Calmegin Is Required for Fertilin <Alpha>/<Beta> Heterodimerization

Developmental Biology 240, 254–261 (2001) doi:10.1006/dbio.2001.0462, available online at http://www.idealibrary.com on View metadata, citation and similar papers at core.ac.uk brought to you by CORE Calmegin Is Required for Fertilin ␣/␤ provided by Elsevier - Publisher Connector Heterodimerization and Sperm Fertility

Masahito Ikawa,* Tomoko Nakanishi,* Shuichi Yamada,*,† Ikuo Wada,‡ Katsuya Kominami,§ Hiromitsu Tanaka,¶ Masami Nozaki,¶ Yoshitake Nishimune,¶ and Masaru Okabe*,‡,1 *Genome Information Research Center, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565- 0871, Japan; †Program for Promotion of Basic Research Activities for Innovative Biosciences; ‡Department of Biochemistry, Sapporo Medical School of Medicine, Sapporo 060-8556, Japan; §Research and Development Center, Fuso Pharmaceutical Industries, Osaka 536-8523, Japan; and ¶Research Institute for Microbial Diseases, Osaka University, Osaka 565-0871, Japan

Loss of the endoplasmic reticulum resident chaperone calmegin leads to the production of sterile sperm that do not bind to the egg zona pellucida (M. Ikawa et al., 1997, Nature 387, 607–611). In the present study, we demonstrate that calmegin ؊/؊ sperm were defective in migrating into the oviducts and in binding to the egg plasma membrane. Despite the impaired adhesive function, calmegin ؊/؊ sperm could fertilize eggs when zonae pellucidae were partially dissected, and eggs fertilized in this manner could develop normally to term. Since these sperm characteristics were similar to those found in ,fertilin ␤ ؊/؊ sperm, we investigated the interaction of calmegin with fertilin ␤. Using immunoprecipitation techniques calmegin was found to bind to sperm membrane proteins, fertilin ␣ and ␤, during spermatogenesis. The binding was speciﬁc to calmegin: another endoplasmic reticulum chaperone calnexin, a calmegin homologue, was not able to bind to fertilin ␣ and ␤. In the calmegin ؊/؊ mice, a loss of heterodimerization of fertilin ␣ and ␤ was observed and fertilin ␤ was not detectable in mature sperm. The data not only explain why the calmegin and fertilin ␤ knockout mouse lines share a common infertile phenotype, but also reveal the importance of the maturation of sperm membrane proteins in the endoplasmic reticulum. © 2001 Academic Press Key Words: ADAM; chaperone; calmegin; calnexin; cyritestin; egg plasma membrane; fertilin; fertilization; sperm; zona pellucida.

INTRODUCTION when the calnexin interaction is disrupted, some nascent membrane proteins are incorrectly folded or assembled and Many complex membrane proteins undergo subunit fold- are retrotranslocated to cytosolic proteasomes (Chevet et ing and assembly in the endoplasmic reticulum (ER) before al., 1999; Fayadat et al., 2000). transport to the cell surface. Among the ER resident pro- Calmegin was first identified as a spermatogenic cell- teins, calnexin, a molecular chaperone for N-linked glyco- specific antigen which is present from pachytene spermato- proteins, functions as a component of ER quality control cytes to round spermatids (Watanabe et al., 1992). Later it machinery (Ellgaard et al., 1999; Trombetta and Helenius, was proved to be a Ca2ϩ binding protein and to be a calnexin 1998). Mutations in membrane proteins are known to cause homologue in the testis (Watanabe et al., 1994). In our some diseases such as cystic fibrosis, where improperly previous paper, we showed that calmegin is essential for folded mutant protein (cystic fibrosis transmembrane con- sperm fertility since the sperm from calmegin Ϫ/Ϫ mice ductance regulator, CFTR) is retained by calnexin inside the lack the ability to bind the zonae pellucidae of unfertilized ER and fails to be transported to the plasma membrane eggs (Ikawa et al., 1997). Our data suggested that the (Chevet et al., 1999; Pind et al., 1994). On the other hand, calmegin chaperones a sperm membrane protein required for adhesion to zona pellucida, but it remained unclear 1 To whom correspondence should be addressed. Fax: ϩ81-6- which membrane proteins are missing and how the loss 6879-8376. E-mail: [email protected]. occurs.

Apart from our work, defective zona binding has also microscope. Following the partial zona dissection, eggs were de- been observed in sperm from mice lacking fertilin ␤, also tached from the dish by introducing BSA-containing medium into known as an ADAM2 (a member of the ADAM family, the drop. The eggs were then washed with TYH medium and transmembrane proteins containing adisintegrin and met- subjected to in vitro fertilization. Fertilization events were judged alloprotease domain) (Cho et al., 1998; Primakoff and by pronuclei formation 8 h after sperm insemination. Fertilized eggs were cultivated until the blastocyst stage and then transferred Myles, 2000). Fertilin consists of an ␣ and a ␤ subunit and to pseudopregnant females. has been proposed to be a sperm/egg fusion protein (Blobel et al., 1992; Cho et al., 2000). The defect of fertilin ␤ Ϫ/Ϫ sperm was reported to include impaired migration of sperm Transgenic Mice Expressing Calmegin into the oviduct and disabled binding of sperm to the egg A rescue construct was prepared in the pBluescript SK IIϩ plasma membrane. Here, we show that calmegin Ϫ/Ϫ plasmid. The SacI–BamHI fragment (330 bp) of the calmegin sperm share phenotypes in common with fertilin ␤ Ϫ/Ϫ promoter region (Watanabe et al., 1995) was placed in front of the sperm. Further analysis revealed that mature type fertilin BamHI–EcoRI fragment (2146 bp) containing calmegin cDNA with was lacking on calmegin Ϫ/Ϫ sperm since calmegin was a polyadenylation signal. TgN(CM-CM)Osb transgenic mouse lines important in fertilin ␣/␤ heterodimerization during sper- were produced by injecting the SacI–EcoRI fragment of the rescue construct into pronuclei of B6C3F1 ϫ B6C3F1 fertilized eggs. matogenesis. Offspring carrying the transgene were identified by PCR using Primer C (5Ј-CCTTCCTgCg gCTTgTTCTC T-3Ј) and Primer D (5Ј-TATCATCCTT CTTTgCTTTT g-3Ј) for calmegin cDNA. The MATERIALS AND METHODS endogenous calmegin mutation was detected by PCR using Primer A(5Ј-TCTTACCACA AAgCACCTCC-3Ј) and Primer-B (5Ј- Sperm Migration Analysis gCACAACAgg ATggATgATT-3Ј). B6C3F1 females were superovulated by intraperitoneal injection of 5 units of pregnant mare’s serum gonadotropin followed 48 h Immunoprecipitation and Western Blotting later by 5 units of human chorionic gonadotropin (HCG). Super- Analysis ovulated females were caged together with calmegin mutant males 12 h after HCG injection, and the formation of vaginal plug was Rabbit polyclonal antisera were raised against synthesized pep- observed every 30 min. About 2 h after copulation, oviducts were tides, murine calnexin (C-terminal, EDEILNRSPRNRKPRRE), excised together with a connective part of the uterus. The sperm calmegin (C-terminal, DESPGSGDAPLKSLRKRRVRKD), and fer- ␤ were flushed out from the remainder of the uterus with 1.0 ml of tilin (C-terminal, MKWRMDDFSSEEQFESESESKD). Rabbit anti- ␣ TYH medium and an aliquot was used for counting sperm num- fertilin “metalloprotease and prodomain” antiserum was a gift bers. To detect sperm in the utero-tubal junction, the oviduct with from Dr. Paul Primakoff. These antibodies were used for immuno- attached uterus was fixed in 4% paraformaldehyde–PBS for 6 h, precipitation and Western blotting for testicular lysates. To detect ␤ followed by washing with PBS, and was then prepared for frozen fertilin on sperm, mAb 9D was purchased from CHEMICON sections. International, Inc. Cell lysates or immunoprecipitates, prepared as before (Ikawa et al., 1997), were subjected to SDS–PAGE under nonreducing (0.5% SDS at room temperature) or reducing condi- Sperm Binding Analysis tions (2% SDS, 5% ␤-mercaptoethanol and boiled) and then analyzed by Western blotting. The eggs were collected from oviducts of superovulated mice at 16 h after HCG injection and placed into a drop of TYH medium under mineral oil. The eggs were freed from cumulus cells by Ϫ1 RESULTS incubating with hyaluronidase (500 ␮gml ; Sigma) for 5 min followed by washing. Half of the cumulus-free eggs were treated Defective Sperm Migration into Oviduct in with acidic Tyrode’s solution (Sigma) for about 30 s to remove the ؊ ؊ zonae pellucidae and were then washed with TYH medium. The Calmegin / Sperm zona-intact and zona-free eggs were transferred into 200-␮l drops of To observe the effect of calmegin disruption on sperm TYH medium covered with mineral oil. Sperm prepared from migration from the uterus into the oviduct, cross sections caudae epididymides of 3-month-old male mice were dispersed into of the utero-tubal junction were prepared and observed 2 h TYH medium and, following 120 min of preincubation, were mixed after coitus. We found few sperm from calmegin Ϫ/Ϫ mice with eggs to a final density of 1–2 ϫ 105 sperm mlϪ1. After 15 min into the colliculus, the initial part of the oviduct, while of coincubation, the eggs were washed gently with TYH medium and sperm bound to zona pellucida or egg plasma membrane were numerous sperm were observed in the nearby uterine lu- observed. men (Figs. 1A–1D). There was no difference in the numbers of wild-type and calmegin Ϫ/Ϫ sperm recovered from the uterus (5.9 Ϯ 2.0 ϫ 106 and 5.7 Ϯ 1.9 ϫ 106, respectively). Partial Zona Dissection and in Vitro Fertilization The partial zona dissection was performed following the method Impaired Adhesion to Zona-Free Eggs in Calmegin described by Nakagata et al. (1997). Briefly, the cumulus-free eggs ؊/؊ Sperm were transferred into 0.3 M sucrose/PB1 without BSA on bacterial- grade culture dish. Zonae pellucidae were then dissected manually In our previous report, sperm adhesion to zona pellucida with the tip of a 30-gauge injection needle under a dissecting was observed to be defective (Ikawa et al., 1997). In the

FIG. 1. Defective migration of calmegin Ϫ/Ϫ sperm into oviduct found. Sperm transits from uterus to oviduct were observed by removing the utero-tubal junction 2 h after coitus and making frozen sections stained with hematoxylin and eosin. (A, C) Wild type; (B, D) Ϫ/Ϫ. Boxed areas in (A) and (B) were magniﬁed in (C) and (D). c, colliculus; u, uterus.

present study, sperm adhesion was also deficient when we mice entered the perivitelline space through a slit opened in incubated the calmegin Ϫ/Ϫ sperm with zona-free eggs, the zona and succeeded in fusing with eggs (Fig. 3). Eight hours (Figs. 2A and 2B). To measure sperm adhesion quantita- after the insemination, fertilization was confirmed by pro- tively, the numbers of sperm bound to zona-intact or nuclear formation. As shown in Table 1, fertilization was zona-free unfertilized eggs were counted. As shown in Fig. significantly reduced when eggs were inseminated with 2C, the sperm from calmegin Ϫ/Ϫ mice bound far less not calmegin Ϫ/Ϫ sperm. However, once fertilization occurred, only to zonae pellucidae (0.8% of wild type) but also to eggs fertilized by calmegin Ϫ/Ϫ sperm could successfully plasma membranes of zona-free eggs (5.2% of wild type). develop to the blastocyst stage in vitro. When the blastocysts were transferred to pseudopregnant female, they developed to term comparable to control eggs (Table 1). Subsequent male Rescue of Calmegin ؊/؊ Sperm Infertility by offspring, which were heterozygous for mutant and wild-type Partial Zona Dissection in Vitro calmegin alleles, showed normal fecundity (data not shown). In our previous report, none of the 72 zona-intact eggs was fertilized by calmegin Ϫ/Ϫ sperm (Ikawa et al., 1997). Here, to Rescue of Calmegin ؊/؊ Infertility in Vivo by analyze the fusing ability of calmegin Ϫ/Ϫ sperm with egg Transgenic Approach plasma membranes, we mixed the sperm with partially zona- dissected eggs to circumvent the zona penetration step of To determine whether calmegin was truly responsible for fertilization (Nakagata et al., 1997). Sperm from calmegin Ϫ/Ϫ producing sterile sperm, we produced transgenic mice ex-

FIG. 2. Defective binding of calmegin Ϫ/Ϫ sperm to egg plasma membrane. Zona-intact and zona-removed eggs were incubated with cauda epididymal sperm for 15 min and observed. (A, B) Microscopic view of sperm binding to zona removed eggs. (A) Wild type; (B) calmegin Ϫ/Ϫ sperm. (C) Sperm binding to zona-intact and zona-removed eggs was quantitatively measured by counting sperm bound to eggs. The average numbers of sperm bound per egg (mean Ϯ s.d.) were 65 Ϯ 13 in wild type and 0.5 Ϯ 0.5 in Ϫ/Ϫ sperm on zona pellucida (P Ͻ 0.001) and 33 Ϯ 13 in wild type and 1.7 Ϯ 1.1 in Ϫ/Ϫ sperm on egg plasma membrane (P Ͻ 0.001). pressing calmegin under the control of calmegin promoter Interaction between Calmegin and Fertilin (Fig. 4). The transgene was injected into B6C3F1 ϫ B6C3F1 in the ER fertilized eggs and transgenic pups were determined by ampli- fying a 363-bp fragment by PCR. Four transgenic founder mice Since calnexin, a ubiquitously expressed calmegin homo- (lines 2, 11, 13, and 19) were obtained and were backcrossed to logue, has been reported to be an ER chaperone for secretory calmegin Ϫ/Ϫ. When the testis lysates obtained from calme- proteins to mature (Ellgaard et al., 1999; Trombetta and gin Ϫ/Ϫ and transgenic male were subjected to Western Helenius, 1998), a similar function was postulated for blotting analysis, comparable amounts of calmegin (93 kDa) calmegin in sperm membrane protein maturation during were detected in lines 2 and 19 (Fig. 4C). The germ cell- and spermatogenesis. To determine whether calmegin interacts differentiation stage-speciﬁc expression of the calmegin trans- with ␣ and/or ␤ subunits of fertilin, testis lysates were gene (pachytene–spermatid) was conﬁrmed by immunostain- immunoprecipitated with anti-calmegin antibody and then ing (data not shown). To investigate the restoration of fecun- analyzed by Western blotting using anti-fertilin ␣ and dity, rescued males from these two lines were mated with anti-fertilin ␤ antibodies. As a control, immunoprecipitates wild-type females for 3 months. The rescued males showed with anti-calnexin antibody were used since the calnexin normal fecundity (average litter size/plug observed; TG line 2, also exists in spermatogenic cells which express calmegin 8.0 Ϯ 2.7; TG line 19, 9.1 Ϯ 2.1), while nontransgenic (Ikawa et al., 1997). Calmegin was precipitated only from calmegin Ϫ/Ϫ male littermates showed reduced fecundity wild-type but not from calmegin Ϫ/Ϫ mouse testis. The (average litter size/plug observed, Ͻ0.1). amount of precipitated calnexin was not different in wild-

FIG. 3. Fertilization of calmegin Ϫ/Ϫ sperm after partial zona dissection. Unfertilized eggs were subjected to partial zona dissection before in vitro fertilization. (A) A slit was opened by hand using a 30-gauge injection needle under a stereomicroscope. (B) Pronuclei were observed in the eggs fertilized by calmegin Ϫ/Ϫ sperm 8 h after insemination.

type and calmegin Ϫ/Ϫ testis lysates (Fig. 5A). When the calmegin Ϫ/Ϫ testes, although certain amounts of mono- immunoprecipitates were probed with anti-fertilin ␣ anti- mers still remained (Fig. 6B). When the sperm proteins were body, fertilin ␣ precursor was found only in the anti- separated by SDS–PAGE under reducing conditions and calmegin immune complex (Fig. 5B). Fertilin ␤ precursor probed with anti-fertilin ␤ antibody, the mature form of was also found to be calmegin speciﬁc (Fig. 5C). fertilin ␤, a broad band around 43 kDa, completely disappeared from calmegin Ϫ/Ϫ sperm (Fig. 6C). Loss of Heterodimerization and Sperm Surface Presentation of Fertilin ␣/␤ in Calmegin ؊/؊ DISCUSSION ␣ The effects of calmegin disruption on the fate of fertilin ؊ ؊ and ␤ during spermatogenesis were investigated by Western Calmegin / Sperm Phenotype blotting analysis. Since fertilin ␣ and ␤ were reported to We have previously shown that the disruption of calme- heterodimerize in the testis, SDS–PAGE was performed gin expression results in defective sperm adhesion to egg under reducing conditions to compare the total protein zona pellucida (Ikawa et al., 1997). After our report, fertilin amounts of fertilin ␣ and ␤, respectively. A decrease of Ϫ/Ϫ sperm were also reported to exhibit defective adhesion pre-fertilin ␣ and pre-fertilin ␤ was found in calmegin Ϫ/Ϫ to zona pellucida. Moreover, an impaired migration into testis (Fig. 6A). When we performed the SDS–PAGE under oviduct and a defective adhesion to egg plasma membrane nonreducing conditions, in which complexes of fertilin ␣ were observed for fertilin Ϫ/Ϫ sperm (Cho et al., 1998). As precursor and ␤ precursor could be detected as 160 and 145 shown in the present paper, calmegin Ϫ/Ϫ sperm were also kDa (␣/␤ and ␣Ј/␤ heterodimer, respectively; Cho et al., defective in migration into the oviduct and in adhesion to 2000), we found the disappearance of both bands from the egg plasma membrane. To analyze the egg-fertilizing

TABLE 1 Partial Zona Dissection Restored the Fertilizing Ability of Calmegin Ϫ/Ϫ Sperm in Vitro

No. eggs fertilized No. eggs developed No. pups/transferred Genotype of calmegin No. eggs examined (%) to blastocyst (%) (%)

ϩ/ϩ or ϩ/Ϫ 179 148 (83 Ϯ 6) 135 (92 Ϯ 2) 17/61 (29 Ϯ 20) Ϫ/Ϫ 549 201 (30 Ϯ 14)* 179 (84 Ϯ 9) 30/80 (35 Ϯ 21)

Note. The data indicate means Ϯ SD from three independent experiments. * Fertilization rate in the Ϫ/Ϫ was signiﬁcantly different from control (*, P Ͻ 0.05, unpaired Student’s t test).

FIG. 4. Rescued fertility of calmegin Ϫ/Ϫ male by transgene. (A) Genomic structure of wild-type allele and calmegin knockout allele with the transgene construct for rescue experiment. Neo-resistant gene was inserted into calmegin exon 2. The 330-bp fragment containing calmegin promoter region was used for rescue experiment. (B) Genotyping of rescued mice. PCR from calmegin ϩ/ϩ gave 1 kb and Ϫ/Ϫ gave 2 kb of signal using Primers A and B. PCR from transgene gave a 363-bp fragment using Primers C and D. Nonspeciﬁc signals were observed when transgene was absent. MW, ␾174X HaeIII digest. (C) Western blotting analysis of testis lysates from rescued mice. Comparable amounts of calmegin (93 kDa) were detected in wild-type, TG lines 2 and 19.

ability of calmegin Ϫ/Ϫ sperm further, we performed a of certain gene knockout mice were actually derived from partial zona dissection followed by in vitro fertilization. the indirect disruption of other genes (Olson et al., 1996), Although the fertilization ratio, 30 Ϯ 14%, observed in we examined whether expression of calmegin from a trans- calmegin Ϫ/Ϫ mice sperm was low (36% of wild type), the gene can restore the wild-type phenotype to calmegin Ϫ/Ϫ value was comparable to that reported in fertilin ␤ Ϫ/Ϫ mice. We found that two of four transgenic mouse lines that sperm (50% of wild type) (Cho et al., 1998). From these express calmegin from the transgene during spermatogen- experiments, calmegin was demonstrated to be required for esis regained their fecundity. These results clearly proved sperm migration, zona pellucida adhesion, and egg plasma that the deﬁciency of calmegin was the real cause of membrane adhesion but not essential for egg plasma mem- infertility of calmegin Ϫ/Ϫ mice. brane fusion as described for fertilin ␤ Ϫ/Ϫ mice (Cho et al., 1998, 2000). It is noteworthy that once the eggs were Loss of Fertilin Heterodimerization in Calmegin fertilized with calmegin Ϫ/Ϫ sperm there were no signiﬁ- ؊/؊ Mice cant differences in subsequent development. Since there are some cases in which apparent phenotypes Since calmegin was localized on ER membranes and disappeared from spermatogenic cells during the last stages of spermatogenesis (Watanabe et al., 1992; Yoshinaga et al.,

FIG. 5. Specific binding of fertilin ␣ and ␤ to calmegin. Testicular cell extracts prepared from wild-type and calmegin Ϫ/Ϫ mice were FIG. 6. Impaired fertilin ␣/␤ assembly and disappearance of fertilin immunoprecipitated with anti-calnexin or anti-calmegin antibod- ␤ from sperm surface. Twenty micrograms of testicular cell lysates ies. (A) The immunoprecipitated calnexin (CN) and calmegin (CM) and 15 ␮g of sperm lysates were subjected to Western blotting bands (90 and 93 kDa, respectively) were detectable even by silver analysis. (A, B) Testicular proteins were resolved by SDS–PAGE under staining after SDS–PAGE under reducing conditions. Comparable reducing (A) and nonreducing conditions (B). (C) Sperm proteins were amounts of CN and CM immunoprecipitates were analyzed in the resolved by SDS–PAGE under reducing conditions. Proteins trans- following Western blotting experiments. (B, C) Proteins transferred ferred to PVDF membrane were probed with anti-fertilin ␣- or anti- to PVDF membrane were probed with anti-fertilin ␣ (B) or anti- fertilin ␤-specific antibodies. Open arrowhead, fertilin ␣; solid arrow- fertilin ␤ (C) specific antibodies. head, fertilin ␤; half-solid arrowhead, fertilin ␣/␤ heterodimer.

1999), calmegin could not participate in sperm/egg adhe- REFERENCES sion event directly. In addition, since calmegin is consid- ered to function like calnexin (Ikawa et al., 1997; Watanabe et al., 1994), we speculated that calmegin chaperones sperm Bass, J., Chiu, G., Argon, Y., and Steiner, D. F. (1998). Folding of insulin receptor monomers is facilitated by the molecular chap- membrane proteins required for sperm fertility. Therefore, Ϫ Ϫ ␤ erones calnexin and calreticulin and impaired by rapid dimeriza- the very similar phenotypes of calmegin / and fertilin tion. J. Cell Biol. 141, 637–646. Ϫ Ϫ / mouse lines led us to investigate the relationship Blobel, C. P., Wolfsberg, T. G., Turck, C. W., Myles, D. G., between these two proteins. Immunoprecipitation followed Primakoff, P., and White, J. M. (1992). A potential fusion peptide by Western blot analysis revealed that both fertilin ␣ and ␤ and an integrin ligand domain in a protein active in sperm–egg specifically complexed with calmegin in the ER but not fusion [see comments]. Nature 356, 248–252. calnexin. The disruption of calmegin was shown to cause Chevet, E., Jakob, C. A., Thomas, D. Y., and Bergeron, J. J. (1999). impaired heterodimerization of fertilin ␣/␤ and finally Calnexin family members as modulators of genetic diseases. resulted in the complete absence of fertilin ␤ from mature Semin. Cell Dev. Biol. 10, 473–480. sperm. Since fertilin ␣ was not present in mature sperm Cho, C., Bunch, D. O., Faure, J. E., Goulding, E. H., Eddy, E. M., when the fertilin ␤ gene was disrupted (Cho et al., 1998), Primakoff, P., and Myles, D. G. (1998). Fertilization defects in fertilin ␣ was also predicted to be missing from mature sperm from mice lacking fertilin beta. Science 281, 1857–1859. Cho, C., Ge, H., Branciforte, D., Primakoff, P., and Myles, D. G. calmegin Ϫ/Ϫ sperm. There is precedence for the disappear- (2000). Analysis of mouse fertilin in wild-type and fertilin ance of a membrane protein from the cell surface when the beta(Ϫ/Ϫ) sperm: Evidence for C-terminal modification, alpha/ chaperone function of calnexin (a calmegin homologue) was beta dimerization, and lack of essential role of fertilin alpha in disrupted. For example, in the absence of a functioning sperm–egg fusion. Dev. Biol. 222, 289–295. calnexin, the formation of the insulin receptor homodimer Ellgaard, L., Molinari, M., and Helenius, A. (1999). Setting the was repressed and the receptor failed to be presented on the standards: Quality control in the secretory pathway. Science 286, cell surface (Bass et al., 1998). The present data not only 1882–1888. indicated the importance of calmegin function for the Fayadat, L., Siffroi-Fernandez, S., Lanet, J., and Franc, J. L. (2000). sperm membrane protein maturation but also reinforced Calnexin and calreticulin binding to human thyroperoxidase is the concept of ER chaperone functioning to form dimerized required for its first folding step(s) but is not sufficient to proteins. promote efficient cell surface expression. Endocrinology 141, 959–966. In addition to calmegin Ϫ/Ϫ and fertilin ␤ Ϫ/Ϫ sperm, Hagaman, J. R., Moyer, J. S., Bachman, E. S., Sibony, M., Magyar, the defective sperm–zona pellucida adhesions were reported P. L., Welch, J. E., Smithies, O., Krege, J. H., and O’Brien, D. A. for sperm from mice lacking a testicular isoform of angio- (1998). Angiotensin-converting enzyme and male fertility. Proc. tensin converting enzyme (tACE) (Hagaman et al., 1998; Natl. Acad. Sci. USA 95, 2552–2557. Kessler et al., 2000) and a sperm membrane protein, cyri- Ikawa, M., Wada, I., Kominami, K., Watanabe, D., Toshimori, K., testin (ADAM3) (Shamsadin et al., 1999). Recently, it was Nishimune, Y., and Okabe, M. (1997). The putative chaperone reported that, if fertilin ␤ or cyritestin was disrupted, the calmegin is required for sperm fertility. Nature 387, 607–611. fertilin ␣ also disappeared from the sperm (Nishimura et al., Kessler, S. P., Rowe, T. M., Gomos, J. B., Kessler, P. M., and Sen, 2001). In the present paper, we demonstrated the direct G. C. (2000). Physiological non-equivalence of the two isoforms evidence of the specific interaction of fertilin ␣ and ␤ to of angiotensin-converting enzyme. J. Biol. Chem. 275, 26259– calmegin during spermatogenesis and also indicated that 26264. the heterodimerization of fertilin ␣ and ␤ is important for Liu, D. Y., and Baker, H. W. (2000). Defective sperm-zona pellucida interaction: A major cause of failure of fertilization in clinical the subsequent presentation of fertilin on sperm surface. in-vitro fertilization. Hum. Reprod. 15, 702–708. Further studies clarifying the intermolecular relationships Nakagata, N., Okamoto, M., Ueda, O., and Suzuki, H. (1997). among recently emerged zona-binding related proteins are Positive effect of partial zona-pellucida dissection on the in vitro necessary to understand the basis of sperm fertility in fertilizing capacity of cryopreserved C57BL/6J transgenic mouse molecular terms. These basic investigations also are clini- spermatozoa of low motility. Biol. Reprod. 57, 1050–1055. cally relevant since a defect in zona pellucida adhesion was Nishimura, H., Cho, C., Branciforte, D. R., Myles, D. G., and reported to be the major cause of failure in clinical in vitro Primakoff, P. (2001). Analysis of loss of adhesive function in fertilization programs (Liu and Baker, 2000). sperm lacking cyritestin or fertilin beta. Dev. Biol. 233, 204–213. Olson, E. N., Arnold, H. H., Rigby, P. W., and Wold, B. J. (1996). ACKNOWLEDGMENTS Know your neighbors: Three phenotypes in null mutants of the myogenic bHLH gene MRF4. Cell 85, 1–4. We thank G. Gerton and P. Primakoff for critically reading the Pind, S., Riordan, J. R., and Williams, D. B. (1994). Participation of manuscript. All experiments were performed with the consent of the endoplasmic reticulum chaperone calnexin (p88, IP90) in the the Animal Care and Use Committee of Osaka University. This biogenesis of the cystic fibrosis transmembrane conductance work was supported by grants from the Ministry of Education, regulator. J. Biol. Chem. 269, 12784–12788. Science, Sports, and Culture (11780533 and 11234203), the Minis- Primakoff, P., and Myles, D. G. (2000). The ADAM gene family: try of Health and Welfare (H10-genome-012), and Takeda Science Surface proteins with adhesion and protease activity. Trends Foundation. Genet. 16, 83–87.

Shamsadin, R., Adham, I. M., Nayernia, K., Heinlein, U. A., Watanabe, D., Yamada, K., Nishina, Y., Tajima, Y., Koshimizu, U., Oberwinkler, H., and Engel, W. (1999). Male mice deficient Nagata, A., and Nishimune, Y. (1994). Molecular cloning of a for germ-cell cyritestin are infertile. Biol. Reprod. 61, 1445– novel Ca(2ϩ)-binding protein (calmegin) specifically expressed 1451. during male meiotic germ cell development. J. Biol. Chem. 269, Trombetta, E. S., and Helenius, A. (1998). Lectins as chaperones in 7744–7749. glycoprotein folding. Curr. Opin. Struct. Biol. 8, 587–592. Yoshinaga, K., Tanii, I., and Toshimori, K. (1999). Molecular Watanabe, D., Okabe, M., Hamajima, N., Morita, T., Nishina, Y., chaperone calmegin localization to the endoplasmic reticulum of and Nishimune, Y. (1995). Characterization of the testis-specific meiotic and post-meiotic germ cells in the mouse testis. Arch. gene “calmegin” promoter sequence and its activity defined by Histol. Cytol. 62, 283–293. transgenic mouse experiments. FEBS Lett. 368, 509–512. Watanabe, D., Sawada, K., Koshimizu, U., Kagawa, T., and Nishi- Received for publication July 18, 2001 mune, Y. (1992). Characterization of male meiotic germ cell- Revised September 14, 2001 specific antigen (Meg 1) by monoclonal antibody TRA 369 in Accepted September 18, 2001 mice. Mol. Reprod. Dev. 33, 307–312. Published online October 25, 2001

Calmegin Is Required for Fertilin &lt;Alpha&gt;/&lt;Beta&gt; Heterodimerization

Calmegin Is Required for Fertilin <Alpha>/<Beta> Heterodimerization