.Nfiflfia “.13. .Jzfi - n.Iuc. x904! E? 11:21: a . 1.1,... L. P5133. .w _ _ . um -' » .If) '{ n a :loquvI...l. 10?») 0|. (bl: . l.“ Annunzhmmmtu "hilt. . efiul. bl..- ‘st’rt {- . "4' 5”}...5: a A .4" -200 ‘tfi‘l‘ttb - "V‘!l‘a§, LAtiilfi. I .77!!! .E'; r - l.‘..(.n. - oliun. . . A 266.!!!)3. :l 21:0v . .5. 1.3.0.1... it : . .Jliilo. ‘ ’9‘... ’10. .Il: .nfilo.a1)oll‘t6..tr.bll. . ’32:: Jbvlvnl..$6).roli)l ‘ 4 . 3...: .6:: 2.: ‘75.! . 1905 .51.! ill-3.4.98.3.» {.t‘. . 2.1.! ‘34} i‘.‘ . y. .p . 1| «1 .633! 1:916}, . .11.... 11;... ‘1. 5.0 IIELHI a. u'I-a£. tat-unha'y. .. uflh a n! ’ v f o . . "Ol.—vs Ci 'III :E!’ '- . . . 3|. ........:€.is..}hx..vi!#:.. .. 1.3.. It... gait-r3..- ‘Piuflklssbig‘ . . 7 .l (4...:.i.7.l.se.-...:...:...¥.¢I.l.t .fioluru(..ttlfi.x.1!fl.z§.i£.. n... at... k..1..3..7!.-$l;1.¢s . ..JP... A -,.‘ ‘ . . ..! ‘ t 1 . .‘ . , . l‘ ‘ ‘ .. a .. . . . u . . _,.L. y t..llr. . lull? 'I' ..t‘ uvt.‘ ‘¢\ fill L . 1. . ‘cu .....la .I.l:s!,.lc. r4 .1... . .,.,‘II I ‘Iv x VERS SITY LIBRARI IE IUIHHIIhtlllWlHHUIlllllllIl Hill“ III ll 3 1293 00891 ll This is to certify that the dissertation entitled Investigations into the Mechanism of Cyproheptadine- Induced Inhibition of Proinsulin Biosynthesis and Depletion of Pancreatic Insulin Content. presented by Christopher Paul Miller has been accepted towards fulfillment of the requirements for Ph.D. Pharmacology/Toxicology degree in WWW/M Major professor MSU is an Affirmative Action/Equal Opportunity Institution O~12771 LIBRARY MIchIgen State UnwersIIy PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. INVESTIGATIONS INTO THE MECHANISM OF CYPROHEPTADINE- INDUCED INHIBITION OF PROINSULIN BIOSYNTHESIS AND DEPLETION OF PANCREATIC INSULIN CONTENT BY Christopher Paul Miller A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 1990 8 RACT INVESTIGATIONS INTO THE MECHANISM OF CYPROHEPTADINE- INDUCED INHIBITION OF PROINSULIN BIOSYNTHESIS AND DEPLETION OF PANCREATIC INSULIN CONTENT BY Christopher Paul Miller The diabetogenic drug cyproheptadine has previously been shown to inhibit proinsulin biosynthesis and deplete pancreatic insulirn In initial experiments to investigate the mechanisms of these actions, cyproheptadine reversibly decreased rat pancreatic preproinsulin mRNA. Further experiments were conducted to determine the relationship between cyproheptadine-induced alterations of pancreatic preproinsulin mRNA, proinsulin and insulin levels in 1129, and alterations of proinsulin synthesis, and preproinsulin mRNA and insulin levels in yit;_. Results from in 211g time course studies. and comparative experiments with cyproheptadine analogs in rats, and acute in yitgg exposures of isolated rat islets suggest that cyproheptadine-induced.decreases in preproinsulinszNA levels are correlated with, but do not cause, inhibition of proinsulin synthesis and depletion of pancreatic insulin content. Cyproheptadine-induced decreases in pancreatic proinsulin occurred prior to decreases in preproinsulin mRNA, and acute inhibition of proinsulin synthesis j,_n_ 212:9 occurred without decreases in preproinsulin mRNA. In experiments with. mice, cyproheptadine reversibly decreased pancreatic insulin and preproinsulin mRNA. The insulin depletion occurred more slowly than in rats, while decreases of preproinsulin mRNA occurred as rapidly as in rats. These findings are discussed in relation to known species differences in cyproheptadine metabolism. Experiments were performed to examine the direct actions of cyproheptadine on rat pancreatic islet cells, RINmSF and HIT-T15 cells in ML.- In short-term studies (5 2 hr), cyproheptadine inhibited insulin secretion, and selectively inhibited insulin synthesis. In rat islet cells, cyproheptadine did not acutely inhibit proinsulin to insulin conversion. Longer cyproheptadine exposures (24 and 48 hr) reversibly decreased media and cellular insulin levels. In RINmSF and HIT-T15 cells, cyproheptadine failed to decrease preproinsulin mRNA levels after short or long term exposures to inhibitory concentrations, indicating that in these cells, decreases of preproinsulin mRNA are not required for cyproheptadine actions. Cyproheptadine appears to suppress proinsulin synthesis by inhibiting translation of preformed preproinsulin mRNA. Decreases of preproinsulin mRNA may occur independently, or secondary to, inhibition of proinsulin synthesis. Since cyproheptadine appears to inhibit.several steps.of the insulin biosynthetic pathway rapidly and reversibly, this drug might represent a useful tool for the study of regulation of insulin biosynthesis. To my parents and step-parents, who all have helped me achieve this goal. Especially to my mother, whose love and sacrifice has always been a source of strength and inspiration. iv Lfififlglhfiflfififlfiflifi I would like to express sincere gratitude to my friend and advisor, Dr. Larry Fischer, for providing encouragement, direction and discipline, yet allowing me to develop and learn independently. I am also grateful to the members of my committee, by whose suggestions this dissertation was improved. A special note of thanks should go to Dr. Jerry R. Colca of the Upjohn Company, whose expertise in pancreatic islet biochemistry was particularly helpful. I would also like to thank Kena Chapman, Rhonda Hensley, Barbara Blok and Shannon Mullaney for excellent technical assistance, and Drs. Stephen J. Giddings (Washington Univ., St. Louis, MO) and.Moriko Ito (Dept. of Biochemistry, MSU) for instruction and helpful discussions pertaining to some of the preproinsulin messenger RNA experiments described herein. Finally, I would like to thank the Pharmaceutical Manufacturer's Association Foundation, the MSU Department of Pharmacology and Toxicology, and Center For Environmental Toxicology for financial support during my training here at MSU . ILELE 9? CONTENTS Page LIST OF TABLESOOOOOOOOOOOO ...... 00...... 000000000000 Viii LIST OF FIGURESOOOOOOOOOOOOOO0.0.0.0.... 000000000000 ix ABBREVIATIONSIO0.0.0.0.... 00000000000000000000000000 Xii INTRODUCTION ..... OOOOOOOOOOOOOOOOOOOO... OOOOOOOOOOOO 1 Pancreatic Islet Anatomy and Physiology ........ 2 Pancreatic Islet Pharmacology and Toxicology... 4 Insulin Biosynthesis/Gene Expression.. ......... 6 Insulin Gene Structure ..... ........... ......... 9 Transcriptional and Post-transcriptional ' Control of Insulin Biosynthesis.... ....... 18 Translational and Post-Translational Control of Insulin Biosynthesis ........... 21 Experimental Alteration of Insulin Biosynthesis.......... ............ 29 Effects of CPR on the Endocrine Pancreas ....... 33 Objectives/Rationale.................. ......... 38 MATERIALS AND METHODS....... ........ . ............... 46 materiaISIOOOOOOOOO0.0.0.... ...... O OOOOOOOOOOOO 46 nethOdSOOOOOOOOOOOOOOOOOOOOOO ..... O. ......... O. 48 In yivg Experiments................. ........... 48 In yitzg Experiments......... ...... . ........... 58 RESULTS..0...OOOOOOOOOOCOOOOOOOOOOOO0.0......0.....0 80 In 2110 Time Course of CPH Effects in the Rat................................ 80 Acute In 21:19 Effects of CPR in Isolated Rat Islets.................... 90 In 11:19 Effects of CPR in Cultured Dispersed Rat Islet Cells......... ...... .. 95 In 11;:9 Effects of CPR in RINmSF and HIT-T15 Cells......................... 105 In 2119 Effects of 4-DPMP and 2-DPMP in the Rat......................... 127 In yivg Effects of CPH in the Mouse ............ 129 vi DISCUSSIONOOOOOOOOOOOO0.0.0.000...0.0.0.0... ........ 135 Mechanism of CPR Inhibition of Proinsulin Synthesis and Depletion of Pancreatic Insulin Content................... ........ 135 Cultured Cells as Models for CPH Actions in the Endocrine Pancreas.................... 143 Effects of CPR in Mice................... ...... 153 CPH as a Tool for the Study of Insulin Gene ExpreSSionOOOOOOOOOO0.00... .......... 156 SUMMARY AND CONCLUSIONS............. ........... ..... 158 REFERENCESOOO0.00000000000000000000 OOOOOOOOOOOOOOOOO 162 vii LI§I_QZ_IA§L§§ Effects of a single dose of CPH on rat body weight and pancreatic total RNA...... Trypan blue dye exclusion assay of RINmSF and HIT-T15 cell viability after CPH exposure for 24 and 48 hr................. Effects of CPR on PPImRNA levels in RINmSF and HIT-T15 cells exposed for 24 hr....... Effects of a single dose of 4-DPMP or 2-DPMP on rat body weight, and pancreatic total RNA, PPImRNA and B-actin mRNA..................... ..... Effects of a single dose of CPR on mouse body weight, serum glucose concentration, and pancreatic total RNA, PPImRNA, and insulin content.............. ...... ... viii 119 123 128 134 10. 11. 12. LI§I_QI_ZI§!B§§ Insulin biosynthesis...................... Structures of insulin genes...... ...... ... Structure and amino acid sequence of human proinsulin...... ................. Northern analysis of rat pancreatic PPImRNA at O, 24, and 48 hr after CPH administration...... ...... . ........... Structures of CPR, 4-DPMP and 2-DPMP. ..... Time course of CPR effects on rat pancreatic insulin content........ ...... .. Gel filtration chromatography. (Bio Gel P30) of acetic acid pancreatic extracts from control and CPR-treated rats.......... ............ Time course of CPH effects on rat pancreatic proinsulin..................... Northern analysis of pancreatic PPImRNA and B-actin mRNA levels from control and CPR-treated rats 10 hr after dosing....... Time course of CPR effects on rat pancreatic PPImRNA levels..... ........ Time course of CPM effects on rat pancreatic fi-actin mRNA levels ........ Northern analysis of PPImRNA and fi-actin mRNA levels in isolated rat islets after 30 min exposure tooor IOMCPHEMOOOOCCOOOOOOOOOO ix 11 24 39 45’ 84 85 86 87 88 89 92 Emit 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Effects of 0 or 10 uM CPH on PPImRNA, proinsulin synthesis, and non-proinsulin protein synthesis in cultured rat pancreatic islets exposed for 30 minEMOOOOOCOOOOOOOOOOOOOO ....... Effects of CPR on glucose-stimulated insulin release from cultured rat pancreatic islet cells.................... Effects of CPR on glucose-stimulated insulin synthesis in cultured rat pancreatic islet cells.. .............. Conversion of proinsulin to insulin in cultured rat pancreatic islet cells exposed to CPH or monensin in vitro ....... Cellular insulin content of rat pancreatic islet cells after culture for 48 hr in the presence of CPH.......... Insulin concentration in media from rat pancreatic islet cells after culture for 48 hr in the presence of CPH..... ..... Effects of CPH on stimulated insulin release from HIT-T15 and RINmSF cells..... Effects of CPR on insulin and non-insulin protein synthesis in HIT-T15 and RINmSF cells.....OOOOOO00.0.0000.........O Cellular insulin content of RINmSF and HIT-T15 cells after culture for 48 hr in the presence of CPH.............. Insulin concentration in media from RINmSF and HIT-T15 cells after culture for 48 hr in the presence of CPH.......... Recovery of cellular insulin content in RINmSF and HIT-T15 cells after removal of CPR (10 uM, 48 hr exposure)............... 94 99 100 102 104 105 112 113 114 115 116 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. Recovery of media insulin from RINmSF and HIT-T15 cells after removal of CPR (10 uM, 48 hr exposure).... ........... Lack of effects on RINmSF and HIT-T15 cell growth and division during recovery from CPH exposure...... .......... Northern analysis of PPImRNA and fi-actin mRNA levels in HIT-T15 and RINmSF cells exposed to 0 or 10 pH CPR for 2 hr .............................. Northern analysis of PPImRNA and fi-actin mRNA levels in HIT-T15 and RINmSF cells exposed to CPR for 24 hr ..... Northern analysis of PPImRNA and GAPDH mRNA levels in RINmSF cells exposed to CPH or sodium butyrate. ........ Comparison of the insulin-depleting effects of actinomycin D, cycloheximide, and CPR in RINmSF cells exposed for 24 hr......................... ........ Comparison of the insulin-depleting effects of CPH, 4-DPMP and 2-DPMP in RINmSF cells exposed for 24 hr......... Mouse body weights during and after administration of eight daily doses of CPR or vehicle........... .............. Mouse pancreatic insulin content during and after administration of eight daily doses of CPR or vehicle....... Mouse serum glucose concentrations during and after administration of eight daily doses of CPR or vehicle. ...... xi 117 118 120 121 122 125 126 131 132 133 TIONS ACT-D: Actinomycin D ALX: Alloxan CAT: Chloramphenicol Acetyl Transferase cAMP: Cyclic Adenosine Monophosphate CPH: Cyproheptadine CHX: Cycloheximide DMCPH: Desmethylcyproheptadine DPMP: Diphenylmethylpiperidine GPAIS: Guinea Pig Anti-Insulin Serum ip.: Intraperitoneal IRI: Immunoreactive Insulin po.: Per Osteum PPImRNA: Preproinsulin Messenger Ribonucleic Acid RER: Rough Endoplasmic Reticulum RIA: Radioimmunoassay STZ: Streptozotocin xii INTRODUCTION The disease diabetes mellitus represents a complex variety of disorders characterized by the inability to regulate blood glucose levels within normal physiological limits (Marble, 1971). Although. the etiology of ‘these disorders is poorly understood, it is currently thought that some environmental factors may contribute to the development of the disease (Cahill and McDevitt, 1981: Craighead, 1978). These could include dietary habits, exposures to viruses, and exposures to xenobiotic substances such as pharmaceutical agents, food contaminants and environmental pollutants (Malaisse, 1987). Although there are examples for which each of these factors have been proposed to somehow contribute to the development of diabetes, there is a lack of knowledge regarding’ the relative importance. of these as causative diabetogenic agents. At present there is a need for further research effort to be directed towards estimating the role that exposures to xenobiotic, chemicals may' play in the pathogenesis of diabetes. Research conducted in this laboratory is directed towards elucidating the mechanisms by which certain chemical substances are capable of causing toxic effects in insulin-producing B-cells of the endocrine 2 pancreas. Knowledge of how xenobiotic substances can alter [3- cell function might provide valuable insight into the normal physiology of the insulin-secreting cells, as well as into the etiology of certain types of diabetes mellitus. One denial currently under study is the antihistaminic and antiserotonergic drug cyproheptadine (CPH) . CPH appears to be able to selectively inhibit proinsulin synthesis in pancreatic islet fi-cells leading to the depletion of pancreatic insulin content (Hintze et al. , 1977; Rickert and Fischer, 1975: Rickert et al., 1975). The studies described in this thesis are directed towards elucidating the biochemical mechanism(s) of these actions. a c o Pancreatic islets are small spherical collections of endocrine cells scattered throughout the pancreas. Islets comprise approximately 1-2% of the mass of the pancreas. The remaining 98-99% of the pancreas is composed of exocrine, ductal, connective and neural tissues. Islets are composed of four major cell types, each of which is responsible for the production and release of one of the major islet hormones. 8- cells produce insulin, a-cells produce glucagon, 6-cells produce somatostatin, and PP-cells produce pancreatic polypeptide. fi-cells are located in the center of the islets, and comprise approximately 60% of the islet mass. o-cells and 6-cells make up approximately 25 and 10%, respectively, of the 3 islet mass, and are located mainly around the islet periphery. PP-cells are also located around the islet periphery, and are present in lower numbers than either 0- or 6-cells. The pancreas receives arterial blood from the splenic, hepatic and mesenteric arteries. Venous drainage is into the splenic and mesenteric veins. Individual islets have extensive networks of highly fenestrated capillaries allowing for rapid transfer of the islet hormones across the capillary endothelium into the circulation. Pancreatic islets receive both sympathetic and parasympathetic innervation. Stimulation of the splanchnic nerve (sympathetic) inhibits insulin secretion and stimulates the release of glucagon. Stimulation of the vagus nerve (parasympathetic) stimulates insulin release and inhibits glucagon secretion. Glucose is the principle physiological regulator of insulin and glucagon secretion. IHyperglycemia stimulates the secretion of insulin, while hypoglycemia stimulates the release of glucagon. Insulin stimulates the uptake and utilization of glucose at various sites throughout the body, thereby lowering blood glucose concentration. Glucagon stimulates the mobilization and release of glucose from the liver, thereby raising blood glucose levels. Glucagon also stimulates the release of insulin, while insulin appears to inhibit the release of glucagon. Additionally, somatostatin inhibits the release of both glucagon and insulin. The net 4 result of these relationships is that blood glucose concentrations are maintained within normal physiological limits (80-110 mg/dl) . Many endogenous substances can influence the secretion of insulin and glucagon [gastrin, secretin, cholecystokinin, enteroglucagon, glucagon-like peptides, galanin, gastric inhibitory peptide, free fatty acids, some amino acids (arginine, alanine, leucine and lysine) and sugars (mannose, galactose and glyceraldehyde) and triglycerides], but the contribution of each and the interaction of these substances to augment or attenuate the influence of glucose on hormone secretion is unknown. Exercise and fasting inhibit the release of insulin, while stimulating the release of glucagon. Lastly, many stresses including infection, burns, toxemia, tissue infarction, and major surgery all rapidly increase glucagon secretion. ancreat a o o a d Tox olo There are many drugs that can alter the function of pancreatic fi-cells. These include agents that are used therapeutically to treat the hyperglycemia of non-insulin- dependent diabetes mellitus, and agents used to alleviate hypoglycemia caused by insulin-secreting islet cell tumors. Sulfonylurea compounds such as tolbutamide, chlorpropamide, glipizide, and glyburide are the most commonly used oral hypoglycemic agents (Larner, 1985) . These drugs are believed 5 to alleviate hyperglycemia of non-insulin-dependent diabetes mellitus by stimulating the release of insulin from functional islet fi-cells, and by enhancing the sensitivity of insulin- responsive tissues (liver, muscle, adipose) to the actions of insulin. Mypoglycemia caused by insulin-secreting tumors have been successfully treated by the administration of streptozotocin (STZ) , diazoxide, or somatostatin (Larner, 1985). STZ is an agent that causes the selective destruction of islet B-cells (see below). Diazoxide and somatostatin inhibit the release of insulin from islet B-cells. Phenytoin and some thiazide diuretics have also been used experimentally to produce hyperglycemia due to inhibition of insulin secretion (Larner, 1985). There are a variety of chemical compounds that have been shown to produce toxicity to pancreatic islets. Most of the chemicals in this group produce selective toxicity to insulin- producing fi-cells, while a few substances have been reported to selectively damage the glucagon-producing a-cells. These o-cell toxicants include cobalt, cadmium, iodoacetate, neutral red, potassium xanthate, and diethylthiocarbamate (Cooperstein and Watkins, 1981). The known fi-cell toxicants include the rodenticide VACOR (Prosser and Karam, 1978), the antitrypanosomal drug pentamidine (Bouchard et a1. , 1982) , the marine antifoulant triphenyltin flouride (Manabe and Wada, 1981), the nitrosourea anticancer drug STZ (Rakieten et al., 1963), the experimental diabetogenic agent alloxan (ALX) (Dunn 6 et al. , 1943) , and certain diphenylmethyl-piperidine compounds including CPH (Fischer and Rickert, 1975). u e s o The process of "insulin gene expression" refers to all of the steps involved in thetconversion.of information encoded in the insulin gene to the formation of mature insulin molecules in secretory granules of islet.fi-cells (Figure 1) (Steiner and Tager, 1979; Permutt et al., 1981; Permutt et al., 1984: Steiner et al., 1985; Selden et al., 1987). Information encoded in the insulin gene(s) is transcribed by RNA polymerase II into preproinsulin mRNA (PPImRNA). The primary transcript is processed within the nucleus to form mature PPImRNA. This post-transcriptional processing involves the addition of a 7-methylguanosine residue at the 5' end, polyadenylation at the 3' end, and the splicing out of one or two introns (depending upon the insulin gene; see below). The mature PPImRNA leaves the nucleus, is bound to ribosomes in the cytoplasm, and is translated to form preproinsulin. The pre portion of the molecule is removed rapidly from the nascent polypeptide, as the molecule is sequestered into the membranous vesicles of the endoplasmic reticulum. The resulting proinsulin molecules are transferred to the Golgi region where conversion to insulin, zinc crystal formation and packaging into secretory vesicles takes place. 7 MJCLEUS P "new ——«———-—--— “— t79 M 5' .— ne no —-—-~ 3 "—559.11: a «vs 119— ------ Chromosome II 1 TRANSCRIPTION . i—{Efiq- IcI ~s lug—T“; Removal .c I Transcr'pt mm We: \ /' "V” RNA PROCESSNG new” 4...: 1 1 I—— sse 8 C RN29“ CYTOPLASM M.\ Rough Endoplasmic Reticulum (‘Pre' Segment Cleavage) Proinsulin \, ' Microveslcles Processing pretenses {we Golgi Complex Early Granules @PROGRESSIVEQ G) Insulin in CrystalIoId~@ CONVERSION © Mature Granules C-peptide -——@ C9 0 Membrane Rec of M Y "a, \ ma: ...... - Secreted PORTAL BLOOD - Hohsuh and lnter- Products 8%:th oI Conversion -4 Figure 1. Insulin biosynthesis. Steps involved in insulin gene expression in human pancreatic islet fi-cells. Figure taken from Steiner and Tager (1979). 8 Until recently, it. was. believed. that expression. of insulin genes was limited to B-cells of the islets of Langerhan's (Eng and Yalow, 1981; Giddings et al., 1985). However, it has now been demonstrated that other tissues also may synthesize insulin. Using Northern analysis or in situ hybridization to detect PPImRNA, evidence has been presented for insulin synthesis in such extrapancreatic tissues as the yolk sac of the developing rat fetus (Giddings and Carnaghi, 1989: Rau et al., 1989; Muglia and Lockler, 1984), human placenta (Liu et al., 1985), mouse and rat brain (Lee et al., 1984; Raizada et al., 1979; Raizada, 1983), rat pituitary (Budd et al., 1986), and cultured rabbit neurons and mouse seminal vesicle epithelial cells (Schechter et al., 1988; Stahler et al., 1987). The physiological role of insulin synthesis in these non-pancreatic tissues is unclear at this time. With.the exception of rats and mice, all mammals thus far examined produce a single type of insulin that is coded for by a single insulin gene (Humbel, 1972). Rats and mice, and certain species of fish (tuna and toadfish), synthesize two different insulins that are coded for by two non-allelic insulin genes. The structures of these two insulin genes have been examined in great detail (Lomedico et al., 1979; Cordell et al., 1979), as have the proinsulin and insulin molecules expressed from the two genes (Clark and Steiner, 1966; Smith, 1966). 9 In the rat, the initial ‘translation products, preproinsulins I and II, differ only by 7 amino acid substitutions [3 in the signal peptide sequence (pre-region), 2 in the connecting peptide (C-peptide) region, and 2 in the B-chain of mature insulin] (Lomedico et al., 1979). In rats and.mice, under normal conditions, approximately 60% of total pancreatic insulin is insulin I, while the remaining 40% is insulin II (Clark and Steiner, 1966; Kakita et al., 1982a). The same ratio (60:40) exists for rat and mouse insulin I and II mRNAs (Giddings and Carnaghi, 1989; Koranyi et al., 1989). It is currently believed that expression of the two genes is coordinately regulated at the transcriptional and translational levels (Giddings and Carnaghi, 1988; Rhodes et al., 1937). Inaulia_sssg_§trns&urg Characterization of insulin genes was made possible by cloning of insulin cDNAs. The first successful cloning of rat insulin cDNAs was reported in 1977 (Ullrich et al., 1977). These investigators isolated and cloned cDNAs for most of the mRNAs coding for rat insulins I and II from a cDNA library obtained from isolated rat islets of Langerhan's. In 1978, Villa-Romaroff and co-workers cloned a portion of the cDNA encoding rat insulin I from a cDNA library derived from a transplantable rat insulinoma (Villa-Romaroff et al., 1978). In 1979, Chan and co-workers (Chan et al., 1979) reported the 10 construction of recombinant plasmids containing full length cDNAs corresponding to the mRNAs coding for rat insulins.I and II. The mature mRNAs for rat insulin I and II are approximately 600 bases in length, of which 330 nucleotides code for preproinsulin. There are 60 bases of 5' untranslated mRNA, 55 (rat I) or 56 (rat II) bases of 3' untranslated.mRNA, and approximately 150 adenine residues in the poly-A tail. These rat insulin cDNAs were then utilized.to isolate and characterize the genes that coded for insulin from various species. In 1979, two groups reported the cloning and sequencing of rat insulin genes. Cordell et al. (1979) isolated and characterized the gene encoding rat insulin I, and Lomedico et al. (1979) isolated and characterized the genes encoding both rat insulins I and II. .Also in 1979, Bell and coworkers (Bell et al., 1979) reported the cloning and sequencing of the human insulin gene, and in 1980, Perler et al. described the cloning and sequencing of the chicken insulin gene. Since there is a high degree of homology among insulins and insulin genes from a variety of species, many insulin cDNAs and genes have now been cloned (for a review, see Steiner’et al., 1985). From these studies, many important observations have been made regarding the nature of insulin genes. As shown in figure 2 the gene encoding rat insulin II has two introns, one in the 5' flanking region of the gene (119 bp) and one in the region of the gene that codes for the C- cap site Rat I -l% "pr *%d( 11 pdyAmm ...] e I emi— Rat II Human X**** Figure 2. Structure of insulin genes. ............................... ................................ ............................... ................................ ............................... ................................ ............................... .................................... éfikméielc vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv v vv 'vv ................................... .............................................................. ............................................. ........................................................ .............................. ...-.....o..---.--..-. eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee ................................................. LAA‘AALAAAAAAAAA‘AAA-A‘ALI-IIIAILAAAIlllII'I'Jl0l1""'.‘. and II insulin genes with the human insulin gene. taken from Bell et al. (1980). Comparison of rat I 12 peptide of’proinsulin (499 bp). The gene encoding rat insulin I has a single 119 bp intron in the same location of the 5' non-coding region of the gene as does rat insulin II, but lacks the second intron in the region encoding the C-peptide. The human insulin gene has a 179 bp intron in the 5' non- coding region and a 789 bp intron in the region coding for C- peptide. The chicken insulin gene contains the 119 bp intron in the 5' non-coding region and a 3.5 kb intron in the region encoding the C-peptide. For all of the other insulin genes that have been cloned, the locations of the two introns have been highly conserved but the lengths are quite variable (Steiner et al., 1985). It is presumed that in mammals and birds that possess a single insulin gene, that gene represents the ancestral insulin gene. Since the rat insulin II gene shares the intron-exon organization with the human and chicken genes, it is believed that the rat insulin II gene is the ancestral gene in this species. The second rat insulin gene (rat insulin I) is thought to have arisen from a duplication of the ancestral gene followed by the loss of the second intron (Lomedico et al., 1979). It has been proposed that this occured by RNA-mediated transposition of a cDNA copy of an incompletely processed upstream transcript of the insulin II gene (Soares et al., 1985). It is uncertain why the insulin gene duplication event persisted in the rat and mouse, and why it also occured in the tuna and toadfish. 13 With the exception of the second intron of the rat insulin II gene, there is a very high degree of homology between the rat insulin I and II genes. The homology extends for approximately 500 bases upstream of the 5' ends of the genes, but.does not extend.beyond.the end corresponding to the 3' end of the mRNAs (Lomedico et al., 1979). There is also considerable homology between the rat (II) and human insulin genes (Bell et al., 1980). This homology is evident in the 5' flanking region, the coding regions, and into the 3' untranslated region, but does not extend further in the 3' direction (Bell et al., 1980). The intervening sequences of the rat insulin II gene and the human insulin gene also are quite divergent except for the bases immediately surrounding the splice junctions (Bell et al., 1980). The high degree of homology in the 5' flanking regions of these genes is consistent with the proposal that important regulatory sequences are conserved within these regions (Bell et al., 1980; Lomedico et al., 1979; Cordell et al., 1979). DNA sequence elements located in the 5' flanking region of the insulin gene are essential for directing expression of the gene in pancreatic islet fl-cells. These sequence elements interact with fi-cell-specific nuclear regulatory proteins to facilitate insulin gene expression in islet B-cells. The absence of these nuclear proteins in non-insulin-producing cells allows the gene to remain quiescent in non-fi-cells. 14 Walker et al. (1983) and Episkopou et al. (1984) first demonstrated that regions of the insulin gene 5' flank were involved in cell-specific expression of reporter genes placed under control of the insulin gene 5' flank. These investigators constructed transient expression vectors with reporter genes placed adjacent to the insulin gene 5' flank. Walker et al. (1983) showed that when Chloramphenicol acetyl transferase (CAT) gene expression was placed under the control of the insulin gene 5' flank, CAT gene expression occured in a transformed B-cell line, HIT-T15, but not in a chymotrypsin- producing cell line, AR4-ZJ, or fibroblast-derived Chinese hamster ovary cells. Furthermore, progressive deletions were made into the insulin gene 5' flank so that regulatory sequences could be located. Sequences involved in conferring fi-cell-specific expression to the insulin gene were found to be located within 300 bp upstream of the transcription start site. This region roughly corresponds to the location of a tissue-specific DNase I hypersensitive site (Wu and Gilbert, 1981). It is believed that DNase I hypersensitivity is related to relaxation of chromatin structure which may allow regulatory proteins access to potential control elements. Using a slightly different strategy, Episkopou et al. (1984) placed the gene for a selectable marker (xanthine/ guanosine phosphoribosyltransferase) under insulin promoter control, then characterized the cell—specificity of colony formation under selective pressure. Transduction of a 15 retroviral vector containing insulin gene 5' flank upstream of the xanthine/guanosine phosphoribosyltransferase gene into insulin positive and negative lines of rat insulinoma cells yielded colony formation only in insulin-producing cells. The advent of transgenic mouse technology facilitated an elegant approach in the investigation of cell-specific insulin gene expression in 2122. Hanahan (1985) established lines of transgenic :mice that. harbored. a recombinant fusion. gene containing portions (520-660 bp) of the insulin gene 5' flank linked to the coding region of the SV40 large T-antigen gene. In these mice, expression of SV40 large T-antigen was confined to pancreatic islet fi-cells. This cell-specific expression eventually lead to the production of solid fi-cell tumors. Other investigators have since shown that different genes put under insulin 5' flank control also are expressed exclusively in islet fi-cells of transgenic mice (Epstein et al., 1989: Allison et al., 1988; Sarvetnidk et al., 1988). Also, in transgenic mice harboring the human insulin gene and regions of the 5' flanking DNA (up to 12.5 kb), expression of the human gene is confined to mouse islet fi-cells (Selden et al., 1986: Bucchini et al., 1986; Bucchini et al., 1989). Taken together, results from these in yitrg and in 2119 studies indicate that the insulin gene 5' flanking region contains information.that mediates islet B-cell-specific expression of the gene. 16 Discrete regulatory elements that confer B-cell-specific expression have been identified within the insulin gene 5' flank (Edlund et al., 1985; Karlsson et al., 1987). Systematic block replacement mutations indicated that three distinct regions were required for transient expression of an insulin/CAT recombinant plasmid transfected into HIT-T15 cells (Karlsson et al., 1987). These three regions were the TATA box of the promoter (located at.positions -23 to -32, relative to the transcription start site), and.elements at located at - 104 to -112 (termed the Nir box) and -233 to -241 (the Far box). The TATA box of the promoter is involved in RNA polymerase binding and transcription initiation. The Nir and Far boxes are homologous with each other and function synergistically as transcriptional enhancers. Nuclear extracts from insulin-producing HIT-T15 and RINmSF cells contain DNA-binding proteins that interact with the Nir and Far box motifs, but these proteins are absent in nuclear extracts from non-insulin-producing cells (Ohlsson and Edlund, 1986; Ohlsson et al. , 1988) . Gel shift analyses indicated that the same protein, or similar proteins, appear(s) to bind both the Nir and Far box elements. Other DNA-protein interactions are also likely to be involved in regulating expression of the insulin gene in insulin-producing cells, as over 40 binding species have been detected in HIT cell nuclear extracts by systematic protein binding analysis of sequential overlapping segments of the insulin gene 5' 1'7 flank. (Moss et al., 1988). Some of these DNA-protein interactions are likely ‘to .represent. binding’ of nuclear transcription factors and accessory proteins to regulatory elements of the gene. Negative regulatory elements have also been identified within the insulin gene 5' flank. The results of Nir et al. (1986) suggest that trans-acting factors that repress insulin gene expression in non-insulin-producing cells might do so by binding to negative regulatory elements within the 5' flank. In insulin-producing cells, the effects of these repressors are overridden by the presence of dominant positive trans- acting factors. Insulin gene transcription is suppressed in HIT-T15 cells infected with adenovirus type.5 (Stein.and Ziff, 1987) . This suppression of transcription was mediated by Adenovirus type 5 Ela proteins which may mimic the cellular repressors proposed by Nir et al. (1986) to inhibit insulin gene expression in non-insulin-producing cells. Negative regulation of insulin gene transcription has also been reported to be mediated by "silencer" elements located in repetitive. sequences associated. with ‘the rat insulin I gene locus (Laimins et al., 1986). These silencer elements are thought to function in a manner analogous to enhancer elements (ie. orientation and distance-independent modulation of promoter function) but they will suppress, rather than increase, the transcription of genes. 18 rans t o a ost- ansc i tional Control of Insul n 81222112112111 It is generally accepted that there are multiple levels of control in the regulation of insulin gene expression. Transcriptional regulation of insulin biosynthesis was inferred from early observations that actinomycin D (ACT-D) inhibited glucose-stimulated RNA and proinsulin synthesis in isolated rat islets (Jarrett et al., 1967; Morris and Korner, 1970). Permutt and Kipnis (1972) made the important observation that when ACT-D and glucose (15.3 mM) were added to isolated islets at the same time, glucose stimulation of insulin synthesis was evident after 30 min, but inhibition by ACT-D was not detectable until 60 min. Glucose-stimulation of insulin synthesis appeared.to involve an early phase (up to 60 min) where synthesis of RNA was not involved, and a later phase (beyond 60 min) where synthesis of RNA was involved. Glucose has been shown to directly modulate PPImRNA levels in cultured isolated mouse, rat and human islets (Brunstedt and Chan, 1982; Giddings et al.,l985: Hammonds et al., 1987). Low glucose concentrations (2.8-3.3 mM) lead to decreases in ‘the levels of PPImRNA, while. high. glucose concentrations (20-28 mM) allowed the maintenance of PPImRNA levels, or increased PPImRNA levels after culture at low glucose. Importantly, insulin biosynthetic capacity of islets was correlated with variations in the levels of PPImRNA. 19 It is interesting to note that glucose, leucine, and 2- ketoisocaproate (the deamination product of leucine) are all equally effective at maintaining PPImRNA levels in cultured mouse islets (Welsh et al., 1986). This suggests that the metabolic fluxes induced by glucose and other fuel molecules might be important in insulin gene regulation, rather than the glucose molecule itself. Modulation of PPImRNA levels can be due to alteration of insulin gene transcription and/or PPImRNA degradation. Glucose, dexamethasone, cyclic adenosine monophosphate (cAMP) and cholera toxin have been shown to stimulate insulin gene transcription acutely in isolated rat islets (Nielsen et al., 1985; Welsh et al., 1988). Glucose also causes selective stabilization of PPImRNA relative to total islet RNA (Welsh et al., 1985). PPImRNA is a relatively stable mRNA with a half- life (ti/2) 6-10 times longer than for total poly(A)‘RNA. Elevation of glucose increased the t”? of PPImRNA from 29 hr at 3.3 mM glucose to 77 hr at 17 mM glucose, without increasing the t1/2's of total islet RNA or poly(A)‘RNA. Pancreatic PPImRNA levels can be readily measured in M. and have been shown to be altered by a variety of experimental manipulations. Fasting (Giddings et al., 1981: Giddings et al., 1982; Fukumoto et al., 1986), adrenalectomy (Fiedorek and Permutt, 1989), chronic infusion of insulin (Kruszynska et al., 1988; Chen et al., 1989), dietary manganese deficiency (Baly et al., 1988), coxsackievirus B4 20 infection (Chatterjee and Nejman, 1988) , and administration of STz (Permutt.et.al., 1984) decrease pancreatic PPImRNA.levels. Also, certain strains of mice that are genetically prone to the development of a form of non-insulin-dependent diabetes- mellitus show a progressive decrease in PPImRNA levels (Orland and Permutt, 1987). Injection of glucose or refeeding of fasted rats (Giddings et al., 1982) or administration of dexamethasone to adrenalectomized rats causes elevation or restoration of normal pancreatic PPImRNA levels (Fiedorek and Permutt, 1989) , and partial pancreatectomy causes compensatory increases in PPImRNA levels in the remaining portion of the pancreas in rats (Orland et al., 1985). Rat insulin I and II mRNAs appear to be coordinately regulated in 2129 (Giddings and Carnaghi, 1987). The ratios of rat insulin I and II mRNA remained constant during fasting and sucrose-treatment, during pregnancy, after administration of dexamethasone, in growth hormone tumor-bearing rats, and during fetal pancreatic development, suggesting that the same or similar mechanisms regulate the levels of both mRNAs. Furthermore, in sucrose-treated rats, premRNAs for both insulins I and II were increased in parallel with each other and with the mature insulin I and II mRNAs, suggesting that changes in gene transcription rates or premRNA stability are involved in regulation of rat insulin I and II mRNA levels. 21 o d t- a on Control of Insulin mm Preproinsulin: The immediate product resulting from translation of PPImRNA is preproinsulin. In M translation of RNA isolated from fetal bovine pancreas (Lomedico and Saunders, 1976) or isolated rat islets (Chan et al., 1976) was shown to result in the fermation of insulin immunoreactive proteins with molecular mass between 11,500 and 13,000 daltons. The larger mass of these translation products, relative to proinsulin (approx. 9000 daltons) and insulin (approx. 6000 daltons) suggested that these proteins were precursors for proinsulin and insulin. Chan et al. (1976) immunoprecipitated and sequenced the in 21:39 translation product from rat islet RNA and demonstrated that it was rat proinsulin with 23 amino acids covalently attached to the N-terminal end of the 8 chain. Preproinsulins from several other species have also been characterized and shown to have the same basic structure: lug-preregion-B chain-2 basic amino acids-C peptide-2 basic amino acids-A chain-COOH (Lomedico et al., 1977; Shields and Blobel, 1977; Shields, 1981). The pre-regions of these preproinsulins have similarity to each other, and to the pre- regions of other secretory proteins (Blobel and Dobberstein, 1975; Kemper et al., 1974: Devillers-Thiery et al., 1975). .All pre-sequences appear to have a high proportion of hydrophobic residues, and within the insulin pre-sequences, 22 there is a conservation of location of 6 out of 7 leucine residues. The major portion of rat preproinsulin has a half- life.of approximately'l'min in isolated rat islets, indicating that it is rapidly processed to form proinsulin (Patzelt et al., 1978). This is consistent with the signal hypothesis of Blobel (Blobel and Dobberstein, 1975), which suggests that.the hydrophobic pre-sequence serves to target the nascent polypeptide to be sequestered across microsomal membranes into the cisternae of the rough endoplasmic reticulum (RER). The pre-sequence is thought to be translated then rapidly cleaved from the nascent peptide by microsomal proteases. Proinsulin: It was first shown by Steiner and coworkers that the immediate precursor of insulin is proinsulin (Steiner et al., 1967: Steiner and Oyer, 1967). These investigators demonstrated that human insulinoma slices and isolated rat islets incorporated radioactive amino acids into insulin and a larger, acid soluble, insulin immunoreactive protein. Incorporation of labeled amino acids into the larger protein (9,000 daltons) occured earlier than into insulin, and the label could be chased into insulin in the presence of cycloheximide (CHX) or an excess of unlabeled amino acid. Also, limited hydrolysis of the larger protein lead to the formation of a protein that was indistinguishable from an authentic insulin standard. 23 Proinsulins from a variety of species have been isolated and characterized (for a review, see Permutt, 1980) . The structure and amino acid sequence.of human.proinsulin is Shown in figure 3. All proinsulins have the same basic structure: NHz-B chain-2 basic amino acids-C peptide-2 basic amino acids- A chain-COOK. The amino acid sequences and lengths of the A and B chains have been highly conserved throughout vertebrate proinsulins (Steiner et al., 1985). The sequences of the C peptides exhibit greater variation, but there does appear to be some conservation of C peptide length and overall net charge. The C peptide of proinsulin is thought to function to bring the A and B chains into an alignment that favors the formation of the interchain disulfide bridges and subsequent formation of mature insulin. The C peptide may also function to expand the length of the proinsulin molecule so that it can be transported across the RER membrane by the SRP (Signal Recognition Particle)-docking protein mechanism (see below), which might require a minimum peptide chain length (Steiner et al. , 1985) . Conversion of Proinsulin to Insulin: The conversion of proinsulin to insulin is mediated by the sequential action of two types of proteolytic enzymes (Steiner et al., 1985: Kemmler et al., 1971; Mackin and Noe, 1987). First, a trypsin-like protease cleaves the proinsulin molecule at the basic amino acid residues located at either 24 Figure 3. Structure and amino acid sequence of human proinsulin. Circles in dashes indicate sites of cleavage during conversion to insulin. Figure taken from Oyer et al. (1971). 25 end of the C peptide. Next, a carboxypeptidase B-like enzyme removes the C terminal basic residues remaining after tryptic cleavage. The trypsin-like enzyme is thought to be a thiol protease similar or identical to Cathepsin B (Docherty et al., 1982: Docherty et al., 1983). Conversion of proinsulin to insulin occurs in the lumen of the Golgi and in fi-cell secretory granules, where insulin and C peptide are found in equimolar amounts. Glucose, L-leucine, and tolbutamide have been reported to enhance the rate proinsulin to insulin conversion in isolated rat islets (Nagamatsu et al., 1987: Nagamatsu and Grodsky, 1988: Gold et al., 1986). Mannoheptulose prevented the glucose enhancement of conversion, suggesting that glucose metabolism mediated this effect (Nagamatsu et al., 1987). Conversion of proinsulin to insulin has been reported to be inhibited by exposure of islets to Tris(hydroxymethyl)- aminomethane (Tris) Halban. et al., 1986) or the sodium ionophore monensin (Gold et al., 1984). Glucose-Stimulation of Insulin Biosynthesis: It has been known for almost 25 years that glucose acutely and selectively stimulates insulin biosynthesis relative to total protein synthesis in isolated islets (Parry and Taylor, 1966: Howell and Taylor, 1966: Morris and Korner, 1970a: Tanese et al., 1970: Lin et al., 1971: Berne, 1975: Pipeleers et al., 1973: Ashcroft.et al., 1978). Early studies 26 indicated that acute glucose stimulation of insulin synthesis occured in isolated rat islets in the presence of ACT-D (Morris and Korner, 1970b: Permutt and Kipnis, 1972a). Itoh et al. (1978) and Itoh and Okamoto (1980) later demonstrated that glucose acutely stimulated proinsulin synthesis in isolated rat islets without changing the levels of PPImRNA. This acute stimulation was inhibited by CHX but not by a- amanitin, an inhibitor of RNA polymerase II. Glucose is currently believed to stimulate proinsulin synthesis acutely by three principle mechanisms (Permutt and Kipnis, 1972c: Permutt, 1974; Welsh et al., 1986): 1) Increasing ribosomal binding to PPImRNA leading to stimulation of initiation of translation: 2) Increasing the transfer of ribosome-bound PPImRNA from the cytosol to the RER: and 3) selective enhancement of translation elongation rates of nascent proinsulin molecules (at glucose concentrations less than 3.3 mM). ‘Leucine appears ‘to stimulate (proinsulin biosynthesis by these same mechanisms, but theophylline (and therefore, presumably cAMP) appears to only to stimulate translation initiation (Welsh et al., 1987). Glucose enhanced association of ribosome-bound PPImRNA with membranes of the RER is thought to be be mediated by the signal recognition particle-docking protein mechanism (Walter and Blobel, 1981: Meyer et al., 1982: Gilmore et al., 1982). This mechanism appears to be a key step in the regulation of translation of many secretory polypeptides (Walter and Blobel, 27 1981). Glucose metabolism generates an intracellular signal that stimulates insulin biosynthesis. Whether this signal is a metabolite of glucose or a metabolism-induced alteration in the level or charge state of an as of yet unidentified intracellular second messenger is not yet known. As reviewed by Ashcroft (1980), only sugars that are metabolized by pancreatic islets are capable of stimulating insulin biosynthesis. Glucose, mannose, and N- acetylglucosamine fall into this category, but others such as galactose, ribose, sucrose, xylitol, and sorbitol do not (Ashcroft et al., 1978a and 1978b: Ashcroft et al., 1976: Pipeleers et al., 1973: Lin and Haist, 1969). Insulin biosynthesis in isolated islets can also be stimulated by the triose D-glyceraldehyde (Jain et al., 1975), by pyruvate and lactate (Jain et al., 1978), and by inosine and dihydroxyacetone, two substances which are capable of being metabolized to triose phosphates (Ashcroft et al., 1978). Mannoheptulose (an inhibitor of hexokinase) inhibits the stimulation of insulin synthesis by glucose and mannose (Lin and Haist, 1969: Pipeleers et al., 1973), but not that stimulated by inosine or dihydroxyacetone (Ashcroft et al., 1978), or' D-glyceraldehyde (Jain. et al., 1975). Taken together, these results indicate that metabolism of glucose, rather than the presence of the glucose molecule itself, mediates stimulation of insulin biosynthesis. Furthermore, 28 glucose must be metabolized at least to triose phosphates in order for stimulation of insulin synthesis to occur. Potential intracellular signals produced by the metabolism of glucose include elevation of intracellular calcium or CAMP, altered cytosolic NADH/NAD” or NADPH/NADP‘ ratios, or changes in the activities of protein kinases (Ashcroft, 1980). Cyclic AMP probably does not mediate glucose stimulation of insulin biosynthesis, but might modulate the sensitivity of the insulin biosynthetic response to stimulation by glucose and other stimuli (Maldonato et al., 1977: Ashcroft et al., 1978: Lin and Haist, 1973: Schatz et al., 1973: Sandler et al., 1983). Elevation of intracellular calcium concentration does not appear to mediate the stimulation of insulin synthesis by glucose. Leinweber and Schatz (1982) and Lin and Haist (1973) have shown that W3 calcium concentrations in the medium in short-term labelling experiments results in 1.89m insulin synthesis and gegnegsed insulin release. Also, verapamil treatment inhibited insulin release but was without effect on insulin synthesis. This represents an important contrast to the well-documented role of elevated intracellular calcium in glucose-stimulation of insulin secretion (Hedeskov, 1980: Ashcroft, 1980). Alterations of cytosolic NADPH/NADP’ or NADH/NAD+ ratios generated by glucose metabolism are believed to be involved in 29 the regulation of insulin secretion (Hedeskov et al., 1987; Ashcroft, 1981) . It is uncertain whether the extent of reduction of cytosolic pyridine nucleotides is important in regulation of insulin biosynthesis. There are a number of B-cell proteins that are phosphorylated in response to stimulation by glucose (Christie et al., 1984: Colca et al., 1985). This is thought to occur as a result.of the activation.of protein kinases, although the mechanism of this activation is not understood. The identities of the kinases and phosphorylated proteins have not been well characterized to date. It is likely that some of these are involved in the regulation of insulin biosynthesis. e e a t at o s nthesis A wide variety of biological substances and pharmacological agents can modulate insulin biosynthesis. Arginine and lysine inhibit the synthesis of proinsulin in isolated rat islets, but other cationic amino acids do not (Schatz et al., 1975: Patzelt, 1988). Tolbutamide and glibenclamide inhibit glucose-stimulated proinsulin biosynthesis (Schatz et al., 1975). Agents that modulate the levels of islet polyamines influence insulin biosynthesis. Decreased islet putrescine, spermidine and spermine levels by exposure in M to difluormethylornithine, methylacetylenic putrescine, and ethylglyoxal bis (guanylhydrazone) is associated with decreased 30 proinsulin and total islet protein synthesis, and decreased islet PPImRNA levels (Welsh and Sjoholm, 1988). At present, the role of islet polyamines in the regulation of insulin biosynthesis is not well understood. Insulin biosynthesis is also inhibited by exposure of isolated rat islets to STZ (Maldonato et al., 1976) and ALX (Jain and Logothetopoulos, 1976). These agents are thought to act by a final common mechanism involving the production of DNA strand breaks, activation of the nuclear enzyme ADP- ribosyl transferase (ADPRT), and depletion of cellular NAD levels leading to inhibition of proinsulin biosynthesis (Uchigata et al., 1982). ALX is capable of initiating the production of reactive oxygen intermediates which can cause DNA strand breakage (Uchigata et al., 1982). STZ is believed to be metabolized to species that alkylate DNA (Okamoto, 1985). Yamamoto et al. (1981) demonstrated that ALX (1 mM) and STZ (2 mM) rapidly produced DNA strand breaks and increased ADPRT activity in isolated islets. ADPRT activation was followed by depletion of cellular NAD levels and inhibition of proinsulin synthesis. Inhibitors of ADPRT such as nicotinamide or picolinamide prevented NAD depletion and inhibition of proinsulin synthesis. Although the reversal of the depletion of NAD was essentially complete at the concentrations of ALX, STZ, ADPRT inhibitors used, the reversal of ALx-induced inhibition of proinsulin synthesis was 31 only'partial, suggesting that ALX can inhibit the synthesis of proinsulin by a mechanism other than through depletion of cellular NAD levels. Similar findings were subsequently reported by Uchigata et al. (1983) who examined the in 2i§gg proinsulin biosynthetic capacity of islets isolated from rats administered the same agents in 212g. There is some uncertainty regarding the specificity of .ALX and STZ inhibition of insulin synthesise Gunnarson (1975) isolated islets from STZ and ALX-treated rats and evaluated proinsulin and total islet protein synthesis at low and high glucose concentrations in 213:9. At high glucose (16.7 mM), prior in 2129 exposure to STZ and ALX inhibited proinsulin synthesis and total islet protein synthesis to similar extents. At low glucose (3.3 mM) the inhibition appeared to be more specific for proinsulin synthesis relative to total islet protein synthesis. Jain and Logothetopoulos (1976) examined the ability of ALX to inhibit protein synthesis directly in isolated islets exposed in 2i;;_. These investigators reported that ALX (1.25 mM) appeared to inhibit proinsulin synthesis without decreasing total islet protein synthesis. Maldonato et al. (1976) examined the specificity of STZ inhibition of protein synthesis in isolated rat islets exposed in 21229. When islets were exposed to STZ for 60 min prior to measurement of proinsulin synthesis, concentrations exceeding 0.22 mM produced an apparent specific inhibition of proinsulin synthesis, while concentrations below 0.22 mM 32 inhibited proinsulin synthesis and total islet protein synthesis to similar extents. Shorter exposure durations were associated with decreased proinsulin specificity at each STZ concentration. An important component of this analysis was the correction for the fact that insulin is synthesized only in fl-cells, while total islet protein synthesis reflects biosynthetic activity in all islet cells. By correcting their results in this manner, it became evident that STZ exhibited selectivity for fi-cells relative to non-fi-islet cells, but did not exhibit selectivity for insulin relative to other fi-cell proteins. Exposure of isolated mouse islets to STZ for 30 min followed by culture ‘without STZ for 6 days results in depletion of insulin content and PPImRNA levels, and inhibition of glucose-stimulated proinsulin synthesis, insulin release and oxygen uptake, without alterations of islet DNA, RNA, and total protein.synthesis rates (Eizirik.et al., 1988). The decrease of proinsulin synthesis in these experiments did not represent an.acute inhibitory effect.of STZ. Instead, the initial STZ-induced fi-cell damage leads to a long-term compromise of fi-cell function involving impairment of islet respiration, leading to decreases of PPImRNA, and ultimately to inhibited proinsulin synthesis. The selectivity for decreasing proinsulin synthesis and PPImRNA is consistent with fi-cell specificity in the cytotoxic effects of STZ. 33 It has been proposed that the glucose moiety present within the STZ molecule facilitates the recognition and uptake of STZ by islet fi-cells (Cooperstein and Watkins, 1980), leading to the fi-cell specificity of STZ. The basis for 8- cell specificity of ALX actions is believed to be related to the rapid uptake of ALx by islet fi-cells, coupled with high sensitivity of B-cells to reactive oxygen intermediates (Malaisse et al., 1982). f ts o C t e n Pancreas CPH produces selective structural and functional alterations of insulin-producing fi-cells in the endocrine pancreas. These actions are shared by other structurally- related compounds such as 4-diphenylmethy1piperidine (4-DPMP) , N-methyl-4-DPMP, cyclizine, pizotyline, azacylonol, and the CPI-I metabolite, desmethylcyproheptadine (DMCPH) (Hintze et al., 1977: Fischer et al., 1973). Even though CPH is recognized as an antihistaminic, antiserotonergic agent (Stone et al., 1960), the actions of CPR on insulin-producing cells are not believed to be due to histamine or serotonin receptor antagonist activities. Some of the structurally-related agents listed above have similar actions on insulin-secreting cells, but lack histamine or serotonin receptor antagonist activity (Engelhardt et al., 1965). Administration of a single oral dose of CPR to rats (45 mg/kg, po.) results in a 50% depletion of pancreatic insulin 34 content within 24 hours (Hintze et al., 1977b). Continued daily administration of this same dose for two weeks will maintain a hyperglycemic and.hypoinsulinemic state in treated animals (Rickert et al., 1975). CPR-treated animals exhibit non-fasting hyperglycemia, glucose intolerance and striking morphological alterations in islet fi-cells including loss of secretory granules and vesiculation of the RER leading to the formation of large cytoplasmic vacuoles (Longnecker et al., 1972). Withdrawal of CPR resulted in regranulation of B- cells, a return of pancreatic insulin content and normoglycemia. CPH actions appear to ‘be selective fer pancreatic,B-cells as other islet.endocrine cell types (a-, 6- and PP-cells), and cells of the exocrine pancreas were unaffected. Exposure of whole rat islets in_21§;g to 50 uM CPR for 6 days depleted islet insulin content (Halban et al., 1979), an effect attributed by the authors to inhibition of insulin synthesis, but probably due to non-specific islet cell toxicity. The high concentration of CPI-I used produced irreversible decreases in islet glucagon content, altered 8- cell secretory granule shape, and evidence of intense lysosomal activity in islet a- and.fi-cells. 'These effects are unlike those seen after CPI-I treatment of rats in M (Longnecker et al., 1972). At present there is uncertainty iregarding the ability of CPR to deplete insulin content in 11:19. 35 CPR inhibits insulin release from perfused rat pancreatic segments and intact pancreata (Rickert et al. , 1975: Joost et al., 1974: Joost et al., 1976a), and isolated rat and mouse islets (Joost et al., 1976b: Richardson et al., 1975: Richardson, 1976: Donatsch et al., 1980). This effect is believed to be due to inhibition of voltage-dependent calcium movement into islet B-cells. Donatsch et al. (1980) demonstrated that insulin secretogogues that are dependent upon voltage-dependent entry of extracellular calcium (eg. high K‘, glucose, and tolbutamide) exhibited greater sensitivity to CPI-I inhibition than those not dependent on extracellular calcium entry (eg. veratridine, theophylline, and the calcium ionophore A23187) . CPR-inhibition of insulin secretion may play a role in the non-fasting hyperglycemia seen after repeated oral administration of the drug. CPI! treatment (single dose, 45 mg/kg, po.) causes rat pancreatic proinsulin levels to decline before changes of insulin occur (Hintze et al., 1977b). Pancreatic proinsulin declined to 20% of control at 6 hr after dosing, while insulin levels decreased to 20% of control 18 hr after dosing. These observations suggested that CPH treatment caused the suppression of proinsulin synthesis in islet fi-cells, and that this inhibition caused depletion of insulin. Confirmation of the inhibitory action of CPR on proinsulin synthesis was obtained from experiments involving acute CPH exposures of isolated rat pancreatic islets in L139 (Hintze et al., 36 1977b). Results from these studies indicated that the drug acutely and selectively inhibited proinsulin synthesis. Exposure of islets for 40 min to 16 an CPH decreased the incorporation of 3H-leucine into insulin and proinsulin by 75% while incorporation into ‘total cellular' protein ‘was not affected. It is currently believed that metabolites of CPR might play a role in the pancreatic effects of the drug. In the rat, CPH is N-demethylated to form DMCPH (Wold et al., 1972), which is further modified to DMCPH-epoxide (Hucker et al., 1974: Hintze et al., 1975). DMCPH-epoxide appears to be quite stable, as it can be found intact in rat urine after an oral dose of CPR (Hintze et al., 1975). Pretreatment of rats with SRP-525A or 3,5-diethoxycarbonyl-2,6-dimethyl-4-ethyl-1,4- dihydropyridine has been shown to inhibit the formation of DMCPH-epoxide, and to protect animals from DMCPH and CPH- induced depletion of pancreatic insulin content (Chow et al., 1989). Furthermore, it has been shown that in isolated rat islets, DMCPH and DMCPH-epoxide are 10 and 22 times m potent, respectively, than CPH in inhibiting proinsulin synthesis. Interestingly, DMCPH and DMCPH-epoxide appear to be 2 and 9-fold Legs potent, respectively, than CPH as inhibitors of insulin release (Chow et al., 1990). Taken together, these findings indicate that the DMCPM and DMCPH- epoxide, might mediate CPR-induced pancreatic insulin depletion in the rat. 37 Early studies examining morphological alterations of pancreatic tissue indicated that mice were resistant to CPH- induced islet fi-cell toxicity (Wold et al., 1971). It was later shown that mice metabolize CPH to DMCPH (Wold and Fischer, 1972) but fail to produce DMCPH-epoxide (Hintze et al. , 1975) . These findings suggested that the relative insensitivity of the mouse to the pancreatic effects of CPH might be related to the lack of formation of DMCPH-epoxide. There is some uncertainty regarding the susceptibility of humans to the pancreatic effects of CPH. As CPH is currently used therapeutically to treat a variety of disorders (Douglas, 1985), this point is of some practical concern. Humans do not appear to form DMCPH-epoxide, as this CPH metabolite was not detected in human urine (Hintze et al., 1975). The major urinary metabolite of CPH in humans is a quaternary ammonium glucuronide conjugate (Kennedy et al., 1977). Administration of CPH to children has been associated with abnormal glucose tolerance and.hlunted insulin-secretory responses (Golander and Spirer, 1982). CPH (1-100 uM, 3 hr) also appears to inhibit insulin synthesis in isolated human islets exposed in 21219 (Jahr et al., 1981). However, no information ‘was. given for’ CPH effects on. total protein synthesis, so the CPH sensitivity of insulin biosynthesis in humans remains uncertain. 38 ijectiveszgntionale The overall objective of these studies was to investigate the mechanism of CPH-induced inhibition of insulin biosynthesis and depletion of pancreatic insulin content. Prior to this study, CPH was known to inhibit insulin synthesis in isolated rat islets in vitro, and to decrease proinsulin levels in the rat pancreas prior to depleting insulin content in 2129 (Hintze et al. , 1977) . These results suggested that CPH inhibited proinsulin synthesis in M9, but the mechanism of this inhibition was not understood. Results from initial experiments performed in collaboration with Dr. S.J. Giddings of Washington University, St. Louis, MO, indicated that administration of a single, insulin-depleting dose of CPH to rats (45 mg/kg, po.) produced a rapid and reversible decrease of pancreatic PPImRNA levels. Figure 4 shows a Northern analysis of pancreatic RNA samples from control and CPH-treated animals at 0, 24 and 48 hr after administration of CPH or distilled water. PPImRNA levels in CPH-treated animals were decreased relative to control at 24 hr, but returned towards control at 48 hr after dosing. These results provided the first indication that CPH-treatment decreased PPImRNA levels in the pancreas. This finding lead to the hypothesis that CPH-induced decreases in PPImRNA cause the inhibition of proinsulin synthesis and depletion of insulin content. In order to test this hypothesis, a series of experiments 39 0 24 43 CON CON CPH CON CPH 1 2 3 4 5 6 7 8 9 10 pm» .Q..i§ isgfi h! -_———_-.. Figure 4. Northern analysis of rat pancreatic PPImRNA at 0, 24 and 48 hr after CPH administration. Pancreatic RNA samples were prepared from. control and CPH-treated rats at the indicated times after a single dose of CPH (45 mg/Kg, po) or distilled. wateru Bands represent samples. obtained from separate animals. Lanes 1 and 2: zero-time controls: 3 and 4: 24 hr controls: 5 and 6: 24 hr CPH-treated: 7 and 8: 48 hr controls: 9 and 10: 48 hr CPH-treated. 40 were performed in which CPH-induced alterations of PPImRNA, proinsulin and insulin levels were examined 1n 2129, and PPImRNA, insulin, and proinsulin synthesis 1); m. The specific aims of these experiments are described below. 1.) To examine the time course of CPH-induced alterations of rat pancreatic insulin, proinsulin and PPImRNA. Evaluation of the time course of events following a single oral dose of CPH would provide important information related to the mechanism of CPH actions. If CPH inhibition.of proinsulin synthesis and depletion of insulin content are due to decreases of PPImRNA, proinsulin and insulin levels should decline 9199; a drug- induced decline of PPImRNA. If the chronology of CPH effects occurs otherwise, alterations of PPImRNA levels would not be responsible for the inhibition of insulin synthesis and insulin depletion. 2.) To determine whether acute CPH-inhibition of proinsulin synthesis in isolated rat islets is associated with decreased PPImRNA levels. Measurement of pancreatic proinsulin levels provides an indirect estimate of proinsulin synthesis 1n 2129. Decreases of pancreatic proinsulin levels 1n 2129 are consistent with, but may not be directly related to, CPH inhibition of proinsulin synthesis. In order to test directly whether CPH-inhibition of proinsulin synthesis occurs in concert with, or independent of, changes in PPImRNA levels, 41 experiments were performed in isolated rat pancreatic islets. Use of isolated rat islets permits the direct evaluation of alterations of proinsulin synthesis by measurement of incorporation of radiolabelled amino acids into proinsulin. 3.) To characterize the effects of CPR in cultured dispersed rat pancreatic islet cells in 21999. Although CPH had been shown to inhibit insulin synthesis directly in isolated rat islets (Hintze et al., 1977b), there‘was uncertainty regarding the ability of CPH to deplete fi-cell insulin content 1nj219;_. In.experiments with.cultured whole rat islets, CPH appeared to elevate insulin content at low concentrations (0.05 and 0.5 uM) and deplete insulin at higher concentrations (50 uM and greater) (Halban et al., 1979). The high CPH concentrations necessary to produce insulin depletion in cultured pancreatic islets made interpretation of these findings difficult. Experiments were conducted with dispersed islet cells to examine the ability of CPHI to idirectly inhibit insulin synthesis, secretion, and proinsulin to insulin conversion during acute exposure, and to deplete cellular insulin content during prolonged exposure to CPH 1n 219:9. Use of cultured dispersed islet cells avoided some of the potential problems associated with the use of cultured whole islets [ie. islet necrosis due to inadequate nutrient and oxygen.supply of cells .located in the islet center (Andersson, 1972)]. CPH- inhibition of insulin synthesis and secretion were examined to 42 determine whether the acute CPH effects in dispersed islet cells were similar to those previously reported in whole islets (Hintze et al., 1977b: Chow et al., 1988). Potential effects on proinsulin to insulin conversion were evaluated because this process was one of the steps of the insulin biosynthetic process that had not been previously examined for sensitivity to inhibition by CPH, and inhibition of conversion could contribute to depletion of pancreatic proinsulin and insulin. 4.) To examine the effects of CPR in RINmSF and HIT-T15 cells. RINmSF cells were established in culture from a radiation-induced transplantable rat islet cell tumor (Chick et al., 1977: Gazdar et al., 1980). HIT-T15 cells are an SV40-transformed.hamster*pancreatic islet cell line (Santerre et al., 1981). These cell lines have been used by others as models to study regulation of insulin synthesis and secretion (Praz et al., 1983: Welsh et al., 1985: Wollheim and Pozzan, 1984: Gold et al., 1988: Hammonds et al., 1987: Meglasson et al., 1987). We initially sought to determine whether RINmSF and HIT-T15 cells could serve as useful models with which to examine the mechanism of CPH actions. Experiments were conducted.to confirm that CPH inhibited insulin secretion and synthesis, and depleted cellular insulin content in both cell lines. Additional studies examined the abilities of other DPMP compounds and non-specific inhibitors of protein 43 synthesis to deplete cellular insulin content of RINmSF cells. An important contrast between the two cell lines is that HIT- T15 cells respond. to glucose stimulation. with increased insulin synthesis and.secretion (Hill and Boyd, 1985: Hammonds et al., 1987) , whereas RINmSF cells are unresponsive to glucose (Praz et al., 1983: Welsh et al., 1985). This difference allowed inferences to be made regarding potential interactions between CPH and glucose signalling mechanisms within insulin-producing cells. 5.) To determine whether CPH-induced inhibition of insulin synthesis and depletion of cellular insulin content in RINmSF and HIT-T15 cells is associated with alterations of PPImRNA levels. Results from initial experiments indicated that CPH inhibited insulin synthesis and secretion and depleted insulin content in both cell lines. Additional studies were performed to examine the ability of CPH to alter PPImRNA levels during short and long-term exposures to inhibitory concentrations of the drug. Correlation of potential CPH-induced alterations of insulin synthesis, insulin content, and PPImRNA levels provided additional information regarding the mechanism of CPH actions. 6.) To examine whether other insulin-depleting diphenyl- methylpiperidine compounds also decrease rat pancreatic PPImRNA levels. Oral administration of 4-diphenylmethyl- 44 piperidine (4-DPMP: 45 mg/kg, po., single dose) has been shown to deplete pancreatic insulin content in the rat, while the same dose of 2-DPMP“was without effect (Hintze et al., 1977b). The abilities of 4-DPMP and 2-DPMP (Fig. 5) to decrease PPImRNA levels were evaluated to further investigate the correlation between chemically-induced decreases of PPImRNA and pancreatic insulin content in rats. 7.) To examine the ability of CPH to decrease PPImRNA levels and deplete pancreatic insulin levels in mice. There are species differences in CPH metabolism, and in susceptibility to some effects of CPH in the endocrine pancreas. Rats and mice N-demethylate CPH, but only rats formia stable epoxide of the demethylated metabolite (Wold and Fischer, 1973: Hintze et al., 1975). It has been proposed that this epoxide may be partly responsible for the actions of CPH in rats, as it appears to be approximately 20-times more potent than CPH as an inhibitor of proinsulin synthesis in isolated rat islets (Chow et al., 1988). Mice have been reported to be less sensitive than rats to CPH-induced alterations in fi-cell morphology (Wold et al., 1971), but the insulin depleting effects of CPH in the mouse have not been adequately characterized. Experiments were conducted in mice to gain additional information regarding CPH-induced decreases of .PPImRNA and insulin levels, and to further examine the nature of species differences in CPH action. 45 4-DPMP C PH 2-DPMP Figure 5. Structures of CPH, 4-diphenylmethylpiperidine (4- DPMP) and 2-DPMP. R THODB HAIEBIALfi Animals: Male Sprague-Dawley rats weighing 175-200 g and male Swiss Webster mice weighing 25-30 g were purchased from Charles River Breeding Company (Portage, MI). Rats were housed two or three animals per cage, and mice were housed four or five animals per cage in the Laboratory Animal Care Service facility in the basement of the Life Sciences Bldg., Michigan State University. Upon arrival animals were allowed to acclimate for 4 to 6 days. Animals received Purina Rodent Chow and clean tap water ad libitum until the time of experiments. Clonal Insulin-Producing Cell Lines: RINmSF cells were the generous gift of Dr. Paolo Meda (Geneva, Switzerland), and.HIT-T15 cells were kindly provided by Dr. Robert Santerre (Eli Lilly, Indianapolis, IN). Chemicals and Reagents: Cyproheptadine was obtained as the hydrochloride ‘monohydrate from the Merck Institute for Therapeutic Research 46 47 (West Point, PA). The purity was checked by HPLC as described by Chow and Fischer (1986). 4-diphenylmethylpiperidine (4- DPMP) was obtained from Pfaltz and Bauer (Flushing, NY). 2- DPMP was synthesized by Dr. H. Aboul-Enein (Hintze et al., 1977). Guinea pig anti-insulin serum was obtained from the Department of Pharmacology , Un ivers ity of Indiana (Indianapolis, IN). Highly purified rat insulin, bovine proinsulin, and bovine insulin were purchased from Novo Biolabs (Danbury, CT). Collagenase, bovine serum albumin (BSA), bisbenzimide trihydrochloride (Hoescht No. 33258), and Ficoll were obtained from Sigma Chemical Company (St. Louis, MO) . Diethylpyrocarbonate (DEPC) was purchased from Kodak Co. (Rochester, NY). Reinforced nitrocellulose membranes were purchased from Schleicher and Schuell Co. (Keene, NH). Roswell Park Memorial Institute (RPMI)-1640 culture medium (180 mg/dl D-glucose) , Ham's F-12 Nutrient Mixture (200 mg/dl D-glucose), fetal bovine serum, dialyzed horse serum, penicillin/ streptomycin, and trypsin/EDTA were obtained from Grand Island Biological Company (Grand Island, NY). Trypsin was purchased from DIFCO (Detroit, MI). All other reagents were of the highest quality commercially available. Glassware and Plasticware: Culture plasticware was purchased from Falcon Labware (Oxnard, CA). Polypropylene microfuge tubes were purchased from Sarstedt Co. (Princeton, NJ). Sterile, polypropylene 48 centrifuge tubes (15 and.50 ml) were obtained from.Corning Co. (Corning, NY). All glassware that came into contact with insulin-containing samples was coated with Prosil-28 (PCR Research Chemicals Inc., Gainesville, FL), an organosilane surface-treating solution that reduces protein binding. All glassware that was used in the RNA experiments was treated with 0.1% DEPC to inactivate RNase, then autoclaved prior to use (Maniatis et al., 1982). Radioisotopes: 3H-leucine (NET-135H: spec. act. 40-60 Ci/mM) and ["EI]- labelled porcine insulin (NEx-196: spec. act. 2200 Ci/mM)‘were purchased from New England Nuclear Corporation (Boston, MA). Radiochemical purity of 3H-leucine and [‘ZSIJ-labelled insulin was checked by TLC (Randerath, 1966) and HPLC (Halban et al., 1986) , respectively. [cum-labelled cytidine-5'-triphosphate (cat. #32015H; spec. act. >600 Ci/mM) was purchased from ICN Radiochemicals (Irvine, CA). garages 1- IE_!I!Q_£IREBINENT§1 1. o the at Dosing and Sacrifice: This experiment was designed to closely examine the early time course of CPH effects on pancreatic insulin, proinsulin and PPImRNA. Animals were administered a single dose of CPH 49 (45 mg/kg, po.) at approximately 9 am. This treatment has been shown to decrease rat pancreatic proinsulin and insulin by’6 and 24 hr, respectively (Hintze et al., 1977). Groups of control animals received the same volume of distilled water (1.5 ml/100 g body weight). At 1.5, 3, 6, 10, and 24 hr following dosing, groups of four control and four CPH treated rats were individually weighed and anesthetized with sodium pentobarbital (45 mg/kg, ip). After adequate anesthesia was obtained, pancreata were surgically removed, then the animals were sacrificed. A zero time control group was included, in which four animals received vehicle then were immediately anesthetized, and processed as described above. For all animals, pancreata were removed and weighed, and portions were processed either for acetic acid extraction of insulin and proinsulin (Hintze et al., 1977), or for guanidine isothiocyanate extraction of total pancreatic RNA (Chirgwin et al., 1979). Complete removal of feed was necessary as CPH- treated rats will eat less than untreated controls, and it is known that decreased feed intake is associated with lowering of PPImRNA levels relative to ad libitum fed animals (Giddings et al., 1981). In order to correct for this potential confounding variable, all animals had their feed withdrawn at the time of dosing. 50 Estimation of Pancreatic Insulin and Proinsulin: Portions (approx. 50 mg) of the splenic region of each pancreas were weighed, then quickly homogenized in cold 1N acetic acid (20 mg pancreas per ml). The homogenates were vortexed, heated at 95' for 5 min, stored at 4‘ for 12-15 hr to fully extract proinsulin and insulin, then centrifuged (10,000 x G, 10 min, 4'). The clear supernatents were frozen at -20' until radioimmunoassay (RIA) of total immunoreactive insulin (IRI) (Starr et al., 1977), or Bio-Gel P30 separation of proinsulin and insulin (Hintze et al., 1977), followed by RIA. For measurement of pancreatic proinsulin, 100 pl aliquots of the acetic acid tissue extracts were pooled from the four animals in each treatment group and chromatographed on Bio-Gel P30 columns (.9 x 50 cm) (BioRad Inc., Rockville Center, NY). The Bio-Gel columns were precoated with BSA to minimize non- specific protein binding. This was accomplished by loading and eluting 4 m1 of 30% BSA.in phosphate buffered saline (PBS: pH 7.4) prior to chromatography of samples. The samples were eluted from the columns with.1 N acetic acid at a flow rate of .06-.08 ml/min. Fractions were collected at 10 min intervals (LKB‘Ultrorac II fraction collector) into Prosil 28-coated 16 x 100 mm.borosilicate test tubes containing 100 pl of 30% BSA in phosphate buffered saline (PBS: pH 7.4). Fractions were . mixed, transferred to 1.5 ml polypropylene microfuge tubes, and frozen (-20') until RIA of immunoreactive proteins. Peak 51 identities were confirmed by comparison to retention volumes of purified bovine proinsulin and insulin standards which were chromatographed separately. Recovery for each sample was checked by adding before chromatography approximately 1000 cpm (1.5 pg) of 125I-labelled monocomponent porcine insulin. Radioactivity in fractions eluted from the columns were counted in a gamma radiation counter (Model 4/600 Plus, ICN Micromedic Systems Inc., Cleveland, OH). Recoveries were consistently between 90 and 100 % of the radioactivity loaded onto the columns, and.no corrections of data for recovery'were necessary. Estimation of Total Pancreatic RNA, PreproinsulintmRNA and.n- Actin mRNA: Portions (approx. 100 mg) of the splenic region of each pancreas were weighed then quickly homogenized in 5 ml of chilled 4 M guanidine isothiocyanate, 0.75 mM sodium citrate (pH 7.4) (GIT buffer). Sodium lauroyl sulfate (SLS) was added to the tissue homogenates to a final concentration of 0.1%. The homogenates were then vortexed and kept on ice until all samples for a given time point were obtained. Total RNA was prepared from the tissue homogenates by repeated ethanol precipitations out of progressively smaller volumes of guanidine hydrochloride, followed by water extraction, reprecipitation and final dissolution in DEPC-treated sterile MilliQ water (Millipore Inc., Bedford MA) (10 ug RNA/m1), as 52 described by Chirgwin et al. (1979). Concentration and yield of RNA was calculated from UV absorbance at 260 nm. Purity of RNA preparations was estimated from the ratio of UV absorbance at 260 and 280 nm. Values for this ratio were always between 1.9 and 2.2. Pancreatic PPImRNA levels were quantitated by dot blot hybridization as described by Orland et al. (1985) . Briefly, samples of total RNA (5-20 pg) were heat denatured (60°, 15 min) in 6X SSC (0.9 M sodium chloride, 0.09 M sodium citrate), 7.4% formaldehyde, then applied to reinforced nitrocellulose membranes (Schleicher and Schuell Inc., Keene, NH) using a vacuum manifold (BRL, Life Technologies Division, Gaithersburg, MD). The blots were baked (60 min, 80' under vacuum) , prehybridized then hybridized with [um-labelled cRNA probes for rat insulin I mRNA, or chicken fi-actin mRNA. The prehybridization and hybridization were carried out exactly as describe by Giddings and Rotwein (1984) , except for the addition of 10% dextran sulfate and omission of 1% glycine in the prehybridization buffer, and an incubation temperature of 60' . The dextran sulfate was substituted for glycine as recommended by S.J. Giddings (personal communication) . The incubation temperature of 60' is more suitable for hybridization of cRNA probes to RNA samples than the 42' temperature used for cDNA probes. The labelled probes were produced by 1n 219199 transcription of ‘pSP64 expression vectors (Promega Biotec Inc., Madison, WI) in the presence of [32P]- 53 labelled rCTP. These pSP64 plasmids contain cDNA inserts of portions of rat insulin I mRNA or chicken fi-actin mRNA. The expression vector containing the rat I insulin.cDNA insert was generously supplied by Dr. S.J. Giddings (Washington Univ., St. Louis, MO). As the rat I insulin cRNA probe contains sequences complementary to coding regions present in both rat insulin I and II mRNAs (300 bases of exons II and III), the probe detects both rat insulin mRNAs (collectively referred to as PPImRNA). The plasmid containing the chicken fi-actin cDNA insert was kindly provided by Dr. J. Wang (Dept. of Biochemistry, MSU). The chicken B-actin insert is an almost full-length cDNA. The probe produced from the insert hybridizes with DNA from a variety of species (Cleveland et al., 1980), and detects a single band in rat pancreatic RNA samples corresponding to rat fi-actin mRNA. After hybridization, membranes were washed [twice for 30 min in 2X SSC, 0.1% SDS (sodium dodecyl sulfate) at 65’, followed by once for 30 min in 0.1x SSC, 0.1% SDS at 65'] and dried, then autoradiograms were obtained using intensifying screens (Kodak, Rochester, NY) at -70' (Maniatis et a1, 1982). The relative amounts of PPImRNA in each sample were determined by densitometric analysis of the autoradiograms. In order to confirm that RNA was not degraded during the extraction procedure, aliquots of selected RNA samples were also subjected to Northern analysis (Maniatis et al. , 1982: Fourney et al., 1933). 54 For Northern analysis, RNA samples (10 pg in l or 2 pl sterile DEPC-treated water) were diluted 4-fold with RNA sample buffer (1.0:5.0:0.5:1.5 (v:v), 10X MOPS/EDTA (0.2 M MOPS [3-(N-morpholino)propane sulfonic acid], 50 mM sodium acetate, 10 mM EDTA, pH 7.0): formamide: 37% formaldehyde: DEPC-treated water) and heat denatured (65', 15 min). One pl of ethidium bromide solution (1 mg/ml in DEPC-treated water) and a minimum of 1/6 vol of 6X gel loading buffer (0.25% bromophenol blue, 0.25% xylene cyanol, 30% glycerol) were added to the samples. The RNA was then fractionated by electrophoresis (60-80 V, 3-4 hr) on 1% agarose gels (11 x 14 cm) containing 0.66 M formaldehyde and 1X MOPS/EDTA buffer The electrophoresis buffer was 1X MOPS/EDTA. Following electrophoresis, the gels were photographed over a short wave UV transilluminator using a Kodak Wratten #9 yellow filter. RNA in the gels was transferred to reinforced nitrocellulose membranes by capillary blotting (Maniatis et a1, 1982) in 20x SSC. The blots were then handled as described for the dot blot analysis above. 2. - - n the at Experiments were conducted to determine whether the relative abilities of close structural analogues of CPH to 'deplete insulin content in the rat (Hintze et al., 1977) are correlated with their relative abilities to decrease pancreatic PPImRNA levels. Results from the rat time course 55 experiment described above indicated that the maximum effect of CPH on PPImRNA levels was observed at 10 hr after dosing, while pancreatic IRI was not depleted until 24 hr after dosing. For this reason, the IRI depleting effects of 4-DPMP and 2-DPMP were evaluated at 24 hr after dosing, and the PPImRNA altering abilities were examined at 10 hr after dosing. Groups of five rats were administered single doses of 4- DPMP, or 2-DPMP (45 mg/kg, po.) at approximately 9 am. Similar dose levels of 4-DPMP have been reported by Hintze et a1. (1977) to deplete pancreatic IRI content by 70%, while equimolar doses of 2-DPMP were without effect on pancreatic IRI. Control rats received the same volume of distilled water (1.5 ml/100 g body weight). As for experiments with CPH, all animals had their food withdrawn at the time of dosing. Ten hr after receiving these doses, animals were weighed and sacrificed. The pancreata were removed and processed for estimation of PPImRNA and fi-actin mRNA levels by the same methods that were used in the rat time course experiment described above. Additional groups of animals were administered the same doses of 4-DPMP or 2-DPMP (or vehicle) and sacrificed 24 hr later. For these animals, the pancreata were removed and processed as described above for the estimation of pancreatic IRI. This was done to provide confirmation of the IRI-depleting effects of 4-DPMP and the inactivity of 2-DPMP. Since this experiment was not intended 56 to characterize the time course of events following administration of 4-DPMP and 2-DPMP, pancreatic proinsulin values were not obtained from these animals. 3. cc 0 C e use At present there are no published data on CPH-induced depletion of pancreatic insulin content in mice. Preliminary results from V. Virayotha (M.S. Thesis, Univ. Iowa, 1983) indicated that administration of eight daily doses of CPH (45 mg/kg, po.) caused an 80% depletion of mouse pancreatic insulin content. In. order' to. gain further information regarding the sensitivity of 'the ‘mouse to CPH effects, experiments were performed to evaluate the time course of CPH- induced depletion of mouse pancreatic insulin content. For these experiments, animals received doses of 45 mg CPH/kg (po.) at 9 am on each of eight consecutive days. Groups of control mice received the same volume of distilled water (1.5 ml/100 g body weight). Groups of control and CPH-treated animals (N=6) were sacrificed at 12 hr and 1, 2, 4, 6, and 8 days after starting dosing, and at 12 hr and 1, 2, and 4 days after the eighth daily dose. At each time point, animals were weighed then sacrificed by cervical dislocation. Whole blood was taken from the heart by cardiac puncture. The blood was allowed to clot, then was centrifuged (1500 x G, 5 min), and the serum transferred to 1.5 ml microfuge tubes. Glucose concentrations were determined in the serum samples by the 57 glucose oxidase technique using a Beckman Glucose Analyzer II (Beckman Instruments, Fullerton, CA). Pancreata were removed and immediately homogenized in cold 1 N acetic acid. IRI was extracted from the mouse pancreata by the same procedure outlined above for extraction of pancreatic insulin from the rat. Additional studies were also conducted to examine whether CPH could decrease PPImRNA levels in the mouse pancreas. For these experiments, groups of animals (N=4) were sacrificed at 0, 12 and 24 hr after administration of a single dose of CPH (45 mg/kg, po., 9 am) or vehicle (distilled water, 1.5 ml/100 g body weight). Control and CPH-treated mice were partially feed-restricted to compensate for CPH-induced decreases in food consumption. All animals were fasted during the first 12 hr after dosing (9 am to 9 pm), then received half of the food that control animals normally consume during the second 12 hr period (9 pm to 9 am). Animals were sacrificed in the same manner as for the 8 day experiment described above. However, for this study, portions of the pancreas were taken for acetic acid extraction of immunoreactive insulin, or guanidine isothiocyanate extraction of pancreatic total RNA as described previously for 1n 21V9 experiments in the rat. 58 1. c o ro sulin Bios esis a d 22IEBEA_L1I21£.i£.1£21£§2§.8£§.l§12£§ Islet Isolation and Preculture: Experiments were conducted to measure PPImRNA levels after 30 min exposure of isolated rat islets to concentrations of CPH that inhibit proinsulin biosynthesis. For these experiments, islets were isolated from male Sprague-Dawley rats (175-200 g) by collagenase digestion (Lacy and Kostianovsky, 1967), and placed in culture overnight in RPMI 1640 medium supplemented with 10% fetal bovine serum, 200 pg/ml penicillin and 200 mU/ml streptomycin to recover from the isolation procedure. Typically 1000 to 2500 islets were isolated from 4 to 6 rats. The following day, the islets were distributed into 15 ml conical polypropylene centrifuge tubes (Corning Co., Corning, NY) in batches of 200 per tube for measurement of proinsulin biosynthesis, or 250 per tube for determination of PPImRNA levels. Measurement of Proinsulin Biosynthesis in Isolated Islets: Islets were first washed with 10 ml of a Krebs Ringer bicarbonate buffer (KRB: pH 7.4), and preincubated for 30 min at 37' in a C02 incubator (5 % C02, 95% air) in 5 ml of preincubation buffer (KRB supplemented with 13 essential amino acids (Eagle, 1955) including L-leucine, and 300 mg/dl d- glucose: Preincubation buffer was pre-equilibrated with 59 conditions in the C02 incubator.) After preincubation, islets were briefly pelleted (500 x G, 2-5 min), buffer was removed, and 1 ml of incubation/labelling buffer added. This buffer contained the same constituents as the preincubation buffer except that there was 0 or 10 pM CPH, and 50 pCi per ml of HP- leucine present in the labelling buffer and no unlabelled L- leucine present. This concentration of CPH has previously been shown to inhibit proinsulin synthesis acutely and selectively in isolated rat islets (Chow et al., 1989). Islets were incubated at 37' C02 under 5% C02, 95% air with periodic gentle agitation for 30 min. As the time required for proinsulin to insulin conversion is approximately 45 min (Steiner and Tager, 1979) , the 30 min 3H-leucine incorporation period permitted the measurement of proinsulin synthesis. At the end of the labelling period, islets were washed twice with 10 ml cold chase/stop buffer (KRB containing 50 mM unlabelled L-leucine and no glucose). After the second wash, the islets were resuspended in 200 pl of 1 N acetic acid, heated at 95' for 5 min, then sonicated [3 x 10 sec with a Virsonic 300 sonicator (Virtis, Gardiner, NY) at setting 4 (microprobe)]. The sonicates were stored at 4' for 12-14 hr to fully extract insulin and proinsulin, then centrifuged to pellet insoluble material (10,000 x G, 15 min, 4'). The supernatents were transferred to 1.5 ml polypropylene microfuge tubes and frozen at -20'. Incorporation of 3H-leucine into insulin immunoreactive proteins and total islet proteins was 60 quantitated by Protein A-Sepharose (Pharmacia LKB, Piscataway, NJ) immunoprecipitation (Halban and Wollheim, 1980) and trichloroacetic acid (TCA) precipitation (Berne, 1975), respectively, followed by liquid scintillation counting. Measurement of PPImRNA Levels in Isolated Islets: Islets were incubated in the same manner as for the biosynthesis experiments described above except that there was ru>3H-leucine present during the incubation/labelling period. At the end of the 30 min incubation period, islets were washed twice with cold chase/stop buffer, then resuspended and lysed in 3.2 ml of GIT buffer containing 0.1% SIS. The lysates were drawn five times through a sterile 22-gauge (1 inch long) needle, layered onto a 1.2 ml CsCl cushion in a sterile 4.5 ml polyallomer ultracentrifuge tube (Beckman Instruments, Inc., Palo Alto, CA), and centrifuged for 16 hr at 36,000 rpm in a Beckman SW56 rotor. The supernatent was removed and the RNA pellet was resuspended in 0.5 ml of 95 % ethanol then transferred to a 1.5 ml microfuge tube. The polyallomer tube was washed a second time with 0.5 ml 95 % ethanol, which was added to the microtube containing the first wash. The RNA was precipitated by addition of 0.1 vol. of 2 M potassium acetate and storage at -20' for 8-10 hr. The precipitated RNA was recovered by centrifugation (10,000 x G, 30 min, 4'). The pellet was washed once with 0.5 ml 95 % ethanol, dried by vacuum centrifugation (Speed Vac, Savant Inc. , Farmingdale, 61 NY) then resuspended in 10 pl DEPC-treated MilliQ water. The yield of RNA from 250-300 cultured rat islets was not sufficient to permit quantitation by UV absorbance at 260/280 nm. Therefore, instead of analyzing known equal amounts of RNA by Northern analysis, equal volumes (4 pl) of the samples were analyzed, and the hybridization signals for PPImRNA were normalized to those for fi-actin mRNA. Ethidium bromide staining of RNA samples after electrophoresis indicated that approximately equal amounts of RNA were loaded in each lane. 2. Egggggg 9: 925 9n 99199999 Qispensed Rat Pancreatic 151.: 9.9.1.1.; Cell Culture: Experiments were conducted to examine the direct effects of CPH on cultured dispersed rat islet cells 1n 219;_. Rat pancreatic islets were isolated by collagenase digestion (Lacy and Kostianovsky, 1967) followed by separation on a discontinuous Ficoll density gradient (Sharp et al., 1975), and then dispersed into individual cells by treatment with Trypsin/EDTA (Weir et al., 1984). The cells were cultured in RPMI 1640 media supplemented with 10% fetal calf serum, 200 pg/ml penicillin and 200 mU/ml streptomycin (Ono et al. , 1979) . For all of the experiments to be described, cells were isolated and plated (50,000 cells per 2 ml culture medium in each 35 mm culture dish), then cultured at 37' and 5% CO2 for at least 24 hr prior to the start of experiments. 62 CPR Alteration of Cellular Insulin Content in Rat Islet Cells: After the initial 24 hr culture period to recover from isolation-induced stresses, medium was removed and replaced with fresh RPMI-1640 containing 0, 0.1, 1.0 or 10.0 pM CPH. CPH was added to the culture medium as loo-fold concentrated stock solutions in sterile water. Control dishes received the same volume (20 pl) of sterile distilled water. The islet cells were then cultured for 48 hr in the presence of the drug. At the end of this exposure interval, medium was collected, centrifuged (500 x G, 10 min) and frozen (-20°) until RIA. Cells were harvested and processed for the estimation of cellular immunoreactive insulin content and DNA content. Briefly, cells were gently scraped from the culture dishes with a teflon cell harvester (GIBCO, Grand Island, NY), washed and pelleted in cold phosphate-buffered saline (PBS), then resuspended in 1 ml cold PBS. Aliquots of 450 pl of this cell suspension were pelleted (150 x G, 5 min) and resuspended either in 450 pl of 1 N acetic acid or 450 pl of DNA assay buffer (2 M NaCl, 0.05 M Nazi-1P0“ 2 mM EDTA, pH 7.4) (LaBarca and Paigen, 1980). Cells that were resuspended in 1 N acetic acid were heated at 95' for 5 min, stored at 4' for 12-15 hr, centrifuged (15,000 x G, 10 min, 4') , then the supernatent was frozen (-20') until insulin RIA. Cells that were resuspended in DNA assay buffer were sonicated [3 x 10 sec at setting 4 .(microprobe)] then stored at 4' (8-24 hr) until assayed for DNA . 63 CPR Inhibition of Glucose-stimulated Insulin Secretion from Rat Islet Cells: After recovery and acclimation, cells were washed (3 x 2 ml) then preincubated for 30 min at 37° in a modified Kreb's Ringer bicarbonate buffer containing 10 mM HEPES, 5 mg/ml BSA (KRB/HEPES) and 50 mg/dl glucose. Preincubation in low glucose allowed the cells to reach basal levels of insulin release. They were then tested for insulin secretion by incubation for 30 min in KRB/HEPES containing 300 mg/dl glucose and 0, 0.1, 1.0 or 10.0 pM CPH. Insulin released into the KRB/HEPES was measured by RIA. At the end of the incubation, cells were processed for estimation of cellular IRI content (see above). Secretion was expressed as a percent of cellular IRI content at the end of the stimulation period. CPH Inhibition of Insulin Biosynthesis in Rat Islet Cells: Culture media was removed, and cells were preincubated for 30 min at 37' in RPMI 1640 containing 50 mg/dl glucose. The cells were then incubated for 120 min in a leucine-free RPMI 1640 culture medium containing 300 mg/dl glucose, 50 pCi 3H-leucine/ml, and CPH (o, 0.1, 1.0, 5.0 or 10.0 pM). After the incorporation period, cells were gently scraped from the culture dishes, washed (3 x 2 ml) with cold PBS, then resuspended in 200 pl 1 N acetic acid. The cells were heated at 95' for 5 min, stored at 4' for 8-15 hr, centrifuged (10,000 x G, 10 min, 4'), and the supernatents frozen (-20'). 64 Incorporation of’ iH-leucine into insulin immunoreactive protein and total cellular protein was quantitated by Protein A-Sepharose immunoprecipitation (Halban and Wollheim,1980) and TCA-precipitation (Berne, 1975) , respectively, followed by liquid scintillation counting. Measurement of CPH Effects on Conversion of Proinsulin to Insulin in Rat Islet Cells: Cells were prepared for 3H-leucine incorporation as described above for insulin biosynthesis experiments. After a 30 min preincubation in RPMI 1640 containing 50 mg/dl glucose, the cells were incubated for 30 min in leucine-free RPMI medium containing 200 mg/dl glucose and 50 pCi of 3H- leucine/ml, 219999; the addition of CPH. During this period the proinsulin that was labelled by incorporation of 3H- leucine did not have sufficient time to be converted to insulin (Steiner and Tager, 1979) . The 3H-leucine-containing medium was removed, the cells were gently washed twice with fresh complete RPMI 1640 medium (containing 50 mg/l L-leucine and 200 mg/dl d-glucose) without.§H-leucine. The cells were then allowed to incubate in complete RPMI 1640 (2 ml per 35 mm culture dish) for an additional 120 min to allow conversion.of the labelled proinsulin to labelled insulin. In selected dishes of cells, 10 pM CPH was added at the beginning of the 120 min chase/conversion period. The same volume of sterile distilled water (20 pl) was added.to control dishes during the 65 chase/conversion period. As a positive control, 10 pM monensin was added at the start of the chase/conversion period. Monensin has been shown by Gold et al. (1984) to inhibit proinsulin to insulin conversion in isolated rat islets. After the chase/conversion period the cells were gently washed (3 x 2 ml) with cold fresh complete RPMI 1640 media. Insulin and proinsulin were then extracted from the cells by the same procedure that was used for extraction of IRI for the islet cell biosynthesis experiments previously described. Aliquots (20 pl) of the acetic acid extracts were analyzed by high performance liquid chromatography (HPLC) to separate rat insulin and proinsulin as described by Halban et al. (1986). Ten pg each of unlabelled bovine insulin and proinsulin were co-injected with each sample to serve as carriers for the labelled proteins. The HPLC system consisted of two Waters Associates (Milford, MA) Model 6000A solvent delivery systems controlled by a Waters Model 660 solvent programmer, a Waters Model 06K injector, an ISCO (Lincoln, NB) Model V4 variable wavelength absorbance detector set at 260 nm, and an ISCO Retriever II fraction collector. The system used a Du Pont (Wilmington, DE) Zorbax ODS reverse-phase column (15.0 cm Length x 4.6 mm ID). The mobile phase consisted of a nonlinear gradient (#9 Waters ‘solvent programmer: initial to final conditions over a 20 min period) . of two buffers. Buffer A was 50 mM phosphoric acid, 20 mM triethylamine, 50 mM sodium perchlorate, pH 3.0. Buffer B was 66 90% acetonitrile, 10% distilled, deionized water (MilliQ water, Millipore Inc., Bedford MA). The gradient started at 64% buffer A, 36% buffer B and ended at 58% buffer A, 42% buffer B. This was followed by 20 min at 42% buffer B, then the column was washed at 70% buffer B for an additional 20 min before returning to initial conditions” The flow rate for the mobile phase was 1 ml/min. Fractions were collected at 1 min intervals into prosil 28-coated 12 x 75 mm borosilicate test tubes containing 100 pl of 0.5 mM sodium borate, 10‘mg/ml BSA, pH 9.3 (Halban et al., 1986). Each fraction was mixed and transferred to a 20 ml glass scintillation vial. Fifteen ml of Safety-Solve (Research Products Inc., Mount Prospect, IL) were added to the vials, then the radioactivity in the individual fractions was quantitated by liquid scintillation counting. HPLC analysis of the islet cell acetic acid extracts revealed several major peaks of radioactivity eluting before the column washing step. The first peak corresponded to the column void volume (not shown in the chromatograms presented in Figure 16). The second and third peaks were tentatively identified as rat insulins I and II based on the following criteria: 1) The labelled peaks had similar retention times to the two :major' peaks observed. by' UV absorbance at 260 nm for HPLC analysis of a purified rat insulin standard: 2) The labelled peaks were not present in acetic acid extracts of islet cells labelled for 30 min, but ‘werejpresent in extracts from islet cells labelled for’60 min, 67 or pulse-labelled for 30 min and chased for 120 min: 3) The peaks both demonstrated insulin immunoreactivity when they were lyophilized, resuspended and assayed by RIA: 4) Rat insulin II has a single methionine residue at position 8-29. Rat insulin I has no methionine residues, and has a lysine at position B-29 (Smith, 1966). The third but not the second peak.was present after labelling isolated rat islets with ”S- methionine: The fourth. and fifth. peaks ‘were ‘tentatively identified as rat proinsulins I and II for the following reasons: 1) The peaks were the major peaks seen in acetic acid extracts from islet cells labelled with 3H-leucine for 30 min: 2) The radioactivity incorporated into the fourth and fifth peaks could.be completely tranferred into the second.and.third peaks during the 120 min chase/conversion period. It is possible that the smaller fourth peak could be a conversion intermediate, rather than one of the rat proinsulins. Lack of available rat proinsulin I and II standards make conclusive identification of the fourth and fifth peaks difficult. 3. C d -T 5 Ce Cell Culture: HIT-T15 cells were initially maintained in culture according to the specifications of Santerre et al. (1980). The cells were plated (2 x:105 cells/dish) in 35 mm Falcon tissue culture dishes, and.grown.at 37' in.an atmosphere of 5% CO2 and 95% 02 in Ham's F-12 Nutrient Mixture supplemented with 68 15% dialyzed horse serum, 2.5% fetal bovine serum, 200 pg/ml penicillin and 200 mU/ml streptomycin. These culture conditions were used for the initial studies examining the effects of CPH in HIT-T15 cells (ie. inhibition of insulin synthesis and secretion and depletion of insulin content). However, Ham's F-12 Nutrient Mixture was originally formulated to permit rapid clonal growth of mammalian cells and is not well-suited to support cell populations in excess of 105/ml (Ham, 1964). Therefore, in later experiments (ie. CPH effects on insulin synthesis and insulin content, and measurement of PPImRNA levels), HIT-T15 cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum, 200 pg/ml penicillin and 200 pu/ml streptomycin. Most investigators studying HIT-T15 cells now use RPMI 1640 medium (Gold et al., 1988: Robertson, 1986: Meglasson et al., 1987: Ashcroft and Stubbs, 1987: Boyd et al., 1986) as it is better-suited for maintenance of higher densities of cells. In our laboratory, HIT-T15 cells were equally responsive in either media to the insulin-depleting effects of CPH. (Forty eight hr exposure of cells to 10 pM CPH caused approximately a 75% depletion of cellular insulin content in.RPMI 1640 or Ham's F-12.) In this thesis there are no instances where data were pooled from experiments performed in different media. HIT-T15 cells used for these experiments were from passage number 57-68. During this interval, cells remained responsive to glucose stimulation of insulin release. Later 69 passage numbers (>70) were unresponsive to glucose stimulation and were not used for these experiments. RINmSF cells were cultured as described by Gazdar et al. (1980). Cells were grown in RPMI 1640 medium supplemented with 10% fetal bovine sermm, 200 pg/ml penicillin and 200 mU/ml streptomycin. RINmSF cells were plated (2 x 10" cells/dish) in.35 mm Falcon tissue culture dishesc The RINmSF cells used for these experiments had undergone from 10 to 40 passages in our laboratoryu .Atjpassage number 40, the insulin secretory responses to glyceraldehyde, potassium and alanine were similar in magnitude to the responses observed at passage number 10. For both cell lines, stock cultures were passaged weekly, and received fresh media every 2 days. The cells did not reach confluence during the experimental period. CPH Alteration of Cellular Insulin Content in RINmSF and HIT- T15 Cells: Experiments were first conducted to determine whether CPH would deplete cellular insulin content in RINmSF and HIT-T15 cells. Cells were plated as described above, then cultured for 3 days prior to starting the experiments. After this acclimation period, the media were aspirated and replaced by test media containing 0, 0.1, 1.0 or 10.0 pM CPH. Cells were cultured in test media for 24 or 48 hr, then media was collected and cells processed for analysis of insulin. Culture media were removed, centrifuged (500 x G, 10 min) and 70 frozen until insulin RIA. Cells were harvested by gentle trypsinization [5 min at 37' in 1X Trypsin-EDTA (0.5 g Trypsin (1:250) and 0.2 g EDTA/l HBSS w/o Caz’ and Mg“: 313cc Inc. Grand Island, NY), 0.2 ml/dish], then washed 2 times and resuspended in 1 ml of cold Hank's balanced salt solution. Cells in 450 pl aliquots of this suspension were pelleted (150 x G, 5 min), and resuspended either in 450 pl of 1 N acetic acid or 450 pl of DNA assay buffer. Cells that were resuspended in 1 N acetic acid were heated for 5 min at 95°, sonicated [3 x 10 sec at setting 4 (microprobe)], stored at 4' for 12-15 hr to fully extract insulin and proinsulin, and centrifuged (10,000 x G, 10 min, 4'). The supernatent was frozen until insulin RIA. Cells resuspended in DNA assay buffer were sonicated then stored at 4' (12-24 hr) until assayed for DNA. Since CPH-induced depletion of rat pancreatic insulin content 1n 2129 is reversible upon removal of the drug (Rickert et al., 1975), it was determined whether the CPH- induced.depletion of cellular insulin.content seen.after 48 hr exposures of RINmSF and HIT-T15 cells was also reversible. For these experiments, cells were cultured in test media (0 or 10 pM CPH) for 48 hr (with media Change at 24 hr), then washed (3 x 2 ml) and cultured in CPH-free'media for an additional 48 hr'period. .At the end of the CPH exposure period and after 24 . and 48 hr of recovery, media were collected for insulin RIA, and cells were harvested as above for estimation of cellular 71 insulin and DNA. For'both cell lines, it.has.previously been reported.that there is passage to passage variability in cellular insulin content and secretion (Gazdar et al., 1980: Bathena et al., 1982: Ashcroft et al., 1986: Zhang et al.1989). In our laboratory, we also have observed some passage-dependent variability in these parameters. In order to normalize for this variability, results are expressed as percent of control for each experiment, where all dishes for a given experiment contained cells from the same passage. Over the course of these experiments, control RINmSF cells contained 15.4 i 2.6 ng insulin per pg DNA (mean i SEM, N=l3) and released 1,148 i 167 ng insulin per culture dish during the 48 hr exposure intervals. Control HIT-T15 cells contained 37.4 r 6.9 ng insulin per pg DNA (N=l9), and released 301 i 48 ng insulin per culture dish during the 48 hr exposure intervals. Comparative Experiments in RINmSF Cells With Cycloheximide and Actinomycin D: In order to provide information regarding the specificity of CPH effects, additional experiments were conducted to determine whether non-specific inhibitors of protein synthesis such as CHX and ACT-D would produce CPH-like effects in RINmSF cells. The specific objectives of these experiments was to examine whether these agents could cause depletion of cellular insulin content at concentrations that would not cause non- 72 specific cytotoxic changes. RINmSF cells were exposed for 24 hr to 0, 0.1, 1.0, or 10.0 pM CPH, 0.05, 0.5, or 5.0 pM CHX, or 0.05, 0.5, or 5.0 pg/ml ACT-D. After the exposure period, cells were harvested for estimation of cellular insulin content and DNA content per culture dish as described above. Comparative Experiments in RINmSF Cells With 4-DPMP and 2- DPMP: In order to provide additional information to support the contention that insulin depletion in RINmSF cells could serve as a model for examining the mechanism of CPH-induced depletion of pancreatic insulin content, experiments were conducted to determine the relative abilities of 4-DPMP and 2- DPMP to deplete cellular insulin content in RINmSF cells. For these experiments, cells were exposed for 24 hr to 0, 0.1, 1.0 or 10 pM concentrations of 4-DPMP or 2-DPMP. After the exposure period, cells were harvested for estimation of cellular insulin content and DNA content per culture dish as previously described. CPH Inhibition of Secretogogue-Stimulated Insulin Release From RINmSF and HIT-T15 Cells: The ability of CPH to directly inhibit the stimulated secretion of insulin from the cells was examined. Cells received fresh media on the third day after plating and were 73 used for secretion tests on the following day. Culture medium was aspirated and the cells were washed (3 x 2 ml) and preincubated for 30 min at 37' in.a KRB/HEPES (5 mg BSA/ml for RINmSF cells and 1 mg BSA/ml for HIT-T15 cells), and 2.8 mM glucose (RINmSF cells only). After the preincubation, cells were incubated for 45 or 60 min in 1 ml of KRB/HEPES containing CPH (0, 0.1, 1.0 or 10.0 pM) and selected secretogogues (HIT-T15 cells: 50 mg/ml D-glucose: RINmSF cells: 10 mM DL-alanine, 10 mM DL-glyceraldehyde, or 20 mM potassium Chloride) . Concentrations of secretogogues were selected based on preliminary concentration-response experiments. The test concentrations elicited 80-90% of maximum stimulated release. After the test incubations, the buffer was removed, centrifuged to remove any cells (500 x G, 10 min) and frozen (-20') until analyzed for insulin by RIA. CPH Inhibition of Insulin Biosynthesis in RINmSF and HIT-T15 Cells: Media from cells grown in culture for 3 days were aspirated and replaced with 1 ml of fresh leucine-free culture media containing 0, 0.1, 1.0, 5.0 or 10.0 pM CPH and 25 pCi/ml 3H-leucine (added in 25 pl .01% ethanol). After 2 hr of labelling under culture conditions, cells were washed (3 x 2 ml) with KRB/HEPES, then harvested to extract and measure labelled IRI in the same manner as was used for cellular IRI estimation above. The only difference in the procedure was 74 that the cell suspension was pelleted, then directly resuspended in 500 pl of 1 N acetic acid. The acetic acid extract was heated (95', 5 min), sonicated, then stored at 4' (12-16 hr). The next day the extract was centrifuged (10,000 x G, 30 min, 4'), and the supernatent frozen (-20°) until analysis for labeled insulin-immunoreactive proteins. Incorporation of 3H-leucine into insulin immunoreactive proteins and total cellular proteins was quantitated by Protein A-Sepharose immunoprecipitation (Halban and Wollheim, 1980) and TCA precipitation (Berne, 1975), respectively, followed by liquid scintillation counting. Measurement of Cytotoxicity in RINmSF and HIT-T15 Cells: To determine whether CPH effects were due to general cytotoxicity, cells were exposed to CPH under similar conditions as for the insulin depletion experiments, then harvested by gentle trypsinization and examined for their ability to exclude trypan blue (Tennant, 1964) . Additionally, since both. RINmSF and. HIT-T15 cells rapidly' divide and proliferate in culture, comparison of growth rates between treated and control cells served as another index of cell viability. Growth rates for the cells were estimated by measuring DNA content per culture dish (see above) at various time points following plating. 75 Determination of CPR Effects on PPImRNA Levels in RINm5F and HIT-T15 Cells: Experiments were conducted.to determine whether exposure of RINm5F and HIT-T15 cells to concentrations of CPH that inhibit insulin synthesis and deplete insulin content result in alterations of PPImRNA levels. For these experiments, HIT- T15 and RINm5F cells were plated and harvested in a similar manner as for the depletion and biosynthesis experiments already described. The only difference was that for these experiments the cells were plated in 100 mm diameter culture dishes at 1 x 105 cells per ml in 15 ml of culture medium. Insulin synthesis had been shown to be acutely and selectively inhibited by exposure of both cell lines to 5 and 10 pM CPH for 2 hr (Figure 20). In order to determine if this inhibition was associated.with alterations of PPImRNA levels, cells from the same passage were plated for measurement of insulin synthesis (35 mm dishes) and PPImRNA levels (100 mm dishes). The biosynthesis experiments were conducted as described above. Cells that were plated in the 100 mm dishes were incubated under the same conditions as for the biosynthesis experiments except that there was no 3H-leucine present, and the only CPH concentrations tested were 0 and 10 pM. At the end of the 2 hr incubation period, the medium was aspirated, and the cells were gently washed (3 x 10 ml) with cold phosphate buffered saline (PBS). After the final wash, the dishes were momentarily inverted at a slight angle, and 76 the lip of the culture dish was blotted to completely drain the PBS. The cells were then lysed by addition of 3.2 ml of GIT buffer containing 0.1% sodium lauroyl sulfate. The lysate was drawn five times through.a sterile 1 inch, 22-gauge needle (each time being returned. to the culture dish to ‘wash remaining material free from the surface) , then transferred to a 15 ml conical polypropylene centrifuge tube (Corning Co., Corning, NY). The lysate was then cleared by centrifugation (10 min at 10,000 x G and 4'). The supernatent was layered over a 1.2 ml cushion of 5.7 M CsCl, 0.25 mM sodium citrate (pH 7.4) in a 4.5 ml polyallomer ultracentrifuge tube (Beckman Instruments, Palo Alto, CA). The tubes were balanced (by addition of GIT buffer), loaded into a Beckman SW56 swinging bucket rotor, and centrifuged at 36,000 rpm for 16 hr at 5- 10' . The supernatent was aspirated, and the pelleted material (mostly RNA) was resuspended in 0.5 ml of 8 M guanidine hydrochloride, 0.25 mM sodium Citrate (pH 7.4). The resuspended RNA was transferred to a 1.5 ml polypropylene tube and precipitated by addition of 0.05 vol. of 1N acetic acid followed by 2 vol. of 100% ethanol, and subsequent storage at -20' for 2-4 hr. The RNA was pelleted by centrifugation for 30 min at 10,000 x G, at 4'. The supernatent was aspirated, and the pellet was washed with 1 ml 95 % ethanol, then dried either under nitrogen or by vacuum.centrifugation. The dried pellet was dissolved in 100 pl of sterile, DEPC-treated MilliQ water. This solution was centrifuged (10,000 x G, 10 min, 4') 77 to pellet insoluble material, and the supernatent transferred to a Clean, sterile 1.5 ml polypropylene microfuge tube which was kept on ice. Two additional water extractions were performed on the pelleted material, with the supernatents being pooled in the clean microfuge tube. The RNA in the pooled supernatents was precipitated by addition of 0.1 vol. of 2 M potassium acetate and 2 vol. 100 % ethanol and storage at -20' for 10-12 hr. The precipitated RNA was pelleted by centrifugation (10,000 x G, 30 min, 4'), washed with 95 % ethanol, dried, and resuspended in 20 pl of DEPC-treated MilliQ water. The concentration and yield of RNA was determined by measurement of UV absorbance at 260 nm on 2 or 5 pl aliquots from the 20 pl of RNA solution diluted to 1 ml with 0.1 M Tris-HCl, 0.01 M EDTA, pH 3.0. Typically, 100-200 pg of cellular total RNA was isolated from each 100 mm culture dish of RINm5F or HIT-T15 cells at approximately 75% confluence. The purity of the RNA preparations was determined by calculation of the ratio of UV absorbance at 260 and 280 nm. Values for this ratio were routinely between 2.0 and 2.2. PPImRNA levels were analyzed by Northern analysis and scanning densitometry as described for analysis of PPImRNA and p-actin mRNA levels in RNA samples from the whole pancreas. RNA samples from control and CPH-treated RINm5F cells were also analyzed for glyceraldehyde phosphate dehydrogenase mRNA (GAPDH mRNA) levels. For this analysis, Northern blots were hybridized with a [am-labelled cDNA probe for GAPDH mRNA. 78 This probe was produced by random hexanucleotide primer extension labelling (Feinberg and Vogelstein, 1983) of a cDNA fragment of rat GAPDH mRNA (kindly provided by Dr. R. Schwartz, Dept. Microbiology, MSU). Exposure of RINm5F and HIT-T15 cells for 24 hr to 10 pM CPH resulted in 75 and 50% depletion, respectively, of cellular IRI levels. In experiments to determine whether CPH- induced insulin depletion was associated with alterations of PPImRNA levels, cells were plated either in 100 mm culture dishes for the measurement of PPImRNA levels, or in 35 mm culture dishes for confirmation of CPH-induced insulin depletion. Cells that were plated in 35 mm dishes were exposed to 0, 0.1, 1.0, or 10.0 pM CPH for 24 hr then harvested for estimation of cellular IRI and DNA as previously described. Cells that were plated in 100 mm culture dishes were exposed for 24 hr to the same CPH concentrations, then harvested for estimation of PPImRNA levels as described above for the 2 hr CPH exposures. PPImRNA levels were determined by Northern analysis and scanning densitometry. Statistical Analysis: Data were analyzed by Student's t-Test or randomized block design analysis of variance (Steel and Torrie, 1960). Dunnett's procedure was used to evaluate statistical significance of differences between means when comparisons involved multiple means and a common control mean. 79 Alternatively, Duncan's multiple range test was applied to test differences between treatment means. Statistical significance was evaluated at the p<.05 level. BISULTB V v ct the Rat The time course of CPH-induced alterations of insulin, proinsulin, and PPImRNA.were carefully evaluated over a 24 hr period following administration of a single, insulin-depleting dose of CPH. For this experiment, control and CPH-treated rats were food-deprived at the time of dosing. Both groups of animals lost approximately 10% of their body weights during the 24 hr experimental period (Table 1). CPH produced no additional weight loss over that which was produced by feed withdrawal for 24 hr. Effects of a Single Dose of CPR on Rat Pancreatic Insulin: The time course of CPH-induced.depletion of rat pancreatic IRI is shown in Figure 6. CPH administration caused a loss of pancreatic IRI, reaching a 50% decrease at 24 hr after dosing. Results in control animals showed that fasting alone did not produce alterations of pancreatic IRI content during the 24 hr period. Effects of a Single Dose of CPR on Rat Pancreatic Proinsulin: Extracts of the whole pancreas were subjected to gel filtration.chromatography to separate and quantify proinsulin in control and CPH-treated animals. Representative 80 81 chromatograms from selected samples are shown in Figure 7. Figure 7A is a Chromatogram showing immunodetectable proinsulin and insulin in pancreatic extracts from the zero time control group. The expanded scale clearly exhibits the proinsulin peak. Chromatograms of pancreatic extracts from CPH-treated animals at 1.5 and 3 hr after dosing (Fig. 7B and 7C, respectively) indicate a decrease of the proinsulin peak at both time points. Figure 8 shows the time course of proinsulin decline after CPH treatment. Proinsulin levels in the pancreas are reduced by 50% at 1.5 hr and are decreased by 65% at 3 hr after drug administration. Proinsulin levels remain depressed for the duration of the 24 hr period. Effects of a Single Dose of CPR on Rat Pancreatic Total RNA, PPImRNA and fi-actin mRNA: Neither CPH treatment nor fasting produced any detectable alterations of pancreatic total RNA at any of the time points examined (Table 1). Figure 9 shows an autoradiogram.of a representative Northern blot of pancreatic total RNA hybridized with insulin and actin cRNA probes. Figure 9 provides direct evidence that at 10 hr after dosing, PPImRNA.levels in.CPH-treated.animals were lower’than those in control animals. Also shown are hybridization signals for 8- actin mRNA, indicating that levels of this mRNA are not different in the control and treated animals at 10 hr after drug administration. The time course of CPH effects on rat pancreatic PPImRNA levels is shown in Figure 10. CPH produced a rapid decrease 82 in PPImRNA as levels appeared to be declining at 1.5 and 3 hr after dosing, and a statistically significant decrease was observed at 6 hr. At 6 and 10 hr after CPH administration, PPImRNA levels in the treated animals were 35 and 30%, respectively, of the control values for each time point. PPImRNA returned towards control levels at 24 hr after dosing. Fasting alone also appeared to produce a decrease of PPImRNA levels. At 10 and 24 hr following the removal of feed, PPImRNA levels in the control rats were approximately 70% of the zero time control value. To assess whether the CPH-induced effects were specific for PPImRNA, samples were also analyzed for levels of fi-actin mRNA. As shown in Figure 11, mean levels of fi-actin mRNA in control and CPH-treated animals were not different. The mean levels of B-actin mRNA appear to decrease over the course of the experiment, but these decreases do not reach statistical significance. Table 1. and. pancreatic ‘total RNA. 83 Effects of a single dose of CPH on rat body weight Groups of four animals were sacrificed at the indicated time points after administration of CPH (45 mg/Kg, po.) or vehicle. Data are presented.as mean 2 SEM. Treatment Time Body Weight Pancreatic Total RNA (Hr) (g) tug/m9)° CON 0 222.3 i 3.2 10.4 i 1.9 CON 1.5 227.5 i 2.3 8.0 i 1.0 CPH 1.5 223.0 i 2.8 8.5 i 0.2 CON 3 221.8 i 2.4 8.1 r 0.5 CPH 3 220.0 1 3.6 9.6 i 0.4 CON 6 212.3 i 8.5 9.3 i 0.7 CPH 6 211.0 2 5.6 10.6 i 0.7 CON 10 209.3 2 4.8 9.2 i 0.7 CPH 10 218.0 i 5.6 10.1 t 0.4 CON 24 209.3 1 3.8 10.8 i 0.2 CPH 24 202.8 i 6.8 10.0 i 0.6 a Data expressed as pg total RNA per mg pancreas. 84 ng IRI Per mg Pancreas 200 150- 100- 3,0 50- 0—0 CON "'4 0—0 CPH 0 - - - . - - - . 0 12 24 HOURS AFTER TREATMENT Figure 6. Time course of CPH effects on rat pancreatic insulin content. Data are expressed as ng immunoreactive insulin (IRI) per mg of pancreas (mean 1 SEM). "a" denotes values significantly different from controls for each time point (p<.05) . "b" denotes values significantly different from the zero time control. 85 an , camao.0nn I‘ E in» «sun E I“: g namemuu g 50‘ \ \ 0 . ~1( ‘J . .1 0 10 a: 30 «1 50 mm cHthHn ‘ [3 z . 5 I“) ’ msxm up- 2 z D s so- . Humans J) 0 . -. 0 10 a: «i so an ontsnn ’ (3 z Iflh E x 1m: 0 Ill 0 k E 2 501 . "WU o v A. ' ##— 0 10 a: 30 «I 50 Imunmuuuutn Figure 7. Gel filtration chromatography (Bio Gel P30) of acetic acid pancreatic extracts from control and CPH-treated rats. .A.) Zero time control. B.) CPH-treated, 1.5 hr after dosing. C.) CPH-treated, 3 hr after dosing. Each fraction was assayed for insulin immunoreactive material by RIA. 86 D 8 E 10 °7 - ~o g \ / a) O E 3 o—o con 0. -—- CPH .5 54 3 U) .5 8 O. U. C O . . . . . . T . 0 12 24 HOURS AFTER TREATMENT Figure 8. Time course of CPH effects on rat pancreatic proinsulin. Values were determined by RIA (rat insulin standard) of the proinsulin peaks from gel filtration separation of the pancreatic extracts for each treatment group. Each point represents the mean value from a pooled sample from four rats as described in materials and methods. 87 con/10 CPH/10 1 2 3 4 s a ’7 (3 Actin- ,, is: U «.4 «i i PWDOOOeeee Figure 9. Northern analysis of rat pancreatic PPImRNA and p- actin mRNA from control and CPH-treated rats 10 hr after dosing. Each lane represents RNA samples obtained from individual animals. Lanes 1-4: Control rats 10 hr after vehicle treatment; Lanes 5-8: CPH-treated rats 10 hr after dosing. 150- 3 E 125-. 2:28.95. 5 1001 / O .5 75-\ £- 3 a E 50- T" 5 b i 25- N-4 " " 0- a,h 0 . - - . . . - ' 0 12 24 HOURS AFI'ER TREATMENT Figure 10. Time course of CPH effects on rat panCreatic PPImRNA levels. Data were calculated as integrated peak areas from densitometric scanning of autoradiograms of dot blots, and are expressed as a percent of the value for the zero time control group (mean 1 SEM) . "a" denotes values significantly different from the zero-time controls. "b" denotes values significantly different from controls at each time point (p<.05) . 89 150- O 2 f 125- % .. o 1CN34 (J ”a E 75" /l f 1 s 3 50- :1 § O—OCON o 25' N-4 0—0 CPH n. 0 - - . . - . , 0 12 24 HOURS AFTER TREATMENT Figure 11. Time course of CPH effects on rat pancreatic B- actin mRNA levels. Data were calculated and expressed as for PPImRNA levels in the legend to figure 10. 9O Acute In V1990 ngects of QPE in Isolated Rat Islets Effects of CPH on PPImRNA and Proinsulin Synthesis: Experiments were conducted to measure proinsulin biosynthesis and PPImRNA levels in rat islets exposed 1n 21929 for 30 min to 0 or 10 pM CPH. Figure 12 shows a Northern blot of RNA samples prepared from these islets. The outer pairs of lanes on both sides of the autoradiogram are RNA samples from whole rat pancreas and.RINm5F cells. 'These were included in the gel as control samples to ensure that the transfer and hybridization steps of the analysis were satisfactory. Exposure of rat islets to CPH for 30 min produces n0’ alteration in the levels of PPImRNA or B-actin mRNA. This autoradiogram represents data generated from a single experiment using three groups of 250 control islets and three groups of 250 CPH-treated islets. Similar results were obtained from additional separate experiments to test the reproducibility of these findings (data not shown). The blot shown in Figure 12 was exposed to x-ray film for variable time intervals so that hybridization signals that were suitable for quantitation.by scanning densitometry could be obtained for PPImRNA and fl-actin mRNA. These results are presented in Figure 13A. The peak areas for PPImRNA were normalized to those for B-actin mRNA. In agreement with the autoradiogram shown in Figure 12, the data in Figure 13A indicate that CPH exposure (10 pM, 30 min) produces no decrease of islet PPImRNA levels. 91 The data presented in Figure 138 show that incorporation of 3H-leuCine into proinsulin is inhibited by approximately 50% during the 30 min CPH exposure. Figure 13C demonstrates that the incorporation of 3H-leucine into islet proteins other than proinsulin (cpm incorporated into TCA-precipitable material minus cpm incorporated into proinsulin) was not inhibited.during the 30 min CPH exposure. 'Taken together, the data presented in Figure 13(A-C) indicate that CPH acutely and selectively inhibits the synthesis of proinsulin in isolated rat islets 2199999 decreasing the levels of PPImRNA. 92 % Ratlslets til; 0 OE z: 2:: 2:: 55 on. OIL 00. 5E Ira. oo 00 00 0.0: 12 34 56 78 910 Actin— PPI> Figure 12. Northern analysis of PPImRNA and fi-actin mRNA levels in isolated rat islets after 30 min exposure to 0 or 10 pM CPH 1n 21929. RNA samples were prepared from groups of 250 rat pancreatic islets. Lanes 1 and 10: 5 pg total RNA from RINm5F cells; Lanes 2 and 9: 10 pg total RNA from rat pancreas: Lanes 3, 5, and 7: control islet total RNA; Lanes 4, 6, and 8: total RNA isolated from CPH-treated islets. 93 Figure 13. Effects of 0 or 10 pM CPH on PPImRNA, proinsulin synthesis, and non-proinsulin protein synthesis in cultured rat pancreatic islets exposed for 30 min _i_n vitro. A.) PPImRNA levels. Data were calculated as integrated peak areas from densitometric scanning of the autoradiogram shown in Figure 15. Results are expressed as the integrated areas of PPImRNA (normalized to fi-actin mRNA), and are expressed as a percent of the mean control value. Each bar represents the mean (1 SEM) from.three groups of 250 islets. B.) Proinsulin biosynthesis. Data were calculated as cpm of 3H-leucine incorporated into insulin immunoprecipitable material. C.) Synthesis of islet proteins other than proinsulin. Data were calculated as cpm of 3H-leucine incorporated into TCA- precipitable material minus 3H-leucine incorporation into proinsulin. Asterisks denote values significantly different from controls (p<.05). 94 _ w i O .... ... .... a. el 1. o 0 25.... stone 3 ......ng sees. CPH Conc. (IN) 10 o ‘ 5 0 9.... 8N .2. “.63 E... .819 i i 4i 2 0 0 IO CPH Cone. ON) 0 CPI-(Cone. 0N) 95 In‘v1999 3119999 9: 9:: 1n gnltnred Dispersed Rat Islet Cells One important objective of this project was to Characterize the ability of CPH to deplete cellular insulin content 1n 219;_. Experiments were conducted to examine the ability of CPH to inhibit insulin synthesis and secretion and alter cellular insulin content in cultured dispersed rat islet cells. Effects of CPR on Insulin Secretion and Synthesis: Studies were initially performed to determine the short-term inhibitory effects of CPH on glucose-stimulated insulin secretion and synthesis in cultured dispersed islet cells. Figure 14A depicts the ability of glucose to stimulate the secretion of insulin from the cells. Increasing the glucose concentration from 50 to 300 mg/dl resulted in a 5-fold increase of insulin secretion. Figure 148 demonstrates that CRH inhibits glucose-stimulated insulin secretion from the cells in a concentration-dependent manner. At 0.1 pM CPH, glucose-stimulated insulin release was inhibited by 33%, while at 1 and 10 pM CPH, release was inhibited by 90%. Figure 15A illustrates the ability of glucose to stimulate insulin biosynthesis in cultured dispersed islet cells. Increasing the glucose concentration from 50 to 300 mg/dl produced a 5-fold increase in the incorporation of HP- leucine into insulin and proinsulin (designated (pro) insulin) . Figure ISijrovides evidence that CPH acutely and selectively inhibits the synthesis of (pro) insulin in the cells in a 96 concentration-dependent manner. (Pro) insulin biosynthesis was not affected by exposure to 0.1 or 1.0 pM CPH, but was inhibited by greater than 75% at 5 and 10 pM CPH. Synthesis of non-insulin proteins also appeared to be slightly inhibited by exposure to 10 pM CPH, but this effect did not reach levels that were statistically different from controls. Effects of CPH on Proinsulin to Insulin Conversion: In order to Clarify the potential role that inhibition of proinsulin to insulin conversion might play in CPH-induced insulin depletion, experiments were conducted in the dispersed islet cells to examine directly the ability of CPH to inhibit the conversion process. For these experiments, islet cells were pulse labelled for 30 min with 3H-1eucine to allow incorporation of the label into proinsulin, then chased for 120 min in the presence or absence of 10 pM CPH. Acetic acid extracts were prepared from the cells and analyzed by HPLC to measure labelled insulin and proinsulin separately. Figure 16A shows the HPLC Chromatogram for islet cells labelled for 30 min with 3H-leucine. A single major peak is shown that represents radiolabelled proinsulin. Labelling of islet cells for 30 min followed by a 120 min chase period without drug treatment results in conversion of the radiolabelled proinsulin to radiolabelled insulin (Figure 168). Labelling of islet cells for 30 min followed by a 120 min chase period in the presence of 10 pM CPH, also results in the complete conversion of the radiolabelled proinsulin to 97 form radiolabelled insulin (Figure 16C). As a positive control, the 120 min chase period was performed in the presence of 10 pM monensin, a known inhibitor of proinsulin to insulin conversion (Gold et al., 1984) . The Chromatogram shown in Figure 160 shows that under the conditions of these experiments, monensin inhibited the conversion of proinsulin to insulin. After the 120 min chase period, there was still a substantial portion of radiolabelled proinsulin that had not been converted to insulin. These results indicate that under the conditions of these experiments, CPH does not inhibit the conversion of proinsulin to insulin. Effects of CPH on Cellular Insulin Content: The ability of CPH to alter cellular insulin content of the dispersed islet cells 1n 21999 is shown in Figure 17. Exposure of cells for 48 hr to 1.0 pM CPH elevated cellular insulin content to 160% of control, while exposure to 10 pM CPH depleted cellular insulin content to 50% of control. In separate experiments, 48 hr exposure to 10 pM CPH caused no decrease in islet cell viability as measured by trypan blue dye exclusion assay (91 :l: 1 (N=3) vs. 86 t 2 (N=4) percent viable cells in dishes exposed to 0 or 10 pM CPH, respectively). Media insulin levels were measured from the cells that were exposed to CPH for 48 hr. Results shown in Figure 18 demonstrate that media insulin levels were decreased in a concentration-dependent manner by exposure to CPH. 98 Experiments were conducted to evaluate whether CPH- induced decreases in cellular insulin levels were reversible upon termination of exposure. Islet cells were exposed to 0 or 10 pM CPH then cultured for an additional 48 hr in the absence of the drug; In these experiments, insulin content of CPH-treated cells was decreased to 48% of control after the initial 48 hr exposure (7.3 i 0.5 vs. 3.6 r 0.3 ng IRI/ng DNA (N=4) in control and CPH-treated cells, respectively), and returned to 72% of control after 48 hr of recovery (6.5 r 0.5 vs. 4.7 i 0.3 ng IRI/ng DNA (N=4) in control and CPH-treated cells, respectively). 99 .s' 20~ g 2 5 E) . FL .. ' "\7 'E ‘ ‘ .5 a 15: E 2-0' \t i .3 ‘3 ‘D 1 5~ 'o O: ' .9, 10“ +1 8 5 g E 1.o~ g 54 I .. 0, 0 0.5- C E a: t 3 O 9 o a 0.0 Ly --..-. - "--., - mm. 50 300 ‘1 0 0.1 . -1.0 10.0 Glucose Conc. (mg/dL) CPH Conc. (I‘M) Figure 14. Effects of CPH on glucose-stimulated insulin release from cultured rat pancreatic islet cells. A.) Glucose stimulation of insulin release. Data were calculated as ng insulin released per 50,000 cells per 30 min. Each bar represent the mean (i SEM) of three observations. B.) CPH inhibition of glucose-stimulated insulin release. Results were calculated as ng of IRI released per 30 minutes per 100 ng of cellular IRI. Each point represents the mean value (1 SEM) of 4 separate experiments (2-4 culture dishes per treatment per experiment). Asterisks denote significant differences from controls (p<.05). GPAlS—Precipitoble cpm (x103) Figure 15. Effects of CPH on glucose-stimulated insulin synthesis in cultured rat pancreatic islet cells. A.) Glucose-stimulation of insulin synthesis. Results were calculated as cpm of 3H-leucine incorporated into immunoprecipitable proteins (guinea pig anti-porcine insulin serum, GPAIS) per 50,000 cells per 120 min. B.) CPH inhibition of glucose-stimulated (300 mg/dl) insulin and non- insulin protein synthesis. Results were calculated as above, and were expressed as percents of control (labelled in the absence of CPH) for each experiment. For these experiments, control cells incorporated 3684 t 773 cpm per 50,000 cells per 2 hr into GPAIS-precipitable protein, and 2606 i 742 cpm into non-insulin proteins. Each point represents the mean value (3: SEM) of 3-5 separate experiments. Asterisks denote significant differences from controls (p<.05). . 0 ‘ B E 20‘ 125 N 3 100 <9 L. 0 ‘- - a: 'E T I 6 D. . o C C .L g 15 0 1 1 \l o “o" 75" o 4, l o 10. c 8. 13 so. 8 o (g o—o ITRI c 5. / O—O otol Protein . m 25- " 0. (p (-|Rl) 9 . \. O I y f OJ l ' ' ' ' 'YVVI ' ' ‘ 'fi*‘*I O 100 200 300 0.1 1.0 10.0 Glucose Conc. (mg/dL) CPH Conc. (pM) 101 Figure 16. Conversion of proinsulin to insulin in cultured rat pancreatic islet cells exposed to CPH or monensin in yitgg. HPLC chromatograms of acetic acid extracts of islet cells pulse-labelled with z‘H—leucine for 30 min then chased in the presence of excess leucine for an additional 120 min. A.) Islet cells labelled with 3H-leucine for 30 min in the absence of any experimental treatment (no chase); B.) Islet cells labelled for 30 min then chased for 120 min; 0.) Islet cells labelled for 30 min then chased in the presence of 10 pH CPH (CPH only during chase); D.) Islet cells labelled for 30 min then chased for 120 min in the presence of 10 0M monensin (monensin only during chase). 102 I” can. no. sou-Luann m no. ao- “- um . fi W V I V V j V 8 to Is to as so as 40 a mun-Mint) in m "In. ”mutant-Lean: II- N‘ m 4004 ”a a V V V V I I I 1 I 10 u an as so a do 0: mama-L) I“ ”mu-mo 1m. tummanoumo ”I“ m com «um “i o . f' f . 'A' . I 10 i! Q 28 a 8 N GO manna-L) 1” ”at” an. 1U“M(§10flm m '0‘ m am no- ”1 ‘ V ’—f Y— 1 I f V 1 I "I ‘8 an a an as Q 43 103 5.0- i o l < 40‘ T E o 8‘ :5.0-(Ty///1 E3 1 CL g 2.0~ m o C * 1.C)1 Dow/Hm". . 0 0.1 1.0 10.0 CPH Conc. (pM) Figure 17. Cellular insulin content of rat pancreatic islet cells after culture for 48 hours in the presence of CPH. Values were calculated as ng of cellular IRI per ng of DNA. Each point represents the mean value (1 SEM) of 3 separate experiments (2-4 culture dishes per treatment per experiment) . Asterisks denote significant differences from controls (p<.05). 104 2000 J3) .é 1\/ 2? t 1500- Q) 0. it ‘0 T (D 8 1000- (f 0) '6 O: “ $ g 500- o 0‘ .L C \ 1: o Own/A mum, . “.-.... . ”...", o 0.1 1.0 10.0 CPH Conc. (,uM) Figure 18. Insulin concentration in media from rat pancreatic islet cells after culture for'48 hours in the presence of CPR. Values were calculated as total ng released per 50,000 cells per 48 hours. Each point represents the mean value (i SEM) of 4 separate experiments (2-4 culture dishes per treatment per experiment). Asterisks denote significant differences from controls (p<.05). 105 V t ct 0 SF and n -T15 Cel 8 Effects of CPR on Insulin secretion: The ability of CPH to directly inhibit secretogogue-stimulated insulin release from RINm5F and HIT-T15 cells is shown in Figure 19 (A-D). Inhibition of glucose-stimulated insulin release from.HIT-T15 cells is shown in Figure 19A. CPH inhibited glucose (50 mg/dl) stimulated insulin release from HIT-T15 cells in a concentration-dependent manner with 1 and 10 uM CPH producing significant inhibition. Basal release of insulin (no glucose present) was not affected by CPH exposure (data not shown). The ability of CPH to inhibit insulin release from RINm5F cells when stimulated by 10 mM DL—glyceraldehyde, 10 mM DL- alanine or 20 mM potassium chloride is shown in Figure 19 (B- D) . The CPH concentration-responses for inhibition of release varied somewhat with the secretogogue used. Generally, significant inhibition of release occurred at CPH concentrations between 1.0 and 10.0 pM. As for HIT-T15 cells, basal release from RINm5F cells (no stimulus present) was not altered by CPH (data not shown). Effects of CPR on Insulin Synthesis: Experiments were conducted to determine the ability of CPH to inhibit insulin biosynthesis in RINm5F and HIT-T15 cells during a 2 hr exposure. Results shown in Figure 20 illustrate that CPH selectively inhibited insulin synthesis in both cell lines. In RINm5F cells (Fig. 208), synthesis of insulin- immunoreactive‘ proteins represented about 10% of total 106 cellular protein synthesis, while in HIT-T15 cells (Fig. 20A) , this value was approximately 13%. In both cell lines, incorporation of 3H-leucine into non-insulin immunoreactive proteins was decreased with increasing CPH concentrations, but incorporation into insulin-immunoreactive proteins was inhibited to a greater extent. In RINm5F cells, exposure for 2 hr to 10 an CPH resulted in a 15% inhibition of non-insulin protein synthesis and a 55% decrease in insulin synthesis. In HIT-T15 cells, 10 uM produced a 30% decrease of non-insulin protein synthesis and a 70% decrease of insulin synthesis. Effects of CPR on Cellular Insulin Content: The ability of CPH to alter the cellular insulin content of RINm5F and HIT- T15 cells is shown in Figure 21. Culture of cells for 48 hr with the indicated concentrations of CPH depleted cellular insulin content in both cell lines. This depletion was concentration-dependent, and reached approximately 30% of controls using 10 MM CPH. Exposure for 24 hr to 10 pH CPH also caused depletion of cellular insulin content. HIT-T15 cells were depleted to 51% of control [340 t 7 (N=4) vs. 175 i 18 (N=4) ng IRI/ng DNA in control and CPH-treated cells, respectively], while RINm5F cells were depleted to 25% of control [175 i 11 (N83) vs. 44 t 2 (N=3) ng IRI/ng DNA in control and CPH-treated cells, respectively]. Effects of CPR on.nedie Insulin Levels: Media insulin levels .after 48 hr of exposure of cells to CPH are shown in Figure 22. CPH reduced media insulin levels during culture of both 107 cell lines. The reduction was also concentration dependent, and maximum using 10 MI CPH. Both cell lines appeared equally sensitive to reductions in insulin released into media. Reversibility of CPE Effects: Experiments were conducted to investigate whether the depletion of cellular insulin content caused by CPH was reversible. Results shown in Figure 23 demonstrate that during a 48 hr recovery period following exposure for 48 hr to 10 uM CPH, cellular insulin content in both RINm5F and HIT-T15 cells returned to control levels. Media levels of insulin during recovery of cells are shown in Figure 24. Media insulin from HIT-T15 cells recovered to a greater extent than from RINm5F cells. Media insulin from CPH-treated RINm5F cells was still less than control at the end of the 48 hr recovery period. Evaluation of Potential Cytotoxic Effects: In order to determine if the CPH-induced effects were associated with cytotoxicity, experiments were conducted to characterize the effects of CPH on cell viability and division. Cells exposed to 10 uM CPH showed no reduction in the ability to exclude trypan blue relative to control cells after 24 and 48 hr of exposure (Table 2). Effects of CPH on cell division were estimated by comparison of DNA content per culture dish after 48 hr of exposure to 0 or 10.0 uM CPH. Dishes containing cells exposed to 10.0 uM CPH show no reduction of DNA content ,after 48 hr of exposure (Fig. 25). Dishes containing cells exposed to 10.0 uM CPH for 48 hr then cultured for 24 or 48 hr 108 in CPH-free media (Fig. 25) also showed no reduction of DNA content relative to control. 'These results suggest that.10 uM CPH does not alter viability or inhibit cell division for either cell line. Effects of CPR on PPImRNA Levels. Comparison with Inhibition of Insulin Synthesis: In order to determine whether inhibition of insulin synthesis in the cell lines after a short exposure period was independent of effects on PPImRNA levels, experiments were conducted to examine the effects of a 2 hr CPH exposure on PPImRNA levels in RINm5F and HIT-T15 cells. Figure 26 shows a Northern analysis of RNA samples prepared from RINm5F and HIT-T15 cells exposed to 10 pH CPH. Treatment of both cell lines for 2 hr with 10 uM CPH produced no decrease of PPImRNA levels. The data presented in Figures 20 and 26 indicate that insulin synthesis in RINm5F and HIT- T15 cells can be inhibited by CPH without alterations of PPImRNA levels. Effects of CPR on PPImRNA Levels. Comparison with Insulin Depletion: PPImRNA levels in the cells were measured after exposures to insulin-depleting concentrations of CPH. Figure 27 shows two representative autoradiograms from Northern blots of RNA samples prepared from HIT-T15 cells (Fig. 27A) and RINm5F cells (Fig. 278) exposed to 10 uM CPH for 24 hr. In both cell lines, PPImRNA and p-actin mRNA levels were unaltered by exposure to 10 uM CPH. Results of scanning densitometry of these autoradiograms and others from separate 109 experiments are presented in Table 3. CPH-treatment appears to cause a slight decrease in the mean levels of PPImRNA (normalized to fi-actin mRNA) in both cell lines. However, these decreases do not reach statistical significance. In order to establish that the lack of CPH effect did not represent an inability to detect alterations of PPImRNA levels in the cells, a positive control experiment was performed in which RINm5F cells were exposed to 22mM sodium.butyrate for 24 hr. This treatment has been reported by Philippe et al. (1988) to cause an elevation of PPImRNA levels in RINm5F cells. The autoradiograms shown in Figure 28 demonstrate that while CPH produces no alteration of PPImRNA levels (Fig. 28A) , exposure to 2 mM sodium butyrate increased levels of the mRNA (Fig. 288). This result supports the contention that the lack of CPH effect does not represent an inability to detect alterations of PPImRNA levels in the cells. Figure 28C demonstrates that levels of a control mRNA, glyceraldehyde phosphate dehydrogenase mRNA (GAPDH mRNA), are unaffected by 24 hr CPH exposure in RINm5F cells. Comparison of CPR Effects with Known Protein Synthesis Inhibitors: Experiments were conducted to determine the ability of CHX and ACT-D to deplete cellular insulin content in RINm5F cells after 24 hr exposures. The results of these studies are presented in figure 29 (A-C). Increasing concentrations of these non-specific inhibitors of protein synthesis were found to inhibit cell growth and division, as 110 shown by decreases of DNA content per culture dish (Fig. 29A and 29B) . However, these agents produced no depletion of cellular insulin content, even at concentrations that inhibited cell growth and division. This is in contrast to results produced by 24 hr CPH exposure (Fig. 29C) . Increasing concentrations of CPH cause depletion of cellular insulin content, without inhibiting cell growth and division. These data indicate that in RINm5F cells, the insulin-depleting effects of CPH can not be mimicked by exposure to non-specific inhibitors of protein synthesis. Comparison of CPR Effects with CPH Analogs: Experiments were also conducted to determine the relative potencies of CPH, 4- DPMP and 2-DPMP to deplete pancreatic insulin content in RINm5F cells. Shown in Figure 30 is a summary of these experiments. Exposure of the cells for 24 hr to 1 uM 4-DPMP leads to approximately a 50% depletion of cellular insulin content. Exposure to 10 uM 4-DPMP or CPH led to an 85% depletionn lNone of the concentrations of 2-DPMP used in these experiments produced depletion of cellular insulin levels. None of the agents tested produced significant alterations in DNA content per culture dish over the experimental period (data not shown). 111 Figure 19. Effects of CPH on stimulated insulin release from HIT-T15 and RINm5F cells. Glucose alone produced a 2.5-fold stimulation. of insulin release from IHIT-TlS cells. Glyceraldehyde, alanine, and potassium, each alone, produced 2-, 4-, and 10-fold stimulations, respectively, of insulin release from RINm5F cells (data not shown). A.) D-glucose (50mg/dl)-stimulated immunoreactive insulin release from HIT- T15 cells. B.) DL—glyceraldehyde (10 mM) stimulated release from RINm5F cells. C.) DL-alanine (10 mM) stimulated release from RINm5F cells. D.) Potassium (20 mM KCl) stimulated release from RINm5F cells. Stimulated secretion was calculated from increases above basal release (secretogogue absent, CPH absent) and are expressed as a percent of maximal secretion (secretogogue present, CPH absent) . Each point represents the mean value (iSEM) from 3 to 10 separate experiments (3 culture dishes per treatment per experiment). Asterisks denote values significantly different from.controls for each cell line (p<.05). 112 . ~ . ) n ) M u. w ( 0 C. 0 C. * T i n. . N .X. I l. .u N v C v c H . H P v P C l C *VIO“. .- 4 d6 m 4 0d q q 1' m 0 5 O 5 O O 5 O 5 0 O 7 5 2 O 7 5 2 JOmPZOU no thUmwa JOmHZOU uO szumwm . 0 1| v 1| W \l . \I w M 0 C *T.|. w. m m C *.O.. m. ..l m w 0 V H C . C l c H l W . P v C v C 4| r 1|. d vl...l|. d d. u.- 0. q *«IOII. q d a 0 0 5 0 5 O m n w "a o 0 7 5 2 405.200 .10 hzuumwm Jomhzoo no hzwomua 113 HIT-T15 CELLS RINm5F CELLS A 1sz B 1sz / \ '5 ” \ 1004 l“ g '\6 s 100- ’ \ 3 \ 5 \ 5 P “9 ”a 75« 9X:\ 3 75‘ \\ ‘57 \ 9’9 ‘é \ g 50. \ g 504 * n. \l o. * ‘0 25‘ o—o Non-IRI * an: 25‘ o—o Non-IRI e—e IRI e—«e IRI o I V T o I j 1 f o 0.1 1.0 10.0 o 0.1 1.0 10.0 CPH Cone. (pM) CPH Conc. (nu) Figure 20. Effects of CPH on insulin and non-insulin protein synthesis HIT-T15 cells (A.) and RINm5F cells (8.). Cells were incubated with 3H-leucine (25 uCi/l ml) and CPI-I for 2 hr. Values were calculated as cpm of 3M-leucine incorporated into insulin immunoreactive proteins (IRI) and non-insul in proteins (Non-IRI) , and expressed as a percent of control (no CPH) for each experiment. In these experiments, control RINm5F cells incorporated 894 i 133 cpm per 10‘ cells per 2 hr into GPAIS- precipitable protein and 9051 i 339 cpm into non-insulin proteins. Control HIT-T15 cells incorporated 1017 t 96 cpm per 10" cells per 2 hr into GPAIS-precipitable protein and 7724 i 837 cpm into non-insulin proteins. Each point represents the mean value (1 SEM) of three separate experiments. Asterisks denote values significantly different from controls for each cell line (p<.05). 114 150- 0—0 RINm5F ... 125-1 e—0HIT-T15 a -\1 '2 100-0 ' 0 l C) L5 75- o ,_ 1 E C.) 50“ a: D: 33 6 25- .. 0.1 1.0 10.0 CPH CONC. (pM) Figure 21. Cellular insulin content of RINm5F and HIT-T15 cells after culture for 48 hr in the presence of CPH. Values were calculated as ng of cellular insulin per pg of DNA and are expressed as a percent of control (no CPH) for each experiment. See materials and methods (p. 71) for control cellular insulin content values. Each point represents the mean value (1 SEM) of 3 or 4 separate experiments (5 culture dishes per treatment per experiment) . Asterisks denote values significantly different from controls for each cell line (p<.05). 150- 0—0 RINm5F _1 125- 0—0 HIT-T15 0 L: E 1004 T 8 9 LE. 9 Q 50- o: E . 25- O O * o-. . . . - ...”... 0.1 1.0 10.0 CPH CONC. (pM) Figure 22. Insulin concentration in media from RINm5F and HIT-T15 cells after culture for 48 hr in the presence of CPH. Values were calculated as total ng insulin in the media, and are expressed as a percent of control (no CPH) for each experiment. See materials and methods (p. 71) for control media insulin values. Each point represents the mean value (i-SEM) of 4 separate experiments (5 culture dishes per treatment per experiment). Asterisks denote values significantly different from controls for each cell line (p<.05). 116 _J 1007 o—o RINm5F (I) fCr) O—O HIT-T15 '2 e o 75- O l 0 L1. 0 +2: 50~ 8 C E .. 1 0. 25-6 * C * 0‘ I j I O 24 48 HOURS OF RECOVERY Figure 23. Recovery of cellular insulin content in RINm5F and HIT-T15 cells after removal of CPR (10 uM, 48 hr exposure). Values were calculated as ng of cellular insulin per ug of DNA, and are expressed as a percent of control for each experiment (no CPH during the exposure period). Each point represents the mean value (iSEM) of 3 separate experiments (3 culture dishes per treatment per time point per experiment). Asterisks denote values significantly different from.controls for each cell line at each time point (p<.05). 117 _. 1oo~ g o—o RINm5F E e—+—e HIT—T15 . o 75- . o u- l O '— 50- * E I : o 33 9 0 o. 25- 1 * * 9 0" I1 T ' O 24- 48 HOURS OF RECOVERY Figure 24. Recovery of media insulin from RINm5F and HIT-T15 cells after removal of CPH (10 pM, 48 hr exposure). Values were calculated as ng insulin in media from each 24 hr interval during the recovery period, and are expressed as a percent of control values (no CPH exposure) at each time point. Each point represents the mean value (:1: SEM) of 3 separate experiments (3 culture dishes per treatment per time point per experiment). Asterisks denote values significantly different from controls for each cell line at each time point (p<.05). 118 A B 250 250- 7 [:1 o CPH l :1 o CPH _L -10pMCPH -10pMCPH 2: 20m § 200. E E J 8 150- o ‘50 ‘5 ‘5 E 100« ’2 100- 8 “63 E o E J o a so- . a so . c 8 8 o o- — o- o 24 4a 0 24 48 HOURS OF RECOVERY HOURS OF RECOVERY Figure 25. Lack of effects on RINm5F (A.) and HIT-T15 (8.) cell growth and division during recovery from CPH exposure. Values were calculated as pg of DNA per culture dish and are expressed as a percent of the amount in control dishes (no CPH) at the beginning of the recovery period. Control dishes of RINm5F and HIT-T15 cells contained 52-101 and 25-44 pg of DNA, respectively. Each bar represents the mean value (tSEM) of three separate experiments (3 culture dishes per treatment per time point per experiment). 119 Table 2. Trypan blue dye exclusion assay of RINm5F and HIT- T15 cell viability after CPH exposure for 24 and 48 hr. Each value represents the mean from two culture dishes. Cell Line CPH Conc. Percent Viabilitya (pM) 24 Hr. 48 gr. RINm5F Control 98.5 96.5 " 0.1 96.9 98.6 " 1.0 96.1 97.2 " 10.0 97.7 98.5 HIT-T15 Control 98.9 94.8 " 0.1 96.3 98.1 " 1.0 100.0 95.1 " 10.0 100.0 97.7 ° Data expressed as percent of total cells excluding trypan blue dye. 120 HIT-T15 RINm5F CON CPH CON CPH 1-2 3 45,6 7 §_fi9_fl‘_|07 11 12 A: «Hand. x‘ufl.’ 1.4.35: :i .... ~,. . 1.";sz Ain‘t .umui- w ... CPH: 2hr,10uM Figure 26. Northern analysis of PPImRNA (I) and fi-actin mRNA (A) levels in HIT-T15 and RINm5F cells exposed to 0 or 10 pM CPH for 2 hr. Lanes 1-3: Control HIT-T15 cells; Lanes 4-6: CPH-treated HIT-T15 cells; Lanes 7-9: Control RINm5F cells; Lanes 10-12: CPH-treated RINm5F cells. Bands represent RNA samples prepared from separate culture dishes of cells. 121 A B H lT-T15 ’uM CPH 0 .1 1 10 Figure 27. Northern analysis of PPImRNA (I) and fi-actin mRNA (A) levels in HIT-T15 cells (A.) and RINm5F cells (8.) exposed to CPR for 24 hr. Bands represent RNA samples prepared from individual culture dishes. 122 mM NaB uM CPH 0 .2 2 0 .1 1 1g _¢-—_ GAPDHmRNA . . . . Figure 28. Northern analysis of PPImRNA and GAPDH mRNA in RINm5F cells exposed to CPH or sodium butyrate. A.) PPImRNA levels in RINm5F cells exposed to CPH for 48 hr. 8.) GAPDH mRNA levels in RINm5F cells exposed to CPH for 48 hr. C.) PPImRNA levels in.RINm5F cells exposed to sodium.butyrate for 24 hr. Bands represent RNA samples prepared from individual culture dishes. 123 Table 3. Effects of CPH on PPImRNA levels in RINm5F and HIT- T15 cells exposed for 24 hr. Data were calculated as integrated peak areas from densitometric scanning of autoradiograms of Northern blots. Integrated peak areas for PPImRNA were normalized to those for fi-actin mRNA, and are expressed as a percent of control for each experiment. Each value represents the mean (iSEM) from three separate experiments. Cell Line CPH Conc. PPImRNA/B-actin mRNA (pM) (% of CON) RINm5F Control 100 i 20 n 0.1 74 i 3 " 1.0 97 i 25 n 10.0 75 i 24 HIT-T15 Control 100 t 23 " 0.1 81 i 21 u 1.0 89 i' 12 " 10.0 70 i 14 124 Figure 29. Comparison of the insulin-depleting effects of actinomycin D (A.), cycloheximide (8.), and.CPH (C.) in RINm5F cells exposed for 24 hr. Concentration-response curves are shown for effects on cellular IRI content and DNA content per culture dish. ‘Values were calculated as ng IRI per pg DNA, or pg DNA per culture dish, and are expressed as a percent of control for each experiment. For these experiments, control RINm5F cells contained .202 i .021 ng insulin per pg DNA (N=3) . Each point represents the mean (i SEM) for three separate experiments“ .Asterisks denote ‘values ‘that are significantly different from controls (p<.05). 1125 Acnuoumu o CYCLOHEXIMIDE fects of 125‘ 125‘ n RINmSF ~ 'g 100. E 100. are 5 5 ”es .5 75. .5 7s- .tent per g if so. 50. mm, or E a 254 251 :rcent Of A 1 O I fir I 1 o I v y ' mm o .05 0.5 5.0 0 .05 0.5 5.0 r “9 DNA COM- (09/ ml) ‘ Cone. (pH) of three Lhat are CYPfimflflflflUEMNE 150 125‘ d 1 Percent of Control 13 a N (I O I e o 0.1 110 10:0 ‘ Cone. (pM) 126 1501 I ........ 1,,” 1 .g ..... .. 5 100- 04—;\ ‘3 \ «u \ § \ s. \ a 50‘ o—o CPH 9‘ n— -o 4-DPMP * \\ 4..., 2-DPMP \ ... * o U U U U 0 0.1 1.0 10.0 Conc. (pM) Figure 30. Comparison of the insulin-depleting effects of CPH, 4-DPMP and 2-DPMP in RINm5F cells exposed for 24 hr. Values were calculated as ng IRI per pg DNA and are expressed as a percent of control for each experiment. For these experiments, control RINm5F cells contained .110 2 .039 ng insulin per pg DNA (N=3) . Each point represents the mean value (1 SEM) from three separate experiments. Asterisks denote values that are significantly different from controls. 127 In Vivg Effects of 4-9232 and z-DPMg on Rat Pancreatic PPImRNA eve Effects of a Single Dose of 4-DPMP or 2-DPMP on Rat Pancreatic PPImRNA, p-Actin mRNA and Insulin: This experiment was performed to determine if the insulin-depleting action of CPH analogs was associated with the ability of the compounds to lower PPImRNA levels in rats. Administration of a single dose of 4-DPMP (45 mg/kg, po.) caused a 40% decrease of PPImRNA levels at 10 hr after dosing (Table 4). An equimolar dose of 2-DPMP produced no decrease of PPImRNA levels. Also shown in Table 2, body weights at the time of sacrifice, pancreatiC' total RNA and pancreatic fi-actin mRNA were unchanged by 4-DPMP and 2-DPMP treatment. In animals that were treated with 4-DPMP and sacrificed at 24 hr after dosing, pancreatic IRI levels were decreased to 36% of control (94 i- 13 ng IRI/mg pancreas vs. 274 1’ 45 ng/mg), while pancreatic IRI levels in animals treated with 2- DPMP were not decreased (221 i 51 ng/mg). 128 ..mo.vmv Houucoo acne ucmnmuuau saucmosuacmam . .mnao> Houucoo coma on» no useonQQ o no commounxo can .muoHn you no nfisuvoacsuousn mo mcficccom ofiuumeouwmcoc Bonn moons xooa counsmoucw no nounuaoaoo case a .moouocsn ma non czm Hoaou ma no connoumxw sumo - o.m H «.mm o.oa H H.¢m m.H H ~.oa 6.4 H o.qa~ osmouw o.m a m.em .m.m H s.~o 8.6 a H.HH 8.5 H m.-~ expose «.5 a o.ooa s.HH H o.ooa ~.o a 5.6” m.n H n.mH~ zoo aAzoo we as a“zoo «0 we .Aoe\oav “my azme :fiuomun czmsHmm «zm Hmuoe panama soon ucoeumone caucouocom .zmm w cams on cousomoum one even .0H0«Eo> no A.on .w&\ma may mamalm .mzmnnv no :ofiucnumwcfiacn nouns as ea conflufiuonn mums macawcm noon mo mmoouo .auooouocsaaw on no commonmxm sumo .m sense :a no connoumxo can couoHsono .noouocsu as pan «2% Hana» ma no commouaxo sumo . .0000 ma H use no.8 « 6.44 m.o « s.ma .eo.nH « m.mma m.o u n.Hn ¢~\zmo on a sun so.» H «.mm H.H « s.n~ o.~a H m.aoa o.a a m.Hn ¢~\zoo ma « emu .es.a a n.n~ 6.6 a m.- .en.sa « m.on~ «.6 a n.~n ~H\:ao we a mag o.s~ « 8.8» o.~ a m.- «.8 H v.mna n.a « o.Hn ~H\zoo mm H non m.m H o.ooH o.~ a m.o~ m.~H u H.soa 8.6 « n.Hn o\zoo oxae\mcv nAo\zoo»a .Aoa\ua. xao\oec .68 . OGOU axe omooaao pauses Anal mess HmH «zmsamm Hence aspen zoom \ucmeummue UHHflflHO—uflm .zmm H coma no couconoua one even .0H0finm> no A.oa .ox\ma me. new no sawuouumacaace nouns H: «a use NH .o as coowuquasn ones manages noon no nmsouo .ucoucoo.:«nsmca use .dzmaumm .42“ Honou.oaueou0ceafl.coHusuucoocoo omoooam Bayou .unuaos upon mason no man no anon savage a no mucouum .n canes .z 2:: #u . '2- 9t.° a 2! 3 ., o :_o ns- i; ;.u he: Time Course of CPS Effects in the Rat: It has been known for almost twenty years that administration of CPH to rats causes structural and functional alterations of insulin-producing fi-cells in the endocrine pancreas (Longnecker et al., 1972) . Prior to the present. investigations it was known that CPH inhibited insulin synthesis (Hintze et ial., 1977) and. depleted. pancreatic insulin content (Rickert et al., 1975), but the biochemical and molecular alterations produced by CPH in islet B-cells were not well understood. Results presented in this thesis have provided new information related to the mechanism(s) of CPH actions in insulin-producing cells. These studies are the first to indicate that CPH treatment decreases pancreatic PPImRNA levels. This effect was rapid and reversible, and was also produced by treatment of rats with 4-DPMP, a close structural analog of CPH that also depletes rat pancreatic insulin content. While a number of experimental manipulations decrease PPImRNA levels in the ‘whole animal, including fasting' (Giddings et. al., 1981; 135 136 Giddings et al., 1982; Fukumoto et al., 1986), adrenalectomy (Fiedorek and Permutt, 1989), chronic infusion of insulin (Chen et al., 1989; Kruszynska et al., 1988), manganese deficiency in the diet (Baly et al. , 1988) , and administration of STZ (Permutt et al., 1984), the CPH-induced decrease of PPImRNA is unusual in its rapidity and magnitude. PPImRNA levels were decreased to 30% of controls within 10 hr after CPH administration. The other manipulations listed above require much longer (days to weeks) to produce similarly large decrements of PPImRNA. The reversibility of the PPImRNA response to CPH distinguishes it from the pancreatic response to the cytotoxic diabetogenic agent STZ. Permutt et al. (1984) administered a single dose of STZ to 2 day old rats, then evaluated pancreatic insulin content, proinsulin biosynthesis, and PPImRNA levels at 4 and 7 weeks of age. PPImRNA levels were decreased by 60% and 85% at 4 and 7 (weeks of age, respectively. Proinsulin synthesis and pancreatic insulin content were also decreased in both age groups. The persistence of STZ-induced pancreatic effects at 7 weeks after dosing is likely to be due to drug-induced cytotoxicity, as STZ is known to cause irreversible destruction of fi-cells (Cooperstein and Watkins, 1981) . The maximum CPH-induced decreases in pancreatic proinsulin and PPImRNA were as large as those produced by STZ, but were reversible within 24-48 hr after dosing. 137 The decrease of PPImRNA after an oral dose of CPH could occur by a drug-related inhibition of transcription of insulin genes, or by enhanced degradation of PPImRNA. Evidence was not presented in this thesis that would allow discrimination between these possibilities. However, the rapidity of the PPImRNA response to CPH suggests that the drug might increase degradation of the mRNA. The half-life of PPImRNA in cultured isolated rat islets has been estimated.to be between 30 and 80 hr (Welsh et al., 1985). If the half-life of PPImRNA in 2129 falls within this range, the rapid decline of PPImRNA after CPH treatment cannot. be accounted. for' by inhibition. of transcription alone. In cultured rat islets, glucose appears to selectively stabilize PPImRNA relative to total islet RNA, an effect that may be coupled to glucose-stimulation of PPImRNA translation (Welsh et al., 1985; Welsh et al., 1986). Conversely, perhaps the ability of CPH to inhibit PPImRNA translation may result in a destabilization of PPImRNA. The chronology related to alterations of pancreatic insulin, proinsulin and PPImRNA produced by CPH provide information regarding the possible mechanism of CPH-induced insulin depletion. Depletion of pancreatic insulin content was preceded by decreases of proinsulin and PPImRNA, supporting previous suggestions that CPH causes pancreatic insulin depletion by inhibiting proinsulin synthesis (Hintze et al., 1977; Halban et al., 1979). Maldonato et al. (1976) suggested that proinsulin levels in isolated rat islets can 138 serve as a surrogate measure of proinsulin synthesis rates. This proposal was based on the premise that the islet proinsulin pool is small and turns over rapidly as conversion to insulin takes place. Presumably, the same features are true of pancreatic proinsulin pools in the intact animal. The present results showing that proinsulin levels in the pancreas were decreased by approximately 50% at 1.5 hr after CPH administration indicates that the synthesis of proinsulin in y_i_v_g was inhibited prior to this time point. The rapidity of the decrease in proinsulin observed in this study is consistent with an acute, direct inhibition of proinsulin synthesis by CPH. The CPH-induced decline of PPImRNA in mg lagged behind the decrease of proinsulin levels by several hours. This suggests that the decrease of PPImRNA does not cause the decrease in proinsulin. Instead, it is more likely that the initial action of CPH is suppression of proinsulin synthesis, by inhibiting, in some as yet unknown manner, the translation of preformed PPImRNA. Acute Effects of CPS on Proinsulin Synthesis and PPImRNA Levels in Isolated Rat Islets, RINm5F Cells and HIT-T15 Cells: The conclusion that CPH inhibited proinsul in biosynthesis by a mechanism not associated with a loss of PPImRNA was supported by results from experiments using isolated rat islets, HIT-T15 cells and RINm5F cells. For experiments 139 utilizing isolated rat islets, a short (30 min) labelling period was used to determine whether acute CPH-inhibition of proinsulin synthesis occurred prior to, or coincident with, alterations of PPImRNA levels. Similar experiments performed with HIT-T15 and RINm5F cells employed a longer labelling period (2 hr) since the clonal cells appear to produce substantially less insulin per cell than islet fi-cells (C. Miller, unpublished observations). CPH acutely and selectively inhibited proinsulin synthesis in rat islets without decreasing PPImRNA levels. In RINm5F and HIT-T15 cells, CPH also acutely and selectively inhibited the production of insulin-immunoreactive proteins without decreasing PPImRNA levels. These data are consistent with the results from in yiy_o experiments and support the conclusion that CPH decreases proinsulin levels in y_iyg by a mechanism independent of a drug-related decrease in PPImRNA. The rapidity of the CPH effect to inhibit proinsulin synthesis in \Lim and to reduce levels of the prohormone in m is suggestive of a drug-induced alteration in the translation of PPImRNA. Although CPH-induced decreases of PPImRNA appear not to be involved in acute inhibition of proinsulin synthesis, the eventual lowering of PPImRNA levels in 1&9 may contribute to a prolonged suppression of proinsulin synthesis. Lack of available PPImRNA for translation would be expected to lead to decreased proinsulin synthesis. However, the PPImRNA deficit 140 may not be important in the actions of CPH if the translation process is inhibited. In that case, levels of PPImRNA would not be a factor in rates of proinsulin synthesis. Experiments were performed to examine whether CPH—induced depletion of cellular insulin content in RINm5F and HIT-T15 cells was associated with decreases in PPImRNA. Exposure of RINm5F and HIT-T15 cells for 24 hr to 10 pM CPI-I depleted insulin content to 25 and 50% of controls, respectively, while PPImRNA levels were not decreased by CPH in either cell line. These observations, coupled with the lack of acute (2 hr) CPH effects on PPImRNA levels, indicate that in these cells, . decreases of PPImRNA are not required for inhibition of insulin synthesis or depletion of insulin content. There are several possible reasons for the failure of CPH to decrease PPImRNA levels, while still decreasing proinsulin synthesis and depleting cellular insulin content in RINm5F and HIT-T15 cells. These tumor-derived cells appear to have lost the ability to regulate insulin gene expression in the same manner as primary fi-cells (Nielsen et al., 1985). It is possible that CPH-responsiveness has been lost or altered at the gene or mRNA level, but retained in some form at the level of translation. Alternatively, perhaps a CPH-sensitive PPImRNA pool exists in insulin-producing cells, and is smaller in the clonal cells than in normal islet fi-cells in 1119. Measurement of total cellular PPImRNA levels by the methods used in the present studies would not have detected CPH 141 effects on small subcellular pools of PPImRNA. At present it is.not known‘whether'CPH‘treatment.decreases PPImRNA levels in rat islets exposed for longer periods in m. It is possible that the lack of CPH-effects on PPImRNA in the two clonal cell lines may reflect an inability of CPH to decrease PPImRNA levels in yitrg. Elucidation of the reasons for the lack of CPH-induced.decreases of PPImRNA.in'RINmSF'and.HIT-T15 cells might shed new light on regulation of insulin gene expression in these cells and in normal islet fi-cells. Lack of CPR Inhibition of Proinsulin to Insulin Conversion:‘ Until the present investigation, there had been no information available pertaining to the potential effects of CPH on the conversion of proinsulin to insulin. Inhibition of conversion alone could lead to the depletion of pancreatic insulin content, as mature insulin was released or degraded without being replenished from proinsulin. However, if this were the case, proinsulin pools would not be expected to decrease as rapidly as they have been shown to do after CPH exposure in the whole animal (Hintze et al., 1977; C. Miller, this thesis). Alternatively, proinsulin to insulin conversion could be inhibited in addition to inhibition of proinsulin synthesis. This could lead to decreases in proinsulin and insulin levels. In order to clarify the potential role that inhibition of conversion might play in CPH-induced insulin depletion, experiments were conducted in dispersed rat islet 142 cells to examine directly the ability of CPH to inhibit the conversion process. In these experiments, conversion of proinsulin to insulin was demonstrated in HPLC analysis by the appearance of labelled insulin. peaks and ‘the coincident disappearance of labelled proinsulin peaks. Under the conditions of these studies, CPH produced no inhibition of the conversion process. Monensin (10 pM), a known inhibitor of proinsulin conversion (Gold et al., 1984), was included as a positive control in these analyses, and was shown to inhibit conversion in the islet cells. Lack of CPI-I inhibition of proinsulin to insulin conversion is consistent with the- proposal that suppression of proinsulin synthesis represents the primary action of CPH in the depletion of pancreatic insulin content. The possibility that CPH inhibits the processing of the primary translation product, preproinsulin, to proinsulin was not addressed in the present studies. Preproinsulin is very short-lived [t1]2 of approximately 1 min in isolated rat islets (Patzelt et al., 1978)], as it is rapidly processed to form proinsulin. As such, it is very difficult to measure the levels or synthesis rates of the primary translation product. Using a rapid pulse-chase labelling strategy involving 2-5 min labelling of isolated rat islets, Patzelt et al. (1978) measured the kinetics of preproinsulin synthesis under various experimental conditions. These investigators reported that CPH reduced the incorporation of 3H-leucine into 143 preproinsulin, a finding consistent with the current proposal that CPH inhibits translation of PPImRNA. A separate inhibitory action of CPR on processing of preproinsulin to proinsulin was not evaluated in those studies. .At the present time this possibility seems unlikely, but.can.not.be excluded. lt d e a ode s P8 Actions in the Endocr e Pancreas Effects of CPR in Cultured Dispersed Rat Islet Cells: In order to further investigate the actions of CPH, experiments were conducted to examine the ability of CPH to directly alter the function of insulin-producing cells in yitzg. The use of in 21319 systems allows for the precise control of drug concentrations surrounding the islet cells and ensures that the effects produced by the drug represent direct inhibitory actions on islet cellsw Prior to these studies, it was known that CPH inhibited insulin synthesis and secretion in isolated rat and mouse islets (Hintze et al., 1977; Halban et al., 1979; Chow et al., 1989; Joost et al., 1976), but little information was available regarding the ability of CPH to deplete insulin content in yitzg. In experiments utilizing cultured whole rat islets exposed to CPH in yitng, Halban et al. (1979) reported that CPH depleted islet insulin content, but this effect occurred only at high, potentially cytotoxic drug concentrations (2 50 pM) . While isolated whole islets are suitable models to study 144 acute effects of chemicals on insulin cell function, there is uncertainty regarding the utility of cultured whole islets to examine lgng;tgzm effects of chemicals on islet B-cells. In the pancreas, nutrients and oxygen reach the islet center via the islet vasculature, but with the placement of isolated islets in culture, nutrients and oxygen must diffuse through many layers of cells to reach.the islet centeru This impaired nutrient access and gas exchange causes the fi-cell-rich islet core to become necrotic over'a period of a few days in culture (Andersson, 1976). The present studies utilized cultured dispersed rat islet cells as a model system to evaluate the in Ms; insulin- depleting effects of CPR. Dispersed islet cells have been widely used as model systems for the study of islet function injyitzg (Kostianovsky et al., 1974; Ono et al., 1979: Weir et al., 1984: Nielsen and Lernmark, 1984). For the purposes of these studies, the main.advantage of using these cells is that the problem.of islet.central necrosis was avoided” .All of the cells receive an adequate supply of nutrients, and unimpaired exchange of gases can occur. Experiments were conducted with the cultured dispersed rat islet cells to examine the ability of CPH to inhibit insulin synthesis, secretion, and proinsulin to insulin conversion acutely, and to deplete cellular insulin content after prolonged exposure in 21:19. In the present studies, CPH was shown to inhibit insulin synthesis and secretion from n «2" 145 the islet cells in a similar manner as has been shown for isolated whole islets (Halban et al. , 1979; Chow et al. , 1988: Hintze et al., 1977; Donatsch et al., 1979). The responses were concentration-dependent, with inhibition of insulin secretion occurring at 0.1, 1.0 and 10 pM CPH, and selective inhibition of insulin synthesis occurring at 5 and 10 pM CPH. The inhibitory actions of CPH on insulin synthesis and secretion may occur by separate mechanisms . Chow et al . (1989) have demonstrated that among CPH and its metabolites, a reverse order of potency exists for inhibition of insulin secretion and inhibition of insulin synthesis. CPH is the most potent inhibitor of insulin secretion, and the least potent inhibitor of insulin synthesis. The CPH metabolite DMCPH-epoxide is the most potent inhibitor of insulin synthesis, and the least potent inhibitor of secretion. These findings are not consistent*with a common mechanism of action for CPH inhibition of insulin synthesis and secretion. Exposure of the cells for 48 .hr to CPH produces alterations of cellular insulin content. At 1.0 pM CPH, cellular insulin content was elevated relative to controls, while at 10.0 pM CPH, insulin content was depleted. These observations can be explained by referring to the concentration-responses for CPH inhibition of insulin synthesis and secretion. At 1.0 pM CPH, insulin release was inhibited while synthesis was not affected. Cells that were exposed for 48 hr to 1.0 pM CPH could continue to synthesize _. 146 insulin, but would not. be able to release it, thereby increasing cellular insulin.contents IExposure of cells for 48 hr to 10 pM CPR caused depletion of insulin content. This depletion presumably occurred as the net result of the almost complete inhibition of synthesis seen at this drug concentration, and the normal degradation and turnover of existing insulin stores within fi-cells (Halban and Wollheim, 1980). These results indicate that exposure of islet cells to concentrations of CPH that inhibit insulin synthesis can lead to depletion of cellular insulin content in yi§;_. Media insulin levels were decreased from islet cells after 48 hr CPH exposures. At 0.1 and 1.0 pM CPI-I, media insulin levels were decreased while cellular insulin content was elevated or unchanged. Presumably, these decreases of media insulin are related to the direct inhibition of insulin secretion mentioned above. At 10 pM CPH, cellular insulin content was also depleted. It is possible that in cells exposed for 48 hr to 10 pM CPH, the decreased media insulin might be partly due to reduced amounts of releasable cellular insulin. Experiments were conducted.t0>evaluate the reversibility of CPH-effects after 48 hr exposures of cultured islet cells. In these studies, CPH-induced decreases of islet cell insulin content and media insulin levels were only partially reversible over a 48 hr recovery period. It is possible that full reversibility would have been observed if a longer 147 recovery period was used. In the pancreas in 2129 and in intact islets in yigzg, intra-islet paracrine regulatory mechanisms may play a role in regulation. of‘ fi-cell function (Pipeleers, 1984). These paracrine regulatory mechanisms, which may also modulate the fi-cell response to CPH, are disrupted when islets are dispersed into individual cells (Pipeleers et al., 1985; Weir et al., 1984). For this reason, it was initially anticipated that the dispersed rat islet cells might not respond to CPH in the same manner as whole rat islets. However, the concentration-response relationships for CPH inhibition of insulin synthesis and secretion in dispersed rat islet cells were similar to those previously reported in whole rat islets (Chow et al., 1989). ‘In the present studies, depletion of cellular insulin content was produced by CPH at concentrations lower than those reported by Halban et al. (1979) to deplete insulin content of cultured whole rat islets. Whether this discrepancy reflects differences in exposure conditions (duration of exposure or CPH concentration), or actual differences in CPH sensitivity between whole islets and dispersed islet cells remains uncertain. The observation that CPH inhibited insulin synthesis and secretion and depleted insulin content in dispersed islet cells suggests that the inhibitory actions of CPH are not dependent upon the existence of normal islet architecture and/or intact paracrine regulatory mechanisms, and that the dispersed rat islet cell _— 4r 148 system may be a useful model to further investigate the mechanism of CPH actions. Effects of CPR in RINm5F and HIT-T15 Cells: Two clonal insulin-producing cell lines, RINm5F and HIT- T15, were utilized to further investigate mechanisms of CPH- induced inhibition of insulin synthesis and depletion of insulin content. Experiments were conducted to characterize the ability of CPH to deplete cellular insulin content in the cells after 24 and 48 hr exposures, and to examine the acute (2 hr) inhibitory effects of CPH on insulin production and release. In additional experiments, correlations were made between PPImRNA levels and CPH inhibition of insulin synthesis and depletion of cellular insulin content. Both RINm5F and HIT-T15 cells responded to CPH treatment with reversible losses. of’ cellular insulin. content, and diminished insulin biosynthetic and secretory capacities. CPH-induced responses were concentration-dependent, and occurred at non-cytotoxic drug concentrations. Insulin- depletion was also produced by exposure of RINm5F cells to 4- DPMP, an agent.with structural similarity to CPH that.has.been shown to deplete pancreatic insulin content in the rat (Hintze et al., 1977b: C. Miller, this thesis). Another structural analog of CPH, 2-DPMP, that. was without effect on rat pancreatic insulin content (Hintze et al., 1977b; C. Miller, this thesis), failed to deplete cellular insulin content of 149 RINm5F cells. The finding that these cell lines responded to CPH (and DPMP) treatment in a similar manner to isolated rat islets and.dispersed islet cells exposed in gitrg, and.to rats and mice treated in yiyg, suggested that the cells might be adequate models for the study of CPH actions on islet.fi-cells. The depletion of cellular insulin content in RINm5F and HIT-T15 cells appears to be due to the ability of CPH to inhibit insulin biosynthesis, because the drug reduced incorporation of’ iH-leucine into insulin immunoreactive proteins. It is likely that in these cells, as has been shown for freshly isolated rat islets (Hintze et al., 1977; Chow et al., 1989), CPH inhibits the production of insulin by inhibiting the synthesis of its precursor, proinsulin. Proinsulin synthesis was not measured directly in experiments with.RINm5F and.HIT-T15 cells for two reasons: 1) The kinetics of proinsulin ‘to insulin. conversion. has not been ‘well- characterized in these cells, thereby making it difficult to define a labelling interval that would ensure incorporation of 3H-leucine into proinsulin without conversion to insulin; and 2) the antibody used in the immunoprecipitation assay binds both insulin and proinsulin. Therefore, in RINm5F and HIT-T15 cells, the immunoprecipitation assay measures incorporation of 3H-leucine into total insulin immunoreactive proteins (insulin and proinsulin). In isolated rat islets, newly synthesized proinsulin requires approximately 45 min to be converted to mature insulin (Steiner and Tager, 1979) , and a 30 min TEE—iii ' 150 labelling period permits the direct measurement of proinsulin synthesis. In RINm5F and HIT-T15 cells, cellular insulin content was not elevated as observed in cultured rat islet cells after prolonged exposure to 1.0 pM CPH. At first this might seem somewhat surprising, since insulin release from the clonal cells also appears to be more sensitive to CPH inhibition than insulin synthesis. However, the clonal cells have lower insulin biosynthetic capacities than primary islet fi-cells (Gold et al., 1988; Valverde et al., 1988), and the tendency to gain insulin content while release was inhibited may be less for the clonal cells. CPH-induced depletion of cellular insulin content in RINm5F and. HIT-T15 cells 'was completely reversible upon removal of the drug, and this is consistent with recovery of pancreatic insulin seen after administration of CPH to rats (Rickert et al., 1975). The recovery of cellular and pancreatic insulin content likely reflects reversibility of CPH inhibition of insulin biosynthesis. The reduction of media insulin levels observed after 48 hr exposures of the cells to CPH is probably due to direct inhibition by the drug of insulin secretion, because as shown in these studies, CPH inhibited secretogogue-stimulated insulin release. The lower concentration of insulin in the culture media of CPH-treated cells may also be due to the depleted cellular insulin content, resulting in reduced ’1‘ 151 amounts of releasable insulin in the cells. A complete recovery of media insulin was not observed for RINm5F cells, but this result may not accurately reflect the status of insulin secretion by the cells at the end of the 48 hr recovery period. The media insulin values represent cumulative hormone release over a 24 hr period and full recovery attained during the 24-48 hr period would not have been detected. The fact.that.CPH inhibited insulin secretion in response to stimulation by glyceraldehyde, alanine or potassium depolarization in RINm5F cells is consistent with current knowledge of CPH actions in fi-cells. The insulin secretory responses of RINm5F cells to these stimulants have been shown to be calcium-dependent (Wollheim and Pozzan, 1984) , and it is known that CPH can interfere with calcium.movement into islet fi-cells (Donatsch et al., 1980; Kloppel et al., 1978; Joost et al., 1976). CPH inhibition of glucose-stimulated insulin release from IHIT-Tls cells is also consistent ‘with. the proposed calcium antagonist properties of CPH, as glucose- stimulated insulin release from HIT-T15 cells has also been shown to be calcium-dependent (Ashcroft et al., 1986; Boyd et al., 1986). RINm5F and HIT-T15 cells have important functional differences between them. Within the context of these studies, the most important of these differences concerns the relative glucose-responsiveness of the cell lines. Glucose 152 has been shown to stimulate insulin synthesis and secretion in HIT-T15 cells, but not in RINm5F cells (Hill and Boyd, 1985; Ashcroft et al., 1986; Hammonds et al., 1987; Gold et al., 1988; Meglasson et al., 1987; Praz et al., 1983; Welsh et al., 1985). This difference provided. a means to examine a potential interaction between glucose signalling mechanisms and CPH alteration of insulin cell function. The observation that CPH inhibited insulin synthesis and depleted insulin content in both cell lines suggested either that the mechanism of CPH action may not be dependent upon the existence of operational glucose signalling mechanisms, or that CPH acts at a site that is distal to the abnormality that causes RINm5F cells to be unresponsive to glucose. The biochemical basis for the glucose-unresponsiveness of RINm5F cells is not well understood. It has been suggested that abnormalities of glucose transport (Giroix et al., 1985; Malaisse et al., 1986; Meglasson et al., 1986) and/or metabolism (Halban et al., 1983; Giroix et al., 1986; Vischer et al., 1987) may account for the failure of RIN cells to secrete insulin in response to stimulation by glucose. The reasons for the glucose insensitivity of insulin biosynthesis in RINm5F cells have not been elucidated. RIN cells have been used to investigate a possible link between glucose handling and the selective toxicity of STZ to insulin-secreting cells. Unlike CPH, the actions of STZ appear to be dependent upon the existence of normal glucose 153 signalling mechanisms because RIN cells were found to be less sensitive to the actions of STZ when compared to normal islet fi-cells (LeDoux et al., 1984; Mossman et al., 1986). This lower sensitivity may be due to a reduced ability of RIN cells to transport glucose and the glucose-containing STZ molecule into cells. When compared to published results obtained for isolated rat islets (Hintze et al., 1977; Halban et al., 1979), RINm5F cells appear to be as sensitive as isolated pancreatic islets to the insulin-inhibitory actions of CPH. E fec s s Wold et al. (1971) examined.the ability of CPH to produce morphological alterations in pancreatic islets of several species. These investigators administered repeated oral dose of CPH, then evaluated pancreatic tissue for morphological alterations using the light microscope. CPH caused severe alterations in rat islets, but was without effect in islets from mice, rabbits and hamsters. Since that time it has been believed that the mouse is less sensitive than the rat.to CPH- induced fl-cell toxicity. However, until the present investigations, the insulin-depleting effects of CPH in mice had not been adequately evaluated. Results presented here indicate that the mouse is susceptible to CPH-induced depletion of pancreatic insulin content. Administration of eight daily doses of CPH (45 mg/kg, po.) to mice produced reversible decreases of 154 pancreatic insulin content and elevations of serum glucose concentrations. These effects appeared to occur slightly more slowly than has been documented for CPH effects in the rat. Rickert et al. (1975) reported that CPH depleted rat pancreatic insulin content to 50% of control 24 hr after the first of fourteen daily CPH doses (45 mg/kg, po.), while plasma glucose concentrations were elevated 24 hr after the second daily dose. In the present investigation, mouse ‘pancreatic insulin content.was depleted by 50% 24 hr after the second of eight daily doses, while serum. glucose concentrations were not significantly elevated until 24 hr after the sixth daily dose. The metabolic fate of CPH is different in rats and mice. In rats, CPH is N-demethylated to form DMCPH (Wold et al., 1972), which is further metabolized to DMCPH-epoxide (Hucker et al., 1974; Hintze et al., 1975). Mice also form DMCPH (Wold and Fischer, 1972), but fail to form DMCPH-epoxide (Hintze et al., 1975). It has been proposed that DMCPH- epoxide may be partly responsible for the actions of CPH in rats, as it appears to be approximately 20-times more potent than CPH in inhibiting proinsulin synthesis in isolated rat islets (Chow et al., 1989), and inhibitors of drug metabolism reduce CPH-induced depletion.of pancreatic insulin content in rats (Chow et al.,1988). The lack of formation of DMCPH- epoxide has been suggested to lead to diminished sensitivity of the mouse to the insulin-depleting effects of CPH (Wold.and 155 Fischer, 1972; Chow et al., 1988). This proposal may be correct as mice appear to respond to CPH somewhat more slowly than rats. However, it is important to note that the formation of the CPH metabolite DMCPH-epoxide is not required for CPH-induced depletion of pancreatic insulin content, because as shown in this study, insulin-depletion was produced after CPH administration to mice. Mice also appear to be susceptible to CPH-induced decreases in pancreatic PPImRNA levels. In these studies, administration of a single dose of CPH (45 mg/kg, po.) caused a rapid and reversible decrease of mouse pancreatic PPImRNA levels. PPImRNA in CPH-treated animals were decreased to 26% of controls 12 hr after dosing. Partial restriction of food intake also caused a decrease in PPImRNA levels, but this decrease was not evident until 24 hr after food restriction. At 24 hr after dosing, PPImRNA levels in CPH-treated animals were not significantly different from control. This reversibility was due to decreased PPImRNA levels in control animals and increases in.CPH-treated animals from 12 to 24 hr. Since CPH-induced decrease of PPImRNA in mice appears to be similar to that produced in rats, formation of the metabolite DMCPH-epoxide may not be required for this action. 156 P oo e at one nsulin Bios nthesis Ge 0 CPH appears to rapidly and reversibly inhibit several components of the insulin biosynthetic pathway. The drug suppresses proinsulin synthesis by a mechanism likely to involve inhibition of translation of preformed PPImRNA, and decreases PPImRNA levels Iby :mechanisms that. may involve inhibition of insulin gene transcription and/or enhanced PPImRNA degradation. Interestingly, these effects are opposite to those produced by glucose. Glucose elevates PPImRNA levels by stimulating insulin gene transcription (Nielsen et al., 1985) and stabilizing PPImRNA (Welsh et al., 1985), and stimulates proinsulin synthesis by enhancing the translation.of preformed.PPImRNA (Itoh and.Dkamoto, 1980). It is tempting to speculate that CPH may interfere with the pathway by which glucose stimulates insulin gene expression. As this pathway is not completely understood, CPH may represent a useful tool for the study of translational and transcriptional control of insulin biosynthesis. Use of CPH has already provided new information related to regulation of insulin gene expression. Results from recently completed RNase protection analyses of rat insulin I and II mRNA levels in control and CPH-treated rats indicates that CPH has a more rapid inhibitory action on rat insulin II gene expression than on rat insulin I gene expression (Giddings et al., manuscript in preparation). These results 157' are significant in that they represent the first conclusive evidence that expression of the two rat insulin genes can be differentially altered by experimental manipulation. These findings may help to explain the evolutionary advantage conferred.to:mice and.rats in retaining the duplicated insulin (I) gene. Elucidation of the mechanism(s) by which CPH decreases pancreatic PPImRNA levels may provide new information related to turnover and degradation of PPImRNA, and potential coupling between translation and PPImRNA stability. If CPH inhibits insulin gene transcription without affecting PPImRNA stability, results presented in the present studies suggest that the half-life of PPImRNA in 11:19 may be considerably shorter' (6-10 hr) than. estimated. previously in. cultured isolated rat islets in 21:19 (30-80 hr; Welsh et al., 1985). Alternatively, if CPH enhances PPImRNA degradation, investigation of the mechanism of this action may provide support for the proposal that coupling exists between PPImRNA translation and stability. 158 EQM! £2 QONCLUSIONS The studies described in this thesis have examined the mechanism(s) of CPH-induced inhibition of proinsulin synthesis and depletion of pancreatic insulin content. Results from initial experiments indicated that CPH treatment decreased PPImRNA levels in rats. This finding lead to the hypothesis that CPH-induced decreases in PPImRNA cause inhibition of proinsulin synthesis and depletion of pancreatic insulin content. In order to test this hypothesis, experiments were conducted to examine the ability of CPH and similar compounds to alter pancreatic proinsulin, insulin and PPImRNA levels in 111g, and PPImRNA and insulin levels, and proinsulin synthesis in 13.12.- CPH treatment caused rapid and reversible decreases in pancreatic PPImRNA levels in rats. Decreases of pancreatic PPImRNA and insulin could also be produced by treatment of rats with the structurally-related compound, 4-DPMP. When compared to other experimental manipulations known to decrease PPImRNA levels, the CPH-induced decrease is unusual in rapidity, magnitude and reversibility. In time course experiments with rats, PPImRNA levels were decreased to 35% of control within 10 hr after a single dose of CPH (45 mg/kg, po.) and returned to control levels 24 hr after dosing. In 159 these studies, pancreatic proinsulin levels were decreased to 50% of control within 1.5 hr after CPH‘treatmentm The finding that CPH-induced decreases in proinsulin occurred before decreases in PPImRNA suggests that the decline of PPImRNA can not cause the drop in proinsulin, and that the drug inhibits proinsulin synthesis by a mechanism not involving a decrease PPImRNA levels. This conclusion was supported by results from acute experiments with isolated rat islets, RINm5F cells and HIT-T15 cells exposed to inhibitory concentrations of CPH in mpg. In these experiments, CPH selectively inhibited insulin and proinsulin synthesis without decreasing PPImRNA levels, suggesting that acute CPH suppression of proinsulin synthesis occurs by inhibition of translation of preformed PPImRNA, not by decreasing PPImRNA levels. CPH-induced decreases in PPImRNA and insulin were also shown to occur in mice. The magnitude of the decrease in PPImRNA appeared to be similar to that produced in rats, as CPH decreased PPImRNA levels in mice to 25% of control 12 hr after a single oral dose (45 mg/kg). However, the depletion of pancreatic insulin content in mice occurred more slowly than in rats. In mice, insulin depletion required at least 2 daily doses of CPH (45 mg/kg, po.), whereas in rats, a single dose of CPH (45 mg/kg, po.) caused insulin depletion. It has previously been proposed that the rat-specific metabolite, DMCPH-epoxide, may contribute to the actions of CPH in rats. The finding that CPH-induced insulin depletion occurred more “‘1 "45' 160 slowly in mice, supports this proposal. However, since mice appear to be as sensitive as rats to CPH-induced decreases in PPImRNA, DMCPH-epoxide may not be involved in this action. The direct effects of CPH were evaluated in cultured dispersed rat pancreatic islet cells, and in the clonal insulin-producing cell lines RINm5F and HIT-T15. In these studies, CPH was shown to produce similar effects in each of the cell systems. In short-term experiments (<2 hr), CPH selectively inhibited insulin synthesis and decreased insulin secretion. In 24 and 48 hr exposures in culture, CPH decreased media and cellular insulin levels. These effects were produced at drug concentrations that produced no decreases of cell viability. In rat islet cells, CPH was shown not to inhibit the conversion of proinsulin to insulin, suggesting that this process is not a target of CPH actions in 1119. In RINm5F and HIT-T15 cells, CPH-induced depletion of cellular insulin content was shown not to be associated with decreases in PPImRNA levels. The reasons for the CPH- insensitivity are unclear, but this finding’does lend further support to the contention that.decreases in PPImRNA.may not.be required for CPH-induced depletion of pancreatic insulin content. 0n the basis of the data presented here, the initial hypothesis that CPH-induced decreases in PPImRNA cause the inhibition of proinsulin synthesis and depletion of insulin 161 content must be rejected. CPH-induced inhibition of proinsulin synthesis causes the depletion of insulin content, but the suppression of proinsulin synthesis appears to be due to inhibition of translation of preformed PPImRNA, not to decreases in PPImRNA levels. CPH may decrease PPImRNA levels by inhibiting insulin gene transcription and/or enhancing PPImRNA degradation. Whether the decreases in PPImRNA and proinsulin synthesis represent independent actions of CPH, or separate consequences of alterations in a common initial target is uncertain at present. Since CPH appears to inhibit multiple steps of the insulin biosynthetic pathway (inhibition of PPImRNA translation, and decreases in PPImRNA levels), this islet toxicant might represent a useful tool for the study of regulation of insulin biosynthesis. P NCEB “‘1 162 Allison, J. , Campbell, I.L. , Morahan G. , Mandel T.E. , Harrison L.C., and Miller, J.F.A.P.: Diabetes in transgenic mice resulting from over-expression of class I histocompatibility' molecules in pancreatic B cells. Heggge 333:529-533, 1988. Andersson, A.: Tissue culture of isolated pancreatic islets. 52:21.88922I18211 [Suppl.](Kbh) 205:233-293: 1975- Ashcroft, S.J.H., Crossley, J.R. and Crossley, P.C.: The effect of N-acylglucosamines on the biosynthesis and secretion of insulin in rat pancreatic islets. Biochem. L 154:701-707, 1976. Ashcroft, S.J.H., Bunce, J., Lowry, M., Hansen, 8.8., and Hedeskov, CuJ.: The effect of sugars on (pro)insulin biosynthesis. Bieeheme_ge 174:517-526, 1978. Ashcroft, S.J.H.: Glucoreceptor mechanisms and the control. of insulin release and biosynthesis. Diabegologie 18:5- 15, 1980. Ashcroft, S.J.H., Hammonds, P. and Harrison, D.E.: Insulin secretory responses of a clonal cell line of simian virus 40-transformed B-cells. Qiaeegologia 29:727-733, 1986. Baly, D.L., Lee, I. and Doshi, R.: Mechanism of decreased insulinogenesis in manganese-deficient rats, Decreased insulin mRNA levels. FEBS Lett. 239:55-58, 1988. Bathena, S.J., Oie, H.K., Gazdar,.A.F., Voyles, N.R., Wilkins, S.D. and Recant, L.: Insulin, glucagon, and somatostatin receptors on cultured cells and clones from rat islet cell tumor. Diabetee 31:521-531. 1982. Bell, G.I., Pictet, R.L., and Rutter, W.J.: Sequence of the human insulin gene. Neture 284:26-32, 1980. Bell, G.I., Swain, W.F., Pictet, R., Cordell, 8., Goodman, H.M., and Rutter, W.J.: Nucleotide Sequence of a cDNA clone encoding human preprosinsulin. 3333 282:525- 527, 1979. Berne, C.: Anti-insulin serum coupled to sepharose 48 as a tool for the investigation of insulin biosynthesis in the B-cells of obese hyperglycemic mice. Engeczieol. 97:1241-1247, 1975. "'1’ . .5 .3... 163 Blobel, G. and Dobberstein, B.: Transfer of proteins across membranes I. Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane-bound ribosomes of murine myeloma. LEM 67:835-851, 1975. Bouchard, P., Sai, P., Reach, 6., Caubarrere, I., Ganeval, D. and Assan, R.: Diabetes mellitus following pentamidine- induced hypoglycemia in humans. Diabetes 31:40-45, 1982. Boyd, A.E. III, Hill, R.S., Oberwetter, J.M. and Berg, M.: Calcium dependency and free calcium concentrations during insulin secretion in a hamster beta cell line. Q_§11n Inggg; 77:774-781, 1986. Bucchini, D., Ripoche, M.A., Stinnakre, M.G., Desbois, P., Lor‘es, P., Mbnthioux, E., Absil, J., Lepesant, J.A., Pictet, R. , and Jami, J .: Pancreatic expression of human insulin gene in transgenic mice. grog, Nat, Acad, Sci, 83:2511-2515, 1986. Bucchini, D., Madsen, 0., Desbois, P., Pictet, R., and Jami, J.: B islet cells of pancreas are the site of expression of the human insulin gene in transgenic mice. ,Ezp‘_ge11 Egg; 180:467-474, 1988. Budd, G.C., Pansky, B. and Cordell, 8.: detection of insulin synthesis in mammalian anterior pituitary cells by immunocytochemistry and demonstration of insulin-related transcripts by in, situ RNA-DNA hybridization. J; fll§§22h§E1_Q¥§Q§D§EL 34:573'5731 1985- Brunstedt, J., and Chan, S.J.: Direct effect of glucose on the preproinsulin mRNA level in isolated pancreatic islets. W 106:1383-1389. 1982. Cahill, G.F. and.ncDevitt, H.o.: Insulin-dependent.diabetes: the initial lesion. M 304:1454-1463, 1981. Chan, S.J., Keim, P., and Steiner, D.F.: Cell-free synthesis of rat preproinsulins: Characterization and partial amino acid sequence determination. EIQQI_H§§1_AQ§QL_§211 73: 1964-1968, 1976. Chan, S.J., Noyes, B.E., Agarwal, R.L., Steiner, D.F.: Construction and selection of recombinant plasmids containing full-lenght complementary DNAs corresponding to rat insulins I and II. WM 76:50:36- 5040, 1979. 164 Chatterjee, N.K. and Nejman, C.: Insulin mRNA content in pancreatic beta cells of coxsackievirus B4-induced diabetic mice. W 55:193-202, 1988. Chen, L., Komiya, I., Inman, L., O'Neil, J., Appel, M., Alam, T., and. Unger, R. H.: 1Effects of hypoglycemia and prolonged fasting on insulin and glucagon gene expression. J1_Qlin1_1nxest1 84:711-714, 1989. Chick, W.L., Warren,S., Chute, R.N., Like, A.A., Lauris, V. and Kitchen, K.C.: A transplantable insulinoma in the rat. 2:99. Egg, Aged. Sgi, 74:628-632, 1977. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J.: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemg 18:5294- 5299, 1979. Chow, S.A. and Fischer, L.J.: Metabolism and disposition of cyproheptadine and desmethylcyproheptadine in pregnant and fetal rats. Q;ug_fle§ebe_pi§pe 15:740-748, 1987. Chow, S.A., Falany, J.L. and Fischer, IuJ.: cyproheptadine metabolites inhibit proinsulin and insulin biosynthesis and insulin release in isolated rat pancreatic islets. £211_Bigli_19812911 5:129-143 1989- Chow, S.A. and Fischer, I..J.: Susceptibility of fetal rat endocrine pancreas to the diabetogenic action of cyproheptadine. Iexieel, 5221, Ehermecol, 84:264-277, 1986. Chow, S.A., Rickert, D.E., and Fischer, L.J.: Evidence that drug metabolites are involved in cyproheptadine-induced loss of pancreatic insulin. g, Bhgrm, Exp. The:, 246: 143-149, 1988. Christie, M.R. and Ashcroft, S.J.H.: Cyclic AMP-dependent protein phosphorylation and insulin secretion in intact islets of Langerhans. fiieeheme_1e 218: 87-99, 1984. Clark, J.L. and Steiner, D.F.: Insulin biosynthesis in the rat: Demonstration of two proinsulins. ,1?_J'E'ee_,__;1et_g_5eeg_L £914,62:278-285, 1969. Cleveland, D.W., Lopata, M.A., MacDonald, R.J., Cowan, N.J., Rutter, W.J. and Kirschner, M.W.: Number and evolutionary conservation of a- and fi-tubulin and cytoplasmic B- and gamma-actin genes using specific cloned cDNA probes. Qell 20:95-105, 1980. 165 Colca, J.R., Kotagal, N., Lacy, P.E., Brooks, C.L., Norling, L., Landt, M. and McDaniel, M.L.: Glucose-stimulated protein phosphorylation in the pancreatic islet. fiieehene_le 220:529-537, 1984. Cooperstein, S.J. and Watkins, D.: Action of toxic drugs on islet cells. In: W. Cooperstein, S.J. and Watkins, D. Eds., Academic Press, New York, pp. 387-425, 1981. Cordell, B., Bell, G., Tischer, E., DeNoto, F.M., Ullrich, A., Pictet, R., Rutter, W.J. and Goodman, H.M.: Isolation and characterization of a cloned rat insulin gene. Cell 18:533-543, 1979. Craighead, J.E.: Current Views of insulin-dependent diabetes mellitus. fl, Engl, Q, neg, 299:1439-1445, 1978. Devillers-Thiery, A., Kindt, T., Scheele, G., and Blobel, G.: Homology in amino-terminal sequence of precursors to pancreatic secretory proteins. W _ 72:5016-5020, 1975. Docherty, R., Carroll, R., and Steiner, D.F.: Identification of a 31,500 molecular weight islet cell protease as cathepsin B. W 80:3245-3249, 1983. Docherty, R., Carroll, R., and Steiner, D.F.: Conversion of proinsulin to insulin: Involvement of a 31,500 molecular weight thiol protease. Proc. Natl. Acad. Sci. 79:4613- 4617, 1982. Donatsch, P., Lowe, D.A., Richardson, B.P. and Taylor, P.: Mechanism by which cyproheptadine inhibits insulin secretion. W1 70:355-362, 1980. Douglas, W.: Histamine and 5-hydroxytryptamine and their antagonists- In: Ine___2harm3221291211___82si§___gf W. 7‘" Ed. (A.G. Gilman, L.S. Goodman, T.W. Rall and F. Murad eds.), Macmillan, New York, pp. 605- 638, 1985. Duncan, D.B.: Multiple range and multiple F tests. Biometrics 11:1955 Dunn, J.S., Sheen, H.L. and McLetchie, N.G.B.: Necrosis of islets of Langerhans produced experimentally. Ihugger, 244 I:484-487, 1943. Eagle, H.: Nutrition needs of mammalian cells in tissue culture. Seienee 122:501-505, 1955. 166 Edlund, T., Walker, M.D., Barr, R.J., and Rutter,‘W.J.: Cell- specific expression of the rat insulin gene: Evidence for role of two disticnt 5' flanking elements. Seienee 230:912-916, 1985. Eizirik, D.L., Sandler, 8., Welsh, N. and Hellerstrom, C.: Preferential reduction of insulin production in mouse pancreatic islets maintained in culture after streptozotocin exposure» .EDQQQIianl 122:1242-1249, 1988. Eng, J. and Yalow, R.S.: Evidence against extrapancreatic insulin synthesis. Eroe, Net, Aead. Sei. 78:4576-4578, 1981. Engelhardt, R.L., Zell, H.C., Saari, W.S., Christy, M.E., Colton, C.D., Stone, C.A., Stavorski, J.M., Wenger, H.C. and Ludden, C.T.: Structure-activity relationship in the cyproheptadine series. J. Meg, Chem. 8:829-835, 1965. Episkopou, V., Murphy, A.J.M., and Efstratiadis, A.: Cell- specified expression of a selectable hybrid gene. 2:291 nag. ACQQ §Qi, 81:4657-4661, 1984. Epstein, P.N., Overbeek, P.A., and Means, A.R.: Calmodulin- induced early-onset diabetes in transgenic mice. gel; 58:1067-1073, 1989. Feinberg, A.P. and Vogelstein, B.: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Age1e_fiieeheme 132:6-13, 1983. Fiedorek, F.T. Jr. and Permutt, M.A.: Proinsulin.mRNA levels in fasting and fed ADX rats: evidence for an indirect effect of glucocorticoids. Am. J, Physiel. 256:E303- 8308, 1989. Fischer, L.J., Wold, J.S. and Rickert, D.E.: Structure- activity relationships in drug-induced pancreatic islet cell toxicity- I9212911_89211_£narm12211 26:288-298. 1973. Fourney, R., Miyakoshi, J., Day, R.S. and Paterson, H.C.: Northern blotting: Efficient RNA staining and transfer. zgene 10:5-7, 1988. Fukumoto, H. , Seino, Y. , Koh, G. , Takeda, J. , Tsuji, K. , Kurose, T. , Kitano, N. , Tsuda, K. , Taminato, T. , and Imura, H. : Effects of oral glucose administration on preproinsulin mRNA in viVO- WW 141:1201-1206, 1986. 167 Gazdar, A.F., Chick, W.L., Oie, H.K., Sims, H.L., King, D.L., Weir, G.C. and Lauris, V.: Continuous, clonal, insulin- and somatostatin-secreting cell lines established from a transplantable rat islet cell tumor. Proc. Nat. Aged. £91; 77:3519-3523, 1980. Giddings, S.J. and Carnaghi, L.: Rat insulin II gene expression by extraplacental membranes. J. Biol. Chem. 264:9462-9469, 1989. Giddings, S. J. and Carnaghi, L.: The two nonallelic rat insulin mRNAs are regulated coordinately in vivo. L M 263:3845- -,3849 1988. Giddings, S. J., Chirgwin, J., and Permutt, M. A.: Effects of Glucose on proinsulin messenger RNA in rats in vivo. QM 31: 624- 629, 1982. Giddings, S.J. , Chirgwin, J. , and Permutt, M.A.: Glucose regulated insulin biosynthesis in isolated rat pancreatic islets is accompanied by changes in proinsulin mRNA. W 2371-7511985- Giddings, S.J., Chirgwin, J., and Permutt, M.A.: Evaluation of rat insulin messenger RNA in pancreatic and extrapancreatic tissues. W 28:343-347, 1985. Giddings, S. J. , Chirgwin, J ., and Permutt, M. A. . The effects of fasting and feeding on preproinsulin messenger RNA in rats. W 67: 952-960, 1981. Giddings, S.J., Orland, M.J., Weir, G.C., Bonner-Weir,S., and Permutt, M.A.: Impaired insulin biosynthetic capacity in a rat model for non-insulin-dependent diabetes. Studies with Dexamethasone. niebeteg 34:2-35-240, 1985. Giddings, S. J. and Rotwein, R.S.: Isolation and Quantititation of Insulin Messenger RNA From Pancreata. Ch. V. SeC- E. 99- 347-361. in: W W (J- Larner and S.L. Pohl, ed.) Wiley Interscience, NY, 1984. Gilmore, R. , Blobel, G. and Walter, P.: Protein transport across the endoplasmic reticulum. I. Detection in the microsomal membrane of a receptor for the signal recognition particle. Wm 96:463-469, 1982. Giroix, M.R., Sener, A., Dufrane, S. P., Malaisse-Lagae, F., and Malaisse, W. J .. Glucose metabolism in insulin- producing tumoral cells. Ar9h1__8192h9m1__8192h2§1 241:561-570, 1985. 168 Giroix, M.H., Sener, A. and Malaisse, W.J.: D-glucose transport and concentration in tumoral insulin-producing cells. W 2513C847-C851, 1986. Golander, A. and Spirer, 2.: Inhibition of the insulin response to glucose after treatment with cyproheptadine. A£§§1_£§Qié§21_§£§n§1 71:485-487,1982. Gold, G., Qian, R., and Grodsky, G.M.: Insulin biosynthesis in HIT cells. Effects of glucose, forskolin, IBMX, and dexamethasone. Diabetee 37:160-165, 1988. Gold, G., Pou, J., Nowlain, R.M., and Grodsky, G.M.: Effects of monensin on conversion of proinsulin to insulin and secretion of newly synthesized insulin in isolated rat islet. M5 33:1019-1024, 1984. Gold, G., Pou, J., Gishizky, H.L., Landahl, H.D., and Grodsky, G.M.: Effects of tolbutamide pretreatment on the rate of conversion of newly synthesized proinsulin and the compartmental characteristics of insulin storage in, isolated rat islets. Diabetes 35:6-12, 1986. Gunnarsson, R.: Inhibition of insulin biosynthesis by alloxan, streptozotocin, and. N-nitrosomethylureaJ M91291 W 11:759-765, 1975. Halban, P.A., Wollheim, C.B., Blondel, B., Niesor, E. and Renold, A.E.: Perturbation of hormone storage and release induced by cyproheptadine in rat pancreatic islets in vitro. W 104:1096-1106, 1979. Halban, P. A., Wollheim, C. B., Blondel, B. and Renold, A.E.: Long-term exposure of isolated pancreatic islets to mannoheptulose: Evidence for insulin degredation in the beta cell. 31W 29:2625-2633, 1980. Halban, P.A., Rhodes, C.J. and Shoelson, S.E.: High- performance liquid chromatography (HPLC): A rapid, flexible and sensitive method for separating islet proinsulin and insulin. mm 29:893-896, 1986. Halban, P.A., Amherdt, M., Orci, L., and Renold, A.E.: Tris(hydroxymethyl)aminomethane inhibits the synthesis and processing of proinsulin in isolated rat pancreatic islets without affecting release of insulin stores. mebetee 35:433-439, 1986. Halban, P.A. and Wollheim, C.B. : Intracellular degredation of insulin stores by rat pancreatic islets in _L‘Lre. An alternative pathway for homeostasis of pancreatic insulin content. ”1214—91131; 255:6003-6006,1980. 169 Halban, P.A., Praz, G.A~ and Wollheim, C.Bw: .Abnormal glucose metabolism accompanies failure of glucose to stimulate insulin release from a rat pancreatic cell line (RINm5F) . W 212:439-443 , 1983.- Ham, R.G.: Clonal growth of mammalian cells in a chemically defined, synthetic medium. Exec. Nat, Acad. S91, 53 : 288-293 , 1965. Hammonds, P., Schofield, P.N. and Ashcroft, S.J.H.: Glucose regulates preproinsulin.messenger RNA levels in a clonal cell line of simian virus 40—transformed B-cells. FEES m 213: 149-154 , 1987 . Hammonds, P., Schofield, P.N., Ashcroft, S.J.H., Sutton, R. and Gray, D.W.R.: Regulation and specificity of glucose- stimulated insulin gene exptression in human islets of Langerhans. EE§§_L§EL1 233:131-137, 1987. Hanahan, D.: Heritable formation of pancreatic B-cell tumors in transgenic mice expressing recombinant insulin/simian virus 40 oncogenes. Neenze 315:115-122, 1985. Hedeskov, C.J.: Mechanisms of glucose-induced insulin secretion. EhY§1211_B§!1 60:442-509, 1980. Hedeskov, C.J., Capito, K. and Thams, P.: Cytosolic ratios of free [NADPHJ/[NADP’] and [NADHJ/[NAD’] in mouse pancreatic islets, and nutrient-induced insulin secretion. Bigchsm1_l1 241:161-167, 1987. Helgason, T. and Jonasson, M.R.: Evidence for a food additive as a cause of ketosis-prone diabetes. Legcet 2:716, 1981. Hill, R.S. and Boyd, A.B. III: Perifusion of a clonal cell line of simian virus 40-transformed beta cells. Diabetes 34:115-120, 1985. Hintze, R.L., Wold, J.S. and Fischer, LMJ.: Disposition of cyproheptadine in rats, mice, and humans and identi f ication of a stable epoxide metabol ite . 12mg M2§§21_Di§21 3:1’9: 1975- Hintze, R.L., Grow, A.B. and Fischer, L.J.: Cyproheptadine- induced alterations in rat insulin synthesis. Bieeheme Ehezmeeele 26:2021-2027, 1977. Hintze, R.L., Aboul-Enein, H.Y. and Fischer, L.J.: Isomeric specificity of diphenylmethylpiperidine in the production of rat pancreatic islet cell toxicity. 22312211. 7:133- 140 , 1977 . 170 Howell, S.L. and Taylor, K.W.: Effects of glucose concentration on incorporation of [3H]leucine into insulin using isolated mammalian islets of Langerhans. Biesneml_8122hxsl_AQta1 130:519- -521. 1966. Hucker, H.B., Balletto, A.J., Stauffer, S.C., Zacchei, A.G. and Arison, B. H.: Physiological disposition and urinary metabolites of cyproheptadine in the dog, rat and cat. W 2: 406-415 1974- Humbel, R.E., Bosshard, H.R. and Zahn, H.: Chemistry of insulin. In: s'ol Vo . n o zencrees, (N. Freinkel and. D. Steiner, Edsw) The American Physiological Society, Baltimore, MD, pp. 111- 132, 1972. Itoh, N. and Okamoto, H.: Synthesis of preproinsulin with RNA preparation of streptozotocin-nicotinamide induced B- Cell tumor. EE§§_L§£§1 80:111-114, 1977. Itoh, N. and Okamoto, H.: Translational control of proinsulin . synthesis by glucose. Ne§gre 283:100-102, 1980. Itoh, N., Sei, T., Nose, R., and Okamoto, H.: Glucose stimulation of the proinsulin synthesis in isolated pancreatic islets without increasing amount of proinsulin mRNA. EEE§_L§§L1 93:343-347, 1978. Jahr, H., Beckert, R., Reiher, H., Besch, W. and Zuhlke, H.: Cyproheptadine inhibits proinsulin biosynthesis and secretion of human pancreatic islets. Homone Metab. B£§1 13:359, 1981. Jain, K. and Logothetopoulos, J.: Proinsulin biosynthesis by pancreatic islets of the rat and the study of alloxan cytotoxicity in m. Biechim, Biophys. Acge. 435:145- 151, 1976. Jain, H., Logothetopoulos, J., and Zucker, P.: The effects of D- and L-glyceraldehyde on glucose oxidation, insulin secretion and insulin biosynthesis by pancreatic islets of the rat. ‘flieenine_§iepny§e_5e§ee, 399:384-394, 1975. Joost, H. G., Poser, W. and Panten, U.: Inhibition of insulin release from the rat pancreas by cyproheptadine and tricyclic antidepressants. Naunxnzgshmiedehergs_brch1 EDQIEQQQL1 285:99-102, 1974. Joost, H.G., Beckman, J., Holze, S., Lenzen, S., Poser, W. and Hassleblatt, A.: Inhibition of insulin and glucagon release from the perfused rat pancreas by cyproheptadine (Periactinol, Nuran). Diepegelegie 12:201-206, 1976. 171 Joost, H.G. , Beckmann, J. , Lenzen, S. and Hasselblatt, A.: Inhibition of insulin release by cyproheptadine: Effects on 3 ' 5 ' -cyclic-AMP-content and “Ca-accumulation of incubated mouse islets. A§£Q1_EQQQQI1BQL1 82:121-126, 1976. Kakita, K., O'Connell, K. and Permutt, M.A.: Pancreatic contents of insulins I and II in laboratory rodents: analysis of immunoelectrophoresis. 121 ebefies 31:841- 845, 1982. Karlsson, O., Edlund, T., Barnett Moss, J., and Rutter, W.J.: A mutational analysis of the insulin gene transcription control region: Expression in beta cells is dependent on two related sequences within the enhancer. roc Aeege_§eie 84:8819-8823, 1987. Kemper, B., Habener, J.F., Mulligan, R.C., Potts, J.T. and Rich, A.: Pre-proparathyroid hormone: A direct translation product of parathyroid messenger mRNA. 229.9... mm 71:3731-3735, 1974. Kennedy, K.A. , Halmi, K.A. and Fischer, L.J.: Urinary excretion of a quaternary ammonium glucuronide metabolite of cyproheptadine in humans undergoing chronic drug therapy. M 21: 1813-1820, 1977. Kemmler, W., Peterson, J.D., and Steiner, D.F.: Studies on the conversion of proinsulin to insulin. M 246:6786-6791, 1971. Kloppel, G., Bommer, G., Ruttmann, E. and Schafer, H.J.: Qualitative and semiquantitative calcium cytochemistry in B-cells of mice treated with cyproheptadine and mannoheptulose. 82:11.2n992r1n911 87:786-798, 1978. Koranyi, L., Permutt, M. A., Chirgwin, J. M. and Giddings, S. J.: Proinsulin I and II gene expression in inbred mouse strains. mm 3: 1895- 1902, 1989. Kostianovsky, M., McDaniel, M.L., Still, M.F., Codilla, R.C. and Lacy, P.E.: Monolayer cell culture of adult rat islets of Langerhans. W 10:337-344, 1974. Kruszynska,Y.T., Villa-Komaroff, L., and Halban, P.A.: Islet B-cell dysfunction and the time course of recovery following chronic over insulinisation of normal rats. niggeeglggig 31: 621-626, 1988. Labarca, C. and Paigen, K.: A simple, rapid, and sensitive DNA assay procedure. W 102: 344- 352, 1980. 172 Lacy, P.E. and Kostianovsky, H.: Method for the isolation of intact islets of Langerhans from the rat pancreas. .Difihfitgfi 16:35-39, 1967. Laimins, L., Holmgren-Konig, M., and Khoury, G.: Transcriptional "silencer" element in rat repetitive sequences associated with the rat insulin 1 gene locus. BIQQL-HQSL_AQQQL_§£1L 33:3151'3155: 1985- Larner, J. : Insulin and oral hypoglycemic drugs: Gltucagon. In: Ihe_Eharma29l9g1c11_Ba§1§_gf__ner12_9112§ 7" Ed (A. G- Gilman, L. S. Goodman, T. W. Rall and F. Murad eds. ), Macmillan, New York, pp. 1490-1516, 1985. LeDoux, S. P. and Wilson, G. L.: Effects of streptozotocin on a clonal isolate of rat insulinoma cells. We BiQEnY§1_A§L§1 804: 387- 392, 1984. Lee, C.H., Lin, B.C., Costa, E. and Chuang, D.M.: Detection of mRNA species in the brain and other tissues of rat which hybridize with the pro-insulin gene. Fee. Free, . 43 : 2219A, 1984 . Leinweber, K.P. and Schatz, H.: Long-term and short-term effects of calcium, verapamil and diazoxide on biosynthesis and release of (pro-)insulin in isolated islets of rat pancreas. 82111.3212211n211 99:94- 100, 1982. Lin, B.J., Nagy, B.R., and Haist, R.E.: Efffect of various concentrations of glucose on insulin biosynthesis. W 91: 309-311, 1971. Lin, B.J. and Haist, R.E.: Insulin biosynthesis: The monoaminergic mechanisms and the specificity of "glucoreceptor". Engeezinele 96:1247-1253, 1974. Lin, B.J. and Haist, R.E.: Insulin biosynthesis: Effects of carbohydrates and related compounds. gen, 1, Rhysiel, Bhezmeeele 47:791-801, 1969. Lin, B. J. and Haist, R. E.: Effects of some modifiers of insulin secretion on insulin biosynthesis. Engeerinele 92:735-742,1972. Lin, B.J., Henderson, M.J., Levine, B.B., Nagy, B.R., and Nagy, E.M.: Effects of iodacetate an fluoride on islet respiration and insulin biosynthesis. mm 8:353-358, 1976. 173 Liu, R.S., Wang, C.Y., Mills, N., Gyves, M. and Ilan, J.: Insulin genes expressed in human placenta from normal and diabetic pregnancies. Bree, Nee, Aead. Sci. 82: 3868- 3870, 1985. Lo, D., Burkly, L.C., Widera, G., Cowing, C., Flavell, R.A., Palmiter, R.D., and Brinster, R.L.: Diabetes and tolerance in transgenic mice expressing class II MHC molecules in pancreatic beta cells. Cell 53:159-168, 1988. Lomedico, P. T., Chan, S. J., Steiner, D. F., and Saunders, G. F.: Immunological and chemical characterization of bovine preproinsulin. 1, Biel, Qhem , 252: 7971-7978, 1977. Lomedico, P. T. and Saunders, G. F. . Preparation of pancreatic mRNA: cell.- free translation of an insulin-immunoreactive polypeptide. Nugleig Aeies Res. 3: 381- -391, 1976. Lomedico, P.T., Rosenthal, N., Efstratiadis, A., Golbert, W., Kolodner, R. , and Tizard, R.: The structure and evolution 1 of the two nonallelic rat preproinsulin genes. gel]. 18:545-558, 1979. Longnecker, D.S., Wold, J.S. and Fischer, L.J.: Ultrastructural study of alterations in beta cells of pancreatic islets from cyproheptadine-treated rats. Qigggseg 21:71-79, 1972. Mackin, R. B. and Noe, B.D.: Characterization of an islet carboxypeptidase involved in prohormone processing. W 120: 457-468,1987 . Malaisse, W.J., Malaisse-Lagae, F., Sener, A. and Pipeleers, D.G.: Determinants of the selective toxicity of alloxan to the pancreatic B cell. Bree, Nae, Acad. Sei , 79:927- 930, 1982. Malaisse, W. J. . The endocrine pancreas as a target organ for toxicity. Ch. 8, pp. 143-158, In: Ta e a 'c 291. 11. (G. M. Cohen ed.) CRC Press, Boca Raton, FL, 1986. Malaisse, W.J., Giroix, M.H., Malaisse-Lagae, F., and Sener, A.: 3-O-methyl-D-glucose transport in tumoral insulin- producing cells. W 251: C841-C846, 1986. Maldonato, A. , Trueheart, P.A. , Renold, A.E. , and Sharp, G.W.G.: Effects of streptozotocin in vitro on proinsulin biosynthesis, insulin release and ATP content of isolated rat islets of Langerhans. W13 12:471-481, 1976. 174 Maldonato, A., Renold, A.E., Sharp, G.W.G., and Cerasi, E.: Glucose-induced proinsulin biosynthesis, Role of cyclic AMP. mm 26:538-545, 1977. Manabe, S. and Wada, O.: Triphenyltin fluoride (TPTF) as a diabetogenic agent. TPTF induces diabetic lipemia by inhibiting insulin secretion from morphologically intact rabbit B-cells. mm 30:1013-1021, 1981. Maniatis, T., Fritsch, E. F. and Sambrook., J.: Ch. 6, pp. 188- 209, Extraction, Purification, and Analysis of'mRNA From Eukaryotic cells. In: o : bo Meneel. Cold Spring Harbor Laboratories, 1982. Marble, A.: Ch. 1., pp. 1-10, Current Concepts of Diabetes Mellitus. In: Jeelin'e Qiebetes Mellitus. (A. Marble, P. White, R.F. Bradley, L.P. Krall, ed.) Lea and Febiger, Philadelphia, 1971. Meglasson, M.D., Manning, C.M., Najafi, H. and Matschinsky, F.M.: Fuel-stimulated insulin secretion by clonal. hamster B-cell line HIT-T15. m 36:477-484, 1987. Meglasson, M.D., Manning, C.D., Najafi, H. and Matschinsky, F.Mg: Glucose transport by radiation-induced insulinoma and clonal pancreatic B-cells. Qieeeeee 35:1340-1344, 1986. Meyer, D.I., Louvard, D. and Dobberstein, B.: Secretory protein translocation across membranes- the role of the "docking protein". Neeere 297:503-508, 1982. Miller, L.V., Stokes, J.D. and. Silpipat, C.: Diabetes mellitus and autonomic dysfunction after Vacor rodenticide ingestion. Di§h§E1_QéI§ 1:73, 1978. 1Morris, G. E. and.Korner, A1: The effect of glucose on insulin biosynthesis by isolated islets of Langerhans of the rat. W 208: 404-413 1970. Morris, G.E. and Korner, A.: RNA synthesis and the stimulation of insulin biosynthesis by glucose. Em Legge 10:165-168, 1970. Moss, L.G., Barnett Moss, J., and Rutter, W.J.: Systematic binding analysis of the insulin gene transcription control region: Insulin and Immunoglobulin enhancers utilize similar transactivators. 8:2620-2627, 1988. 175 Mossman, B.T., Ireland, C.M., Filipak, M., LeDoux, S. and Wilson, G.L.: Comparative interactions of streptozotocin and chlorozotocin with DNA of an insulin-secreting cell line (RINr). meetelegje 29:186-191, 1986. Muglia, L. and Locker, J .: Extrapancreatic insulin gene expression in the fetal rat. Proc. Nee. Aced. Sci, 81:3635-3639, 1984. Nagamatsu, S. , Bolaffi, J.L. and Grodsky, G.M.: Direct effects of glucose on proinsulin synthesis and processing during desensitization. 2299021 3101 . 120: 1225-1231, 1987 . Nagamatsu, S. and Grodsky, G.M.: Glucose-regulated proinsulin processing in isolated islets from rat pancreas. Dleeetes 37:1426-1431, 1988. Nielsen ,D.A., Welsh, M., Casadaban, M.J. and Steiner, D. F.: Control of insulin gene expression in pancreatic B-cells and in an insulin-producing cell line, RINSF cells. I.‘ Effects of glucose and cyclic AMP on the transcription Of insulin mRNA. MM 260:13585-13589, 1985. Nielsen, J. H. and Lernmark, A.: Purification of islets and cells from islets. In: a '0 ° WELL Academic Press, New York. pp. 99- 126, 1983. Nir, U., Walker, M.D., and Rutter, W.J.: Regulation of rat insulin 1 gene expression: Evidence for negative regulation in nonpancreatic cells. P oc Nat. a . ' 83:3180-3184, 1986. Ohlsson, H. and Edlund, T.: Sequence-specific interactions of nuclear factors with the insulin gene enhancer. Cell 45:35-44, 1986. Ohlsson,H., Karlsson, O., and Edlund, T.: A beta-cell- specific protein binds to the two major regulatory sequences of the insulin gene enhancer. Bree, Nae, Aced, $_9_1_,_ 85:4228-4231, 1988. Okamoto, H.: The role of poly(ADP-ribose) synthetase in the development of insulin-dependent diabetes and islet beta cell regeneration. W 1:15-20. 1985. . Ono, J. , Takaki, R. , Okano, H. and Fukuma, M. . Long-term culture of isolated pancreatic islet cells with special reference to the beta-cell function. 1n_yi§ze 15: 95- 102, 1979. 176 Orland, M. J., Chyn, R., and Permutt, M. A.: Modulation of proinsulin messenger RNA after partial pancreatecomy in rats, Relationships to glucose homeostasis. ,1, Clin. MES... 75. 2047-2055,1985. Orland, M.J. and Permutt, M.A.: Quantitative analysis of pancreatic proinsulin mRNA in genetically Diabetic (db/db) mice. Dieeeeee 36:341-347, 1987. Orland, M.J. and Permutt, M.A.: Genetic susceptibility to diabetes in inbred strains of mice: measurements of proinsulin mRNA and response to dexamethasone. mm 30:934-939,1987. Oyer, P.E., Cho, 8., Peterson, J.D. and Steiner, D.F.: Studies on human proinsulin. Isolation and amino acid sequence of the human pancreatic C-peptide. M1211. glam. 246:1375-1386, 1971. Parry, D. G. and Taylor, K. W.: The effects of sugars on incorporation of [3H]leucine into insulins. MM~ 100:2c-4c, 1966. Patzelt, C.° . Differential ingibitory action of cationic amino acids on protein synthesis in pancreatic rat islets. Diem-lg 31:241-246,1988. Patzelt, C. , Labrecque, A.D. , Duguid, J.R. , Carroll, R.J. , Keim, P.S. , Keinrikson, R.L. , and Steiner, D.F. : Detection and kinetic behavior of preproinsul in in pancreatic islets. W 75:1260-1264, 1978. Permutt, M. A.° . Effect of glucose on initiation and elongation rates in isolated rat pancreatic islets. Meme 248:2738-2742,1974. Permutt, M.A., Kakita, K., Malinas, P., Karl, I., Bonner- Weir, S., Weir, G., and Giddings, S.J.: An in vivo analysis of pancreatic protein and insulin biosynthesis in a rat model for non-insulin-dependent diabetes. ,1; W 73:1344-1350, 1984. Permutt, M. A.: Biosynthesis of insulin. Ch. 4., pp. 75-95 In: le .- - 0 .-.!°.‘ 121:” 30 ru- 3 '1 _ 0 0- an. £§£h21291~ (S.J. Cooperstein and D. Watkins, ed.) Academic Press, NY, 1981. Permutt, M. A. and Kipnis, D. M.: Insulin Biosynthesis I. On the mechanism of glucose stimulation. J.:—3.1M 247: 1194- 1199, 1972. 177 Permutt, M.A. and Kipnis, D.M.: Insulin Biosynthesis II. Effect. of' glucose on .ribonucleic {acid synthesis in isolated rat islets. ,1, Biol, ghee. 247:1200-1207, 1972. Permutt, M.A. and Kipnis, D.M.: Insulin Biosynthesis: Studies of islet polyribosomes. Proc. Net, AQQQ: gel, 69:505- 509 , 1972 . Perler, F., Efstratiadis, A., Lomedico, P., Gilbert, W., Kolodner, R., and Dodgson, J.: The evolution of genes: the chicken preproinsulin gene. Cell 20:555-566, 1980. Philippe, J., Drucker, D.J., Chick, W.L. and Habener, J.F.: Transcriptional regulation of genes encoding insulin, glucagon, and angiotensinogen by sodium butyrate in a rat islet cell line. Mol. gell. Biel. 7: 560- 563, 1987. Pipeleers, D.G., Marichal, M. and Malaisse, W.J.: The stimulus-secretion coupling of glucose-induced insulin release: XIV. Glucose regulation of insular biosynthetic ' activity. W 93:1001-1011, 1973. Pipeleers, D.G., Schuit, F.C., in't Veld, P.A., Maes, E., Hooghe-Peters, E.L., Van De Winkel, M. and Gepts, W.: Interplay of nutrients and hormones in the regulation of insulin release. EDQQQIin211 117:824-833, 1985. Pipeleers, D.G.: Islet cell interactions with pancreatic B- cells. Expereeeie 40:1114-1124, 1984. Praz, G.A. , Halban, P.A. , Wollheim, C.B. , Blondel, B. , Strauss, A.J . and Renold, A.E.: Regulation of immunoreactive-insulin release from a ‘ rat cell line (RINm5F) . ELM 210:345-352, 1983. Prosser, P.R. and Karam, J .K.: Diabetes mellitus follwing rodenticide ingestion in man. J . M, nee, Aseee, 239:1148-1150, 1978. Raizada, M.R., Weyhenmeyer, J.A. and Fellows, R.E.: Cells cultured from rat brain synthesize immunoreactive insulin. m 15:185A, 1979. Raizada, M.K.: Localization of insulin-like immunoreactivity in neurons from primary cultures of rat brain. Ml Beg; 143:351-357, 1983. Rakieten, N., Rakieten, M.L. and Nadkarni, M.V.: Studies on the diabetogenic actions of streptozotocin (NSC 37917). W 29: 91‘93 1953- I El 178 Randerath, K.: Amino acids, amino acid derivatives, peptides and proteins. In: 1111' n-leye; Chromangraehy, 2"d Edn. Academic Press, New York, pp. 110-128, 1966. Rau, K., Muglia, L., and Locker, J.: Insulin-gene expression in extrafetal membranes of rats. pfieeeeee 38:39-43, 1989. Rhodes, C.J., Lucas, C.A. and Halban, P.A.: Glucose stimulates the biosynthesis of rat I and II insulin to an equal extent in isolated pancreatic islets. Em Leeel 215:179-182, 1987. Rickert, D.E. and Fischer, L. J.: Cyproheptadine and beta cell function in the rat: Insulin secretion from pancreas segments in vitro. oc. Soc. E B' . Me . 150:1-6, 1975. Rickert, D.E., Burke, J. and Fischer, L.J.: Cyproheptadine- induced depletion of insulin in the rat. J. P a aco . E§EL_Ih§11 193:585-593, 1975. Sample, C.B. and Steiner, D.F.: Tissue-specific binding of a nuclear factor to the insulin gene promoter. ffififi Leeel 222:332-336, 1987. Sandler, 8., Welsh, M. and Andersson, A.: Streptozotocin- induced impairment of islet B-cell metabolism and its prevention by a hydroxyl radical scavenger and inhibitors of poly(ADP-ribose) synthetase. A§§§1__EDQ£m12911 IQXIEQLL 53:392-400, 1982. Santerre, R.F., Cook, R.A., Crisel, R.M.D., Sharp, J.D., Schmidt, R.J., Williams, C.D. and Wilson, C.P.: Insulin synthesis in a clonal cell line of simian virus 40- transformed hamster pancreatic beta cells. Ezoc, Nee, AQ§Q1_§EIL 78:4339-4343, 1981. Sarvetnick, N., Liggitt, D., Pitts, S.L., Hansen, S.E., and Stewart, T.A.: Insulin-dependent diabetes mellitus induced in transgenic mice by ectopic expression of class II MHC and interferon-gamma. Qell 52:773-782, 1988. Schatz, H., Maier, V., Hinz, M., Nierle, C. and Pfeiffer, E.F.: Stimulation of H-3-leucine incorporation into the proinsulin and insulin fraction of isolated pancreatic mouse islets in the presence of glucagon, theophylline and cyclic AMP. nieeeeee 22:433-441, 1973. .1 179 Schatz, H., Nierle, C., and Pffeiffer, E.F.: (Pro-)insulin biosynthesis and release of newly synthesized (pro)insulin from isolated islets of rat pancreas in the presence of amino acids and sulphonylureas. Egg, ,1, £11n1_1nxe§tl 5:477-485. 1975- Schechter, R., Holtzclaw, L., Sadiq, F., Kahn, A., and Devaskar , S . : Insulin synthesis by isolated rabbit neurons. Eneeerlnele, 123:505-513, 1988. Selden, R.F., Skoskiewicz, M.J., Russell, P.S., Goodman, H.M.: Regulation of insulin-gene expression, Implications for gene therapy. N, Engl, J, Med, 317:1067-1076, 1987. Selden, RJF., Skoskiewicz, M.J., Burke Hewie, K., Russell, P.S., and Goodman, H.M.: Regulation of human insulin gene expression in transgenic mice. NaEure 321:525- 528, 1986. Shields, D. and Blobel, G.: Cell-free synthesis of fish preproinsulin, and processing by heterologous mammalian" microsomal membranes. Bree, Nee, Acad. Sci. 74:2059- 2063, 1977. Shields, D.: Cell- free synthesis of angler fish preproinsulin: Complete amino acid sequence of the signal peptide. W 98: 242- 249 1981- Soares, M.B., Schon, E., Henderson, A., Karathanasis, S.K., Cate, R., Zeitlin, S., Chirgwin, J. and Efstratiadis, A.: RNA-mediated gene duplication: the rat preproinsulin I gene is a functional retroposon. o . e . o 5:2090-2103,1985. Smith, L.F.: Species variation in the amino acid sequence of insulin. Am1411_flsfi1 40:662-666, 1966. Stahler, M.S., Budd, G.C., and Pansky, B.: Evidence for insulin synthesis in normal mouse seminal vesicle based on in situ RNA-DNA.hybridization. £1211_B§R;2§1.36=999- 1006, 1987. Starr, J.I., Horwitz, D.L.,Rubenstein, A.H. and Mako, M.E.: Insulin, proinsulin, and Czpeptide. In: W W. 2 Ed- (B-M- Jaffe and H.R. Behrman, eds.). Academic Press, N.Y. pp 613-642, 1979. Steel, R. G. D. and Torrie, J. H.: §§§315L12§1 McGraw-Hill, New York, 1960. 180 Stein, R.W. and Ziff, E.B.: Repression of insulin gene expression by adenovirus type 5 Ela proteins. MILL. 3121; 7:1164-1170, 1987. Steiner, D.F., Chan, S.J., Welsh, J.M., Kwok, S.C.M.: Structure and evolution of the insulin gene. M gene_t_,_ 19:463-484, 1985. Steiner, D. F., Cunningham, D., Spigelman, L., and Aten, B.: Insulin biosynthesis: evidence for a precursor. m 157:697-700,1967. Steiner, D.F. and Oyer, P.E.: The biosynthesis of insulin and a probable precursor of insulin by a human islet cell adenoma. Bree. Neg, Acee. §cl. 57:473-480, 1967. Steiner, D.F. and Tager, H.S.: Biosynthesis of insulin and glucagon. In: Eegloeeineleey, (DeGroot et al., eds.). Grunne and Stratton, Inc., N.Y., Vol. 2, pp. 921-934, 1979. _ Stone, C.A., Wenger, N.C., Ludden, C.T., Stavorski, J.M. and Ross, C.A.: Antiserotonin-antihistaminic properties of cyproheptadine - W 131 = 73-84 . 1961. Sugden, M.C. , Ashcroft, S.J.H. and Sugden, P.H. : Protein kinase activities in rat pancreatic islets of Langerhans. 5122h§m1_l1 180:219-229, 1979. Takaki, R., Ono, J., Fukuma, M. and Ikegama, T.: Long-term culture of isolated pancreatic islet cells. 2:221_§Q§1 £321_fli_ll_H£§1 149: 402-406 1975- Tanese, T., Lazarus, N.R., Devrim, S., and Recant, L.: Synthesis and release of proinsulin and insulin by isolated rat islets of langerhans. MW 49:1394-1404, 1970. Tennant, J.R.: Evaluation of the trypan blue technique for determination of cell viability. mneelengeeien 2:685- 694, 1964. Uchigata, Y. , Yamamoto, H. , Kawamura, A. and Okamoto, H. : Protection by superoxide dismutase, catalase, and poly(ADP-ribose) synthetase inhibitors against alloxan- and streptozotocin-induced islet DNA strand breaks and against the inhibition of proinsulin synthesis. M121... Queue 257: 6084- 6088, 1982. 181 Uchigata, Y., Yamamoto, H., Nagai, H. and.Okamoto, H.: Effect of poly(ADP-ribose) synthetase inhibitor administration to rats before and. after injection. of alloxan and streptozotocin on islet proinsulin synthesis. Qiebetes 32 : 316-318 , 1983 . Ullrich, A., Shine, J., Chirgwin, J., Pictet, R., Tischer, E., Rutter, W.J., and Goodman, H.M.: Rat insulin genes: Construction of plasmids containing the coding sequences. fieieeee 196:1313-1319, 1977. Valverde, I., Barreto, M. and.Malaisse, W.J.: Stimulation by D-glucose of protein biosynthesis in tumoral insulin- producing cells (RINm5F Line). Endocrleol. 122: 1443- 1448 , 1988 . Villa-Komaroff, L., Efstratiadis, A., Broome, S., Lomedico, P., Tizard, R., Naber, S.P., Chick, W.L., and Gilbert, W.: A bacterial clone synthesizing proinsulin. Pgoc, W 75:3727-3731, 1978. Vischer, U., Blondel, B., Wollheim, C.B., Hoppner, W., Seitz, H.J. and Iynedjian, P.: Hexokinase isoenzymes of RIN- m5F insulinoma cells. Expression of glucokinase gene in insulin-producing cells. W 241:249-255, 1987. Walker, M.D., Edlund, T., Boulet, A.M., and Rutter, W.J.: Cell-specific expression controlled by the 5'-flanking region of insulin and chymotrypsin genes. HQLBIQ 306:557-561, 1983. Walter, P. and Blobel, G.: Translocation of proteins across the endoplasmic reticulum. II. Signal recognition protein (SRP) mediates the selective binding to microsomal membranes of in 2.1.1129 assembled polysomes synthesizing secretory proteins. Qe_gell_Blele, 91:551- 556,1981. Weir, G.C., Halban, P.A., Meda, P., Wollheim, C.B., Orci, L. and Renold, A.E.: Dispersed adult rat pancreatic islet cells in culture: A, B, and D cell function. neteeeliem 33:447-453, 1984. Welsh, M., Scherberg, N., Gilmore, R. and Steiner, D.F.: Translational control of insulin.biosynthesis. Evidence for regulation of elongation, initiation and signal- recognition-particle-mediated.‘translational arrest. by glucose. BiQ£h§m1_l1 235:459-467, 1986. 182 Welsh, M., Brunstedt, J., and Hellerstrom, C.: Effects of D- glucose, L-leucine, and 2-ketoisocaproate on insulin mRNA levels in mouse pancreatic islets. Biebehes 35:228- 231, 1986. Welsh, M., Nielson, D.A., MacKrell, A.J. and Steiner, D.F.: Control of insulin gene expression in pancreatic fi-cells and in an insulin-producing cell line, RINSF cells. II. Regulation of insulin mRNA stability. ,1, Biel, ghee, 260:13590-13594, 1985. Welsh, M., Weber, T., Wrange, O., Nielsen, D.A., Matthieu, M., Steiner, D.F. : Regulation of insulin gene expression by dexamethasone, a , and a phorbol ester. Biomee, Qieehim. ACEfi: 47:299-303, 1988. Welsh, N., Welsh, M., Steiner, D.F., and Hellerstrom, C.: Mechanisms of leucine- and theophylline-stimulated insulin biosynthesis in isolated rat pancreatic islets. Bieehehl_el 246:245-248, 1987. Welsh, N. and Sjoholm, A.: Polyamines and insulin production in isolated mouse pancreatic islets. Bloehee. g, 252: 701-707, 1988. Wold, J.S. and Fischer L.J.: The tissue distribution of cyproheptadine and it metabolites in rats and mice. 11 W 133:133-195: 1972 - Wold, J .S., Longnecker, D.S. and Fischer L.J.: Species dependent pancreatic islet toxicity produced by cyproheptadine: Alterations in fi-cell structure and function. 121W]... 19:188-201. 1971- Wollheim, C. B. and Pozzan, T.: Correlation between cytosolic free Ca” and insulin release in an insulin-secreting cell line. MW 259. 2262- 2267, 1984. Wu, C. and Gilbert, W.: Tissue-specific exposure of chromatin structure at the 5' terminus of the rat preproinsulin II gene. W13. 73:1577'1530: 1931- Yamamoto, H. , Uchigata, Y. and Okamoto, H.: Streptozotocin and alloxan induce DNA strand breaks and poly(ADP- ribose) synthetase in pancreatic islets. Hem 294:284-286, 1981. Zhang, H.J., Walseth, T.E. and Robertson, R.P.: Insulin secretion and cAMP metabolism in HIT cells. Reciprocal and serial passage-dependent relationships. Qieheeee 38:44-48, 1989. HICHIGQN T WM 31 I mIIISIMI‘IIIMMIILMQIMMET 2930