Mouse Sperm Lacking ADAM1b/ADAM2 Fertilin Can Fuse with the Egg Plasma Membrane and Effect Fertilization*

Fertilin, a heterodimeric protein complex composed of α (ADAM1) and β (ADAM2) subunits on the sperm surface, is believed to mediate adhesion and fusion between the sperm and egg plasma membranes. Here we have shown that mutant male mice lacking ADAM1b are fertile and that the loss of ADAM1b results in no significant defect in sperm functions such as migration from the uterus into oviduct, binding to egg zona pellucida, and fusion with zona pellucida-free eggs. ADAM1b-deficient epididymal sperm showed a severe reduction of ADAM2 on the cell surface, despite the normal presence of ADAM2 in testicular germ cells. The appearance of ADAM1b and ADAM2 on the sperm surface depended on formation and abundance of ADAM1b/ADAM2 fertilin in testicular germ cells. These results suggest that mouse ADAM1b/ADAM2 fertilin may play a crucial role not in the sperm/egg fusion but in the appearance of these two ADAMs on the sperm surface.

Fertilin, a heterodimeric protein complex composed of ␣ (ADAM1) and ␤ (ADAM2) subunits on the sperm surface, is believed to mediate adhesion and fusion between the sperm and egg plasma membranes. Here we have shown that mutant male mice lacking ADAM1b are fertile and that the loss of ADAM1b results in no significant defect in sperm functions such as migration from the uterus into oviduct, binding to egg zona pellucida, and fusion with zona pellucida-free eggs. ADAM1b-deficient epididymal sperm showed a severe reduction of ADAM2 on the cell surface, despite the normal presence of ADAM2 in testicular germ cells. The appearance of ADAM1b and ADAM2 on the sperm surface depended on formation and abundance of ADAM1b/ADAM2 fertilin in testicular germ cells. These results suggest that mouse ADAM1b/ADAM2 fertilin may play a crucial role not in the sperm/ egg fusion but in the appearance of these two ADAMs on the sperm surface.
Members of the ADAM (a disintegrin and metalloprotease) 3 family are multifunctional, transmembranous proteins consisting of pro-, metalloprotease, disintegrin, cysteine-rich, epidermal growth factor-like, transmembrane, and cytoplasmic tail domains (1)(2)(3). The ADAM proteins are involved in the regulation of membrane fusion, cell-to-cell adhesion, cell migration, and shedding of cytokines, growth factors, and their receptors in the processes such as fertilization, neurogenesis, myogenesis, cancer, and inflammation (3). Although at least 40 ADAM genes have been identified in a variety of species, 4 only the roles of some ADAMs have been elucidated. In particular, there are many members of the ADAM family that are exclusively or predominantly expressed in the testis but are functionally uncharacterized yet.
Mammalian fertilization requires sperm to adhere/bind to and penetrate through the zona pellucida (ZP), an extracellular glycoprotein matrix surrounding the egg, and then to fuse with the egg membrane (4,5). Fertilin, a heterodimeric protein complex composed of ␣ (ADAM1) and ␤ (ADAM2) subunits present on the sperm surface, has been believed to mediate adhesion and fusion between the sperm and egg plasma membranes, because the cysteine-rich and disintegrin domains of ADAM1 and ADAM2 contain the putative sequences that resemble the fusogenic peptide of viral fusion proteins and the integrin binding Arg-Gly-Asp tripeptide of snake venom toxins, respectively (6 -10). In mouse, two different isoforms of ADAM1, ADAM1a and ADAM1b, are synthesized in the testis, and only ADAM1b is present on the plasma membrane of epididymal sperm (11,12). The functional roles of ADAM1a (13), a protein resident within the endoplasmic reticulum (ER) of testicular germ cells (TGC), ADAM2 (14,15), and ADAM3 known as cyritestin (15,16) have been evaluated by analysis of null mutant mice. The loss of ADAM1a or ADAM2 resulted in male infertility; sperm of ADAM1a-or ADAM2-deficient mice were incapable of ascending from the uterus into the oviduct through the uterotubal junction and of binding to the egg ZP (13,14). The inability of these mutant sperm to bind the ZP may be explained by a negligibly low level of ADAM3 on the sperm surface (13,15). Indeed, ADAM3-deficient male mice were infertile, and the mutant sperm showed the defect in the ZP binding (15,16). Interestingly, the loss of ADAM2 in TGC resulted in the lack of both ADAM1b and ADAM3, in addition to ADAM2, on the sperm surface (15,17). Because ADAM2-deficient mouse sperm were also defective in adhesion to and fusion with the egg membrane (14,15), the functional roles of ADAM1b, ADAM2, and ADAM1b/ADAM2 fertilin complex in fertilization are still controversial.
In this study, we have produced mice carrying a null mutation in Adam1b using homologous recombination. Unexpectedly, the male mice lacking ADAM1b are fertile, and ADAM1b-deficient epididymal sperm are functionally normal in the migration into the oviduct, binding to the ZP, and fusion with the egg. Despite the normal presence of ADAM2 in TGC, only ADAM2 among membranous proteins tested is severely reduced on epididymal sperm of the ADAM1b-deficient mice. On the basis of the experimental results on four lines of mutant mice lacking ADAM1a, ADAM1b, ADAM2, or ADAM3, we conclude that, at least in the mouse, ADAM1a/ADAM2 fertilin is responsible for the appearance of ADAM3 on the sperm surface, whereas ADAM1b/ ADAM2 fertilin is responsible for the appearance of ADAM1b and ADAM2 on the sperm surface.

EXPERIMENTAL PROCEDURES
Generation of Mutant Mice Lacking ADAM1b-A targeting vector containing an expression cassette of the neomycin-resistance gene, neo, flanked by 7.5-and 1.45-kbp genomic regions of Adam1b was constructed by using a mouse genomic clone, mFAG1 (11), encoding ADAM1b (Fig. 1A). For negative selection, the MC1 promoter-driven herpes simplex virus thymidine kinase gene was inserted at the 3Ј-end of the targeting vector. The vector was linearized by digestion with SalI and electroporated into mouse D3 embryonic stem cells, and homologous recombinants were selected by using G418 and ganciclovir as described previously (18). Seven embryonic stem cell clones carrying the targeted mutation were identified from 450 clones resistant to G418 and ganciclovir and injected into C57BL/6 mouse blastocysts. Chimeric male mice were crossed to C57BL/6 female mice (Japan SLC Inc.) to establish heterozygous mutant lines. Homozygous mice were obtained by mating of heterozygous males and females. All animal experiments were carried out according to the Guide for the Care and Use of Laboratory Animals at the University of Tsukuba.
Blot Hybridization-Genomic DNA was prepared from mouse tail, digested by EcoRI, separated by agarose gel electrophoresis, and transferred onto Hybond-N ϩ nylon membranes (Amersham Biosciences). Total cellular RNA was prepared from testicular tissues using Isogen (Nippon Gene, Toyama, Japan) as described previously (19). The RNA samples were glyoxylated, separated by agarose gel electrophoresis, and transferred onto the nylon membranes. The blots were probed by 32 Plabeled DNA fragments and analyzed by a BAS-1800II Bio-Image analyzer (Fuji Photo Film, Tokyo, Japan) as described (20).
Immunochemical Analysis-Testicular tissues from 3-to 4-monthold mice were minced with a razor blade in 4 mM Hepes-NaOH, pH 7.4, containing 140 mM NaCl, 4 mM KCl, 10 mM glucose, and 2 mM MgCl 2 , filtered through a nylon mesh, and centrifuged at 1,200 ϫ g for 10 min at 4°C as described (15). The cell pellet was suspended in the same buffer, and the suspension was put on a 52% Percoll gradient (Amersham Biosciences) in the above buffer and centrifuged at 11,000 ϫ g for 10 min at 4°C. TGC was then recovered from a white band near the top of the gradient and washed three times with phosphate-buffered saline. TGC and cauda epididymal sperm were suspended in a lysis buffer consisting of 20 mM Tris-HCl, pH 7.4, 1% Triton X-100, 150 mM NaCl, and 1% protease inhibitor mixture (Sigma-Aldrich), kept on ice for 20 min, and centrifuged at 10,000 ϫ g for 10 min at 4°C as described (12). Proteins in the supernatant solution were denatured by boiling for 5 min in the presence of 1% SDS and 1% 2-mercaptoethanol, separated by SDS-PAGE, and transferred onto Immobilon-P membranes (Millipore). After blocking with 1% skim milk, the blots were incubated with primary antibodies for 2 h and then with horseradish peroxidase-conjugated secondary antibodies for 1 h. Immunoreactive proteins were detected by an ECL Western blotting detection kit (Amersham Biosciences), and the intensities of the immunoreactive protein bands were quantified by using a Fuji Film Science Lab Image Gauge software (version 3.4). Protein concentration was determined using a Coomassie protein assay reagent kit (Pierce). Immunoprecipitation analysis was carried out according to the previously published method (13).
Sperm Migration into Oviduct-Female B6C3F1 mice were caged with male mice 12 h after intraperitoneal injection of human chorionic gonadotropin (Teikoku Zoki Co.), and formation of vaginal plug was observed every 30 min as described previously (13,27). The oviducts were excised with a connective part of the uterus ϳ2 h after copulation, fixed in phosphate-buffered saline containing 4% paraformaldehyde, washed with phosphate-buffered saline, and frozen in OCT compound (Sakura Finetechnical Co., Tokyo, Japan). Sections were stained with hematoxylin and observed under an Olympus BX50 microscope (Tokyo, Japan) equipped with an HC-2500 camera (Fuji Film, Tokyo, Japan).
In Vitro Fertilization-Eggs tightly packed with cumulus cells were collected from the oviductal ampulla of superovulated BDF1 mice 14 to 15 h after human chorionic gonadotropin injection and placed in a 0.1-ml drop of a modified Krebs-Ringer bicarbonate solution (TYH medium) (28) covered with mineral oil. Fresh cauda epididymal sperm from 3-month-old mice were capacitated by incubation for 2 h in a 0.2-ml drop of TYH medium at 37°C under 5% CO 2 in air. An aliquot (1.5 ϫ 10 4 sperm/10 l) of the capacitated sperm suspension was mixed with the eggs in a 90-l drop of TYH medium. After incubation at 37°C under 5% CO 2 in air, the eggs were treated with bovine testicular hyaluronidase (3 units/ml, Sigma-Aldrich) for 10 min to remove cumulus cells. The female and male pronuclei in the eggs were fixed in 4% paraformaldehyde for 15 min, stained with Hoechst 33342 (2 g/ml) for 30 min, and then viewed under an Olympus BX70 epifluorescence microscope as described (13).
Sperm-ZP Binding-Superovulated eggs were briefly treated with bovine hyaluronidase (3 units/ml), washed, and placed in a 0.1-ml drop of TYH medium covered with mineral oil. An aliquot (1.5 ϫ 10 4 sperm/10 l) of capacitated sperm suspension was added to a 90-l drop containing cumulus-free eggs and two-cell embryos in TYH medium, and the mixture was incubated for 30 min at 37°C under 5% CO 2 in air. The eggs were transferred to 100 l of fresh TYH medium, washed by pipetting, and fixed in 0.25% glutaraldehyde. The number of sperm tightly bound to the egg ZP was counted under an Olympus BX71 microscope equipped with a DP-12 camera as described (29). The twocell embryos were used as an internal negative control for nonspecific binding, and the average number of bound sperm/two-cell embryo was Ͻ1.0 under the above conditions.
Sperm-Egg Fusion-ZP-free eggs were mechanically prepared by using a piezo-driven micromanipulator (Prime Tech Ltd., Ibaraki, Japan) without the use of ␣-chymotrypsin as described previously (30). The ZP-free eggs previously treated with Hoechst 33342 (2 g/ml) were inseminated with capacitated sperm (1.5 ϫ 10 5 cells/ml) followed by incubation for 30 min at 37°C under 5% CO 2 in air. After fixation in 0.25% glutaraldehyde, the eggs were observed under the above fluoromicroscope.

RESULTS
To explore the functional role of ADAM1b in fertilization, we produced mutant mice lacking ADAM1b by homologous recombination in embryonic stem cells. The targeting construct was designed to replace the 455-residue protein-coding region containing the pro-, metalloprotease, disintegrin, and cysteine-rich domains of ADAM1b at positions 79 -533 with neo (Fig. 1A). The genotypes of wild-type (Adam1b ϩ/ϩ ), heterozygous (Adam1b ϩ/Ϫ ), and homozygous (Adam1b Ϫ/Ϫ ) mice for the targeted mutation of Adam1b were identified by Southern blot analysis of genomic DNA (Fig. 1B). Northern blot analysis indicated the absence of ADAM1b mRNA in Adam1b Ϫ/Ϫ testis. In addition, protein extracts of Adam1b Ϫ/Ϫ TGC completely lacked a 120-kDa protein corresponding to the precursor form of ADAM1b. These data demonstrate the absence of ADAM1b in the Adam1b Ϫ/Ϫ testis.
To examine the interaction of Adam1b Ϫ/Ϫ mouse sperm with eggs, in vitro fertilization assays were carried out using capacitated cauda epididymal sperm. When cumulus-intact eggs were used, the fertilization rate 2 (data not shown) and 6 h after insemination was normal in Adam1b Ϫ/Ϫ mouse sperm (Fig. 1C). No significant difference was found either in the sperm binding to ZP or in the fusion of sperm with ZP-free eggs between Adam1b ϩ/Ϫ and Adam1b Ϫ/Ϫ mice (Fig. 1, D and E). Moreover, microscopic analysis of frozen sections of the uterotubal junction 2 h after copulation indicated the ability of Adam1b Ϫ/Ϫ mouse sperm to ascend from the uterus into the oviduct (Fig. 2). Thus, the loss of ADAM1b causes no significant defect in sperm function.
Although the level of ADAM2 in TGC was similar between Adam1b ϩ/ϩ and Adam1b Ϫ/Ϫ mice (Fig. 3A), epididymal sperm of Adam1b Ϫ/Ϫ mice showed the loss of ADAM2 as well as of ADAM1b (Fig. 3D). To verify the loss of ADAM2 on Adam1b Ϫ/Ϫ mouse sperm, we carried out immunoblot analysis of sperm extracts from Adam1b ϩ/ϩ and Adam1b Ϫ/Ϫ mice using two different antibodies (Fig. 4A). As described above, anti-ADAM2 polyclonal antibody raised against the C-terminal 22-residue peptide (23) was not immunoreactive with the 45-kDa mature form of ADAM2 in the Adam1b Ϫ/Ϫ sperm extracts. However, the mature protein in the Adam1b Ϫ/Ϫ extracts was slightly but significantly recognized by anti-ADAM2 monoclonal antibody 9D2.2. Moreover, the Adam1b ϩ/ϩ and Adam1b Ϫ/Ϫ sperm extracts con-  tained a very small amount of the 100-kDa ADAM2 precursor. Thus, the precursor and mature forms of ADAM2 are present on Adam1b Ϫ/Ϫ mouse sperm. The cytoplasmic tail of mature ADAM2 on Adam1b Ϫ/Ϫ mouse sperm may be partially processed by a proteolytic enzyme, because the C-terminal region is not recognized by anti-ADAM2 polyclonal antibody. The level of the mature form of ADAM2 on Adam1b Ϫ/Ϫ mouse sperm was ϳ4% of that on Adam1b ϩ/ϩ mouse sperm (Fig. 4B).
As shown in Fig. 5, levels of both ADAM1b and ADAM2 on Adam1b ϩ/Ϫ mouse sperm were reduced to a similar extent despite the normal level of ADAM2 in Adam1b ϩ/Ϫ TGC (108% of that in Adam1b ϩ/ϩ TGC). Other proteins, ADAM3, PH-20, a testis-specific isoform of angiotensin-converting enzyme, and Izumo, were normally present on Adam1b Ϫ/Ϫ mouse sperm (97-103% of the levels on Adam1b ϩ/ϩ mouse sperm, see Fig. 3D). These data suggest that the appearance of ADAM1b and ADAM2 on the sperm surface may depend on the formation and abundance of ADAM1b/ADAM2 fertilin in TGC.
ADAM1b/ADAM2 fertilin formation in TGC for the appearance of these two ADAMs on the sperm surface (Figs. 3 and 5). Table 1 summarizes the correlation between the abundance of ADAM proteins and sperm function in four mouse lines lacking ADAM1a, ADAM1b, ADAM2, or ADAM3. Our data on Adam1b Ϫ/Ϫ mouse sperm strongly support the previous finding that ADAM3 is crucial in the sperm binding to egg ZP (15,16).
We have previously demonstrated that the loss of ER-resident ADAM1a results in the lack of ADAM1a/ADAM2 fertilin in TGC and ADAM3 on epididymal sperm, despite the normal presence of ADAM1b/ADAM2 fertilin and ADAM3 in TGC (13). The levels of ADAM1b and ADAM2 on the sperm surface are also normal in Adam1a Ϫ/Ϫ mice (13). In the present study, Adam1b Ϫ/Ϫ mice lack ADAM1b/ADAM2 fertilin in TGC and contain a severely reduced level of ADAM2 on epididymal sperm, whereas the ADAM3 level is normal on Adam1b Ϫ/Ϫ mouse sperm (Figs. 3 and 5). Thus, the formation of the ADAM1a/ADAM2 and ADAM1b/ADAM2 fertilins in TGC is presumably essential for the appearance of ADAM3 and both ADAM1b and ADAM2 on the sperm surface, respectively. This probability is consistent with the fact that the loss of ADAM1a/ADAM2 and ADAM1b/ ADAM2 fertilins in Adam2 Ϫ/Ϫ TGC leads to the lack of ADAM1b and ADAM3, in addition to ADAM2, on the sperm surface (13,15). The defective transport of ADAM3 onto the sperm surface in Adam1a Ϫ/Ϫ mice, and of both ADAM1b and ADAM2 in Adam1b Ϫ/Ϫ mice, may simultaneously occur in Adam2 Ϫ/Ϫ mice. It is also possible that Adam1a Ϫ/Ϫ mouse sperm are functionally identical to Adam2 Ϫ/Ϫ mouse sperm, because Adam1b Ϫ/Ϫ mouse sperm show no functional defect ( Figs. 1 and 2). Indeed, the Adam2 Ϫ/Ϫ mouse sperm show the same phenotype as Adam1a Ϫ/Ϫ mouse sperm except for a reduced rate of sperm/egg fusion in Adam2 Ϫ/Ϫ mouse sperm (Table 1).
Very small amounts of the precursor and mature forms of ADAM2 are still present on Adam1b Ϫ/Ϫ mouse sperm (Figs. 4 and 5). There are several possibilities that may explain the presence of ADAM2 on the surface of Adam1b Ϫ/Ϫ mouse sperm. ADAM2 is possibly transported onto the Adam1b Ϫ/Ϫ sperm surface by a protein complex with ERresident ADAM1a or a truncated form of ADAM1b or by ADAM2 itself. Immunoblot analysis of Adam1b Ϫ/Ϫ sperm extracts indicated the absence of the precursor, processing intermediate, and mature forms of ADAM1a (Fig. 3D). No signal for ADAM1b mRNA was found when total testicular RNA of Adam1b Ϫ/Ϫ mice on blots was probed by 140and 815-bp DNA fragments encoding the 5Ј-untranslated region and 18-nucleotide protein-coding region (data not shown) and the proteincoding region corresponding to the cysteine-rich, epidermal growth factor-like transmembrane and cytoplasmic tail domains of ADAM1b (Fig. 1, Probe N), respectively. Thus, it is conceivable that some ADAM2 in TGC may be capable of appearing on the sperm surface without the complex formation with ADAM1a or ADAM1b. Moreover, ADAM2 on Adam1b Ϫ/Ϫ mouse sperm may compensate for the loss of ADAM1b and ADAM1b/ADAM2 fertilin, because Adam1b Ϫ/Ϫ mouse sperm are functionally normal (Figs. 1 and 2). This possibility seems unlikely but cannot be ruled out completely at the present time.
Male mice lacking an ER-resident molecular chaperone, calmegin (Clgn Ϫ/Ϫ ), share the sterile phenotype with Adam1a Ϫ/Ϫ and Adam2 Ϫ/Ϫ mice: the defects of sperm both in binding egg ZP and in migrating from the uterus into oviduct (23,24,27). Importantly, calmegin is required for heterodimerization between ADAM1a and ADAM2 and/or between ADAM1b and ADAM2 in TGC (23). The defect of Clgn Ϫ/Ϫ mouse sperm in the ZP binding may be explained by the lack of ADAM3 on the sperm surface because of a possible absence of ADAM1a/ADAM2 fertilin in TGC. Clgn Ϫ/Ϫ mouse sperm, indeed, lack ADAM3 on the cell surface. 5 At any rate, the mechanism of the ADAM3 transport onto the sperm surface probably regulated by ADAM1a/ADAM2 fertilin and/or calmegin remains to be clarified. Recently, Stein et al. (17) reported that the loss of ADAM3 in Adam2 Ϫ/Ϫ mice may result, at least in part, from a disruption of protein trafficking.
Supposing the set of membranous proteins, except ADAM1b, ADAM2, and ADAM3, on the sperm surface is identical among Adam1a Ϫ/Ϫ , Adam1b Ϫ/Ϫ , Adam2 Ϫ/Ϫ , and Adam3 Ϫ/Ϫ mice, Adam3 Ϫ/Ϫ mouse sperm may not enter from the uterus into oviduct ( Table 1). The reason why Adam3 Ϫ/Ϫ mouse sperm normally migrate into the oviduct (15,16) is unknown at present. However, at least neither ADAM1b nor ADAM2 nor ADAM1b/ADAM2 fertilin presumably functions in the sperm migration, because Adam1b Ϫ/Ϫ mouse sperm are capable of migrating into the oviduct despite the loss or severe reduction of these two ADAM proteins (Fig. 2). Comparative studies using mutant sperm lacking ADAM3 are required for identification of a protein(s) involved in sperm migration into the oviduct.
Although ADAM1/ADAM2 fertilin (ADAM1b/ADAM2 fertilin in mouse) on the sperm surface was initially discovered as a protein complex responsible for the fusion between sperm and egg plasma membranes (6, 7), direct evidence for the functional role of this complex has not been provided to date. We have shown here that in the mouse, sperm fertilin plays no crucial role in fertilization. In fact, none of the egg integrins as ADAM receptors has been demonstrated to be essential for sperm/egg binding and fusion (31). Rather, another ADAM1a/ADAM2 fertilin is required for fertilization, because this ER-resident protein complex is implicated in the appearance of ADAM3 on the sperm surface that functions in sperm/egg ZP binding (13,15,16). Further characterization of ADAM1b/ADAM2 fertilin will be necessary to evaluate 5 M. Okabe, personal communication.

TABLE 1
Correlation between the abundance of ADAM proteins and sperm functions in four mouse lines lacking ADAM1a, ADAM1b, ADAM2, or ADAM3 whether the mouse sperm fertilin is indeed a non-functional protein complex or still contributes to an unknown sperm function(s).