingly, only testis showed the presence of an activator isoform (2), whereas the band migrating at ~390 bp in most tissues represents a repressor composed of exons B, E + F, and H, and in the lowest band, exon B is spliced directly to H. Promoters P3 and P4 therefore underlie highly tissue-specific control, and promoter P1 gives rise to an activator isoform only in the testis.
|
Analysis of protein products of CREM
transcriptsIn order to analyse in more detail the modular composition of CREM-
1 and CREM-2 mRNA and to generate expression vectors for functional studies, full length cDNA were amplified from human testis with sense primers anchored in the 5'-untranslated regions of exons 1 or 2 respectively, paired with an antisense primer to the 3'-untranslated region incorporating a FLAG epitope (Figure 5A). Four families of clones were obtained and sequenced. All contained the KID, the basic region and DBD II and either lacked or included the 2. The cDNA were inserted into the eukaryotic expression vector pcDNA3.1 and were designated CREM-1-2-ß, CREM-1-ß, CREM-2-2-ß and CREM-2-ß (the suffix ß indicating presence of DBD II, while 
indicates presence of DBD I).
|
Exons
1 and 2 provide an in-frame methionine codon when spliced to exons E + F encoding the KID. We wished to determine whether these putative start methionines were utilized in CREM-1 and CREM-2 transcripts. In addition, we have previously reported that three ATG codons in exons F, G and H can give rise to alternative translation initiation, resulting in N-terminally truncated repressor isoforms (S-CREM and SS-CREM) (Gellersen et al., 1997). Potential translation products of CREM-1 and CREM-2 mRNA and their estimated sizes are shown in Figure 5A. Expression vectors for the indicated cDNA were subjected to in-vitro-transcription/translation in the presence of [35S]methionine, followed by immunoprecipitation with FLAG antiserum, SDS–PAGE and autoradiography. CREM-2- and CREM- cDNA were included for comparison. They have been characterized by us previously and are composed of the leader exon B, exons E + F, H and Ia (encoding the DBD I) (Gellersen et al., 1997). CREM-2- is a transcriptional activator containing the 2 (exon G) while the repressor CREM- lacks such a domain. The autoradiograph in Figure 5B demonstrates that full length proteins (32, 24.5, 29, 22 kDa) were translated from all four test cDNA, indicating that the ATG codons in exons 1 and 2 can be utilized by the translational machinery. In addition, translation is initiated at the ATG in exon F generating S-CREM-2-ß (20 kDa) and S-CREM-ß (13 kDa). These are in fact the predominant products of the CREM-2 cDNA, suggesting that the ATG in exon 2 provides a less favourable initiation context than the ATG in exon 1. The CREM-2- and CREM- control cDNA produced predominantly full length proteins (32.5 and 25 kDa respectively) and additional truncated forms S-CREM-2- (20 kDa), S-CREM- (13 kDa) and SS-CREM-2- (16 kDa) initiated in exons F and G, as reported previously (Gellersen et al., 1997). The shortest form, SS-CREM-/ß, was not observed. In addition, we analysed translation products of these CREM transcripts in transfected cells. COS-7 cells were chosen because they can be transfected with high efficiency. Cell extracts were subjected to Western blot analysis with CREM antibody (Figure 5C). In cells transfected with the empty expression vector pcDNA3.1, a faint signal was obtained with an endogenous CREM protein migrating at ~25 kDa. Translation of the CREM-2 cDNA yielded only the truncated S-CREM products initiated in exon F, confirming poor utilization of the ATG start codon in leader exon 2 seen by in-vitro translation (compare Figure 5B, left panel). As already observed in vitro (Figure 5B, right panel), the start codon in leader exon 1 is much more efficiently used for translation initiation in vivo, giving rise to roughly equal amounts of full length and truncated product. Notably, both repressor transcripts CREM-1-ß and CREM-2-ß are much more highly expressed than their corresponding activator counterparts CREM-1-2-ß and CREM-2-2-ß. The CREM-2- and CREM- control transcripts, products of promoter P1, yielded predominantly full length proteins in COS-7 cells as well as in vitro. Essentially the same results were obtained by immunodecoration of the Western blot with an antibody against the FLAG epitope (data not shown).
Transcriptional effects of the novel CREM isoforms were then tested by transfection analysis in the SKUT-1B cell line (Figure 6). A luciferase reporter construct driven by a CRE (pCRE/-36rPRL/luc3) was activated by co-transfection of an expression vector for the catalytic subunit of protein kinase A, PKA-Cß. Controls received the inactive mutant PKA-Cßmut. Co-transfection of CREM-1 or CREM-2 vectors did not significantly affect basal activity of pCRE/-36rPRL/luc3, but PKA-mediated activation was repressed in the presence of pcDNA/CREM-2-ß. The lack of stimulatory activity of the supposed activator transcripts CREM-1-2-ß and CREM-2-2-ß is likely due to co-translation of the truncated 20 kDa isoform S-CREM-2-ß (see Figure 5B and C) which lacks the PKA phosphorylation site, or to low overall levels of expression (see Figure 5C). The positive control construct pcDNA/CREM-2- enhanced PKA-stimulated promoter activity in accordance with the predominant translation of full length activator protein (see Figure 5C). Interestingy, the supposed repressor transcripts CREM-1-ß and CREM- did not inhibit PKA-mediated transcriptional activation; this coincided with a high level of expression of their respective 24.5 and 25 kDa full length protein products. These lack the 2 but still comprise the KID; the question arises as to whether such molecules, possibly dimerized with an activator, do not counteract PKA-mediated activation.
|
Regulation of promoters P3 and P4
We next wished to investigate transcriptional activity of the novel promoters P3 and P4. P3 promoter constructs were generated by PCR on human genomic DNA. The resultant 2.9 kb fragment was cloned into the luciferase reporter vector pGL3-Basic. 5'-truncated constructs with 0.4 and 0.3 kb of 5'-flanking sequence were also produced. A 3.2 kb fragment flanking P4 was obtained by Genome Walker PCR, and a 0.6 kb deletion construct was generated by enzymatic digest (Figure 7A
). A computerized search for transcription factor binding sites revealed two CRE-like elements both in the P3 and the P4 5'-flanking regions (at positions –791 and –208 relative to the start ATG in exon 1, and at –1860 and –1129 relative to the start ATG in exon 2). The P3 and P4 promoter/reporter constructs were transfected into SKUT-1B cells to test responsiveness to PKA activation (Figure 7B). CREM-P3 constructs had a slightly higher basal activity compared with CREM-P4 constructs. However, while the positive control construct pCRE/-36rPRL/luc3 was induced 17-fold by PKA-Cß, none of the CREM-P3 or CREM-P4 promoter constructs responded to this stimulus.
|
Finally, in the absence of primate testicular tissue from different postnatal stages, developmental activation of promoters P1, P3 and P4 was assessed in testes of mice between 3 and 70 days of age (Figure 8
). RT–PCR with primers specific for exons B, 1 or 2, paired with an antisense primer to exon H, showed low abundance of numerous splice variants up to day 14 for transcripts derived from P1 and P4 (initiated at exons B and 2). The level of transcripts encoding presumed activator isoforms sharply increased at day 16 (appearance of pachytene spermatocytes), continued to increase until day 20 (appearance of haploid spermatids) and remained high thereafter until at least day 70 (full spermatogenesis). In contrast, activator transcript abundance from P3 (initiated at exon 1) was low until day 18 and strongly elevated from day 20 onwards. While two types of activator transcripts were present among the CREM-B and CREM-1 isoforms, containing either one or two transactivation domains, only one type of activator splice form including both transactivation domains was present among the CREM-2 transcripts.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We identified novel CREM exons in the primate testis by screening a cynomolgus monkey testis cDNA library and performing 5'-RACE on cDNA from human testis, and partially characterized the corresponding novel human promoters P3 and P4. Promoters P3 and P4 give rise to CREM-
1 and CREM-2 transcripts, carrying novel leader exons 1 and 2. These promoters, P3 and P4, of the CREM gene and the novel exons, 1 and 2, have recently also been described in the rat testis (Daniel et al., 2000). However, human exons 1 and 2 were found to be significantly longer than their rat counterparts. The sizes of rat exons 1 and 2 have been predicted from the longest mouse EST clones (Daniel et al., 2000). We found human exon 1 to extend at least 150 bp further upstream than rat exon 1. The remarkably high homology between the 5'-UTR of human exon 1 and the corresponding rat genomic DNA suggests that the transcribed region of the rat gene may also extend further upstream. Human and monkey exons 2 are at least 79 and 64 bp longer than the corresponding rat exon. A search of the human EST database with exon 1 returned five clones, all from human testis. Three of those extend to position –265, and two extend to position –235 relative to the 3'-end of exon 1, and were thus somewhat shorter than our longest 5'-RACE clone extending to –289. In all EST clones, exon 1 was followed by exons E + F and G. A search with exon 2 retrieved three EST clones, again all from human testis. They extend to positions –108 and –113 and therefore suggest transcription start sites similar to our 5'-RACE clones. In two of the EST clones, exon 2 was spliced to exons E + F and G; in one clone, exon 2 was followed by exon G which disrupts the ORF.
As in rodents, the primate exon 1 also adds a start codon and an ORF of 26 amino acids, while exon 2 provides only an in-frame start codon to the downstream sequences. There appear to be species-specific splicing differences; whereas rat and mouse exons 2 are preferentially spliced to exon C (encoding 1) and give rise to the CREM- isoforms including both transactivation domains, human exon 2 is predominantly spliced to exon E to produce the 2 activator isoform or repressors. The presence of exon C in human CREM-1 or CREM-2 transcripts is rarely detectable as shown by Southern blot analysis of RT–PCR products. The first 14 bp of exon C reported with the cloning of the human CREM cDNA (Masquilier et al., 1993) deviate from the human genome working draft sequence (accession no. NT_0088583.4); this region has been suggested to be polymorphic and result from the insertion of a partial Alu element (Daniel et al., 2000).
The DBD II is the predominant DNA binding region included in human testis CREM transcripts; a similar preferential splicing of exon Ib over exon Ia has also been observed in the rat testis (Daniel et al., 2000). In contrast, we have previously cloned CREM isoforms from human endometrial stromal cells and found a prevalence of DBD I (Gellersen et al., 1997). The significance of differential inclusion of the alternative DBD is not known.
The fact that searches of mouse and human EST databases for exons 1 and 2 retrieved exclusively testicular transcripts (Daniel et al., 2000; and this report) suggested a high tissue-specificity of P3 and P4 promoter activity. This was confirmed by us using a panel of 19 different monkey tissues. CREM-1 expression was entirely restricted to the testis; CREM-2 was abundantly expressed in the testis and very faintly in prostate and seminal vesicle. Interestingly, in addition to these male reproductive tissues, monkey uterus displayed weak expression of CREM-2 mRNA. When we examined tissues of the human utero-placental unit, however, we could not detect the presence of exons 1 or 2 in CREM transcripts (data not shown).
Promoter P1 of the CREM gene is believed to be constitutively active, and our screen of a panel of monkey tissues confirmed ubiquitous expression of P1 products. However, we found a highly testis-specific splicing pattern generating an activator isoform exclusively in this tissue. The P2 promoter giving rise to the repressor ICER is highly responsive to cAMP signalling (Molina et al., 1993; Walker and Habener, 1996). Daniel et al. also demonstrated very strong PKA inducibility of rat P3 reporter constructs transfected into human JEG-3 choriocarcinoma cells, while P4 was much more weakly induced by PKA (Daniel et al., 2000). We constructed reporter constructs of the human P3 and P4 promoters extending ~3 kb upstream of the presumed transcriptional start sites. When transfected into JEG-3 cells, primary cultures of human myometrial cells (data not shown) or the myometrial cell line SKUT-1B (Figure 7), P3 and P4 displayed very low basal activities. This lack of activity in a heterologous cell system may reflect the stringent tissue-specific control of P3 and P4, possibly achieved by the requirement for testis-specific transcription factors and/or co-factors. In contrast to the rat promoters, human P3 and P4 were not induced by co-transfected PKA catalytic subunit in spite of two putative CRE in each promoter. This lack of response is particularly significant because SKUT-1B cells very efficiently activated a CRE-driven control construct. Whether the difference in PKA responsiveness between the rat and human promoters is due to different cellular backgrounds in the transfection systems employed or represents a relevant species-specific difference in promoter control remains open at present.
In the monkey and human testis we previously localized CREM mRNA in late pachytene spermatocytes and early round spermatids (Steger et al., 1999; Behr et al., 2000). However, the probe used for in-situ hybridization did not allow differentiation between transcripts generated from the alternative promoters P1, P3 and P4. It is conceivable that the different CREM promoters are necessary for the promotion of different stages of spermatogenesis. The distribution of CREM clones obtained by 5'-RACE PCR and the RT–PCR analysis of mouse postnatal development suggest that all three CREM promoters exhibit substantial activity in the testis. We have ruled, using RT–PCR, that exon B is spliced to exons 1 or 2 or that exon 1 is spliced to exon 2 (data not shown). Daniel et al. have used semi-quantitative RT–PCR to analyse stage-specific CREM transcript abundance in the rat testis (Daniel et al., 2000). These authors found distinct expression profiles for the exons B, 1 and 2. Although the data showed slight differences in the abundance of exons B and 2, the transcripts containing these exons seem to be consistently present throughout the entire spermatogenic cycle in the rat. In contrast, transcripts generated from P3 including exon 1 were barely detectable during stages IX–XII.
In order to follow the developmental activation of promoters P1, P3 and P4, we analysed 18 different time points of postnatal testicular development in the mouse by semi-quantitative RT–PCR. Transcripts generated from all three promoters were already detectable in the early postnatal testis. Whereas from P4 almost exclusively activator transcripts were generated, activator as well as repressor transcripts were produced from P1 and P3. The transcript abundance from all promoters increased between postnatal days 16 and 20. However, our data suggest that activation of promoters P1 and P4 during testicular development precedes that of P3 by several days. Previously a developmental switch from repressor to activator CREM transcript expression from promoter P1 around day 14 of postnatal mouse testicular development has been reported (Foulkes et al., 1992). However, our data shown in Figure 8 do not completely confirm this switch model. In the mouse testes from day 3 to 14, we detected several repressor as well as activator isoforms transcribed from P1. The signal intensity of the bands representing the different transcripts was rather uniform, indicating similar levels of transcript abundance. From day 16 onwards we observed an increase in the abundance of the activator transcripts. However, at least one repressor was still detectable in the adult mouse testis. Furthermore, activator transcripts were generated from P3 and P4 already in the prepubertal testis. In summary, our data suggest a gradual change of CREM expression rather than an abrupt switch as reported elsewhere (Foulkes et al., 1992).
The biological relevance of the alternative promoters and their specific transcripts for mammalian spermatogenesis is an unresolved issue as yet, since the inactivation of the CREM gene in mice ablated the transcripts from all alternatively used promoters P1, P3 and P4 (Blendy et al., 1996; Nantel et al., 1996). The most conclusive approach would be the selective inactivation of each single promoter in the mouse to answer the question whether the concerted action of all three promoters is necessary for normal spermatogenesis or whether there is a redundancy of CREM promoter function. A related question arising in the context of alternative promoter utilization is that of putative functional relevance of the specific N-terminal regions provided to the CREM protein isoforms by each individual promoter. So far, no specific function has been assigned to the sequence encoded by exon B which adds 40 amino acids N-terminal to the 1 or the KID. In this report, we demonstrated that the in-frame methionine codons in exons 1 and 2 allow initiation of translation, with the ATG in exon 1 being more effectively used than that in 2. Exon 1 adds 26 N-terminal residues while exon 2 only provides an in-frame start codon to the downstream exons. It may be speculated that B- or 1-specific N-termini provide additional epitopes for protein–protein interactions. It is well established that activation of CREM results from phosphorylation at Ser117 in the KID and interaction with the co-activator CREB binding protein (CBP). Interestingly, activation of CREM in germ cells can occur without phosphorylation, and the testicular factor ACT (activator of CREM in testis) has been shown to activate CREM by interaction with the KID in the absence of phosphorylation at Ser117 (Fimia et al., 1999).
Our data indicate, however, that the presence of exons B, 1 or 2 in CREM transcripts does not necessarily imply presence of the corresponding amino acid sequences in the translated products. We demonstrated here and in a previous report that alternative downstream translation initiation gives rise to truncated repressor S-CREM isoforms (Gellersen et al., 1997). These are preferentially initiated at an in-frame ATG in exon F of the KID, while a minority of translation products, SS-CREM, is generated from a start codon in exon G (2). As a consequence, presumed activator transcripts encoding the KID and one or two Q-rich regions can give rise to a substantial proportion of repressor peptides irrespective of the promoter-specific N-terminal region. When we expressed CREM-1-2-ß and CREM-2-2-ß transcripts in the SKUT-1B cell line, they failed to enhance PKA-mediated activation of a CRE-driven promoter, in contrast with the CREM-2- transcript. This is most likely due to co-translation of truncated CREM isoforms in the transfected cells as also observed in in-vitro-translation experiments and confirmed by Western blot analysis. An important issue to resolve is a putative tissue-specific regulation of translational initiation, which may potentially lead to predominant generation of full length activator isoforms in the testis. However, only the generation of exon-specific antibodies, particularly to B, 1 and E, would allow unambiguous assignment of testicular CREM proteins detected by Western blotting or immunohistochemistry and permit conclusions as to the biological role of individual isoforms.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Beate Hartung and Tanja Schneider-Merck for excellent technical assistance, and Prof. Dr F.Leidenberger and Prof. Dr E.Nieschlag for continuous support. We are grateful to Drs R.Maurer and G.E.DiMattia for expression vectors. This work was supported by the Deutsche Forschungsgemeinschaft (grant Ni-130/15-2).
|
Notes |
|---|
3 Present address: Covance Laboratories GmbH, D-48163 Münster, Germany
4 Present address: Institute of Anatomy, Developmental Biology, University of Essen, D-45122 Essen, Germany
5 To whom correspondence should be addressed at: Institute of Anatomy, Developmental Biology, University of Essen, Hufelandstraße 55, D-45122 Essen, Germany. E-mail: ruediger.behr{at}uni-essen.de
* This work was presented in part at the 83nd Annual Meeting of the Endocrine Society, June 2001, in Denver, CO, USA.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Behr, R. and Weinbauer, G.F. (1999) Germ cell-specific cyclic adenosine 3',5'-monophosphate response element modulator expression in rodent and primate testis is maintained despite gonadotropin deficiency. Endocrinology, 140, 2746–2754.
Behr, R. and Weinbauer, G.F. (2000) CREM activator and repressor isoforms in human testis: sequence variations and inaccurate splicing during impaired spermatogenesis. Mol. Hum. Reprod., 6, 967–972.
Behr, R. and Weinbauer, G.F. (2001) cAMP response element modulator (CREM): an essential factor for spermatogenesis in primates? Int. J. Androl., 24, 126–135.[Web of Science][Medline]
Behr, R., Hunt, N., Ivell, R., Wessels, J. and Weinbauer, G.F. (2000) Cloning and expression analysis of testis-specific cyclic 3', 5'-adenosine monophosphate-responsive element modulator activators in the nonhuman primate (Macaca fascicularis): comparison with other primate and rodent species. Biol. Reprod., 62, 1344–1351.
Blendy, J.A., Kaestner, K.H., Weinbauer, G.F., Nieschlag, E. and Schütz, G. (1996) Severe impairment of spermatogenesis in mice lacking the CREM gene. Nature, 380, 162–165.[Medline]
Bodor, J., Spetz, A.L., Strominger, J.L. and Habener, J.F. (1996) cAMP inducibility of transcriptional repressor ICER in developing and mature human T lymphocytes. Proc. Natl. Acad. Sci. USA, 93, 3536–3541.
Daniel, P.B., Rohrbach, L. and Habener, J.F. (2000) Novel cyclic adenosine 3',5'-monophosphate (cAMP) response element modulator isoforms expressed by two newly identified cAMP-responsive promoters active in the testis. Endocrinology, 141, 3923–3930.
de Groot, R.P. and Sassone-Corsi, P. (1993) Hormonal control of gene expression: multiplicity and versatility of cyclic adenosine 3',5'-monophosphate-responsive nuclear regulators. Mol. Endocrinol., 7, 145–153.
Delmas, V., van der Hoorn, F., Mellstrom, B., Jegou, B. and Sassone-Corsi, P. (1993) Induction of CREM activator proteins in spermatids: down-stream targets and implications for haploid germ cell differentiation. Mol. Endocrinol., 7, 1502–1514.
Fimia, G.M., De Cesare, D. and Sassone-Corsi, P. (1999) CBP-independent activation of CREM and CREB by the LIM-only protein ACT. Nature, 398, 165–169.[Medline]
Foulkes, N.S., Mellström, B., Benusiglio, E. and Sassone-Corsi, P. (1992) Developmental switch of CREM function during spermatogenesis: from antagonist to activator. Nature, 355, 80–84.[Medline]
Fujimoto, T., Fujisawa, J. and Yoshida, M. (1994) Novel isoforms of human cyclic AMP-responsive element modulator (hCREM) mRNA. J. Biochem., 115, 298–303.
Gellersen, B., Kempf, R. and Telgmann, R. (1997) Human endometrial stromal cells express novel isoforms of the transcriptional modulator CREM and up-regulate ICER in the course of decidualization. Mol. Endocrinol., 11, 97–113.
Lamas, M., Monaco, L., Zazopoulos, E., Lalli, E., Tamai, K., Penna, L., Mazzucchelli, C., Nantel, F., Foulkes, N.S. and Sassone-Corsi, P. (1996) CREM: a master-switch in the transcriptional response to cAMP. Phil. Trans. R. Soc. Lond. B, 351, 561–567.[Web of Science][Medline]
Laoide, B.M., Foulkes, N.S., Schlotter, F. and Sassone-Corsi, P. (1993) The functional versatility of CREM is determined by its modular structure. EMBO J., 12, 1179–1191.[Web of Science][Medline]
Masquilier, D., Foulkes, N.S., Mattei, M.-G. and Sassone-Corsi, P. (1993) Human CREM gene: evolutionary conservation, chromosomal localization, and inducibility of the transcript. Cell Growth Differ., 4, 931–937.[Abstract]
Maurer, R.A. (1989) Both isoforms of the cAMP-dependent protein kinase catalytic subunit can activate transcription of the prolactin gene. J. Biol. Chem., 264, 6870–6873.
Molina, C.A., Foulkes, N.S., Lalli, E. and Sassone-Corsi, P. (1993) Inducibility and negative autoregulation of CREM: an alternative promoter directs the expression of ICER, an early response repressor. Cell, 75, 875–886.[Web of Science][Medline]
Müller, F.U., Bokník, P., Knapp, J., Neumann, J., Vahlensieck, U., Oetjen, E., Scheld, H.H. and Schmitz, W. (1998) Identification and expression of a novel isoform of cAMP response element modulator in the human heart. FASEB J., 12, 1191–1199.
Nantel, F., Monaco, L., Foulkes, N.S., Masquilier, D., LeMeur, M., Henriksén, K., Dierich, A., Parvinen, M. and Sassone-Corsi, P. (1996) Spermiogenesis deficiency and germ-cell apoptosis in CREM-mutant mice. Nature, 380, 159–162.[Medline]
Russell, L.D., Ettlin, R.A., Sinha-Hikim, A.P. and Clegg, E.D. (1990) Histological and Histopathological Evaluation of the Testis. Cache River Press, Clearwater, CRP, Vienna, IL, USA.
Sanborn, B.M. (2000) Increasing the options | new 3',5' cyclic adenosine monophosphate (cAMP)-responsive promoters and new exons in the cAMP response element modulator gene. Endocrinology, 141, 3921–3922.
Steger, K., Klonisch, T., Gavenis, K., Behr, R., Schaller, V., Drabent, B., Doenecke, D., Nieschlag, E., Bergmann, M. and Weinbauer, G.F. (1999) Round spermatids show normal testis-specific H1t but reduced cAMP-responsive element modulator and transition protein 1 expression in men with round-spermatid maturation arrest. J. Androl., 20, 747–754.
Walker, W.H. and Habener, J.F. (1996) Role of transcription factors CREB and CREM in cAMP-regulated transcription during spermatogenesis. Trends Endocrinol. Metab., 7, 133–138.[Medline]
Walker, W.H., Sanborn, B.M. and Habener, J.F. (1994) An isoform of transcription factor CREM expressed during spermatogenesis lacks the phosphorylation domain and represses cAMP-induced transcription. Proc. Natl. Acad. Sci. USA, 91, 12423–12427.
Weinbauer, G.F., Behr, R., Bergmann, M. and Nieschlag, E. (1998) Testicular cAMP responsive element modulator (CREM) protein is expressed in round spermatids but is absent or reduced in men with round spermatid maturation arrest. Mol. Hum. Reprod., 4, 9–15.
Wilson, P.A. and Melton, D.A. (1994) Mesodermal patterning by an inducer gradient depends on secondary cell–cell communication. Curr. Biol., 4, 676–686.[Web of Science][Medline]
Submitted on May 10, 2002; accepted on August 12, 2002.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. L.-A. Nguyen, S. de Walque, E. Veithen, A. Dekoninck, V. Martinelli, Y. de Launoit, A. Burny, R. Harrod, and C. Van Lint Transcriptional Regulation of the Bovine Leukemia Virus Promoter by the Cyclic AMP-response Element Modulator {tau} Isoform J. Biol. Chem., July 20, 2007; 282(29): 20854 - 20867. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Jaspers, B. Gellersen, R. Kempf, A. Samalecos, M. Bergmann, and K. Steger Functional Characterization of Male Germ Cell-Specific CREM Isoforms J Androl, January 1, 2007; 28(1): 59 - 66. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Osada, M. Hirata, R. Tanuma, J. Kusuda, M. Hida, Y. Suzuki, S. Sugano, T. Gojobori, C.-K. J. Shen, C.-I Wu, et al. Substitution Rate and Structural Divergence of 5'UTR Evolution: Comparative Analysis Between Human and Cynomolgus Monkey cDNAs Mol. Biol. Evol., October 1, 2005; 22(10): 1976 - 1982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Vouk, P. Hudler, L. Strmsnik, M. Fink, G. Majdic, B. Zorn, E. Lalli, P. Sassone-Corsi, N. Debeljak, R. Komel, et al. Combinations of genetic changes in the human cAMP-responsive element modulator gene: a clue towards understanding some forms of male infertility? Mol. Hum. Reprod., August 1, 2005; 11(8): 567 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Steger, R. Behr, I. Kleiner, G. F. Weinbauer, and M. Bergmann Expression of activator of CREM in the testis (ACT) during normal and impaired spermatogenesis: correlation with CREM expression Mol. Hum. Reprod., February 1, 2004; 10(2): 129 - 135. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||