| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Center for Research on Reproduction and Women's Health, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104-6080
ABSTRACT
The mRNA that encodes the testis-specific protein phosphoglycerate kinase (PGK2) is a long-lived mRNA that is transcribed in meiotic and postmeiotic male germ cells. Pgk2 mRNA is present in germ cells for up to 2 wk before its protein product is detected. Using affinity chromatography with the 3'-UTR of the Pgk2 mRNA, several proteins, including the RNA-binding protein, polypyrimidine tract binding protein 2 (PTBP2), were identified in mouse testis extracts. Coimmunoprecipitation experiments confirmed that PTBP2 binds to Pgk2 mRNA in the testis and RNA gel shifts demonstrated that PTBP2, but not PTBP1, binds to a specific region of the Pgk2 3'-UTR. Recombinant PTBP2 increased the stability of reporter constructs that contained the 3'-UTR Pgk2 sequence element in both testis extracts and transfected HeLa cells. We propose that PTBP2 is a trans-acting factor that helps to stabilize Pgk2 mRNA in male mouse germ cells.
gametogenesis,, gene regulation, posttranscriptional control, RNA-binding proteins, spermatogenesis
Posttranscriptional processing events that regulate mRNA stability, transport, and translation play prominent roles in controlling the spatial and temporal patterns of protein synthesis in many developmental systems, including embryogenesis, neurogenesis, and spermatogenesis [1]. In the testis, posttranscriptional mechanisms play prominent roles in controlling the timing of protein synthesis, owing to the cessation of transcription during spermiogenesis [2, 3]. Many proteins that are essential for spermatozoan assembly and function are synthesized in late-stage male germ cells and are dependent upon the translation of stored mRNAs [3, 4]. To date, most of the germ cell mRNAs that are known to be translationally regulated are initially transcribed in postmeiotic cells and stored as translationally inactive ribonucleoprotein particles (RNP) in the cytoplasm of haploid round spermatids before their translation in the transcriptionally inactive later stage spermatids. Recently, 35 translationally delayed mRNAs that are initially transcribed during meiosis have been identified using a microarray approach [5]. The gene for the glycolytic enzyme phosphoglycerate kinase (PGK) 2, the protein product of which is synthesized only in male germ cells, is a member of this group of genes.
PGK is a highly conserved and widely expressed enzyme that catalyzes the reversible conversion of 1,3-diphosphoglycerate to 3-phosphoglycerate, generating ATP in the process [6]. PGK has also been reported to function as a disulfide reductase in tumor expansion and metastasis [7]. The mammalian genome contains two functional Pgk genes: X-linked Pgk1, which encodes a ubiquitously expressed PGK1, and a second autosomal gene that encodes PGK2, which is expressed only in the testis [8]. In the testis, as germ cells differentiate, PGK1 is expressed before X-chromosome inactivation occurs in meiosis. PGK2 starts to replace PGK1 during meiosis and is the sole phosphoglycerate kinase in postmeiotic germ cells and spermatozoa [9–14]. This changeover of PGK isozymes is initiated at the transcriptional level in meiotic spermatocytes, with the transcriptional regulation of Pgk2 being regulated both positively and negatively by cis-acting DNA elements [15–17].
In mice, Pgk2 mRNAs are initially detected at low levels in the early stages of meiotic spermatocytes, and the mRNA levels increase dramatically in haploid spermatids [8, 9, 12]. PGK2 protein is first detected many days later in the late-stage (step 12) spermatids [18]. Consistent with this discontinuity between mRNA and protein synthesis, Pgk2 mRNAs are sequestered as RNPs in meiotic spermatocytes and move into polysomes in later-stage postmeiotic cells [5]. This temporal separation for more than a week of Pgk2 transcription and translation requires long-term Pgk2 mRNA stabilization in male germ cells. In the present study, we show that polypyrimidine tract-binding protein 2 (PTBP2), which is a member of the polypyrimidine tract-binding protein (PTBP) family that is known to stabilize mRNAs in somatic cells [19–22], regulates the stability of Pgk2 mRNA by binding to a regulatory element in its 3'-UTR.
Preparation of Testis Extracts
The present investigations were conducted in accordance with the guide for Care and Use of Laboratory Animals (1966), and the Institutional Animal Care and Use Committee approved in advance all procedures involving animals. Testis extracts were prepared from sexually mature (60–80-day-old) CD-1 mice (Charles River Breeding Laboratories, Wilmington, MA) using a modified procedure from Han et al. [23, 24]. Briefly, testes were decapsulated, washed, and resuspended in buffer A (10 mM Hepes [pH 7.9], 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 1x protease inhibitor cocktail) (Roche, Indianapolis, IN) and homogenized in a Teflon glass homogenizer on ice until most of the cells were lysed. The homogenates were filtered through a 40-µm cell strainer (BD Falcon, San Jose, CA), to remove debris and tissue clumps, homogenized again, and centrifuged for 15 min at 5000 rpm in a Sorvall SS-34 rotor. For cytoplasmic preparations, extracts were centrifuged for 1.5 h at 34 000 rpm in a Beckman SW 50.1 rotor and the supernatants were stored at –80°C. For nuclear extracts, pellets from the centrifugation at 5000 rpm were washed with buffer A, centrifuged at 3000 rpm, resuspended in buffer B (20 mM Hepes, [pH 7.9] 25% glycerol, 0.5 M NaCl, 1.5 mM MgCl2, 0.1 mM EDTA, 0.5 mM DTT, 1x protease inhibitor cocktail), and homogenized using Tissue Grinders (Kontes, Vineland, NJ). The homogenates were centrifuged at 20 000 rpm at 4°C for 30 min. Supernatants were dialyzed against buffer C (20 mM Hepes, [pH 7.9], 20% glycerol, 100 mM KCl) and stored at –80 °C until use. Total testis extracts in RIPA buffer (10 mM sodium phosphate [pH 7.2], 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 150 mM NaCl) were prepared from adult CD-1 mice (Charles River Laboratories, Wilmington, MA), as previously described [25].
The 3'-UTR of Pgk2 and subclones thereof (F1, F2, F3, F1–1, F1–2, F1–3 and F12, F23) were subcloned into the pGEM-3Z vector using the EcoRI and HindIII sites, which facilitated T7-directed in vitro transcription. The complete open reading frame (ORF) of Ptbp2 was amplified by RT-PCR from testis RNA and cloned into a T-vector. Recombinant mouse full-length His-tagged PTBP2 was expressed from pET-42a (Novagen, San Diego, CA), which was grown in Escherichia coli BL21-CodonPlus (DE3)-RIL cells (Stratagene, La Jolla, CA). The PCR product of the Ptbp2 cDNA was digested with HindIII and SalI and cloned into the HindIII/SalI site of pEGFP-C2 (Clontech, Palo Alto, CA), to generate the expression plasmid pEGFP-PTBP2. A truncated form of PTBP2 that lacked 55 amino acids of the N-terminus was inserted into pEGFP-C2 using the same sites. The pEGFP-C2 constructs, in which the 3'-UTR of the vector was replaced with the 3'-UTR of Pgk2 or its F3 region using the XbaI and MluI restriction sites, were also used in reporter assays.
To synthesize sense 5'-capped and 3'-polyadenylated mRNAs for in vitro decay assays, plasmids that contained the F1 or F2 sequence of the 3'-UTR of Pgk2 with a poly(A)30 tail (pSP64-F1-polyA and pSP64-F2-polyA) were generated using the HindIII and SacI sites. After cutting with EcoRI and purification by phenol extraction, linearized DNA (0.5 µg) was transcribed in vitro using the mMESSAGE mMACHINE SP6 Kit (Ambion, Austin, TX) and the RNAs were purified with a Megaclear Kit (Ambion).
UV Cross-Linking and Gel Shift Assays
UV cross-linking was carried out in vitro using adult testis cytoplasmic extracts (30 µg) in a 25-µl reaction volume that contained 30 mM Tris-HCl (pH 7.6), 2 mM MgCl2, 60 mM KCl, 5 mg/ml heparin, and [32P]-labeled RNA. The reaction mixtures were incubated at room temperature for 20 min, and then exposed to UV light for 11 min (Stratalinker 1800; Stratagene), followed by RNase T1 digestion (2 µg/ml) for 20 min at room temperature. The RNA-protein complexes were separated by 10% SDS-PAGE.
For gel shift assays, cytoplasmic extracts or recombinant PTBP2 were incubated with [32P]-labeled RNA probes, as described above, and fractionated in 4% polyacrylamide gels in TBE buffer. For supershift assays, extracts and antibody were preincubated for 5 min before addition of the radiolabeled RNA.
Affinity Purification of F1 RNA-Binding Proteins
Affinity purification was performed as described previously [26]. Testis cytoplasmic extracts from sexually mature CD-1 mice were prepared as described above. Biotinylated RNA was synthesized using the T7 Mega transcription kit (Ambion) with CTP that contained 40% biotin-16-CTP. In order to determine the optimal ratio of biotin-16-CTP to CTP, several different combinations were tested. A trace amount of [32P]-UTP was added to the transcription reaction for ease of monitoring the biotinylated RNA in the UV-cross-linking assay.
Briefly, 300 µl of streptavidin-conjugated paramagnetic particles (Promega, Madison, WI) were washed three times with 1.0 ml of 1x PBS, resuspended in 500 µl of binding buffer (BB) (20 mM Tris-HCl [pH 7.5], 2.5 mM MgCl2, 60 mM KCl, 5% glycerol, 1 mM DTT, 1x protease inhibitor cocktail) and stored at 4°C. To eliminate nonspecific protein binding to streptavidin particles, cytoplasmic extracts (4 mg), prepared as described above, were diluted 1:1 with 2x BB, mixed with the paramagnetic particles, and incubated for 1 h at 4°C. After centrifugation at 1000 x g for 5 min to remove the particles, the supernatants were mixed with biotinylated RNA and incubated for 2 h at 4°C. Washed particles (500 µl) were added to the mixture and after 60 min, the particles were recovered using a magnetic stand (Promega), and washed three times with 1.0 ml of binding buffer and once with wash buffer (20 mM Tris-HCl, [pH 7.5], 2.5 mM MgCl2, 120 mM KCl, 5% glycerol, 1 mM DTT). The bound proteins were recovered using elution buffer (20 mM Tris-HCl [pH 7.5], 1 M KCl), dialyzed against buffer A, and concentrated with a Microcon YM-10 (Millipore, Billerica, MA). The concentrated eluates were electrophoresed in 10% SDS-PAGE and detected by silver staining. Protein bands that were present in the F1 sample but not in the 18S rRNA control were digested with trypsin and spotted onto a MALDI (matrix-assisted laser desorption/ionization) plate. The peptide mass fingerprint was determined using a MALDI mass spectrometer (Applied Biosystems). The acquired MS-MS data were compared to the nonredundant protein database (National Center for Biotechnology Information), using the SEQUEST software to identify the proteins (Proteomics Core Facility, University of Pennsylvania).
Testis cytoplasmic extracts from sexually mature CD-1 mice were prepared as described above in the presence of SUPERase-in (1 U/µl) (Ambion). To immunoprecipitate PTBP2 complexes, testis extracts (5 mg) were precleared with protein A-agarose beads (100 µl), preimmune serum (10 µl), and yeast tRNA (100 µg/ml) in 5 ml of TKM-T buffer (20 mM Tris, [pH 7.6], 40 mM KCl, 1.5 mM MgCl2, 0.1% Tween-20, 0.1% Empigen BB) that contained 500 U/ml of SUPERase-in at 4°C for 2 h. After centrifugation at 2000 rpm for 2 min, the supernatants were incubated overnight at 4°C with protein A-agarose beads (20 µl) and anti-PTBP2 (a gift from R. Darnell; raised against the full-length PTBP2 [27], 10 µg). The mixtures were centrifuged at 2000 rpm for 5 min, and the pellets were washed four times with TKM-T (5 ml). The pellets were extracted in 250 µl of Trizol Regent (Sigma, St. Louis, MO), followed by chloroform extraction. Immunoprecipitated RNA was reverse-transcribed with random primers with the Superscript II reverse transcription kit (Invitrogen, Carlsbad, CA) and PCR assays were performed.
Immunohistochemistry and Western blot Analyses
Testes from sexually mature CD-1 mice were fixed and processed by the Histological Core Facility of the Children's Hospital of Pennsylvania. Immunohistochemistry was performed as previously described [25]. Anti-PTBP2 (the same antibody used in the immunoprecipitation assay) was diluted 1:200 and anti-PTBP1, designated PTBP1-NT (a gift from D. Black; prepared against a peptide within the N terminus of PTBP1 [28]), was diluted 1:100. Protein concentrations were determined with the BCA Protein Assay Kit (Pierce, Rockford, IL). For Western blotting, aliquots from total testis (20 µg) were separated by 10% SDS-PAGE and blotted onto polyvinylidene difluoride membranes (Millipore), and anti-PTBP2 (1:3000 dilution, the same antibody as used above), anti-PTBP1 (PTBP1-NT or PTBP1-CT, which was a gift from D. Black, was raised against a peptide within the C-terminus of PTBP1[29]), 1:1000 dilution), and anti-ß-actin (1:10 000 dilution) (Sigma) were used as primary antibodies.
Immunodepletion and In Vitro Decay Assay
PTBP2 was removed from testis extracts by overnight immunoprecipitation at 4°C with the same antibody as used above but prebound to protein A-Sepharose. Equal amounts of supernatant from the preimmune- and anti-PTBP2-depleted reactions were analyzed by immunoblotting and used in RNA decay assays.
Messenger RNA was examined in an in vitro decay assay, as described previously [30]. In brief, [32P]-labeled capped and polyadenylated RNAs were incubated with cytoplasmic extracts from testes (15 µg protein/assay) in 25 µl of a buffer that contained 100 mM KCH3COOH, 2 mM Mg(CH3COOH)2, 10 mM Tris-HCl (pH 7.6), 2 mM DTT, 10 mM creatine phosphate, 1 µg creatine phosphokinase, 1 mM ATP, 0.4 mM GTP, and 0.1 mM spermine. Reactions were incubated at 30°C and terminated with 100 µl stop buffer (400 mM NaCl, 25 mM Tris–HCl [pH 7.6], 1% SDS, 5 mM EDTA]. An in vitro-transcribed F12 RNA (100 000 cpm) was added as a normalization control for purification losses. The RNAs were phenol/chloroform extracted, ethanol-precipitated, and analyzed on denaturating urea 6% polyacrylamide gels.
HeLa cells were grown in 6-well tissue culture dishes that containing Dulbecco modified Eagle medium that was supplemented with 10% fetal bovine serum and streptomycin. The pEGFP-C2-UTR or pEGFP-C2-F3 plasmid was transiently transfected using the Liposome 2000 reagent (Invitrogen) following the manufacturers protocol. The Renilla luciferase phRL-TK vector was used as an internal control for transfection efficiency. Cotransfection assays of pEGFP-C2 reporter vectors with vectors that expressed PTBP2, truncated PTBP2 or translin were performed at a ratio of 1:4. For mRNA stability assays, actinomycin D (10 µM) was added 24 h after transfection, and the RNAs were purified using the Trizol reagent. RNA samples were separated on formaldehyde agarose gels and transferred overnight to a Hybond-N+ membrane. The membrane was hybridized with a [P32]-labeled EGFP DNA probe at 68°C for 1 h using the Quikhyb kit (Stratagene).
Identification of Proteins Binding to the 3'-UTR of Pgk2 mRNA
The stability and time of translation of transcripts are often controlled by trans-acting factors. To identify and characterize cellular factors that can stabilize and/or temporally control PGK2 expression, [32P]-labeled Pgk2 3'-UTR was incubated with adult testis cytoplasmic extracts and the RNA-protein complexes were cross-linked with UV light. Proteins with estimated molecular masses of 90 kDa, 55 kDa, and 50 kDa were detected (Fig. 1A).
|
To localize the primary binding sites of the proteins within the 3'-UTR of Pgk2, the 3'-UTR was subcloned into three regions (F1, F2, and F3) (Fig. 1B) and the UV cross-linking assays were repeated. We found the strongest RNA-protein interaction within the 93-nucleotide (nt) F1 transcript, which suggests that the proteins bind preferentially to an upstream sequence of the 3'-UTR of Pgk2 (Fig. 1A).
To identify the proteins that bind to the F1 RNA, affinity chromatography was performed with biotin-labeled F1 RNA and adult testis cytoplasmic extracts. After washing and elution of the bound proteins, they were trypsinized and the peptide sequences were determined by capillary liquid chromatography-electrospray mass spectrometry analysis. Several proteins, including FUBP1 (far upstream element-binding protein 1), KSRP (KH-type splicing regulatory protein), PTBP2 (polypyrimidine tract-binding protein 2), EEF1A (eukaryote elongation factor 1
), as well as fragments of PTBP2 and EEF1A (data not shown), were identified (Fig. 2A).
|
Comparing UV-cross-linked proteins of similar electrophoretic mobilities to those identified by affinity chromatography (the radiolabeled phosphates did not change significantly protein migration), PTBP2 was suspected to be the 55-kDa cross-linked protein, leading us to investigate further PTBP2 (see below). A slightly faster migrating protein band encoded a truncated form of the RNA-binding protein KSRP. Efforts to identify and purify the 90-kDa protein have not been successful, perhaps due to the low amount of protein available.
Mouse Testis Germ Cells and Somatic Cells Contain Two Isoforms of PTBP2
Two isoforms of PTBP (PTBP1 and PTBP2) have been identified, and PTBP2 is the abundant form in human and mouse brains and testes [27, 29]. In the brain, two isoforms of PTBP2 that differ by one amino acid, i.e., a deletion of R495 in the fourth RNA-binding motif, have been detected [31]. Recently, multiple isoforms of PTBP2, likely the result of alternative splicing, have also been identified in nonneuronal cells [31].
To determine which PTBP2 isoforms are expressed in the testis, we sequenced the complete ORFs of several Ptbp2 cDNA clones that were generated from the RNA of total testes, enriched populations of pachytene spermatocytes (meiotic cells) or round spermatids (postmeiotic cells). The two variants of PTBP2 found in the brain were found in mouse testis with slight enrichment for the PTBP2 that lacks R495 (3:2 ratio). Both isoforms of Ptbp2 were detected in cDNAs from total testis RNA, meiotic pachytene spermatocytes, postmeiotic round spermatids, and the testes of Kit-mutant mice (the testes of the latter lack germ cells), which suggests that neither isoform is germ cell- or somatic cell-specific in the testis. Since the Ptbp2 cDNAs that lack an arginine residue at amino acid position 495 are more abundant and this isoform appears to be upregulated in the testis (D.L. Black, personal communication, 31), we chose to use it in the subsequent experiments.
PTBP2 Is Present in a Pgk2 mRNA-Protein Complex in Testis Extracts
To establish whether PTBP2 binds to Pgk2 mRNA in vivo, anti-PTBP2 was used to immunoprecipitate RNA-protein complexes from adult testes extracts. Using a highly purified polyclonal antibody specific for PTBP2 [27] and specific primers to detect Pgk2 and control mRNAs, Pgk2 mRNA was selectively immunoprecipitated using the anti-PTBP2 antibody (Fig. 2B). In contrast, two highly abundant and also translationally regulated germ cell mRNAs that encode protamine 2 and transition protein 2 (TNP2) [32] were not precipitated by the anti-PTBP2 antibody. None of the three mRNAs was precipitated with preimmune serum.
Expression Patterns of PTBP1 and PTBP2 Proteins in Mouse Testes
To define the cellular functions of PTBP2 during spermatogenesis, we examined the cellular and subcellular locations of this protein in mouse testes using a rabbit polyclonal antibody raised against recombinant mouse PTBP2 [27] (Fig. 3A). As a control, the cellular and subcellular locations of PTBP1 were determined with a rabbit polyclonal antibody (PTBP1-NT) raised against a unique 15-amino acid N-terminus peptide of PTBP1 [28].
|
Immunohistochemical staining revealed that PTBP2 was abundant in the nuclei and cytoplasm of germ cells from pachytene spermatocytes to spermatids, whereas PTBP1 was most abundant in spermatogonia, indicating generally nonoverlapping patterns of expression of PTBP1 and PTBP2 in the testis (Fig. 3A). The abundance of PTBP2, but not of PTBP1, in the cytoplasm is consistent with a significant RNA-binding function for PTBP2. The antibody specificity was confirmed by Western blotting of total testis extracts (Fig. 3B).
To confirm the subcellular distributions of PTBP1 and PTBP2 during spermatogenesis, nuclear and cytoplasmic extracts from the testes of prepubertal (17-day-old and 22-day-old) mice and adult mice were analyzed by Western blotting (Fig. 3C). The testes of 17-day-old mice contain pachytene spermatocytes but lack postmeiotic germ cells, whereas the early stages of round spermatids are abundant in the testes of 22-day-old mice. Consistent with many studies of PTBP1 in mammalian somatic cells [28, 33], PTBP1 is primarily a nuclear protein in the male germ cells of both prepubertal and adult mice (Fig. 3C). Although PTBP2 is also abundant in nuclei, we detected increasing expression of PTBP2 in the cytoplasm as the testes differentiated from prepubertal to adulthood. Considering that Pgk2 mRNA is detectable in meiotic germ cells, while its protein is first detected in postmeiotic germ cells [18], we hypothesize that PTBP2, which is abundant at the time of Pgk2 mRNA synthesis and storage, plays a role in regulating the stability of Pgk2 mRNA.
PTBP2, But Not PTBP1, Specifically Binds to the F1 Region of the 3'-UTR of Pgk2 mRNA
To determine whether PTBP2 binds to the F1 region of the 3'-UTR of Pgk2 mRNA, gel-shift assays were performed with [P32]-labeled RNA probes (F1, F2 and F3) that were incubated with adult testis cytoplasmic extracts. Consistent with the results from the UV-cross-linking assay (Fig. 1), the F1 RNA, but not the F2 and F3 RNAs, formed a strong complex with the cytoplasmic proteins (Fig. 4A, lane 1). Moreover, this RNA-protein complex was strongly shifted with anti-PTBP2 antibody (Fig. 4A, lane 2), which indicates that PTBP2 binds to the F1 region of the 3'-UTR of Pgk2.
|
To investigate whether PTBP2 binds directly to the F1 RNA, gel shift assays were performed with recombinant PTBP2. The recombinant PTBP2 bound to F1 (Fig. 4B, lane 2), but not to F2 (Fig. 4B, lane 4) or F3 (Fig. 4B, lane 6). These RNA-binding assays confirm that PTBP2 binds directly and selectively to the F1 region of Pgk2 mRNA.
Since the testis contains both PTBP1 and PTBP2 and both can bind to polypyrimidine-rich regions [29], we performed similar gel shift and supershift assays with two different antibodies that selectively detect PTBP1 (Fig. 4C). Although the control anti-PTBP2 antibody produced a supershift (Fig. 4C, lane 2), no supershifts were detected with either antibody to PTBP1 (Fig. 4C, lanes 3, 4), which indicates that PTBP2, but not PTBP1, binds to the F1 RNA. The addition of ATP, which has been reported to dissociate PTBP1-RNA complexes [29], failed to dissociate the PTBP2-F1 complex, further supporting the specificity of a PTBP2-F1 RNA complex (data not shown).
Binding of PTBP2 Requires More than One CU-Rich Element of F1
To define further the part of F1 of the 3'-UTR of Pgk2 that binds PTBP2, RNAs from five different subclones of F1 (F1–1, F1–2, F1–3, F12, and F23), were transcribed in vitro and used in mobility shift assays (Fig. 4, D–F). The F1–1 and F1–3 regions contain CU-rich elements, which are believed to be binding sequences with preference for the PTBP class of proteins [29]. Although a strong F1-protein complex was detected (Fig. 4E), no specific RNA-PTBP2 complexes were seen with F1 subclones F1–1, F1–2, F1–3, F12 or F23 (the latter contains a CU-rich like sequence) (Fig. 4, E and F). Thus, only the 93-nt F1 region can form stable PTBP2-RNA complexes.
PTBP2 Helps to Stabilize Pgk2 mRNA in Extracts and Cells
Since PTBP proteins are known to be involved in mRNA stabilization and initiation of IRES-directed translation [19–22, 34, 35], we wished to determine whether PTBP2 plays a role in Pgk2 mRNA stabilization in the mouse testis by binding to the 3'-UTR of the Pgk2 mRNA.
Two complementary approaches were taken to determine whether PTBP2 binding to Pgk2 mRNA alters its stability. First, we conducted an in vitro decay assay in testis extracts using 5'-capped and 3'-polyadenylated transcripts. The 32P-labeled F1 or a negative control (F2 RNA) was incubated with testis extract and recombinant PTBP2. The addition of 100 nM or 200 nM PTBP2 increased the estimated half-life of the F1 RNA from about 20 min to 100 min (Fig. 5A). In contrast, the stability of the F2 RNA, which does not bind PTBP2, was not increased by the addition of the same amounts of PTBP2 (Fig. 5A). Second, we used anti-PTBP2 polyclonal antibody to deplete PTBP2 from testis extracts (Fig. 5B). The decay of 32P-labeled F1 or F2 RNAs was monitored in the testis extracts prepared using preimmune serum or anti-PTBP2 antibody. F1 RNA was rapidly degraded in the PTBP2-depleted testis extract. The addition of recombinant PTBP2 to the PTBP2-depleted testis extract restored F1 RNA stability (Fig. 5C). In contrast, the decay of control F2 RNA was not affected by the depletion of PTBP2. These results suggest that PTBP2 is an important factor for Pgk2 mRNA stability in vitro.
|
To extend these in vitro experiments to cells, a second group of experiments using transient transfection assays was performed. Since PTBP2 is predominantly located in the nucleus following transfection into HeLa cells (data not shown), we prepared an N-truncated construct of PTBP2 that lacks the nuclear localization sequence by removing 55 amino acids from the N-terminus. This change does not affect the RNA binding of PTBP, as reported previously [33–38], and it directs the N-truncated PTBP2 to the cytoplasm in HeLa cells (data not shown). The endogenous PTBP2 levels in HeLa cells are very low [29] and the overexpression of PTBP2 was confirmed by Western blotting (data not shown).
To determine whether PTBP2 can stabilize mRNAs in living cells, an EGFP construct that contained the 3'-UTR of Pgk2 mRNA was cotransfected with vectors that expressed the full-length PTBP2, the N-truncated PTBP2 or translin, which is another RNA-binding protein. Translin was used as the negative RNA-binding protein control for cotransfection because it does not bind to Pgk2 mRNA in vivo [39], and the full-length PTBP2 was used as a subcellularly mislocalized protein control. A similar EGFP construct that contained the F3 region of Pgk2 mRNA was used as a nonbinding control. The N-truncated cytoplasmic PTBP2, but not the full-length nuclear PTBP2, clearly stabilized the EGFP mRNA that contained the F1 element of the 3'-UTR of Pgk2 mRNA, which indicates that PTBP2 overexpression in the cytoplasm can stabilize the reporter gene mRNA (Fig. 6). This stabilization was specific for the F1-binding region, as there was no stabilization when the F1 region was replaced with F3 in the reporter mRNA when cotransfected with either the PTBP2 proteins or translin (Fig. 6). These in vitro decay and transfection assays demonstrate that PTBP2 can stabilize Pgk2 mRNA.
|
Both male and female germ cells rely heavily upon posttranscriptional regulation to control the timing of synthesis of specific proteins. During the haploid phase of spermiogenesis, transcription ceases, which creates the need for mRNA storage and delayed translation for many postmeiotic mRNAs that are essential for spermatozoan formation. Many of the well-characterized posttranscriptionally regulated mRNAs of the mammalian testis are first transcribed in postmeiotic germ cells by the spermatid transcription factor CREM-tau, and then stored in the cytoplasm for up to 7 days [3, 4].
Pgk2 is a member of a group of posttranscriptionally regulated mRNAs that are transcribed during meiosis in primary spermatocytes and stored as ribonucleoprotein particles (RNPs) until they are translated up to 14 days later in postmeiotic spermatids [5]. In fact, since Pgk2 mRNAs are first detected in early-stage spermatocytes and PGK2 protein first appears in late-stage spermatids, Pgk2 mRNA appears to be one of the most translationally delayed germ cell mRNAs in the mammalian testis. The general failure to radiolabel and to culture male germ cells makes it difficult to estimate precisely the turnover rate and half-life of Pgk2 mRNA in vivo.
Using RNA affinity chromatography to identify trans-acting factors that might regulate Pgk2 mRNA, we have identified several proteins, including PTBP2, a proteolytic product of PTBP2, FUBP1, and KSRP, which bind to a 93-nt region of the 259-nt 3'-UTR of Pgk2 mRNA. All of the identified proteins are biologically reasonable regulators of the Pgk2 mRNA. FUBP1 is a transcriptional activator of c-myc, which is a component of a complex that binds to AU-rich elements [40]. In somatic cells, KSRP is a 75-kDa protein that is involved in pre-mRNA splicing in the nucleus and in an exosome-directed mRNA decay pathway in the cytoplasm [41, 42]. In the testis, we detected a KSRP of about 52 kDa, which is perhaps a proteolytic digestion product of a larger KSRP or a protein that arises from an alternate translation initiation site. Northern blotting of total testis RNA revealed a single KSRP mRNA of about 4 kb, which suggests that the smaller KSRP is not derived from a novel spliced mRNA in the testis (data not shown).
PTBP2, which was previously identified as a form of PTBP that is abundant in the brain [27], belongs to a multi-functional family of proteins involved in mRNA splicing, localization, and stability, as well as IRES-directed translation initiation [19–22, 34, 35, 43, 44]. PTBP contains four RNA-binding domains and binds to target RNA sequences that contain one or more pyrimidine-rich motifs, e.g., UCUU or CUCUCU. All four RNA-binding motifs of the protein are capable of specific interactions with RNA, making it difficult to define a specific RNA consensus sequence [45]. Moreover, PTBP is not only an RNA-binding protein that binds to CU tracts, but it also serves as an RNA chaperone that brings together nonadjacent pyrimidine tracts. In the present study, we show that PTBP2 binds to a subregion (the F1 region) of the 3'-UTR of Pgk2, which contains two CU-rich regions that differ from the consensus pyrimidine-rich motifs. The requirement of both CU-rich regions in the F1 region for PTBP2 binding (Fig. 4) is consistent with the observation that strong interactions involving PTBP2 usually depend on its binding to multiple CU-rich regions [45–47].
PTBP protein, which was initially identified as a splicing factor, is a nuclear protein in most cells, although Xenopus oocytes contain significant amounts of PTBP in the cytoplasm [48, 49]. In the present study, we report that in the testis, PTBP1 is primarily located in the nucleus, while PTBP2 is more abundant in the cytoplasm (Fig. 3). Transfection of EGFP-fused PTBP2 into HeLa cells revealed that PTBP2 was also predominantly found in the nuclei (data not shown), which suggests that interactions with other proteins or its modification may alter the localization of PTBP2 in male germ cells. In somatic cells, phosphorylation of a serine in the N-terminus of PTBP1 directs it to the cytoplasm [49]. The presence of PTBP2 in the cytoplasm allows it to play an important role in the posttranscriptional control of target mRNAs, such as Pgk2 mRNA in the testis.
PTBP1 has been recently identified as the factor that stabilizes CD154, insulin, VGFP, and iNOS mRNAs by binding to their 3'-UTRs [19–22]. In the present study, we demonstrate by in vitro decay and transient transfection assays that PTBP2 increases the stability of Pgk2 mRNA by binding to its 3'-UTR. We also show that during spermatogenesis, the expression of PTBP2 protein parallels the mRNA expression pattern of Pgk2, a prerequisite for it to regulate the stability/translation of Pgk2 mRNA. Recently, it has been proposed that KSRP and PTBP1 complexes that bind to the 3'-UTR of iNOS mRNA regulate the turnover of the mRNA [22]. Our isolation of PTBP2 and a truncated form of KSRP by affinity chromatography (Fig. 2) raises the exciting possibility that interactions between KSRP and PTBP2 modulate the stabilization, translation, and turnover of Pgk2 mRNA.
ACKNOWLEDGMENTS
We thank R.B. Darnell (Rockefeller University) for anti-PTBP2, and D.L. Black (University of California, Los Angeles) for the anti-PTBP1-NT and anti-PTBP1-CT antibodies.
FOOTNOTES
1Supported by NIH grant HD 28832. 
Correspondence: 2Norman B. Hecht, Center for Research on Reproduction and Women's Health, University of Pennsylvania School of Medicine, 1310 Biomedical Research Building, 421 Curie Boulevard, Philadelphia, PA 19104-6080. FAX: 215 573 7627; e-mail: nhecht{at}mail.med.upenn.edu
Received: 10 January 2007.
First decision: 26 January 2007.
Accepted: 27 February 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  Z. Yu and N. B. Hecht The DNA/RNA-Binding Protein, Translin, Binds microRNA122a and Increases Its In Vivo Stability J Androl, September 1, 2008; 29(5): 572 - 579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Sartini, H. Wang, W. Wang, C. F. Millette, and D. L. Kilpatrick Pre-Messenger RNA Cleavage Factor I (CFIm): Potential Role in Alternative Polyadenylation During Spermatogenesis Biol Reprod, March 1, 2008; 78(3): 472 - 482. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
