Novel pharmacological actions of trequinsin hydrochloride improve human sperm cell motility and function

Background and Purpose Asthenozoospermia is a leading cause of male infertility, but development of pharmacological agents to improve sperm motility is hindered by the lack of effective screening platforms and knowledge of suitable molecular targets. We have demonstrated that a high‐throughput screening (HTS) strategy and established in vitro tests can identify and characterise compounds that improve sperm motility. Here, we applied HTS to identify new compounds from a novel small molecule library that increase intracellular calcium ([Ca2+]i), promote human sperm cell motility, and systematically determine the mechanism of action. Experimental Approach A validated HTS fluorometric [Ca2+]i assay was used to screen an in‐house library of compounds. Trequinsin hydrochloride (a PDE3 inhibitor) was selected for detailed molecular (plate reader assays, electrophysiology, and cyclic nucleotide measurement) and functional (motility and acrosome reaction) testing in sperm from healthy volunteer donors and, where possible, patients. Key Results Fluorometric assays identified trequinsin as an efficacious agonist of [Ca2+]i, although less potent than progesterone. Functionally, trequinsin significantly increased cell hyperactivation and penetration into viscous medium in all donor sperm samples and cell hyperactivation in 22/25 (88%) patient sperm samples. Trequinsin‐induced [Ca2+]i responses were cross‐desensitised consistently by PGE1 but not progesterone. Whole‐cell patch clamp electrophysiology confirmed that trequinsin activated CatSper and partly inhibited potassium channel activity. Trequinsin also increased intracellular cGMP. Conclusion and Implications Trequinsin exhibits a novel pharmacological profile in human sperm and may be a suitable lead compound for the development of new agents to improve patient sperm function and fertilisation potential.


| INTRODUCTION
Asthenozoospermia (low sperm motility) has been reported as the leading cause of male infertility (Kumar & Singh, 2015).
Intracytoplasmic sperm injection (ICSI) is the most common and successful treatment for male infertility. While it is a pragmatic solution, it involves invasive treatment of the female partner and bypasses all natural sperm selection processes. There are concerns that ICSI may be associated with long-term health issues for the children born, particularly in cases where the spermatozoa are predominately immotile and do not have the capacity to fertilise under natural conditions (Esteves, Roque, Bedoschi, Haahr, & Humaidan, 2018;Hanevik, Hessen, Sunde, & Breivik, 2016). Therefore, the development of novel direct treatments for male infertility is desirable, although this represents a significant challenge because of the limited understanding of the regulation of normal and dysfunctional sperm (Barratt et al., 2017).

Intracellular calcium concentration ([Ca 2+ ] i ) is an established regu-
lator of sperm function, and a wealth of evidence suggests that the principal cation channel in sperm (CatSper) influences sperm function and fertilisation potential through regulation of extracellular calcium influx (Singh & Rajender, 2015;Strünker et al., 2011;Tamburrino et al., 2014;Williams et al., 2015). CatSper is confined to the principal piece of the flagellum and is modulated by intracellular pH (pHi) and membrane potential. It is sensitive to progesterone (Lishko, Botchkina, & Kirichok, 2011;Strünker et al., 2011), which stimulates cell penetration into a viscous medium (used as an in vitro model for regions of the female reproductive tract; Barratt & Publicover, 2012). [Ca 2+ ] i also plays a significant role in the regulation of soluble cyclases that drive the production of cyclic nucleotides. These key secondary messengers have been shown to be fundamental for human sperm cell motility, cell capacitation, and acrosome reaction. Cyclic nucleotides are actively enzymatically degraded by PDEs, and PDE inhibitors can positively affect sperm cell motility and function (Maréchal et al., 2017;Tardif et al., 2014;Willipinski-Stapelfeldt et al., 2004).
Identifying CatSper agonists to improve sperm motility and function is a logical approach to drug discovery for male infertility. We have previously described the development of a high-throughput screening (HTS) system to identify compounds that increase [Ca 2+ ] i and thereafter have assessed the functional consequence of in vitro application of two compounds (Martins da Silva et al., 2017). However, sperm motility is multiform and adaptive, and not every patient sample responded to treatment in vitro. As such, there remains a clear need to continue to identify potential therapeutic compounds.
In this study, we hypothesised that novel CatSper agonists could be identified by screening a library of small molecules with defined molecular targets (chemogenomic library). This library was assembled from well-characterised, commercially available ligands (Tocris) for a range of validated drug targets including enzymes, receptors, and transporters. We demonstrate that trequinsin hydrochloride, a PDE3 inhibitor (PDE3i; Degerman, Belfrage, & Manganiello, 1997;Lal, Dohadwalla, Dadkar, D'Sa, & de Souza, 1984), is highly effective at inducing

What is already known
• There is an unmet clinical need for compounds to treat asthenozoospermia (poor sperm motility).

What this study adds
• Trequinsin hydrochloride raised intracellular calcium and cyclic GMP in human sperm and improved motility.
What is the clinical significance • Trequinsin hydrochloride has clinically relevant positive effects on human sperm motility.
• Thus, trequinsin hydrochloride has the potential to be a novel treatment for male infertility.
F I G U R E 1 Experimental plan. Systematic functional and mechanistic screening strategy for the identification of the molecular and functional effects of trequinsin an increase in [Ca 2+ ] i , which corresponded with improved sperm motility. Detailed characterisation of the mechanism of action of trequinsin suggests that these effects are achieved through complex and novel pharmacological activities in human spermatozoa.
The study aimed to investigate hit compounds from a chemogenomic drug library screen for effects on sperm motility and to determine the mechanism responsible. This was achieved in three phases. Phase 1 employed HTS of compounds for their ability to increase sperm [Ca 2+ ] i relative to a saturating concentration of progesterone (P4). Phase 2 involved detailed sperm function tests, and Phase 3 involved molecular analysis of trequinsin hydrochloride, which was selected due to its high efficacy in Phase 1 and its purported PDE3i activity. An outline of the experimental approach is shown in Figure 1 (Cooper et al., 2010) were used in this study under the same ethical approval. All obtained samples for research were analysed in line with suggested guidance for human semen studies where appropriate (Björndahl, Barratt, Mortimer, & Jouannet, 2015).

| Preparation of donor and patient sperm samples
All donors and patients adhered to an abstinence period of 2-5 days before sample collection by masturbation into a sterile plastic container. The sample was placed in a 37 C incubator for 30 min to allow liquefaction. Semen samples from patients were categorised according to World Health Organization guidelines (Cooper et al., 2010).
Donor and andrology semen samples were prepared by density gradient centrifugation (DGC) as described by Martins da Silva et al.

| Chemogenomics library high-throughput screen
Dundee University Drug Discovery Unit in-house chemogenomics library was screened for compounds that increase [Ca 2+ ] i in human sperm. The compound library is composed of a set of 223 commercially available small molecules and drugs (Tocris), each with a welldefined mechanism of action, potency at the primary target, and selectivity. The compounds were selected as representative ligands for a diverse range of drug targets including enzymes, GPCRs, ion channels, and transporters. The compound library was initially screened on a single 384 well assay plate, at a single concentration of 40 μM. HTS and data analysis were performed as previously described (Martins da Silva et al., 2017). In brief, spermatozoa from two to four different donors were pooled together after preparation by DGC, diluted to a density of 2.2 × 10 7 ·ml −1 in Flexstation assay buffer (1 X HBSS [Invitrogen], 20-mM HEPES, 0.5-mM probenecid, pH 7.4), and incubated for 60 min (37 C) with 2 x Calcium 3 dye (Molecular Devices). Spermatozoa were washed following incubation, resuspended in Flexstation assay buffer, and plated in 384 well clear bottom, black well assay plates (Greiner Bio One) at a density of 2.5 × 10 5 cells/50 μl per well. [Ca 2+ ] i was measured using a Flexstation 3 (Molecular Devices). Baseline calcium-dependent fluorescence (excitation wavelength = 485 nm, emission wavelength = 525 nm, and cutoff = 515 nm) was measured for 18 s; 12.5 μl of each test compound was transferred to the assay plate using an internal 16-channel robotic pipette head, and the resulting change in fluorescence was monitored for a further 82 s. Follow-up assays were performed to determine the potency of hit compounds. All assay plates in the screen were subject to quality control analysis.
Preliminary analysis of all HTS primary and potency raw data was performed using the AUC function within the SoftMax Pro analysis software (SoftMax Pro Data Acquisition and Analysis Software, RRID: SCR_014240) to quantitate agonist-evoked fluorescence as previously described. Data were exported as a text file for further data processing and analysis in Activity Base version 7.3.1.4 (IDBS), and the percentage effect for each compound was normalised to the paired positive control (10-μM P4). Compounds were pragmatically classified on the basis of calcium fluorescence elicited and designated as low responder (blue 20-49%), mild responder (orange 50-89%), and high responder (green 90-120%) relative to progesterone (Table 1).
Data were normalised to paired controls and expressed as a Note. Twenty-seven U.S. Food and Drug Administration-approved active compounds were identified from the DDU Chemogenomics library screen following Flexstation assay testing, and categorised based on their ability to increase [Ca 2+ ] i (low to high percentage increase relative to 10-μM progesterone [positive control]). Trequinsin hydrochloride was selected for this study as it was highly efficacious and a PDE inhibitor (the compound library screen of all 223 commercially available small molecules and drugs [Ca 2+ ] i is shown in Figure S12).

| Flow cytometry analysis
Following 3-hr incubation in capacitating conditions, two aliquots containing 2 × 10 −6 sperm were centrifuged at 0.3 x g for 5 min. The Desensitisation experiments were carried in accordance with an established methodology (Brenker et al., 2018;Schaefer, Hofmann, Schultz, & Gudermann, 1998;Strünker et al., 2011). The first compound addition was added after 1-min recording of baseline fluorescence, followed by addition of the second compound after 5 min.
Control experiments were conducted to demonstrate that progesterone and PGE 1 do not cross-desensitise. Control experiments to demonstrate desensitisation, involved either addition of progesterone followed by 17α-hydroxyprogesterone or addition of PGE 1 followed by PGE 2 . The protocol used to assess the mode of action of trequinsin was similar. Cells were first challenged with either progesterone or PGE 1 followed, after 5 min, by trequinsin. Readings from an additional time control well (baseline) were taken as were readings from a well that was exposed to a single agonist at the time point that matched the time point of addition of the second agonist in the desensitisation experiments. All compounds were used at a final concentration of 10 μM.

| Measurement of pHi
After 3 hr in CM, spermatozoa (4 × 10 −6 per ml −1 ) were incubated with 2-μM 2 0 ,7 0 -bis(2-carboxyethyl)-5,6-carboxyfluorescein (ThermoFisher, Paisley, UK) for 30 min at 37 C. The cells were centrifuged for 3 min at 500 x g, the supernatant was removed, and the cells were then resuspended in sEBSS. A FLUOstar Omega reader (BMG Labtech) was used to detect the emitted fluorescence (excitation wavelength ratio of 440/490 nm and emission wavelength of 530 nm). Cell calibration was achieved following cell lysis by the addition of 1% Triton X-100, a reading was taken from each well, and a calibration curve was constructed using 1-M HCl and 1-M NaOH.
Fluorescence measurements for control (cells +1% DMSO) and trequinsin (10 μM) were recorded, as well as ammonium chloride (NH 4 Cl), which was used as a positive control (10-mM final concentration).

| Electrophysiology
The effect of trequinsin on individual sperm plasma membrane ion channels was investigated using whole-cell patch clamp electrophysiology (Brown et al., 2016). Sperm were allowed to settle on a glass coverslip prior to being placed in the recording chamber that was per- Cs + -based divalent-free intracellular solution to study membrane slope conductance (Gm) that is predominantly carried by K + ions (Brown et al., 2016) and CatSper channels, respectively (Supporting Information). Transition to whole-cell configuration was achieved by applying brief suction. To study outward membrane conductance, a depolarising ramp protocol was imposed (−92 to 68 mV) over 2,500 ms, and membrane potential was held at −92 mV between test pulses.
The effect of trequinsin on reversal potential and membrane slope conductance of outward currents was assessed by regression analysis over the voltage range where membrane current crosses the x axis (I = 0) and outward current from 20 to 68 mV, respectively (Brown et al., 2016).
After achieving the whole-cell configuration, monovalent CatSper currents were recorded by superfusing sperm with Cs + -based divalent-free bath solution (Supplementary Solutions in the Supporting Information). Currents were evoked by a ramp protocol (−80 to 80 mV over 1 s). Membrane potential was held at 0 mV between ramps. Data were sampled at 2 kHz and filtered at 1 kHz (PClamp 10 software, Axon Instruments, USA). The post-recording analysis was conducted as described previously to adjust for liquid junction potential and normalise for cell size (Brown et al., 2016).
2.10 | Detection of cyclic nucleotides by reversedphase HPLC
The samples were centrifuged (5 min, 300 x g), the supernatant was removed, and the pellet was resuspended in 0.5 ml of 100-mM sodium acetate (pH 4), sonicated for 1 min in a water bath, briefly vortexed, and centrifuged again (5 min, 3,000 g). The supernatant was removed and placed in a fresh Eppendorf, snap frozen in liquid nitrogen, and stored on dry ice until solid phase extraction.

| Standards and stock solutions
Stock solutions of cyclic nucleotides (cAMP and cGMP) were prepared to 3 mol·L −1 in mobile phase (see Section 2.10.4). From the stock solutions, five 10-fold serial dilutions were produced to achieve a 6-point standard curve (peak AUC). This was used for quantification of cyclic nucleotides in sperm samples.

| Data and statistical analysis
The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2018).
This research did not include the use of animals. Statistical power analysis was conducted to ensure that the group size was sufficient to measure an effect for each experiment using R pwr package (R Project for Statistical Computing, RRID:SCR_001905; pwr.2p.test) and Cohen's effect size analysis (control vs. treatment, sig. level = 0.05, power = 0.8). N numbers refer to data from independent samples.
Donor samples were allocated randomly by the technical team, and For the analysis of individual patient sperm motility results, statistical significance was recorded when the ±SD did not overlap for control and treatment conditions (Tardif et al., 2014). HPLC data were extracted using Breeze 2 software and analysed using Microsoft Excel. Results are expressed as pmol per 10 6 cells.  Figure S1). The functional and molecular profile of trequinsin was studied in further detail, as presented in this report.

| Nomenclature of targets and ligands
The other three compounds eliciting >90% increase in [Ca 2+ ] i did not promote motility (as assessed by CASA, data not shown) and were therefore not studied further.

| Donor sperm assessment
It is well accepted that activation of CatSper and elevation of cyclic nucleotides are fundamental for sperm motility and function (Ahmad et al., 2015;Alasmari, Costello, et al., 2013;Lefièvre, de Lamirande, & Gagnon, 2002). Motility and kinematic parameters of capacitated spermatozoa from healthy volunteer donors (80% DGC fraction) after 20-min exposure to trequinsin were studied using CASA. Trequinsin (0.1-100 μM) had no significant effect on TM or PM ( Figure S2).
However, a bell-shaped dose-response curve was obtained for HA (- Figure S2). As ≥30 μM had no effect on HA, 10-μM trequinsin was used in subsequent experiments. Capacitated sperm from the 80% DGC fraction exposed to 10-μM trequinsin showed no change in TM or PM ( Figure S3A,B) over a 2-hr period. However, the percentage of HA cells sperm was significantly increased (Figure 2). We also assessed the ability of trequinsin to stimulate penetration into viscous medium (Kremer test) as a measure of functional motility in the same spermatozoa population (80%, capacitated). Trequinsin, progesterone, and IBMX all significantly and similarly increased cell penetration into viscous medium at 1 cm. However, trequinsin and progesterone were significantly better than IBMX at stimulating penetration at 2 cm ( Figure 3). Trequinsin did not induce premature acrosome reaction in capacitated cells ( Figure S4B). In contrast, trequinsin had no effect on the motility parameters of cells from the 80% fraction in noncapacitating conditions ( Figure S5A-C).
Cells with poor motility isolated from 40% fraction after DGC preparation of ejaculates from healthy volunteer sperm donors were initially used as a surrogate for patient sperm, as previously described (Tardif et al., 2014). In contrast to donor 80% fraction cells incubated in capacitating conditions, 40% fraction capacitated cells showed a trequinsin-induced significant increase in PM 40 min after initial exposure, which was maintained for the duration of the assay period F I G U R E 2 Effect of trequinsin on capacitated donor sperm cell hyperactivation. Trequinsin significantly increased HA in all donor cells from the 80% DGC fraction exposed to capacitating conditions. *P<.05, significantly different from start (time 0); two-way ANOVA with Sidak's multiple comparison analysis. The increase in hyperactivation was sustained for a 2-hr period after initial exposure. A minimum of 200 cells were counted at each time point. Hyperactivation classified by CASA parameters: VCL >150 μM·s −1 , linearity <50%, and amplitude of lateral head displacement >7 μM. In the same sample set, %TM and %PM were unaffected by the addition of trequinsin ( Figure S3) F I G U R E 3 Sperm penetration assay. The ability of trequinsin to stimulate sperm penetration into viscous medium was assessed using capacitated sperm from the 80% DGC fraction (n = 5). A significant increase in cell penetration was observed in the presence of trequinsin in comparison to control, but not in comparison to cells stimulated with progesterone (P4). Cell penetration at 1 cm was not significantly different between trequinsin-and IBMX-treated cells. However, trequinsin stimulated a significantly greater cell number to penetrate at 2 cm compared to IBMX. *P<.05, significantly different from control; two-way ANOVA with Sidak's multiple comparison analysis ( Figure 4a). Although there was no effect on TM ( Figure S6A), hyperactivation was also significantly increased, similar to donor 80% fraction cells (Figure 4b). Under non-capacitating conditions, trequinsin significantly improved PM of 40% fraction sperm for the entire experimental period (Figure 4c) but had no effect on TM or hyperactivation ( Figure S6B,C). The significant changes in motility seen in sperm from the 40% DGC fraction in capacitating conditions provided proof of concept that trequinsin may similarly boost sperm motility in poorly motile sperm from patients. To investigate this further, we assessed patient sperm motility over a 2-hr period in response to treatment with trequinsin exposed to capacitating conditions.

| Patient sperm assessment
A total of 25 patients attending the ACU for routine andrology assessment, in vitro fertilisation, ICSI, and sperm study patients (Table 2) T A B L E 2 Effect of trequinsin on patient sperm motility Note. Summary of motility changes in patient samples (in vitro fertilisation, ICSI, and andrology) treated with 10-μM trequinsin. The motility of 25 patient samples was assessed using CASA over a 2-hr period at regular intervals (see Section 2), and an average for each parameter was taken overall. A minimum of 200 cells were counted at each time point. ", significant increase; -, no change; #, significant decrease. Significant means and SD (control vs. treatment at each time point) do not or do overlap for increase and decrease, respectively (TM, total motility; PM, progressive motility; HA, hyperactivated motility). Patient samples are categorised based on semen World Health Organization (WHO) parameters (see Section 2). ✓ represents a WHO guideline criterion met; × represents a criterion not meeting WHO guidelines.
F I G U R E 4 Effect of trequinsin on 40% DGC fraction (poor motility) donor sperm motility. Trequinsin significantly increased the percentage of progressively motile sperm in (a) capacitating (n = 5) and (b) non-capacitating (n = 6) conditions. Hyperactivation was also significantly increased when sperm were incubated in (c) capacitating conditions (n = 8). *P<.05, significantly different compared to time 0; two-way ANOVA with Sidak's multiple comparison analysis.
Corresponding motility data can be found in Figure S6 added on 27 December 2019, after first online publication: Figure 4 parts B and C have been relabelled. ] CatSper is a ligand-activated, pHi and voltage-sensitive channel.
Therefore, to investigate the mechanism by which trequinsin causes an increase in [Ca 2+ ] i , we utilised whole-cell patch clamp electrophysiology to examine the drug's ability to modulate ion channel function directly and monitored changes in pHi using the ratiometric dye 2 0 ,7 0bis(2-carboxyethyl)-5,6-carboxyfluorescein. Predictably, trequinsin significantly potentiated inward and outward CatSper currents, to a degree not significantly different from progesterone (Figure 6a,b).
Therefore, we exploited this phenomenon to investigate the mechanism of the trequinsin-induced increase in [Ca 2+ ] i . Pretreatment with progesterone caused desensitisation of the response to 17-OHprogesterone but not PGE 1 or trequinsin (Figure 7). Fittingly, pretreatment with PGE 1 caused desensitisation of the trequinsin, but not the progesterone response (Figure 7). Trequinsin is a potent PDE3i (Tinsley et al., 2009). PDE enzymes control the hydrolysis of cyclic nucleotides, specifically cAMP and cGMP, both of which are substrates for PDE3 (Lefièvre et al., 2002). In contrast to the non-specific PDEi IBMX, trequinsin did not significantly induce elevation of cAMP ( Figure 8a). However, it induced a significant~4-fold increase of cGMP in capacitated cells (Figure 8b).

| Patient [Ca 2+ ] i profile in response to trequinsin
Given that poor motility and impaired fertilisation potential are associated with impaired CatSper function (Kelly et al., 2018), it is important  Figure S10).

| DISCUSSION
Male infertility is a significant health challenge that is estimated to affect one in 10 men (Datta et al., 2016). In up to 40% of these cases, the cause may be due to reduced sperm motility (asthenozoospermia; van der Steeg et al., 2011). However, as there are currently no licensed agents to treat infertile men, ICSI remains the only viable treatment option to ensure oocyte-spermatozoon interaction. A fundamental reason for the shortfall in progression in the field of male fertility therapeutics has been the lack of knowledge regarding a F I G U R E 7 Examination of agonist cross-desensitisation. Population average [Ca 2+ ] i trace using capacitated donor sperm from the 80% DGC fraction (n = 5) showing initial agonist addition of either a saturating concentration of 10-μM progesterone (P4) (a, c, e) or 10-μM PGE 1 (b, d, f), followed by the second agonist addition. A baseline control shown in blue was included in each experiment and a blank (sEEBS, represented as "B") followed by the addition of the second agonist green. Cross-desensitisation experiments are shown in red. (g) Bar chart showing cell exposed to 10-μM 17-OH-progesterone (17OHP4) did not produce a significant Ca 2+ response compared to that of PGE 1 (10 μM) and trequinsin (10 μM). Effects of PGE 1 and trequinsin were not significantly different. *P<.05, significantly different as indicated; two-way ANOVA with Sidak's multiple comparison analysis. (h) Cells pre-exposed to 10-μM PGE 2 had significantly lower Ca 2+ responses (<2%) compared to progesterone exposure. Effects of PGE 2 and trequinsin were not significantly different. *P<.05, significantly different as indicated; two-way ANOVA with Sidak's multiple comparison analysis suitable molecular target in sperm, thereby limiting the opportunity for implementing drug discovery strategies (Barratt, De Jonge, & Sharpe, 2018;Hughes, Rees, Kalindjian, & Philpott, 2011). However, a wealth of studies now demonstrate that CatSper is a key determinant of sperm motility and fertilisation competence (Alasmari, Costello, et al., 2013;Brown et al., 2018;Kelly et al., 2018;Ren et al., 2001;Smith et al., 2013;Strünker et al., 2011;Williams et al., 2015) and therefore represents a plausible target for the development of novel therapeutics for male infertility. We have previously described a highthroughput drug screening methodology in conjunction with relevant in vitro tests to identify compounds that increase functional sperm motility (Martins da Silva et al., 2017). While this study validated our drug discovery strategy, there continues to be a significant unmet clinical need to identify efficacious compounds that influence different forms of sperm motility and function. In this study, we utilised an HTS strategy to screen an in-house drug discovery library and identified trequinsin hydrochloride, a putative selective PDE3i, which signif- CatSper is the primary calcium-conducting plasma membrane ion channel in sperm that is activated by intracellular alkalinisation, membrane depolarisation, and physiological ligands such as progesterone and PGE 1 (Singh & Rajender, 2015;Strünker et al., 2011;Tamburrino et al., 2014). It can also be manipulated by compounds, including endocrine disrupting chemicals, that may compromise sperm function (Schiffer et al., 2014;Tavares et al., 2013).  (Miller et al., 2016a(Miller et al., , 2016b. We exploited observations that these mechanisms exhibit limited cross-desensitisation (Brenker et al., 2018;Lishko et al., 2011;Miller et al., 2016aMiller et al., , 2016bSchaefer et al., 1998;Shimizu et al., 1998;Strünker et al., 2011) to show that the trequinsin crossdesensitisation profile is indistinguishable from that of PGE 1 . As trequinsin did not alter pHi, we conclude that trequinsin increases [Ca 2+ ] i by a combination of direct activation of CatSper as well as by membrane potential depolarisation through a partial blocking effect on the sperm potassium channel. However, we cannot rule out additional direct actions on pathways that regulate intracellular stores (Correia, Michelangeli, & Publicover, 2015) or extracellular calcium entry (De Blas et al., 2009;De Toni et al., 2016;Kumar et al., 2016).
Trequinsin is a potent (subnanomolar IC 50 ) inhibitor of recombinant PDE3 (Tinsley et al., 2009). As cyclic nucleotides are essential second messengers for sperm motility (Balbach, Beckert, Hansen, & Wachten, 2018;Jansen et al., 2015;Mukherjee et al., 2016), we utilised HPLC to measure cAMP and cGMP changes in sperm exposed to trequinsin and demonstrated that only cGMP was significantly increased. Given that PDE3 enzymes metabolise cAMP and cGMP (Ahmad, Degerman, & Manganiello, 2012), this result is surprising because pharmacological and immunological evidence supports the presence of PDE3 in human sperm, localised to the postacrosomal region of the head (Lefièvre et al., 2002). In contrast, PDE3 isoforms were not among the seven PDE enzymes identified in a study analysing the human sperm proteome (Wang et al., 2013). Therefore, our data may reflect the inhibitory activity of cGMP-PDE, PDE6D. The notion of non-selective PDE-inhibitory activity of trequinsin is supported by the relatively high concentration that is required to increase HA. In fact, 10-μM trequinsin is above the IC 50 at the cAMPspecific PDE2A and PDE4 and cGMP-specific PDE5A (Souness & F I G U R E 8 Measurement of cyclic nucleotide levels in capacitated 80% DGC fraction donor sperm using RP-HPLC. (a) Trequinsin did not alter intracellular cAMP in comparison to control (cells +1% DMSO; n = 11). (b) Trequinsin significantly increased intracellular cGMP (n = 11). IBMX, a non-specific PDEi, was used as a positive control. IBMX significantly increased both cAMP and cGMP (n = 11). *P<.05, significantly different as indicated; two-way ANOVA with Sidak's multiple comparison analysis Rao, 1997;Tinsley et al., 2009;Wunder, Gnoth, Geerts, & Barufe, 2009). However, proteomic data do not support their expression in human sperm (Wang et al., 2013). It is notable that micromolar concentrations of PDEi are generally required to induce improvements in human sperm motility Lefièvre et al., 2002;Maréchal et al., 2017;Tardif et al., 2014) but the reason for this is unknown.
Given that NO donor compounds can modify sperm kinematic parameters, including VCL and straight line velocity, it is entirely plausible that an effect of trequinsin on cGMP levels may contribute to the changes seen in VCL and straight line velocity ( Figure S11; Miraglia et al., 2011). In further support for this mode of action, trequinsin increased the percentage of sperm exhibiting HA and penetration into viscous medium under capacitating conditions in all donor samples. Reassuringly, premature acrosome reaction was not induced in these samples; implying sperm-zona pellucida binding would not be hindered.
As expected, the increase in hyperactivation was dependent upon cell capacitation status. Hyperactivation was unaltered in cells maintained in non-capacitating conditions, despite trequinsin giving a robust [Ca 2+ ] i increase in these cells ( Figure S9). Although trequinsin was highly effective at increasing hyperactivation in patient sperm samples incubated in capacitating conditions (22/25), two were unresponsive, and all motility parameters were reduced in one. The reason for this profile is unknown, but we could demonstrate that one unresponsive case (R2947) was not due to defective [Ca 2+ ] i signalling (- Figure S10). Biological variability is certainly seen within human sperm populations. Indeed, not all patients respond to drugs, and this finding may not be uncommon (Alvarez et al., 2003;Moohan, Winston, & Lindsay, 1993).
Consequently, the same level of exposure to a drug, for example, trequinsin, may result in different levels of biological effects in individual patients. This is the key concept encompassed by the term "individualised medicine." Determining the reasons for the biological variability seen in this and other studies (Martins da Silva et al., 2017) is an important consideration for future drug development and is dependent upon robust screening strategies and phenotypic assays to identify and treat specific molecular and functional impairment. Additionally, the development of multi-target compounds could be advantageous. For example, it would be interesting to determine if trequinsin could restore the fertilising potential of sperm affected by CatSper and sperm potassium channel dysfunction (Brown et al., 2016;Kelly et al., 2018;Williams et al., 2015).
In summary, we have shown that trequinsin hydrochloride is an efficacious CatSper agonist that suppresses sperm potassium channel activity, elevates cGMP (but not cAMP), and induces similar kinetics of [Ca 2+ ] i increase as progesterone through a mechanism that crossdesensitises with PGE 1 . This novel pharmacological profile results in a phenotype of increased hyperactivation and penetration into viscous medium, which is relevant to sperm function required for natural conception. We conclude that the pharmacological profile of trequinsin in human sperm is unique in terms of effect on multiple key intracellular mediators that influence sperm function (Esposito et al., 2004;Hess et al., 2005;Martins da Silva et al., 2017;Tardif et al., 2014;Williams et al., 2015) and holds promise as a novel agent to treat male infertility. Wunder, F., Gnoth, M. J., Geerts, A., & Barufe, D. (2009)

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of this article.