Sources of human urinary epinephrine

Kidney International, Vol. 51(1997), pp. 324—327

Sources of human urinary epinephrine MICHAEL G. ZIEGLER, Mo AUNG, and BRIAN KENNEDY Department of Medicine, Division of Nephrology, University of California San Diego, San Diego, California, USA

Sources of human urinary epinephrine. The kidney is a likely source for

some urinary epinephrine (E) since adrenalectomized animals and humans continue to excrete urinary E and the human kidney contains E synthesizing enzymes. We studied subjects during an intravenous infusion of 3H-E to determine the fraction of urinary E derived from the kidney. Eight normal subjects (CON) and 5 older, heavier hypertensives (OHH)

ate a light breakfast along with ascorbic acid supplementation and had intravenous and arterial lines placed. They received an infusion of 3H-E and had an oral water load. During the final hour of 3H-E infusion, urine

too low to alter blood pressure increase renal vascular resistance and renin release, and decrease sodium, potassium and chloride levels in urine [7]. Although the kidney can make E, the fraction of urinary E made by the kidney has not been measured. Several studies suggest that the fraction of urine E derived from nonad-

renal sources is substantial. Adrenal medullectomy failed to

3H-E infusion was abruptly discontinued, arterial blood samples were

decrease rat urinary E levels [8]. In another study, urinary E levels in the rat failed to decrease after removal of the adrenal medullae and bilateral renal denervation [4]. Bilaterally adrenalectomized

collected to measure 3H-E kinetics. The total body clearance of 3H-E was about 2,500 mI/mm from plasma and clearance of 3H-E to urine was about 170 mI/mm. CON had plasma E levels of 43 4 pg/mI. Their predicted rate of clearance of E from plasma to urine of 7,471 865 pg/mm was less than (P = 0.018) the actual urinary E excretion of 15,037 2,625 pg/mm. Thus, 43 9% of urinary E in CON was apparently derived from renal sources and not filtered from blood. Among OHH 85 4% of urinary E was derived from the kidney, significantly (P < 0.01) different from CON.

humans are reported to have normal urinary E levels [9]. We carried out a preliminary study of three subjects that suggested that urinary E may not simply be filtered from the bloodstream [31. Since the kidney seems a likely source for urinary E not derived from the adrenal, we measured the clearance of tritiated epinephrine (3H-E) into urine and actual urinary E excretion to estimate the amount of urinary E of renal origin in normal

plasma 3H-E clearance into urine (P = 0.03). A major fraction of urinary E is not filtered from the blood stream but is apparently derived from the kidney. A small fraction of urinary E may be derived from E stored in nerve endings along with norepinephrine, but this probably represents less than 2% of urinary E. Renal cleavage of E sulfate into E may be another

subjects and in a group of older, heavier hypertensives.

and arterial blood samples were collected for 3H-E and E levels. After the

The OHH also produced much more urinary E than predicted from

potential source of urinary E. Some, and perhaps most, urinary E not filtered from the bloodstream is derived from renal N-methylation of norepinephrine as the human kidney has two enzymes capable of convert-

ing norepinephrine to E. In conclusion, a major portion of urinary E is derived from the kidney and not filtered from the bloodstream. This is an important factor in the interpretation of urine E levels. Renal E could alter renal blood flow, electrolyte reabsorption, and renin release prior to excretion into urine.

Rat [1, 21 and human kidney [3] can synthesize epinephrine (E) in vitro. The kidney contains two N-methyltransferases that can convert norepinephrine to E [4]. One of these enzymes is phenylelhanolaminc-N-mcthyltransferase, the enzyme which syrithesizes E in the adrenal. PNMT is located in the renal glomeruli and tubules where urine is formed. Epinephrine sulfate (E-S04) is excreted by the kidney and renal sulfatase could convert E-S04 to E [5]. Epinephrine regulates many renal junctions. The kidney has a, a7, p, and adrenergic receptors that regulate renin release, vascular tone, and sodium absorption [6]. Epinephrine infusions

,

Received for publication March 5, 1996 and in revised form August 13, 1996 Accepted for publication August 18, 1996

© 1997 by the International Society of Nephrology

Methods

Subjects for the study were normal volunteers (CON) or patients who were older, heavier, and who gave a history of uncomplicated essential hypertension (OHH). Subjects first pro-

vided written informed consent for the study and then had a medical history, physical examination, electrocardiogram, complete blood count, and serum chemistries performed. Subjects were excluded if they had significant concomitant disease. Screening tests were normal in all subjects except for some who had elevated cholesterol, triglycerides, or left ventricular hypertrophy on electrocardiogram. The subjects took no medication for two weeks prior to the study and had no alcohol, tobacco, or caffeine on the day of the study. On the morning of the study, subjects were admitted to the Clinical Research Center and had a breakfast of a bread roll, a pat of margarine, a piece of fruit, a caffeine free beverage and took one gram of ascorbic acid. They had an

intravenous line placed for 3H-E administration and a radial artery catheter placed for blood sampling. The subjects received a water load of 1 liter followed by 250 ml per half-hour to increase urine flow.

lI-E with a specific activity of 50 Ci/mrnol was sterilized by passage through a Millipore filter and tested for sterility and pyrogenieily. The 3H-E was diluted in 250 ml of 5% dextrose with 50 mg ascorbic acid to serve as an antioxidant. The H-E infusion was begun at 120 j.rCi/hr for 10 minutes and then 60 pCi/hr for 170 minutes. Prior to the last hour of the infusion subjects voided and

that urine was discarded. Throughout the final hour of the infusion, arterial blood samples were collected for 3H-E and E levels. The subjects voided at the end of the final hour of infusion

324

325

Ziegler Ct al: Sources of human urinaiy epinephrine

and the urine volume was measured and an aliquot frozen. The 3H-E infusion was abruptly stopped and arterial blood samples were collected at 0.5, 1, 2, 4, 8, and 16 minutes after termination of the infusion. Plasma E levels increase in response to stress and hypoglycemia. In order to maintain constant plasma E levels, the subjects were fed a breakfast with complex carbohydrates and were kept in a single room on the Clinical Research Center without distractions. They had vascular catheters inserted three hours prior to blood and urine sampling and they remained supine and resting during the 3H-E infusion.

Table 1. Subject characteristics (mean

8

Age Ponderosity kg/si2 Systolic BP mm Hg Diastolic BP mm Hg Triglycerides mg%

5

32±3 24.6 120

75

59±5 29.9

0.5 2

156

36

Heart rate

69 3

Creatinine mg%

0.9

0.05

0.8 10

90 2 66 2

1

148

P<

OHH

CON

N

SEM)

290 53 1.2

0.05

NS NS

0.1

3H-E clearance to urine is calculated by the formula:

Assays

(3H-E m11) (ml urine min1) 3H-E was isolated from 1 ml samples of plasma and urine and urine clearance = 0.1 ml of infusate by alumina chromatography. Most catecholplasma H-E ml amine metabolites do not bind to alumina. We checked for acidic If we assume that 3H-E and E are handled alike, then E should and glycol metabolites that might interfere with the measurement be cleared from plasma into urine at the same rate as 3H-E. The of 3H-E and found none. Plasma and urine E levels were measured by a sensitive expected rate of appearance of E from plasma to urine can be radioenzymatic assay [10]. Plasma and urine E samples were calculated as:

extracted into a lipid solvent with a chelating agent to eliminate

expected E in urine = 3H-E urinaiy clearance plasma E

calcium and other agents which interfere with catecholamine assay in urine. The assay has a sensitivity of 6 pg/mI for E in The percent of urinary E derived from the kidney was calculated plasma. The coefficient of variation for measurement of E was as: 13% at the low levels found in plasma from resting subjects. Prior studies have demonstrated identical assay efficiency in urine and plasma [10].

Percent renal E =

Actual urine E -

expected urine ,

Actual unne E

E 100

Statistical analysis

Data analysis

Data were analyzed by two-tailed t-test. Non-normally distribWe analyzed the decay of 3H-E in plasma using the model of uted data were analyzed by the Wilcoxon signed ranks test. Bufano, Vaonag and Starcich [11] and Linares et al [12], who have Repeatedly sampled data were analyzed by ANOVA for repeated shown that 3H-NE kinetics followed a biexponential decay in measures. Data are expressed as mean SEM.

human plasma. We found that 3H-E kinetics fit well to the

Results

biexponential decay equation:

3H-E = (A

et) =

(B

e)

The 13 subjects were divided into two groups. The CON

subjects comprised S normal young men who were not obese and who had normotensive blood pressures and normal lab chemisThis equation describes the rate of decay of 3H-E in a two tries. The OHH subjects comprised 5 men nearly twice the age of compartment model. The volume of distribution of the first the CON group, who had significantly higher blood pressures, and who were heavier than the men in the CON group. The two compartment is then: groups had similar heart rates and serum creatinine levels (Table

where A and B are constants and ci and /3 are the rate constants.

Volume =

clearance a

and the half-life of 3H-E in the first compartment is: 0.693

1).

All subjects received a 3H-E infusion for three hours. Arterial 3H-E plasma levels remained stable during the final hour of the infusion while urine and blood samples were collected. At the end of three hours the 3H-E infusion was abruptly discontinued and

arterial plasma 3F1-E levels fell precipitously with a hiexponential decay pattern (Fig. 1). Total body clearance of 3H-E from arterial plasma was about 2,5 liter per minute in both groups of subjects. The rate of decay of 3H-E was fitted to a biexponential decay After the 3H-E infusion was discontinued, the rapid decay half-life formula by the program R-STRIP (Micro Math Scientific Softof 3H-E in CON was 1.5 0.3 minutes, which did not differ from ware, Salt Lake City, UT, USA). the OHH half-life of 0.9 0.1 minute. The volume of distribution We used the following formulae to calculate 3H-E plasma and for this rapid decay component of plasma 3H-E was 4.9 1.0 liter urinary clearances: for CON and 3.5 0.7 liter for OHH. Plasma 3H-E levels fell to about ¼ of plateau 3l-l-E levels four minutes after termination of 3H-E clearance from plasma is calculated by: the 3H-E infusion (Fig. 1). 3H-E inñised per minute Subjects had arterial blood sampled for endogenous E levels 3H-E clearance = four times during the final hour of 3H-E infusion and these plasma H-E t½

a

326

Ziegler et al: Sources of human urinaiy epinephrine

Table 2. E and 3H-E levels and clearance rates (mean

200 E Cl)

Arterial E pg/ml Arterial 3H-E clearance

C

43 2,495

4 123

27 2,669

P<

3 180

0.05

24

NS NS

NS

mI/mm

I

100

0 0

50

Lii

OHH

CON

150

SCM)

3H-E to urine clearance

174

13

172

mI/mm

C,,

Predicted arterial to urine

0

7,471

865

4,692

980

15,037

2625

39,783 85

12,991

0.05

4%

(LOl

E pg/mm

Actual urine Epg/min

% RenalE

I,I, 0

120

140

160

43

9%

180 196

Time, minutes Fig. 1. Levels of 3H-E in plasma from OHH and CON subjects. 3H-E was

infused for 180 minutes when the infusion was abruptly stopped and blood sampled at 180 plus 0.5, 2, 4, 8 and 16 minutes. Levels of 3H-E did not differ between OHH and CON by ANOVA for repeated measures.

undetectable to 1.4% of the amount of norepinephrine released [141. Following '4C-E injections, urinary 14C-E excretion in women fell to trace levels by 20 minutes [15]. Thus, it is likely that a very small amount of urinary E is derived from renal rioradren-

ergic nerves when the E is co-released with norepinephrine. However, this amount of E is probably less than 2% of norepinephrine levels, an amount too small to be detected under the

samples were assayed in duplicate. The mean of these eight conditions of this experiment. plasma E level measurements for each subject ranged from 16 to 54 pg/mi. Plasma E levels were slightly higher in CON than in OHH (Table 2) even though 3H-E clearance was similar in the two groups. The amount of 3H-E excreted into urine was much smaller than total body 3H-E disappearance. 3H-E clearance to

The kidney can synthesize E from the methylation of norepinephrine and may be able to derive E from the action of sulfatase

approximates normal renal glomerular filtration rate. Making the assumption that arterial E is handled by the kidney the same as arterial 3H-E permits calculation of the amount of E that should be excreted into urine from arterial blood. In CON this calculated rate of 7471 865 pglml was significantly lower than the actual excretion rate of 15,037 2625 pg/ml (P = 0.018 Wilcoxon). The difference between predicted and measured excretion of arterial E was even more extreme in OHH so that only 15% of urine E in OHH was calculated to have come from arterial blood, significantly less than actual urine E excretion (P = 0.029, Wilcoxon test, Table 2). Although 3H-E clearance into urine was similar between CON and OHH, plasma E levels were somewhat higher in CON and urine E excretion was higher in OHH. As a consequence, a greater fraction of urinary E was calculated to be derived from the kidney in OI-IH than in CON.

plasma dopamine sulfate and norepinephrine sulfate levels exceed the levels of dopamine and of norepinephrine in plasma. Plasma E-S04 levels are much lower than dopamine sulfate and norepinephrine sulfate but similar to plasma E levels [16, 171. Plasma levels of E-S04 could theoretically produce the amount of urine E

on E-SO4. In humans, catecholamines can be inactivated by sulfation to catecholamine sulfates. In the rat, dopamine sulfate can be converted to a dopamine in the kidney by sulfatase. The urine was about 170 mI/mm in both groups, a number which human kidney may be able to convert E-SO4 to E. In humans,

Discussion A major fraction of urinary E is derived from some source other

than the clearance of plasma E. Potential sources of this E in urine are adrenal venous E, E previously taken up by the kidney and excreted very slowly, or the synthesis of E by the kidney. A very small fraction of urinary E is probably derived from E taken up by the kidney and released later. The sympathetic nerves have an active uptake process for norepinephrine and very small amounts of E are taken up by active transport into noradrcncrgic nerve terminals. We did not measure urine E until after two hours of 3H-E infusion. E was cleared very quickly from the plasma hy uptake-i and uptake-2 mechanisms and subsequent metabolism.

of renal origin found in this study if there were a high rate of conversion of F-SO4 to E in human kidney. The adrenal gland is located on top of the kidney. If blood from the adrenal vein penetrated into the kidney it might be a source of urinary E. However, the adrenal sits outside the renal capsule and there are no known anatomic connections between the adrenal veins and the renal parenchyma in humans. Urine is formed in the renal glomerulus where pressures are higher than venous pres-

sure. It thus seems unlikely that adrenal venous E serves as a direct source of urinary E for both anatomic and hydrostatic reasons. The renal synthesis of norepinephrine to E is a likely source of some urinary E. Rat kidney can convert norepincphrine to urinary E using two different N-methylating enzymes [4]. These enzymes

are present in human kidney in even higher levels [3]. These

methylating enzymes can use norepinephrine derived from nerve endings or from the bloodstream as substrate. Buu et a! [5] have

also demonstrated that adrenalectomized rats can produce urinary 3H-E from injected 3H-dopamine sulfate. The adrenal medulla uses phenylethanolamine-N-methyltransferase (PNMT) to methylate norepinephrine into E. Since PNMT

is specific for -hydroxylated amines, it does not methylate dopamine very well. The kidney contains both PNMT and an N-methyltransferase which can methylate amines other than norepinephrine so the kidney could methylate dopamine into

The precipitous decay of tracer levels of plasma 3l-l-E in this experiment appeared to be quite similar to the decay of much higher levels of infused E in a study by Clutter et al [31 During epinine, which could then be converted to E. the stimulation of cardiac sympathetic nerves in humans, the Although the source of excess urinary E is not known, intrareamount of E released from sympathetic nerves ranged from nal synthesis of F from E-S04 or from norepinephrine seem likely

Ziegler et al: Sources of human urinaty epinephrine

sources of urinary E of renal origin. E synthesized by the kidney could potentially alter renal blood flow, sodium handling, and renin release but the actual physiologic role of renally synthesized E is not known. Young riormotensive control subjects in this study derived 43%

of their urinary E from the kidney. A group of older, heavier hypertensive men derived 85% of urinary E from the kidney. The older group had both a higher level of renal E synthesis and lower

plasma E levels. Others have reported that plasma E levels are lower in older subjects [18—21, reviewed in 22]. The older hypertensive group not only had lower plasma E levels, they had higher urinary E levels. Increased urinary E levels have been found in some studies of hypertensives [23—25] but this is by no means a uniform finding. The possibility of enhanced renal E production in hypertensives is intriguing and deserves further

327

inhibitor of phenylethanolamine-N-methyltransferase upon stimulated adrenal catecholamine release and excretion in the rat. NaunynSchmiedeberg's Arch Phar,nacol 97:245—25 0, 1977 9. VON EULER US, Koss D, LUFr R: Adrenaline excretion during resting

conditions and after insulin in adrenalectomized human subjects. Acta Endocrinol 38:441—448, 1961 10. KENNEDY B, ZIEGLER MG: A more sensitive and specific radioenzymatic assay for catecholamines. Life Sci 47:2143—2153, 1990 11. BUFANO G, VAONAG G, STARCICH R: Approach to the pharmacokinetics of DL-7-3H-noradrenaline. Eon C/in Pharmacol 6:88—92, 1973 12. LINARES OA, JACQUES JA, ZECH LA, SMITH MJ, SANFIELD JA, MORROW LA: Norepinephrine metabolism in humans. Kinetic analysis and model. J Clin Invest 80:1332—1341, 1987 13. CLUTrER WE, BIER DM, SHAH SD, CRYER PE: Epinephrine plasma

metabolic clearance rates and physiologic thresholds for metabolic and hemodynamic actions in man. J C/in Invest 66:94—101, 1980 14. ESLER M, KAYE D, THOMPSON J, JENNINGS G, Cox H, TURNER A, LAMBERT G, SEALS D: Effects of aging on epinephrine secretion and

regional release of epinephrine from the human heart. J Clin Endo-

investigation.

crinol Metab 80:435—442, 1995 E is present in human plasma and urine following bilateral C, ALTON H: Urinary excretion of adrenaline metaboadrenalectomy [26] and a major fraction of human urinary E 15. MCG000ALL lites in man during intervals of 2 minutes, 5 minutes, and 10 minutes comes from the kidney. Medline files from the past 10 years list after intravenous injection of adrenaline. Biochem Pharmaco/ 14: 800 papers indexed under both "epinephrine" and "urine." A 1595—1604, 1965

great many of these papers used urinary E as a guide to

16. FRANCIS GS, GOLDSMITH SR, PIERPONT G, COHN JN: Free and

conjugated plasma catecholamines in patients with congestive heart adrenomedullaty activity. E synthesized in the kidney is a major failure. J Lab C/in Med 103:393—398, 1984 source of urinary E and may heavily influence urinary E levels. E 17. RATGE D, GEHRKE A, MELZNER I, WISSER H: Free and conjugated stimulates renal renin release and sodium reabsorption, so human catecholamines in human plasma during physical exercise. Clin Exp renal E production may play a role in blood pressure regulation. Pharmacol Pharin Physiol 13:543—553, 1986 18. KJELDSEN SE, EIDE I, CHRISTENSEN C, WESTHEIM A, MULLER 0:

Acknowledgments

This study was supported by NIH Grants HL 35924 and Clinical Research Center Grant #MOI RR 00827. Reprint requests to Michael G. Ziegler, M.D., UCSD Medical Center, 200 West Arbor Drive, San Diego, California 92103-8341, USA.

Renal contribution to plasma catecholamines-effect of age. Scand J Clin Lab Invest 42:461—466, 1982 19. MESSERLI FH, FROHLICH ED, SUAREZ DH, REISIN E, DRESLINSKI GR,

DUNN FG, COLE FE: Borderline hypertension: Relationship between age, hemodynamics and circulating catecholamines. Circulation 64: 760—764, 1981 20. FRANCO-MORSELLI R, ELGHOZI JL, JOLY E, DIGIUILIO S, MEYER P:

Increased plasma adrenaline concentration in benign essential hypertension. Br Mcdi 2:1251—1254, 1977

E-mail: [email protected]

References 1. ZIEGLER MG, KENNEDY B, ELAYAN H: A sensitive radioenzymatic

assay for epinephrine-forming enzymes. Life Sci 43:2117—2122, 1988 2. ZIEGLER MG, KENNEDY B, ELAYAN H: Extraadrenal adrenaline formation by two separate enzymes. Experientia 45:718—720, 1989 3. KENNEDY B, BIGBY TD, ZIEGLER MG: Non-adrenal epinephrine forming enzymes in man: Characteristics, distribution, regulation and relationship to epinephrine levels. J Clin invest 95:2896—2902, 1995 4. ZIEGLER MG, KENNEDY B, ELAYAN H: Rat renal epinephrine synthesis. J Clin invest 84:1130—1133, 1989 5. Buu NT, NAIR G, KUCHEL 0, GENEST J: The extra-adrenal synthesis of epinephrine in rats. J Lab C/in Med 98:527—535, 1981 6. SNAVELY MD, ZIrv;I.EP. MG, INSEL PA: Subtype selective down-

regulation of rat renal cortical a and p adrellergic receptors by

21. BERTEL 0, BUHLER FR, KIOWSKI W, LUTOLD BE: Decreased beta-

adrenoceptor responsiveness as related to age, blood pressure and plasma catecholamines in patients with essential hypertension. Hypertension 2:130—138, 1980

22. GOLDSTEIN DS: Plasma catecholamines and essential hypertension. An analytical review. Hypertension 5:86—99, 1983 23. FREDICKSON M, TUOMISTO M, BERGMAN-LOSMAN B: Neuroendocrine

and cardiovascular stress reactivity in middle-aged normotensive adults with parental history of cardiovascular disease. Psychophysiology 28:656—664, 1991 24. Mo R, MYKING OL, LUND-JOHANSEN P, OMVIK P: The Bergen Blood

Pressure Study: Inappropriately low levels of circulating atrial natriuretic peptide in offspring of hypertensive families. Blood Pressure 3:223—230, 1994

25. BROWN MJ, CAUSON RC, BARNES VF, BRENNAN P, BARNES G,

catecholamines. Endocrinology 117:2182—2189, 1985 7. SMYTHE C, MCNICKEL JF, BRADLEY SE: The effect of epinephrine,

GREENBERG G: Urinary catecholamines in essential hypertension: Results of 24-hour urine catecholamine analyses from patients in the

L-epincphrine and norepinephrine on glomerular filtration rate, renal

Medical Research Council trial for mild hypertension and from

plasma flow, and the urinary excretion of sodium, potassium, and

matched controls. Quart i Med 57:637—65 1, 1985 26. SHiu- SD, TSE TF, CLU1FER WE, CREYER PE: The human sympathochromaffin system. Am J Physiol 247:E380—E384, 1984

water in normal man. J C/in Invest 31:499—506, 1952 8. PENDELTON RG, WEINER G, JENKINS B, GESSNER G: The effects of an