Zinc is an intracellular signal during sperm activation in Caenorhabditis elegans

Infertility is a prevalent medical problem that affects 8-12% of couples, and infertility treatment can be stressful and expensive. In ∼20% of couples that struggle with infertility, sperm defects or deficiencies appear to be the source of the problem (Agarwal et al., 2021; Anderson et al., 2009; Thonneau et al., 1991; Vander Borght and Wyns, 2018). Thus, treatments that can increase sperm activity are desirable. In addition to clinical relevance, understanding sperm function is an important goal in biology. Sperm represent a fascinating example of a differentiated cell with a highly specialized function: delivering the male pronucleus to the egg to produce a fertilized zygote. Sperm from many species are kept in an inactive state to conserve energy until mobility is needed (Aitken and Nixon, 2013; Ellis and Stanfield, 2014). The immotile spermatid then undergoes a dramatic post-meiotic transition known as sperm activation or spermiogenesis whereby it becomes a mature, motile sperm. Because sperm chromatin is compacted for delivery into the egg, it is transcriptionally silent (Baker, 2016; Kimmins and Sassone-Corsi, 2005). Therefore, sperm activation must be accomplished without new gene transcription, and it is important to characterize the signaling events that promote this rapid cell differentiation event.

The nematode Caenorhabditis elegans is a powerful experimental system for studies of sperm differentiation and activation (reviewed by L'Hernault, 2009; Chu and Shakes, 2013; Ellis and Stanfield, 2014). In C. elegans, inactive spermatids are round and immotile, with specialized endoplasmic reticulum-derived membranous organelles located just beneath the plasma membrane (Ward et al., 1981). During activation, these organelles fuse with the plasma membrane, and the nematode cytoskeletal Major Sperm Protein is assembled into filaments that ultimately produce a pseudopod for crawling. Forward genetic screens have identified many genes essential for one or more steps of sperm development; they have been characterized into two signal transduction pathways that regulate sperm activation (Ellis and Stanfield, 2014). One pathway involves the secreted protease TRY-5, which probably functions extracellularly to activate sperm by cleaving a protein on the plasma membrane of sperm cells. A candidate for this target is the transporter protein SNF-10. A protease inhibitor, SWM-1, appears to play a regulatory role by inhibiting the protease TRY-5 (Fenker et al., 2014; Smith and Stanfield, 2011; Stanfield and Villeneuve, 2006). The other pathway includes the genes spe-8, spe-12, spe-19, spe-27 and spe-29; mutations of these genes result in self-sterile hermaphrodites, yet largely fertile males (Geldziler et al., 2005; L'Hernault et al., 1988; Minniti et al., 1996; Muhlrad et al., 2014; Nance et al., 2000; Nance et al., 1999). Mutations in spe-4, spe-6, spe-46 and spe-47 suppress the self-sterile phenotype caused by a mutation of spe-8; these genetic interactions suggest that these four genes function in the pathway at a position downstream of spe-8 (Gosney et al., 2008; LaMunyon et al., 2015; Liau et al., 2013; Muhlrad and Ward, 2002).

Zhao et al. (2018) identified zipt-7.1 as a positive regulator of sperm activation in C. elegans. A deletion mutation in zipt-7.1 results in a sperm-activation defect and causes partial or complete sterility in both sexes. ZIPT-7.1 is a ZIP family cation transporter that is homologous to the mammalian zinc transporter ZIP7 (also known as SLC39A7). When expressed in cultured cells, C. elegans ZIPT-7.1 selectively enhances zinc uptake. Moreover, ZIPT-7.1 is expressed in the germ line, consistent with a function in sperm. Based on genetic analysis, zipt-7.1 interacts with previously identified genes involved in sperm activation and functions downstream of spe-6. Finally, biochemical studies show that ZIPT-7.1 can bind to the presenilin SPE-4. Based on these findings, Zhao et al. (2018) proposed a new model for sperm activation. In spermatids, ZIPT-7.1 is localized to membranous organelles, which contain a higher level of zinc than the cytosol, and ZIPT-7.1 is not active. When sperm activation is triggered, a signal transduced by the SPE-8 group of proteins suppresses the function of SPE-4 and SPE-6, allowing ZIPT-7.1 to become active and transport zinc into the cytosol. The increase in cytosolic zinc promotes phenotypic changes associated with activation, including extension of the pseudopod. This model proposed that release of zinc from internal stores is part of the signal transduction process that mediates sperm activation.

The hypothesis that zinc signaling mediates sperm activation makes two key predictions: (1) dynamic changes in zinc occur during sperm activation; and (2) forced zinc entry will activate sperm and suppress the defect of zipt-7.1(lf) mutants. To test this zinc signaling model, we investigated these predictions. Here, we describe the development of a method for dynamic imaging of labile zinc during sperm activation using the zinc-specific fluorescence probe FluoZin-3 AM and time-lapse confocal imaging. We observed two phases of dynamic changes in labile zinc levels that coincide with sperm activation, consistent with aspects of the first prediction. Consistent with the second prediction, we discovered that the zinc ionophore pyrithione activated sperm in vitro and suppressed the defects of zipt-7.1(lf) mutants. To place zinc signaling in the context of other sperm activation pathways, we compared activation by zinc pyrithione to activation by extracellular zinc, the Na⁺/H⁺ antiporter monensin and the protease cocktail pronase in multiple mutant backgrounds (Nelson and Ward, 1980; Ward et al., 1983). These results indicate that the protease pathway does not require zinc signaling, whereas the zipt-7.1 pathway includes zinc signaling as an essential step in the process.

In C. elegans, sperm are produced by both hermaphrodites and males. Hermaphrodites only produce sperm early in life; the spermatids are activated soon after formation and retained in the spermathecae. Males produce sperm throughout adult life, and sperm are activated upon transfer to hermaphrodites during copulation. Male sperm are also stored in the hermaphrodite spermathecae, where fertilization occurs (Ward and Carrel, 1979). Our model in Fig. 1A makes two main predictions regarding dynamic changes in zinc levels during sperm activation: (1) in response to the sperm-activation signal and before pseudopod extension is complete, zinc levels will decrease in the vesicle lumen and increase in the cytosol as a result of ZIPT-7.1 mediated transport; and (2) following pseudopod extension, overall zinc levels will decrease as a result of exocytosis of vesicles that release zinc into the extracellular space. To test these predictions, we established an assay system that permits dynamic imaging of labile zinc levels in live sperm undergoing the activation process. Previous reports by Liu et al. (2013) and Zhao et al. (2018) demonstrated that zinc dyes such as Zinpyr-1 can be used to infer the distribution and level of labile zinc in live sperm using static images. Here, we sought to extend these methods by developing an approach to visualize dynamic changes in labile zinc in individual sperm cells during the process of sperm activation by using confocal microscopy to collect a time sequence of static images. To visualize labile zinc, we explored three dyes with a range of binding affinities for zinc: Newport Green DCF (Kd≈1 μM), FluoZin-3 AM (Kd≈15 nM) and Zinpyr-1 (Kd≈1 nM) (Walkup et al., 2000). These dyes fluoresce when bound to zinc, thus indicating the level of labile zinc. Sperm were dissected from males, incubated with one dye in sperm medium, and then visualized with fluorescent light. All three dyes stained sperm cells, although we observed some differences in the pattern of fluorescence. Furthermore, each dye displayed some sperm-to-sperm variability in the staining pattern, which may reflect trafficking of the dye within the sperm cell following uptake, which affects the subcellular localization of the dye.

These pilot studies revealed two sources of toxicity. First, all three zinc dyes caused dose-dependent toxicity as measured by the ability of sperm to undergo activation. Thus, high concentrations of dyes decreased the response to pronase-mediated sperm activation. Second, fluorescent light caused dose-dependent toxicity, also measured by the ability of sperm to be activated by pronase. These two toxic effects were additive, so that sperm exposed to dyes and fluorescent light displayed less activation than sperm exposed to either treatment alone. To minimize toxicity, we used two approaches. First, we chose to use FluoZin-3 AM because it caused the least toxicity, and we used the lowest concentration of FluoZin-3 AM that still allowed visualization. Second, fluorescent light exposure was minimized by only imaging the samples periodically rather than continuously. We developed the following protocol to obtain robust visualization of labile zinc with minimal toxicity: (1) dissected sperm were incubated with 10 μM FluoZin-3 AM in 1× Sperm Medium (SM) buffer for 10 min on a cover slide; (2) a smaller cover slide was placed over the sample, supported by either Vaseline strips or double-sided tape, creating a chamber; (3) excess FluoZin-3 AM was removed by flowing SM buffer through the chamber; (4) fluorescence and differential interference contrast images were captured every 10 s on a spinning disk confocal microscope with a 100× objective; and (5) during the imaging phase, pronase solution was flowed through the chamber to activate sperm.

To analyze the data, we used ImageJ software to measure fluorescence intensity for the region of the cell body. Fig. 1B shows images for one sperm, and Fig. 1C shows quantified fluorescence for that sperm at intervals of 10 s. In response to pronase treatment, individual sperm displayed rapid pseudopod extension; because each sperm initiates pseudopod extension at a different time, we defined as time 0 the last time point before pseudopod extension. This made it possible to plot the data from multiple sperm on the same axis and determine average values and standard deviation at each time (Fig. 1E).

For individual sperm before pseudopod extension, the FluoZin-3 fluorescence was distributed across the cell body, and quantified fluorescence intensity was stable over time (Fig. 1B,C,E; Fig. S2A, time −120 to 0 s). This signal might derive from labile zinc in the cytosol, the lumen of membranous organelles, or both; this technique does not resolve signal derived from the cytosol versus the lumen of organelles. We did not observe an increase in cytosolic zinc levels prior to pseudopod extension, in contrast to the model predictions. There are several explanations for this negative result, including a problem with the model or technical issues that may have prevented us from observing such a change.

During the period of initial pseudopod extension, which lasts ∼60 s, the cell body area of the sperm decreased significantly (Fig. 1D) and the fluorescence intensity of that area increased significantly (Fig. 1E, time ∼0 to ∼60 s); the pseudopod extension lacked FluoZin-3 fluorescence (Fig. 1B). Over the next 60 s, the area of the cell body did not display a significant change (Fig. 1D), whereas the fluorescence intensity decreased significantly by about 42% (Fig. 1E, time ∼60 to ∼120 s). Thus, two dynamic changes were observed: from 0-60 s the cell body area shrinks, and fluorescence intensity increases, and from 60-120 s fluorescence intensity decreases without a change in cell body area (Fig. 1F).

The increase in FluoZin-3 fluorescence occurring at the same time as the reduction in the area of the cell body suggests that FluoZin-3:zinc complexes might be compressed into a smaller area as a result of pseudopod extension, resulting in increased fluorescence intensity. To investigate this hypothesis, we used the dye CellTracker Red as a control instead of FluoZin-3 AM. The initial increase in fluorescence intensity and decrease in cell body area was also observed in sperm loaded with CellTracker Red, suggesting that this increased fluorescence intensity is caused by the dye becoming restricted to a smaller area (Fig. 2A-D, Fig. S2B). By contrast, the CellTracker Red fluorescence did not display the second dynamic change; after the increase in fluorescence intensity, the CellTracker Red signal remained relatively constant, whereas the FluoZin-3 fluorescence intensity decreased to a new steady state (Fig. 2E).

Although the model predicts an increase in the concentration of cytosolic zinc during the phase of pseudopod extension, it does not predict that the mechanism would be concentration in a smaller volume. This unexpected finding might reveal an alternative mechanism of dynamic regulation of zinc levels, or it might be a consequence rather than a cause of sperm activation. Several explanations might account for the decrease in FluoZin-3 fluorescence: (1) FluoZin-3::zinc complexes in the lumen of membranous organelles are secreted from sperm cells as a result of exocytosis during the final stages of pseudopod extension; (2) a zinc transporter, such as a member of the CDF protein family, promotes efflux of zinc from the cytosol across the plasma membrane to the extracellular space; (3) labile zinc is transferred from FluoZin-3 to alternative binding sites, such as proteins, thus decreasing fluorescence; or (4) photobleaching of the FluoZin-3::zinc complex results in reduced fluorescence intensity.

Liu et al. (2013) discovered that extracellular zinc can activate wild-type C. elegans sperm in vitro. Zhao et al. (2018) showed that zipt-7.1 is necessary for the full effect of extracellular zinc, suggesting that ZIPT-7.1 mediates the transport of extracellular zinc into the cytosol. zipt-7.1 is part of the spe-8 pathway of sperm activation (Fig. 3A), and zipt-7.1(lf) mutants display a cold-sensitive sterile phenotype as a result of defective sperm activation (Fig. S1A). Consistent with the model that zipt-7.1 functions in the germline, zipt-7.1(lf) mutants did not display hypersensitivity to growth defects caused by zinc deficiency (Fig. S1B). To characterize how activation by extracellular zinc interacts with genes that regulate sperm activation, we performed a dose-response analysis with zipt-7.1(lf), spe-8(lf) and snf-10(lf) mutants. Wild-type sperm displayed dose-dependent activation, with ∼20% activated by 0.5 mM zinc and ∼70% activated by 2.0 mM zinc (Fig. 3C). These results are generally consistent with previous findings of Liu et al. (2013), but those investigators reported >80% sperm activation by 0.4 mM zinc. This discrepancy in the dose response might reflect differences in methodology or buffer conditions.

Extracellular zinc weakly activated the sperm of zipt-7.1(lf) mutant animals: 1.0 mM zinc caused ∼10% sperm activation in zipt-7.1(lf) mutants, significantly lower than the ∼80% sperm activation in zipt-7.1(+) controls and significantly higher than untreated zipt-7.1(lf) mutants (Fig. 3B, Table 1). These results are consistent with previous findings (Liu et al., 2013; Zhao et al., 2018) and suggest that zipt-7.1 is necessary for most, but not all, of the extracellular zinc-mediated sperm activation. Thus, these data suggest that a ZIPT-7.1-independent transport mechanism also exists. The defect was more severe in spe-8(lf) mutant sperm, which displayed no detectable activation in response to extracellular zinc (Fig. 3C, Table 1). Thus, spe-8 is necessary for all extracellular zinc-mediated sperm activation, consistent with previous findings (Liu et al., 2013). These results suggest that spe-8 positively regulates zipt-7.1 and the ZIPT-7.1-independent transport mechanism, perhaps by functioning upstream of these processes in a signaling pathway (Fig. 3D). By contrast, extracellular zinc robustly activated snf-10(lf) mutant sperm: a small defect was observed at 0.5 mM zinc, but mutant and wild type displayed similar activation at 1.0 and 2.0 mM zinc (Fig. 5G, Table 1). Thus, snf-10 is not necessary for sperm activation caused by extracellular zinc.

Based on the hypothesis that zinc functions downstream of zipt-7.1 to activate sperm, we predict that methods that increase the concentration of cytosolic zinc would bypass zipt-7.1 and cause sperm activation even in the zipt-7.1(lf) mutant. Pyrithione is a well-characterized zinc ionophore that can bind zinc and transport it across cellular membranes (Fig. 4A,B); in the presence of extracellular zinc, pyrithione can increase the cytosolic concentration of zinc in cells (Chen and Ai, 2016). In the presence of 1 mM supplemental extracellular zinc, as little as 10 nM pyrithione increased the percentage of zipt-7.1(lf) mutant sperm that displayed activation from ∼5% to ∼40% (Fig. 4C). A dose-response curve comparing zipt-7.1(lf) and zipt-7.1(+) control sperm showed that the activation percentage was similar to or only slightly lower than control (Fig. 4C,E). Thus, pyrithione was sufficient to bypass the activation defect of zipt-7.1(lf) mutant sperm in response to extracellular zinc. To investigate the role of the supplemental zinc, we performed the experiment without adding supplemental zinc. Because the sperm are dissected from animals, there is likely to be residual zinc in the dissected samples. For the zipt-7.1(+) control sperm, the dose-response curve shifted such that there was less activation at low concentrations and more activation at high concentrations; these sperm did not display activation in the absence of pyrithione, displayed partial activation of ∼25% with 0.1 µM pyrithione, and displayed full activation of ∼85% at 1 µM pyrithione (Fig. 4D,E). The dose-response curve for zipt-7.1(lf) mutants was shifted in a similar manner (Fig. 4D,E). These results suggest that the dissected sperm have residual zinc in the preparation, but the concentration is less than 1 mM, so that a higher concentration of pyrithione is necessary for full activation. In the presence of 1 mM supplemental extracellular zinc, even very low concentrations of pyrithione were adequate for the full response.

To test rigorously whether pyrithione-induced sperm activation requires zinc, we performed the in vitro activation experiment with pyrithione in the presence of either the cell-permeable zinc chelator TPEN [N,N,N′,N′-tetrakis (2-pyridinylmethyl)-1,2-ethanediamine] or the extracellular general cation chelator EDTA. Treatment with either 100 µM TPEN or 100 µM EDTA abolished sperm activation caused by pyrithione (Fig. 4F). These results suggest that pyrithione activated sperm by functioning as a zinc ionophore.

zipt-7.1 appears to function at the end of the spe-8 sperm activation pathway, because the zipt-7.1(lf) phenotype is epistatic to the spe-6(lf) mutant phenotype (Fig. 3A) (Zhao et al., 2018). Based on this pathway, we predict that using pyrithione to increase cytosolic zinc levels should bypass the requirement for all of the spe-8 pathway genes. Consistent with this model, pyrithione effectively activated the sperm of spe-8(lf) mutant animals, with a similar dose-response curve to wild type (Fig. 4G, Table 1). Thus, whereas spe-8 is necessary for sperm activation by extracellular zinc, it is not necessary for sperm activation by pyrithione. These results indicate that spe-8 functions downstream of extracellular zinc and upstream of cytosolic zinc, consistent with the suggestion that spe-8 promotes the ability of extracellular zinc to enter the cytosol. Furthermore, 1 µM pyrithione activated the sperm of spe-12(lf), spe-27(lf) and spe-29(lf) mutants (Fig. 4H). The snf-10(lf) mutant can be activated by high concentrations of extracellular zinc. As expected, pyrithione effectively activated the snf-10(lf) mutant sperm, similar to the wild-type control (Fig. 5H, Table 1). Overall, these results indicate that the zinc ionophore pyrithione effectively activated sperm in all mutant backgrounds that were analyzed (Table 1).

Stanfield and colleagues identified a protease pathway that regulates sperm activation (Fenker et al., 2014; Smith and Stanfield, 2011; Stanfield and Villeneuve, 2006). Loss-of-function mutations of swm-1, which encodes a protease inhibitor, result in prematurely activated sperm in the males. This premature activation is alleviated by the loss of try-5, which encodes a secreted protease, suggesting that TRY-5 activates sperm by functioning extracellularly to cleave a protein on the surface of sperm. The substrate of TRY-5 has not been established, but SNF-10 emerged as a candidate based on genetic studies. Similar to try-5(lf) mutations, snf-10(lf) mutations suppress premature sperm activation caused by swm-1(lf) mutations, and snf-10 encodes an SLC6 family transporter protein predicted to be on the surface of sperm (Fig. 5A). To define the relationship between zinc signaling and the protease pathway, we analyzed sperm activation by pronase. Pronase is a cocktail of proteases isolated from the bacteria Streptomyces griseus. This commonly used reagent is highly effective in activating wild-type C. elegans sperm (Ward et al., 1983). An emerging model based on these genetic and pharmacological observations is that pronase mimics the endogenous protease TRY-5, functioning extracellularly to cleave a cell-surface protein on sperm, perhaps SNF-10 (Fig. 5B).

To quantify the response of mutant sperm to pronase, we performed a dose-response analysis. For wild-type sperm, treatment with 50 µg/ml pronase caused ∼40% of sperm to display activation, whereas treatment with 100 µg/ml caused over 95% of sperm to display activation (Fig. 5E). Pronase robustly activated sperm of zipt-7.1(lf) mutant animals: 100 μg/ml pronase caused ∼75% sperm activation in zipt-7.1(lf) mutants and ∼95% in zipt-7.1(+) controls (Fig. 5C, Table 1). This small but statistically significant difference suggests that zipt-7.1 makes a minor contribution to pronase-mediated sperm activation. By contrast, pronase caused only a small amount of activation of spe-8(lf) mutant sperm: 100 μg/ml pronase caused <10% sperm activation in spe-8(lf) mutants compared with over 95% in wild-type controls (Fig. 5E, Table 1). These results are consistent with previous findings (Shakes and Ward, 1989) and suggest that spe-8 is necessary for the majority of the pronase-mediated sperm activation. snf-10(lf) mutant sperm displayed no activation with any tested pronase concentration (Fig. 5F, Table 1), consistent with previous findings (Fenker et al., 2014). Thus, snf-10 is necessary for all pronase-mediated sperm activation.

These genetic studies indicate that the pronase pathway may not require an increase in cytosolic zinc concentration to activate sperm. To test this model directly, we determined the effect of the zinc chelators TPEN and EDTA on pronase-mediated sperm activation. At concentrations that effectively blocked pyrithione-mediated sperm activation, these chelators had no significant effect on pronase-mediated sperm activation (Fig. 5D). This finding is consistent with results of Liu et al. (2013) and suggests that protease-induced sperm activation does not require zinc signaling.

Monensin is a complex organic molecule isolated from the bacteria Streptomyces cinnamonensis (Fig. 6A) (Pinkerton and Steinrauf, 1970). Monensin functions as an ionophore and is widely used for its antibiotic activity. Monensin was the first compound reported to activate C. elegans sperm in vitro (Nelson and Ward, 1980). The proposed mechanism is that monensin treatment facilitates sodium and proton exchange across the sperm plasma membrane, thereby increasing the pH of the sperm cytosol (Fig. 6B) (Pressman, 1976; Ward et al., 1983). Consistent with this model, the pH increase in the sperm precedes the visible morphological changes of the activation process. Furthermore, the weak base triethanolamine (TEA) can also activate sperm in vitro, consistent with the suggestion that an increase in pH can activate C. elegans sperm (Ward et al., 1983).

To elucidate the relationships between monensin and the genes necessary for sperm activation, we tested the ability of monensin to activate sperm of zipt-7.1(lf), spe-8(lf) and snf-10(lf) mutants. To establish assay conditions, we used 0.1 μM monensin at a range of extracellular pH levels from 6.6 to 9.0. Monensin did not activate sperm in the pH range 6.6-7.4 and displayed increasing effectiveness as pH increased with maximum induction at pH 8.6 (Fig. 6C). Thus, a basic pH environment was necessary for monensin to induce sperm activation, and pH 8.6 was used for subsequent experiments. To determine the dose-response curve to monensin, we investigated concentrations from 0 to 10 µM. Wild-type sperm displayed the maximum response with 0.1-0.2 μM monensin, with 50-60% of the sperm displaying activation; higher doses were progressively less effective (Fig. 6D,E).

The dose-response curve of zipt-7.1(lf) mutant sperm was shifted compared with control: there was a significantly lower response to low doses of monensin (0.1-0.2 μM), and a small but significantly higher response to the higher dose of 5 μM monensin (Fig. 6D). The maximum response, at 0.5 μM monensin, was about 40% sperm activation, which was lower than the maximum response for control sperm. Sperm of spe-8 pathway mutants have been reported to respond to increased intracellular pH (Minniti et al., 1996; Shakes and Ward, 1989). Consistent with these findings, wild-type and spe-8(lf) sperm displayed a similar response to low doses of monensin; interestingly, spe-8(lf) mutants displayed a higher response to high doses, such as 2 μM monensin (Fig. 6E). The dose-response curve of snf-10(lf) mutant sperm was shifted compared with wild type: there was a significantly lower response to the low dose of 0.1 μM monensin, and a small but significantly higher response to the higher dose of 2 μM monensin (Fig. 6F). The maximum response, at 0.5 μM monensin, was about 60% sperm activation, which is similar to the maximum response for wild-type sperm. These results are consistent with previous reports that snf-10(lf) sperm can be activated with monensin (Fenker et al., 2014). These results indicate that zipt-7.1(lf), spe-8(lf) and snf-10(lf) sperm can all respond to the pH change caused by monensin treatment, but the dose-response curves are shifted (Table 1).

Zinc has well-established biochemical functions; stable binding of zinc to proteins can influence tertiary structure or facilitate catalysis (Kambe et al., 2015). More recently, zinc was proposed to function as a second messenger similar to calcium (Yamasaki et al., 2007). Because zinc signaling is a dynamic process and techniques to monitor zinc fluxes are still primitive, zinc signaling is just beginning to be explored, and crucial questions remain. Here, we propose that sperm activation in C. elegans is mediated by zinc signaling; specifically, that ZIPT-7.1 mediates release of zinc from intracellular organelles in spermatids, and the increased level of cytosolic zinc promotes sperm activation.

If zinc signaling regulates sperm activation, then we predict there will be dynamic changes in zinc concentration during this process. To test this prediction, we developed methods to monitor dynamic changes in labile zinc during sperm activation triggered by pronase. Although our model predicts an increase in cytosolic zinc prior to and during pseudopod extension, no change in FluoZin-3 fluorescence was observed prior to pseudopod extension. There are several possible explanations for this negative result. If this is a true negative result, then the model is not supported. However, this might be a false negative result for several reasons: (1) FluoZin-3 may be present in the cytosol and the lumen of the membranous organelles, and thus a movement of zinc from the organelle lumen to the cytosol would not cause a change in overall FluoZin-3 fluorescence; (2) the kinetics of zinc binding to FluoZin-3 might be too slow to detect this dynamic change, or the change might be too small compared with background; or (3) pronase might not initiate zinc signaling but rather activate a parallel pathway of sperm activation. Consistent with this possibility, zinc chelators did not inhibit sperm activation caused by pronase.

During the first 60 s following initiation of pseudopod extension, the concentration of labile zinc in the cell body increased, as measured by FluoZin-3 fluorescence. This is likely caused by a reduction in the volume of the cell body, as the pseudopod appears to lack labile zinc and a control cytosolic dye displayed a similar pattern. Although our model predicted an increase in FluoZin-3 fluorescence during this period, it did not predict that the mechanism would be concentration in a reduced volume. This unexpected mechanism might constitute a zinc signaling event, or it might contribute to a prior zinc signaling event mediated by ZIPT-7.1. If this is a zinc signaling event, then an alternative interpretation of the zipt-7.1 mutant phenotype is that zipt-7.1 functions in immature sperm to promote zinc accumulation in the cells as a necessary precursor to this concentration event. Alternatively, this might be a consequence of pseudopod extension that does not have a signaling function.

During the second phase, which occurs 60-120 s following the initiation of pseudopod extension, the concentration of labile zinc decreases in the cell body; a control cytosolic dye did not display this pattern. Several models might explain this observation. We favor the interpretation that this decrease results from exocytosis from intracellular vesicles. It has been well documented that sperm display fusion of membranous organelles during the activation process, which may provide the membrane necessary for the pseudopod extension (Washington and Ward, 2006). Our results are consistent with the model that residual zinc in the lumen of the membranous organelles is released into the extracellular space by exocytosis, reducing the level of labile zinc in the cell body. Other possible interpretations for the mechanism of reduced labile zinc include: zinc efflux across the plasma membrane, perhaps by a member of the CDF family; transfer of zinc from FluoZin-3 to another binding molecule, such as a protein; and photobleaching of the FluoZin-3 fluorescence.

The system we developed for dynamic imaging of labile zinc during sperm activation has the potential to be modified to address additional questions. First, we used pronase as the activator; by examining alternative activators, such as zinc, pyrithione and monensin, it would be possible to compare the pattern of zinc dynamics. Second, we used wild-type sperm; examining mutant sperm such as zipt-7.1 with pronase as the activator or spe-8 and snf-10 with pyrithione or monensin as the activator could elucidate the function of these genes. Third, we used FluoZin-3 AM as the zinc dye; by using an extracellular zinc dye, it would be possible to test directly the hypothesis that the decrease in labile zinc is due to exocytosis or efflux. By using a protein-based zinc sensor that is localized to only the cytosol or only the lumen of the membranous organelles, it would be possible to directly test whether zinc is transported from the vesicle lumen to the cytosol (Pratt et al., 2021).

Zinc signaling has been proposed to occur both extracellularly and intracellularly. Zinc is concentrated in synaptic vesicles and released into the synaptic cleft to cause a transient increase in the concentration of zinc. The precise role of this dynamic change in extracellular zinc levels has not been established, but high levels of zinc might modulate the activity of neurotransmitter receptors (Anderson et al., 2017; McAllister and Dyck, 2017). Zinc is released from mammalian oocytes during fertilization, a phenomenon described as ‘zinc sparks’ because it causes a transient increase in the level of extracellular zinc (Kim et al., 2011; Que et al., 2015). In both cases, zinc is released by exocytosis of zinc-containing vesicles.

Several examples of intracellular zinc fluxes have been reported. In vertebrate mast cells stimulated to undergo degranulation by an extracellular ligand, Yamasaki et al. (2007) visualized an intracellular ‘zinc wave’. The endoplasmic reticulum appears to be the origin of this zinc wave. Mast cells, like nematode sperm, mount a rapid response to external signals that includes fusion of intracellular vesicles to the plasma membrane, which appears to be mediated by zinc signaling (Ellis and Stanfield, 2014; Krystel-Whittemore et al., 2015). In T lymphocytes and leukocytes responding to extracellular signals, a rapid increase in cytosolic zinc has been documented (Haase et al., 2008; Yu et al., 2011).

C. elegans ZIPT-7.1 is homologous to vertebrate ZIP7. Hogstrand et al. (2009) proposed that vertebrate ZIP7 mediates a zinc signal in breast epithelial cells and is localized to intracellular membranes. Furthermore, Taylor et al. (2012) reported that ZIP7 activity is regulated by phosphorylation. Our results raise the possibility that ZIPT-7.1 is regulated by the spe-8 group of genes, perhaps by a post-translational event, such as phosphorylation.

To evaluate the functional significance of cytosolic zinc, we analyzed sperm activation in vitro. Liu et al. (2013) showed that high levels of extracellular zinc can activate C. elegans sperm, which is consistent with the model that zinc functions extracellularly or intracellularly. To distinguish between these possibilities, we analyzed the zinc ionophore pyrithione. Pyrithione was a potent activator, indicating that zinc functions in the cytosol to activate sperm. Consistent with this interpretation, the extracellular zinc chelator EDTA and the intracellular zinc chelator TPEN blocked the activity of pyrithione. Although a number of chemicals have been demonstrated to activate C. elegans sperm in vitro, pyrithione has not previously been reported to have this effect. These results indicate that exposure to extracellular zinc can activate sperm because some zinc crosses the plasma membrane and enters the cytosol. The zinc transporter ZIPT-7.1 plays an important role in promoting zinc entry, as zipt-7.1(lf) mutants display only a small response to extracellular zinc. The finding that some response remains in zipt-7.1(lf) mutants suggests that there is an alternative route of entry for zinc, perhaps another zinc transporter in the ZIP family.

There is a long history of analyzing the effects of zinc on animal sperm, including that of vertebrates (human, mouse, hamster), sea urchin and C. elegans (Andrews et al., 1994; Aonuma et al., 1978; Clapper et al., 1985; Kerns et al., 2018; Lishko et al., 2010; Yamaguchi et al., 2009; Zhao et al., 2016). Sea urchin spermatids are activated by sea water, and zinc is the active ingredient in sea water (Clapper et al., 1985); zinc causes an increase in intracellular calcium levels and intracellular pH, which triggers the acrosome reaction, a membrane fusion event (Beltrán et al., 2014; Clapper et al., 1985). In vertebrate sperm, the role of zinc has been controversial; experiments with the addition of supplemental zinc and zinc chelators suggest a variety of roles for zinc, indicating its importance, but specific functions have not been well defined (Zhao et al., 2016). The C. elegans model is similar and distinct from mechanisms observed in sea urchins. In both cases, an increase in the level of zinc stimulates sperm activation. However, extracellular zinc in sea water is the physiological source in sea urchins, whereas intracellular stores are likely to be the physiological source in C. elegans.

A variety of molecules have been identified that can activate C. elegans sperm in vitro (Bae et al., 2009; Liu et al., 2014; Machaca et al., 1996; Sakamoto and Imai, 2017). These activators are likely to mimic physiological triggers or affect endogenous signal transduction pathways; thus, the identification and characterization of such activators can provide clues to the physiological process. Here, we used dose-response curves to analyze quantitatively the effectiveness of four activators (zinc, pyrithione, pronase and monensin) in three genetic backgrounds (zipt-7.1, spe-8 and snf-10).

Both spe-8 and snf-10 mutants displayed strong defects in responding to pronase, and snf-10 mutants were completely unresponsive, consistent with previous reports (Fenker et al., 2014; Shakes and Ward, 1989). By contrast, zipt-7.1 mutants responded effectively with only minor differences compared with control. Thus, although Zhao et al. (2018) placed zipt-7.1 in the spe-8 pathway, zipt-7.1 mutants are phenotypically distinct from other spe-8 group mutants with respect to their response to pronase. zipt-7.1 functions at the end of the pathway to regulate zinc transport, suggesting that the protease pathway does not require zinc signaling. Consistent with this interpretation, we showed that pronase can induce sperm activation in the presence of zinc chelators. By contrast, other members of the spe-8 group function further upstream and appear to be involved in the protease pathway as well as the zinc pathway, indicating that the spe-8 pathway may diverge upstream of zipt-7.1 (Fig. 7). Although the pronase pathway does not require zinc signaling, zinc and pyrithione were sufficient to activate snf-10 mutant sperm, indicating that cytosolic zinc can bypass the pronase pathway.

Monensin induces sperm activation by increasing intracellular pH. In many species, alkalinization of the cytosol is a crucial aspect of the regulation of sperm mobility (Hamamah and Gatti, 1998). In sea urchins, sperm mobility is closely associated with intracellular pH (Christen et al., 1982), and these sperm can also be activated with monensin (Hansbrough and Garbers, 1981). In both sea urchins and humans, intracellular pH levels control sperm mobility at least partially by regulating intracellular calcium levels, as the sperm-specific calcium channel CatSper is activated by alkaline pH (Kirichok et al., 2006; Ren et al., 2001; Seifert et al., 2015). In both species, voltage-gated proton channels have been identified and could be key to the alkalinization process (Lishko et al., 2010; Windler et al., 2018).

Here, we tested sperm activation by monensin in both a pH- and dose-dependent manner in different genetic backgrounds. Responsiveness to monensin was slightly but significantly altered in all mutants; compared with wild type, mutants were less sensitive at lower concentrations and more responsive at higher concentrations. However, none of the mutants effectively blocked sperm activation by monensin, indicating that alkalinization of the cytosol can bypass the requirement for the protease pathway and the zinc pathway. These results suggest two possible interpretations: (1) alkalinization of the cytosol is caused by an unidentified sperm activation pathway that functions in parallel to the zinc and protease pathways; and (2) alkalinization of the cytosol is a downstream effect of the zinc and protease pathways. Taken together, these results indicate that extracellular triggers for sperm activation, including protease and zinc, require certain genes and intracellular pathways to be effective. By contrast, pyrithione and monensin change ion concentrations in the cytosol and can largely bypass known genetic regulators of sperm activation. Thus, high levels of zinc and low levels of protons may be downstream effectors of sperm activation pathways (Fig. 7).

Because sperm defects are a common cause of human infertility, the identification of methods to activate sperm in vitro may establish a foundation of knowledge that leads to new approaches to treating infertility. Zinc has been demonstrated to play a role in human fertility, and zinc treatment has been shown to increase sperm motility in some patients (Allouche-Fitoussi and Breitbart, 2020; Omu et al., 1998; Salas-Huetos et al., 2019; Salas-Huetos et al., 2018). In this regard, the identification of pyrithione as an activator of C. elegans sperm may lead to new methods to activate human sperm.

C. elegans strains were cultured on Nematode Growth Medium (NGM) dishes with a small lawn of Escherichia coli strain OP50 at 20°C unless otherwise indicated. All strains were derived from the wild-type strain Bristol N2 (Brenner, 1974). Mutations used in this study were: spe-8(hc50) I, spe-12(hc76) I, spe-27(it110) IV, spe-29(it127) IV, zipt-7.1(ok971) IV, him-5(e1490) V, snf-10(hc194) V. To obtain males, we generated the zipt-7.1(ok971); him-5(e1490) double mutant; him-5(e1490) males were used as the control strain in these cases. him-5 loss-of-function mutations do not affect sperm activation, but increase nondisjunction of the sex chromosome, thereby increasing the frequency of XO male progeny (Hodgkin et al., 1979; Nelson et al., 1982).

Sperm were dissected from males as described below into 10 μl of 1× SM buffer with 10 μM FluoZin-3 AM (Invitrogen) (excitation: 494 nm; emission: 516 nm) or 1 μM CellTracker Red CMTPX (Invitrogen). After 10 min of incubation in a moisture chamber, a coverslip was placed over the sample, supported by a Vaseline strip or double-sided tape, which creates a chamber on the slide. The remaining space of the chamber was filled with 1× SM buffer by perfusion through one opening. Excess dye was washed away by perfusing 200 μl of 1× SM buffer through the chamber. Perfusion was accomplished by adding liquid medium to one opening of the chamber and removing liquid at the other opening with a Kimwipe. After imaging began, sperm were activated by perfusing 200 µl of 200 μg/ml pronase in 1× SM buffer. Images were obtained with a Nikon Spinning Disk Confocal microscope with 100× objective. To normalize the signal for individual sperm, we subtracted the background value, set the fluorescence intensity at time 0 equal to 0, and set the maximum intensity during the experiment to 1.0. Time 0 was defined as the last time point prior to obvious morphological change (pseudopod extension).

Sperm activation experiments were performed as described by Singaravelu et al. (2011) with minor modifications. Briefly, virgin L4 males were cultured on NGM dishes with a lawn of E. coli and no hermaphrodites for 2 days, briefly transferred to dishes with no bacteria, and dissected to release sperm in 10 μl of 1× SM buffer (50 mM HEPES, 25 mM KCl, 45 mM NaCl, 1 mM MgSO₄, 5 mM CaCl₂, 10 mM D-glucose) on a glass microscope slide. In experiments with compounds that promote sperm activation, sperm medium included monensin (Sigma-Aldrich, M5273), pronase (Roche), zinc chloride (Sigma-Aldrich, 229997) and/or pyrithione sodium salt (2-mercaptopyridine N-oxide sodium salt) (Sigma-Aldrich, H3261). In experiments using monensin and pyrithione, 1% DMSO was present in the sperm medium. In experiments with compounds that might inhibit sperm activation, sperm medium included TPEN (Sigma-Aldrich) or EDTA (Corning). In experiments using TPEN, 1% ethanol was present in the sperm medium. The pH of the sperm medium was 7.4 unless indicated otherwise, as in Fig. 6. In experiments using TPEN or EDTA, we prepared sperm medium without MgSO₄ and CaCl₂. To assay sperm activation, which we defined as the presence of a pseudopod, we incubated dissected sperm in a moisture chamber for 20 min, covered the sample with a coverslip, observed with a Zeiss Axioplan 2 microscope, scored each sperm for the presence of a pseudopod by morphological criteria (yes or no), and reported the percentage activated sperm. In the dose-response assay shown in Fig. 3B, we observed a small amount of precipitation when the concentration of zinc was 2000 µM, suggesting that the solution was saturated.

The analysis of growth rate in zinc-deficient medium was performed as described by Dietrich et al. (2017). Briefly, synchronized L1 stage worms were cultured on Noble Agar Minimal Media (NAMM) dishes seeded with 5× concentrated E. coli OP50 as the food source and supplemented with various concentrations of TPEN dissolved in ethanol. After 3 days of culture, images of individual worms were obtained using a Leica M80 microscope equipped with a Leica KL 300 LED light source and a Leica IC80 HD Digital Camera Module. Worm length was measured using ImageJ software (NIH) by drawing a line from head to tail through the midline of the worm as described by Roh et al. (2012).

L4 stage hermaphrodites developed at 15°C, 20°C or 25°C were placed on individual NGM dishes seeded with E. coli OP50, cultured at the appropriate temperature, and transferred to fresh dishes daily. Progeny on the dishes were evaluated after 2 days. We defined hermaphrodites with no progeny that hatched or fewer than 30 progeny that hatched as ‘sterile’ and ‘low fertility’, respectively.

Values were compared using two-tailed, unpaired Student's t-tests, except for the data shown in Fig. S1, for which the Chi-square test was used. All data represent mean±s.d. Statistical significance is indicated as *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

We thank Mike Shih and the Washington University Center for Cellular Imaging (WUCCI) for microscopy assistance. We thank Zuzana Kocsisova, David Piston, Tim Schedl and members of the Schedl lab (Washington University in St. Louis), and Gillian Stanfield and Kristin Fenker (University of Utah) for technical assistance. Some stains were obtained from the Caenorhabditis Genetics Center (CGC), which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440).