Group X secreted phospholipase A₂ specifically decreases sperm motility in mice

Biological preparations

Computer-assisted motility analysis

Noncapacitated or capacitated sperm suspension was immediately placed onto an analysis chamber (100 µm depth, Leja Products B.V., Nieuw-Vennep, the Netherlands) and kept at 37°C for microscopic quantitative study of sperm movement. Sperm motility parameters were measured at 37°C using a sperm analyzer (Hamilton Thorn Research Inc, Beverly MA). The settings employed for analysis were as follows: acquisition rate: 60 Hz; number of frames: 100; minimum contrast: 25; minimum cell size: 10; low static-size gate: 2.4; high static-size gate: 2.4; low static-intensity gate: 1.02; high static-intensity gate: 1.37; minimum elongation gate: 12; maximum elongation gate: 100; magnification factor: 0.70. The motility parameters measured were curvilinear velocity (VCL), average path velocity (VAP) and amplitude of lateral head displacement (ALH). At least, 100 motile sperm were analyzed for each assay. Progressive hyperactivated sperm were characterized by VCL > 250 µm/sec, VAP > 70 µm/sec, ALH > 18 µm, LIN < 45 and non-progressive hyperactivated sperm (star spin) by 120 < VCL < 250 µm/sec, VAP < 50 µm/sec and LIN < 30.

Method to trace the theoretical mGX-treated distribution: the sperm subpopulation belonging to the bin will shift to the bin having a velocity of .

Production of recombinant sPLA₂s

Recombinant mouse sPLA₂s group IIA, IID, IIE, V, and X and the H48Q mutant of mGX-sPLA₂ were produced as described previously (Singer et al., 2002; Rouault et al., 2007). Pro-mGX sPLA₂ was produced as for mature mGX sPLA₂ using the pAB3 vector in which the full-length cDNA coding for PromGX sPLA₂ was inserted in frame with the ▵GST protein and the factor Xa cleavage site, which were removed from PromGX sPLA₂ by using the factor Xa protease (Rouault et al., 2007).

Phospholipids staining/viability test

Control untreated sperm cells (noncapacitated or capacitated for 90 min) or sperm treated for 10 min with 200 nM mGX sPLA₂ in M16 medium were washed twice with PBS, and then incubated in a staining buffer (from Apoptosis detection kit, Sigma–Aldrich, Saint Quentin Fallavier, France) containing either carboxyfluorescein diacetate (6-CFDA) or 1 µg/ml annexin V-Cy3 for 15 min in the dark at room temperature. After incubation with annexin V-Cy3, sperm was washed twice with the staining buffer and fixed in PBS with 4% paraformaldehyde. AnnexinV-Cy3 staining was analyzed using a confocal microscope (Leica Microsystems GmbH, Wetzlar, Germany).

Chemical compounds

M2 and M16 media, fatty acid free BSA (fraction V) and annexin V-Cy3 (from Apoptosis detection kit) were purchased from Sigma-Aldrich, Saint Quentin Fallavier, France. LY329722 was a generous gift from Pr. M. Gelb.

Statistical analyses

Statistical analyses were performed with SigmaPlot. Paired t-test were used to compare the effect of various compounds on AR. Data presented represent mean ± SEM. Statistical tests with a two-tailed P values ≤0.05 were considered as statistically significant. Linear regression was calculated with SigmaPlot.

mGX sPLA₂ specifically inhibits sperm motility among five different mouse sPLA₂s

During its short life, sperm become exposed to several sPLA₂s, including those present in the seminal plasma and originating from epididymis and prostate or present in the female genital tract (Valentin et al., 1999; Masuda et al., 2004; Lessig et al., 2008; Sato et al., 2010; Escoffier et al., 2010b). We thus evaluated the effect of five different mammalian sPLA₂s known to be present in male or female genital tracts on sperm motility using the computer-assisted sperm analysis (CASA) system. These sPLA2s are mGIIA, mGIID, mGIIE, mGV, and mGX. Sperm tracks are characterized with the CASA system by four basic parameters evaluating sperm motility: the track speed (VCL), the progressive velocity (VSL), the path velocity (VAP) and the lateral displacement of the head (ALH). We first focused on noncapacitated sperm and because noncapacitated sperm have mostly linear progressive tracks, only the VCL and the ALH parameters were studied. Sperm were incubated with the 5 different recombinant sPLA₂s at 200 nM during 10 min in M16 culture medium at 37°C. Figure 1A,B clearly shows that only mGX sPLA₂ was a potent modulator of VCL and ALH: in the presence of 200 nM mGX sPLA₂, the mean VCL was decreased by 23%, from 165.55 ± 5.44 (n = 19) to 124.73 ± 6.65 µm/sec (n = 16), P < 10⁻⁴ while the mean ALH was decreased by 14%, from 11.45 ± 0.35 (n = 19) to 9.81 ± 0.54 µm (n = 16), P = 0.01. The potency of mGX was variable between different males: no correlation between initial VCL values of the sample and the mGX sPLA₂ potency was noticed (Fig. 1A), contrary to ALH values, which appeared to be negatively correlated with control levels (Fig. 1B, correlation coefficient 0.69). In order to better characterize the inhibitory effect of mGX sPLA₂ on sperm motility, we determined the minimal concentration of mGX sPLA₂ triggering the inhibitory effect in 10 min. Figure 1C shows mGX sPLA₂ dose–response curves for VCL and ALH and indicates that concentrations of mGX sPLA₂ as low as 20 nM affect significantly sperm motility, and more particularly VCL. sPLA2, as an enzyme hydrolyzing the lipids of the plasma membrane, may cause a premature cell death, affecting the sperm motility. To test this hypothesis, sperm were first treated with 200 nM mGX sPLA₂ for 10 min and subsequently incubated with 6-carboxyfluorescein diacetate (6-CFDA) for 15 min. This compound is used as a marker of sperm viability: it enters freely into cells and produces green fluorescence after de-esterification by endogenous esterases only in living cells (Silva and Gadella, 2006). Both control (83.02% ± 1.2, n = 2) and treated sperm (85.42% ± 2.4, n = 2) showed a strong green fluorescence signal (Fig. 1D), indicating that the sperm motility decrease induced by the sPLA₂ was not due to premature cell death during the duration of experiment.

The effect of mGX sPLA₂ is dependent on its enzymatic activity

To determine whether the enzymatic activity of mGX sPLA₂ is required for modulation of sperm motility, we first analyzed the effect of its H48Q mutant which has a very low enzymatic activity (Surrel et al., 2009). The mGX H48Q mutant was unable to affect sperm VCL (Fig. 2A) or sperm ALH (Fig. 2B). Second, since mGX sPLA₂ is produced as a zymogen (pro-mGX sPLA₂) which has a very low enzymatic activity (Morioka et al., 2000), we tested the effect of pro-mGX sPLA₂ on sperm motility and found no change on VCL and ALH (Fig. 2A,B respectively). These results indicate that the enzymatic activity plays a crucial role in the effect of mGX sPLA₂ on sperm motility. To further confirm these results, we tested the effect of the specific sPLA₂ inhibitor LY329722. This compound is known to inhibit mGX enzymatic activity with an IC₅₀ of 75 nM (compound A in Smart et al., 2006). Preincubation of mGX sPLA₂ with LY329722 at 1 µM prevented its inhibitory effect on VCL (Fig. 2C).

Finally, we measured the kinetic of VCL inhibition by mGX sPLA₂ at 200 nM. Sperm were incubated with mGX sPLA₂ from 1 to 20 min: the VCL inhibition was proportional to the incubation period up to 10 min, then reached a plateau after 10 min (Fig. 2D). This result reinforces the importance of sPLA₂ enzymatic activity in sperm velocity decrease and suggests that all substrates were hydrolyzed in 10 min at 200 nM.

Heterogeneous response of sperm to mGX sPLA₂ treatment

The sperm population presents a high degree of heterogeneity in terms of motility, even in a cell population obtained from a single male. This heterogeneity is displayed by a wide range of VCL and ALH values, indicating different subpopulations of sperm. For instance, values for VCL ranged from 7.5 to 450 µm/sec in a control cell population (Fig. 3A, blue bars/curve). We noticed the same degree of heterogeneity for ALH, with values ranging between 0 and 40 µm (Fig. 1D). In this study, static sperm cells and slow cells were not counted and only sperm cells with a VAP > 7.5 µm/sec were analyzed. An important question was to determine whether mGX sPLA₂ inhibits slow (VCL between 50 and 150 µm/sec) or fast sperm subpopulations (VCL between 200 and 300 µm/sec) with a similar potency (Fig. 3A). To address this question, we first compared the VCL distribution of control and mGX sPLA₂-treated sperm populations, and then the VCL distribution of mGX sPLA₂-treated sperm population versus a theoretical-treated distribution obtained after applying a identical reduction factor for the speed of each bin corresponding to the decrease of mean VCL calculated from the whole population: for this set of experiments we obtained a decrease of 30% (Fig. 3B, n = 3). As shown in Figure 3A, adding mGX sPLA₂ modified the VCL distribution of sperm population, essentially by increasing the percentage of sperm with low speed (black bars/curve) and decreasing by a twofold factor the fast sperm subpopulation. We then traced a theoretical VCL distribution mimicking the effect of mGX sPLA₂-treated sperm (Fig. 3C, red bars/line) by applying to all bins of control sperm an identical reduction factor (see Materials and Methods Section) corresponding to the reduction measured on the whole population, which was 30% (Fig. 3B). Comparing the distributions of measured and theoretical mGX sPLA₂-treated sperm clearly shows that the theoretical distribution does not fit with that corresponding to mGX sPLA₂-treated sperm (Fig. 3C, black vs. red curves). A similar discrepancy was observed when the size of the bins was reduced by a twofold factor (Size of 25, data not shown). This result suggests that some sperm subpopulations are differentially affected by the sPLA₂ treatment. In particular, fast sperm were less inhibited by mGX sPLA₂ than expected from the theoretical calculation. Because ALH and VCL amplitudes are correlated (correlation coefficient 0.6994 ± 0.025; Fig. 3D, n = 5), this result is consistent with our above result showing that mGX-inhibition of ALH appeared to be negatively correlated with its mean initial ALH amplitude (Fig. 1B). The heterogeneity of sperm population is illustrated by the tracks of sperm representing the position of the head of sperm over a period of 1.7 sec measured in control sperm and mGX sPLA₂-treated sperm, respectively (Fig. 3E,F).

mGX sPLA₂ is involved in phospholipid redistribution at the plasma membrane of the flagellum during capacitation

We have previously demonstrated that endogenous mGX sPLA₂ is released via spontaneous AR during capacitation, and reaches after 45 min a concentration able to induce AR in a paracrine mode (Escoffier et al., 2010b). This process amplifies spontaneous AR by likely hydrolyzing phospholipids of the sperm plasma membrane right next to the acrosomal membrane. Because mGX sPLA₂ is a potent modulator of sperm motility, we were wondering whether it may target phospholipids of the plasma membrane of the flagellum as well. To address this hypothesis, we first chose to measure the effect of exogenous mGX sPLA₂ treatment on sperm phosphatidylserine distribution because (i) phosphatidylserine is readily hydrolyzed by mGX sPLA₂ (Singer et al., 2002) and (ii) among phospholipids, phosphatidylserine is easily identified because it can be specifically stained by annexin V conjugated with the Cy3 fluorescent dye. Sperm incubated in a capacitating medium for 35 min present a strong annexin V-Cy3 staining in their midpiece. We thus compared the intensity of annexin V-Cy3 stainings of untreated capacitated sperm and capacitated sperm subsequently incubated with 200 nM mGX sPLA₂ during 10 min (Fig. 4A). This experiment shows that recombinant mGX sPLA₂ strongly decreased the annexin V-Cy3 staining of sperm capacitated during 35 min and demonstrates the ability of mGX sPLA₂ to hydrolyze phospholipids of the sperm flagellum membrane.

Sperm capacitation leads to important changes in the lipid composition of the plasma membrane (Gadella and Harrison, 2000), including caspase-independent redistribution of phosphatidylserine in the external leaflet of the plasma membrane (de Vries et al., 2003). We confirmed that phosphatidylserine distribution changes during capacitation in our conditions. We observed that annexin V-Cy3 staining was absent or weak on noncapacitated sperm (Fig. 4B). After 35 min of capacitation, the staining was strong and localized mainly in the midpiece of flagella. On the other hand, sperm heads and principal pieces were weakly or not stained. At 80 min, a decrease in the annexinV-Cy3 staining was observed. This result confirms that capacitation, induced by cholesterol efflux and/or bicarbonate influx, leads to a modification of phospholipid asymmetry, promoting externalization of anionic phospholipids at the plasma membrane of the flagellum and specifies its variation throughout capacitation. The fact that the intensity of annexinV-Cy3 staining decreased between 35 and 80 min of capacitation (Fig. 4B) is in good accordance with both mGX sPLA₂ enzymatic activity on sperm phospholipids (Fig. 4A) and the increase of mGX sPLA₂ concentration in the capacitation medium after 45 min of capacitation (Escoffier et al., 2010b) and suggests that released mGX sPLA₂ would be able to control sperm motility of capacitated sperm.

Endogenous mGX sPLA₂ released during spontaneous AR decreases sperm velocity

We thus analyzed whether endogenous mGX sPLA₂ was able to modulate sperm motility during the capacitation process. To test this hypothesis, we capacitated sperm in the presence of LY329722, the specific inhibitor of mGX sPLA₂ tested in Figure 2C, during various times of capacitation (0, 55, and 90 min) and compared the VCL of LY329722-treated versus non-treated sperm (control). There was no statistically significant difference of the VCL between control and LY329722-treated sperm at 10 and 55 min of capacitation (Fig. 5A). In contrast, sperm treated during 90 min with LY329722 were faster than non-treated sperm: the mean VCL increased from 182.7 ± 5.1 to 191.1 ± 5.0 µm/sec (P = 0.036, n = 11) for LY329722-treated sperm and control sperm respectively (Fig. 5A). The fact that LY329722 treatment was effective only at 90 min is consistent with the kinetics of release of endogenous mGX sPLA₂ during spontaneous AR, as demonstrated previously (Escoffier et al., 2010b).

To confirm that mGX sPLA₂ released during capacitation was able to reduce the sperm velocity, we compared the potency of recombinant mGX sPLA₂ on the mobility of noncapacitated versus sperm capacitated for 90 min. Indeed, a 10 min application of 200 nM recombinant sPLA₂ on capacitated sperm, a protocol sufficient to obtain a maximal sperm velocity inhibition (Fig. 2D), should produce a weaker inhibition than that obtained with uncapacitated sperm, since the capacitated cells are already partially inhibited by released mGX sPLA₂. Noncapacitated and capacitated sperm were treated with mGX sPLA₂ in the same conditions: the effect of mGX sPLA₂ on sperm motility was measured after 10 min of incubation in M16 medium. Capacitated sperm were not treated in the capacitating medium (containing 2% fatty acid free BSA) in order to avoid an artefactual effect of BSA, which may chelate lipid metabolites produced by mGX sPLA₂. Capacitated sperm were thus washed with M16 before mGX sPLA₂ incubation in order to remove BSA. In this set of experiments, mGX sPLA₂ inhibited the VCL of noncapacitated sperm by ∼18%; P < 10⁻³, n = 6 (Fig. 5B). On the other hand, the velocity of the same sperm, capacitated for 90 min and then treated for 10 min with mGX sPLA₂, was inhibited by ∼9% (P = 0.036, n = 6). A similar result showing a decrease of mGX sPLA₂ potency on ALH was observed at 90 min of capacitation (Fig. 5C). These results are in good accordance with the initial release of mGX sPLA₂ and a subsequent phospholipid hydrolysis. Sperm were capacitated at a concentration of 200,000 sperm/ml. We have previously shown that after 90 min of capacitation, the amount of released mGX sPLA₂ is around 1 ng/10⁶ sperm (Escoffier et al., 2010b). Thus, during this experiment, sperm were swimming in a medium containing 0.02 nM of sPLA₂. This concentration is very low when compared to those used in Figure 1, and thus the effect of LY329722 on sperm motility at 90 min of capacitation suggests that a long incubation (45 min) with a low concentration of mGX sPLA₂ has an effect similar to the one observed after a short application of higher concentrations. This hypothesis was confirmed by comparing the effect of 200 nM mGX sPLA₂ applied during 10 min versus the effect of 20 nM applied either during 60 or 120 min in a medium that does not promote capacitation (Fig. 5D): similar levels of inhibition were obtained at 200 nM during 10 min (12%) or at 20 nM during 120 min (13%), n = 5.

Finally, capacitation leads to a modification of sperm movements, and a small fraction of sperm switches from a linear movement to a more undulatory movement, and eventually to a circular movement. It is accepted that sperm presenting such movement changes are fully capacitated and are described as progressive and nonprogressive (or star spin) hyperactivated sperm, respectively (Mortimer, 2000). LY329722 treatment did not modify the hyperactivated sperm subpopulations (Fig. 5E) and this result suggests that released mGX sPLA₂ does not interfere with the capacitation process leading to changes in sperm movements.

Sperm motility is controlled by mGX sPLA₂

To our knowledge, this is the first report showing the modulation of sperm motility by an sPLA₂ during capacitation. Platelet activating factor (PAF), a potent lipid mediator, has been shown to increase sperm chemotactism, but its effect on sperm motility parameters has not been investigated (Wu et al., 2001). Disruption of the gene for the intracellular PLA₂ group VIA (iPLA₂β) leads to a severe impairment of sperm motility, but the direct contribution of this PLA₂ to sperm flagellum beat has not been demonstrated (Bao et al., 2004). Similarly, disruption of the Pla2g3 gene (coding for mGIII sPLA₂) leads to a series of profound defects in sperm maturation, including asthenozoospermia (Sato et al., 2010). These effects appear to be due to maturation defects rather than to a defective control of sperm flagellum beat during capacitation.

We have previously shown that mGX sPLA₂ is a strong inducer of AR (Escoffier et al., 2010b) and an obvious question concerns the relationship between the mGX-induced motility decrease and the modification of the acrosomal status of sperm. In Figure 6, we have plotted on the same graph the dose–response curves for mGX sPLA₂-induced AR level and sperm motility decrease as a function of mGX sPLA₂ concentrations. It clearly appears that AR did not induce a sperm motility decrease since there was not correlation between the two phenomena. The absence of correlation is obvious, especially at low and high concentrations of mGX sPLA₂. For instance, at 2 nM, the motility was not affected whereas this concentration produced a 30% increase in the number of acrosome reacted sperm.

We have not yet determined the molecular pathways explaining how mGX sPLA₂ enzymatic activity reduces sperm motility. Since various fatty acids and lysophospholipids, the two products of sPLA₂ enzymatic activity, can inhibit sperm motility (Hong et al., 1986; Siegel et al., 1986), the effect of mGX sPLA₂ is likely due to the release of such metabolites. Reactive oxygen species (ROS) are produced when sperm are incubated with unsaturated fatty acids (Aitken et al., 2006) and sperm motility is known to be sensitive to oxidative damage (Agarwal et al., 2008). Although the production of ROS by sPLA₂ on sperm remains to be determined, different sPLA₂s have been involved in ROS production in several cell types (Adibhatla and Hatcher, 2008; Kim et al., 2009). An alternative hypothesis may involve changes of the biophysical properties of proteins embedded in the plasma membrane. Indeed, it has been recently shown that the functions of various membrane proteins, including ion channels, are specifically modulated by different types of lipids via protein–lipid interactions. For instance, the voltage sensor of K_v2.1 K⁺ channels is modulated by charged phospholipids and sphingomyelinase alters the biophysical properties of this channel (Xu et al., 2008). However the fact that LY329722 did not modify hyperactivated sperm subpopulations (Fig. 5D) suggests that mGX sPLA₂ does not interfere with cell signaling involved in hyperactivation, at least in vitro. Hyperactivation is mostly controlled in vitro by calcium signaling (Suarez, 2008) and this result suggests that the mGX sPLA₂ effect is not related to changes in calcium influx.

Physiological relevance of mGX sPLA₂ action on sperm motility

By using CASA, we identified several subpopulations based on sperm velocity (Fig. 3). Because the CASA system takes a snapshot of the sperm population, the true existence of these subpopulations may still be debated. However, the existence of highly motile sperm subpopulation, sorted by swim-up techniques is well accepted for human sperm (Mortimer and Mortimer, 1992) and we obtained similar results by swim-up techniques with mouse sperm (not shown). This notion of sperm population heterogeneity is clearly illustrated by the histograms of sperm velocity showing the presence of different subpopulations in a single sperm sample. Moreover, we have demonstrated that the different subpopulations are not equally inhibited by mGX sPLA₂, the slow speed sperm being more sensitive to mGX sPLA₂ treatment. The stronger effect of mGX sPLA₂ on slow sperm may be due to different lipid compositions of the plasma membrane since it has been shown that the heterogeneity of sperm population velocity is associated with a difference in lipid composition of slow versus fast sperm subpopulations (Buffone et al., 2009; Zalata et al., 2009). Although the effect of mGX sPLA₂ on ALH was dependent of its initial amplitude (Fig. 1B), confirming therefore that less motile cells are more affected by mGX sPLA₂, the absence of correlation between the mGX sPLA₂ potency and the initial VCL amplitude of the sample (Fig. 1A) may appear controversial. However, the reasons for differences of sperm velocity between the different males are likely not exclusively related to lipid composition differences since the flagellum beat is controlled by numerous factors as described at the beginning of this article. Another facet of sperm heterogeneity has been recently described, showing that a fraction of the sperm population is resistant to AR induced by mGX sPLA₂ (Escoffier et al., 2010b). This mGX resistant subpopulation of sperm gives higher fertility outcome than control sperm in in vitro fertilization experiments, and we have proposed that mGX sPLA₂ may sort different sperm subpopulations presenting different levels of fertility. The fact that high-speed sperm (the likely most competent sperm subpopulation for fertilization) are less sensitive than low speed sperm to sPLA₂ treatment, suggests that mGX sPLA₂ may further contribute to sort competent sperm cells by slowing-down the slow, likely suboptimal sperm, and thus decreasing the chance for these latter to reach oocytes on time for fertilization.

From these results, we can propose that mGX sPLA₂ released by sperm during spontaneous AR may sort competent sperm cells by two ways: first by promoting AR of a specific sperm subpopulation with a low fertility potency and second by decreasing the sperm motility of slow and likely suboptimal sperm.