- Email: [email protected]
Epinephrine synthesis by an N-methyltransferase in rat liver
GASTROENTEROLOGY
1990;98:152-155
Epinephrine Synthesis by an N-Methyltransferase in Rat Liver HAMZEH ELAYAN, BRIAN KENNEDY, and MICHAEL G. ZIEGLER Department of Medicine, Division of Nephrology, University of California San Diego Medical Center, San Diego, California
We investigated if liver can synthesize epinephrine in vitro and in vivo. Homogenates of rat liver readily synthesized [‘Hlepinephrine from [3H]S-adenosylmethionine and norepinephrine. Liver homogenates also N-methylated dopamine at more than twice the rate that they N-methylated norepinephrine. In contrast, adrenal homogenates, which AJ-methylate norepinephrine to form epinephrine using the enzyme phenylethanolamine-N-methyltransferase (PNMT), methylated dopamine only about 1% as well as norepinephrine. Synthesis of epinephrine by liver homogenates was not significantly inhibited by the PNMT inhibitor SKF 29661 at a concentration that inhibited adrenal homogenate epinephrine synthesis by nearly 99%. These findings indicate that liver can synthesize epinephrine in vitro using an enzyme other than PNMT.. Adrenal demedullation of rats reduced plasma epinephrine levels to 7% of control values, but left liver epinephrine and epinephrineforming enzyme levels unchanged. Treatment of demedullated rats with 6-hydroxydopamine plus reserpine also resulted in dramatically reduced plasma epinephrine levels but no change in hepatic epinephrine and N-methylating enzyme levels. We conclude that the liver synthesizes its own epinephrine.
E
pinephrine (E) has a profound effect on liver function. Its most important action is an increase in glucose output (l-3).This is primarily due to enhanced hepatic glycogen breakdown into glucose (Z-5). Epinephrine-induced increases in glycogenolysis are mediated in part by elevations in adenylate cyclase activity (3,6), cyclic adenosine monophosphate (2.77, and phosphorylase activity (2,7-9), and decreases in glycogen synthase activity (3.74). Epinephrine increases liver RNA and protein synthesis (111,and decreases hepatocyte triglyceride and triglycerol re-
lease (12), carbamyl phosphate synthetase synthesis (13), and leucine incorporation in hepatocytes (14). During stress or hypoglycemia sympathetic nerves cause release of E into the bloodstream from the adrenal. However, the adrenal is not the only source of circulating E. It has been found in the plasma and urine of bilaterally adrenalectomized humans (15,161 and adrenodemedullated rats (17,18). The source of this E is unknown but the enzyme phenylethanolamine-N-methyltransferase (PNMT), which synthesizes E from norepinephrine (NE) in the adrenal (19), is present in several other organs including the brain, heart, and lung (20).A nonspecific N-methyltransferase (NMT) that could potentially synthesize E from NE is also present in several organs including the liver (20,21). The liver contains NE and the methyl donor S-adenosylmethionine in its sympathetic nerves (22). Thus, the liver contains all the reactants necessary to synthesize E. Prior assays for NMT activity have used substrates such as phenylethanolamine. Although these assays had the advantage of speed and sensitivity, they could not measure actual E formation nor evaluate whether NMTs other than PNMT might synthesize E. We used a new assay (23)that measures conversion of NE into E or dopamine (DA] into epinine. This assay, combined with measures of liver E in animals with low circulating E levels, provides a reasonable guide to hepatic E synthesis. Materials and Methods Male Sprague-Dawley rats weighing 180-200 g were subjected to either bilateral adrenal demedullation (20 rats) Abbreviations used in this paper: DA, dopamine; E, epinephrine; NE, norepinephrine; NMT, N-methyltransferase; PNMT. phenylethanolamine-N-methyltransferase. 0 1990 by the American Gastroenterological Association 0018-5085/80/$3.00
January 1990
EPINEPHRINE
or sham operation (9 rats) under ether anesthesia. The rats were allowed to recover for 1 wk, then each of 9 bilaterally demedullated rats received one dose of 6-hydroxydopamine [Sigma Chemical Co, St. Louis, MO.), 20 mg/kg i.p., followed 24 h later by four doses on successive days of reserpine (Sigma), 5 mg/kg i.p. dissolved in 10% ascorbic acid in 0.9% saline. This treatment was designed to lower E in the circulation by demedullation, destroy sympathetic nerve terminals with 6-hydroxydopamine, and eliminate catecholamine storage in any nerve terminals that survived 6hydroxydopamine treatment. Rats treated with demedullation, 6-hydroxydopamine and reserpine are referred to as D-6-R rats. We thus established three groups of rats for experimentation. Nine rats that underwent sham adrenal medullectomy constitute the control group. Eleven rats underwent adrenal medullectomy and are referred to as AMX rats. Nine rats underwent D-6-R treatment. The data shown in Figures 1 and 2 come from the control animals, and data in Figures 3-5 from all three groups of rats. All rats were killed by decapitation 24 h after the last dose of reserpine was given and a blood sample was collected from each rat into a heparinized tube. Decapitation is a potent stimulus to adrenal E release, so trunk blood was collected after decapitation and E levels were measured to determine the adequacy of adrenal demedullation. Liver samples weighing about 500 mg were separated, washed with ice-cold saline, and stored at -70%. The brainstem, adrenals, and salivary glands were separated from the sham-operated rats, washed with ice-cold saline, and stored at -70°C. Tissues were homogenized in 1 ml of 0.1 M tris-HCl buffer with 0.1% Triton (pH 7) then diluted to form 100 mg/ml, and centrifuged for 10 min at 5800 g. The supernatant was stored at -70°C. Values are expressed per gram wet weight of liver tissue. Catecholamines were assayed in duplicate lOO-~1samples of tissue homogenates or plasma according to the catechol-omethyltransferase-based radioenzymatic method of Ziegler
LIVER
SYNTHESIS
ADRENAL
BY RAT LIVER
153
BRAINSTEM
Figure 2. Rate of N-methylation of DA by homogenates of liver, adrenal, and brainstem. Either NE or DA were present in the incubation mix at lo-’ M. Values are expressed as percentage of rate of N-methylation of NE, and are the mean + SEM of four to five determinations. The three tissues differ from each other by analysis of variance, p ~0.001. kp to.05 vs. liver by Duncan’s test.
al. (24). Liver N-methylating enzyme activities were assayed in 50-J duplicates of tissue homogenates according to the radioenzymatic method of Ziegler et al. (23). using 10e3 M NE as substrate in the presence or absence of the PNMT inhibitor SKF 29661 at 10m4M (kindly supplied by Dr. Paul Hieble, Smith Kline Beckman Corp.). Norepinephrine and DA, both at the concentration of 10e4 M, were used as substrates in certain determinations, and the ratios of epinine formed from DA to E formed from NE were determined. Liver N-methylating enzyme was partially characterized by comparing the effect of SKF 29661 and the DA/NE ratio of the liver homogenates with the corresponding values of the adrenal glands, brainstem, and salivary glands. et
300 60
50
e," 0
B a
20
10 2
50
K
*
*
a
> I
LIVER
ADRENAL
0I 30
SHAM
AMX
D6R
BRAINSTEM
Figure 1. Percentage of E-forming activity remaining in the presence of lo-’ M SICP29661 of homogenates from liver, adrenal, and brainstem. Norepinephrine was present in the incubation mix at 10m3M. Values are expressed as percentage of activity in the absence of inhibitor and are the mean + SEM of four to live determinations. The three tissues differ from each other by analysis of variance, p
Figure 9. The rate of E formation from NE by liver homogenates horn three groups of rats. Sham-operated (SHAM) animals served as the control group. Adrenal demedullated (AMX) animals had their adrenal medulIae removed, and D-6-R at&n& also received 6-hydroxydopamine and reset-pine. There was no significant difference in the rate of E formation among groups. Values are shown as picomoles of E formed per gram of liver per hour and are the mean * SEM of 9-11 rats.
154
ELAYAN
ET AL.
GASTROENTEROLOGY
BOO0
I-
8000 7000
t
/
z
5000..
g
4000.-
zooo-1000-OT
*
Discussion
E
DA
Figure 4. Plasma catecholamine levels in sham-operated (SHAM), adrenal demedullated (AMX), and demedullated rats that also received &hydroxydopamine and reserpine (D6Rj Values are expressed in picograms per milliliter and are the mean + SEM of S-11 rats. Treatment groups are signiikantly different from each other by analysis of variance with respect to NE (p c 0.0011,DA (p < 0.05j and E (p < 0.001). ti ~0.05 vs. SHAM by Duncan’s test. 0, SHAM; R$AMX; @ D6R.
The results were expressed as mean values +SEM. Differences among groups were determined by analysis of variance and Duncan’s test. Results Liver homogenates readily synthesized [3H]E from NE and [3H]SAM. This N-methylating activity was not reduced by the PNMT inhibitor SKF 29661 at 1W4 M (Figure 1). In contrast, the E-forming activity of adrenal homogenate was inhibited by 88% using this concentration of SKF 29661 (Figure 1). The liver enzyme methylated DA more than twice as well as 2
0
NE
DA
NE, whereas adrenal and brainstem homogenates methylated DA only about 1% as well as NE in control animals (Figure 2). Adrenal demedullation and D-6-R treatment tended to increase hepatic N-methylating activity (Figure 3). Adrenal demedullation reduced plasma E levels to 7% of control values (p < 0.05). D-6-R treatment also significantly reduced plasma E [p < 0.05) (Figure 4). In contrast, adrenal demedullation and D-6-R treatment did not reduce liver E levels from those observed in sham-operated controls (Figure 5).
* l!iiikA I
3000.~
NE
Vol. 98, No. 1
E
Figure 5. Liver catecholamine levels in sham-operated I(SHAW adrenal demedullated (AMX), and demedullated rats tlhat also received &hydroxydopamine and reserpine (DORj Values are expressed in nanograms per gram tissue and are the mean + SEM of 6-11 rats. Treatment groups are significantly different from each other by analysis of variance with respect to NE (p -C0.001). *p < 0.05 vs. SHAM by Duncan’s test. 0, SHAM; RI,AMX; ?? , D6R.
Liver homogenates methylated NE to form E. Earlier investigators reported that liver homogenates could N-methylate phenylethanolamine (20,21). However, the liver N-methylating enzyme does not appear to be PNMT. It was not inhibited by the PNMT inhibitor SKF 29661 at concentrations that greatly inhibit adrenal PNMT. Also the liver enzyme methylated the phenylethylamine DA more than twice as well as the phenylethanolamine NE. In contrast, adrenal PNMT methylates DA only about one onehundredth as well as NE. Other investigators have also provided evidence that the dominant liver Nmethylating enzyme is not PNMT (20,21). It is a less specific N-methyltransferase that can methylate both NE and DA. The liver N-methylating enzyme appears to synthesize E in vivo. After adrenal demedullation, plasma E levels were reduced to 7% of control values. In contrast, liver E was unaffected by removal of the adrenal medullae. Compared with organs such as the heart and spleen, the liver accumulates E from the bloodstream with only moderate efficiency (25). Axelrod et al. [25) found that 2 h after a [3H]E infusion, only 40% of the E taken up by the liver remained unmetabolized. Thus, the E that we find in liver 11 days after adrenal medullectomy is unlikely to be of adrenal origin. Epinephrine levels in the liver were also unchanged in D-6-R rats, despite the fact that liver NE was lower and N-methylating activity was not significantly increased. This suggests that in normal rats nearly all liver E is locally synthesized. Liver E does not appear to be stored in sympathetic nerves. D-6-R treatment was designed to destroy sympathetic nerves with 6-hydroxydopamine and prevent catecholamine storage in any remaining nerve terminals with reserpine. The D-6-R treatment effectively lowered hepatic NE, but failed to lower hepatic E (Figure 5). Epinephrine differs from NE in its effects on liver function. Epinephrine has a greater effect on liver adenylate cyclase activity (6) and cyclic adenosine monophosphate levels (2) than NE. Norepinephrine increases glycerol and fatty acid release from liver
January 1990
whereas E does not (26). Thus, hepatic conversion of NE to E could have important effects on body chemistry. The liver N-methylating enzyme may also synthesize epinine from DA in vivo. Dopamine is present in the liver at about one-fifth of the level of NE, and liver homogenates readily N-methylate DA. The effects of epinine on hepatic function have not been studied. However, epinine stimulates CX-,p-, and DA-receptors and it differs in pharmacologic profile from DA (27). Liver NMT methylates DA to form epinine more readily than it methylates NE to form E (Figure 2). The amount of epinine and E formed in the liver will depend on the availability of DA and NE to serve as substrates. Most tissue NE is bound in sympathetic nerve vesicles so the amount of free NE available in hepatic tissue may be relatively low. Local E synthesis may underlie the maintenance of normal hepatic E levels while this organ is cultured in vitro. It may also provide the mechanism by which resting rats without adrenal medullae maintain normal liver glycogen and blood glucose levels (29,301. References I. Usami M, Seino Y. Taminato
T, et al. Effect of DG5128 on epinephrine and glucagon induced glucose output from the isolated perfused rat liver. Life Sci 1988;42:129-36. 2. Brighenti L. Puviani AC, Gavioli ME, Fabbri E, Ottolenghi C. Catecholamine effect on cyclic adenosine 3’:5’-monophosphate level in isolated catfish hepatocytes. Gen Comp Endocrinol 1987;68:216-23. 3. Exton JH. Mechanisms of hormonal regulation of hepatic glucose metabolism. Diabetes Metab Rev 1987;3:163-83. 4. Graham SM, Herring PA, Arinze IJ. Age-associated alterations in hepatic beta-adrenergic receptor/adenylate cyclase complex. Am J Physioll987;253:E277-82. 5. James AE, Burleigh KM, Kraegen EW. In vivo glucose metabolism in individual tissues of the rat. Interaction between epinephrine and insulin. J Biol Chem 1986;261:6366-74. 6. Sanae F. Matsunaga T, Miyamoto K, Koshivra R. Studies on responsiveness of hepatoma cells to catecholamines III. Differences between the receptor-adenylate cyclase regulating systems in AH130 cells and cultured normal rat liver cells. Jpn J Pharmacol1986;42:281-8. 7. Nemeth S, Viskupic E. Kvetnansky R, Kolena J. Effect of epinephrine infusions imitating plasma epinephrine levels in immobilization stress and of fasting on the activity of key enzymes in catfish liver and muscles. Physiol Bohemoslov 1986;61:469-75. 8. Nemeth S, Viskupic E. Dose-dependent relationships between the effects of epinephrine and glucose on the activity of key enzymes of glycogen metabolism in the rat liver. Physiol Bohemoslov 1987:36:X-20. 9. Nemeth S, Viskupic E, Kvetnansky R, Kolena J. Progressive weakening of the response of key enzymes of liver glycogen metabolism to permanently increased plasma epinephrine levels. Physiol Bohemoslov 1987;36:441-6. 10. Ottolenghi C, Puviani KC, Gavioli ME, Brighenti L. Epinephrine effect on glycogen phosphorylase activity in catfish liver and muscles. Gen Comp Endocrinoll986;61:469-75. 11. Breton L. Maissiat R. Clot JP, Baudry M. Effects of glucagon and epinephrine on the rat liver: oxidative phosphorylation and ultrastructure. C R Sot Biol [Paris] 1986;180:613-21.
EPINEPHRINE
SYNTHESIS BY RAT LIVER
155
12. Brindle NP. Ontko JA. Suppression of triglyceride secretion by epinephrine in isolated rat hepatocytes. Biochem Biophys Res Commun 1986;141:191-7. 13. Kitigawa Y. Hormonal regulation of carbamoyl-phosphate synthetase I synthesis in primary cultured hepatocytes and Reuber hepatoma H-35. Defective regulation in hepatoma cells. Eur J Biochem 1987;167:19-25. 14. Brostrom CO, Bocckino SB, Brostrom MA, Galuska EM. Regulation of protein synthesis in isolated hepatocytes by calcium mobilizing hormones. Mol Pharmacoll986;29:104-11. 15. Von Euler US, Ikkos D, Luft R. Adrenaline excretion during resting conditions and after insulin in adrenalectomized human subjects. Acta Endocrinol 1961;38:441-8. 16. Hollifield JW, Nadeau JJ. Vasodilator therapy. In: Hunter JC, ed. Hypertension update. Bloomfield, N.J.: Health Learning Systems, 1980:235-46. 17. Pendleton RG, Weiner G, Jenkins B, Gessner G. Effects of an inhibitor of phenylethanolamine N-methyltransferase upon stimulated adrenal catecholamine release and excretion in the rat. Naunyn Schmiedebergs Arch Pharmacoll977;297:245-50. 18. Ricordi C. Shah SD, Lacy PE, Clutter WE, Cryer PE. Delayed extra-adrenal epinephrine secretion after bilateral adrenalectomy in rats. Am J Physioll988;254:E52-3. 19. Axelrod J. Purification and properties of phenylethanolamine N-methyltransferase. J Lab Clin Med 1981;98:527-35. 20. Pendleton RG, Gessner G. Sawyer J. Studies on the distribution of phenylethanolamine N-methyltransferase. Res Commun Chem Path01 Pharmacol1978:21:315-25. 21. Saavedra JM, Coyle JT. Axelrod J. The distribution and properties of the non-specific N-methyltransferase in brain. J Neurothem 1973;20:743-52. 22. Baldessarini RJ, Kopin IJ. S-adenosylmethionine in brain and other tissues. J Neurochem 1966:13:769-77. 23. Ziegler MG, Kennedy B, Elayan H. A sensitive radioenzymatic assay for epinephrine-forming enzymes. Life Sci 1989;43:211722. 24. Ziegler MG, Woodson LC, Kennedy B. Radioenzymatic assay of catecholamines, metabolites and related enzymes. In: Krstulovic AM, ed. Quantitative analysis of catecholamines and related compounds. New York: Wiley, 1986:263-80. 25. Axelrod J. Weil-Malherbe H, Tomchick R. The physiological disposition of 3H-epinephrine and its metabolite metanephrine. J Pharmacol Exp Ther 1959;127:251-6. 26. Sheridan MA. Effects of epinephrine and norepinephrine on lipid mobilization from coho salmon liver incubated in vitro. Endocrinology 1987;120:2234-9. 27. Nichols AJ, Ruffolo PR Jr. Evaluation of the alpha and beta adrenoceptor-mediated activities of the novel, orally active inotropic agent, ibopamine. in the cardiovascular system of the pithed rat: comparison with epinine and dopamine. J Pharmacol Exp Ther 1987:242:455-63. 28. Janssens PA, Lowry P. Hormonal regulation of hepatic glycogenolysis in the carp, Cyprinis carpio. Am J Physiol 1987;252(4, part 2]:R653-60. 29. Arnall DA. Marker JC, Conlee RK. Winder WW. Effect of infusing epinephrine on liver and muscle glycogenolysis during exercise in rats. Am J Physioll986:250(6, part l]:E641-9. 30. Marker JC, Arnall DA. Conlee RK, Winder WW. Effect of adrenodemedullation on metabolic responses to high intensity exercise. Am J Physiol1986;251(3, part l):R552-9.
Received March 13.1989. Accepted May 19,1989. Address requests for: Michael G. Ziegler, M.D., LJCSD Medical Center, H-781-B. San Diego, California 92103. This work was supported by grant HL 35924 from the National Institutes of Health.
1990;98:152-155
Epinephrine Synthesis by an N-Methyltransferase in Rat Liver HAMZEH ELAYAN, BRIAN KENNEDY, and MICHAEL G. ZIEGLER Department of Medicine, Division of Nephrology, University of California San Diego Medical Center, San Diego, California
We investigated if liver can synthesize epinephrine in vitro and in vivo. Homogenates of rat liver readily synthesized [‘Hlepinephrine from [3H]S-adenosylmethionine and norepinephrine. Liver homogenates also N-methylated dopamine at more than twice the rate that they N-methylated norepinephrine. In contrast, adrenal homogenates, which AJ-methylate norepinephrine to form epinephrine using the enzyme phenylethanolamine-N-methyltransferase (PNMT), methylated dopamine only about 1% as well as norepinephrine. Synthesis of epinephrine by liver homogenates was not significantly inhibited by the PNMT inhibitor SKF 29661 at a concentration that inhibited adrenal homogenate epinephrine synthesis by nearly 99%. These findings indicate that liver can synthesize epinephrine in vitro using an enzyme other than PNMT.. Adrenal demedullation of rats reduced plasma epinephrine levels to 7% of control values, but left liver epinephrine and epinephrineforming enzyme levels unchanged. Treatment of demedullated rats with 6-hydroxydopamine plus reserpine also resulted in dramatically reduced plasma epinephrine levels but no change in hepatic epinephrine and N-methylating enzyme levels. We conclude that the liver synthesizes its own epinephrine.
E
pinephrine (E) has a profound effect on liver function. Its most important action is an increase in glucose output (l-3).This is primarily due to enhanced hepatic glycogen breakdown into glucose (Z-5). Epinephrine-induced increases in glycogenolysis are mediated in part by elevations in adenylate cyclase activity (3,6), cyclic adenosine monophosphate (2.77, and phosphorylase activity (2,7-9), and decreases in glycogen synthase activity (3.74). Epinephrine increases liver RNA and protein synthesis (111,and decreases hepatocyte triglyceride and triglycerol re-
lease (12), carbamyl phosphate synthetase synthesis (13), and leucine incorporation in hepatocytes (14). During stress or hypoglycemia sympathetic nerves cause release of E into the bloodstream from the adrenal. However, the adrenal is not the only source of circulating E. It has been found in the plasma and urine of bilaterally adrenalectomized humans (15,161 and adrenodemedullated rats (17,18). The source of this E is unknown but the enzyme phenylethanolamine-N-methyltransferase (PNMT), which synthesizes E from norepinephrine (NE) in the adrenal (19), is present in several other organs including the brain, heart, and lung (20).A nonspecific N-methyltransferase (NMT) that could potentially synthesize E from NE is also present in several organs including the liver (20,21). The liver contains NE and the methyl donor S-adenosylmethionine in its sympathetic nerves (22). Thus, the liver contains all the reactants necessary to synthesize E. Prior assays for NMT activity have used substrates such as phenylethanolamine. Although these assays had the advantage of speed and sensitivity, they could not measure actual E formation nor evaluate whether NMTs other than PNMT might synthesize E. We used a new assay (23)that measures conversion of NE into E or dopamine (DA] into epinine. This assay, combined with measures of liver E in animals with low circulating E levels, provides a reasonable guide to hepatic E synthesis. Materials and Methods Male Sprague-Dawley rats weighing 180-200 g were subjected to either bilateral adrenal demedullation (20 rats) Abbreviations used in this paper: DA, dopamine; E, epinephrine; NE, norepinephrine; NMT, N-methyltransferase; PNMT. phenylethanolamine-N-methyltransferase. 0 1990 by the American Gastroenterological Association 0018-5085/80/$3.00
January 1990
EPINEPHRINE
or sham operation (9 rats) under ether anesthesia. The rats were allowed to recover for 1 wk, then each of 9 bilaterally demedullated rats received one dose of 6-hydroxydopamine [Sigma Chemical Co, St. Louis, MO.), 20 mg/kg i.p., followed 24 h later by four doses on successive days of reserpine (Sigma), 5 mg/kg i.p. dissolved in 10% ascorbic acid in 0.9% saline. This treatment was designed to lower E in the circulation by demedullation, destroy sympathetic nerve terminals with 6-hydroxydopamine, and eliminate catecholamine storage in any nerve terminals that survived 6hydroxydopamine treatment. Rats treated with demedullation, 6-hydroxydopamine and reserpine are referred to as D-6-R rats. We thus established three groups of rats for experimentation. Nine rats that underwent sham adrenal medullectomy constitute the control group. Eleven rats underwent adrenal medullectomy and are referred to as AMX rats. Nine rats underwent D-6-R treatment. The data shown in Figures 1 and 2 come from the control animals, and data in Figures 3-5 from all three groups of rats. All rats were killed by decapitation 24 h after the last dose of reserpine was given and a blood sample was collected from each rat into a heparinized tube. Decapitation is a potent stimulus to adrenal E release, so trunk blood was collected after decapitation and E levels were measured to determine the adequacy of adrenal demedullation. Liver samples weighing about 500 mg were separated, washed with ice-cold saline, and stored at -70%. The brainstem, adrenals, and salivary glands were separated from the sham-operated rats, washed with ice-cold saline, and stored at -70°C. Tissues were homogenized in 1 ml of 0.1 M tris-HCl buffer with 0.1% Triton (pH 7) then diluted to form 100 mg/ml, and centrifuged for 10 min at 5800 g. The supernatant was stored at -70°C. Values are expressed per gram wet weight of liver tissue. Catecholamines were assayed in duplicate lOO-~1samples of tissue homogenates or plasma according to the catechol-omethyltransferase-based radioenzymatic method of Ziegler
LIVER
SYNTHESIS
ADRENAL
BY RAT LIVER
153
BRAINSTEM
Figure 2. Rate of N-methylation of DA by homogenates of liver, adrenal, and brainstem. Either NE or DA were present in the incubation mix at lo-’ M. Values are expressed as percentage of rate of N-methylation of NE, and are the mean + SEM of four to five determinations. The three tissues differ from each other by analysis of variance, p ~0.001. kp to.05 vs. liver by Duncan’s test.
al. (24). Liver N-methylating enzyme activities were assayed in 50-J duplicates of tissue homogenates according to the radioenzymatic method of Ziegler et al. (23). using 10e3 M NE as substrate in the presence or absence of the PNMT inhibitor SKF 29661 at 10m4M (kindly supplied by Dr. Paul Hieble, Smith Kline Beckman Corp.). Norepinephrine and DA, both at the concentration of 10e4 M, were used as substrates in certain determinations, and the ratios of epinine formed from DA to E formed from NE were determined. Liver N-methylating enzyme was partially characterized by comparing the effect of SKF 29661 and the DA/NE ratio of the liver homogenates with the corresponding values of the adrenal glands, brainstem, and salivary glands. et
300 60
50
e," 0
B a
20
10 2
50
K
*
*
a
> I
LIVER
ADRENAL
0I 30
SHAM
AMX
D6R
BRAINSTEM
Figure 1. Percentage of E-forming activity remaining in the presence of lo-’ M SICP29661 of homogenates from liver, adrenal, and brainstem. Norepinephrine was present in the incubation mix at 10m3M. Values are expressed as percentage of activity in the absence of inhibitor and are the mean + SEM of four to live determinations. The three tissues differ from each other by analysis of variance, p
Figure 9. The rate of E formation from NE by liver homogenates horn three groups of rats. Sham-operated (SHAM) animals served as the control group. Adrenal demedullated (AMX) animals had their adrenal medulIae removed, and D-6-R at&n& also received 6-hydroxydopamine and reset-pine. There was no significant difference in the rate of E formation among groups. Values are shown as picomoles of E formed per gram of liver per hour and are the mean * SEM of 9-11 rats.
154
ELAYAN
ET AL.
GASTROENTEROLOGY
BOO0
I-
8000 7000
t
/
z
5000..
g
4000.-
zooo-1000-OT
*
Discussion
E
DA
Figure 4. Plasma catecholamine levels in sham-operated (SHAM), adrenal demedullated (AMX), and demedullated rats that also received &hydroxydopamine and reserpine (D6Rj Values are expressed in picograms per milliliter and are the mean + SEM of S-11 rats. Treatment groups are signiikantly different from each other by analysis of variance with respect to NE (p c 0.0011,DA (p < 0.05j and E (p < 0.001). ti ~0.05 vs. SHAM by Duncan’s test. 0, SHAM; R$AMX; @ D6R.
The results were expressed as mean values +SEM. Differences among groups were determined by analysis of variance and Duncan’s test. Results Liver homogenates readily synthesized [3H]E from NE and [3H]SAM. This N-methylating activity was not reduced by the PNMT inhibitor SKF 29661 at 1W4 M (Figure 1). In contrast, the E-forming activity of adrenal homogenate was inhibited by 88% using this concentration of SKF 29661 (Figure 1). The liver enzyme methylated DA more than twice as well as 2
0
NE
DA
NE, whereas adrenal and brainstem homogenates methylated DA only about 1% as well as NE in control animals (Figure 2). Adrenal demedullation and D-6-R treatment tended to increase hepatic N-methylating activity (Figure 3). Adrenal demedullation reduced plasma E levels to 7% of control values (p < 0.05). D-6-R treatment also significantly reduced plasma E [p < 0.05) (Figure 4). In contrast, adrenal demedullation and D-6-R treatment did not reduce liver E levels from those observed in sham-operated controls (Figure 5).
* l!iiikA I
3000.~
NE
Vol. 98, No. 1
E
Figure 5. Liver catecholamine levels in sham-operated I(SHAW adrenal demedullated (AMX), and demedullated rats tlhat also received &hydroxydopamine and reserpine (DORj Values are expressed in nanograms per gram tissue and are the mean + SEM of 6-11 rats. Treatment groups are significantly different from each other by analysis of variance with respect to NE (p -C0.001). *p < 0.05 vs. SHAM by Duncan’s test. 0, SHAM; RI,AMX; ?? , D6R.
Liver homogenates methylated NE to form E. Earlier investigators reported that liver homogenates could N-methylate phenylethanolamine (20,21). However, the liver N-methylating enzyme does not appear to be PNMT. It was not inhibited by the PNMT inhibitor SKF 29661 at concentrations that greatly inhibit adrenal PNMT. Also the liver enzyme methylated the phenylethylamine DA more than twice as well as the phenylethanolamine NE. In contrast, adrenal PNMT methylates DA only about one onehundredth as well as NE. Other investigators have also provided evidence that the dominant liver Nmethylating enzyme is not PNMT (20,21). It is a less specific N-methyltransferase that can methylate both NE and DA. The liver N-methylating enzyme appears to synthesize E in vivo. After adrenal demedullation, plasma E levels were reduced to 7% of control values. In contrast, liver E was unaffected by removal of the adrenal medullae. Compared with organs such as the heart and spleen, the liver accumulates E from the bloodstream with only moderate efficiency (25). Axelrod et al. [25) found that 2 h after a [3H]E infusion, only 40% of the E taken up by the liver remained unmetabolized. Thus, the E that we find in liver 11 days after adrenal medullectomy is unlikely to be of adrenal origin. Epinephrine levels in the liver were also unchanged in D-6-R rats, despite the fact that liver NE was lower and N-methylating activity was not significantly increased. This suggests that in normal rats nearly all liver E is locally synthesized. Liver E does not appear to be stored in sympathetic nerves. D-6-R treatment was designed to destroy sympathetic nerves with 6-hydroxydopamine and prevent catecholamine storage in any remaining nerve terminals with reserpine. The D-6-R treatment effectively lowered hepatic NE, but failed to lower hepatic E (Figure 5). Epinephrine differs from NE in its effects on liver function. Epinephrine has a greater effect on liver adenylate cyclase activity (6) and cyclic adenosine monophosphate levels (2) than NE. Norepinephrine increases glycerol and fatty acid release from liver
January 1990
whereas E does not (26). Thus, hepatic conversion of NE to E could have important effects on body chemistry. The liver N-methylating enzyme may also synthesize epinine from DA in vivo. Dopamine is present in the liver at about one-fifth of the level of NE, and liver homogenates readily N-methylate DA. The effects of epinine on hepatic function have not been studied. However, epinine stimulates CX-,p-, and DA-receptors and it differs in pharmacologic profile from DA (27). Liver NMT methylates DA to form epinine more readily than it methylates NE to form E (Figure 2). The amount of epinine and E formed in the liver will depend on the availability of DA and NE to serve as substrates. Most tissue NE is bound in sympathetic nerve vesicles so the amount of free NE available in hepatic tissue may be relatively low. Local E synthesis may underlie the maintenance of normal hepatic E levels while this organ is cultured in vitro. It may also provide the mechanism by which resting rats without adrenal medullae maintain normal liver glycogen and blood glucose levels (29,301. References I. Usami M, Seino Y. Taminato
T, et al. Effect of DG5128 on epinephrine and glucagon induced glucose output from the isolated perfused rat liver. Life Sci 1988;42:129-36. 2. Brighenti L. Puviani AC, Gavioli ME, Fabbri E, Ottolenghi C. Catecholamine effect on cyclic adenosine 3’:5’-monophosphate level in isolated catfish hepatocytes. Gen Comp Endocrinol 1987;68:216-23. 3. Exton JH. Mechanisms of hormonal regulation of hepatic glucose metabolism. Diabetes Metab Rev 1987;3:163-83. 4. Graham SM, Herring PA, Arinze IJ. Age-associated alterations in hepatic beta-adrenergic receptor/adenylate cyclase complex. Am J Physioll987;253:E277-82. 5. James AE, Burleigh KM, Kraegen EW. In vivo glucose metabolism in individual tissues of the rat. Interaction between epinephrine and insulin. J Biol Chem 1986;261:6366-74. 6. Sanae F. Matsunaga T, Miyamoto K, Koshivra R. Studies on responsiveness of hepatoma cells to catecholamines III. Differences between the receptor-adenylate cyclase regulating systems in AH130 cells and cultured normal rat liver cells. Jpn J Pharmacol1986;42:281-8. 7. Nemeth S, Viskupic E. Kvetnansky R, Kolena J. Effect of epinephrine infusions imitating plasma epinephrine levels in immobilization stress and of fasting on the activity of key enzymes in catfish liver and muscles. Physiol Bohemoslov 1986;61:469-75. 8. Nemeth S, Viskupic E. Dose-dependent relationships between the effects of epinephrine and glucose on the activity of key enzymes of glycogen metabolism in the rat liver. Physiol Bohemoslov 1987:36:X-20. 9. Nemeth S, Viskupic E, Kvetnansky R, Kolena J. Progressive weakening of the response of key enzymes of liver glycogen metabolism to permanently increased plasma epinephrine levels. Physiol Bohemoslov 1987;36:441-6. 10. Ottolenghi C, Puviani KC, Gavioli ME, Brighenti L. Epinephrine effect on glycogen phosphorylase activity in catfish liver and muscles. Gen Comp Endocrinoll986;61:469-75. 11. Breton L. Maissiat R. Clot JP, Baudry M. Effects of glucagon and epinephrine on the rat liver: oxidative phosphorylation and ultrastructure. C R Sot Biol [Paris] 1986;180:613-21.
EPINEPHRINE
SYNTHESIS BY RAT LIVER
155
12. Brindle NP. Ontko JA. Suppression of triglyceride secretion by epinephrine in isolated rat hepatocytes. Biochem Biophys Res Commun 1986;141:191-7. 13. Kitigawa Y. Hormonal regulation of carbamoyl-phosphate synthetase I synthesis in primary cultured hepatocytes and Reuber hepatoma H-35. Defective regulation in hepatoma cells. Eur J Biochem 1987;167:19-25. 14. Brostrom CO, Bocckino SB, Brostrom MA, Galuska EM. Regulation of protein synthesis in isolated hepatocytes by calcium mobilizing hormones. Mol Pharmacoll986;29:104-11. 15. Von Euler US, Ikkos D, Luft R. Adrenaline excretion during resting conditions and after insulin in adrenalectomized human subjects. Acta Endocrinol 1961;38:441-8. 16. Hollifield JW, Nadeau JJ. Vasodilator therapy. In: Hunter JC, ed. Hypertension update. Bloomfield, N.J.: Health Learning Systems, 1980:235-46. 17. Pendleton RG, Weiner G, Jenkins B, Gessner G. Effects of an inhibitor of phenylethanolamine N-methyltransferase upon stimulated adrenal catecholamine release and excretion in the rat. Naunyn Schmiedebergs Arch Pharmacoll977;297:245-50. 18. Ricordi C. Shah SD, Lacy PE, Clutter WE, Cryer PE. Delayed extra-adrenal epinephrine secretion after bilateral adrenalectomy in rats. Am J Physioll988;254:E52-3. 19. Axelrod J. Purification and properties of phenylethanolamine N-methyltransferase. J Lab Clin Med 1981;98:527-35. 20. Pendleton RG, Gessner G. Sawyer J. Studies on the distribution of phenylethanolamine N-methyltransferase. Res Commun Chem Path01 Pharmacol1978:21:315-25. 21. Saavedra JM, Coyle JT. Axelrod J. The distribution and properties of the non-specific N-methyltransferase in brain. J Neurothem 1973;20:743-52. 22. Baldessarini RJ, Kopin IJ. S-adenosylmethionine in brain and other tissues. J Neurochem 1966:13:769-77. 23. Ziegler MG, Kennedy B, Elayan H. A sensitive radioenzymatic assay for epinephrine-forming enzymes. Life Sci 1989;43:211722. 24. Ziegler MG, Woodson LC, Kennedy B. Radioenzymatic assay of catecholamines, metabolites and related enzymes. In: Krstulovic AM, ed. Quantitative analysis of catecholamines and related compounds. New York: Wiley, 1986:263-80. 25. Axelrod J. Weil-Malherbe H, Tomchick R. The physiological disposition of 3H-epinephrine and its metabolite metanephrine. J Pharmacol Exp Ther 1959;127:251-6. 26. Sheridan MA. Effects of epinephrine and norepinephrine on lipid mobilization from coho salmon liver incubated in vitro. Endocrinology 1987;120:2234-9. 27. Nichols AJ, Ruffolo PR Jr. Evaluation of the alpha and beta adrenoceptor-mediated activities of the novel, orally active inotropic agent, ibopamine. in the cardiovascular system of the pithed rat: comparison with epinine and dopamine. J Pharmacol Exp Ther 1987:242:455-63. 28. Janssens PA, Lowry P. Hormonal regulation of hepatic glycogenolysis in the carp, Cyprinis carpio. Am J Physiol 1987;252(4, part 2]:R653-60. 29. Arnall DA. Marker JC, Conlee RK. Winder WW. Effect of infusing epinephrine on liver and muscle glycogenolysis during exercise in rats. Am J Physioll986:250(6, part l]:E641-9. 30. Marker JC, Arnall DA. Conlee RK, Winder WW. Effect of adrenodemedullation on metabolic responses to high intensity exercise. Am J Physiol1986;251(3, part l):R552-9.
Received March 13.1989. Accepted May 19,1989. Address requests for: Michael G. Ziegler, M.D., LJCSD Medical Center, H-781-B. San Diego, California 92103. This work was supported by grant HL 35924 from the National Institutes of Health.