Specifically, IZUMO1 protein is a testis-specific cell-surface protein belonging to the class of immunoglobulin type-I cell superfamily, characterized by a cytoplasmic C-terminal tail, a transmembrane region and a conserved ‘Izumo domain’, which is linked to an extracellular immunoglobulin-like (Ig-like) C2-type domain. It has been demonstrated that each of these domains plays a critical role in gamete adhesion
[10][16][17]. After AR in mammalians, IZUMO1 is localized in the EQ
[18] to mediate gamete adhesion with the oocyte plasma membrane. The absence of IZUMO1 leads to the impairment of gamete adhesion and to an accumulation of sperm in the perivitelline space
[10]. However, despite being a protein necessary to carry out membrane fusion, it lacks a fusogenic peptide or SNARE-like structure, and thus IZUMO1 could be one of the components that form the fusogenic machinery in spermatozoa, acting at the level of organization or stabilization of a multiprotein complex
[19].
2. Analysis of TMEM95
TMEM95 encodes a highly conserved single pass type I transmembrane protein consisting of 176 amino acids. The presence of a Pfam:IZUMO domain—database of protein families and domains—and a leucine rich repeat C-terminal domain (LRRCT) has been predicted, the latter being involved in a multitude of biological functions, such as signal transduction and cell adhesion
[25]. Its expression differs between species, being exclusive to the testis in rodents, while in cattle, it is also expressed in the brain. In addition, two splicing variants have been found in bovines
[26]. In mature mammalian sperm, this protein relocalizes from the outer acrosomal membrane to the EQ after undergoing AR. It has also been localized to the neck of bovine spermatozoa, indicating an additional role of TMEM95 during early embryonic cell divisions
[23]. This hypothesis was confirmed after observing a lower cleavage rate in embryos obtained from spermatozoa with mutated TMEM95
[24].
The protein sequence of TMEM95 is highly conserved among mammalian species, such as rodents, boars, bovines and primates, including humans
[20]. Therefore, genetic variants altering TMEM95 are likely to induce male subfertility. In humans, numerous polymorphic sites have been identified, including several potential loss-of-function variants
[23]. In other species (bovine and rodents), ablation or mutation of TMEM95 prevents penetration and fusion between the sperm and oocyte membranes, leading to sperm accumulation in the perivitelline space. This blockade of fusion does not generate structural defects in spermatozoa; therefore, it is suggested that TMEM95 may not play an important architectural role. TMEM95 deficiency is not detectable in a routine seminogram as it does not affect sperm motility or morphology.
Due to the normality of the seminogram, it might be advisable to develop functional assays, e.g., for the integrity of sperm-surface proteins or for effective prospective monitoring of male fertility
[22][23][24].
Assays performed in HEK293T cells, noted by their adherent growth, suggest that TMEM95 requires one or more partners as, by itself, it does not promote cell fusion after IZUMO1 and JUNO binding
[22]. Coimmunoprecipitation analyses suggest interactions of TMEM95, SPACA6, SOF1 and FIMP with IZUMO1
[20]. However, the nature of this cooperation is unknown as ablation of any of these genes does not affect the amount or localization of IZUMO1
[20][22][27]. It should be noted that the expression of these five proteins is not sufficient for sperm–oocyte fusion.
On the other hand, no interactions between TMEM95 and IZUMO1 or JUNO have been detected
[22]. Therefore, it is suggested that TMEM95, SOF1 and SPACA6 could directly or indirectly regulate membrane fusion through an IZUMO1-independent pathway or act as fusion mediators downstream of IZUMO1 and JUNO interaction. Membrane fusion is a highly coordinated and dynamic process, involving tight timing and proper interactions between molecules to allow the sperm head to internalize into the ooplasm
[20].
Interestingly, TMEM95 disappears in bovine spermatozoa that have undergone the AR
[24] implying a different function than IZUMO1. It is possible that TMEM95 does not act as a fusogenic protein that directly mediates membrane fusion between spermatozoa and oocytes. Rather, it could mask the actual fusogens that mediate fusion prior to AR; the release of TMEM95 could facilitate exposure of the functional domain of the fusogenic protein to the cell surface, which is necessary for fusion to occur. Alternatively, it is possible that sperm and oocyte membrane fusion is a bilateral fusion process, in which the fusogen must be present on both membranes; during acrosomal exocytosis, TMEM95 may be released and transferred to the oocyte plasma membrane, thus, ensuring that fusion occurs bilaterally
[20].
3. Analysis of IZUMO1
3.1. Characterization and Localization of IZUMO1
IZUMO1 encodes a single-pass type I transmembrane protein consisting of 350 amino acids with a 21 amino acid extracellular N-terminal signal peptide, a 20 amino acid transmembrane domain (amino acid position 293 to 313) and a 36 amino acid intracellular C-terminal domain (human Uniprot IZUMO1) containing several potential phosphorylation sites
[28]. It belongs to the immunoglobulin superfamily with an IZUMO domain and an extracellular immunoglobulin domain
[29]. The IZUMO domain includes a receptor-binding platform IZUMO1 to JUNO, which is a glycosylphosphatidylinositol (GPI)-anchored oolemma protein
[15][16][17]. An isoform of IZUMO1 (IZUMO1_v2) encoded by a different exon (exon 1b) of the Izumo1 through alternative splicing has been observed in mice. This isoform exhibits functional properties identical to the original protein
[30].
The IZUMO family consists of four proteins: IZUMO1, IZUMO2, IZUMO3 and IZUMO4. All of them share a region of residues of about 150 amino acids of high homology between them called IZUMO domain, located between the signal peptide and the immunoglobulin domain. Experiments with native and recombinant IZUMO proteins suggest that the IZUMO domain is involved in homodimer formation. These proteins, originally discovered in the mouse, have homologues in several mammalian species, thus it is a conserved family
[28]. Western blotting has detected IZUMO1 as a 56.4 kDa protein in mice, whereas in human weights 37.2 kDa
[10].
All members of the IZUMO family show a pattern of eight conserved cysteines and a similar predicted secondary structure consisting of four α-helices between cysteine motifs. Notably, IZUMO4 lacks a transmembrane domain
[28]. IZUMO1, a stable monomeric protein with extensive mixed α-β secondary structural features, was characterized in 2016. The overall structure consists of two domains: a four-helical N-terminal rod-shaped bundle of four helices (4HB; residues 22–134) and an immunoglobulin-like domain (Ig-like; residues 167–254). Two antiparallel β-strands (β1 and β2) function as a hinge between the 4HB and Ig-like domains. The four helices in the IZUMO1 4HB domain (α1, α2, α3 and α4) vary from 14 to 30 residues in length. The helices are amphipathic in character with a solvent-exposed polar surface and hydrophobic residues packed in a core
[16].
Generally, in immunolocalization studies, the samples were divided in capacitated or reacted (spontaneous or induced) spermatozoa and fresh spermatozoa as control group. Comparison between this groups shows the relocation of IZUMO1 during the AR. A few articles
[29][31] compare cryopreserved and fresh spermatozoa (control). These studies indicate the effects of the cryopreservation, such as changes in localization and density of IZUMO1.
There are slight differences in the localization of IZUMO1 between species. In mouse spermatozoa IZUMO1 is originally localized in the acrosomal membrane (inner and outer acrosomal membranes)
[18]. While in bull spermatozoa it is detected along the border between the principal and EQ of the acrosomal region
[29]. During the AR IZUMO1 gradually moves with the help of the actin cytoskeleton and testis-specific serine kinase (TSSK6) towards the EQ
[32][33]. It has been observed in rodents that the beginning of IZUMO1 relocation does not depend on the beginning of the AR, however, later stages of relocation correlate positively with the status of the AR. The complete translocation in the EQ coincides with the completion of the AR
[32].
IZUMO1 relocation in individual sperm populations during the spontaneous AR correlates with species-specific promiscuity behaviour, exhibiting faster relocation. Notably, in rodents, induction of the AR by calcium ionophore provides comparable or identical results to the spontaneous AR, whereas induction by progesterone completes relocation 20 min faster than in the spontaneous AR
[32]. Therefore, it is suggested that IZUMO1 relocation is independent of external or internal calcium ion triggering the AR. While progesterone, a hormone secreted by cumulus oophorus cells responsible for the induction of the AR, is able to initiate IZUMO1 relocation.
The location of IZUMO1 can be affected by several factors. One of them is the loss of sperm equatorial segment protein 1 (SPESP1), an acrosomal protein that localizes to the EQ after the AR. Its loss in both Spesp1
−/− and Spesp1
−/+ mice results in an altered localization of IZUMO1, being distributed over a wider area and in an irregular manner. This event results in an inhibition of sperm–oocyte fusion, observed in both mouse and human sperm
[34].
The effect of cryopreservation on IZUMO1 location has been observed in bull spermatozoa
[29]. Cryopreservation generates severe damage to the acrosome, resulting in aberrant translocation of IZUMO1. After induction of the AR, a higher loss of acrosomes is observed in cryopreserved spermatozoa compared to freshly ejaculated samples. Consequently, the percentage of cryopreserved spermatozoa with normal or abnormal acrosome and a normal IZUMO1 distribution pattern decreases about a 9 ± 5%. Although the detailed mechanism for IZUMO1 translocation in cryopreserved sperm is still unclear, Fukuda et al. suggest that cryopreserved sperm undergo changes, such as the capacitation process induced by the freezing and thawing process and by exposure to components such as egg yolk
[29].
3.2. Role of IZUMO1 in Mammalian Fertilization
The molecular mechanism of recognition between IZUMO1 and JUNO was recently discovered, in which IZUMO1 interacts with JUNO through its N-terminal domain
[35]. The crystal structure indicates that the residues of the three IZUMO1 regions (4HB, hinge and Ig-like) contact JUNO 20-228 through extensive van der Waals, hydrophobic and aromatic interactions. There are also two intermolecular salt bridges and eight hydrogen bonding interactions. However, all these interactions are weak
[16]. On the other hand, Inoue et al. described two types of dimer configuration in IZUMO1: open and closed. The open dimer could increase JUNO affinity to increase the fertilization rate. While the closed dimer directly participates in membrane fusion after binding to JUNO
[36].
First, JUNO binds to monomeric IZUMO1, gradually accumulating at the sperm contact site to induce its dimerization, which is followed by a tight junction phase in which IZUMO1 folds the entire structure to the sperm membrane side through a thiol-disulphide exchange reaction. Following this, IZUMO1 stops binding to JUNO to bind to a second putative oocyte receptor.
Thus, the role of JUNO is to rearrange IZUMO1 so that it can generate the necessary strength to collapse the repulsion between the juxtaposing membranes through an unidentified receptor on the oocyte. However, this is not sufficient to fuse both membranes, requiring other proteins, such as CD9 in the oocyte and SPACA6 in the sperm
[12][13][37][38]. CD9, IZUMO1 and JUNO are involved in gamete recognition and membrane adhesion, but do not induce membrane fusion
[15]. Interestingly, CD9 has been shown to regulate the interaction between JUNO and IZUMO1 in wild-type mice
[39].
Recently, the structure of the human JUNO/IZUMO1 complex has been reported and critical amino acids have been identified during the two molecules interaction
[16][17]. According to structural data, mutations in W148, H157 and R160 of IZUMO1 and W52 and L81 of JUNO are essential for sperm and oocyte plasma membrane recognition
[17].
After fusion, the sperm membrane comprises a continuous single membrane plane with a complicated invaginated structure. However, sperm–oocyte fusion is not completely achieved at this point, as electron microscopy images have showed, the “internalization” of the invaginated inner acrosomal membrane occurs later in the fertilization process
[40].