Glucagon167644/FULLTEXT01.pdf · Glucagon ! " # $ ! %! % ! & ' & & & ( ! $ ! " # $ $ % & ' () * # ) ) +* * ,$ )! -./* 0 1 # *. 2 1. &.3 # / )4 # 3 .5

Glucagon

To my mother Ana Marília and my beloved Pablo

List of papers

This thesis is based on the following papers, which will be referred to in the text by their roman numerals.

I Liu Y.J., Vieira E., Gylfe E. A store-operated mechanism deter-mines the activity of the electrically excitable glucagon-secreting pancreatic -cell. Cell Calcium, 35: 357-365 (2004)

II Vieira E., Liu Y.J., Gylfe E. Involvement of 1 and -adrenoceptors in adrenaline stimulation of glucagon-secreting mouse -cell. Naunyn Schmiedebergs Arch Pharmacol, 369: 179-183 (2004)

III Vieira E., Salehi A., Gylfe E. Glucose inhibits glucagon secretion from mouse pancreatic -cells independent of KATP channels and paracrine -cell influence (submitted article).

IV Salehi A., Vieira E., Gylfe E. Paradoxical stimulation of gluca-gon secretion by high glucose concentrations (submitted article)

Reprints were made with the permission of the publishers.

Contents

Introduction.....................................................................................................9Diabetes mellitus, a bihormonal disease ....................................................9Adrenergic control of glucagon secretion ................................................10Signal Transduction of Glucagon Secretion.............................................10Store-operated Ca2+ entry .........................................................................13

Aims..............................................................................................................15

Materials and Methods..................................................................................16Preparation of pancreatic islets and cells .................................................16Loading with indicators............................................................................17Parallel measurements of [Ca2+]i and membrane potential by digital imaging fluorometry.................................................................................17Paracrine influence...................................................................................19Identification of -cells............................................................................20Measurements of glucagon and insulin secretion.....................................20

Results and Discussion .................................................................................21Adrenaline stimulates glucagon secretion via 1 and adrenoceptors.....21Adrenaline stimulation and glucose inhibition of the -cell involve a store-operated current...............................................................................22KATP channels are functionally active in mouse -cells ...........................25The elevation of [Ca2+]i underlying glucagon secretion from mouse

-cells is due to opening of L-type Ca2+ channels....................................28Glucose inhibition of glucagon secretion in mouse islets occurs independently of KATP channels and products secreted from pancreatic

-cells .......................................................................................................28High glucose concentrations paradoxically stimulate glucagon release from mouse islets and clonal -cells ........................................................30

Conclusions...................................................................................................32

Svensk sammanfattning ................................................................................33

Acknowledgements.......................................................................................35

References.....................................................................................................37

Abbreviations

[Ca2+]i Cytoplasmic Ca2+ concentration cAMP Cyclic adenosine monophosphate EN Endoplasmatiskt nätverk ER Endoplasmic reticulum GABA -aminobutyric acid IP3 Inositol 1,4,5-trisphosphate KATP channel ATP-sensitive K+ channel PLC Phospholipase C PTX Pertussis toxin SERCA Sarco(endo)plasmic reticulum Ca2+ ATPase SOCs Store-operated channels

9

Introduction

Diabetes mellitus, a bihormonal diseaseThe plasma glucose concentration is normally maintained within a relatively narrow range (3-8 mmol/l) by hormonal and neural mechanisms. The hor-monal mechanisms for glucose homeostasis involve the glucose-lowering action of insulin and the glucose-elevating effects of glucagon and adrena-line.

Insulin secreted by pancreatic -cells reduces blood glucose by inhibiting glycogenolysis and gluconeogenesis in the liver and by stimulating glucose uptake, storage, and utilization in muscle and fat. Glucagon is secreted by pancreatic -cells and is primarily acting on the liver. It is a potent activator of glycogenolysis and gluconeogenesis and increases blood glucose levels within minutes. The hyperglycaemic effect of the adrenomedullary hormone adrenaline is more complex. Adrenaline has a glucagon-like action on the liver increasing glycogenolysis and gluconeogenesis and the hormone also stimulates the secretion of glucagon and inhibits that of insulin. A change in the balance between these hormones may consequently impair glucose ho-meostasis.

Diabetes mellitus is primarily a disease with hyperglycemia due to lack or inappropriate secretion of insulin. However, diabetes has long been viewed as a bihormonal disorder, since hypersecretion of glucagon contributes to the hyperglycemia. Both types 1 and 2 diabetes are associated with abnormali-ties of glucagon secretion. In diabetic subjects, hyperglycemia is aggravated because glucose fails to suppress the secretion of glucagon (Gerich et al., 1976) and a glucose challenge has been found to paradoxically stimulate glucagon release (Ohneda et al., 1978; Mitrakou et al., 1990). Moreover, a defective glucose counter-regulation with failing stimulation of glucagon release when the blood glucose concentration falls below the normal fasting level makes diabetic patients treated with insulin or sulphonylureas suscepti-ble to dangerous hypoglycemia. This is a potentially life-threatening condi-tion and hypoglycemia is a significant cause of deaths in insulin-treated dia-betes (Cryer, 2002). To improve the treatment of diabetic subjects it is there-fore important to clarify the signal transduction underlying both insulin and glucagon secretion.

10

Adrenergic control of glucagon secretion Like the secretion of most other hormones, exocytosis of glucagon from the

-cells is triggered by an increase of the cytoplasmic Ca2+ concentration ( Ca2+

i) (Barg et al., 2000). It is well established that adrenaline stimulation of glucagon secretion involves a -adrenergic mechanism with rise of cyclic adenosine monophosphate (cAMP) (Lacey et al., 1991). Studies in dispersed islet cells from the guinea pig indicated that this effect is exerted at the level of the -cell (Johansson et al., 1987; Johansson et al., 1989). Adrenaline was thus found to elevate Ca2+

i by activation of a -adrenergic mechanism in-volving rise of cAMP. Later studies on rat -cells have clarified that the stimulatory effect of adrenaline on exocytosis of glucagon is partly ex-plained by a protein kinase A-dependent increase of Ca2+ influx through voltage-dependent L-type channels (Gromada et al., 1997). Whereas it is generally accepted that adrenaline inhibition of insulin secretion involves an

2-adrenergic mechanism (Nakaki et al., 1980; Schuit and Pipeleers, 1986), there are different opinions about the involvement of -adrenergic mecha-nisms in adrenaline stimulation of glucagon secretion. Studies of the per-fused rat pancreas did not indicate an essential role of -adrenoceptors(Filipponi et al., 1986). On the other hand, in vivo experiments on rat and mouse have favoured the involvement of both 1 and 2 adrenoceptors (Skoglund et al., 1987; Saito et al., 1992).

Signal Transduction of Glucagon Secretion Nutrients and hormones are the most important regulators of glucagon secre-tion. Whereas, amino acids and adrenaline stimulate glucagon release, glu-cose is inhibitory (Pipeleers et al., 1985). Consistent with the roles of gluca-gon and insulin to maintain normoglycemia, glucagon secretion is inhibited when the glucose concentrations is raised to 4-6 mM and insulin secretion is stimulated at higher glucose concentrations of the sugar. Studies in the mouse have shown that stimulation of -cells results in depolarization-dependent large amplitude Ca2+

i oscillations similar to those in -cells pro-viding an explanation for pulsatile release of glucagon (Berts et al., 1995; Berts et al., 1996b; Berts et al., 1997). It is well established that -cells are electrically excitable and generate action potentials in the absence of glucose (Rorsman and Hellman, 1988; Gromada et al., 1997; Göpel et al., 2000a). The membrane conductances involved in action potential generation have been characterized in guinea pig (Rorsman, 1988; Rorsman and Hellman, 1988), mouse (Göpel et al., 2000b), and rat (Gromada et al., 1997) -cells. These studies indicate that -cells are equipped with voltage dependent T-, N-, and L-type Ca2+ currents. They also contain a prominent voltage-gated and tetrodotoxin-sensitive Na+ current that is activated during action poten-

11

tials (Bokvist et al., 1999; Göpel et al., 2000b). However, the mechanisms for stimulus-secretion coupling of glucagon release are unclear.

In the insulin releasing -cell, ATP-sensitive K+ (KATP) channels have a central role in stimulus-secretion coupling. During glucose stimulation it is the closure of these channels, which causes depolarisation leading to influx of Ca2+ through voltage-dependent channels, rise of Ca2+

i and exocytosis of insulin (Ashcroft and Rorsman, 1989). In pancreatic -cells, the role of KATP channels is controversial and their closure has been suggested to result in both stimulation and inhibition of glucagon secretion. Evidence for a stimulatory role of KATP channels in glucagon secretion has been obtained in rat -cells, which have much higher channel density (Bokvist et al., 1999) than mouse -cells (Quesada et al., 1999; Göpel et al., 2000b), even exceed-ing the density in rat -cells (Bokvist et al., 1999). Accordingly, tolbu-tamide-induced closure of the KATP channels, stimulates the electrical activ-ity (Bokvist et al., 1999) and exocytosis of glucagon (Høy et al., 2000) in isolated rat -cells, and glucose was recently found to paradoxically stimu-late glucagon release from purified rat -cells by closing the KATP channels (Olsen et al., 2005). Studies of the clonal hamster -cell line R1G9 have shown that closure of the KATP channels by tolbutamide tends to elevate Ca2+

i whereas the KATP channel opener diazoxide has a pronounced lower-ing effect (Bode et al., 1999). Ablation of KATP channels by knockout of the regulatory sulphonylurea receptor 1 subunit provided support for a stimula-tory role of these channels in the -cells by showing that glucagon secretion is diminished (Muñoz et al., 2005) or absent (Shiota et al., 2005) during ex-posure to low glucose concentrations. Apart from the role in the pancreatic

-cells, the KATP channels in the hypothalamus may be important for stimu-lation of glucagon release from the pancreas. Due to the presence of KATPchannels, low blood glucose concentrations are sensed by hypothalamic neu-rons, resulting in increased autonomic stimulation of the -cells (Miki et al., 2001; Evans et al., 2004).

Despite the above-mentioned evidence for a stimulatory role of KATPchannel closure in -cells and -cells, the KATP channel closure in mouse -cells has been suggested to lead to inhibition of glucagon release (Göpel et al., 2000b; Gromada et al., 2004). In this model, depolarization by glucose-induced closure of KATP channels inhibits glucagon release by inactivating voltage-dependent ion channels involved in the action potential firing (Göpel et al., 2000b). However, such a role is difficult to reconcile with observa-tions that glucose inhibits glucagon secretion from mice lacking KATP chan-nels after knockout of the channel forming inward rectifier K+ channel Kir6.2 (Miki et al., 2001) and inhibits the blunted glucagon secretion from SUR1 knockout mice (Muñoz et al., 2005). Glucose has also been proposed to inhibit glucagon secretion by a KATP channel independent mechanism. Based on early studies of guinea-pig -cells it was suggested that glucose inhibits glucagon release by lowering [Ca2+]i after promoting intracellular

12

sequestration and outward transport of the ion (Johansson et al., 1987; Jo-hansson et al., 1989). However, this mechanism cannot account for the inac-tivation of voltage-dependent Ca2+ entry associated with inhibition of gluca-gon secretion (Gromada et al., 1997; Barg et al., 2000). In another study of clonal hamster R1G9 -cells, the inhibitory effect of glucose is attributed to hyperpolarization by activation of the electrogenic Na/K-ATPase (Bode et al., 1999).

Apart from a direct effect of glucose on -cells, inhibition of glucagon re-lease may be due to stimulated secretion of paracrine factors from the other cells types within the pancreatic islet. It has been proposed that insulin (Östenson, 1979; Berts et al., 1996b; Diao et al., 2005; Ravier and Rutter, 2005), -aminobutyric acid (GABA) (Rorsman et al., 1989) and Zn2+

(Ishihara et al., 2003; Franklin et al., 2005) from the -cells or somatostatin from the -cells (Cejvan et al., 2003) are important inhibitors of glucagon release. The situation is complicated by observations indicating that different mechanisms may be involved depending on species. Recently, it was pro-posed that glucose inhibits glucagon secretion from rat islets but stimulates secretion in purified rat -cells (Franklin et al., 2005), indicating that inhibi-tion is accounted for paracrine factors in this species. However, in another study, glucose was found to inhibit glucagon secretion from purified rat -cells (Pipeleers et al., 1985). Studies of rat islets and -cells have provided evidence that secretion of GABA from the -cells is involved in glucose inhibition of glucagon release (Rorsman et al., 1989; Wendt et al., 2004). Insulin and Zn2+, which is co-secreted with insulin, independently inhibit glucagon secretion from rat -cells/islets, perhaps by activating KATP chan-nels in -cells (Ishihara et al., 2003; Franklin et al., 2005). However, in an-other study, insulin had no effect on glucagon secretion from purified rat -cells (Pipeleers et al., 1985). Also somatostatin released from cells has been proposed to mediate glucose inhibition of glucagon release by acting on somatostain receptor subtype 2 (Cejvan et al., 2003). In another study, soma-tostatin was found to inhibit exocytosis in isolated rat -cells by activating the serine/threonine protein phosphatase calcineurin (Gromada et al., 2001).

The situation is different in the mouse. GABA has no effect on Ca2+i in

isolated mouse -cells (Berts et al., 1996b). Although GABA slightly re-duces glucagon secretion from mouse islets, this effect is too small to ex-plain the inhibitory effect of glucose (Gilon et al., 1991). In contrast to the inhibitory effect of Zn2+ on rat -cells, this cation has been found to activate mouse -cells (Ravier and Rutter, 2005). However, there is evidence for an inhibitory role of insulin in both primary (Berts et al., 1996b; Ravier and Rutter, 2005) and clonal (Ravier and Rutter, 2005) mouse -cells. Confus-ingly, studies of primary mouse and clonal TC-6 cells have indicated that insulin receptors are critical for stimulation of glucagon secretion at low glucose concentrations (Diao et al., 2005). The role of somatostatin to inhibit glucagon secretion may be similar in the mouse and rat. Somatostatin was

13

thus found to hyperpolarize mouse -cells by activating G-protein-gated K+

channels (Yoshimoto et al., 1999). Although GABA is released from -cells vesicles other than the insulin-

zinc complex, secretion is regulated in similar manner (Braun et al., 2004). A major objection against the involvement of paracrine -cells factors in the regulation of glucagon secretion is that the physiologically important inhibi-tion occurs at lower glucose concentrations than those stimulating insulin secretion (Gerich et al., 1976; Gylfe, 1990).

Store-operated Ca2+ entry The [Ca2+]i signals regulating a variety of cellular functions often result from a combination of Ca2+ influx through the plasma membrane and release from intracellular stores. In excitable cells like the -cell, voltage-dependent Ca2+

influx is most important for providing the elevation of [Ca2+]i, which triggers exocytosis (Barg et al., 2000). The store-operated or capacitative pathway for Ca2+ entry was first discovered in non-excitable cells lacking voltage-dependent Ca2+ channels (Putney, 1986; Putney, 1990). Activation of G-protein-coupled receptors linked to phospholipase C (PLC), results in generation of inositol 1,4,5-trisphosphate (IP3), which mobilizes Ca2+ from the endoplasmic reticulum (ER). The resulting calcium depletion of the ER activates a store-operated influx of Ca2+ required for store refilling and re-petitive release. It is the depletion rather than IP3 or an IP3 metabolite that triggers influx, since the store-operated channels (SOCs) can also be acti-vated by exposing the cells to Ca2+-deficient medium or to inhibitors of the sarco(endo)plasmatic reticulum Ca2+ ATPase (SERCA) (Putney, 1990). De-spite intense research the Ca2+ sensor in the ER and the signal activating SOCs remain elusive. According to the conformational coupling model emp-tying of Ca2+ from the ER leads to a conformation change in the IP3 recep-tors, which is transmitted to SOCs in the plasma membrane by direct pro-tein-protein interaction (Berridge, 1990; Irvine, 1990). Another model in-volves a diffusible calcium influx factor, which is released from the ER and is acting on the SOCs (Randriamampita and Tsien, 1993). Recently, a novel ER protein STIM1 was found to have a central role in the activation of SOCs (Draber and Draberova, 2005; Liou et al., 2005; Roos et al., 2005; Zhang et al., 2005). The exact function of STIM1 is unclear. It was suggested to act as Ca2+ sensor in the ER (Liou et al., 2005; Roos et al., 2005; Zhang et al., 2005), induce calcium influx factor production (Draber and Draberova, 2005) and even take part in the formation of SOCs (Zhang et al., 2005). However, other data indicate that SOCs belong to a family of transient re-ceptor potential channels (Hardie and Minke, 1993). The transient receptor potential channels can be classified into three major subfamilies: classical, vanilloid and melastatin. Expression of the classical and vanilloid type 6

14

channels has thus been found to induce Ca2+ entry in response to depletion of intracellular Ca2+ stores in different cell types (Parekh and Putney, 2005).

The store-operated Ca2+ influx plays a role also in excitable cells (Moffatt and Cocks, 2004; Parekh and Putney, 2005), where it has a depolarizing effect. The first evidence for a store-operated mechanism in islets was ob-tained in mouse pancreatic -cells with the observation that carbachol in-duces influx-dependent sustained elevation of [Ca2+]i even when the voltage-dependent Ca2+ channels are blocked (Gylfe, 1991). Later studies clarified that the emptying of the ER activates a voltage-independent influx pathway (Liu and Gylfe, 1997) and that the rate of influx is inversely proportional to the filling state of the ER (Dyachok and Gylfe, 2001). The Ca2+ sequestra-tion in the ER is stimulated by glucose (Gylfe, 1991; Tengholm et al., 1999; Tengholm et al., 2001), which consequently shuts off the store-operated pathway (Liu and Gylfe, 1997). Even when maximally activated, the current through the store-operated pathway is not sufficiently pronounced to depo-larize the -cell from the resting -70 mM to the -40 to -50 mV required for opening of the voltage-dependent L-type Ca2+ channels (Worley et al., 1994; Chow et al., 1995). Nevertheless it has been suggested that a store-operated current carried by Na+ or Ca2+ may be important for generating fast [Ca2+]ioscillations when the -cells are somewhat depolarized (Worley et al., 1994; Bertram et al., 1995; Gilon et al., 1999). Also influx of Ca2+ through the store-operated pathway may be of physiological significance for maintained PLC activity during exposure of -cells to G-protein coupled receptor ago-nists (Thore et al., 2005).

Due to a high input resistance the membrane potential of -cells is more sensitive to small currents than in -cells (Barg et al., 2000). Since the action potentials of -cells start at voltages as negative as -60 mV (Rorsman and Hellman, 1988; Bokvist et al., 1999; Göpel et al., 2000b), a small store-operated current can be expected to have much more dramatic effects in -cells than in -cells. Because adrenaline mobilizes intracellular Ca2+ in -cells (Johansson et al., 1989) activation of a store-operated current may un-derlie the depolarization, which activates voltage-dependent Ca2+ influx.

15

Aims

The aims of the present study were to clarify:

1. The possible involvement of KATP channels in signal transduction of glucagon secretion in mouse -cells

2. Whether a calcium store-operated mechanism can explain adrena-line stimulation and glucose inhibition of glucagon secretion in mouse -cells

3. Receptors and signalling pathways involved in adrenaline stimu-lation of glucagon secretion from mouse -cells

4. The possible involvement of KATP channels and paracrine release from pancreatic -cells in glucose inhibition of glucagon secre-tion from mouse -cells

5. The involvement of [Ca2+]i in the paradoxical glucose stimulation of glucagon release

16

Materials and Methods

Preparation of pancreatic islets and cells Islets of Langerhans were isolated with collagenase (Boehringer Mannheim GmbH, Mannheim, Germany) from NMRI and C57/BL6 mice. The Uppsala Ethics Committee approved the experimental procedures. The animals were killed by decapitation under anesthesia with CO2. The lower duodenal part of the pancreas was rejected to avoid cells producing pancreatic polypeptide (Liu et al., 1999). The freshly isolated islets were either used for studies of glucagon secretion or preparation of free cells. Free cells were obtained by incubating the islets for 4 min at 37 C in Ca2+-deficient medium containing 0.5 mM EDTA and 0.05% trypsin (Invitrogen, Carlsbad, CA) followed by brief shaking. The cells were suspended in RPMI 1640 medium (Gibco Ltd., Paisley, Scotland) supplemented with 10% fetal calf serum (Gibco), 100 IU/ml penicillin, 100 µg/ml streptomycin and 30 µg/ml gentamicin. Small samples of this suspension (15 l) were applied to the centers of poly-L-lysine-coated (Sigma Chemical Co., St. Louis, MO) circular 25 mm cover slips. The cover slips were then kept for 60 min in an incubator at 37°C with a humidified atmosphere of 5% CO2 to allow cells to settle and begin attachment. More medium was then cautiously added and the cells were cultured for 1-3 days. In some experiments 100 ng/ml pertussis toxin (PTX, Sigma) was present during the last 20-24 h. Rat basophilic leukemia cells (kindly provided by Professor Tobias Meyer, Stanford University, Stan-ford, USA) used in control experiments were cultured in Dulbecco’s modi-fied essential medium (Gibco) supplemented with 10% fetal calf serum. These cells were allowed to grow close to confluence on 25 mm cover slips. R1G9 hamster glucagonoma cells were kindly provided by Dr. Björn Olde, Lund University, Sweden with permission from Dr Jacques Philippe, Uni-versity of Geneva, Switzerland. The cells were cultured in plastic dishes or culture bottles in GlutaMAXTM-containing RPMI 1640 medium (Gibco) supplemented with 10% fetal calf serum, 100 U/ml penicillin, and 100 µg/ml streptomycin. For secretion studies the R1G9 cells were seeded at a density of 100 000 cells/well in 48-well plates and cultured for 3 days as described above.

17

Loading with indicatorsLoading of cells with the Ca2+ indicator fura-2 was performed during 40 min incubation at 37°C in a buffer containing 0.5 mg/ml bovine serum albumin (Sigma), 125 mM NaCl, 4.8 mM KCl, 1.2 mM MgCl2, 1.28 mM CaCl2, 3 mM glucose, 1 µM fura-2 acetoxymethyl ester (Molecular Probes Inc., Eugene, OR; 0.1% dimethylsulphoxide, Sigma) and 25 mM HEPES (Sigma) with pH adjusted to 7.4 with 13 mM NaOH. The cells were then equili-brated for 10 min at room temperature with 450 nM of the potential-sensitive probe bis-oxonol (Molecular Probes) in the same medium but lacking fura-2 acetoxymethyl ester. When the effects of higher concentrations of KCl were tested osmotic compensation was made by reduction of NaCl. The cover slips with the attached cells were used as exchangeable bottoms of an open chamber. The chamber volume was 0.16 ml and the cells were superfused at a rate of 1 ml/min with a medium containing 450 nM bis-oxonol. Thapsigar-gin (Sigma), which sticks to plastic, was added directly to the superfusion chamber with a pipette. The superfusion flow was then interrupted for 2-3 min to ascertain an effect of the drug.

Parallel measurements of [Ca2+]i and membrane potential by digital imaging fluorometry The superfusion chamber was placed on the stage of an inverted Nikon Dia-phot microscope equipped with an epifluorescence illuminator and a 40x oil immersion fluorescence objective. The chamber holder and the objective were maintained at 37°C by custom-built thermostats. The epifluorescence illuminator was connected through a 5 mm diameter liquid light guide to an Optoscan monochromator (Cairn Research Ltd., Faversham, UK) with rapid grating and slit width adjustment and a 150W xenon arc lamp. The mono-chromator provided excitation light at 340 nm (2.8 nm half-bandwidth), 380 nm (2.5 nm half-bandwidth) and 485 nm (2.5 nm half-bandwidth) and emis-sion was measured at >515 nm by an intensified CCD camera. The Meta-fluor software (Universal Imaging Corp., Downingtown, PA) controlled the monochromator acquiring fluorescence images of 30 accumulated frames at 340, 380 and 485 nm every 4 s. [Ca2+]i images were calculated from 340/380 nm ratio images as previously described (Liu et al., 2004). Increases in bis-oxonol fluorescence indicated depolarization and decreases hyperpolariza-tion (Epps et al., 1994).

Tolbutamide (Hoechst Marion Roussel AB, Stockholm, Sweden) in-creased the bis-oxonol fluorescence more than expected when comparing with K+ depolarization. Even the hyperglycaemic and hyperpolarizing sul-phonamide diazoxide increased bis-oxonol fluorescence in - and -cells, indicating that this class of drugs interacts with the indicator fluorescence

18

independently of membrane potential. Control experiments were therefore performed on non-excitable rat basophilic leukaemia cells in which sul-phonamides are not expected to affect the membrane potential. Fig. 1 indi-cates that 500 M tolbutamide, which was used in most experiments, in-creased the bis-oxonol fluorescence equivalent to a depolarization with about 50 mM K+. When comparing 500 M tolbutamide with of 50 M glipizide (Figs. 1-2) and 100 M glibenclamide (Fig. 2) it became apparent that the magnitude of the effect on bis-oxonol fluorescence was dependent on sul-phonamide concentration. Fig. 2 shows that none of the drugs affected [Ca2+]i in these non-excitable cells. Ideally the membrane potential re-cordings should be performed with a more potent sulphonamide than tolbu-tamide. This was not done because the interference was discovered at a rela-tively late stage of the studies. Bis-oxonol is a slow indicator and depolariza-tion in response to K+ or adrenaline seemingly lagged the [Ca2+]i responses by 25-30 sec.

Fig 1. Effect of 50 M glipizide (G), 500 M tolbutamide (T) and different concen-trations of K+ on the bis-oxonol fluorescence from rat basophilic leukemia (RBL) cells. The fluorescence change induced by tolbutamide corresponded to depolariza-tion by about 50 mM K+.

19

Fig 2. Effect of 500 M tolbutamide (T), 100 M glibenclamide (Glib) and 50 Mglipizide (G) on bis-oxonol fluorescence and [Ca2+]i of rat basophilic leukemia (RBL) cells.

Paracrine influence The procedures for concentrating cells to the central area of the cover slips resulted in an average of 13 cells per measured image field of 0,085 mm2.Since image fields with high cell density were selected to obtain measure-ments from more than one -cell, this density is probably an overestimate for the central region. Few cells were found in the periphery and the average cell density in the entire chamber with 64 mm2 bottom area was considerably smaller. The chamber volume (0.16 ml) was exchanged 6 times per min by the superfusion medium. It is apparent that the concentrations of paracrine factors released from -cell and -cells are much smaller than those obtained in the narrow extracellular space of islets when measuring insulin secretion in batch incubations. The concentration of insulin in the superfusion medium was determined by ultra sensitive ELISA (Bergsten et al., 1994) in 10 ex-periments. Under conditions, which stimulate insulin secretion maximally

20

(20 mM glucose), the effluent from the superfusion chamber contained <1 pM insulin. Pretreatment with PTX, which blocks the inhibitory effect of somatostatin on glucagon secretion (Göpel et al., 2004), and exposure to 0.1

M of the phosphatidylinositol 3 kinase inhibitor wortmannin (Sigma), which prevents insulin inhibition of the -cell (Ravier and Rutter, 2005), were used to clarify whether the effects of glucose on membrane potential and [Ca2+]i could be explained by paracrine influence from neighboring -and -cells.

Identification of -cellsThe -cells were initially selected by their small size and [Ca2+]i response to adrenaline (Sigma) (Johansson et al., 1989; I), which is not shared by - (I) and -cells (Berts et al., 1996a). Moreover, each experiment was terminated by immunostaining the cells in the experimental chamber. The cells were superfused with albumin-free medium and fixed with 95% ethanol. After rinsing with distilled water and Tris buffer (0.5 M, pH 7.6), normal goat serum (diluted 1:10; DAKO Corp.) was added to reduce background stain-ing. After 10 min, rabbit anti-glucagon (1:200; Zymed Laboratories INC) was added for 20-30 min followed by rinsing with Tris buffer. Biotinylated goat anti-rabbit immunoglobulin (1:500; DAKO) was then introduced for 10 min, followed by rinsing and addition of alkaline phosphatase-conjugated streptavidine (1:200; DAKO Corp.) for a further 10 min. The BCIP/NBT color reagent (DAKO) was then added for 2-5 min.

Measurements of glucagon and insulin secretion Batches of 8–12 islets and 3-day cultured R1G9 cells were pre-incubated for 30 min at 37 C in 1 ml of Krebs-Ringer buffer (pH 7.4) supplemented with 10 mM HEPES, 0.1% bovine serum albumin and 1 mM glucose. Each incu-bation vial was gassed with 95% O2 and 5% CO2 to obtain constant pH and oxygenation. The islets were then incubated for 1 h at 37 C in a Krebs-Ringer buffer supplemented with different glucose concentrations, 500 Mtolbutamide, 4.8 or 8 mM K+ and 50 M cyclopiazonic acid (Calbiochem, La Jolla, CA). At the end of the incubation, aliquots of the medium were removed and frozen pending the radioimmunoassays (Panagiotidis et al., 1992).

21

Results and Discussion

Adrenaline stimulates glucagon secretion via 1 and adrenoceptorsAmong blood glucose elevating hormones adrenaline is important both by mobilizing hepatic glycogen directly and by stimulating glucagon secretion. It is clear from the present study that the adrenaline effect on the -cell con-sisted of initial mobilization of intracellular Ca2+, accompanied by voltage-dependent influx of the ion (I). The presence of early and late effects of adrenaline raised the question whether different types of receptors are in-volved. Using guinea pig -cells it was previously demonstrated that

-adrenergic activation with elevation of cAMP is involved in adrenaline-induced increase of Ca2+

i required for glucagon secretion (Johansson et al., 1989). The existence of such a mechanism has been confirmed in rat -cells (Gromada et al., 1997). In the present study, the -adrenergic antagonist propanolol inhibited the late Ca2+

i response to adrenaline and had a much smaller effect on the early adrenaline response in isolated -cells (II). In support for a -adrenergic mechanism with elevation of cAMP, the late Ca2+

i response to adrenaline was partially or completely inhibited by the protein kinase A inhibitor Rp-cAMPS. The role of cAMP was evident also from the observation that the adenylate cyclase activator forskolin mimicked the effect of adrenaline by inducing slow Ca2+

i oscillations (II). Our data consequently indicated that cAMP is preferentially involved in the late Ca2+

i response to adrenaline, which can be attributed to voltage-dependent influx of Ca2+ in being inhibited by hyperpolarizing diazoxide (II). This con-clusion is consistent with the observation in rat (Gromada et al., 1997) and mouse (Ma et al., 2005) -cells that cAMP amplifies Ca2+

i signaling by increasing the voltage-dependent Ca2+ current. Apart from the involvement in the late Ca2+

i response to adrenaline, cAMP may contribute to the initial mobilization of intracellular Ca2+. Studies of pancreatic -cells have indi-cated that sensitization of the IP3 receptors by protein kinase A-mediated phosphorylation promotes Ca2+ mobilization (Liu et al., 1996; Dyachok and Gylfe, 2004).

Based on in vivo experiments in rats it was suggested that 2 receptor ac-tivation leads to stimulation of glucagon secretion (Saito et al., 1992). Since

2 receptors couple to inhibitory G-proteins and lower cAMP, such stimula-tions may be due to reduced release of inhibitory paracrine factors rather

22

than to a direct effect on -cells. In the present study the 2-agonist clonidine lacked effect on the -cell, and the 2-antagonist yohimbine marginally in-hibited of the Ca2+

i response to adrenaline (II). Therefore, it seems unlikely that 2-adrenoceptors on the -cells are important for stimulation of gluca-gon secretion.

1-Adrenergic activation involves the classical PLC pathway resulting in formation of Ca2+-mobilizing IP3 (García-Sáinz et al., 1999). The participa-tion of 1 adrenoceptors in the action of adrenaline on Ca2+

i was evident from complete inhibition of the initial as well as the late Ca2+

i response by the 1 receptor antagonists phentolamine and prazosin (II). The complete inhibition of the late adrenaline response is intriguing considering the in-volvement of a -adrenergic component in this phase. Interestingly, in vivoexperiments in mice have shown that -blockage with phentolamine inhibits

-adrenergic activation of glucagon secretion (Ahrén and Lundquist, ).Our data indicate that such an inhibition does not necessarily involve paracrine or endocrine mediators, since it is present also in isolated -cells.

Adrenaline stimulation and glucose inhibition of the -cell involve a store-operated current

As discussed above, the adrenaline effect on -cells consisted of an initial mobilization of Ca2+ from intracellular stores attributed to formation of both IP3 and cAMP. The subsequent late Ca2+

i response to adrenaline depended on depolarization, since it was usually prevented by diazoxide and always inhibited by the voltage-dependent Ca2+ channel blocker methoxyverapamil (I). Although cAMP has been found to enhance the Ca2+ current through L-type channels in depolarized rat (Gromada et al., 1997) and mouse (Ma et al., 2005) -cells, such a mechanism cannot account for the depolarization, which opens the channels. The present results indicated that emptying of the Ca2+ stores in the ER is the mechanism causing the depolarization by activat-ing a store-operated depolarizing current. In support for the involvement of a store-operated mechanism leading to depolarization and voltage-dependent Ca2+ influx, the late effect of carbachol, which mobilizes intracellular Ca2+

via muscarinic receptors (Berts et al., 1997), was inhibited by diazoxide (I). Moreover, depletion of the ER by SERCA inhibition depolarized the -cells and caused Ca2+

i oscillations, which were inhibited by diazoxide and meth-oxyverapamil (I). In mouse pancreatic -cells Ca2+-mobilizing agonists acti-vate a store-operated influx of Ca2+, which is blocked by the inhibitor 2-aminoethoxydiphenyl borate (Dyachok and Gylfe, 2001). In accordance with such actions we found that 2-aminoethoxydiphenyl borate had little effect on the initial -cell response to adrenaline but blocked the small sus-tained elevation, which remains when preventing Ca2+ influx through the voltage-dependent channels (I).

23

It is well established that glucose stimulates Ca2+ sequestration in the ER of -cells (Gylfe, 1991; Chow et al., 1995; Tengholm et al., 1999), and that such filling turns off the store-operated entry of Ca2+ (Liu and Gylfe, 1997; Dyachok and Gylfe, 2001). The present study extends previous observation in guinea-pig -cells, that Ca2+ incorporated in response to glucose is mobi-lized with adrenaline (Johansson et al., 1989), by showing that glucose-induced Ca2+ sequestration results in hyperpolarization after shutting off a store-operated current (I). This current is, at least in part, mediated by Ca2+,since omission of the sugar resulted in depolarization and increase of basal Ca2+

i under conditions preventing Ca2+ influx through the L-type channels. Glucose-induced Ca2+ sequestration in the -cell is associated with lowering of [Ca2+]i below the basal level and both the sequestering and lowering ef-fects are maximal at 20 mM of the sugar (Gylfe, 1988; Gylfe, 1991; Tengholm et al., 1999). Such a glucose dependence was confirmed in the present study, showing modest lowering of basal [Ca2+]i in hyperpolarized

-cells when the glucose concentration was raised from 0 to 7 mM and a more pronounced effect after further rise of glucose to 30 mM (IV). The glucose dependence of -cells studied in parallel was different with maximal lowering of [Ca2+]i at 7 mM. Indeed, experiments performed with 0, 3 and 20 mM glucose indicated that both the [Ca2+]i lowering and Ca2+ sequestering effects of glucose were close to maximal at 3 mM of the sugar (I). Consistent with a role for the store-operated mechanism in glucose inhibition of gluca-gon secretion, glucose regulation of Ca2+ sequestration in the -cells was much left-shifted as compared to the -cell and coincided with 0-7 mM range inhibiting secretion (III).

Measurements of glucagon provided additional evidence for a role of a store-operated pathway in glucose inhibition of secretion. SERCA inhibition consequently stimulated glucagon release in the 0-20 mM glucose range and eliminated the inhibitory effect of the sugar (III). These actions were paral-leled by abolishment of the hyperpolarizing and [Ca2+]i lowering effect of glucose on individual -cells. The pronounced effect of SERCA inhibition on glucagon release contrasted dramatically to a modest amplification of glucose-stimulated insulin secretion without significant effect at sub-threshold concentration of the sugar. The insulin data are consistent with little effect of SERCA inhibition on [Ca2+]i and membrane potential in un-stimulated -cells (Chow et al., 1995; Dyachok and Gylfe, 2001). It is only when the -cell is somewhat depolarized that additional depolarization by the store-operated pathway is sufficient to trigger a voltage-dependent rise of [Ca2+]i (Worley et al., 1994) and insulin release (Cruz-Cruz et al., 2005).

Apparently the -cells show an exceptional sensitivity to variation in the store-operated current. Due to the presence of a depolarizing cascade with activation of low threshold T-type Ca2+, voltage-dependent Na+ as well as L-type Ca2+ channels the -cells respond to injection of small currents with electrical activity (Göpel et al., 2000b). The present data indicated that mobi-

24

lization of intracellular Ca2+ by adrenaline induces a store-operated current sufficient for -cell activation. A model for adrenaline stimulation and glu-cose inhibition of glucagon secretion is presented in Fig.3. Adrenaline binds to both 1 and -adrenergic receptors in the -cells. The resulting formation of IP3 and cAMP are involved in the release of Ca2+ from the ER that acti-vates a depolarizing store-operated influx of cations, which eventually trig-gers voltage-dependent Ca2+ influx and glucagon secretion. The role of glu-cose is to activate Ca2+ sequestration in the ER and shut off the stimulatory cascade.

Fig. 3. Model for adrenaline stimulation and glucose inhibition of glucagon secre-tion. Adrenaline (Adr) endoplasmic reticulum (ER), store-operated (SOC) voltage-dependent (VOC) and sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA)

25

KATP channels are functionally active in mouse -cellsStudying mouse -cells exposed to 3 mM glucose we found that only 6 % of previously silent cells responded to tolbutamide-induced closure of KATPchannels with elevation of [Ca2+]i (I). However, the KATP channel activator diazoxide inhibited adrenaline-stimulated [Ca2+]i signalling and this effect was readily reversed by tolbutamide (I). These data were taken to indicate that the KATP channels are functionally inactive in most mouse -cells. Sub-sequent studies in 1 mM glucose showed that 21% of the previously silent

-cells responded with [Ca2+]i elevation to tolbutamide and 8 % to slight depolarization by raising the K+ concentration from 4.8 to 8 mM (III). How-ever, all -cells responded when combining 8 mM K+ with tolbutamide. It therefore seems that the KATP channels are functionally active in -cells but that their closure alone is insufficient to trigger voltage-dependent Ca2+ in-flux. This was an unexpected finding considering the high input resistance of

-cells making the membrane potential sensitive to small current variations (Barg et al., 2000).

As mentioned above depolarization by closure of KATP channels in the -cells has been proposed to result both in inhibition (Göpel et al., 2000b;

Gromada et al., 2004) and stimulation (Bode et al., 1999; Bokvist et al., 1999; Høy et al., 2000; Muñoz et al., 2005; Olsen et al., 2005; Shiota et al., 2005) of glucagon secretion. The most compelling arguments for an inhibi-tory effect was obtained with the observation that depolarization by 10-16 mM K+ or by tolbutamide inhibit glucagon secretion from mouse islets (Gromada et al., 2004). We found inhibitory effects of either 8 mM K+ or 0.5 mM tolbutamide on glucagon secretion from mouse islets (Fig. 4). However, 8 mM K+ or tolbutamide stimulated only isolated mouse -cells as evident from a tendency to depolarise and raise [Ca2+]i (III). Moreover, the elevation of [Ca2+]i, which was always observed in response to 8 mM K+ plus tolbu-tamide (III), was associated with stimulation of glucagon secretion as com-pared to 8 mM K+ or tolbutamide alone (Fig.4). We therefore think that the inhibitory effects of depolarization with tolbutamide or 8 mM K+ are medi-ated by paracrine factors released from the non- -cells within the islets.

A stimulatory role of KATP channel closure in mouse -cells was sup-ported from most results (I, III, IV) including the observation that activation of the KATP channels was associated with lowering of [Ca2+]i in isolated

-cells and inhibition of glucagon secretion at low glucose concentrations (IV). Evidence for a stimulatory role of KATP channel closure is provided also from studies of [Ca2+]i (Olsen et al., 2005), electrical activity (Bokvist et al., 1999), exocytosis (Høy et al., 2000) and secretion (Olsen et al., 2005) of glucagon from the KATP channel-rich rat -cells, and by the observation that tolbutamide raises [Ca2+]i in clonal hamster glucagonoma cell line R1G9 (Bode et al., 1999). Moreover, the most salient feature in mice lacking func-tional KATP channels after knockout of the regulatory sulphonylurea receptor

26

1 subunit is a low glucagon secretion with lacking (Shiota et al., 2005) or diminished (Muñoz et al., 2005) stimulation at low glucose concentrations.

Fig.4 Effects of glucose, tolbutamide and 8 mM K+ on glucagon secretion from isolated mouse pancreatic islets. Glucagon secretion was measured after 60 min incubation. Glucose (1 mM), tolbutamide (0 or 500 M) and K+ (4.8 or 8 mM) were present as indicated. Data are presented as mean values ± SEM of 8 experiments. **P=0.002, ***P<0.001 by paired t-test. Data obtained in collaboration with Albert Salehi from Lund University.

The majority of the isolated mouse -cells did not show any spontaneous activity in 1 mM glucose (III), which is consistent with poor glucagon secre-tion from purified rat -cells exposed to similar concentrations of the sugar (Pipeleers et al., 1985; Olsen et al., 2005). These observations contrast to the stimulated glucagon secretion from mouse (III, IV) and rat (Franklin et al., 2005) islets exposed to 0-1 mM glucose. We therefore speculate that apart from the inhibitory influence from the neighbouring cells in the islets, there is also a stimulatory component. This component may be depolarizing amino acids originating from the islet cells. It is likely that the interstitial space

40

20

0Glucose (mM)

K+ (mM)

Tolbutamide ( M)

1 1 1 1

0 0500 500

8 84.8 4.8

Glu

cago

n se

cret

ion

pg/(

isle

t · h

) *** ***

*** ***

27

within the islets contains amino acids during static incubation. Considerable amounts of free amino acids are lost from islets during perfusion with nutri-ent-free medium but not during static incubation (Gylfe, 1974).

Under physiological conditions the -cells are exposed to a depolarizing mixture of amino acids. We have found that isolated -cells exposed to RPMI 1640 medium containing 1 mM glucose show spontaneous [Ca2+]ioscillations, which are inhibited by glucose elevation (Vieira and Gylfe, 2004). Consistent with these observations RPMI 1640 medium containing 1 mM glucose doubled glucagon secretion from mouse islets as compared to amino acid-free medium with the same sugar concentration (Fig. 5). More-over, 20 mM glucose readily inhibited glucagon secretion in the RPMI 1640 medium. Instead of using a depolarizing mixture of amino acids to activate [Ca2+]i signalling in single -cells, the -cells were depolarized in a more controlled manner by exposure to 8 mM K+ plus 500 M tolbutamide (III). With presence of tolbutamide it became possible to study KATP channel-independent effects of glucose on individual -cells.

50

0

-Conotoxin ( M)

Glu

cag

on

Sec

reti

on

pg

/isle

ts/h

Nifedipine ( M)

***

RPMI 1640

******

Glucose (mM)- + + + +

11 1 1 200 0 0.1 0 00 0 0 10 0

Fig. 5 Effects of RPMI 1640 medium, glucose concentration, -conotoxin and nifedipine on glucagon secretion from isolated mouse islets. Glucagon secretion was measured in a balanced salt solution containing 1mM glucose or in RPMI medium with 1 mM or 20 mM glucose. The effects of -conotoxin (0.1 M) and nifedipine (10 M) were tested in the RPMI medium containing 1mM glucose. Data are pre-sented as mean values ± S.E.M. of 12 experiments. *P ‹ 0,001. Data obtained in collaboration with Dr. Albert Salehi, Lund University, Sweden

28

The elevation of [Ca2+]i underlying glucagon secretion from mouse -cells is due to opening of L-type Ca2+

channelsThe elevation of [Ca2+]i causing glucagon secretion from the depolarized

-cell has generally been attributed to opening of L-type Ca2+ channels (Rorsman and Hellman, 1988; Berts et al., 1996b; Barg et al., 2000; Göpel et al., 2000b; Muñoz et al., 2005). However, it was recently proposed that stimulation of glucagon release in 1 mM glucose is due to activation of N- rather than L-type Ca2+ channels but that the L-type channels become domi-nating after increase of cAMP (Gromada et al., 1997; Gromada et al., 2004). In mouse -cells the N-type Ca2+ channels have been proposed to mediate tonic glucagon release and the L-type channels cAMP-dependent secretion (Ma et al., 2005). We found that [Ca2+]i signalling induced by 8 mM K+ plus tolbutamide was always inhibited by the L-type Ca2+ channel blocker nifedipine but never by the N-type channel blocker -conotoxin (III). Like-wise [Ca2+]i signalling in -cells exposed to RPMI 1640 medium containing 1 mM glucose was always inhibited by nifedipine but rarely by -conotoxin (Vieira and Gylfe, 2004). Fig. 4 shows that glucagon secretion in RPMI 1640 medium with 1 mM glucose was unaffected by -conotoxin but abol-ished by nifedipine. Our data indicate that L-type Ca2+ channels are most important for stimulated glucagon secretion under physiological conditions.

Glucose inhibition of glucagon secretion in mouse islets occurs independently of KATP channels and products secreted from pancreatic -cells It is well established that inhibition of glucagon secretion is more sensitive to the glucose concentration than is stimulation of insulin release (Gerich et al., 1974). Our data indicated that maximal inhibition of glucagon secretion from mouse islets occurs at the glucose threshold for stimulation of insulin release (III, IV). Although this finding does not support a role of inhibitory

-cell factors (Gylfe, 1990), it cannot be excluded that such factors contrib-ute to glucose inhibition of glucagon release, when -cell secretion is stimu-lated (Östenson, 1979; Rorsman et al., 1989). It is also possible that inhibi-tion of glucagon secretion results from stimulated release of somatostatin (Cejvan et al., 2003) or from a direct effect of glucose on the pancreatic

-cell (Pipeleers et al., 1985; Unger, 1985; Johansson et al., 1987; Bode et al., 1999). However, wortmannin, which prevents the inhibitory effect of insulin (Ravier and Rutter, 2005), and PTX pretreatment, which blocks the inhibitory action of somatostatin (Göpel et al., 2004) did not prevent glucose reduction of [Ca2+]i in the individual -cells (III). GABA, which may act as a

29

paracrine inhibitor of glucagon secretion in the guinea-pig (Rorsman et al., 1989) and rat (Wendt et al., 2004), but has little effect on glucagon secretion from mouse islets (Gilon et al., 1991), failed to affect [Ca2+]i and membrane potential in the mouse -cells (III). Zn2+, which is co-secreted with insulin and inhibits glucagon secretion from rat islets (Ishihara et al., 2003; Franklin et al., 2005), was recently found to stimulate mouse -cells (Ravier and Rut-ter, 2005). In accordance with the latter observation we found that Zn2+ de-polarized all -cells and activated Ca2+ signaling (III). Against this back-ground it seems unlikely that the actions of glucose on [Ca2+]i and membrane potential in individual -cells should be mediated by paracrine factors re-leased from non- -cells attached to the cover slip.

Closure of KATP channels in -cells have been proposed to couple and in-crease of glucose concentration to inhibit glucagon secretion (Göpel et al., 2000b; Gromada et al., 2004). This scenario predicts that closure of the KATPchannels depolarizes the -cells and lowers [Ca2+]i. We instead determined that depolarization obtained with tolbutamide was associated with increase of [Ca2+]i in about 21% of previously silent mouse -cells exposed to 1 mM glucose and 80 % of the -cells with spontaneous [Ca2+]i oscillations. More-over tolbutamide always activated -cells exposed to 8 mM K+. The depo-larization and [Ca2+]i signalling in -cells exposed to 8 mM K+ plus tolbu-tamide were counteracted by glucose concentrations ranging from 1-20 mM, indicating that these actions of the sugar occurred independently of KATPchannel closure (III). Contrary to a glucose-induced depolarization predicted by the KATP channel dependent model for inhibition of glucagon release (Göpel et al., 2000b; Gromada et al., 2004), the sugar always hyperpolarized mouse -cells (I, III). The dose-response relationship for glucose-inhibited glucagon secretion from mouse islets showed maximal effect at 7-8 mM glucose both under control conditions and in the presence of 8 mM K+ plus tolbutamide, further supporting a KATP channel-independent mechanism (III). Aditional evidence that glucose inhibits glucagon secretion independ-ently of KATP channels and products secreted from -cells was obtained with the observation that some inhibition of glucagon secretion from mouse islets remained in the presence of the KATP channel opener diazoxide. Under these conditions all types of islets cells are hyperpolarized, which should minimize the release of paracrine factors (IV). Consistent with a direct inhibitory ef-fect of glucose on -cells, glucose inhibited glucagon secretion from hamster R1G9 glucagonoma cells (IV).

30

High glucose concentrations paradoxically stimulate glucagon release from mouse islets and clonal -cellsA remarkable observation in the present study was that the dose-response relationship for glucose-regulated glucagon secretion from mouse islets is U-shaped with inhibition followed by stimulation. Increase of the glucose con-centration from 0 to 4 mM inhibited glucagon secretion and the maximal effect was obtained at 7-8 mM. The inhibitory effect then gradually de-creased and above 20 mM inhibition was dramatically reversed into a small stimulation at 25-30 mM. There was an U-shaped dose-response relationship for glucose-regulated glucagon secretion also in clonal hamster R1G9 gluca-gonoma cells with maximal inhibition at 7 mM. In this case the stimulatory component was left-shifted with significant stimulation already at 12 mM glucose (IV). The presence of an U-shaped dose-response relationship for glucose-regulated glucagon secretion in hamster glucagonoma cells is a strong argument that glucose both inhibits and stimulates glucagon secretion by direct effects on the -cells.

To clarify the role of Ca2+ in the signal transduction underlying glucose-regulated glucagon secretion, [Ca2+]i was measured in individual mouse

-cells. -Cells with spontaneous [Ca2+]i oscillations in 1 mM glucose re-acted with lowering of [Ca2+]i after introduction of 20 mM of the sugar. This observation implies that the inhibitory part of the dose-response relationship for glucose-regulated glucagon secretion is due to lowering of [Ca2+]i. At 30 mM glucose there was only a lowering of [Ca2+]i, indicating that the stimula-tory effect of high sugar concentrations cannot be explained by rise of [Ca2+]i. These conclusions were supported by measurements of glucagon secretion from mouse islets and [Ca2+]i in mouse -cells during hyperpolari-zation with diazoxide (IV). Hyperpolarization inhibited glucagon secretion at 0-1 mM glucose in parallel with lowering of [Ca2+]i. However, hyperpolari-zation did not prevent high concentrations of glucose to stimulate glucagon release although the sugar lowered [Ca2+]i to below the basal levels (IV).

A recent observation that glucose stimulates glucagon secretion from pu-rified rat -cells has been taken to indicate that secretion at the level of the

-cell is regulated in a -cell-like manner, and that glucose inhibition re-quires release of paracrine factors from non- -cells within the islets (Olsen et al., 2005). The stimulation of glucagon secretion from mouse islet is prin-cipally different from that in purified rat -cells. Whereas the latter is attrib-uted to a -cell-like pathway with closure of KATP channels leading to depo-larization and voltage-dependent Ca2+ influx, glucose stimulation of gluca-gon secretion from mouse islets occurred independent of KATP channels and did not require elevation of [Ca2+]i (IV). A reason for the different [Ca2+]iresponses is apparently that rat -cells have a high density of KATP channels, even exceeding that in rat -cells (Bokvist et al., 1999), whereas mouse

-cells show little KATP channel expression (Quesada et al., 1999; Göpel et

31

al., 2000b) corresponding to only 2% of that in mouse -cells (Barg et al., 2000). The lack of accompanying [Ca2+]i elevation in glucose-stimulated glucagon secretion from mouse islets does not imply that secretion is Ca2+-insensitive under these conditions. Clamping the membrane potential of mouse -cells at a depolarized level by combining diazoxide with 30 mM K+

induced a steady elevation of [Ca2+]i, which was not further affected by the glucose concentration. The rise of [Ca2+]i markedly stimulated glucagon release at all glucose concentrations, but high concentrations of the sugar still had an additional stimulatory effect (IV). This phenomenon is reminis-cent of the amplifying effect of glucose on insulin secretion (Gembal et al., 1992), and a similar amplifying effect has recently been found in rat -cells (Olsen et al., 2005). However, it is striking that the stimulatory effect of high glucose concentrations on glucagon secretion did not require elevation of [Ca2+]i. In pancreatic -cells there is evidence that increased metabolism can to some extent stimulate insulin release independent of any [Ca2+]i elevation (Westerlund et al., 1996; Ravier et al., 1999). A similar mechanism may be more pronounced in the -cells. The signal transduction underlying this in-triguing phenomenon remains to be elucidated.

Hypersecretion of glucagon is an important feature in diabetes contribut-ing to glucose dysregulation (Gerich et al., 1976; Jiang and Zhang, 2003). The mechanisms underlying this phenomenon are unclear. The present study of mouse islets and hamster glucagonoma cells indicate that paradoxical hypersecretion of glucagon during hyperglycemia may depend on an U-shaped dose-response-relationship for the glucose regulation of secretion. If present in humans, the U-shaped dose response relationship for glucose-regulated glucagon secretion may explain hyperglucagonemia in diabetes.

32

Conclusions

1. KATP channels are functionally active in mouse -cells.

2. A calcium store-operated mechanism can explain adrenaline stimulation and glucose inhibition of glucagon secretion in mouse

-cells

3. Adrenaline stimulation of glucagon secretion involves both 1and -adrenoceptor mechanisms with mobilization of intracellular Ca2+ as well as depolarization and voltage-dependent Ca2+ influx through L-type channels.

4. Glucose inhibition of glucagon secretion occurs independently of KATP channels and products secreted from pancreatic -cells.

5. Glucose paradoxically stimulates glucagon secretion at high con-centrations independently of [Ca2+]i

33

Svensk sammanfattning

Brist på det blodsockersänkande hormonet insulin kännetecknar de flesta former av sockersjuka (diabetes). Förhöjda nivåer av socker (glukos) är en viktig orsak till diabeteskomplikationer som fotsår, njurskada och blindhet. Vid diabetes förvärras den ökade glukoskoncentrationen av förhöjda nivåer av det blodsockerhöjande hormonet glukagon. Glukagons viktigaste funktion är att frisätta glukos från levern vid låga blodsockerhalter. Även denna funk-tion är störd hos diabetiker. Eftersom diabetiker försöker hålla blodsockerni-våerna på en låg nivå riskerar de att överdosera insulin varvid blodsockret kan sjunka alltför mycket. Detta är ett farligt tillstånd som ibland orsakar dödsfall. Om glukagon frisattes på ett normalt sätt skulle diabetiker vara mindre känsliga för överdosering av insulin. Studier av glukagonfrisättning-ens reglering kan därför förväntas leda till förbättrad behandling av diabetes.

Insulin frisätts från -celler och glukagon från -celler, som båda finns i bukspottkörtelns Langerhanska öar. I avhandlingen studerades såväl gluka-gon- och insulinfrisättning från musens Langerhanska öar som glukagonfri-sättning från en -cellinje från hamster. Dessutom mättes mus- -cellernas membranpotential och koncentrationen av kalcium (Ca2+) i cellernas cyto-plasma. Ca2+ är en viktig faktor, som styr glukagonfrisättningen, och mem-branpotentialen påverkar i sin tur Ca2+-koncentrationen. Glukos befanns hämma glukagonsekretionen från musöar med maximal effekt vid 7 mM. Eftersom denna glukoskoncentration obetydligt stimulerar insulinsekretionen är det inte troligt att frisättning av faktorer från -cellerna förklarar den hämmade glukagonsekretionen. Då glukos också hämmande glukagonfri-sättningen från -cellslinjen är det istället sannolikt att sockret även har en direkt hämmande effekt på normala -celler. En oväntad upptäckt var att höga glukoskoncentrationer hade en paradoxalt stimulerande effekt på glu-kagonfrisättningen. Medan glukos hämmande effekt kunde förklaras av en Ca2+-sänkning så skedde stimuleringen oberoende av Ca2+.

Både - och -celler har en speciell typ av K+ kanaler, som är känsliga för ATP. Dessa kanaler, som har stor betydelse för glukosstimulerad insulin-frisättning har även föreslagits vara viktiga för glukoshämmad glukagonsek-retion. Avhandlingen visar att glukos hämmande effekt på glukagonfrisätt-ningen kvarstår även då de ATP-känsliga kanalerna inaktiveras. Adrenalin, som är en kraftig stimulator av glukagonfrisättningen, ökade Ca2+ i cytop-lasman både genom frisättning från det endoplasmatiska nätverket (EN) och genom inflöde över cellmembranet. Medan den Ca2+-frisättande effekten

34

kunde tillskrivas bildning av inositol 1,4,5-trisfosfat och cykliskt AMP, akti-verades Ca2+-inflödet genom en minskning av membranpotentialen via akti-vering av specifika katjonkanaler. Dessa katjonkanaler, som är beroende av Ca2+-innehållet i EN, inaktiveras via glukosstimulerat upptag av Ca2+ i denna organell. EN har sålunda en nyckelroll i frisättningen av glukagon genom att via sitt Ca2+-innehåll påverka -cellernas membranpotential. Upptäckten att höga glukoskoncentrationer också kan initiera en Ca2+-oberoende stimule-ring av glukagonsekretionen kan förklara varför patienter med diabetes ofta har en skadlig ökning av detta hormon.

35

Acknowledgements

I would like to give my thanks to several people that have helped me to complete this thesis. First of all, I would like especially to thank my supervi-sor, Professor Erik Gylfe, who provided me with the opportunity to pursue PhD studies within his research group, for his continuous help with research, his professional guidance in my studies and his friendship. Docent Eva Grapengiesser, my co-supervisor and Professor Bo Hellman have contrib-uted with inspiring suggestions and interesting discussions.

During theses years I have had contact with many people that helped me with unconditional advice, company, lively discussions regarding research and teaching. Thanks to Peter Bergsten, Tea Sundsten, Henrik Ortsäter, E-ri Sol, Hanna Nyblom, Kristofer Thörn, Jenny Sågetorp, Olof Idevall, Anders Tengholm, Oleg Dyachok, Andreea Barbu, Vitek Kriz, Göran Mattsson, Sophia Thore and Ludwina Redtzer.

Technical support has been very important from the beginning till the end of my thesis work. A very important person is responsible for that: Helen Dansk. Thanks for your help in the laboratory and as well for being my “Swedish mother” during this period.

The secretarial assistance was also very valuable for my studies; I would like to thank Karin Öberg, Agneta Bäfwe, GunBritt Lind and Birgitta Jön-zén. Thanks to all members of the Department of Medical Cell Biology. Some friends did not participate directly of my studies but were essential for my life: Nanna-Elisa Lönnholm, Lais Malavasi, Andres Martineli, Katarina Kindahl and Ilda Hansson.

I want to express my gratitude to the commission of the Postgraduate Training Program Islet Research European Network (IREN) for the opportu-nity to start a training program in Sweden. Thanks for my previous supervi-sor Antonio Carlos Boschero and previous co-supervisor Everardo Magal-haes Carneiro for putting up with my wish to continue my studies in Swe-den.

Thanks are due to my family and friends in Brazil, especially to my mother Ana Marilia Pereira da Silva who have always been there for me in spite of the geographical distance between us.

Finally, I would like to thank the most important person in my life, my beloved Pablo Giambiagi.

The studies were supported by grants from the Swedish Medical Research Council (12X-6240), the Swedish Foundation for Strategic Research, the

36

Swedish Foundation for International Cooperation in Research and Higher Education, the Wenner-Gren Center Foundation, the Swedish Diabetes As-sociation, Novo-Nordisk Foundation, Family Ernfors foundation and the Swedish Society for Medical Research.

37

References

Ahrén, B. and Lundquist, I. ( ). -Adrenoceptor blockade by phentolamine inhibits -adrenergically and cholinergically induced glucagon secretion in the mouse. Hormone and Metabolic Research 19, 600-603.

Ashcroft, F. M. and Rorsman, P. (1989). Electrophysiology of the pancreatic -cell. Progress in Biophysics and Molecular Biology 54, 87-143.

Barg, S., Galvanovskis, J., Göpel, S. O., Rorsman, P. and Eliasson, L. (2000). Tight coupling between electrical activity and exocytosis in mouse glucagon-secreting -cells. Diabetes 49, 1500-1510.

Bergsten, P., Grapengiesser, E., Gylfe, E., Tengholm, A. and Hellman, B.(1994). Synchronous oscillations of cytoplasmic Ca2+ and insulin release in glu-cose-stimulated pancreatic islets. Journal of Biological Chemistry 269, 8749-8753.

Berridge, M. J. (1990). Calcium oscillations. Journal of Biological Chemistry 265,9583-9586.

Bertram, R., Smolen, P., Sherman, A., Mears, D., Atwater, I., Martin, F. and Soria, B. (1995). A role for calcium release-activated current (CRAC) in cho-linergic modulation of electrical activity in pancreatic -cells. Biophysical Jour-nal 68, 2323-2332.

Berts, A., Ball, A., Dryselius, S., Gylfe, E. and Hellman, B. (1996a). Glucose stimulation of somatostatin-producing islet cells involves oscillatory Ca2+ sig-nalling. Endocrinology 137, 693-697.

Berts, A., Ball, A., Gylfe, E. and Hellman, B. (1996b). Suppression of Ca2+ oscil-lations in glucagon-producing 2-cells by insulin/glucagon and amino acids. Biochimica et Biophysica Acta 1310, 212-216.

Berts, A., Gylfe, E. and Hellman, B. (1995). Ca2+ oscillations in pancreatic islet cells secreting glucagon and somatostatin. Biochemical and Biophysical Re-search Communications 208, 644-649.

Berts, A., Gylfe, E. and Hellman, B. (1997). Cytoplasmic Ca2+ in glucagon-producing pancreatic -cells exposed to carbachol and agents affecting Na+

fluxes. Endocrine 6, 79-83. Bode, H. P., Weber, S., Fehmann, H. C. and Göke, B. (1999). A nutrient-

regulated cytosolic calcium oscillator in endocrine pancreatic glucagon-secreting cells. Pflügers Archiv. European Journal of Physiology 437, 324-334.

Bokvist, K., Olsen, H. L., Høy, M., Gotfredsen, C. F., Holmes, W. F., Buschard, K., Rorsman, P. and Gromada, J. (1999). Characterisation of sulphonylurea and ATP-regulated K+ channels in rat pancreatic A-cells. Pflügers Archiv. European Journal of Physiology 438, 428-436.

Braun, M., Wendt, A., Birnir, B., Broman, J., Eliasson, L., Galvanovskis, J., Gromada, J., Mulder, H. and Rorsman, P. (2004). Regulated exocytosis of GABA-containing synaptic-like microvesicles in pancreatic -cells. Journal of General Physiology 123, 191-204.

38

Cejvan, K., Coy, D. H. and Efendic, S. (2003). Intra-islet somatostatin regulates glucagon release via type 2 somatostatin receptors in rats. Diabetes 52, 1176-1181.

Chow, R. H., Lund, P. E., Löser, S., Panten, U. and Gylfe, E. (1995). Coinci-dence of early glucose-induced depolarization with lowering of cytoplasmic Ca2+ in mouse pancreatic -cells. Journal of Physiology 485, 607-617.

Cruz-Cruz, R., Salgado, A., Sanchez-Soto, C., Vaca, L. and Hiriart, M. (2005). A Thapsigargin-sensitive cationic current leads to membrane depolarization, calcium entry and insulin secretion in rat pancreatic ß-cells. American Journal of Physiology 289, E439-445.

Cryer, P. E. (2002). Hypoglycaemia: the limiting factor in the glycaemic manage-ment of Type I and Type II diabetes. Diabetes 45, 937-948.

Diao, J., Zeenat, A., Chan, C. B. and Wheeler, M. B. (2005). Glucose-regulated glucagon secretion requires insulin receptor expression in pancreatic alpha-cells. Journal of Biological Chemistry 280, 33487-33496.

Draber, P. and Draberova, L. (2005). Lifting the fog in store-operated Ca2+ entry. Trends in Immunology 26, 621-624.

Dyachok, O. and Gylfe, E. (2001). Store-operated influx of Ca2+ in the pancreatic beta cells exhibits graded dependence on the filling of the endoplasmic reticu-lum. Journal of Cell Science 114, 2179-2186.

Dyachok, O. and Gylfe, E. (2004). Ca2+-induced Ca2+ release via inositol 1,4,5-trisphosphate receptors is amplified by protein kinase A and triggers exocytosis in pancreatic -cells. Journal of Biological Chemistry 29, 45455-45461.

Epps, D. E., Wolfe, M. L. and Groppi, V. (1994). Characterization of the steady-state and dynamic fluorescence properties of the potential-sensitive dye bis-(1,3-dibutylbarbituric acid)trimethine oxonol (Dibac4(3)) in model systems and cells. Chemistry and Physics of Lipids 69, 137-150.

Evans, M. L., McCrimmon, R. J., Flanagan, D. E., Keshavarz, T., Fan, X., McNay, E. C., Jacob, R. J. and Sherwin, R. S. (2004). Hypothalamic ATP-sensitive K+ channels play a key role in sensing hypoglycemia and triggering counterregulatory epinephrine and glucagon responses. Diabetes 53, 2542-2551.

Filipponi, P., Gregorio, F., Ferrandina, C., Nicoletti, I., Mannarelli, C., Pippi, R. and Santeusanio, F. (1986). Alpha-adrenergic system in the modultaion of pancreatic A and B cell function in normal rats. Diabetes Research and Clinical Practice 2, 325-336.

Franklin, I., Gromada, J., Gjinovci, A., Theander, S. and Wollheim, C. B.(2005). -cell secretory products activate -cell ATP-dependent potassium channels to inhibit glucagon release. Diabetes 54, 1808-1815.

García-Sáinz, J. A., Vázquez-Prado, J. and Villalobos-Molina, R. (1999). 1-Adrenoceptors: subtypes, signaling, and roles in health and disease. Archives of Medical Research 30, 449-458.

Gembal, M., Gilon, P. and Henquin, J. C. (1992). Evidence that glucose can con-trol insulin release independently from its action on ATP-sensitive K+ channels in mouse B cells. Journal of Clinical Investigation 89, 1288-1295.

Gerich, J. E., Charles, A. and Grodsky, G. M. (1974). Characterization of the effects of arginine and glucose on glucagon and insulin release from the per-fused rat pancreas. Journal of Clinical Investigation 54, 833-841.

Gerich, J. E., Charles, A. and Grodsky, G. M. (1976). Regulation of pancreatic insulin and glucagon secretion. Annual Review of Physiology 38, 353-388.

39

Gilon, P., Arredouani, A., Gailly, P., Gromada, J. and Henquin, J. C. (1999). Uptake and release of Ca2+ by the endoplasmic reticulum contribute to the os-cillations of the cytosolic Ca2+ concentration triggered by Ca2+ influx in the electrically excitable pancreatic B-cell. Journal of Biological Chemistry 274,20197-20205.

Gilon, P., Bertrand, G., Loubatières-Mariani, M. M., Remacle, C. and Hen-quin, J. C. (1991). The influence of -aminobutyric acid on hormone release by the mouse and rat endocrine pancreas. Endocrinology 129, 2521-2529.

Gromada, J., Bokvist, K., Ding, W. G., Barg, S., Buschard, K., Renström, E. and Rorsman, P. (1997). Adrenaline stimulates glucagon secretion in pancre-atic A-cells by increasing the Ca2+ current and the number of granules close to the L-type Ca2+ channels. Journal of General Physiology 110, 217-228.

Gromada, J., Høy, M., Buschard, K., Salehi, A. and Rorsman, P. (2001). Soma-tostatin inhibits exocytosis in rat pancreatic -cells by Gi2-dependent activation of calcineurin and depriming of secretory granules. Journal of Physiology 535,519-532.

Gromada, J., Ma, X., Hoy, M., Bokvist, K., Salehi, A., Berggren, P. O. and Rorsman, P. (2004). ATP-sensitive K+ channel-dependent regulation of gluca-gon release and electrical activity by glucose in wild-type and SUR1-/- mouse

-cells. Diabetes 53 (Supplement 3), S181-S189. Gylfe, E. (1974). Changes of free amino acids in pancreatic -cells after starvation

and substrate deprivation. Acta Endocrinologica 75, 105-118. Gylfe, E. (1988). Glucose-induced early changes in cytoplasmic calcium of pancre-

atic -cells studied with time-sharing dual-wavelength fluorometry. Journal of Biological Chemistry 263, 5044-5048.

Gylfe, E. (1990). How secretion is inhibited. Nature 344, 300. Gylfe, E. (1991). Carbachol induces sustained glucose-dependent oscillations of

cytoplasmic Ca2+ in hyperpolarized pancreatic -cells. Pflügers Archiv. Euro-pean Journal of Physiology 419, 639-643.

Göpel, S. O., Kanno, T., Barg, S. and Rorsman, P. (2000a). Patch-clamp charac-terisation of somatostatin-secreting -cells in intact mouse pancreatic islets. Journal of Physiology 528.

Göpel, S. O., Kanno, T., Barg, S., Weng, X. G., Gromada, J. and Rorsman, P.(2000b). Regulation of glucagon secretion in mouse -cells by KATP channels and inactivation of TTX-sensitive Na+ channels. Journal of Physiology 528,509-520.

Göpel, S. O., Zhang, Q., Eliasson, L., Ma, X., Galvanovskis, J., Kanno, T., Salehi, A. and Rorsman, P. (2004). Capacitance measurements of exocytosis in mouse pancreatic -, - and -cells within intact islets of Langerhans. Jour-nal of Physiology 556, 783-799.

Hardie, R. C. and Minke, B. (1993). Novel Ca2+ channels underlying transduction in Drosophila photoreceptors: implications for phosphoinositide-mediated Ca2+ mobilization. Trends Neuroscience 16, 371-376.

Høy, M., Olsen, H. L., Bokvist, K., Buschard, K., Barg, S., Rorsman, P. and Gromada, J. (2000). Tolbutamide stimulates exocytosis of glucagon by inhibi-tion of a mitochondrial-like ATP-sensitive K+ (KATP) conductance in rat pancre-atic A-cells. Journal of Physiology 15, 109-120.

Irvine, R. F. (1990). 'Quantal' Ca2+ release and the control of Ca2+ entry by inositol phosphates--a possible mechanism. FEBS Letters 263, 5-9.

Ishihara, H., Maechler, P., Gjinovci, A., Herrera, P. L. and Wollheim, C. B.(2003). Islet -cell secretion determines glucagon release from neighbouring -cells. Nature Cell Biology 5, 330-335.

40

Jiang, G. and Zhang, B. B. (2003). Glucagon and regulation of glucose metabo-lism. American Journal of Physiology Endocrinology and Metabolism 284,E671-E678.

Johansson, H., Gylfe, E. and Hellman, B. (1987). The actions of arginine and glucose on glucagon secretion are mediated by opposite effects on cytoplasmic Ca2+. Biochemical and Biophysical Research Communications 147, 309-314.

Johansson, H., Gylfe, E. and Hellman, B. (1989). Cyclic AMP raises cytoplasmic calcium in pancreatic 2- cells by mobilizing calcium incorporated in response to glucose. Cell Calcium 10, 205-211.

Lacey, R. J., Berrow, N. S., Scarpello, J. H. and Morgan, N. G. (1991). Selective stimulation of glucagon secretion by beta 2-adrenoceptors in isolated islets of Langerhans of the rat. British Journal of Pharmacology 103, 1824-1828.

Liou, J., Kim, M. L., Heo, W. D., Jones, J. T., Myers, J. W., Ferrell, J. E. and Meyer, T. (2005). STIM is a Ca2+ sensor essential for Ca2+-store-depletion-triggered Ca2+ influx. Current Biology 15, 1235-1241.

Liu, Y. J., Grapengiesser, E., Gylfe, E. and Hellman, B. (1996). Crosstalk be-tween the cAMP and inositol trisphosphate signalling pathways in pancreatic -cells. Archives of Biochemistry and Biophysics 334, 295-302.

Liu, Y. J. and Gylfe, E. (1997). Store-operated Ca2+ entry in insulin-releasing pan-creatic -cells. Cell Calcium 22, 277-286.

Liu, Y. J., Hellman, B. and Gylfe, E. (1999). Ca2+ signaling in mouse pancreatic polypeptide cells. Endocrinology 140, 5524-5529.

Liu, Y. J., Vieira, E. and Gylfe, E. (2004). A store-operated mechanism determines the activity of the electrically excitable glucagon-secreting pancreatic -cell. In Cell Calcium, vol. 35, pp. 357-365.

Ma, X., Vikman, J. and Eliasson, L. (2005). Stimulation of glucagon release by cAMP in mouse pancreatic A-cells depends on calcium influx through L-type Ca2+ channels. Diabetologia 46, A166.

Miki, T., Liss, B., Minami, K., Shiuchi, T., Saraya, A., Kashima, Y., Horiuchi, M., Ashcroft, F. M., Minokoshi, Y., Roeper, J. et al. (2001). ATP-sensitive K+ channels in the hypothalamus are essential for the maintenance of glucose homeostasis. Nature Neuroscience 4, 507-512.

Mitrakou, A., Kelley, D., Veneman, T., Jenssen, T., Pangburn, T., Reilly, J. and Gerich, J. E. (1990). Contribution of abnormal muscle and liver glucose me-tabolism to postprandial hyperglycemia in NIDDM. Diabetes 39, 1381-1390.

Moffatt, J. D. and Cocks, T. M. (2004). Coactivation of capacitative calcium entry and L-type calcium channels in guinea pig gallbladder. American Journal Physiol Gastrointest Liver Physiol 286, 1090-10100.

Muñoz, A., Hussain, K., Bryan, J., Aguilar-Bryan, L. and Rajan, A. S. (2005). Regulation of glucagon secretion at low glucose concentrations: evidence for adenosine triphosphate-sensitive potassium channel involvement. Endocrinol-ogy 146, 5514-5521.

Nakaki, T., Nakadate, T. and Kato, R. (1980). 2-adrenoceptors modulating insu-lin release from isolated pancreatic islets. Naunyn-Schmiedeberg's Archives of Pharmacology 313, 151-153.

Ohneda, A., Watanabe, K. and Horigome, K. (1978). Abnormal response of pan-creatic glucagon to glycemic changes in diabetes mellitus. Journal of Clinical Endocrinology and Metabolism 46, 504-510.

Olsen, H. L., Theander, S., Bokvist, K., Buschard, K., Wollheim, C. B. and Gromada, J. (2005). Glucose stimulates glucagon release in single rat -cellsby mechanisms that mirror the stimulus-secretion coupling in -cells. Endocri-nology 146, 4861-4870.

41

Östenson, C. G. (1979). Regulation of glucagon release: effects of insulin on the pancreatic A2-cell of the guinea pig. Diabetologia 17, 325-330.

Panagiotidis, G., Salehi, A., Westermark, P. and Lundquist, I. (1992). Homolo-gous islet amyloid polypeptide: effects on plasma levels of glucagon, insulin and glucose in the mouse. Diabetes research and clinical practice 18, 167-171.

Parekh, A. B. and Putney, J. W., Jr. (2005). Store-operated calcium channels. Physiological Reviews 85, 757-810.

Pipeleers, D. G., Schuit, F. C., Van Schravendijk, C. F. H. and Van de Winkel, M. (1985). Interplay of nutrients and hormones in the regulation of glucagon re-lease. Endocrinology 117, 817-823.

Putney, J. W., Jr. (1986). A model for receptor-regulated calcium entry. Cell Cal-cium 7, 1-12.

Putney, J. W., Jr. (1990). Capacitative calcium entry revisited. Cell Calcium 11,611-624.

Quesada, I., Nadal, A. and Soria, B. (1999). Different effects of tolbutamide and diazoxide in -, -, and -cells within intact islets of Langerhans. Diabetes 48,2390-2397.

Randriamampita, C. and Tsien, R. Y. (1993). Emptying of intracellular Ca2+

stores releases a novel small messenger that stimulates Ca2+ influx. Nature 364,809-814.

Ravier, M. A., Gilon, P. and Henquin, J. C. (1999). Oscillations of insulin secre-tion can be triggered by imposed oscillations of cytoplasmic Ca2+ or metabo-lism in normal mouse islets. Diabetes 48, 2374-2382.

Ravier, M. A. and Rutter, G. A. (2005). Glucose or insulin, but not zinc ions, in-hibit glucagon secretion from mouse pancreatic -cells. Diabetes 54, 1789-1797.

Roos, J., DiGregorio, P. J., Yeromin, A. V., Ohlsen, K., Lioudyno, M., Zhang, S., Safrina, O., Kozak, J. A., Wagner, S. L., Cahalan, M. et al. (2005). STIM1, an essential and conserved component of store-operated Ca2+ channel function. Journal of Cell Biology 169, 435-445.

Rorsman, P. (1988). Two types of Ca2+ currents with different sensitivities to or-ganic Ca2+ channel antagonists in guinea pig pancreatic 2-cells. Journal of General Physiology 91, 243-254.

Rorsman, P., Berggren, P. O., Bokvist, K., Ericson, H., Möhler, H., Östenson, C. G. and Smith, P. A. (1989). Glucose-inhibition of glucagon secretion in-volves activation of GABAA-receptor chloride channels. Nature 341, 233-236.

Rorsman, P. and Hellman, B. (1988). Voltage-activated currents in guinea pig pancreatic 2 cells. Evidence for Ca2+-dependent action potentials. Journal of General Physiology 91, 223-242.

Saito, M., Saitoh, T. and Inoue, S. (1992). Alpha 2-adrenergic modulation of pan-creatic glucagon secretion in rats. Physiology and Behavior 51, 1165-1171.

Schuit, F. C. and Pipeleers, D. G. (1986). Difference in adrenergic recognition by pancreatic A and B cells. Science 232, 875-877.

Shiota, C., Rocheleau, J. V., Shiota, M., Piston, D. and Magnuson, M. A. (2005). Impaired glucagon secretory responses in mice lacking the type 1 sulfonylurea receptor. American Journal of Physiology 289, E570-E577.

Skoglund, G., Lundquist, I. and Ahrén, B. (1987). Alpha 1- and alpha 2-adrenoceptor activation increases plasma glucagon levels in the mouse. Euro-pean Journal of Pharmacology 143, 83-88.

Tengholm, A., Hellman, B. and Gylfe, E. (1999). Glucose regulation of free Ca2+

in the endoplasmatic reticulum of mouse pancreatic beta cells. Journal of Bio-logical Chemistry 274, 36883-36890.

42

Tengholm, A., Hellman, B. and Gylfe, E. (2001). The endoplasmic reticulum is a glucose-modulated high affinity sink for Ca2+ in mouse pancreatic -cells. Jour-nal of Physiology 530, 533-540.

Thore, S., Dyachok, O., Gylfe, E. and Tengholm, A. (2005). Feedback activation of phospholipase C via intracellular mobilization and store-operated influx of Ca2+ in insulin-secreting -cells. Journal of Cell Science 118, 4463-4471.

Unger, R. H. (1985). Glucagon physiology and pathophysiology in the light of new advances. Diabetologia 28, 574-578.

Wendt, A., Birnir, B., Buschard, K., Gromada, J., Salehi, A., Sewing, S., Rors-man, P. and Braun, M. (2004). Glucose inhibition of glucagon secretion from rat -cells is mediated by GABA released from neighboring -cells. Diabetes53, 1038-1045.

Westerlund, J., Hellman, B. and Bergsten, P. (1996). Pulsatile insulin release from mouse islets occurs in the absence of stimulated entry of Ca2+. Journal of Clinical Investigation 97, 1860-1863.

Vieira, E. and Gylfe, E. (2004). Physiological activation of the glucagon-releasing mouse -cell by a low glucose concentration involves opening of voltage-dependent L-type Ca2+ channels. Diabetologia 47, A155.

Worley, J. F., III, McIntyre, M. S., Spencer, B., Mertz, R. J., Roe, M. W. and Dukes, I. D. (1994). Endoplasmic reticulum calcium store regulates membrane potential in mouse islet -cells. Journal of Biological Chemistry 269, 14359-14362.

Yoshimoto, Y., Fukuyama, Y., Horio, Y., Inanobe, A., Gotoh, M. and Kurachi, Y. (1999). Somatostatin induces hyperpolarization in pancreatic islet cells by activating a G protein-gated K+ channel. FEBS Letters 12, 265-269.

Zhang, S. L., Yu, Y., Roos, J., Kozak, J. A., Deerinck, T. J., Ellisman, M. H., Stauderman, K. A. and Cahalan, M. (2005). STIM1 is a Ca2+ sensor that acti-vates CRAC channels and migrates from the Ca2+ store to the plasma mem-brane. Nature 437, 902-905.