The molecular mechanisms underlying Ca2+ entry into sperm are now much more well defined thanks to direct recordings of mature sperm cells. This article reviews the function of a sperm-specific ion channel, CatSper. CatSpers have a clearly defined function in sperm's hyperactivated motility and are essential for male fertility. We propose that physiological stimuli such as zona pellucida and cyclic nucleotides induce Ca2+ entry through CatSper channels instead of acting on CaV and CNG channels as previously thought.
What Is the Role of Ca2+ Signaling in Sperm Function?
During natural fertilization, the goal of a sperm is simple and fundamental: to fuse with the egg and deliver genetic material. Despite this simple goal, there are many obstacles a sperm has to overcome, with Ca2+ involved in nearly every step (99). For example, Ca2+ is essential for sperm capacitation, chemotaxis, and egg penetration. Only with completion of capacitation are freshly ejaculated sperm able to fertilize an egg. This “maturation” process is completed in the female reproductive tract. Capacitation has been replicated in vitro for several different species using defined media that include calcium, bicarbonate, and bovine serum albumin (BSA) (5, 19, 99). Capacitation is accompanied by several cellular changes including membrane hyperpolarization, changes in membrane lipid composition, intracellular alkalinization, increased level of protein tyrosine phosphorylation, and increases in intracellular concentrations of Ca2+ and cAMP (10, 83, 86). In addition, some of the changes (e.g., intracellular alkalinization and increased cAMP levels) can elicit Ca2+ entry on their own, suggesting an interplay between extracellular Ca2+ and intracellular signaling molecules during the capacitation process.
Most of the sperm that are deposited into the female reproductive tract are trapped in the sperm reservoir in the fallopian tube by their attachment to epithelial cells (83, 99). On ovulation, it is hypothesized that sperm are “attracted” to the eggs by the follicular fluid released during ovulation as well as by chemoattractants that are released from the cumulus-oocyte complex (26). Although the molecular mechanisms of chemotaxis are largely unclear, Ca2+ plays a central role. Consistent with a role for Ca2+ entry in the fertilization process, several of the chemoattractants associated with fertilization, such as follicular fluid and progesterone, as well as the molecules/events downstream of these chemoattractants, such as cyclic nucleotides (cAMP/cGMP) and membrane depolarization, have been shown to induce Ca2+ influx, presumably through Ca2+-permeable ion channels (14, 23, 26, 31, 33, 46, 71, 89).
Of the sperm that are deposited, only a small fraction will successfully migrate from the sperm reservoir to the site of fertilization, and even then the sperm will encounter a final barrier before reaching the egg: the cumulus layer. Substances secreted from cumulus cells, such as progesterone, trigger Ca2+ influx into sperm. Specifically in the case of progesterone, the acrosome reaction (AR) will also be induced in sperm (26, 71).
The egg is surrounded by a thick layer of zona pellucida (ZP) glycoproteins (89). ZP glycoproteins are able to induce an increase in [Ca2+] in sperm, initiate changes in sperm motility, and promote the AR (33, 34, 36, 50). Each of these responses involve Ca2+ entry and are hypothesized to help sperm penetrate the egg coating to reach the egg membrane through a combination of mechanical forces and chemical digestion. On fusion with the egg membrane, the entry of a sperm releases sperm “factors” that trigger a wave of Ca2+ as a signaling event in the egg leading to the initiation of egg activation and the creation of a new life (89, 99).
Where Does Ca2+ Come From?
Similar to somatic cells, Ca2+ entry from the extracellular space and Ca2+ release from intracellular stores are responsible for increases in [Ca2+]i (11, 14, 23, 33, 71). However, mechanisms associated with the release of Ca2+ from intracellular stores remain largely unclear. Unlike most cells, mature sperm do not contain an ER, which is typically a major source for Ca2+ release, and the long principle piece region of sperm does not contain intracellular organelles. Therefore, potential areas for Ca2+ stores in sperm include the acrosome in the head, a redundant nuclear envelope (RNE) that colocalizes with the IP3 receptor in the neck region, and mitochondria packed in the midpiece (21, 44). The low basal [Ca2+]i level is maintained by Ca2+ absorption by mitochondria and active Ca2+ extrusion by the plasma membrane Ca2+ pump (PMCA) (90). PMCA4 found in the tail of sperm is required for normal sperm motility and male fertility (70, 78).
Ca2+-Permeable Channel Proteins in Sperm
How extracellular Ca2+ enters sperm is also poorly understood. A large number of plasma membrane Ca2+-permeable ion channel proteins and mRNAs have been found in testis and sperm over the past decades (23). These include high-voltage-activated and low-voltage-activated (T-type) Ca2+-selective channels (CaVs), TRP channels, cyclic nucleotide-gated (CNG) channels, and CatSper channels. In support of possible functions of these channels, stimuli that can open them by, for example, depolarization for CaVs (8, 32, 91), menthol/low temperature for TRPM8 (25), and cAMP/cGMP for CNG channels led to an increase in [Ca2+]i (94). Several plasma membrane Ca2+ transporters (e.g., PMCA4) and neurotransmitter receptors coupled to Ca2+ entry (e.g., nAChRs) have also been found in sperm (64, 70, 78). Given the relative high homogeneity of a sperm cell population and the arguably “single-minded” function of sperm compared with neurons where most channel proteins were initially discovered, it is intriguing that so many Ca2+-permeable ion channel proteins/mRNAs found in neurons are also present in testes and sperm. For example, 9 of the 10 CaV α1 subunits have been detected in testis (11). Therefore, a key question is: Does the mere existence of a protein in sperm indicate a functional significance for sperm physiology? We propose that if a protein meets the following criteria, it is clearly a sperm ion channel with physiological importance: 1) the protein is detectable in sperm, ideally with knockout sperm as a negative control for antibody specificity; 2) the ion channel current generated by the protein is detectable by patch-clamp recording; 3) blocking the channel with highly selective blockers impairs sperm function; 4) mutations in the gene encoding the protein leads to defects in sperm action and male fertility. The last criterion can be complicated by genetic redundancy and misregulation of gene expression in knockout models. Therefore, a protein that does not meet this criterion could still be potentially important for sperm function in vivo. Similar criteria have been used to evaluate ion channels in the nervous system and the cardiovascular system. In our opinion, most of the Ca2+-permeable channel proteins reported so far from testis and sperm, as listed in the beginning of this section, do not meet these criteria, and their specific roles in sperm Ca2+ signaling remain established.
The CatSper Channel Complex
In sperm, a family of ion channels that clearly meets most of the criteria proposed above is the CatSper channel family. There are four pore-forming CatSper channel proteins, CatSper1–4. Like KV, TRP, and CNG channels, CatSpers contain six transmembrane-spanning (TM) domains (48, 62, 67, 72, 73, 75). In contrast, CaVs and NaVs contain 24 TM domains (4 × 6 TM). The overall sequences of CatSpers, however, are similar to those of CaVs; and they have the T/S-x-E/D-x-W signature sequence in the ion selectivity filter region, a feature shared by CatSpers and CaVs. The intracellular NH2 terminus of CatSper1 is rich in histidines (49 out of 250 amino acids in the mice; Ref. 75), which may be related to the pHi sensitivity of the channel (see below). In addition, the S4 segments of CatSpers have charged residues (R/K) at every third position, a signature of voltage-gated ion channels.
When the CatSper protein complex was purified from mouse testis, a testis-specific HSP70 protein, HSP70-2, an intracellular protein involved in protein-folding, and at least two auxiliary subunits, CatSperβ and CatSperγ, were also identified (60, 88). CatSperβ is predicted to contain two TM domains, whereas CatSperγ is predicted to contain a single type I TM domain (FIGURE 1). Both of the auxiliary subunits have large segments of ∼1,000 amino acids that are presumably extracellular, although they do not contain any recognizable protein domains to suggest function.
CatSpers, CatSperβ, and CatSperγ have been found in a wide spectrum of animals, from humans and mice to sea urchins and sea squirts. The existence of CatSpers in echinoderms (e.g., sea urchins Strongylocentrotus purpuratus), tonicates (e.g., Ciona intestinalis), and cnidarians (e.g., Nematostella vectensis) suggests that this complex arose early in metazoan evolution (16). So far, CatSper proteins have not been identified in protostomes such as Drosophila melanogaster and Caenorhaditis elegans, and it is intriguing that some vertebrates (e.g., zebrafish) do not appear to express CatSper proteins. The lack of CatSpers in some animals suggests that fertilization steps common to the animals, such as sperm-egg fusion, do not require CatSpers. Like many other fertilization-specific molecules, the degree of sequence identity between CatSper orthologs, even between human and mice, is strikingly low (∼50%) compared with that of many other ion channel proteins found in the brain (∼90%).
In mice, CatSper proteins 1–4, β and γ appear to have testis-specific expression patterns (48, 60, 72, 73, 75, 88). Using knockout mice as a control, immunostaining assays have found that the CatSper proteins are localized in the principal piece of sperm (60, 72, 75, 88) (FIGURE 1). The mechanisms of such a striking subcellular localization pattern are not clear but are perhaps based on interactions among the CatSper proteins and the auxiliary subunits. For example, in the CatSper1 knockout, CatSper proteins 2, β, and γ are all expressed at normal levels in testis yet are undetectable in sperm (18, 60, 72, 88). Similarly, CatSper1 is expressed at normal levels in Catsper2 knockout testis, but is absent in sperm (18). These findings suggest that the synthesis of each CatSper protein is independent of the other CatSper proteins in sperm precursors, but trafficking of the proteins to the sperm tail and/or the stability of them in the location is dependent on each other.
Biophysical Properties and Activation Mechanisms of CatSper Channels
The biophysical properties of CatSper channels have not been well characterized since these complexes, like several other sperm-specific ion channels/transporters, have been resistant to functional reconstitution in heterologous systems for reasons that remain unknown. In contrast, several single 6TM proteins with similarity to CatSper proteins that have been identified in bacteria form voltage-sensitive Na+ and Ca2+-permeable channels following heterologous expression, suggesting that 6TM proteins, like the 24TM CaV and NaV proteins, are able to form Ca2+ and Na+ channels (56, 76, 101).
CatSper knockout sperm do not have a Ca2+-permeable channel current (ICatSper) that can be detected from the principal piece of wild-type sperm, suggesting that CatSper channels are responsible for the current (55). Under physiological conditions, ICatSper is permeable to Ca2+, and, similar to CaVs, CatSper channels become permeable to monovalent ions when extracellular cations are removed.
The S4 helices of all the CatSper proteins have charged residues (K/R), and the conductance of CatSper proteins has some voltage dependence, although it is weak. For example, the slope factor for CatSper proteins is ∼30 mV, compared with several millivolts exhibited by other voltage-gated channels (55). This suggests that voltage is less likely to be a trigger for the opening of CatSper channels. Furthermore, voltage-independent activation is supported by the observation that physiological stimuli such as ZP glycoproteins and BSA are able to readily elicit CatSper-dependent increases in [Ca2+]i when Vm is “clamped” by the K+ ionophore valinomycin (97).
A direct activator of CatSper is hypothesized to be intracellular alkalinization. For example, when pHi is increased from 6.0 to 7.5, the V1/2 values of the conductance (G)-voltage (V) curve shift from 87 to 11 mV. This makes CatSper channels more accessible at physiological membrane potentials (55). However, the mechanism(s) by which intracellular alkalinization activates CatSper channels remain unknown.
There are likely additional activators of CatSper channels. For example, when CatSper-dependent increases in [Ca2+]i is elicited by BSA, the [Ca2+]i increase is maintained when both intracellular pH and membrane potential are “clamped” (97), suggesting potential CatSper activators distinct from intracellular alkalinization. It is also possible that many of the G-protein-coupled receptors (GPCRs) that are found in sperm, and the signaling molecules that are downstream of these receptors (29), could potentially modulate/activate CatSper channels. Finally, CatSperβ and γ have large extracellular domains (60, 88), which raises the possibility that direct “ligands” for CatSper channels may exist.
CatSper Channels in Male Fertility
The importance of CatSper proteins in male fertility has been clearly established: disruption of the CatSper genes leads to complete male infertility in both humans and mice. In mice, all four CatSper genes have been knocked out, and each of these knockout models share the phenotype of male infertility (47, 72, 74, 75). Consistent with the restricted expression of CatSper proteins in sperm, the defects present affect sperm physiology rather than spermatogenesis or spermiogenesis. The identical phenotype of the four knockout models and the physical association among the proteins suggest that the four CatSper proteins form a heteromeric channel, and disruption of any of the four proteins will result in a nonfunctional channel (72). This interpretation, however, is complicated by the fact that knocking out one protein destabilizes the others.
CatSper mutations have also been found in infertile humans. Mutation in CatSper1 and 2 are associated with asthenoteratozoospermia and male fertility (6, 7). It is not clear whether suboptimal function of CatSper proteins caused by genetic polymorphisms or environmental influences are associated with male infertility, and no mutations in CatSperβ or γ have been reported.
CatSper Channels in Sperm Hyperactivated Motility
The most dramatic behavior defect in CatSper knockout sperm is a lack of hyperactivated motility. In addition to the more conventional form of basic motility, hyperactivated motility is characterized by an asymmetric vibrant beating of a sperm tail, which occurs during sperm capacitation and when sperm are stimulated (43, 100). This type of motility is thought to generate a more powerful force. However, the function of this motility in vivo has been hard to evaluate. Since CatSper knockout sperm do not undergo hyperactivated motility yet maintain their basal motility function, they represent a model in which to test the in vivo function of hyperactivated motility. In in vitro fertilization (IVF) assays, CatSper knockout sperm are unable to penetrate the ZP surrounding an egg but are able to fertilize the eggs when the zona pellucida are enzymatically removed (74, 75). In addition, the mutant sperm are not able to move beyond the oviductal reservoir, presumably because they do not have sufficient mechanical force to elude the adherent epithelial cells in the oviducts or, alternatively, because they have a lack of chemotaxis (45).
Physiological Stimuli Leading to CatSper-Dependent Ca2+ Entry
CatSper channels are a convergent point for several Ca2+ signaling cascades. Some of the most important stimuli in sperm function have been shown to induce CatSper-dependent increases in [Ca2+]i, which include cyclic nucleotides, alkaline depolarization, ZP glycoproteins, and BSA.
Cyclic Nucleotide-Induced Ca2+ Entry
The importance of cyclic nucleotides (e.g., cAMP, cGMP) in sperm physiology is perhaps best illustrated by the complete male sterile phenotype exhibited by mice deficient in sACY, a soluble adenylyl cyclase that generates cAMP in sperm in response to physiological stimuli such as bicarbonate (20, 28, 42, 61). Cyclic AMP is also proposed to be a mediator during sperm chemotaxis triggered by odorant receptor activation in human sperm (79). Both cAMP and cGMP elicit increases in [Ca2+]i in sperm as demonstrated in assays where cell-permeable cAMP/cGMP is applied or when caged-cGMP is uncaged using UV flashing techniques (1, 75, 94).
The cAMP/cGMP-induced increases in [Ca2+]i are dependent on extracellular Ca2+, suggesting that Ca2+-permeable channels are involved. In photoreceptors and olfactory neurons, cAMP/cGMP activates CNG channels by binding to the carboxy-termini of the channel proteins (22, 51). Several CNG channel proteins have been detected in sperm and are hypothesized to mediate cAMP/cGMP-induced Ca2+ responses (93, 94). However, whole-cell sperm patch-clamp recordings have failed to detect any current directly activated by cell-permeable cGMP (54). In addition, mice deficient in the CNG channel proteins are fertile and have not been shown to exhibit any defect in sperm function (12, 65). Similarly, humans carrying mutations in the CNG genes have impaired vision but not male infertility (12). These observations seem to be inconsistent with the presumed roles of CNGs in the cAMP/cGMP-induced Ca2+ signaling in sperm physiology. Another model for the cAMP/cGMP-induced [Ca2+]i increases proposes that cyclic nucleotides activate the hyperpolarization-activated cation channels (HCN), resulting in a depolarization and a subsequent opening of CaVs (24, 53). Two HCN channel proteins have been found in sea urchin sperm (39, 40). In mammals, there are four HCNs (HCN1–4). HCNs 1, 2, and 4 have been knocked out in mice, and no defect in sperm function or fertility has been reported, although these mice have profound phenotypes in cardiac and neuronal function (13, 77). These data, in our opinion, suggest that the presumed cAMP/cGMP-HCN-CaV pathway has no significant in vivo function in sperm physiology and male fertility.
In contrast with wild-type sperm, application of cell-permeable cAMP/cGMP (e.g., 8-Br-cAMP, 8-Br-cGMP) to sperm lacking CatSper channels does not elicit increases in [Ca2+]i (75, 96). However, Ca2+ responses can be restored in CatSper-deficient sperm with exogenous expression of a GFP-CatSper fusion protein, suggesting that the increases in [Ca2+]i induced by cAMP/cGMP are mediated by CatSper channels. Moreover, the localization of CatSper channels is consistent with the cGMP-induced increases in [Ca2+]i observed in the principal piece of sperm (96). Although it is not clear how cAMP and cGMP induce a Ca2+-influx through CatSper channels, it is hypothesized that the role of cAMP and cGMP is indirect since application of 8-Br-cGMP does not appear to influence ICatSper during whole-cell voltage-clamp recording assays where intracellular pH is buffered (55).
Zona Pellucida-Induced Ca2+ Entry
Consistent with the model that contact between a sperm and the egg coat triggers Ca2+ entry into the sperm, bath application of solubilized ZP glycoproteins from bovine or mouse induces an extracellular Ca2+-dependent increase in [Ca2+]i in sperm, suggesting an entry of Ca2+ (9, 30, 35). In mice, the proteins that constitute the zona pellucida matrix are ZP1, ZP2, and ZP3 (89). Experiments using gel-purified ZPs have indicated that ZP3 is a Ca2+ influx-inducing protein (4). The function of ZP-induced increases in [Ca2+]i is presumably twofold: 1) to change sperm motility and 2) to induce the AR. Both of these events assist sperm in overcoming their final obstacle of penetrating the egg coat (99).
How contact between a sperm and the ZP triggers increases in [Ca2+]i has been an active area of research. Although the molecular details remain unclear, “sensing” of the ZP by sperm is hypothesized to be mediated by GPCR(s) since ZP-induced increases in [Ca2+]i can be blocked by G-protein inhibitors such as PTX and phospholipase inhibitor neomycin (35). In addition, males deficient in a phospholipase C isoform (PLCδ4) are subfertile, and their sperm have a reduced capacity for ZP-induced increases in [Ca2+]i (37, 38). Using labeled ZP glycoproteins as “ligands,” several sperm proteins that are able to bind ZP glycoproteins have been identified. However, the receptors mediating the G protein-dependent increases in [Ca2+]i triggered by ZP glycoproteins have not been established (66, 89).
In capacitated mouse and bovine sperm, ZP glycoproteins induce both a fast [Ca2+]i response that occurs on the millisecond to second scale as well as a slower/sustained response that extends over several minutes (30). Based on experiments that demonstrated signal transduction by ZP glycoproteins could be mediated by GPCRs and that increases in [Ca2+]i could be induced by thapsigargin (an IP3 receptor antagonist commonly used to deplete intracellular Ca2+ stores), the sustained response was proposed to be mediated via store-operated channels (SOCs) that are activated on Ca2+ store depletion (68). Additional experiments need to be performed to determine whether store depletion per se can activate channels, preferably with patch-clamp recordings where the “degree” of intracellular store depletion can be better controlled and channel activities can be directly measured. A channel that is proposed to mediate the sustained Ca2+ response is TRPC2, a member of the TRPC ion channel family (49, 84). Polyclonal antibodies raised against TRPC2-derived peptides have been shown to block ZP-induced [Ca2+]i increases in mouse sperm, as well as the ZP-induced AR. Trpc2, however, is a pseudogene in humans (85), and in TPRC2 knockout mice, despite having strong behavioral phenotypes, no defect in sperm function or fertility has been reported (58, 81). Possible interpretations of these data include 1) TRPC2 does not encode the channel in vivo, and the polyclonal antibodies used had nonspecific actions; 2) there is upregulation of another TRP in TRPC2 knockout sperm; 3) humans use a different TRP channel, and/or the sustained rise in [Ca2+]i is nonessential for sperm function and male fertility. However, the latter is unlikely given that the AR strongly correlates with the sustained increase in [Ca2+]i observed (35). In addition, it has recently been discovered that the most well characterized store depletion-activated channel, CRAC, is formed by ORAI and STIM, two proteins unrelated to TRP channels, and such a channel activation involves a close interaction between the ER and plasma membrane, which may not exist in sperm (15, 69). In summary, whether store depletion directly activates channels in sperm and the identities of the channel constituents remain established.
The molecular mechanisms underlying the early, rapid rise in [Ca2+]i have also been extensively studied. Consistent with the hypothesis that an increase in [Ca2+]i is caused by membrane depolarization and intracellular alkalinization elicited by ZP proteins, incubation of sperm in alkaline medium with a high [K+] (which depolarizes cells) has been shown to mimic the increases in [Ca2+]i elicited by ZP glycoproteins, and such a treatment can bypass the requirement of G proteins (4, 8, 32). The [Ca2+]i responses can be inhibited by several different CaV blockers with varying efficacy, suggesting a role for CaVs in the [Ca2+]i response (8, 32, 92). There have been several high- and low-voltage-gated CaV mRNAs and proteins found in testis and sperm (11, 23). Since mature sperm normally do not synthesize proteins, a reasonable assumption was made that ion channel currents in sperm should also exist in their precursor cells. Indeed, patch-clamp recordings of sperm precursor cell spermatocytes from mice (2, 59) and rats (41) found CaV currents with only low voltage-gated (T-type) channels detectable. These data support a T-type CaV-based Ca2+ signaling model where depolarization of sperm caused by physiological stimuli opens T-type Ca2+ channels and increases [Ca2+]i (23, 31). Other signaling pathways can presumably affect this pathway as well by, for example, removing the inactivation of T-type channels via membrane hyperpolarization that has been observed during capacitation in mammalian sperm (3, 102), during motility activation by osmotic shock in carp sperm (57), and during chemotaxis guided by egg peptide in sea urchin sperm (52, 82).
However, data from several recent experiments challenge the CaV-based model. For example, whole-cell patch-clamp recordings using uncapacitated mouse corpus epididymal sperm found that sperm do not have a detectable CaV current, and although a CaV current can be detected in spermatocytes, it gradually disappears during spermiogenesis and becomes undetectable before spermatocytes become testis sperm (98). Second, in a bath solution with high [K+] (up to ∼60 mM) to depolarize membrane potential up to approximately −20 mV and inactivate T-type CaVs, ZP glycoproteins still readily elicit increases in [Ca2+]i in capacitated sperm (98). Third, the T-type CaV current that is present in wild-type spermatocytes and presumed to also be in sperm is absent in spermatocytes from CaV3.2 knockout mice, yet the mutant mice have a normal response to ZPs in sperm and no defect in fertility despite a distinct cardiovascular phenotype (27, 80). We believe that the simplest interpretation of these data is that functional T-type CaV does not exist in mature sperm, or if it is present at very low levels beyond the detection limit, it is not required for the [Ca2+]i responses or sperm function (98).
Recent data suggest a role for CatSper proteins in the early phase of ZP-induced increases in [Ca2+]i. In CatSper1-deficient sperm, the early [Ca2+]i response (within 2 min) induced by solublized ZP glycoproteins is absent and can be restored with exogenous expression of a GFP-tagged CatSper1 protein (98). The ZP-induced increases in [Ca2+]i originate in the principal piece of sperm where CatSper proteins are localized and propagate to the sperm head within a few seconds. This time interval can be extended if the sperm are cooled from 37 to 16°C. The “slow” [Ca2+]i response (after 2 min) that occurs after the initial fast response is still present in some knockout sperm, suggesting that an unidentified CatSper-independent pathway exists (98).
The mechanisms involved in transducing ZP-induction to the opening of CatSper channels are currently unknown. However, the ZP receptors are hypothesized to be GPCRs that are located in the sperm head. If so, what are the signals that diffuse over >10 μm to the principal piece of sperm to activate CatSper channels? Both cell depolarization and intracellular alkalinization are downstream of ZP binding and represent possible mechanisms. Since the ZP-induced [Ca2+]i changes persist when membrane potential is “clamped” with the K+ ionophore valinomycin and the voltage dependence of CatSper channels is weak (55, 98), membrane depolarization is not likely to be directly responsible for the opening of CatSper channels by the ZP. Intracellular alkalinization, on the other hand, has been shown to potentiate ICatSper and is therefore a potential messenger coupling ZP binding to CatSper opening.
CatSper-Dependent Ca2+ Entry Induced by Other Stimuli
CatSper channels are required for several other well characterized changes in [Ca2+]i induced by stimuli such as BSA and alkaline depolarization (97). BSA is a key component for the successful in vitro capacitation of sperm for several mammalian species (99). Similar to ZP glycoproteins, acute application of BSA induces [Ca2+]i increases that often include both a rapid phase as well as a sustained phase, with the former requiring CatSper channels (97). Extracellular Ca2+ is also needed for in vitro capacitation, which results in biochemical changes such as increased levels of tyrosine phosphorylation, the AR, and changes in motility characterized by the appearance of hyperactivated motility. In CatSper knockout sperm, hyperactivated motility does not develop during capacitation, yet the other biochemical changes such as protein tyrosine phosphorylation and the AR appear to be intact (74, 96, 98). These results suggest that there are different requirements for Ca2+ during sperm capacitation, with the change in motility requiring a CatSper-associated Ca2+ influx and the other cellular changes using CatSper-independent mechanisms (63, 97).
There are clearly CatSper-independent Ca2+-entry pathways. The ATP-induced increases in [Ca2+]i, for example, are intact in CatSper1 knockout sperm (97). In addition, the sustained phase of increases in [Ca2+]i induced by ZP glycoproteins are present in sperm deficient in CatSpers (97, 98). Additional studies are needed to identify the sources responsible for these increases in [Ca2+]i.
Ca2+ Propagation Induced by Ca2+ Entry Through CatSper Channels
CatSper channel-associated proteins appear to be strictly localized to the principal piece of sperm. In consistence with this localization pattern are patch-clamp recordings that have detected ICatSper in the same region yet not in the head or midpiece (55). In addition, all CatSper-dependent [Ca2+]i increases induced by cyclic nucleotides, ZPs, or BSA start in the principal piece. However, these [Ca2+]i changes are not restricted to the initial region but instead propagate through the midpiece and head within a few seconds (96–98) (FIGURE 2). Similarly, tail-to-head Ca2+ propagation has also been observed in human sperm stimulated with odorant (79), as well as in sea urchin sperm induced by egg peptides (95).
A subsequent question would be: What are the cellular functions of the Ca2+ influx-induced Ca2+ propagation? All of the physiological stimuli that lead to CatSper-dependent Ca2+ influxes are associated with the AR and motility change, both of which require an increase in [Ca2+]i. Since hyperactivated motility is not induced in CatSper knockout sperm during capacitation, it is hypothesized that CatSper-initiated increases in [Ca2+]i in the principal piece are for motility. In the midpiece of sperm where mitochondria are abundant, an increase in [Ca2+]i has the potential to increase production of ATP via Ca2+-sensitive dehydrogenases and NADH production. In support of this model, an [ATP] deficiency has been observed in CatSper knockout sperm (96). However, the function of increased [Ca2+]i in the head induced by Ca2+ influx from the tail through CatSper channels remains unclear. Increases in [Ca2+]i in sperm head are usually associated with the AR; the overall AR rates induced by ZP glycoproteins, alkaline depolarization, and cyclic nucleotides in the CatSper knockout sperm, however, do not appear to differ from those of wild-type sperm (96, 98), although subtle differences in kinetics cannot be ruled out. It is likely that the sustained phase of [Ca2+]i increases present in CatSper mutant sperm contribute to the AR.
Ca2+ signaling during fertilization is a fascinating problem and is of fundamental physiological significance. Yet, in contrast with what is known about other Ca2+ signaling processes such as neuronal communication and muscle contraction, little is known at the molecular level concerning Ca2+ signaling in sperm. For example, it is known that sperm detect environmental stimuli such as oviductal fluid during their journey toward the egg and respond with Ca2+ signaling. However, the receptors for the stimuli remain largely unknown. The identities of potential intracellular stores also need to be further studied, along with characterization of the ion channels responsible for the sustained [Ca2+]i increases.
As discussed in the previous sections, CaV and CNG channels do not have a significant role in Ca2+ signaling in mammalian sperm. The currently known sperm Ca2+-permeable ion channels with clearly established functions in fertilization are the CatSper channels. They are required for the Ca2+ entry during capacitation and during the final interaction between sperm and the egg coat before fusion. Their function at the whole animal level is also clear: CatSper genes appear to be specifically evolved for sperm function and male fertility. Based on their unique structure, restricted expression patterns, and physiological functions, development of drugs specific for CatSpers, and/or the auxiliary subunits, have the potential to treat motility-related infertility or serve as contraceptives (17).
Many questions remain to be answered regarding the signaling pathways related to CatSper proteins (FIGURE 3). First, of the hundreds of ion channels cloned, the CatSper channels remain one of the few that have not been reconstituted in heterologous systems, despite efforts that have spanned almost a decade. One possible reason for this is that there are more subunits involved than the six proteins that have already been characterized (i.e., CatSper proteins 1–4, β, and γ). Perhaps more likely, the “right” cellular environment for the functional reconstitution of CatSper channels still needs to be identified, which may differ significantly from the membrane composition of cells currently used in heterologous expression models. Second, what are other physiological stimuli of CatSper channels in addition to ZP glycoproteins, BSA, and cyclic nucleotides? Follicular fluids and progesterone secreted from the cumulus around the egg have also been shown to elicit [Ca2+]i increases in sperm, although it has not been determined whether these stimuli are CatSper-dependent. Third, the activation mechanisms for CatSper channels are poorly understood. For example, CatSper conductance increases with intracellular alkalinization, yet if this is the result of direct channel activation, what is the pH-sensing mechanism in the CatSper channel? One possibility is that the histidine-rich domain of CatSper1 senses changes in pH. If changes in pH are indeed a physiological activation mechanism for CatSper channels, how do stimuli such as ZP lead to intracellular alkalinization? The principal piece-localized sperm Na+/H+ exchanger (sNHE, SLC9A10) required for male fertility has been proposed to be a possible mediator (87). In addition, since BSA induces increases in [Ca2+]i when both voltage and pHi are “clamped,” other activation mechanisms besides alkalinization are probably present. Finally, what is the mechanism for the tail-to-head propagation of Ca2+ mediated by the CatSper-dependent Ca2+ entry? Is it based on simple Ca2+ diffusion, or does it require Ca2+ release from intracellular stores and/or additional Ca2+ entry in the midpiece and head regions? What is the functional significance of such a directional Ca2+ signaling? Does a tail-to-head Ca2+ propagation rather than a head-to-tail propagation generate a more powerful mechanical force? Only with further studies can these questions be answered and our understanding of mammalian fertilization improved.
We thank Drs. David Clapham, George Gerton, Yuriy Kirichok, Betsy Navarro, Bayard Storey, and members of the Ren Laboratory for discussion. We have used reviews when possible, and we apologize to those whose original work was not cited.
The fertilization-related research in our laboratory has been supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
- ©2010 Int. Union Physiol. Sci./Am. Physiol. Soc.