Abstract

Paraspeckle protein 1 (PSP1) in humans is a recently identified component protein of a novel nuclear body, paraspeckle. The protein has a DBHS (Drosophila behavior, human splicing) motif that is found in PSF and p54nrb/NonO proteins. These DBHS-containing proteins have been reported to be involved in various nuclear events such as DNA replication, transcription, and mRNA processing. Here we show that mouse paraspeckle protein 1 (mPSP1; encoded by the Pspc1 gene) has two isoforms with different C-termini lengths. Abundant expression of the longer isoform (mPSP1-α) and the shorter one (mPSP1-β) were observed in testis and kidney, respectively. Transiently expressed mPSP1-α was localized in nuclei, but mPSP1-β was localized in both nuclei and cytoplasm. These observations suggest that alternative splicing regulates tissue distribution and subcellular localization. Like other DBHS-containing proteins, mPSP1 has RNA-binding activity. In mouse testis, mPSP1-α was found in the nuclear matrix fraction. Furthermore, by coimmunoprecipitation, we confirmed that mPSP1 interacts with other DBHS-containing proteins, PSF and p54nrb/NonO. Therefore, we conclude that mPSP1 may regulate multiple phases of important nuclear events during spermatogenesis.

Introduction

Spermatogenesis is a complex process of cell development and differentiation during which extensive changes in cell morphology and intracellular and intranuclear organization occur. This process requires the coordinate expression of a large array of testis-specific genes. It is becoming evident that early mRNA processing events such as 5′ capping, splicing, and polyadenylation in the nucleus occur while transcription proceeds, suggesting coupling of transcriptional and posttranscriptional regulation. A growing number of multifunctional proteins are being found that participate in both processes. PSF and p54nrb/NonO are RNA- and DNA-binding proteins that have been implicated in the regulation of both processes of gene expression in nuclei.

PSF is cloned as an interacting protein with a polypyrimidine tract binding (PTB) protein that binds the polypyrimidine tract in most introns of higher eukaryotes [1]. The region is important for the definition of the 3′ splice site. Several proteins bind the region and participate in constitutive pre-mRNA splicing. Human p54nrb and its mouse ortholog, NonO, share the DBHS (Drosophila behavior, human splicing) motif [2], which comprises 2 × RRM (RNA recognition motif) and HTH (helix-turn-helix) DNA-binding domains, with PSF. Although there is no direct evidence for involvement of pre-mRNA splicing, p54nrb/NonO binds RNA via its N-terminus [3] and can bind to β-globin pre-mRNA and several mRNAs [4], the intronic pyrimidine-rich sequence in β-tropomyosin pre-mRNA [5, 6].

In addition to the RNA-binding properties described above, PSF binds DNA [7] and p54nrb/NonO binds single-strand (ss) DNA (and RNA) through its N-terminus and double-strand (ds) DNA through its C-terminus [3]. These proteins bind the insulin-like growth factor response element in the porcine P450scc gene, DNA-binding domain (DBD) of thyroid hormone receptors (TRs) and retinoid X receptors (RXRs) [8], to form a complex with SF-1, and together, bind the CYP17 promoter. Repression of transcription is achieved by the binding of the Sin3A repressor to PSF along with class I histone deacetylase. The protein p54nrb/NonO binds to an enhancer element in the long terminal repeats of murine intracisternal A particles and activates transcription [4]. Recently, it was reported that PSF and p54nrb/NonO bind the androgen receptor and function as a coactivator that potentiates transcription [9]. These data demonstrate that these proteins are involved not only in mRNA splicing, but also in transcriptional activation and repression. The coordination of transcription and mRNA processing does not fully account for the multiple functions of these proteins.

Another aspect of the function of these proteins is a nuclear retention of defective RNAs, DNA unwinding and DNA pairing during DNA replication [10, 11], annealing of DNA [1113], and nonclassical carbonic anhydrase activity [14]. These proteins are usually identified as monomers. However, PSF and p54nrb/NonO intrinsically can form homodimers and heterodimers. In fact, the PSF-p54nrb/NonO heterodimer is able to enhance the DNA topoisomerase I activity better than PSF alone [10]. These proteins might mediate different functions and monomer-dimer transitions depending on the intracellular compartments (nuclear matrix, nucleoplasm, nuclear foci, nucleolus, and cytoplasm [1521]) in which they are located, and the localization might be regulated by phosphorylation [16, 20, 22].

Human PSP1 (paraspeckle protein encoded by PSPC1 gene), which contains a DBHS motif similar to p54nrb/NonO and PSF, was recently shown to be a component of a novel nuclear domain termed paraspeckles [19]. PSP1, PSF, and p54nrb/NonO are found in the complex that binds to the transactivation domain of the androgen receptor [9]. We independently identified mouse PSP1 (mPSP1) as a novel RNA-binding protein with an abundant expression in mouse testis. Here we report the expression of mPSP1 during spermatogenesis. Moreover, we show that mPSP1 is an RNA-binding protein in the nuclear matrix and can dimerize with PSF and NonO. From all data presented here, we suggest that mPSP1 is concerned with transcriptional regulation and mRNA processing and may participate in chromatin remodeling and nuclear shaping during spermatogenesis.

Materials and Methods

Cloning of Pspc1 cDNA and Antibody Production

Two alternative splicing forms of mouse Pspc1 cDNAs that encode mPSP1-α (523 amino acids [aa]) and mPSP1-β (391 aa) were cloned by using a degenerate polymerase chain reaction (PCR) strategy essentially as previously described [23]. On the basis of the consensus sequence of the RRM motif, degenerate primers (RNP-2 primer, 5′-CGCCTTGTACAT(C/A/T)GGIGGN(C/A/T)T-3′ and RNP-1 primer, 5′-GCAITCIACGAAI(G/C/A)C(G/A)(G/A/T)ANCC-3′) were designed. These primers were used for PCR amplification from mouse ovary cDNA. The PCR-amplified products were cloned into the PCR II TOPO vector. By cDNA and genome database survey, two of the clones were identified as alternative splice variants of mouse Pspc1. The predicted amino acid sequences of the two clones are identical from the N-terminus to Asn386 at the amino acid level, but have different C-termini. The protein encoding a common region of mPSP1 (38–386 aa residues) was produced as a thioredoxin fusion protein by cloning the corresponding cDNA into the BamHI and NotI sites of pET-32a(+) (Novagen, Madison, WI). A rabbit polyclonal antibody was raised against the bacterially expressed recombinant protein, and the anti-mPSP1 antibody that recognized both isoforms was affinity-purified. The specificity of the antibody was confirmed by the presence of appropriate signals as shown by Western blot analysis on mouse testis extracts and bacterially expressed recombinant mPSP1.

Preparation of Cell Lysates and Western Blot Analysis

Protocols for the use of animals in this experiment were in accordance with National Institutes of Health standards established in the Guidelines for the Care and Use of Experimental Animals. All tissues were removed from BALB/cAJcl mice deeply anesthetized with Ketalar 50 (Sankyo, Tokyo, Japan). Tissues were homogenized in PBS buffer containing Complete proteinase inhibitor cocktail (Roche Diagnostics, Basel, Switzerland) and 10 M urea. Homogenates were mixed with 5× lysis buffer (5% SDS, 5% dithiothreitol [DTT], 1.5% Empigen BB [Calbiochem, San Diego, CA]/PBS) and sonicated using a Sonifier 250 (Branson Ultrasonics, Danbury, CT) at 40% of its maximum output five times for 10 sec. Each protein extract was diluted with 2× sample buffer (3% SDS, 20% glycerol, 0.12 M Tris-HCl pH 6.8, 5% β-mercaptoethanol, and 0.01% bromophenol blue), boiled for 3 min, and centrifuged at 17 000 × g for 20 min to remove cell debris. Western blot analysis was performed essentially according to the manufacturer’s instructions using the enhanced chemiluminescence (ECL) Western blotting detection reagents (Amersham Pharmacia Biotech, Piscataway, NJ). Supernatants were electrophoresed on 12% SDS-PAGE gels and transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA). Following blocking with 5% skim milk in Tris-buffered saline-0.05% Tween-20 (TBS-T), the blots were incubated with rabbit anti-mPSP1 polyclonal antibody (1:1000) in TBS-T containing 5% skim milk for 1 h. After washing with TBS-T, the membrane was incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit immunoglobulin G antibody (1:5000) (Zymed Laboratories, Inc. San Francisco, CA) for 1 h. After further washing with TBS-T, the immunoreactive bands were visualized with ECL reagent (Amersham Pharmacia Biotech). The blots were stained with Comassie Brilliant Blue to evaluate equal loadings of the proteins in each lane.

Immunohistochemistry

BALB/cAJcl adult mouse (9-wk-old) testes were fixed overnight in Bouins fixative. The testes were embedded in paraffin and cut into 5-μm sections. After deparaffinization and rehydration by xylene and serial dilutions of ethanol, the slides were blocked with 5% skim milk in PBS for 1 h. Endogenous peroxidases were blocked with 3% H2O2 for 15 min at room temperature. The sections were incubated with a 1:100 dilution of anti-mPSP1 polyclonal antibody in 5% skim milk/PBS overnight at 4°C. Then, testis sections were incubated with 1:100 dilutions of HRP-conjugated anti-rabbit secondary antibody (Zymed Laboratories) in 5% skim milk/PBS for 1 h at room temperature. Immunohistochemical staining for peroxidase was performed using diaminobenzidine (DAB) as chromogen. Sections were counterstained with Mayers hematoxylin (Muto Pure Chemicals, Tokyo, Japan). Serial sections were stained with periodic acid-schiff (PAS) to identify the stages of the seminiferous tubules. The sections were examined by light microscopy (Olympus, Tokyo, Japan).

Construction and Cell Transfection of pGFP-mPSP1-α and pGFP-mPSP1-β Eukaryotic Expression Vectors

Coding sequences of two isoforms of mPSP1 were PCR-amplified and subcloned into pEGFP-C2 vector (Clontech Laboratories, Inc., Franklin Lakes, NJ). NIH/3T3 cells were transiently transfected with the pGFP-mPSP1-α and pGFP-mPSP1-β eukaryotic expression vectors using PolyFect transfection reagent (Qiagen, Hilden, Germany). At 18 h posttransfection, cells were fixed in 4% paraformaldehyde in PBS at room temperature, and the subcellular localization of the exogenous proteins was visualized by fluorescence microscopy (Olympus). Hoechst 33258 (0.6 μg/ml) was used for nuclear staining 30 min prior to fixation.

Nuclear Matrix Preparation

Nuclear matrices were prepared as described in [2426]. A mouse testis was homogenized using Downs homogenizer in buffer A (10 mM Hepes/KOH pH 7.4, 15 mM KCl, 1 mM EDTA, 0.25 M sucrose, 1 μg/μl aprotinin, 1 μg/μl pepstatin, 1 μg/μl leupeptin, 1 mM DTT, 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 50 mM NaF, 1 mM Na3VO4), and centrifuged at 2500 × g for 10 min. The supernatant was used as the cytoplasmic fraction. The precipitate was dissolved in buffer A, overlaid onto 0.5 M sucrose in buffer A, and centrifuged at 2500 × g for 10 min. This was repeated three times. To remove the chromatin, the nuclear fraction was digested with 250 U/ml RNase-free DNase I in CSK buffer (10 mM Pipes pH 6.8, 100 mM NaCl, 300 mM sucrose, 3 mM MgCl2, 1 mM EGTA, 0.5% Triton X-100, 1 μg/μl aprotinin, 1 μg/μl pepstatin, 1 μg/μl leupeptin, 1 mM DTT, 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride, 50 mM NaF, 1 mM Na3VO4, and 2 mM ribonucleoside vanadyl complexes) at 37°C for 30 min. After centrifugation, the residual nuclear pellet was resuspended in 0.25 M ammonium sulfate in CSK buffer and incubated at 4°C for 5 min. The washing with ammonium sulfate was repeated three times, and the pellet was again washed with 2 M NaCl in CSK buffer at 4°C for 5 min. The residual pellet was solubilized in 8 M urea in CSK buffer and used as a nuclear matrix fraction. From these procedures, we extracted a 6-mg cytoplasmic fraction and a 96-μg nuclear matrix fraction from one testis. The proteins (20 μg) from each fraction were eletrophoresed on 12% SDS-PAGE gel and subjected to Western blot analysis by anti-mPSP1 and anti-lamin B1 (Santa Cruz Biotechnologies, Santa Cruz, CA) antibodies.

In Vitro RNA-Binding Analysis

In vitro RNA-binding analysis occurred as described in [23]. Bacterial expression vectors that produce two RRM domains of mPSP1 were constructed by subcloning appropriate cDNA fragments into pET28(+) (Novagen). After purification of the recombinant proteins by Ni-NTA resin, 100 pmol of purified fusion protein extracts were electrophoresed on 12% SDS-PAGE gels and transferred onto PVDF membranes (Millipore). The protein on the membrane was renatured in PBS-0.05% Tween-20 (PBS-T) containing 1 mM DTT at room temperature for 30 min. The RNA homopolymers poly(A), poly(G), poly(C), and poly(U) were [32P]-labeled using T4 polynucleotide kinase (Takara, Tokyo, Japan). To reduce the background, the membrane was processed overnight in blocking buffer (10 mM Tris-HCl pH 7.5, 1% BSA, 1 mM DTT, and 100 mM NaCl) at 4°C, and then incubated with 105 cpm of [32P]-labeled RNA homopolymers and tRNA (20 μg) in binding buffer (10 mM Tris-HCl pH 7.5, 100 mM NaCl, and 1 mM DTT) for 90 min at room temperature. After extensive washing with the binding buffer, the blots were analyzed by an imaging system of FLA-2000 (Fuji Photo Film, Tokyo, Japan).

Coimmunoprecipitation Assay

The expression plasmids (pFLAG-mPSP1-α, pHA-PSF, and pHA-NonO) were prepared by subcloning mPSP1-α cDNAs into the pFLAG-CMV2 vector (Sigma-Aldrich, St. Louis, MO), and the mouse PSF and NonO cDNAs into the pcDNA3 vector (Invitrogen, Carlsbad, CA). NIH/3T3 cells that were 70% confluent in a 10-cm dish were cotransfected with 7 μg of indicated plasmids (pFLAG-mPSP1-α and either pHA-PSF or pHA-NonO) using PerFectin Reagent (Gene Therapy Systems, San Diego, CA) according to the manufacturer’s instructions. At 30 h posttransfection, the cells were lysed with lysis buffer (50 mM Tris-HCl pH 7.5, 0.5% NP-40, 150 mM NaCl, 50 mM NaF, 1 mM Na3VO4, 1 mM DTT, and protease inhibitor cocktail [Roche Diagnostics]). All lysates from cells in the 10-cm dish were immunoprecipitated with 1 μg of mouse anti-FLAG M2 monoclonal antibody (Sigma-Aldrich) or anti-HA rabbit antibody (MBL, Nagoya, Japan). The precipitates were resolved by SDS-PAGE and analyzed by immunoblotting with the anti-FLAG M2 monoclonal antibody and mouse anti-HA monoclonal antibody (12CA5).

Results

Tissue and Developmental Expression of mPSP1 Protein

We obtained two clones using degenerate primers on the basis of a degenerate PCR strategy. By homology search and comparison of the genome sequence of human PSP1, we identified two clones as alternative splicing variants.

Two isoforms, mPSP1-α and mPSP1-β, differ in the length of their C-terminus. In addition to the common region for both isoforms, the shorter form (mPSP1-β) has an additional 5 residues, and the longer form (mPSP1-α) has an additional 137 residues (Fig. 1). We raised a rabbit antiserum against the common region of mPSP1 fused with thioredoxin and a hexahistidinyl segment. The antibody was affinity-purified by immobilized antigen affinity chromatography. To determine the expression of mPSP1 protein in various mouse tissues, Western blot analysis was performed. Two major bands that corresponded well with the expected molecular masses of the isoforms (58 kDa and 45 kDa) were observed. The longer isoform, mPSP1-α, was most abundant in testis (Fig. 2, lane 9), and the shorter one, mPSP1-β, was most abundant in kidney (Fig. 2, lane 7). A lower level of expression of both isoforms was observed in other tissues examined, suggesting that the expression level and the isoform expression were regulated in a tissue-specific manner.

Fig. 1

Comparison of protein structures between two mPSP1 isoforms. Mouse PSP1-α (523 aa) and mPSP1-β (391 aa) share an N-terminal fragment and a DBHS domain, and differ in the length of their C-termini. In addition to the common region, the shorter form (mPSP1-β) has an additional 5 residues (VVETA), and the longer form (mPSP1-α) has an additional 137 residues. These sequence data are available from GenBank, the European Molecular Biology Laboratory (EMBL), and the DNA Data Bank of Japan (DDBJ) (accession numbers AK017689)

Fig. 1

Comparison of protein structures between two mPSP1 isoforms. Mouse PSP1-α (523 aa) and mPSP1-β (391 aa) share an N-terminal fragment and a DBHS domain, and differ in the length of their C-termini. In addition to the common region, the shorter form (mPSP1-β) has an additional 5 residues (VVETA), and the longer form (mPSP1-α) has an additional 137 residues. These sequence data are available from GenBank, the European Molecular Biology Laboratory (EMBL), and the DNA Data Bank of Japan (DDBJ) (accession numbers AK017689)

Fig. 2

Western blot analysis of mPSP1 in protein extracts from adult mouse tissues. Arrowheads indicate the position of mPSP1 isoforms. The molecular size markers are indicated at left. 1, Cerebrum; 2, cerebellum; 3, salivary gland; 4, lung; 5, liver; 6, spleen; 7, kidney; 8, ovary; 9, testis

Fig. 2

Western blot analysis of mPSP1 in protein extracts from adult mouse tissues. Arrowheads indicate the position of mPSP1 isoforms. The molecular size markers are indicated at left. 1, Cerebrum; 2, cerebellum; 3, salivary gland; 4, lung; 5, liver; 6, spleen; 7, kidney; 8, ovary; 9, testis

Next, to determine the subcellular localization of mPSP1 isoforms, NIH/3T3 cells were transfected with pGFP-mPSP1-α and pGFP-mPSP1-β. The GFP-mPSP1-α demonstrated punctate foci dispersed throughout, but exclusively in the nuclei (Fig. 3a). Conversely, most GFP-mPSP1-β uniformly localized to the nuclei, and significant amounts of the shorter isoform were also observed in the cytoplasm (Fig. 3d). Green fluorescent protein alone did not give any specific localization (Fig. 3g).

Fig. 3

Localization of GFP-mPSP1 in NIH/3T3 cells. NIH/3T3 cells were transiently transfected with the pGFP-mPSP1-α (37Μ–523ϒ) and pGFP-mPSP1-β (37Μ–391Α) eukaryotic expression vectors. Panels a, d, and g) GFP-mPSP1-α, GFP-mPSP1-β and control GFP. Panels (b, e, and h) show nuclear staining by Hoechst 33258 for the same samples in (a, d, and g), respectively. Panels (c, f, and i) are phase-contrast microscopy for the same samples in (a, d, and g), respectively. Bars = 50 μm

Fig. 3

Localization of GFP-mPSP1 in NIH/3T3 cells. NIH/3T3 cells were transiently transfected with the pGFP-mPSP1-α (37Μ–523ϒ) and pGFP-mPSP1-β (37Μ–391Α) eukaryotic expression vectors. Panels a, d, and g) GFP-mPSP1-α, GFP-mPSP1-β and control GFP. Panels (b, e, and h) show nuclear staining by Hoechst 33258 for the same samples in (a, d, and g), respectively. Panels (c, f, and i) are phase-contrast microscopy for the same samples in (a, d, and g), respectively. Bars = 50 μm

Because mPSP1-α protein was highly expressed in testis, we examined the expression of mPSP1 in mouse testis by immunohistochemistry. Immunostaining of mPSP1 protein on mouse testis sections was first detected in the nuclei of leptotene spermatocytes (Fig. 4, e and f). After meiosis, strong signals were observed in the nuclei of round spermatids (Fig. 4a). Very faint signals remained at early stages of elongating spermatids (Fig. 4c), and its expression was not detectable at stage 10 (Fig. 4d).

Fig. 4

Localization of mPSP1 protein in testis. Sections of adult mouse testis were immunostained with affinity-purified anti-mPSP1 rabbit polyclonal antibody. Sections were counterstained with hematoxylin. a) Stage III, (b) stage VI∼VII, (c) stage IX, (d) stage X, (e) stage XI, (f) stage XII. Bars = 50 μm. Arrow indicates dotted staining of a round spermatid. An early elongated spermatid indicated by an arrowhead exhibits partially faded staining

Fig. 4

Localization of mPSP1 protein in testis. Sections of adult mouse testis were immunostained with affinity-purified anti-mPSP1 rabbit polyclonal antibody. Sections were counterstained with hematoxylin. a) Stage III, (b) stage VI∼VII, (c) stage IX, (d) stage X, (e) stage XI, (f) stage XII. Bars = 50 μm. Arrow indicates dotted staining of a round spermatid. An early elongated spermatid indicated by an arrowhead exhibits partially faded staining

Mouse PSP1 Associates with Other DBHS-Containing Proteins PSF and p54nrb/NonO

The PSF and p54nrb/NonO proteins have been usually identified as monomers. However, PSF is able to interact directly with p54nrb/NonO, thus forming a heterodimer [6, 7]. To determine the heterodimerization activities of mPSP1, we performed coimmunoprecipitation experiments on transfected cells. Protein extracts from cells transfected with two plasmids encoding FLAG-mPSP1 and HA-PSF or HA-NonO were immunoprecipitated by either anti-FLAG antibody or anti-HA antibody, followed by immunoblot analysis for the resulting immunocomplex using either anti-FLAG antibody or anti-HA antibody as a probe. The expressions derived from the transfected plasmids were verified, as shown in Figure 5, a and b, lanes 1–3. HA-PSF and HA-NonO were detected in the immunocomplexes by anti-FLAG antibody (Fig. 5, a and b, lane 6). Reciprocally, the immunocomplex with anti-HA antibody contained FLAG-mPSP1 (Fig. 5, a and b, lane 9), demonstrating that mPSP1 binds to PSF and NonO, forming heterodimers. The interactions between mPSP1 and PSF or NonO were also confirmed using the yeast two-hybrid assay (data not shown). These data show that mPSP1, PSF, and p54nrb/NonO have the ability to form heterodimers.

Fig. 5

mPSP1 associates with PSF and p54nrb/NonO. NIH/3T3 cells were cotransfected with pFLAG-mPSP1 and either pHA-PSF (a) or pHA-p54nrb/NonO (b). The pluses and minuses represent plasmids included in the transfections. IP, immunoprecipitation; IB, immunoblotting

Fig. 5

mPSP1 associates with PSF and p54nrb/NonO. NIH/3T3 cells were cotransfected with pFLAG-mPSP1 and either pHA-PSF (a) or pHA-p54nrb/NonO (b). The pluses and minuses represent plasmids included in the transfections. IP, immunoprecipitation; IB, immunoblotting

Mouse PSP1 Protein Associates with Nuclear Matrix

Recently, it was demonstrated that matrin4, a nuclear matrix protein, is identical to PSF [20]. Immunohistochemical analysis showed that intracellular localization of mPSP1 was very similar to that of PSF, so we tested whether mPSP1 was also included in the nuclear matrix fraction. The nuclear matrix fraction was prepared from mouse testis by biochemical procedures described in Materials and Methods. Preparation of the nuclear matrix was evaluated by the existence of lamin B1 in the fraction (Fig. 6, lower panel). mPSP1-α was observed in cytoplasmic and nuclear matrix fractions (Fig. 6, upper panel), suggesting that a significant amount of mPSP1-α was tightly associated with the nuclear matrix. Mouse PSP1 and exogenously expressed GFP-mPSP1-α were solely localized in the nucleus. Therefore, cytoplasmic mPSP1 would have come from the soluble nuclear fraction, which was eluted from disrupted nuclei by the homogenization procedure.

Fig. 6

Existence of mPSP1 in nuclear matrix fraction. Nuclei isolated from mouse testis were treated with DNase I and extracted in 2.5 M ammonium sulfate and 2 M NaCl. Western blotting was performed using anti-mPSP1 and anti-Lamin B1. 1, Whole extract; 2, cytoplasmic extract; 3, 2.5 M ammonium sulfate fraction; 4, 2 M NaCl fraction; 5, nuclear matrix fraction. The signals in the upper and lower panels indicate mPSP1-α (58.7 kDa) and Lamin B1 (68.9 kDa), respectively

Fig. 6

Existence of mPSP1 in nuclear matrix fraction. Nuclei isolated from mouse testis were treated with DNase I and extracted in 2.5 M ammonium sulfate and 2 M NaCl. Western blotting was performed using anti-mPSP1 and anti-Lamin B1. 1, Whole extract; 2, cytoplasmic extract; 3, 2.5 M ammonium sulfate fraction; 4, 2 M NaCl fraction; 5, nuclear matrix fraction. The signals in the upper and lower panels indicate mPSP1-α (58.7 kDa) and Lamin B1 (68.9 kDa), respectively

Mouse PSP1 Binds to RNA

The RNA-binding ability of mPSP1 was confirmed by Northwestern analysis. The RNA binding domain of mPSP1 protein bound to poly(A), poly(G), and poly(U) (Fig. 7). These results verified the RNA-binding abilities of mPSP1, and furthermore suggested that mPSP1 might have some sequence preference on its RNA-binding property.

Fig. 7

RNA-binding properties of mPSP1 by in vitro RNA-binding assay. Purified recombinant mPSP1 was electrophoresed on 12% SDS-PAGE gels and transferred onto PVDF membranes. The membranes were incubated with [32P]-labeled RNA homopolymers in binding buffer containing 100 mM NaCl. After washing with binding buffer, the membranes were dried and exposed. Arrowhead indicates the signal of the mPSP1 protein complex and the RNA homopolymer

Fig. 7

RNA-binding properties of mPSP1 by in vitro RNA-binding assay. Purified recombinant mPSP1 was electrophoresed on 12% SDS-PAGE gels and transferred onto PVDF membranes. The membranes were incubated with [32P]-labeled RNA homopolymers in binding buffer containing 100 mM NaCl. After washing with binding buffer, the membranes were dried and exposed. Arrowhead indicates the signal of the mPSP1 protein complex and the RNA homopolymer

Discussion

Mouse PSP1 is the mouse ortholog of PSP1, a component of a novel nuclear body, paraspeckle, identified in HeLa cells by Fox et al. [19]. In this study, we show that mPSP1 was abundantly expressed in testis and kidney. As expected from the primary amino acid sequence, mPSP1 bound RNA and formed heterodimers with other members of the protein family, PSF and p54nrb/NonO. Moreover, we showed that mPSP1 is present in the nuclear matrix fraction, suggesting that mPSP1 may play an essential role in various aspects of nuclear events during spermatogenesis.

The deduced amino acid sequence of mPSP1 shows that mPSP1 has two RRM domains, followed by an HTH motif. The combination of these domains is shared with PSF and p54nrb/NonO. As described by Fox et al. [19], we also observed two isoforms, mPSP1-α and mPSP1-β, in mouse tissues. Similarly, PSF comprises short and long isoforms. The longer isoform of PSF possesses two nuclear localization signals (NLSs) in the C-terminal, and the shorter isoform lacks one NLS at the distal C-terminus. Deletion of the distal NLS led to a diffuse cytoplasmic localization [27]. Therefore, two NLSs are required for complete nuclear localization of PSF. We examined the subcellular localization of exogenously expressed mPSP1 isoforms in NIH/3T3 cells. Similar to the PSF isoforms, mPSP1-α is localized in the nucleus (punctate pattern), whereas mPSP1-β was distributed in both the nucleus (punctate pattern) and cytoplasm. This observation was also observed in HeLa cells (data not shown). However, Fox et al. [19] mentioned that localization of both isoforms of exogenously expressed human PSP1-α and PSP1-β in HeLa cells is nuclear. The inconsistency of the localization between human and mouse PSP1 isoforms does not simply result from difference of the cells used for the transfections. Preliminarily, we found that mPSP1 in the cytoplasm of the urinary tubule in kidney, in which mPSP1-β was exclusively expressed. This would support cytoplasmic localization of mPSP1-β. P54nrb/NonO is shown to have carbonic anhydrase activity [14]. The enzyme catalyzes interconversion of carbon dioxide and bicarbonate. The activity is crucial for pH regulation in urinary tubules of kidney, where mPSP1-β expresses. Mouse PSP1-β may also be involved in maintenance of homeostasis. Therefore, subcellular localization, and possibly function, of mPSP1 may be controlled by alternative splicing.

Proteins with DBHS domains are usually identified as monomeric forms. However, it is often observed that PSF and p54nrb/NonO are found in the same protein complex. In addition, the PSF-p54nrb/NonO heterodimer is able to enhance DNA topoisomerase I activity better than PSF alone [10]. Recently, Hrp65, a DBHS-containing protein in Chironomus tentans, interacts with itself and other proteins with DBHS domains, human PSF, and Drosophila NonA, through the DBHS domain [28]. PSP1 has been found in association with DBHS-containing proteins in a novel nuclear body, paraspeckle [19], and in a complex that binds the transactivation domain of androgen receptor [9]. Here we demonstrated that mPSP1 directly interacts with PSF and p54nrb/NonO. These observations suggest that these proteins may exist and function mainly as dimers via interactions between their DBHS domains. The finding of a new member of DBHS-containing proteins would increase the repertoire as a dimerization partner among DBHS-containing proteins. Therefore, three DBHS proteins, their dimerization, and their multiple isoforms may produce many combinations as functional units and expand the spectrum of their biological functions. This will account for the multifunctional properties in a variety of nuclear and cytoplasmic processes.

We demonstrated that mPSP1 is included in nuclear matrix fractions. Recently, PSF was shown to be identical to Matrin 4, a major nuclear matrix protein [2, 20]. The nuclear matrix has been operationally regarded as the insoluble structure remaining in the nucleus after a series of biochemical extraction steps [29, 30]. Although the existence and the importance of the nuclear matrix is controversial [31], a growing number of studies have proposed the idea that many nuclear events, such as DNA replication, transcription, and multiple steps of RNA processing, are accomplished on the nuclear matrix [32]. Nuclear matrix proteins can be divided into two classes; one is ubiquitously expressed, and the other is tissue-specific [33]. Commonly expressed nuclear matrix proteins might participate in basal events in the nuclei. On the other hand, tissue and stage-specific nuclear matrix proteins [34, 35] play roles not only in general nuclear events, but in regulation of tissue and stage-specific gene expression [36]. The expression of PSF and p54nrb/NonO [3, 37] is widespread in mouse tissues, but we showed that expression of mPSP1-α, a nuclear matrix protein, was highly tissue- and stage-specific. Therefore, we would propose that the mPSP1-α isoform has specific roles in mouse spermatogenesis.

The RRM is required for RNA and ssDNA recognition, and the HTH motif is responsible for binding to DNA [38, 39]. As expected, PSF and p54nrb/NonO bind not only RNA, but to ssDNA and dsDNA as well [1, 3, 40, 41]. By Northwestern analysis using RNA homopolymers, we demonstrated that mPSP1 bound RNA with little preference for poly(G), poly(U), and poly(A). We did not examine the DNA-binding, but PSP1 was identified in the complex of DNA transcriptional machinery, suggesting a DNA-binding activity of the protein. Therefore, in addition to other DBHS family members, mPSP1 may perform important roles in various nuclear events by binding to both DNA and RNA.

In spermatogenesis, germ cell-specific molecular mechanisms that differ from somatic cells may govern development. This includes germ cell-specific alternative splicing and posttranscriptional regulation, transcriptional silencing, and chromatin remodeling in the nucleus. Because mPSP1-α is abundantly expressed in the nuclei of specific stages of spermatogenesis, we expect that the protein plays a specific role in germ cells. PSP1 and p54nrb/NonO act as transcriptional coactivators of the human androgen receptor [9]. Recently, it was reported that the complex formation of PSF with VL30 RNA repressed steroidogenesis [42]. PSF and p54nrb/NonO were also identified as interacting proteins with the carboxyl-terminal domain of the largest subunit of eukaryotic RNA polymerase II, which plays an important role in promoting steps of pre-mRNA processing. Therefore, we speculate that mPSP1 functions to regulate transcription and early mRNA processing of genes in germ cells. Moreover, mPSP1 could be involved in chromatin remodeling and nuclear shaping in spermatogenesis. During spermatogenesis, chromatin-associated histones are replaced by basic transition proteins (TP1 and TP2), and condensation of chromatin and nuclear shaping occur. The timing of mPSP1 expression corresponds well with that of chromatin remodeling and a dramatic change in the nuclear environment. It is suggested that the proteins associated with the nuclear matrix are concerned with chromatin remodeling and nuclear shaping in spermatogenesis [4345]. So, as a nuclear matrix protein, mPSP1 may also contribute to chromatin remodeling through the DNA-binding activity in the nuclei of haploid cells.

Acknowledgment

The authors thank Miyuki Matsuda for providing samples.

References

1.
Patton
JG
,
Porro
EB
,
Galceran
J
,
Tempst
P
,
Nadal-Ginard
B
.
Cloning and characterization of PSF, a novel pre-mRNA splicing factor
.
Genes Dev
 
1993
7
:
393
406
2.
Nakayasu
H
,
Berezney
R
.
Nuclear matrins: identification of the major nuclear matrix proteins
.
Proc Natl Acad Sci U S A
 
1991
88
:
10312
10316
3.
Yang
YS
,
Hanke
JH
,
Carayannopoulos
L
,
Craft
CM
,
Capra
JD
,
Tucker
PW
.
NonO, a non-POU-domain-containing, octamer-binding protein, is the mammalian homolog of Drosophila nonAdiss
.
Mol Cell Biol
 
1993
13
:
5593
5603
4.
Basu
A
,
Dong
B
,
Krainer
AR
,
Howe
CC
.
The intracisternal A-particle proximal enhancer-binding protein activates transcription and is identical to the RNA- and DNA-binding protein p54nrb/NonO
.
Mol Cell Biol
 
1997
17
:
677
686
5.
Hallier
M
,
Tavitian
A
,
Moreau-Gachelin
F
.
The transcription factor Spi-1/PU.1 binds RNA and interferes with the RNA-binding protein p54nrb
.
J Biol Chem
 
1996
271
:
11177
11181
6.
Peng
R
,
Dye
BT
,
Perez
I
,
Barnard
DC
,
Thompson
AB
,
Patton
JG
.
PSF and p54nrb bind a conserved stem in U5 snRNA
.
RNA
 
2002
8
:
1334
1347
7.
Zhang
WW
,
Zhang
LX
,
Busch
RK
,
Farres
J
,
Busch
H
.
Purification and characterization of a DNA-binding heterodimer of 52 and 100 kDa from HeLa cells
.
Biochem J
 
1993
290
:Pt 1
267
272
8.
Mathur
M
,
Tucker
PW
,
Samuels
HH
.
PSF is a novel corepressor that mediates its effect through Sin3A and the DNA binding domain of nuclear hormone receptors
.
Mol Cell Biol
 
2001
21
:
2298
2311
9.
Ishitani
K
,
Yoshida
T
,
Kitagawa
H
,
Ohta
H
,
Nozawa
S
,
Kato
S
.
p54nrb acts as a transcriptional coactivator for activation function 1 of the human androgen receptor
.
Biochem Biophys Res Commun
 
2003
306
:
660
665
10.
Straub
T
,
Grue
P
,
Uhse
A
,
Lisby
M
,
Knudsen
BR
,
Tange
TO
,
Westergaard
O
,
Boege
F
.
The RNA-splicing factor PSF-p54 controls DNA-topoisomerase I activity by a direct interaction
.
J Biol Chem
 
1998
273
:
26261
26264
11.
Bertrand
P
,
Akhmedov
AT
,
Delacote
F
,
Durrbach
A
,
Lopez
BS
.
Human POMp75 is identified as the pro-oncoprotein TLS/FUS: both POMp75 and POMp100 DNA homologous pairing activities are associated to cell proliferation
.
Oncogene
 
1999
18
:
4515
4521
12.
Akhmedov
AT
,
Bertrand
P
,
Corteggiani
E
,
Lopez
BS
.
Characterization of two nuclear mammalian homologous DNA-pairing activities that do not require associated exonuclease activity
.
Proc Natl Acad Sci U S A
 
1995
92
:
1729
1733
13.
Akhmedov
AT
,
Lopez
BS
.
Human 100-kDa homologous DNA-pairing protein is the splicing factor PSF and promotes DNA strand invasion
.
Nucleic Acids Res
 
2000
28
:
3022
3030
14.
Karhumaa
P
,
Parkkila
S
,
Waheed
A
,
Parkkila
AK
,
Kaunisto
K
,
Tucker
PW
,
Huang
CJ
,
Sly
WS
,
Rajaniemi
H
.
Nuclear NonO/p54(nrb) protein is a nonclassical carbonic anhydrase
.
J Biol Chem
 
2000
275
:
16044
16049
15.
Shav-Tal
Y
,
Lee
BC
,
Bar-Haim
S
,
Schori
H
,
Zipori
D
.
Reorganization of nuclear factors during myeloid differentiation
.
J Cell Biochem
 
2001
81
:
379
392
16.
Shav-Tal
Y
,
Cohen
M
,
Lapter
S
,
Dye
B
,
Patton
JG
,
Vandekerckhove
J
,
Zipori
D
.
Nuclear relocalization of the pre-mRNA splicing factor PSF during apoptosis involves hyperphosphorylation, masking of antigenic epitopes, and changes in protein interactions
.
Mol Biol Cell
 
2001
12
:
2328
2340
17.
Andersen
JS
,
Lyon
CE
,
Fox
AH
,
Leung
AK
,
Lam
YW
,
Steen
H
,
Mann
M
,
Lamond
AI
.
Directed proteomic analysis of the human nucleolus
.
Curr Biol
 
2002
12
:
1
11
18.
Dreger
M
,
Bengtsson
L
,
Schoneberg
T
,
Otto
H
,
Hucho
F
.
Nuclear envelope proteomics: novel integral membrane proteins of the inner nuclear membrane
.
Proc Natl Acad Sci U S A
 
2001
98
:
11943
11948
19.
Fox
AH
,
Lam
YW
,
Leung
AK
,
Lyon
CE
,
Andersen
J
,
Mann
M
,
Lamond
AI
.
Paraspeckles: a novel nuclear domain
.
Curr Biol
 
2002
12
:
13
25
20.
Zhang
Z
,
Carmichael
GG
.
The fate of dsRNA in the nucleus: a p54(nrb)-containing complex mediates the nuclear retention of promiscuously A-to-I edited RNAs
.
Cell
 
2001
106
:
465
475
21.
Hirano
K
,
Erdodi
F
,
Patton
JG
,
Hartshorne
DJ
.
Interaction of protein phosphatase type 1 with a splicing factor
.
FEBS Lett
 
1996
389
:
191
194
22.
Otto
H
,
Dreger
M
,
Bengtsson
L
,
Hucho
F
.
Identification of tyrosine-phosphorylated proteins associated with the nuclear envelope
.
Eur J Biochem
 
2001
268
:
420
428
23.
Kurihara
Y
,
Tokuriki
M
,
Myojin
R
,
Hori
T
,
Kuroiwa
A
,
Matsuda
Y
,
Sakurai
T
,
Kimura
M
,
Hecht
NB
,
Uesugi
S
.
CPEB2, a novel putative translational regulator in mouse haploid germ cells
.
Biol Reprod
 
2003
69
:
261
268
24.
Kashiwabara
S
,
Zhuang
T
,
Yamagata
K
,
Noguchi
J
,
Fukamizu
A
,
Baba
T
.
Identification of a novel isoform of poly(A) polymerase, TPAP, specifically present in the cytoplasm of spermatogenic cells
.
Dev Biol
 
2000
228
:
106
115
25.
Gerner
C
,
Holzmann
K
,
Grimm
R
,
Sauermann
G
.
Similarity between nuclear matrix proteins of various cells revealed by an improved isolation method
.
J Cell Biochem
 
1998
71
:
363
374
26.
He
DC
,
Nickerson
JA
,
Penman
S
.
Core filaments of the nuclear matrix
.
J Cell Biol
 
1990
110
:
569
580
27.
Dye
BT
,
Patton
JG
.
An RNA recognition motif (RRM) is required for the localization of PTB-associated splicing factor (PSF) to subnuclear speckles
.
Exp Cell Res
 
2001
263
:
131
144
28.
Kiesler
E
,
Miralles
F
,
Ostlund Farrants
AK
,
Visa
N
.
The Hrp65 self-interaction is mediated by an evolutionarily conserved domain and is required for nuclear import of Hrp65 isoforms that lack a nuclear localization signal
.
J Cell Sci
 
2003
116
:Pt 19
3949
3956
29.
Berezney
R
,
Coffey
DS
.
Identification of a nuclear protein matrix
.
Biochem Biophys Res Commun
 
1974
60
:
1410
1417
30.
Nickerson
J
.
Experimental observations of a nuclear matrix
.
J Cell Sci
 
2001
114
:Pt 3
463
474
31.
Martelli
AM
,
Falcieri
E
,
Zweyer
M
,
Bortul
R
,
Tabellini
G
,
Cappellini
A
,
Cocco
L
,
Manzoli
L
.
The controversial nuclear matrix: a balanced point of view
.
Histol Histopathol
 
2002
17
:
1193
1205
32.
Pederson
T
.
Thinking about a nuclear matrix
.
J Mol Biol
 
1998
277
:
147
159
33.
Korosec
T
,
Gerner
C
,
Sauermann
G
.
Common nuclear matrix proteins in rat tissues
.
Electrophoresis
 
1997
18
:
2109
2115
34.
Lakshmanan
Y
,
Subong
EN
,
Partin
AW
.
Differential nuclear matrix protein expression in prostate cancers: correlation with pathologic stage
.
J Urol
 
1998
159
:
1354
1358
35.
Dworetzky
SI
,
Fey
EG
,
Penman
S
,
Lian
JB
,
Stein
JL
,
Stein
GS
.
Progressive changes in the protein composition of the nuclear matrix during rat osteoblast differentiation
.
Proc Natl Acad Sci U S A
 
1990
87
:
4605
4609
36.
Getzenberg
RH
.
Nuclear matrix and the regulation of gene expression: tissue specificity
.
J Cell Biochem
 
1994
55
:
22
31
37.
Shav-Tal
Y
,
Lee
B
,
Bar-Haim
S
,
Vandekerckhove
J
,
Zipori
D
.
Enhanced proteolysis of pre-mRNA splicing factors in myeloid cells
.
Exp Hematol
 
2000
28
:
1029
1038
38.
Johnston
SD
,
Lew
JE
,
Berman
J
.
Gbp1p, a protein with RNA recognition motifs, binds single-stranded telomeric DNA and changes its binding specificity upon dimerization
.
Mol Cell Biol
 
1999
19
:
923
933
39.
Wintjens
R
,
Rooman
M
.
Structural classification of HTH DNA-binding domains and protein-DNA interaction modes
.
J Mol Biol
 
1996
262
:
294
313
40.
Shav-Tal
Y
,
Zipori
D
.
PSF and p54(nrb)/NonO—multi-functional nuclear proteins
.
FEBS Lett
 
2002
531
:
109
14
41.
Urban
RJ
,
Bodenburg
Y
,
Kurosky
A
,
Wood
TG
,
Gasic
S
.
Polypyrimidine tract-binding protein-associated splicing factor is a negative regulator of transcriptional activity of the porcine p450scc insulin-like growth factor response element
.
Mol Endocrinol
 
2000
14
:
774
782
42.
Song
X
,
Sui
A
,
Garen
A
.
Binding of mouse VL30 retrotransposon RNA to PSF protein induces genes repressed by PSF: effects on steroidogenesis and oncogenesis
.
Proc Natl Acad Sci U S A
 
2004
101
:
621
626
43.
Chen
JL
,
Guo
SH
,
Gao
FH
.
Nuclear matrix in developing rat spermatogenic cells
.
Mol Reprod Dev
 
2001
59
:
314
321
44.
Chen
JL
,
Longo
FJ
.
Expression and localization of DNA topoisomerase II during rat spermatogenesis
.
Mol Reprod Dev
 
1996
45
:
61
71
45.
Wu
JY
,
Ribar
TJ
,
Cummings
DE
,
Burton
KA
,
McKnight
GS
,
Means
AR
.
Spermiogenesis and exchange of basic nuclear proteins are impaired in male germ cells lacking Camk4
.
Nat Genet
 
2000
25
:
448
452

Author notes

1
This work was supported by grants from the Ministry of Education, Science, Sports, and Culture of Japan; the Kanagawa Academy of Science and Technology Research; and the Kihara Memorial Yokohama Foundation. R.M. and S.K. contributed equally to this work.