Regulation and roles of Ca2+ stores in human sperm

[Ca2 +]i signalling is a key regulatory mechanism in sperm function. In mammalian sperm the Ca2 +-permeable plasma membrane ion channel CatSper is central to [Ca2 +]i signalling, but there is good evidence that Ca2 + stored in intracellular organelles is also functionally important. Here we briefly review the current understanding of the diversity of Ca2 + stores and the mechanisms for the regulation of their activity. We then consider the evidence for the involvement of these stores in [Ca2 +]i signalling in mammalian (primarily human) sperm, the agonists that may activate these stores and their role in control of sperm function. Finally we consider the evidence that membrane Ca2 + channels and stored Ca2 + may play discrete roles in the regulation of sperm activities and propose a mechanism by which these different components of the sperm Ca2 +-signalling apparatus may interact to generate complex and spatially diverse [Ca2 +]i signals.

Cellular activity is constantly regulated by environmental cues and signals from other cells. Long-term regulation of cell function is normally achieved by control of gene expression, changing the complement and levels of proteins in the cell, but rapid or short-term changes are achieved by 'post-translational' protein modification, such as phosphorylation, sumoylation and nitrosylation, which alter the function/activity of proteins already present. Ca 2C -signalling is a key regulator of such post-translational modifications, with changes in cytoplasmic Ca 2C concentration ([Ca 2C ] i ) controlling the activities of key enzymes and proteins. Large changes in [Ca 2C ] i can be achieved 'instantaneously' by flux of Ca 2C into the cytoplasm from the extracellular fluid or from storage organelles (primarily the endoplasmic reticulum) within the cell (Fig. 1a). The rapidity with which [Ca 2C ] i -signals can be generated is crucial for 'instantaneous' cellular responses such as activation of muscle contraction and secretion of neurotransmitters that are achieved by rapid posttranslational modification of protein function.
The highly condensed nucleus of sperm is transcriptionally silent (Miller et al. 2005, Miller & Ostermeier 2006 and translational activity is also negligible (though evidence has been presented for translation occurring at mitochondrial ribosomes; Gur & Breitbart 2008, Zhao et al. 2009, Chandrashekran et al. 2014a. Regulation of sperm function is therefore dependent primarily on post-translational processes. [Ca 2C ] i signalling is pivotal to this regulation, and in mammalian sperm it plays a central role in controlling the cell's behaviour (motility type and potentially chemotaxis), the induction of acrosome reaction (AR) and the process of capacitation , Darszon et al. 2007. The importance for sperm function of membrane Ca 2Cchannels and Ca 2C -influx is well established (Darszon et al. 2011) but there is also good evidence for the existence and functional importance of intracellular Ca 2C -storage organelles in sperm (Darszon et al. 2007). Previously we reviewed the identities and functions of Ca 2C stores in sperm, focussing on the evidence for the existence of such stores, their components (pumps and channels) and their possible roles in the regulation of function in the mature sperm cell (Costello et al. 2009). Since then considerable progress has been made in understanding the central role of Ca 2C signalling in the regulation of mammalian and non-mammalian sperm function and the mechanisms by which sperm [Ca 2C ] i signals are generated. In particular successful application of whole cell patch clamp technique, in human as well as mouse sperm, has revealed the central importance of Ca 2C influx through CatSper, a sperm specific, Ca 2C -permeable channel in the membrane of the flagellar principal piece. Male mice null for CatSper are infertile (Ren et al. 2001) and their sperm show defective motility (Carlson et al. 2003). Here we review recent progress in understanding the diversity of mechanisms for the regulation of Ca 2C store activity and the evidence for their involvement in controlling sperm function.

Ca 2C stores and their regulation
The importance of Ca 2C stores in generating complex Ca 2C signals in somatic cells has long been recognized. Until relatively recently the endoplasmic reticulum Ca 2C store has been the major focus for research as this was the first organelle to show controllable mobilization of Ca 2C through second messengers acting upon intracellular Ca 2C channels, as well as being able to be refilled via Ca 2C pumps. Additionally, these Ca 2C signals could also be re-modelled through the regulation of these Ca 2C transporters to generate complex spatial and temporal Ca 2C transients (Berridge et al. 2003). It has now become clear that many other organelles such as mitochondria, endosomes, lysosomes and Golgi complexes also contribute to the generation and propagation of these complex Ca 2C signals within cells . Furthermore, novel Ca 2C transporters have also been identified within these other organelles and several have recently been identified in sperm (Costello et al. 2009).

Intracellular Ca 2C channels
The major intracellular Ca 2C channels that have been identified and appear to be almost ubiquitously distributed within mammalian cells, especially on the endoplasmic reticulum, include the inositol-1,4,5-trisphosphate-(IP 3 )-sensitive Ca 2C channel (or IP 3 receptor; IP 3 R) and the ryanodine receptor (RyR)   (Fig. 1a). The IP 3 receptor, as the name implies, is activated by the second messenger IP 3 that is generated through the hydrolysis of phosphatidylinositol-4,5-bisphosphate. This channel has a specific IP 3 binding site that is located towards the N-terminus of the protein (Seo et al. 2012) and also has a requirement for Ca 2C which acts as a co-agonist in order for the channel to open (Bezprozvanny et al. 1991). The activation of RyR is likely to be through a mechanism involving Ca 2C induced Ca 2C release (CICR) and by the action of the putative second messenger cyclic-adenosine diphospho-ribose (cADPR) (Ogunbayo et al. 2011) (Fig. 1a). cADPR is made from NAD by the action of an ADP-ribosyl cyclase enzyme such as CD38 (Cosker et al. 2010), although other as yet unidentified enzymes may also be involved in catalysing this reaction (Guse 2015). It is as yet unclear whether, unlike the IP 3 R, cADPR binds directly to RyR or whether it binds to accessory proteins such as calmodulin or FK506-binding protein, that then interact with the RyR (Guse 2015).
Another metabolite of NAD which is believed to have Ca 2C mobilizing ability is nicotinic acid adenine dinucleotide phosphate (NAADP; Genazzani et al. 1997

R66 J Correia and others
NAADP is made from NADP through the action of either CD38 acting as a base-exchanger, swapping the nicotinamide group for nicotinic acid or via an unidentified NADP-deaminase (Guse 2015). NAADP is believed specifically to mobilize Ca 2C from acidic stores such as lysosomes (Churchill et al. 2002, Menteyne et al. 2006, which can then induce CICR at RyRs and IP 3 Rs in mammalian cells (Cancela et al. 1999;Fig. 1a). Results initially presented by Calcraft et al. (2009), indicated that NAADP specifically activates Ca 2C -specific two-pore channels (TPC) within the acidic organelles, these channels being first described in plants (Peiter et al. 2005). However, in kinetic studies there is a prominent lag between addition of NAADP and Ca 2C mobilization (Genazzani et al. 1997). Combined with the observation that photo-affinity labelling with azido-NAADP (Lin-Moshier et al. 2012) showed labelling of only low molecular weight proteins, not consistent with TPCs, it suggests that NAADP might function by binding to accessory proteins rather than directly to the channel. Recently there has been considerable controversy as to whether the NAADP-sensitive Ca 2C channel is a TPC (Morgan & Galione 2014). Data from two studies (Wang et al. 2012, Cang et al. 2013 suggested that TPCs are in fact Na C -specific channels with very low Ca 2C selectivity that are activated by phosphoinositide lipids and modulated by mTOR, but not by NAADP. However, recently published work with cells from mice null for TPC1 and TPC2 provided strong evidence that TPCs are similarly permeable to Ca 2C and Na C and are NAADP-gated through binding to an accessory protein (Ruas et al. 2015). Numerous kinases have been shown to modulate the activity of both the IP 3 Rs and RyRs, including several ubiquitous ser/thr kinases such as PKA, PKG and CaMKII (Yule et al. 2010, Camors & Valdivia 2014. Indeed, some of these kinases such as PKA appear to have both stimulatory and inhibitory effects on the IP 3 R, dependent upon isoform subtype and the presence of multiple kinase-dependent phosphorylation sites on the same receptor (Dyer et al. 2003). Less ubiquitous ser/thr kinases such as Akt and polo kinases as well as tyrosine kinases such as fyn kinase have also been shown to affect these channels (Yule et al. 2010, Camors & Valdivia 2014.
Both the RyRs and the IP 3 Rs are modulated by changes in their oxidation states caused by reactive oxygen species (ROS) and reactive nitrogen species (RNS), and this occurs mainly through modification of specific cysteine (cys) amino acid residues. Oxidation of these cys residues in RyRs occurs both by S-glutathionylation as well as S-nitrosylation by the second messenger nitric oxide (NO; Csordas & Hajnoczky 2009) and promotes the activity of the channel by enhancing RyR subunit interactions and also by reducing the efficacy of inhibitory modulators (Hamilton & Reid 2000). In IP 3 Rs the effects of oxidative stress are complex: low levels of cys oxidation caused by low concentrations of thimerosal (a cys-modifying mercuric compound) and naturally generated ROS cause sensitization of this channel, while higher concentrations of thimerosal inhibit channel activity (Missiaen et al. 1991, Sayers et al. 1993. Currently, however, there is little evidence that NO can affect the activity of the IP 3 Rs.

Intracellular Ca 2C pumps
The major transporter involved in refilling Ca 2C stores within the endoplasmic reticulum is the sarcoplasmic/endoplasmic reticulum Ca 2C ATPase (SERCA; Fig. 1a), and these pumps occur abundantly in all somatic cells. Their role is to pump Ca 2C back into the storage organelles to help terminate Ca 2C signals , Michelangeli & East 2011. There are three isoforms of this Ca 2C ATPase, each encoded by a different gene and each isoform can exist in a variety of spliced variants that differ in size and regulatory properties (Michelangeli & East 2011). SERCA1 is mainly confined to skeletal muscle, while SERCA2 is widely distributed in most other tissues and organs and type 3 has a limited expression. Another related Ca 2C ATPase that is also found ubiquitously within somatic cells is the secretory pathway Ca 2C ATPase (SPCA), which is localized to the Golgi apparatus (Wootton et al. 2004). SPCA exists in two isoforms with the expression of type 1 being far more widespread than type 2, which appears to be mainly located within glandular tissues (Vanoevelen et al. 2005). Recently there has been evidence to suggest that SPCA2 can interact with and regulate the plasma membrane located ORAI Ca 2C channels that are implicated in store-operated Ca 2C entry (Feng et al. 2010), which may indicate a dual function for this Ca 2C ATPase in cells that express it.
There is currently some debate as to which type of intracellular Ca 2C ATPase is expressed in mature sperm. We have highlighted that SPCA1 is present in human sperm, where it appears to be mainly localized to the neck region of the cell where the redundant nuclear envelope (RNE) and calreticulin-containing vesicles are situated (Harper et al. 2005). This study also found no evidence for expression of SERCA in human sperm as no cross-reactivity was observed with a pan-isoform SERCA antibody and no effects on [Ca 2C ] i were observed with specific but saturating concentrations of the SERCAinhibitor thapsigargin. However, a more recent study (Lawson et al. 2007) detected SERCA2, mainly localized to the acrosome and mid-piece, using a SERCA2-specific antibody.
Unlike the intracellular Ca 2C channels, there is no strong evidence to suggest that either SERCA or SPCA can be directly phosphorylated and regulated by protein kinases, although some Ca 2C ATPase modulatory proteins like phospholamban (that is found almost exclusively in heart) are regulated through phosphorylation by PKA, PKG and CamKII (Colyer 1998). There is considerable evidence indicating that oxidative stress can modulate SERCA activity (although no studies have yet been undertaken on SPCA). Again a number of critical cys residues such as cys674 can be S-glutathionylated to cause an increase in SERCA pump activity (Adachi et al. 2004). Modifications of other cys residues on the Ca 2C ATPase, however, can have inhibitory effects (Sayers et al. 1993, Sharov et al. 2006, Csordas & Hajnoczky 2009).
Ca 2C stores, mechanisms for store mobilisation and store-operated Ca 2C channels in sperm During the later stages of their development spermatozoa shed much of their cytoplasm including intracellular organelles. Thus mammalian sperm contain no organised endoplasmic reticulum. However, studies on the expression of Ca 2C store components and on the generation [Ca 2C ] i signals suggest that the remaining intracellular organelles function as Ca 2C -stores and play a significant role in the regulation of cellular function (Costello et al. 2009). In particular, the acrosomal vesicle at the apex of the head and the collection of vesicular membranous structures that occur at the sperm neck and anterior midpiece (including the cytoplasmic droplet of human sperm) appear to be functionally important Ca 2C -stores ( Fig. 1b; shown in green). At both these locations IP 3 Rs have been detected in human and in bovine sperm by immuno-staining (Dragileva et al. 1999, Kuroda et al. 1999, Ho & Suarez 2001, Naaby-Hansen et al. 2001. RyRs have also been detected in human and rodent sperm (Trevino et al. 1998, Lefievre et al. 2007). Staining of human sperm with anti-RyR1, anti-RyR2, pan-RyR and BODIPY-FLX ryanodine is localised primarily to the neck region, though some acrosomal staining was also observed (Harper et al. 2004, Lefievre et al. 2007, Park et al. 2011. In contrast, other authors (Ho & Suarez 2001) have reported no staining of bovine sperm with BODIPY-FLX ryanodine (see Costello et al. (2009) for further discussion). Thus mobilisation of stored Ca 2C in mammalian sperm may occur in response to generation of IP 3 by activity of phospholipase C and by CICR at IP 3 Rs or RyRs. These processes can be sensitised by effects such as oxidative stress and S-nitrosylation (see 'Ca 2C stores and their regulation'). For instance, exposure of human sperm to NO% at levels equivalent to those produced by explants of reproductive tract lining mobilises stored Ca 2C and modifies flagellar activity (Lefievre et al. 2007, Machado-Oliveira et al. 2008.
In addition to generation of IP 3 in sperm, there is evidence that other Ca 2C mobilising messengers (NAADP and cADPR) are synthesised in sperm and/or produced in response to stimulation. Sea urchin sperm contain significant levels of both cADPR and NAADP, which may contribute to oocyte activation (Chini et al. 1997, Billington et al. 2002. Human sperm have been shown to contain cADPR at micromolar concentrations but NAADP was not detected (Billington et al. 2006). Interestingly, this study also demonstrated synthesis of cADPR by human sperm but the ecto-enzyme CD38 (an enzyme present on mammalian cells that synthesises both cADPR and NAADP; see 'Ca 2C stores and their regulation') could not be detected by western blotting. In contrast, Park et al. (2011), reported detection of CD38 in human sperm after co-incubation with prostasomes (prostate-derived membrane vesicles; see below). Furthermore, the presence of a novel NAADP synthase, which lacks the cyclase activity of CD38, has been described both in sea urchin (Vasudevan et al. 2008) and human sperm (Sanchez-Tusie et al. 2014). In sea urchin sperm this enzyme is strongly Ca 2C -regulated and most active at acid pH whereas the human enzyme shows only weak Ca 2C -regulation and activity is maximal at pH 7-8 (Vasudevan et al. 2008, Sanchez-Tusie et al. 2014.
Recent findings have supported the idea that NAADP is functional in human sperm. Sanchez-Tusie et al.
(2014) investigated the effects of cell-permeant (AMester) derivatives of NAAPD and cADPR. No effects were observed with cADPR, consistent with previous pharmacological investigation by Billington et al. (2006), but NAADP caused elevation of [Ca 2C ] i both in cells incubated under standard conditions and also when [Ca 2C ] o was buffered to 100 nM, conditions under which Ca 2C influx is negligible and [Ca 2C ] i signalling depends solely on mobilisation of stored Ca 2C . Staining of NAADP receptors using the fluorescent NAADP receptor ligand Ned-19 and identification of acidic organelles using lysotracker highlighted both an anterior store (potentially the acrosome) and a store at the sperm neck (Fig. 1b). Consistent with these findings, Arndt et al. (2014), studying AR (see below), provided evidence for involvement in this process of NAADP and TPCs, which have been proposed to be the NAADP receptor/Ca 2C channel of acidic Ca 2C storage organelles (Calcraft et al. 2009 ; Fig 1a; see 'Ca 2C stores and their regulation'). Park et al. (2011) investigated the incorporation into human sperm of proteins from prostasomes (prostatederived vesicles which are normally added to sperm during ejaculation) and their effects on [Ca 2C ] i signalling. They concluded that CatSper channel proteins were present in the differentiated sperm, but other Ca 2C signalling 'tools' including RyRs and CD38 were added to the freshly-ejaculated sperm upon mixing with prostasomes, by fusion with the membrane of the midpiece. They also examined the effects of stimulation with progesterone on [Ca 2C ] i and motility of sperm exposed to prostasomes and sperm that had been rapidly removed from semen to minimise mixing with prostasomes. Their data suggest that the generation of R68 J Correia and others sustained [Ca 2C ] i signals (such as the second component of the biphasic progesterone-induced [Ca 2C ] i signal) and consequent effects on motility may depend, at least partly, upon generation of cADPR by prostasomederived enzymes. Interestingly, CD38-null mice proved to be fertile, but analysis showed that 20% of normal ADPR cyclase activity remained in prostasomes from these animals, indicating the presence of a non-CD38 ADPR-cyclase, potentially that described by Sanchez-Tusie et al. (2014). Thus both NAADP and cADPR are potentially synthesised by sperm and involved in regulation of sperm Ca 2C store activity but their roles are not yet clear.
In somatic cells mobilisation of stored Ca 2C induces secondary Ca 2C influx through channels at the cell membrane (store-operated channels, SOCs) by the process of capacitative Ca 2C entry (CCE) (Fig. 1a). CCE both prolongs Ca 2C signals that are induced by store mobilisation and provides Ca 2C for re-charging of the storage organelles. Recently great progress has been made in elucidating the key players and mechanisms in this process. Stromal interaction molecule (STIM) has been identified as the sensor molecule present in the membrane of the Ca 2C store. The intraluminal part of STIM includes a Ca 2C -binding EF hand that detects depletion of stored Ca 2C . STIM then redistributes, moving to a position adjacent to the plasma membrane where it activates channel proteins (ORAI and possibly members of the TRPC (transient receptor potential canonical) family (Cahalan 2009)). [Ca 2C ] i signals in human and other mammalian sperm induced by agonists and by treatments designed to mobilise stored-Ca 2C show characteristics consistent with the occurrence of CCE (Blackmore 1993, Dragileva et al. 1999, O'Toole et al. 2000, Park et al. 2011, Lefievre et al. 2012. STIM1, ORAI and TRPC proteins have been detected in human sperm (Castellano et al. 2003, Darszon et al. 2012, Lefievre et al. 2012, STIM1 being localised primarily to the neck region/midpiece and the acrosome where Ca 2C stores are present (Lefievre et al. 2012). To date the application of whole-cell patch clamp has not provided evidence for the occurrence of CCE in human sperm (Lefievre et al. 2012) so these findings must be interpreted cautiously, but [Ca 2C ] i signals generated by mobilisation of Ca 2C stores in sperm may be amplified by activation of CCE. Induction of CCE in somatic cells can have a latency of tens of seconds due to the need for STIM to migrate to the peripheral portions of the endoplasmic reticulum where it can interact with SOC proteins ), but in sperm the storage organelles are close to the plasma membrane and STIM proteins are localised here, such that CCE could be near 'instantaneous'. Pre-treatment of human sperm with low concentrations of 2-aminoethoxydiphenyl borate, which potentiates CCE by promoting the interaction of STIM with SOCs (Navarro-Borelly et al. 2008, Wang et al. 2009, Yamashita et al. 2011) significantly enhanced the amplitude of the progesterone-induced Ca 2C transient at the sperm neck (where secondary release of stored Ca 2C may occur; Fig. 1b; see 'Model for interaction of CatSper channels and Ca 2C -stores') but did not affect the response in the flagellum, where progesterone activates CatSper channels (Fig. 1b), or the kinetics of the signal at either location (Lefievre et al. 2012). Conversely, when sperm were pre-treated with a cellpenetrating peptide that mimics part of the key SOAR region of STIM1 (potentially preventing auto-inhibitory folding of STIM upon store-refilling) there was a marked prolongation of the progesterone-induced [Ca 2C ] i transient in a subset of cells (Morris et al. 2015).

Mobilisation of sperm Ca 2C stores by agonists
In the majority of somatic cells mobilisation of stored Ca 2C occurs upon agonist-induced synthesis of Ca 2C mobilising intracellular messengers. Thus agonistinduced synthesis of inositol trisphosphate, cADPR and NAADP can lead to rapid release of stored Ca 2C and generation of local, global and complex spatio-temporal signals (Fig. 1a). Is there evidence that such processes occur and are functionally significant in responses to agonist stimulation of sperm?
The best-characterised agonist-induced [Ca 2C ] i signals in sperm are responses to solubilised zona pellucida/zona proteins in mouse cells and progesterone in human. Application of patch clamp has clearly shown that the primary action of progesterone in human sperm is to activate CatSper channels, leading to Ca 2C -influx (Lishko et al. 2011, Strunker et al. 2011 Espino et al. (2009)), suggesting that any mobilisation of stored Ca 2C is a secondary response. Synthesis of IP 3 is reported to occur downstream of progesteroneinduced Ca 2C influx (Thomas & Meizel 1989), an important observation that should be pursued. Stimulation of mouse sperm with zona proteins induces AR, which requires elevation of [Ca 2C ] i in the sperm head (Florman et al. 2008) and is dependent on mobilisation of Ca 2C from the acrosomal store (De Blas et al. 2002;see below). The nature of the Ca 2C influx following stimulation is not clear and several channels may be involved (Florman et al. 2008, Cohen et al. 2014), but Ca 2C signals are sensitive to inhibition of G-protein signalling (using pertussis toxin) and inhibition of PLC (Florman et al. 2008, Ren & Xia 2010. Furthermore, in sperm from mice null for PLCd4 (in which males' fertility is severely impaired) the [Ca 2C ] i response is reduced and zona-induced AR does not occur (Fukami et al. 2001(Fukami et al. , 2003. Thus conventional IP 3 -induced mobilisation of stored Ca 2C is apparently central to this essential aspect of mammalian sperm physiology. Evidence for the existence of other store-mobilising agonists is largely preliminary, but there are a number of candidates, of which the best-studied is vitamin D (Blomberg Jensen 2014). Human sperm have been shown to express vitamin D receptor (VDR; Aquila et al. 2009, Blomberg Jensen et al. 2010, the enzymes CYP2R1 and CYP27B (which produce the active compound (1,25(OH) 2 D 3 ) cholecalciferol) and the inactivating enzyme CYP24A1 (Blomberg Jensen et al. 2010. All are expressed in the neck region of the sperm and staining of cells for VDR and CYP24A1 shows a strong association. In sub-fertile patients the proportion of cells expressing CYP24A1 varies greatly and is significantly correlated with semen quality (sperm count, concentration, morphology and motility;Blomberg Jensen et al. 2011. Stimulation of human sperm with 1,25(OH) 2 D 3 (100 pM-1 mM) induced a [Ca 2C ] i response, including a transient and plateau, that was blocked by pre-treatment with the non-genomic VDR antagonist 1b,25(OH) 2 D 3 but was insensitive to the nuclear VDR antagonist ZK159222 (Blomberg Jensen et al. 2011). This effect was greatly reduced by pre-treatment with the phospholipase C inhibitor U73122 (2 mM) but was also inhibited by incubation in EGTA-buffered medium for up to 20 min prior to stimulation. Both motility and AR were significantly increased upon stimulation with 1,25(OH) 2 D 3 (Blomberg Jensen et al. 2011).
Kisspeptin, a peptide agonist of the G-protein coupled receptor GPR54/KISS1R, has also been shown to cause sustained, dose-dependent elevation of [Ca 2C ] i in human and in mouse sperm (Pinto et al. 2012, Hsu et al. 2014. In neurons binding of kisspeptin to its receptor activates PLC and results in generation of IP 3 and diacyglycerol, leading to mobilisation of stored Ca 2C and also depolarisation (Liu et al. 2008, Pielecka-Fortuna et al. 2008, Beltramo et al. 2014. In human sperm the effect of kisspeptin on [Ca 2C ] i did not occlude the response to stimulation with the CatSper agonist progesterone and was not reduced when applied in the presence of progesterone (Pinto et al. 2012). Both KISS1R and kisspeptin itself were detected in the head of human sperm, suggesting that an autocrine action of the peptide may occur. Motility parameters of kisspeptintreated cells were significantly altered, including an increase in lateral movement of the head and a decrease in linearity of the sperm path, characteristics of hyperactivated sperm (Pinto et al. 2012). Ghrelin, another peptide hormone that also acts through mobilisation of stored Ca 2C (Camina et al. 2003), has also been detected in human sperm (Moretti et al. 2014). Micromolar concentrations of ghrelin have been shown to increase [Ca 2C ] i and motility in rat sperm (Lukaszyk et al. 2012), but expression of ghrelin receptors or effect of ghrelin on human sperm [Ca 2C ] i have not been investigated.

Functional significance of Ca 2C -stores
The acrosome AR is the fusion between the outer acrosomal membrane and the overlying plasma membrane. Fusion occurs at multiple points, resulting in vesiculation and loss of the fused outer acrosomal membrane/plasmalemma so that the acrosomal content is released and the inner acrosomal membrane becomes the new cell surface. Membrane fusion proteins from the SNARE family are present in the acrosomal region and may be integrated into microdomains that facilitate Ca 2C -regulated membrane fusion in a manner that has been compared with events at presynaptic terminals (De Blas et al. 2005, Mayorga et al. 2007, Zitranski et al. 2010. Zona pellucida proteins interact with sperm surface receptors to activate a signalling cascade leading to AR (Florman et al. 2008) and release of acrosomal content at the surface of the zona may, in combination with hyperactivated motility, facilitate zona penetration. However, observation of mouse IVF using sperm with GFP-labelled acrosomes showed that, in addition to cells that undergo AR at the surface of the zona, sperm which arrive having already lost their acrosome (probably within the cumulus) may go on to penetrate the zona and fertilise (Jin et al. 2011). Physiological inducers of AR that have been studied (primarily mouse ZP3 and progesterone) induce Ca 2C influx across the plasma membrane and a sustained rise in [Ca 2C ] i . O'Toole et al. (2000) provided pharmacological evidence that ZP3-induced AR in mouse sperm involved activation of store-operated Ca 2C influx downstream of Ca 2C store mobilisation. De Blas et al. (2002) showed that in streptolysinpermeabilised human sperm, mobilisation of the acrosomal Ca 2C store was a requirement for AR even when it was directly induced by introduction of Rab3A into the cytoplasm. Further studies using this permeabilised sperm model have provided information about the mechanisms by which fusion of the plasma and outer acrosomal membranes is regulated, resulting in a detailed model in which mobilisation of the acrosomal store is a central and necessary event (Ruete et al. 2014). Stimulation of PLC, leading to generation of IP 3 and activation of IP 3 Rs in the outer acrosomal membrane may be key to this process (Fukami et al. 2001(Fukami et al. , 2003, but there is also evidence that the acrosomal membrane contains the NAADP-sensitive, Ca 2C -permeable TPC (Calcraft et al. 2009) and that NAADP mobilises acrosomal Ca 2C in mouse sperm (Arndt et al. 2014).

The RNE and calreticulin-containing vesicles
A second area where Ca 2C storage organelles have been identified in mammalian sperm is at the sperm neck and midpiece (Fig. 1b). Mitochondria have mechanisms for accumulation and release of Ca 2C (Drago et al. 2011, Pizzo et al. 2012) and therefore may contribute to Ca 2C R70 J Correia and others buffering and signalling in this part of the sperm. Inhibition of mitochondrial function in sea urchin sperm, using respiratory inhibitors or uncouplers, causes a rise in [Ca 2C ] i and leads to activation of Ca 2C influx that has characteristics consistent with SOCs (Ardon et al. 2009). Treatment with mitochondrial uncouplers (2,4-dinitrophenol, carbonyl cyanide-4-(trifluoromethoxy)-phenyl-hydrazone) also increases [Ca 2C ] i in human sperm (J Morris and S Publicover, unpublished observations). Mitochondria may thus contribute to shaping of Ca 2C signals in sperm. However, the primary stimulus-regulated Ca 2C storage in this part of the sperm is in the RNE and/or a second, apparently separate group of calreticulin-containing vesicular structures, both of which are sited at the sperm neck region and cytoplasmic droplet (Ho & Suarez 2001, Naaby-Hansen et al. 2001. Mobilisation of Ca 2C stored in these compartments regulates flagellar activity and treatment of mouse sperm with thimerosal stimulates hyperactivated motility by activating Ca 2C release from these organelles (Ho & Suarez 2001, Marquez et al. 2007). This effect occurs in the absence of extracellular Ca 2C and can be induced in sperm that are null for CatSper (Marquez et al. 2007). In mouse sperm the direction of the major, high-amplitude flagellar bend of hyperactivated sperm can be clearly characterised by reference to the hooked acrosomal cap (pro-hook or antihook). Sperm that became hyperactivated during capacitation in vitro (due to activation of CatSper) show pro-hook bends whereas those activated by store mobilisation (using thimerosal) show anti-hook bends (Chang & Suarez 2011). When sperm were observed interacting with the lining of isolated mouse oviducts, most hyperactivated cells showed anti-hook bending of the type that is elicited by store mobilisation (Chang & Suarez 2012).
In human sperm a similar effect of store mobilisation is observed. Thimerosal greatly increases the proportion of cells showing hyperactivated motility and 4-aminopyridine, which both alkalinises the cytoplasm (thus activating CatSper) and mobilises stored Ca 2C , is similarly potent (Alasmari et al. 2013a,b). In contrast, manipulations that should activate CatSper (elevation of pH i , stimulation with progesterone or prostaglandin E 1 ) elevate [Ca 2C ] i but have only minor stimulatory effects on the proportion of hyperactivated cells. Instead, these manipulations significantly increase penetration into viscous media (Alasmari et al. 2013a,b, Luo et al. 2014.

Model for interaction of CatSper channels and Ca 2C -stores
Patch clamp recordings have provided no evidence that conventional voltage-operated Ca 2C channels contribute to Ca 2C influx in mature mammalian sperm. In mouse sperm null for CatSper1 and the K C channel Slo3, only a small leak current was recorded even at high intracellular pH and strong depolarisation (Zeng et al. 2013). CatSper channels in mouse and human sperm are pH-and (weakly) voltage-sensitive, but in human sperm the channel is also ligand-sensitive. Established Ca 2Cmobilising agonists of human sperm such as progesterone and prostaglandin E 1 have been shown to activate CatSper but also a range of other small molecules including environmental pollutants such as 1,1,1-trichloro-2,2-bis(4-chlorophenyl)ethane, 1,1bis(4-chlorophenyl)-2,2,2-trichloroethane (4,4 0 -DDT), p,p 0 -dichlorodiphenyldichloroethylene and 4-methylbenzylidene camphor are potent agonists (Tavares et al. 2013, Schiffer et al. 2014). In addition, agents used to demonstrate cyclic-nucleotide-activated Ca 2C influx (such as 8-Br-AMP) have been shown directly to activate CatSper by binding at the extracellular surface (Brenker et al. 2012). Thus it is possible that a significant proportion of the pharmacological data that apparently support the existence of multiple Ca 2C influx pathways in sperm are misleading and in fact reflect actions of the drugs on Ca 2C flux through CatSper channels (Brenker et al. 2012). Furthermore, experiments using CatSper null mice provide strong evidence that [Ca 2C ] i elevation induced by solubilised ZP is dependent on Ca 2C influx through the CatSper channel in the flagellum, which then propagates to the head ; though see Cohen et al. (2014)). Interestingly, the ability of solubilised zona to induce AR was not diminished in CatSper-null sperm. These findings not only suggest that CatSper is the primary Ca 2C influx pathway in mammalian sperm, but also that in human sperm it may act as a Ca 2C -signalling 'hub' or 'node', such that the effects of diverse agonists are summated/integrated in the rate of Ca 2C influx into the flagellum (Brenker et al. 2012). This is an elegant and simple model for which there is already a significant body of data, but in its basic form it does not address the question of how a sperm can generate and use diverse [Ca 2C ] i signals to control diverse Ca 2C -sensitive functions.
Mouse sperm null for CatSper are unable to hyperactivate (Carlson et al. 2003) and evidence from clinical cases suggests that CatSper is also required for normal levels of motility in human sperm (Avenarius et al. 2009, Smith et al. 2013. Why, then, is manipulation of Ca 2C stores more effective in inducing hyperactivated motility than treatments targeted to CatSper (Alasmari et al. 2013b)? We have proposed that CatSper activation acts as a trigger and consequent elevation of flagellar [Ca 2C ] i stimulates secondary release of stored Ca 2C at the sperm neck, either by stimulating synthesis of IP 3 or by CICR, leading to hyperactivation (Alasmari et al. 2013b). Mathematical modelling of the Ca 2C signals induced by CatSper activation in mouse sperm suggests that diffusion of Ca 2C from the flagellum cannot explain the [Ca 2C ] i increase that occurs at the sperm head upon activation of CatSper and that such a secondary Ca 2C release at the neck region must occut (Olson et al. 2010). Recently we have investigated the occurrence of such secondary responses in human sperm by uncaging Ca 2C in the principal piece of the flagellum. Uncaging induces a clear [Ca 2C ] i transient in the flagellum that decays within 5-10 s. At the neck region of the sperm the transient is truncated and rises more slowly, consistent with diffusion of Ca 2C from the uncaged pool, but in a small proportion of cells (w10%) we have observed a late [Ca 2C ] i response at the neck region, often including multiple peaks (Fig. 2). The low incidence of this secondary Ca 2C -mobilisation is consistent with our observation that, though direct release of stored Ca 2C can induce hyperactivated motility in the majority of human sperm, only a small proportion of cells hyperactivate upon activation of CatSper (Alasmari et al. 2013a,b).
Ca 2C -store-mediated [Ca 2C ] i oscillations occur more readily in sperm incubated for a prolonged period (O24 h) under capacitating conditions (Kirkman-Brown et al. 2004). Capacitation involves generation of ROS and RNS (Herrero et al. 1999, Aitken & Nixon 2013 and we have observed that store mobilisation is sensitised and induced by low concentrations of NO% donors, through a mechanism that involves protein S-nitrosylation (Machado-Oliveira et al. 2008). RyRs were detected in the human sperm nitrosoproteome (Lefievre et al. 2007) and it is well-established that IP 3 Rs and RyRs are sensitised by oxidative stress (Bootman et al. 1992, Sayers et al. 1993, Stoyanovsky et al. 1997, Meissner 2004, Bansaghi et al. 2014) (see 'Ca 2C stores and their regulation'). We propose that CICR from the sperm neck Ca 2C -store is regulated during capacitation, perhaps through the effects of oxidative stress on Ca 2C release channels (Alasmari et al. 2013b) (Fig. 3).

Final remarks
The central role of [Ca 2C ] i signalling in the physiology of mammalian sperm and the pivotal importance of CatSper in this process are well established -mice null for CatSper are infertile (Ren et al. 2001) and in men CatSper lesions are associated with impaired sperm  Susceptibility of the store to CICR is potentially regulated/sensitised by processes occurring during capacitation, including cAMP signalling, oxidative stress and S-nitrosylation as well as Ca 2C store filling and effects of agonists on Ca 2C -store release channels.

R72 J Correia and others
function (Avidan et al. 2003, Avenarius et al. 2009, Smith et al. 2013. The available evidence suggests that Ca 2C -stores also play important roles in both AR and the regulation of motility. Future studies should address the mechanisms by which store mobilisation is achieved (both by CICR and by agonistinduced generation of Ca 2C -mobilising 2nd messengers) and regulated, particularly the significance of capacitation in Ca 2C -store filling and in sensitising Ca 2C release mechanisms. Also, similarly to the important species differences in expression and function of sperm ion channels between human and mouse sperm , Miller et al. 2015, there may be differences in store-regulation and/or function between species. An intriguing possibility is that, at least in human sperm, it may prove possible to bypass the effects on motility of lesions in the expression, function or regulation of CatSper channels by pharmacological activation of stored Ca 2C release.

Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the review.
Funding J Correia was supported by the Wellcome Trust (grant number 086470).