- Email: [email protected]
Glipizide stimulates sympathetic outflow in diabetes-prone mice
Life Sciences, Vol. 56, No. 9, pp. 661-661995 Cmnieht 0 1995 Eltier Science Ltd Print2 ii the USA. All rights reserved cm-3205/95 $950 t .oa
Pergamon
0024-3205(94)00499-4
GLIPIZIDE STIMULATES SYMPATHETIC OUTFLOW IN DIABETES-PRONE MICE C.M. Kuhn, R.S. Surwit and M.N. Feinglos Departments of Pharmacology, Psychiatry and Medicine, Durham N.C. 27710.
Duke University Medical Center,
(Received in final form December 10, 1994) Summarv
The purpose of the present study was to determine if the oral hypoglycemic agent glipizide influenced sympathetic outflow in diabetes-prone mice. C57BL/6 (diabetes-prone) and diabetes-resistant (A/J) were treated with saline or glipizide, and sympathetic outflow determined by the fall in organ norepinephrine content after synthesis inhibition with cr-methyl-para-tyrosine. Sympathetic outflow to the liver and pancreas were slower in B1/6 mice than in control A/J. Glipizide increased sympathetic outflow to the pancreas in both strains of mice, but did not influence outflow to other organs significantly. The results of this study suggest that glipizide can influence central glucoregulatory mechanisms after peripheral administration. Key Wordr:
glipizide, diabetes, C57BL/6 mice, hypoglycemic
agent
The mechanism by which oral hypoglycemic agents decrease blood glucose is still controversial. Several actions have been described, most importantly enhanced insulin release and improvement in insulin action (1,2). One major action of these agents is thought to result from blockade of specific ATP-sensitive Kt channels on the beta cell of the pancreas which leads to beta cell depolarization and increased insulin release (3). However, the clinical benefit derived from these agents seems to exceed that predicted on the basis of these actions. The central nervous system has not been widely considered as a site of action, as these drugs show low lipophilicity and only slight entry into the brain. However, the recent description of a wide distribution within the CNS of the ATP-sensitive Kt channels on which these agents are thought to act (4) has raised the possibility that actions within the CNS might contribute to the observed clinical benefit derived from these agents. Furthermore, recent studies both & ti and in vitro suggest that the second generation sulfonylurea hypoglycemic agents can influence behavior through actions in the CNS (5,6). Despite the limitation of access, oral hypoglycemic agents could act within the CNS through at least two mechanisms. They could bind to ATP-sensitive K+ channels present in sites that are not behind the blood brain barrier. Alternatively, they could elicit reflex actions within the CNS derived from a peripheral mechanism of action that could ultimately influence insulin and/or glucagon release. Direct influence on parasympathetic or sympathetic control of insulin secretion and/or glucose release from the liver represent the two most likely mechanisms by which such drugs could act within the CNS to influence blood glucose. Numerous studies demonstrate an important role for both limbs of the autonomic nervous system in glucoregulation (7,8). Studies in this laboratory and elsewhere suggest that altered sympathetic outflow represents a particularly likely candidate (9,lO). In mice, as in most other mammals, the Corresponding author: Dr. Cynthia Kuhn, Dep’t of Pharmacology, Duke University Medical Center, Durham, NC 27710
662
Glipizide Action in Diabetes-prone Mice
Vol. 56, No. 9, 1995
sympathetic innervation of the pancreas stimulates glucagon secretion and inhibits insulin secretion via beta and alpha-2 adrenergic receptors respectively (9-13). Numerous studies support a role for the SNS in stress-induced hyperglycemia, which is thought to contribute to loss of glucose homeostasis in stable diabetics (14). Furthermore, enhanced sympathetic inhibition of insulin secretion has been postulated to contributed to the etiology of NIDDM. Patients with NIDDM have abnormal insulin responses to the alpha adrenergic antagonist phentolamine which suggest the presence of excessive sympathetic inhibition of insulin secretion (15). The purpose of the present study was to test the hypothesis that the oral hypoglycemic agent glipizide is capable of decreasing sympathetic outflow in mice. The ability of glipizide to decelerate the decrease in organ norepinephrine content following synthesis inhibition with alpha-methyl-para-tyrosine (AMPT) has been used to assess this possibility. In mice, both anatomic and neurochemical studies have shown that the majority (> 90%) of pancreatic norepinephrine content reflects innervation of the endocrine islets (17,18). Furthermore, evaluation of pancreatic NE content after synthesis inhibition has been shown to vary inversely with insulin secretion in several models with altered sympathetic outflow to the liver including animals with VMH lesions, diet-induced obesity and genetic models of obesity (18-22). Therefore,this strategy can provide useful insight into the activity We have compared the action of glipizide in a strain of mice, the C57BL/6, which develops diabetes when allowed to eat a diet high in fat and simple carbohydrate, to its actions in a strain of mice, the A/J, which does not develop diabetes when allowed to eat the same diet (23). Both strains were tested to evaluate the possibility that glipizide might exert greater actions in the strain (BL/6) that is prone to develop diabetes. Methods Animals: Twelve week-old mice of the C57BL/6 and A/J strain were obtained from Jackson Laboratories (Bar Harbor, MA). They were housed in a vivarium with a 12 hr lightdark cycle, and allowed to eat Purina rodent chow ad libitum. Drun Treatment: Animals were treated subcutaneously with saline or glipizide (1 mg/kg) followed immediately by saline or AMPT (250 mglkg) ip. The dose of glipizide was based on previous animal studies showing significant hypoglycemic actions after chronic administration of 1 mg/kg (24,25). Animals were killed by decapitation at t = 0 (immediately after injection), 1, 2 or 4 hrs later (immediately after injection) or 4 hrs later. These time points have been used to evaluate sympathetic nervous system activity in diabetic mice in previous studies (20,21,26). Heart, liver, pancreas and fat were frozen at -20” until assayed for norepinephrine. Blood glucose and insulin was not influenced by the single dose of glipizide received by these animals 4 hrs before collection (data not shown). Organ Noreoineohrine Analysis: Tissue norepinephrine content was determined by on-line trace enrichment high pressure liquid chromatography followed by electrochemical detection. Tissues were homogenized in 0.2 N perchloric acid containing 0.5 mM EDTA and 0.5 mM sodium metabisulfite, centrifuged at 20,000 x G for 30 minutes, and supernatant used for analysis. The internal standard dihydroxybenyzlamine was added to an aliquot of the PCA extract, which was neutralized by addition of Tris buffer (1M pH 8.4) and catecholamines extracted onto alumina. Adsorbed catecholamines were eluted with the same perchloric acid solution used for homogenization, and injected onto the HPLC.Catecholamines were adsorbed onto a 3 cm Nucleosil 5Sa cation exchange precolumn with mobile phase of 0.02 M citric acid, 0.02 M potassium acetate, 0.5 mM EDTA, and eluted onto the Spherisorb 0DS2 Cl8 reverse phase analytic column with a mobile phase of 0.02 M citric acid, 0.2 M potassium acetate, 0.5 mM EDTA, 0.65 mM octylsodium sulfate, 1% acetonitrile. NE was quantitated by comparison to a standard curve extracted and run in parallel with each set of samples,
Glipizide Action in Diabetes-prone Mice
Vol. 56, No. 9, 1995
463
and individual recovery corrected with the internal standard. Drugs: Glipizide was a gift of PRATT Pharmaceuticals, Division of Pfizer Pharmaceuticals. AMPT methyl ester, catecholamines and DHBA for standards were obtained from Sigma Chemical Company (St. Louis, MO). Statistics: Results are expressed as mean + SEM. Results were analyzed by 2 way ANOVA (strain x treatment, comparing vehicle, (t=O), AMPT and AMPT + glipizide). The rate of decline of NE content was analyzed by least squares analysis of the regression line calculated for fall of NE content between t=O and t = 4. Results The purpose of the first experiments was to conduct a time-course for norepinephrine depletion following synthesis inhibition with AMPT to determine the time period over which depletion was linear, and presumably reflective of neuronal impulse flow in catecholamine neurons. The two strains of mice were both tested to validate that the half life of norepinephrine was similar enough in the two that a a similar time period could be used to test the effects of glipizide in both strains. Mice were injected with AMPT-methyl ester in saline, and killed at t =0 (immediately after injection), or t = 1,2 or 4 hours. Norepinephrine content was evaluated at each time point, (N= Wgroup) and a linear regression analysis of the rate of depletion used to calculate the half life of norepinephrine in each organ. As shown in Figure 1, norepinephrine depletion was linear over four hours in all tissues observed. Table 1 shows the half lives calculated from the rates of depletion. The half life of norepinephrine in pancreas and liver of the C57 BL/6 animals was significantly slower than the half life observed in tissues of the A/J mice. In contrast, the half life of norepinephrine in heart was similar in both strains. Pancreas 400
Norepinephrine
Content .
B”6
l
AIJ
300 9 E 200 Y
0
1
2
3
4
5
Time (hrs,
Heart
Norepinephrine
Content
Fig. 1 Decline in Organ Norepinephrine after AMPT. Animals were killed at the indicated time, and as described in methods. Results are expressed as mean+ SEM. N = 8lgroup.
NE determined
664
Glipizide Action in Diabetes-prone
Vol. 56, No. 9, 1995
Mice
The results of this experiment indicated that norepinephrine depletion was linear for at least 4 hrs. This time was selected for evaluation of glipizide effects, as catecholamine depletion was linear throughout this time period, glipizide effects should be maintained after a single dose for at least this time period, and AMPT inhibition of tyrosine hydroxylation should still be maintained. Evaluation at shorter times would not permit accurate quantitation of norepinephrine depletion in organs with slow turnover rates (C57BL/6 liver, for example, while with longer time periods the inhibition of NE synthesis cannot be maintained. The effect of glipizide on the AMPT-induced NE decline in pancreas, liver and heart was assessed by measuring NE content at 4 hours after synthesis inhibition. As norepinephrine depletion was proceeding linearly during this time period, norepinephrine content at this time provides an approximation of effects on sympathetic outflow. As shown in Figure 2, norepinephrine depletion in the pancreas was greater in animals treated with glipizide than those treated with saline. However, glipizide did not accelerate turnover in other organs. This finding suggests that glipizide specifically accelerated sympathetic outflow to the pancreas in both diabetes prone C57BL/6 and non diabetes-prone A/J mice. 0 100
AMPT AMPT + Glip
80
A/J
A/J Pancreas
Liver
Bu6
Heart
Fig. 2 Tissue NE content 4 hours after treatment with AMPT or AMPT + glipizide. Animals were treated with AMPT and vehicle or glipizide as described above, and killed 4 hrs later. Organ NE content is expressed as mean + SEM. N = 8/group. * indicates p < .05 or better relative to AMPT-treated control. Discussion The results of this study were the opposite of that predicted: the oral hypoglycemic agent glipizide enhanced rather than diminished sympathetic outflow as assessed with AMPTinduced decline of NE. Theoretically, this action should suppress insulin secretion rather than enhance it, and elevate rather than lower blood glucose. Furthermore, stimulation of glucagon secretion could also result from such stimulation. Levels of glucoregulatory hormones could not be assessed accurately in the present experiment, as all animals were treated with AMPT, an agent which itself can significantly influence insulin secretion. If changes in insulin reflect the increased sympathetic output, then the effect reported here could oppose the clinical action of glipizide.
GlipizideActionin Diabetes-proneMice
Vol.56, No. 9, 1995
The specificity
of this action for pancreas supports the possibility
645
that glipizide
influences regulation of glucose homeostasis through actions in the CNS. It is unlikely that the
action observed in the present study represents a reflex sympathetic response to a rapid lowering of insulin by the peripheral actions of glipizide. Although hypoglycemia is a potent stimulus for sympathetic outflow (27), such a counterregulatory response would involve a global increase in sympathetic outflow rather than being restricted to pancreatic innervation (28). We cannot rule out the possibility, however that this action could represent a transient activation that is reversed with chronic treatment. The few animal studies reporting blood glucose levels in animal models of NIDDM report significant effects after chronic drug treatment, but the effects of a single dose are rarely reported. The strain difference in basal sympathetic tone represents the other significant finding of the present study. Substantially slower NE turnover was observed in pancreas and liver of the “diabetes-prone” C57BL16 mice than was observed in the “diabetes-resistant” A/J strain. This apparent sympathetic withdrawal is consistent with previous studies demonstrating dietinduced sympathetic withdrawal and supersensitivity of adrenergic receptor populations in liver and/or pancreas that produce hyperglycemia following administration of exogenous catecholamines (21,22,29,30). We have speculated previously that this relative sympathetic withdrawal might reverse markedly when animals are fed a high fat, high simple carbohydrate diet, and contribute to the hyperglycemia which results (9). One possibility suggested by the present finding is that the glipizide-induced increase in sympathetic outflow serves to desensitize adrenergic receptors in the pancreas and attenuate adrenergic inhibition of insulin release. This effect would occur in addition to any direct stimulatory effect of glipizide on insulin secretion and insulin sensitivity. In summary, the present results suggest that the oral hypoglycemic agent glipizide influences sympathetic outflow to the pancreas in mice. The enhanced sympathetic outflow observed with a single dose of glipizide could reflect a direct drug action on the brain. These findings suggest the CNS-mediated glucoregulatory events might contribute to the clinical actions of these agents. In future studies, we plan to address the important question of the contribution of this change to glucoregulation following administration of clinically-effective doses to diabetic animals, and to evaluate structure-activity study of various sulfonylureas. Acknowledgements: This research was supported by a grant from Pfizer Pharmaceuticals to MNF and MH 00303 to RS. TABLE 1 Half Life of Norepinephrine in Peripheral Organs Organ
A/J
C57BL/6
Pancreas
2.7 hrs
7.7 hrs*
Liver
2.2 hrs
6.7 hrs*
Heart
5.7 hrs
8.6 hrs
Animals were treated with AMPT, killed at various times and norepinephrine determined as described in methods. Half life was calculated from least squares regression line of norepinephrine disappearance. N = 9-12Jgroup. * indicates p < .05 relative to A/J.
666
Glipizide Action in Diabetes-prone
Mice
Vol. 56, No. 9, 1995
REFERENCES 1. H. BECK-NIELSEN, 0. HOTHER-NIELSEN, 0. PEDERSEN. Diabetic Med. 3 613-620 (1988). 2. J. JACKSON, R. BRESSLER. CLINICAL PHARMACOLOGY OF SULPHONYLUREA HYPOGLYCAEMIC AGENTS: PART 1. Drugs 22 21 l-245 (198 1). FM. ASHCROFT Ann. Rev. Neurosci. 1197-xX (1990). :: C. MOURRE, C. WIDMANN AND M LAZDUNSKI. Brain Res. 519 29-43 (1990) D. LEVESQUE AND S.A. GREENFIELD. Neuropharmacol. 30 359-365 (1991). :: M. AMALRIC, C. HEURTEAUX, A. NIEOULLON AND M LAZDUNSKI. Eur. J. Pharmacol. 213 71-77 (1991). 7. D. PORTE AND SC. WOODS in --a Diabetes Mellitus Eds. M. Ellenberg and H. Rifkin. Med. Exam. Publ. pp 267-294. (1983) L. LANDSBERG. Clin. Exp. Hyperten. 12 817-830 (1990) :: R.S. SURWIT AND M.N. FEINGLOS. sabetes Care 1183-85 (1988) 10. R.S. SHERWIN, H. SHAMOON, R. HENDLER, L. SACCA, N. EIGLER, AND M. WALESKY. Metabolism 29 1146-I 154 (1980) 11. R.E. MILLER Endocrine Reviews 2 471-494 (1981) 12. N. YAMAGUCHI Canad. J. Physiol. Pharmacol. 70 167-206 (1991) 13. B. AHREN, I. LUNDQUIST Horm. Met. Res. 19 600-603 (1987) 415-420 (1986) 14. T. NAKADATE, T NAKAKI, T. MURAKI, R. KATO Eur. J. Pharmacol. 65 421-424 (1980) 15. D.G. BRUCE, D.J. CHISHOLM, L.H. STORLIEN, E.W. KRAEGEN AND G.A. SMYTHE. Diabetologia 35 835-843 (1992). 16. V.L. BROADSTONE, M.A. PFEIFER, V. BAJAJ, J.I. STAGNER, E. SMOLS. Diabetes 36 932-937 (1987). 17. I. LUNDQUIST, B. AHREN, C. HANSSON AND R. HAKANSON. Pancreas 6 662667 (1989) 18. I. LUNDQUIST, L.E. ERICSON. Cell and Tissue Res. 193 73-85 (1978) 19. A, TAKAHASHI, Y. IKARASHI, H. ISHIMARU AND Y. MARUYAMA. Life Sci. 53 1567-1572 (1993) 20. B.E. LEVIN, J. TRISCARI, A.C. SULLIVAN Amer. J. Physiol. 244 R347-R355 (1983) 21. B.E. LEVIN, J. TRISCARI, A.C. SULLIVAN. Amer. J. Physiol. 245 R367-R371 (1983) 22. T. YOSHIDA, H. NISHIOKA, Y. NAKAMURA, M. KONDO. Metabolism 33 10601063 (1984) 23. R.S. SURWIT, CM. KUHN, C. COCHRANE, J.A. MCCUBBIN, M.N. FEINGLOS. Diabetes 37 1163-1167 (1988). 24. A.S. REDDI. C.A. VELASCO. P.R. REDDY. M.Y.. R.A. CAMERINI-DAVALOS. Exp. Molec. Pathol. 53 140-151 (1990). ’ 25. H.E. LEBOVITZ. Pharmacotherapy 5 63-77 (1985). 26. B.E. LEVIN, J. TRISCARI, S. HOGAN AND A.C. SULLIVAN. Am. J. Physiol. 252 R471-R478 (1987). 27. J.B. YOUNG, R.M. ROSA, L LANDSBERG. Am. J. Physiol. 247 E35E40 (1984). 28. B.E. LEVIN AND A.C. SULLIVAN. J. Aut. Nerv. Sys. 20 233-242 (1987). 29. A.W. KNEHANS AND D.R. ROMSOS. Am. J. Physiol. 244 E567-E574 (1983) 30. C.J. LYNCH, M/L. STEER, M.R. CONNORS, R.A. SCHATZ, R.C. DETH. Biochem. Pharmacol. 34 623-629 (1985).
Pergamon
0024-3205(94)00499-4
GLIPIZIDE STIMULATES SYMPATHETIC OUTFLOW IN DIABETES-PRONE MICE C.M. Kuhn, R.S. Surwit and M.N. Feinglos Departments of Pharmacology, Psychiatry and Medicine, Durham N.C. 27710.
Duke University Medical Center,
(Received in final form December 10, 1994) Summarv
The purpose of the present study was to determine if the oral hypoglycemic agent glipizide influenced sympathetic outflow in diabetes-prone mice. C57BL/6 (diabetes-prone) and diabetes-resistant (A/J) were treated with saline or glipizide, and sympathetic outflow determined by the fall in organ norepinephrine content after synthesis inhibition with cr-methyl-para-tyrosine. Sympathetic outflow to the liver and pancreas were slower in B1/6 mice than in control A/J. Glipizide increased sympathetic outflow to the pancreas in both strains of mice, but did not influence outflow to other organs significantly. The results of this study suggest that glipizide can influence central glucoregulatory mechanisms after peripheral administration. Key Wordr:
glipizide, diabetes, C57BL/6 mice, hypoglycemic
agent
The mechanism by which oral hypoglycemic agents decrease blood glucose is still controversial. Several actions have been described, most importantly enhanced insulin release and improvement in insulin action (1,2). One major action of these agents is thought to result from blockade of specific ATP-sensitive Kt channels on the beta cell of the pancreas which leads to beta cell depolarization and increased insulin release (3). However, the clinical benefit derived from these agents seems to exceed that predicted on the basis of these actions. The central nervous system has not been widely considered as a site of action, as these drugs show low lipophilicity and only slight entry into the brain. However, the recent description of a wide distribution within the CNS of the ATP-sensitive Kt channels on which these agents are thought to act (4) has raised the possibility that actions within the CNS might contribute to the observed clinical benefit derived from these agents. Furthermore, recent studies both & ti and in vitro suggest that the second generation sulfonylurea hypoglycemic agents can influence behavior through actions in the CNS (5,6). Despite the limitation of access, oral hypoglycemic agents could act within the CNS through at least two mechanisms. They could bind to ATP-sensitive K+ channels present in sites that are not behind the blood brain barrier. Alternatively, they could elicit reflex actions within the CNS derived from a peripheral mechanism of action that could ultimately influence insulin and/or glucagon release. Direct influence on parasympathetic or sympathetic control of insulin secretion and/or glucose release from the liver represent the two most likely mechanisms by which such drugs could act within the CNS to influence blood glucose. Numerous studies demonstrate an important role for both limbs of the autonomic nervous system in glucoregulation (7,8). Studies in this laboratory and elsewhere suggest that altered sympathetic outflow represents a particularly likely candidate (9,lO). In mice, as in most other mammals, the Corresponding author: Dr. Cynthia Kuhn, Dep’t of Pharmacology, Duke University Medical Center, Durham, NC 27710
662
Glipizide Action in Diabetes-prone Mice
Vol. 56, No. 9, 1995
sympathetic innervation of the pancreas stimulates glucagon secretion and inhibits insulin secretion via beta and alpha-2 adrenergic receptors respectively (9-13). Numerous studies support a role for the SNS in stress-induced hyperglycemia, which is thought to contribute to loss of glucose homeostasis in stable diabetics (14). Furthermore, enhanced sympathetic inhibition of insulin secretion has been postulated to contributed to the etiology of NIDDM. Patients with NIDDM have abnormal insulin responses to the alpha adrenergic antagonist phentolamine which suggest the presence of excessive sympathetic inhibition of insulin secretion (15). The purpose of the present study was to test the hypothesis that the oral hypoglycemic agent glipizide is capable of decreasing sympathetic outflow in mice. The ability of glipizide to decelerate the decrease in organ norepinephrine content following synthesis inhibition with alpha-methyl-para-tyrosine (AMPT) has been used to assess this possibility. In mice, both anatomic and neurochemical studies have shown that the majority (> 90%) of pancreatic norepinephrine content reflects innervation of the endocrine islets (17,18). Furthermore, evaluation of pancreatic NE content after synthesis inhibition has been shown to vary inversely with insulin secretion in several models with altered sympathetic outflow to the liver including animals with VMH lesions, diet-induced obesity and genetic models of obesity (18-22). Therefore,this strategy can provide useful insight into the activity We have compared the action of glipizide in a strain of mice, the C57BL/6, which develops diabetes when allowed to eat a diet high in fat and simple carbohydrate, to its actions in a strain of mice, the A/J, which does not develop diabetes when allowed to eat the same diet (23). Both strains were tested to evaluate the possibility that glipizide might exert greater actions in the strain (BL/6) that is prone to develop diabetes. Methods Animals: Twelve week-old mice of the C57BL/6 and A/J strain were obtained from Jackson Laboratories (Bar Harbor, MA). They were housed in a vivarium with a 12 hr lightdark cycle, and allowed to eat Purina rodent chow ad libitum. Drun Treatment: Animals were treated subcutaneously with saline or glipizide (1 mg/kg) followed immediately by saline or AMPT (250 mglkg) ip. The dose of glipizide was based on previous animal studies showing significant hypoglycemic actions after chronic administration of 1 mg/kg (24,25). Animals were killed by decapitation at t = 0 (immediately after injection), 1, 2 or 4 hrs later (immediately after injection) or 4 hrs later. These time points have been used to evaluate sympathetic nervous system activity in diabetic mice in previous studies (20,21,26). Heart, liver, pancreas and fat were frozen at -20” until assayed for norepinephrine. Blood glucose and insulin was not influenced by the single dose of glipizide received by these animals 4 hrs before collection (data not shown). Organ Noreoineohrine Analysis: Tissue norepinephrine content was determined by on-line trace enrichment high pressure liquid chromatography followed by electrochemical detection. Tissues were homogenized in 0.2 N perchloric acid containing 0.5 mM EDTA and 0.5 mM sodium metabisulfite, centrifuged at 20,000 x G for 30 minutes, and supernatant used for analysis. The internal standard dihydroxybenyzlamine was added to an aliquot of the PCA extract, which was neutralized by addition of Tris buffer (1M pH 8.4) and catecholamines extracted onto alumina. Adsorbed catecholamines were eluted with the same perchloric acid solution used for homogenization, and injected onto the HPLC.Catecholamines were adsorbed onto a 3 cm Nucleosil 5Sa cation exchange precolumn with mobile phase of 0.02 M citric acid, 0.02 M potassium acetate, 0.5 mM EDTA, and eluted onto the Spherisorb 0DS2 Cl8 reverse phase analytic column with a mobile phase of 0.02 M citric acid, 0.2 M potassium acetate, 0.5 mM EDTA, 0.65 mM octylsodium sulfate, 1% acetonitrile. NE was quantitated by comparison to a standard curve extracted and run in parallel with each set of samples,
Glipizide Action in Diabetes-prone Mice
Vol. 56, No. 9, 1995
463
and individual recovery corrected with the internal standard. Drugs: Glipizide was a gift of PRATT Pharmaceuticals, Division of Pfizer Pharmaceuticals. AMPT methyl ester, catecholamines and DHBA for standards were obtained from Sigma Chemical Company (St. Louis, MO). Statistics: Results are expressed as mean + SEM. Results were analyzed by 2 way ANOVA (strain x treatment, comparing vehicle, (t=O), AMPT and AMPT + glipizide). The rate of decline of NE content was analyzed by least squares analysis of the regression line calculated for fall of NE content between t=O and t = 4. Results The purpose of the first experiments was to conduct a time-course for norepinephrine depletion following synthesis inhibition with AMPT to determine the time period over which depletion was linear, and presumably reflective of neuronal impulse flow in catecholamine neurons. The two strains of mice were both tested to validate that the half life of norepinephrine was similar enough in the two that a a similar time period could be used to test the effects of glipizide in both strains. Mice were injected with AMPT-methyl ester in saline, and killed at t =0 (immediately after injection), or t = 1,2 or 4 hours. Norepinephrine content was evaluated at each time point, (N= Wgroup) and a linear regression analysis of the rate of depletion used to calculate the half life of norepinephrine in each organ. As shown in Figure 1, norepinephrine depletion was linear over four hours in all tissues observed. Table 1 shows the half lives calculated from the rates of depletion. The half life of norepinephrine in pancreas and liver of the C57 BL/6 animals was significantly slower than the half life observed in tissues of the A/J mice. In contrast, the half life of norepinephrine in heart was similar in both strains. Pancreas 400
Norepinephrine
Content .
B”6
l
AIJ
300 9 E 200 Y
0
1
2
3
4
5
Time (hrs,
Heart
Norepinephrine
Content
Fig. 1 Decline in Organ Norepinephrine after AMPT. Animals were killed at the indicated time, and as described in methods. Results are expressed as mean+ SEM. N = 8lgroup.
NE determined
664
Glipizide Action in Diabetes-prone
Vol. 56, No. 9, 1995
Mice
The results of this experiment indicated that norepinephrine depletion was linear for at least 4 hrs. This time was selected for evaluation of glipizide effects, as catecholamine depletion was linear throughout this time period, glipizide effects should be maintained after a single dose for at least this time period, and AMPT inhibition of tyrosine hydroxylation should still be maintained. Evaluation at shorter times would not permit accurate quantitation of norepinephrine depletion in organs with slow turnover rates (C57BL/6 liver, for example, while with longer time periods the inhibition of NE synthesis cannot be maintained. The effect of glipizide on the AMPT-induced NE decline in pancreas, liver and heart was assessed by measuring NE content at 4 hours after synthesis inhibition. As norepinephrine depletion was proceeding linearly during this time period, norepinephrine content at this time provides an approximation of effects on sympathetic outflow. As shown in Figure 2, norepinephrine depletion in the pancreas was greater in animals treated with glipizide than those treated with saline. However, glipizide did not accelerate turnover in other organs. This finding suggests that glipizide specifically accelerated sympathetic outflow to the pancreas in both diabetes prone C57BL/6 and non diabetes-prone A/J mice. 0 100
AMPT AMPT + Glip
80
A/J
A/J Pancreas
Liver
Bu6
Heart
Fig. 2 Tissue NE content 4 hours after treatment with AMPT or AMPT + glipizide. Animals were treated with AMPT and vehicle or glipizide as described above, and killed 4 hrs later. Organ NE content is expressed as mean + SEM. N = 8/group. * indicates p < .05 or better relative to AMPT-treated control. Discussion The results of this study were the opposite of that predicted: the oral hypoglycemic agent glipizide enhanced rather than diminished sympathetic outflow as assessed with AMPTinduced decline of NE. Theoretically, this action should suppress insulin secretion rather than enhance it, and elevate rather than lower blood glucose. Furthermore, stimulation of glucagon secretion could also result from such stimulation. Levels of glucoregulatory hormones could not be assessed accurately in the present experiment, as all animals were treated with AMPT, an agent which itself can significantly influence insulin secretion. If changes in insulin reflect the increased sympathetic output, then the effect reported here could oppose the clinical action of glipizide.
GlipizideActionin Diabetes-proneMice
Vol.56, No. 9, 1995
The specificity
of this action for pancreas supports the possibility
645
that glipizide
influences regulation of glucose homeostasis through actions in the CNS. It is unlikely that the
action observed in the present study represents a reflex sympathetic response to a rapid lowering of insulin by the peripheral actions of glipizide. Although hypoglycemia is a potent stimulus for sympathetic outflow (27), such a counterregulatory response would involve a global increase in sympathetic outflow rather than being restricted to pancreatic innervation (28). We cannot rule out the possibility, however that this action could represent a transient activation that is reversed with chronic treatment. The few animal studies reporting blood glucose levels in animal models of NIDDM report significant effects after chronic drug treatment, but the effects of a single dose are rarely reported. The strain difference in basal sympathetic tone represents the other significant finding of the present study. Substantially slower NE turnover was observed in pancreas and liver of the “diabetes-prone” C57BL16 mice than was observed in the “diabetes-resistant” A/J strain. This apparent sympathetic withdrawal is consistent with previous studies demonstrating dietinduced sympathetic withdrawal and supersensitivity of adrenergic receptor populations in liver and/or pancreas that produce hyperglycemia following administration of exogenous catecholamines (21,22,29,30). We have speculated previously that this relative sympathetic withdrawal might reverse markedly when animals are fed a high fat, high simple carbohydrate diet, and contribute to the hyperglycemia which results (9). One possibility suggested by the present finding is that the glipizide-induced increase in sympathetic outflow serves to desensitize adrenergic receptors in the pancreas and attenuate adrenergic inhibition of insulin release. This effect would occur in addition to any direct stimulatory effect of glipizide on insulin secretion and insulin sensitivity. In summary, the present results suggest that the oral hypoglycemic agent glipizide influences sympathetic outflow to the pancreas in mice. The enhanced sympathetic outflow observed with a single dose of glipizide could reflect a direct drug action on the brain. These findings suggest the CNS-mediated glucoregulatory events might contribute to the clinical actions of these agents. In future studies, we plan to address the important question of the contribution of this change to glucoregulation following administration of clinically-effective doses to diabetic animals, and to evaluate structure-activity study of various sulfonylureas. Acknowledgements: This research was supported by a grant from Pfizer Pharmaceuticals to MNF and MH 00303 to RS. TABLE 1 Half Life of Norepinephrine in Peripheral Organs Organ
A/J
C57BL/6
Pancreas
2.7 hrs
7.7 hrs*
Liver
2.2 hrs
6.7 hrs*
Heart
5.7 hrs
8.6 hrs
Animals were treated with AMPT, killed at various times and norepinephrine determined as described in methods. Half life was calculated from least squares regression line of norepinephrine disappearance. N = 9-12Jgroup. * indicates p < .05 relative to A/J.
666
Glipizide Action in Diabetes-prone
Mice
Vol. 56, No. 9, 1995
REFERENCES 1. H. BECK-NIELSEN, 0. HOTHER-NIELSEN, 0. PEDERSEN. Diabetic Med. 3 613-620 (1988). 2. J. JACKSON, R. BRESSLER. CLINICAL PHARMACOLOGY OF SULPHONYLUREA HYPOGLYCAEMIC AGENTS: PART 1. Drugs 22 21 l-245 (198 1). FM. ASHCROFT Ann. Rev. Neurosci. 1197-xX (1990). :: C. MOURRE, C. WIDMANN AND M LAZDUNSKI. Brain Res. 519 29-43 (1990) D. LEVESQUE AND S.A. GREENFIELD. Neuropharmacol. 30 359-365 (1991). :: M. AMALRIC, C. HEURTEAUX, A. NIEOULLON AND M LAZDUNSKI. Eur. J. Pharmacol. 213 71-77 (1991). 7. D. PORTE AND SC. WOODS in --a Diabetes Mellitus Eds. M. Ellenberg and H. Rifkin. Med. Exam. Publ. pp 267-294. (1983) L. LANDSBERG. Clin. Exp. Hyperten. 12 817-830 (1990) :: R.S. SURWIT AND M.N. FEINGLOS. sabetes Care 1183-85 (1988) 10. R.S. SHERWIN, H. SHAMOON, R. HENDLER, L. SACCA, N. EIGLER, AND M. WALESKY. Metabolism 29 1146-I 154 (1980) 11. R.E. MILLER Endocrine Reviews 2 471-494 (1981) 12. N. YAMAGUCHI Canad. J. Physiol. Pharmacol. 70 167-206 (1991) 13. B. AHREN, I. LUNDQUIST Horm. Met. Res. 19 600-603 (1987) 415-420 (1986) 14. T. NAKADATE, T NAKAKI, T. MURAKI, R. KATO Eur. J. Pharmacol. 65 421-424 (1980) 15. D.G. BRUCE, D.J. CHISHOLM, L.H. STORLIEN, E.W. KRAEGEN AND G.A. SMYTHE. Diabetologia 35 835-843 (1992). 16. V.L. BROADSTONE, M.A. PFEIFER, V. BAJAJ, J.I. STAGNER, E. SMOLS. Diabetes 36 932-937 (1987). 17. I. LUNDQUIST, B. AHREN, C. HANSSON AND R. HAKANSON. Pancreas 6 662667 (1989) 18. I. LUNDQUIST, L.E. ERICSON. Cell and Tissue Res. 193 73-85 (1978) 19. A, TAKAHASHI, Y. IKARASHI, H. ISHIMARU AND Y. MARUYAMA. Life Sci. 53 1567-1572 (1993) 20. B.E. LEVIN, J. TRISCARI, A.C. SULLIVAN Amer. J. Physiol. 244 R347-R355 (1983) 21. B.E. LEVIN, J. TRISCARI, A.C. SULLIVAN. Amer. J. Physiol. 245 R367-R371 (1983) 22. T. YOSHIDA, H. NISHIOKA, Y. NAKAMURA, M. KONDO. Metabolism 33 10601063 (1984) 23. R.S. SURWIT, CM. KUHN, C. COCHRANE, J.A. MCCUBBIN, M.N. FEINGLOS. Diabetes 37 1163-1167 (1988). 24. A.S. REDDI. C.A. VELASCO. P.R. REDDY. M.Y.. R.A. CAMERINI-DAVALOS. Exp. Molec. Pathol. 53 140-151 (1990). ’ 25. H.E. LEBOVITZ. Pharmacotherapy 5 63-77 (1985). 26. B.E. LEVIN, J. TRISCARI, S. HOGAN AND A.C. SULLIVAN. Am. J. Physiol. 252 R471-R478 (1987). 27. J.B. YOUNG, R.M. ROSA, L LANDSBERG. Am. J. Physiol. 247 E35E40 (1984). 28. B.E. LEVIN AND A.C. SULLIVAN. J. Aut. Nerv. Sys. 20 233-242 (1987). 29. A.W. KNEHANS AND D.R. ROMSOS. Am. J. Physiol. 244 E567-E574 (1983) 30. C.J. LYNCH, M/L. STEER, M.R. CONNORS, R.A. SCHATZ, R.C. DETH. Biochem. Pharmacol. 34 623-629 (1985).