J. Endocrinol. Invest. 13: 575-580, 1990
Chronobiology of catecholamine excretion in normal and diabetic men G. Del Rio, C. Carani, A. Baldini, P. Marrama, and L. Della Casa Dipartimento di Endocrinologia e Metabolismo, Universita di Modena, 41100 Modena, Italy
ABSTRACT. The adrenomedullaryresponseto stimuli is often elevated in poorly controlled' insulin dependent diabetic patients, and it is controversial whether the adrenomedullary hyperactivity induces the suppression of the circadian rhythm of catecholamines. We have studied the urinary excretion of catecholamines in 11 diabetic patients during 48 h in 4-h collections. Eleven age and weight matched normal subjects served as controls. A circadian rhythm was detected for adrenaline and noradrenaline excretion both in normal and diabetic subjects, with the highest value for
both catecholamines in the early afternoon. The mean daily adrenaline levels were significantly higher in diabetic than in control subjects (p < 0.05). The dopamine excretion was correlated with noradrenaline excretion in normal subjects but did not show a definite circadian rhythm. We conclude that the adrenomedullary hyperactivity does not affect the rhythmic fluctuations of adrenaline ·and noradrenaline. The dopamine excretion does not show circadian variations and this probably reflects the absence of a single controlling oscillator.
INTRODUCTION Most of physiological processess in man, including the endocrine system, vary in an approximately sinusoidal fashion over the 24 h circadian span (1 ) and their loss is often associated with pathological conditions (2). It has been shown that the circadian rhythm in catecholamine excretion is lost in condition of sustained physical activity and after myocardial infarction, Le., in conditions of high sympathoadrenal activity (3). We have recently document-
pamine excretion from the innervated kidney (7). It is therefore important to establish if dopamine oscillations reflect sympathetic nervous system activation in normal and pathological conditions. The direct relationship between sodium intake and dopamine excretion (8) indicates that other factors may influence the rhythmicity of dopamine excretion. The primary aims of this study were to determine 1) whether the adrenomedullary hyperactivity affects the rhythmic fluctuations of catecholamines; 2) whether there is a circadian variation in dopamine excretion and its relationship with noradrenaline and adrenaline excretion in normal and diabetic subjects.
ed an hyperactivity of the adrenal medulla in type I diabetic subjects in the absence of parallel changes in noradrenaline excretion, suggesting a dissociation between the adrenal medulla and sympathetic nervous system activity (4). Other authors have reported a concomitant increase in adrenaline and dopamine, without changes in noradrenaline excretion in insulin dependent diabetic patients (5). Very little is known about the urinary excretion of dopamine in diabetic patients. Most of the urinary dopamine derives from the extraneuronal decarboxilation of L-Dopa (6); however the baroreceptor stimulation increases both noradrenaline and do-
PATIENTS AND METHODS Eleven insulin dependent diabetic patients (9 males and 2 females), aged 22-41, with normal blood pressure and creatinine clearance more than 1.20 ml/s were studied. Their mean duration of diabetes was of 12 ± 2 yr. They were selected on the basis of absent sympathetic or parasympathetic neuropathy. We have retained as normal R-R variations to deep breathing> 15 beats/min and a drop in systolic blood pressure less than 20 mmHg two min after shifting from the supine to an upright posture (9). None of the patients had nephropathy or was taking medication other than insulin (Total dose 48
Key-words: Catecholamines,100M, chronobiology. Correspondence: Dr. G. Del Rio, Istituto Patologia Medica, Via del Pozzo 71, 41100 Modena, Italy.
Received September 19, 1989; accepted April 26, 1990.
575
G. Del Rio, C. Carani, A. Baldini, et al.
± 4 U daily of intermediate- and short-acting insu-
tate buffer and 200 J-LI of the internal standard (IS, N-methyl- dopamine). The pH was adjusted to 6.07.0, thenthe mixture poured overa disposable microcation column prefilled with Bio-Rex 70 (Bio-Rad, Richmond, USA). Under these conditions catecholamines will complex readily with borate and can be selectively eluted from the column. Compounds were detected by their electrochemical activity and their concentrations determined by comparing NAilS and AilS peak height ratio in the unknown sample to the ratio in the urine standard. The results are reported as nanomol per 4 hours. The intrassay CV of A, NA and DA were 5°1o, 3.2°/0 and 8°1o, respectively. The interassay CV of A, NA and DA were 8.1 %, 7.0% and 9.0%, respectively. All urine specimens for an individual subject were analyzed in the same assay. The results of the same 4-h periods on each of the two consecutive days were averaged in each men. Data are expressed as mean ± SE. HBA1 was measured using ion exchange chromatography (normal range 5-8°/0) with final spectrophotometrical determination (Bio-Rad, Richmond, Usa). The Kolmogorov-Smirnov test was usedto evaluate the distribution of data. Differences in cathecholamine levels at different times were analysed using Duncan's multiple comparison analysis technique and the repeated samples analysis of variance method (11). Cosinoranalysis was used to determine whether any variations with time reflected a true circadian rhythm. The results of this analysis, giving mean level (Mesor) rhythm amplitude and time of crest (Acrophase) were summarized for the group as a whole using the procedure of Nelson et al. (12). A comparison between chronobiological characteristics of catecholamines in diabetic and control subjects was performed using the mesor test and the rhythm test (13). A P value less than 0.05 was considered significant.
lin). Seven point blood glucose profiles were performed on two separate days before entry the study, using glucostix read on a Glucometer " (Ames, Elkhart, USA). The mean glucose profiles and HbA1 were 11.4 ± 0.7 mmol/L and 10.8 ± 0.7°/0, respectively. Eleven healthy volunteers (8 males and 3 females) matched for age and weight servedas control group. They did not show medical problems, and none was taking any medication. STUDY PROTOCOL All partecipants were hospitalized at least 24 h before the study and placed on a normocaloric diet (carbohydrates 55%, lipids 300/0, proteins 15%). Meals and drink were provided at 07:30, 12:00, 19:00 and a little snack at 16:00 and 23:00 h. Tobacco, coffee, tea, banana, vanilla were prohibited during the study. All the patients and the controls were asleep by 23:30 and were roused at 07:00 h; when not in bed they keptto a fixed activity regimen. On the second day the subjects began to void into their 4-h urine collection containers. The volume of' each then in 4-h period on two consecutive days (12 samples) was recorded. A 50 ml of urine was retained for determinations of urinary free adrenaline, noradrenaline, dopamine and creatinine excreation. Since a small decrease inblood glucose concentration constitutes a stimulus foradrenal medullary secretion (1 0) we have performed in the patients frequent capillary glucose measurements (at 07:00, 11 :00, 15:45, 18:45 and 22:30 h) in order to exclude hypoglycemicreactions. The range of glucose concentrations at these time points was 9.7-10.4 mrnol/L The daily urinary sodium excretion was measured in all the subjects examined before and after the collection of urines for catecholamine determinations.
RESULTS In normal subjects, repeated measures ANOVA revaled significant differences between the 4-h time periods for urinary A, NA and DA excretion (p < 0.01, P < 0.001 and p < 0.001, respectively). In diabetic men a significant difference between the 4-h periods was detected only for A and NA excretion (p < 0.0001 and p < 0.004, respectively). Both in normal and diabetic subjects, after the onset of sleep, mean urinary A and NA values were always lowerthan during walking hours (Fig. 1). A circadian
ANALYTICAL METHODS The urinefor measurements of free catecholamines was immediately acidified with 6 N HCI and stored at 4 C until analysis. Urinary excretion rate of noradrenaline (NA), adrenaline (A) and dopamine (DA) was measured in the timed "4- hour urine specimens by reverse phase high pressure liquid chromatography with electrochemical detection as previously described (4). The details of the assay can be summarized as follows: 3 ml acidified urine were combined with 5 ml 0.03 M ammonium ace-
576
Catecholamines in 100M
a a
700
...co co ~
Chronobiology of catecholamine excretion in normal and diabetic men G. Del Rio, C. Carani, A. Baldini, P. Marrama, and L. Della Casa Dipartimento di Endocrinologia e Metabolismo, Universita di Modena, 41100 Modena, Italy
ABSTRACT. The adrenomedullaryresponseto stimuli is often elevated in poorly controlled' insulin dependent diabetic patients, and it is controversial whether the adrenomedullary hyperactivity induces the suppression of the circadian rhythm of catecholamines. We have studied the urinary excretion of catecholamines in 11 diabetic patients during 48 h in 4-h collections. Eleven age and weight matched normal subjects served as controls. A circadian rhythm was detected for adrenaline and noradrenaline excretion both in normal and diabetic subjects, with the highest value for
both catecholamines in the early afternoon. The mean daily adrenaline levels were significantly higher in diabetic than in control subjects (p < 0.05). The dopamine excretion was correlated with noradrenaline excretion in normal subjects but did not show a definite circadian rhythm. We conclude that the adrenomedullary hyperactivity does not affect the rhythmic fluctuations of adrenaline ·and noradrenaline. The dopamine excretion does not show circadian variations and this probably reflects the absence of a single controlling oscillator.
INTRODUCTION Most of physiological processess in man, including the endocrine system, vary in an approximately sinusoidal fashion over the 24 h circadian span (1 ) and their loss is often associated with pathological conditions (2). It has been shown that the circadian rhythm in catecholamine excretion is lost in condition of sustained physical activity and after myocardial infarction, Le., in conditions of high sympathoadrenal activity (3). We have recently document-
pamine excretion from the innervated kidney (7). It is therefore important to establish if dopamine oscillations reflect sympathetic nervous system activation in normal and pathological conditions. The direct relationship between sodium intake and dopamine excretion (8) indicates that other factors may influence the rhythmicity of dopamine excretion. The primary aims of this study were to determine 1) whether the adrenomedullary hyperactivity affects the rhythmic fluctuations of catecholamines; 2) whether there is a circadian variation in dopamine excretion and its relationship with noradrenaline and adrenaline excretion in normal and diabetic subjects.
ed an hyperactivity of the adrenal medulla in type I diabetic subjects in the absence of parallel changes in noradrenaline excretion, suggesting a dissociation between the adrenal medulla and sympathetic nervous system activity (4). Other authors have reported a concomitant increase in adrenaline and dopamine, without changes in noradrenaline excretion in insulin dependent diabetic patients (5). Very little is known about the urinary excretion of dopamine in diabetic patients. Most of the urinary dopamine derives from the extraneuronal decarboxilation of L-Dopa (6); however the baroreceptor stimulation increases both noradrenaline and do-
PATIENTS AND METHODS Eleven insulin dependent diabetic patients (9 males and 2 females), aged 22-41, with normal blood pressure and creatinine clearance more than 1.20 ml/s were studied. Their mean duration of diabetes was of 12 ± 2 yr. They were selected on the basis of absent sympathetic or parasympathetic neuropathy. We have retained as normal R-R variations to deep breathing> 15 beats/min and a drop in systolic blood pressure less than 20 mmHg two min after shifting from the supine to an upright posture (9). None of the patients had nephropathy or was taking medication other than insulin (Total dose 48
Key-words: Catecholamines,100M, chronobiology. Correspondence: Dr. G. Del Rio, Istituto Patologia Medica, Via del Pozzo 71, 41100 Modena, Italy.
Received September 19, 1989; accepted April 26, 1990.
575
G. Del Rio, C. Carani, A. Baldini, et al.
± 4 U daily of intermediate- and short-acting insu-
tate buffer and 200 J-LI of the internal standard (IS, N-methyl- dopamine). The pH was adjusted to 6.07.0, thenthe mixture poured overa disposable microcation column prefilled with Bio-Rex 70 (Bio-Rad, Richmond, USA). Under these conditions catecholamines will complex readily with borate and can be selectively eluted from the column. Compounds were detected by their electrochemical activity and their concentrations determined by comparing NAilS and AilS peak height ratio in the unknown sample to the ratio in the urine standard. The results are reported as nanomol per 4 hours. The intrassay CV of A, NA and DA were 5°1o, 3.2°/0 and 8°1o, respectively. The interassay CV of A, NA and DA were 8.1 %, 7.0% and 9.0%, respectively. All urine specimens for an individual subject were analyzed in the same assay. The results of the same 4-h periods on each of the two consecutive days were averaged in each men. Data are expressed as mean ± SE. HBA1 was measured using ion exchange chromatography (normal range 5-8°/0) with final spectrophotometrical determination (Bio-Rad, Richmond, Usa). The Kolmogorov-Smirnov test was usedto evaluate the distribution of data. Differences in cathecholamine levels at different times were analysed using Duncan's multiple comparison analysis technique and the repeated samples analysis of variance method (11). Cosinoranalysis was used to determine whether any variations with time reflected a true circadian rhythm. The results of this analysis, giving mean level (Mesor) rhythm amplitude and time of crest (Acrophase) were summarized for the group as a whole using the procedure of Nelson et al. (12). A comparison between chronobiological characteristics of catecholamines in diabetic and control subjects was performed using the mesor test and the rhythm test (13). A P value less than 0.05 was considered significant.
lin). Seven point blood glucose profiles were performed on two separate days before entry the study, using glucostix read on a Glucometer " (Ames, Elkhart, USA). The mean glucose profiles and HbA1 were 11.4 ± 0.7 mmol/L and 10.8 ± 0.7°/0, respectively. Eleven healthy volunteers (8 males and 3 females) matched for age and weight servedas control group. They did not show medical problems, and none was taking any medication. STUDY PROTOCOL All partecipants were hospitalized at least 24 h before the study and placed on a normocaloric diet (carbohydrates 55%, lipids 300/0, proteins 15%). Meals and drink were provided at 07:30, 12:00, 19:00 and a little snack at 16:00 and 23:00 h. Tobacco, coffee, tea, banana, vanilla were prohibited during the study. All the patients and the controls were asleep by 23:30 and were roused at 07:00 h; when not in bed they keptto a fixed activity regimen. On the second day the subjects began to void into their 4-h urine collection containers. The volume of' each then in 4-h period on two consecutive days (12 samples) was recorded. A 50 ml of urine was retained for determinations of urinary free adrenaline, noradrenaline, dopamine and creatinine excreation. Since a small decrease inblood glucose concentration constitutes a stimulus foradrenal medullary secretion (1 0) we have performed in the patients frequent capillary glucose measurements (at 07:00, 11 :00, 15:45, 18:45 and 22:30 h) in order to exclude hypoglycemicreactions. The range of glucose concentrations at these time points was 9.7-10.4 mrnol/L The daily urinary sodium excretion was measured in all the subjects examined before and after the collection of urines for catecholamine determinations.
RESULTS In normal subjects, repeated measures ANOVA revaled significant differences between the 4-h time periods for urinary A, NA and DA excretion (p < 0.01, P < 0.001 and p < 0.001, respectively). In diabetic men a significant difference between the 4-h periods was detected only for A and NA excretion (p < 0.0001 and p < 0.004, respectively). Both in normal and diabetic subjects, after the onset of sleep, mean urinary A and NA values were always lowerthan during walking hours (Fig. 1). A circadian
ANALYTICAL METHODS The urinefor measurements of free catecholamines was immediately acidified with 6 N HCI and stored at 4 C until analysis. Urinary excretion rate of noradrenaline (NA), adrenaline (A) and dopamine (DA) was measured in the timed "4- hour urine specimens by reverse phase high pressure liquid chromatography with electrochemical detection as previously described (4). The details of the assay can be summarized as follows: 3 ml acidified urine were combined with 5 ml 0.03 M ammonium ace-
576
Catecholamines in 100M
a a
700
...co co ~
