- Email: [email protected]
Testosterone stimulates the biosynthesis of m-aconitase and citrate oxidation in prostate epithelial cells
ELSEVIER
Molecular and Cellular Endocrinology 112 (1995) 45-51
Testosterone stimulates the biosynthesis of m-aconitase and citrate oxidation in prostate epithelial cells Leslie C. Costello*, Yiyan Liu, Renty B. Franklin Department of Physiology, Uniuersity of Maryland Dental School, 664 W Baltimore St., Baltimore, MD
21201, USA
Received 20 April 1995;revision received, accepted 12 May 1995
Abstract Mitochondria (m-Iaconitase is a rate-limiting regulatory enzyme in prostate epithelial cells which minimizes citrate oxidation by these cells. This unique metabolic characteristic is responsible for the ability of the prostate to accumulate and secrete extraordinarily high levels of citrate. Testosterone is a major regulator of prostate growth and function, and stimulates citrate oxidation. Therefore, an important action of testosterone might be its stimulation of m-aconitase in prostate epithelial cells. Studies were conducted with rat ventral prostate (VP) epithelial cells to establish the effect of testosterone on the level of m-aconitase and corresponding citrate oxidation. Physiological concentrations (10-7-10-‘o M) of testosterone in vitro markedly increased the level of m-aconitase in freshly prepared isolated prostate epithelial cells. This increase was apparent within 3 h of exposure to the hormone. The stimulatory effect of testosterone on m-aconitase was abolished by actinomycin D and by cycloheximide. Both the level of m-aconitase enzyme and the level of m-aconitase activity were similarly increased by
testosterone treatment. Correspondingly, testosterone increased the rate of mitochondrial citrate oxidation while having no effect on the rate of isocitrate oxidation, thereby demonstrating that the action of testosterone is specific&y targeted at the m-aconitase reaction. In vivo studies revealed that castration markedly decreased and testosterone administration increased the m-aconitase level of prostate epithelial cells. In contrast, neither liver nor kidney m-aconitase level was altered by castration. These studies demonstrate that testosterone regulates the biosynthesis of m-aconitase in prostate epithelial c&s. It appears that this is a cell-specific effect since neither kidney nor liver m-aconitase was affected. The studies reveal that prostate epithelial cells contain a constitutive and an androgen-induced m-aconitase component; whereas, kidney and liver contain a constitutive but no androgen-induced m-aconitase. Unlike essentially all other cells, m-aconitase is a regulatory and regulated enzyme in prostate epitheliai cells. Keywords: m-Awn&e;
Citrate oxidation; Prostate cells; Testosterone
1. Introduction The prostate glands of humans and other animals have the unique function of accumulating and secreting enormously high levels of citrate. The metabolic and regulatory relationships of prostate citrate production were reviewed and summarized in our recent review articles (Costello and Franklin, 1991a,b,
*Corresponding author, Tel.: +l 410 7067257, Fax: +1 410 7060193.
1994a,b). Prostate secretory epithelial cells possess unique metabolic relationships which permit this capability of net citrate production (i.e. citrate accumulation and secretion), and which are not associated with any other cells in the body. The most important metabolic characteristic of these cells is their low capability to oxidize citrate. There is now convincing evidence that prostate epithelial cells contain a uniquely limiting m-aconitase (EC 4.2.1.33 which minimizes citrate entry into the Krebs cycle. Consequently, in these cells, the rate of citrate synthesis exceeds the rate of citrate oxidation, resulting in the
0303-7207/95/$09.50 0 1995 Elsevier Science Ireland Ltd. AU rights reserved. SSDI
0303-7207(95)03582-R
46
L.C. Costelloet al. /Molecular and CellularEndocrinology112 (1995) 45-51
accumulation and ultimate secretion of citrate. This is a unique relationship since m-aconitase is not typically considered to be a regulatory or regulated enzyme in the intermediary metabolism of eukaryotic cells. Unlike essentially all other cells, in prostate epithelial cells, citrate is a major end-product of intermediary metabolism. Prostate growth and function are regulated by testosterone. In regard to citrate metabolism, testosterone has a dual role. It stimulates net citrate production by increasing citrate synthesis, and it stimulates citrate oxidation (Harkonen et al., 1982; Franklin et al., 1986; Costello and Franklin, 1991a,b). Earlier studies (Franklin et al., 1986) demonstrated that both citrate oxidation and m-aconitase activity of rat ventral prostate (VP) mitochondria were increased by testosterone. However, the mechanism by which testosterone increases m-aconitase activity of prostate epithelial cells has not been established. In this report, we will provide evidence that (1) testosterone directly stimulates the biosynthesis of m-aconitase enzyme, (2) that this effect occurs under physiological conditions, (3) that this stimulator-y effect of testosterone is specific for prostate epithelial cells, and (4) that this effect is accompanied by a corresponding testosterone-induced increase in mitochondrial citrate oxidation. To the best of our knowledge, this represents the first report of a direct stimulatory effect of any hormone on the biosynthesis of m-aconitase in any eukaryotic cells. The report also demonstrates that m-aconitase can be an important regulated and regulatory enzyme in citrate metabolism. 2. Materials and methods Young adult male rats (275-350 g) were employed in these studies. Castration and sham operation were performed as previously described (Costello and Franklin, 1993; Costello et al., 1994). The preparation of VP isolated epithelial cells and mitochondria has also been described (Costello and Franklin, 1993; Costello et al., 1994). In vitro experiments were conducted in the following manner. Freshly prepared prostate epithelial cells were generally obtained from 6 to 10 rats. The cells were washed and suspended in Hank’s Balance Salt Solution G-IBSS). Aliquots of cell suspension (generally 1.0 ml) were pipetted into 15-ml centrifuge tubes. Experimental tubes were treated by the addition of 50 ~1 of an appropriate concentration of testosterone. Control tubes were treated by the addition of 50 ~1 of vehicle (50% methanol). The tubes were capped and placed on a rotary shaker in a 37°C incubator. After an appropriate incubation period, the tubes were centrifuged at 600 x g for 5 min to
harvest the cells. The cells were suspended in buffered sucrose medium (250 mM sucrose containing 50 mM HEPES buffer, pH 7.2) and were ruptured in a motor-driven conical glass homogenizer. Mitochondria were harvested from the cell homogenate. The mitochondrial pellet was suspended in buffered sucrose medium and sonicated (Biosonic Ultrasonicatar) by three 20-s bursts maintained at 4°C. An aliquot of sonicate was assayed for protein concentration (Bradford, 1976). The remaining sonicate was assayed for m-aconitase. The level of m-aconitase enzyme was determined by Western blots using the procedures previously described (Costello et al., 1994). The m-aconitase antibody was kindly provided by Dr. Paul Srere. This antibody was raised against purified bovine heart maconitase. The results of this study demonstrate that the antibody cross-reacts with rat tissue m-aconitase. We employed highly purified m-aconitase (kindly provided by Dr. Mary Claire Kennedy) and partially purified m-aconitase (Sigma) as standards for the identification of m-aconitase in the rat tissue. An 82-kDa band, which is consistent with the molecular weight of m-aconitase, was identified as m-aconitase. In addition, we performed Western blots on the cell homogenate, the cytosol fraction and on the mitochondrial fraction. The 82-KDa band was readily detected in cell homogenate and mitochondrial preparation but was negligible in the cytosol fraction, thereby confirming that we were detecting m-aconitase. The immunoblots were subjected to optical densitometry, and the relative optical density units were used to estimate the levels of m-aconitase enzyme. The enzyme levels are presented as O.D. units/mg mitochondrial protein. The calculated O.D. units were in the linear range of detection. Duplicate experimental and control tubes were employed in all experiments, and all experiments were repeated at least once to confirm the results. The data presented represent the typical results obtained in these studies. The m-aconitase specific activity of sonicated mitochondria was determined spectrophotometrically by the reduction of 2,6-dichlorophenolindophenol (DCIP). The reaction system (1.0 ml) consisted of 20 mM KCN, 0.2 mM DCPIP, 2 mM cti-aconitate, 0.2 mM NADP, 0.01 mM phenazine methosulfate, 2.0 units isocitrate dehydrogenase (Sigma) and 80 ~1 mitochondrial sonicate in a final volume of 1.0 ml 50 mM HEPES buffer, pH 7.4. The reduction of DCIP was assayed at 600 nm. n-Aconitase specific activity is expressed in nmols of DCIP reduced by cisaconitate/mitochondrial protein per min. Mitochondrial oxidation of citrate and isocitrate was determined as follows. VP epithelial cells were exposed to testosterone or vehicle for 3 h, after which
47
L.C. Costello et al. /Molecular and Cellular Endocrinology 112 (1995) 45-51
O.D.
1.25
1.41
2.58
2.50
c2
Tl
T2
1.08
1.79
2.81
2.7%
Cl
c2
Tl
T2
I
I Mitochondria
Cells
Fig. 1. The in vitro stimulatoty effect of 10-s M testosterone on the level of m-aconitase in prostate epithehal cells. The cells were incubated for 3 h at 37°C in the presence of testosterone or vehicle (control). Immunoblots were performed with 100 pg protein from sonicated cell extracts and sonicated mitochondrial extracts.
the cells were harvested and mitochondria were prepared as described above. Sonicated mitochondrial preparations were incubated with 0.4 mM citrate or enzymatically active isocitrate, 0.2 mM NADP, and 8 x lop4 mM cytochrome C in a final volume of 0.5 ml buffered sucrose, pH 7.2. The reactions were performed in microfuge tubes placed on a rotary mixer at 26°C for up to 60 min. Trichloroacetic acid (TCA, final cont. 7%) was added to terminate the reaction, after which the tubes were centrifuged at 1000 x g for 5 min. The extract was assayed for citrate and isocitrate by fluoroenzymatic assays. Zero-time tubes were prepared by the addition of TCA prior to incubation. Citrate and isocitrate oxidation is expressed as nmols substrate/mg mitochondrial protein per h. 3. Results In the first series of studies, we attempted to determine if a physiological level of testosterone stimulated the level of m-aconitase in freshly isolated VP epithelial cells. The cells were prepared from VP glands resected from 18-h castrated rats. The isolated cells were exposed for 3 h to either 10-s M testosterone or vehicle. The m-aconitase levels of sonicated whole cell extract and sonicated mitochondria extract were determined by Western blots. The results (Fig. 1) demonstrate that testosterone treatment resulted in a significant increase in the m-aconitase level de-
O.D.
1.41
1.72
termined for the cell extract (+90%) and for the mitochondrial extract (+53%). We do not know how much of the immunoreactive protein in cell extract is represented by m-aconitase, precursor m-aconitase, or possibly c-(cytosolic) aconitase. Because of the clear and definitive results obtained with the mitochondrial preparation, all subsequent studies involved the determination of m-aconitase in sonicated mitochondrial extracts. In the next series of studies, we determined the time course of this in vitro stimulatory effect of testosterone on m-aconitase. The conditions were identical to those described above except that the exposure time to lo-* M testosterone was varied over a 5-h period. The results (Fig. 2) demonstrate that the stimulatory effect of testosterone was evident at 3 h and 5 h. At 1 h, no testosterone-induced increase in m-aconitase was evident. Physiological levels of testosterone range from approximately 10-8-10-‘o M. Consequently, it was important to determine the effects of testosterone over this range. With the use of the same experimental conditions described above, we determined the effects of 3-h exposure of 10-7-10-‘o M testosterone on m-aconitase levels. Fig. 3 demonstrates a dose-response increase over the range of 10-7-10-9 M testosterone. At lo-” M, testosterone elicited little or no increase in m-aconitase. It was then important to ascertain if the stimulatory effect of testosterone on m-aconitase was dependent
2.58
2.50
”
Ohr
lhr
3hr
5hr
Fig. 2. Time-course of lo-’ M testosterone on the stimulation of m-aconitase in prostate epithelial cells. The conditions were the same as described in Fig. 1 for sonicated mitochondrial extracts.
L.C. Costello et al. /Molecular and Cell&r Endocrinology 112 (1995) 45-51
48
Fig. 3. The dose response of testosterone stimulation of the level of m-aconitase in prostate epithelial cells. The conditions were the same as described in Fig. 1 for sonicated mitochondrial extracts except for the variation in the concentration of testosterone.
upon protein and/or RNA synthesis. For this, we determined the effect of cycloheximide and actinomycin D on testosterone stimulation of m-aconitase. In these studies, we employed lo-* M testosterone and 3 h exposure time. Fig. 4 demonstrates that both cycloheximide (3 x 10e5 M) and actinomycin (8 x 10T6 M) treatment abolished the stimulatory effect of testosterone. In contrast, neither cycloheximide nor actinomycin had any effect on the control (endogenous) level of m-aconitase. The results support the contention that the stimulatory effect of testosterone on the level of m-aconitase involves both RNA and protein synthesis, i.e. de novo biosynthesis of maconitase. All the studies described above involved the identification of immunoreactive m-aconitase protein. It was important to determine if differences in the level of m-aconitase protein induced by testosterone correlated with changes in m-aconitase enzyme activity. Consequently, we determined the in vitro effect of 3-h exposure of cells to lo- 7 M testosterone on the specific activity of m-aconitase. Mitochondrial sonicates from testosterone-treated cells exhibited a maconitase specific activity of 0.18 nmols/min per mg protein as compared to 0.11 for untreated control cells; i.e. testosterone resulted in a 65% increase in m-aconitase activity. This increase in enzyme activity correlates with the testosterone increase in immunoreactive m-aconitase enzyme level. These results indicated that the rate of mitochondrial citrate oxidation should be significantly increased by testosterone. Consequently, the oxidation rates of citrate and isoci-
O.D.
2.97
6.09
3.21
C
T
T+AD
trate were determined for mitochondrial preparations from testosterone treated and control VP epithelial cells. The results (Fig. 5) demonstrate that testosterone significantly increased ( + 87%) mitochondrial citrate oxidation while having no significant effect on isocitrate oxidation. The absence of any effect on isocitrate oxidation demonstrates that testosterone stimulation of citrate oxidation occurs specifically at the m-aconitase step. This is consistent with our earlier studies which demonstrated that m-aconitase rather than isocitric dehydrogenase was the limiting step of citrate oxidation in VP cells (Costello et al., 1976). The studies described above involved the in vitro effects of testosterone on isolated epithelial cells and demonstrate a direct effect of testosterone on isolated prostate epithelial cells. However, it was also important to determine if the testosterone status in vivo would elicit a regulatory effect on the m-aconitase level of VP epithelial cells. At the same time, we attempted to determine if the in vivo effect would be specific for prostate epithelial cells. In these studies, rats were either castrated or sham-operated and maintained for 18 h. The rats were then sacrificed and the VP glands, liver, and kidney were resected. The mitochondria from each tissue were harvested and assayed for m-aconitase. The results (Fig. 6) clearly demonstrate that the m-aconitase level of VP was markedly decreased by castration. In contrast, castration had no effect on the m-aconitase level of kidney or liver. The results indicate that the constitutive level of m-aconitase in VP is significantly less than the
T+CH
AD
CH
Fig. 4. The effects of cycloheximide (3 X 1O-5 MI and actinomycin D (8 X 10e6 MI on testosterone stimulation of the level of m-aconitase in prostate epithelial cells. The conditions were the same as described in Fig. 1.
49
L.C. Costello et al. /Mohxular and Cellular Enhcrinobgy 112 (199s) 45-51
I!WCIT
was determined. Castrated rats were injected subcutaneously with either 1 mg testosterone or vehicle and sacrificed after 18 h. The results (Fig. 7) demonstrated that testosterone administration increased the maconitase level when compared to the vehicle-injected castrate, and the m-aconitase level was of the same magnitude as the sham (normal) animal. Consequently, testosterone in vivo exhibits the same stimulatory effect on VP m-aconitase as observed in vitro.
60 50 40
CITRATE
0
30
d
20 10 0
+T
-T
4. Discussion
-T
+T
Fig. 5. The in vitro effects of testosterone on mitochondrial citrate and isocitrate oxidation. VP epithelial cells were incubated for 3 h at 37°C in the presence of lo-* M testosterone or vehicle after which the mitochondria were isolated for determination of citrate and isocitrate oxidation. P < 0.01.
constitutive level in kidney or liver. In the normal (sham-operated) animal, the androgen-induced level of m-aconitase approached the constitutive level found in liver and kidney. In order to confirm that the effect of castration was due to testosterone depletion, the effect of testosterone administration to castrated rats
O.D.
0.32
2.23
C
S VP
The results of this study provide evidence that testosterone regulates the level of m-aconitase in prostate epithelial cells. The fact that inhibitors of RNA and protein synthesis abolished this effect of testosterone indicates the likelihood that the mechanism of testosterone regulation involves the biosynthesis of m-aconitase. The in vitro stimulatory effect of physiological levels of testosterone and the in vivo decrease by castration demonstrate that testosterone stimulation of prostate m-aconitase is a physiological and direct hormonal action. In previous studies (Franklin et al., 1987, 1990), we demonstrated that testosterone exerts a similar effect on mitochondrial aspartate aminotransferase (m&AT) of prostate epithelial cells. Moreover, more current information (Qian et al., 1993) has revealed that testosterone increases transcription of the mAAT gene in these cells. This effect is not due to a generalized stimula-
5.63
5.07
1.72
C
S
C
K
1.79
s L
Fig. 6. The in vivo effect of castration on the level of m-aconitase in ventral prostate, liver, and kidney mitochondria. 18-h castrated rats were compared with sham-operated control rats.
O.D.
2.58
4.69
Fig. 7. The effect of testosterone administration to castrated rats on the level of m-aconitase in ventral prostate mitochondria. Castrated rats were injected S.C.with either 1 mg testosterone or vehicle and sacrificed after 18 h. Sham-operated rats served as the control group.
50
L.C. Costello et al. /Molecular
and Cellular Endocrinology 112 (1995) 45-51
tory effect of testosterone on protein synthesis. Earlier studies have demonstrated that stimulation of mAAT was a specific effect since neither the isozyme cytosoic AAT nor mitochondrial malate dehydrogenase were altered by testosterone treatment (Franklin et al., 1982, 1987). Furthermore, testosterone had no effect on cytosolic aconitase activity, whereas maconitase activity was increased (Franklin et al., 1986). Consequently, we believe that the m-aconitase gene, like the mAAT gene, is an androgen-regulated gene in prostate epithelial cells. Studies concerning this possibility are now in progress. m-Aconitase catalyzes the equilibrium reaction by which citrate is converted to isocitrate. This is the entry step of citrate into the Krebs cycle which is necessary for the subsequent oxidation of citrate. m-Aconitase is not generally considered to be a regulatory enzyme in the intermediary metabolism of eukaryotic cells. However, in attempting to determine the biochemical mechanism responsible for the low capability of prostate epithelial cells to oxidize citrate, we reported that m-aconitase activity was the ratelimiting step. Our earlier studies demonstrated that VP tissue and mitochondrial preparations readily utilized isocitrate and contained sufficient isocitrate dehydrogenase activity, whereas citrate oxidation was negligible (Costello et al., 1976, 1978; Franklin et al., 1977, 1978). Additional evidence of a limiting maconitase was provided by the characteristically high citrate/isocitrate ratio (30-40/l) of VP, human prostate and pig prostate (Franklin et al., 1977, unpublished information; Kavanagh, 1994). This is in contrast to the more typical citrate/isocitrate ratio of 9-10/l observed in essentially all other soft tissues. Harkonen et al. (1982) reported that, unlike other tissues, fluoroacetate-treatment of rats had virtually no effect on the citrate oxidation or tissue citrate levels of VP. This can best be explained on the basis that m-aconitase activity of VP is extremely low so that fluoroacetate (converted to fluorocitrate) inhibition of m-aconitase was uniquely ineffective when compared to other tissues. Moreover, the direct assay of m-aconitase activity of VP and kidney mitochondrial preparations revealed that prostate m-aconitase activity was extremely low (Costello and Franklin, 1981). Consequently, there is mounting evidence that m-aconitase is a rate-limiting step for citrate oxidation in prostate epithelial cells. There now exists studies with other cells (Boquist et al., 1985; Hemanz and de la Fuente, 1988; Hall and Hen&son, 1993) which indicate that under certain conditions maconitase activity might be rate-limiting. At this time, the mechanism responsible for the low m-aconitase activity in prostate epithelial cells has not been elucidated. Evidence exists in support of the possibility
that zinc inhibition of m-aconitase might be an important factor (Costello and Franklin, 1981). Earlier studies (Harkonen et al., 1982; Franklin et al., 1986) indicated that testosterone stimulates citrate oxidation and m-aconitase activity of VP. Moreover, in contrast to m-aconitase, c-aconitase activity was unaltered by testosterone. Consequently, the current studies revealing that testosterone increases the level of m-aconitase enzyme are consistent with our previous report of testosterone stimulation of m-aconitase activity of VP mitochondria. This represents the first report of a direct stimulatory effect of testosterone on the biosynthesis of m-aconitase in any cells. To our knowledge, this also represents the first report of direct regulation of m-aconitase biosynthesis by any hormones under physiological conditions. The possibility that this testosterone effect is specific for prostate epithelial cells, as indicated by the absence of the effect on either kidney or liver cells, will be most important. The results of this study reveal that prostate epithelial cells contain a constitutive m-aconitase component and an androgen-induced component; whereas, other cells contain only the constitutive level. Since m-aconitase is not normally a rate-limiting enzyme in kidney, liver, and other tissues, it is clear that the constitutive level of enzyme is sufficient for normal functioning of the Krebs cycle. Consequently, there would be no metabolic or physiologic purpose or role of a hormonally-induced increase in the level of m-aconitase in those tissues. This is evident from the fact that the cellular steady-state citrate/isocitrate ratio is established at the same ratio of the equilibrium established by m-aconitase. In contrast, the constitutive level of m-aconitase in prostate epithelial cells imposes a rate-limiting activity. Testosterone-induced increase in the biosynthesis of maconitase and in m-aconitase activity would provide a mechanism resulting in increased citrate oxidation. Therefore, it makes sense that in concert with its function of accumulating and secreting citrate, maconitase is a rate-limiting regulatory enzyme in prostate epithelial cells; whereas, in essentially all other cells, citrate oxidation is a major pathway of metabolism. It follows that testosterone regulation of m-aconitase is essential to the regulation of citrate oxidation in prostate epithelial cells. It is becoming evident that testosterone exerts a dual function in regard to prostate citrate metabolism. It increases net citrate production and it increases citrate oxidation. The increase in net citrate production results from testosterone stimulation of citrate synthesis. This effect of testosterone is achieved via its stimulation of mAAT and pyruvate dehydrogenase (PDH) El (Y, the two key regulatory enzymes associated with citrate synthesis in prostate epithelial cells
L.C. Costello et al. /Molecular and Cellular Endocrinology 112 (1995) 45-51
(Franklin et al., 1990; Qian et al., 1993; Costello et al., 1994). The increase in citrate oxidation is a result of testosterone stimulation of citrate oxidation. However, under the influence of testosterone, the rate of citrate synthesis still exceeds the rate of citrate oxidation (Franklin et al., 1986). This dual effect is important since it augments citrate production and secretion; and, at the same time, increases energy production required to support other cellular activities which are increased by testosterone. Acknowledgements The authors express deep appreciation to Dr. Paul Srere, VA Medical Center, Dallas, TX for his generosity in providing us with the m-aconitase antibody employed in these studies. We are also appreciative of his many helpful comments and input during the course of this investigation and in the preparation of this paper. We wish to thank Dr. Mary Claire Kennedy, Medical College of Wisconsin, for providing us with purified m-aconitase and for her helpful comments concerning these studies. This research was supported by NIDDK research grant DK20815. References Boquist, L., Ericsson, I., Lorentzon, R. and Nelson, L. (1985) Alterations in mitochondrial aconitase activity and respiration, and in concentration of citrate in some organs of mice with experimental or genetic diabetes. FEBS Lett. 183, 173-176. Bradford, M. (1976) Rapid and sensitive method for quantitation of microgram quantities of protein utihzing the principal of protein-dye binding. Anal. Biochem. 72, 248-254. Costello, L.C., Fad&a, G. and FrankIin, R.B. (1978) Citrate and isocitrate utilization by rat ventral prostate mitochondria. Enzyme 23, 176-181. Costello, L.C. and Franklin, R.B. (1981) Aconitase activity, citrate oxidation, and zinc inhibition in rat ventral prostate. Enzyme 26, 281-287. Costello, L.C. and Franklin, R.B. (1991aJ Concepts of citrate production and secretion by prostate. 1. Metabolic relationships. Prostate 18, 25-46. Costello, L.C. and Franklin, R.B. (1991b) Concepts of citrate production and secretion. 2. Hormonal relationships in normal and neoplastic prostate. Prostate 19, 181-205. Costello, L.C. and Franklin, R.B. (1993) Testosterone regulates pyruvate dehydrogenase activity of prostate mitochondria. Horm. Metab. Res. 25, 268-270.
51
Costello, L.C. and FrankIin, R.B. (1994a) Effect of prolactin onthe prostate. Prostate 24, 162-166. Costello, L.C. and Franklin, R.B. (1994b) The bioenergetic theory of prostate malignancy. Prostate 25, 162-166. Costello, L.C., Franklin, R.B. and Liu, Y. (1994) Testosterone regulates pyruvate dehydrogenase Ela in prostate. Endocrinol. J. 2, 147-151. Costello, L.C., Franklin, R.B. and Stacey, R. (1976) Mitochondrial isocitrate dehydrogenase and isocitrate operation of rat neutral prostrate. Enzyme 21, 495-506. Costello, L., Liu, Y. and Franklin, R.B. (1995) Prolactin specihcahy increases pyruvate dehydrogenase Ela in lateral prostate epithelial cells. Prostate, 26: 189-193. Costello, L.C. and Q’NeilI, J. (1969) A simplified and sensitive method for citrate determination in biological samples. J. Appl. Physiol. 27, 120-122. Franklin, R.B., BrandIy, R.L. and Costello, L.C. (1982) Mitochondrial aspartate aminotransferase and the effect of testosterone on citrate production in rat ventral prostate. J. Ural. 172, 798-802. Franklin, R.B. and Costello, L.C. (1978) Citrate uptake and citrate production by rat ventral prostate fragments. Invest. Urol. 16, 44-47. Franklin, R.B., Costello, L. and Littleton, G. (1977) Citrate uptake and oxidation by fragments of rat ventral prostate. Enzyme 22, 45-51. Franklin, R.B., Kahng, M.W., Akuffo, V. and Costello, L.C. (1986) The effect of testosterone on citrate synthesis and citrate oxidation and a proposed mechanism for regulation of net citrate production in prostate. Harm. Metab. Res. 18, 177-181. Franklin, R.B., Kukoyi, B., Akuffo, V. and Costello, L.C. (1987) Testosterone stimulation of mitochondrial aspartate aminotransferase levels and biosynthesis in rat ventral prostate. J. Steroid Biochem. 28,247-256. Franklin, R.B., Qian, K. and Costello, L.C. (1990) Regulation of aspartate aminotransferase messenger ribonucleic acid level by testosterone. J. Steroid B&hem. 5, 569-574. Hall, R. and Her&son, K. (1993) Mitochondrial myopathy with succinic dehydrogenase and aconitase deficiency. 3. Chn. Invest. 92, 2660-2666. Harkonen, P.L., Kostian, M.L. and Santti, R.S. (1982) Indirect androgenic control of citrate accumulation in rat ventral prostate. Arch. Androl. 8, 107-116. Hemanz, A. and de la Fuente, M. (1988) Characterization of aconitate hydratase from mitochondria and cytoplasm of ascites tumor cells. B&hem. CeU Biol. 66,792-795. Kavanagh, J.P. (1994) Isocitric and citric acid in human prostatic and seminal fluid: Implications for prostatic metabolism and secretion. Prostate 24, 138-142. Qian, K., Franklin, R.B. and Costello, L. (1993) Testosterone regulates mitochondrial aspartate aminotransferase gene expression and mRNA stability in prostate. J. Steroid B&hem. Mol. Biol. 44, 13-19.
Molecular and Cellular Endocrinology 112 (1995) 45-51
Testosterone stimulates the biosynthesis of m-aconitase and citrate oxidation in prostate epithelial cells Leslie C. Costello*, Yiyan Liu, Renty B. Franklin Department of Physiology, Uniuersity of Maryland Dental School, 664 W Baltimore St., Baltimore, MD
21201, USA
Received 20 April 1995;revision received, accepted 12 May 1995
Abstract Mitochondria (m-Iaconitase is a rate-limiting regulatory enzyme in prostate epithelial cells which minimizes citrate oxidation by these cells. This unique metabolic characteristic is responsible for the ability of the prostate to accumulate and secrete extraordinarily high levels of citrate. Testosterone is a major regulator of prostate growth and function, and stimulates citrate oxidation. Therefore, an important action of testosterone might be its stimulation of m-aconitase in prostate epithelial cells. Studies were conducted with rat ventral prostate (VP) epithelial cells to establish the effect of testosterone on the level of m-aconitase and corresponding citrate oxidation. Physiological concentrations (10-7-10-‘o M) of testosterone in vitro markedly increased the level of m-aconitase in freshly prepared isolated prostate epithelial cells. This increase was apparent within 3 h of exposure to the hormone. The stimulatory effect of testosterone on m-aconitase was abolished by actinomycin D and by cycloheximide. Both the level of m-aconitase enzyme and the level of m-aconitase activity were similarly increased by
testosterone treatment. Correspondingly, testosterone increased the rate of mitochondrial citrate oxidation while having no effect on the rate of isocitrate oxidation, thereby demonstrating that the action of testosterone is specific&y targeted at the m-aconitase reaction. In vivo studies revealed that castration markedly decreased and testosterone administration increased the m-aconitase level of prostate epithelial cells. In contrast, neither liver nor kidney m-aconitase level was altered by castration. These studies demonstrate that testosterone regulates the biosynthesis of m-aconitase in prostate epithelial c&s. It appears that this is a cell-specific effect since neither kidney nor liver m-aconitase was affected. The studies reveal that prostate epithelial cells contain a constitutive and an androgen-induced m-aconitase component; whereas, kidney and liver contain a constitutive but no androgen-induced m-aconitase. Unlike essentially all other cells, m-aconitase is a regulatory and regulated enzyme in prostate epitheliai cells. Keywords: m-Awn&e;
Citrate oxidation; Prostate cells; Testosterone
1. Introduction The prostate glands of humans and other animals have the unique function of accumulating and secreting enormously high levels of citrate. The metabolic and regulatory relationships of prostate citrate production were reviewed and summarized in our recent review articles (Costello and Franklin, 1991a,b,
*Corresponding author, Tel.: +l 410 7067257, Fax: +1 410 7060193.
1994a,b). Prostate secretory epithelial cells possess unique metabolic relationships which permit this capability of net citrate production (i.e. citrate accumulation and secretion), and which are not associated with any other cells in the body. The most important metabolic characteristic of these cells is their low capability to oxidize citrate. There is now convincing evidence that prostate epithelial cells contain a uniquely limiting m-aconitase (EC 4.2.1.33 which minimizes citrate entry into the Krebs cycle. Consequently, in these cells, the rate of citrate synthesis exceeds the rate of citrate oxidation, resulting in the
0303-7207/95/$09.50 0 1995 Elsevier Science Ireland Ltd. AU rights reserved. SSDI
0303-7207(95)03582-R
46
L.C. Costelloet al. /Molecular and CellularEndocrinology112 (1995) 45-51
accumulation and ultimate secretion of citrate. This is a unique relationship since m-aconitase is not typically considered to be a regulatory or regulated enzyme in the intermediary metabolism of eukaryotic cells. Unlike essentially all other cells, in prostate epithelial cells, citrate is a major end-product of intermediary metabolism. Prostate growth and function are regulated by testosterone. In regard to citrate metabolism, testosterone has a dual role. It stimulates net citrate production by increasing citrate synthesis, and it stimulates citrate oxidation (Harkonen et al., 1982; Franklin et al., 1986; Costello and Franklin, 1991a,b). Earlier studies (Franklin et al., 1986) demonstrated that both citrate oxidation and m-aconitase activity of rat ventral prostate (VP) mitochondria were increased by testosterone. However, the mechanism by which testosterone increases m-aconitase activity of prostate epithelial cells has not been established. In this report, we will provide evidence that (1) testosterone directly stimulates the biosynthesis of m-aconitase enzyme, (2) that this effect occurs under physiological conditions, (3) that this stimulator-y effect of testosterone is specific for prostate epithelial cells, and (4) that this effect is accompanied by a corresponding testosterone-induced increase in mitochondrial citrate oxidation. To the best of our knowledge, this represents the first report of a direct stimulatory effect of any hormone on the biosynthesis of m-aconitase in any eukaryotic cells. The report also demonstrates that m-aconitase can be an important regulated and regulatory enzyme in citrate metabolism. 2. Materials and methods Young adult male rats (275-350 g) were employed in these studies. Castration and sham operation were performed as previously described (Costello and Franklin, 1993; Costello et al., 1994). The preparation of VP isolated epithelial cells and mitochondria has also been described (Costello and Franklin, 1993; Costello et al., 1994). In vitro experiments were conducted in the following manner. Freshly prepared prostate epithelial cells were generally obtained from 6 to 10 rats. The cells were washed and suspended in Hank’s Balance Salt Solution G-IBSS). Aliquots of cell suspension (generally 1.0 ml) were pipetted into 15-ml centrifuge tubes. Experimental tubes were treated by the addition of 50 ~1 of an appropriate concentration of testosterone. Control tubes were treated by the addition of 50 ~1 of vehicle (50% methanol). The tubes were capped and placed on a rotary shaker in a 37°C incubator. After an appropriate incubation period, the tubes were centrifuged at 600 x g for 5 min to
harvest the cells. The cells were suspended in buffered sucrose medium (250 mM sucrose containing 50 mM HEPES buffer, pH 7.2) and were ruptured in a motor-driven conical glass homogenizer. Mitochondria were harvested from the cell homogenate. The mitochondrial pellet was suspended in buffered sucrose medium and sonicated (Biosonic Ultrasonicatar) by three 20-s bursts maintained at 4°C. An aliquot of sonicate was assayed for protein concentration (Bradford, 1976). The remaining sonicate was assayed for m-aconitase. The level of m-aconitase enzyme was determined by Western blots using the procedures previously described (Costello et al., 1994). The m-aconitase antibody was kindly provided by Dr. Paul Srere. This antibody was raised against purified bovine heart maconitase. The results of this study demonstrate that the antibody cross-reacts with rat tissue m-aconitase. We employed highly purified m-aconitase (kindly provided by Dr. Mary Claire Kennedy) and partially purified m-aconitase (Sigma) as standards for the identification of m-aconitase in the rat tissue. An 82-kDa band, which is consistent with the molecular weight of m-aconitase, was identified as m-aconitase. In addition, we performed Western blots on the cell homogenate, the cytosol fraction and on the mitochondrial fraction. The 82-KDa band was readily detected in cell homogenate and mitochondrial preparation but was negligible in the cytosol fraction, thereby confirming that we were detecting m-aconitase. The immunoblots were subjected to optical densitometry, and the relative optical density units were used to estimate the levels of m-aconitase enzyme. The enzyme levels are presented as O.D. units/mg mitochondrial protein. The calculated O.D. units were in the linear range of detection. Duplicate experimental and control tubes were employed in all experiments, and all experiments were repeated at least once to confirm the results. The data presented represent the typical results obtained in these studies. The m-aconitase specific activity of sonicated mitochondria was determined spectrophotometrically by the reduction of 2,6-dichlorophenolindophenol (DCIP). The reaction system (1.0 ml) consisted of 20 mM KCN, 0.2 mM DCPIP, 2 mM cti-aconitate, 0.2 mM NADP, 0.01 mM phenazine methosulfate, 2.0 units isocitrate dehydrogenase (Sigma) and 80 ~1 mitochondrial sonicate in a final volume of 1.0 ml 50 mM HEPES buffer, pH 7.4. The reduction of DCIP was assayed at 600 nm. n-Aconitase specific activity is expressed in nmols of DCIP reduced by cisaconitate/mitochondrial protein per min. Mitochondrial oxidation of citrate and isocitrate was determined as follows. VP epithelial cells were exposed to testosterone or vehicle for 3 h, after which
47
L.C. Costello et al. /Molecular and Cellular Endocrinology 112 (1995) 45-51
O.D.
1.25
1.41
2.58
2.50
c2
Tl
T2
1.08
1.79
2.81
2.7%
Cl
c2
Tl
T2
I
I Mitochondria
Cells
Fig. 1. The in vitro stimulatoty effect of 10-s M testosterone on the level of m-aconitase in prostate epithehal cells. The cells were incubated for 3 h at 37°C in the presence of testosterone or vehicle (control). Immunoblots were performed with 100 pg protein from sonicated cell extracts and sonicated mitochondrial extracts.
the cells were harvested and mitochondria were prepared as described above. Sonicated mitochondrial preparations were incubated with 0.4 mM citrate or enzymatically active isocitrate, 0.2 mM NADP, and 8 x lop4 mM cytochrome C in a final volume of 0.5 ml buffered sucrose, pH 7.2. The reactions were performed in microfuge tubes placed on a rotary mixer at 26°C for up to 60 min. Trichloroacetic acid (TCA, final cont. 7%) was added to terminate the reaction, after which the tubes were centrifuged at 1000 x g for 5 min. The extract was assayed for citrate and isocitrate by fluoroenzymatic assays. Zero-time tubes were prepared by the addition of TCA prior to incubation. Citrate and isocitrate oxidation is expressed as nmols substrate/mg mitochondrial protein per h. 3. Results In the first series of studies, we attempted to determine if a physiological level of testosterone stimulated the level of m-aconitase in freshly isolated VP epithelial cells. The cells were prepared from VP glands resected from 18-h castrated rats. The isolated cells were exposed for 3 h to either 10-s M testosterone or vehicle. The m-aconitase levels of sonicated whole cell extract and sonicated mitochondria extract were determined by Western blots. The results (Fig. 1) demonstrate that testosterone treatment resulted in a significant increase in the m-aconitase level de-
O.D.
1.41
1.72
termined for the cell extract (+90%) and for the mitochondrial extract (+53%). We do not know how much of the immunoreactive protein in cell extract is represented by m-aconitase, precursor m-aconitase, or possibly c-(cytosolic) aconitase. Because of the clear and definitive results obtained with the mitochondrial preparation, all subsequent studies involved the determination of m-aconitase in sonicated mitochondrial extracts. In the next series of studies, we determined the time course of this in vitro stimulatory effect of testosterone on m-aconitase. The conditions were identical to those described above except that the exposure time to lo-* M testosterone was varied over a 5-h period. The results (Fig. 2) demonstrate that the stimulatory effect of testosterone was evident at 3 h and 5 h. At 1 h, no testosterone-induced increase in m-aconitase was evident. Physiological levels of testosterone range from approximately 10-8-10-‘o M. Consequently, it was important to determine the effects of testosterone over this range. With the use of the same experimental conditions described above, we determined the effects of 3-h exposure of 10-7-10-‘o M testosterone on m-aconitase levels. Fig. 3 demonstrates a dose-response increase over the range of 10-7-10-9 M testosterone. At lo-” M, testosterone elicited little or no increase in m-aconitase. It was then important to ascertain if the stimulatory effect of testosterone on m-aconitase was dependent
2.58
2.50
”
Ohr
lhr
3hr
5hr
Fig. 2. Time-course of lo-’ M testosterone on the stimulation of m-aconitase in prostate epithelial cells. The conditions were the same as described in Fig. 1 for sonicated mitochondrial extracts.
L.C. Costello et al. /Molecular and Cell&r Endocrinology 112 (1995) 45-51
48
Fig. 3. The dose response of testosterone stimulation of the level of m-aconitase in prostate epithelial cells. The conditions were the same as described in Fig. 1 for sonicated mitochondrial extracts except for the variation in the concentration of testosterone.
upon protein and/or RNA synthesis. For this, we determined the effect of cycloheximide and actinomycin D on testosterone stimulation of m-aconitase. In these studies, we employed lo-* M testosterone and 3 h exposure time. Fig. 4 demonstrates that both cycloheximide (3 x 10e5 M) and actinomycin (8 x 10T6 M) treatment abolished the stimulatory effect of testosterone. In contrast, neither cycloheximide nor actinomycin had any effect on the control (endogenous) level of m-aconitase. The results support the contention that the stimulatory effect of testosterone on the level of m-aconitase involves both RNA and protein synthesis, i.e. de novo biosynthesis of maconitase. All the studies described above involved the identification of immunoreactive m-aconitase protein. It was important to determine if differences in the level of m-aconitase protein induced by testosterone correlated with changes in m-aconitase enzyme activity. Consequently, we determined the in vitro effect of 3-h exposure of cells to lo- 7 M testosterone on the specific activity of m-aconitase. Mitochondrial sonicates from testosterone-treated cells exhibited a maconitase specific activity of 0.18 nmols/min per mg protein as compared to 0.11 for untreated control cells; i.e. testosterone resulted in a 65% increase in m-aconitase activity. This increase in enzyme activity correlates with the testosterone increase in immunoreactive m-aconitase enzyme level. These results indicated that the rate of mitochondrial citrate oxidation should be significantly increased by testosterone. Consequently, the oxidation rates of citrate and isoci-
O.D.
2.97
6.09
3.21
C
T
T+AD
trate were determined for mitochondrial preparations from testosterone treated and control VP epithelial cells. The results (Fig. 5) demonstrate that testosterone significantly increased ( + 87%) mitochondrial citrate oxidation while having no significant effect on isocitrate oxidation. The absence of any effect on isocitrate oxidation demonstrates that testosterone stimulation of citrate oxidation occurs specifically at the m-aconitase step. This is consistent with our earlier studies which demonstrated that m-aconitase rather than isocitric dehydrogenase was the limiting step of citrate oxidation in VP cells (Costello et al., 1976). The studies described above involved the in vitro effects of testosterone on isolated epithelial cells and demonstrate a direct effect of testosterone on isolated prostate epithelial cells. However, it was also important to determine if the testosterone status in vivo would elicit a regulatory effect on the m-aconitase level of VP epithelial cells. At the same time, we attempted to determine if the in vivo effect would be specific for prostate epithelial cells. In these studies, rats were either castrated or sham-operated and maintained for 18 h. The rats were then sacrificed and the VP glands, liver, and kidney were resected. The mitochondria from each tissue were harvested and assayed for m-aconitase. The results (Fig. 6) clearly demonstrate that the m-aconitase level of VP was markedly decreased by castration. In contrast, castration had no effect on the m-aconitase level of kidney or liver. The results indicate that the constitutive level of m-aconitase in VP is significantly less than the
T+CH
AD
CH
Fig. 4. The effects of cycloheximide (3 X 1O-5 MI and actinomycin D (8 X 10e6 MI on testosterone stimulation of the level of m-aconitase in prostate epithelial cells. The conditions were the same as described in Fig. 1.
49
L.C. Costello et al. /Mohxular and Cellular Enhcrinobgy 112 (199s) 45-51
I!WCIT
was determined. Castrated rats were injected subcutaneously with either 1 mg testosterone or vehicle and sacrificed after 18 h. The results (Fig. 7) demonstrated that testosterone administration increased the maconitase level when compared to the vehicle-injected castrate, and the m-aconitase level was of the same magnitude as the sham (normal) animal. Consequently, testosterone in vivo exhibits the same stimulatory effect on VP m-aconitase as observed in vitro.
60 50 40
CITRATE
0
30
d
20 10 0
+T
-T
4. Discussion
-T
+T
Fig. 5. The in vitro effects of testosterone on mitochondrial citrate and isocitrate oxidation. VP epithelial cells were incubated for 3 h at 37°C in the presence of lo-* M testosterone or vehicle after which the mitochondria were isolated for determination of citrate and isocitrate oxidation. P < 0.01.
constitutive level in kidney or liver. In the normal (sham-operated) animal, the androgen-induced level of m-aconitase approached the constitutive level found in liver and kidney. In order to confirm that the effect of castration was due to testosterone depletion, the effect of testosterone administration to castrated rats
O.D.
0.32
2.23
C
S VP
The results of this study provide evidence that testosterone regulates the level of m-aconitase in prostate epithelial cells. The fact that inhibitors of RNA and protein synthesis abolished this effect of testosterone indicates the likelihood that the mechanism of testosterone regulation involves the biosynthesis of m-aconitase. The in vitro stimulatory effect of physiological levels of testosterone and the in vivo decrease by castration demonstrate that testosterone stimulation of prostate m-aconitase is a physiological and direct hormonal action. In previous studies (Franklin et al., 1987, 1990), we demonstrated that testosterone exerts a similar effect on mitochondrial aspartate aminotransferase (m&AT) of prostate epithelial cells. Moreover, more current information (Qian et al., 1993) has revealed that testosterone increases transcription of the mAAT gene in these cells. This effect is not due to a generalized stimula-
5.63
5.07
1.72
C
S
C
K
1.79
s L
Fig. 6. The in vivo effect of castration on the level of m-aconitase in ventral prostate, liver, and kidney mitochondria. 18-h castrated rats were compared with sham-operated control rats.
O.D.
2.58
4.69
Fig. 7. The effect of testosterone administration to castrated rats on the level of m-aconitase in ventral prostate mitochondria. Castrated rats were injected S.C.with either 1 mg testosterone or vehicle and sacrificed after 18 h. Sham-operated rats served as the control group.
50
L.C. Costello et al. /Molecular
and Cellular Endocrinology 112 (1995) 45-51
tory effect of testosterone on protein synthesis. Earlier studies have demonstrated that stimulation of mAAT was a specific effect since neither the isozyme cytosoic AAT nor mitochondrial malate dehydrogenase were altered by testosterone treatment (Franklin et al., 1982, 1987). Furthermore, testosterone had no effect on cytosolic aconitase activity, whereas maconitase activity was increased (Franklin et al., 1986). Consequently, we believe that the m-aconitase gene, like the mAAT gene, is an androgen-regulated gene in prostate epithelial cells. Studies concerning this possibility are now in progress. m-Aconitase catalyzes the equilibrium reaction by which citrate is converted to isocitrate. This is the entry step of citrate into the Krebs cycle which is necessary for the subsequent oxidation of citrate. m-Aconitase is not generally considered to be a regulatory enzyme in the intermediary metabolism of eukaryotic cells. However, in attempting to determine the biochemical mechanism responsible for the low capability of prostate epithelial cells to oxidize citrate, we reported that m-aconitase activity was the ratelimiting step. Our earlier studies demonstrated that VP tissue and mitochondrial preparations readily utilized isocitrate and contained sufficient isocitrate dehydrogenase activity, whereas citrate oxidation was negligible (Costello et al., 1976, 1978; Franklin et al., 1977, 1978). Additional evidence of a limiting maconitase was provided by the characteristically high citrate/isocitrate ratio (30-40/l) of VP, human prostate and pig prostate (Franklin et al., 1977, unpublished information; Kavanagh, 1994). This is in contrast to the more typical citrate/isocitrate ratio of 9-10/l observed in essentially all other soft tissues. Harkonen et al. (1982) reported that, unlike other tissues, fluoroacetate-treatment of rats had virtually no effect on the citrate oxidation or tissue citrate levels of VP. This can best be explained on the basis that m-aconitase activity of VP is extremely low so that fluoroacetate (converted to fluorocitrate) inhibition of m-aconitase was uniquely ineffective when compared to other tissues. Moreover, the direct assay of m-aconitase activity of VP and kidney mitochondrial preparations revealed that prostate m-aconitase activity was extremely low (Costello and Franklin, 1981). Consequently, there is mounting evidence that m-aconitase is a rate-limiting step for citrate oxidation in prostate epithelial cells. There now exists studies with other cells (Boquist et al., 1985; Hemanz and de la Fuente, 1988; Hall and Hen&son, 1993) which indicate that under certain conditions maconitase activity might be rate-limiting. At this time, the mechanism responsible for the low m-aconitase activity in prostate epithelial cells has not been elucidated. Evidence exists in support of the possibility
that zinc inhibition of m-aconitase might be an important factor (Costello and Franklin, 1981). Earlier studies (Harkonen et al., 1982; Franklin et al., 1986) indicated that testosterone stimulates citrate oxidation and m-aconitase activity of VP. Moreover, in contrast to m-aconitase, c-aconitase activity was unaltered by testosterone. Consequently, the current studies revealing that testosterone increases the level of m-aconitase enzyme are consistent with our previous report of testosterone stimulation of m-aconitase activity of VP mitochondria. This represents the first report of a direct stimulatory effect of testosterone on the biosynthesis of m-aconitase in any cells. To our knowledge, this also represents the first report of direct regulation of m-aconitase biosynthesis by any hormones under physiological conditions. The possibility that this testosterone effect is specific for prostate epithelial cells, as indicated by the absence of the effect on either kidney or liver cells, will be most important. The results of this study reveal that prostate epithelial cells contain a constitutive m-aconitase component and an androgen-induced component; whereas, other cells contain only the constitutive level. Since m-aconitase is not normally a rate-limiting enzyme in kidney, liver, and other tissues, it is clear that the constitutive level of enzyme is sufficient for normal functioning of the Krebs cycle. Consequently, there would be no metabolic or physiologic purpose or role of a hormonally-induced increase in the level of m-aconitase in those tissues. This is evident from the fact that the cellular steady-state citrate/isocitrate ratio is established at the same ratio of the equilibrium established by m-aconitase. In contrast, the constitutive level of m-aconitase in prostate epithelial cells imposes a rate-limiting activity. Testosterone-induced increase in the biosynthesis of maconitase and in m-aconitase activity would provide a mechanism resulting in increased citrate oxidation. Therefore, it makes sense that in concert with its function of accumulating and secreting citrate, maconitase is a rate-limiting regulatory enzyme in prostate epithelial cells; whereas, in essentially all other cells, citrate oxidation is a major pathway of metabolism. It follows that testosterone regulation of m-aconitase is essential to the regulation of citrate oxidation in prostate epithelial cells. It is becoming evident that testosterone exerts a dual function in regard to prostate citrate metabolism. It increases net citrate production and it increases citrate oxidation. The increase in net citrate production results from testosterone stimulation of citrate synthesis. This effect of testosterone is achieved via its stimulation of mAAT and pyruvate dehydrogenase (PDH) El (Y, the two key regulatory enzymes associated with citrate synthesis in prostate epithelial cells
L.C. Costello et al. /Molecular and Cellular Endocrinology 112 (1995) 45-51
(Franklin et al., 1990; Qian et al., 1993; Costello et al., 1994). The increase in citrate oxidation is a result of testosterone stimulation of citrate oxidation. However, under the influence of testosterone, the rate of citrate synthesis still exceeds the rate of citrate oxidation (Franklin et al., 1986). This dual effect is important since it augments citrate production and secretion; and, at the same time, increases energy production required to support other cellular activities which are increased by testosterone. Acknowledgements The authors express deep appreciation to Dr. Paul Srere, VA Medical Center, Dallas, TX for his generosity in providing us with the m-aconitase antibody employed in these studies. We are also appreciative of his many helpful comments and input during the course of this investigation and in the preparation of this paper. We wish to thank Dr. Mary Claire Kennedy, Medical College of Wisconsin, for providing us with purified m-aconitase and for her helpful comments concerning these studies. This research was supported by NIDDK research grant DK20815. References Boquist, L., Ericsson, I., Lorentzon, R. and Nelson, L. (1985) Alterations in mitochondrial aconitase activity and respiration, and in concentration of citrate in some organs of mice with experimental or genetic diabetes. FEBS Lett. 183, 173-176. Bradford, M. (1976) Rapid and sensitive method for quantitation of microgram quantities of protein utihzing the principal of protein-dye binding. Anal. Biochem. 72, 248-254. Costello, L.C., Fad&a, G. and FrankIin, R.B. (1978) Citrate and isocitrate utilization by rat ventral prostate mitochondria. Enzyme 23, 176-181. Costello, L.C. and Franklin, R.B. (1981) Aconitase activity, citrate oxidation, and zinc inhibition in rat ventral prostate. Enzyme 26, 281-287. Costello, L.C. and Franklin, R.B. (1991aJ Concepts of citrate production and secretion by prostate. 1. Metabolic relationships. Prostate 18, 25-46. Costello, L.C. and Franklin, R.B. (1991b) Concepts of citrate production and secretion. 2. Hormonal relationships in normal and neoplastic prostate. Prostate 19, 181-205. Costello, L.C. and Franklin, R.B. (1993) Testosterone regulates pyruvate dehydrogenase activity of prostate mitochondria. Horm. Metab. Res. 25, 268-270.
51
Costello, L.C. and FrankIin, R.B. (1994a) Effect of prolactin onthe prostate. Prostate 24, 162-166. Costello, L.C. and Franklin, R.B. (1994b) The bioenergetic theory of prostate malignancy. Prostate 25, 162-166. Costello, L.C., Franklin, R.B. and Liu, Y. (1994) Testosterone regulates pyruvate dehydrogenase Ela in prostate. Endocrinol. J. 2, 147-151. Costello, L.C., Franklin, R.B. and Stacey, R. (1976) Mitochondrial isocitrate dehydrogenase and isocitrate operation of rat neutral prostrate. Enzyme 21, 495-506. Costello, L., Liu, Y. and Franklin, R.B. (1995) Prolactin specihcahy increases pyruvate dehydrogenase Ela in lateral prostate epithelial cells. Prostate, 26: 189-193. Costello, L.C. and Q’NeilI, J. (1969) A simplified and sensitive method for citrate determination in biological samples. J. Appl. Physiol. 27, 120-122. Franklin, R.B., BrandIy, R.L. and Costello, L.C. (1982) Mitochondrial aspartate aminotransferase and the effect of testosterone on citrate production in rat ventral prostate. J. Ural. 172, 798-802. Franklin, R.B. and Costello, L.C. (1978) Citrate uptake and citrate production by rat ventral prostate fragments. Invest. Urol. 16, 44-47. Franklin, R.B., Costello, L. and Littleton, G. (1977) Citrate uptake and oxidation by fragments of rat ventral prostate. Enzyme 22, 45-51. Franklin, R.B., Kahng, M.W., Akuffo, V. and Costello, L.C. (1986) The effect of testosterone on citrate synthesis and citrate oxidation and a proposed mechanism for regulation of net citrate production in prostate. Harm. Metab. Res. 18, 177-181. Franklin, R.B., Kukoyi, B., Akuffo, V. and Costello, L.C. (1987) Testosterone stimulation of mitochondrial aspartate aminotransferase levels and biosynthesis in rat ventral prostate. J. Steroid Biochem. 28,247-256. Franklin, R.B., Qian, K. and Costello, L.C. (1990) Regulation of aspartate aminotransferase messenger ribonucleic acid level by testosterone. J. Steroid B&hem. 5, 569-574. Hall, R. and Her&son, K. (1993) Mitochondrial myopathy with succinic dehydrogenase and aconitase deficiency. 3. Chn. Invest. 92, 2660-2666. Harkonen, P.L., Kostian, M.L. and Santti, R.S. (1982) Indirect androgenic control of citrate accumulation in rat ventral prostate. Arch. Androl. 8, 107-116. Hemanz, A. and de la Fuente, M. (1988) Characterization of aconitate hydratase from mitochondria and cytoplasm of ascites tumor cells. B&hem. CeU Biol. 66,792-795. Kavanagh, J.P. (1994) Isocitric and citric acid in human prostatic and seminal fluid: Implications for prostatic metabolism and secretion. Prostate 24, 138-142. Qian, K., Franklin, R.B. and Costello, L. (1993) Testosterone regulates mitochondrial aspartate aminotransferase gene expression and mRNA stability in prostate. J. Steroid B&hem. Mol. Biol. 44, 13-19.