Physiological Reviews

Calcium Channels in the Development, Maturation, and Function of Spermatozoa

Alberto Darszon, Takuya Nishigaki, Carmen Beltran, Claudia L. Treviño


A proper dialogue between spermatozoa and the egg is essential for conception of a new individual in sexually reproducing animals. Ca2+ is crucial in orchestrating this unique event leading to a new life. No wonder that nature has devised different Ca2+-permeable channels and located them at distinct sites in spermatozoa so that they can help fertilize the egg. New tools to study sperm ionic currents, and image intracellular Ca2+ with better spatial and temporal resolution even in swimming spermatozoa, are revealing how sperm ion channels participate in fertilization. This review critically examines the involvement of Ca2+ channels in multiple signaling processes needed for spermatozoa to mature, travel towards the egg, and fertilize it. Remarkably, these tiny specialized cells can express exclusive channels like CatSper for Ca2+ and SLO3 for K+, which are attractive targets for contraception and for the discovery of novel signaling complexes. Learning more about fertilization is a matter of capital importance; societies face growing pressure to counteract rising male infertility rates, provide safe male gamete-based contraceptives, and preserve biodiversity through improved captive breeding and assisted conception initiatives.


Fertilization is a fundamental and convoluted process, necessary to unite two haploid gametes, the male spermatozoon and the female oocyte/egg, to produce a unique individual. Mature and competent gametes are needed to achieve this impeccably choreographed process. We are still far from fully understanding the gamete intracellular molecular dialogue necessary for fertilization. Spermatozoa are specialized cells able to reach, recognize, and fuse to the egg or oocyte despite being essentially incapable of gene transcription or protein synthesis. Improving our understanding of fertilization is essential as it is at the heart of creating life. Furthermore, it is urgently needed to tackle increasing male infertility in developed societies, to provide safer male gamete-based contraceptives, to improve animal breeding, and to preserve biodiversity (37, 521).

Ca2+ is critical in different cellular signaling processes acting at diverse spatial sites, and its role as second messenger has been long recognized (51, 79). More recently, the molecular cloning of the calcium sensing receptor (CaSR) established its role as a first messenger (61). To function in these roles, Ca2+ concentration must be finely regulated both intra- and extracellularly. The intracellular Ca2+ concentration ([Ca2+]i) is maintained within strict limits by a variety of cellular mechanisms that buffer, sequester, extrude, and accumulate Ca2+, with concentration changes often occurring in highly localized areas within the cell (79). Indeed, Ca2+ plays a pivotal role in fertilization participating in the main functions of both invertebrate and vertebrate spermatozoa such as maturation, motility, and the acrosome reaction (AR), an exocytotic process required for fertilization in many species (571). The fundamental importance of ion channels, particularly those for Ca2+, for sperm physiology and fertilization has been consolidated in the last 10 years due to significant advances in the tools available to study them (135, 292, 519). This is emphasized by the fact that certain Ca2+ channel blockers inhibit these functions (139, 178, 381, 419).

Cells use a significant portion of their energetic resources to build and maintain ion concentration gradients across their membranes using ion pumps and transporters. These ionic gradients are a source of information about the state of the cell and its surroundings. Ion channels, which upon activation can alter the electric potential of the cell and the concentrations of internal second messengers within a wide time-range, depending on the modes of channel regulation, translate this information into cell behavior to contend with the changes within and outside of the cell. Therefore, it is not surprising that genomes from the fruit fly to the human encode hundreds of different channels.

Spermatozoa are small, morphologically complex cells (Fig. 1) and differentiated (Fig. 2) to accomplish a concrete but fundamental goal: to deliver their genetic material to the egg, activate its development, and generate a new individual. To achieve this goal, a matter of life and death, they must contend with myriad environmental changes while they mature and then steer in search of the egg (Fig. 3). In this regard, it would seem that the multiple signaling processes needed for spermatozoa to mature, properly swim, and reach the egg at the right time, in a proper condition and fuse with it, would require a rich set of ion channels and transporters. This review will attempt to describe and discuss the Ca2+ channels that participate in these key sperm processes. It is organized considering the development and travel of spermatozoa towards the egg. In general, the characteristics of each channel family appear when first involved in a process or function. Thereafter, each new section provides further information regarding properties of specific channels participating in a particular sperm function.

Figure 1.

Diagram of sea urchin, mouse, and human spermatozoa. The major sperm components for each species are indicated. A: sea urchin sperm; entire cell body (left), enlarged head cross-section (top right) showing the position of the acrosome, nucleus, centriole, and nuclear envelope remnants (above and below the nucleus). Note a single doughnutlike mitochondrion that resides below the nucleus. Enlarged cross-section of flagellum (bottom right) shows the classic structure of 9+2 doublets microtubules, in which the nine doublets surround a central pair. Outer and inner dynein arms as well as the radial spokes are illustrated. The flagellum of this species does not have additional cytoskeleton around the axoneme. B: entire cell bodies of mouse and human spermatozoa. The shape of the head and the acrosome are quite different in these species. Mouse spermatozoa have a hook and crescent moon-shaped head and acrosome, respectively. In contrast, human spermatozoa have a rounded and cap-shaped head and acrosome, respectively. Mammalian flagella can be divided into mid, principal, and end pieces. The midpiece is separated from the principal piece by the annulus. Equatorial and postacrosomal segments are also indicated. In mouse, a cytoplasmic droplet is usually observed in testicular and epididymal spermatozoa, but not in ejaculated mature spermatozoa, while it is frequently observed in human mature spermatozoa. C: enlarged human sperm (head and midpiece region) cross-sections viewed from two different angles (90° rotation). The nucleus occupies most of the head space, and it is overlaid by the acrosome; there is very little cytosol in these cells. The centriole and the redundant nuclear envelope are located at the base of the head. The axoneme is surrounded by outer dense fibers in the midpiece and the principal piece. The outer dense fibers are surrounded by mitochondria in the midpiece, but by the fibrous sheath in the principal piece. D: enlarged cross-section of mammalian sperm flagellum. The axoneme runs along all the flagella; it is surrounded by mitochondria and outer dense fibers in the midpiece, by the outer dense fibers, and the fibrous sheath in the principal piece (2 out of 9 outer dense fibers are fused with the fibrous sheath). The end piece contains only the axoneme.

Figure 2.

Diagram of the mammalian male reproductive organs and spermatogenesis. Left: spermatogenesis diagram. Spermatogenesis advances from the base to the lumen of the seminiferous tubule with structurally and functionally essential nurse cells, called Sertoli cells. Spermatogonia (germinal stem cells) reside in the basal compartment and proliferate. Primary spermatocytes (leptotene) differentiate from spermatogonia and transverse the blood-testis barrier formed by tight junction of adjacent Sertoli cells. Secondary spermatocytes (pachytene and diplotene) are formed by the first meiosis and converted into round spermatids by the second meiosis. Elongated spermatids are cells in the process of spermiation undergoing the remarkable morphological transformation of the nucleus and cell body. In the final process of spermiation, most of the cytoplasm will be removed (phagocytized by Sertoli cells) to form fully developed spermatozoa (the residual cytoplasm is known as cytoplasmic droplet). Middle: a seminiferous tubule and its cross-section. Newly formed spermatozoa are transported towards the epididymis through the lumen of the tubule (L). Right: testis and epididymis. The testis is filled with tightly packed seminiferous tubules; one of them is represented as a gray line inside the testis. The epididymis consists of a single highly coiled tubules. This organ can be roughly divided into three portions: caput, corpus and cauda; sperm transit along these tubules is required for sperm maturation.

Figure 3.

Diagram of the mammalian female reproductive system. Left: the architecture and major components of the female reproductive tract are shown. The vagina is the site of ejaculation and the start point of the sperm path towards the oocyte. Swimming and muscle contractions allow spermatozoa to pass through the cervix and the uterus to reach the isthmus. The cervix separates the vagina from the uterus, which is the site of fetus development. The isthmus or sperm reservoir is located in the Fallopian tubes. In the absence of oocytes, spermatozoa remain attached to the epithelium that lines the tubal lumen for several days, depending on the species. The ampulla is the second segment of the Fallopian tube and the site of fertilization. During ovulation, the oocyte migrates to the ampulla, and spermatozoa detach from the oviductal epithelium (sperm can attach and detach several times) to reach the egg. Middle: cross-section of the isthmus and ampulla regions showing the vast invaginations of these segments. Right: diagram of the oocyte and the surrounding zona pellucida and cumulus cells.

Ion channels in general constitute a very minor percentage of the total membrane protein. In this regard, demonstrating their functional presence and location in a specific cell is an arduous and sometimes frustrating enterprise. Because spermatozoa are basically considered transcriptionally silent, their ion channels are synthesized by spermatogenic cells, their progenitors, during spermatogenesis. However, the presence of transcripts of a specific ion channel, and even of its protein and currents in spermatogenic cells, does not necessarily mean that the protein will be functionally expressed in mature spermatozoa. The physiological relevance of the transcripts detected in mature spermatozoa is still unclear and cannot be taken as proof that the proteins they code for are present in spermatozoa. Therefore, demonstrating that an ion channel is functionally present in mature spermatozoa requires proving either by controlled immunological or proteomic strategies that the protein is there, and showing that it is functional either electrophysiologically or using ion-sensitive dyes or proteins. Pharmacology and ultimately elimination of the specific ion channel from spermatozoa, in the species where it can be done (i.e., null mice), will yield clear evidence of the role of the channel in sperm physiology.

The authors are aware of their limitations and preferences, which most likely have led to some biased proposals and regrettably to exclude some important contributions. Several excellent reviews are provided, hoping they will complement and balance the subject matter of this review (47, 178, 418, 433, 548).


Spermatogenesis is the transformation of diploid spermatogonial stem cells into haploid spermatozoa that takes place inside the seminiferous tubules (Fig. 2). It is a continuous, complex, and highly regulated process. Functional units along the tubule's epithelial compartments (basal and adluminal) repeat a cycle that can be divided in four main phases: mitosis, meiosis, spermiogenesis, and spermiation. In the basal compartment, primary spermatogonia proliferate and after a species-specific fixed number of mitotic divisions (253) differentiate into leptotene spermatocytes that are the only cell type that transverses the blood-testis barrier into the adluminal compartment. Once in the adluminal compartment, cells differentiate to pachytene and then diplotene spermatocytes that undergo two rounds of meiotic division to produce haploid spermatids. The process continues with spermiogenesis in which spermatids experience a strong morphological transformation including, among others, the formation of the acrosome, chromatin condensation, and the elimination of the excess cytoplasm (residual body) with the simultaneous formation of the cytoplasmic droplet (see below), to produce elongated spermatids that migrate towards the edge of the lumen (118). During spermiation, fully developed spermatozoa are released into the lumen. Two interesting structures are formed in the final stages of sperm differentiation. During nuclear volume reduction, the excess material is packaged into the redundant nuclear envelope (RNE), whose name derives from the lack of an established function for this organelle (Fig. 1). This structure has been detected in spermatozoa from several mammals and in sea urchin (210), and it is now proposed that it functions as a Ca2+ store. In support of this proposal, inositol trisphosphate (IP3) receptors (IP3Rs) have been detected in the RNE (261). The second structure found in mammalian sperm is the cytoplasmic droplet (see Fig. 1) that is produced during the formation of the residual body which concentrates excess cytoplasm, mitochondria, lipid droplets, the protein synthesis machinery, residual structures, etc. The residual body is detached from sperm during spermiogenesis, and it is phagocytosed by Sertoli cells, but the cytoplasmic droplet remains attached to the flagella, and depending on the species, this droplet is lost in the epididymis or after ejaculation (247). There are several proposed functions for the cytoplasmic droplet, such as the modification of the sperm plasma membrane (394) or volume regulation (173).

Spermatogenic cells do not accomplish this process independently; instead, they are nursed by Sertoli cells (Fig. 2) inside the tubules and hormonal stimulation is provided by Leydig cells from the interstitial compartment where peritubular myoid cells are also present (212). The germ cell cycle and movement along the tubule is a process under tight control including signaling that involves several families of kinases and phosphatases (246). In spite of the importance of spermatogenesis, the regulation as well as biochemical and molecular mechanisms underlying this process have remained largely unexplored (118). In particular, very little is known about the role of Ca2+ during the cellular events required to produce mature spermatozoa. Considering that Ca2+ is one of the most common messengers inside the cell and has such a predominant role in almost every signaling cascade, it is likely that it would be involved in this process. Indeed, spermatogenic cells regulate their Ca2+ homeostasis with a fine balance between Ca2+ uptake and release by intracellular stores (250). In fact, postmeiotic cells (round spermatids) and meiotic cells (pachytene spermatocytes) differentially regulate their [Ca2+]i upon external supply of oxidative glycolytic substrates such as glucose and lactate (434). For example, exposure of mouse spermatogenic cells to maitotoxin (a potent marine toxin) produced a significantly greater elevation in [Ca2+]i in cells incubated in glucose-containing media than in cells incubated in lactose-containing media (435). Glucose and lactate are normally secreted by Sertoli cells into the adluminal compartment. It was proposed that these differences in Ca2+ signaling, mediated primarily by internal Ca2+ stores, could be a differentiation-related process (434). Intracellular Ca2+ channels (see sect. IIID), namely, IP3Rs (378) and ryanodine receptors (RyRs), have been detected in spermatogenic cells, and for the latter, a role during differentiation has been proposed (119).

To ensure a high-quality supply of gametes, spermatogenic cells undergo apoptosis, a form of programmed cell death, to discard excess and unfit cells (518); however, the signaling mechanism leading to apoptosis is not completely understood. Apoptosis can also be induced by environmental factors such as temperature, toxins, exposure to chemicals, etc. Herrera et al. (250) reported that high temperature conditions can induce programmed cell death in rat spermatogenic cells mediated through a Ca2+-initiated mechanism. Mishra et al. (363) treated rat spermatogenic cells in vitro with 2,5-hexanedione (a common industrial solvent) to induce apoptosis. They characterized an apoptotic mechanism regulated by changes in the [Ca2+]i that translate into a differential expression of members of the Bcl protein family. These proteins can be either pro- or antiapoptotic factors, and their expression ratio determines the survival or death of the cells (363). In accordance with the observed pharmacology, Mishra et al. (363) proposed that the channels involved belong to the CaV3 family (see sect. IIIA).

As mentioned before, very little is known about Ca2+ regulation and its contribution during spermatogenesis. Because mature spermatozoa are basically transcriptionally silent and spermatozoa were not easily patch clamped (162) until very recently (292), spermatogenic cells were the preferred model to study Ca2+ channels using molecular biology and electrophysiological approaches (11, 14, 234, 324, 451). The main interest was as a proxy to understand their role in mature sperm, but their participation during differentiation has not been investigated. In particular, currents from voltage-dependent Ca2+ (CaV) channels were detected and characterized in mouse spermatogenic cells (134, 234, 479, 514), and their function was extrapolated to mature sperm.

Although spermatozoa are in principle transcriptionally silent, as mentioned, certain mRNAs have been detected in these cells. It has been proposed that these RNAs are long lived and can be delivered to the egg for later expression (174, 300). A few papers have reported that some sperm mRNAs can be translated to protein via the mitochondria machinery (231, 232). In any case, mRNAs from either spermatogenic cells or mature sperm coding for several CaV α-subunits and auxiliary subunits have been isolated, and the presence of the corresponding proteins has been confirmed with specific antibodies (see sect. IIIA and Table 1).

View this table:
Table 1.

Expression of Cav channels in sperm cells


A. Voltage-Gated Ca2+ Channels

Voltage-gated Ca2+ (CaV) channels are responsible for [Ca2+]i increases induced by membrane potential (Em) changes that regulate processes such as contraction, secretion, neurotransmission, and gene expression in diverse cell types. These channels have been catalogued into two major functional classes: high voltage-activated (HVA) and low voltage-activated (LVA) channels (89). HVA channels require strong depolarizations to open, after which they inactivate slowly. Considering the biophysical and pharmacological characteristics of their macroscopic currents, they have been classified into L, N, P/Q, and R types. On the other hand, LVA channels need weaker depolarizations to open, and they inactivate faster and at more negative potentials than HVA channels. LVA currents were initially named T type (181) because they open transiently.

The ion-conducting pore of CaV channels is formed by the α1 subunit of ∼200 kDa, which is encoded by a family of 10 genes grouped into three subfamilies: CaV1 with four members CaV1.1-CaV1.4 all conducting L-type currents, CaV2 with three members conducting P/Q- (CaV2.1), N- (CaV2.2), and R-type (CaV2.3) currents, and CaV3 with three members (CaV3.1-CaV3.3) all conducting T-type currents. In this review, we use CaV1 and/or CaV2 for HVA and CaV3 for LVA channels.

The topology of the α1 subunit consists of four repeated domains (I–IV) each containing six transmembrane (TM) segments (S1–S6) surrounding a central pore (87, 91). Auxiliary subunits for CaV1 and CaV2 channels form another protein family with four members (CaVα2, CaVβ, CaVγ, and CaVδ) (88). CaV channels in different tissues may display unique properties due to alternative splicing, posttranslational modifications (10), and diverse regulation mechanisms. In what follows, the principal regulation modes of CaV channels will be summarized, paying attention to those more relevant to sperm physiology. Table 2 sums up the pharmacology of CaV channels.

View this table:
Table 2.

Calcium channel blockers inhibit acrosomal reaction (AR)

Several protein kinases such as protein kinase A (PKA), protein kinase C (PKC), and Ca2+/calmodulin (CaM)-dependent protein kinase II (CaMKII) phosphorylate and regulate CaV channels. In many instances, these phosphorylation events stimulate CaV channels by diverse mechanisms depending on the particular isoform (38, 87, 89, 128, 551).

CaV channels are also regulated by G proteins activated through different pathways including those activated by neurotransmitters and hormones (140). This regulation can be direct, via physical interactions between the G protein and channel subunits or indirect via second messengers and/or by protein kinases (140, 152). Recently, Perez-Reyes (405) published an excellent review on the properties of CaV3 channels. This work describes three different mechanisms involving activation of G protein-coupled receptors that lead to particular inhibition of CaV3.2 channels: 1) a direct inhibition by Gβ2γ2 subunits; 2) a pathway independent of Gβ2γ2 but involving phospholipase C (PLC), diacylglycerol (DAG), and PKC; and 3) a mechanism involving Gαs that does not involve any kinases.

Ca2+ itself regulates CaV channel inactivation by associating with CaM. The mechanisms involved in this regulation are best understood for CaV1 and CaV2 channels, although CaV3 channel function may also be influenced by CaM (336). Other related Ca2+ sensor proteins resembling CaM can displace it from shared binding sites in the α1 subunits of CaV channels, but are different enough to confer distinct forms of regulation (89, 149, 310).

Only recently has the modulation of CaV3 channels by endogenous ligands and second messenger pathways been explored (113). Molecules such as anandamide, arachidonic acid, and polyunsaturated fatty acids (micromolar range) inhibit CaV3 channels. For example, lipoaminoacids such as N-arachidonoylglycine (NAGly), a molecule closely related to endocannabinoids that does not activate cannabinoid receptors or TRPV1 channels, has been reported to block CaV3.2 channels (34).

Interestingly, Zn2+, an important divalent cation present at millimolar concentrations in seminal plasma (in mammals, from 0.2 to 2.6 mM depending on the species; Refs. 331, 349), preferentially inhibits CaV3.2 (IC50 0.8 μM), while CaV3.1 and Cav3.3 are 100- to 200-fold less sensitive. Apparently, some of the Zn2+ inhibitory effects are mediated through binding to an extracellular histidine residue in the S3–S4 linker of domain I (265).

Truly selective CaV3 blockers are urgently needed to help characterize these channels, both in heterologous expression and in native cells. Approaches aimed at developing specific blockers such as the one undertaken by Uebele et al. (523) are welcomed. They described a specific CaV3 antagonist: 2-(4-cyclopropylphenyl)-N-((1R)-1-{5-[(2,2,2-trifluoroethyl)oxo]pyridin-2yl}ethyl) acetamide (TTA-A2) that is a selective and state-dependent CaV3 channel antagonist, with a potency of ∼10 nM in the depolarized (inactivated) state and 400 nM in the hyperpolarized (closed) state. Sensitivity of L-, N-, P/Q-, and R-type currents to TTA-A2 was above 6 μM. The authors used this compound therapeutically to treat mice with obesity and sleep disorders that were correlated with CaV3.1 activity.

The regulatory properties of auxiliary subunits have been described mainly for CaV1 and CaV2 (HVA) channels. Although there is conflicting evidence showing that CaV1 and CaV2 auxiliary subunits may or may not modulate CaV3 channels, recent reports show that among the eight members of the CaV auxiliary channel γ subunits, γ6 inhibits CaV3.1 currents (115, 326) and that the amino acid sequence GxxxA (where x is any amino acid) in its first transmembrane domain plays an important role in this modulation. Chen and Best (115) proposed that the inhibitory action of γ6 on CaV3 channels could explain the observation that despite the strong expression of CaV3.1 and CaV3.2 subunit mRNA in adult ventricular myocytes, no CaV3 currents are detectable in these cells. Postinfarcted ventricular myocytes often display reoccurrence of CaV3 currents with higher expression of mRNA for CaV3.1 and CaV3.2 channels (115).

1. CaV channels in spermatogenic cells and spermatozoa

To establish the presence and activity of CaV channels in spermatozoa, researchers have used diverse experimental approaches to detect mRNA, protein, and currents (Table 1).


Mouse spermatogenic cells express several transcripts coding for CaV channels. In the case of CaV1 and CaV2 channels, RT-PCR experiments revealed the presence of CaV1.2, CaV2.1, and CaV2.3 (164, 324), while CaV2.2, CaV2.3, and CaV1.2 (219, 400) transcripts were reported in mature human spermatozoa. Additionally, specific spliced transcripts for CaV1.2 are expressed in human sperm (219, 220). All three transcripts for CaV3 channels are present in mouse and human spermatogenic cells (164, 276, 400, 463, 474). With respect to auxiliary subunits, transcripts for all four known genes encoding the β subunits were reported in mouse spermatogenic cells (463) and α2δ, β1, β2, and β4 in mature human spermatozoa (282).


Immunocytochemistry and/or Western blot experiments using specific antibodies for the different CaV subunits confirmed the presence of CaV1.2 and CaV2.1 proteins in spermatogenic cells (463), and all the members of CaV1 and CaV2 subfamilies, with the exception of CaV1.1, CaV1.3, and CaV1.4, in mature sperm (514, 554, 556). All three CaV3 proteins were found both in mouse and human mature sperm (164, 276, 400, 463, 475, 514, 556). Importantly, Western blot and immunohistochemistry experiments using CaV3.1 or CaV3.2 null mice as controls corroborated that CaV3.2 channels are present in mature spermatozoa (160). Despite the presence of several CaV proteins, the function of each of these channels has not been fully corroborated (see below).


Patch-clamp studies, the classical approach to examine cell currents, have not revealed CaV1 or CaV2 (HVA) currents in spermatogenic cells, including those from CaV3.1 and 3.2 null mice (134, 160). However, channel antagonists to CaV1 or CaV2 channels interfere with certain sperm Ca2+ responses (554) and the AR (46, 221), suggesting that these channels may be present in an electrophysiologically inactive state and activated during sperm differentiation and maturation.

In contrast, CaV3 currents have been well documented in spermatogenic cells by patch-clamp experiments (14, 234, 324, 449). These currents display the properties of somatic cell CaV3 currents, such as low voltage thresholds for activation and inactivation, and near-equivalent Ba2+ selectivity relative to Ca2+. The pharmacology of the currents is also consistent with CaV3 channels as they are inhibited by the anticipated concentrations of amiloride, pimozide, mibefradil, kurtoxin, and low Ni2+ concentrations (14, 396) (Table 2). Drugs normally considered as CaV1 and CaV2 antagonists such as nifedipine and related 1,4-dihydropyridines (DHPs), at lower (μM) concentrations can also block spermatogenic T-type currents (CaV3) (14, 449) (Table 2). The pharmacological profile of the spermatogenic cell currents (137), as well as the current recordings from CaV3.1 and Cav3.2 null spermatogenic cells (160), are consistent with CaV3.2 being the principal isoform functionally present in these cells. Additionally, CaV3 channels in spermatogenic cells appear to be regulated by protein kinases (13), albumin (163), and CaM (336).

As mentioned before, patch-clamping mature sperm was extremely difficult (162) until Kirichok used the cytoplasmic droplet as the contact site for the pipette, which allows giga-seal formation and electrical continuity with the cell (292) (reviewed in Refs. 329, 433). Since then, several ion channels have been detected in testicular and epididymal mouse spermatozoa (137, 346, 381, 431, 450, 566). Intriguingly, CaV3 currents such as those recorded in spermatogenic cells were detected in testicular but not in epididymal sperm (137, 566). CaV1 or CaV2 currents were not recorded in either cell type, under the experimental conditions tested (137, 160). On the other hand, currents from CatSper channels (see sect. IVD) have been consistently detected in mouse epididymal spermatozoa and not in the cells from CatSper null mice. These findings challenge previous functional, molecular, and pharmacological data indicating the involvement of CaVs and other channels in diverse sperm functions (see discussion in sect. V, C and D) and imply that the CatSper channel is one of the major Ca2+-conducting channels in spermatozoa (433).

B. Cyclic Nucleotide-Gated Channels

Cyclic nucleotide-gated (CNG) channels have a well-established role in sensory transduction processes (vision and olfaction); their function in other tissues such as heart, kidney, brain, or spermatozoa is not so clear (285, 353). CNG channels function as heterotetramers resulting from the combination of A (CNGA1–4) and B (CNGB1 and 3) subunits with a stoichiometry that varies in different tissues. Heterologous expression of CNG channels has shown that homomeric A (with the exception of A4; Ref. 109) but not B subunits can form functional channels (107). Each subunit contains six TM segments and a single cyclic nucleotide-binding site near the COOH terminus (74, 285). CNG channels open upon cAMP or cGMP binding, have low ion selectivity, are weakly voltage sensitive, and unlike other channels, they do not inactivate or desensitize (74). Depending on the particular subunit combination, they exhibit differential selectivity to cAMP and cGMP (412). As CNG channels are nonselective under physiological conditions, they usually carry inward Na+ and Ca2+ currents (353). The first sperm ion channel cloned from mouse testis was a CNG channel (557); thereafter, others have been cloned (285). The A3 and B1 subunits were localized within the flagellum of mature mouse spermatozoa (285). Addition of permeable cGMP to mouse spermatozoa provoked [Ca2+]i increases dependent on [Ca2+]e, while cAMP was less effective (295, 558). Curiously, spermatozoa from CatSper null mice do not present this response (432), and the A3 null mice are fertile (285). These observations throw into question the relevance of these channels in sperm physiology. In spite of this, their role in sea urchin sperm chemotaxis is well established (see sect. VI and Refs. 135, 284).

C. Transient Receptor Potential Channels

Initial evidence for the existence of transiet receptor potential (TRP) channels came from a Drosophila melanogaster mutant, reported by Cosens and Manning back in 1969 (103). These authors reported a particular phenotype present in the trp mutant fly, which exhibited defects in electroretinogram recordings. Exposure to constant light produced a sustained response in the wild type, whereas the trp-deficient fly displayed a transient response, producing the name “transient receptor potential” or TRP (368). After 20 years, Montell and Rubin (370) reported the sequence of the trp gene, which was later confirmed to encode a novel class of ion channels with homology to voltage-gated channels (361, 411). Since then, this family has been extensively studied, and it is now well established that they are involved in diverse cellular functions.

Members of the TRP channel family are considered to be polymodal cellular sensors playing important roles in vision, taste, hearing, touch, olfaction, thermal perception, and nociception (307). They are quite versatile and diverse in their features, selectivity, gating, and regulatory mechanisms and depending on the species, an organism can express as many as 30 different genes. Based on sequence similarity they can be divided in seven groups: 1) TRPC (canonical), 2) TRPV (vanilloid), 3) TRPM (melastatin), 4) TRPA (ankyrin), 5) TRPML (mucolipin), 6) TRPP (polycistin), and 7) TRPN (only one member found in zebrafish, worms, and flies but not in mammals). TRP subunits contain six transmembrane (TM) segments, with both the COOH and NH2 termini facing the cytosol. Different biochemical and optical methods suggest that active channels assemble as homo- or heterotetramers (307). Heteromultimers have been described for all TRP subfamilies except TRPA and TRPN (483, 531). The modulation and structural properties of TRP channels are quite diverse; for example, three members of the TRPM subfamily possess an enzymatic activity linked to the COOH terminus; one channel exhibits an ADP-ribose pyrophosphatase, and two more contain an atypical protein kinase (369), while others can act as mechanical (157) or humidity sensors (333). The localization of TRP channels is also varied; some reside in the plasma membrane (PM) and some in intracellular organelles, or both (155). In spite of these divergences, there are certain features shared by several members of the TRP superfamily, and in general, they are believed to integrate signals, as the response to one agonist is modified by another. For example, the TRPV1 threshold for heat sensitivity (43°C) decreases if pHi drops to 6. Its sensitivity to endocannabinoids, anandamide, piperine (black pepper), and allicin (garlic) is modified by the presence of ethanol, nicotine, or changes in phosphatidylinositol 4,5-bisphosphate (PIP2) levels (531).

Most TRPs are relatively nonselective Ca2+ channels with a PCa/PNa <10, with some exceptions such as TRPM4 and TRPM5 that are monovalent-selective and do not permeate Ca2+. In contrast, TRPV5 and TRPV6 are highly Ca2+ selective with PCa/PNa >100 (92). The TRP domain located at the proximal COOH-terminal end (25 amino acids) contains the TRP box (EWKFAR) once considered as a signature of TRP channels. This domain is only present in all TRPC members, whereas TRPM, TRPN, and TRPV members display a smaller consensus sequence. Only W is present in all TRP members in this particular domain (reviewed in Ref. 307). The function of this domain is not quite clear; it may have a role in PIP2 binding (a modulator of TRP channels, see below) (439), or participate in channel assembly (205) as well as in tetramerization and gating (206).

PIP2 can modulate activation or inactivation of several ion channels such as members of the TRPC, TRPM, and TRPV subfamilies, and it probably modulates other TRPs too (494). PIP2 is present primarily in the cytoplasmic leaflet of the plasma membrane but not in trafficking vesicles or organelles, and therefore, its abscence would silence the channels as they are being synthesized or in transit (168). Alternatively, the activity of the channels could be controlled indirectly after PLC transient activation and PIP2 depletion from the plasma membrane (494). Furthermore, since TRP channels are sensitive to several stimuli (pH, temperature, phosphorylation, mechanical and osmotic changes, etc.), PIP2 can serve as a means of desensitization or sensitization to such factors (494).

D. Store-Operated Channels

Frequently, [Ca2+]i rises result from a combination of Ca2+ release from intracellular stores followed by Ca2+ entry via plasma membrane Ca2+ channels, a concept termed capacitative Ca2+ entry or store-operated channel (SOC) activity (422). Several research groups tested this notion primarily using compounds that empty intracellular stores (thapsigargin and cyclopiazonic acid, etc.) and Ca2+ fluorescence imaging techniques (421, 494). The first Ca2+ current regulated by store depletion was measured by Hoth and Penner in mast cells (264) and was named ICRAC (Ca2+ release-activated Ca2+ current). It was a very small Ca2+-selective current, not voltage activated and inward rectifying. ICRAC was found and studied in several cell types (263), but its molecular identity and sensing mechanism remained unknown until recently. TRP channels were for a long time the best candidates to produce SOC activity (and probably ICRAC).

The recent discovery of two protein families, namely, stromal interaction molecules (STIM) sensors and ORAI channels, introduced a new concept for the molecular composition of SOC activity and ICRAC currents. This discovery also helped to explain the contradictory results obtained with the heterologous expression of some TRP channels that did not have CRAC characteristics yet displayed SOC activity (reviewed in Refs. 151, 421). However, some TRPC subunits are still recognized as part of the SOC machinery, a notion supported by a wealth of experimental evidence generated before the discovery of STIM and ORAI proteins (reviewed in Ref. 54). SOC activity caused by TRPC channels seems to be cell or condition specific (314). On the other hand, there is also evidence for the interaction of TRPCs with STIM and ORAI, leading to the proposal that SOC activity requires a multicomponent complex (see below). Alternatively, it has also been considered that ORAI proteins are not channels, but auxiliary subunits that confer particular properties to the SOC machinery, such as Ca2+ selectivity (322). Despite the discovery of STIM and ORAI, the exact molecular composition of the SOC machinery is still a matter of debate and an interesting area of research.

1. ORAI and STIM

The STIM and ORAI protein families with two and three genes, respectively, are essential components of SOC activity. STIM1 and 2 are membrane proteins located primarily in the ER where they act as Ca2+ sensors. STIM1 is a multidomain protein with a single TM segment, which senses the Ca2+ store content using a classical two EF-hand motif located luminally at the NH2 terminus. It then transmits the information to ORAI channels via domains in the COOH terminus and thus regulates SOC channel opening (314). At resting conditions, STIM1 exists as a dimer with Ca2+ bound to the EF-hand (Kd of ∼0.2–0.6 mM) (481). Depletion of the ER causes Ca2+ dissociation from the EF-hand and a rapid spatial reorganization of STIM1, reported as oligomerization and/or translocation to the PM or to ER-PM junctions (151, 281). This movement allows STIM1 to communicate with ORAI channels located at the PM. Despite its very similar architecture to STIM1, STIM2 apparently does not translocate or oligomerize, and its role during SOC activity is still controversial (281). ORAI channels were named after the three heaven's gate keepers Eunomia (Order or Harmony), Dike (Justice) and Eirene (Peace) in Greek mythology (172). They are small proteins (28–33 kDa) with four TM segments that tetramerize to form the ion channel pore. ORAI proteins are highly glycosylated with cytoplasmic NH2 and COOH termini (436). Coexpression of STIM1 with ORAI1, -2, and -3 reconstitutes CRAC currents with different properties in terms of Ca2+ and Na+ selectivity, current amplitude, and pharmacological sensitivity to drugs such as 2-APB (184). ORAI3 is activated by 2-APB independently of STIM1 or store depletion. Low concentration of 2-APB (5–10 μM) stimulates STIM1-ORAI1 currents, but higher concentrations (50 μM) completely inhibit ORAI1 and partially inhibit ORAI2 currents (184). Based on these properties, Goto et al. (222) developed 2-APB derivatives to differentially activate and inhibit SOCs. The pharmacology of SOC activity began before the discovery of ORAI channels; therefore, data interpretation and usage of these compounds can be complicated; there are excellent reviews concerning this issue (436, 500).

As mentioned before, the discovery of ORAI channels and the establishment of the identity of the ICRAC current did not eliminate TRP channels as participants in SOC activity. In fact, there is evidence that TRPCs can interact with STIM1, and it has been suggested that native SOC channels may be formed by a combination of STIM, ORAI, and TRPC proteins (314). Moreover, Vaca (524) suggested that SOC activity requires the formation of a macromolecular complex named SOCIC (store-operated calcium influx complex) that includes additional players such as the sarco/endoplasmic reticulum Ca2+-ATPase (SERCA) and EB1 (end tracking protein). The assembly (or disassembly) of the complex regulates SOC activity and does not occur randomly in the membrane, but in specialized domains known as lipid rafts (524). This is a new and fascinating field that is just beginning to unfold, for example, after store depletion STIM1 recruits adenylate cyclase, thereby altering cAMP levels (316). At least two different functions for STIM1 at the plasma membrane have been described: 1) a cell-cell interaction mediator via its NH2 terminal that protrudes extracellularly and also 2) as a modulator of arachidonic acid-regulated channels, which are different from SOCs and independent of store depletion (357).

E. Intracellular Ca2+ Channels

IP3Rs and RyRs are the major intracellular Ca2+ channels located primarily in ER-type intracellular stores. In addition, two-pore channels (TPCs) were recently described as intracellular Ca2+ channels located in acidic organelles such as lysosomes (199, 585, 586). The activity of intracellular Ca2+ channels depends on the action of second messengers produced by different stimuli and diverse signaling pathways (IP3, cADPR, and NAADP, respectively). The diversity of these channels and their mobilizing agents allows the cell to produce specific signals due in part to spatiotemporal differences in the signaling cascades.

1. IP3Rs

There are three genes coding for IP3Rs with differential tissue expression and distinct properties. The generation of IP3 together with DAG occurs via stimulation of G protein-coupled receptors (GPCR) or tyrosine kinase-coupled receptors which activate PLCs that in turn cleave PIP2 to produce IP3 and DAG. The IP3R is regulated by phosphorylation via Ca2+/CaMKII and cGMP-dependent protein kinase (359). The three-dimensional structure of IP3R is tetrameric with three functional domains: the ligand-binding domain (NH2 terminal), the channel-forming domain with six membrane-spanning regions (COOH terminal), and the modulatory/coupling domain, separating the two regions (6). There are several proteins that bind to IP3R and modulate its activity such as cytochrome c, huntingtin, CaM, Homer, CARP (carbonic anhydrase-related protein), PKA, and RACK1 (substrate for the Src protein-tyrosine kinase), among others (121, 359). This diverse set of proteins allows the production of unique spatiotemporal patterns in Ca2+ intracellular signals (121). Particularly interesting is the 530-amino acid long protein named IRBIT (inositol-1,4,5-trisphosphate receptors binding protein released with IP3) that was recently described as an IP3R binding protein (6). IRBIT acts as a pseudoligand (competitive inhibitor) because it does not activate IP3Rs and dissociates from them upon IP3 binding, regulating Ca2+ signaling (6). After its release, IRBIT was found to interact and activate the pancreatic but not the kidney Na+/HCO3 cotransporter 1 (NBC1), suggesting a role in HCO3 secretion and pH regulation in epithelial tissues (467). IRBIT has a multisite strict phosphorylation requirement for its dual modulation activities (289). Regulation of HCO3 secretion and Cl absorption is an important but incompletely resolved issue for different epithelia. For example, in the pancreatic duct, it involves the concerted function of cystic fibrosis transmembrane regulator (CFTR) and SLC26 transporters, although the molecular mechanism is still under study. IRBIT has emerged as a potential coordinator of this interaction, linking fluid and electrolyte transport with cAMP and Ca2+ signaling (572). Recently, it has been shown that IRBIT can also regulate Na+/H+ exchanger type 3 (NHE3) activity in kidney proximal tubules (245). Interestingly, although IRBIT mRNA was found to be ubiquitously expressed, the highest levels were detected in brain, kidney, and reproductive tissues such as testis (6). Members of the SLC26 family as well as CFTR channels have also been reported in spermatozoa from different species (117, 249).

2. RyRs

There are three genes encoding RyRs with several splice variants and differential expression levels depending on the tissue. RyR1 and RyR2 are present in skeletal and cardiac muscle, respectively; all three isoforms are present in brain (469). RyRs are very large proteins that form homotetramers with four subunits of 565 kDa each, and the huge cytosolic NH2 terminal (80% of the protein) is a scaffold for proteins that regulate channel function (303). The modulation of RyRs is quite diverse, including CaM, phosphorylation (PKA and CaMKII), oxidation, nitrosylation, ions (Ca2+ and Mg2+), pH, and several regulatory proteins (reviewed in Ref. 580). In muscle cells, activation by Ca2+ plays an important role, a mechanism known as Ca2+-induced Ca2+ release (CICR). In nonmuscle cells, RyR2 and -3 are also activated by Ca2+ and/or by cADPR (580). The diverse regulatory mechanisms allow the participation of RyRs in several cellular functions such as muscle contraction, exocytosis, secretion, neurotransmitter release, and gene transcription. During muscle contraction, there is a well-known excitation-contraction coupling and physical interaction between CaV1 channels and RyR1. Similar interactions have been described in smooth muscle and neurons that are probably mediated by the Homer protein. Interestingly, RyRs have been reported to interact with other Ca2+ channels such as members of the TRPC family that can also bind Homer (416).

3. TPCs

TPCs are encoded by either two or three genes, depending on the species (62). This novel class of Ca2+ channels is found in acidic organelles and was first described in plants (188) and in rats (274). Structurally, TPCs function presumably as dimers; they resemble one-half of NaV and CaV channels, since they have two (instead of four) domains with the classical six TM regions. TPCs possibly represent evolutionary intermediates of NaV and CaV channels (586). Plant TPCs possess two EF-hand Ca2+ binding motifs, a feature lost in animal TPCs, consistent with their respective regulation by Ca2+ and nicotinic acid adenine dinucleotide phosphate (NAADP) (404). NAADP was long known as a potent Ca2+-mobilizing agent (nanomolar range), but the target remained elusive until TPCs were discovered. TPCs have just arrived to the experimental arena; therefore, there is still much to be done to discover their properties, topology, and modes of regulation. For example, recently Calcraft et al. (78) proposed a mechanism by which the Ca2+ signal generated by NAADP targeting lysosomal TPC2 could be further amplified by CICR via IP3R. The initial NAADP-induced Ca2+ signal may be small, but precisely localized to couple with neighboring IP3R or RyR systems, to trigger larger or global Ca2+ changes, if a particular threshold is reached. This coupling of signals provides a versatile and spatially organized regulatory mechanism for cellular functions (78).

Interestingly, cADPR and NAADP, the agonists of RyR and TPCs, respectively, are synthesized by the same enzyme, ADP-ribosyl cyclases, a family of multifunctional enzymes that catalyze several reactions including the synthesis of cADPR and NAADP, as well as their hydrolysis (443). These reactions are differentially regulated by several factors that include pH, cGMP, cAMP, Zn2+, etc. (443).

It is worth mentioning that sea urchin sperm expresses a particular type of NAADP synthase, and a possible role for NAADP has been proposed for the sea urchin AR (529, 530).

F. SOC (or SOCIC) Machinery Components in Spermatogenic Cells and Spermatozoa

1. mRNA detection

As mentioned earlier, SOC channels have important roles in sperm motility and the AR (see sects. IVB and V). Accordingly, the presence of different components of the SOC complex has been reported in spermatogenic and sperm cells (Table 3). Mouse spermatogenic cells express transcripts for all three IP3Rs and RyRs (515). Additionally, transcripts for trpc1–7 and trpc1, -3, -6, and -7 were found in mouse and human spermatogenic cells, respectively (85, 516). More recently, several transcripts from other TRP subfamilies [TRPV (V1, V5) and TRPM (M3, M4, M7)] have been documented in rat spermatogenic cells (53, 319). In particular, TRPV1 (boar) (49, 50) and TRPM8 (human and mouse) transcripts have been found in spermatozoa (143, 347).

View this table:
Table 3.

Expression of TRP and intracellular channels in sperm cells

2. Protein detection

IP3Rs have been immunolocalized to the acrosome of several species (539) and in some cases also detected in the RNE (260). RyR3 protein was also found in mouse spermatozoa (119, 239, 515). Several TRPC proteins were identified in mature mouse and human spermatozoa. TRPC1, -2, -3, and -6 were immunolocalized predominantly in the flagella, and TRPC2 in the overlying region of the mouse sperm acrosome (283, 516). In human spermatozoa, TRPC1, -3, -4, and -6 were detected in the flagella (85, 516). From the newly discovered SOC components, only STIM1 has been immunolocalized to the neck region of human sperm (105), although it is likely that ORAI proteins are also present and functional in spermatozoa.

There are also reports of the presence of TRP channels from other subfamilies. For example, western blot analysis verified the expression of TRPM4, TRPM7, and TRPV5 in rat spermatogenic cells and spermatozoa (318), and TRPM8 expression was confirmed by western blot and immunocytochemistry in human and mouse spermatozoa (143, 347). The physiological role of these so-called sensory channels may be in sperm guidance processes such as chemo- or thermotaxis.


A. The Epididymis

General transcription and translation of sperm proteins cease in the late spermatid stage in the testis. Upon leaving the testis, although morphologically differentiated, mammalian spermatozoa are unable to swim or fertilize eggs. Their suppressed ability to move forward, possibly related to the changing levels of 2-arachidonoylglycerol throughout the epididymis, is recovered as they further mature (94) (Fig. 2). In contrast, spermatozoa of invertebrates and lower vertebrates have full fertilizing capacity when released from the testis (571). Transit from the epididymal caput to the cauda via the corpus advances sperm functional maturation and takes from 7 to 11 days depending on the species. Different epididymal regions secrete factors that mediate sperm maturation (100). For example, certain mammalian cysteine-rich secretory proteins (CRISPS) are incorporated into, and onto, sperm during spermatogenesis and posttesticular sperm maturation. CRISPS contain two domains, a CAP domain and a characteristic cysteine-rich COOH-terminal domain, referred to as the Crisp domain (96, 214). Epididymal principal cells secrete high concentrations of CRISP1 and CRISP4 into the epididymal lumen where they surround the spermatozoon. CRISPs have been found to regulate voltage-gated K+ channels (KV), CaV channels, CNGA channels, and RyRs (96, 214, 570).

Epididymal fluids from several species seem to be hyperosmolal (∼280–420 mmol/kgH2O) to blood serum that is isotonic to female tract fluids (∼300 mmol/kgH2O) (574). When ejaculated, spermatozoa mix with accessory gland fluids that constitute most of the ejaculate and that are hyposmolal to epididymal fluid. Therefore, ejaculated sperm face a hyposmotic challenge upon ejaculation. Indeed, it is worth emphasizing that spermatozoa develop their ability for osmotic regulation while traveling through the epididymis (101, 496).

During posttesticular maturation, rodent spermatozoa take up small water-soluble components of the epididymal fluid (i.e., taurine, l-carnitine, myo-inositol, and glutamate) (522). Human spermatozoa seem to use inorganic osmolytes instead (574). This process of slow isovolumetric regulation is likely to help spermatozoa contend with the hyposmolal fluids of the male accessory glands and female tract that they encounter upon ejaculation.

It is interesting that male infertility of apparently healthy animals able to mate has, in some instances, been correlated with spermatozoa displaying angulated flagella when released from the epididymis. Furthermore, several transgenic mice with normal sperm production that are infertile also display flagellar angulation after epididymal transit, which is uncommon in normal mice (101). Angulation seems to result from permeability problems and often occurs at the end of the midpiece where the cytoplasmic droplet is located (see sect. I for definition and Figs. 1 and 2). The capacity of mouse and human spermatozoa to properly regulate their volume has been correlated with their ability to fertilize (173, 409).

As the droplet is most likely a significant cytoplasm compartment, it is therefore an important site for osmolyte and water permeation. Morphological changes at this site minimize membrane damage due to stretching. In most species, the cytoplasmic droplet migrates from the interface between the head and flagellum to the annulus as spermatozoa travel from the caput to the corpus of the epididymis (Fig. 2) (101). Irrespective of its position, the role of the cytoplasmic droplet in osmolyte regulation is consistent with the localization of K+ and Cl channels involved in volume regulation at this site in mouse and human spermatozoa (36, 573). The droplet is usually not lost within the epididymis; it is lost at ejaculation in many species though apparently not in humans (99). This loss seems to improve fertility in some domestic species (101).

B. Motility

The fundamental goal of spermatozoa is to deliver their genetic material to the female gamete. These highly specialized cells have evolved sophisticated strategies to regulate their motility and succeed in finding the egg or oocyte. Thus motility is one of the most important functions of male gametes. Therefore, it is not surprising that sperm motility defects have often been found to cause sterility in many gene-targeted transgenic mice (PKA, CatSper, etc.) (165, 358, 390, 395, 459, 472, 542, 584), although they display no morphological defects. Most male gametes, including some species of the plant kingdom, use a flagellum as the driving force for motility. As seen in Figure 1, the axoneme is the core structure that propels flagella (328). It is typically formed by nine microtubule doublets around a central pair of microtubules (9 + 2 structure). Each outer doublet is connected to its neighbor by nexin protein links (not shown) and also has a series of projections, called the radial spokes, that act as spacers to position the doublets in a circle around the central pair of microtubules. All doublets are anchored to the base of the flagellum, which usually terminates in a centriole-like basal body or in the connecting piece in mammalian spermatozoa. The force that slides microtubules producing flagellar beating is generated by the axonemal motor protein dynein ATPases that walk unidirectionally towards the base of the flagellum. Dyneins from one doublet transiently interact with the following doublet. During the beat cycle, dyneins on one side of the axoneme tend to bend the flagellum in one direction, and dyneins on the opposite side bend it in the other direction (reviewed in Ref. 72). Hence, the beat consists of alternate episodes of activation of the two opposing dynein bridge sets, where each set is switched “on” and the other “off” alternatively at appropriate mechanical set points. The control of the activation switch of the two dynein-bridge sets resides in the axoneme itself (see Ref. 328 for a comprehensive review). Dynein ATPase activity is known to be modulated by pH, ATP, ADP, Ca2+, and phosphorylation mediated for instance by cAMP/PKA (123, 270, 271, 328, 372).

1. Adenylate cyclases in spermatozoa

cAMP is an essential second messenger in sperm physiology (571) that is synthesized by two types of adenylyl cyclases (ACs): a soluble adenylyl cyclase (sAC) (330) and transmembrane ACs (mACs, nine isoforms: mAC1–9). The sAC is regulated by bicarbonate and Ca2+, and the mACs are activated mainly by Gsα (the α-subunit of heterotrimeric G proteins), and differently regulated by [Ca2+], PKA, and PKC (236). The sAC is essential for sperm motility regulation and fertility (165, 252), is structurally and functionally similar to ACs from cyanobacteria, and is not activated by G proteins or forskolin, the typical activators of mACs. Although in mouse sperm sAC is located in the flagellum (252), it may affect plasma membrane changes in the anterior part of the sperm head (189). In sea urchin spermatozoa, it is distributed along the cell and also participates in the AR (44). sAC was thought to function only in male germ cells, but it is also present in somatic cells localized at distinct sites (169, 588, 589). In human airway cells, sAC is localized to cilia where it regulates their motility (457).

Initially it was thought that cAMP in spermatozoa was produced only by sAC and that Gsα was absent. However, evidence has suggested that several mACs and Gsα may be present in sea urchin and mammalian spermatozoa (39, 44, 183, 478). Further experiments must corroborate that mACs are found in spermatozoa. There are experimental lines that indicate that both AC types are important for sperm maturation, motility, and the AR. mAC3 null mice are subfertile as their spermatozoa show spontaneous AR alteration (334).

2. Flagellar form modulation by Ca2+

It is known that, in an appropriate medium supplied with ATP, the sperm flagellum can be maintained motile after removal of the plasma membrane by detergent treatment (213). With the use of demembranated sea urchin sperm, the importance of Ca2+ for sperm flagellar form was first reported in 1974. The curvature of sperm swimming trajectories became larger as the Ca2+ concentration was elevated (73), due to the increase in flagellar asymmetry (71). These discoveries led to a central dogma that the degree of flagellar beat asymmetry is determined by the global Ca2+ concentration. The first Ca2+ imaging of the moving sperm flagellum was performed in 1993 using hamster spermatozoa (493). This experiment revealed that the [Ca2+]i of hyperactivated spermatozoa was higher than that of activated spermatozoa (see sect. IVC). Hyperactivated sperm motility occurs during capacitation (see sect. IVF) and is generally characterized by high amplitude and asymmetric flagellar bends in nonviscous media (Fig. 4). However, we now know that the relationship between [Ca2+]i and flagellar asymmetry is not always proportional, as described later in sea urchin and sea squirt (ascidians) spermatozoa (465, 562) (see sect. VI). Although several efforts have been made, the mechanism of how Ca2+ controls the form of flagellar bending is not fully understood. Pharmacological evidence indicates that CaMKII and/or CaMKIV is involved in this process in mammalian sperm (269, 344). Recently, a novel Ca2+-binding protein, named calaxin, was identified as an axoneme component in sea squirt sperm (365). Bioinformatics revealed that calaxin belongs to the neuronal calcium sensor protein family. Immunoelectron microscopy and biochemical analysis showed that it interacts with the dynein heavy chain at the outer arm dynein in a Ca2+-dependent manner. These results suggest that calaxin may be important for flagellar form regulation by Ca2+ (365).

Figure 4.

Properties and physiological roles of hyperactivation. A: activated (left) and hyperactivated (right) flagellar shape and motility direction (arrows) displayed by spermatozoa in nonviscous experimental media. Activated spermatozoa advance showing a symmetric flagellar bend, while hyperactivated spermatozoa tumble in one place and do not advance efficiently due to an asymmetric flagellar beating pattern. B: importance of hyperactivated sperm motility: i) To advance in highly viscoelastic fluids in the female genital tract more effectively than activated sperm, ii) to detach spermatozoa from the isthmus reservoir and advance towards the ampulla (site of fertilization), and iii) to facilitate sperm penetration through the cumulus matrix and the zona pellucida.

C. Hyperactivation

In mammals, spermatozoa become motile when ejaculated into the female genital tract. The ejaculated spermatozoa show symmetric flagellar beating with low curvature and swim progressively with an almost straight path (Fig. 4). This swimming pattern is called “activated motility.” Furthermore, a subpopulation of sperm recovered from the oviduct shows vigorous asymmetric flagellar beating with large amplitude and high curvature in a standard low-viscosity medium, which is called “hyperactivated motility.” Hyperactivation is an essential process for spermatozoa to successfully fertilize oocytes, and Ca2+ is a fundamental regulatory factor; it is not known how it is triggered under physiological conditions (reviewed in Ref. 488). Possibly distinct factors such as bicarbonate (sect. IVD), progesterone (sect. IVE), or a temperature decrease (sect. IVG) transiently induce hyperactivation at different sites and occasions while the spermatozoa search for the oocyte.

Ca2+ is a fundamental regulatory factor for sperm hyperactivation. With the use of demembraneted bull spermatozoa, it was demonstrated that Ca2+ but not cAMP can induce hyperactivation (259). For instance, 80% of spermatozoa show hyperactivation at 400 nM Ca2+. As already mentioned, [Ca2+]i is higher in hyperactivated hamster spermatozoa (100–300 nM) than in activated spermatozoa (10–40 nM) (493). Ca2+ channels at the plasma membrane (CatSper) and from internal Ca2+ stores (IP3R and RyR) seem involved in achieving hyperactivation (see below). In addition to Ca2+ channels, sperm [Ca2+]i homeostasis is controlled by several types of Ca2+ transporters (552). A plasma membrane Ca2+-ATPase 4 (PMCA4), localized in the principal piece of the flagellum, is crucial for sperm function since its elimination affects sperm motility (459) and hyperactivation (395) resulting in male infertility. Mitochondrial abnormalities found in PMCA4-deficient spermatozoa (395) suggest Ca2+ overload due to defective Ca2+ extrusion.

1. Physiological roles of hyperactivation

Hyperactivated spermatozoa are less progressive than activated spermatozoa in low viscous media. Some hyperactivated cells are completely nonprogressive; they just tumble in the same place. In contrast, hyperactivated spermatozoa swim more progressively than activated spermatozoa in high viscous and/or high viscoelastic media (489, 491) (Fig. 4). This experimental observation indicates that one of the physiological roles of hyperactivation is to ensure progressive movement in the high viscous fluids encountered in the female reproductive tract, such as the cervical mucus which is several hundredfold more viscous than aqueous experimental media (560). Hyperactivation is also important to facilitate sperm penetration of the extracellular matrix of cumulus cells and of the oocyte, known as the zona pellucida (ZP) (Figs. 3 and 4) (482). Certain transgenic mice are infertile due to sperm motility defects, but in some cases their spermatozoa can fertilize ZP-free oocytes, but not intact oocytes (334, 426, 432). This defect has been attributed to a specific inability to initiate hyperactivated motility. On the other hand, it is known that the head of mammalian spermatozoa attaches to epithelial cells in the isthmus of the oviduct, a site known as the sperm reservoir (150, 492) (Fig. 3). Spermatozoa detach from the reservoir by hyperactivated motility and swim forward to the site of fertilization, the ampulla (150). Therefore, a third function of hyperactivation is to allow spermatozoa to detach from the sperm reservoir in the oviduct which may occur in multiple occasions (Fig. 4).

D. CatSper Channels

1. Structure of CatSper

A combination of bioinformatics and gene-targeted transgenic mouse strategies revealed that spermatozoa possess a sperm-specific cation channel named CatSper (cation channel of spermatozoa), which is essential for hyperactivated sperm motility. CatSper is a novel type of Ca2+ channel most likely composed of four separate pore-forming subunits, CatSper1–4 (9, 279, 335, 425, 432), and three additional auxiliary subunits, CatSperβ, CatSperγ, and CatSperδ (124, 332, 543) (Fig. 5).

Figure 5.

Structure of CatSper channel. A, left: protein topology of CatSper1 channel showing six transmembrane segments (S1-S6) with S4 containing several positively charged residues, the pore loop, and cytoplasmic COOH and NH2 termini; the latter has a histidine (His)-rich region. Middle: Catsper 2, 3, and 4 have the same CatSper1 architecture except for a shorter COOH and NH2 termini, lacking the His-rich region. Right: topology of the β, γ, and δ CatSper subunits with one, two, and one trasmembrane segments, respectively. B: sequence of the mouse CatSper1 NH2-terminal showing the histidine-rich region (51 of 250 histidine residues are shown in bold). C, top: amino acid sequence alignment of the human (h) and mouse (m) S4 segment of CatSper 1–4, the putative voltage-sensor. Bottom: the S4 segments of each domain (I-IV) for mouse CaV3.1 channel is also shown for comparison. Positively charged residues (arginine and lysine) are shown in bold. D, top: amino acid sequence alignment of the human (h) and mouse (m) pore region of CatSper1–4. Bottom: the loop forming sequence of each domain (I-IV) for mouse CaV3.1 channel is also shown for comparison. The pore consensus sequence for Ca2+-selective channels is [T/S]x[D/E]xW (shown in bold).

The pore-forming subunits of CatSper (CatSper1–4) structurally resemble CaV channels, including the selectivity consensus sequence for Ca2+ (T/S X D/E X W) (335) (Fig. 5). Each of the 4 separate CatSper subunits has 6 TM segments, similar to KV channels, in contrast to CaV channels which are composed of 24 TM segments in a single polypeptide. In all known voltage-gated cation channels, the 4th TM segments (S4) have several positively charged residues (R/K). This is the case of CatSper1 and CatSper2, although CatSper3 and CatSper4 have only two (335), which may explain the mild voltage dependence of its gating. Remarkably, CatSper1 has a histidine-rich large cytoplasmic terminal domain (432) that varies in sequence and length among species, even among primates (414).

In contrast to CatSper1–4, auxiliary subunits were identified by proteomics as CatSper1-associated proteins. Expressing a fully functional HA-EGFP (green fluorescent protein)-tagged CatSper1 in a CatSper1-null background, allowed immunopurification of protein complexes containing CatSper1 and novel proteins CatSperβ and CatSperγ (332, 543). CatSperβ is a 126-kDa protein containing two TM segments with two short cytoplasmic domains and a large extracellular domain. CatSperγ is a 131-kDa protein containing a single TM segment with a large extracellular domain and a short cytoplasmic tail. Notably, CatSper proteins including CatSperβ and CatSperγ are practically absent in mature spermatozoa from CatSper1-null mice (332, 543). This is strong experimental evidence that CatSperβ and CatSperγ are auxiliary proteins for the CatSper channel. The large extracellular domains of CatSperβ and CatSperγ suggest possible channel modulation by external signals such as ligands and cell-cell or extracellular matrix-cell interactions during sperm maturation in the epididymis and/or the female genital tract. As indicated earlier, all CatSper-related genes (except for CatSperδ) were found in the sea squirt (Ciona intestinalis) and sea urchin (Strongylocentrotus purpuratus) genomes, but not in those of flies (Drosophila melanogaster) or worms (Caenorhabditis elegans) (77, 332, 543). Just recently, CatSperδ, the newest auxiliary protein, was identified in the CatSper molecular complex (124). CatSperδ has a single TM segment with a large extracellular domain and a short cytoplasmic tail, a similar protein topology as CatSperγ. Notably, CatSperδ-null transgenic male mice are infertile due to a hyperactivation defect. The amount of CatSper1 in spermatozoa is remarkably reduced in the CatSperδ-null mice, indicating that CatSperδ is an essential auxiliary subunit for expression of the functional CatSper channel (124).

2. Biophysical properties of CatSper

In spite of intensive efforts by many researchers, functional CatSper channels have not been successfully expressed in heterologous systems, preventing their full biophysical characterization. They have only been studied in spermatozoa of wild-type (WT) and mutant mice. Male infertility was observed in CatSper1–4 gene-targeted transgenic mice with an identical phenotype (278, 423, 426, 432), which indicates that all pore-forming CatSper genes are essential for the functional channel. It seems likely that the CatSper heterotetramer must be composed of the four distinct subunits, differing from other six TM channels such as KV, TRP, and CNG channels that are frequently composed of homotetramers, although in some instances these latter classes of channels may also form heterotetramers under physiological conditions (285).

The recording of CatSper ionic currents necessitated the introduction of a strategy to facilitate obtaining whole cell patch-clamp recordings. Kirichok et al. (292) found that the cytoplasmic droplet (see Fig. 1) of epididymal sperm is a suitable region to form giga-seals with the patch pipette (292). In WT mouse spermatozoa, these authors recorded a large current in divalent cation-free (DVF) medium, that was absent in spermatozoa from CatSper1-null mice. This large monovalent cation current recorded in epididymal spermatozoa was defined as ICatSper. A striking feature of ICatSper is that it is potentiated by intracellular alkalinization (292). Since the NH2-terminal cytoplasmic domain of CatSper1 is histidine rich, it is speculated that this domain functions as a pH sensor.

The size of the monovalent ICatSper was reduced by increasing extracellular Ca2+ concentrations with an IC50 of 65 nM (292), a typical property of Ca2+-selective channels. Ca2+ inward currents were recorded with external solutions containing 2 mM Ca2+ and with pipette solutions at pH 8. The Ca2+ current was much smaller than the monovalent cation current observed in the DVF medium and was absent in spermatozoa from CatSper1-null mouse, indicating its identity with ICatSper.

As for Ca2+-selective channels, Ba2+ can pass through CatSper channels without causing Ca2+-dependent inactivation (292). With the use of Ba2+ tail currents, it was observed that ICatSper is much less voltage-dependent than typical voltage-gated channels. It is worth pointing out that the voltage at which ICatSper can be activated shifts toward negative potentials, closer to sperm Em values, as the pipette pH is raised from 6.0 to 7.5 (292), the physiological range for sperm pHi. This property reasonably explains why ICatSper is stimulated by intracellular alkalinization. It was previously reported that a medium of high K+ and pH 8.6 (K8.6) induced an increase in [Ca2+]i in mouse spermatozoa (554). This cell-depolarizing and -alkalinizing stimulus is an accidentally appropriate condition to activate CatSper channels. Indeed, this stimulus does not increase [Ca2+]i in CatSper1-null mouse spermatozoa (83).

Electrophysiological recordings (292) as well as immunolocalization of CatSper1 (432), CatSperβ (332), CatSperγ (543), and CatSperδ (124) locate this channel in the principal piece of the flagellum. Furthermore, the NH4Cl-induced Ca2+ increase commences in the principal piece of the flagellum and spreads relatively slowly to the entire mouse WT spermatozoa, but not in CatSper1-null mouse spermatozoa (565).

3. Physiological functions of CatSper

The unquestionable importance of CatSper channels is ascertained by the fact that null mice lacking any of the CatSper isoforms are infertile (278, 423, 426, 432), and human CatSper1 and -2 mutations have been associated with asthenoteratozoospermia (19, 20).

CatSper 1-null mouse spermatozoa cannot fertilize ZP-intact oocytes but do fertilize ZP-free oocytes, indicating that CatSper channels are required to allow sperm to penetrate ZP (432). A careful comparison of the flagellar beating form of WT and CatSper1-null spermatozoa revealed that mutant spermatozoa were unable to display hyperactivated motility; their flagellar beating remains symmetric with small amplitude and low curvature even in capacitating medium (see sect. IVF) (83). In agreement with the involvement of CatSper1 in motility, CatSper2-null mouse spermatozoa swim less progressively than WT mouse spermatozoa in a high viscous medium (426). Observations 1 h postcoitum of sperm migration in the oviduct documented the presence of both WT and CatSper1-null spermatozoa in the lower isthmus, the sperm reservoir in the oviduct (262). However, CatSper1-null spermatozoa did not reach the ampullary-isthmic junction, the upper region of the oviduct where fertilization takes place, even 4 h postcoitum (262). In vivo video-recordings revealed that WT spermatozoa in the isthmus migrated toward the upper region of the oviduct by repeatedly detaching and reattaching to epithelial cilia using vigorous flagellar beating. In contrast, CatSper1-null spermatozoa remained attached to the epithelial cilia (262). These findings show that hyperactivation is essential for sperm migration in the oviduct, particularly to escape from the sperm reservoir. CatSper-null mouse spermatozoa manifest clear defects in all physiological roles previously proposed for hyperactivation (Fig. 4): 1) progressive swimming in high viscous fluid, 2) penetration through egg extracellular matrix, and 3) escape from the reservoir in the isthmus. These findings combine to establish the importance of hyperactivation for fertilization and the key involvement of CatSper in these motility-related processes.

In addition to their role in Ca2+ signaling, capacitated CatSper1-null mouse spermatozoa have been found to have higher cAMP levels (82) and lower ATP levels (565) than WT sperm. It is not clear how these changes occur. Considering that glycolysis is fundamental for mammalian sperm ATP synthesis (358) and that glycolytic enzymes are localized in the principal piece of the flagellum, CatSper might participate in regulating glycolysis. In demembranated bovine spermatozoa, ATP dose-dependently potentiates high flagellar curvature, indicating its importance in hyperactivation (259). It is therefore clear that CatSper is essential for hyperactivation due to its influence on Ca2+ influx and indirectly on ATP levels. On the other hand, the lack of CatSper does not reduce tyrosine phosphorylation during capacitation (see sect. IVF) (83, 426) or interfere with the AR (see sect. V) (565).

4. Regulation of CatSper under physiological conditions

Cyclic nucleotides are known to induce an increase in [Ca2+]i in WT spermatozoa (295), but not in CatSper1-null spermatozoa (432). However, cyclic nucleotides do not directly activate CatSper channels (292). Considering that ICatSper is potentiated by alkalinization, cAMP might indirectly stimulate CatSper by a pHi increase (381, 433) (Fig. 6). Uniquely, spermatozoa possess a Na+/H+ exchanger (sNHE) (542) containing a consensus cyclic nucleotide binding domain (CNBD) sequence. In addition, sNHE seems to directly interact with sAC (541), and both sNHE and CatSper are localized in the principal piece of the flagellum (432, 542). With all information taken into account, it is likely that a functional link between sNHE and CatSper exists, a matter worth exploring. There is as yet no experimental evidence that cAMP elevates sperm pHi, and it has not been possible to functionally express sNHE in heterologous systems efficiently, impeding its direct characterization. Future studies will surely determine how sNHE is regulated and its functional connection to the CatSper channel.

Figure 6.

Ion fluxes during mammalian sperm hyperactivation and capacitation. Both processes take place during sperm transit through the female tract and have parallel but also intercrossing pathways. Albumin present in the female tract removes cholesterol from the sperm plasma membrane modifying several membrane properties; it may also directly activate CatSper channels, increasing [Ca2+]i. Activation of a Na+/HCO3 (NBC) cotransporter (possibly activated by external Ca2+) increases HCO3 levels activating a soluble adenylyl cyclase (sAC) and producing cAMP. This cAMP may activate the sperm Na+/H+ exchanger (sNHE) and together with the activation of the proton channel (HV) (by Zn2+ removal) would raise pHi, another cellular parameter that activates CatSper and SLO3 channels. The overall Ca2+ increase may influence glycolysis and the axoneme activity promoting hyperactivation of motility. Several Ca2+ mobilizing pumps (PMCA4, SPCA1) and channels [IP3 receptor (IP3R), ryanodine receptor (RyR)] may also participate during Ca2+ signaling. Concomitantly, the increase in cAMP levels activates PKA, which after several unknown steps stimulates tyrosine kinases to produce the tyrosine phosphorylation associated with capacitation. Possible connections between hyperactivation and capacitation signaling cascades are indicated. In some species, capacitation is accompanied by membrane hyperpolarization; the channels and transporters involved during this process are indicated.

It is worth noting that CatSper3,4-null spermatozoa keep moving in Ca2+-free medium (<10 nM) (278), in which mammalian spermatozoa are known to lose their motility. This observation can be explained by the massive Na+ influx through CatSper channels that occurs in the absence of external Ca2+ in WT spermatozoa (511), but not in CatSper-null spermatozoa (80). In mouse (161) and human spermatozoa (511), it was demonstrated that external Ca2+ removal induces a Na+-dependent depolarization and repolarization by readdition of Ca2+.

5. HVs may modulate pHi and CatSper

The regulation of pHi is fundamental for sperm function; in mammals, these cells face significant external pH changes during maturation and in their transit through the female genital tract (571). Recently, it was reported that voltage-gated proton channels (HVs) are functionally expressed in mammalian spermatozoa (329). Interestingly, sperm HV currents are much larger in human than in mouse, suggesting that pHi regulation may be somehow different in the two species. In mouse, sNHE (see previous section) could dominate sperm pHi regulation rather than the HV. These differences may be related to the fact that human spermatozoa do not seem to hyperpolarize during capacitation according to fluorescence sperm population measurements (216), although flow cytometry results have indicated a small hyperpolarization may occur (70). As indicated earlier, seminal fluid contains a high Zn2+ concentration (mM), and micromolar concentrations of this divalent cation block HV channels (329). Zn2+ concentrations are lower in the female genital tract; therefore, the levels of this cation may contribute to the regulation of sperm responsiveness at different stages by affecting HV activity. In addition, lipids such as arachidonic acid potentiate HV activity (329). It has been proposed that the depolarization-induced [Ca2+]i increase in human spermatozoa may be due to CaV channels (327). Alternatively, a depolarizing stimulus could increase pHi through the HV channel, which would result in CatSper channel activation (433).

E. Progesterone Induces Flagellar Responses and Ca2+ Release From Ca2+ Stores

Progesterone, a steroid hormone, is synthesized by the cumulus/granulosa cells and regulates the female sexual cycle. This steroid was identified as the major active component of the human follicular fluid that induces the AR (see sect. V) (398). While genomic action of this steroid is established through the nuclear receptor (167), progesterone binds to an unidentified receptor on the sperm plasma membrane, provoking a rapid increase in the [Ca2+]i (57), referred to as a nongenomic action (337). The signaling mechanisms involved in the nongenomic progesterone response in mammalian spermatozoa remain ill-defined. The progesterone-induced [Ca2+]i increase depends on extracellular Ca2+ (55), although Ca2+ release from intracellular store(s) may be involved (41). Stopped-flow fluorometry revealed that progesterone-induced [Ca2+]i increases occur with a minimal time lag (<50 ms) (290), indicating that the Ca2+ transporter should be regulated by the receptor via a small number of intermediary steps, in contrast to the response to chemoattractants (200 ms time lag) in echinoderm spermatozoa (286, 388) (see sect. VI). Nevertheless, the molecular identity of the Ca2+ channel involved in the progesterone-induced rapid Ca2+ influx is still unknown (see note added in proof).

A G protein-coupled receptor for progesterone was discovered in fish oocytes in 2003 (587), and expression of a homologous membrane progesterone receptor (mPRα) was detected in human testis (587). Western blotting revealed mPRα in human sperm membranes and immunostaining localized this protein mainly in the midpiece of the sperm flagellum (507). Progesterone treatment of sperm membranes increases GTPγS binding, suggesting it activates a G protein (507). These findings point to mPRα as a candidate nongenomic receptor for progesterone in human spermatozoa. Progesterone induced higher [Ca2+]i transient increases in capacitated than in noncapacitated human spermatozoa (see sect. IVF) (511). Phosphodiesterase inhibitors potentiated this transient increase in noncapacitated spermatozoa and a PKA antagonist inhibited the response, suggesting that a PKA-dependent phosphorylation upregulates the progesterone signaling cascade (512). In contrast, a PLC inhibitor, U73122, did not affect the progesterone-induced increase in [Ca2+]i, indicating that PLC is not involved in the signaling cascade (4). However, progesterone has been reported to stimulate PLC activity and hence IP3 production (506). Reexamination of these matters is necessary. Recently, a caged progesterone analog was developed (290). This compound is a promising tool to unveil the signaling mechanism and physiological function of this steroid.

The physiological function of progesterone had been ascribed mainly to the AR (see sect. V). Recent observations indicate that this steroid is also important for the modulation of sperm motility. Progesterone induces [Ca2+]i oscillations in a fraction of human spermatozoa (10–36%), and these responses are accompanied by increases in flagellar bend amplitude (239), one of the important characteristics of hyperactivation. While the initial Ca2+ transient induced by progesterone is caused mainly by Ca2+ influx, the subsequent Ca2+ oscillations are likely due to Ca2+ release from Ca2+ store(s) localized in the base of the flagellum (239) (Fig. 6). Once the Ca2+ oscillations started, removal of external Ca2+ (<5 μM) did not affect the oscillations, although further addition of a Ca2+ chelator inhibited them. Low concentrations (50–100 μM) of ryanodine accelerated the oscillations, while high doses (500 μM) reduced them. TMB-8 and tetracaine, inhibitors of Ca2+ release from Ca2+ stores, arrested the oscillations, suggesting the importance of store emptying/refilling cycles in this process (238). The progesterone-induced oscillations are relatively insensitive to thapsigargin or cyclopiazonic acid, inhibitors of SERCA, but sensitive to bis-phenol, a nonspecific Ca2+-ATPase inhibitor. Western blotting detected secretory pathway Ca2+-ATPase (SPCA1) in human spermatozoa, but not SERCA, and immunostaining showed that SPCA1 is localized in the head-tail junction, where the Ca2+ oscillations are observed (238). A fluorescently labeled ryanodine binds to the same sites (238). These results strongly suggest that SPCA1 is involved in refilling Ca2+ during the Ca2+ oscillations. Similar Ca2+ oscillations were observed in resting conditions in a smaller fraction of the sperm population (293, 386), and this fraction increased upon the addition of nitric monoxide (341) or two CaV1 agonists, BAY K 8644 or FPL64176 (293). Therefore, the Ca2+ oscillations might be induced by different types of stimuli independently of their signaling mechanisms. The progesterone-induced Ca2+ oscillations occurring in a sperm fraction of WT mouse spermatozoa are not observed in PLCδ4-null mice, suggesting the involvement of this PLC in the Ca2+ oscillations (186).

It is believed that mammalian spermatozoa have at least two Ca2+ stores: the acrosome and the RNE (105). Calreticulin, a Ca2+ binding protein generally found in Ca2+ stores, and IP3Rs were detected in these two stores in spermatozoa of various mammal species (260, 261, 377, 539).

Interestingly, thapsigargin (a Ca2+-ATPase antagonist) and thimerosal (IP3R and RyR agonist) can induce hyperactivation in bull spermatozoa (260). Ca2+ imaging revealed that these reagents immediately induce [Ca2+]i increases at the base of the sperm flagellum, where the RNE is localized (261). The effects of thapsigargin and thimerosal are independent of external Ca2+ (261). Thimerosal can induce hyperactivation even in CatSper2-null spermatozoa (345), indicating the importance of Ca2+ release from the Ca2+ store to modulate the flagellar bend. On the other hand, procaine, a reagent which triggers hyperactivation only in the presence of extracellular Ca2+, failed to induce hyperactivation in CatSper2-null spermatozoa (345), indicating that the procaine effect on sperm motility is mediated by CatSper channels.

F. Capacitation

After ejaculation into the female reproductive tract, spermatozoa must further mature, through a process referred to as capacitation, before they can fertilize the egg (18, 112, 571). Although five decades have gone by since its discovery, the molecular mechanisms of this intricate process are still not well understood. Recent findings indicate that the functional changes involved in capacitation result from a combination of sequential and in some instances parallel processes (7, 190). In light of this, capacitation can be viewed as happening in both a faster and slower track, possibly involving distinct events in the sperm head and flagella (534). Fast events such as sperm motility initiation (81) and modification of membrane lipid architecture and organization (190) begin when spermatozoa are released from the epididymis and are exposed to increased HCO3 concentrations. Slower events occur with a delay and are responsible for endowing spermatozoa with the ability to fertilize the egg, such as the acquisition of hyperactivated motility (see sect. IVC) and the capacity to undergo the physiologically relevant AR (see sect. V).

As spermatozoa travel from the epididymis and through the female tract, they encounter significant changes in osmolarity and in the concentrations of various ions. After mixing with the seminal fluid, stored spermatozoa in the caudal epididymis are ejaculated into the female tract where extracellular [K+] ([K+]e) is reduced, and extracellular [HCO3] and [Na+] are significantly increased. HCO3 concentrations elevate from ∼4 mM in the cauda epididymis (16) to >20 mM in the seminal plasma and female tract (340, 393). These [HCO3]e increases immediately stimulate AC (393) and mouse sperm flagellar beating and enhance the cell's sensitivity to depolarization by external K+ (553). Elevated HCO3 also triggers a rapid decrease in sperm membrane phospholipid asymmetry mediated by scramblases that externalize phosphatidylethanolamine and phosphatidylserine (189, 241). The evidence indicates that HCO3 stimulates sAC (330) and increases cAMP levels (7, 68, 144, 397), which in turn activate PKA, resulting in the rapid phosphorylation of a subset of proteins (7, 242). As expected, the ion concentration changes experienced by spermatozoa as they enter the female tract alter their Em (i.e., Ref. 148).

Two crucial ionic parameters for capacitation are [Ca2+]i and pHi (571), and both parameters increase during this process (29, 141, 192, 329, 402, 583). It is known that among other components, Ca2+, serum albumin, and HCO3 are necessary for sperm capacitation in vitro (538, 571). In mouse spermatozoa this process requires a minimum [Ca2+]e of 100–200 μM (182, 343), although 2 mM [Ca2+]e is used in capacitating medium. Interesting results have shown that in addition to its critical role as second messenger, Ca2+ can act as a first messenger, and a CaSR has been cloned (61). The presence of such a receptor and its involvement in the HCO3-induced acceleration of flagellar beat frequency has been postulated in mouse spermatozoa (81).

It is thought that albumin is needed to remove cholesterol from the sperm plasma membrane to induce its reorganization and change its fluidity (108, 513, 538). These albumin-induced sperm plasma membrane changes could modulate Ca2+ and/or HCO3 ion fluxes leading to the activation of sAC and increases in cAMP levels (538) (Fig. 6). Albumin can act independently of the cAMP pathway by inducing proline-directed serine/threonine phosphorylation (277). Notably, this protein can also induce rapid (seconds) [Ca2+]i increases that were recently shown to depend on CatSper and partially on pHi, and to be independent of G proteins and PLC (566). However, electrophysiological recordings revealed that albumin actually hyperpolarizes corpus epididymal spermatozoa instead of depolarizing them (566), an interesting finding worth further study. Here again we find that albumin, a component important for capacitation, causes fast and slow responses in spermatozoa.

A distinguishing feature associated with sperm capacitation is the increase in protein tyrosine phosphorylation. Interestingly, under capacitating conditions, mouse spermatozoa first reach maximum cAMP levels in 60 s and soon after PKA-dependent phosphorylation occurs (446). Thereafter, at least 30 min must pass to detect tyrosine phosphorylation (536). As these tyrosine phosphorylation events occur later and are blocked by PKA inhibition, they are mediated by cAMP and PKA (536). These initial findings in mouse spermatozoa (536) have been corroborated in a wide range of species (193, 309, 535).

Recent reports have suggested that the Src family of protein tyrosine kinases (SFKs) mediate the capacitation-associated increase in tyrosine phosphorylation, hyperactivation, and the AR in mammalian spermatozoa (166, 308, 364, 528). However, newly found evidence using Src-null mice and inhibitors of SFKs and phosphatases revealed that c-Src is either not important for these processes or that the lack of c-Src can be compensated by other members of the SFKs. More importantly, these new findings indicate that capacitation is regulated by two parallel pathways. One of them requires activation of PKA, and the other involves inactivation of Ser/Thr phosphatases (299).

As indicated above, capacitation in mammalian spermatozoa is HCO3 dependent. Aside from activating sAC (330), HCO3 influx hyperpolarizes mouse spermatozoa in a [Na+]e-dependent manner (148). The ionic dependence of this hyperpolarization is consistent with the presence of an electrogenic Na+/HCO3 cotransporter in these cells that is important for capacitation. Furthermore, HCO3 uptake may contribute to the known pHi increase observed during sperm capacitation (148, 402, 583).

Regulation of sperm pHi is fundamental for motility, capacitation and the AR. Spermatozoa possess a specific Na+/H+ exchanger, sNHE, that is a member of the mammalian NHE superfamily (84, 427, 542). The sNHE protein is predicted to contain 14 TM segments and is located on the principal piece of the sperm flagellum. sNHE contains a distinctive consensus sequence for a putative cyclic nucleotide-binding domain near its intracellular COOH terminus and four putative TM segments analogous to the voltage-sensing domain of voltage-gated ion channels (87). These unique features suggest that sNHE may be regulated by cyclic nucleotides and Em (541), in addition to phosphorylation changes and protein interactions (84). sNHE null male mice are infertile, and their spermatozoa are immotile (542). Addition of NH4Cl or permeable analogs of cAMP (541) rescues these sperm functions. It turns out that full-length expression of sAC and its proper regulation by HCO3 require sNHE. sNHE and sAC are apparently part of a transduction complex that could efficiently regulate pHi, HCO3, cAMP, and sperm motility (541).

As mentioned in section IVD, human spermatozoa possess a high density of HV channels in the flagella while those of mouse have less (329). This suggests that sNHE is more important for mouse sperm pHi regulation. This exchanger may be stimulated by the hyperpolarization that occurs when these cells undergo capacitation. HV channels appear to be stimulated during capacitation, possibly by PKC-dependent phosphorylation (329), and may contribute to the increase in pHi which is important for this process.

In addition to HCO3, Ca2+, and albumin, capacitation in vitro depends on the presence of K+, Na+ and Cl in the medium (7, 139). We still do not fully understand which sperm ion channels and transporters are involved in the Em, pHi, and [Ca2+]i changes necessary to achieve this maturational process. Bovine and mouse spermatozoa hyperpolarize during capacitation (12, 148, 374, 582). Noncapacitated mouse spermatozoa are relatively depolarized (Em approximately −35 to −45 mV) and hyperpolarize to approximately −70 mV during capacitation (12, 148, 374). Voltage-gated K+ channels (447), Ca2+-activated K+ channels (110), and inwardly rectifying K+ (Kir) channels (3, 170, 374) have been detected in testis, spermatogenic cells, and spermatozoa using molecular, electrophysiological, and biochemical tools. There is pharmacological evidence that Kir channels could contribute to the capacitation associated with hyperpolarization (3, 374).

Importantly, the second case of infertility caused by the knockout of an ion channel, SLO3, was recently reported (450). The SLO3 channel belongs to the high-conductance slowpoke (Slo) K+ channel family and is only found in mammalian spermatogenic cells and spermatozoa. It was cloned a while ago from a mouse testis library and heterogeneously expressed (458). The mouse SLO3 channel is activated by voltage change and intracellular alkalinization and displays a K+/Na+ permeability ratio (PK/PNa) of ∼5. Navarro et al. (380) reported a pHi-sensitive current in the principal piece of epididymal mouse sperm, sharing some of the SLO3 characteristics, which they named KSPER. When expressed in Xenopus oocytes, SLO3 is also stimulated by elevated cAMP levels through a PKA-dependent phosphorylation. Patch-clamp recordings in testicular mouse spermatozoa revealed that a K+ current that is time and voltage dependent is activated by intracellular alkalinization, has a PK/PNa ≥5, is weakly blocked by TEA, and is very sensitive to Ba2+ (346). These currents are absent in testicular spermatozoa from Slo3 null mice, confirming their identity. Under capacitating conditions, spermatozoa from Slo3 null mice are unable to perform progressive motility, to hyperpolarize, and to undergo the AR. Surprisingly, not even Ca2+ ionophore treatment can trigger the AR in these spermatozoa. These findings indicate that SLO3 channel activation plays a preponderant role in the capacitation-associated processes necessary for fertilization; this channel is thus an attractive candidate for a male contraceptive drug (450). Very recently, it was shown that SLO3 channels heterologously expressed and in epididymal mouse spermatozoa are stimulated by PIP2. Epidermal growth factor (EGF) acting through the EGF receptor inhibits SLO3 currents in a PIP2-dependent manner (501). There are some discrepancies regarding the selectivity and voltage dependence between SLO3 channels expressed in Xenopus oocytes and in spermatozoa that have to be worked out to fully understand the contribution of SLO3 to the capacitation-associated hyperpolarization.

Other ion channels and transporters may contribute to the capacitation-associated hyperpolarization. Amiloride-sensitive epithelial Na+ channels (ENaCs) have been reported to be present in mouse spermatozoa. Noncapacitated mouse spermatozoa are depolarized by the addition of external Na+ and hyperpolarized by ENaC blockers and by a cAMP/PKA-dependent process (248, 297). ENaCs and CFTR have been found to colocalize in several cell types where they regulate each other (48). CFTR is a cAMP-modulated, ATP-dependent, Cl channel, and many of its mutations cause cystic fibrosis (CF), the most prevalent human genetic disease (438). ENaCs and CFTRs have been reported to be present in the midpiece of mouse and human sperm flagella (249, 569). Experimental evidence suggests that cAMP induced activation of CFTR leads to the closing of ENaCs, contributing to the capacitation-associated hyperpolarization (249).

Cl can be transported across the plasma membrane by different Cl channels, exchangers, and cotransporters; it is important for the regulation of cell Em, volume, pHi, and HCO3 transport (254). It has been shown that external Cl is necessary for sperm capacitation and fertilization (117, 555). Considering that [Cl]i increases during capacitation (249), it has been hypothesized that its influx is coupled to the activity of a Cl/HCO3 antiporter. HCO3 influx through this exchanger would stimulate sAC and increase the cAMP levels (555).

Why is the capacitation-associated hyperpolarization needed in spermatozoa? Initially it was proposed that the hyperpolarization was necessary to remove inactivation from CaV channels so they would be ready to open upon ZP stimulation to achieve the AR (12, 15, 324). As we have discussed earlier, the functional involvement of CaV channels in mature spermatozoa is now being debated (see sect. III). Alternatively, this hyperpolarization may be required to activate sNHE and increase pHi, which would in turn stimulate SLO3 and CatSper channels (381), and possibly sAC (541). A hyperpolarization would also increase the driving force for Ca2+ influx through other Ca2+-permeable channels such as TRPs, some of which are also stimulated by alkalinization.

An increase in [Ca2+]i seems to be important for capacitation (Fig. 6), a matter that deserves further study. This increase could result from 1) a fast stimulation of CatSper (566) and/or CaV3 channels caused by the exposure of spermatozoa to albumin (163). Serum albumin induces an immediate increase in the CaV3 currents of spermatogenic cells, and increases their Ca2+ window current by shifting the voltage dependence of both steady-state activation and inactivation (163). The increase could also result from 2) a slower stimulation of CatSper and other pHi-sensitive channels caused by the capacitation-associated alkalinization and 3) inhibition of the Ca2+-ATPase and/or the Na+/Ca2+ exchanger resulting from cholesterol removal and/or increases in cAMP (190).

Many important questions remain unanswered about the complex process of capacitation; some related to Ca2+ are 1) as spermatozoa from CatSper null mice undergo the normal changes in tyrosine phosphorylation and the AR, this channel seems unnecessary for capacitation, but needed for hyperactivation. If Ca2+ influx is needed for capacitation, other Ca2+ transporters must participate in this process and in the AR. What is their identity and modes of regulation? 2) Are any of the proteins important in Ca2+ transport during capacitation regulated by their phosphorylation state? 3) What is the interplay between Em, pHi, [Ca2+]i, and cAMP? 4) Do the lipid changes associated with capacitation regulate Ca2+ transport?

G. Other Aspects of Mammalian Sperm Motility

1. Chemotaxis

Although sperm chemotaxis is well established in marine animals (see sect. VI), its relevance in mammalian spermatozoa has been a matter of debate for a long time (284). Follicular fluid was found to attract capacitated human spermatozoa (95, 428, 533). The relatively low number of responding cells (∼10% or less) complicates the interpretation of the results. Mammalian spermatozoa have been shown to express olfactory receptors (401), although their physiological function is not fully understood. One of these human olfactory receptors, hOR17–4, was detected in human testicular cDNA, and bourgeonal, its agonist, is a chemoattractant for human spermatozoa (477). Accordingly, bourgeonal elevates human sperm [Ca2+]i, and extracellular Ca2+ is indispensable for both chemotaxis and the [Ca2+]i increase. Pharmacological evidence suggests that AC, but not PLC, is involved in the signaling cascade of bourgeonal (477). All membrane forms of ACs (mAC1–9) have been detected in human and mouse spermatozoa (39, 478). Furthermore, SQ22536 (ATP competitor of mAC) blocks the bourgeonal-induced [Ca2+]i increase and the human sperm motility responses (chemotaxis and an increase in flagellum beat frequency). These results are consistent with the involvement of the cAMP cascade in bourgeonal signaling in human spermatozoa. Ca2+ imaging in immobilized human spermatozoa revealed that the bourgeonal response initiates in the flagellar midpiece and travels to the head in seconds. Human spermatozoa display turns accompanied by asymmetric flagellar beating while swimming in the descending bourgeonal gradient, which resembles sea urchin sperm chemotaxis (478) (see sect. VI). The Ca2+ channels responsible for the Ca2+ influx induced by bourgeonal have not been determined, and the physiological relevance of bourgeonal is a matter of debate (22, 284). It is a perfume odorant apparently not present in the female reproductive tract. Therefore, bourgeonal may mimic the function of an unidentified natural agonist. Bourgeonal can attract 60% of the sperm population (478), while in another report, human follicular fluid attracts only 10% (95), suggesting that the signaling mechanisms are different (284). Alternatively, if both chemoattracts share the same mechanism, the natural agonist concentration in the follicular fluid is very low.

In mouse spermatozoa, lyral, another perfume odorant, was also reported to function as a chemoattractant (187). In WT animals, 30% of the spermatozoa responded to lyral by increasing the [Ca2+]i. This is consistent with the observation that 30% of the spermatids express the lyral receptor (MOR23). Spermatozoa from transgenic mice which overexpressed MOR23 respond better (up to 50%) to lyral than those of WT. As the lyral concentration used in the sperm chemotaxis assay was high (50 mM in capillary), distinguishing between real chemotaxis or trapping due to lyral toxicity must be considered.

2. Thermotaxis

The temperature of the female reproductive tract varies depending on the region and the sexual cycle (267). It has been speculated that these intriguing temperature differences could be caused by local variations in macromolecules, irrigation microfluidics, and morphology (25, 268). The distal isthmus temperature is significantly lower than that of the ampulla region in the pig (268). Furthermore, before ovulation the temperature in the lower isthmus of rabbits, the site of the sperm reservoir, is 0.8°C lower than that in the isthmic-ampullary junction, the site of the fertilization. Temperature differences between two regions can be up to 1.6°C after ovulation, mainly due to a temperature decrease in the sperm reservoir (25). Rabbit and human spermatozoa display thermotaxis, that is, they preferentially swim toward the warmer site (26). Studies on human sperm thermotaxis recently found unexpectedly that external Ca2+ is not essential for thermotaxis (24). In agreement with this result, blockers for many types of Ca2+ channels did not inhibit thermotaxis, namely, Ni2+ and nifedipine (CaV), ruthenium red (TRPV1), BCTC (TRPM8), and Gd3+ (SOCs). In contrast, PLC-dependent intracellular Ca2+ mobilization seems important for thermotaxis, since BAPTA-AM (intracellular Ca2+ chelator), U73122 (PLC inhibitor), and 2-APB (IP3R antagonist) inhibited the process. Recently, we identified TRPM8 channels in human spermatozoa and confirmed their functionality through recording [Ca2+]i increases induced by menthol and cooling, which were inhibited by specific TRPM8 antagonists BCTC and capsazepine (143). Although the TRPM8 temperature threshold for activation is lower than the temperature of the female genital tract, many parameters such as lipid compositions, particularly phosphoinositides (439), phosphorylation state (1), and the [Ca2+]i (439) can modify this threshold. Considering that the temperature of the sperm reservoir decreases after ovulation, TRPM8 channels could alternatively modulate sperm motility (i.e., hyperactivation).


A. Generalities

As illustrated in Figure 1, the acrosome, a single membrane-delimited specialized organelle, overlies the nucleus of many sperm species. The outer acrosomal membrane and the overlying plasma membrane fuse and vesiculate when spermatozoa are presented with physiological inducers from the female gamete, its vicinity, or to appropriate pharmacological stimuli. This unique, tightly regulated, irreversible, single exocytotic process releases the acrosomal contents, modifies membrane components, and exposes the inner acrosomal membrane to the extracellular medium (130, 131, 571). Sperm egg-coat penetration and fusion with the eggs' plasma membrane require the release and exposure of cell components resulting from this exocytotic process, named the acrosome reaction (AR). The egg envelope, an important element of the gamete recognition and signaling process, is also a protective layer (146, 549).

External and internal Ca2+ are essential for AR (130, 571), and a group of ion channel antagonists that block its flow also inhibit this process. These findings emphasize the preponderant role certain Ca2+-permeable channels play in AR (134, 178, 418). As in other secretory cell types, AR requires soluble N-ethylmaleimide-sensitive attachment protein receptors (SNAREs). The presence of sperm SNAREs has been documented in sea urchins (461, 462) and mammals (reviewed in Ref. 352). Significant advancement in our comprehension of how the fusion machinery operates during the sperm AR has been made utilizing permeabilized human spermatozoa. These studies propose a working model for regulated sperm acrosomal exocytosis. In resting spermatozoa, SNAREs are assembled in inactive complexes in the same membrane (cis-complexes). Sperm AR stimulation triggers external Ca2+ uptake leading to RAB3A activation and SNARE complex disassembly. Free SNAREs are then able to reassemble in trans-complexes where plasma membrane SNAREs interact with SNAREs in the outer acrosomal membrane. This process may be facilitated by synaptotagmin and complexin. At this stage, the trans-complexes are arrested and resistant to tetanus toxin but sensitive to botulinum neurotoxin B, until a local Ca2+ increase coming from the acrosome through IP3Rs activates the synaptotagmin-dependent relief of the complexin block, and acrosomal exocytosis is completed (86, 352).

The acrosome functions as a Ca2+ store whose contents are released during AR. The molecular identity of the physiological AR inductor(s); when, where, and how this reaction takes place; and the main actors participating in this indispensable process have been the subject of intense research for years, and a current matter of lively debate (i.e., Ref. 191).

The classical view, although presently under scrutiny, proposes that the zona pellucida (ZP), the extracellular coat surrounding the egg, is the physiological inductor of the mammalian sperm AR. ZP is composed of a matrix of four glycoproteins (ZP1-ZP4), except for mouse ZP that contains only three glycoproteins (367). The importance and site of ZP protein glycosylation is also under revision (191). ZP3 is considered the main physiological inductor of the mammalian sperm AR (177, 550) though in the human system ZP4 and ZP1 can induce this reaction to a lesser extent (76, 200, 282). In spite of these, recent experiments in which oocytes expressing genetically modified mouse ZPs are used show that very few spermatozoa bound to ZP have undergone AR after 1 h (28, 191). This puzzling observation reopens questions about the nature of the physiological inducer of AR and the site at which this process occurs. Examination of gamete interactions within the first minutes under physiological conditions using oocytes expressing genetically modified ZPs and labeled spermatozoa to detect AR are required to answer these fundamental questions. It is fortunate that genetically modified mouse spermatozoa that express GFP in their acrosome and red fluorescent protein in their mitochondria are now available. These spermatozoa fertilize normally in vivo and in vitro and can be observed exciting with light through the dissected uterine and oviductal walls as they migrate through the female reproductive tract, as well as during AR (244). On the other hand, as gamete interaction and the AR are species specific, it is believed spermatozoa possess specialized receptors for ZP (417). Although several ZP3 receptor candidates have been proposed (159), their physiological relevance is still unclear, since knockout mice for these candidates are fertile or subfertile (21, 470, 550). Evidence is accumulating that gamete interaction may involve multiple recognition sites (389, 527).

At the present time, Izumo is the only sperm protein known to be necessary for sperm-egg interaction. It is a member of the immunoglobulin superfamily; knocking it out results in sterile mice whose spermatozoa are not able to fuse with the egg plasma membrane (273). It is not known how Izumo intervenes in the fusion process. Recently it was reported that Izumo relocalizes from the anterior acrosome where it is found in intact mouse spermatozoa, to other regions in the head engaged in fusion with the egg plasma membrane. Tssk6, a member of the testis-specific serine kinase family, is necessary for the movement of Izumo, and its role is mediated by the actin cytoskeleton (362, 476).

Other cellular ligands can induce the AR such as progesterone (31, 55, 355, 506), γ-aminobutyric acid (GABA) (280, 355, 559), glycine (301), EGF (133), ATP (339, 442), acetylcholine (66), and sphingosine-1-phosphate (495). The role of many of these “alternative” agonists is not clear. For example, progesterone has been proposed to facilitate capacitation (35), promote sperm hyperactivation (490), induce chemotaxis (158), and finally potentiate the ZP-induced AR (440) and trigger AR itself (32). The [Ca2+]i rise and AR induced by progesterone are insensitive to pertussis toxin (PTX), suggesting different signaling mechanisms to those involved in the ZP-triggered AR (30, 375, 504).

Soluble ZP and/or ZP3 initiate a set of responses in capacitated spermatozoa that will lead to a sustained [Ca2+]i increase needed to achieve AR. Important among these is a pHi elevation (179), apparently regulated by G proteins (11). Gi, Gq, and Gs have been reported to be present in mammalian sperm (40, 539, 544). In mouse spermatozoa ZP can activate Gi (545) and PTX, a specific inactivator of Gi, inhibits the ZP-induced pHi change and the AR (15). PTX also inhibits AR and many of the ion fluxes associated with it in bovine and human spermatozoa (179, 315). These results indicate the importance of the pHi change induced by ZP for the AR. In addition, this process involves Em changes (15) that require further investigation.

Also associated with AR triggered by ZP and other inductors are increases in cAMP, protein phosphorylation, and phospholipid turnover (69, 178, 190, 441). The presence and activation of PLCs, PKC, and PKA during AR have been documented (30, 68, 185). As anticipated, the ZP and progesterone-induced Ca2+ permeability changes associated with AR are also influenced by PKC and PKA inhibitors in mammalian spermatozoa (67, 178, 180, 243). As described in the capacitation section (sect. IVF), protein tyrosine kinases are present in mammalian spermatozoa, and their antagonists can inhibit ZP (317, 420), progesterone (294, 338, 354, 429, 504), and thapsigargin-induced AR (156). Tyrosine kinases such as Src (detected in bovine spermatozoa) were shown to participate together with PKA and ouabain (a specific Na+-K+-ATPase inhibitor present in the female tract) in the transactivation of EGF receptor (EGFR) leading to [Ca2+]i increases and the AR (133). Tyrosine kinases and phosphatases have indeed been proposed as key elements for AR. Findings indicate that dephosphorylation of synaptotagmin by calcineurin must occur at a particular time during this process in human spermatozoa to achieve exocytosis (86). It is worth noting that syntaxin, synaptotagmin, and SNAP25 can associate with CaV2.1 and CaV2.2 channels. Therefore, both types of channels have been found as components of the exocytotic vesicle docking/fusion machinery (17, 89). Further studies are required to fully understand the interrelationship between [Ca2+]i and kinase/phosphatase activity during intermediate steps of the physiologically relevant AR that involve the fusion machinery.

B. Ca2+ and the Acrosome Reaction

A common and fundamental feature of physiological and pharmacological AR inducers is that they provoke intracellular multicomponent Ca2+ rises. At least three separate yet linked Ca2+ channels have been proposed to participate in these responses. Triggering of the sperm AR involves signaling bifurcations that regulate G protein activation, cAMP, and pHi. The complexity of the [Ca2+]i changes that trigger AR results from different levels of integration of the signaling events regulating these fluxes. Molecular identification of the channels involved in this process has been the subject of study of several research groups using diverse experimental approaches (105, 134, 178, 433).

The proposal that CaV channels participate in AR evolved from functional and pharmacological observations (reviewed in Refs. 136, 178) (Tables 1 and 2). For example, the spermatozoa of many mammalian species undergo [Ca2+]i increases sensitive to CaV channel blockers when depolarized by elevating [K+]e. These changes depend on external pH and Ca2+ and can lead to AR (15, 23, 63, 176, 327).

C. Initial Transient Ca2+ Rise

When exposed to solubilized ZP3, individual mouse spermatozoa undergo an initial fast (∼200 ms) and transient [Ca2+]i elevation compatible with the kinetic and pharmacological properties of CaV3 channels. Thereafter, a significantly slower and sustained [Ca2+]i increase occurs lasting up to minutes (12). As mentioned earlier, patch-clamp studies in mouse spermatogenic cells revealed mainly CaV3 currents, and all isoforms were reported to be present (11, 324) (Table 1). To distinguish the CaV3 family member responsible for these currents, two of three isoforms were knocked out. CaV3.1 knock-out mice are fertile, and their spermatogenic cells display CaV3 currents similar to those of WT animals (479), suggesting that CaV3.2 channels are responsible for the currents. This finding is consistent with the Ni2+ blocking profile of the CaV3 currents in spermatogenic cells and their location in the head area of spermatozoa (514). Surprisingly, CaV3.2 null male mice are also fertile even though their spermatogenic cells basically do not display CaV3 currents (114, 160, 479). To discard compensatory processes, it will be necessary to use the double (CaV3.1, CaV3.2) knockout mice.

CaV3.3 has not been knocked out; however, the slower kinetics of its currents (406) makes it a less likely candidate to produce the CaV3 currents observed in mouse spermatogenic cells. In addition, CaV3.3 channels immunolocalized to the flagella, whereas CaV3.1 and CaV3.2 are found at the sperm head, the site of AR (514). Additionally, CaV3 currents in spermatogenic cells display sensitivity to compounds such as urocortin, fenvalerate, and gossypol that may explain their contraceptive properties (27, 502, 567). It is noteworthy that none of the individual CaV channels that have been knocked out generated an infertile mouse phenotype (114, 291, 445) (see Table 1).

CaV3 channels, including those in spermatogenic cells and testicular sperm, inactivate at potentials more positive than −60 mV (11, 137, 407). Therefore, in cells such as noncapacitated mouse spermatozoa, in which the resting potential is more positive than −60 mV, removal of CaV3 inactivation requires hyperpolarization of Em. Taking this into account, it was proposed that the capacitation-associated hyperpolarization in mouse spermatozoa (see sect. IVF) is important to render CaV3 channels ready to open during AR (12, 138). Still, once inactivation is removed, CaV3 channel opening requires a depolarization. The section on capacitation presents a discussion of possible alternative purposes of the hyperpolarization that occurs in mouse spermatozoa during this process.

If CaV3 channels participate in AR, how does ZP induce a depolarization in capacitated mouse spermatozoa to open them? A nonselective, cation channel could mediate this depolarization (15). Poorly selective mammalian sperm cation channels registered in planar bilayers (111, 305) and in patch-clamp studies (162) are candidates for causing this depolarization. Additionally, as mouse and human spermatozoa have a Cl equilibrium potential of approximately −30 mV (209, 248), they would depolarize upon Cl channel opening. These cells possess the niflumic acid-sensitive anion channels documented in patch-clamp studies (162). Furthermore, ionotropic GABA and glycine receptors have been detected in mammalian spermatozoa, and their ligands can induce the AR. Strychnine, an antagonist of the glycine receptor, inhibits the AR induced by recombinant human ZP3 (65), and a null mutant of the glycine receptor is subfertile and cannot undergo ZP-induced AR (453). These findings suggest that the opening of Cl channels could depolarize capacitated spermatozoa to open CaV channels and trigger AR, a possibility worth further exploration.

Altogether, molecular and pharmacological evidence supports the role of CaV channels during AR, particularly in mouse spermatozoa (Tables 1 and 2). However, although now it is possible to perform patch-clamp recordings in testicular and epididymal mouse spermatozoa as well as in ejaculated human spermatozoa, CaV currents have only been detected in testicular sperm, raising the question as to whether these channels are functional in epididymal and ejaculated sperm. It is surprising that of all the ion channels whose message and proteins described as being present in mouse spermatozoa, only CatSper and SLO3 currents, present in the principal piece, have been detected by patch clamping the cytoplasmic droplet of epididymal spermatozoa. Furthermore, from all mice lacking ion channels, only mice null for CatSper and Slo3 are infertile. However, it is reasonable to speculate whether, under these conditions, there are modes of regulation in spermatozoa that functionally silence the activity of these channels. As discussed, CaV currents disappear from testicular to epididymal spermatozoa, and in addition, neither CNG channels nor Kirs have been detected in the latter. Is this due to potential technical problems, such as difficulties in clamping the voltage throughout this morphological complex cell and/or a relative scarcity of these channels? It must be stressed that capacitation involves membrane reorganization and a complex set of phosphorylation changes that might unmask the activity of CaV and other channels in preparation for a role during AR, whereas novel modes of CaV3 channel regulation (see sect. IIIA) as well as possible CaV3 channel interactions with as yet undefined partners may inhibit the activity of these channels during patch-clamp experiments in epididymal spermatozoa prior to capacitation. For instance, as indicated earlier γ6, a CaV auxiliary subunit, inhibits CaV3 channels (116). Although it is not known if this subunit is present in spermatozoa, it is tempting to speculate that inhibition by γ6 could maintain CaV3 channels silent in epididymal spermatozoa, and its removal during capacitation would reactivate them.

In spite of these arguments, the observation that spermatozoa from CaV3.2 null mice basically display no residual CaV3 currents, yet remain fertile, is difficult to reconcile with their participation in AR. Unless spermatogenesis is particularly rich and versatile in its ability to compensate for missing CaV channels and for their expression time, the role of these channels in fertilization is open to question. In spite of this, there is evidence suggesting that CaV3 channels may participate in sea urchin sperm chemotaxis (284, 563).

CaV channels are also present in human spermatozoa, but their role during AR is even more controversial, due in part to the lack of native ZP for experimentation (reviewed in Ref. 134). There is indirect evidence of their activity in human spermatozoa (47, 56, 276, 400, 418, 466). Fluorescence determinations simultaneously measuring [Ca2+]i and Em showed that the addition of K+ in the presence of valinomycin depolarizes human sperm and induces [Ca2+]i increases; however, these [Ca2+]i increases were insensitive to classical CaV blockers (nifedipine and verapamil) but sensitive to Ni2+ (327).

Recently, recombinant proteins as well as native ZP isolated from human oocytes recovered after in vitro fertilization procedures have helped to define better the participation of CaV channels in AR. Pharmacological studies measuring AR induced with either native ZP or recombinant ZP3 also suggest the presence and participation of CaV channels during human AR (consult Table 2 and references therein).

Progesterone at concentrations found in female genital fluids (μM) influence several aspects of sperm behavior (see sect. IVE), AR among them (reviewed in Ref. 32). The nature of the channels involved in the progesterone-induced increase in [Ca2+]i is not clear, although it may involve a combination of CaVs, SOCs, and intracellular channels in an intertwined network of signaling cascades (233, 327, 418, 505, 512). Single-cell Ca2+ imaging experiments show that almost all cells respond to progesterone with a biphasic Ca2+ increase; however, only ∼20% undergo AR, indicating that other mechanisms must be activated concomitantly to achieve exocytosis (32). For instance, Src inhibitors decrease the Ca2+ rise and the AR induced by progesterone (528). The proposal is that Src is activated by progesterone upstream of the Ca2+ increase and that it may be involved in Ca2+ release from intracellular stores, such as the RNE, considering its postacrosomal localization. These results are consistent with the inhibition of the progesterone-induced [Ca2+]i plateau phase caused by general tyrosine kinase antagonists (59). Src is known to regulate [Ca2+]i in different cell types either by affecting Ca2+ release from intracellular stores or by phosphorylation of CaV channels (528).

D. Sustained Ca2+ Rise

The initial and transient Ca2+ elevation triggered by solubilized ZP or ZP3 is followed by a slow and prolonged Ca2+ elevation that can last several minutes. Since the early days of studying sperm Ca2+ transport, it was known that a high and sustained Ca2+ influx is necessary for the AR to occur (11, 15, 179, 306, 454, 456, 571). How is this second Ca2+ elevation generated? The current view is that the initial and transient Ca2+ rise activates a Ca2+-sensitive PLC-δ (185) that hydrolyzes PIP2 to produce DAG and IP3 (539). The fact that PLC-δ4 null mice are infertile and unable to undergo the ZP3-induced AR supports this model (185). Indeed, it has been shown that ZP/ZP3 induces IP3 increases in mouse spermatozoa (510). Binding of IP3 to its receptor (located in the acrosome and the RNE, Fig. 1 and Table 3) releases Ca2+ from one or both stores (142, 251, 260, 391, 539), which in turn, as in somatic cells, would induce the opening of Ca2+ channels at the plasma membrane (SOC activity). As mentioned before, several TRPCs are present in spermatozoa, and in particular TRPC2 has been implicated in mouse AR (283). Considering that TRPC2 null mice are fertile (484), that human TRPC2 is a pseudogene (575), and that the molecular composition of the SOC machinery seems to be more complex than previously thought, it is likely that other components such as STIM, ORAI, and other TRPCs may be also involved in the sustained Ca2+ entrance. This contention is supported by the reported presence of STIM1 in human sperm (105) (A. Sánchez-Tusie, unpublished results).

Additionally, the possible regulation of the Ca2+ channels responsible for the sustained increase in [Ca2+]i needed for AR by proteins like homer (350), enkurin (498), junctate (480), PKDREJ (497) etc., and mechanisms that affect their activity (PIP2 levels, phosphorylation, pH, etc.) clearly require more attention.

Figure 7 summarizes a working signaling model of the components and events leading to the mammalian sperm AR. Solubilized ZP3 induces a fast and transient increase in [Ca2+]i that could involve CaV channels, although other Ca2+-permeable channels may be responsible for this influx. After IP3 production by PLC-δ, Ca2+ release from IP3-sensitive intracellular stores would activate SOCs responsible for the sustained [Ca2+]i elevation. The figure also summarizes our view of parallel and concomitant steps necessary to achieve AR that occur thereafter. This signaling cascade involves preparation of the fusion machinery that includes SNAREs, α-SNAP/NSF, synaptotagmin, Rab3A, and recently incorporated, exchange protein directly activated by cAMP (Epac) (64), a GEF protein that links cAMP and Ca2+ signaling (reviewed in Ref. 352). It has been proposed that a sAC is activated by Ca2+ during the sustained ZP3-induced Ca2+ rise, that the resulting cAMP activates Epac and that the signaling cascade branches in at least two arms that come back together at the end of the cascade. Dissecting the roles of sAC and mACs during AR requires further work. One branch prepares the tethering of the SNAREs fusion machinery involving Rab3A activation, and the other one activates PLC-ε via Rap stimulation, to again produce IP3 and release Ca2+ from internal stores, the final trigger for membrane fusion (64). The existence and regulation of two waves of IP3 production and Ca2+ release from internal stores require further investigation.

Figure 7.

Calcium mobilization elicited upon ZP3 binding to sperm receptor(s) during the mammalian acrosome reaction. ZP3 binding activates a Gi protein likely involved in the pHi increase, and may also activate CaV channels inducing a membrane depolarization (caused by GABA or an unidentified cation channel). This initial and transient Ca2+ rise stimulates PLC-δ that catabolizes PIP2 to produce IP3 and DAG. IP3 binds to its receptor located in the acrosome and/or in the RNE, liberating stored Ca2+ and activating SOCs (TRPCs and/or ORAI), possibly through STIM aggregation. DAG and PIP2 may also stimulate TRPCs channels. The resulting sustained Ca2+ increase activates sAC, which elevates cAMP turning EPAC on. EPAC signaling branches in two arms: 1) activation of Rab3A, triggering the tethering of the acrosomal and plasma membranes through the stimulation of RIM and αSNAP/NSF to render the SNARE machinery ready for fusion, and 2) activation of Rap, which in turn stimulates PLC-ε to generate IP3 and liberate Ca2+, the final trigger for membrane fusion. Low and high Ca2+ levels are maintained in the cytosol (PMCA) and Ca2+ stores (SPCA1 and SERCA) by ATP-driven Ca2+ pumps.

E. CatSper Channels and the Acrosome Reaction

CaV3 currents have not been recorded from mouse epididymal spermatozoa (433), and both the CaV3.1 and CaV3.2 null mice are fertile (122, 479). These findings have led to the consideration of the participation of CatSper channels in the solubilized ZP induced elevation of [Ca2+]i (566). Ren and Xia (433) reported that solubilized ZP initiated an early [Ca2+]i elevation, mostly within 20 s, in the principal piece of the sperm flagella, where CatSper proteins are present, that propagated to the sperm head within a few seconds. Around 35% of the cells exposed to ZP displayed an additional delayed response separated from the early one. Spermatozoa from CatSper1 null mice lose the early [Ca2+]i increase (within 2 min) induced by solubilized ZP. This response was rescued by spermatozoa from mice expressing a functional GFP tagged CatSper1. These results indicate that solubilized ZP activates CatSper channels provoking the early [Ca2+]i increase. It is worth noting that the time resolution employed in this work (2 frame/s) would not allow reliable detection of the fast ZP-induced opening of CaV3 channels (∼100 ms) reported (11). Interestingly, spermatozoa from CatSper null mice still undergo the solubilized ZP-induced AR as well as those from WT mice, and 20% of them display the delayed [Ca2+]i elevation. These findings suggest that the early [Ca2+]i increase induced by solubilized ZP mediated by CatSper1 is not essential for AR (433). Why then is AR inhibited by CaV3 channel blockers? It would be informative to test CaV blockers (i.e., nimodipine, mibrefadil, etc.) to corroborate the pharmacological profile of the AR and the participation of CaV channels in CatSper null spermatozoa.

It is also notable that spermatozoa from CatSper null mice undergo the normal tyrosine phosphorylation associated with capacitation and can fertilize zona-free eggs (83, 426, 432, 565). Furthermore, the [Ca2+]i rise induced by solubilized ZP (or progesterone and thapsigargin; Ref. 355) has been reported to commence in the mammalian sperm head (175, 185, 186, 468) and not in the principal piece as reported by Ren and Xia (433). The correlation with high spatiotemporal resolution [Ca2+]i changes and AR in single spermatozoa is important to understand this fundamental process required for fertilization. Now it is possible to image the acrosome status using a transgenic mouse which expresses GFP in the acrosome (379) or fluorescence probes (186, 240, 509).


In contrast to terrestrial animals, most marine animals release their gametes into the ocean where unlimited dilution will increase the distance separating them. In these species, successful fertilization against all odds requires efficient sperm chemotaxis towards the egg. To improve the probability of gamete interaction, female gametes of many marine animals have evolved to exploit diffusible molecules that activate and attract their homologous spermatozoa (360). Sperm chemotaxis is a fascinating biological phenomenon in which spermatozoa rapidly and precisely recognize a chemoattractant gradient and regulate their swimming direction toward the source of the chemoattractant (227). It is known that Ca2+ plays fundamental roles in this process without exception (284).

A. Sea Urchin Spermatozoa

Although the first molecular identification of a sperm chemoattractant was achieved in the 1880s in the plant kingdom (bracken ferns) (410), it is in sea urchins where our understanding of the mechanisms underlying sperm chemotaxis has advanced furthest (98, 135, 227, 284). Sea urchin eggs are covered with an extracellular matrix, called egg jelly (EJ), mainly composed of polysaccharides, which function as species-specific ligands to induce the AR (see sect. VII). The EJ also contains small diffusible peptides, which activate sperm motility in slightly acidified seawater (392). These peptides, collectively named sperm-activating peptides (SAPs) (499), induce rapid physiological changes in sperm cyclic nucleotide levels, Em, pHi, and [Ca2+]i (135, 284). Sperm chemotaxis in response to a ligand was first documented in the animal kingdom using the purified ligand resact (CVTGAPGCVGGGRL-NH2), a SAP derived from Arbacia punctulata (546). Conversely, speract (GFDLNGGGVG) was the first purified and structurally identified SAP from Hemicentrotus pulcherrimus (499) and Strongylocentrotus purpuratus (237).

As a consequence of intensive efforts in time-resolved fluorometry (286, 351, 387, 388, 485), Ca2+ imaging (58, 227, 561563), and molecular identification (60, 195, 211, 223, 486, 487), our knowledge about the signaling cascade triggered by SAPs has advanced significantly in the last 10 years. Figure 8 depicts a proposed general scheme for SAP signaling. SAPs may bind directly to guanylyl cyclase [i.e., resact (471) and asterosap (351)] or to its associated receptor [i.e., speract (132)] in the flagellar plasma membrane stimulating the synthesis of cGMP (286). The elevation of cGMP induces K+ efflux by activating a cyclic nucleotide-gated K+-selective channel (tetraKCNG/CNGK channel) (60, 194, 195). The tetraKCNG/CNGK channel is a new type of K+ channel composed of a single polypeptide that contains 24 TM segments (4 subunits of 6 TM segments). Recently, it was demonstrated that the KCNG channel of A. punctulata spermatozoa is activated when cGMP binds to the third cyclic nucleotide binding domain (CNBD), with high affinity (K1/2: 26 nM for cGMP and 17 μM for cAMP) (60). The tetra-KCNG/CNGK channel does not have an intrinsic inactivation mechanism; therefore, its activation will depend on cGMP levels. The SAP-induced hyperpolarization affects several important Em-dependent membrane proteins such as CaV channels (removal of inactivation) (223, 485), hyperpolarization-activated cyclic nucleotide-gated (HCN) channels (Na+ influx) (197, 211), Na+/H+ exchangers (NHE) (pHi increase) (437), and K+-dependent Na+/Ca2+ exchangers (NCKX) (Ca2+ extrusion) (487). Elevation of pHi together with the hyperpolarization activates AC and increases cAMP level (45), which potentiates Na+ influx (211). The pHi increase also inactivates the guanylyl cyclase through dephosphorylation of its catalytic domain (547) and enhances the activity of phosphodiesterase-5 (486), which is activated by cGMP itself (444) and also by phosphorylation mediated by PKA upon EJ treatment (486).

Figure 8.

Signaling events regulating [Ca2+]i and sea urchin sperm swimming. A: after binding to its receptor, speract stimulates a membrane guanylate cyclase (GC) and elevates cGMP that activates tetrameric cGMP-regulated K+ channels (tetraKCNG) leading to membrane potential (Em) hyperpolarization. This Em change stimulates hyperpolarization-activated and cyclic nucleotide-gated channels (HCN), removes inactivation from voltage-activated Ca2+ channels (Cav), facilitates Ca2+ extrusion by K+-dependent Na+/Ca2+ exchangers (NCKX), and activates Na+/H+ exchangers (NHE) and adenylate cyclase (AC) activity. HCN opening and Na+ influx contribute to Em depolarization, and increases in [Ca2+]i and Na+ further depolarize Em. Elevated [Ca2+]i enhances flagellar bending, leading the spermatozoon to turn. Possibly, the [Ca2+]i increase also opens Ca2+-regulated Cl channels (CaCC) and/or Ca2+-regulated K+ channels (CaKC), which then contribute to hyperpolarize again the Em, removing inactivation from Cav channels and opening HCN channels. This mechanism is then cyclically repeated to generate a train of Ca2+ increases that produce a sequence of turns. The sequence continues until one or more of the molecular components in the pathway are downregulated. cAMP activates a poorly characterized Ca2+ influx pathway, which may contribute to a tonic [Ca2+]i increase. B: a spermatozoon (green) swimming in a circular trajectory in a chemoattractant gradient (gray background) is periodically stimulated due to changes in the rate of chemoattractant capture. Top panel: Chemotactic behavior of a single L. pictus sperm. When swimming in an ascending gradient, the onset of [Ca2+]i fluctuations is suppressed until the spermatozoon senses an ascending to descending gradient inversion (red circle). After a ∼200-ms delay, the spermatozoon undergoes a [Ca2+]i fluctuation just before reaching the gradient minima (white circles, fluctuations from many experiments). Consequently, the spermatozoon executes a turn-and-run episode that brings it closer to the chemoattractant source (black arrow). Bottom panel: nonchemotactic behavior of a single S. purpuratus sperm. The onset of [Ca2+]i fluctuations is not suppressed by an ascending chemoattractant gradient. They occur at random; therefore, the turn-and-run episodes promote sperm relocalization but not a directed approach to the chemoattractant source. C: model of chemotaxis in sea urchin spermatozoa. A chemotactic spermatozoon swimming in a chemoattractant gradient (background) undergoing cyclic changes in Em from resting to a hyperpolarized state (Hyp, green shadow) and then to a depolarized state (Dep, blue shadow) that control Cav activity. The spermatozoon path is depicted as a black arrow as it swims in a chemoattractant gradient (background). The incremental activation of speract receptors in spermatozoa experiencing an ascending gradient leads to extended hyperpolarization which suppresses Ca2+ fluctuations (i). The hyperpolarization reverses once sperm enter a negative speract gradient, which after ∼200-ms delay (red line) generates a chemotactic turn that reorients the spermatozoon towards the gradient source (ii). The onset of each [Ca2+]i fluctuation is denoted by the black circles. Repolarization due to GC inactivation, decrease in cGMP levels by degradation, Na+ influx through HCN channels, and other unknown depolarizing elements open Cav channels, possibly of the T and L type (ii). The black (low) to red (high) pseudo-color shade around the trajectory illustrates the [Ca2+]i changes in the flagella. Note that straight swimming coincides with elevated [Ca2+]i. At some point during the straighter swimming phase in the positive speract gradient, a hyperpolarized Em is reestablished and extended by continuous speract receptor recruitment (iii), which once again reverts to depolarized Em as sperm leave the positive gradient (iv). This sets up a sequence of chemotactic turns, triggered by hyperpolarization/depolarization cycles that serve as the primary translators of the state of the extracellular chemoattractant gradient.

After hyperpolarizing, Em is displaced towards the resting level as HCN channels open and possibly tetra-KCNG/CNGK channels close. This Em change initially activates CaV3 channels and then possibly CaV1 and CaV2 channels. Transcripts for CaV3.3, CaV1.2, and CaV2.3 were obtained from S. purpuratus testis (223, 485) (Table. 1). Immunostaining indicated that CaV1.2 channels are localized on the sperm flagellum (223) (Table 1). Initially CaV channels were not considered important for SAP-induced Ca2+ changes, mainly because blocking them did not inhibit the speract-induced Ca2+ increase measured in sperm populations by fluorometry (455). Thereafter, Ca2+ imaging of individual immobilized S. purpuratus spermatozoa revealed that speract induced rapid [Ca2+]i fluctuations superimposed with a tonic [Ca2+]i increase originating from the flagellum, both being dependent on external Ca2+ (561). CaV channel blockers, such as Ni2+ and nimodipine, inhibited only the sperm [Ca2+]i fluctuations, but not the tonic [Ca2+]i increase (561), explaining the previous observations. Time-resolved fluorometry using stopped-flow or caged compounds showed a significant time delay (150–600 ms) for the [Ca2+]i increase after spermatozoa were stimulated by SAPs or caged cGMP (286, 351, 387, 388, 503). This time lag well corresponds to the time required to achieve the hyperpolarization induced by SAPs or cGMP (485) that removes inactivation followed by activation of the CaV3 channels. This delay for the [Ca2+]i increase must be fundamental for the mechanism of sperm chemotaxis as described later. It is reasonable to consider that the [Ca2+]i fluctuations induced by speract (561) are caused by regulated hyperpolarization-depolarization Em cycles. Technically challenging single sea urchin sperm Em measurements are needed to corroborate this explanation. The current hypothesis is that Ca2+-dependent Cl and K+ channels might be involved in the rehyperpolarization of the sperm Em after Ca2+ influx through CaV channels. These elements could generate Em oscillations without requiring cGMP fluctuations, which anyway are likely to occur by repetitive stimuli with SAPs during sperm chemotaxis under experimental conditions where spermatozoa swim in two dimensions in circles close to the surface of the coverslip (58, 227).

Although the mechanism responsible for the tonic Ca2+ influx is unknown, it has been proposed that cAMP regulates it (97, 387). As all CatSper-related genes were found in the sea urchin genome (see sect. IVD), this channel, which is activated by the pHi and cAMP increases induced by speract, is a good candidate. Further experiments are required to advance our understanding of the SAP signaling cascade.

Time-resolved Ca2+ imaging using caged cGMP revealed that only the Ca2+ transient sensitive to Ni2+ and nimodipine induces asymmetric flagellar bending that results in a highly curved trajectory altering the sperm swimming direction; the remaining tonic [Ca2+]i increase does not (562). Furthermore, detailed analysis of the relationship between flagellar [Ca2+]i and the degree of its asymmetry revealed a positive correlation only during the very early phase of the [Ca2+]i increase, but not for the entire process (135, 563). In other words, after a short asymmetric period, the flagellum bend returns to a less asymmetric form while the [Ca2+]i is still elevated. Therefore, the observations indicate that a proportional relationship between [Ca2+]i and the degree of flagellar asymmetry does not hold in intact sea urchin spermatozoa. What determines the return to less asymmetric flagellar form despite continuing high [Ca2+]i levels is a new and open question.

In spite of the fact that sea urchin SAPs induce very similar physiological responses in spermatozoa from their corresponding species, chemotaxis had been demonstrated only in A. punctulata spermatozoa towards resact (286, 546). Although speract redirects the swimming paths of S. purpuratus spermatozoa with a stereotypical turn-and-run pattern (Fig. 8), under laboratory conditions it has not been shown to be a chemoattractant (reviewed in Ref. 135). Recently our group discovered that speract is chemotactic towards L. pictus spermatozoa (227). Notably, we now have been able to characterize spermatozoa of two phylogenetically closely related sea urchins that react to speract gradients with turn-and-run type motility responses, yet only one of these species is chemotactic. A detailed analysis was carried out for individual sperm trajectories and flagellar [Ca2+]i in response to a chemoattractant gradient generated by photoactivating caged speract (503). The most striking difference between the two species was that L. pictus spermatozoa selectively undergo Ca2+ fluctuations and turn when swimming in descending speract gradients, close to where the gradient changes polarity, while S. purpuratus spermatozoa generate Ca2+ fluctuations and turn without distinguishing the direction of the chemoattractant gradient. This different behavior emanates from the ability of L. pictus spermatozoa to selectively suppress their Ca2+ fluctuations when swimming toward the source of the chemoattractant. A central feature of sea urchin sperm chemotaxis, and possibly of sperm chemotaxis in general, is tuning of the Ca2+ fluctuations and the associated turning episodes in the direction of the chemoattractant gradient. The physiological significance of the distinct behavior of spermatozoa from these two sea urchin species may stem from differences in their respective habitats such as varying hydrodynamic environments and/or in their reproductive cycles, animal density, and gamete spawning synchronization (227).

So far chemotaxis has been mostly studied in spermatozoa swimming in two dimensions, that is, spermatozoa undergoing thigmotaxis on the coverslip surfce (104). Under physiological conditions, sea urchin spermatozoa swim in three dimensions describing helical trajectories with geometric characteristics distinct from those they display in two dimensions (102). Spermatozoa swim 25% slower in two dimensions, and their circular trajectories have ∼130% larger radius of curvature than those swimming in three dimensions. These results point out the necessity of characterizing the chemotactic response of spermatozoa in three dimensions.

As usual, novel findings generate new questions, and this is the case for sea urchin sperm chemotaxis. 1) Does sperm Em oscillate during the [Ca2+]i fluctuations triggered by SAPs? 2) What is a molecular identity of the Ca2+ channels that contribute to the tonic Ca2+ influx induced by SAPs? 3) What determines the symmetric flagellar form at high [Ca2+]i during sperm chemotaxis? 4) What is the molecular explanation for the behavioral difference between S. purpuratus and L. pictus? 5) How do spermatozoa respond to SAPs in three dimensions?

B. Sea Squirt (Ascidians) Spermatozoa

Spermatozoa of the sea squirt Ciona intestinalis display clear chemotaxis in response to gradients of a sulfated steroid named SAAF (sperm-activating and attracting factor), which is released only from unfertilized mature oocytes (296, 576, 578). Since the SAAF receptor has not been identified, the signaling mechanism is not well characterized. It is known that an Em hyperpolarization is an important initial step (275) and the tetraKCNG/CNGK channel has been identified in the C. intestinalis genome (60), suggesting that the initial signaling steps might be similar to those of sea urchin spermatozoa. Also, the general sperm behavior during chemotaxis is almost identical to that described in the sea urchin; that is, spermatozoa change swimming direction upon facing a descending SAAF gradient which triggers a Ca2+ spike that generates a chemotactic turn (465). The degree of flagellar asymmetry and the [Ca2+]i are positively correlated only at the initial phase (465), as observed in sea urchin spermatozoa. However, CaV blockers do not affect chemotaxis; instead, SKF96365, a general TRP channel blocker (see sect. IIIC), inhibits chemotaxis (577). Interestingly, the first enzyme containing a functional voltage sensor domain besides voltage-gated channels, a voltage-sensitive phosphatase that hydrolyzes plasma membrane PIP2, was discovered in C. intestinalis (376). This phosphatase, named Ci-VSP, is localized in the sperm tail plasma membrane. Considering that PIP2 is a crucial modulator of TRPs and other channels (494) and Ci-VSP hydrolyzes PIP2 in a voltage-dependent manner (376), possibly SAAF may utilize a TRP channel combined with Ci-VSP instead of voltage-gated channels. Further studies are required to confirm this hypothesis. C. intestinalis is an attractive animal model to study sperm chemotaxis because it can be genetically manipulated (452).


A. Generalities

The sea urchin is a marine organism that has been a preferred model to study fertilization and AR. Jean C. Dan originally discovered the AR using sea urchin gametes in 1952 (131). The AR in sea urchin spermatozoa results in the exocytosis of the acrosomal vesicle and the pHi-dependent polymerization of actin which leads to the formation of the acrosomal tubule (see Fig. 9) (508). This exocytotic process exposes membrane covered by bindin that will fuse with the egg plasma membrane (43, 134, 383). External Ca2+ and its uptake are strictly required for sperm AR. After signaling initiation upon contact with the egg vestments, there is a complex and sustained influx of Ca2+ required for AR (130, 226, 456). This reaction is triggered species-specifically by the EJ, sulfated polysaccharides that surround the oocyte (255, 532). The EJ component that induces AR in sea urchin spermatozoa is a fucose-sulfated polymer or FSP that interacts with a sperm plasma membrane receptor (suREJ1). The characteristics of the EJ receptor and its homologs are discussed in detail below.

Figure 9.

Hypothetical signaling pathway of the AR in sea urchin spermatozoa. The fucose sulfate polymer from egg jelly (FSP) binds to its receptor (suREJ1) transiently activating, by unknown mechanisms, a TEA+-sensitive K+ channel (K+C). Activation of the K+ channel hyperpolarizes spermatozoa removing CaV inactivation and stimulating voltage- and Ca2+-dependent Na+/H+ exchange increasing pHi and CaV. A suREJ3/PC2 complex could participate in Ca2+ uptake and/or in the depolarization needed to open CaVs. The transient [Ca2+]i elevation may stimulate in parallel PLC-δ to produce IP3 that would bind to the IP3R, and possibly an ADP-ribosyl cyclase (ARC) increasing NAADP levels that would activate its receptor (two-pore channel, TPC; Ref. 572). The S. purpuratus genome contains the predicted sequences for three TPC isofoms (TPC1–3; Ref. 63). SOCs would then activate (SOC) allowing a sustained [Ca2+]i increase fundamental to trigger the AR. As CatSper channels are present in sea urchin sperm, the pHi increase would activate them. They may contribute to the [Ca2+]i increases associated with the AR in an unknown manner. cAMP elevation (produced by sAC and mACs) may be needed to stimulate EPAC, to induce fusion that involves SNAREs as in mammalian sperm (344, 85), and is likely to regulate different channels. DIDS-sensitive Cl channels (ClC) possibly regulate osmotic changes necessary for the AR.

Binding of EJ or FSP to S. purpuratus sea urchin spermatozoa induces Ca2+ and Na+ influx, and K+ and H+ efflux (202, 203, 208, 454, 456). These fluxes result in Em changes (215, 218) and elevations of [Ca2+]i (224, 225, 257), pHi (225, 312), [Na+]i (437), cAMP (202, 204), and IP3 (153). These physiological AR agonists also stimulate the activity of PKA (207, 415), phospholipase D (154), and nitric oxide synthase (302). In mammalian spermatozoa, PLC-δ4 is required for the ZP- and progesterone-induced AR (186). PLC-δ has been cloned (39% identity with bovine PLCδ2) and identified by Western blot analysis (80 kDa) in S. purpuratus (suPLCδ) sperm extracts with an anti-PLCδ2 antibody against the mammalian isoform (106). Although PLC-δ regulation is not well understood, PLC-δ1 (the most similar to PLC-δ2) is regulated by Ca2+, the GTP-binding proteins Rho and Gh, as well as by the levels of PIP2 and IP3 (430). In sea urchin spermatozoa, immunofluorescence and immunogold experiments showed that Rho localizes in the acrosomal region, the middle piece of the head, and in the flagellum. Rho was also found in a preparation enriched in acrosomal and plasma membranes colocalizing (in a continuous density gradient) with bindin, the adhesive protein characteristic of the acrosome (526). Altogether these data suggest that the initial fast, transient [Ca2+]i increase triggered by FSP binding to the sea urchin spermatozoon activates suPLC-δ to trigger AR.

Very recently it was shown that binding of FSP to sea urchin spermatozoa also increases NAADP levels (530) and activates both sAC and mAC (44). Although both types of ACs are involved in the sea urchin sperm AR, the contribution of the sAC is more important (44).

In L. pictus spermatozoa, binding of EJ or FSP within seconds triggers a transient K+-dependent hyperpolarization, followed by a depolarization. This hyperpolarization precedes and may lead to the activation of a Ca2+-dependent Na+/H+ exchange that in turn increases pHi (218). In these cells, as in S. purpuratus spermatozoa, [Ca2+]i and pHi increases, and indirectly plasma membrane hyperpolarization, elevate cAMP levels (45) required for diverse and not fully understood signaling events. Tetraethylammonium (TEA) or an increase in [K+]e (30–40 mM) completely inhibit the initial Em changes, and the [Ca2+]i, pHi, and cAMP increases as well as the AR (215, 454). TEA-sensitive K+ channels (218, 304, 323) have been recorded in planar bilayers with incorporated S. purpuratus sperm plasma membranes. These findings clearly indicate that K+ channels are essential for the AR to occur. It is remarkable that little is known about their molecular identity.

Planar bilayer reassembly of sea urchin sperm plasma membrane also revealed the presence of anion channels sensitive to DIDS. Interestingly, DIDS inhibits the sea urchin sperm AR, which suggests that anion channels and transporters may participate in this important exocytotic process (371). In addition, a cAMP-activated cationic channel sensitive to Ba2+ and two other cationic channels of larger conductance were detected in planar bilayers with incorporated hybrid vesicles shed by the fusion of plasma and acrosomal membranes during AR (460).

B. Acrosome Reaction-Associated [Ca2+]i Changes

Sea urchin sperm population experiments with Ca2+-sensitive fluorescent dyes have shown that the FSP-induced AR is accompanied by Ca2+ uptake mediated by at least two different Ca2+ channels. The initial [Ca2+]i increase is fast, transient, and sensitive to the CaV channel blockers verapamil and DHPs (224, 437) that also inhibit the AR (454). Approximately five seconds later another Ca2+ channel (permeable to Mn2+ and Na+, insensitive to DHPs, and pHi dependent) opens, sustaining an elevated [Ca2+]i for several minutes and allowing sperm AR (225, 437). It is worth pointing out that inhibition of the first channel blocks the second, suggesting that their activation is linked. In addition, opening of the second channel and the AR are inhibited if the FSP-induced pHi increase associated with AR is prevented, for instance, by elevating external K+ to 50 mM. It was reported that a lower molecular mass (∼60 kDa) hydrolyzed form of FSP (hFSP), that increases pHi but cannot induce AR, activates the second channel (256).

Classical experiments that empty internal stores with thapsigargin or cyclopiazonic acid hint that the second channel opened during the sea urchin sperm AR could be a SOC (217), as postulated earlier for mammalian spermatozoa (283, 391, 451). The proposed intracellular Ca2+ store was the acrosome (217, 391). Later experiments showed that SOC activation resulted in acrosomal exocytosis even in the absence of actin polymerization (258). The pivotal importance of Ca2+ efflux from the acrosome to achieve acrosomal exocytosis for other sperm species has become apparent in recent years (86, 127, 142, 157, 248, 250, 251, 352, 391).

In many cell types, IP3, resulting from the hydrolysis of PIP2 by PLC, binds to the IP3R, releases Ca2+ from intracellular Ca2+ stores such as the endoplasmic reticulum, and activates SOCs (421). An IP3R-like protein was localized on the sperm acrosome region and less intensely in the flagella (581). As IP3 production had been reported during the AR (153), it was proposed that IP3 binding to the IP3R in the acrosome empties this organelle and activates SOCs. At the time it was thought that TRPCs may mediate the sustained uptake of Ca2+ through the second Ca2+ channel opened during the AR, which could be a SOC (217). In somatic cell types, DAG stimulates PKC (33), activating a subset of the TRP family (TRPC3, -6, -7) (54). The direct interaction between TRPCs and IP3Rs, as well as depletion of Ca2+ from internal Ca2+ stores, have been proposed as mechanisms for TRPC activation (reviewed in Ref. 2). As CatSper channels are present in sea urchins (433) and they, as well as the second channel opened by FSP binding to sea urchin sperm, are modulated by pHi (225, 437), it is worth exploring if CatSper participates in the AR.

C. Ca2+ Channels Implicated in the Sea Urchin Sperm Acrosome Reaction

1. A Ca2+-permeable and voltage-dependent channel

Initially a voltage-dependent multistate high-conductance Ca2+ channel was recorded from S. purpuratus flagellar membranes fused into black lipid membranes (BLMs). The channel displayed a high main-state conductance of 172 pS in 50 mM CaCl2 solutions with voltage-dependent decay to smaller conductance states at negative Em. At voltages more positive than −40 mV, the main-state open probability had values of ∼1.0 and decreased at more negative potentials, following a Boltzmann function with an E0.5 = −72 mV and an apparent gating charge value of 3.9. The channel poorly discriminated for divalent over monovalent cations (PCa/PNa = 6). La3+, Co2+, and Cd2+ which inhibit the EJ-induced AR at millimolar concentrations blocked the channel with a similar potency (325). These findings suggested that this channel could participate in the AR (325). Channels with similar characteristics were transferred directly from S. purpuratus and L. pictus sea urchin as well as from mouse intact spermatozoa to BLMs, corroborating the relevance of this Ca2+ channel in sperm physiology (42).

2. CaV channels

As described in section VIA, two CaV channels were identified in S. purpuratus testes. Their sequences were found to be most similar to mammalian CaV1.2 and CaV2.3, respectively. Antibodies against the CaV2.3 channel and a general antibody that detects a domain present in all CaV1 and CaV2 channels labeled the acrosome area of sea urchin spermatozoa. It is worth noting that the predicted sequence for the suCaVNL (M_001186699.1) has a polycystic kidney disease (PKD) domain. Furthermore, the CaV channel antagonists nifedipine (30 μM) and nimodipine (30 μM), which inhibit the AR, diminished (20–30% of the control) the [Ca2+]i elevation induced by a K+-induced depolarization in valinomycin-treated L. pictus spermatozoa, consistent with the presence of functional CaV channels in sea urchin spermatozoa (223). The functional participation of sea urchin sperm CaV channels in the AR has been suggested by Em and [Ca2+]i measurements in sperm populations, and by extrapolating the mammalian CaV pharmacology; therefore, further work is needed to establish the functional role of these channels (Table 2).

3. The inductor of the acrosome reaction and its receptor

The natural AR inductor FSP interacts with suREJ1, a sperm plasma membrane receptor of 210 kDa with one transmembrane domain and high homology to the human autosomal dominant polycystic kidney disease (ADPKD) protein, PKD1/PC1 (373, 525). Monoclonal antibodies against suREJ1 induce the AR by opening Ca2+ channels (136, 373, 517).

Most cases of ADPKD are caused by naturally occurring mutations in two genetically interacting loci, pkd1 (85%) and pkd2 (∼15%) (320). pkd1 encodes a large multispanning membrane protein [PKD1 or polycystin-1 (PC1)], while pkd2 encodes a protein [PKD2 or polycystin-2 (PC2) or TRPP2], now known to belong to the TRP superfamily of ion channels (520). The products of the pkd genes play key roles in renal and vascular mechanosensory transduction, in primary cilia of renal, nodal, and endothelial cells (reviewed in Ref. 403).

Sea urchin spermatozoa possess two additional homologs of suREJ1, suREJ2 and suREJ3. These three proteins contain the “REJ module,” a >900 amino acid sequence shared by PKD1/PC1 and by PKDREJ, a testis-specific protein in mammals (266) whose function is discussed in the mammalian sperm AR section (497). suREJ2 is present throughout the entire sperm plasma membrane, has two transmembrane segments, is mainly located over the sperm mitochondrion, and seems not to be exposed to the extracellular media. Due to its localization, it was proposed that it functions as an anchor between the mitochondrion and the plasma membrane during spermiogenesis (196). Then again, suREJ3 is a multidomain orphan receptor with 11 putative TM segments resembling latrophilins, G protein-coupled receptors involved in exocytosis (356). Its COOH-terminal transmembrane region includes a sequence that is homologous to CaV channels and which has been implicated in association with PC2 (424, 568). Interestingly, both suREJ1 and suREJ3 are found in the plasma membrane over the acrosomal vesicle. Considering the homology between the sea urchin sperm suREJ proteins and mammalian PKDREJ, it was suggested that both animal groups may share fertilization signaling pathways (356).

Sea urchin spermatozoa also possess PC2 (suPC2), a six-pass transmembrane protein containing a COOH-terminal cytoplasmic EF hand and coiled-coil domains (382), equivalent to mammalian TRPP2 (PKD2 or PC2) (129) which interacts with suREJ3 and localizes exclusively as a thin band on the sperm plasma membrane overlying the acrosomal vesicle (356, 382). There is a long-standing debate about the subcellular localization of TRPP2, as it has been found in the plasma membrane and in the endoplasmic reticulum. In the plasma membrane it can interact for example with PC1, TRPC1, TRPC4, and TRPV4. In the endoplasmic reticulum it interacts with IP3Rs, syntaxin, and RyRs (321, 448, 520). Fascinating questions remain open as to the function of these proteins in the sea urchin spermatozoa.

It is known that PC1 and PC2 interact to produce Ca2+-permeable nonselective currents that transduce extracellular stimuli (235). Since channel activities resembling PC1/PC2 complexes have been measured in sea urchin sperm membranes (42, 325), REJ and suPC proteins may participate in the ion fluxes triggered by FSP that lead to the AR. S. purpuratus sea urchins contain 10 SpREJ proteins within the PKD1 gene family with low similarity to each other and distinct patterns of expression during embryogenesis but, except for AR-associated roles of SpREJ1 and SpREJ3, the functions of the other SpREJ proteins remain unknown (228).

4. Transient receptor potential channels

The S. purpuratus genome contains the predicted sequences for TRPC3, TRPC5, and TRPC6. Accordingly, RNAs for TRPC3 and TRPC6 were found in S. purpuratus testis (G. Granados-González, unpublished data). Using commercial antibodies against the mammalian proteins, our group found differential staining in the acrosomal and the mitochondrial areas, as well as in the flagella of mature sea urchin spermatozoa, for TRPC3, TRPC5, and TRPC6. The signal for TRPC3 was stronger in the acrosomal area than in the mitochondrial area and weaker in the flagella. Moreover, consistent with the presence of the mentioned TRPCs in sea urchin spermatozoa, Western blot experiments with sperm membranes showed different relative mobility bands (TRPC3 ∼109 kDa; TRPC5 ∼115 kDa; TRPC6 ∼140 kDa). Since the differential distribution of TRPCs suggests that these channels could participate in the AR as well as in sperm motility, functional studies are needed. Unpublished results from our group by C. Wood showed that 20 μM SKF96365, a TRPC channel blocker, inhibits sea urchin sperm motility.

5. Ca2+ stores and Ca2+ clearance mechanisms

Since spermatozoa lack an endoplasmic reticulum, the acrosome (217, 530), the mitochondrion (8, 105, 230), and possibly remnants from the nuclear envelope (210) may act as intracellular Ca2+ stores. In sea urchin spermatozoa, two Ca2+-ATPases have been described, a plasma membrane Ca2+-ATPase located in the head (229) and a SPCA, surprisingly restricted to the mitochondrion. It is known that Ca2+-ATPases are responsible for ∼75% of the Ca2+ extrusion, and as anticipated, their inhibition completely blocks AR, stressing the importance of these enzymes in fertilization. The remaining Ca2+ efflux is carried out by a Na+/Ca2+ exchanger (229).

It was recently reported that agents that alter the mitochondrial function via differing mechanisms (CCCP, a proton gradient uncoupler; antimycin, a respiratory chain inhibitor; oligomycin, a mitochondrial ATPase inhibitor; and CGP37157, a Na+/Ca2+ exchange inhibitor) induce [Ca2+]i increases that depend on external Ca2+. The plasma membrane permeation pathways activated by the mitochondrial inhibitors are permeable to Mn2+, sensitive to SOC blockers (Ni2+, SKF96365, and Gd2+) (399) as well as to internal-store ATPase inhibitors (thapsigargin and bisphenol) (8). These observations taken together indicate that the functional status of the sea urchin sperm mitochondrion regulates cell Ca2+ homeostasis, including Ca2+ influx possibly through SOCs.

A plethora of studies on Ca2+ increases in sea urchin sperm populations triggering the AR either with the natural inducer (FSP or EJ) or artificially [i.e., raising external pH to 9, depolarization, nigericin, Ca2+ ionophore, etc. (147, 201)] have been reported. However, it is fundamental to learn the location and temporal characteristics of the [Ca2+]i changes that occur under these conditions, particularly on the sperm head, to fully understand the mechanisms that regulate the AR signaling pathways. Therefore, Ca2+ imaging studies at the single-cell level of sea urchin spermatozoa undergoing the AR are urgently needed.

6. Cationic channels activated by NAADP

NAADP-induced Ca2+ release was discovered in the sea urchin egg (93, 311). Indeed, experimental results indicated that sperm produce NAADP that is injected into the sea urchin egg and that it participates in its activation (125, 198). Fertilization experiments in the starfish are consistent with this proposal (366). More recently, it has been found that NAADP may play a role in the sperm AR, as the acrosome is a lysosome-related organelle and NAADP triggers Ca2+ release from lysosome-related organelles of many cell types (126, 198).

Experiments using Mn2+ quenching of fura 2 fluorescence in sperm vesicles and 45Ca2+ uptake in digitonin-permeabilized sea urchin spermatozoa revealed that NAADP directly regulates acrosomal cation channels (530). Ca2+ uptake induced by NAADP was partially inhibited by bafilomycin (V-ATPase inhibitor) and by glycylphenylalanine 2-naphthylamide (a lysosomal disruptor), and significantly decreased by thapsigargin and ionomycin, indicating that the Ca2+ uptake was directly into the acrosome. It is worth pointing out that two-pore channels (TPCs/TPNCs; discussed in sect. IIID) may be the acrosomal channels responding to NAADP (530).


These are exciting times for the field of fertilization, as new genetic, electrophysiological, and imaging tools are opening new avenues of research and allowing the revision of basic concepts in gamete physiology. For marine organisms like the sea urchin, cameras with better temporal and spatial resolution will facilitate examination of swimming spermatozoa and their response to chemoattractants in three dimensions, and how [Ca2+]i fluctuations modulate flagellar form. Perhaps soon it will be possible to measure Em changes in individual spermatozoa, as this would have an enormous impact on our understanding of many aspects of sperm motility and physiology. Learning more about how pHi is regulated and its influence on motility, capacitation, and the sperm AR is urgently needed. Experiments in mice with genetically modified ZP eggs (191) and GFP-labeled spermatozoa (244) could begin to explore the site and time when the AR occurs and, with anticipated new probes, when and where [Ca2+]i increases, and how ZP3 participates and if it has to be glycosylated and where.

Fertilization is a complex process that generates a unique individual resulting from the fusion of the male and female gametes. Spermatozoa face a myriad of environmental and maturational changes as they travel to find the egg. The authors may be wrong, but it seems unlikely that these specialized cells could succeed in their mission with a very restricted set of ion channels and transporters. Until today, from all the ion channels that have been knocked out, only two sperm specific channels, CatSper, a putative Ca2+ channel (423), and SLO3, a K+ channel (416), result in male infertility, although sperm production is normal. Does this imply that significant redundancy is essential to safeguard this critical event for life? Alternatively, as has been suggested by some authors (i.e., Ref. 431), nature has evolved a unique minimal set of specialized channels and transporters for spermatozoa to achieve their apparently simple but crucial goal. Oddly, the only mammalian sperm ionic currents that have been reported by patch clamping of the mouse cytoplasmic droplet (see sects. III and IVD) are from the two channels mentioned above, plus those recently recorded in human spermatozoa that are likely derived from HV channels (329) (see sect. IVD). However, functional, pharmacological, biochemical, and molecular biological strategies indicate that other channels are present in spermatozoa. It is clear from the research in sea urchin sperm chemotaxis that more than two Ca2+ channels are required, in addition to the tetraKCNG (see sect. VI), a crucial channel for this process that is from a new family (60, 194). Future experiments will explain why patch-clamp recordings in epididymal mouse sperm so far have only revealed a restricted set of ion channels. Are CaVs and other channels functionally silenced during maturation and turned on during capacitation?

We can anticipate fruitful and challenging times ahead for research in reproduction and sperm physiology that will certainly contribute to our better understanding of the wonder of life and hopefully to its preservation in all its treasured forms.


During the process of publication, two groups (329a, 484a) independently reported that CatSper is the principal Ca2+ channel activated by progesterone in human spermatozoa. A long-lasting mystery has been solved.


Our work is supported by DGAPA Grants IN211809 (to A. Darszon), IN204109 (to C. L. Treviño), IN217409 (to C. Beltran), and IN2211103 (to T. Nishigaki); CONACyT Grants 49113 (to A. Darszon), 56660 (to T. Nishigaki), and 99333 (to C. L. Treviño); and National Institutes of Health Grant R01 HD038082-07A1 (to A. Darszon).


No conflicts of interest, financial or otherwise, are declared by the authors.


We are in debt to Drs. Jorge Carneiro, Chris Wood, Shirley Ainsworth, Richard Horn, and Victor Vacquier for their suggestions and comments that significantly influenced the structure of the review. The reviewers of our work have made a major contribution that we appreciate enormously. We deeply thank Adan Guerrero for providing Figure 8, its legend, and valuable comments. We also thank Jose Luis de la Vega and Yoloxochitl Sánchez for technical assistance.

Address for reprint requests and other correspondence: A. Darszon, Departamento de Genética del Desarrollo y Fisiología Molecular, Instituto de Biotecnología, UNAM, Avenida Universidad #2001 Col. Chamilpa, CP 62210, Cuernavaca, Mor., México (e-mail: darszon{at}


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68.
  69. 69.
  70. 70.
  71. 71.
  72. 72.
  73. 73.
  74. 74.
  75. 75.
  76. 76.
  77. 77.
  78. 78.
  79. 79.
  80. 80.
  81. 81.
  82. 82.
  83. 83.
  84. 84.
  85. 85.
  86. 86.
  87. 87.
  88. 88.
  89. 89.
  90. 90.
  91. 91.
  92. 92.
  93. 93.
  94. 94.
  95. 95.
  96. 96.
  97. 97.
  98. 98.
  99. 99.
  100. 100.
  101. 101.
  102. 102.
  103. 103.
  104. 104.
  105. 105.
  106. 106.
  107. 107.
  108. 108.
  109. 109.
  110. 110.
  111. 111.
  112. 112.
  113. 113.
  114. 114.
  115. 115.
  116. 116.
  117. 117.
  118. 118.
  119. 119.
  120. 120.
  121. 121.
  122. 122.
  123. 123.
  124. 124.
  125. 125.
  126. 126.
  127. 127.
  128. 128.
  129. 129.
  130. 130.
  131. 131.
  132. 132.
  133. 133.
  134. 134.
  135. 135.
  136. 136.
  137. 137.
  138. 138.
  139. 139.
  140. 140.
  141. 141.
  142. 142.
  143. 143.
  144. 144.
  145. 145.
  146. 146.
  147. 147.
  148. 148.
  149. 149.
  150. 150.
  151. 151.
  152. 152.
  153. 153.
  154. 154.
  155. 155.
  156. 156.
  157. 157.
  158. 158.
  159. 159.
  160. 160.
  161. 161.
  162. 162.
  163. 163.
  164. 164.
  165. 165.
  166. 166.
  167. 167.
  168. 168.
  169. 169.
  170. 170.
  171. 171.
  172. 172.
  173. 173.