Hormonal Control of Rat Adrenal Phenylethanolamine N-Methyltransferase Enzyme Activity, the Final Critical Pathway Dona L. Wong, Ph.D., Brenda Siddall, BSc., and Wei Wang, M.S. To examine whether glucocorticoids control rat adrenal but returned corticosterone to hypophysectomized levels phenylethanolamine N-methyltransferase (PN MT) and decreased PNMT mRNA to 50% of normal. Despite through gene transcription, the effects of hypophysectomy the decline in PNMT mRNA and its partial utilization and acute and chronic glucocorticoid replacement on for protein synthesis, PNMT enzymatic activity was fully PNMT mRNA and enzymatic activity were detennined. restored. These findings indicate that glucocorticoids Glucocorticoid depletion through hypophysectomy did not exert marked but complex influences on PNMT gene alter PNMT mRNA, whereas PNMT activity declined to transcription. In addition, corticosteroids appear to rv25% of normal. A single dose of ACTH (4 JU SC) posttranscriptionally regulate PNMT protein expression, rapidly induced PNMT mRNA, with a six-fold peak at 6 underscoring the uncoupling between the expression of hours postinjection. The short-term rise in PNMT PNMT mRNA and active enzyme. Thus, glucocorticoid mRNA was accompanied by an increase in corticosterone control of gene transcription and protein synthesis do not and elevated levels of glucocorticoid receptor mRNA. fully account for changes in PNMT expression, Ribosomal loading experiments suggested that available consistent with the previous observation that PNMT mRNA was fully utilized for protein synthesis. glucocorticoid control of PNMT proteolysis is also However, PNMT activity did not increase commensurately. important in PNMT regulation and the potential for Chronic ACTH treatment (4 JU SC daily for 7 days) epinephrine biosynthesis. [Neuropsychopharmacology sustained elevated levels of glucocorticoid receptor mRNA 13:223-234, 1995] KEY WORDS: Phenylethanolamine N-methyltransferase; nephrine, is a critical determinant of epinephrine ex­ Rat adrenal gland; Honnonal control; Gene expression; pression during acute and chronic stress. In response Enzyme activity to stress, large quantities of glucocorticoids and epi­ The catecholamine biosynthetic enzyme phenylethanol­ nephrine are released from the adrenal gland. Further, amine N-methyltransferase (PNMT, E.C. 2.1.1.28), the glucocorticoids have been demonstrated to affect the final enzyme in the pathway converting tyrosine to epi- expression of PNMT, and it has been suggested that the primary source of this control occurs through PNMT gene expression (Evinger et al. 1992). However, the hormones regulate PNMT still From the Nancy Pritzker Laboratory of Developmental and Mo­ mechanism(s) by which lecular Neurobiology, Department of Psychiatry and Behavioral remains a subject of much controversy. Sciences, Stanford University School of Medicine, Stanford, Cali- Early studies in the adult rat showed that depletion fornia. ' of glucocorticoids through hypophysectomy markedly Address correspondence to: Dona L. Wong, Ph.D., Nancy Pritz­ ker Laboratory of Developmental and Molecular Neurobiology, MSLS reduced PNMT enzymatic activity to ~20% of normal Bldg., Rm P-106, Department of Psychiatry and Behavioral Sciences, values (Ciaranello et al. 1975; Wong et al. 1985; Evinger Stanford University School of Medicine, Stanford, CA 94305-5485. et al. 1992). Basal levels of the enzyme could be restored Received November 29, 1994; revised March 15, 1995; accepted March 28, 1995. by adrenocorticotropin (ACTH) or glucocorticoid ad- NEUROPSYCHOPHARMACOLOGY 1995-VOL. 13, NO. 3 © 1995 American College of Neuropsychopharmacology Published by Elsevier Science Inc. 0893-133X/95/$9 .50 6"'5 Avenue· of the Americas, New York, l\Y 10010 SSDI 0893-133X(95)00066-M 224 D.L. Wong et al. NEUROPSYCHOPHARMACOLOGY 1995-VOL. 13, NO. 3 ministration. However, PNMT enzymatic activity could gene and endogenous PNMT gene expression, al­ never be elevated beyond normal through corticoste­ though dexamethasone, by itself, does not appear to roid manipulation (Ciaranello et al. 1975). Protein turn­ activate either the PNMT-reporter gene construct or the over studies indicated that hormones exerted their endogenous PNMT gene. Deletion mutation and site effects by controlling the rate of protein degradation directed mutagenesis studies demonstrated that Egr-1 without affecting protein synthesis (Ciaranello et al. activation is mediated through two Egr-1 consensus se­ 1975). In the presence of corticosteroids, levels of the quences, a perfect match sequence and an 8/9 match cosubstrate S-adenosylmethionine (SAM), the methyl sequence ( -165 bp and - 45 bp upstream of the site donor for the enzymatic reaction catalyzed by PNMT, of transcription initiation), whereas the glucocorticoid were sustained, and the binding of SAM to PNMT pro­ response is mediated through the GRE (-513 bp) in tected the enzyme against proteolytic degradation the 5' upstream PNMT promoter/regulatory region. (Ciaranello 1978; Berenbeim et al. 1979; Wong et al. It is also evident from in vivo studies that PNMT 1982; Wong et al. 1985). transcriptional activity is likely uncoupled from PNMT More recently it has been demonstrated that PNMT translational activity. For example, when intact rats gene expression is also hormonally regulated. Whereas were administered the type II specific glucocorticoid the effects of hypophysectomy on PNMT mRNA have agonist, RU28362, or the mixed type I and type II glu­ been variable, no change (Wong et al. unpublished; cocorticoid agonist, dexamethasone, PNMT enzymatic Evinger et al. 1992) or marked depletion (Jiang et al. activity declined, falling to 70% of normal levels at a 1989), corticosteroids have consistently been shown to dose of 30 µg daily of dexamethasone or 100 µg daily restore (Jiang et al. 1989) and/or elevate PNMT mRNA of RU28362 (Wong et al. 1992). Whereas higher doses beyond normal levels (Wong et al. 1992; Evinger et al. of RU28362 sustained the attenuation of PNMT, higher 1992) in the adult rat. doses of dexamethasone restored PNMT activity to nor­ Glucocorticoid control of PNMT transcription is mal. PNMT mRNA showed a markedly different pat­ consistent with the presence of a glucocorticoid response tern. Whereas low doses of RU28362 (100 µg daily) did element (GRE) in the 5' upstream promoter/regulatory not change PNMT mRNA levels, high doses (~100 µg sequences of the PNMT gene (Baetge et al. 1988; Kaneda daily) elevated mRNA levels as much as eleven-fold et al. 1988; Batter et al. 1988; Ross et al. 1990; Morita (1,000 µg daily). A similar pattern was observed with et al. 1992). Furthermore, the rat GRE appears func­ dexamethasone except that the rise in PNMT mRNA tional based on the induction of chloramphenicol occurred at a lower dose (30 µg daily) and the incre­ acetyltransferase (CAT) reporter gene activity from a ment in mRNA expression was much greater ( rv2Q fold PNMT-CAT reporter gene construct transiently ex­ at 1,000 µg daily). pressed in bovine chromaf:fm cells (Ross et al. 1990). The present studies were undertaken to further ex­ However, the induction of reporter gene expression by amine the apparent independent effects of glucocorti­ dexamethasone occurred only in cells expressing the coids on PNMT gene and protein expression using a endogenous PNMT gene and required very high corti­ paradigm of glucocorticoid depletion and restoration costeroid concentrations (20 µM). and to determine the relative importance of these mech­ This observation, coupled with evidence demon­ anisms in controlling the critical biologic endpoint, strating that corticosteroids are unable to initiate PNMT PNMT enzymatic activity. Our findings confirm the un­ expression prematurely in the developing rat adrenal coupling of glucocorticoid control of PNMT mRNA and gland or cultured pheochromoblasts (Bohn et al. 1981; protein expression. In response to glucocorticoid deple­ Michelson and Anderson 1992), suggested that gluco­ tion through hypophysectomy, PNMT mRNA expres­ corticoids might be incapable of independently activat­ sion was not significantly reduced, whereas PNMT en­ ing the PNMT gene. Rather, a transcriptional factor, zymatic activity markedly declined. Acute corticosteroid other than the glucocorticoid receptor (GR), might be replacement rapidly elevated PNMT mRNA well be­ necessary to initiate PNMT gene expression, a factor yond normal levels without a concommitant elevation with which the activated GR could then interact to facili­ in PNMT enzymatic activity. In contrast, chronic glu­ tate PNMT gene transcription. cocorticoid replacement therapy restored PNMT enzy­ Consistent with this hypothesis, we have shown matic activity to normal levels although PNMT mRNA recently using transient cotransfection assays that the was reduced to half of normal values. Although PNMT immediate early gene transcriptional protein, Egr-1 mRNA appeared fully and comparably utilized for pro­ (Sukhatme et al. 1988), activates both a PNMT pro­ tein translation in the case of acute corticosteroid moter-luciferase reporter gene construct and the endog­ replacement, it did not appear to be fully utilized in the enous PNMT gene in a PC12 derivative cell line, RSl case of chronic replacement therapy. Clearly, PNMT (Ebert et al. 1994). Further, dexamethasone (1 µM) in­ gene transcription alone is not the limiting event in duces an additional two-fold increment in both reporter PNMT enzymatic expression. NEUROPSYCHOPHARMACOLOGY 1995-VOL. 13, NO. 3 Hormonal Control of