Once the sperm cross the entire ZP and enter the space between the matrix and the oocyte membrane, known as the perivitelline space, they face the molecules responsible for the binding and fusion of membranes for fertilization. The first molecule in the oolemma described as essential for fertilization was CD9 (Le Naour et al., 2000; Miyado et al., 2000), a cell surface protein belonging to the tetraspanin superfamily, which participates in cell migration, adhesion, proliferation, differentiation, and signal transduction (Table 1). The tetraspanins are associated with several molecules located in the cell surface such as β1 integrins (Le Naour et al., 2000), and CD9 might be implicated in the assembly or the stabilization of these adhesion molecules (Kaji et al., 2000). To study the role of this protein in fertilization, CD9-deficient mice were generated (CD9^–/–) and ZP-enclosed oocytes collected and inseminated in vitro to elucidate in which step of fertilization the failure occurred. Whereas the ZP was penetrated by wild-type sperm, it was unable to fuse with the oolemma, penetrate into ooplasm, and form pronuclei (Miyado et al., 2000). Insemination of CD9^–/– ZP-free oocytes was then performed observing the same pattern as with ZP-enclosed oocytes and yielding 98% fusion rates for CD9^+/+ ZP-free oocytes and 4% for CD9^–/– ZP-free oocytes. When the sperm were injected into oocytes, however, similar implantation and embryo development rates were obtained from eggs derived from wild-type and CD9^–/– mice, thus confirming the implication of CD9 in sperm–egg fusion (Miyado et al., 2000). Later, it would be discovered that CD9 is enriched on the oocyte microvillar membrane, generating fusion-competent sites for fertilization (Jégou et al., 2011), and it is crucial for a normal shape and distribution of the membrane microvilli (Runge et al., 2007). Indeed, a recent analysis of CD9 crystal structure suggests that the highly asymmetric shape of this tetraspanin is directly involved in the clustering-induced curvature generation and its localization at high curvature regions (Umeda et al., 2020), whereas a large extracellular loop would be responsible for the critical role of CD9 in the sperm–egg fusion (Zhu et al., 2002; Umeda et al., 2020).

The oocytes express more tetraspanins in addition to CD9. For instance, CD81 has been described in mouse and human oocytes (Takahashi et al., 2001), and CD151 is also present in human eggs (Ziyyat et al., 2006). CD81-deficient female mice (CD81^–/–) presented a 40% reduction in fertility, and in vitro fertilization (IVF) assays suggested that this phenotype is due to a failure in sperm–egg fusion. CD9 and CD81 might play a complementary role in gamete fusion since double knockout mice (CD9^–/– and CD81^–/–) were completely infertile (Rubinstein et al., 2006; Ohnami et al., 2012). The presence of CD9 and CD81 has been recently described also in the sperm head, over the acrosome and in the plasma membrane, respectively, suggesting once more the critical role of these proteins in sperm–egg fusion (Frolikova et al., 2018). Also CD9 and CD81 have been detected in bovine and porcine oocytes’ plasma membrane, and their role in fertilization has been studied by in vitro approaches blocking these receptors with antibodies (Jankovicova et al., 2019). The crystal structure of CD81 has revealed that cholesterol binds CD81 within the intramembrane cholesterol-binding pocket, thus modulating conformation of the molecule (Zimmerman et al., 2016). CD81 may exist in both close and open conformations, and the equilibrium between both shifted toward the close state when cholesterol is bound, disfavoring its binding to CD19 partner (Zimmerman et al., 2016). This conformational change has also been proposed for CD9, and molecular dynamics simulation suggests that binding lipid molecules to the central cavity may disturb structural conformation, affecting binding partner affinity to EWI-2 (Umeda et al., 2020). Undoubtedly, the role of these proteins, whose activity can be modulated by their binding to oolemma components, is essential to mediate the gamete membrane binding and fusion processes for a successful fertilization.

In 2014, more than a decade after was discovered, a glycosylphosphatidylinositol (GPI)-anchored protein was identified in the oocyte as vital for fertilization (Bianchi et al., 2014). This protein encoded by the Folr4 gene was known as Folate receptor 4 due to its putative function in folate uptake. The expression of this protein in mice CD4⁺CD25⁺ regulatory T-lymphocytes, as well as its function as a mediator of responses to dietary folate and its use in anti-tumor therapy, had been widely studied, but its presence and function in mammalian oocytes were unknown until 2014. In this study, it was demonstrated that Folr4 protein was unable to bind folate, but it was essential for fertilization, so the authors renamed this protein “JUNO,” in honor of the fertility and marriage Roman goddess (Bianchi et al., 2014). In this work, the key role of JUNO in the interaction between the sperm membrane and the oolemma was assessed by IVF assays in mice. Firstly, when IVF was performed pre-incubating for 10 min zona-free eggs with an anti-JUNO antibody, no fertilization occurred. Secondly, female JUNO-deficient mice (JUNO^–/–) were generated, being infertile. No differences were observed in the oocyte’s morphology nor in the number of ovulated oocytes when compared with wild-type mice. However, eggs from JUNO^–/– mice were not fertilized after natural mating nor IVF. Moreover, ZP-free-JUNO^–/– oocytes were also tested in IVF, obtaining the same result. Altogether, these experiments confirmed the involvement of JUNO in sperm–egg binding. Another interesting observation from Bianchi and Wright’s study (Bianchi et al., 2014) was that JUNO seems to be involved in the block of polyspermy occurring after fertilization by protein shedding from the oolemma within 40 min after sperm penetration.

Two years after JUNO’s identification as the IZUMO1 binding partner in the sperm, several research groups revealed the tertiary structure of JUNO and JUNO–IZUMO1 complex, thus allowing the identification of the precise amino acids involved in this recognition through van der Waals interactions, suggesting that W148 of IZUMO1 interacts with L81, L82, M83, and P84 of JUNO and that W62 of JUNO interacts with R160, K161, S162, and Y163 of IZUMO1, being responsible for sperm–oocyte plasma membrane interaction (Aydin et al., 2016; Han et al., 2016; Kato et al., 2016; Ohto et al., 2016; Inoue, 2017). Besides, studies focused on knowing the precise time of JUNO expression during oocyte maturation have been performed in mice, concluding that JUNO expression starts when the oocyte reaches 13- to 22-μm diameter (Suzuki et al., 2017), coincident with the time that oocytes gain the ability to fuse with the sperm, at 20 μm (Zuccotti et al., 1994). It was also found that JUNO was expressed at the prophase I, germinal vesicle stage; and its expression continued at similar levels until metaphase II stage, thus being accumulated in oolemma during oocyte maturation (Suzuki et al., 2017). Also in humans, a recent study confirmed that JUNO is involved in human gamete recognition via interaction with human IZUMO1 and that its expression progressively increases during oocyte maturation, starting at prophase I and significantly increasing through metaphase I and II stages (Jean et al., 2019). In order to determine how CD9 and JUNO cooperate, a recent work performed a detailed analysis of the area where the sperm–egg fusion occurs, revealing that whereas CD9 is still fully present, JUNO disappears from the oocyte’s surface in concentric circles, suggesting that the molecular dynamics of CD9 and JUNO are differentially regulated (Inoue et al., 2020). Moreover, JUNO localization was abnormal in oocytes obtained from CD9^–/– mice since JUNO was localized in the cortical actin cap region lacking microvilli, whereas it was nearly absent from the cortical actin cap in the wild-type oocytes, suggesting that CD9 may play a crucial role in the molecular compartmentalization of GPI-anchored proteins (Inoue et al., 2020).

To date, only the few proteins in the oolemma mentioned above have been demonstrated to be crucial for gamete binding, thus contributing to fertilization process. Almost all of them have been described thanks to the use of gene editing technologies by generating gene-deficient/modified mice. However, we could define the criterion of “essentiality” too strict to be able to address the molecules that may be involved in fertilization process mainly due to a redundant effect of the participating molecular cascades and that some candidates may have a complex and fundamental activation in other cellular processes so that its deficiency makes the organism unviable. This may be the case of the proteins involved in the membrane fusion process, where no candidate with a specific function in fusion has been described in the oocyte and that is totally essential (Bianchi and Wright, 2020). Indeed, recently, it has been defined that the role of a phosphatidylserine (PtdSer) is exposed on the head region of the sperm as a key player in sperm–egg fusion (Rival et al., 2019). Interestingly, the authors in this study have found similarities in gene expression pattern between proteins involved in cell fusion on skeletal myoblasts and egg, so that several PtdSer receptors described in myoblast have been detected in oocytes, i.e., CD36, brain-specific angiogenesis inhibitor 1 (BAI1), brain-specific angiogenesis inhibitor 3 (BAI3), T-cell immunoglobulin and mucin domain containing 4 (Tim-4), and Mer tyrosine kinase (Mer-TK). Using antibodies that bind to PtdSer (BAI1/BAI3- and CD36) or to PtdSer-binding domain (Annexin V) and single- or double-knockout mice (for Tim-4, BAI1, and Mer-TK), this study concludes that even with the considerable redundancy among the PtdSer receptors, they contribute to sperm–egg fusion.

Despite extensive research, the identity and presence of oolemma receptors and fusogens remain ambiguous, and our comprehension of the exact mechanisms mediating gamete binding and fusion is incomplete due to the transient nature of the binding event and the highly orchestrated and dynamic fusion mechanism. Therefore, the concomitant use of different methods can better inform us about the role of a specific molecule involved in fertilization. The methods used up to now can be classified into in vivo models, i.e., gene-edited animals, and in vitro models, i.e., solubilized, purified, and recombinant proteins (Caballero-Campo et al., 2006; Chiu et al., 2008; Bhandari et al., 2010), and antibodies against the protein of interest (Jankovicova et al., 2019). Undoubtedly, gene-edited mice have been of the outmost importance to reveal new molecules involved on the different fertilization stages. However, this technology is expensive, time-consuming, and not easily transferrable to livestock species, where it is not as much developed as in the murine model. So alternative strategies for studying molecular fertilization have been used. We anticipate that numerous candidates may emerge thanks to application of high-sensibility new screening strategies such as low-input proteomics (Virant-Klun et al., 2016), large-scale mutagenesis screens, and protein-homology analysis from diverse taxa. However, efficient and low-time-consuming experiments should be applied to optimize the experimental conditions to recapitulate gamete binding and fusion. The most recent technology applied to screen the binding and fusion of gametes is here reviewed.