Plasma Metanephrines Are Markers of Pheochromocytoma Produced ...

24 downloads 2468 Views 763KB Size Report
free metanephrines in patients with pheochromocytoma are derived from metabolism before .... Upper reference limits of normal (URLs) for plasma concentra-.
0021-972X/98/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1998 by The Endocrine Society

Vol. 83, No. 6 Printed in U.S.A.

Plasma Metanephrines Are Markers of Pheochromocytoma Produced by Catechol-OMethyltransferase Within Tumors GRAEME EISENHOFER, HARRY KEISER, PETER FRIBERG, EVA MEZEY, THANH-TRUC HUYNH, BHARGAVA HIREMAGALUR, TODD ELLINGSON, SUSHIL DUDDEMPUDI, AGNES EIJSBOUTS, AND JACQUES W. M. LENDERS Clinical Neuroscience Branch, National Institute of Neurological Disorders and Stroke (G.E., E.M., T.-T.H., B.H., T.E., S.D.) and Hypertension Endocrine Branch, National Heart Lung and Blood Institute (H.K.), National Institutes of Health, Bethesda, Maryland 20892-1424; Department of Clinical Physiology, University of Go¨teborg, Go¨teborg, Sweden (P.F.); and Departments of Rheumatology (A.E.) and Internal Medicine (J.W.M.L.), St. Radboud University Hospital, Nijmegen, The Netherlands ABSTRACT This study examined whether the high sensitivity of plasma free metanephrines for diagnosis of pheochromocytoma may result from production of free metanephrines within tumors. Presence in pheochromocytomas of catechol-O-methyltransferase (COMT), the enzyme responsible for conversion of catecholamines to metanephrines, was confirmed by Western blot analysis, enzyme assay, and immunohistochemistry. Western blot analysis and enzyme assay indicated that membrane-bound and not soluble COMT was the predominant form of the enzyme in pheochromocytoma. Immunohistochemistry revealed colocalization of COMT in the same chromaffin cells where catecholamines are translocated into storage vesicles by the vesicular monoamine transporter. Levels of free metanephrines in pheochromocytoma over 10,000 times higher than plasma concentrations in the same patients before removal of tumors indicated production of

P

HEOCHROMOCYTOMAS are tumors of chromaffin cells most commonly arising from the adrenal medulla and usually presenting with hypertension. Unequivocal diagnosis of the tumor typically requires confirmation by several tests. Perhaps the most important diagnostic test is biochemical evidence of excessive catecholamine production by the tumor. This is usually achieved using measurements of catecholamines in urine or plasma (1–3) or of certain catecholamine metabolites in urine (3–5). However, none of these commonly used tests reliably confirms or excludes the presence of a tumor in every patient with pheochromocytoma (6). A new test with more promise than other tests involves measurements of free metanephrines (normetanephrine and metanephrine) in plasma (7). These O-methylated metabolites of norepinephrine and epinephrine are produced by the enzyme, catechol-O-methyltransferase (COMT) and are rapidly metabolized to sulfate conjugates by the enzyme, monoamine-preferring phenolsulfotransferase. The resulting sulfate conjugates are present in plasma and urine in conReceived January 22, 1998. Revision received February 23, 1998. Accepted March 2, 1998. Address all correspondence and requests for reprints to: Graeme Eisenhofer, Building 10, Room 4D20, National Institutes of Health 10 Center Drive, Bethesda, Maryland 20892-1424. E-mail: [email protected].

metanephrines within tumors. Comparisons of the production of metanephrines in patients with pheochromocytoma with production from catecholamines released or infused into the circulation indicated that more than 93% of the consistently elevated levels of circulating free metanephrines in patients with pheochromocytoma are derived from metabolism before and not after release of catecholamines into the circulation. The data indicate that the elevated plasma levels of free metanephrines in patients with pheochromocytoma are derived from catecholamines produced and metabolized within tumors. Some tumors do not secrete catecholamines, but all appear to metabolize catecholamines to free metanephrines, thus explaining the better sensitivity of plasma free metanephrines over other tests for diagnosis of pheochromocytoma. (J Clin Endocrinol Metab 83: 2175–2185, 1998)

centrations more than 25-fold higher than those of the free metanephrines (8). Assays of urine metanephrines typically employ a deconjugation step so that sulfate-conjugated metanephrines comprise the bulk of these measurements. Thus, these commonly used assays of urinary metanephrines are largely measurements of metabolites (i.e. sulfate-conjugated derivatives) that are different from those of the free metanephrines (7). In a study of over 50 patients with pheochromocytoma, every patient had elevated plasma concentrations of free unconjugated metanephrines, showing that this test reliably excludes the presence of a tumor, whereas measurements of plasma catecholamines or urine metanephrines do not (9). A question arising from this study is why measurements of free metanephrines should provide a more superior method for diagnosis of pheochromocytoma than measurements of the parent amines or other metabolites? Our hypothesis for the superiority of plasma free metanephrines over other tests is that in patients with pheochromocytoma, free metanephrines are produced mainly from metabolism of catecholamines within the tumor, not from catecholamines metabolized after their secretion into the circulation. Because some tumors do not secrete catecholamines or secrete them episodically (10 –13), ongoing metabolism of catecholamines to free metanephrines within tumors would explain the bet-

2175

2176

EISENHOFER ET AL.

ter sensitivity of free metanephrines compared with the parent amines and other metabolites for diagnosis of the tumor. In the present study, the hypothesis outlined above was examined by several procedures designed to assess the source of plasma metanephrines in patients with pheochromocytoma. The presence of COMT in tumor tissue was examined using Western blot analysis and immunohistochemical localization of the enzyme. The capacity of pheochromocytomas to produce free metanephrines was further examined from measurements of COMT enzyme activity and the presence of O-methylated amine metabolites in tumor tissue. Finally, comparisons of the production of metanephrines from the parent amines in patients with pheochromocytoma vs. production in other subjects receiving insulin tolerance tests or infusions of radiolabeled catecholamines enabled estimation of amounts of metanephrines derived from metabolism of circulating catecholamines, as distinct from those derived from metabolism of catecholamines within tumors. Materials and Methods Subjects The patient database for this report includes 82 patients with histologically proven pheochromocytomas, half of whom were studied at the National Institutes of Health (Bethesda, MD), and the rest of whom were examined at several other centers within the United States or in The Netherlands (St. Radboud University Hospital, Nijmegen) or Sweden (Sahlgrenska Hospital, University of Go¨teborg, Go¨teborg). Other data from 52 of these patients have been reported previously in a study that examined the sensitivity and specificity of plasma metanephrines in the diagnosis of pheochromocytoma (9). The present study examines the source of the elevated plasma metanephrines in patients with pheochromocytoma. Blood samples were obtained from all patients, and samples of tumor tissue were obtained at surgery from 34 patients. Additional data includes the results of glucagon stimulation tests or multiple blood sampling in selected patients. For comparison purposes, the data in this report are also derived from subject groups without pheochromocytoma (healthy normotensives and patients with angina pectoris or congestive heart failure) who received iv infusions of 3H-labeled catecholamines or bolus injections of insulin. Other subjects provided samples of normal liver or adrenal medulla during surgical procedures for conditions other than pheochromocytoma. Upper reference limits of normal (URLs) for plasma concentrations of metanephrines and catecholamines are the same as those established in our previous report (9). All studies were approved by the appropriate institutional review boards, and all patients and subjects gave their informed consent to participate in studies.

Insulin tolerance tests Insulin tolerance tests were carried out in seven healthy volunteers (two males and five females; mean 6 sd; age, 51.9 6 8.5 yr) to examine plasma metanephrine responses to release of catecholamines from the adrenals. These studies were carried out between 0800 and 1100 h after an overnight fast with subjects supine. A catheter inserted into an antecubital vein was used for administration of insulin and sampling of blood. Thirty minutes later, insulin was administered as an iv bolus injection at a dose of 0.1 U/kg. Blood samples (10 mL) were obtained immediately before and at 30, 45, 120, and 180 min after the injection of insulin.

Infusions of 3H-labeled catecholamines Infusions of 3H-labeled norepinephrine and epinephrine were carried out in 13 patients with angina pectoris, 25 patients with congestive heart failure, and 10 normal control subjects to examine production of meta-

JCE & M • 1998 Vol 83 • No 6

nephrines from circulating catecholamines. All subjects were supine during infusion of radiotracers. The radiotracers (L-2,5,6-[3H]norepinephrine, 40 – 60 Ci/mmol; L-N-methyl-[3H]epinephrine, 65–75 Ci/ mmol; New England Nuclear Corp., Boston, MA) were infused simultaneously into a forearm vein at 1.0 –1.5 mCi/min. Blood samples were taken at intervals between 15 and 146 min after the start of infusions.

Glucagon stimulation tests Plasma concentrations of metanephrines and catecholamines were examined in six pheochromocytoma patients with a positive test result to glucagon (i.e. a more than 3-fold increase in plasma norepinephrine) and one patient with a negative test result but a large increase in plasma epinephrine. For this test, patients received a 1-mg iv bolus dose of glucagon. Venous blood was sampled in the supine position immediately before the bolus of glucagon and at 2 min after the bolus. In the patient with the negative test result but the large increase in plasma epinephrine, blood was sampled at 5 min before and immediately before the glucagon and at intervals of 1, 2, 3, 4, 5, 10, and 20 min after glucagon. Two other blood samples were taken from this patient on 2 separate days preceding the day of the glucagon stimulation test.

Patient with paroxysmal hypertension Blood samples were taken from an antecubital vein on three occasions in one 42-yr-old female patient with a pheochromocytoma and paroxysmal attacks of flushing, palpitations, and chest pain. On one occasion a blood sample was taken and measurements of blood pressure and heart rate were recorded when the patient was free of the above symptoms. On a second occasion blood samples and hemodynamic measurements were taken during a paroxysmal attack lasting 10 min. On a third occasion a blood sample was taken and hemodynamic measurements recorded within 10 min after the end of a paroxysmal attack.

Collection of blood and tissue samples All blood samples were transferred into tubes containing an anticoagulant (heparin or EDTA) and immediately placed on ice until centrifuged (4 C) to separate the plasma. Plasma samples were stored at 280 C until assayed for concentrations of catecholamines and metanephrines. Samples of pheochromocytoma tissue were obtained within 40 min of removal of tumors from 34 patients. For comparison purposes in assays of COMT enzyme activity, samples of normal liver were obtained from three patients who underwent partial hepatectomies. Samples of normal adrenal tissue were also obtained from another patient who underwent an adrenalectomy for a small adrenal cortical adenoma. Samples of excised tissue were placed on ice immediately after removal, extraneous tissue removed, and small 50- to 400-mg pieces were dissected apart and placed on dry ice before storage at 280 C.

Plasma and tissue catecholamines and metanephrines Plasma and tissue concentrations of catecholamines and metanephrines were quantified by liquid chromatography with electrochemical detection. Tissue concentrations were determined in weighed samples of tissue that were homogenized in at least 5 vol 0.4 m perchloric acid with 0.5 mm EDTA. Homogenized samples were centrifuged (1500 3 g for 15 min at 4 C), and supernatants were collected and stored at 280 C until assayed. Concentrations of catecholamines (norepinephrine and epinephrine) were determined after extraction from plasma or perchloric acid tissue supernatants using alumina absorption as described previously (14). Concentrations of metanephrines were determined using a different liquid chromatography procedure after extraction onto solid phase ion exchange columns (7). Concentrations of 3H-labeled norepinephrine, epinephrine, normetanephrine, and metanephrine were determined by scintillation counting of mobile phase eluants collected from the outlet of the electrochemical cell after separation by liquid chromatography. Intraassay coefficients of variation (CVs) were 1.9% for norepinephrine, 3.0% for epinephrine, 4.2% for normetanephrine, and 3.3% for metanephrine. Interassay CVs were 6.5% for norepinephrine, 11.4% for epinephrine, 6.5% for normetanephrine, and 5.7% for metanephrine.

METANEPHRINES IN PHEOCHROMOCYTOMA Western blot analysis of COMT Antibody to human COMT, raised in guinea pig against purified recombinant (Escherichia coli) soluble-COMT was kindly provided as a gift by Dr. Ismo Ulmanen (Orion Corp., Orion Pharma, Espoo, Finland). The antibody is highly selective for human COMT and does not crossreact with rat COMT (our unpublished data). Samples of human pheochromocytoma, adrenal medulla and liver tissue were homogenized in 2% SDS buffer (50 mm Tris, pH 6.8, 2.5% glycerol, 2% mercaptoethanol), boiled immediately for 10 min at 100 C, and then chilled for 5 min on ice. Samples of tissue preparations, with equivalent amounts of proteins (15 mg), were resolved on a 12% SDS-polyacrylamide gel and transferred to nitrocellulose membranes using a Bio-Rad transblot apparatus (BioRad, Hercules, CA). After transfer, nitrocellulose membranes were blocked in Tris buffered saline (50 mm Tris, pH 7.4, 0.9% NaCl, 0.05% Tween-20 and 5% BSA) for 1 h. Nitrocellulose membranes were then incubated for 1 h with antihuman COMT primary antibody at a 1:1000 dilution. Membranes were then washed three times for 5 min with Tris buffered saline and incubated for 45 min with secondary antibody (antiguinea pig IgG) conjugated to horseradish peroxidase at 1:20,000 dilution. Detection of the two bands corresponding to the 30-kDa membrane-bound and 24-kDa soluble species of COMT was achieved by the enhanced chemiluminescence method.

COMT enzyme assay Tissue samples (50 –700 mg) were homogenized in at least 4 vol phosphate buffer (50 mm sodium phosphate and 0.5 mm dithiothreitol, pH 7.5). Homogenates were centrifuged at 100,000 3 g for 30 min (4 C). Supernatant fractions were used for assays of soluble COMT activity, and the washed and resuspended pellets used for assays of membranebound COMT activity. The above method does not completely purify the two species of COMT (the supernatant fraction includes some of the larger membrane form of the enzyme), but does provide fractions enriched with either soluble or membrane-bound COMT. COMT activity was assessed by incubation (37 C for 30 min) of 100 mL tissue preparation in a total volume of 500 mL containing 50 mm sodium phosphate buffer (pH 7.5), 5 mm MgCl2, 2 mm S-adenosylmethionine, and 0.4 mm 3,4-dihydroxybenzylamine as substrate (Sigma Chemical Co., St. Louis, MO). The reaction was terminated by addition of 100 mL 2 m perchloric acid. Samples were then centrifuged at 1500 3 g for 15 min to separate the supernatant, which was used for measurements of the 3-methoxy-4-hydroxybenzylamine (MHBA) produced from 3,4-dihydroxybenzylamine. Amounts of MHBA in 10-mL samples of supernatant were determined by liquid chromatography with electrochemical detection. COMT activity was determined from the rate of production of MHBA and expressed as picomole of MHBA produced per minute per milligram of protein. These activities were determined under saturating conditions of substrate and therefore represent maximum velocity (Vmax) values. Division of these values by MichaelisMenten affinity constant (Km) values previously determined by Lotta and colleagues (15) for human membrane-bound COMT (24.1 mm) and human soluble COMT (369 mm) provided additional estimates of COMT activity. The intra- and interassay CVs were 4.7% and 16.6%, respectively.

2177

Mountain View, CA), were both diluted to 1:2000. The use of these antibodies in the immunohistochemical localization of VMAT-1 and VMAT-2 to endocrine cells and neurons has been documented previously (16). Tissue sections were first incubated overnight at 4 C in the guinea pig antibody to COMT and then incubated for 1 h at room temperature in the rabbit antibody to either VMAT-1 or VMAT-2. Sections were then incubated for consecutive 1-h periods at room temperature with a 1:2000 dilution of antiguinea pig IgG conjugated to CY3 fluorochrome followed by 1:100 dilution of antirabbit IgG conjugated to fluorescein isothiocyanate (Jackson Immunologicals, West Grove, PA). After application of secondary antibodies, sections were washed in distilled water, air dried, and coverslipped using Cytoseal 60 mounting medium (Stephens Scientific, Riverdale, NJ). Immunofluorescent labeling was viewed with a Leitz Dialux 20 fluorescence microscope (Leitz, Wetzlar, Germany) using appropriate filters. Negative controls included incubations in nonimmune mouse IgG or normal rabbit serum instead of the primary antibody or incubations without primary and/or the secondary antibodies.

Statistical analyses The distributions of plasma and tumor tissue concentrations of catecholamines and metanephrines in patients with pheochromocytoma were highly skewed. Normal distributions were obtained after logarithmic transformation of the data. Mean values for these variables are therefore provided as geometric means. Corresponding confidence intervals and standard errors were established from the normalized data. All other results for normally distributed data are expressed as arithmetic means 6 sem. Where data showed nonnormal distributions, statistical tests of significance were carried out on normalized data. These tests included paired t tests and ANOVA for repeated measures with post hoc tests carried out using Scheffe’s F test. Relationships between tissue concentrations of catecholamines and metanephrines or between tumor mass and plasma catecholamines or metanephrines were examined by linear regression analysis, significance established using Pearson’s correlation coefficient.

Results COMT immunoblots

Western blot analysis showed two forms of COMT corresponding to the larger 30-kDa membrane-bound form and the smaller 25-kDa soluble form of the enzyme (Fig. 1). Both soluble and membrane-bound COMT were detected consis-

Immunohistochemistry Localizations of COMT in different cells or the same cells where catecholamines are stored were examined by double staining of tissue sections with antibodies raised against COMT and against the vesicular monoamine transporters 1 and 2 (VMAT-1 and VMAT-2). Samples of pheochromocytoma tissue—frozen on dry ice soon after collection and stored at 280 C—were cut into 12-mm-thick sections using a Frigocut E 2800 cryostat (Reichert, Heidelberg, Germany). The frozen sections were mounted onto silanized slides and fixed in 4% paraformaldehyde fixative for 10 min, followed by multiple washes in PBS. To decrease nonspecific staining, the fixed sections were incubated at room temperature in PBS (pH 7.4) with 0.6% Triton X-100 and 1% BSA (diluent) for 30 min, and then washed in PBS. Primary antibody to COMT, raised in guinea pig (the same antibody used for Western blot analysis), was diluted 1:300 in the diluent described above. Primary antibodies to VMAT-1 and VMAT-2, raised in rabbits (Phoenix Pharmaceuticals,

FIG. 1. Western blot analysis of membrane-bound and soluble forms of COMT in samples from six pheochromocytomas (PHEO 1– 6), one sample of human liver, and one sample of normal human adrenal medulla (AD MD). Positions of membrane-bound COMT (MB) and soluble COMT (S) polypeptides are depicted on right and molecular mass standards on left. Amount of total protein used was 12 mg for each sample.

2178

JCE & M • 1998 Vol 83 • No 6

EISENHOFER ET AL.

tently in pheochromocytoma tissue, with the membranebound form of the enzyme showing a much greater abundance than the soluble form. In a single sample of normal adrenal medulla, COMT was almost totally present in the membrane-bound form. Soluble and membrane-bound COMT were also detected in human liver where the pattern was reversed with a considerably greater abundance of the soluble over the membrane-bound form of the enzyme. The membrane-bound form of COMT was considerably more abundant in both adrenal medulla and pheochromocytoma tissue than in liver. COMT enzyme activity

The pattern of distribution of membrane-bound and soluble COMT observed by Western blot analysis was mirrored by the results of COMT enzyme activity assays (Table 1). These assays showed higher (P , 0.05) Vmax values for enzyme activity of COMT in the preparation of pheochromocytoma tissue enriched with the membrane-bound form of the enzyme (particulate fraction) than in that enriched with the soluble form of the enzyme (supernatant fraction). In liver, the pattern was reversed with considerably lower enzyme activity in the particulate fraction than in the supernatant fraction. The difference in activities of membranebound vs. soluble COMT in pheochromocytoma tissue was considerably enhanced when activities were calculated by dividing Vmax values with respective Km values for membrane-bound and soluble forms of the enzyme. This additional analysis also suggested relatively higher activities of membrane-bound COMT in pheochromocytoma tissue than activities of both soluble and membrane-bound COMT in liver. COMT immunohistochemistry

Sections of pheochromocytoma tissue showed clear immunohistochemical staining with antibodies to COMT as well as antibodies to both the VMATs (VMAT-1 and VMAT-2) involved in translocation of biogenic amines into vesicular storage vesicles (Fig. 2). The morphology of the two tissue sections taken from the benign adrenal tumors shown in Fig. 2 illustrates an arrangement of stained cells in clusters or interconnecting cords, with a polygonal shape and large nuclei typical of pheochromocytoma chromaffin cells. Chromaffin cells stained for COMT in the cytoplasmic area surrounding the nuclei and all chromaffin cells that stained for COMT also stained for VMAT-1 or VMAT-2, indicating colocalization of COMT in the same cells that contained the VMAT responsible for translocation of catecholamines into storage vesicles.

Metanephrines in pheochromocytoma tissue

High concentrations of norepinephrine and epinephrine within the micromole per gram of tissue range were present in pheochromocytoma tissue (Table 2). Normetanephrine and metanephrine were also readily detected in samples of pheochromocytoma tissue. Tumor tissue showed a considerable range in concentrations of catecholamines, with those of epinephrine in particular showing the widest range. Separate analysis of tumors with either high and low concentrations of epinephrine (.1600 nmol/g and ,400 nmol/g) showed that tumors with high epinephrine concentrations also had high metanephrine concentrations, and those with low epinephrine concentrations had low metanephrine concentrations (Table 2). Corresponding with the above, there was a highly significant positive relationship (r 5 0.85, P , 0.0001) between tumor tissue concentrations of epinephrine and metanephrine (data not shown). There was also a positive relationship (r 5 0.58, P , 0.003) between tumor tissue concentrations of norepinephrine and normetanephrine (data not shown). Among the tumors, there were four that were taken from extraadrenal locations. These tumors had on average about 10-fold lower tissue concentrations of epinephrine and metanephrine than tumors located in the adrenal glands. Plasma metanephrines in relation to tumor concentrations of epinephrine and etanephrines and tumor size

Patients who had tumors with low epinephrine concentrations showed normal plasma concentrations of epinephrine (0.05– 0.54 pmol/mL) and also tended to show normal plasma metanephrine levels (0.21– 0.61 pmol/mL) before removal of tumors. Patients who had tumors with high concentrations of epinephrine all showed elevated plasma concentrations of metanephrine (1.13–16.52 pmol/mL) and also tended to show elevated plasma epinephrine levels (0.12– 4.80 pmol/mL). Although tumor tissue concentrations of metanephrines were much lower than those of the parent amines, they were nevertheless several orders of magnitude higher (P , 0.0001) than corresponding concentrations of metanephrines in the plasma of the same patients before removal of tumors (Fig. 3). Tumor tissue concentrations of normetanephrine averaged 23,000 times higher than the corresponding plasma concentrations and tissue concentrations of metanephrine averaged 13,000 times higher than plasma concentrations. In patients with pheochromocytomas, plasma concentrations of normetanephrine showed highly significant positive relationships with mass of tumors (r 5 0.66, P , 0.0001), whereas there was no relationship between tumor mass and

TABLE 1. COMT activity in tissue preparations enriched with membrane-bound (m-COMT) and soluble (s-COMT) forms of enzyme n

Pheochromocytoma Liver

6 3

Vmax values (pmolzminzmg)

Vmax/Km values (mlzminzmg)

m-COMT

s-COMT

m-COMT

s-COMT

16.9 6 4.8 2.0 6 1.5

7.1 6 0.6 66.3 6 12.4

798 6 216 84 6 63

21 6 2 180 6 33

Values represent means 6 SEM activities. Vmax values are calculated as picomoles MHBA produced per minute per milligram protein in COMT-enriched preparations. Vmax/Km values were estimated by dividing Vmax values by respective Km values previously estimated elsewhere (15) for human m-COMT (Km 5 24.1 mM) and human s-COMT (Km 5 369 mM), and represented in units of milliliters per minute per milligrams protein.

METANEPHRINES IN PHEOCHROMOCYTOMA

2179

FIG. 2. Immunohistochemical stainings of COMT and VMAT-1 (A, B, and C) and COMT and VMAT-2 (D, E, and F) in two tissue sections from an excised pheochromocytoma. A, B, and C, Same section in which COMT (A) is shown in red (CY3-labeled secondary antibody viewed through a rhodamine filter), VMAT-1 (B) is shown in green (fluorescein isothiocyanate-labeled secondary antibody viewed through a green filter), and both COMT and VMAT-1 (C) are shown colocalized in yellow/orange as a double-exposure overlay of two (A and B) images. Majority of cells are double labeled. D, E, and F, Similar immunostaining of another section for COMT (D), VMAT-2 (E), and overlapping image of both COMT and VMAT-2 (F). Similar to colocalization of COMT with VMAT-1, COMT is also present in majority of VMAT-2positive cells. Scale bar: 40 mm

plasma concentrations of norepinephrine (Fig. 4). Similarly, there was a positive relationship between tumor mass and plasma concentrations of metanephrine (r 5 0.46, P , 0.0005), but no relationship with plasma epinephrine (r 5 0.24) (data not shown). The regression line for the relationship between plasma normetanephrine and tumor size indicated that a patient with a small 10-g tumor might be expected to show a 3-fold increase in plasma concentrations of normetanephrine above the URL for normal compared with a 20-fold increase for a large 200-g tumor.

Plasma metanephrines during insulin tolerance tests

In normal subjects who received an iv bolus injection of insulin, plasma glucose levels fell from baseline concentrations of 4.8 6 0.1 mmol/L to 1.2 6 0.1 mmol/L at 30 min after insulin. Plasma concentrations of epinephrine increased (P , 0.0001) by 27-fold, and concentrations of norepinephrine increased (P , 0.0001) by 3-fold above baseline levels at 45 min after insulin administration (Fig. 5). In contrast to the large insulin-induced increases in plasma catecholamines, plasma concentrations of normetanephrine remained unchanged and plasma concentrations of metanephrine increased (P ,

2180

EISENHOFER ET AL.

JCE & M • 1998 Vol 83 • No 6

TABLE 2. Pheochromocytoma tissue concentrations of catecholamines and metanephrines in tumors with low and high concentrations of epinephrine

Norepinephrine Normetanephrine Epinephrine Metanephrine

High epinephrine (nmol/g) n 5 14

Low epinephrine (nmol/g) n 5 16

16,971 (10,864 –26,485) 124 (69 –223) 16,490 (8710 –31,189) 86 (49 –152)

7,567 (4,529 –12,647) 104 (64 –170) 61 (26 –143) 2 (1–3)

Values represent geometric means and 95% confidence intervals (in parentheses) of tissue concentrations (nanomoles per gram tissue). Tumors are divided into two groups with concentrations of epinephrine above 1600 nmol/g (high epinephrine) or below 400 nmol/g (low epinephrine).

FIG. 3. Comparison of concentrations of normetanephrine (A) and metanephrine (B) in pheochromocytoma tissue with concentrations in plasma from same subjects before removal of tumors. Vertical axi show concentrations of metanephrines in tissue (picomoles per gram) or plasma (picomoles per milliliter) as a logarithmic scale.

0.0001) by only 2.8-fold above baseline concentrations. The increase in absolute plasma concentrations of metanephrine after insulin was only 7.8 6 2.2% of the increase in plasma concentrations of epinephrine. Plasma metanephrines during iv infusions of 3H-labeled catecholamines

In normal control subjects and patients with angina or congestive heart failure who received iv infusions of 3Hlabeled catecholamines, plasma concentrations of [3H]nore-

FIG. 4. Relationships between mass of tumors and plasma concentrations normetanephrine (A) and norepinephrine (B) in same pheochromocytoma patients before removal of tumors. Both horizontal and vertical axi are shown in logarithmic scales. URLs for plasma concentrations of normetanephrine or norepinephrine are shown by horizontal dotted lines. Two patients had plasma normetanephrine concentrations below URLs of normal. Both, however, had elevated plasma concentrations of metanephrine. One of these patients has been described previously (9) and results for the other patient are shown in Table 3.

pinephrine and [3H]epinephrine reached steady state levels close to 1000 dpm/mL within 30 min after the start of radiotracer infusions (Fig. 6, A and C). In line with the rapid circulatory clearance and short half-lives of plasma metanephrines (17), steady state plasma levels of 3H-labeled metanephrines were also reached rapidly. However, steady state plasma concentrations of [3H]normetanephrine were only 3.3 6 0.1% those of plasma [3H]norepinephrine concentrations, whereas concentrations of [3H]metanephrine were only 7.7 6 0.4% of [3H]epinephrine concentrations. There were no differences among subject groups in the formation of 3H-labeled metanephrines from iv infused 3H-labeled

METANEPHRINES IN PHEOCHROMOCYTOMA

FIG. 5. Plasma concentrations (mean 6 SEM) of epinephrine and metanephrine (A) and of norepinephrine and normetanephrine (B) before and at intervals after iv bolus injection of insulin in seven normal subjects. Time of injection of insulin is denoted by dotted vertical line and arrow. E, epinephrine; MN, metanephrine; NE, norepinephrine; NMN, normetanephrine. *, Denotes a significantly higher concentration than that before insulin and at 180 min after insulin.

catecholamines as reflected by steady state ratios of plasma 3 H-labeled metanephrines to catecholamines. Plasma metanephrines during glucagon stimulation

In six pheochromocytoma patients who showed a positive glucagon-stimulation test result for the presence of a tumor (.3-fold increase in plasma norepinephrine), plasma norepinephrine and epinephrine concentrations both increased (P , 0.005) by an average of 7-fold above baseline values (Fig. 7, A and C). Despite these large increases in plasma catecholamines, plasma concentrations of normetanephrine and metanephrine showed little change and were not consistently elevated by glucagon (Fig. 7, B and D). In one patient with a pheochromocytoma in whom blood was sampled over 3 days and at repeated intervals before and for 20 min after glucagon, there was no increase in plasma norepinephrine after glucagon (negative test result), but there was a large 10-fold increase in plasma epinephrine peaking at 2 min after glucagon (Fig. 7, F and G). The 10-fold increase in plasma epinephrine was asso-

2181

FIG. 6. Comparison of formation metanephrines from iv infused 3 H-labeled catecholamines in subjects without pheochromocytoma (A and C) with formation of metanephrines from catecholamines in patients with pheochromocytoma (B and D). Steady state plasma levels of [3H]normetanephrine ([3H]NMN) formed from iv infused [3H]norepinephrine ([3H]NE) were only 3.3 6 0.1% those of plasma [3H]norepinephrine concentrations (A) and steady state plasma levels of [3H]metanephrine ([3H]MN) formed from iv infused [3H]epinephrine ([3H]E) were only 7.7 6 0.4% those of plasma [3H]epinephrine concentrations (C). Dashed horizontal arrows show small amounts of plasma normetanephrine (NMN) or metanephrine (MN) expected to be derived from metabolism of norepinephrine (NE) or epinephrine (E) released into circulation of patients with pheochromocytoma (solid portions of bars). This assumes similar formation of metanephrines from circulating catecholamines as shown in A and C for subjects without pheochromocytoma (see Fig. 7 and Table 3 for validity). Large amounts of plasma normetanephrine or metanephrine not accounted for by metabolism of norepinephrine or epinephrine released into circulation are shown by percent values in solid blocks bounded by vertical arrows. These amounts represent proportions of normetanephrine or metanephrine derived from metabolism of norepinephrine (93.9%) or epinephrine (97.7%) before and not after release of catecholamines into circulation of patients with pheochromocytoma (i.e. presumably amounts derived from metabolism within tumors).

ciated with a transient small (,20% above baseline) increase in plasma metanephrine peaking at 2–3 min after glucagon. This nonsymptomatic patient, with multiple endocrine neoplasia type II syndrome, was diagnosed by chance with an adrenal pheochromocytoma following routine work-up for another condition. The patient repeatedly had normal plasma concentrations of norepinephrine on the 1st and 3rd days of blood sampling and normal baseline concentrations of epinephrine on all 3 days of blood sampling. Despite these normal plasma levels of cat-

2182

JCE & M • 1998 Vol 83 • No 6

EISENHOFER ET AL.

Plasma metanephrines in a pheochromocytoma patient with a hypertensive paroxysm

In one pheochromocytoma patient who suffered from paroxysmal attacks of flushing, palpitations, and chest pain, plasma concentrations of catecholamines increased substantially during a paroxysmal attack, whereas plasma concentrations of free metanephrines increased by only a small amount in comparison (Table 3). In particular, plasma concentrations of epinephrine increased by 83-fold, whereas plasma concentrations of free metanephrine increased by only 3.6-fold. The absolute increase in plasma concentrations of free metanephrine was only 5% of the increase in plasma epinephrine, and the increase in free normetanephrine was less than 1% of that of norepinephrine. Production of metanephrines in tumors and from metabolism of circulating catecholamines in patients with pheochromocytoma

FIG. 7. Plasma concentrations of norepinephrine (NE), epinephrine (E), normetanephrine (NMN), and metanephrine (MN) in individual pheochromocytoma patients before (BL) and after bolus iv injection of glucagon (GLU). URLs of normal are shown by horizontal dotted lines. Plasma concentrations of norepinephrine (A), normetanephrine (B), epinephrine (C), and metanephrine (D) in six patients (each patient shown with a different symbol) with positive glucagon stimulation tests (i.e. .3-fold increases in norepinephrine) showed consistently large increases in plasma catecholamines compared with absent or much smaller increases in plasma metanephrines (note: logarithmic scale of vertical axi). In another patient with a negative glucagon stimulation test, characterized by no increase in plasma norepinephrine (F), there was a large transient 10-fold increase in plasma epinephrine (G). Increase in epinephrine peaked at 2 min and ended by 10 min after glucagon and was accompanied by a smaller transient increase in plasma metanephrine. This patient showed normal plasma concentrations of catecholamines, but consistently elevated plasma concentrations of normetanephrine and metanephrine on 3 separate days before glucagon stimulation test.

echolamines, plasma concentrations of both normetanephrine and metanephrine were consistently elevated well above the URLs of normal at all time points and, unlike plasma catecholamines, showed little day-to-day variation.

In patients with pheochromocytoma, plasma concentrations of normetanephrine were 53% those of plasma concentrations of norepinephrine, and in patients with elevated epinephrine or metanephrine concentrations, plasma concentrations of metanephrine were 3.38-fold higher than epinephrine concentrations (Fig. 6, B and D). Comparison of the relative plasma concentrations of normetanephrine to norepinephrine in pheochromocytoma patients (0.53) with the relative increase in plasma [3H]normetanephrine to [3H]norepinephrine during iv infusion of radiolabeled catecholamines (0.033) indicated that only 6.1% (0.033/0.53) of the normetanephrine in plasma of pheochromocytoma patients was derived from metabolism of norepinephrine after its release into the circulation (Fig. 6, A and B). Similarly, comparison of the relative plasma concentrations of metanephrine to epinephrine in pheochromocytoma patients (3.38) with the relative increase in plasma [3H]metanephrine to [3H]epinephrine during iv infusion of radiolabeled catecholamines (0.077) indicated that only 2.3% (0.077/3.38) of the elevated metanephrine in plasma of pheochromocytoma patients was derived from metabolism of epinephrine after its release into the circulation (Fig. 6, C and D). Most of the elevated plasma concentrations of normetanephrine (93.9%) and metanephrine (97.7%) in patients with pheochromocytoma were derived from metabolism of norepinephrine and epinephrine in tissue before entry into the circulation, not from metabolism after entry of precursor catecholamines into the circulation. Discussion

This study establishes that pheochromocytomas contain considerable amounts of the membrane-bound form of COMT, resulting in local metabolism of catecholamines to free metanephrines. The large increases in plasma concentrations of free metanephrines in patients with pheochromocytoma are thus almost entirely caused by metabolism of catecholamines within tumors, with only small amounts produced from catecholamines released into the circulation. COMT exists in two forms: one form is present as a soluble species in the cytoplasm, the other as a larger molecular

METANEPHRINES IN PHEOCHROMOCYTOMA TABLE 3. Hemodynamics and plasma concentrations of catecholamines and metanephrines in a pheochromocytoma patient with paroxysmal attacks of flushing, palpitations, and chest pain

Hemodynamics Blood pressure (mmHg) Heart rate (beats/min) Plasma catecholamines Norepinephrine (pmol/mL) Epinephrine (pmol/mL) Plasma metanephrines Normetanephrine (pmol/mL) Metanephrine (pmol/mL)

No symptoms

During paroxysm

After paroxysm

125/75 68

180/92 92

125/70 72

4.79a 0.31

22.84a 25.86a

4.08a 1.98a

0.55 0.54a

0.61 1.92a

0.63 0.89a

Values are shown for an individual pheochromocytoma patient on three occasions. First, when there were no symptoms present; second, during a paroxysmal attack of flushing, palpitations, and chest pain; and third, within 10 min after end of a paroxysmal attack. a Denotes a plasma concentration above URL of normal.

weight species bound to membranes (18, 19). Both are coded for by the same gene, but differ in the presence or absence of a peptide tail that anchors the larger species to membranes. Variations in the presence of the two forms of COMT among tissues are dependent on transcriptional regulation by two promoters as well as posttranscriptional regulation at the translation step (20). Because the soluble form of COMT is by far the more abundant species in most tissues, particularly in the liver and kidneys (18), it was long considered the enzyme responsible for catecholamine metabolism. Other considerations now suggest that this notion may be incorrect (19). In particular, Km values for catecholamines of the membranebound form of COMT are much lower than those for the soluble form (15–25 vs. 200 –370 mm), making the membranebound enzyme most likely responsible for metabolism of the low cytoplasmic concentrations of catecholamines found in vivo (15). Additionally, compared with the soluble form of COMT, the membrane-bound enzyme has a more than 10fold higher selectivity for 3-O- than 4-O-methylation of catecholamines (15). Because the O-methylated derivatives of catecholamines are almost exclusively present in vivo as 3-methoxy metabolites, it seems likely that their formation reflects selective 3-O-methylation by the membrane-bound form of COMT. The above suggestion that the membrane-bound form of COMT is the enzyme species largely responsible for metabolism of catecholamines in vivo is supported by the results of the present study in addition to those from some of our previous work (8, 21). This other work showed that over 90% of circulating metanephrine and between 23% and 40% of circulating normetanephrine are produced by metabolism of the parent catecholamines within the adrenal glands (8, 21). This makes the adrenals the single largest tissue source of circulating metanephrines in the body, with production surpassing by far that of metanephrines by the liver (21). The present findings that COMT in pheochromocytoma and adrenal medulla tissue exists predominantly in the membranebound form makes chromaffin tissue distinct from most other tissues, such as the liver, where the soluble form is the most predominant form of the enzyme (18, 19). The high rates of production of metanephrines by the adrenals together with the abundance of membrane-bound COMT within

2183

chromaffin tissue is consistent with the suggestion that it is the membrane-bound form of the enzyme that is responsible for in vivo metabolism of catecholamines (19). Although Western blot and Vmax values from COMT enzyme assays both indicated higher amounts of membranebound than soluble COMT in pheochromocytoma tissue, this difference appeared more pronounced by Western blot analysis. This likely reflects two factors: first, in enzyme assays, although the supernatant fraction is enriched with the smaller soluble form of COMT, it also contains some of the larger species of COMT in unbound form, and, second, enzyme activities are determined under saturating conditions of substrate favoring the soluble species of COMT, which has a higher Vmax than the membrane-bound form (15, 19). As described by Roth (19), a more realistic method to compare likely relative activities in vivo of membrane-bound and soluble forms of COMT can be achieved by dividing Vmax values by the corresponding Km values for each form of the enzyme. Doing this by using the Km values of recombinant human soluble and membrane-bound COMT determined by Lotta et al. (15) yields activities of the membranebound form of COMT in pheochromocytoma tissue many fold higher than activities of the soluble form. This additional analysis also revealed higher activity of membrane-bound COMT in pheochromocytoma tissue than both soluble and membrane-bound COMT activity in liver. VMAT-1 is preferentially localized to large dense core vesicles where it is involved in the translocation and packaging of biogenic amines into secretory granules (22). Although VMAT-1 is predominantly located within endocrine cells, such as chromaffin cells of the adrenal medulla, these cells also often contain the neuronal VMAT-2 (16). Immunohistochemical colocalization of COMT with both VMAT-1 and VMAT-2 indicates the presence of the metabolizing enzyme within the same chromaffin cells where catecholamines are stored. This colocalization may seem surprising, but is not without precedence. In sympathetic neurons, the catecholamine- metabolizing and mitochondrial-bound enzyme, monoamine oxidase, is responsible for most norepinephrine turnover (23). The majority of this is secondary to norepinephrine leaking from vesicular stores into the axoplasm where the transmitter is available for metabolism by monoamine oxidase (23, 24). Presumably this high rate of intraneuronal metabolism serves to safeguard against high toxic axoplasmic concentrations of catecholamines. The considerable intraneuronal metabolism of catecholamines leaking from storage vesicles also serves as a gearing mechanism to minimize the requirement for increases in the activity of the rate-limiting enzyme in catecholamine synthesis, tyrosine hydroxylase, to match increases in exocytotic transmitter release (24). Perhaps similar reasons apply in the adrenal medulla where metabolism to metanephrine almost equals epinephrine release (21). Immunohistochemical staining of COMT in the cytoplasm of pheochromocytoma chromaffin cells is in agreement with recent findings on the intracellular localization of membranebound COMT to the rough endoplasmic reticulum (25). Perhaps the presence of COMT bound to these structures in chromaffin cells may provide distinct locations within the cytoplasm where catecholamines are O-methylated. This

2184

EISENHOFER ET AL.

may serve to compartmentalize intracellular sites of catecholamine metabolism away from those of synthesis and storage. Apart from the presence in tumor tissue of membranebound COMT, the capacity of COMT in pheochromocytomas to metabolize catecholamines is also reflected by the high concentrations of metanephrines in tumor tissue. Although these concentrations are less than 2% those of the parent amines, it must be recognized that catecholamines are sequestered in chromaffin granules at concentrations 104–105 times higher than those in the cytoplasm (26), where they are available for metabolism by COMT. The high tissue levels of metanephrines in pheochromocytomas are in agreement with previous findings (27–29). However, the additional findings of the present study that these levels are over 10,000 times higher than plasma concentrations of metanephrines illustrate how these high tissue levels may easily make a major contribution to the elevated levels of plasma metanephrines found in patients with pheochromocytoma. Estimation of the proportion of circulating metanephrines produced from metabolism of catecholamines released into the circulation can be achieved by examination of the production of metanephrines from iv infused catecholamines or from catecholamines released directly into the circulation by the adrenals (30). The results of the insulin tolerance tests showed that plasma metanephrines are relatively insensitive to large increases in adrenal release of catecholamines, with the increase in plasma metanephrine only 7.8% that of the increase in plasma epinephrine. This result is remarkably close to the findings during iv infusions of 3H-labeled catecholamines, where the increase in plasma [3H]metanephrine was 7.7% of the increase in [3H]epinephrine. Assuming similar metabolism of circulating catecholamines in patients with pheochromocytoma (an assumption dealt with in the paragraph below), if all the metanephrine in plasma was derived from epinephrine released into the circulation, then plasma concentrations of metanephrine would be expected to be close to 8% of those of epinephrine. Instead, the elevated plasma concentrations of metanephrine in patients with pheochromocytoma were over 200% higher than concentrations of epinephrine. A similar situation was observed for normetanephrine indicating that only small proportions of both plasma-free normetanephrine (,7%) and metanephrine (,3%) are derived from metabolism of catecholamines after their release into the circulation from pheochromocytomas. The independence of metanephrine production on metabolism of catecholamines released into the circulation from tumors was further illustrated by the relative lack of increase in plasma metanephrines after glucagon in patients with a positive glucagon stimulation test characterized by large increases in plasma catecholamines. Similarly, in one patient with an episodically active tumor, plasma concentrations of norepinephrine increased by nearly 5-fold and epinephrine by more than 80-fold during a paroxysmal hypertensive attack. Despite these large increases, plasma concentrations of free metanephrine increased by only 6% of the increase in epinephrine, and plasma concentrations of free normetanephrine remained stable. The relative increases in plasma free metanephrines after increased catecholamine release into the circulation in patients with pheochromocytoma were

JCE & M • 1998 Vol 83 • No 6

slight and thus similar to those in normal subjects during insulin tolerance tests and iv infusions of catecholamines. This indicates little difference in metabolism of circulating catecholamines that could account for the high plasma concentrations of free metanephrines relative to catecholamines in patients with pheochromocytoma. Thus, in patients with pheochromocytoma, more than 93% of the free normetanephrine and 97% of the free metanephrine in plasma are derived from metabolism of catecholamines before release into the circulation, almost all of this representing metabolism within tumors. A similar conclusion to the above was reached by Crout and Sjoerdsma (29) many years ago when they observed that the urinary excretion of catecholamines and their metabolites in patients with pheochromocytoma showed a completely different profile from that in subjects receiving iv infusions of catecholamines. These investigators also showed that tumor size was not a determinant of catecholamine excretion, but did influence the excretion of catecholamine metabolites. Similarly, in the present study highly significant positive relationships between tumor size and plasma concentrations of free metanephrines contrasted with complete lack of relationships between tumor size and plasma concentrations of catecholamines. These findings provide further evidence for the independence of elevations in plasma levels of metanephrines on catecholamine release from pheochromocytomas. More importantly, the findings also show how the magnitude of increases in plasma free metanephrines above normal can provide information about tumor size, whereas elevations in catecholamines do not. Although the above analysis establishes that elevated circulating levels of free metanephrines in patients with pheochromocytoma are largely produced independently of catecholamine release from tumors, this does not in itself explain why plasma free metanephrines should provide a better diagnostic test for a tumor than the parent amines themselves. A plausible explanation, however, follows from other findings that a proportion of patients with pheochromocytoma harbor tumors that are nonfunctional or quiescent and do not secrete catecholamines or secrete catecholamines episodically (10 –13). In these patients, plasma and urinary catecholamines typically yield false negative tests. In contrast, plasma free metanephrines (normetanephrine or metanephrine) appear to be much more consistently elevated (9), indicating that even when tumors are not secreting catecholamines, they are actively metabolizing catecholamines to the free metanephrines. An example of a catecholamine producing and metabolizing, yet nonsecreting tumor, is illustrated by the present results for the patient shown in Fig. 7, F and G. This patient, with multiple endocrine neoplasia type II syndrome, was only diagnosed with a tumor after a routine computed tomography (CT) scan revealed an adrenal lesion. Plasma catecholamines were normal or near normal in every sample, whereas plasma concentrations of normetanephrine and metanephrine were both consistently elevated well above the URLs of normal in each of 11 separate samples taken over the course of 3 separate days. Clearly this patient had a tumor that produced and metabolized catecholamines to metanephrines, but did not secrete

METANEPHRINES IN PHEOCHROMOCYTOMA

catecholamines into the bloodstream in amounts sufficient to cause a positive test result. In conclusion, the present study shows that pheochromocytomas have an unusually higher content of the membranebound than the soluble form of COMT. However, it appears that the membrane-bound and not the soluble form of COMT is the enzyme species responsible for metabolism of catecholamines in vivo. The high content of membrane-bound COMT in the same tumor cells where catecholamines are synthesized and stored leads to considerable production of free metanephrines within pheochromocytomas. Thus, the elevated plasma levels of free metanephrines in patients with pheochromocytoma are produced independently of catecholamine release by tumors. Because some tumors are quiescent or nonfunctional and do not readily or continually secrete catecholamines, measurements of free metanephrines provide a more reliable marker for the presence of a pheochromocytoma than do the parent amines. Acknowledgments We are grateful Ms. Courtney Holmes, Ms. Anneli Ambring, Mr. Jacques J. Willemson, and Mr. Douglas Hooper for expert technical assistance; and Dr. Albert Verhofstad, Ms. Joan Folio, RN, and Ms. Janet Ward Sakol, RN, MSN for assistance in obtaining patient samples and information from patient records. Thanks are also extended to the many clinicians and their patients who participated in this study.

References 1. Bravo EL, Tarazi RC, Gifford RW, Stewart BH. 1979 Circulating and urinary catecholamines in pheochromocytoma. Diagnostic and pathophysiologic implications. N Engl J Med. 301:682– 686. 2. Duncan MW, Compton P, Lazarus L, Smythe GA. 1988 Measurement of norepinephrine and 3,4-dihydroxyphenylglycol in urine and plasma for the diagnosis of pheochromocytoma. N Engl J Med. 319:136 –142. 3. Manger WM, Gifford Jr RW. 1993 Pheochromocytoma: current diagnosis and management. Cleve Clin J Med. 60:365–378. 4. Manu P, Runge LA. 1984 Biochemical screening for pheochromocytoma. Superiority of urinary metanephrines measurements. Am J Epidemiol. 120:788 –790. 5. Heron E, Chatellier G, Billaud E, Foos E, Plouin PF. 1996 The urinary metanephrine-to-creatinine ratio for the diagnosis of pheochromocytoma. Ann Intern Med. 125:300 –303. 6. Bravo EL. 1994 Evolving concepts in the pathophysiology, diagnosis, and treatment of pheochromocytoma. Endocr Rev. 15:356 –368. 7. Lenders JWM, Eisenhofer G, Armando I, Keiser HR, Goldstein DS, Kopin IJ. 1993 Determination of plasma metanephrines by liquid chromatography with electrochemical detection. Clin Chem. 39:97–103. 8. Eisenhofer G, Friberg P, Pacak K, et al. 1995 Plasma metadrenalines: do they provide useful information about sympatho-adrenal function and catecholamine metabolism? Clin Sci. 88:533–542.

2185

9. Lenders JW, Keiser HR, Goldstein DS, et al. 1995 Plasma metanephrines in the diagnosis of pheochromocytoma. Ann Intern Med. 123:101–109. 10. Manger WM, Gifford RW. 1996 Clinical and experimental pheochromocytoma. Cambridge: Blackwell Science. pp 333–365. 11. Sinclair D, Shenkin A, Lorimer AR. 1991 Normal catecholamine production in a patient with a paroxysmally secreting phaeochromocytoma. Ann Clin Biochem. 28:417– 419. 12. Stewart MF, Reed P, Weinkove C, Moriarty KJ, Ralston AJ. 1993 Biochemical diagnosis of phaeochromocytoma: two instructive case reports. J Clin Pathol. 46:280 –282. 13. Shawar L, Svec F. 1996 Pheochromocytoma with elevated metanephrines as the only biochemical finding. J La State Med Soc. 148:535–538. 14. Eisenhofer G, Goldstein DS, Stull R, et al. 1986 Simultaneous liquid-chromatographic determination of 3,4-dihydroxyphenylglycol, catecholamines, and 3,4-dihydroxyphenylalanine in plasma, and their responses to inhibition of monoamine oxidase. Clin Chem. 32:2030 –2033. 15. Lotta T, Vidgren J, Tilgmann C, et al. 1995 Kinetics of human soluble and membrane-bound catechol O-methyltransferase: a revised mechanism and description of the thermolabile variant of the enzyme. Biochemistry. 34:4202– 4210. 16. Weihe E, Schafer MK, Erickson JD, Eiden LE. 1994 Localization of vesicular monoamine transporter isoforms (VMAT1 and VMAT2) to endocrine cells and neurons in rat. J Mol Neurosci. 5:149 –164. 17. Eisenhofer G. 1994 Plasma normetanephrine for examination of extraneuronal uptake and metabolism of noradrenaline in rats. Naunyn Schmiedebergs Arch Pharmacol. 349:259 –269. 18. Guldberg HC, Marsden CA. 1975 Catechol-O-methyl transferase: pharmacological aspects and physiological role. Pharmacol Rev. 27:135–206. 19. Roth JA. 1992 Membrane-bound catechol-O-methyltransferase: a reevaluation of its role in the O-methylation of the catecholamine neurotransmitters. Rev Physiol Biochem Pharmacol. 120:1–29. 20. Tenhunen J, Salminen M, Lundstrom K, Kiviluoto T, Savolainen R, Ulmanen I. 1994 Genomic organization of the human catechol O-methyltransferase gene and its expression from two distinct promoters. Eur J Biochem. 223:1049 –1059. 21. Eisenhofer G, Rundqvist B, Aneman A, et al. 1995 Regional release and removal of catecholamines and extraneuronal metabolism to metanephrines. J Clin Endocrinol Metab. 80:3009 –3017. 22. Liu Y, Schweitzer ES, Nirenberg MJ, Pickel VM, Evans CJ, Edwards RH. 1994 Preferential localization of a vesicular monoamine transporter to dense core vesicles in PC12 cells. J Cell Biol. 127:1419 –1433. 23. Eisenhofer G, Friberg P, Rundqvist B, et al. 1996 Cardiac sympathetic nerve function in congestive heart failure. Circulation. 93:1667–1676. 24. Eisenhofer G, Rundqvist B, Friberg P. 1998 Determinants of cardiac tyrosine hydroxylase activity during exercise-induced sympathetic activation in humans. Am J Physiol. 274:R626 –R634. 25. Ulmanen I, Pera¨nen J, Tenhunen J, et al. 1997 Expression and intracellular localization of catechol O-methyltransferase in transfected mammalian cells. Eur J Biochem. 243:452– 459. 26. Schuldiner S, Shirvan A, Linial M. 1995 Vesicular neurotransmitter transporters: from bacteria to humans. Physiol Rev. 75:369 –392. 27. Sjoerdsma A, King WM, Leeper LC, Udenfriend S. 1957 Demonstration of the 3-methoxy analog of norepinephrine in man. Science. 127:876. 28. Kopin IJ, Axelrod J. 1960 Presence of 3-methoxy-4-hydroxyphenylglycol and metanephrine in phaechromocytoma tissue. Nature. 185–788. 29. Crout JR, Sjoerdsma A. 1964 Turnover and metabolism of catecholamines in patients with pheochromocytoma. J Clin Invest. 43:94 –102. 30. Lenders JW, Kvetnansky R, Pacak K, Goldstein DS, Kopin IJ, Eisenhofer G. 1993 Extraneuronal metabolism of endogenous and exogenous norepinephrine and epinephrine in rats. J Pharmacol Exp Ther. 266:288 –293.