Adenylate kinase activity in ejaculated bovine sperm flagella.

Adenylate kinase (ATP:AMP phosphotransferase, EC 2.7.4.3) activity was detected in the flagella of ejaculated bovine spermatozoa. This activity provided sufficient ATP to produce normal motility in cells permeabilized with digitonin and treated with 0.5 mM MgADP. In the presence of ADP, adenylate kinase activity was inhibited by P1,P5-di(adenosine 5')-pentaphosphate (Ap5A), an adenylate kinase-specific inhibitor, and motility was stopped. ATP-supported motility was not affected by Ap5A. Mitochondrial adenylate kinase activity allowed AMP to stimulate respiration in permeabilized sperm. Adenylate kinase activity in tail fragments was most active in a pH range from 7.6 to 8.4, and a similar pH sensitivity was observed for this enzyme activity in a hypotonic extract of whole sperm. The apparent km of adenylate kinase activity in permeabilized tail fragments was about 1.0 mM ADP in the direction of ATP synthesis. The fluctuation of nucleotide concentrations in normal and metabolically stimulated sperm suggested that adenylate kinase was most active when the cell was highly motile, although adenylate kinase activity did not appear to be coupled strictly with motility.

phosphate diffuses down the tail where ATP is formed by another creatine kinase isozyme, presumably in close proximity to dynein ATPases. No such system has been described for bovine sperm, in which ATP apparently diffuses down the tail in quantities sufficient to provide energy for motility as well as ion pumps and other ATP-utilizing systems. Unlike sea urchin sperm, which exhibit full motility when introduced into sea water and continue at that rate until their energy supply is exhausted, mammalian sperm motility changes pattern and speed after capacitation and the acrosome reaction (4). In these cells, respiration is tightly coupled to motility (5); however, because mammalian sperm lack a system for transporting ATP down the tail in response to increased * This work was supported by Grant AM 10334 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. demand, changes in the speed of motility appear to be limited by the efficiency of diffusion.
We report here the presence of adenylate kinase (ATP:AMP phosphotransferase, EC 2.7.4.3) in the bull sperm tail. When operating in the direction of ATP synthesis, adenylate kinase would allow the sperm to exploit the energy of two high energy phosphate bonds in each molecule of ATP.
This capacity would double the energy available for use by the dynein ATPase without necessitating increased ATP diffusion.

MATERIALS AND METHODS
Preparation of Bovine Sperm-Bovine ejaculates were provided by American Breeders Service (De Forest, WI). Semen was centrifuged at 300 X g for 15 min, the seminal plasma was discarded, and the sperm were washed twice by centrifugation at 300 X g for 15 min in a medium containing 120 mM NaCl, 10 mM KCI, 1 mM MgCI2, and 10 mM MOPS, ' pH 7.4 (NKM). The pelleted sperm were suspended at a concentration of 1-2 X lo9 celIs/mI in NKM and kept at 20 "C until use.
Preparation of Tail Fragments-Washed sperm suspended in 3 ml of NKM at a concentration of 5 X 10' cells/ml were disrupted using a Polytron (Brinkmann Instruments) at a power setting of nine for three five-s bursts separated by 1 min on ice. One-ml aliquots of disrupted sperm were centrifuged for 2 s at 15,600 X g in an Eppendorf centrifuge (model 5414, Brinkmann Instruments), the pellet was resuspended in 0.75 ml of NKM, and the centrifugation was repeated. The supernatant fractions were layered on 2 ml of NKM-buffered Percoll (1.065 g/ml) and centrifuged at 30,000 X g in a fixed angle rotor for 15 min. A band at the buffer-Percoll interface, containing tail fragments with very little contamination by midpieces, heads, or whole sperm, was suspended in NKM and centrifuged at 30,000 X g for 5 min, after which the pellet was suspended in minimal NKM and stored at -70 "C until use.
Hypotonic Extraction of Whole Sperm-Washed sperm were suspended at 5 X IO8 cells/ml in a hypotonic medium containing 2 mM MgS04, 0.2 mM EDTA, and 10 mM MOPS, pH 7.4, for 1 min at 4 "C and then centrifuged at 30,000 X g for 15 min. The supernatant fraction was collected and kept on ice until use.
ATP Assay-Washed sperm were prepared as in the permeabilization and reactivation procedures using MgADP, except that they were incubated at 30 'C. After 2 min, cold trichloroacetic acid was added (10% final concentration). The acid extract was collected by centrifugation, and a mixture of freon and trioctyIamine ( 4 1 ) was added (solvent:extract, 2:1), agitated vigorously for 15 s, and centrifuged at 15,600 X g for 15 s. The organic layer was removed, and the neutralized extract was used immediately or stored at -70 "C.
These conditions were also appropriate for continuous monitoring of ATP production by sperm tail fractions or whole sperm (7). Tail fraction was added to an assay mixture, and then 10-22.5 pM ADP was added. Light emission was recorded on a chart recorder, and ATP content was determined by comparison with ATP standards. Digitonin, which was used in some assays, had no effect on the luciferase assay.
When ATP, ADP, and AMP were analyzed in the same extract, all three nucleotides were measured in coupled enzymatic assays. ATP was measured by change of absorbance at 340 nm in an assay mixture containing 50 mM Tris, pH 8.0, 5.25 mM MgC12, 0.22 mg/ml NAD, 8 units/ml glucose-6-phosphate dehydrogenase, 37.5 mM glucose, and 7 units/ml hexokinase. ADP and AMP were analyzed in a mixture containing 200 mM Tris, pH 7.6, 11.0 mM MgSO4, 94.0 mM KCl, 0.65 mM EDTA, 0.043 mM ATP, 0.22 mg/ml NADH, 12.2 mM phosphoenolpyruvate, and 16.5 units/ml lactate dehydrogenase. ADP titers were measured after incubation with 4 units/ml pyruvate kinase, and AMP was subsequently measured following incubation with 3.6 units/ml myokinase.
Adenylate Kinase and Nucleoside Diphsphkinase Assays-Adenylate kinase activity in sperm sonicate fractions was assayed in a medium containing 70 mM glycylglycine, pH 8.0, 10 mM glucose, 5 mM MgCIZ, 3 mM ADP, 1 mM NADP, 5 units/ml hexokinase, and 1 unit/ml glucose-6-phosphate dehydrogenase. Nucleoside diphosphokinase activity was monitored in the same medium with the addition of 3 mM dCTP to act as a substrate and 3 mM AMP to inhibit adenylate kinase. Enzyme activity was correlated with absorbance change at 340 nm.
Sperm Respiration-Oxygen uptake was analyzed polarographically in a Gilson Oxygraph (Gilson Medical Electronics, Middleton, WI) fitted with a Clarke-type oxygen electrode.  Table I). With the addition of CAMP, both the number of motile cells and the quality of motility increased (Table I). The pattern of tail movement became rhythmic, arc amplitudes increased, and the cells generally appeared to move with increased vigor. Because, under these conditions, nearly all of the cells were stuck by their heads to the glass slide or cover slip, speed and directional patterns were not examined.

Reactivation of Motility in Ejaculated Bovine
In the presence of CAMP, ADP stimulated motility that was qualitatively and quantitatively indistinguishable from that produced by MgATP (Table I). Although ADP preparations usually contain some contaminating ATP, this was less than 1.0% of the ADP concentration in these studies, far below the minimum necessary to reactivate full motility (0.2-0.3 mM, data not shown). Mammalian spermatozoa contain enzymes that could transform ADP into the ATP necessary for dynein activity, most likely adenylate kinase and nucleoside diphosphokinase (8). In order for nucleoside diphosphokinase to provide ATP from added ADP, another nucleoside triphosphate must donate a terminal phosphate. Inosine 5'triphosphate, which was included in these preparations to act as a substrate for nonspecific phosphatases, could act as a phosphate donor. However, exclusion of ITP did not affect the stimulatory action of ADP (Table I). Another phosphate donor might be the structural guanine nucleotides bound to the axoneme (8). This possibility has not been investigated. No other added nucleotides, including GTP, GDP, GMP, AMP, or IDP, supported motility in permeabilized sperm.
Adenylate kinase is presumed to salvage adenine nucleotides by converting AMP to ADP at the expense of ATP. The resulting diphosphate is then rephosphorylated by the mitochondrial F1-ATPase. However, the equilibrium constant of adenylate kinase approaches unity, indicating that the direction of catalyzed reaction is determined by substrate concentrations. Thus, ATP could easily be produced from excess ADP. Adenylate kinase is inhibited by ApsA (9), which is considered specific because it does not inhibit nucleoside diphosphokinase (10) or other phosphotransferases. Adding low concentrations of ApA to permeabilized sperm had no effect on sperm reactivated by MgATP but inhibited motility in sperm reactivated by MgADP (Table I). In some trials, sperm exposed to ADP and ApsA recovered some motility after several minutes, albeit of lower quality (i.e. jerky, small amplitude strokes). This recovery of motility by Ap6P-treated cells may indicate that the inhibitor was degraded by sperm phosphatases. Production of ATP by Sperm Taik Using ADP as a Substrate-A sperm tail fraction, practically devoid of contaminating mitochondria, produced ATP when incubated with ADP (Table 11). ATP synthesis was not affected by fluoride, oligomycin, or vanadate but was inhibited by ApA. This tail fraction appeared to retain at least partial membrane integrity because the addition of digitonin increased the rate of ATP production (Fig. lA). The ATP-sparing function of ITP is illustrated in Fig. lA, indicating increased accumulation of ATP in the presence of ITP. ITP is not a substrate for the luciferin-luciferase reaction, and this nucleotide preparation contained a negligible amount of contaminating ATP (about 1%). In addition, ATP was not synthesized from IDP or GDP Indistinguishable from control Sperm were added to medium at 20 "C containing digitonin and, where indicated, cAMP and Ap,A. After 1.5 min, ATP or ADP was added, incubated 1 min, and then motility was videorecorded for approximately 4 min.
Motility was analyzed by reviewing the characteristics of 400-600 cells. Motility of sperm with ATP + CAMP was used as 100% (actual values were 68.5 f 9.9%, mean f S.D. for four trials) and as control for qualitative comparisons.
'ATP, 0.5 mM; ADP, 0.5 mM; CAMP, 5 pM; ApsA, 50 p~. (not shown). ApsA inhibition of adenylate kinase was also observed in a continuous ATP assay (Fig. 1, B and C). The sperm tail adenylate kinase exhibited an increase in activity from pH 6.4 to about 7.6, above which activity was stable (Fig. 2). To verify that the observed pH sensitivity was due to effects on the adenylate kinase itself and not to effects on the remainder of the tail structure, a brief hypotonic extraction was performed on whole sperm, and adenylate kinase activity in the crude extract was analyzed for pH sensitivity. The resulting activity profile was similar to that observed in tail fragments (Fig. 2, inset); however, the tail fragment adenylate kinase activity was sensitive to freezing. Tail fragments stored at -70 "C for several days lost about 70% of their adenylate kinase activity, and the remaining enzyme became unresponsive to changes in pH. No loss of activity was observed in hypotonic fractions with similar treatment (not shown).
Activity of Adenylate Kinase in Intact Sperm-As illustrated in Fig. 3A, the concentrations of adenosine phosphates were constant and roughly equal over 30 min in motile sperm; however, treatment with 2.5 mM caffeine (Fig. 3B), which stimulated motility and generated a 2-fold increase in respiration (ll), drastically changed the adenosine phosphate distribution, decreasing ATP and increasing AMP. This shift in stimulated cells is consistent with the proposal that sperm tail adenylate kinase produces ATP from ADP, thus accumulating AMP. The equilibrium constant for these nucleotides at peak stimulation of sperm metabolism is about 1 (Table 111), which corresponds to the reported &of adenylate kinase. Thus, the measured reaction is not invalidated by contaminating enzymes that might alter the adenine nucleotides. In sperm treated with 30 mM NaF, which blocked motility but did not affect oxidative phosphorylation (111, ATP titers were very high and remained stable for an extended period (Fig. 3C), while ADP and AMP titers were extremely low. Adding oligomycin to fluoride-inhibited sperm (Fig, 3 0 ) caused a precipitous drop in ATP levels and increased titers of both ADP and AMP, suggesting that the sperm adenylate kinase activity was not coupled exclusively to motility but also provided ATP for other metabolic processes.
The sperm tail adenylate kinase had an apparent k, of 1.1 mM for ADP measured in situ (Fig. 4). Using a water volume of 14.3 p3 for ejaculated bull sperm (12) and amounts of adenosine phosphates from Fig. 3A, we calculate the concentration of ADP in untreated and caffeine-stimulated sperm as 7.7 mM and 8.1 mM, respectively.
Stimulation of Respiration by AMP-In washed bull sperm that had been permeabilized by filipin, AMP was as effective as ADP at stimulating State 3 respiration (Table IV) FIG. 1. ATP production by sperm tails using ADP as a substrate. Sperm tail fragments lacking mitochondria incubated with 10 p~ ADP ( A ) in reactivation medium produced ATP measured in a continuous luciferase assay. Digitonin (0.15 mM) and ITP (1.0 mM) were added as indicated. Incubation of tail fragments with ADP, followed by ApsA (50 p~) ( B ) , or incubation of tails with Ap6A (50 NM) for 3 min prior to ADP addition (C) resulted in the inhibition of ATP synthesis. The initial increase in photon production after addition of ADP and Ap& reflects ATP contamination in these compounds. ATP concentrations in ADP preparations are never more than 1%. at the mitochondria. Addition of A p d blocked stimulation by AMP but did not affect stimulation by ADP. To preclude the possibility that tail adenylate kinases had migrated out of permeabilized flagella, filipin-treated sperm were centrifuged and resuspended in new media. These cells retained their responsiveness to AMP, indicating that the adenylate kinase activity was mitochondrial and not flagellar.

DISCUSSION
Adenylate kinase is a ubiquitous enzyme that, by virtue of an equilibrium constant approaching 1, can produce either ADP or stoichiometric amounts of ATP and AMP depending on the concentrations of the three nucleotides. Between the

Sperm Tail Adenylate
Kinase 6089 inner and outer mitochondrial membranes, adenylate kinase probably uses ATP to salvage AMP generated by cellular metabolism (13). However, in all compartments, the direction of adenylate kinase activity is determined by substrate concentrations (13, 14), and where ATP is being used rapidly, the reaction is in the direction of ATP production from ADP. Adenylate kinase activity has been detected in a variety of mM Tris at the indicated pH levels. After 5 min at 30 "C, the reactions were stopped by adding cold perchloric acid, which was subsequently neutralized by adding cold KOH. The neutralized extract obtained after centrifugation was analyzed for ATP content as detailed under "Materials and Methods." Inset, to verify that the observed changes in activity were primarily the result of pH sensitivity of the tail adenylate kinase and not due to sensitivity of other tail structures, an identical experiment was performed on a hypotonic extract of whole sperm.  3,8, 18, 19). The function and location of this enzyme in vivo have not been determined, although Brokaw and Gibbons (18) noted that adenylate kinase activity was recovered with the tail fraction of homogenized sea urchin sperm. The ability of ADP to support motility in Triton-treated cell models was variable (18, 20), but there appeared to be a correlation between the presence of the cell membrane and the ability of ADP to stimulate motility. Adenylate kinase activity has also been implied previously in bull sperm flagella in novel experiments reported by Lindemann and Rikmenspoel (21). ADP added to punctured or dissected bull sperm supported slow, but apparently normal, motility. Although no attempt was made to identify the responsible activity, the authors' conclusion that adenylate kinase was present in bovine sperm has been borne out by the evidence reported here.
In the flagella of bull sperm relying on oxidative phosphorylation, dynein and other ATPases consume ATP that is produced only in the midpiece and must diffuse down the tail to power motility. We suggest that, in the sperm tail, adenylate kinase forms ATP in response to high ADP concentrations that develop when motility is stimulated after capacitation or prior to fertilization (4). AMP formed by the reaction diffuses back to the mitochondria to be rephosphorylated to ADP by mitochondrial adenylate kinase. Because this mitochondrial phosphorylation takes place at the expense of ATP, the net energy expenditure of a system using a tail adenylate kinase would be equal to a conventional system. The advantage of the tail adenylate kinase pathway is that one diffusing nucleotide could supply energy from two high energy phosphate bonds instead of just one. A theoretical treatment by Raff and Blum (26) lends support to this proposal with the conclusion that adenylate kinase could increase the effective length of a cilium relying on ATP diffusion from mitochon-  FIG. 4. Lineweaver-Burk plot of adenylate kinase activity in sperm tail fragments. Sperm tail fragments were incubated at 30 "C in the medium described in Fig. 2 with varying amounts of ADP. After 2.5 min, cold trichloroacetic acid was added, and ATP was extracted and measured in a luciferase assay as under "Materials and Methods." Least squares linear regression analysis of ATP production versus substrate supply indicated an apparent k, of approximately 0.94 mM in the direction of ATP synthesis.
is that adenylate kinase is distributed evenly throughout the tail and provides additional energy for ATP-utilizing systems under all circumstances. Alternatively, adenylate kinase might be localized in a specific region of the tail, such as the distal portion, where diffusion of mitochondrial ATP might be less efficient and the increased energy available from each ATP could compensate for a deficient nucleotide supply. A third possibility is that adenylate kinase is located in close conjunction with dynein ATPase. In this case, which would be loosely analogous to the system of creatine kinase isozymes in sea urchin sperm (3), adenylate kinases located in the vicinity of dynein ATPases would ensure the availability of additional energy at times of increased demand. The possibility also exists that adenylate kinase is a regulated enzyme in the sperm tail and that it participates in the stimulation of sperm motility specifically at capacitation or fertilization by increasing the available energy.
It is obvious that bovine sperm have a large amount of adenylate kinase activity, but separating the activity in the mitochondria (midpiece) and tail is difficult because the midpiece is an integral part of the tail. Several pieces of data presented here do, however, suggest that adenylate kinase is located both in sperm tails and mitochondria. The primary evidence for tail adenylate kinase is that permeabilized sperm become motile in the presence of ADP and CAMP. It is well established that dyneins use ATP exclusively (22, 23), so the

Respiration stimulated by ADP and AMP
Respiration of washed sperm was measured polarographically in an oxygraph (Gilson Medical Electronics, Middleton, WI) fitted with a Clarke-type electrode in NKM containing 10 mM pyruvate, 10 mM malate, 5 mM KHzPO,, and 0.2 mg/ml bovine serum albumin (State 4 respiration with substrate present but no nucleotides) to which was added either 2.5 mM ADP or 2.5 mM AMP (State 3 respiration). Sperm (3 X 10' cells) were treated with filipin (5 p~) , added in the oxygraph chamber (filipin), or added and followed by centrifugation at 300 X g for 5 min (filipin-wash) prior to oxygraph recording. Sperm from different ejaculates exhibited wide variation in both State 3 and State 4 respiration. Trials 1-3 and 4-5 were performed on sperm from separate sets of three pooled ejaculates. Percent of control values were calculated after subtraction of State 4 respiration from State 3 respiration in each trial. ADP-stimulated respiration was used as control in noninhibited experiments. Respiration without inhibitors was used as control in inhibitor studies. Ap6A, 50 p~. added ADP must be converted to ATP before it can be used for motility. In permeabilized sperm treated with 30 mM sodium fluoride, which inhibited sperm motility (11) and blocked nucleotide-consuming phosphatases' but did not inhibit adenylate kinase, added ADP would produce enough ATP to make the medium concentration about 12 PM. If adenylate kinase were located exclusively in the mitochondria, ATP produced there would have to diffuse through the medium to the tail to stimulate motility. However, these studies also indicate that bovine sperm require about 0.25 mM ATP to achieve full motility. Thus, in order for ADP to produce motility, it must be converted to ATP in the tail itself.
In order for our hypothesis to be valid, mitochondrial adenylate kinase must rephosphorylate AMP returned from the tail at the mitochondria. The participation of this mitochondrial enzyme is indicated in experiments in which AMP stimulated respiration in permeabilized sperm. The possibility that tail adenylate kinase may have been released and was redeposited in the vicinity of the mitochondria is lessened by the fact that the AMP effect persisted after the permeabilized cells were washed, which would probably remove any free tail enzyme.
Although theoretical calculations indicate that diffusion of mitochondrial ATP should be sufficient to produce maximal motility in both sea urchin and bull sperm (24, 25), empirical observations of sea urchin motility (3, 27) indicate that ATP diffusion would not support full motility, particularly in the distal portions of the flagellum. Sea urchin sperm rely on a facilitated diffusion of high energy phosphates in the form of phosphocreatine to supply adequate energy to all portions of the tail. Although sea urchin sperm also contain adenylate kinase, its activity is insufficient to support full motility in the absence of the phosphocreatine shuttle (27). It may be postulated that sea urchin sperm retain adenylate kinase activity to salvage AMP at the mitochondria or to reduce ~ P. K. Schoff, unpublished observations.