Functional Relationships between Capacitation-dependent Cell Signaling and Compartmentalized Metabolic Pathways in Murine Spermatozoa*

Spermatozoa are highly polarized cells with specific metabolic pathways compartmentalized in different regions. Previously, we hypothesized that glycolysis is organized in the fibrous sheath of the flagellum to provide ATP to dynein ATPases that generate motility and to protein kinases that regulate motility. Although a recent report suggested that glucose is not essential for murine sperm capacitation, we demonstrated that glucose (but not lactate or pyruvate) was necessary and sufficient to support the protein tyrosine phosphorylation events associated with capacitation. The effect of glucose on this signaling pathway was downstream of cAMP, and appeared to arise indirectly as a consequence of metabolism as opposed to a direct signaling effect. Moreover, the phosphorylation events were not affected by uncouplers of oxidative respiration, inhibitors of electron transfer, or by a lack of substrates for oxidative respiration in the medium. Further experiments aimed at identifying potential regulators of sperm glycolysis focused on a germ cell-specific isoform of hexokinase, HK1-SC, which localizes to the fibrous sheath. HK1-SC activity and biochemical localization did not change during sperm capacitation, suggesting that glycolysis in sperm is regulated either at the level of substrate availability or by downstream enzymes. These data support the hypothesis that ATP specifically produced by a compartmentalized glycolytic pathway in the principal piece of the flagellum, as opposed to ATP generated by mitochondria in the mid-piece, is strictly required for protein tyrosine phosphorylation events that take place during sperm capacitation. The relationship between these pathways suggests that spermatozoa offer a model system for the study of integration of compartmentalized metabolic and signaling pathways.

Mammalian spermatozoa are highly differentiated cells that display extreme polarization of cellular architecture and function. For example, the sperm head has evolved to interact with the egg's extracellular matrix and plasma membrane, and contains the paternal genetic material, whereas the sperm flagellum acts to pro-vide motility for these cells. In regard to this polarization of function, sperm have two major constraints. First, they have little cytoplasm, and therefore have a reduced ability to translocate metabolic intermediates or substrates from one region to another. In addition, they are transcriptionally inactive, and therefore cannot make new proteins in response to changing needs. To overcome these constraints, we, along with others, have hypothesized that sperm possess compartmentalized metabolic and signaling pathways in specific regions of the cell poised to function in a localized fashion (1)(2)(3)(4).
The most obvious example of metabolic compartmentalization in spermatozoa is that of oxidative respiration. This pathway is restricted to the mid-piece of the flagellum, because mitochondria are located solely in this region. Oxidative respiration provides the most efficient generation of ATP, yet the major sites of ATP consumption in sperm are the dynein ATPases found associated with the axoneme throughout the entire length of the flagellum (5). In some species such as the sea urchin, a phosphorylcreatine shuttle has been observed to transfer high-energy phosphate from the mitochondria to the principal piece (6). However, in most mammalian species, this system either is poorly developed or is absent entirely (see Kaldis et al. (7) for a review). How then do the principal piece and end-piece of the flagellum meet their energy needs?
We, along with others, have hypothesized that glycolysis is compartmentalized in a cytoskeletal element in the principal piece of the flagellum known as the "fibrous sheath" (FS). 1 This structure would then be able to provide localized ATP production to the dynein ATPases along the entire length of the principal piece of the flagellum (3,4,8). This theory was first suggested following observations that the enzymes of glycolysis downstream from aldolase are all associated with a cytoskeletal structure of the rabbit sperm flagellum (2). It was later demonstrated that one of these enzymes, glyceraldehyde-3phosphate dehydrogenase (GAPDH), localizes to the FS in several mammalian species (3,9). Concurrently, a germ cell-specific isoform of type 1 hexokinase (HK1-SC), the first enzyme in glycolysis, was also found to target to the FS (4,8). This localization demonstrated for the first time that the proximal portion of the glycolytic pathway was also associated with the FS.
In addition to tethering enzymes of glycolysis, the FS can also act as a scaffold for signaling molecules involved in the control of motility. A major component of the FS has been identified as an A-kinase anchoring protein (AKAP) (10). This protein, AKAP4, has been shown to bind protein kinase A (10), which plays a role in many of the phosphorylation events postulated to govern motility (11). Therefore, glycolysis organized in the FS could be used not only to provide motility by generating ATP for the dynein ATPases, but also to help regulate motility by providing ATP to protein kinases (4).
Changes in sperm motility occur not only within the epididymis as sperm acquire motility, but also within the female reproductive tract as sperm mature functionally and become competent to fertilize an egg. This process, known as "capacitation," is associated with the activation of a novel signal transduction cascade that includes the efflux of cholesterol from membranes, alterations in membrane phospholipid content, alterations in intracellular calcium and bicarbonate ions, alterations in membrane potential, and elevations in cAMP that ultimately result in the downstream phosphorylation of a number of proteins on tyrosine residues (see Kopf et al. (12) for a review). It has long been recognized that capacitated sperm display increased energy demands (13), presumably to effect these changes in sperm signaling and function. However, the relationship between the signaling events associated with capacitation and changes in sperm energy metabolism is poorly understood, despite the co-compartmentalization of these pathways in regions such as the FS.
Several studies using different model systems have been published on the role of glucose in sperm capacitation and fertilization. In some species (e.g. the mouse and human), glucose is necessary for capacitation, the onset of hyperactivated motility, and acrosomal exocytosis (14 -17). Contrary to the stimulatory role seen in the mouse and human, glucose inhibits capacitation in the bull (18) and guinea pig (19). These species differences might be related to the fact that glucose can have other effects on sperm unrelated to the generation of ATP. For example, glycolysis might affect sperm function by regulating intracellular pH (18). Alternatively, glucose might function as a signaling molecule by acting on regulatory enzymes such as cyclic nucleotide phosphodiesterases (20). Finally, glucose might be used in alternative metabolic pathways such as the pentose phosphate pathway (PPP).
A possible role for the PPP in capacitation/fertilization is pertinent to a discussion of metabolic compartmentalization in sperm in that we have hypothesized that this pathway might be localized to the mid-piece and/or sperm head (4). Although GAPDH is found in the FS, it and most other glycolytic enzymes have not been found elsewhere in the sperm. Yet the same germ cell-specific HK found in the FS, HK1-SC, is also found in the mid-piece and head (4). The presence of HK in compartments lacking GAPDH suggests that HK might be supplying glucose 6-phosphate (Glu-6-P) to a pathway other than glycolysis, such as the PPP, in these regions. In the first step unique to this pathway, glucose-6-phosphate dehydrogenase (G6PDH) produces an alternative type of energy, reducing power in the form of NADPH. Of interest, reducing power produced from glucose metabolism has been implicated in several steps of fertilization including fusion of the sperm with the egg and sperm head decondensation (21)(22)(23).
A prime candidate to regulate glucose metabolism in sperm is HK1-SC. HK1 enzymatic activity is altered during male germ cell development in the rat (24), and during epididymal maturation in the bull (25). Because HK functions to generate Glu-6-P, and this intermediate can be used by both glycolysis and the PPP, HK1-SC is in an ideal position to regulate two pathways of glucose metabolism. Such regulation during sperm capacitation might occur at several levels. For example, enzyme activities might vary absolutely, or shift in subcellular or biochemical localization (e.g. shift from a particulate to a soluble phase). Alternatively, substrate availability might be altered, or pathway inhibition might be regulated by the consumption/removal of end products.
The highly differentiated structure of sperm, the co-localization of metabolic and signaling pathways to specific structures, and the importance of these pathways in sperm function, make these cells a potentially valuable model for studying the crosstalk between metabolic and signaling pathways. In the present report, we have therefore investigated the functional relationships between metabolic and signaling pathways during murine sperm capacitation.

EXPERIMENTAL PROCEDURES
Reagents and Media-All reagents were purchased from Sigma, unless otherwise noted. A polyclonal antiserum against HK type 2 (␣-HK2) was generously provided by Dr. John Wilson (Michigan State University). Anti-phosphotyrosine antibodies (␣-PY, clone 4G10) were purchased from Upstate Biotechnology (Lake Placid, NY), and used at a 1:10,000 dilution for immunoblots. A polyclonal antiserum against the germ cell-specific domain of HK1-SC (␣-qcs) had been generated previously and was used as described by Travis  3) was used for most assays involving mature spermatozoa, except where noted. This medium was designed to support sperm functions yet not interfere with spectrophotometric assays.
Collection and Capacitation of Spermatozoa-Cauda epididymal sperm were collected from retired breeder CD1 mice (Charles River, Wilmington, MA) by a swim-out procedure in 2 ml of ModW at 37°C. Epididymal tissue was removed, and the sperm washed at 100 ϫ g for 1 min in a clinical centrifuge to remove any gross tissue debris. The sperm were resuspended in a final volume of 3-8 ml of ModW, and then centrifuged at 500 ϫ g for 8 min in a round-bottomed tube. The resultant "fluffy" pellet was counted, assessed for motility, and diluted for use. In all cases, large-bore plastic transfer pipettes or large-orifice pipette tips were used to minimize damage to the sperm membranes.
After collection and washing, sperm were incubated under noncapacitating (ModW medium alone) or capacitating conditions (ModW with 10 mM NaHCO 3 , and either 3 mM 2-hydroxypropyl-␤-cyclodextrin (2-OH-␤-CD) or 1.0 mM dibutyryl-cAMP and 0.1 mM IBMX) for 1 h in a 37°C water bath at a final concentration of 2 ϫ 10 6 sperm/300 l. In place of bovine serum albumin, 2-OH-␤-CD was used as a cholesterol acceptor so as to use a completely defined medium (27). Depending upon the experiment, the sperm were incubated with varying concentrations of glucose, lactate, pyruvate, and 2-deoxyglucose (2-DG) as metabolic substrates. Uncouplers of oxidative respiration and inhibitors of electron transfer were used at or above concentrations used in previously published experiments with spermatozoa (28 -30). For experiments in which protein tyrosine phosphorylation was assessed, sperm were concentrated by centrifugation at 10,000 ϫ g and then washed in 1 ml of ModW containing 0.2 mM Na 3 VO 4 to inhibit phosphatase activity during extraction.
Electrophoresis and Immunoblotting-Protein extracts from spermatozoa were prepared by boiling the cells in sample buffer with ␤-mercaptoethanol (31), and then centrifuging at 10,000 ϫ g to yield a supernatant fraction. Cardiac tissue was obtained from male CD1 mice, minced with dissecting scissors, and homogenized in a Teflon/glass homogenizer 20 times on ice. Sample buffer was added to the extract, which was sonicated on ice prior to boiling and centrifugation to yield a soluble fraction. Proteins were separated under reducing conditions by SDS-polyacrylamide gel electrophoresis using 10% polyacrylamide gels (31), and transferred to Immobilon-P membranes (Millipore, Bedford, MA). Blocking, immunoblotting, and detection of the immunoreactive proteins by chemiluminescence (Renaissance, PerkinElmer Life Sciences, Boston, MA) were performed as described in Travis et al. (4).
Spectrophotometric Analysis of HK1-SC and G6PDH Activity-Spermatozoa were incubated for 1 h at 37°C under noncapacitating or capacitating conditions as described above. After incubation, they were split into 4 treatment groups for different methods of solubilization or permeabilization, with each sample ultimately being resuspended in a final volume of 800 l of either ModW or ModW ϩ 10 mM NaHCO 3 depending upon the capacitation status. Pellet and supernatant fractions were obtained for each treatment group.
The first group was solubilized with 0.5% Triton X-100 in ModW (Ϯ NaHCO 3 ) for 15 min at room temperature with occasional hand vortexing. The sperm were centrifuged for 2 min at 5000 ϫ g, and the supernatant collected. The pellet was resuspended and centrifuged again to remove any trace amounts of supernatant, and then resuspended. The second group was referred to as "intact" sperm, because they were merely washed by centrifugation and split into supernatant and pellet fractions. These sperm were not truly intact, because centrifugation has been shown to induce sublethal membrane damage in human (32), and murine 2 sperm. In the third group, sperm were concentrated by centrifugation at 3000 ϫ g for 1 min and then resuspended in 790 l of reagent-grade water (Milli-Q, Millipore water purification system). These "hypotonically-treated" sperm were incubated at room temperature for 15 min with occasional hand vortexing, prior to separation into pellet and supernatant fractions. The last group was permeabilized with streptolysin-O (SLO) (33) (Murex Diagnostics Limited, Dartford, United Kingdom) in ModW (Ϯ NaHCO 3 ) at a final concentration of 0.6 IU/10 6 sperm. These "SLO-treated" sperm were incubated at 37°C for 15 min. Again, supernatant and pellet fractions were obtained by centrifugation.
After treatment, all of the resuspended pellet and supernatant fractions were loaded into individual cuvettes for spectrophotometric determination of hexokinase (HK) activity. HK activity was measured by means of a coupled enzyme reaction pathway that led to the reduction of NADP ϩ to NADPH. The rate of this reduction was measured as a change in absorbance at 365 and 395 nm using an American Instrument Co. (AMINCO) dual wavelength spectrophotometer (Travenol Laboratories, Inc., Silver Springs, MD). The coupled reactions were as follows: glucose ϩ ATP 3 Glu-6-P ϩ ADP and Glu-6-P ϩ NADP ϩ 3 NADPH ϩ 6-phospho-glucono-␦-lactone, with the first reaction catalyzed by endogenous HK, and the second catalyzed by exogenous G6PDH. These reactions were carried out using an ATP-regenerating system of creatine phosphate (CP) and creatine phosphokinase (CPK). Exogenous reagents (with accompanying final concentrations in parentheses) were added in the following order: glucose (5 mM), CP (20 mM), CPK (20 IU/ml), NADP ϩ (0.1 mM), G6PDH (7 IU/ml), and ATP (1 mM). Measurements of activity were determined from slopes taken from within the linear range, and were recorded using two different gain settings to provide optimal resolution. After recording enzyme activity, the samples not treated with Triton X-100 had this detergent added to a final concentration of 0.5%, to effect complete solubilization of the HK and obtain a maximal reading from each sample. This procedure was used as an internal control for equal loading of extracts into the cuvettes. In addition, this step demonstrated whether identical amounts of HK were either solubilized or retained in the particulate fraction under noncapacitating and capacitating conditions. The K m and V max with respect to glucose for the germ cell-specific HK1-SC were deduced by quantifying HK activity at various concentrations of glucose, and fitting the data to a hyperbolic function.
Measurements of endogenous G6PDH activity were performed in a manner similar to the above methods, except that only a single reaction, not a coupled reaction pathway, was utilized. Exogenous NADP ϩ and varying concentrations of Glu-6-P were added to the sperm extracts (Triton X-100 treatment was utilized to provide maximal solubilization) and reduction of the NADP ϩ was quantified. Unlike earlier measurements of HK activity, additional glucose, CP, and CPK were excluded to reduce de novo generation of Glu-6-P within the cuvette. The K m and V max with respect to Glu-6-P for G6PDH were deduced by quantifying G6PDH activity at various concentrations of Glu-6-P, and fitting the data to a hyperbolic function.

Glucose Was Critical for Protein Tyrosine Phosphorylation
Events Associated with Capacitation-Despite an extensive literature on sperm metabolism, the basic question of which metabolic substrates are required to support the maturational process of capacitation remains a matter of controversy. Species-specific differences have further complicated the issue, but even within the murine model system there is some disagreement over the role of glucose. To determine potential roles for glucose in supporting capacitation, we first determined whether glucose was required for protein tyrosine phosphorylation events associated with capacitation. Epididymal sperm were washed free of glucose and then incubated under conditions that support capacitation, with the inclusion of glucose, lactate, and pyruvate, alone or in combination (Fig. 1). Using these substrates at concentrations typically present in media supporting capacitation, only glucose was found to be both necessary and sufficient to support the full pattern of protein tyrosine phosphorylation associated with capacitation (34,35).
We next investigated the concentration of glucose required to support this signaling pathway by incubating sperm with varying concentrations of glucose (Fig. 2). Two different methods of capacitation were employed to test whether glucose was acting either upstream in the signaling pathway at the level of cholesterol efflux (incubation with 2-OH-␤-CD) (27), or downstream at the level of the second messenger cAMP (incubation with the cell-permeable dibutyryl-cAMP and IBMX) (35). Regardless of the method of capacitation, a glucose concentration between 10 and 100 M was sufficient to support the same degree of protein tyrosine phosphorylation as that seen with 5 mM glucose in ModW. Similar findings were observed using sperm from both CD-1 and B6SJLF1/J males, showing that the phenomenon was not strain-specific (data not shown).
These results suggested an absolute requirement for glucose in supporting the protein tyrosine phosphorylation signaling pathway associated with capacitation. Yet they did not indicate whether the effect of glucose was by a direct action on a signaling molecule, or whether the effect was mediated by some product of glucose metabolism. To resolve this distinction, sperm were incubated under capacitating conditions in the absence of glucose, but in the presence of various concentrations of a nonmetabolizable glucose analog, 2-DG (Fig. 3). This analog is phosphorylated by HK, but will not enter glycolysis or the PPP. If 2-DG stimulated tyrosine phosphorylation to the same degree as glucose, then the effect on signaling would result from cross-talk between the signaling pathway leading to protein tyrosine phosphorylation and 1) the transporters bringing 2-DG into the sperm, 2) HK which phosphorylates the 2-DG, or 3) the 2-DG itself. Any of these possibilities would suggest that glucose might directly regulate signaling leading to protein tyrosine phosphorylation. If the 2-DG did not stimulate the increase in protein tyrosine phosphorylation, then the Proteins were then extracted, separated by electrophoresis, transferred to Immobilon P membranes, and immunoblotted with anti-phosphotyrosine (␣-PY). A control ("Ϫ"), with no metabolic substrate, revealed a minimal pattern of protein tyrosine phosphorylation. The one phosphorylated band visualized in this lane represented HK1-SC, which is constitutively phosphorylated on these residues in mature sperm (44), and can thereby function as a loading control. All immunoblots presented in this paper were performed with sperm extracts made from CD-1 males, and are representative examples of experiments that were performed a minimum of three times with similar results. effect of glucose on signaling would most likely result from some product or intermediate of glucose metabolism.
Spermatozoa incubated with increasing concentrations of 2-DG did not show an increase in protein tyrosine phosphorylation (Fig. 3), suggesting that the effect of glucose on protein tyrosine phosphorylation in murine sperm probably resulted from a product of glucose metabolism. Moreover, with concentrations of 2-DG Ͼ10 M, a decrease in protein tyrosine phosphorylation was observed, even below that seen in sperm incubated in the absence of 2-DG. This finding suggested that 2-DG could function as a "phosphorylation sink," and that the phosphorylation events associated with capacitation were not terminal. If these events had been terminal, there would not have been a decrease in protein tyrosine phosphorylation below that seen in the uncapacitated state. Thus the protein substrates appeared to be alternately phosphorylated and dephosphorylated during capacitation.
The failure of lactate and/or pyruvate to support signaling events associated with capacitation suggested a minimal role for ATP produced by the oxidation of these compounds in this cascade. However, these experiments did not completely rule out input from oxidative respiration, as ATP may also be derived from other endogenous sources such as fatty acid oxidation. To eliminate any ATP production from the mitochondria, sperm were incubated with uncouplers of oxidative phosphorylation and inhibitors of electron transfer, alone and in combination (Fig. 4). Regardless of the presence of these compounds, the pattern of protein tyrosine phosphorylation was not decreased so long as glucose was present in the medium. These experiments were also performed using a ModW medium lacking lactate and pyruvate (with 2.4 mM CaCl 2 added to compensate for the removal of the hemi-calcium salt). Results were identical to those shown in Fig. 4 (data not shown).
Quantitation of Sperm HK and G6PDH Activities during Capacitation-These data confirmed a critical role for ATP derived from glycolysis as opposed to oxidative respiration to support signaling events associated with capacitation. A prime candidate for regulating glucose metabolism in sperm is HK, because of its upstream position in glycolysis and the PPP. To investigate whether HK activity was regulated during capacitation in murine sperm, the first step was to ensure that only one type of HK was present. The germ cell-specific HK found in murine spermatozoa has been identified as HK1-SC, and has been localized to the FS of the flagellum, as well as the mitochondria and the membranes of the sperm head (4). No evidence for the somatic HK1 has been found in these cells (4). However, other HK family members also exist, and the first studies to document the existence of a germ cell-specific HK suggested that HK2 might be found associated with male reproductive tissues (36). To rule out the possibility that HK2 might also be found in spermatozoa, thereby complicating the interpretation of data, sperm proteins were probed with an antiserum against HK2 (Fig. 5). No evidence for HK2 was found in mature murine spermatozoa, although a positive control (cardiac tissue) demonstrated the efficacy of the antiserum.
HK activity during capacitation could be regulated in two ways. The first would be an absolute increase in overall activity, and the second would be a biochemical shift in activity, for example, from an insoluble form to a soluble form, as occurs in somatic cells as HK can be bound to mitochondrial porins or may be found in the cytosol. To investigate these possibilities, we incubated sperm under noncapacitating or capacitating conditions and then subjected them to a variety of treatments resulting in different degrees of permeabilization or solubilization (Fig. 6). In no test group was overall HK activity altered during the process of capacitation. Moreover, HK was not seen to shift in regard to its solubility characteristics with capacitation. These findings were true whether sperm were capacitated with 2-OH-␤-CD, or with dibutyryl-cAMP and IBMX (Fig. 6).
Although HK enzymatic activity appeared to be equal in noncapacitated and capacitated sperm, these activities might represent the action of different amounts of protein in the supernatant and pellet fractions. For example, half the amount of an enzyme that was twice as active would give the same overall activity as twice the amount of enzyme that was half as active. To rule out this possibility, amounts of HK protein in the supernatant and pellet fractions from the different treatment groups were quantified by immunoblotting and found not to vary in three replicate experiments (data not shown). In addition, treatment groups that were not subjected to solubilization with Triton X-100 had this detergent added after initial measurements were recorded, and then additional measurements of HK activity were taken. Triton X-100 provided maximal solubilization and revealed approximately equivalent amounts of HK activity in the appropriate fractions (data not shown). These data suggested that HK activity neither was regulated with capacitation, nor underwent a shift in solubility characteristics.
Finally, the K m with respect to glucose and V max for HK1-SC were determined by spectrophotometric analysis at 24°C utilizing supernatant fractions from sperm treated with Triton X-100. The K m was equal to 35.4 Ϯ 5.6 M (S.D.) (n ϭ 6) and the V max was equal to 3.93 Ϯ 0.54 nmol/min/10 6 sperm (S.D.) (n ϭ 6). Our measurement of the K m is similar to the estimation made by Katzen (36) of 100 M for HK1 in spermatozoa, and is very close to his quantification of 30 M for the K m of brain HK1. The high affinity and rate of HK activity in sperm underscores the physiological importance of this enzyme in the function of these cells.
Although exogenous G6PDH was added in excess in the previous experiments as part of the coupled reaction pathways, endogenous G6PDH represented the first enzyme specific to the PPP in the sperm. This alternative pathway of glucose metabolism has been suggested to play a role in murine fertilization by generating NADPH. The enzymatic activity of endogenous G6PDH with respect to Glu-6-P was assessed spectrophotometrically at 24°C utilizing supernatant fractions from noncapacitated sperm treated with Triton X-100. The K m was equal to 5.4 Ϯ 4.8 M (S.D.) (n ϭ 3) and the V max was equal to 0.069 Ϯ 0.011 nmol/min/10 6 sperm (S.D.) (n ϭ 3). These values were obtained from noncapacitated sperm that were incubated and solubilized with Triton X-100 in the absence of glucose. Analysis of G6PDH in sperm incubated under capacitating conditions did not reveal remarkable differences versus noncapacitated sperm. 3 However, strict values for G6PDH activity in capacitated sperm are problematic because capacitation requires the presence of glucose in the medium. Determination of exact kinetic values for G6PDH was impossible given the unknown intracellular concentration of Glu-6-P in the sperm, which resulted from their incubation condition.

ATP Produced by Glycolysis Was Required for Signaling
Events during Murine Sperm Capacitation-We have demonstrated that of the three metabolic substrates (glucose, lactate, and pyruvate) commonly included in media designed to support capacitation, only glucose was necessary and sufficient to support the protein tyrosine phosphorylation signaling events associated with capacitation. Moreover, glucose concentrations as low as 10 -100 M were able to support the same level of tyrosine phosphorylation as higher concentrations of this sugar. Glucose could exert this effect on the signaling pathways associated with capacitation in one of two ways. The first would be a direct effect on a signaling molecule such as an adenylyl cyclase or a cyclic nucleotide phosphodiesterase, whereas an alternative possibility would be an indirect effect through some intermediate or end product of glucose metabolism. Regardless, the fact that the response to glucose was identical whether capacitation was induced at the level of cholesterol efflux, or at the level of cAMP, suggests that the effect of glucose was downstream to the rise in cAMP associated with capacitation.
Recently, Redkar and Olds-Clarke (37) suggested that in the absence of glucose, the binding of sperm to the plasma membrane of zona pellucida-free eggs was significantly retarded, but still occurred. In contrast to previous reports suggesting the importance of glucose in sperm function (14 -16), these authors inferred that glucose was not required for stages of capacitation leading up to and including acrosomal exocytosis (37). One methodological difference that might explain the disparity between the results is that in the present study sperm were washed free of glucose, whereas Redkar and Olds-Clarke (37) utilized a swim-out method of sperm collection without any washes. While presumably designed to avoid potential sublethal membrane damage induced by centrifugation (32), their methodology did not ensure a glucose-free condition; based on their methods the medium may have actually contained glucose at concentrations Ն10 M. In light of our findings, even a concentration as low as this might have a significant impact on signaling events associated with capacitation. In the present study, care was taken to minimize membrane damage by the use of large-bore transfer pipettes, low-speed centrifugations in swinging-bucket rotors, and round-bottomed tubes. The ultimate assessment of membrane damage in sperm is testing fertilizing ability, and this end point was not affected by these methods. 4 Incubation of sperm with the nonmetabolizable glucose analog, 2-DG, did not support the pattern of protein tyrosine phosphorylation observed with glucose. While it could be argued that 2-DG might not elicit an identical signaling response as glucose, this possibility is unlikely given the fact that this analog is brought into the cell through the same transporters and with the same affinity as glucose (38). Rather, these data suggest that the role of glucose in supporting protein tyrosine phosphorylation was to produce a specific metabolite, or an end product of metabolism such as ATP or NADPH. The phosphorylation sink effect noted with higher concentrations of 2-DG also speaks toward the importance of glucose metabolism in supporting these dynamic post-translational modifications.
Intuitively, the most straightforward connection between glucose metabolism and phosphorylation events is the production of ATP and its subsequent utilization by protein kinases.
To confirm that in sperm, ATP produced by glycolysis is not equivalent to ATP produced by oxidative respiration, we incubated sperm with uncouplers of oxidative respiration. These compounds allow respiration to occur, but do not allow the concomitant production of ATP (26,29). Whether sperm were incubated with uncouplers of oxidative respiration or inhibitors of electron transfer, we observed no significant decreases in patterns of protein tyrosine phosphorylation, suggesting that glycolysis provides the ATP for these signaling events.
Interestingly, spermatozoa remained motile in the presence of uncouplers of oxidative respiration, but had significant decreases in percent motility when uncouplers and electron transfer inhibitors were added together. 5 Although surprising, there is precedence in canine sperm for the maintenance of motility in the presence of uncouplers of respiration (1). These observations suggest in sperm that, unlike uncouplers which enhance flux through glycolysis, inhibitors of electron transfer feedback negatively on glycolysis, perhaps through the build-up of intracellular lactate. In addition, it has been suggested that inhibitors such as antimycin A might have effects on sperm motility distinct from inhibiting electron transfer (1). Although the spermatozoa remained visibly motile in the presence of the uncouplers 1799 and pentachlorophenol, it remains unclear whether the mitochondria have some input into the pattern or frequency of flagellar motility. Differentiating the input from the various compartmentalized metabolic pathways in sperm motility will form the basis of future investigations.
The phosphorylation of individual proteins on tyrosine residues has been correlated with the onset of hyperactivated FIG. 6. Hexokinase enzyme activity in noncapacitated and capacitated sperm prepared under a variety of extraction conditions. HK enzyme activity was not observed to vary in a statistically significant fashion whether sperm were incubated under noncapacitating conditions (white bars), in the presence of 3 mM 2-OH-␤-CD and 10 mM NaHCO 3 (diagonally hatched bars), or 1 mM dibutyryl-cAMP, 0.1 mM IBMX, and 10 mM NaHCO 3 (black bars). This result was true for both supernatant (S) and particulate (P) fractions, whether the sperm were extracted with Triton X-100 (TX-100), washed with ModW (intact), incubated in a hypotonic medium (hypotonic), or permeabilized with SLO (SLO). Note that the extremely low values for the hypotonic supernatants resulted as both an artifact of their preparation (they were centrifuged to remove the original medium, thereby removing HK as revealed in the "intact" supernatants) as well as the resistance of murine sperm to hypotonic treatment (45). One-way analysis of variance was employed to evaluate differences between values, with p Ͻ 0.05 used as the criterion for significance. Error bars denote S.E. of the mean. The JMP (version 3) statistical package (JMP Statistical Visualization Software, SAS Institute, Inc., Carey, NC) was used for these analyses. motility in hamster sperm (39). The link between signaling events and changes in motility is further strengthened by the fact that most all of the tyrosine-phosphorylated proteins identified thus far localize to the sperm flagellum (39,40). Taken together, these findings support the hypothesis offered by Travis et al. (4), that a glycolytic pathway organized in the FS would be positioned not only to supply ATP to the flagellar ATPases, but also to supply ATP to the protein kinases believed to be involved in the regulation of capacitation and motility. These results suggest a functional relationship between the glycolytic and signaling pathways co-localized to the FS.
Hexokinase Activity Was Unchanged during Murine Sperm Capacitation-Given the importance of glycolysis in sperm function, and the need to regulate changes in energy production as sperm mature, it becomes important to investigate potential regulators of sperm glucose metabolism. HK generates Glu-6-P from glucose, supplying this substrate to both glycolysis and the PPP. Because both of these pathways have been suggested to be of importance in sperm function, and because HK activity is regulated at different stages of male germ cell development (24,25), we investigated whether HK activity was regulated with capacitation. A potential problem with several earlier studies on enzymatic activities in sperm is that these activities were only tested in soluble extracts or homogenates (41,42). Given the cytoskeletal attachment of the glycolytic enzymes and the close apposition of the overlying plasma membrane, this methodology might produce nonphysiologic results as the enzymes are removed from the influence of potential regulators. Therefore, we quantified HK activity in noncapacitated and capacitated sperm that were permeabilized or solubilized to varying degrees.
Despite these precautions, no change in HK activity or solubility was noted with capacitation. The value for the K m of HK1-SC of 30 M was similar to the estimates made by Katzen (36) for both the germ cell-specific and brain HK1. The HK activity detected was high, and would provide rapid phosphorylation of glucose as it was taken into the cell. This reaction prevents loss of the Glu-6-P from the cell by means of the negative charge on the sugar phosphate group.
The V max for G6PDH, the first enzyme specific to the PPP, suggests an important role for this enzyme in murine sperm. Standardization of units to 10 6 cells allows a direct comparison of V max for G6PDH in human and murine sperm. The V max for G6PDH in human spermatozoa, 0.012 Ϯ 0.002 nmol/min/10 6 cells (43), is several times lower than our calculation of V max for G6PDH in murine sperm, 0.069 Ϯ 0.011 nmol/min/10 6 cells. Most likely, this difference resulted from increased activity or increased protein abundance in the mouse. If the G6PDH is functioning in the sperm head, this difference in activity might reflect the fact that the murine sperm head is larger than the human sperm head. Regardless, these values corroborate the suggested important role for this alternative pathway of glucose metabolism in murine sperm function (21)(22)(23).
The highly polarized structure and function of mammalian sperm dictates that these cells compartmentalize specific metabolic and signaling pathways to regions where they are needed. These pathways must be integrated to support normal cellular function. This inter-relationship underscores the fact that in these cells, metabolic pathways often regarded as "housekeeping" in nature cannot be dissociated from very specific changes in maturation and function. We propose to extend the use of spermatozoa as a model system not only to study compartmentalized metabolic pathways, but also to study the relationship between signaling and metabolic pathways.