0023-972X/79/4803-0415$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1979 by The Endocrine Society
Vol. 48, No. 3 Printed in U.S.A.
Regulation of Lipolysis by Human Adipose Tissue in Hyperthyroidism* PETER ARNER,f ANDERS WENNLUND, AND JAN OSTMAN Karolinska Institute, Department of Internal Medicine, Huddinge Hospital, S-141 86 Huddinge, Sweden
ABSTRACT. The effects of noradrenaline (NA) and isopropylnoradrenaline (ISNA) on glycerol release and cAMP levels in sc adipose tissue were studied in vitro in 27 patients with hyperthyroidism. In 11 patients, the studies were repeated after 6-12 months of treatment for hyperthyroidism. A third group comprised 21 euthyroid patients otherwise healthy except for morbid obesity. The lipolytic response to ISNA, observed in untreated thyrotoxic patients, was found to be reduced by 30% when the patients were reexamined after treatment for thyrotoxicosis. This reduction was attributable to a decrease in the cAMP level. This was observed whether adipose tissue was incubated in the presence or absence of a phosphodiesterase inhibitor, theophylline. Both NA and ISNA induced 50% more rapid glycerol release
A
CCELERATED mobilization of lipids in thyrotoxic L patients has been demonstrated by measurements of the turnover and concentration of FFA and glycerol (1-4). A synergism between thyroid hormones and catecholamines has been postulated on the basis of an observed enhanced mobilization of plasma FFA during infusion of catecholamines in patients with hyperthyroidism, whereas hypothyroid patients showed a diminished response (2). The mechanism of this interaction between thyroid hormones and catecholamines in thyrotoxicosis is not fully understood. The lipolytic response to catecholamines in adipose tissue has been studied only in rats which were made hyperthyroid by the administration of thyroid hormones (5-12). Species differences probably render this model inapplicable to man. Here, in contrast to the situation in the rat (13), catecholamines are the only hormones to exert a pronounced lipolytic action in vitro. Furthermore, in human adipose tissue, catecholamines can produce both a- and /?-adrenergic actions of lipolysis (14, 15), while in the rat only the /? effect has been observed (13). With a view to elucidating the adrenergic regulation of Received May 2, 1978. * This work was supported by grants from the Swedish Medical Research Council, the Karolinska Institute, the Wiberg Foundation, ihe Jeansson Foundation, and the Memory of Lars Hierta Foundation. f To whom requests for reprints should be addressed.
and 4 times higher cAMP levels in adipose tissue of the thyrotoxic subjects than in the obese euthyroid patients. A positive correlation between tissue cAMP and glycerol release, on one hand, and mean fat cell size, on the other hand, was observed in treated thyrotoxic patients and obese euthyroid patients but was not recorded in the untreated hyperthyroid patients. The basal rate of lipolysis was not altered in thyrotoxicosis. The results suggest that the enhanced lipolytic response to catecholamines in adipose tissue of hyperthyroid patients is due to increased /?-adrenergic responsiveness. In addition, a disruption in subsequent stages of the regulatory pathway at the level of protein kinase or hormone-sensitive lipase also seems possible. (J Clin Endocrinol Metab 48: 415, 1979)
lipolysis in thyrotoxicosis, we have conducted an in vitro study of sc adipose tissue obtained from untreated hyperthyroid patients. Some of them were reexamined after treatment for hyperthyroidism at a time when they had become euthyroid. A comparison was made with observations in adipose tissue from euthyroid patients suffering from morbid obesity but otherwise healthy. The effects of isopropylnoradrenaline (ISNA) and noradrenaline (NA) on lipolysis and levels of cAMP in adipose tissue were examined. Materials and Methods Clinical data relating to all patients are presented in Table 1. Twenty-seven thyrotoxic patients were in-patients. The diagnosis of hyperthyroidism was made on clinical findings combined with elevated values for total serum T4, the free T4 index, and serum T3. Seventeen of these patients had diffuse toxic goiter (Grave's disease) and 7 had toxic multinodular goiter (Plummer's disease). Three patients had hyperthyroidism unaccompanied by goiter. None of the patients had received any type of drug in the relevant period before the study. In 11 of the patients, a reexamination was performed 6-12 months after introduction of the treatment with propylthiouracil or methimazole and T4 (added after 4-8 weeks). At the time of reexamination, the patients had been euthyroid for at least 3 months, as judged on clinical and laboratory findings. Another group comprised 21 patients admitted to the hospital with morbid, nonendocrine obesity. They were otherwise healthy and thyroid function tests gave normal values. The
415
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 04:55 For personal use only. No other uses without permission. . All rights reserved.
416
ARNER, WENNLUND, AND OSTMAN
JCE & M • 1979 V0U8 • No 3
TABLE 1. Clinical data of the patients Age (yr)
Sex
BW (% of average)
T4 (nmol/liter)
FTI
T3 (nmol/liter)
Thyrotoxic Before treatment (n = 27) During treatment (n = 11)
49 (24-46) 46 (25-64)
22 F, 5 M 10 F, 1 M
100 (75-155) 113 (92-171)
240 (175-323) 142 (89-187)
291 (180-458) 138 (77-183)
6.8 (4.1-7.9) 2.4 (1.7-3.5)
Control (n = 21)
34 (18-57)
16 F, 5 M
168 (120-295)
102 (59-168)
94 (50-171)
Patients
Average body weight was obtained from tables computed by Documenta Geigy (37). Values are means; ranges are given in parentheses. T3 was analyzed in those of the untreated thyrotoxic patients that were reinvestigated during treatment for thyrotoxicosis. FTI, Free T4 index. subjects were fed a regular hospital diet before the study. A sc fat biopsy was taken from the upper lateral part of the thigh at 0800 h after an overnight fast. The examination and biopsy procedures were approved by the Ethics Committee of the Karolinska Institute. The patients were informed of the study and their consent was obtained.
Methods Local anesthesia was induced with prilocaine chloride, as described elsewhere (16). Sections of sc adipose tissue, weighing about 50 mg, were preincubated for 30 min in Krebs-Henseleit bicarbonate (KHB) buffer, pH 7.4, which contained 40 mg/ml bovine serum albumin (Armour Pharmaceutical Co., Eastbourne England; lot R-970). They were then incubated in 1 ml fresh buffer for 2 h, and aliquots of the medium were removed for glycerol determination (17). The buffer contained no glucose. Air was used as the gas phase. In separate experiments, the concentration of cAMP in adipose tissue was determined by a modification (18) of the protein-binding method of Gilman (19). Two incubation procedures were used. For patients that were lean or of normal weight, a small amount of adipose tissue was available (less than 1.5 g wet wt). Fifty milligrams of tissue were preincubated for 30 min in 2.5 ml KHB buffer, pH 7.4, containing 10 mmol/liter phosphodiesterase inhibitor (theophylline; ACO, Sweden) and then incubated for 10 min in fresh buffer (2.5 ml). From obese patients, larger amounts of adipose tissue were obtained (1.5-2 g) and, therefore, cAMP could be measured without a phosphodiesterase inhibitor (20). Then, 200 mg adipose tissue were preincubated and incubated in 10 ml KHB buffer without theophylline. Albumin was not added, since it interferes with the assay of cAMP (18). The presence of theophylline did not bias the results, since the drug increases the cAMP level without affecting the ratio between basal and catecholamine-stimulated cAMP. Under the described conditions, 1) the catecholamine-induced cAMP levels reach a peak after 10 min of incubation (20), 2) insignificant amounts of cAMP in intact human adipose tissue originate from stromal cells (21, 22), and 3) omission of albumin from the buffer does not lead to accumulation of intracellular FFA in amounts large enough to interfere with either the production or the degradation of cAMP. NA-bitartrate (Astra, Sweden) and ISNA-hydrochloride (Winthrop, England) were added in vitro in the concentration of 6 jumol/liter, at which level the drugs are known to cause maximum stimulation of lipolysis and production of cAMP under normal conditions (20, 23). cAMP-binding protein and
[3H]cAMP were obtained from Boehringer-Mannheirn (West Germany) and protein kinase inhibitor was obtained from Sigma. All of the incubations were run in quadruplicate. The glycerol and cAMP estimations were performed in duplicate. The fat cell diameter was determined (24) in two fat specimens, each weighing about 30 mg. One hundred cells were measured and mean fat cell volume and mean cellular triglyceride content were calculated using the formula evolved by Hirsch and Gallian (25). The number of fat cells incubated was calculated using the mean cellular triglyceride content and the total triglyceride content of the fat portion. The values given are the mean ± SEM. The differences between results were tested for statistical significance using the paired or unpaired t test. For regression analysis, the method of least squares was used. Mean lal eel volume. mm •10''
Glycerol release. /jmol / 2 h / io' cells
30
Ji t
Glycerol release. ^imol / 2 hi g lipid
4'
1000
20
2
500
10
0'
0
Basal
Mea n mm •
pmol / 1 0 ' cells
0
Basal
Wi
0-
Isopropyl noradrenaline
pmol 110' cells
1000-1 20000' 500-
0000
oTheophylline
Cyclic AMP. pmol / g lipid iOO
200 Q Basal
Isopropyl noradrenaline
cell volume,
Cyclic AMP, p m o l / g lipid
«s
it •-
ill
2000
•
Theophylline
isopropyl noradrena
isopropyl -
noradrenaline
10"*
Mea n l a l mm • 1 0 '
Cyc.tc AMP,
Theophylline
Basal
lal eel volume.
c
1000 4000-
^
•if]
e
Cyclic AMP.
2000-
0'
Iscpropyl -
noradrenalir
Theophylline Isopropyl -
ne
noradrenaline
FIG. 1. Fat cell size, glycerol release, and cAMP levels in sc adipose tissue obtained from the same patients with hyperthyroidism before (•) and after (D) treatment. The adipose tissue was incubated with or without ISNA (6 jumol/liter). Top row, Glycerol experiments (nine patients); middle row, cAMP levels in theophylline-free buffers (six patients); bottom row, cAMP levels using buffers containing 10 mmol/liter theophylline (five patients). The values are the mean ± SE and the results are expressed in terms of lipid weight (right column) and number of cells [left column). Comparison before and after treatment was by Student's paired t test. XX, P < 0.01, n, Number of patients.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 04:55 For personal use only. No other uses without permission. . All rights reserved.
417
LIPOLYSIS REGULATION IN HYPERTHYROIDISM
Results In the thyrotoxic patients examined before and after therapeutic normalization of thyroid function, there was no change in fat cell size or basal rate of glycerol release (Fig. 1). On the other hand, ISNA-induced glycerol release was significantly reduced (30%) by treatment for thyrotoxicosis (P < 0.01; Fig. 1). The cAMP level in adipose tissue incubated in catecholamine-free buffer was not influenced by the treatment regardless of whether theophylline was present in the buffer. During treatment for thyrotoxicosis, however, the ISNA-induced cAMP was significantly reduced. This decrease was significant both in the presence and absence of theophylline in the buffer (P < 0.01; Fig. 1). The results were similar whether the values for glycerol release and cAMP levels were expressed in terms of cell number or lipid weight. The means for glycerol and cAMP in adipose tissue of thyrotoxic and obese euthyroid patients matched for fat cell volume are given in Fig. 2. The mean fat cell volume for each patient group was about 800 X 10~6 mm3. The basal glycerol production was nearly the same for the
two groups. Both for NA (P < 0.05) and ISNA (P < 0.01), the rate of glycerol release was about 50% greater for the thyrotoxic than for the obese euthyroid patients. When only theophylline was present in the buffer, the tissue cAMP level was higher in the thyrotoxic patients (P < 0.05). Addition of either NA or ISNA to the buffer resulted in a nearly 4-fold higher level of cAMP in the adipose tissue of the thyrotoxic patients than in that of the obese euthyroid patients (P < 0.05 and P < 0.01, respectively). The results were the same whether the values for glycerol and cAMP were expressed in terms of lipid weight or number of fat cells. The results were also similar when all of the thyrotoxic and obese, euthyroid patients were compared. Before treatment was given for hyperthyroidism, there was no correlation between either cAMP or glycerol release and fat cell size. After treatment (Fig. 3), there was a positive correlation in respect to both basal {P < 0.05) and NA-induced glycerol release (P < 0.001). After treatment, when buffer containing theophylline was used, Glycerol release, /umol / 2 h / lO7 cells
Glycerol release, / j m o l / 2 h / 1O7 cells
BASAL
NORADRENALINE
Glycerol release. fjmoll
Cyclic AMP
2 h/K)' cells
pmol 110' cells
500
1000
500
1000
Mean fat cell v o l u m e , m m 1 • 10''
II Basal
Theophylline Theophylline Theophylline Cyclic AMP, pmol /10 Noradrenaline Isopropyl Noradrenaline
cells
Cyclic AMP, pmol / 10
THEOPHYLLINE
THEOPHYLLINE
•
cells NORADRENALINE
noradrenalir
Isopropylnoradrenalme
Glycerol release, /umol / 2 h / g hpi
Vol. 48, No. 3 Printed in U.S.A.
Regulation of Lipolysis by Human Adipose Tissue in Hyperthyroidism* PETER ARNER,f ANDERS WENNLUND, AND JAN OSTMAN Karolinska Institute, Department of Internal Medicine, Huddinge Hospital, S-141 86 Huddinge, Sweden
ABSTRACT. The effects of noradrenaline (NA) and isopropylnoradrenaline (ISNA) on glycerol release and cAMP levels in sc adipose tissue were studied in vitro in 27 patients with hyperthyroidism. In 11 patients, the studies were repeated after 6-12 months of treatment for hyperthyroidism. A third group comprised 21 euthyroid patients otherwise healthy except for morbid obesity. The lipolytic response to ISNA, observed in untreated thyrotoxic patients, was found to be reduced by 30% when the patients were reexamined after treatment for thyrotoxicosis. This reduction was attributable to a decrease in the cAMP level. This was observed whether adipose tissue was incubated in the presence or absence of a phosphodiesterase inhibitor, theophylline. Both NA and ISNA induced 50% more rapid glycerol release
A
CCELERATED mobilization of lipids in thyrotoxic L patients has been demonstrated by measurements of the turnover and concentration of FFA and glycerol (1-4). A synergism between thyroid hormones and catecholamines has been postulated on the basis of an observed enhanced mobilization of plasma FFA during infusion of catecholamines in patients with hyperthyroidism, whereas hypothyroid patients showed a diminished response (2). The mechanism of this interaction between thyroid hormones and catecholamines in thyrotoxicosis is not fully understood. The lipolytic response to catecholamines in adipose tissue has been studied only in rats which were made hyperthyroid by the administration of thyroid hormones (5-12). Species differences probably render this model inapplicable to man. Here, in contrast to the situation in the rat (13), catecholamines are the only hormones to exert a pronounced lipolytic action in vitro. Furthermore, in human adipose tissue, catecholamines can produce both a- and /?-adrenergic actions of lipolysis (14, 15), while in the rat only the /? effect has been observed (13). With a view to elucidating the adrenergic regulation of Received May 2, 1978. * This work was supported by grants from the Swedish Medical Research Council, the Karolinska Institute, the Wiberg Foundation, ihe Jeansson Foundation, and the Memory of Lars Hierta Foundation. f To whom requests for reprints should be addressed.
and 4 times higher cAMP levels in adipose tissue of the thyrotoxic subjects than in the obese euthyroid patients. A positive correlation between tissue cAMP and glycerol release, on one hand, and mean fat cell size, on the other hand, was observed in treated thyrotoxic patients and obese euthyroid patients but was not recorded in the untreated hyperthyroid patients. The basal rate of lipolysis was not altered in thyrotoxicosis. The results suggest that the enhanced lipolytic response to catecholamines in adipose tissue of hyperthyroid patients is due to increased /?-adrenergic responsiveness. In addition, a disruption in subsequent stages of the regulatory pathway at the level of protein kinase or hormone-sensitive lipase also seems possible. (J Clin Endocrinol Metab 48: 415, 1979)
lipolysis in thyrotoxicosis, we have conducted an in vitro study of sc adipose tissue obtained from untreated hyperthyroid patients. Some of them were reexamined after treatment for hyperthyroidism at a time when they had become euthyroid. A comparison was made with observations in adipose tissue from euthyroid patients suffering from morbid obesity but otherwise healthy. The effects of isopropylnoradrenaline (ISNA) and noradrenaline (NA) on lipolysis and levels of cAMP in adipose tissue were examined. Materials and Methods Clinical data relating to all patients are presented in Table 1. Twenty-seven thyrotoxic patients were in-patients. The diagnosis of hyperthyroidism was made on clinical findings combined with elevated values for total serum T4, the free T4 index, and serum T3. Seventeen of these patients had diffuse toxic goiter (Grave's disease) and 7 had toxic multinodular goiter (Plummer's disease). Three patients had hyperthyroidism unaccompanied by goiter. None of the patients had received any type of drug in the relevant period before the study. In 11 of the patients, a reexamination was performed 6-12 months after introduction of the treatment with propylthiouracil or methimazole and T4 (added after 4-8 weeks). At the time of reexamination, the patients had been euthyroid for at least 3 months, as judged on clinical and laboratory findings. Another group comprised 21 patients admitted to the hospital with morbid, nonendocrine obesity. They were otherwise healthy and thyroid function tests gave normal values. The
415
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 04:55 For personal use only. No other uses without permission. . All rights reserved.
416
ARNER, WENNLUND, AND OSTMAN
JCE & M • 1979 V0U8 • No 3
TABLE 1. Clinical data of the patients Age (yr)
Sex
BW (% of average)
T4 (nmol/liter)
FTI
T3 (nmol/liter)
Thyrotoxic Before treatment (n = 27) During treatment (n = 11)
49 (24-46) 46 (25-64)
22 F, 5 M 10 F, 1 M
100 (75-155) 113 (92-171)
240 (175-323) 142 (89-187)
291 (180-458) 138 (77-183)
6.8 (4.1-7.9) 2.4 (1.7-3.5)
Control (n = 21)
34 (18-57)
16 F, 5 M
168 (120-295)
102 (59-168)
94 (50-171)
Patients
Average body weight was obtained from tables computed by Documenta Geigy (37). Values are means; ranges are given in parentheses. T3 was analyzed in those of the untreated thyrotoxic patients that were reinvestigated during treatment for thyrotoxicosis. FTI, Free T4 index. subjects were fed a regular hospital diet before the study. A sc fat biopsy was taken from the upper lateral part of the thigh at 0800 h after an overnight fast. The examination and biopsy procedures were approved by the Ethics Committee of the Karolinska Institute. The patients were informed of the study and their consent was obtained.
Methods Local anesthesia was induced with prilocaine chloride, as described elsewhere (16). Sections of sc adipose tissue, weighing about 50 mg, were preincubated for 30 min in Krebs-Henseleit bicarbonate (KHB) buffer, pH 7.4, which contained 40 mg/ml bovine serum albumin (Armour Pharmaceutical Co., Eastbourne England; lot R-970). They were then incubated in 1 ml fresh buffer for 2 h, and aliquots of the medium were removed for glycerol determination (17). The buffer contained no glucose. Air was used as the gas phase. In separate experiments, the concentration of cAMP in adipose tissue was determined by a modification (18) of the protein-binding method of Gilman (19). Two incubation procedures were used. For patients that were lean or of normal weight, a small amount of adipose tissue was available (less than 1.5 g wet wt). Fifty milligrams of tissue were preincubated for 30 min in 2.5 ml KHB buffer, pH 7.4, containing 10 mmol/liter phosphodiesterase inhibitor (theophylline; ACO, Sweden) and then incubated for 10 min in fresh buffer (2.5 ml). From obese patients, larger amounts of adipose tissue were obtained (1.5-2 g) and, therefore, cAMP could be measured without a phosphodiesterase inhibitor (20). Then, 200 mg adipose tissue were preincubated and incubated in 10 ml KHB buffer without theophylline. Albumin was not added, since it interferes with the assay of cAMP (18). The presence of theophylline did not bias the results, since the drug increases the cAMP level without affecting the ratio between basal and catecholamine-stimulated cAMP. Under the described conditions, 1) the catecholamine-induced cAMP levels reach a peak after 10 min of incubation (20), 2) insignificant amounts of cAMP in intact human adipose tissue originate from stromal cells (21, 22), and 3) omission of albumin from the buffer does not lead to accumulation of intracellular FFA in amounts large enough to interfere with either the production or the degradation of cAMP. NA-bitartrate (Astra, Sweden) and ISNA-hydrochloride (Winthrop, England) were added in vitro in the concentration of 6 jumol/liter, at which level the drugs are known to cause maximum stimulation of lipolysis and production of cAMP under normal conditions (20, 23). cAMP-binding protein and
[3H]cAMP were obtained from Boehringer-Mannheirn (West Germany) and protein kinase inhibitor was obtained from Sigma. All of the incubations were run in quadruplicate. The glycerol and cAMP estimations were performed in duplicate. The fat cell diameter was determined (24) in two fat specimens, each weighing about 30 mg. One hundred cells were measured and mean fat cell volume and mean cellular triglyceride content were calculated using the formula evolved by Hirsch and Gallian (25). The number of fat cells incubated was calculated using the mean cellular triglyceride content and the total triglyceride content of the fat portion. The values given are the mean ± SEM. The differences between results were tested for statistical significance using the paired or unpaired t test. For regression analysis, the method of least squares was used. Mean lal eel volume. mm •10''
Glycerol release. /jmol / 2 h / io' cells
30
Ji t
Glycerol release. ^imol / 2 hi g lipid
4'
1000
20
2
500
10
0'
0
Basal
Mea n mm •
pmol / 1 0 ' cells
0
Basal
Wi
0-
Isopropyl noradrenaline
pmol 110' cells
1000-1 20000' 500-
0000
oTheophylline
Cyclic AMP. pmol / g lipid iOO
200 Q Basal
Isopropyl noradrenaline
cell volume,
Cyclic AMP, p m o l / g lipid
«s
it •-
ill
2000
•
Theophylline
isopropyl noradrena
isopropyl -
noradrenaline
10"*
Mea n l a l mm • 1 0 '
Cyc.tc AMP,
Theophylline
Basal
lal eel volume.
c
1000 4000-
^
•if]
e
Cyclic AMP.
2000-
0'
Iscpropyl -
noradrenalir
Theophylline Isopropyl -
ne
noradrenaline
FIG. 1. Fat cell size, glycerol release, and cAMP levels in sc adipose tissue obtained from the same patients with hyperthyroidism before (•) and after (D) treatment. The adipose tissue was incubated with or without ISNA (6 jumol/liter). Top row, Glycerol experiments (nine patients); middle row, cAMP levels in theophylline-free buffers (six patients); bottom row, cAMP levels using buffers containing 10 mmol/liter theophylline (five patients). The values are the mean ± SE and the results are expressed in terms of lipid weight (right column) and number of cells [left column). Comparison before and after treatment was by Student's paired t test. XX, P < 0.01, n, Number of patients.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 04:55 For personal use only. No other uses without permission. . All rights reserved.
417
LIPOLYSIS REGULATION IN HYPERTHYROIDISM
Results In the thyrotoxic patients examined before and after therapeutic normalization of thyroid function, there was no change in fat cell size or basal rate of glycerol release (Fig. 1). On the other hand, ISNA-induced glycerol release was significantly reduced (30%) by treatment for thyrotoxicosis (P < 0.01; Fig. 1). The cAMP level in adipose tissue incubated in catecholamine-free buffer was not influenced by the treatment regardless of whether theophylline was present in the buffer. During treatment for thyrotoxicosis, however, the ISNA-induced cAMP was significantly reduced. This decrease was significant both in the presence and absence of theophylline in the buffer (P < 0.01; Fig. 1). The results were similar whether the values for glycerol release and cAMP levels were expressed in terms of cell number or lipid weight. The means for glycerol and cAMP in adipose tissue of thyrotoxic and obese euthyroid patients matched for fat cell volume are given in Fig. 2. The mean fat cell volume for each patient group was about 800 X 10~6 mm3. The basal glycerol production was nearly the same for the
two groups. Both for NA (P < 0.05) and ISNA (P < 0.01), the rate of glycerol release was about 50% greater for the thyrotoxic than for the obese euthyroid patients. When only theophylline was present in the buffer, the tissue cAMP level was higher in the thyrotoxic patients (P < 0.05). Addition of either NA or ISNA to the buffer resulted in a nearly 4-fold higher level of cAMP in the adipose tissue of the thyrotoxic patients than in that of the obese euthyroid patients (P < 0.05 and P < 0.01, respectively). The results were the same whether the values for glycerol and cAMP were expressed in terms of lipid weight or number of fat cells. The results were also similar when all of the thyrotoxic and obese, euthyroid patients were compared. Before treatment was given for hyperthyroidism, there was no correlation between either cAMP or glycerol release and fat cell size. After treatment (Fig. 3), there was a positive correlation in respect to both basal {P < 0.05) and NA-induced glycerol release (P < 0.001). After treatment, when buffer containing theophylline was used, Glycerol release, /umol / 2 h / lO7 cells
Glycerol release, / j m o l / 2 h / 1O7 cells
BASAL
NORADRENALINE
Glycerol release. fjmoll
Cyclic AMP
2 h/K)' cells
pmol 110' cells
500
1000
500
1000
Mean fat cell v o l u m e , m m 1 • 10''
II Basal
Theophylline Theophylline Theophylline Cyclic AMP, pmol /10 Noradrenaline Isopropyl Noradrenaline
cells
Cyclic AMP, pmol / 10
THEOPHYLLINE
THEOPHYLLINE
•
cells NORADRENALINE
noradrenalir
Isopropylnoradrenalme
Glycerol release, /umol / 2 h / g hpi
