Gen. Pharmac. Vol. 23, No. 4. pp. 769-773, 1992
0306-3623/92$5.00+ 0.00 Copyright © 1992PergamonPress Lid
Printed in Great Britain. All rights reserved
EFFECT OF LITHIUM ON GASTRO-INTESTINAL COMPLICATIONS IN ALLOXAN-DIABETIC RATS YusuF ~)ZT~RK,l* NURAY YILDIZO~LU-ARI,2 N1LGONALTAN3 and V. MELIHALTAN2 l Department of Pharmacology, Faculty of Pharmacy, University of Anadolu, 26470 Tepeba~l, Eski~ehir, 2Department of Pharmacology, Faculty of Pharmacy, University of Ankara, 06100 Tando[an, Ankara and 3Department of Biochemistry, Faculty of Medicine, University of Gazi, 06330 Etiler, Ankara, Turkey (Received 14 October 1991)
Abstract--1. Decreased fl-adrenergic and serotonergic responses have been reported in gastro-intestinal tract of experimentally diabetic rats. Effects of lithium on the decreased fl-adrenergic and serotonergic responsiveness of the gastro-intestinal tract due to diabetes were investigated using gastric fundus strips and proximal duodenum from alloxan diabetic rats. 2. A 6-day treatment with lithium chloride (2 mEq/kg i.p. in saline) normalized the decreased gastro-intestinal responses of the alloxan-diabetic rats, whereas the lithium treatment did not affect the elevated blood glucose levels due to experimental diabetes. 3. Furthermore, the lithium treatments of control and alloxan-diabetic rats did not alter the relaxing effect of manganese chloride on the isolated duodenum. 4. These results strongly suggest that the improving effect of lithium is not related to adenylate cyclase activation and may be as a consequence of its direct action on the diabetic gastro-intestinal smooth muscles.
INTRODUCTION Lithium salts have been used as a psychotropic agent in the treatment and prophylaxis of certain psychiatric disorders (Schou, 1976, 1989, 1990). There have been many clinical studies of the effect of lithium giving contradictory results on glucose tolerance. In manic-depressive patients improved, unchanged and impaired glucose tolerance during lithium treatment have been observed (Henninger and Mueller, 1970; Shopsin et al., 1972; Van der Velde, 1970). Nevertheless, reports of acute onset diabetes mellitus in patients receiving lithium and diabetes have been postulated (Craig et al., 1977; Johnston, 1977). Animal experiments revealed that lithium affects lipid and carbohydrate metabolism in rat liver (Vendsborg and Vilstrup, 1976; Viegut and Jefferson, 1990). Moreover, lithium therapy exerts diabetogenic effects in non-diabetic rats while it is not able to cause any progression in the diabetic state of the alloxandiabetic rats (Altan et al., 1986). In contrast, it has been suggested that lithium may have antidiabetic action and improve glucose tolerance of patients with mild, maturity-onset diabetes mellitus (Mannisto and Koivisto, 1972; Mellerup et al., 1972). Lithium ion has a weak action on carbohydrate metabolism that resembles somewhat that of insulin. It has been demonstrated that lithium, like insulin, stimulates both glucose utilization and glycogen synthesis in the isolated rat diaphragm and increases the incorporation of (14C)glucose into CO2, glycogen and lipid biosynthesis in isolated rat adipocytes. The effects of lithium ion on glucose oxidation and lipid synthesis are the results of lithium-enhanced glucose transport, *To whom all correspondence should be addressed.
since these effects have been blocked in the presence of cytochalasin B (Cheng et al., 1983). On the other hand, diabetes mellitus causes gastrointestinal complications such as dysphagla, abdominal pain, nausea and vomiting, malabsorption, fecal incontinence and diarrhea (Ogbonmaya and Arem, 1990; Yang et al., 1984). Disturbances of gastrointestinal tract are among the first of autonomic syndromes related to diabetes to achieve prominence, and, of these, "diabetic diarrhea" has attracted most attention (Bargen et al., 1936; Sheridan and Bailey, 1946). The diarrhea is often intermittent, sometimes alternating with periods of constipation. The major clinical problem concerns the conflict between diabetic diarrhea and constipation (Hosking et al., 1978; Katz and Spiro, 1966; Malins and French, 1957). Another diabetic complication related to the alimentary tract is atonic dilatation of the stomach (Kassander, 1958). Relatively few studies using animal models of diabetes have investigated the gastro-intestinal manifestations. Recently, decreased fl-adrenerglc and serotonergic responses in the gastro-intestinal tract of diabetic rats have been reported as long-term diabetic complications which are closely related to clinical manifestations of diabetes mellitus (Altan et al., 1987; Mathison and Davison, 1988; (~ztiirk et al.., 1990; Yfldizo~lu-An et al., 1988). These changes in the gastro-intestinal /~-adrenergic and serotonergic responsiveness have been attributed to a decrease in the number of fl-adrenoccptors and in the affufity of serotonin receptors resulting from diabetes. Nevertheless, no specific changes have been observed in the gastro-intestinal muscarinic responses of diabetic rats. Slight but not significant decrease in the gastrointestinal responses to manganese chloride has been 769
770
YUSUF OZTORK et al.
observed in the alloxan- a n d streptozotocin-diabetic rats as well (Altan et aL, 1987). It has been s h o w n t h a t the gastro-intestinal complications related to fladrenergic a n d serotonergic responsiveness are i m p r o v e d following in vivo insulin treatment. The decreased fl-adrenergic responses, b u t n o t serotonergic responses, are also i m p r o v e d following in vitro insulin t r e a t m e n t (Ytldlzo~lu-An et al., 1988). Despite the extensively investigated effects of insulin o n the diabetic complications, less is k n o w n a b o u t the effect o f lithium o n the long-term diabetic complications. Moreover, there is no report concerning the effect o f lithium o n the gastro-intestinal c o m p l i c a t i o n s o f diabetes mellitus. Therefore, this study was designed to investigate the effects o f lithium o n the long-term diabetic c o m p l i c a t i o n s of the gastro-intestinal tract using alloxan-diabetic rats as a n experimental model.
Rat stomach fundus strips
The rat stomach fundus strips were prepared in a manner consistent with a procedure described earlier (Altan et al., 1987; Yddizo~lu-An et al., 1988). Stomach fundus strips isolated from rats were set up at 37°C in a 10-ml organ bath containing Krebs' solution and aerated with 5 0 CO2 in oxygen. Before testing, the fundus strips were allowed to equilibrate for 60 min and during this period bathing fluid was changed every 15 min. In each assay two non-cumulative concentration-response curves were obtained with serotonin (Merck, Fed. Rep. Germany); the doses were administered at 10-min intervals. Then, the same concentration-response curves repeated for the stomach fundus strips from diabetic, lithium treated control and lithium treated diabetic rats. The responses of rat stomach fundus strips were expressed as a percentage of maximal contraction obtained by serotonin. The contractions of the rat fundus strips were recorded with the same system as that used for the relaxations of rat duodenum. The tension on the fundus strips was 1.0 g.
MATERIALS AND METHODS Animal preparatwns
Male Wistar rats (230-300 g) from a local strain were used. Rats were housed in a room with controlled temperature (21 + 2°C) and humidity (65-70%) care being taken of the day and night cycle. The rats were divided into 4 groups and housed separately in metabolic cages: group I, control; group 2, lithium-treated control; group 3, diabetic; group 4, lithium-treated diabetic. During the experiments, all rats had free access to food and water. Diabetes was produced by a single i.v. injection of alloxan monohydrate (Sigma, U.S.A.) (60 mg/kg/body wt) into lateral tail vein. Alloxan was dissolved in physiological saline. Control rats were injected with only the same physiological saline solution. On the other hand, 48 hr after diabetes induction, some of the control and diabetic rats received 2 mEq/kg lithium chloride (Sigma, U.S.A.) each morning for 8 weeks. At the end of this period, all rats killed by cervical dislocation for tissue isolation. In each case, the injection volume did not exceed 0.1 ml. Blood samples were taken by retro-orbital puncture after a 10 hr overnight fast for blood glucose and serum lithium measurements. Isolated rat duodenum
The isolated rat duodenum was prepared according to the method described earlier (Altan et al., 1987; Yfldlzo~lu-An et al., 1988). Briefly, the proximal 2crn of the duodenum from rats were removed after killing and kept in Krebs' solution at 6°C for 2 hr. Atropine sulfate (Sigma, U.S.A.) was added to Krebs' solution at the concentration of 0.143mmol/I. Then, the duodenum was suspended in a 10-ml organ bath filled with the atropinized Krebs' solution, at 31°C, and gassed with 5% CO 2 in oxygen. The suspended duodenum was allowed to equilibrate for 60rain. During this period, the tissue was rinsed every 15min. In each assay two non-cumulative concentration-response curves obtained with salbutamol hemisulfate (Sigma, U.S.A.) and manganese chloride (Merck, Fed. Rep. Germany) on the same tissue; the doses were administered at 10-min intervals. Later, the same concentration-response curves repeated for the duodena from diabetic, lithium treated control and lithium treated diabetic rats. The manganese chloride- and salbutamol-induced relaxations were expressed as a percentage of the maximal relaxations obtained by these relaxant agents in the duodenum from control rats. The relaxations of the isolated rat duodenum were recorded with an isotonic transducer (Ugo Basile, No. 7006) connected to a recording microdynamometer (Ugo Basile, No. 7050). The load on the tissue was 1 g.
Assessment o f diabetes and serum lithium levels In the present study, 24--48 hr after alloxan treatment, rats demonstrated characteristic symptoms of diabetes such as polyphagia, polydipsia, polyuria, stable hyperglycaemia etc. Blood glucose levels were regularly measured by glucose oxidase method (Fales et al., 1961). Serum lithium levels were determined by flame photometry (Nevius and Lachattin, 1965). Analysis o f data and statistics
To evaluate the effect of serotonin on the rat stomach fundus strips and the effects of salbutamol and manganese chloride on the isolated rat duodenum, apparent affinity (pD2) and intrinsic activity (~E) values were calculated (Ari6ns and Van Rossum, 1957; Ari6ns and Simonis, 1964) and linear regression analysis was applied (Goldstein, 1964). All values reported represent the mean of the results obtained from individual experiments. When indicated, the significance between the mean values was determined by one-way variance analysis (Finney, 1978). RESULTS Results o f the b l o o d glucose m e a s u r e m e n t s o f the 4 g r o u p s are s u m m a r i z e d in T a b l e 1. As can be seen in the table, b l o o d glucose levels were significantly e n h a n c e d after the lithium t r e a t m e n t o f control rats. A slight e n h a n c e m e n t in the b l o o d glucose levels was noticed in the alloxan-diabetic rats treated with lithium. However, this e n h a n c e m e n t was too small to be significant. In addition, cachexia was also observed in some o f the lithium-treated non-diabetic rats a n d it h a d n o effect in the s m o o t h muscle experiments. C o n c e n t r a t i o n - r e s p o n s e curves to serotonin obtained from the s t o m a c h fundus strips o f control, alloxan-diabetic, lithium-treated control a n d lithiumtreated diabetic rats are s h o w n in Fig. I. The contractile responses o f rat s t o m a c h fundus to serotonin were significantly reduced in alloxan-induced diabetes. The decrease in the s e r o t o n i n responses was not observed in the lithium-treated diabetic rats, while the lithium t r e a t m e n t in the same conditions showed n o significant effect on the s e r o t o n i n - i n d u c e d c o n t r a c t i o n s o f s t o m a c h f u n d u s strips from a g e - m a t c h e d control rats. T h e c o n c e n t r a t i o n - r e s p o n s e curves to m a n g a n e s e chloride o f the isolated d u o d e n u m from control,
Lithium and gastrointestinal diabetic complications Table I. Effectof lithiumchloride (2 mEq/kg i.p. in saline)on blood
(o)
glucose levels (mg/100ral) in diabetic and non-diabetic rats (values4- SEM, n = 8 in each case) Control
Control + Li
Diabetic
Diabetic + Li
105.0 + 8.4
170,0 4. 15.2
365.6 ± 14.3
357.8 +_ 10.7
alloxan-diabetic, lithium-treated control and lithiumtreated diabetic rats are shown in Fig. 2. As can be noticed in the figure, no significant change was determined resulting from both alloxan-induced diabetes and lithium therapy. In contrast, the relaxant responses to salbutamol were reduced depending on the experimentally-induced diabetes. It was found that lithium treatment reverses the changes in the responsiveness to salbutamol of rat duodenum due to alloxan-induced diabetes without showing any change in the relaxant responses to salbutamol of the nondiabetic rat duodenum. Table 2 summarizes apparent affinity (pD2) and intrinsic activity (aE) values calculated from control, alloxan-diabetic, lithium-treated control and lithium-treated diabetic rats.
771
IO0
== C
05o n-
'
1;o
'
Salbutamol doses (nM)
100
DISCUSSION
Gastroenteropathy is one of the long-term complications of diabetes mellitus (Hosking el aL, 1978; Katz and Spiro, 1966; Ogbonmaya and Arem, 1990; Yang et al., 1984). Experimental models of diabetes are increasingly being used for the investigations of gastro-intestinal complications as well as the other complications owing to diabetes mellitus. Megacolon has been reported in streptozotocin-diabetic rats (Nelson el al., 1976). Recently, decreased contractile responses to calcium have been observed in the K+-depolarized duodenum isolated from alloxandiabetic rats (Oztfirk el al., 1987). Decreased //adrenergic responsiveness in the gastro-intestinal tract of experimentally-diabetic rats has also been observed (Altan el al., 1987; Mathison and Davison, 1988; Oztiirk et al., 1990; Ylldlzo~lu-An et al., 1988). It has been proposed that the decreased gastrointestinal fl-adrenergic responsiveness might have resulted from the decrease in the number of /~-adrenergic receptors (Altan et al., 1987; Yddizo~luA n el al., 1988). The decreased fl-adrenergic responses in the rat gastro-intestinal tract is improved following in vitro and in vivo insulin treatments (Yddlzo~lu-An et al., 1988). Serotonergic responsiveness of the rat gastro-intestinal tract is also decreased depending on the alloxan and streptozotocin diabetes (Altan et al., 1987; Mathison and Davison, 1988; Yddmo~lu-An el al., 1988) and the decreased gastrointestinal responses to serotonin are improved following/n vivo insulin treatment (Yfldlzo~lu-An el al., 1988). In this study, mean serum lithium levels were found to be not higher than 0.80 mmol/i in all rats treated with lithium. By assuming that the therapeutic lithium levels in man (Baldessarini and Stephans, 1970; Cooper and Simpson, 1978) are very similar to those in experimental animals, it seems logical to accept that this study was carried out within therapeutic lithium concentration.
== C 0 CI.
50
0
re
o/lo
o. o
o.: o o. o
1. o
M a n g a n a s e chloride doses (raM)
(c) ,-, 100
O C o
el
50
o.;5
o'.4 o'.s 1'.6 6'.4 2 .2 s;.4 Serotonin doses (14M)
Fig. 1. Effects of lithium treatment on the relaxant action of
salbutamol (A) and manganese chloride (B) in the duodenum and on the contractile action of scrotonin (C) in the rat fundus strips. Control, IB; lithium treated control, 17; diabetic, 0 ; lithium treated diabetic, 0 ; (n = 8 in each c~se).
772
YUSUFOZ'I~RK et al. Table 2. Changes in pD2 and 0[ E values of the gastro-intestinal tract obtained from lithium treated- and non-treated diabetic and non-diabetic rats (values 5- SEM, n = 8 in each case) Isolated o r g a n Agonist Treatment pD2 ~E Rat fundus S e r o t o n i n Control 8.31 _+0.03 1.00 _+0.02 Control + Li 7.58 * 0. I0 1.02 _.+0.05 Alloxan 7.56 + 0.21 0.45 + 0.06 AIIoxan + Li 8.25 + 0.06 0.99 _.+0.04 Rat duodenum Salbutamol Control 6.18 5- 0.04 1.00 5- 0.03 Control + Li 6.19 + 0.08 1.03 5- 0.03 Alloxan 6.22 5- 0.09 0.56 + 0.06 AIIoxan + Li 6.16 5- 0.05 1.03 + 0.03 Manganese Control 3.50 _+0.04 1.00 _+0.06 Control + Li 3.46 + 0.06 1.04 + 0.04 AIIoxan 3.41 5-0.18 0.89_+0.11 Alloxan + Li 3.49 + 0.60 0.99 5- 0.01
Neither an improvement nor a progression was observed in the blood glucose levels of alloxandiabetic rats treated with lithium. This finding is also in agreement with our previous observations (Altan et aL, 1986). The exact mechanism of diabetogenic action elicited by lithium is not known yet. Recent investigations, however, have shown that lithium chloride potentiates tumor necrosis factor ( T N F ) induced and interleukin-I (IL-I) induced cytokine and cytokine receptor expression (Beyaert et al., 1991). Such a cytokine activation has been demonstrated to play an important role in pancreatic /%cell damage in the onset of autoimmune diabetes (Mandrup-Poulsen, 1990; Rabinovitch and Baquerizo, 1990). Cachexia which was observed in some of the lithium-treated rats is also an indirect evidence in favor of cytokinemediated E-cell damage. Since cachexia is one of the major symptoms o f T N F activation (Jones and Selby, 1989; Theologldes, 1986), it might be speculated that lithium-induced diabetogenic action may be closely related to the cytokine activation. The findings obtained in the present study also reveal that lithium has an insulin-like improving effect on the decreased fl-adrenergic and serotonergic responsiveness of the gastro-intestinal tract from alloxan-diabetic rats. The improvement in the rat gastro-intestinal tract following lithium treatment might be unrelated with the adenylate cyclase activation in the smooth muscles, since manganese chloride, an adenylate cyclase activator (Wiemer et al., 1978) caused relations o f rat d u o d e n u m from all experimental groups with the similar magnitude. There is a possibility that the improving effect o f lithium on the decreased serotonergic responsiveness o f the gastro-intestinal tract may be related to increased inositol phospholipid turnover in the smooth muscle. In fact, lithium increases inositol phospholipid turnover in various tissues (Rana and Hokin, 1990). It is well-known that inositol phospholipids play a central role in smooth muscle reactivity (Chilvers et al., 1990a, b). It would, thus, be of interest to investigate the possible relationship of phospholipid turnover with the insulin-like improving effect of lithium on the diabetic gastro-intestinal complications. Acknowledgement--The authors are grateful to Professor K. H. C. Ba~r, director of Medicinal Plant Research Center, Eski~hir for the computer facilities used in this study.
REFERENCES
Altan N., Arslan M., Altan V. M., Oztiirk Y. and Daryavuz M. (1986) The effect of lithium on carbohydrate metabolism in normal and alloxan-diabetic rats. In Balkan Contribution to Endocrinology and Metabolism (Edited by Kolo~lu S. and Kolo~,lu L. B.), Vol. 1, pp. 373-377. Ankara University Press, Ankara. Altan V. M., Ytldtzo~lu N. and Oztflrk Y. (1987) Decreased gastro-intestinal responses to certain agonists in streptozotocin- and alloxan-diabetic rats in vitro. Pharmacology 34, 143-148. Ariens E. J. and Simonis A. M. (1964) Drug receptor interaction: interaction of one or more drugs with a receptor system. In Molecular Pharmacology (Edited by Ariens E. J.), Vol. 1, pp. 119-286. Academic Press, New York. Ariens E. J. and Van Rossum J. M. (1957) pD,, pA~ and pD'~ values in analysis of pharmacodynamics. Archs int. Pharmacodyn. Thbr. 110, 275-299. Baldessarini R. J. and Stephans J. H. (1970) Clinical pharmacology and toxicology of lithium salts. Archs Gen. Psychiat. 22, 72-80. Bargen J. A., Bollman J. L. and Kepler E. J. (1936) "Diarrhea and diabetes" and statorrhea of pancreatic insufficiency. Mayo Clin. Proc. !1, 737-742. Beyaert R., Schulzeosthoff K., Vanroy F. and Fiers W. (199 I) Lithium chloride potentiates tumor necrosis factor induced and interleukin-induced cytokine and cytokine receptor activation. Cytokine 3, 284-291. Cheng K., Creacy S. and Lamer J. (1983) qnsulin-like' effects of lithium ion on isolated rat adipocytes. II. Specific activation of glycogen synthase. Molec. Cell. Biochem. 56, 183-188. Chilvers E. R., Batty I. H., Barnes P. J. and Nahorski S. R. (1990a) Formation of inositol polyphosphates in airway smooth muscle after muscarinic receptor stimulation. J. Pharmac. exp. Ther. 252, 786-791. Chilvers E. R., Challiss R. A. J., Willcocks A. L., Potter B. V. L., Barnes P. J. and Nahorski S. R. (1990b) Characterization of stereospecific binding sites for inositol 1,4,5triphosphate in airway smooth muscle. Br. J. Pharmac. 99, 297-302. Cooper T. B. and Simpson G. M. (1978) Kinetics of lithium and clinical response. In Psychopharmacology: A Generation of Progress (Edited by Lipton M. A., Di Mascio A. and Killam K. F.), pp. 923-930. Raven Press, New York. Craig J., Abu-Saleh M., Smith B. and Evans I. (1977) Diabetes mellitus in patients on lithium. Lancet ii, 1028. Fales F. W., Russel J. A. and Fain J. N. (1961) Some applications and limitations of the enzymatic, reducing (Somogyi) and anthrone methods for estimating sugars. Clin. Chem. 7, 389-395. Finney D. J. (1978) Statistical Methods in Biological Assay. Griffin, London. Goldstein A. (1964) Biostalistics. An Introductory Text. Macmillan, New York.
Lithium and gastrointestinal diabetic complications Henninger G. and Mueller P. S. (1970) Carbohydrate metabolism in mania. Archs Gen. Psychiat. 23, 310-316. Hosking D. J., Bennett T. and Hampton J. R. (1978) Diabetic autonomic neuropathy. Diabetes 27, 1043-1055. Johnston B. B. (1977) Diabetes mellitus in patients on lithium. Lancet il, 935-936. Jones A. L. and Selby P. (1989) Clinical applications of turnout necrosis factor. Prog. Growth Factor Res. !, 107-122. Kassander P. (1958) Asymptomatic gastric retention of diabetes (gastroparesis diabeticorum). Ann. intern. Med. 48, 797-812. Katz L. A. and Spiro H. (1966) Gastrointestinal manifestations of diabetes. New Engl. J. Med. 275, 1350-1361. Malins J. M. and French J. M. (1957) Diabetic diarrhea. Q. JI Med. 26, 461-480. Mandrup-Poulsen T. 0990) Cytokine-mediated beta-cell destruction: the molecular effector mechanism causing IDDM (Abstract). J. Autoimmun. 3, 77. Mannisto P. and Koivisto V. (1972) Antidiabetic effects of lithium. Lancet fl, 1031. Mathison R. and Davison J. S. (1988) Modified smooth muscle responses of jejunum in streptozotocin-diabetic rats. J. Pharmac. exp. Ther. 244, 1045-1050. Mellerup E. T., Plenge P., Vandsborg P., Rafaelson O. J., Kjeldsen H, and Agerbaek H. (1972) Antidiabetic effect of lithium. Lancet il, 1367-1368. Nelson J. S., Lacy P. E. and Hirschberg G. E. (1976) Megacolon and autonomic neuropathy in diabetic rats. J. Neuropath. exp. Neurol. 35, 335. Nevius D. B. and Lachattin G. F. (1965) Operation of the technic on flame photometer with natural gas. Use in the determination of serum lithium and the semi-micro analysis of serum sodium and potassium. Clin. Chem. 11, 633-639. Ogbonmaya K. I. and Arem R. (1990) Diabetic diarrhea: pathophysiology, diagnosis and management. Archs intern. Med. 150, 262-267. Oztfirk Y., Yfldlzo~,lu N., Altan V. M. and Karasu (~. (1987) Altered responses to calcium and trifluoperazine in K +-depolarized duodenum from alloxan-diabetic rats. Pharmac. Res. Commun. 19, 597--608.
773
Oztiirk Y., Ylldlzo~lu-An N., Oziian A., Oz~likay A. T. and Altan V. M. (1990) Decreased//-adrenergic responses of rat small intestine due to non-insulin-dependent diabetes. Diabetes Res. clin. Practice 9, 123-127. Rabinovitch A. and Baquerizo H. (1990) Immunological mechanisms of pancreatic //-cell destruction: effects of cytokines (Abstract). J. Autoonmun. 3, 85. Rana R. S. and Hokin L. E. (1990) Role of phosphoinositides in transmembrane signalling Physiol. Rev. 70, 115-164. Schou M. (1976) Pharmacology and toxicology of lithium. A. Rev. Pharmac. Toxic. 16, 231-243. Schou M. (1989) Lithium prophylaxis: myths and realities. Am. J. Psychiat. 146, 573-576. Schou M. (1990) Lithium and treatment-resistant depressions: a review. Lithium 1, 3-8. Sheridan E. P. and Bailey C. C. (1946) Diabetic nocturnal diarrhea. J. Am. med. Ass. 130, 632-634. Shopsin B., Stern S. and Gershon S. (1972) Altered carbohydrate metabolism during treatment with lithium carbonate. Archs Gen. Psychiat. 26, 566-571. Theologides A. (1986) Cachexia is a feature of cancer patients. Am. J. Med. 81, 696-698. Van der Velde C. D. (1970) Effectiveness of lithium carbonate in the treatment of manic depressive illness. Am. J. Psychiat. 127, 345-351. Vendsborg P. B. and Vilstrup H. (1976) The influence of lithium on carbohydrate and lipid metabolism in the perfused rat liver. Acta pharmac, tox. 38, 10-16. Viegut V. and Jefferson J. W. (1990) Lithium and liver. Lithium 1, 9-13. Wiemer G., Kaiser G. and Palm D. (1978) Effects of Mg 2+, Mn 2* and Ca 2+ on adenylate cyclase activity: evidence for a metallic site. Archs Pharmac. 303, 145-152. Yang R., Arem R. and Chan U (1984) Gastrointestinal tract complications of diabetes mellitus. Archs intern. Med. 144, 1251-1254. Yildlzo~lu-An N., Altinkurt O., Oztfirk Y,, Altan V. M., Pekiner C. and Karasu (~. (1988) Decreased gastro-intestinal responses to salbutamol and serotonin in streptozotocin-induced diabetes: improving effect of insulin in vivo and in vitro. Gen. Pharmac. 19, 665-668.
0306-3623/92$5.00+ 0.00 Copyright © 1992PergamonPress Lid
Printed in Great Britain. All rights reserved
EFFECT OF LITHIUM ON GASTRO-INTESTINAL COMPLICATIONS IN ALLOXAN-DIABETIC RATS YusuF ~)ZT~RK,l* NURAY YILDIZO~LU-ARI,2 N1LGONALTAN3 and V. MELIHALTAN2 l Department of Pharmacology, Faculty of Pharmacy, University of Anadolu, 26470 Tepeba~l, Eski~ehir, 2Department of Pharmacology, Faculty of Pharmacy, University of Ankara, 06100 Tando[an, Ankara and 3Department of Biochemistry, Faculty of Medicine, University of Gazi, 06330 Etiler, Ankara, Turkey (Received 14 October 1991)
Abstract--1. Decreased fl-adrenergic and serotonergic responses have been reported in gastro-intestinal tract of experimentally diabetic rats. Effects of lithium on the decreased fl-adrenergic and serotonergic responsiveness of the gastro-intestinal tract due to diabetes were investigated using gastric fundus strips and proximal duodenum from alloxan diabetic rats. 2. A 6-day treatment with lithium chloride (2 mEq/kg i.p. in saline) normalized the decreased gastro-intestinal responses of the alloxan-diabetic rats, whereas the lithium treatment did not affect the elevated blood glucose levels due to experimental diabetes. 3. Furthermore, the lithium treatments of control and alloxan-diabetic rats did not alter the relaxing effect of manganese chloride on the isolated duodenum. 4. These results strongly suggest that the improving effect of lithium is not related to adenylate cyclase activation and may be as a consequence of its direct action on the diabetic gastro-intestinal smooth muscles.
INTRODUCTION Lithium salts have been used as a psychotropic agent in the treatment and prophylaxis of certain psychiatric disorders (Schou, 1976, 1989, 1990). There have been many clinical studies of the effect of lithium giving contradictory results on glucose tolerance. In manic-depressive patients improved, unchanged and impaired glucose tolerance during lithium treatment have been observed (Henninger and Mueller, 1970; Shopsin et al., 1972; Van der Velde, 1970). Nevertheless, reports of acute onset diabetes mellitus in patients receiving lithium and diabetes have been postulated (Craig et al., 1977; Johnston, 1977). Animal experiments revealed that lithium affects lipid and carbohydrate metabolism in rat liver (Vendsborg and Vilstrup, 1976; Viegut and Jefferson, 1990). Moreover, lithium therapy exerts diabetogenic effects in non-diabetic rats while it is not able to cause any progression in the diabetic state of the alloxandiabetic rats (Altan et al., 1986). In contrast, it has been suggested that lithium may have antidiabetic action and improve glucose tolerance of patients with mild, maturity-onset diabetes mellitus (Mannisto and Koivisto, 1972; Mellerup et al., 1972). Lithium ion has a weak action on carbohydrate metabolism that resembles somewhat that of insulin. It has been demonstrated that lithium, like insulin, stimulates both glucose utilization and glycogen synthesis in the isolated rat diaphragm and increases the incorporation of (14C)glucose into CO2, glycogen and lipid biosynthesis in isolated rat adipocytes. The effects of lithium ion on glucose oxidation and lipid synthesis are the results of lithium-enhanced glucose transport, *To whom all correspondence should be addressed.
since these effects have been blocked in the presence of cytochalasin B (Cheng et al., 1983). On the other hand, diabetes mellitus causes gastrointestinal complications such as dysphagla, abdominal pain, nausea and vomiting, malabsorption, fecal incontinence and diarrhea (Ogbonmaya and Arem, 1990; Yang et al., 1984). Disturbances of gastrointestinal tract are among the first of autonomic syndromes related to diabetes to achieve prominence, and, of these, "diabetic diarrhea" has attracted most attention (Bargen et al., 1936; Sheridan and Bailey, 1946). The diarrhea is often intermittent, sometimes alternating with periods of constipation. The major clinical problem concerns the conflict between diabetic diarrhea and constipation (Hosking et al., 1978; Katz and Spiro, 1966; Malins and French, 1957). Another diabetic complication related to the alimentary tract is atonic dilatation of the stomach (Kassander, 1958). Relatively few studies using animal models of diabetes have investigated the gastro-intestinal manifestations. Recently, decreased fl-adrenerglc and serotonergic responses in the gastro-intestinal tract of diabetic rats have been reported as long-term diabetic complications which are closely related to clinical manifestations of diabetes mellitus (Altan et al., 1987; Mathison and Davison, 1988; (~ztiirk et al.., 1990; Yfldizo~lu-An et al., 1988). These changes in the gastro-intestinal /~-adrenergic and serotonergic responsiveness have been attributed to a decrease in the number of fl-adrenoccptors and in the affufity of serotonin receptors resulting from diabetes. Nevertheless, no specific changes have been observed in the gastro-intestinal muscarinic responses of diabetic rats. Slight but not significant decrease in the gastrointestinal responses to manganese chloride has been 769
770
YUSUF OZTORK et al.
observed in the alloxan- a n d streptozotocin-diabetic rats as well (Altan et aL, 1987). It has been s h o w n t h a t the gastro-intestinal complications related to fladrenergic a n d serotonergic responsiveness are i m p r o v e d following in vivo insulin treatment. The decreased fl-adrenergic responses, b u t n o t serotonergic responses, are also i m p r o v e d following in vitro insulin t r e a t m e n t (Ytldlzo~lu-An et al., 1988). Despite the extensively investigated effects of insulin o n the diabetic complications, less is k n o w n a b o u t the effect o f lithium o n the long-term diabetic complications. Moreover, there is no report concerning the effect o f lithium o n the gastro-intestinal c o m p l i c a t i o n s o f diabetes mellitus. Therefore, this study was designed to investigate the effects o f lithium o n the long-term diabetic c o m p l i c a t i o n s of the gastro-intestinal tract using alloxan-diabetic rats as a n experimental model.
Rat stomach fundus strips
The rat stomach fundus strips were prepared in a manner consistent with a procedure described earlier (Altan et al., 1987; Yddizo~lu-An et al., 1988). Stomach fundus strips isolated from rats were set up at 37°C in a 10-ml organ bath containing Krebs' solution and aerated with 5 0 CO2 in oxygen. Before testing, the fundus strips were allowed to equilibrate for 60 min and during this period bathing fluid was changed every 15 min. In each assay two non-cumulative concentration-response curves were obtained with serotonin (Merck, Fed. Rep. Germany); the doses were administered at 10-min intervals. Then, the same concentration-response curves repeated for the stomach fundus strips from diabetic, lithium treated control and lithium treated diabetic rats. The responses of rat stomach fundus strips were expressed as a percentage of maximal contraction obtained by serotonin. The contractions of the rat fundus strips were recorded with the same system as that used for the relaxations of rat duodenum. The tension on the fundus strips was 1.0 g.
MATERIALS AND METHODS Animal preparatwns
Male Wistar rats (230-300 g) from a local strain were used. Rats were housed in a room with controlled temperature (21 + 2°C) and humidity (65-70%) care being taken of the day and night cycle. The rats were divided into 4 groups and housed separately in metabolic cages: group I, control; group 2, lithium-treated control; group 3, diabetic; group 4, lithium-treated diabetic. During the experiments, all rats had free access to food and water. Diabetes was produced by a single i.v. injection of alloxan monohydrate (Sigma, U.S.A.) (60 mg/kg/body wt) into lateral tail vein. Alloxan was dissolved in physiological saline. Control rats were injected with only the same physiological saline solution. On the other hand, 48 hr after diabetes induction, some of the control and diabetic rats received 2 mEq/kg lithium chloride (Sigma, U.S.A.) each morning for 8 weeks. At the end of this period, all rats killed by cervical dislocation for tissue isolation. In each case, the injection volume did not exceed 0.1 ml. Blood samples were taken by retro-orbital puncture after a 10 hr overnight fast for blood glucose and serum lithium measurements. Isolated rat duodenum
The isolated rat duodenum was prepared according to the method described earlier (Altan et al., 1987; Yfldlzo~lu-An et al., 1988). Briefly, the proximal 2crn of the duodenum from rats were removed after killing and kept in Krebs' solution at 6°C for 2 hr. Atropine sulfate (Sigma, U.S.A.) was added to Krebs' solution at the concentration of 0.143mmol/I. Then, the duodenum was suspended in a 10-ml organ bath filled with the atropinized Krebs' solution, at 31°C, and gassed with 5% CO 2 in oxygen. The suspended duodenum was allowed to equilibrate for 60rain. During this period, the tissue was rinsed every 15min. In each assay two non-cumulative concentration-response curves obtained with salbutamol hemisulfate (Sigma, U.S.A.) and manganese chloride (Merck, Fed. Rep. Germany) on the same tissue; the doses were administered at 10-min intervals. Later, the same concentration-response curves repeated for the duodena from diabetic, lithium treated control and lithium treated diabetic rats. The manganese chloride- and salbutamol-induced relaxations were expressed as a percentage of the maximal relaxations obtained by these relaxant agents in the duodenum from control rats. The relaxations of the isolated rat duodenum were recorded with an isotonic transducer (Ugo Basile, No. 7006) connected to a recording microdynamometer (Ugo Basile, No. 7050). The load on the tissue was 1 g.
Assessment o f diabetes and serum lithium levels In the present study, 24--48 hr after alloxan treatment, rats demonstrated characteristic symptoms of diabetes such as polyphagia, polydipsia, polyuria, stable hyperglycaemia etc. Blood glucose levels were regularly measured by glucose oxidase method (Fales et al., 1961). Serum lithium levels were determined by flame photometry (Nevius and Lachattin, 1965). Analysis o f data and statistics
To evaluate the effect of serotonin on the rat stomach fundus strips and the effects of salbutamol and manganese chloride on the isolated rat duodenum, apparent affinity (pD2) and intrinsic activity (~E) values were calculated (Ari6ns and Van Rossum, 1957; Ari6ns and Simonis, 1964) and linear regression analysis was applied (Goldstein, 1964). All values reported represent the mean of the results obtained from individual experiments. When indicated, the significance between the mean values was determined by one-way variance analysis (Finney, 1978). RESULTS Results o f the b l o o d glucose m e a s u r e m e n t s o f the 4 g r o u p s are s u m m a r i z e d in T a b l e 1. As can be seen in the table, b l o o d glucose levels were significantly e n h a n c e d after the lithium t r e a t m e n t o f control rats. A slight e n h a n c e m e n t in the b l o o d glucose levels was noticed in the alloxan-diabetic rats treated with lithium. However, this e n h a n c e m e n t was too small to be significant. In addition, cachexia was also observed in some o f the lithium-treated non-diabetic rats a n d it h a d n o effect in the s m o o t h muscle experiments. C o n c e n t r a t i o n - r e s p o n s e curves to serotonin obtained from the s t o m a c h fundus strips o f control, alloxan-diabetic, lithium-treated control a n d lithiumtreated diabetic rats are s h o w n in Fig. I. The contractile responses o f rat s t o m a c h fundus to serotonin were significantly reduced in alloxan-induced diabetes. The decrease in the s e r o t o n i n responses was not observed in the lithium-treated diabetic rats, while the lithium t r e a t m e n t in the same conditions showed n o significant effect on the s e r o t o n i n - i n d u c e d c o n t r a c t i o n s o f s t o m a c h f u n d u s strips from a g e - m a t c h e d control rats. T h e c o n c e n t r a t i o n - r e s p o n s e curves to m a n g a n e s e chloride o f the isolated d u o d e n u m from control,
Lithium and gastrointestinal diabetic complications Table I. Effectof lithiumchloride (2 mEq/kg i.p. in saline)on blood
(o)
glucose levels (mg/100ral) in diabetic and non-diabetic rats (values4- SEM, n = 8 in each case) Control
Control + Li
Diabetic
Diabetic + Li
105.0 + 8.4
170,0 4. 15.2
365.6 ± 14.3
357.8 +_ 10.7
alloxan-diabetic, lithium-treated control and lithiumtreated diabetic rats are shown in Fig. 2. As can be noticed in the figure, no significant change was determined resulting from both alloxan-induced diabetes and lithium therapy. In contrast, the relaxant responses to salbutamol were reduced depending on the experimentally-induced diabetes. It was found that lithium treatment reverses the changes in the responsiveness to salbutamol of rat duodenum due to alloxan-induced diabetes without showing any change in the relaxant responses to salbutamol of the nondiabetic rat duodenum. Table 2 summarizes apparent affinity (pD2) and intrinsic activity (aE) values calculated from control, alloxan-diabetic, lithium-treated control and lithium-treated diabetic rats.
771
IO0
== C
05o n-
'
1;o
'
Salbutamol doses (nM)
100
DISCUSSION
Gastroenteropathy is one of the long-term complications of diabetes mellitus (Hosking el aL, 1978; Katz and Spiro, 1966; Ogbonmaya and Arem, 1990; Yang et al., 1984). Experimental models of diabetes are increasingly being used for the investigations of gastro-intestinal complications as well as the other complications owing to diabetes mellitus. Megacolon has been reported in streptozotocin-diabetic rats (Nelson el al., 1976). Recently, decreased contractile responses to calcium have been observed in the K+-depolarized duodenum isolated from alloxandiabetic rats (Oztfirk el al., 1987). Decreased //adrenergic responsiveness in the gastro-intestinal tract of experimentally-diabetic rats has also been observed (Altan el al., 1987; Mathison and Davison, 1988; Oztiirk et al., 1990; Ylldlzo~lu-An et al., 1988). It has been proposed that the decreased gastrointestinal fl-adrenergic responsiveness might have resulted from the decrease in the number of /~-adrenergic receptors (Altan et al., 1987; Yddizo~luA n el al., 1988). The decreased fl-adrenergic responses in the rat gastro-intestinal tract is improved following in vitro and in vivo insulin treatments (Yddlzo~lu-An et al., 1988). Serotonergic responsiveness of the rat gastro-intestinal tract is also decreased depending on the alloxan and streptozotocin diabetes (Altan et al., 1987; Mathison and Davison, 1988; Yddmo~lu-An el al., 1988) and the decreased gastrointestinal responses to serotonin are improved following/n vivo insulin treatment (Yfldlzo~lu-An el al., 1988). In this study, mean serum lithium levels were found to be not higher than 0.80 mmol/i in all rats treated with lithium. By assuming that the therapeutic lithium levels in man (Baldessarini and Stephans, 1970; Cooper and Simpson, 1978) are very similar to those in experimental animals, it seems logical to accept that this study was carried out within therapeutic lithium concentration.
== C 0 CI.
50
0
re
o/lo
o. o
o.: o o. o
1. o
M a n g a n a s e chloride doses (raM)
(c) ,-, 100
O C o
el
50
o.;5
o'.4 o'.s 1'.6 6'.4 2 .2 s;.4 Serotonin doses (14M)
Fig. 1. Effects of lithium treatment on the relaxant action of
salbutamol (A) and manganese chloride (B) in the duodenum and on the contractile action of scrotonin (C) in the rat fundus strips. Control, IB; lithium treated control, 17; diabetic, 0 ; lithium treated diabetic, 0 ; (n = 8 in each c~se).
772
YUSUFOZ'I~RK et al. Table 2. Changes in pD2 and 0[ E values of the gastro-intestinal tract obtained from lithium treated- and non-treated diabetic and non-diabetic rats (values 5- SEM, n = 8 in each case) Isolated o r g a n Agonist Treatment pD2 ~E Rat fundus S e r o t o n i n Control 8.31 _+0.03 1.00 _+0.02 Control + Li 7.58 * 0. I0 1.02 _.+0.05 Alloxan 7.56 + 0.21 0.45 + 0.06 AIIoxan + Li 8.25 + 0.06 0.99 _.+0.04 Rat duodenum Salbutamol Control 6.18 5- 0.04 1.00 5- 0.03 Control + Li 6.19 + 0.08 1.03 5- 0.03 Alloxan 6.22 5- 0.09 0.56 + 0.06 AIIoxan + Li 6.16 5- 0.05 1.03 + 0.03 Manganese Control 3.50 _+0.04 1.00 _+0.06 Control + Li 3.46 + 0.06 1.04 + 0.04 AIIoxan 3.41 5-0.18 0.89_+0.11 Alloxan + Li 3.49 + 0.60 0.99 5- 0.01
Neither an improvement nor a progression was observed in the blood glucose levels of alloxandiabetic rats treated with lithium. This finding is also in agreement with our previous observations (Altan et aL, 1986). The exact mechanism of diabetogenic action elicited by lithium is not known yet. Recent investigations, however, have shown that lithium chloride potentiates tumor necrosis factor ( T N F ) induced and interleukin-I (IL-I) induced cytokine and cytokine receptor expression (Beyaert et al., 1991). Such a cytokine activation has been demonstrated to play an important role in pancreatic /%cell damage in the onset of autoimmune diabetes (Mandrup-Poulsen, 1990; Rabinovitch and Baquerizo, 1990). Cachexia which was observed in some of the lithium-treated rats is also an indirect evidence in favor of cytokinemediated E-cell damage. Since cachexia is one of the major symptoms o f T N F activation (Jones and Selby, 1989; Theologldes, 1986), it might be speculated that lithium-induced diabetogenic action may be closely related to the cytokine activation. The findings obtained in the present study also reveal that lithium has an insulin-like improving effect on the decreased fl-adrenergic and serotonergic responsiveness of the gastro-intestinal tract from alloxan-diabetic rats. The improvement in the rat gastro-intestinal tract following lithium treatment might be unrelated with the adenylate cyclase activation in the smooth muscles, since manganese chloride, an adenylate cyclase activator (Wiemer et al., 1978) caused relations o f rat d u o d e n u m from all experimental groups with the similar magnitude. There is a possibility that the improving effect o f lithium on the decreased serotonergic responsiveness o f the gastro-intestinal tract may be related to increased inositol phospholipid turnover in the smooth muscle. In fact, lithium increases inositol phospholipid turnover in various tissues (Rana and Hokin, 1990). It is well-known that inositol phospholipids play a central role in smooth muscle reactivity (Chilvers et al., 1990a, b). It would, thus, be of interest to investigate the possible relationship of phospholipid turnover with the insulin-like improving effect of lithium on the diabetic gastro-intestinal complications. Acknowledgement--The authors are grateful to Professor K. H. C. Ba~r, director of Medicinal Plant Research Center, Eski~hir for the computer facilities used in this study.
REFERENCES
Altan N., Arslan M., Altan V. M., Oztiirk Y. and Daryavuz M. (1986) The effect of lithium on carbohydrate metabolism in normal and alloxan-diabetic rats. In Balkan Contribution to Endocrinology and Metabolism (Edited by Kolo~lu S. and Kolo~,lu L. B.), Vol. 1, pp. 373-377. Ankara University Press, Ankara. Altan V. M., Ytldtzo~lu N. and Oztflrk Y. (1987) Decreased gastro-intestinal responses to certain agonists in streptozotocin- and alloxan-diabetic rats in vitro. Pharmacology 34, 143-148. Ariens E. J. and Simonis A. M. (1964) Drug receptor interaction: interaction of one or more drugs with a receptor system. In Molecular Pharmacology (Edited by Ariens E. J.), Vol. 1, pp. 119-286. Academic Press, New York. Ariens E. J. and Van Rossum J. M. (1957) pD,, pA~ and pD'~ values in analysis of pharmacodynamics. Archs int. Pharmacodyn. Thbr. 110, 275-299. Baldessarini R. J. and Stephans J. H. (1970) Clinical pharmacology and toxicology of lithium salts. Archs Gen. Psychiat. 22, 72-80. Bargen J. A., Bollman J. L. and Kepler E. J. (1936) "Diarrhea and diabetes" and statorrhea of pancreatic insufficiency. Mayo Clin. Proc. !1, 737-742. Beyaert R., Schulzeosthoff K., Vanroy F. and Fiers W. (199 I) Lithium chloride potentiates tumor necrosis factor induced and interleukin-induced cytokine and cytokine receptor activation. Cytokine 3, 284-291. Cheng K., Creacy S. and Lamer J. (1983) qnsulin-like' effects of lithium ion on isolated rat adipocytes. II. Specific activation of glycogen synthase. Molec. Cell. Biochem. 56, 183-188. Chilvers E. R., Batty I. H., Barnes P. J. and Nahorski S. R. (1990a) Formation of inositol polyphosphates in airway smooth muscle after muscarinic receptor stimulation. J. Pharmac. exp. Ther. 252, 786-791. Chilvers E. R., Challiss R. A. J., Willcocks A. L., Potter B. V. L., Barnes P. J. and Nahorski S. R. (1990b) Characterization of stereospecific binding sites for inositol 1,4,5triphosphate in airway smooth muscle. Br. J. Pharmac. 99, 297-302. Cooper T. B. and Simpson G. M. (1978) Kinetics of lithium and clinical response. In Psychopharmacology: A Generation of Progress (Edited by Lipton M. A., Di Mascio A. and Killam K. F.), pp. 923-930. Raven Press, New York. Craig J., Abu-Saleh M., Smith B. and Evans I. (1977) Diabetes mellitus in patients on lithium. Lancet ii, 1028. Fales F. W., Russel J. A. and Fain J. N. (1961) Some applications and limitations of the enzymatic, reducing (Somogyi) and anthrone methods for estimating sugars. Clin. Chem. 7, 389-395. Finney D. J. (1978) Statistical Methods in Biological Assay. Griffin, London. Goldstein A. (1964) Biostalistics. An Introductory Text. Macmillan, New York.
Lithium and gastrointestinal diabetic complications Henninger G. and Mueller P. S. (1970) Carbohydrate metabolism in mania. Archs Gen. Psychiat. 23, 310-316. Hosking D. J., Bennett T. and Hampton J. R. (1978) Diabetic autonomic neuropathy. Diabetes 27, 1043-1055. Johnston B. B. (1977) Diabetes mellitus in patients on lithium. Lancet il, 935-936. Jones A. L. and Selby P. (1989) Clinical applications of turnout necrosis factor. Prog. Growth Factor Res. !, 107-122. Kassander P. (1958) Asymptomatic gastric retention of diabetes (gastroparesis diabeticorum). Ann. intern. Med. 48, 797-812. Katz L. A. and Spiro H. (1966) Gastrointestinal manifestations of diabetes. New Engl. J. Med. 275, 1350-1361. Malins J. M. and French J. M. (1957) Diabetic diarrhea. Q. JI Med. 26, 461-480. Mandrup-Poulsen T. 0990) Cytokine-mediated beta-cell destruction: the molecular effector mechanism causing IDDM (Abstract). J. Autoimmun. 3, 77. Mannisto P. and Koivisto V. (1972) Antidiabetic effects of lithium. Lancet fl, 1031. Mathison R. and Davison J. S. (1988) Modified smooth muscle responses of jejunum in streptozotocin-diabetic rats. J. Pharmac. exp. Ther. 244, 1045-1050. Mellerup E. T., Plenge P., Vandsborg P., Rafaelson O. J., Kjeldsen H, and Agerbaek H. (1972) Antidiabetic effect of lithium. Lancet il, 1367-1368. Nelson J. S., Lacy P. E. and Hirschberg G. E. (1976) Megacolon and autonomic neuropathy in diabetic rats. J. Neuropath. exp. Neurol. 35, 335. Nevius D. B. and Lachattin G. F. (1965) Operation of the technic on flame photometer with natural gas. Use in the determination of serum lithium and the semi-micro analysis of serum sodium and potassium. Clin. Chem. 11, 633-639. Ogbonmaya K. I. and Arem R. (1990) Diabetic diarrhea: pathophysiology, diagnosis and management. Archs intern. Med. 150, 262-267. Oztfirk Y., Yfldlzo~,lu N., Altan V. M. and Karasu (~. (1987) Altered responses to calcium and trifluoperazine in K +-depolarized duodenum from alloxan-diabetic rats. Pharmac. Res. Commun. 19, 597--608.
773
Oztiirk Y., Ylldlzo~lu-An N., Oziian A., Oz~likay A. T. and Altan V. M. (1990) Decreased//-adrenergic responses of rat small intestine due to non-insulin-dependent diabetes. Diabetes Res. clin. Practice 9, 123-127. Rabinovitch A. and Baquerizo H. (1990) Immunological mechanisms of pancreatic //-cell destruction: effects of cytokines (Abstract). J. Autoonmun. 3, 85. Rana R. S. and Hokin L. E. (1990) Role of phosphoinositides in transmembrane signalling Physiol. Rev. 70, 115-164. Schou M. (1976) Pharmacology and toxicology of lithium. A. Rev. Pharmac. Toxic. 16, 231-243. Schou M. (1989) Lithium prophylaxis: myths and realities. Am. J. Psychiat. 146, 573-576. Schou M. (1990) Lithium and treatment-resistant depressions: a review. Lithium 1, 3-8. Sheridan E. P. and Bailey C. C. (1946) Diabetic nocturnal diarrhea. J. Am. med. Ass. 130, 632-634. Shopsin B., Stern S. and Gershon S. (1972) Altered carbohydrate metabolism during treatment with lithium carbonate. Archs Gen. Psychiat. 26, 566-571. Theologides A. (1986) Cachexia is a feature of cancer patients. Am. J. Med. 81, 696-698. Van der Velde C. D. (1970) Effectiveness of lithium carbonate in the treatment of manic depressive illness. Am. J. Psychiat. 127, 345-351. Vendsborg P. B. and Vilstrup H. (1976) The influence of lithium on carbohydrate and lipid metabolism in the perfused rat liver. Acta pharmac, tox. 38, 10-16. Viegut V. and Jefferson J. W. (1990) Lithium and liver. Lithium 1, 9-13. Wiemer G., Kaiser G. and Palm D. (1978) Effects of Mg 2+, Mn 2* and Ca 2+ on adenylate cyclase activity: evidence for a metallic site. Archs Pharmac. 303, 145-152. Yang R., Arem R. and Chan U (1984) Gastrointestinal tract complications of diabetes mellitus. Archs intern. Med. 144, 1251-1254. Yildlzo~lu-An N., Altinkurt O., Oztfirk Y,, Altan V. M., Pekiner C. and Karasu (~. (1988) Decreased gastro-intestinal responses to salbutamol and serotonin in streptozotocin-induced diabetes: improving effect of insulin in vivo and in vitro. Gen. Pharmac. 19, 665-668.
