- Email: [email protected]
Catecholamines in depression: a cumulative study of urinary norepinephrine and its major metabolites in unipolar and bipolar depressed patients versus healthy volunteers at the NIMH
Psychiatry Research 87 Ž1999. 21]27
Catecholamines in depression: a cumulative study of urinary norepinephrine and its major metabolites in unipolar and bipolar depressed patients versus healthy volunteers at the NIMH Fred GrossmanU , William Z. Potter Lilly Corporate Center, Drop Code 2033, Indianapolis, IN 46285, USA Received 6 August 1998; received in revised form 20 May 1999; accepted 7 June 1999
Abstract Studies comparing urinary norepinephrine ŽNE. and its metabolites in unipolar or bipolar depressed patients and healthy volunteers have not yielded consistent findings. However, in unipolar depressed patients, most studies in non-elderly populations consistently report elevated concentrations of plasma NE, at least following an orthostatic challenge. Expanding upon previous studies which showed elevated plasma NE in depression, we compared the urinary excretion of NE, normetanephrine ŽNMN., 3-methoxy-4-hydroxyphenylglycol ŽMHPG., and vanillylmandelic acid ŽVMA. in age- and sex-matched unipolar and bipolar depressed patients versus healthy volunteers hospitalized at an inpatient unit at the National Institute of Mental Health. Only depressed subjects with a minimum 4-week drug-free period were included. Total turnover ŽNEq NMNq MHPG q VMA. was reduced in this sample of unipolar and bipolar depressed patients. MHPG concentration did not distinguish unipolar from bipolar depressed patients and was not significantly different from that in healthy volunteers. A construct of the average fractional extraneuronal concentration of NE ŽNEq NMNrNEq NMNq MHPG q VMA. was significantly higher in unipolar and bipolar depressed patients than in healthy volunteers. This finding extends data suggesting that unmedicated unipolar and bipolar depressed patients have a ‘hyperresponsive’ noradrenergic system and provides a framework which ties together plasma and urinary findings. Q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Norepinephrine; Normetanephrine; 3-Methoxy-4-hydroxyphenylglycol; Vanillylmandelic acid
U
Corresponding author. Tel.: q1-317-433-4311; fax: q1-317-276-6026. E-mail address: [email protected] ŽF. Grossman.
0165-1781r99r$ - see front matter Q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 9 9 . 0 0 0 5 5 - 4
22
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
1. Introduction Studies that suggest reduced noreprinephrine ŽNE. output in bipolar vs. unipolar depressed patients rely on either 3-methoxy-4-hydroxyphenylglycol ŽMHPG. in urine or NE in plasma ˚ ŽSchildkraut et al., 1978; Agren, 1982; Muscettola et al., 1984; Roy et al., 1985, 1987; Rudorfer et al., 1985; deVilliers et al., 1987; Veith et al., 1988; reviewed in Goodwin and Jamison, 1990.. This, however, is a simplistic interpretation of the data since concentrations of NE in plasma and MHPG in urine are determined by multiple processes which are embedded in what is loosely termed ‘output’. For instance, desipramine, an NE uptake inhibitor, increases plasma NE and decreases urinary MHPG in humans consistent with an increase in extraneuronal ‘parent’ neurotransmitter and a shift away from intraneuronal metabolism andror a decrease in total synthesis ŽLinnoila et al., 1982.. It is unclear whether this represents increased or decreased output. Very recent studies with moclobemide, a type A monoamine oxidase inhibitor, reveal reductions of the major deaminated metabolites ŽMHPG and DHPG. and increases of NMN, the major methylated metabolite of NE in urine, consistent with earlier studies of clorgyline ŽPotter et al., 1981.. This specific pattern of change is explained by the preferential intraneuronal localization of monoamine oxidase ŽMAO. and the extraneuronal location of catechol-O-methyltransferase ŽCOMT. ŽScheinin et al., 1998.. Thus, by investigating the pattern of drug-induced alterations on NE and its metabolites in urine, one can specify whether there is a shift from intraneuronal to extraneuronal metabolism using an integrated measure not provided by the types of plasma studies done to date. Studies in which the noradrenergic system is ‘stressed’ such as with an orthostatic challenge or cold pressor test have shown unipolar and bipolar depressed patients to have higher plasma NE levels when compared to healthy age- and sexmatched volunteers ŽRoy et al., 1985, 1987; Rudorfer et al., 1985; Veith et al., 1988.. A subgroup of bipolar depressed patients show reduced resting NE ŽRudorfer et al., 1985.. These studies suggest a dysregulation of peripheral release of
NE in affective illness. Taken together, bipolar and unipolar depressed patients have been shown to exhibit an exaggerated NE response to ‘stress’ with a suggestion that bipolar depressed patients at rest show reduced NE output. Such investigations of plasma NE Žhalf-life s 10 min. at one or two points in time are likely to reflect very recent states including levels of anxiety or activity. Plasma NE concentrations also depend on volume of distribution, rate of entry, and clearance of released NE ŽKopin, 1985., variables addressed by determining ‘spillover’ rates which are increased in depressed unipolar patients ŽEsler et al., 1982; Veith et al., 1985.. Bipolar patients have not been studied with this technique. Even if ‘spillover’ is determined with this radiolabeled tracer technique, the interpretation is complicated since the ‘spillover’ measure may or may not reflect firing of peripheral sympathetic nerves. Measures of NE and its major metabolites in urine offer an alternative technique for studying components of NE regulation in humans over a 24-h Žor longer. period presenting an ‘average’ of these multiple measures underlying point-in-time plasma measures. As briefly described above, and in other pharmacologic studies, these urinary measures provide robust indices with absolute and relative changes Ži.e. ratios. consistent with specific pharmacologic actions ŽLinnoila et al., 1986.. Thus, reliable information about the average production of NE and its metabolites within an individual can be determined from only two consecutive 24-h urine collections ŽLinnoila et al., 1986.. Furthermore, since all major metabolites are measured, 24-h urinary measures of NE and its metabolites can account for interindividual differences in the relative metabolism of NE as well as its turnover Žamount formed and excreted per 24 h at steady state.. Investigators have reported elevated urinary excretion of NE and its major extraneuronal metabolite normetanephrine ŽNMN. in depressed patients ŽRoy et al., 1985, 1988; Maas et al., 1987; Davis et al., 1988.. These findings are more impressive when the excretion of NE and NMN in depression is examined relative to total NE excretion. Maas et al. Ž1987. reported marked increases in urinary NE Ž57%. and NMN Ž42%. with a modest increase in uri-
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
nary catecholamine excretion Ž16%. in depressed patients. These results suggest a shift toward extraneuronal pathways and are consistent with suggestions of increased NE release and ‘spillover’ in depression. We, therefore, sought to investigate possible unipolarrbipolar changes with a comparison of baseline excretion of urinary NE and its metabolites and the fractional extraneuronal concentration of NE Žderived from the ratio of the average concentration of NE q NMNrNE q NMNq MHPG q VMA. in unmedicated unipolar and bipolar depressed patients and healthy volunteers. The basis for employing such a ratio rests on the distinction between intraneuronal and extraneuronal paths of NE referred to above.
2. Methods Patients received diagnoses of bipolar or unipolar depression without psychosis or comorbid psychiatric or medical conditions on the basis of Research Diagnostic Criteria ŽSpitzer et al., 1978. and longitudinal observation on the NIMH inpatient ward. All patients had been referred because of histories of poor response to antidepressant medications andror their need for ECT in the past. The diagnosis of bipolar or unipolar depression was confirmed by consensus following evaluation by a physician, nurse, and social worker after the patient’s admission to the ward and drug washout. During an extended medication-free period of 4]8 weeks, patients remained on the NIMH ward. To be included in biochemical studies, patients were required to show moderate to severe depression as documented by daily ratings on the Bunney-Hamburg Scale ŽBunney and Hamburg, 1963.. Patients participated in a variety of studies during stays at the NIMH that averaged 6 months. Patients and healthy volunteers were free of any medical, neurological, or endocrinological disorders and were placed on a low monoamine, low caffeine, and alcohol-free diet. Urine aliquots from two or more 24-h collections with a minimum volume of 900 cm3 were obtained. Healthy volunteers provided informed consent and had no personal nor family
23
history of psychiatric disorders and no first-degree relatives with a history of substance abuse. Healthy volunteers did not reside on the ward for more than a few days. 2.1. Urine NE and metabolites Urine samples were collected, stored at y208C and cumulatively assayed for total Žfree q conjugated. NE, MHPG, VMA, and NMN using previously described gas chromatography with mass fragmentographic detection methods ŽGordon et al., 1974; Karoum and Neff, 1982.. Values from at least two 24-h urine collections were averaged. 2.2. Sample size The sample comprised 34 patients and healthy volunteers from a larger group of 51 subjects on whom urinary measures were obtained. Since higher concentrations of MHPG are found in males, and in some studies age Žmore than a 40-year difference between groups. correlates positively with concentrations of NE and MHPG ŽPotter et al., 1983., age- and sex-matched groups were identified as closely as possible Žage within 5 years.. Groups were formed in such a way as to achieve the best match between bipolar patients and healthy volunteers based upon previous findings that the greatest difference might be obtained in comparing these groups. This reduced the sample size available for comparison to 12 bipolar depressed patients, 10 unipolar depressed patients, and 12 healthy volunteers. There were eight females in each group, four males in the bipolar depressed and healthy volunteer groups, and two males in the unipolar depressed group. Mean age ŽS.D.. was 36.9 Ž20. for the bipolar group, 42.1 Ž13. for the unipolar depressed group, and 37.3 Ž11. for the healthy volunteer group. We did not recruit patients to extend the age- and sex-matched unipolar depressed group to an equal sample size since that would have violated our ‘random’ cumulative database strategy. For statistical comparison, Student’s t-test was used on the paired data. In order to minimize the chance of a Type II error, uncorrected P values for multiple comparisons are shown. If the Bonferroni correction were used, these should be divided by six.
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
24
3. Results
4. Discussion
Table 1 shows values of the average urinary excretion of NE, NMN, MHPG, and the fractional extraneuronal concentration and total turnover of NE. As shown in this table and in Figs. 2 and 3, the average fractional extraneuronal concentration of NE ŽRATIO. in bipolar and unipolar depressed patients was significantly greater than in healthy volunteers Ž P F 0.001.. The total turnover of NE ŽSUM. was significantly lower in the bipolar and unipolar depressed patients than in the healthy volunteers Ž P- 0.05.. The average 24-h urinary excretion of NE taken by itself Žfree q conjugated. was significantly greater in the bipolar depressed patients Ž1.04" 0.43 mMr24 h. than in healthy volunteers Ž0.73" 0.15 mMr24 h. Ž P- 0.05.. Average 24-h excretion of NE in unipolar depressed patients was somewhat higher Ž0.89" 0.29 mMr24 h. but not significantly different than in healthy volunteers. Urinary excretion of MHPG in bipolar and unipolar depressed patients was not significantly different from that in healthy volunteers ŽFig. 1.. There were no significant differences in the fractional extraneuronal concentrations, total turnover, or average urinary excretion of NE between the bipolar and unipolar depressed patients. Similarly, there were no significant differences in average 24-h urinary concentrations of MHPG between the unipolar and bipolar depressed patients.
Urinary output of NE and fractional extraneuronal NE derived from urinary NE and its metabolites were found to be elevated in both bipolar and unipolar depressed patients when compared to healthy volunteers. These findings extend data from previous studies which have suggested that unmedicated unipolar and bipolar depressed patients, when compared to healthy volunteers, have a ‘hyperresponsive’ noradrenergic system at baseline and in response to a variety of stimuli. Interestingly, the total turnover of NE ŽNEq NMNq MHPG q VMA. was significantly lower in the unipolar and bipolar depressed patients compared to healthy volunteers, suggesting a reduction of tyrosine hydroxylase activity in sympathetic neurons for these groups. A possible explanation is that drug-free patients who remain markedly depressed over weeks to months ‘adjust’ to what must constitute a chronic stress by reducing the total synthesis and sympathetic output of NE, analogous to what is observed in animals ŽRoth et al., 1982.. In contrast to previous studies, we found significantly elevated concentrations of urinary NE in bipolar not unipolar depressed patients compared to healthy volunteers and a trend towards decreased MHPG in both patient groups Ž P0.054.. Several possible methodological factors may account for these differences. Variations in absolute values for patients and control groups
Table 1 Average 24-h urinary excretion of NE and its major metabolites, fractional extraneuronal concentrations, and turnover of NE in bipolar and unipolar depressed patients and healthy volunteers
NE ŽmMr24 h. NMN ŽmMr24 h. MHPG ŽmMr24 h. VMA ŽmMr24 h. SUM ŽmMr24 h.a RATIOb a b
Healthy volunteers Ž n s 12.
Bipolar depressed Ž n s 12.
Unipolar depressed Ž n s 10.
0.73" 0.15 1.38" 0.32 16.62" 7.19 28.01 " 10.23 46.74" 16.94 0.05" 0.01
1.04" 0.43UUU 1.58" 0.34 12.40" 3.19 20.50" 4.42 35.53" 6.93UUU 0.07" 0.01U
0.87" 0.29 1.59" 0.39 11.50" 3.68 18.62" 6.34 32.56" 8.15UUU 0.08" 0.02UU
NEq NMNq MHPG q VMA. NE q NMNrNEq NMNq MHPG q VMA: U P- 0.001 vs. healthy volunteers;
UU
P- 0.001; UUU P- 0.05.
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
Fig. 1. Scatter plot of average 24-h urinary excretion of MHPG in unipolar and bipolar depressed patients and healthy volunteers.
across studies are great enough to implicate differences in assays as perhaps more important than differences in populations ŽPotter and Linnoila, 1989.. For instance, Koslow et al. Ž1983. and Davis et al. Ž1988. utilized a fluorometric assay for NE in contrast to the gas chromatography with mass fragmentographic detection methods used in the present study with substantially different estimates of 24-h urinary NE in bipolar depressed patients Ž0.18 mMr24 h vs. 1.04 mMr24 h.. Koslow et al. Ž1983. reported a greater elevation of urinary NE in unipolar patients. This can be explained by an important methodological difference. Patients in our study were medicationfree for a minimum of 4 weeks verified under hospital conditions while in the other study patients were medication-free for 7]10 days prior to urine collection. Previous reports have demonstrated that abrupt discontinuation of antidepressants leads to an increase in catecholamines for at least 2 weeks ŽCharney et al., 1982.. Moreover, patients in our study had time to accommodate to the research milieu which reduced the possibility of an ‘ordering’ effect based on ‘novelty’ and
25
Fig. 2. Scatter plot of NE turnover in unipolar and bipolar depressed patients and health volunteers derived from the sum of the average 24-h urinary excretion of NE, NMN, MHPG, and VMA.
stress from an inpatient ward in the early stages of the study. Urinary MHPG levels, in both patient groups, were found to be reduced but not significantly different from those in healthy volunteers. This represents a shift from previous studies in which the relative excretion of MHPG in unipolar depressed patients approaches, or in some cases, surpasses levels found in healthy volunteers. This may reflect methodological differences such as patient population shifts Žtertiary centers tend to admit more refractory patients. or assay techniques and study design. However, as stated previously, urinary MHPG, by itself, has not proved to be a sufficiently robust or consistent measure to be accepted as a useful tool in differentiating diagnostic groups. Current models are inadequate to reveal precise mechanisms that would provide the desired ‘higher level’ explanation for our findings. The derived output measure is not an absolute measure with clear functional meaning but a con-
26
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
Fig. 3. Scatter plot of the fractional extraneuronal concentration of NE ŽRATIO. in unipolar and bipolar depressed patients and healthy volunteers determined from a construct ŽNEq NMNrNEq NMNq MHPG q VMA. using the average 24-h urinary excretion of NE, NMN, MHPG, and VMA.
struct reflecting both extraneuronal Žprimarily NE and NMN. and intraneuronal Žprimarily MHPG and VMA. metabolism of released or ‘leaked’ NE. Taken in conjunction with the measure of total output ŽNEq NMNq MHPG q VMA., our findings in depressed patients may best be explained as a reduced total synthesis of NE in sympathetic neurons associated with a relative Žand perhaps absolute. increase in the rate of released NE. The interpretation of increased release is supported by the repeated demonstrations of elevated plasma NE andror ‘spillover’ in previous studies mentioned above. Indeed, the pattern of exaggerated NE increase following an orthostatic challenge in depressed unipolar and bipolar patients compared to healthy volunteers is very similar to the pattern of findings shown in Fig. 3. The elevated ratio suggests that in depression, for a given amount of NE synthesized ŽNE q NMNq MHPG q VMA., there is a greater release rate. A mechanism which would explain
reduced synthesis and increased firing of sympathetic neurons could involve central and peripheral mechanisms. For instance, if central regulation of sympathetic outflow were altered in such a way as to allow for exaggerated NE release to a variety of physiological, environmental, and psychological stimuli, perhaps peripheral feedback mechanisms might reduce tyrosine hydroxylase activity in sympathetic nerves. Clearly such discussion is quite speculative and suggests a need for studies of peripheral sympathetic firing in depressed patients. In conclusion, the present study adds to the cumulative body of ‘difficult to reconcile’ data implicating some abnormality of NE function in both unipolar and bipolar depression. The data do not support the suggestion that unipolar and bipolar patients can be distinguished on the basis of urinary measures of NE andror its metabolites. The application of a construct such as a ratio reflecting relative extraneuronal NE has allowed for a possible explanation that ties together much of our plasma and urinary data. References ˚ gren, H., 1982. Depressive symptom patterns and urinary A MHPG excretion. Psychiatry Research 6, 185]196. Bunney, W.E., Jr., Hamburg, D.A., 1963. Methods for reliable longitudinal observation of behavior. Archives of General Psychiatry 9, 114]128. Charney, D., Heninger, G., Sternberg, D., Landis, H., 1982. Abrupt discontinuation of tricyclic antidepressant drugs: evidence for noradrenergic hyperactivity. British Journal of Psychiatry 141, 377]386. Davis, J.M., Koslow, S.H., Gibbons, R.D., Maas, J.W., Bowden, C.L., Casper, R., Hanin, I., Javaid, J.I., Chang, S.S., Stokes, P.E., 1988. Cerebrospinal fluid and urinary biogenic amines in depressed patients and healthy controls. Archives of General Psychiatry 45, 705]717. deVilliers, A.S., Russell, V.A., Carstens, M.E., Aalbers, C., Gagiano, C.A., Chalton, D.O., Taljaard, J.J., 1987. Noradrenergic function and hypothalamic-pituitary-adrenal axis activity in primary unipolar major depressive disorder. Psychiatry Research 22, 127]140. Esler, M., Turbott, J., Schwarz, R., 1982. The peripheral kinetics of norepinephrine in depressive illness. Archives of General Psychiatry 39, 295]300. Goodwin, F.K., Jamison, K.R., 1990. Biochemical and pharmacologic studies. In: Goodwin, F.K., Jamison, K.R. ŽEds.., Manic Depressive Illness. Oxford University Press, New York, pp. 416]495.
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27 Gordon, E., Olover, J., Black, K., Kopin, I.J., 1974. Simultaneous assay by mass fragmentography of vanillylmandelic acid, homovanillic acid, and 3-methoxy-4-hydroxyphenylethyleneglycol in cerebral fluid and urine. Biochemical Medicine 11, 32]40. Karoum, F., Neff, N., 1982. Quantitative gas chromatographymass spectrometry ŽGC-MS. of biogenic amines: theory and practice. In: Spector, S., Beck, N. ŽEds.., Modern Methods in Pharmacology. Alan Liss, Inc., New York, pp. 39]54. Kopin, I.J., 1985. Catecholamine metabolism: basic aspects and clinical significance. Pharmacology Review 37, 333]364. Koslow, S.H., Maas, J.W., Bowden, C.L., Davis, J.M., Hanin, I., Javaid, J., 1983. CSF and urinary biogenic amines and metabolites in depression and mania. A controlled, univariate analysis. Archives of General Psychiatry 40, 999]1010. Linnoila, M., Karoum, F., Calil, H.M., Kopin, I.J., Potter, W.Z., 1982. Alteration of norepinephrine metabolism with desipramine and zimelidine in depressed patients. Archives of General Psychiatry 39, 1025]1028. Linnoila, M., Guthrie, S., Lane, E.A., Karoum, F., Rudorfer, M., Potter, W.Z., 1986. Clinical studies on norepinephrine metabolism: how to interpret the numbers. Psychiatry Research 17, 229]239. Maas, J.W., Koslow, S.H., Davis, J., Katz, M., Frazer, A., Bowden, C.L., Berman, N., Gibbons, R., Stokes, P., Landis, H., 1987. Catecholamine metabolism and disposition in healthy and depressed subjects. Archives of General Psychiatry 44, 337]344. Muscettola, G., Potter, W.Z., Pickar, D., Goodwin, F.K., 1984. Urinary 3-methoxy-4-hydroxyphenylglycol and major affective disorders. Archives of General Psychiatry 41, 337]342. Potter, W.Z., Linnoila, M., 1989. Biochemical classifications of diagnostic subgroups and D-type scores. Archives of General Psychiatry 46, 269]271. Potter, W.Z., Calil, H.M., Extein, I., Gold, P.W., Wehr, T.A., Goodwin, F.K., 1981. Specific norepinephrine and serotonin uptake inhibitors in man: a cross-over study with pharmacokinetic, biochemical, neuroendocrine and behavioral patterns. Acta Psychiatrica Scandinavica Suppl. 290, 152]165. Potter, W.Z., Muscettola, G., Godwin, F.K., 1983. Sources of variance in clinical studies of MHPG. In: Maas, J.W. ŽEds.., MHPG: Basic Mechanisms and Psychopathology. Academic Press, New York.
27
Roth, K.A., Mefford, I.M., Barchas, J.O., 1982. Epinephrine, norepinephrine, dopamine, and serotonin: differential effects of acute and chronic stress on regional brain amines. Brain Research 239, 417]424. Roy, A., Pickar, D., Linnoila, M., Potter, W.Z., 1985. Plasma norepinephrine level in affective disorders: relationship to melancholia. Archives of General Psychiatry 42, 1181]1185. Roy, A., Guthrie, S., Pickar, D., Linnoila, M., 1987. Plasma norepinephrine responses to cold challenge in depressed patients and normal controls. Psychiatry Research 21, 161]167. Roy, A., Pickar, D., De Jong, J., Karoum, F., Linnoial, M., 1988. Norepinephrine and its metabolites in cerebrospinal fluid, plasma, and urine. Relationship to hypothalamicpituitary-adrenal axis function in depression. Archives of General Psychiatry 45, 849]857. Rudorfer, M.V., Ross, R.J., Linnoila, M., Sherer, M.A., Potter, W.Z., 1985. Exaggerated orthostatic responsivity of plasma norepinephrine in depression. Archives of General Psychiatry 42, 1186]1192. Scheinin, M., Illi, A., Koulu, M., Ojala-Karlsson, O., 1998. Norepinephrine metabolites in plasma as indicators of pharmacological inhibition of monoamine oxidase and catechol-O-methyltransferase. Advances in Pharmacology 42, 367]370. Schildkraut, J.J., Orsulak, P.J., Schatzberg, A.F., Cole, J.O., Rosenbaum, A.H., 1978. Biochemical discrimination of subgroups of depressive disorders based on differences in catecholamine metabolism. In: Usdin, E., Hanin, I. ŽEds.., Biological Markers in Psychiatry and Neurology. Pergamon Press, New York, pp. 23]33. Spitzer, R.L., Endicott, J., Robins, E., 1978. Research Diagnostic Criteria: rationale and reliability. Archives of General Psychiatry 35, 773]782 Veith, R.C., Halter, J.B., Murburg, M.M., Barnes, R.F., Villacre, E., Bachus, F.I., 1985. Increased plasma NE appearance rate in dexamethasone resistant depression. Fourth World Congress of Biological Psychiatry, pp. 186. Veith, R.C., Barnes, R.F., Villacres, R.C., Murburg, M.M., Raskind, M.A., Borson, S., 1988. Plasma catecholamines and norepinephrine kinetics in depression and panic disorder. In: Belmaker, R. ŽEd.., Catecholamines: Clinical Aspects. Alan Liss, Inc., New York, pp. 197]202.
Catecholamines in depression: a cumulative study of urinary norepinephrine and its major metabolites in unipolar and bipolar depressed patients versus healthy volunteers at the NIMH Fred GrossmanU , William Z. Potter Lilly Corporate Center, Drop Code 2033, Indianapolis, IN 46285, USA Received 6 August 1998; received in revised form 20 May 1999; accepted 7 June 1999
Abstract Studies comparing urinary norepinephrine ŽNE. and its metabolites in unipolar or bipolar depressed patients and healthy volunteers have not yielded consistent findings. However, in unipolar depressed patients, most studies in non-elderly populations consistently report elevated concentrations of plasma NE, at least following an orthostatic challenge. Expanding upon previous studies which showed elevated plasma NE in depression, we compared the urinary excretion of NE, normetanephrine ŽNMN., 3-methoxy-4-hydroxyphenylglycol ŽMHPG., and vanillylmandelic acid ŽVMA. in age- and sex-matched unipolar and bipolar depressed patients versus healthy volunteers hospitalized at an inpatient unit at the National Institute of Mental Health. Only depressed subjects with a minimum 4-week drug-free period were included. Total turnover ŽNEq NMNq MHPG q VMA. was reduced in this sample of unipolar and bipolar depressed patients. MHPG concentration did not distinguish unipolar from bipolar depressed patients and was not significantly different from that in healthy volunteers. A construct of the average fractional extraneuronal concentration of NE ŽNEq NMNrNEq NMNq MHPG q VMA. was significantly higher in unipolar and bipolar depressed patients than in healthy volunteers. This finding extends data suggesting that unmedicated unipolar and bipolar depressed patients have a ‘hyperresponsive’ noradrenergic system and provides a framework which ties together plasma and urinary findings. Q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Norepinephrine; Normetanephrine; 3-Methoxy-4-hydroxyphenylglycol; Vanillylmandelic acid
U
Corresponding author. Tel.: q1-317-433-4311; fax: q1-317-276-6026. E-mail address: [email protected] ŽF. Grossman.
0165-1781r99r$ - see front matter Q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 9 9 . 0 0 0 5 5 - 4
22
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
1. Introduction Studies that suggest reduced noreprinephrine ŽNE. output in bipolar vs. unipolar depressed patients rely on either 3-methoxy-4-hydroxyphenylglycol ŽMHPG. in urine or NE in plasma ˚ ŽSchildkraut et al., 1978; Agren, 1982; Muscettola et al., 1984; Roy et al., 1985, 1987; Rudorfer et al., 1985; deVilliers et al., 1987; Veith et al., 1988; reviewed in Goodwin and Jamison, 1990.. This, however, is a simplistic interpretation of the data since concentrations of NE in plasma and MHPG in urine are determined by multiple processes which are embedded in what is loosely termed ‘output’. For instance, desipramine, an NE uptake inhibitor, increases plasma NE and decreases urinary MHPG in humans consistent with an increase in extraneuronal ‘parent’ neurotransmitter and a shift away from intraneuronal metabolism andror a decrease in total synthesis ŽLinnoila et al., 1982.. It is unclear whether this represents increased or decreased output. Very recent studies with moclobemide, a type A monoamine oxidase inhibitor, reveal reductions of the major deaminated metabolites ŽMHPG and DHPG. and increases of NMN, the major methylated metabolite of NE in urine, consistent with earlier studies of clorgyline ŽPotter et al., 1981.. This specific pattern of change is explained by the preferential intraneuronal localization of monoamine oxidase ŽMAO. and the extraneuronal location of catechol-O-methyltransferase ŽCOMT. ŽScheinin et al., 1998.. Thus, by investigating the pattern of drug-induced alterations on NE and its metabolites in urine, one can specify whether there is a shift from intraneuronal to extraneuronal metabolism using an integrated measure not provided by the types of plasma studies done to date. Studies in which the noradrenergic system is ‘stressed’ such as with an orthostatic challenge or cold pressor test have shown unipolar and bipolar depressed patients to have higher plasma NE levels when compared to healthy age- and sexmatched volunteers ŽRoy et al., 1985, 1987; Rudorfer et al., 1985; Veith et al., 1988.. A subgroup of bipolar depressed patients show reduced resting NE ŽRudorfer et al., 1985.. These studies suggest a dysregulation of peripheral release of
NE in affective illness. Taken together, bipolar and unipolar depressed patients have been shown to exhibit an exaggerated NE response to ‘stress’ with a suggestion that bipolar depressed patients at rest show reduced NE output. Such investigations of plasma NE Žhalf-life s 10 min. at one or two points in time are likely to reflect very recent states including levels of anxiety or activity. Plasma NE concentrations also depend on volume of distribution, rate of entry, and clearance of released NE ŽKopin, 1985., variables addressed by determining ‘spillover’ rates which are increased in depressed unipolar patients ŽEsler et al., 1982; Veith et al., 1985.. Bipolar patients have not been studied with this technique. Even if ‘spillover’ is determined with this radiolabeled tracer technique, the interpretation is complicated since the ‘spillover’ measure may or may not reflect firing of peripheral sympathetic nerves. Measures of NE and its major metabolites in urine offer an alternative technique for studying components of NE regulation in humans over a 24-h Žor longer. period presenting an ‘average’ of these multiple measures underlying point-in-time plasma measures. As briefly described above, and in other pharmacologic studies, these urinary measures provide robust indices with absolute and relative changes Ži.e. ratios. consistent with specific pharmacologic actions ŽLinnoila et al., 1986.. Thus, reliable information about the average production of NE and its metabolites within an individual can be determined from only two consecutive 24-h urine collections ŽLinnoila et al., 1986.. Furthermore, since all major metabolites are measured, 24-h urinary measures of NE and its metabolites can account for interindividual differences in the relative metabolism of NE as well as its turnover Žamount formed and excreted per 24 h at steady state.. Investigators have reported elevated urinary excretion of NE and its major extraneuronal metabolite normetanephrine ŽNMN. in depressed patients ŽRoy et al., 1985, 1988; Maas et al., 1987; Davis et al., 1988.. These findings are more impressive when the excretion of NE and NMN in depression is examined relative to total NE excretion. Maas et al. Ž1987. reported marked increases in urinary NE Ž57%. and NMN Ž42%. with a modest increase in uri-
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
nary catecholamine excretion Ž16%. in depressed patients. These results suggest a shift toward extraneuronal pathways and are consistent with suggestions of increased NE release and ‘spillover’ in depression. We, therefore, sought to investigate possible unipolarrbipolar changes with a comparison of baseline excretion of urinary NE and its metabolites and the fractional extraneuronal concentration of NE Žderived from the ratio of the average concentration of NE q NMNrNE q NMNq MHPG q VMA. in unmedicated unipolar and bipolar depressed patients and healthy volunteers. The basis for employing such a ratio rests on the distinction between intraneuronal and extraneuronal paths of NE referred to above.
2. Methods Patients received diagnoses of bipolar or unipolar depression without psychosis or comorbid psychiatric or medical conditions on the basis of Research Diagnostic Criteria ŽSpitzer et al., 1978. and longitudinal observation on the NIMH inpatient ward. All patients had been referred because of histories of poor response to antidepressant medications andror their need for ECT in the past. The diagnosis of bipolar or unipolar depression was confirmed by consensus following evaluation by a physician, nurse, and social worker after the patient’s admission to the ward and drug washout. During an extended medication-free period of 4]8 weeks, patients remained on the NIMH ward. To be included in biochemical studies, patients were required to show moderate to severe depression as documented by daily ratings on the Bunney-Hamburg Scale ŽBunney and Hamburg, 1963.. Patients participated in a variety of studies during stays at the NIMH that averaged 6 months. Patients and healthy volunteers were free of any medical, neurological, or endocrinological disorders and were placed on a low monoamine, low caffeine, and alcohol-free diet. Urine aliquots from two or more 24-h collections with a minimum volume of 900 cm3 were obtained. Healthy volunteers provided informed consent and had no personal nor family
23
history of psychiatric disorders and no first-degree relatives with a history of substance abuse. Healthy volunteers did not reside on the ward for more than a few days. 2.1. Urine NE and metabolites Urine samples were collected, stored at y208C and cumulatively assayed for total Žfree q conjugated. NE, MHPG, VMA, and NMN using previously described gas chromatography with mass fragmentographic detection methods ŽGordon et al., 1974; Karoum and Neff, 1982.. Values from at least two 24-h urine collections were averaged. 2.2. Sample size The sample comprised 34 patients and healthy volunteers from a larger group of 51 subjects on whom urinary measures were obtained. Since higher concentrations of MHPG are found in males, and in some studies age Žmore than a 40-year difference between groups. correlates positively with concentrations of NE and MHPG ŽPotter et al., 1983., age- and sex-matched groups were identified as closely as possible Žage within 5 years.. Groups were formed in such a way as to achieve the best match between bipolar patients and healthy volunteers based upon previous findings that the greatest difference might be obtained in comparing these groups. This reduced the sample size available for comparison to 12 bipolar depressed patients, 10 unipolar depressed patients, and 12 healthy volunteers. There were eight females in each group, four males in the bipolar depressed and healthy volunteer groups, and two males in the unipolar depressed group. Mean age ŽS.D.. was 36.9 Ž20. for the bipolar group, 42.1 Ž13. for the unipolar depressed group, and 37.3 Ž11. for the healthy volunteer group. We did not recruit patients to extend the age- and sex-matched unipolar depressed group to an equal sample size since that would have violated our ‘random’ cumulative database strategy. For statistical comparison, Student’s t-test was used on the paired data. In order to minimize the chance of a Type II error, uncorrected P values for multiple comparisons are shown. If the Bonferroni correction were used, these should be divided by six.
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
24
3. Results
4. Discussion
Table 1 shows values of the average urinary excretion of NE, NMN, MHPG, and the fractional extraneuronal concentration and total turnover of NE. As shown in this table and in Figs. 2 and 3, the average fractional extraneuronal concentration of NE ŽRATIO. in bipolar and unipolar depressed patients was significantly greater than in healthy volunteers Ž P F 0.001.. The total turnover of NE ŽSUM. was significantly lower in the bipolar and unipolar depressed patients than in the healthy volunteers Ž P- 0.05.. The average 24-h urinary excretion of NE taken by itself Žfree q conjugated. was significantly greater in the bipolar depressed patients Ž1.04" 0.43 mMr24 h. than in healthy volunteers Ž0.73" 0.15 mMr24 h. Ž P- 0.05.. Average 24-h excretion of NE in unipolar depressed patients was somewhat higher Ž0.89" 0.29 mMr24 h. but not significantly different than in healthy volunteers. Urinary excretion of MHPG in bipolar and unipolar depressed patients was not significantly different from that in healthy volunteers ŽFig. 1.. There were no significant differences in the fractional extraneuronal concentrations, total turnover, or average urinary excretion of NE between the bipolar and unipolar depressed patients. Similarly, there were no significant differences in average 24-h urinary concentrations of MHPG between the unipolar and bipolar depressed patients.
Urinary output of NE and fractional extraneuronal NE derived from urinary NE and its metabolites were found to be elevated in both bipolar and unipolar depressed patients when compared to healthy volunteers. These findings extend data from previous studies which have suggested that unmedicated unipolar and bipolar depressed patients, when compared to healthy volunteers, have a ‘hyperresponsive’ noradrenergic system at baseline and in response to a variety of stimuli. Interestingly, the total turnover of NE ŽNEq NMNq MHPG q VMA. was significantly lower in the unipolar and bipolar depressed patients compared to healthy volunteers, suggesting a reduction of tyrosine hydroxylase activity in sympathetic neurons for these groups. A possible explanation is that drug-free patients who remain markedly depressed over weeks to months ‘adjust’ to what must constitute a chronic stress by reducing the total synthesis and sympathetic output of NE, analogous to what is observed in animals ŽRoth et al., 1982.. In contrast to previous studies, we found significantly elevated concentrations of urinary NE in bipolar not unipolar depressed patients compared to healthy volunteers and a trend towards decreased MHPG in both patient groups Ž P0.054.. Several possible methodological factors may account for these differences. Variations in absolute values for patients and control groups
Table 1 Average 24-h urinary excretion of NE and its major metabolites, fractional extraneuronal concentrations, and turnover of NE in bipolar and unipolar depressed patients and healthy volunteers
NE ŽmMr24 h. NMN ŽmMr24 h. MHPG ŽmMr24 h. VMA ŽmMr24 h. SUM ŽmMr24 h.a RATIOb a b
Healthy volunteers Ž n s 12.
Bipolar depressed Ž n s 12.
Unipolar depressed Ž n s 10.
0.73" 0.15 1.38" 0.32 16.62" 7.19 28.01 " 10.23 46.74" 16.94 0.05" 0.01
1.04" 0.43UUU 1.58" 0.34 12.40" 3.19 20.50" 4.42 35.53" 6.93UUU 0.07" 0.01U
0.87" 0.29 1.59" 0.39 11.50" 3.68 18.62" 6.34 32.56" 8.15UUU 0.08" 0.02UU
NEq NMNq MHPG q VMA. NE q NMNrNEq NMNq MHPG q VMA: U P- 0.001 vs. healthy volunteers;
UU
P- 0.001; UUU P- 0.05.
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
Fig. 1. Scatter plot of average 24-h urinary excretion of MHPG in unipolar and bipolar depressed patients and healthy volunteers.
across studies are great enough to implicate differences in assays as perhaps more important than differences in populations ŽPotter and Linnoila, 1989.. For instance, Koslow et al. Ž1983. and Davis et al. Ž1988. utilized a fluorometric assay for NE in contrast to the gas chromatography with mass fragmentographic detection methods used in the present study with substantially different estimates of 24-h urinary NE in bipolar depressed patients Ž0.18 mMr24 h vs. 1.04 mMr24 h.. Koslow et al. Ž1983. reported a greater elevation of urinary NE in unipolar patients. This can be explained by an important methodological difference. Patients in our study were medicationfree for a minimum of 4 weeks verified under hospital conditions while in the other study patients were medication-free for 7]10 days prior to urine collection. Previous reports have demonstrated that abrupt discontinuation of antidepressants leads to an increase in catecholamines for at least 2 weeks ŽCharney et al., 1982.. Moreover, patients in our study had time to accommodate to the research milieu which reduced the possibility of an ‘ordering’ effect based on ‘novelty’ and
25
Fig. 2. Scatter plot of NE turnover in unipolar and bipolar depressed patients and health volunteers derived from the sum of the average 24-h urinary excretion of NE, NMN, MHPG, and VMA.
stress from an inpatient ward in the early stages of the study. Urinary MHPG levels, in both patient groups, were found to be reduced but not significantly different from those in healthy volunteers. This represents a shift from previous studies in which the relative excretion of MHPG in unipolar depressed patients approaches, or in some cases, surpasses levels found in healthy volunteers. This may reflect methodological differences such as patient population shifts Žtertiary centers tend to admit more refractory patients. or assay techniques and study design. However, as stated previously, urinary MHPG, by itself, has not proved to be a sufficiently robust or consistent measure to be accepted as a useful tool in differentiating diagnostic groups. Current models are inadequate to reveal precise mechanisms that would provide the desired ‘higher level’ explanation for our findings. The derived output measure is not an absolute measure with clear functional meaning but a con-
26
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27
Fig. 3. Scatter plot of the fractional extraneuronal concentration of NE ŽRATIO. in unipolar and bipolar depressed patients and healthy volunteers determined from a construct ŽNEq NMNrNEq NMNq MHPG q VMA. using the average 24-h urinary excretion of NE, NMN, MHPG, and VMA.
struct reflecting both extraneuronal Žprimarily NE and NMN. and intraneuronal Žprimarily MHPG and VMA. metabolism of released or ‘leaked’ NE. Taken in conjunction with the measure of total output ŽNEq NMNq MHPG q VMA., our findings in depressed patients may best be explained as a reduced total synthesis of NE in sympathetic neurons associated with a relative Žand perhaps absolute. increase in the rate of released NE. The interpretation of increased release is supported by the repeated demonstrations of elevated plasma NE andror ‘spillover’ in previous studies mentioned above. Indeed, the pattern of exaggerated NE increase following an orthostatic challenge in depressed unipolar and bipolar patients compared to healthy volunteers is very similar to the pattern of findings shown in Fig. 3. The elevated ratio suggests that in depression, for a given amount of NE synthesized ŽNE q NMNq MHPG q VMA., there is a greater release rate. A mechanism which would explain
reduced synthesis and increased firing of sympathetic neurons could involve central and peripheral mechanisms. For instance, if central regulation of sympathetic outflow were altered in such a way as to allow for exaggerated NE release to a variety of physiological, environmental, and psychological stimuli, perhaps peripheral feedback mechanisms might reduce tyrosine hydroxylase activity in sympathetic nerves. Clearly such discussion is quite speculative and suggests a need for studies of peripheral sympathetic firing in depressed patients. In conclusion, the present study adds to the cumulative body of ‘difficult to reconcile’ data implicating some abnormality of NE function in both unipolar and bipolar depression. The data do not support the suggestion that unipolar and bipolar patients can be distinguished on the basis of urinary measures of NE andror its metabolites. The application of a construct such as a ratio reflecting relative extraneuronal NE has allowed for a possible explanation that ties together much of our plasma and urinary data. References ˚ gren, H., 1982. Depressive symptom patterns and urinary A MHPG excretion. Psychiatry Research 6, 185]196. Bunney, W.E., Jr., Hamburg, D.A., 1963. Methods for reliable longitudinal observation of behavior. Archives of General Psychiatry 9, 114]128. Charney, D., Heninger, G., Sternberg, D., Landis, H., 1982. Abrupt discontinuation of tricyclic antidepressant drugs: evidence for noradrenergic hyperactivity. British Journal of Psychiatry 141, 377]386. Davis, J.M., Koslow, S.H., Gibbons, R.D., Maas, J.W., Bowden, C.L., Casper, R., Hanin, I., Javaid, J.I., Chang, S.S., Stokes, P.E., 1988. Cerebrospinal fluid and urinary biogenic amines in depressed patients and healthy controls. Archives of General Psychiatry 45, 705]717. deVilliers, A.S., Russell, V.A., Carstens, M.E., Aalbers, C., Gagiano, C.A., Chalton, D.O., Taljaard, J.J., 1987. Noradrenergic function and hypothalamic-pituitary-adrenal axis activity in primary unipolar major depressive disorder. Psychiatry Research 22, 127]140. Esler, M., Turbott, J., Schwarz, R., 1982. The peripheral kinetics of norepinephrine in depressive illness. Archives of General Psychiatry 39, 295]300. Goodwin, F.K., Jamison, K.R., 1990. Biochemical and pharmacologic studies. In: Goodwin, F.K., Jamison, K.R. ŽEds.., Manic Depressive Illness. Oxford University Press, New York, pp. 416]495.
F. Grossman, W.Z. Potter r Psychiatry Research 87 (1999) 21]27 Gordon, E., Olover, J., Black, K., Kopin, I.J., 1974. Simultaneous assay by mass fragmentography of vanillylmandelic acid, homovanillic acid, and 3-methoxy-4-hydroxyphenylethyleneglycol in cerebral fluid and urine. Biochemical Medicine 11, 32]40. Karoum, F., Neff, N., 1982. Quantitative gas chromatographymass spectrometry ŽGC-MS. of biogenic amines: theory and practice. In: Spector, S., Beck, N. ŽEds.., Modern Methods in Pharmacology. Alan Liss, Inc., New York, pp. 39]54. Kopin, I.J., 1985. Catecholamine metabolism: basic aspects and clinical significance. Pharmacology Review 37, 333]364. Koslow, S.H., Maas, J.W., Bowden, C.L., Davis, J.M., Hanin, I., Javaid, J., 1983. CSF and urinary biogenic amines and metabolites in depression and mania. A controlled, univariate analysis. Archives of General Psychiatry 40, 999]1010. Linnoila, M., Karoum, F., Calil, H.M., Kopin, I.J., Potter, W.Z., 1982. Alteration of norepinephrine metabolism with desipramine and zimelidine in depressed patients. Archives of General Psychiatry 39, 1025]1028. Linnoila, M., Guthrie, S., Lane, E.A., Karoum, F., Rudorfer, M., Potter, W.Z., 1986. Clinical studies on norepinephrine metabolism: how to interpret the numbers. Psychiatry Research 17, 229]239. Maas, J.W., Koslow, S.H., Davis, J., Katz, M., Frazer, A., Bowden, C.L., Berman, N., Gibbons, R., Stokes, P., Landis, H., 1987. Catecholamine metabolism and disposition in healthy and depressed subjects. Archives of General Psychiatry 44, 337]344. Muscettola, G., Potter, W.Z., Pickar, D., Goodwin, F.K., 1984. Urinary 3-methoxy-4-hydroxyphenylglycol and major affective disorders. Archives of General Psychiatry 41, 337]342. Potter, W.Z., Linnoila, M., 1989. Biochemical classifications of diagnostic subgroups and D-type scores. Archives of General Psychiatry 46, 269]271. Potter, W.Z., Calil, H.M., Extein, I., Gold, P.W., Wehr, T.A., Goodwin, F.K., 1981. Specific norepinephrine and serotonin uptake inhibitors in man: a cross-over study with pharmacokinetic, biochemical, neuroendocrine and behavioral patterns. Acta Psychiatrica Scandinavica Suppl. 290, 152]165. Potter, W.Z., Muscettola, G., Godwin, F.K., 1983. Sources of variance in clinical studies of MHPG. In: Maas, J.W. ŽEds.., MHPG: Basic Mechanisms and Psychopathology. Academic Press, New York.
27
Roth, K.A., Mefford, I.M., Barchas, J.O., 1982. Epinephrine, norepinephrine, dopamine, and serotonin: differential effects of acute and chronic stress on regional brain amines. Brain Research 239, 417]424. Roy, A., Pickar, D., Linnoila, M., Potter, W.Z., 1985. Plasma norepinephrine level in affective disorders: relationship to melancholia. Archives of General Psychiatry 42, 1181]1185. Roy, A., Guthrie, S., Pickar, D., Linnoila, M., 1987. Plasma norepinephrine responses to cold challenge in depressed patients and normal controls. Psychiatry Research 21, 161]167. Roy, A., Pickar, D., De Jong, J., Karoum, F., Linnoial, M., 1988. Norepinephrine and its metabolites in cerebrospinal fluid, plasma, and urine. Relationship to hypothalamicpituitary-adrenal axis function in depression. Archives of General Psychiatry 45, 849]857. Rudorfer, M.V., Ross, R.J., Linnoila, M., Sherer, M.A., Potter, W.Z., 1985. Exaggerated orthostatic responsivity of plasma norepinephrine in depression. Archives of General Psychiatry 42, 1186]1192. Scheinin, M., Illi, A., Koulu, M., Ojala-Karlsson, O., 1998. Norepinephrine metabolites in plasma as indicators of pharmacological inhibition of monoamine oxidase and catechol-O-methyltransferase. Advances in Pharmacology 42, 367]370. Schildkraut, J.J., Orsulak, P.J., Schatzberg, A.F., Cole, J.O., Rosenbaum, A.H., 1978. Biochemical discrimination of subgroups of depressive disorders based on differences in catecholamine metabolism. In: Usdin, E., Hanin, I. ŽEds.., Biological Markers in Psychiatry and Neurology. Pergamon Press, New York, pp. 23]33. Spitzer, R.L., Endicott, J., Robins, E., 1978. Research Diagnostic Criteria: rationale and reliability. Archives of General Psychiatry 35, 773]782 Veith, R.C., Halter, J.B., Murburg, M.M., Barnes, R.F., Villacre, E., Bachus, F.I., 1985. Increased plasma NE appearance rate in dexamethasone resistant depression. Fourth World Congress of Biological Psychiatry, pp. 186. Veith, R.C., Barnes, R.F., Villacres, R.C., Murburg, M.M., Raskind, M.A., Borson, S., 1988. Plasma catecholamines and norepinephrine kinetics in depression and panic disorder. In: Belmaker, R. ŽEd.., Catecholamines: Clinical Aspects. Alan Liss, Inc., New York, pp. 197]202.