伊藤 千鶴

Outer dense fibre 2 (Odf2 or ODF2) is a cytoskeletal protein required for flagella (tail)-beating and stability to transport sperm cells from testes to the eggs. There are infertile males, including human patients, who have a high percentage of decapitated and decaudated spermatozoa (DDS), whose semen contains abnormal spermatozoa with tailless heads and headless tails due to head-neck separation. DDS is untreatable in reproductive medicine. We report for the first time a new type of Odf2-DDS in heterozygous mutant Odf2+/- mice. Odf2+/- males were infertile due to haploinsufficiency caused by heterozygous deletion of the Odf2 gene, encoding the Odf2 proteins. Odf2 haploinsufficiency induced sperm neck-midpiece separation, a new type of head-tail separation, leading to the generation of headneck sperm cells or headnecks composed of heads with necks and neckless tails composed of only the main parts of tails. The headnecks were immotile but alive and capable of producing offspring by intracytoplasmic headneck sperm injection (ICSI). The neckless tails were motile and could induce capacitation but had no significant forward motility. Further studies are necessary to show that ICSI in humans, using headneck sperm cells, is viable and could be an alternative for infertile patients suffering from Odf2-DDS.

A number of sperm proteins are involved in the processes from gamete adhesion to fusion, but the underlying mechanism is still unclear. Here, we established a mouse mutant, the EQUATORIN-knockout (EQTN-KO,<italic> Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic>) mouse model and found that the EQTN-KO males have reduced fertility and sperm–egg adhesion, while the EQTN-KO females are fertile. <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> sperm were normal in morphology and motility. <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> <italic>-</italic>Tg (<italic>Acr-Egfp</italic>) sperm, which were produced as the acrosome reporter by crossing <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> with <italic>Eqtn</italic> <italic> ^+/+ </italic>-Tg(<italic>Acr-Egfp</italic>) mice, traveled to the oviduct ampulla and penetrated the egg zona pellucida of WT females. However, <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> males mated with WT females showed significant reduction in both fertility and the number of sperm attached to the zona-free oocyte. Sperm IZUMO1 and egg CD9 behaved normally in <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> sperm when they were fertilized with WT egg. Another acrosomal protein, SPESP1, behaved aberrantly in <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> sperm during the acrosome reaction. The fertility impairment of EQTN/SPESP1-double KO males lacking <italic>Eqtn</italic> and <italic>Spesp1</italic> (<italic>Eqtn</italic>/<italic>Spesp1</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic>) was more severe compared with that of <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic> males. <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic>-Tg (<italic>Eqtn</italic>) males, which were generated to rescue <italic>Eqtn</italic> <italic> ⁻ </italic> <italic> ^/ </italic> <italic> ⁻ </italic>males, restored the reduced fertility.

Spermatids must precisely integrate specific molecules into structurally supported domains that develop during spermatogenesis. Once established, the architecture of the acrosome contributes to the acrosome reaction, which occurs prior to gamete interaction in mammals. The present study aims to clarify the morphology associated with the integration of the mouse fertilization-related acrosomal protein equatorin (mEQT) into the developing acrosome. EQT mRNA was first detected by in situ hybridization in round spermatids but disappeared in early elongating spermatids. The molecular size of mEQT was approximately 65 kDa in the testis. Developmentally, EQT protein was first detected on the nascent acrosomal membrane in round spermatids at approximately step 3, was actively integrated into the acrosomal membranes of round spermatids in the following step and then participated in acrosome remodeling in elongating spermatids. This process was clearly visualized by high-resolution fluorescence microscopy and super-resolution stimulated emission depletion nanoscopy by using newly generated C-terminally green-fluorescent-protein-tagged mEQT transgenic mice. Immunogold electron microscopy revealed that mEQT was anchored to the acrosomal membrane, with the epitope region observed as lying 5-70 nm away from the membrane and was associated with the electron-dense acrosomal matrix. This new information about the process of mEQT integration into the acrosome during spermatogenesis should provide a better understanding of the mechanisms underlying not only acrosome biogenesis but also fertilization and male infertility.

Recently we reported that an oocyte activation ability in human and mouse sperm is associated with head flatness or the presence of perinuclear theca (PT) substance, MN13, which is an oocyte activation-related protein localized on the post-acrosomal sheath (PAS). As such, we hypothesize that the appearance of oocyte activation ability is stage-specifically regulated and depends on the formation of the acrosome or PAS/PT in spermatids. We monitored the appearance and movement of MN13 as a PT-specific molecule during spermatogenesis and analysed how the MN13 localization is affected in mouse and human globozoospermic acrosomeless sperm. MN13 was first detected on the surface of acrosomic vesicles, i.e. on the nascent outer acrosomal membrane of step 5-6 round spermatids (Sb1 spermatids in human), and it was then translocated via the outer acrosomal membrane surface to the most distal region of the acrosome in step 7 round spermatids (Sb2 spermatids). As spermatids elongated, MN13 was translocated via the cytoplasmic space between the nuclear envelope and the overlying plasma membrane towards the post-acrosomal region, and it was organized on the top of the nascent PAS that was typically found in step 14 elongated spermatids (Sd1 spermatids). In contrast, MN13 was not found in any GOPC-deficient spermatids, which completely lack the acrosome but have manchettes (microtubule bundles), nor in mouse and human acrosomeless sperm. The MN13 appearance or the MN13-related PAS/PT formation is highly dependant on acrosome formation; the MN13-related oocyte activation factor/ability is stage-specifically acquired in elongating/elongated spermatids.

A tetraspanin family protein, CD9, has not previously been identified in sperm cells. Here, we characterize sperm CD9 in the mouse, including its unique localization in sperm, appearance during spermatogenesis, and behavior and fate during mouse fertilization. In sperm, CD9 is an inner acrosomal membrane-associated protein, not a plasma membrane-associated protein. Its molecular weight is approximately 24 kDa throughout its processing, from testicular germ cells to acrosome-reacted sperm. A temporal difference was found between mRNA and protein expression; CD9 mRNA was detected in the stages from spermatogonia through round spermatids showing the strongest levels in midpachytene spermatocytes. CD9 protein was detected in the cytoplasm throughout the stages from spermatogonia to spermatocytes. While CD9 was weakly expressed in the spermatids from step 1 through step 14, the signals became clearly positive at the marginal region of the anterior acrosome in elongated spermatids. After the acrosome reaction, the majority of sperm CD9 was retained in the inner acrosomal membrane, but some quantity of CD9 was found on the plasma membrane covering the equatorial segment as detected by immunogold electron microscopy using anti-CD9 antibody. CD9 was maintained on the sperm head after reaching the perivitelline space of CD9-deficient eggs that were recovered after natural mating with wild males. Thus, this study characterizes CD9 in sperm development and fertilization.

Recent studies indicate that round-headed sperm cannot activate oocytes and lack the postacrosomal sheath (PAS) or perinuclear theca (PT), although normal flat-headed sperm can activate oocytes and do have PAS (PT). In this study, we investigated how oocyte activation ability correlates with sperm head morphology (round and flat) and the presence of PT, by studying MN13, a representative molecule of the PT. We analyzed sperm with flat and round heads from infertile patients with globozoospermia (n = 1) and teratozoospermia (n = 1), and also from GOPC(-/-) mice, an animal model of human globozoospermia. Differential interference contrast image analysis, immunocytochemistry with MN13 antibody, transmission electron microscopy and an oocyte activation assay (assessing pronucleus formation) with ICSI were used. Flat-headed (control) sperm from both a healthy fertile volunteer man and wild-type mice had MN13 and PAS (PT). Flat-headed sperm (&lt; 5% of the population) from GOPC(-/-) mice also had both MN13 and PAS (PT), and they showed high oocyte activation ability. In contrast, round-headed sperm from a globozoospermia patient (100%) and GOPC(-/-) mice (&gt; 95% of the population) had neither MN13, nor PAS (PT), nor oocyte activation ability. Oocyte activation was higher in flat- versus round-headed sperm from GOPC(-/-) mice (P &lt; 0.05). Oocyte activation ability may be related to sperm head flatness and presence of MN13 and PAS (PT) in human and mouse sperm. This information is a first step towards the possibility of selecting good-quality sperm with high oocyte activation ability for ICSI.

Deletion of the GOPC gene encoding mouse GOPC (Golgi-associated PDZ- and coiled-coil motif-containing protein) causes infertile round-headed spermatozoa. which have acrosome-less round heads and deformed tails (Yao et at, 2002). This study investigated how GOPC deficient spermatids fail to assemble the peri-nuclear structures in round-headed spermatozoa during spermiogenesis in GOPC knockout mouse testes. In step 1-8 spermatids, Golgi-derived proacrosomal vesicles that are transported to the perinuclear region formed acrosome-like vesicles of various sizes, called pseudoacrosomes. The marginal ring of the acroplaxome, which is generally formed between the descending edge of a developing acrosome and nuclear envelope in a wild spermatid, was poorly formed between the pseudoacrosome and nuclear envelope. In step 9-11 elongating spermatids, a majority of pseudoacrosomes were detached from the nucleus and disappeared from the perinuclear region by spermiation. Concomitantly, several failures occurred on the nucleus, manchette, postacrosomal sheath (perinuclear theca), and posterior ring. Ectoplasmic specializations were poorly formed, and did not always associate with developing spermatids. Consequently, spermatid nuclear elongation to form round-headed spermatozoa developed was impaired. In addition to these sequential failures. the posterior ring deficiency was attributed to the tail deformation destined to occur during epididymal maturation as reported in an accompanying paper (Suzuki-Toyota et al., 2004 in this issue), its eventual phenotype being reminiscent of the round-headed spermatozoa of human infertile globozoospermia.