VHY, a novel myristoylated testis-restricted dual specificity protein phosphatase related to VHX.

The human DUSP15 gene encodes an uncharacterized 235-amino acid member of the subfamily of small dual specificity protein phosphatases related to the Vaccinia virus VH1 phosphatase. Similar to VHR-related MKPX (VHX) (DUSP22), the predicted protein has an N-terminal myristoylation recognition sequence, and we show here that both are indeed modified by the attachment of a myristate to Gly-2. In recognition of this relatedness to VHX, we refer to the DUSP15-encoded protein as VH1-related member Y (VHY). We report that VHY is expressed at high levels in the testis and barely detectable levels in the brain, spinal cord, and thyroid. A VHY-specific antiserum detected a protein with an apparent molecular mass of 26 kDa, and histochemical analysis showed that VHY was readily detectable in pachytene spermatocytes (midstage of meiotic division I) and round spermatids and weakly in Leydig cells (somatic cells outside of the seminiferous tubules). When expressed in 293T or NIH-3T3 cells, VHY was concentrated at the plasma membrane with some staining of vesicular structures in the Golgi region. Mutation of the myristoylation site Gly-2 abrogated membrane location. Finally, we demonstrate that VHY is an active phosphatase in vitro. We conclude that VHY is a new member of a subgroup of myristoylated VH1-like small dual specificity phosphatases.

The human DUSP15 gene encodes an uncharacterized 235-amino acid member of the subfamily of small dual specificity protein phosphatases related to the Vaccinia virus VH1 phosphatase. Similar to VHR-related MKPX (VHX) (DUSP22), the predicted protein has an N-terminal myristoylation recognition sequence, and we show here that both are indeed modified by the attachment of a myristate to Gly-2. In recognition of this relatedness to VHX, we refer to the DUSP15-encoded protein as VH1related member Y (VHY). We report that VHY is expressed at high levels in the testis and barely detectable levels in the brain, spinal cord, and thyroid. A VHY-specific antiserum detected a protein with an apparent molecular mass of 26 kDa, and histochemical analysis showed that VHY was readily detectable in pachytene spermatocytes (midstage of meiotic division I) and round spermatids and weakly in Leydig cells (somatic cells outside of the seminiferous tubules). When expressed in 293T or NIH-3T3 cells, VHY was concentrated at the plasma membrane with some staining of vesicular structures in the Golgi region. Mutation of the myristoylation site Gly-2 abrogated membrane location. Finally, we demonstrate that VHY is an active phosphatase in vitro. We conclude that VHY is a new member of a subgroup of myristoylated VH1-like small dual specificity phosphatases.
Phosphate is removed from phosphoproteins by several unrelated classes of protein phosphatases, including the serine/ threonine-specific phosphatases (PP1, PP2, etc.) 1 and three families of cysteine-based protein tyrosine phosphatases (PTPs) (reviewed in Refs. [1][2][3]. There is also a newly discovered family of metal ion-dependent aspartic acid-based PTPs represented by the four Eyes Absent proteins (4 -6). The largest family of cysteine-based (7) phosphatases is made up of those related to the first sequenced PTP, PTP1B (8), and the receptor-like enzyme, CD45. Within this family are also 19 phosphatases with predominantly nonprotein substrates, 17 of them specific for the D3-phosphate of inositol phospholipids (9,10) and two acting on RNA during the mRNA capping process (11,12). Another subgroup of 38 enzymes, referred to as the "classical" PTPs (13), have a deep catalytic pocket (ϳ9 Å), which only allows the long side chain of phosphotyrosine to reach the catalytic machinery (14). Accordingly, these PTPs are strictly tyrosine-specific. Other members of the PTP family have a more shallow catalytic pocket, and many of them dephosphorylate both phosphoserine/threonine and phosphotyrosine residues, at least in vitro (15)(16)(17)(18)(19)(20). These "dual specificity" phosphatases (DSPs) include the 11 MAP kinase phosphatases (MKPs) that dephosphorylate the mitogen-activated protein kinases Erk, Jnk, and p38 (21,22) at their dually phosphorylated TXY activation loop motif.
Another subgroup of DSPs, which we have referred to as the "atypical" DSPs (22) (as opposed to the "typical" DSPs, the MKPs), includes a number of poorly known enzymes that lack specific MAP kinase-targeting motifs and tend to be much smaller enzymes (less than 250 amino acid residues). The first DSP to be cloned, the VH1 protein from the vaccinia virus (23), is a member of this group, as are the human VHR (24) and a number of genes given the genomic designations DUSP14, DUSP19, DUSP20, DUSP21, and DUSP22. Although VH1 has been reported to dephosphorylate both MAP kinases and Stat1 (25), and VHR can dephosphorylate Erk and Jnk in 293T cells (26) and T cells (27,28), it appears that at least some of these small atypical DSPs also have functions related to MAP kinases.
We recently published (29) a new atypical DSP, which we termed VHX, for VHR-related MKPX, to honor a GenBank TM / NCBI submission of its sequence by Gu and co-workers (Gen-Bank TM /NCBI accession numbers AF165519/gi9294744) 2 during the course of our work. This gene is now designated as DUSP22, and it has also been published by others under the names JKAP (30) and JSP1 (31). Unfortunately, the first name applied to this gene, MKPX, cannot be given priority because the name was already occupied by another gene, DUSP7 (also known as MKP-X, PYST2, and B59), only adding more confusion to the already crowded PTP nomenclature. Here we report the identification and initial characterization of another DSP, encoded by the DUSP15 gene, which encodes a 235-amino acid DSP closely related to VHX. The new protein, which we refer to as VHY, has the same N-terminal consensus myristoylation * This work was supported by National Institutes of Health Grants AI35603, AI48032, AI53585, AI55741, and CA96949 (to T. M.). 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  signal as VHX; and we show that it is indeed myristoylated and thereby targeted to the cytoplasmic face of the plasma membrane, where it co-localizes with filamentous actin in membrane ruffles and filop-odia. VHY expression is highest in the testis, where it is mainly found in pachytene spermatozytes.

EXPERIMENTAL PROCEDURES
Antibodies-The 9E10 anti-Myc epitope tag mAb was from Roche Applied Science. The anti-phosphotyrosine mAb 4G10 was from Upstate Biotechnology Inc. (Lake Placid, NY). Two rabbits were immunized with a GST fusion protein expressing the entire full-length VHY protein. The resulting antisera were used at 1:1,000 dilution.
Plasmids and Site-directed Mutagenesis-The cDNA for VHY was cloned into the 5Ј-EcoRI-3Ј-XhoI site of pcDNA3.1myc-his with the tags in the C terminus of the insert. The cDNA for VHX was in the same vector. Point mutants of VHY and VHX were generated using the Transformer site-directed mutagenesis kit (Clontech and verified by sequencing.
Northern Blots-These were performed with Northern blot filters and multiple tissue arrays from Clontech, and the full-length VHY cDNA as a probe. The blot for mouse mRNA was prepared by standard protocol using 10 g of total RNA.
Southern Blots-Genomic DNA samples from human prostate cancer cell line PC3 and mouse embryonic stem cell line D3 were digested with several common restriction enzymes with sites in both the human DUSP15 (within the VHY cDNA) and the mouse Dusp15 gene (clone BB616533) and analyzed with Southern blot hybridization using the coding region of the human VHY cDNA as a probe.
Histology-Testis samples were collected from adult 129 ϫ C57/Bl6 hybrid mice and immediately fixed in 4% paraformaldehyde in phosphate-buffered saline at 4°C for 2 days. The tissue was then washed in 50% ethanol in phosphate-buffered saline and processed for standard paraffin sectioning. After deparaffinization and dehydration, the sections were immersed in 8 M guanidine, HCl (pH 10) overnight as an unmasking treatment. Immunostaining was performed using an ABC kit according to the manufacturer's instructions (Vector Laboratories, Burlingame, CA), except that a 1:200 dilution of rabbit anti-VHY serum or normal rabbit serum was included for 2 h at room temperature as a primary antibody. Mayer's hematoxylin was used for counterstaining.
p-Nitrophenyl Phosphate Dephosphorylation Assay-The hydrolysis of p-nitrophenyl phosphate was assayed at 30°C for 5 min with 20 g of enzyme in 100 l of 0.1 M BisTris, pH 6.0, 150 mM NaCl, 1 mM dithiothreitol. The reaction was initiated by the addition of various concentrations of p-nitrophenyl phosphate (ranging from 0.125 to 5 mM) to the reaction mixtures. Nonenzymatic hydrolysis of the substrate was corrected by measuring the control without the addition of enzyme. The amount of product p-nitrophenol was determined from the absorbance at 405 nm detected by a PowerWaveX340 microplate spectrophotometer (Bio-Tek Instruments, Inc.) using a molar extinction coefficient of 18,000 M Ϫ1 cm Ϫ1 . The K m and V max were determined by fitting the data to the Michaelis-Menten equation using nonlinear regression and the program GraphPad Prism® (version 4.0).
Cells and Transfections-293T kidney epithelium cells were kept at logarithmic growth in RPMI 1640 medium supplemented with 10% fetal calf serum, 300 mg/ml L-glutamine, and 10 g/ml gentamycin. They were transiently transfected with a total of 5-15 g of DNA by lipofection as described previously (32). Empty vector was added to the control samples to make a constant amount of DNA in each sample. The cells were used for experiments 24 h after transfection.
Immunoblotting-Proteins resolved by SDS-PAGE were transferred electrophoretically to nitrocellulose filters, which were immunoblotted as described previously (27)(28)(29)32) with optimal dilutions of mAbs followed by anti-mouse Ig-peroxidase, and the blots were developed by the enhanced chemiluminescence technique (ECL kit, Amersham Biosciences) according to the manufacturer's instructions.
Confocal Microscopy-Double immunofluorescence staining was done as described previously (28,33,34). Briefly, cells were washed in phosphate-buffered saline and fixed in freshly made 3.7% formaldehyde. The fixed cells were permeabilized with 0.1% saponin in phosphate-buffered saline, blocked in 2.5% normal goat serum in 0.1% saponin in phosphate-buffered saline for 30 min at room temperature, and then incubated with primary and secondary antibody diluted in the same buffer for 1 h each at room temperature. After three washes with phosphate-buffered saline, the cells were mounted onto glass slides and viewed under a confocal laser scanning microscopy MRC-1024 (Bio-Rad). A differential interference contrast image was also taken of most of the cells.

RESULTS
The Protein Encoded by DUSP15-A number of human cDNAs have been deposited in the GenBank TM data base as dual specificity protein phosphatase 15Ј (gi:30316387), dual specificity phosphatase 15-like (gi:29826315, gi:29826317), or other designations (gi:14031085; gi:21757185), many of them differing in N-terminal and C-terminal sequences. Mouse and rat orthologs were also present. The full-length protein encoded by DUSP15 is a 235-amino acid residue protein with the PTP signature motif, HCX 2 GX 2 R, at positions 87-94. It displays 51% identity and 66% similarity with human VHX (DUSP22) over a 184-amino acid alignment (Fig. 1A). Because VHX is its closest relative, we refer to the DUSP15-encoded protein as VHY (for VH1-like member Y). The two proteins are particularly similar in their extreme N termini and around the catalytic core residues, but VHY has a longer C terminus than VHX. The first 10 amino acid residues bear similarity to the consensus myristoylation sequence of Src family kinases (35).
Expression of VHY Is Highest in the Testis-To determine in which tissues and cell types VHY is expressed, we performed a Northern blot using multiple tissue panels (Clontech Inc.) with mRNA from many different tissues and the full-length VHY cDNA as a probe. As shown in Fig. 1C, a prominent ϳ1.4-kb mRNA was seen in the testis but only weakly in some other tissues such as brain, spinal cord, and thyroid. A similar result was obtained with a multiple tissue array (Clontech) probed with full-length VHY; the highest expression was found in testis, and a low (but detectable) level was found in many other tissues (not shown). Thus, VHY is expressed at high levels in the testis but can also be detected in some other tissues. To determine whether the corresponding mouse gene shows a corresponding expression pattern, we first performed a Southern blot to ensure that the human VHY cDNA reacts with the mouse DUSP15 gene. Both human and mouse genomic DNA was digested with EcoRV, PstI, HindIII, and KpnI, resolved electrophoretically, and hybridized with the human VHY cDNA probe. Based on the positive result (Fig. 1C), we then performed a Northern blot with mRNA from mouse testis or an embryonic stem cell line, which showed a strong ϳ1.4-kb mRNA in the testis sample only. A control L32 ribosomal protein cDNA probe detected an equal band in both samples. Thus, the mouse DUSP15 gene is highly expressed also in mouse testis.
Detection of Endogenous VHY Protein in Cell Lines and in the Testis-A GST fusion protein containing the entire open reading frame of VHY was generated, purified, and used for immunization of two rabbits. To test the resulting antisera, we expressed wild-type or catalytically inactive Myc-tagged VHY in 293T cells and then immunoblotted lysates of these cells with the antisera, which reacted well with the proteins (Fig. 2A). The antisera also detected an endogenous protein of 26 kDa in testis but not in mouse embryonic cells, spleen, or human 293T cells (Fig. 2B).
Immunohistochemistry of paraformaldehyde-fixed paraffin sections of mouse testis showed an abundant staining of the cytoplasm of pachytene spermatocytes (midstage of meiotic division I) and round spermatids and a weaker staining of Leydig cells, which are somatic cells outside of the seminiferous tubules (Fig. 2C, middle panels and right panel). A normal rabbit serum control staining performed in parallel did not reveal any specific signal (Fig. 2C, left panels). Together with the demonstration that shows that the mouse DUSP15 gene is expressed in the testis, these findings indicate that VHY is present at readily detectable levels in mouse testis and is found mostly in developing spermatocytes. This is likely to be true also in humans.
VHY Is an Active Phosphatase-To formally demonstrate that VHY is a catalytically active phosphatase, as predicted for its amino acid sequence, we measured its ability to dephosphorylate the general PTPase substrate p-nitrophenyl phosphate.
Although the catalytic cysteine mutant VHY-C88S was inactive in these assays, the GST fusion protein of wild-type VHY readily converted p-nitrophenyl phosphate into the yellow pnitrophenol, with a K m of 0.199 Ϯ 0.0137 mM and K cat of 0.136 Ϯ 0.002 s Ϫ1 (Fig. 3A). Thus, VHY is a catalytically active phosphatase that requires the conserved catalytic residue Cys-88 for activity.
Myristoylation of VHY-The first five amino acid residues in VHY are identical to those in VHX, and both proteins have a Lys at position 7. This sequence is similar to the extreme N terminus of Src family kinases (Fig. 3B). Therefore, we decided to determine whether VHY is post-translationally modified by the attachment of myristic acid to Gly-2. 293T cells were transfected with empty vector or the expression plasmids for VHY or VHX, and metabolically labeled with [ 14 C]myristic acid for 6 h. The proteins were then immunoprecipitated, resolved on SDS gels, and processed for fluorography, which revealed that both VHY and VHX were readily labeled (Fig. 3C). No other bands were detected, and the vector control lane was devoid of bands. Thus, both VHX and VHY are myristoylated.
Myristoylation Determines the Subcellular Location of VHY-Although myristoylation of Src family kinases is required for plasma membrane targeting of these proteins (35), other myristoylated proteins, e.g. the phosphatase calcineurin (36), are cytosolic. To determine where in the cell VHY is located, we first transfected 293T cells with the VHY expression plasmid and stained them with a fluorescein isothiocyanate-conjugated anti-Myc mAb. When these cells were viewed under a confocal microscope, it was clear that most of the fluorescence was enriched at the plasma membrane (Fig. 4). The staining was not even but was concentrated in ruffles and other membrane features. Some fluorescence was also consistently seen on internal membranes. Cells transfected with empty vector did not display any fluorescence at all (Fig. 4). Finally, cells transfected with the myristoylation site mutant VHY-G2A did not show any membrane enrichment (Fig. 4,  bottom panels). Instead, the mutant VHY was located diffusely throughout the cell. Very similar results were obtained for VHX (data not shown); the unmodified protein was partly located at the plasma membrane, whereas the VHX-G2A mutant was diffusely cytoplasmic. Thus, myristoylation of both VHX and VHY at Gly-2 is required for their membrane targeting. DISCUSSION In this paper, we describe and begin the characterization of a novel DSP, VHY, which is encoded by the DUSP15 gene on human chromosome 20q11.21 and is related to the VHX (29) encoded by the DUSP22 gene on 6p25.3. The DUSP15 gene consists of 7 exons, whereas the DUSP22 gene contains 9 exons. The two enzymes are particularly similar in their extreme N termini, which in both proteins are encoded by a similar exon 1 (amino acid residues 1-7 in VHY). The two proteins are also very similar within the cores of their catalytic domains, and the main difference between them is that VHY has a longer C-terminal tail. They also differ strikingly in tissue distribution; VHY expressed at high levels only in the testis and at low levels in other tissues and cell types, whereas VHX is broadly expressed (29). The N termini of VHY and VHX contain a consensus myristoyl transferase recognition motif (35)(36)(37)(38) similar to that found in Src family kinases (35,39,40) and calcineurin B (36) and indeed are modified by myristic acid, presumably at Gly-2 (41). This modification correlates with the plasma membrane enrichment of VHY. VHX is also modified by myristic acid (Fig.   3C) and is enriched at the plasma membrane, where it colocalizes with membrane ruffles. Mutation of Gly-2 to prevent myristoylation also resulted in the loss of this membrane enrichment. Thus, VHX and VHY define a new subgroup of small DSPs anchored at the plasma membrane and possibly other membranes by an N-terminal myristic acid moiety. Both VHY and VHX also belong to a larger group of 18 atypical DSPs characterized by their small size (less than 250 amino acid residues) and a lack of protein-protein interaction domains or motifs, such as the Cdc25 homology region (CH2) present in the larger MAP kinase-specific typical DSPs (23). The atypical DSP group also contains VHR (DUSP3) (24), MKP6 (DUSP14) (42), BEDP (DUSP13A; AAK77966, gi15072533), TMDP (DUSP13B) (43), LMW-DSP20 (DUSP18) (44), SKRP1 (DUSP19) (45,46), BJ-HCC-26 or LMW-DSP21 (DUSP21) (44), and several unpublished open reading frames, which we are currently characterizing. Interestingly, many of these genes have been reported to be expressed at their highest levels in the testis, whereas others are found in muscle tissue, hematopoietic cells, or broadly in other tissues and cell types. The physiological functions of most of these enzymes are unknown, and most of them have not yet been addressed experimentally at all. Some data exist for VHR, which dephosphorylates the MAP kinases Erk and Jnk (26 -28) and requires tyrosine phosphorylation at Tyr-138 for biological activity (28). MKP6 was cloned as a protein physically associated with the CD28 co-receptor in T cells, where MKP6 was proposed to dephosphorylate Erk and Jnk (42). SKRP1 is involved in the regulation of Jnk and interacts with its upstream regulators MKK7 and ASK1 (apoptosis signal-regulating kinase 1) but not with MEKK1 (44,45). Finally, VHX also has been linked to MAP kinase regulation, with a proposed role as a positive regulator of Jnk activation (31,47). VHX associates with p38 and Jnk in vivo but not in vitro (30,47), suggesting that the interaction requires additional components. Candidates include MKK7 or c-Jun, both of which interact with VHX (31,47). Together, these results indicate that VHX may play a regulatory role in the Jnk or p38 pathways, but they do not yet reveal what phosphoprotein(s) VHX dephosphorylates. It should be noted that most studies on VHX, including our own (29), used N-terminally tagged constructs that prevent myristoylation of Gly-2. This presumably hindered the correct subcellular targeting of the protein. It now remains to be determined whether the physiological function(s) of VHX and VHY are related and whether they truly are connected to the Jnk kinase pathway.