Disruption of protein kinase A localization induces acrosomal exocytosis in capacitated mouse sperm

Protein kinase A (PKA) is a broad-spectrum Ser/Thr kinase involved in the regulation of several cellular activities. Thus, its precise activation relies on being localized at specific subcellular places known as discrete PKA signalosomes. A-Kinase anchoring proteins (AKAPs) form scaffolding assemblies that play a pivotal role in PKA regulation by restricting its activity to specific microdomains. Because one of the first signaling events observed during mammalian sperm capacitation is PKA activation, understanding how PKA activity is restricted in space and time is crucial to decipher the critical steps of sperm capacitation. Here, we demonstrate that the anchoring of PKA to AKAP is not only necessary but also actively regulated during sperm capacitation. However, we find that once capacitated, the release of PKA from AKAP promotes a sudden Ca2+ influx through the sperm-specific Ca2+ channel CatSper, starting a tail-to-head Ca2+ propagation that triggers the acrosome reaction. Three-dimensional super-resolution imaging confirmed a redistribution of PKA within the flagellar structure throughout the capacitation process, which depends on anchoring to AKAP. These results represent a new signaling event that involves CatSper Ca2+ channels in the acrosome reaction, sensitive to PKA stimulation upon release from AKAP.

Mammalian sperm require two post-testicular maturation steps to become fertile: epididymal maturation, occurring in the male epididymis, and capacitation, occurring after ejaculation in the female tract. This process is associated with changes in the sperm motility pattern known as hyperactivation, and with the preparation to undergo the acrosome reaction, an exocytotic process, upon exposure to physiological agonists (1). Capacitation is a highly complex process orchestrated by protein kinase A (PKA) 3 activity (2). Its activation appears to be necessary for: 1) membrane potential (Em) hyperpolarization (3,4), which originates after 15-30 min; 2) development of hyperactivated motility after 30 -45 min; and 3) triggering of the acrosome reaction, once capacitation is complete, i.e. after 60 min for mouse sperm. Even though these processes differ in their kinetics, all of them involve phosphorylation cascades initiated by PKA. Moreover, one of the first signaling pathways observed during capacitation is the HCO 3 Ϫ -dependent stimulation of cAMP synthesis, which rapidly activates PKA (5).
Being PKA, a promiscuous Ser/Thr kinase, its precise activation at specific localizations, known as discrete PKA signalosomes, is necessary (6). However, PKA localization can vary over time, and these changes have been proposed to regulate different processes (7,8). PKA associates to cellular structures through A-Kinase anchoring proteins (AKAPs), which form scaffolding assemblies that play a pivotal role in PKA regulation. AKAPs restrict PKA activity to specific microdomains dependent on the regulation of PKA-AKAP interaction. Most AKAPs also anchor other signaling enzymes such as cyclic nucleotide phosphodiesterases that degrade cAMP and allow PKA activation to be locally regulated in amplitude and duration. Synthetic peptides bearing an amphipathic helix domain that interacts with the PKA regulatory II subunits (PKARII, also known as PRKAR2) impair PKA binding to AKAPs (9). These peptides, commonly used as tools to disrupt subcellular PKA localization, contain an N-terminal modification with a stearic acid moiety that confers membrane permeability. One example of this type of amphipathic peptide is sHT31, a 24-residue peptide that binds to PKARII with high affinity (9) preventing their interaction with AKAPs.
In this article, we used the peptide sHT31 to investigate how disruption of PKA-AKAP interaction affects sperm parameters associated to fertilizing capacity. We observed that the release of PKA from AKAP in capacitated sperm promotes a sudden influx of Ca 2ϩ that starts in the principal piece, and propagates to the head triggering the acrosome reaction. This process was not observed in cells lacking the sperm-specific voltage-dependent Ca 2ϩ channel complex CatSper. In addition, we used three-dimensional stochastic optical reconstruction microscopy (3D STORM) to study spatial distribution of PKA catalytic subunit ␣2 (cPKA, also known as PRKACA2) in mouse sperm, which provides 3D information with resolution of ϳ20 nm (10,11). We observed that upon capacitation, the catalytic subunit redistributes within the flagellum. This movement was completely abrogated when sperm were exposed to sHT31. Altogether, these data indicate that PKA localization and anchoring to AKAPs are actively regulated during sperm capacitation. In addition, our findings show that once the sperm is capacitated, delocalization of the catalytic subunit alters PKA-dependent signaling pathways, elevates intracellular Ca 2ϩ , and, in turn, induces the acrosome reaction. This work provides evidence for the involvement of flagellar CatSper channels in the acrosome reaction that takes place in the head.

sHT31 disrupts sperm capacitation-associated pathways
The exposure of cells to PKA-anchoring disruptor peptides has been shown to promote an ablation of compartmentalized PKA signaling (12). To study the role of PKA anchoring in the sperm capacitation-associated pathways, we used the peptide sHT31, which inhibits the interaction between PKARII and AKAP (9). sHT31 is the permeable stearated form of a short peptide composed of 24 residues (N-stearate-DLIEEAASRIV-DAVIEQVKAAGAY) from the PKA-anchoring domain of AKAP. To disrupt PKA-AKAP interaction, sperm were preincubated for 10 min with different sHT31 concentrations before exposure to conditions that support capacitation. Western blot analysis shows that this preincubation with sHT31 blocked the phosphorylation of PKA substrates associated with sperm capacitation (13), with an IC 50 below 1 M (Fig. 1A left panel). These data indicates that PKA anchoring is crucial for proper PKA activity. Tyrosine phosphorylation, which occurs downstream of PKA activation (2), was also blocked (Fig. 1A, right  panel). As a negative control, a second peptide was used (sHT31p), in which two isoleucine residues are replaced by pro- Figure 1. Inhibition of PKA-AKAP interaction abrogates phosphorylation cascades associated with sperm capacitation. A-C, Western blot analysis with ␣-pPKAs (left panel) and ␣-pY antibodies (right panel) of protein extracts obtained from the following conditions: A, mouse sperm incubated in the absence or presence of increasing sHT31 concentrations for 60 min in capacitating (CAP) or noncapacitating (NC) media. B, sperm were incubated in capacitating medium (CAP) in the absence or presence of either 3 M sHT31 (HT) or 6 M sHT31p (HTp, the inactive sHT31 analogue) as a specificity control. C, sperm incubated for 60 min in CAP containing 1 M sHT31 in the presence or absence of 1 mM Bt 2 cAMP, 0.1 mM IBMX (dbcA) or 1 mM 8Br-cAMP, 0.1 mM IBMX (8Br). D, analysis of PKA activity in cell-free assays. Sperm were incubated for 30 min in medium containing 40 M ATP, 1 mM Bt 2 cAMP (dbcA), or 1 mM 8Br-cAMP (8Br), 0.1 mM IBMX, 10 mM MgCl 2 , and 1% Triton X-100, in the absence or presence of 3 M sHT31 or 15 M sPKI. Negative control lane (Ϫ) does not contain B5 2 cAMP. Each condition was processed for Western blot analysis and immunodetected with ␣-pPKAs antibody and ␣-actin, which was used as a loading control. E, sperm PKA activity was measured as described under "Experimental procedures" using Kemptide as PKA substrate. Kinase buffer contained the indicated amounts of sHT31, 30 M H89, or 15 M sPKI, maintaining a constant concentration of DMSO. Negative control lane (white bar) does not contain Bt 2 cAMP. Data represent mean Ϯ S.E. of three independent experiments performed in triplicates.

PKA relocation in sperm capacitation and acrosome reaction
line residues, rendering this peptide unable to disrupt the PKARII-AKAP interaction (9). This control peptide did not block PKA activity even when used at twice the maximum concentration of sHT31 (6 M) (Fig. 1B).
The effect of sHT31 on capacitation-induced phosphorylation pathways in live sperm suggests that disruption of PKA-AKAP anchoring inhibits the ability of this kinase to find its physiological substrates. However, these results could also be explained by a direct effect on cAMP synthesis or PKA catalytic activity. To evaluate the extent by which the sHT31 effect was due to reduced levels of cAMP, sperm incubated with sHT31 were also treated with the cAMP permeable agonists Bt 2 cAMP or 8Br-cAMP in the presence of the broad-spectrum inhibitor of cyclic nucleotide phosphodiesterases IBMX. In these conditions, the sHT31 peptide still blocked phosphorylation of PKA substrates (Fig. 1C, left panel) and the increase in tyrosine phosphorylation (Fig. 1C, right panel) indicating that the sHT31 target is downstream of cAMP synthesis. To evaluate whether sHT31 has a direct effect on PKA catalytic activity, we used two different approaches. First, we prepared whole cell-free sperm extracts by incubating sperm in 1% Triton X-100 buffer (see "Experimental procedures"). These extracts were treated with Bt 2 cAMP and IBMX in the presence of either 3 M sHT31 or 15 M sPKI (a permeable pseudo-substrate of PKA that specifically inhibits PKA). Western blots revealed that Bt 2 cAMP induced phosphorylation of PKA substrates in cell-free extracts, which was blocked with sPKI, but not with sHT31 (Fig. 1D). Moreover, when PKA activity was measured using radioactive ATP and Kemptide as a substrate (13), sHT31 did not produce an inhibitory effect, even at 30 M (Fig. 1E), whereas two wellestablished PKA inhibitors (H89 and sPKI) blocked Kemptide phosphorylation, as expected. These data ruled out sHT31 offtarget effects on cAMP synthesis and PKA activity.

Disruption of PKA localization blocks sperm functional parameters
The main functional events of sperm acquisition of fertilizing capacity are hyperactivated motility and the preparation for the acrosome reaction. The cAMP/PKA pathway in sperm is one of the initial events of capacitation. To evaluate the effect of dis-rupting the PKA-AKAP interaction on the acquisition of fertilizing capacity, sperm were incubated in the presence of increasing sHT31 concentrations. We observed that whereas sperm motility was not affected by sHT31 concentrations below 3 M ( Fig. 2A), sperm treated with sHT31 were unable to hyperactivate (Fig. 2B), undergo the acrosome reaction (Fig.  2C), and fertilize in vitro (Fig. 2D).

Inhibition of PKA-AKAP interaction in capacitated sperm promotes the acrosome reaction
Capacitation-induced hyperpolarization of the plasma Em has been shown to be essential to prepare sperm for the acrosome reaction (16). In addition, Em hyperpolarization was reported to be dependent on PKA activity (3). Thus, considering the diminished acrosome reaction obtained (Fig. 2C), we addressed the effect of PKA-AKAP inhibition on Em hyperpolarization, using the carbocyanine dye DISC 3 (5) in a fluorimetric population assay (5). Fig. 3A shows that disruption of the PKA-AKAP interaction with sHT31 effectively blocked the capacitation-induced Em hyperpolarization at 1 M. Taking all these results into consideration, we analyzed whether PKA-AKAP interaction is still necessary for the acrosome reaction once Em hyperpolarization is achieved. To this aim, sperm were allowed to capacitate for 60 min, then exposed to 1 M sHT31, and Em was measured at 70, 80, and 90 min. To our surprise, a time-dependent Em depolarization was observed (Fig. 3, B and C). This depolarization correlated to the triggering of the acrosome reaction by sHT31, as seen by cytometric analysis using transgenic mice expressing EGFP in their acrosomal vesicles (Fig. 4, A-C) or by acrosomal staining of WT mice (Fig. 4D). However, the acrosome reaction was only induced by sHT31 (at concentrations 1 and 3 M) in capacitated sperm, but not in noncapacitated sperm, nor when capacitated sperm were exposed to the inactive analogue sHT31p (Fig. 4, C and D). This result points toward the need of active PKA for the induction of the acrosome reaction, indicating that a sudden delocalization of active PKA in capacitated sperm triggers the acrosome reaction. Accordingly, when capacitated sperm were exposed to either 30 M KT5720 or 15 M sPKI, two mechanistically different PKA inhibitors, sHT31 failed to induce the acrosome PKA relocation in sperm capacitation and acrosome reaction reaction (Fig. 4E). These data indicate that when PKA anchoring is blocked in capacitated sperm, PKA activity promotes the acrosome reaction.

Acrosome reaction triggered by inhibition of PKA-AKAP interaction in capacitated sperm correlates with Ca 2؉ influx through CatSper channels
Pharmacologically hyperpolarized noncapacitated sperm undergo the acrosome reaction when depolarized (e.g. addition of high K ϩ concentrations), or exposed to solubilized zona pellucidae (16). In view of these findings, we questioned the relationship between Em depolarization triggered by sHT31 (Fig.  3B) and the onset of acrosome reaction (Fig. 4C). Two hypotheses arose: 1) depolarization triggers the acrosome reaction; and 2) depolarization is the consequence of a Ca 2ϩ influx, which triggers the acrosome reaction. To test the first hypothesis, noncapacitated sperm were pharmacologically hyperpolarized with valinomycin before being exposed to sHT31. As seen in Fig. 5A, sHT31 was not capable of inducing the acrosome reaction on hyperpolarized noncapacitated sperm. Furthermore, when sperm were capacitated in medium with valinomycin and 70 mM extracellular K ϩ , abrogating sperm Em hyperpolarization (16), sHT31 was still able to trigger the acrosome reaction (Fig. 5A). In a control experiment, capacitated sperm clamped at depolarized Em lost the ability to undergo the acrosome reaction when challenged with progesterone (Fig.  5A). These results imply that Em hyperpolarization is neither necessary nor sufficient for sHT31 to induce the acrosome reaction, and argues against the first hypothesis. To test the second hypothesis, we exposed capacitated sperm to the Ca 2ϩ channel blocker mibefradil before adding sHT31. This procedure blocked the depolarization (Fig. 5B) and the acrosome reaction ( Fig. 5C) induced by sHT31, supporting the second hypothesis. Moreover, when 3.5 mM EGTA (free Ca 2ϩ of 138 nM as measured with MaxChelator software (17)) was added before sHT31, the acrosome reaction was also blocked (Fig.  5C), further substantiating the role of Ca 2ϩ influx as the cause of depolarization and acrosome reaction.
We performed intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) measurements of Fluo3-AM-loaded spermatozoa on laminincoated coverslips, to allow analysis of live sperm where the flagellum remains free and moving. Fig. 6 shows that [Ca 2ϩ ] i increased upon disruption of PKA-AKAP interaction, only in capacitated sperm ( Fig. 6 and Movie S1). As expected, this [Ca 2ϩ ] i increase was absent in noncapacitated sperm ( Fig. 6 and Movie S2) and in capacitated sperm exposed to the inactive analogue sHT31p ( Fig. 6 and Movie S3). Moreover, 5 M mibefradil affected the amplitude of the signal, further supporting the hypothesis that a Ca 2ϩ channel (either T-type or CatSper) is involved in the Ca 2ϩ entry ( Fig. 6 and Movie S4). In agreement with the inhibition of acrosome reaction by PKA inhibitors, when sperm were preincubated with sPKI for 10 min before exposure to sHT31, the [Ca 2ϩ ] i increase caused by PKA-AKAP disruption was absent ( Fig. 6 and Movie S5).
To analyze [Ca 2ϩ ] i in the different cellular domains we performed measurements of [Ca 2ϩ ] i using concanavalin A-coated coverslips (attaching the whole cell to the coverslip) and increased frame rates (10 Hz, 100 ms/frame). The [Ca 2ϩ ] i increase was first evidenced at the principal piece, followed by an increase in the mid-piece and later propagation to the head, as shown by a representative Ca 2ϩ kymograph in Fig. 7A and Movie S6. The mean of Ca 2ϩ fluctuation in the different subregions is shown in Fig. 7B, indicating the delays of Ca 2ϩ signal propagation. Considering that CatSper channel locates to the principal piece (18), and is blocked by low mibefradil concentrations, these results substantiate the role of CatSper in sHT31-promoted Ca 2ϩ influx. Accordingly, CatSper is the only Ca 2ϩ channel in sperm that is supported by solid electrophysiological evidence and the Ca 2ϩ channel-blocker mibefradil has been shown to inhibit CatSper currents (19). As shown by previous works, when extracellular medium is deprived of Ca 2ϩ and Mg 2ϩ , CatSper becomes permeable to Na ϩ , which can be clearly evidenced by the Na ϩ -dependent depolarization upon Ca 2ϩ chelation (20,21). The magnitude of this depolarization depends on the extent of CatSper opening. Therefore, by studying sperm Em in population assays using the fluorophore DISC 3 (5), we addressed whether PKA-AKAP disruption triggers CatSper opening after EGTA addition. When capacitated sperm loaded with DISC 3 (5) were first exposed to EGTA and then to 1 M sHT31, CatSper opening was evidenced by Em depolarization (Fig. 7, C and D). When 0.1 M sHT31 was used, depolarization was observed to a lesser extent, whereas no depolarization was detected when noncapacitated sperm were exposed to 1 M sHT31, consistent with the requirement of active PKA upon inhibition of AKAP anchoring. To further demonstrate the role of CatSper in the Ca 2ϩ influx conductive

PKA relocation in sperm capacitation and acrosome reaction
to the acrosome reaction, CatSper KO mice were used. sHT31 failed to trigger acrosome reaction in CatSper KO (Fig. 7E). As a control, WT siblings (Ϫ/ϩ) underwent normal acrosome reaction induced by sHT31. It is worth noticing that the Ca 2ϩ ionophore A23187 promoted normal acrosome reaction both in WT and CatSper KO sperm.

PKA localization changes revealed by super-resolution imaging
To analyze PKA localization dynamics during capacitation, 3D STORM was used, enabling the study of spatial distribution of PKA catalytic subunit ␣2 (cPKA, also known as PRKACA2) in mouse sperm. This type of microscopy allows 3D reconstruction with resolution of ϳ20 nm (10,11), to study protein distribution within the narrow sperm tail, which is less than 1 m in diameter. The spatial distribution of cPKA was determined in both noncapacitated and capacitated sperm. As shown in Fig. 8A, noncapacitated sperm exhibited a dense and compact rod-like radial distribution of cPKA. The facilitative glucose transporter 3 (GLUT3), which is present throughout the flagellar plasma membrane (22), showed a continuous ring staining cross-section. This was used as a control, showing a distribution compatible with previous reports (23) (Fig. 8, A and  B) and validating the structural information obtained by 3D STORM.
Interestingly, upon capacitation, cPKA localized to a broader cylinder-like distribution. Fig. 8B shows the radial localization of cPKA in capacitated and noncapacitated sperm as well as the distribution of GLUT3. The distribution of cPKA in sperm incubated in capacitating media showed a shoulder at ϳ110 nm that was not present in noncapacitated sperm. Fig. 8C shows the distribution of cPKA for individual cells that were incubated only in capacitating media, where the effect is much more evident. As already reported, sperm in capacitating conditions consist of two heterogeneous populations (3), half of the cells capacitate and exhibited a distribution shift from 70 to 110 nm,

PKA relocation in sperm capacitation and acrosome reaction
whereas the other half of the cells showed a distribution of cPKA that is not distinguishable from the cells in noncapacitating media. Addition of 1 M sHT31 to capacitated sperm promoted a retraction of cPKA to its original dense and compact rod-like radial distribution, clearly showing by microscopy for the first time that disruption of PKA-AKAP interaction alters PKA localization. To further confirm sHT31 as a disruptor of PKA-AKAP interaction, PKARII immunoprecipitations from capacitated sperm in the absence or presence of sHT31 were performed. This experiment aimed to analyze whether PKARII pulls down AKAP4, the most abundant AKAP in the principal piece where PKA localizes. As shown in Fig. 8E, the amount of AKAP4 that co-immunoprecipitates with PKARII was reduced in the presence of sHT31 (Fig. 8E), further supporting that sHT31 disrupts PKA-AKAP interaction.

Discussion
For many years, it has been accepted that molecules involved in cAMP signaling (i.e. phosphodiesterases and adenylyl cyclases), as well as its effectors and targets are carefully compartmentalized in sperm (2). This compartmentalization helps assure a tight regulation of different sperm processes associated with capacitation. Although some biological effects of capacitation occur in the head (e.g. acrosome reaction), others occur in the tail (e.g. hyperactivation). Biochemical immunolocalizations, supported by enzymatic assays, have conclusively demonstrated that PKA in mouse sperm is restricted to the flagellum (24). Further subcompartmentalization would be allowed by local cAMP generation within independently regulated microdomains. These cAMP modules could be in turn modulated by AKAPs, which dock not only PKA but also enzymes and substrates to specific cell locations (25). In this study, we demonstrated that when PKA-AKAP interaction was disrupted by addition of sHT31 before incubation in capacitating conditions, both HCO 3 Ϫ (that stimulates ADCY10) and permeable cAMP analogues failed to trigger capacitation, due to impaired in vivo PKA activity, as analyzed by Western blotting. This result indicates that PKA needs proper localization to start capacitation events.
Surprisingly, disruption of active PKA from AKAP triggered the acrosome reaction, but only in capacitated sperm. On the contrary, noncapacitated sperm did not undergo exocytosis, indicating that disruption of PKA-AKAP interaction can only induce acrosome reaction in cells that have gained fertilizing capacity, associating this phenomenon to a particular physiological state of the cells, rather than to an off-target effect. In addition, the inactive analogue sHT31p did not induce any changes in either capacitated or noncapacitated sperm, further supporting the use of this tool to study specific AKAP disruption effects. The acrosome reaction, triggered by the blockade of PKA-AKAP interaction, involved influx of extracellular Ca 2ϩ that started in the principal piece followed by propagation to the mid-piece and the head. Our data substantiate the role of CatSper channels in this Ca 2ϩ influx as follows: 1) Ca 2ϩ influx started in the principal piece, where CatSper localizes (26), excluding initiation through Cav3.2 or Cav2.3 channels, which localize in the head (27) where Cav2.3 plays a role in the acrosomal reaction (28); 2) pharmacological inhibition of CatSper abolished both Ca 2ϩ influx and sHT31-induced acrosome reaction; 3) sperm from CatSper KO mice did not undergo acrosome reaction when challenged with sHT31, but exhibited normal acrosome reaction upon Ca 2ϩ ionophore stimulation. This work represents the first report showing that a Ca 2ϩ influx through CatSper is conclusively linked to the acrosome reaction. Strikingly, we found that activation of CatSper was promoted by delocalized PKA activity, because blocking the kinase before impairing PKA-AKAP interaction abolished both the acrosome reaction and Ca 2ϩ influx. The potential contribution of other proteins to the effects seen upon addition of sHT31 cannot be ruled out, because R2D2 domains, which bind AKAPs, are also found in ROPN1, ASP, SP17, and CABYR (29). However, the fact that the induction of the acrosome reaction by sHT31 was completely blocked by a PKA-specific inhibitor (sPKI) as well as by KT5720, shows that this kinase is a key player in the induction mechanism. Thus, it is compelling to think that this could strike as a novel mechanism for in vivo acrosome reaction induction, where sperm cells are exposed to hormones (such as progesterone) and prostaglandins, which could trigger PKA re-localization.
Mice sperm are endowed with the sperm-specific catalytic PKA subunit C␣2 (30), lacking myristoylated PKA catalytic subunit C␣1 (31), which targets to the plasma membrane. Absence of this isoform abolishes canonical PKA signaling events on the membrane. Then, it is possible that anchoring of PKA to different AKAPs restricts the activity of C␣2 to specific locations. Using 3D STORM we show that in noncapacitated sperm, inactive cPKA localizes to a narrow compact rod along the flagellum with a mean radius of 70 nm. However, after capacitation, cPKA acquired a mean radius of 105 nm. Upon disruption of the PKA-AKAP interaction with sHT31 in capacitated sperm, the cPKA subunit returned to its original location, indicating that PKA anchoring to AKAP is required for re-localization of the enzyme throughout capacitation. This represents the first clear evidence that PKA localization changes during this process.
The regulation of CatSper by PKA awaits further work but opens the possibility to new regulation points of CatSper during

PKA relocation in sperm capacitation and acrosome reaction
sperm capacitation and acrosome reaction. It could be hypothesized that PKA anchoring prevents premature acrosome reaction. However, it could also be possible that in vivo, the acrosome reaction is regulated by modification of the PKA-AKAP affinity. To date, apart from the use of high nonphysiological progesterone concentrations, no conclusive physiological inducer of the acrosome reaction has arisen for the mouse model (32). Recent reports have demonstrated that sperm undergo acrosome reaction in the upper segments of the oviduct, prior to contacting the zona pellucida or the progesterone-rich cumulus matrix (33)(34)(35). This new pathway where a sudden impairment of the PKA-AKAP interaction triggers the acrosome reaction through CatSper Ca 2ϩ influx has the potential to open new courses of research.

Mouse sperm preparation
Cauda epididymal mouse sperm were collected from C57BL/6 young adult male mice (8 -13 weeks old) and sacrificed under supervision of the Animal Care and Use Committee of the Facultad de Ciencias Bioquímicas y Farmacéuticas de Rosario (UNR) (protocols approved numbers 7298/532). Each minced cauda epididymis was placed in 500 l of a modified Krebs-Ringer solution H-TYH HEPES-buffered medium containing 119.3 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2 PO 4 , 1.2 mM MgSO 4 , 5.6 mM glucose, 0.5 mM sodium pyruvate, 1.7 mM Ca 2ϩ , and 20 mM HEPES (pH 7.4). The H-TYH medium accounts for noncapacitating medium. After 20 min, epididymides were removed, and the suspension was adjusted with noncapacitating (NC) medium to a final concentration of 1-2 ϫ 10 7 cells/ml. For capacitation (CAP medium), BSA and NaHCO 3 were added to final concentrations of 5 mg/ml and 15 mM, respectively. Sperm were then incubated at 37°C for 60 min. To test the effect of the different inhibitors on capacitation, sperm were preincubated with the respective reagents in noncapacitating medium for 10 min prior to the beginning of the capacitating period.

SDS-PAGE and immunoblotting
Sperm were incubated under either noncapacitating or capacitating conditions for 60 min in the presence or absence of different concentrations of sHT31. When the experiment required it, sperm were also incubated with 1 mM Bt 2 cAMP ϩ 0.1 mM IBMX or 1 mM 8Br-cAMP ϩ 0.1 mM IBMX (36). After treatment, sperm were twice collected by centrifugation and washed in 1 ml of TBS (150 mM NaCl, 20 mM Tris, pH 7.4). Afterward, cell pellets were re-suspended in Laemmli sample buffer (37) without ␤-mercaptoethanol, vortexed for 5 s, and boiled for 3 min. After centrifugation, 5% ␤-mercaptoethanol was added to the supernatants and boiled again for 5 min. Protein extracts equivalent to 1-2 ϫ 10 6 sperm/lane were subjected to SDS-PAGE and electrotransferred to PVDF membranes (Bio-Rad) at 250 mA for 60 min on ice. Membranes were blocked with 3% BSA (Sigma) in TBS containing 0.1% Tween 20 (T-TBS). Antibodies were diluted in T-TBS as follows: ␣-pY 1/10,000, ␣-pPKAs 1/3,000, ␣-PKARII 1/3,000, and ␣-AKAP4 1/3,000. Secondary antibodies were diluted 1/15,000 in T-TBS and developed using an enhanced chemiluminescence detection kit (ECL Plus, Amersham Biosciences, GE Healthcare) according to the manufacturer's instructions. When necessary, PVDF membranes were stripped at 60°C for 15 min in 2% SDS, 0.74% ␤-mercaptoethanol, and 62.5 mM Tris (pH 6.5) and washed six times for 5 min each time in T-TBS. In all experiments, molecular masses were expressed in kilodaltons (kDa) (38).

PKARII and AKAP4 co-immunoprecipitation
Sperm were capacitated for 90 min. When indicated, 3 M sHT31 was added at 60 min, and maintained throughout incubations. Samples were centrifuged at 1,700 ϫ g for 1 min. The resulting pellet was resuspended in RIPA buffer (10 mM Tris-HCl, pH 7.2, 50 mM NaCl, 0.1% SDS, 1% Triton X-100, 1 mM EDTA, 20 mM sodium fluoride, 40 mM glycerol 3-phosphate and protease inhibitors), incubated on ice for 30 min, and centrifuged at 4°C for 5 min at 2500 ϫ g. Supernatants containing 1 ϫ 10 8 cells in a final volume of 500 l were incubated with 3 g of ␣-PKARII antibody for 2 h at room temperature (23-25°C) with constant rocking. After the addition of 30 l of protein G-Sepharose (GE Healthcare), the reactions were further rocked for 1 h at room temperature. The immune complex was recovered by centrifugation, washed four times in RIPA buffer, and subjected to SDS-PAGE and Western blotting. To avoid immunoreactive signals from IgG heavy chains, rabbit and mouse ␣-IgG light chain were used.

Mouse eggs collection and in vitro fertilization assays
Female mouse eggs arrested in metaphase II were collected from 6-to 8-week-old super-ovulated CD1 female mice (Charles River Laboratories) at 13 h after human chorionic gonadotrophin (Sigma) intraperitoneal injection. Cumulus cells were removed by brief incubation (Ͻ5 min) in H-TYH medium containing 7 mM NaHCO 3 , 5 mg/ml of BSA, and 0.02% type IV-S hyaluronidase (Sigma). After cumulus cells removal, eggs were placed in a drop of H-TYH medium containing 22 mM NaHCO 3 and 5 mg/ml of BSA and allowed to recover for 30 min in an incubator with 5% CO 2 at 37°C. Sperm were capacitated in media containing different concentrations of sHT31 for 60 min, and concentrated by centrifugation at 600 ϫ g for 3 min. Fertilization drops (200 l each) containing 10 -20 eggs were inseminated with treated sperm from CD1 male mice (final concentration of 2.5 ϫ 10 6 sperm/ml). Drops contained 0.1 M sHT31, which did not affect fertilization. After 4 h of insemination, eggs were washed through brief passages in three drops of H-TYH medium containing 22 mM NaHCO 3 and 15 mg/ml of BSA using a thin bore pipette to detach any loosely attached sperm. After 3 h of further incubation, eggs were fixed with 3.7% paraformaldehyde/PBS for 15 min, washed, and stained with Hoechst 33342 (Sigma, 10 g/ml) in PBS for 10 min at room temperature. Fertilization was assessed by visualization of the formation of the male and female pronuclei.

Membrane potential assay in cell populations
After sperm treatment, cells were collected by centrifugation (700 ϫ g, 5 min) and concentration was adjusted to 2.7 ϫ 10 6 sperm/ml. Then, sperm were loaded with 1 M of the membrane potential-sensitive dye DISC 3 (5) (Molecular Probes) for 5 min. No mitochondrial un-couplers were used because their contribution to the resting potential has been determined to be

PKA relocation in sperm capacitation and acrosome reaction
insignificant (39). The sperm were transferred to a gently stirred cuvette at 37°C, and the fluorescence was monitored with a Varian Cary Eclipse fluorescence spectrophotometer at 620/670 nm excitation/emission wavelengths. Recordings were initiated when steady-state fluorescence was reached and calibration was performed at the end of each measure by adding 1 M valinomycin and sequential additions of KCl as previously described (40). Sperm Em was obtained from the initial fluorescence (measured as arbitrary fluorescence units) by linearly interpolating it in the theoretical Em values for the calibration curve against arbitrary fluorescence units of each trace. This internal calibration for each determination compensates for variables that influence the absolute fluorescence values.
For experiments where CatSper opening was assessed, extracellular Ca 2ϩ was chelated using EGTA allowing sodium influx through CatSper to occur. The magnitude of the depolarization caused by Na ϩ influx relates to the extent of channel opening. Cells were loaded with DISC 3 (5) and fluorescence was recorded as detailed above. Calcium was chelated with 3.5 mM EGTA to a value of free calcium of 138 nM (MaxChelator) (17) followed by sHT31 addition. The fluorescence change after and before addition of sHT31 was presented as (F max Ϫ F)/(F Ϫ F 0 ), where F represents fluorescence intensity after EGTA addition, F 0 is the mean of 1 min of acquisition before addition of EGTA, and F max represents fluorescence intensity after sHT31 addition (20,21).

Stochastic optical reconstruction microscopy (3D STORM)
After sperm incubation in the appropriate condition (NC, CAP, or CAP plus sHT31), they were washed twice with NC media by centrifugation at 2 ϫ 2,000 rpm for 5 min and finally resuspended in NC media. Sperm were seeded in poly-L-lysinecoated coverslips (Corning number 1.5), air-dried for 10 min, fixed with 4% (v/v) fresh paraformaldehyde in PBS for 10 min at room temperature, and followed by 3 ϫ 5-min PBS washes. Cells were then permeabilized with 0.5% (v/v) Triton X-100 in PBS for 5 min at room temperature and washed 3 ϫ 5 min with PBS. Samples were blocked with 3% (w/v) BSA/PBS for 1 h at room temperature and then incubated with ␣-cPKA antibody (1:200) or anti-GLUT3 antibody (1:100) diluted in 1% (w/v) BSA/PBS, overnight at 4°C in a humidifier chamber. Cells were washed with PBS containing 0.1% (v/v) Tween 20 (T-PBS) at 3 ϫ 5 min, and further incubated with Alexa Fluor 647-conjugated anti-mouse (1/500) or rabbit (1/1,000) secondary antibody diluted in 1% BSA/PBS for 1 h at room temperature. Cells were then washed with T-PBS twice and PBS once for 5 min each, and immediately mounted in STORM imaging buffer (50 mM Tris-HCl, pH 8, 10 mM NaCl, 0.56 mg/ml of glucose oxidase, 34 g/ml of catalase, 10% (w/v) glucose, and 1% (v/v) ␤-mercaptoethanol). Images were acquired using Andor IQ 2.3 software in a custom-built microscope equipped with an Olympus PlanApo ϫ100 NA1.45 objective and a CRISP ASI autofocus system (41,42). Alexa Fluor 647 was excited with a 638-nm laser (DL638-050, CrystaLaser, Reno, NV) under continuous illumination. Initially the photoswitching rate was sufficient to provide a substantial fluorophore density. However, as fluorophores were irreversibly photobleached, a 405-nm laser was introduced to enhance photoswitching. The intensity of the 405-nm laser was adjusted in the range 0.01-0.5 milliwatts to maintain an appropriate density of active fluorophores. Axial localization was achieved via astigmatism using a MicAO 3DSR adaptive optics system (Imagine Optic, Orsay, France) inserted into the emission pathway between the microscope and the EMCCD camera (43,44). This technology achieves 3D localization via shaping the axial point spread function, allowing both the correction of spherical aberrations and the introduction of astigmatism. User-controlled introduction of astigmatism enabled the accurate detection of single-molecules over a thickness of 1 m and, in turn, 3D reconstruction (45). A calibration curve for axial localization was generated with 100-nm Tetra-Speck microspheres (Invitrogen) immobilized on a coverslip (46). The images were acquired in a water-cooled, back-illuminated EMCCD camera (Andor iXon DU-888) operated at Ϫ 85°C at a rate of 23 frames/s. 50,000 frames were collected to generate a super resolution image. Single-molecule localization, drift correction using image cross-correlation and reconstruction were performed with Thunder STORM (47).
To find the molecular radial distributions, we selected regions of interest of the flagellum that were found to lie in a straight line. The center of the flagellar cross-section was first found by Gaussian fitting of the localization histograms along x and y. The coordinates of the localized molecules were then transformed into cylindrical coordinates to obtain the radial position r.

Cell-free assay of PKA substrate phosphorylation
Sperm cells (5 ϫ 10 6 in 60 l of final volume) were incubated in the presence of different inhibitors (3 M sHT31 or 15 M sPKI) for 30 min at 37°C in H-TYH medium supplemented with: 1% Triton X-100, 40 M ATP, 1 mM Bt 2 cAMP, 0.1 mM IBMX, Sigma protease inhibitor mixture, 100 M sodium orthovanadate, 5 mM p-nitrophenyl phosphate, 40 mM ␤-glycerophosphate, and 10 mM MgCl 2 . The presence of Triton X-100 in the assay buffer is responsible for the cell lysis, rendering a "cell-free" assay, and differentiates the experiment from the regular incubation procedure (13). Samples were further subjected to SDS-PAGE and Western blotting followed by immunodetection with an anti-pPKA substrates antibody (clone 100G7E), as described above.

PKA in vitro activity by radioactive assay
Sperm in H-TYH medium were centrifuged at 10,000 ϫ g for 3 min and re-suspended in buffer containing 25 mM Tris-HCl, 150 mM NaCl, protease inhibitor mixture (Roche Applied Science) and 1% Triton X-100 (pH 7.4). Suspensions were incubated on ice for 30 min and then centrifuged at 10,000 ϫ g for 10 min at 4°C. The supernatant was discarded, and the pellet (Triton insoluble fraction) was re-suspended in 330 l of the same buffer. This Triton-insoluble fractions were used as source of PKA and its activity was measured as the amount of radioactive phosphate (␥-32 P) incorporated into "Kemptide" (Sigma), which is a specific and synthetic PKA-substrate. The assay mixture was either supplemented or not with inhibitors as described under "Results," while maintaining constant DMSO concentration (Յ0.5%).

Sperm motility analysis
Sperm suspensions were loaded on a 20 M chamber slide (Leja Slide, Spectrum Technologies) and placed on a micro-PKA relocation in sperm capacitation and acrosome reaction scope stage at 37°C. Sperm movements were examined using the CEROS computer-assisted semen analysis (CASA) system (Hamilton Thorne Research, Beverly, MA). Parameters used were as follows: 30 frames acquired, frame rate of 60 Hz, minimum cell size of 4 pixels, low average path velocity cutoff of 5 mm/s, static head size of 0.2-2.99, static head intensity of 0.26 -1.31, and static head elongation lower than 100. At least 20 microscopy fields corresponding to a minimum of 200 sperm were analyzed in each experiment. Hyperactivated sperm were classified according to CASAnova (48).

Acrosomal status assays
After the appropriate treatment, sperm were seeded on 8-well glass slides and air-dried for 15 min. Then, sperm cells were fixed with 3.7% paraformaldehyde in phosphate-buffered saline (PBS, 137 mM NaCl, 10 mM phosphate, 2.7 mM KCl, pH 7.4) for 15 min at room temperature, washed with PBS (four times of 5 min each time), and permeabilized with 0.5% Triton X-100 for 5 min. After washing with PBS, cells were treated with 10% BSA in PBS for 1 h at room temperature and then incubated with Alexa Fluor 488-conjugated peanut agglutinin (1/500) in PBS containing 1% BSA for an extra hour at room temperature. Before mounting with Slow-Fade Light reagents (Molecular Probes), samples were washed with PBS (four times for 5 min each time). Epifluorescence microscopy was performed using a BH2 Olympus microscope. Differential interference contrast images were taken in parallel and they served as the control for sperm morphology.
The acrosomal status was also evaluated by flow cytometry using sperm from transgenic male mice (BDF1-Tg (CAG-mtDsRed2, Acr-EGFP) RBGS0020sb), displaying acrosomal vesicles expressing green EGFP fluorescence and mid-pieces (mitochondria) expressing red Ds-Red2 fluorescence (49). Sperm were analyzed on a flow cytometer (FACSCanto II flow cytometer, BD Bioscience) after adding propidium iodide to the sperm suspension to discriminate dead cells. A 515-545-nm band path filter and 650-nm long path filter were used for GFP and propidium iodide, respectively.

Live imaging of intracellular Ca 2؉ levels
After capacitation, sperm were incubated with 2 M Fluo3-AM (Molecular Probes) in 0.05% Pluronic acid for 30 min at 37°C in the dark (15). Then, sperm were washed once at 700 ϫ g for 5 min and re-suspended in capacitating medium. Once loaded, sperm were immobilized on mouse laminin (0.1 mg/ml)-coated coverslips to allow recordings. The chamber was filled with recording medium (noncapacitating H-TYH) in the absence or presence of 15 M sPKI or 5 M mibefradil.
All recordings were made at 37°C and Ca 2ϩ imaging was performed before, during, and after addition of recording medium with or without sHT31/sHT31p. Ionomycin (10 M) was added at the end of each experiment as a control for cell viability/response, and for posterior normalization. A Zeiss LSM880 scan head on an axio observer Z1 inverted microscope with a ϫ60 1.4 NA oil immersion objective was used. Excitation was performed at 488 nm using an argon ion laser, combined with a GaAsP spectral detector with 508 -588 nm bandwidth. Bidirectional scanning was used with a dwell time of 1.03 s at 512 ϫ 512 pixels resulting in an acquisition of 2-3 frames/s for periods of 8 -10 min. Movies were processed and analyzed in ImageJ (version 1.38, NIH). Regions of interest (ROIs) were drawn on each sperm for fluorescence quantification. The Fluo3 [Ca 2ϩ ] i changes are presented as (F Ϫ F 0 )/(F max Ϫ F 0 ) ϫ 10, where F represents fluorescence intensity at time t, and F 0 is the mean of 1 min of acquisition before any addition. Response to sHT31 or sHT31p for each sperm was normalized against its own ionomycin response (100% response). Mean response of all cells in the experiment was used to compare responses between experiments, and n values represent the number of experiments. Cells with peak [Ca 2ϩ ] i changes of Ͼ15% relative to the ionomycin changes were counted as responsive. At least 40 cells were analyzed each time, using 3 male mice.
For determination of calcium kinetics, sperm were loaded with Fluo3-AM as described above. Afterward, they were attached to concanavalin A (1 mg/ml)-coated coverslips (instead of laminin) to adhere sperm cells through their head and tail (14). Fluorescence signals were registered with an Eclipse TE300 Nikon inverted microscope using a LED-based pulsed light excitation system and an oil immersion ϫ60 objective (PlanApo N, numerical aperture 1.42). Recordings where captured with AndorIxon-DU8970 EMCCD camera, controlled by iQ software (Andor Technology, Belfast, UK), at 10 images/s with a 2-ms pulse excitation using a single color Cyan LED. To select the desired excitation/emission wavelength we used a bandpass excitation (HQ 485/ 25X) filter, dichroic mirror (Q-505 LP), and emission (HQ 510 LP) filters (Chroma Technology, Bellows Falls, VT). Image acquisition was done at 10 frames/s for periods no longer than 5 min. Steady initial fluorescence was recorded for 30 s, then 1 M sHT31 was added to the recording chamber and fluorescence was registered for 3-4 min more. Ionomycin (10 M) was added at the end of the recordings as a control of cell response. Using ImageJ software, we determined separate ROIs for sperm head, midpiece, and principal piece. Fluorescence from each ROI was background-corrected and normalized using ⌬F ϭ ((F Ϫ F 0 )/ F 0 ) ϫ 100%, where ⌬F is the percentage change in intensity. For each trace, we then determined the time when fluorescence started to rise and estimated the delay in time-response for each part of the sperm.

Statistical analysis
Paired Student's t test was used to compare mean values between control and tested groups. The difference between mean values of multiple groups was analyzed by one-way analysis of variance followed by Holm-Šidák test. Statistical significances are indicated in the figure legends. At least 100 sperm were analyzed in each case, unless specified.