Mitochondrial metabolism of pyruvate in bovine spermatozoa.

Treatment with the polyene antibiotic, filipin, renders the spermatozoan cell membrane permeable to small molecules, but not to the intracellular enzymes aldolase and lactate dehydrogenase. Pyruvate (10 mM) as the sole substrate was metabolized very slowly. L-Carnitine increased pyruvate metabolism 3- to 4-fold and allowed limited rates of oxidative phosphorylation. When spermatozoa treated with filipin were supplemented with malate, there was a rapid, almost linear rate of pyruvate metabolism which was slightly increased by L-carnitine. In the absence of malate, 20 to 30% of the pyruvate used was reduced to lactate; this increased to 57% in the presence of malate. Without malate, about 90% of the pyruvate metabolized was converted to lactate and acetate or L-acetylcarnitine. Rutamycin or rotenone increased both the rate of pyruvate use and the delta lactate/deltapyruvate ratio. Under all treatments, L-carnitine consistently reduced the percentage of pyruvate converted to lactate by about 10%; part of the pyruvate was preferentially shunted into L-acetylcarnitine rather than lactate. The mitochondrial inhibitors, rotenone or rutamycin, did not change the amount of pyruvate that was converted to metabolites other than lactate, or L-acetylcarnitine, or both. Pyruvate-supported State 3 respiration was linear only if L-carnitine, or malate, or both, were added to the incubation medium. Added malate was necessary to produce a rapid State 3 respiratory rate and was also required for significant respiratory activity in the presence of rotenone or rutamycin. From cells metabolizing [2-14C]pyruvate (1.4 mM), 14C-labeled acid-extractable metabolites were separated by ion exchange column chromatography. All of the [2-14C]pyruvate (+/-5%) used was recovered in 14C-labeled metabolites and 14CO2. In the presence of malate, citrate accumulation was significant, and was always large in comparison to flux through the citric acid cycle. Glutamate, beta-hydroxybutyrate, acetoacetate, fumarate, aspartate, and alpha-ketoglutarate did not accumulate in significant amounts. Some 14C-labeled succinate was produced but only in the presence of malate. Alkaline hydrolysis of a fraction containing carnitine esters yielded acetate and a compound tentatively identified as beta-hydroxybutyrate or lactate. As in intact cells, intramitochondrial lactate dehydrogenase competes successfully with the electron transport system for the NADH generated by pyruvate metabolism. The role of lactate and L-carnitine, and conclusions suggested by the accumulation of certain metabolites are discussed in relation to control of citric acid cycle activity.

with the polyene antibiotic, filipin, renders the spermatozoan cell membrane permeable to small molecules, but not to the intracellular enzymes aldolase and lactate dehydrogenase.
Pyruvate (10 mM) as the sole substrate was metabolized very slowly. L-Carnitine increased pyruvate metabolism 3-to 4-fold and allowed limited rates of oxidative phosphorylation.
When spermatozoa treated with filipin were supplemented with malate, there was a rapid, almost linear rate of pyruvate metabolism which was slighty increased by L-carnitine.
In the absence of malate, 20 to 30% of the pyruvate used was reduced to lactate; this increased to 57% in the presence of malate. Without malate, about 90% of the pyruvate metabolized was converted to lactate and acetate or L-acetylcarnitine.
Rutamycin or rotenone increased both the rate of pyruvate use and the Alactatelbpyruvate ratio. Under all treatments, L-carnitine consistently reduced the percentage of pyruvate converted to lactate by about 10%; part of the pyruvate was preferentially shunted into L-acetylcarnitine rather than lactate. The mitochondrial inhibitors, rotenone or rutamycin, did not change the amount of pyruvate that was converted to metabolites other than lactate, or L-acetylcarnitine, or both. Pyruvate-supported State 3 respiration was linear only if L-carnitine, or malate, or both, were added to the incubation medium. Added malate was necessary to produce a rapid State 3 respiratory rate and was also required for significant respiratory activity in the presence of rotenone or rutamytin.
From cells metabolizing [2-14C]pyruvate (1.4 mM), 14C-labeled acid-extractable metabolites were separated by ion exchange column chromatography. All of the [2-*4C]pyruvate (-~5%) used was recovered in 14C-labeled metabolites and 14C02. In the presence of malate, citrate accumulation was significant, and was always large in comparison to flux through the citric acid cycle. Glutamate, /3hydroxybutyrate, acetoacetate, fumarate, aspartate, and (Yketoglutarate did not accumulate in significant amounts. Some 14C-labeled succinate was produced but only in the presence of malate. Alkaline hydrolysis of a fraction containing carnitine esters yielded acetate and a compound tentatively identified as /?-hydroxybutyrate or lactate. As in intact cells, intramitochondrial lactate dehydrogenase competes successfully with the electron transport system for the NADH generated by pyruvate metabolism.
The role of lactate and L-carnitine, and conclusions suggested by the accumulation of certain metabolites are discussed in relation to control of citric acid cycle activity.
The natural substrates available to spermatozoa include endogenous lipids (l), fructose which is in the seminal plasma (2), and glucose and lactate which are in the fluids of the female reproductive tract. However, pyruvate, a product of aerobic glycolysis, is also found in bull semen at 4.7 mg/lOO ml (2). Seminal plasma from vasectomized bulls contains 17 to 47 mg of pyruvate/lOO ml, indicating that pyruvate rapidly disappears in the presence of viable spermatozoa (3). More recently, it has been shown that pyruvate is necessary for in vitro capacitation of guinea pig sperm (4) and for maintenance of mouse gametes (5).
In ejaculated porcine spermatozoa and bovine epididymal spermatozoa, the stoichiometry of pyruvate metabolism does not fit that of a simple dismutation (6-8). In intact bovine spermatozoa, the intramitochondrial location of lactate dehydrogenase allows pyruvate to oxidize the NADH produced from its own oxidation (7,8). It also allows pyruvate to serve as an energy source in the presence of rotenone, an inhibitor of NADH oxidation, and of rutamycin, an inhibitor of oxidative phosphorylation.
Of several substrates tested, only pyruvate or a glycolyzable sugar were able to restore motility in the presence of these inhibitors.
Spermatozoa contain a high concentration of L-carnitine and high carnitine acetyltransferase activity (9, 10). Milkowski et al. (11) and Van Dop et al. (7,8) have shown that added pyruvate rapidly acetylates the free carnitine pool, even under anaerobic conditions (8). The effects of L-carnitine on pyruvate metabolism in heart and blowfly flight muscle mitochondria have been well documented (12-15).
In view of the possible importance of pyruvate in intact spermatozoan metabolism, we have investigated mitochondrial pyruvate metabolism using the polyene antibiotic, Elipin, to render the spermatozoan cell membrane permeable to small molecules (16). The results confirm the important role of intramitochondrial lactate dehydrogenase postulated for intact cells (8,17). In addition, L-carnitine was found to influence pyruvate metabolism via pyruvate dehydrogenase and lactate dehydrogenase. The distribution of 14C label in acid-  (16). Although fluoride enhances the P/O ratios (181, it was not included in the present experiments because it partially inhibited pyruvate metabolism. Fluoride inhibition of several ATP-requiring processes would allow more ATP to pass from the mitochondria to the trapping system. The deoxyglucose-hexokinase trapping was not a quantitative measure of ATP synthesis in filipin-treated cells (even after two wash steps), because of ATPase activity.
Washed, filipin-treated cells contained about three-fourths of their original cytosolic enzyme, aldolase. Lactate dehydrogenase, which is believed to be both mitochondrial and cytoplasmic in bovine spermatozoa (Refs. 21,22,for review,Ref. 23) was almost completely retained by these cells (93%). Small molecules like L-carnitine and L-acetylcarnitine were lost during the washing procedure. The small amount of L-carnitine remaining (approximately 10%) may be intramitochondrial. In addition, nucleotides and citric acid cycle intermediates were low or undetectable after treatment with filipin and washing.
It is not surprising that State 3 pyruvate metabolism (ADP present) in these cells was limited. Fig. 1A shows plots of pyruvate disappearance, and lactate and acetate accumulation. In the absence of 4-carbon dicarboxylic acids that permit more active metabolism, at least 70 to 80% of the pyruvate used was converted to acetate and lactate.
Significant amounts of acetate were also produced in intact cells when electron transport was blocked with inhibitors or by anaerobiosis (81, but lactate production was much lower than in intact cells. This agrees with the data of Van Dop et al. (7,8) which show that lactate production is an outlet for the mitochondrial NADH produced during pyruvate metabolism.
Effects of L-Carnitine and Malate on Pyruvate Metabolism -GCarnitine increases decarboxylation of [1-'Qpyruvate in rat heart mitochondria (13). In filipin-treated bovine epididymal spermatozoa, L-carnitine addition enhanced the rate of pyruvate metabolism nearly 3-fold (Fig. 1B ). There was a large accumulation of L-acetylcarnitine, but no measurable acetate. A concentration of 50 mM L-carnitme was used, because it approximates the intracellular carnitine concentration (about 90 mM). The intracellular carnitine concentration was estimated using the specific gravity of the droplet, midpiece, and tail portions, the weight of a spermatozoon (2) and the total cellular L-carnitine (8,24). This open system can only approximate intracellular conditions, because metabolites like L-acetylcarnitine can freely diffuse out of the filipin-treated cell. Lactate production was low but the sum of the lactate and L-acetylcarnitine accounted for more than 80% of the total pyruvate metabolized.
Malate addition, which provides a ready source of oxalacetate for citrate synthetase, resulted in a rapid, almost linear rate of pyruvate disappearance (Fig. 10 1 (left). Effects of L-carnitine and malate on pyruvate metabolism. Twice washed tilipin-treated bovine epididymal spermatozoa were incubated in the standard incubation medium with pyruvate (10.0 mM) (see "Materials and Methods"). L-Camitine was 50 mM U3 andl)) and malate was 0.77 mM (C andD). The sperm concentrations were 7 x lo8 (A), 8.3 x lo* (B), and 8.1 x lo* (C andD) cells/ml. The following symbols were used: pyruvate, (0); acetate, (A); lactate, the percentage of pyruvate converted to L-acetylcarnitine was significantly decreased by malate (compare Fig. 1, B and D). Lactate production was dramatically increased in the presence of malate (Fig. 1, C and D).
The effects of L-carnitine and malate on 2-deoxyglucose 6phosphate synthesis are shown in Fig. 2. In filipin-treated cells metabolizing pyruvate alone, little phosphate was incorporated into 2-deoxyglucose unless L-carnitine was also added (Fig. 2, A and B). However, there was a rapid linear rate of synthesis with cells metabolizing pyruvate-malate (Fig. 2C); this rate was not influenced by L-carnitine (data not shown).
Mitochondrial Inhibitors - Fig. 3 illustrates the effects of rutamycin and rotenone on State 3 pyruvate-malate metabolism and lactate accumulation.
The rate of pyruvate use was significantly increased over the corresponding control shown in Fig. 1C. As in untreated bovine epididymal spermatozoa (7, 81, the presence of an inhibitor of oxidative phosphorylation (rutamycin) or a Site I inhibitor of electron transport chain (rotenone) increased the percentage of lactate formed from pyruvate. Both rotenone and rutamycin greatly decreased ATP synthesis, although there was a small net accumulation of 2-deoxyglucose 6-phosphate (data not shown). In cells oxidizing 10 mM malate in the absence of pyruvate, either rutamycin or rotenone eliminated all 2-deoxyglucose 6-phosphate accumulation.
Malate alone, at the concentrations (about 0.8 mM) used in the experiments with pyruvate, did not support the synthesis of significant amounts of 2-deoxyglucose 6-phosphate. L-Carnitine had no effect on malate oxidation or ATP synthesis (data not shown).
The results of the metabolic experiments are summarized in Table I where the pyruvate used and the lactate, acetate, and Effects of L-carnitine and malate on 2-deoxyglucase 6-phosphate synthesis. Experimental conditions were the same as in Fig. 1, A to C. Curve A was pyruvate alone; B, pyruvate plus Lcarnitine; C, pyruvate plus malate. acetylcarnitine produced at 15 min are compared. The dramatic effect of carnitine on the rate of pyruvate use (p < 0.001) is evident. The pyruvate converted to metabolites other than lactate and acetylcarnitine or acetate is quite small. Carnitine appeared to increase pyruvate use even in the presence of malate (p < 0.051, and the difference largely was accounted for as L-acetylcarnitine. In the absence of L-carnitine, rotenone or rutamycin increased the rate of pyruvate use (p < 0.05). The percentage of pyruvate converted to lactate in the presence of inhibitors approached that seen in untreated cells (8). The lactate data support our hypothesis that pyruvate metabolism is intramitochondrial, with the intramitochondrial location of LDH-X' (21-23) allowing pyruvate to oxidize the NADH produced during its metabolism via the citric acid cycle.
In the intact cell, the intracellular carnitine is acetylated within the first 5 min after pyruvate addition (8). In the filipin-treated cells, the rate of acetylation of added free carnitine is linear throughout the 20-min incubation period. In cells oxidizing pyruvate, L-carnitine consistently reduced the percentage of pyruvate converted to lactate by about 10%. It appears that part of the pyruvate was preferentially shunted into L-acetylcarnitine, rather than to lactate. The data also indicate that during State 3, pyruvate dehydrogenase is probably not operating maximally in the absence of L-carnitine. Similarly, in the absence of malate, L-carnitine can activate pyruvate decarboxylation, possibly by freeing coenzyme A (15). This increased flux through pyruvate dehydrogenase produces NADH, which can be used to reduce pyruvate to lactate, or oxidized via the electron transport chain to produce ATP (see Fig. 2B). Without tcarnitine or malate, disposal of acetyl units produced by pyruvate dehydrogenase is limited and free acetate accumulates.

Effects on Respiration
-Data obtained on the oxygraph support the data from the metabolic studies (see Fig. 4). In these experiments, the pyruvate concentration was lowered to 1.4 mM; however, the State 3 respiratory rate (ADP present) was the same as with 10 mM pyruvate. State 3 respiration sup ported by pyruvate was slow and nonlinear without L-carnitine (Fig. 4A). With L-carnitine, the respiratory rate increased 2-to 3-fold and remained linear until anaerobiosis. This agrees with earlier work by Bremer (12) the respiratory rate (3-fold) and this enhanced rate was also linear until anaerobiosis (Fig. 4B); this rate was not affected by ccarnitine.
Although the respiratory rate was low with rutamycin or rotenone, respiration decreased even further within 6 to 10 min. A second addition of pyruvate reinitiated respiration. This is predictable based on the rapid rate of pyruvate disappearance in the presence of rutamycin or rotenone (see Table  I), but the oxygen consumed accounts for only a small portion of the pyruvate used (Fig. 4). It is also somewhat surprising that the addition of rotenone or rutamycin did not influence the amount of pyruvate that was converted to metabolites other than lactate, or acetylcarnitine, or both ( In addition to acetylcarnitine, Peak I contained a minor component (less than 1% of the counts added) that resisted alkaline hydrolysis and rechromatographed on Dowex 1 as Peak I. It ranged in size from 0.4 to 1.8 nmol/lO* cells (calculation based on the specific activity of the added pyruvate). It was most likely alanine (19) and was not characterized further. The major product was usually acetate, derived from L-acetylcarnitine.
A third product of alkaline hydrolysis of Peak I co-chromatographed with lactate; it usually ranged from 1 to 2 nmol and occasionally reached concentrations as high as 7 nmol/lOs cells (3.5% of added disintegrations per min; Table III). This product appeared to be unaffected by borohydride reduction, which ruled out acetoacetate or keto acids. The chromatographic pattern suggested it was lactate or possibly P-hydroxybutyrate, although L-carnitine esters of either of these compounds have not been reported previously in biological systems. Peak II was identified as containing mainly acetate, some glutamate and small amounts of aspartate. Acetate was identified by co-chromatography with YClacetate standard and volatilization. Acetate  4. Effect of various treatments on State 3 respiration of filipin-treated spermatozoa. In A to C, the sperm concentration was 3.2 x lOa cells/ml, pyruvate was 1.5 mM, malate was 0.8 mM, and, when added, L-carnitine was 50 mM. In D, the sperm concentration was 1.9 x 10s cells/ml and malate was 8 mM. The numbers in parentheses are respiratory rates expressed as nanogram atoms of oxygen/4 X lOa cells/ min. Inhibitors were added at the following concentrations: rotenone, 15 ELM; rutamycin, 13 PM; antimycin A, 2 PM. In C, rotenone, rutamycin, or antimycin A was added immediately after the substrates.  (19). The elution rate was 2.0 ml/min and the eluent was collected in 2.0 ml fractions. I, acetylcarnitine (see "Results"); ZZ, acetate, glutamate, aspartate; ZZZ, lactate; IV, acetoacetatc (see "Results"); V, succinate; VI, malate; VII, pyruvate; VIII, citrate, cu-ketoglutarate, fumarate (see "Results").
comprised 50 to 60% of this peak and in the absence of malate or L-carnitine was nearly the sole component. p-Hydroxybutyrate which should have chromatographed in this region was not detected under any treatments tested, including prolonged incubation. Glutamate (II), aspartate (II), lactate (III), malate (VI), pyruvate (VII), citrate (VIII), and a-kemglutarate (VIII) were measured enzymatically.
One hundred per cent of Peak V was identified as succinic acid by recrystallization of the pooled material with carrier succinic acid. Occasionally, a small peak (0.2% of added disintegrations per min) was found near Fraction 40 (Peak IV). It was probably acetoacetate (19) but in such small amounts that it was not detectable by enzymatic analysis of the perchloric acid extracts of any treatment group. Citrate comprised the major portion of Peak VIII. Fumarate was too low to be measured enzymatically in the pooled peak. Similarly, oc-ketoglutarate never exceeded 1 to 2 nmol/lO" cells, the limit of detectability. Marker a-ketoglutar-  Table II. Essentially all of the initial [2-Y!]pyruvate (+5%) used was recovered in "C-labeled metabolites and WO,. The labeling pattern for the major metabolites fits that seen with 10 mM pyruvate. In the absence of malate, the rate of pyruvate metabolism was low and only about 30% of the added pyruvate was metabolized.
In the short period of 4 min, and with the lower pyruvate concentration (Table II), the effect of L-carnitine on pyruvate use was not obvious. Fig. 1A and oxygraph data (Fig. 4) showed that the enhancement by carnitine of pyruvate disappearance and oxidation was most apparent after the first 5 min. Again, lactate production was low while acetate and acetylcarnitine have the reciprocal relationship seen in Fig. 1, A andB.
In agreement with the data of Table I (10 mM pyruvate),  rutamycin or rotenone enhanced the rate of pyruvate disappearance. As in intact cells (81, neither antimycin A nor antimycin A plus rotenone blocked pyruvate use. L-Carnitine did not influence pyruvate disappearance in the presence of inhibitors, but did increase the counts recovered in Peak I (see Table II), partly at the expense of lactate. At least 10 to 20% of the pyruvate used was recovered as L-acetylcarnitine. The results of typical alkaline-hydrolysis patterns (Peak I) for rutamycin or rotenone treatments are shown in Table  III. The small amounts of endogenous free tcarnitine present at the time of substrate addition were acetylated within 4 min. Incorporation of 14C-label into L-acetylcarnitine was significant in the presence of L-carnitine; the unknown (lactate?) fraction was highest in the rotenone-treated samples.
Citrate Production-Citrate accumulation was significant in the presence of malate. In a nonradioactive experiment, the average amount of citrate that accumulated in 4 min (44 nmol/ 10' cells) closely approximated malate used (58 nmol/108 cells).
In the presence of inhibitors, citrate accumulation was 36 nmol versus 42 nmol of malate used. In all cases, citrate accumulation was large in comparison to flux through the citric acid cycle as indicated by the negligible release of 14C0, (Table II).

Pyruvate
Metabolism in Sperm Mitochondria Also, the specific activity of citrate formed at 4 min was essentially the same as that of the added pyruvate. Therefore, a continuous flux of pyruvate through the citric acid cycle did not occur at a significant rate. Citrate accumulation was not influenced by any mitochondrial inhibitors, except possibly rotenone (Table II).
'GLabeZilzg of Mulate Pool -The data in Table II suggest that limited W-label is entering the malate pool. This was measured quantitatively in several experiments. For example, aRer 4 min of incubation, the specific activity of the remaining malate was 7 to 9% of that of the added pyruvate. Even after 20 min, the specific activity of the malate was only one-third of the pyruvate. Malate use was similar under all conditions tested. At zero time neither glutamate, fumarate, nor malate were detectable and aspartate was less than 5 nmol/108 cells. Aspartate never increased above 6 nmol/lOs cells during incubation with pyruvate-malate. DISCUSSION The importance of carnitine in fatty acid oxidation has been recognized since the report of Fritz (26). Later, carnitine was shown to have a significant function in carbohydrate use, specifically pyruvate metabolism, in blowfly flight muscle (15). During flight, pyruvate is the end product of glycolysis, neither ol-glycerophosphate nor lactate accumulate, but a large share of the pyruvate is transaminated to alanine (27). Reports from Casillas' (10, 24) and this laboratory (7, 8, I'l) suggest that carnitine and acetylcarnitine are important in bovine spermatozoan carbohydrate and fatty acid metabolism. Pyruvate metabolism in filipin-treated bovine epididymal spermatozoa is quite different from that in blowfly flight muscle (15) or heart mitochondria (l&28) because of the intramitochondrial location of LDH-X.
The observations made on intact spermatozoa oxidizing pyruvate have been confirmed with filipin-treated cells in which the mitochondria are readily accessible to added metabolites and co-factors. The data indicate that intramitochondrial lactate dehydrogenase competes successfully with the electron transport system for the NADH generated by pyruvate metabolism. Malate addition increased the percentage of pyruvate converted to lactate as a result of increased NADH production by malate dehydrogenase and some increase in pyruvate flux through the citric acid cycle. Increased lactate production was also observed in ejaculated human spermatozoa oxidizing pyruvate plus fumarate or succinate (29). The percentage of pyruvate converted to lactate was greater under aerobic than anaerobic conditions in ejaculated porcine spermatozoa (6).
Addition of rutamycin or rotenone to intact bovine epididy-ma1 spermatozoa resulted in a Alactate/Apyruvate ratio of 0.8 which indicates that the pyruvate not reduced to lactate was completely oxidized via the citric acid cycle and that all of the NADH produced during pyruvate oxidation was used to reduce pyruvate via lactate dehydrogenase (8). In filipin-treated cells, with rotenone present to block NADH oxidation at Site I, the Alactate/Apyruvate ratio would be 0.66 if all the pyruvate used which is not reduced to lactate is converted exclusively to citrate. This is close to the observed value of 0.69. If one NADH is allowed for each L-acetylcarnitine produced, this calculation also holds when L-carnitine is present. This calculation for 10 mM pyruvate is tentative, because the initial malate concentration was not large enough to permit a continuous rate of citrate accumulation. At low pyruvate concentra-