Pituitary Tumor-transforming Gene Protein Associates with Ribosomal Protein S10 and a Novel Human Homologue of DnaJ in Testicular Cells*

Pituitary tumor-transforming gene (PTTG) is a recently characterized proto-oncogene that is expressed specifically in adult testis. In this study, we have used in situhybridization and developmental Northern blot assays to demonstrate that PTTG mRNA is expressed stage-specifically in spermatocytes and spermatids during rat spermatogenic cycle. We have used the yeast two-hybrid system to identify proteins that interact with PTTG in testicular cells. Two positive clones were characterized. One of the clones is the ribosomal protein S10, the other encodes a novel human DnaJ homologue designated HSJ2. Northern blot analysis showed that testis contains higher levels of HSJ2 mRNA than other tissues examined, and the expression pattern of HSJ2 mRNA in postnatal rat testis is similar to PTTG. S10 mRNA levels do not vary remarkably among different tissues and remains unchanged during testicular germ cell differentiation. In vitro binding assays demonstrated that both S10 and HSJ2 bind to PTTG specifically and that PTTG can be co-immunoprecipitated with S10 and HSJ2 from transfected cells. Moreover, the binding sites for both proteins were located within the C-terminal 75 amino acids of the PTTG protein. These results suggest that PTTG may play a role in spermatogenesis.

Pituitary tumor-transforming gene (PTTG) is a recently characterized proto-oncogene that is expressed specifically in adult testis. In this study, we have used in situ hybridization and developmental Northern blot assays to demonstrate that PTTG mRNA is expressed stage-specifically in spermatocytes and spermatids during rat spermatogenic cycle. We have used the yeast two-hybrid system to identify proteins that interact with PTTG in testicular cells. Two positive clones were characterized. One of the clones is the ribosomal protein S10, the other encodes a novel human DnaJ homologue designated HSJ2. Northern blot analysis showed that testis contains higher levels of HSJ2 mRNA than other tissues examined, and the expression pattern of HSJ2 mRNA in postnatal rat testis is similar to PTTG. S10 mRNA levels do not vary remarkably among different tissues and remains unchanged during testicular germ cell differentiation. In vitro binding assays demonstrated that both S10 and HSJ2 bind to PTTG specifically and that PTTG can be co-immunoprecipitated with S10 and HSJ2 from transfected cells. Moreover, the binding sites for both proteins were located within the C-terminal 75 amino acids of the PTTG protein. These results suggest that PTTG may play a role in spermatogenesis.
Pituitary tumor-transforming gene (PTTG) 1 was originally isolated from a rat pituitary tumor cell line by RNA differential display (1). Overexpression of PTTG resulted in cell transformation in fibroblasts and tumor formation in nude mice (1). PTTG mRNA was found to be highly expressed in a variety of tumor cell lines, suggesting a possible role of PTTG in tumorigenesis outside the pituitary gland. 2 Testis is one of the few normal adult tissues where a shorter form of PTTG mRNA is expressed (1). The genomic structure of the rat PTTG has recently been characterized (2). A transcriptional enhancer element was identified in the PTTG 5Ј-flanking region that was required for transcriptional activation of PTTG in a testicular germ cell line, GC2 (2). This enhancer element also contains the binding sites for multiple nuclear proteins, two of which are germ cell-specific (2).
Although PTTG mRNA is expressed in various testicular cell lines, including Sertoli, Leydig, and germ cells (2), the expression pattern of PTTG in normal testis is unknown. In this study, we have used in situ hybridization and developmental Northern blot assays to determine the cell type and the differentiation stage in which PTTG mRNA is expressed during the spermatogenic cycle. Our results showed that PTTG mRNA is expressed predominantly in spermatocytes and spermatids in a stage-specific manner, suggesting that PTTG may play a role in spermatogenesis. Because the amino acid sequence of PTTG does not reveal any identifiable structural or functional motifs, it is difficult to deduce its possible functions. To further our understanding of the role of PTTG in spermatogenesis, an in vivo strategy was employed to identify proteins capable of physically associating with PTTG. The yeast two-hybrid system developed by Fields and Song (3) was used to provide a physiological environment in which to detect potential interactions involving the PTTG protein. This system has been used to screen cDNA libraries for clones encoding proteins capable of binding a protein of interest (4 -6). Using this approach, we have identified several cDNA clones encoding proteins that interact with PTTG. One clone isolated in this screen encodes a novel human homologue of the bacterial heat-shock protein DnaJ (designated HSJ2). Eukaryotic DnaJ proteins have been implicated in various cellular processes, including correct protein folding (7-9), intracellular vesicle traffic (10,11), activity control of regulatory proteins (12), promotion of translation initiation (13), and tumor suppression (14). Another cDNA clone isolated encodes a previously characterized protein, ribosomal protein S10. The cloning and characterization of HSJ2 and S10 is described, as well as their association with PTTG and the possible functional implications of this interaction.

MATERIALS AND METHODS
In Situ Hybridization-Paraffin sections of rat testis were obtained from Novagen. 35 S-Labeled antisense and sense riboprobes were synthesized with T7 and T3 polymerase, respectively, and hydrolyzed in mild alkali to a length of 100 -200 nucleotides. The slides were hydrated, post-fixed in paraformaldehyde followed by treatment with proteinase K and acetic anhydride. Slides were hybridized at 50°C overnight in a buffer containing 0.3 M NaCl and 60% formamide and a probe concentration of 10 7 cpm/slide. Washing was performed as follows. Wash 1: 2ϫ SSC (1ϫ SSC ϭ 0.15 M NaCl and 0.015 M sodium citrate) at 50°C for 30 min. RNase A digestion: 20 g/ml RNaseA in 2ϫ SSC at 37°C for 30 min. Wash 2: 2ϫ SSC, 50% formamide at 50°C for 30 min. Wash 3: 0.2ϫ SSC, 0.05% sodium pyrophosphate at 55°C for 30 min. Slides were exposed to ␤-Max hyperfilm for 2 days and were then * This work was supported by National Institutes of Health Grant DK-02346 (to L. Pei.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AF080569.
The polymerase chain reaction product was cloned into TA vector (Invitrogen), and the insert was excised from EcoRI site and cloned in frame into pAS2 vector (CLONTECH). The resulting plasmid was designated pAS2-PTTG.
␤-Galactosidase Assays-Transformants grown on ϪLeu/ϪTrp/ ϪHis/ϩ3AT SD plates were screened further by colony-lift filter assay. For each plate of transformants, a Whatman No. 5 filter was presoaked in 2.5 ml of Z buffer/5-bromo-4-chloro-3-indolyl b-D-galactopyranoside (16.1 g/liter Na 2 HPO 4 ⅐7 H 2 O, 5.5 g/liter NaH 2 PO4⅐H 2 O, 0.75 g/liter KCl, 0.246 g/liter MgSO 4 ⅐7 H 2 O, 0.34 mg/ml 5-bromo-4-chloro-3-indolyl b-Dgalactopyranoside, 38 mM ␤-mercaptoethanol ). A dry filter was placed over the surface of the plate of colonies, and after the filter was oriented to the plate, it was lifted off and transferred to a pool of liquid nitrogen for 10 s. The filter was thawed at room temperature and placed on the presoaked filter. The filters were incubated at 30°C for 30 min to 4 h, and the blue colonies were identified.
The blue colonies were characterized further by quantitative liquid ␤-galactosidase assays. A single colony was used to inoculate a 5-ml liquid medium, and the culture was incubated at 30°C overnight. 2 ml of the overnight culture was inoculated into 8 ml yeast/peptone/glucose medium and grown at 30°C mid-log phase. The exact A 600 was recorded. 1.5 ml of the culture was centrifuged, and the cell pellet was resuspended in 300 l of Z buffer. 100 l of the cell suspension was transferred to a fresh tube, frozen in liquid nitrogen, and thawed at 37°C. 0.7 ml of Z buffer was added to the tube. After recording the time, 0.16 ml of O-nitrophenyl-␤-D-galactopyranoside was added and the tube was incubated at 30°C. After the yellow color developed, 0.4 ml of 1 M Na 2 CO 3 was added to the reaction, and the elapsed time was recorded. After removing cell debris by centrifugation, the A 420 was measured. The ␤-galactosidase units were calculated as the following: ␤-gal units ϭ 1,000 ϫ A 420 /(t ϫ V ϫ A 600 ), where t ϭ elapsed time (in min) of incubation, V ϭ 0.1 ml ϫ concentration factor, and A 600 is of 1 ml of culture.
Characterization of Positive Clones-Yeast plasmid DNA was isolated using Yeastmaker yeast plasmid isolation kit (CLONTECH) according to manufacturer's protocol. Plasmid isolated from yeast was used to transform HB101 Escherichia coli cell using electroporation. Cells were grow on M9 agar plates containing 50 g/ml ampicillin, 40 mg/ml proline, and 1 mM thiamine hydrochloride. Plasmid DNA was isolated from Leu ϩ , Amp r HB101 transformants using a standard plasmid mini-prep procedure. Positive clones were divided into groups based on restriction enzyme digestion patterns. The insert of each independent positive clone was sequenced using Gal4 AD sequencing primer (CLONTECH) and Sequenase (U. S. Biochemical Corp.). A homology search was performed using BLAST system. The inserts of several positive clones containing cDNAs of interest were excised from pAC2 vector by BglII digest and were cloned at the BamHI site into the pBCMV eukaryotic expression vector (Stratagene) for further analysis.
In Vitro Transcription and Translation-The above-characterized cDNAs were transcribed from T3 promoter and translated in reticulocyte lysate using TNT-coupled reticulocyte lysate system (Promega). A typical reaction contains 25 l of rabbit reticulocyte lysate, 2 l of reaction buffer, 20 M amino acid mixture, 1 g of DNA template, 40 units of ribonuclease inhibitor, 10 units of T7 RNA polymerase, and 1 l of Transcent TM Biotin-lysyl-tRNA (Promega) in a total volume of 50 l. The reactions were carried out at 30°C for 1 h.

Construction of Glutathione S-Transferase (GST)-PTTG Fusion Protein and in Vitro
Binding-To construct GST-PTTG, plasmid TA-PTTG was digested with EcoRI, and the insert was subcloned in frame at the EcoRI site of pGEX-4T-1 (Amersham Pharmacia Biotech). Expression of GST fusion protein was induced with 0.5 mM isopropyl-␤-thiogalactopyranoside at 37°C for 90 min. Cells were centrifuged, and the resulting pellet was resuspended in a sonication buffer containing 150 mM KCl, 40 mM HEPES, pH 7.9, 0.5 mM EDTA, 5 mM MgCl2, 1.0 mM dithiothreitol, 0.05% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, and 1 g/ml aprotinin. Cells were lysed by sonication. Cell debris was removed by centrifugation, and the supernatant was added to Sepharose 4B beads (Amersham Pharmacia Biotech).
The in vitro binding assay was performed as follows. 20 l of the in vitro translated protein was incubated with beads containing 200 ng of GST or GST-PTTG fusion protein in the sonication buffer for 90 min at 4°C. Complexes were washed extensively with the sonication buffer, boiled in loading buffer, and separated on 10% SDS-polyacrylamide gels. Gels were transferred to nylon membranes and blocked by incubation with Tris-buffered saline containing 0.5% Tween 20 (TBSB). The membranes were incubated with Streptavidin-horseradish peroxidase conjugate in TBST for 45 min, washed 4 times with TBST and 3 times with TBS. The membranes were then incubated with the chemiluminescent substrate mixture for 1 min and exposed to Kodak x-ray film for 20 min.
Immunoprecipitation-S10 and HSJ2 were subcloned into pCMV eukaryotic expression vector with HA tag. pCMV-S10 and pCMV-HSJ2 were transiently transfected into COS-7 cells either alone or together with pCMV-PTTG. 48 h post transfection, whole cell extracts were made from transfected cells. Cells were lysed in NET-N buffer containing 50 mM Tris, pH 7.6, 5 mM EDTA, 0.3 M NaCl, 1 mM dithiothreitol, 0.1% Nonidet P-40, 0.2 mM phenylmethylsulfonyl fluoride, and 1 g/ml each of leupeptin, pepstatin, and aprotinin, and centrifuged at 30,000 ϫ g for 1 h at 4°C. The supernatant was stored at Ϫ80°C. 1 mg of cell extract was incubated with 2 g of anti-HA (clone 12CA5, Borhinger Mannheim) for 2 h at 4°C. Immune complexes were then precipitated with protein A/G-Sepharose at 4°C for 1 h. The precipitate was collected by centrifugation and washed extensively in NET-N buffer. Protein complexes were separated by SDS-polyacrylamide gel electrophoresis and Western blotted using 1:5000 dilution of the anti-PTTG polyclonal antibody.
Mutagenesis-The N-terminal deletion mutant of pAS2-PTTG were generated using ExSite polymerase chain reaction-based site-directed mutagenesis kit (Stratagene) following manufacturer's instructions. The upper strand primers were The lower strand primer was: 5Ј-CATCCTGAAGCCGAATC-CGGCCTCCATGGCAT-5Ј.
Northern Blot Analysis-The human RNA Master Blot and Multitissue Blot were purchased from CLONTECH. Total RNA was isolated from the testes of rats of various ages using RNAeasy RNA isolation kit (Qiagen). The full-length cDNAs of PTTG, S10, and HSJ2 were labeled with [ 32 P]dCTP using Prime-It II Random Primer labeling kit (Stratagene). Hybridization was performed in ExpressHyb hybridization solution (CLONTECH) at 65°C overnight for RNA Master Blot and total RNA blots or for 1 h for Multi-tissue blot. The membrane was washed three times in 2ϫ SSC and 0.05% SDS at 65°C, followed by two washes in 0.1ϫ SSC and 0.1% SDS at 55°C. Each wash was 20 min. The membrane was exposed to x-ray film for 3 to 24 h. It was then stripped by boiling for 10 min in 0.5% SDS and was hybridized to the second probe as described above.
Southern Blotting-DNA was extracted from human lymphocytes. Cells were incubated in lysis buffer (100 mM NaCl, 25 mM EDTA, 1% SDS, and 100 g/ml proteinase K) at 50°C overnight. Protein contaminants were removed by phenol/chloroform extractions, and DNA was precipitated with ethanol and resuspended in Tris-EDTA. 10 g of DNA was digested with either BamHI, EcoRI, HindIII, or PstI at 37°C overnight. The DNA was separated on 0.8% agarose gel. The gel was denatured and neutralized, and the DNA was transferred to nylon membrane. The membrane was probed with the HSJ2 cDNA using standard Southern blot hybridization and washing conditions. The membrane was exposed to x-ray film for 48 h before developing.

Cellular Localization of PTTG mRNA in Testis-Because
testis is composed of multiple cell types, including somatic Sertoli and Leydig cells, as well as proliferating and differentiating germ cells, it is important to determine in what cell type and in which differentiation stage during the spermotogenic cycle PTTG is expressed. In situ hybridization was performed on paraffin sections of adult rat testis using 35 S-labeled riboprobe. As shown in Fig. 1A, when the antisense PTTG probe was hybridized to adult testis sections, a whole range of signal strengths could be seen between tubule cross-sections, indicative of spermatogenic stage-specific expression. Fig. 1B shows two cross-sections probed with the sense probe. No hybridization signal was detected. Fig. 1C shows a group of seminiferous tubules at various stages of the cycle at higher magnification. Identification of the cycle of the seminiferous epithelium was determined by parameters such as the number of layers of spermatids, their position in the epithelium, and their nuclear shape. Fig. 1D compares two tubules at early (stage 3 or 4, left) or late (stage 12 or 13, right) stage of the cycle. The tubule in the later stage expressed PTTG mRNA to much higher levels than the tubules at the early stage (Fig. 1D). The hybridization signals were strongest over the cell layers representing spermatocytes and spermatids (Fig. 1D). Fig. 1E shows a stage 14 tubule hybridized to PTTG antisense probe. Strong hybridization signals were again detected in spermatocytes and spermatids. In contrast, there was no hybridization signal to the sense probe in a tubule of the same differentiation stage (Fig. 1F).
To confirm these results, developmental Northern blot analysis was performed using testis RNA from rat of various ages. As shown in Fig. 2, PTTG mRNA was barely detectable in the testis of 7-day-old rat. At this time, testis contained somatic cells and spermatogonia cells only (17). By day 14, when leptotene spermatocytes appeared (17), low level PTTG expression was detected. Increasing amount of PTTG mRNA was detected on day 21 when zygotene spermatocytes are present (17). By day 28, PTTG mRMA levels increased dramatically and reached the expression levels in adult animals. At this stage, both pachytene spermatocytes and spermatids are present in rat testis. These results indicate that PTTG mRNA is expressed in both spermatocytes and spermatids.
Screening PTTG Interactive Proteins Using Yeast Two-hybrid System-Stage and cell type-specific expression of PTTG suggests that it may play a role in spermatogenesis. Because the amino acid sequence of PTTG does not provide any functional information, we sought to identify cDNAs that encode proteins able to interact with PTTG using a yeast two-hybrid screening strategy. The entire coding region of the rat PTTG was cloned into a Gal4 DNA binding domain vector pAS2 (18). This hybrid plasmid and plasmid from a human testis cDNA library constructed in the activation domain vector were cotransformed into the yeast host strand CG-1945. The transformants were plated on minimal medium lacking Leu, Trp, and His to select for those containing both types of plasmids (i.e. Leu ϩ , Trp ϩ ) and that also express interacting hybrid proteins (His ϩ ). Primary His ϩ transformants were then tested for expression of the second reporter gene, LacZ, using a filter assay for ␤-galactosidase activity (19).
Approximately 10 6 transformants were placed under selection. Of the transformants plated on SD/ϪHis/ϪLeu/ϪTrp/ϩ5  A and B show the entire sections of testis hybridized to antisense (A) or sense (B) probe, respectively. The slides were exposed to ␤-Max hyperfilm for 2 days. Hybridization signals appear as dark grain. Panel C, low power bright-field view of testis cross-section. Leydig cells are indicated by a bar. The hybridization signal is visible as dark grains. Panel D, high magnification, brightfield view of two tubules at earlier (left) or later (right) stage of the spermatogenic cycle for comparison of PTTG mRNA expression levels. Panel E, high magnification, bright-field view of a stage 14 tubule; strong hybridization signals over spermatocytes and spermatids appear as dark grains. Panel F, high magnification, bright-field view of a stage 14 tubule, probe with sense control. mM 3AT medium, about 2000 grew into colonies within 3 to 5 days. Seventy of the His ϩ colonies were also blue when screened for their ability to produce ␤-galactosidase using the filter lift assay. These His ϩ blue colonies were considered positive in the initial screen and were used for additional studies.
To further test whether the phenotype observed in the original screen was reproducible and dependent on the pAS2-PTTG hybrid, plasmid DNA from each of the seventy clones was used to transform CG-1945 either alone or with pAS2-PTTG. Transformants were assayed for ␤-gal activity, and those showing activity only in the presence of pAS2-PTTG were considered positive. Of the 70 initial isolates, 10 of the recovered plasmids induced the expression of lacZ only in the presence of pAS2-PTTG, and their relative ␤-activity was shown in Table I.
Characterization of Positive Clones Dependent on PTTG Hybrid Expression-Sequence analysis of the 10 positive clones revealed that eight clones showed no significant homology to known protein sequences reported to GenBank. One cDNA insert (y9) encoded a protein with a predicted molecular mass of 19 kDa and was identical to human S10 ribosomal protein (20). Another positive clone (y3) contained an open reading frame of 217 amino acids (Fig. 3A) and was 62 and 47% identical to MSJ-1 (mouse DnaJ) (21) and HSJ1 (homosapiens DnaJ 1) (22), respectively. The highest identity shared by the three proteins was at the N termini, where the 82 amino acids were 70% identical (Fig. 3B). This similarity was shared by several other DnaJ proteins, including the E. coli DnaJ (23,24) and yeast, homologous of DnaJ, SCJ1(25), YDJ1 (26), and SIS1 (27). The similarity between y3 and other DnaJ proteins suggested that y3 is a new member of the human DnaJ family of proteins, and therefore was designated HSJ2 (homosapiens DnaJ 2).
Tissue Distribution of S10 and HSJ2-A human RNA master blot containing mRNA from 50 human tissues was probed with the S10 or HSJ2 cDNA probe. Both S10 and HSJ2 mRNAs were detected in all the tissues examined, although the expression level of HSJ2 mRNA varies among different tissues (data not shown). To determine the transcript size of S10 and HSJ2, Northern blot analysis was performed on selected tissues. The HSJ2 cDNA detected two bands of about 2.8 and 1.7 kb (Fig.  4A). The highest level of expression was detected in testis for the 1.7-kb transcript (normalized to loading control, Fig. 4C). Other tissues, including spleen, intestine, ovary, thymus, prostate, and leukocytes showed much lower level of HSJ2 mRNA expression (Fig. 4A). The S10 probe detected a single transcript of about 600 base pairs, and the expression levels of this transcript were similar in the tissues examined (Fig. 4B).
Developmental Regulation of HSJ2 and S10 Expression-To determine whether HSJ2 and S10 expression undergo developmental regulation, we have used postnatal rat testis as a model. RNAs were isolated from testes of postnatal rats of various ages and used in Northern blot analysis. As shown in Fig. 5A, very low level HSJ2 mRNA was detected at day 7. Increasing levels of HSJ2 mRNA were detected on day 14 and 21. By day 28, there was a more than 10-fold increase in HSJ2 expression (Fig. 5A) (after normalizing to actin control, Fig.  5C). There was no further increase in HSJ2 mRNA levels in adult rat testis (Fig. 5A). S10 mRNA, on the other hand, was expressed to similar levels from postnatal day 7 to adult (Fig.  5B). These results indicate that although HSJ2 expression undergoes developmental regulation, S10 is constitutively FIG. 3. Amino acid sequence of  HSJ2. A, alignment of HSJ2 to MSJ1, HSJ1, and SIS1. B, amino acid sequence of HSJ2 is aligned to amino acids of MSJ1, HSJ1, and SIS1. Identical amino acid sequences to HSJ2 are highlighted.

expressed.
Genomic Analysis of HSJ2-As an initial step to characterize HSJ2 gene structure, human genomic DNA analysis was performed. As shown in Fig. 6, the HSJ2 probe detected two bands in human genomic DNA digested with BamHI, EcoRI, HindIII, and PstI. Because these enzymes do not cut within the cDNA, these results suggest that either the sites for these enzymes are present within the introns of the gene or there are two different HSJ2 genes if the enzymes do not cut within the genes.
Both S10 Ribosomal Protein and HSJ2 Bind PTTG in Vitro-To confirm and extend the protein interaction data obtained in yeast, PTTG was expressed as a GST fusion protein in E. coli (28). To test the ability of S10 and HSJ2 proteins to bind PTTG in vitro, S10 and HSJ2 was in vitro transcribed and translated in the presence of Transcent TM Biotin-lysyl-tRNA (Promega). After translation, GST-PTTG immobilized on glutathione-Sepharose was added. The beads were washed extensively, and the bound proteins were analyzed by SDS-polyacrylamide gel electrophoresis. After electroblotting, the bound, biotinylated protein was visualized by binding to streptavidine-horseradish peroxidase, followed by chemiluminescent detection. As shown in Fig. 7, both S10 and HSJ2 were retained on the glutathione beads when they were incubated with GST-PTTG, whereas neither protein was retained when incubated with GST alone. These results indicated that S10 and HSJ2 protein were retained on the matrix through the PTTG segment of the fusion protein.
Both S10 and HSJ2 Proteins Can Be Coimmunoprecipitated with PTTG Protein-To further characterize the association of PTTG protein with S10 and HSJ2 proteins, we examined whether PTTG protein could be coimmunoprecipitated from transfected cells. Cos-7 cells were transiently transfected with HA-tagged HSJ2 and S10 expression vectors, either alone or together with PTTG expression plasmid. Cells were lysed, and the resulting extracts were immunoprecipitated with anti-HA monoclonal antibody (29). The immunoprecipitates were then resolved by SDS-polyacrylamide gel electrophoresis, and the gel was immunoblotted and probed with an antibody specific for the N-terminal 17 amino acids of PTTG (1). A band corresponding to PTTG was seen only in cells that were cotransfected with both S10 and PTTG or HSJ2 and PTTG expression vectors (Fig. 8). These results indicated that PTTG could form a complex with either S10 or HSJ2.
The C Terminus of PTTG Protein Is Required for Binding of S10 and HSJ2-To determine the region of PTTG protein involved in interaction with S10 and HSJ2, sequential N-terminal deletion mutants (Fig. 9A) of PTTG were subcloned into pAS2 and used to co-transform yeast CG-1945 with the Gal4 activation domain-S10 or HSJ2 fusion plasmid. The resulting transformants were then assayed for ␤-gal activity. As shown in Fig. 9B, transformants resulting from co-transformation of Gal4-S10 or Gal4-HSJ2 with PTTG mutants containing deletions up to 123 N-terminal amino acids (i.e. mutant 1 to 4) grew on SD/ϪTrp/ϪLeu/ϪHis/ϩ3AT plates and showed ␤-gal activity similar to transformants of the wild type PTTG. When the N-terminal deletion was made to amino acid 153 (mutant 5), however, there was no growth on SD/ϪTrp/ϪLeu/ϪHis/ϩ3AT plates, suggesting that this PTTG mutant was unable to bind to S10 and HSJ2. These results suggested that both S10 and HSJ2 bind to PTTG within the 75 amino acids at the C terminus and that amino acid residues 124 -153 play an essential role for PTTG interaction with these proteins. DISCUSSION Previous studies indicated that PTTG is specifically expressed in adult rat testis (1). In this study, the precise cellular location of PTTG mRNA has been determined. Our results show that PTTG messenger is predominantly expressed in spermatocytes and spermatids and in a stage-specific manner during the rat spermatogenic cycle. This expression pattern of PTTG is similar to that of proto-oncogene c-kit, in that both transcripts are present in germ cells. However, expression of c-kit mRNA starts in spermatogonia and continues into spermatocytes (30). c-kit is known to play a role in supporting spermatogonia proliferation and survival. Spermatogonia proliferate to replace themselves and differentiate to give rise to primary spermatocytes in which DNA synthesis is carried out to double chromosomes, after which they enter the first meiotic prophase. Expression of PTTG mRNA in nonproliferating spermatocytes and spermatids suggests that it may play a role in survival and differentiation of these germ cells.
Because the primary structure of PTTG protein does not provide functional information, we have used the yeast twohybrid system to identify proteins that interact with PTTG in testicular cells. We have used the rat PTTG to screen a human library, because the human PTTG cDNA was not available. Because of the species difference, it is possible that PTTGinteractive proteins that do not share high homology between human and rat may not be detected using this screen. However, using this method, we identified and characterized two proteins capable of interacting with PTTG. We demonstrated that both proteins were able to bind to PTTG in vitro and in vivo and that the C-terminal 75 amino acids of PTTG protein were required for this interaction.
One of the cDNA clones isolated from this screen encodes a novel human homologue of the bacterial heat-shock protein DnaJ (HSJ2). HSJ2 shares highest sequence similarity with the previously characterized MSJ-1 (21) and HSJ1 protein (22). Like these proteins, the similarity between HSJ2 and E. coli DnaJ is restricted to the N-terminal region of the molecule. MSJ-1 was shown to have two transcripts of 1.2 and 1 kb that are specifically expressed in testis (21). HSJ1 has two alternatively spliced transcripts of 2 and 3 kb, which are preferentially expressed in neurons (22). In contrast, HSJ2 mRNA is more widely expressed, although testis shows the highest level of expression. They are also two HSJ2 transcripts (1.7 and 2.8 kb) that may be the result of alternative splicing, use of alternative promoters, or polyadenylation sites. Like MSJ-1, HSJ2 expression undergoes developmental regulation in testis. However, although MSJ-1 expression is restricted in spermatids (21), HSJ2 is expressed in germ cells in the earlier stages of differentiation pathway as well as in spermatids. The developmental expression pattern of HSJ2 in postnatal rat is similar to PTTG mRNA expression profile. Both transcripts show very low lev- FIG. 7. PTTG binds to HSJ2 and S10 in vitro: GST pull-down assay. HSJ2 and S10 were transcribed and translated in vitro and labeled with Transcent TM Biotin-lysyl-tRNA. The proteins were allowed to bind to either GST alone or GST-PTTG immobilized on Sepharose 4B beads. The bound proteins were analyzed on SDS-polyacrylamide gels in the lanes indicated, blotted, and visualized by chemiluminescent detection.
FIG. 8. PTTG associates with HSJ2 and S10 proteins in the cell: immunoprecipitation assay. Cos7 cells were transfected with the indicated plasmids. 48 h post-transfection, whole cell lysates were immunoprecipitated with anti-HA monoclonal antibody. The immuno complexes were separated by SDS-polyacrylamide gel electrophoresis and blotted onto a nylon membrane. The blot was probed with anti-PTTG antibody and visualized by chemiluminescent. els expression on postnatal day 7, with increasing levels of expression on days 14 and 21 and dramatic increase on day 28. It is likely that HSJ2 mRNA is expressed in similar germs cells where PTTG messenger is found. Future in situ hybridization studies of HSJ2 in rat testis will confirm these results.
HSJ2 also shows high sequence similarity (59%) to the yeast DnaJ homologue, SIS1 protein (27). The SIS1 gene is essential for viability. In the absence of normal SIS1 function, the polysome levels decrease, and 80S ribosomes accumulate to high levels (13). It is believed that SIS1 functions to dissociate 80S ribosome into 40S and 60S subunits. SIS1 was shown to be associated with 40S ribosomal subunits and the smaller polysomes (13). This evidence indicates that SIS1 is required for the normal initiation of translation (13).
Another protein that interacts with PTTG protein is the human ribosomal large subunit protein S10 (20). The mammalian ribosomal is a complex structure assembled from approximately 80 proteins and four ribosomal RNA molecules (31). The ribosomal proteins are thought to facilitate the folding and maintenance of an optimal configuration of the rRNAs (32), perhaps in this way conferring speed and accuracy on protein synthesis. Recently, the involvement of the ribosomal proteins in cell proliferation (33)(34)(35)(36)(37), differentiation (38), and apoptosis (39) has been demonstrated. It has been show that S10 protein can be cross-linked to eukaryotic initiation factor 3 (40), an observation suggesting that S10 protein forms part of the domain involved in binding of the initiation factor to the 40S subunit at the start of the translation (40). The binding of PTTG protein to S10 protein suggests that PTTG protein may associate with ribosomes through this interaction.
What is the significance of the association of PTTG with HSJ2 and S10 proteins? Both S10 and SIS1 proteins have been shown to be associated with 40S ribosomal subunit and are required for the initiation of translation (13,40). Although the association of HSJ2 with ribosome is unknown at the present, because DnaJ has been shown to mediate the dissociation of several protein complexes, it is possible that HSJ2 might be able to mediate the dissociation of a specific protein complex of the translation. Interestingly, PTTG also interacts with S10, a ribosome protein. One possibility is that PTTG is targeted to the ribosome through interaction with S10 protein, and association with HSJ2 results in dissociation of PTTG-S10 complex and separates it from the ribosome. These hypotheses will be tested in future studies.
In summary, we demonstrated in this study that PTTG is expressed in a stage-specific manner in spermatocytes and spermatids during the spermatogenic cycle. Through interaction with ribosomal protein S10 and a novel human homologue of DnaJ protein, PTTG may play a role in male germ cell differentiation.