SAMP 14 , a Novel , Acrosomal Membrane-associated , Glycosylphosphatidylinositol-anchored Member of the Ly-6 / Urokinase-type Plasminogen Activator Receptor Superfamily with a Role in Sperm-Egg Interaction *

We report a new member of the Ly-6/urokinase-type plasminogen activator receptor (uPAR) superfamily of receptors, SAMP14, which is retained on the inner acrosomal membrane of the human spermatozoan following the acrosome reaction and may play a role in fertilization. The SAMP14 sequence predicted a glycosylphosphatidylinositol (GPI)-anchored protein with a signal peptide, a transmembrane domain near the carboxyl terminus, and a putative transamidase cleavage site in the proprotein. Attachment of SAMP14 to the membrane by a lipid anchor was confirmed by its sensitivity to phosphatidylinositol phospholipase C. SAMP14 has a single functional domain similar to the Ly-6 and urokinase plasminogen activator receptor superfamily of proteins, and the gene mapped to 19q13.33, near the PLAUR locus for uPAR at 19q13.2. Northern and dot blotting showed that SAMP14 expression was testis-specific. Indirect immunofluorescence and immunoelectron microscopy with antisera to purified recombinant SAMP14 localized the protein to outer and inner acrosomal membranes as well as the acrosomal matrix of ejaculated human sperm. Acrosome-reacted sperm demonstrated SAMP14 immunofluorescence, indicating its retention on the inner acrosomal membrane following the acrosome reaction. However, SAMP14 localized to the entire sperm when unwashed swim-up sperm from the ejaculate were stained, indicating that some SAMP14 is loosely associated with the plasma membrane. Antibodies against recombinant SAMP14 inhibited both the binding and the fusion of human sperm to zona free hamster eggs, suggesting that SAMP14 may have a role in sperm-egg interaction. SAMP14 represents a GPI-anchored putative receptor in the Ly-6/uPAR family that is exposed on the inner acrosomal membrane after the acrosome reaction. Members of the Ly-6/urokinase-type plasminogen activator receptor (uPAR) family possess one or several repeats of a 100-amino acid Ly-6/uPAR domain that is defined by conservation of distinct disulfide bonds between 8 or 10 cysteine residues (1). The family can be divided into two subfamilies. One comprises GPI-anchored glycoprotein receptors with 10 cysteine residues. Members of this subfamily include uPAR, CD59, the Ly-6 family, thymocyte B cell antigen, and prostate stem cell antigen. Among these proteins, uPAR contains three tandem copies of the Ly-6/uPAR domain, whereas the others contain a single domain (2). The second subfamily contains snake and frog cytotoxins, principally three-finger neurotoxins. Its members lack the GPI-anchoring signal sequence and possess only eight cysteines. The Ly-6/uPAR receptors mediate several physiological functions including cellular activation, macrophage migration, trophoblast implantation, angiogenesis, and tumor cell invasion (3, 4). During mammalian fertilization, the acrosomal and plasma membranes of the sperm interact with the egg vestments. Sperm surface molecules on the plasma membrane are involved in the recognition of the zona pellucida protein (ZP3) during the early process of sperm egg interaction (primary binding) (5). Following sperm binding to the egg the acrosomal membranes are involved in a critical cascade of events that includes the following: 1) the acrosome reaction, in which the outer acrosomal membrane fuses with the plasma membrane; 2) secondary binding of the inner acrosomal membrane to the zona pellucida; and 3) tertiary binding of the equatorial segment and subsequent fusion to the vitelline membrane. After the acrosome reaction, the inner acrosomal membrane becomes the limiting membrane on the external surface of the anterior principal acrosomal segment, whereas the plasma membrane remains continuous with the outer acrosomal membrane at the anterior tip of the equatorial segment (6). The acrosome contains many enzymes including proteases, such as acrosin (7), esterases, acid phosphatases, glycohydrolases, and aryl sulfatases. Enzymes released from the acrosome are believed to play an important role in digesting the zona and paving the way for penetration of the egg vestments by the sperm (8). Knowledge of the components of the acrosomal membranes and matrix is essential to understanding the biology of acroso* This work was supported by D43 TW/HD 00654 from the Fogarty International Center, National Institutes of Health Grant HD U54 29099, the Andrew W. Mellon Foundation, Schering A.G., and Office of Justice Programs, National Institute of Justice, United States Department of Justice Grant 2000-IJ-CX-K013. 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/EBI Data Bank with accession number(s) AF353721. ‡ To whom correspondence should be addressed: Dept. of Cell Biology, Center for Research in Contraceptive and Reproductive Health, P.O. Box 800732, University of Virginia, Charlottesville, VA 22908. Tel.: 434-924-2007; Fax: 434-982-3912; E-mail: jch7k@virginia.edu. 1 The abbreviations used are: uPAR, urokinase-type plasminogen activator receptor; EST, expressed sequence tag; PIPLC, phosphatidylinositol-specific phospholipase C; aa, amino acid(s); recSAMP14, recombinant SAMP14; PBS, phosphate-buffered saline; TRITC, tetramethylrhodamine isothiocyanate; NHS, N-hydroxysuccinimide. THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 278, No. 33, Issue of August 15, pp. 30506–30515, 2003 © 2003 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

Members of the Ly-6/urokinase-type plasminogen activator receptor (uPAR) 1 family possess one or several repeats of a 100-amino acid Ly-6/uPAR domain that is defined by conservation of distinct disulfide bonds between 8 or 10 cysteine residues (1). The family can be divided into two subfamilies. One comprises GPI-anchored glycoprotein receptors with 10 cysteine residues. Members of this subfamily include uPAR, CD59, the Ly-6 family, thymocyte B cell antigen, and prostate stem cell antigen. Among these proteins, uPAR contains three tandem copies of the Ly-6/uPAR domain, whereas the others contain a single domain (2). The second subfamily contains snake and frog cytotoxins, principally three-finger neurotoxins. Its members lack the GPI-anchoring signal sequence and possess only eight cysteines. The Ly-6/uPAR receptors mediate several physiological functions including cellular activation, macrophage migration, trophoblast implantation, angiogenesis, and tumor cell invasion (3,4).
During mammalian fertilization, the acrosomal and plasma membranes of the sperm interact with the egg vestments. Sperm surface molecules on the plasma membrane are involved in the recognition of the zona pellucida protein (ZP3) during the early process of sperm egg interaction (primary binding) (5). Following sperm binding to the egg the acrosomal membranes are involved in a critical cascade of events that includes the following: 1) the acrosome reaction, in which the outer acrosomal membrane fuses with the plasma membrane; 2) secondary binding of the inner acrosomal membrane to the zona pellucida; and 3) tertiary binding of the equatorial segment and subsequent fusion to the vitelline membrane. After the acrosome reaction, the inner acrosomal membrane becomes the limiting membrane on the external surface of the anterior principal acrosomal segment, whereas the plasma membrane remains continuous with the outer acrosomal membrane at the anterior tip of the equatorial segment (6). The acrosome contains many enzymes including proteases, such as acrosin (7), esterases, acid phosphatases, glycohydrolases, and aryl sulfatases. Enzymes released from the acrosome are believed to play an important role in digesting the zona and paving the way for penetration of the egg vestments by the sperm (8).
Knowledge of the components of the acrosomal membranes and matrix is essential to understanding the biology of acroso-mal functions including the fusion of the outer acrosomal membrane with the plasma membrane during the acrosome reaction, the interaction of the inner acrosomal membrane with zona pellucida during penetration of the zona (secondary binding) and tertiary egg binding and fusion. Many details of sperm capacitation, sperm-egg interaction, and the roles of constituent proteins of the acrosome in these events are still unknown. Definition of the molecular constituents of the acrosomal matrix and acrosomal membranes is clearly of importance to understanding the molecular mechanisms that mediate induction of the acrosome reaction and sperm-egg interactions. Molecules identified as key players in these events are candidates for targeting by rational drug design or for inclusion in a spermbased contraceptive vaccine.
A major step toward identification of novel sperm molecules that may be crucial players in the molecular events leading to fertilization was initiated by Naaby-Hansen et al. (9) during a comprehensive analysis of human sperm proteins utilizing twodimensional gel electrophoresis to create a computerized data base (sperm protein encyclopedia). However, membrane molecules represent only a small subset of proteins in total cellular extracts, and specialized methods are required for their enrichment, isolation, and characterization. To increase the likelihood of identifying sperm membrane molecules that may be involved in the acrosome reaction and/or sperm-egg interaction, we adopted the strategy of Triton X-114 phase partitioning prior to two-dimensional gel electrophoresis and microsequencing by mass spectrometry (10). Several spots in this previous study yielded novel peptide sequences. From a protein spot of pI 4.8 and molecular mass of 14 kDa that yielded novel peptide sequences by mass spectrometry, we have cloned and characterized a membrane-associated protein localized to the acrosome and, transiently, to the plasmalemma of human spermatozoa. This protein, named SAMP14 (sperm acrosomal membrane-associated protein 14) on the basis of its localization, appears to be a member of the Ly-6 and urokinase plasminogen activator receptor family and play a role in sperm-egg binding and fusion.

EXPERIMENTAL PROCEDURES
Semen Samples-All semen samples for this study were obtained under informed consent using forms approved by the University of Virginia Human Investigation committee.
Animals-We used adult female Sprague-Dawley virgin rats of 6 -9 weeks for immunization. All experiments were performed in accordance with the Guidelines for the Use and Care of Laboratory Animals (National Institutes of Health Publication 85-23).
Surface Labeling, Phase Partitioning, and Two-dimensional Gel Electrophoresis-Human semen samples were obtained from healthy donors as described earlier (11). Triton X-114 phase partitioning was employed to enrich potential membrane proteins from human spermatozoa before or after surface labeling of sperm with sulfo-NHS-LCbiotin. Biotinylation of sperm was performed by vectorial labeling conditions described previously (10). To enrich for hydrophobic membrane proteins, Triton X-114 phase separation was performed as described by Bordier (12) and modified by Shetty et al. (10).
Two-dimensional gel electrophoresis was performed according to the method of O'Farrell et al. (13), with modifications (10). Isoelectric focusing was performed in 15 ϫ 0.15-cm acrylamide tube gels.
One-dimensional SDS-PAGE and Western Blot-Protein electrophoresis was performed on discontinuous polyacrylamide gels (4% stacking and 15% separating gel) on a Bio-Rad minigel apparatus or in 0.15-cm-thick 16 ϫ 16-cm slab gels. Gels were stained with Coomassie Blue or blotted. For Western blots, proteins were transferred to nitrocellulose membrane and processed as described previously (10).
Reverse Transcription-PCR-Oligonucleotides designed from an expressed sequence tag (EST) (AA778671) were manufactured by Invitro-gen. The forward primer (5Ј-TACCATATTTCATGGACTCTAAAGTGC-TAC-3Ј) and reverse primer (5Ј-GTCCGGTCATCAACAAAGGCTGCC-TG-3Ј) were utilized to generate a 475-bp probe. For subsequent PCRs, human testicular RNA (Clontech, Palo Alto, CA) was reverse transcribed as described by Wolkowicz et al. (14). 1-l aliquots of the cDNA solution were amplified for 40 cycles with denaturation at 94°C, annealing at 50°C for 30 s, and extension at 68°C for 3 min, as specified by the manufacturer of the recombinant Thermophilus thermophilus polymerase (PerkinElmer Life Sciences). Sequencing was performed by the University of Virginia Biomolecular Research Facility.
Northern and Dot Blot Analysis-Human multiple tissue northern membranes containing eight tissues and multitissue array RNA dot blots containing 76 tissues were purchased from Clontech. The PCRamplified 372-bp cDNA corresponding to the entire open reading frame of SAMP14 was used as a probe for both analyses. Both multiple tissue Northern and multitissue array RNA dot blot analysis were done as described previously (15).
Screening of cDNA Library-A DR2 cDNA library from human testis was purchased from Clontech and screened according to the user manual and as described earlier (15). A 475-bp cDNA fragment was PCR-amplified from the human testicular cDNA using two primers based on GenBank TM sequence AA778671 and used to probe filters. Probes were labeled with [ 32 P]deoxycytidine triphosphate by the random primer method using an oligolabeling kit (Amersham Biosciences). Filters were washed once in 2ϫ SSC, 0.2% SDS at 22°C; once in 0.2ϫ SSC, 0.2% SDS at 42°C; and once in 0.2ϫ SSC, 0.2% SDS at 50°C, each for 20 min. After washing, filters were exposed to x-ray film at Ϫ80°C for 24 -48 h. Secondary screening was repeated as above to isolate single, pure phage plaques. Finally, the cDNA inserts were recovered from the phage particles as plasmids in the pDR2 vector by transforming the Escherichia coli, AM1 strain. Sequencing was performed by the University of Virginia Biomolecular Research Facility in both directions, and the nucleotide and amino acid data were analyzed using the Genetics Computer Group and SEQWeb (Madison, WI) program.
Expression of Recombinant Protein and Production of Antiserum-To express SAMP14 in E. coli, a DNA fragment encoding amino acids (aa) 20 -100 (lacking both the putative signal peptide and COOH-terminal transmembrane domain) of the open reading frame was amplified by PCR. The amplified fragment was fused in frame with His tag at the COOH terminus of pET 28b using the NcoI and XhoI sites. The plasmid was then transformed into the host strain E. coli BL21 DE3. The recombinant protein was affinity-purified on a Ni 2ϩ -nitrilotriacetic acid column under denaturing conditions (16). The recombinant protein was further purified on a model 491 Prep Cell (Bio-Rad) to a single band, which was verified by SDS-PAGE and used for immunizations.
Female virgin Lewis rats (three rats per each group), weighing 160 -200 g, were injected with purified recombinant SAMP14 (recSAMP14) in PBS emulsified with equal amount of Freund's complete adjuvant. Each rat received subcutaneously 100 g of protein in 0.3 ml of PBS for the primary injection and each subsequent injection 3 and 6 weeks later. Antibody titers were checked by Western blot analysis 10 days after each injection, and the animals were sacrificed 10 days after the last boost. Specificity of the rat antisera for SAMP14 was tested by Western blotting against recSAMP14 as well as human sperm protein extracts.
Immunofluorescence and Electron Microscopic Analysis-Spermatozoa were purified by the swim-up method as described by Mandal et al. (16) and were air-dried onto poly-L-lysine-coated slides (Polysciences, Warrington, PA). Following three washes with PBS for 10 min each, slides were blocked in PBS containing 10% normal goat serum for 1 h with or without 0.1% saponin for permeabilization. The sperm were then incubated with a 1:200 dilution of anti-SAMP14 primary antiserum or preimmune serum for 1 h at 22°C, followed by washing in PBS three times for 10 min each at 22°C. Following incubation with the secondary goat anti-rat IgG conjugated with TRITC at 1:100 dilution for 1 h, the slides were washed in PBS, incubated for 5 min with equilibration buffer followed by Slow Fade (Molecular Probes, Eugene, OR), and mounted with coverslips. Images were captured with an Axioplan fluorescence microscope (Zeiss, Oberkochen, Germany).
Processing of pooled sperm for electron microscopy was done as described earlier (15). To stain the sections, they were first blocked in undiluted normal goat serum for 15 min at 22°C. They were then incubated for 16 h at 4°C with either preimmune or rat anti-SAMP14 serum at a dilution of 1:50 containing 1% normal goat serum, 1% bovine serum albumin, and 0.1% Tween 20. After washing, the sections were incubated with a 1:100 dilution of 5-nm gold-conjugated goat-anti-IgG (Goldmark Biologicals, Phillipsburg, NJ) for 1.5 h at 22°C. The sections were washed in distilled water, stained with uranyl acetate, and ob-served with a JEOL 100CX electron microscope.
Acrosome Reaction and Double Staining with Concanavalin A-Human semen samples were obtained from healthy donors as described earlier (11). Live "swim-up" sperm were prepared as described by Mandal et al. (16). Capacitation was conducted in BWW medium (Irvine Scientific, Santa Ana, CA) containing sodium bicarbonate and bovine serum albumin at 37°C overnight (17). Following incubation in capacitating medium, 0.25 M progesterone in absolute ethanol was added to a final concentration of 0.5 mM to induce the acrosome reaction. Aliquots were incubated for 30 min at 37°C in a CO 2 incubator. The cells were washed three times with PBS and air-dried onto poly-L-lysinecoated glass slides. Slides were washed three times in PBS, overlaid with 2.5 g/ml fluorescein isothiocyanate-conjugated concanavalin A (ConA; Sigma) in PBS, and incubated for 30 min in a humid chamber at room temperature. In cases where double staining was done, fluorescein isothiocyanate-conjugated ConA was mixed with the secondary goat-anti-rat IgG conjugated to TRITC and overlaid after the slide was incubated with anti-SAMP14 primary antibody and washed with PBS. The slides were washed with PBS, overlaid with equilibration buffer, and mounted with Slow-Fade reagent as described earlier.
Hamster Egg Penetration Assay-Gamete incubations were carried out in microdrops under paraffin oil at 37°C and 5% CO 2 . The swim-up sperm prepared as described above were washed twice by centrifugation (8 min at 600 ϫ g) in 8-ml volumes of BWW medium containing 5 mg/ml human serum albumin (Sigma) in 15-ml centrifuge tubes. Sperm were capacitated at a concentration of 20 ϫ 10 6 sperm/ml overnight in 250-l drops of BWW containing 30 mg/ml human serum albumin. Hamster ova were obtained from Golden Syrian hamsters by supraovulation (18). Cumulus cells were removed by treating eggs with 0.05% hyaluronidase (Sigma) for 3 min. The oocytes were then pooled and washed through 20-l drops of medium using a pulled, heat-polished Pasteur pipette. Zona pellucidae were removed by treating eggs with 1 mg/ml trypsin for 30 s followed by five washes. The eggs were then randomly allotted into treatment groups.
Rat serum IgG was isolated using Nab TM protein G spin chromatography kit (Pierce) according to the manufacturer's protocol. Protein concentration was measured by reading the OD at 280 nm. Following overnight capacitation, 2 ϫ 10 6 sperm/ml were treated for 1 h in a 20-l drop containing 100, 50, 20, 10, or 5 g/ml of IgG from rat anti-SAMP14 serum in BWW containing 30 mg/ml human serum albumin. One hundred micrograms/ml of preimmune IgG was used as a negative control and 100 g/ml of SAMP32 (15) IgG was used as a positive control. Following sperm/antibody incubation, zona free hamster oocytes were added directly to the sperm suspension, and the gametes were coincubated for 3 h. Following gamete co-incubation, loosely bound sperm were removed from the oocytes by gentle pipetting.
The eggs were then treated with 1 mM acridine orange in 3% Me 2 SO (Sigma) for 15 s to stain the chromatin and washed through three 20-l microdrops. To quantitate binding, oocytes were placed between a microscope slide and an elevated coverslip, and the number of sperm bound per oocyte was recorded. The number of sperm fused per egg was scored by counting the number of acridine orange-stained decondensed sperm heads within each oocyte using fluorescent microscopy. The assay was repeated at least three times for each concentration of IgG used. Specific effect of the antibody was verified by preabsorbing the antibody with an excess amount of recombinant SAMP14. Experimental and control group averages were reported as means Ϯ S.E. Student's t test was used to identify differences in the mean numbers of bound or fused spermatozoa. p Ͻ 0.05 was considered to be statistically significant.
Treatment of Spermatozoa with PIPLC-Percoll-harvested sperm (200 ϫ 10 6 ) were incubated with 1.0 unit/10 8 sperm/ml of PIPLC from Bacillus cereus (Molecular Probes, Inc., Eugene, OR) or without enzyme for 20 min at 22°C. At the end of the incubation time, sperm were pelleted by centrifugation at 1000 ϫ g for 30 min. To prevent proteolysis, 2 mM phenylmethylsulfonyl fluoride and 5 mM EDTA were added to the supernatants, which were collected and dialyzed against distilled water at 4°C overnight. The dialyzed solution was concentrated by lyophilization, and the proteins were reconstituted in Celis lysis reagent (19) before loading for isoelectric focusing.

RESULTS
Phase Partitioning and Surface Labeling-Among several proteins that were selectively partitioned into the Triton X-114 detergent phase, a major protein spot of 14 kDa and pI 4.8 was identified on a silver-stained two-dimensional gel (Fig. 1A) and was determined to be reactive with streptavidin horseradish peroxidase following sperm surface labeling with sulfo-NHS-LC-biotin (Fig. 1B). These results suggested that the spot was a hydrophobic protein and was accessible to cell surface biotinylation, characteristic features of a membrane protein. Microsequencing of the protein spot using mass spectrometry yielded two peptides, ATSCGLEEPVSYR and ---XSDSMEC--- (Fig. 1A).
SAMP14 Encodes a Novel Protein Showing Similarity to Ly-6/uPAR Family of Proteins-Use of the peptide sequences described above to search the nonredundant protein data bases revealed no known gene, suggesting that this protein was novel. However, an exact match was found in the EST data base (entry AA778671). PCR was performed to amplify a hybridization probe from a testicular cDNA library using primers based on the EST sequences. After screening the DR2 human testis cDNA library, six clones were identified at the secondary screening stage. DNA sequencing of the largest clone (1-2) yielded a 475-bp cDNA, which contained the EST AA778671 C, two-dimensional blot of human sperm proteins probed with rat serum against purified recSAMP14. The 14-kDa region (pI ϳ4.8) originally microsequenced bound the antibody (arrow). In addition, the antibody reacted with several spots on the basic and acidic sides of the original excised spot, indicating the existence of SAMP14 charge variants in sperm. The preimmune serum did not react with any proteins (data not shown). D and E, two-dimensional gel electrophoresis followed by Western blotting of supernatants from human sperm treated with PIPLC using rat anti-recSAMP14. Sperm purified by swim-up were separated into two groups (ϳ150 ϫ 10 6 sperm/group) and treated with or without 3 units/10 8 /ml sperm of PIPLC. The control group that was not exposed to PIPLC (D) released a small amount of one isoform of immunoreactive SAMP14 into the medium. However, treatment with PIPLC (E) released additional SAMP14 isoform into the medium including several more charge variants (arrowheads). An acidic shift in the pI of the PIPLC-released isoforms accompanies loss of the GPI anchor (arrowheads in E correspond to the location of different isoforms in C).
sequence embedded within and a contiguous open reading frame encoding 124 amino acids (Fig. 2). The predicted molecular mass and pI of 13 kDa and 5.4, respectively, closely matched the molecular weight and pI of the excised spot. The peptide microsequence ATSCGLEEPVSYR obtained by mass spectrometry was present in the cloned protein. A comparison of the deduced sequence with the GenBank TM data base using BLAST and FASTA as well as Prosite and Prodom revealed that the predicted protein had the closest amino acid similarity to proteins from the Ly-6 and uPAR superfamily. Fig. 3 shows a multiple sequence alignment of SAMP14 and other Ly-6/ uPAR superfamily members with highest homology to SAMP14. One of the unique features of these molecules is the presence of a cysteine-rich domain that is conserved among all of the members (see asterisks). The highly conserved motif Cys-Cys-Xaa-Xaa-Asp-Leu-Cys-Asn, which is a signature of proteins belonging to this family, was located at the COOH terminus of many of these sequences, including SAMP14. Alignment of SAMP14 with three tandem Ly-6 domains of uPAR (uPAR-1, uPAR-2, and uPAR-3) showed that two residues (tyrosine and leucine) located in the uPAR-1 functional epitope involved in ligand binding (20) are conserved in SAMP14, suggesting that SAMP14 may be closely related to uPAR. site between aa 19 and 20; 2) a Ly-6/uPAR-like domain from aa 20 to 112; and 3) a transmembrane domain from aa 101 to 124 (Fig. 2). Based on the presence of a putative transamidase cleavage site following aa 97 in the COOH terminus of the protein, SAMP14 was predicted to be a GPI-anchored protein.
In addition, putative threonine (aa 69, 83, and 88) and serine (aa 78) phosphorylation sites were identified. A scan against the protein profile data base including Prosite predicted protein kinase C phosphorylation sites at aa 78 -80 and 92-94.
Expression of SAMP14 Is Testis-abundant if Not Testis-specific-Northern blot (Fig. 4A) and dot blot (Fig. 4B) hybridization were used to analyze the expression of SAMP14 mRNA in different human tissues. SAMP14 was detected only in testis as a 1.3-kb single transcript among eight tissues including spleen, thymus, prostate, ovary, small intestine, colon, and peripheral lymphocytes on a multitissue Northern blot. This specificity was further confirmed by RNA dot blot hybridization of 76 human tissues, in which only testis yielded a positive signal (Fig. 4B). A search against the EST data base revealed many matches to ESTs of testis origin and a single match to an EST of hippocampal origin.
SAMP14 Has a Mouse Orthologue-When the cDNA sequence of SAMP14 was searched against the human genome data base, SAMP14 was localized to chromosome 19 (19q 13.33). A BLAST search against the mouse nucleotide sequence data base identified an orthologous mouse testis cDNA (AK005760) cloned by the Riken mouse genome project. Human SAMP14 showed 81% identity to the putative mouse homologue. The mouse genomic sequence for SAMP14 was localized to chromosome 7B2, which is syntenic to locus 19q13 of the human (21). In cooperation with the Human Genome Nomenclature Committee, SAMP14 was given the locus name SPACA4, which stands for sperm acrosome membrane-associated 4.

Expression of SAMP14 and Western Blot Analysis-
The cDNA sequence encoding a processed SAMP14 protein from residue 20 to 100 (lacking the signal peptide and putative transmembrane domain) was cloned into the bacterial expression vector pET28b. After induction with isopropyl-1-thio-␤-Dgalactopyranoside for 4 h, recSAMP14 with a molecular size of ϳ10.5 kDa was produced (data not shown). The recombinant protein was affinity-purified with a Ni 2ϩ -nitrilotriacetic acid column under denaturing conditions. The affinity-purified rec-SAMP14 was further purified on a Prep Cell and used as an immunogen to inoculate rats. The resulting antibody to rec-SAMP14 recognized the "native" SAMP14 present in sperm extracts. When the two-dimensional blots containing human sperm Triton X-114 detergent phase extracts were probed with recSAMP14 antibody (Fig. 1C), the protein spot that was cored for microsequencing was recognized. The antiserum also recognized several spots on either side of the original cored spot, which are probably charge variants of SAMP14. Interestingly, these "charge variants" were poorly stained on a silver-stained gel (see Fig. 1A), indicating that they may be present in low abundance.
SAMP14 Is Associated with the Human Sperm Acrosome-Indirect immunofluorescence analysis of fresh air-dried, ejaculated swim-up human spermatozoa using rat serum against recSAMP14 localized SAMP14 to the entire sperm when sperm were washed only once after the swim-up procedure (Fig. 5B). However, when the sperm were washed at least 3-5 times after the swim-up, the immunofluorescent staining was restricted to the acrosomal region ( Fig. 5F) with some sperm showing more intense staining at the equatorial region (arrows). This experiment suggested that some of the antigen is loosely associated with the sperm surface and removed during the washing procedure.

FIG. 3. Multiple amino acid sequence alignment of SAMP14 and other members of the Ly-6/uPAR superfamily.
Residues shaded with a gray background are identical in all sequences. The conserved cysteine residues are marked by asterisks. Numbers on the side refer to amino acid positions relative to the first residue. In the case of the urokinase receptor, uPAR, the three sequential Ly-6 domains have been placed in alignment and labeled UPAR1, -2, and -3. For SP10, the COOH-terminal region of the protein having a Ly-6 domain (30) is aligned. The residues in black form a part of the functional epitope involved in ligand binding in uPAR1 (20). Note that these residues in addition to the adjoining amino acids are conserved in human SAMP14 (boxed). Accession numbers are gi5701858 (G6E_hum), gi21312278 (SAMP14_mouse), gi70151 (CD59_hum), gi 65003 (uPAR_hum), gi10720240 (prostate stem cell antigen; PSCA_human), and gi1042049 (SP-10_hum).
Double staining showed that acrosome-reacted sperm, as determined by more intense staining with ConA (Fig. 6B, boxed  sperm), also showed a cap-shaped reactivity with recSAMP14 antibody (Fig. 6C, boxed sperm). This suggested that a substantial portion of SAMP14 is intra-acrosomal and that a population of SAMP14 is retained on the inner acrosomal membrane of acrosome-reacted sperm. At the ultrastructural level, the antiserum stained mainly the outer and inner acrosomal membranes, although some gold particles were also seen in the acrosomal matrix (Fig. 7A). In the majority of the cases, the gold particles were lined up along the inner acrosomal membrane when acrosome reacted sperm were examined (Fig. 7B).
we tested whether anti-recSAMP14 IgG would block penetration of zona free hamster eggs by capacitated human sperm (Fig. 8). A rat antibody to SAMP32 (15), an acrosomal antigen previously shown to inhibit the hamster egg penetration test, was used as a positive control in the assay. This control showed a significant inhibition of binding (48%) and fusion (57%) to the eggs at 100 g/ml concentration. Significant effects on binding were observed at 100, 50, and 20 g/ml rat anti-SAMP14 IgG and on fusion at 100 and 50 g/ml. This dependence on antibody dose suggests specificity of the inhibition. In three independent assays, for example, the antibody suppressed sperm binding to hamster eggs by roughly 47% (p Ͻ 0.001) at the 50 g/ml concentration, and blocking of sperm fusion was more pronounced at about 63% inhibition (p Ͻ 0.001). The inhibition of binding was completely abolished when the antibody was preincubated with a 10-fold excess of recombinant protein. DISCUSSION The results of this study indicate that SAMP14 is an acrosomal membrane-associated member of the GPI-anchored Ly-6/uPAR family of glycoprotein receptors. In its NH 2 terminus, SAMP14 contained a single Ly-6/uPAR domain including 10 cysteine residues. At the COOH end of this domain, the highly conserved sequence of this family, Cys-Cys-Xaa-Xaa-Asp-Leu-Cys-Asn, was present with arginine substituted for aspartic acid at position 94. PIPLC treatment released a series of No immunostaining was observed with preimmune sera (D and H) or with sera obtained from rats injected with adjuvant alone (data not shown).
FIG. 6. Double staining of acrosome-reacted sperm with ConA and anti-recSAMP14 antibody. Phase-contrast (A) and indirect immunofluorescence staining (B and C) of SAMP14 in capacitated and acrosomereacted, unpermeabilized human spermatozoa. Following incubation with primary antibody, the sperm were double-stained with ConA-fluorescein isothiocyanate (B) and a secondary antibody conjugated to TRITC (C) to visualize the staining of SAMP14 in sperm that have undergone a complete acrosome reaction. Acrosome-reacted sperm as shown by ConA staining (boxed sperm) also show staining with SAMP14 antibody (boxed in C), indicating that SAMP14 is located in the acrosomal region and is probably retained on the inner acrosomal membrane. Note in A that the width of the sperm heads is narrower in those sperm stained with ConA. SAMP14 charge variants, indicating that several SAMP14 isoforms are GPI-anchored. This finding is in concert with the presence of a canonical transamidase cleavage site, a characteristic of GPI-anchored proteins, between amino acids 97 and 98.
In humans, several members of the Ly-6 family containing a single Ly-6/uPAR domain map to chromosome 8, whereas the uPAR (CD87) locus PLAUR, encoding three tandem Ly-6/ UPAR domains, is found at 19q13.2 (22). SAMP14 mapped to the SPACA4 locus at 19q13.33, close to the PLAUR locus. Further, the transcript size of SAMP14, as shown by Northern blot analysis, is ϳ1.3 kb. This is larger than other human Ly-6 family members such as E48 with a transcript size of 0.8 kb (23) or prostate stem cell antigen with a transcript size of ϳ1 kb (24) and is closer to the uPAR transcript of 1.4 kb (25). Last, 2 of the 4 residues of the functional epitope of uPAR that were previously identified to be involved in ligand binding (20) are also conserved in the Ly-6 domain of SAMP14. All of these observations suggest that SAMP14 is more closely related to uPAR than other family members, and functions assigned to uPAR should be considered in hypotheses regarding the role of SAMP14 in acrosomal biology. uPAR binds to urokinase plasminogen activator through the NH 2 terminus (2), the region conserved in SAMP14. uPAR has a role in generating proteolytic activity that is spatially restricted to the plasma membrane, is involved in cell-cell and cell matrix interactions, and participates in signal transduction events (26). Of further interest is the fact that neutrophils contain reservoirs of uPAR in intracellular granules that are translocated to the plasma membrane upon activation (27), a process that may be considered analogous to the exposure of SAMP14 on the inner acrosomal membrane following the acrosome reaction. Together these considerations suggest that SAMP14 may function as an acrosomal membrane receptor in the orphan class, with its ligand yet undefined, and may participate in proteolytic or adhesive events involving the acrosomal membrane.
A search of the Riken mouse genome data base revealed a putative orthologous gene at chromosome 7B2 and a cDNA from mouse testis. Comparison of the mouse and human SAMP14 sequences suggests that both have a Ly-6/uPAR domain with a potential GPI anchor at the COOH terminus, and evolutionarily there is little change in this sequence. The tight linkage of all other murine Ly-6 proteins on chromosome 15 in contrast to the location of murine uPAR on chromosome 7, which is syntenic to human chromosome 19, further confirms that SAMP14 is closely related to uPAR.
One GPI-anchored Ly-6 family member, CD59, has been identified previously on the human sperm surface (28). In lymphocytes, CD59 is a multifunctional protein involved in complement regulatory activity and co-stimulatory activity for human T cells mediated via the CD2 T cell antigen. Apart from having a probable function of protection of sperm from complement-mediated damage, CD59 is also predicted to be involved in cell-cell adhesion during gamete interaction (29). Another Ly-6 family member that is specifically expressed in testis is SP-10, an intra-acrosomal protein that is involved in secondary binding to the egg (30 -32). Among acrosomal membrane proteins, PH-20 is the only GPI-anchored protein characterized previously (33)(34)(35)(36)(37)(38). On acrosome-intact human sperm, PH-20 is located on the plasma membrane over the entire head. However, after the acrosome reaction, PH-20 is also located on the inner acrosomal membrane (38). There are also several GPIanchored proteins reportedly acquired by the sperm membrane during post-testicular transport. Among them is CD52, an antigen that is secreted by epithelial cells of the distal epididymis and ductus deferens and acquired by sperm during their passage through the tract (39,40). Several other GPI-anchored antigens found on the cells of the immune system (namely CD62, CD55, and CD73) are also found anchored to the sperm surface (41). The functions of these GPI-anchored molecules in sperm is largely unknown.
A Northern blot of eight human tissues initially indicated that the RNA transcript is about 1.3 kb long and that SAMP14 is expressed only in testis. This testis specificity of SAMP14 expression was further reinforced by our comprehensive study of RNA expression by RNA dot blot hybridization to 76 human tissues. In agreement with this designation of SAMP14 as a testis-specific gene, EST data base searches during the research phase of this project did not reveal any other tissues expressing SAMP14, but a recent search revealed a brain EST from hippocampus. The report of an EST from brain contrasts with the absence of reactivity in the dot blot that contained RNA from the same tissue. Because of the low incidence of brain ESTs to testis ESTs in the data base, this single brain EST may or may not indicate translated protein in the brain. Antibody to recSAMP14 reacted with the original protein spot cored for peptide sequencing on two-dimensional Western blots. This result, coupled with the finding of one identical microsequence in the protein sequence derived from the cloned cDNA, indicates that the protein corresponding to the microsequenced spot has been successfully cloned. However, lack of the other microsequence in the predicted protein sequence suggests that there may be another protein co-migrating with SAMP14. The antiserum to SAMP14 recognized several additional bands on the basic and acidic sides of the original cored spot that appear to be charge variants of SAMP14, suggesting that post-translational modifications of SAMP14 are likely. Based on the two-dimensional gels of PIPLC-treated sperm, at least three SAMP14 isoforms appear to be GPI-anchored. These were not visualized on a Coomassie-or silver-stained gel, indicating that they are present in lower amounts compared with the major form of SAMP14 at pI 4.8.
In the initial experiments, SAMP14 was enriched in the detergent phase, suggesting that it is a hydrophobic protein and thus possibly associated with a membrane. Upon cloning the cDNA, a putative transmembrane domain (aa 101-124) was found in the predicted protein, and bioinformatic analysis of this domain revealed a potential transamidase cleavage site predicting a GPI-anchored protein. Further, SAMP14 was found to be sensitive to PIPLC, confirming that it is a GPIanchored protein, as are several other members of the Ly-6/ uPAR family. However, treatment with PIPLC did not release all SAMP14, since some of it remained in the sperm pellet following the separation of the supernatant (data not shown), suggesting that 1) there may be a population of GPI-anchored SAMP14 that is resistant to PIPLC; 2) all of the SAMP14 is not accessible to the enzyme; or 3) a non-GPI-anchored population exists. The presence of some soluble SAMP14 in the control samples in the PIPLC release studies suggests a non-GPIanchored population of SAMP14. Regarding the presence of PIPLC-resistant SAMP14, it is well documented that many of the GPI-anchored proteins have a PIPLC-resistant subpopulation (42).
Immunofluorescence studies of ejaculated sperm, immunolocalization on enzyme-dissociated testicular cells, and electron micrographic studies all showed that SAMP14 is mainly located in the acrosome. However, indirect immunofluorescence study of unwashed, air-dried sperm showed the presence of the protein on the entire surface of the sperm. There are at least two possible explanations for this surface staining. First, SAMP14 might be produced in another location along the male reproductive system so that it coats the sperm when they traverse the tract. Second, SAMP14 may be expressed only in the acrosome, but some of the protein may leach into the seminal plasma due to sperm damage or during spontaneous acrosome reactions. In any case, coating the sperm with SAMP14 in this way is in concert with the known adhesive property of many members of Ly-6 and uPAR family of proteins. The labeling of the SAMP14 spot with biotin may be due to this population of SAMP14 on the plasma membrane.
The hamster egg penetration assay demonstrated that antibody against SAMP14 significantly inhibited both the binding and fusion of sperm with zona free hamster eggs, which suggests that SAMP14 may have a role in one or more events leading to fertilization. It is well known that the equatorial region of the sperm contacts the vitelline membrane prior to sperm internalization. Further, following the acrosome reaction, the inner acrosomal membrane is known to come directly in contact with the egg, leading to the later events of egg penetration. Considering that the concentrations of SAMP14 were highest in the posterior acrosome close to the equatorial region and that it is retained and localized in the inner acrosomal membrane, it is not surprising that the antibody to SAMP14 inhibited both sperm-egg binding and fusion. It is appropriate to add here that another acrosomal membrane protein SAMP32, which was discovered recently (15), also localized to the inner acrosomal membrane of the principal piece and equatorial segment. Further, anti-SAMP32 blocked the penetration of zona free hamster eggs by human sperm.
Although a large number of potential candidate binding partners for sperm-egg interaction have been described, including galactyltransferase, Sp56, zona receptor kinase, spermadhesins, PH20, proacrosin, Sp38, and Sp17, the precise roles of these molecules have not been firmly established (43). Hence, the identification of additional molecules that participate in gamete interactions, particularly those retained after the remodeling events of acrosomal exocytosis, could greatly enhance the understanding of this process. SAMP14 is an attractive addition to this group. SAMP14 may be a signaling molecule like PH20 (44), or a SAMP14 ligand may exist on the oolemma. By analogy to uPAR, SAMP14 may play a role in spatial restriction of proteolytic processes at the surface of the inner acrosomal membrane. As a Ly-6/uPAR family member, FIG. 8. Sperm binding and fusion in the hamster egg penetration assay. Capacitated human sperm, preincubated with different concentrations of affinity-purified IgG from immune (IMM) serum against SAMP14 or 100 g/ml of IgG from preimmune serum, were co-incubated with zona free hamster eggs. The numbers of sperm bound to or fused to the eggs were counted under a fluorescent microscope. Affinity-purified IgG from anti-SAMP32 (15) serum at a 100 g/ml concentration was used as a positive control. Bars represent means Ϯ S.E. of 3-5 individual experiments. n ϭ total number of oocytes per group. *, p Յ 0.05; **, p Յ 0.001 (Student's t test). SAMP14 merits further evaluation as a GPI-linked receptor that is retained on the inner acrosomal membrane after acrosomal exocytosis.