PKC (cid:1) II, a New Isoform of Protein Kinase C Specifically Expressed in the Seminiferous Tubules of Mouse Testis*

Protein kinase C (PKC) (cid:1) , a Ca 2 (cid:2) -independent isoform of PKC, has been known to be expressed in skeletal muscle and T cells. In the present study, we isolated and characterized a smaller transcript expressed in the mouse testis, the cDNA of which is referred hereafter as PKC (cid:1) II and the original PKC (cid:1) as PKC (cid:1) I. The cDNA clone of PKC (cid:1) II has 2184 base pairs and 464 amino acids in the possible open reading frame, consisting of the 5 (cid:1) unique sequence of 20 amino acids and the PKC (cid:1) I sequence of 444 amino acids. Genomic DNA analysis revealed that transcription of PKC (cid:1) II is initiated from the PKC (cid:1) II-specific exon, which is located between exons 7 and 8 of the PKC (cid:1) gene, indicating that alternative splicing is the mechanism by which PKC (cid:1) II is generated. PKC (cid:1) II

Protein kinase C (PKC) 1 is a family of serine/threonine kinases that plays crucial roles in signal transduction (1). Members of the PKC family are divided into three groups based on their molecular structures and activating mechanisms: conventional PKC (␣, ␤, and ␥ isoforms) requiring calcium, phosphatidylserine (PS), and diacylglycerol (DG) for activation; novel PKC (␦, ⑀, , and isoforms) activated independent of calcium; and atypical PKC ( and / isoforms), which are independent of both calcium and DG. Alternative splicing-derived variants of PKC were reported for the ␤ and ␦ isoforms, i.e. PKC␤I, PKC␤II, PKC␦II, and PKC␦III (2)(3)(4).
Among these isoforms, we isolated PKC and PKC from a cDNA library of mouse skin (5,6). Our series of studies elucidated that PKC is expressed in squamous epithelia in a close association with differentiation (7) and that it induces terminal differentiation in keratinocytes by inducing transglutaminase 1 and binding to the cyclin E⅐cdk2⅐p21 complex (8,9).
Unlike PKC, PKC was found to be expressed in muscle cells, suggesting the possibility that it is derived from the muscle layer incidentally present in the skin preparation. Besides muscle cells, PKC is known to play an important role in T cells; it is colocalized in the T cell receptor and CD3 complex at the contact region between antigen-specific T cells and antigen-presenting cells (10,11). In T cells, PKC, in synergy with calcineurin, activates JNK and the interleukin-2 promoter and induces cytokine-response modifier A-sensitive apoptosis (12)(13)(14)(15). Tang et al. (16) reported that PKC is also required for angiogenesis and wound healing. Cellular functions mediated by PKC were reviewed by Meller et al. (17).
Expression of a small-size PKC mRNA in the testis was reported by Mischak et al. (18). We also noted the presence of a small-size transcript of PKC in the testis. 2 However, this small-sized PKC has not been characterized yet.
In the present study, we report on PKCII, a new PKC isoform derived by alternative splicing of the PKC gene, which lacks the V1 domain and a zing finger motif in the C1 regulatory domain and is expressed exclusively in the seminiferous tubules of the mouse testis.

EXPERIMENTAL PROCEDURES
Library Screening and DNA Sequencing-The mouse testis cDNA library prepared from the CD-1 mouse testis (Stratagene, CA) was screened by a standard protocol (19) using PCR products of the V3 region (875-1289 bp) of PKCI as a probe (probe A in Fig. 3A). A pair of the primers 1/2 used is shown below in Table I. Among the ϳ1 ϫ 10 6 independent colonies screened, two clones were isolated, which were then subjected to DNA sequencing with the ALFexpress system (Amersham Pharmacia Biotech, UK).
5Ј-Rapid Amplification of cDNA Ends-The 5Ј-end of PKCII mRNA was determined using the 5ЈRACE kit (version 2, Life Technologies, Inc., Gaithersburg, MD) according to the conditions suggested by the manufacturer. PCR products were subcloned into pBluescriptII(Ϫ) and sequenced with the ALFexpress system.
Northern Blot Analysis-The Mouse Multiple Tissue Northern * This work was supported by a Grant-in-aid for Cancer Research from the Ministry of Education, Science, Sports and Culture of Japan. 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) AB062122.
Genomic PCR-Genomic DNA of an 8-week-old female C57BL/6 mouse was prepared from the tail by a method described previously (20) and subjected to genomic PCR using the Expand Long Template PCR system (Roche Molecular Biochemicals, Germany) under conditions recommended by the manufacturer. Pairs of the primers 3/5, 4/5, and 6/7 were used for amplification of exon 2/II exon, exon 4/II exon, and II exon/exon 8, respectively (Table I). PCR products were subcloned into pBluescriptII(Ϫ) and sequenced by Nippon Flour Mills Co., Ltd.
RT-PCR of Testis-Total RNA of the testis from 0-, 1-, 2-, 3-, 4-, 8and 26-week-old mice was purified with a QuickPrep Total RNA extraction kit (Amersham Pharmacia Biotech, UK). Five micrograms of total RNA was reverse-transcribed with SuperScriptII (Life Technologies, Inc.) and amplified in AmpliTaq Gold under the following PCR conditions: 5 min at 95°C, 25 cycles of 1 min at 95°C, 1 min at 56°C, and 2 min at 72°C, followed by 7 min at 72°C. Pairs of primers 8/10 and 9/10 were used for PKCI and PKCII, respectively. The PCR products were subjected to electrophoresis on 2% NuSieve 3:1 agarose gel and visualized with SYBR Green I (Molecular Probes, OR). cDNA concentrations were normalized to the G3PDH content.
In Situ Hybridization-The cDNAs for in situ hybridization were prepared by PCR using a pair of primers 11/12 for PKCI and 13/14 for PKCII. The PCR products were digested with BglII and KpnI and cloned into pLITMUS 28 (New England BioLabs, Beverly, MA). Sense and antisense digoxigenin-labeled RNA probes were transcribed from a plasmid containing PKCI or PKCII cDNA linearized by T7 RNA polymerase in vitro. The testis of the 6-month-old mice (ICR) was fixed in 4% paraformaldehyde, embedded in the Tissue Tek O. C. T. compound (VWR Scientific, San Francisco, CA) and frozen-sectioned. Sections were treated with proteinase K (1 g/ml) at 37°C for 20 min and refixed in 4% paraformaldehyde. They were hybridized in a hybridization buffer (50% formamide, 5ϫ SSC, pH 4.5, 1% SDS, 50 g/ml yeast tRNA, 50 g/ml heparin, and 2.5 g/ml digoxigenin-labeled RNA probe) at 50°C overnight. The hybridized RNA was detected with alkaline phosphatase-conjugated anti-digoxigenin antibody (Roche Molecular Biochemicals, Germany) according to the procedure described by Wilkinson (21).
Laser Microdissection and RT-PCR-The testis of 2-and 4-week-old mice was rapidly embedded in the Tissue Tec OCT medium and frozensectioned at 8-m thickness. Frozen sections were fixed in 100% methanol for 3 min and then stained with toluidine blue. Once air-dried, the sections were laser-microdissected with a CRI-337 (Cell Robotics, Inc., Albuquerque, NM) (22). Approximately 20 -200 cells were laser-transferred from the seminiferous tubules and the interstitium. Contamination with nontarget components was monitored morphologically. Total RNA was extracted from laser-transferred cells according to a modified FLAG-tagged Expression Plasmids-FLAG-tagged expression plasmids of PKCI and -II were constructed with pFLAG-CMV5 (Eastman Kodak, Rochester, NY). The stop codon was removed from cDNAs of PKCI and PKCII, and the EcoRV site was introduced by PCR. Both cDNAs were subcloned into the EcoRV site of pFLAG-CMV5. The cDNAs were transfected using the LipofectAMINE 2000 reagent (Life Technologies, Inc.) into HEK293 cells grown in Dulbecco's modified Eagle's medium plus 10% fetal calf serum. The cells were harvested 24 h after transfection for immunoprecipitation.
Immunoprecipitation and Immunoblotting-Full-length cDNAs of PKCI and -II were cloned into pEF-BOS expression plasmid (23) and transfected into COS7 cells, which were grown in Dulbecco's modified Eagle's medium plus 10% fetal calf serum for 24 h. Transfected COS7 cells as well as testis from 26-week-old mouse (50 mg each) were sonicated in homogenizing buffer, and aliquots of each samples were subjected to immunoblotting with anti-PKC antibody against the Cterminal peptide for PKCI/II (nPKC (C-18), Santa Cruz Biotechnology, Santa Cruz, CA). Transfected HEK293 cells were sonicated in a homogenizing buffer (20 mM Tris-HCl, pH 7.5, 50 mM KCl, 2 mM MgCl 2 , 5 mM EGTA, 1% Nonidet P-40, 1 mM PMSF, 5 g/ml leupeptin, 5 g/ml antipain, and 5 g/ml aprotinin). The supernatants obtained by centrifugation at 21,000 ϫ g for 10 min were incubated with Protein G-Sepharose beads harboring the anti-FLAG antibody (Sigma Chemical Co., St. Louis, MO), followed by washing with buffer A (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Triton X-100, 2.5 mM EGTA, 1 mM PMSF, 5 g/ml leupeptin, 5 g/ml antipain, and 5 g/ml aprotinin) three times, buffer B (the same as buffer A without EGTA) once, and buffer C (20 mM Tris-HCl, pH 7.5, 5 mM MgCl 2 , 1 mM PMSF, 5 g/ml leupeptin, 5 g/ml antipain, and 5 g/ml aprotinin) once. An aliquot of the immunoprecipitated beads was separated using 8% SDS-polyacrylamide gel electrophoresis and transferred onto a Hybond-C membrane (Amersham Pharmacia Biotech, UK). Transferred products on membranes were blocked with 5% skim milk and 3% bovine serum albumin in phosphatebuffered saline and incubated with the anti-PKC antibody, followed by immunodetection using Western blot Chemiluminescence Reagent Plus (PerkinElmer Life Sciences). Expression at the protein level was mon-
In Vitro Kinase Assay-Kinase activity of pFLAG-PKCI and -PKCII was assayed in vitro using myelin basic protein (Sigma) as a substrate. Immunoprecipitated proteins were normalized by immunoblotting and incubated in 50 l of the reaction mixture (50 mM Tris-HCl, pH 7.5, 5 mM MgCl 2 , 100 M ATP, 1 Ci of [␥-32 P]ATP, 1 mM PMSF, 5 g/ml leupeptin, 5 g/ml antipain, and 5 g/ml aprotinin) in the presence or absence of the PKC activators, e.g. 1 mM CaCl 2 or 1 mM EGTA, 50 g/ml PS, and 100 ng/ml TPA at 25°C for 10 min. The reaction was stopped by adding 30 l of SDS sample buffer and boiling for 3 min. Following a brief centrifugation, 20-l aliquots of supernatants were subjected to 15% SDS-polyacrylamide gel electrophoresis, and after drying, the gels were exposed to an x-ray film. Fig. 1, Northern blot analysis of PKC in mouse tissues demonstrated the existence of at least four transcripts corresponding to bands of 3.5-, 3.0-, 2.5-, and 1.2-kb mRNA when a DNA fragment of the V3 region was used as a probe (probe A in Fig. 3A). The two large transcripts (3.5 and 3.0 kb) were detected in all the tissues examined except for the testis, whereas the testis expressed smaller transcripts of 2.5 and 1.2 kb.

Isolation of PKCII cDNA-As shown in
To identify the transcript of 2.5 kb, we screened a mouse testis cDNA library with probe A and isolated two independent clones containing the entire coding region. Throughout this paper, the 2.5-kb transcript is referred to as PKCII and the original PKC as PKCI. The transcript of 1.2 kb remains to be identified.
The cDNA clone of PKCII has 2184 bp, to which 18 nucleotides were added by 5Ј-RACE of mouse testis mRNA. As shown in Fig. 2, PKCII cDNA consists of a unique sequence of 276 bp at the 5Ј-end and a sequence identical to PKCII at the 3Ј-end, of which the boundary is located at 879 bp of the PKCI cDNA (GenBank accession number D11091). The deduced amino acid sequence of PKCII contains four methionines, as a candidate of the starting amino acid for translation, although none of them conform to the Kozak consensus sequence (Fig. 2). Hereafter, we adopted the first methionine as a starting site because of the presence of an in-frame termination codon upstream.
A possible open reading frame initiated from the first ATG codes 464 amino acid, consisting of the 5Ј unique sequence of 20 amino acids and the PKCI sequence of 444 amino acids (Fig.  2). The latter sequence contains a part of the C1 regulatory domain lacking the zinc finger motif, the V3 domain, the C3/ V4/C4 catalytic domain, and the C-terminal V5 domain. The N-terminal 20-amino acid sequence does not show significant homology to any sequence in the existing cDNA data base. The structure of the PKCII cDNA is compared with that of PKCI in Fig. 3A.
Testis-specific Transcription of PKCII-Northern blot analysis showed that the unique sequence of PKCII (probe B in Fig. 3A) hybridized only the smaller PKCII transcripts expressed in the testis (Fig. 1B), whereas a probe of the 5Ј-end of PKCI (probe C in Fig. 3A) hybridized mRNA of PKCI expressed in the heart, brain, spleen, lung, liver, skeletal muscle, and kidney (Fig. 1C). A common probe for PKCI and PKCII (probe A) hybridized both PKCI and PKCII (Fig. 1A). In addition, we found that PKCII was not expressed in the ovary (data not shown). These results indicate that PKCII is specifically transcribed in the testis whereas PKCI is expressed ubiquitously except in the testis.
Genomic Structure of PKC Gene-The above sequencing data suggest that the 2.5-kb transcript detected in the testis could be derived from alternative splicing of the PKC gene (Pkcq). The Pkc q was mapped to chromosome 10p15 in human (24) and chromosome 2 in mice (25). Its genomic structure in human was reported (GenBank accession number AL158043).
PCR analysis of the mouse genomic DNA indicated that the sizes of PCR products from the PKCII-specific domain to exons 2, 4, and 8 were 20, 5.5, and 2.3 kb, respectively. Exons 4, 5, 6, 7, and 8 were found to be well conserved in mouse and human genomes. Based on these results, we located the PKCII-specific exon composed of 276 nucleotides (hereafter, the II exon) between exons 7 and 8 (Fig. 3B). Exon II can be accepted into exon 8, because sequences at the exon-intron boundaries fit with the donor-acceptor splicing rule. However, no splicing acceptor sequence was found around the 5Ј-boundary of the II exon, suggesting that the II exon is the first exon of PKCII cDNA. Fig. 3B summarizes the generation of PKCI and PKCII cDNAs from the PKC gene by alternative usage of exons: The V1 and C1 domains of PKCI cDNA are generated from exons 1-8, whereas the PKCII unique sequence is derived from the II exon. A testis-specific promoter may be located upstream of the II exon. This genomic structure indicates that alternative splicing is the mechanism by which PKCII is generated.
Age-dependent Expression of PKCII-In the testis, spermatogenesis and androgen production proceed in an age-dependent manner, becoming mature at about 4 weeks after birth. Age-dependent expression of PKCII was analyzed by RT-PCR using total testis RNA isolated from newborn, 1-, 2-, 3-, 4-, 8-, and 26-week-old mice. Expression levels were normalized to GAPDH, and developmental stages were monitored based on the expression of c-Kit, protamine-1, and acrosin (data not shown).
When the primers for PKCII-specific sequences were used, expression of PKCII was first detected at 3 weeks and increased thereafter with age until adulthood (Fig. 4). In contrast, PKCI was not expressed in the testis at any age when amplified using the PKCI-specific primers.
This observation suggests that PKCII is involved in the maturation of a cell function(s) in the testis, e.g. spermatogenesis and/or androgen production, and responsiveness.
Expression of PKCI and PKCII at the Protein Level-PKCI and PKCII were expressed at the protein level in the mouse testis and in the cells overexpressing these cDNAs. In Fig. 5A, total proteins of the testis from 26-week-old mouse as well as from COS7 cells transfected with PKCI and PKCII cDNAs were subjected to immunoblotting with the antibody against to the common C-terminal peptide for PKCI and PKCII. A band corresponding to the molecular mass of PKCII (50 kDa) was detected in the testis and PKCII-transfected cells, whereas a band corresponding to PKCI (80 kDa) was seen only in the PKCI-expressing cells. Expression at the protein level was further confirmed by immunoprecipitationimmunoblotting of HEK293 cells transfected with pFLAG-PKCI and -PKCII (Fig. 5B). The transfected HEK293 cells were immunoprecipitated with the anti-FLAG antibody, followed by immunoblotting with the PKCI/PKCII antibody. Again, products of PKCI and PKCII were detected in the cells transfected with the corresponding cDNAs.
Localization of PKCII in Testis-The testis is composed of two important functional components: the seminiferous tubules producing male germ cells and the interstitium producing androgen. To localize the expression of PKCII in the testis, we performed in situ hybridization and RT-PCR of laser-microdissected tissues. As shown in Fig. 6, mRNA transcribed by the II exon was detected by in situ hybridization exclusively in the seminiferous tubules, not in the interstitial cells. A negative control with a sense probe did not show any signal. Furthermore, PKCI was found not to be expressed in the testis when the PKCI-specific sequence was used as a probe (data not shown). It is not certain from in situ hybridization whether Sertoli cells expressed PKCII, although a Sertoli cell-derived cell line, TM-4, did not express it (data not shown).
These specific features of PKCII expression were further confirmed by RT-PCR of the microdissected seminiferous tubules and interstitial cells from 2-and 4-week-old mice (Fig. 7). The expression of PKCII was monitored based on ␤-actin mRNA, and cell type specificities were examined morphologically as well as expression of protamine-1 and selenoprotein-1 in the seminiferous tubules and interstitium, respectively (26,27). As shown in Fig. 7B, PKCII was expressed in the seminiferous tubules of 4-week-old mice, but not in those of 2-weekold mice, whereas no expression was observed in the interstitium. PKCI was not detected in any testis component at 2 or 4 weeks.
These data clearly indicate the exclusive expression of PKCII in seminiferous tubules in an age-dependent manner.
Kinase Activity of PKCI and PKCII-The lack of the complete C1 regulatory domain in PKCII suggests the possible independence of PS and DG or phorbol esters for its activation. To address this question, cDNAs of PKCI and PKCII were subcloned into the expression vector pFLAG-CMV5, transfected into HEK293 cells, and immunoprecipitated with the anti-FLAG antibody. After normalizing the expression level of PKCI and PKCII by immunoblotting with the anti PKC antibody (Fig. 5B), immunoprecipitates were subjected to in vitro kinase assay in the presence or absence of known activators of PKC. Myelin basic protein was used as a substrate. As seen in Fig. 8A, PKCII showed a certain level of kinase activity in the absence of activators and no further activation was observed following the addition of Ca 2ϩ , PS, or TPA, suggesting the possibility that PKCII is constitutively active independent of the PKC activators. Activity of PKCI, however, was dependent on PS and TPA, which is in agreement with previous report (6,28).
In contrast to phosphorylation of myelin basic protein, autophosphorylation was observed only with PKCI in an activatordependent manner (Fig. 8B). No autophosphorylation was noted with PKCII. DISCUSSION PKCII is unique in that it is mainly composed of a catalytic domain and it is specifically expressed in the seminiferous tubules of the mouse testis. Genomic DNA analysis revealed that PKCII is generated by alternative splicing of the PKC gene. The cDNA of PKCII consists of a unique N-terminal sequence encoding 20 amino acids and the C-terminal sequence identical to that of PKCI encoding a part of the C1 regulatory domain followed by the V3 domain, C3/V4/C4 catalytic domain, and V5 domain.
Alternative splicing is a common transcriptional mechanism, by which functionally diverse polypeptides are produced from a single gene. Based on a minimum estimate, 35% of human genes show variably spliced products (29,30). Alternative splicing yields a wide variety of the encoded proteins by addition or deletion of a sequence(s), which may be involved in a certain stage of development and differentiation.

FIG. 6. In situ hybridization of PKCII mRNA in mouse testis.
Six-month-old mouse testis was frozen-sectioned and hybridized for PKCII with sense (A) and antisense probes (B). Note that PKCII was expressed in seminiferous tubules, not in interstitial cells.
As for PKC, alternatively spliced isoforms of the ␤ and ␦ isoforms are known. PKC␤I and -␤II are produced by alternative splicing of the C-terminal V5 domain, which may function in intracellular localization (31,32). Recently, two alternatively spliced isoforms of PKC␦ have been reported, i.e. PKC␦II and PKC␦III. PKC␦II contains a 78-bp insertion at the caspase-3 recognition sequence at the V3 domain (GenBank accession number AB011812). PKC␦III is generated by an 83-bp inser-tion at the same V3 domain, causing in-frame termination, thereby yielding a truncated form of PKC␦ without the catalytic domain (4). In the present study, we reported another alternatively spliced isoform of PKC, i.e. PKCII.
Functions of certain proteins are switched on/off by phosphorylation and dephosphorylation. Switching enzymes, therefore, should be strictly regulated. There are several ways of regulating protein kinases. These include binding of the regulatory proteins, a good example being cdk kinases that are regulated by binding to cyclins and cdk inhibitors, and phosphorylation of a specific residue, such as serine 15 of p53 and threonine 160 of cdk2 (33,34). PKC is well known to be activated by metabolic turnover of polar head groups of membrane phospholipids (35). Other activation mechanisms for PKC include phosphorylation of a serine/threonine residue at the activation loop (36) and cleavage between the regulatory and catalytic domains (37). The latter mechanism is the case for PKCII.
When assayed in vitro using immunoprecipitated PKCII, a certain level of background kinase activity was noted in the absence of PS and TPA. Addition of activators did not enhance the activity, which is consistent with its molecular structure, i.e. lacking the zinc finger motif in the C1 regulatory domain. These data suggest that PKCII is constitutively active independent of PKC activators. PKCII may be regulated at the transcription level or by a yet-unidentified mechanism. Absence of autophosphorylation in PKCII suggests possible absence of an autophosphorylation site(s) or its regulating mechanisms in PKCII.
PKCII is most unique in that it is expressed exclusively in germ cells in seminiferous tubules. There is no expression of PKCII in any tissue other than the testis, and in the testis the expression is limited exclusively to seminiferous tubules. This was demonstrated by in situ hybridization and RT-PCR of the microdissected tissues using the PKCII-specific probe or primer, respectively. In contrast, Kim and Shin (38) reported that signals of PKC were detected in the interstitial cells of mouse testis by in situ hybridization and immunohistochemical staining. However, the 5Ј-end probe that they used for in situ hybridization was specific to PKCI, not to PKCII, and the commercially available antibody used recognized both PKCI and PKCII. We also found that the same antibody stained interstitial cells not seminiferous tubules.
Spermatogenesis occurs in seminiferous tubules and consists of three phases: proliferation and differentiation of spermato-  7. Exclusive expression of PKCII in the microdissected seminiferous tubules. A, seminiferous tubules (a-c) and interstitium (d-f) of mouse testis at 2 and 4 weeks old were microdissected. Specimens before (a, d) and after (b, e) microdissection, and after laser pressure cell transfer (c, f) are shown. B, RT-PCR of the microdissected tissues from 2-week-old (top) and 4-weekold (bottom) mice using PKCIand PKCII-specific primers. Tissue specificities were monitored based on morphology (A) and expression of marker genes (p, protamine-1 for seminiferous tubules; s, selenoprotein-P for interstitium; a, ␤-actin for amount of RNA). gonia, meiotic division of spermatocytes, and development of post-haploid spermatids to sperms. Age-dependent expression of PKCII after 3 weeks coincides with differentiation of haploid germ cells, suggesting its crucial involvement in spermatogenesis. However, Sun et al. (39) reported that PKC null mice seemed normal and were fertile. These mice were generated by homologous deletion of the exon encoding the ATP binding site of the C3 domain, which corresponds to amino acid residues 154 -207 in PKCII. Fertility of these mice may be due to redundancy of PKC in testis, in which ␣, ␦, II, and isoforms are present (18). Further study is needed for elucidating the mechanism by which PKCII mediates signals for spermatogenesis.