Murine male germ cells show a marked upregulation of over 1500 genes at the onset of meiosis and the expression of around 350 of these genes is restricted to the male germ line, many of them directly necessary for the fertility of sperm (Schultz et al., 2003). However, many testis-specific genes have been found to be dispensable, most likely due to the robustness and redundancy of the system (Miyata et al., 2016; Lu et al., 2019; Park et al., 2020). This illustrates the complexity and evolutionary necessity of the tight regulation of spermatogenesis. Even though the cell biology of spermatogenesis has been well described, the molecular mechanism and variety of involved factors remain poorly characterized.

Spermatogenesis takes place in the seminiferous tubules of the testis. The spermatogonia undergo mitosis and develop into spermatocytes. In meiotic spermatocytes of humans and mice, heavily glycosylated soluble hydrolases, zymogens, transmembrane proteins, and other acrosome-specific cargo mature through the ER-Golgi networks and are packed into proacrosomal vesicles (pro-AVs) (Anakwe, 1990; Kashiwabara et al., 1990; Escalier et al., 1991). In mice, after the completion of meiosis, spermatids develop in a 16-step process (Oakberg, 1956). In steps 1-3, pro-AVs fuse and attach to the nuclear membrane (Escalier et al., 1991; Bermudez et al., 1994). How this process is initiated and how premature fusion is inhibited, is so far unknown. Trafficking between the Golgi apparatus and the acrosome is dynamic and bidirectional, with several Golgi proteins found on acrosomes that are later retrieved (Moreno et al., 2000). In steps 3-4, F-actin forms the acroplaxome adjacent to the nuclear membrane to which keratin V is added in steps 4–5. This cytoskeletal structure forms the site of pro-AV attachment to the nuclear lamina, initiating acrosome formation (Kierszenbaum et al., 2004). Adaptor protein 1 (AP-1) has been proposed to be the clathrin-adaptor protein mediating pro-AV trafficking towards the acrosome (Kang-Decker et al., 2001). However, it is unknown if AP-1 is present at the time of fusion and how the pro-AVs merge. Clathrin and adaptins are expressed in steps 1-5, whereas syntaxins and vesicle-associated membrane proteins (VAMPs) which mediate membrane fusion, have been found to localize to the pro-AVs in steps 1-2 and later to the acrosome (Ramalho-Santos et al., 2001). The acrosomal shape has been used to describe the characteristic spermatid development stages: the Golgi phase, when the acrosomal material is located near the Golgi apparatus, the cap phase, when the acrosome grows and spreads onto the nuclear surface, the acrosomal stage, when the nucleus and acrosome elongate, and finally the maturation stage. The spermatozoon continues to mature throughout its passage through the epididymis and female reproductive tract.

The insulin inhibitory receptor (in short: inceptor, gene Iir, also known as Elapor1) has been previously described in the context of cancer prognosis and progression (Deng et al., 2005; Kang et al., 2015; Stinnesbeck et al., 2021; Wissmiller et al., 2023). Moreover, we have shown that inceptor desensitizes insulin and insulin-like growth factor (IGF) receptor signaling in pancreatic beta cells (Ansarullah et al., 2021). Inceptor contains an adaptor protein 2 (AP-2) sorting signal and interacts with the µ subunit of AP-2 (Ansarullah et al., 2021). From our most recent findings in pancreatic beta cells, we propose that inceptor is a sorting and degradation receptor found in clathrin-coated vesicles budding from the plasma membrane and trans-Golgi membrane that routes insulin and proinsulin to lysosomal degradation (Siehler et al., unpublished results). Additionally, inceptor has been found to regulate secretory granule formation in zymogenic cells in the stomach (Cho et al., 2022). Early spermatids also express inceptor, and interestingly, the acrosome formation process has similarities with both regulated secretion and lysosome biogenesis (Khawar et al., 2019). Therefore, we hypothesized that inceptor might play a role in acrosomal trafficking and acrosome formation.

The acrosome is a large granule at the anterior end of the sperm head essential for sperm-oocyte recognition and fusion (Khawar et al., 2019). It is a low-pH organelle that contains Golgi-derived cargo, which is either acrosome-specific or lysosome-like. Therefore, the acrosome has been proposed to be a lysosome-related organelle (Berruti et al., 2010). It contains several enzymes, zymogens, and receptors necessary for oocyte recognition. At the time of oocyte-sperm contact, the acrosome is exocytosed as the outer acrosomal membrane fuses with the plasma membrane, enabling the sperm to penetrate the zona pellucida around the oocyte. Then, the inner acrosomal membrane fuses with the oocyte’s plasma membrane. The acrosome reaction has been reviewed in detail (Aldana et al., 2021).

Some of the factors which play a role in insulin granule formation or secretion are also essential for acrosome formation, such as Protein interacting with C kinase 1 (PICK1), Golgi-associated PDZ- and coiled-coil motif-containing protein (GOPC), or Mysoin Va (Yao et al., 2002; Kierszenbaum et al., 2003; Varadi et al., 2005; Xiao et al., 2009; Wilhelmi et al., 2021; Andersen et al., 2022). Moreover, Golgi and lysosomal proteins are retrieved from acrosomes in pathways resembling retrieval from maturing secretory granules in neuroendocrine cells, mediated by AP-1 and STX6 (Klumperman et al., 1998). The acrosome also displays similarities with the lysosome, but it is unknown whether it can carry out lysosomal metabolic functions in sperm. Even though the acrosome contains several unique proteins, it also contains typical lysosomal proteins, such as cathepsin D and H (Moreno and Alvarado, 2006). In addition to the acrosome’s similarities to lysosomes and secretory granules, recent studies point to mitochondrial engagement in acrosome formation. Condensed mitochondria were found to act as donors to the AV, and several mitochondrial enzymes were shown to localize to the acrosome (Ren et al., 2019; 2022; Otčenášková et al., 2023). Altogether, this shows that there is a substantial gap in knowledge regarding the acrosomal origin and the molecular mechanism of its biogenesis.

Here, we demonstrate the importance of inceptor in male fertility in a whole-body inceptor knockout (KO) mouse model, showing that inceptor is essential for the formation of morphologically normal sperm. Inceptor KO spermatids are characterized by acrosomal malformation, an irregular or round nucleus, and reduced motility. We propose that inceptor is tightly involved in acrosome development in early spermatids and is directly involved in cargo transport and delivery to the forming AV.

Inceptor is highly expressed in several tissues, such as the digestive tract, pancreas, prostate, female reproductive tract, lung, brain, and testis, specifically enriched in glandular cell types, endocrine cells, and early spermatids (Supplementary Figure S1A) (The Human Protein Atlas). We confirmed inceptor protein expression in the stomach, salivary gland, colon, and lung by immunofluorescence (Supplementary Figure S1B).

To analyze the function of inceptor, we generated the Iir ^+/− mouse line by crossing the previously generated GeneTrap line (Ansarullah et al., 2021) to FLPe and RosaCre animals to obtain a germline deletion of exon 3 (Supplementary Figure S1C). By heterozygous intercrossing of the Iir ^+/− mouse line, Mendelian ratios of offspring genotypes were observed at weaning age (Supplementary Figure S1D). We confirmed the absence of inceptor in Iir ^−/− testis lysates by Western blotting (Supplementary Figure S1E). Intriguingly, the Iir ^−/− genotype could not be maintained on a homozygous background. While Iir ^−/− females delivered similar litter sizes as Iir ^+/− females with Iir ^+/− males, matings with Iir ^−/− males were unsuccessful (Figure 1A). These results are consistent with previously described infertility of Iir ^−/− males (Tang et al., 2010; Cho et al., 2022), however, the underlying cellular and molecular mechanism was unknown.

To further investigate the function of inceptor in male fertility, we analyzed testis sections in Iir ^+/+ and Iir ^−/− mice. Interestingly, we noticed partially or totally missing PNA staining, indicating failure in acrosome formation (Figure 1B). The round nuclei of Iir ^−/− spermatozoa are indicative of a condition referred to as globozoospermia (Figure 1C). Globozoospermia and acrosome malformation are often accompanied by motility and mitochondria defects (Xiao et al., 2009; Han et al., 2017). Indeed, Iir ^−/− spermatozoa are characterized by reduced motility, as well as defective mitochondrial organization (Figures 1D–F).

Iir ^+/+ and in Iir ^−/− males did not differ in body weight (Supplementary Figure S2A), while Iir ^−/− males had a slightly increased testis weight without a significant difference in caudal sperm count (Supplementary Figures S2B, C). There was no difference in gross morphology of the testis and epididymis (Supplementary Figure S2D). By analyzing testicular cross sections by immunofluorescence, there were no differences between Iir ^+/+ and Iir ^−/− testes in germ cells marked by DDX4 and Sertoli cells marked by GATA-4 (Supplementary Figure S2E), as well as tubule diameter (Supplementary Figure S2F).

In the seminiferous tubules of the testis, inceptor is expressed in spermatocytes and in Golgi-phase and cap-phase spermatids but not in elongated spermatids in the acrosomal phase (Figure 2A). By confocal microscopy, we found inceptor in the vesicles of pachytene spermatocytes when the pro-AVs are harbored close to the Golgi apparatus and not yet attached to the nuclear membrane (Figure 2B). After meiosis, spermatids develop in 16 steps into spermatozoa (Oakberg, 1956; Wakayama et al., 2022) (Figure 2C). Confocal microscopy and TEM of immunogold labeled Tokuyasu-cryosections showed that inceptor is localized to pro-AVs and other vesicles before stage 3 (S3), before the pro-AVs attach to the nucleus (Figure 2D). Next, pro-AVs start to fuse into a cap-like structure (up to S7), when inceptor is predominantly present in the vesicles trafficking between the Golgi and the AV, and to a lesser extent in the inner and outer acrosomal membranes (Figures 2E, F). Then, the nucleus starts to elongate while the AV expands to cover most of the nuclear surface. The cell discards the Golgi apparatus and other organelles at this stage, while the inceptor immunosignal is completely lost (Figure 2G). In conclusion, in the early stages of acrosome formation, inceptor is localized to trafficking vesicles, pro-AVs, and the inner and outer acrosomal membranes of the developing AV.

As inceptor expression coincides with the early steps of acrosome formation, we used TEM to visualize the acrosomal shape in Iir ^+/+ and Iir ^−/− mice. As expected, in Iir ^+/+ mice, we observed electron-dense pro-AVs before acrosome formation, that are around 200-μm large in diameter (Figure 3A, white arrowhead). During pro-AV fusion, the single AV attaches to the nucleus at the beginning of the cap phase. At the nuclear membrane and AV contact site, the electron-dense acroplaxome is visible (Figure 3B, black arrows). As the AV expands and covers more nuclear surface, the acroplaxome also grows along the contact site (Figure 3C). After that, the nucleus and AV elongate, and the nucleus becomes more electron-dense (Supplementary Figures S3A–C). In Iir ^−/− testis, pro-AVs localize to the trans-Golgi network as in the Iir ^+/+ controls before acrosome formation (Figure 3D). At the beginning of the cap phase, however, pro-AVs do not form one large AV but remain segregated into around 200-μm large vesicles (Figure 3E). These vesicles are in close proximity to the nucleus, suggesting the correct delivery but failure to fuse at the acroplaxome. In later stages, the electron-dense acroplaxome expands around the nucleus, but the AV consists of small, irregularly distributed electron-dense vesicles in Iir ^−/− mice (Figure 3F). In further stages of acrosome formation, the nucleus fails to elongate and remains round or irregular in shape (Supplementary Figures S3D–F). By immunofluorescence, we observed some of the scattered acrosomal content marked by PNA localized to LAMP2-positive lysosomes, whereas in Iir ^+/+ cap-phase spermatids, LAMP2-positive lysosomes mainly localized to the posterior side of the spermatid (Figure 3G). Due to the similarity between inceptor and M6PRs, we investigated the localization of the cation-independent (CI-)M6PR and confirmed that CI-M6PR localization resembles a more diffuse vesicular localization than inceptor, with occasional colocalization with LAMP2-positive lysosomes (Figure 3H, white arrows).

As we observed small, deformed, or lacking acrosomes in Iir ^−/− spermatids, we performed whole-testis proteome analysis on Iir ^+/+ and Iir ^−/− testes to provide an unbiased picture of differential regulation of spermatocyte development. For the analysis, we considered 5389 considered protein groups, from which there were 28 significantly downregulated and one significantly upregulated protein (Figure 4A, Supplementary Table S1). We performed a GO term analysis of the significant hits on the background of all 5389 considered protein groups, which shows significantly deregulated terms related to acrosomes, secretion, and fertilization (Figure 4B). Specifically, we see significant downregulation of proteins localized to the acrosome, either soluble or transmembrane proteins. Many well-known proteins essential for fertilization are strongly downregulated in Iir ^−/− testes. Among these proteins were Acrosin (ACR), Acrosin-binding protein (ACRBP), Acrosomal vesicle protein 1 (ACRV1), Sperm acrosome membrane-associated 1 (SPACA1), and Equatorin (EQTN), which are among the main constituents of the acrosome. Additionally, Izumo sperm-egg fusion protein 4 (IZUMO4), and the Zona-pellucida binding protein (ZPBP), which have direct functions in sperm-egg fusion, were also downregulated. CD46 and TMEM190 are proteins specifically localized to the inner acrosomal membrane and we also found them downregulated in Iir ^−/− testes. Interestingly, Insulin-like factor 6 (INSL6) is also downregulated, a member of the insulin superfamily expressed specifically in the testis with expression onset in pachytene spermatocytes (Burnicka-Turek et al., 2009). We confirmed the downregulation of SPACA1 and Lysozyme-like 4 (LYZL4) by Western blot (Figure 4C).

As inceptor localizes between the Golgi and AV and in its absence the acrosome does not fully develop, we hypothesized that inceptor is necessary for trafficking between the Golgi and AV in developing spermatids. Therefore, the interaction partners and potential cargo of inceptor became of interest in deciphering the underlying mechanism of inceptor’s role in acrosome formation. We performed an inceptor co-IP experiment followed by mass spectrometry analysis (Figure 5A). We analyzed GO terms related to biological processes (GO:BP) on the 128 potential interactors (Supplementary Table S2). The significant GO:BP terms were reduced to eleven parental terms (Supplementary Figure S4). This GO:BP enrichment analysis revealed that inceptor interacts with proteins related to protein localization, organelle organization, catabolic process and vesicle-mediated transport (Figure 5B). From the list of candidate interactors, we validated the binding of inceptor to proteins that are well-known for their function in vesicle trafficking and fusion, such as syntaxin 7 (STX7) and microtubule-associated protein 1B (MAP1B) (Figure 5C). We also confirmed the interaction with the significantly downregulated LYZL4 and the M6P-containing cathepsin Z (Figure 5C). We have previously shown inceptor’s YXXθ to bind the µ subunit of AP-2 in pancreatic beta cells (Ansarullah et al., 2021), as this motif is known to bind the µ subunits of AP1-4 (Braulke and Bonifacino, 2009). Interestingly, we found the adaptor protein 3 μ2 subunit (AP3M2) to interact with inceptor (Supplementary Table S1), so we investigated adaptor protein and inceptor’s interaction in testes. We found that inceptor is bound by the AP complexes 1, 2, and 3, which likely mediate its subcellular localization and trafficking (Figure 5D). In summary, we propose that inceptor mediates vesicle trafficking between the Golgi and AV compartments and facilitates the transport and function of various proteins involved in cellular trafficking, such as proteins involved in vesicle fusion (Figure 5E).

Around seven percent of the human male population has a form of infertility but the underlying disease mechanism remains unknown in most cases (Krausz and Riera-Escamilla, 2018). Moreover, discovering factors that could serve as targets for on-demand male contraception is sought after (Kent et al., 2020). For this purpose, factors essential for spermatogenesis specifically expressed in spermatocytes and spermatids but not in spermatogonia are promising candidates.

Here, we described inceptor expression, localization, and function in the murine testis. Inceptor is essential for the development of morphologically intact, motile, and fertile spermatozoa. Inceptor’s expression onset is in primary spermatocytes, which arise from germ cells by meiosis. The expression is retained until the elongation stage of maturing spermatids. By immunofluorescence and immunogold labeling, we found that inceptor localizes to vesicles shuttling between the Golgi and acrosome compartments, and in its absence, pro-AVs do not fuse to form the acrosome. Together with the downregulation of proteins localized to the acrosome in Iir ^−/− testes and the enrichment of transport and localization-related proteins in inceptor’s interactome, these results indicate, that inceptor mediates intracellular transport to ensure pro-AV fusion. Indeed, we confirm inceptor to bind membrane-fusion and vesicle-trafficking factors, such as STX7 and MAP1B. As inceptor likely does not contain catalytically active domains and we described domains important for interacting with other proteins (Ansarullah et al., 2021), we propose that inceptor does not mediate the membrane fusion of the pro-AVs directly, but has a function in delivering factors that aid the membrane fusion and cargo delivery to the acrosome.

Interestingly, Pick1 ^−/− and Gopc ^−/− mice also show defects in acrosome formation and a failure of pro-AV fusion, similarly to Iir ^−/− mice (Yao et al., 2002; Xiao et al., 2009). From these published immunofluorescence images, PICK1 and GOPC seem to localize more broadly in spermatids compared to inceptor and the TEM images suggest that inceptor and PICK1 could be present in the same vesicles. Other mouse KOs also show globozoospermia and defects in acrosome formation, albeit with some differences. For example, Gm130 ^−/− spermatids form an AV which then fails to grow, whereas Dpy19l2 ^−/− spermatids show a destabilized nuclear membrane (Pierre et al., 2012; Han et al., 2017). Spaca1 ^−/− spermatids have a destabilized AV and lack the acroplaxome, whereas Zpbp ^−/− mice have a fragmented acrosome due to failure in acrosome compaction (Lin et al., 2007; Fujihara et al., 2012). These functional studies suggest, that each step of acrosome formation is tightly regulated by a variety of proteins. However, these regulatory steps seem to interconnected, as Gopc ^−/− mice also show reduced ZPBP1 and SPACA1 proteins, and Zpbp ^−/− mice also show a reduction in SPACA1 (Fujihara et al., 2012). Interestingly, neither PICK1, GOPC, nor DPY19L2 levels were changed in Iir ^−/− testes, whereas SPACA1 and ZPBP1 were among the most significantly downregulated proteins compared to Iir ^+/+ testes (Supplementary Table S1).

Inceptor binds insulin and proinsulin in pancreatic beta cells and regulates their degradation (Siehler et al., unpublished results). Intriguingly, INSL6, a relaxin-like protein and one of insulin-like factors specifically expressed in testes, was also downregulated in testes of Iir ^−/− mice. INSL6 levels have been found to correlate positively with male fertility in humans (Ivell and Grutzner, 2009; Chen et al., 2011; Ivell et al., 2017; Gumus et al., 2022). Inceptor also shares similarities to CI-M6PR, which transports IGF2 towards lysosomes, as well as acts as transport receptors for hydrolases toward late endosomes and lysosomes. As the acrosome shares similarities in origin and content to lysosomes, one hypothesis could be that M6PRs also contribute towards regulating the acrosomal content. Interestingly, the CI- and cation-dependent (CD)-M6PR have only been found to associate with LAMP1-positive lysosomes and not with acrosin-containing vesicles or acrosomes in spermatids (Martínez-Menárguez et al., 1996), similarly to our results (Figure 3H). In later stages, CI-M6PR has been found to transiently co-localize with acrosomes and has been proposed to contribute to shaping the acrosome (Moreno, 2003). However, in CI- and CD-M6PR mutants, the acrosomal content of selected hydrolases was not affected (Chayko and Orgebin-Crist, 2000). We have confirmed that the M6PR colocalizes with LAMP2-positive lysosomes rather than the acrosome, whereas inceptor is more closely located to the acrosome. In summary, we propose that inceptor is an important lysosomal trafficking receptor in different cell types, and an acrosomal trafficking receptor in developing spermatids.