Sulfonylurea-Mediated Stimulation of Insulin Exocytosis via an ATP-Sensitive K⁺ Channel–Independent Action Free

The priming reaction in exocytosis is a fundamental event in the fate of the secretory granule. Although the details are not fully elucidated, it seems clear that it involves pairing of soluble N-ethylmaleimide sensitive fusion protein attachment protein receptor (SNARE) proteins on the vesicles and in the target membrane and ATP-hydrolysis. Insulin secretion is indeed highly dependent on ATP (8), but the exact nature of the priming reaction in pancreatic β-cells remains elusive. Different priming enzymes have been proposed such as N-ethylmaleimide sensitive factor (NSF) (9,10) and phosphatidylinositol-4-phosphate 5-kinase (PtdInsP5K) (11). The role of NSF in insulin secretion (12), and secretion in general, is debated, and the enzyme is currently regarded as playing more a role in postfusion disassembly of the SNARE complex (13). Recently, we have identified the v-type H⁺-ATPase in the granule membrane as one candidate priming enzyme (14). This ATPase is responsible for generation of the acidific granule interior in mature insulin granules. The importance of a low intragranular pH for processing of the prohormone is established (15). However, that intragranular pH also affects the release competence of the insulin granule was not previously known. H⁺ pumping into the granule requires simultaneous Cl⁻ uptake via ClC-3 Cl⁻ channels to maintain electroneutrality, reminiscent of the situation in pancreatic zymogen granules (16). Otherwise, a positive granule potential quickly develops that prevents granule acidification (17). This Cl⁻ conductance determining the rate of acidification is under metabolic control. It is inhibited by decreases in the intracellular ATP/ADP ratio, and under such circumstances, granule acidification is arrested. The involvement of ClC-3 channels in acidification of synaptic vesicles was recently demonstrated in ClC-3 knockout mice (18). Modulation of ClC-3 channel activity also represents the mechanism whereby SU potentiates exocytosis, as will be discussed below.

SU binding to a 65-kd protein in the granule fraction was demonstrated by photoaffinity labeling with [³H]-glibenclamide in normal mouse islets (Fig. 1) and in insulin-releasing RINm5F cells (not shown). The SU-mediated actions on granular ClC-3 Cl⁻ channel activity and cotransport of H⁺ over the granule membrane were explored using the standard whole-cell configuration of the patch clamp technique. The cells were infused with an ATP-containing intracellular solution supplemented with the protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP) (100 μmol/l). Deprotonation of the insulin granules via this artificial pathway was monitored as a rapid ∼60% decrease in LysoSensor fluorescence 1 min after establishment of the standard whole-cell configuration (Fig. 2A–B). Deprotonation did not occur when the ClC-3 conductance was inhibited by the broad Cl⁻ channel blocker 4,4′-diisothiocyanostilbene-2,2′-disulfonic acid (DIDS) (100 μmol/l) or a functional antibody against the ClC-3 channel. Similar effects were observed when the cytosolic ATP/ADP ratio was decreased (3 mmol/l Mg-ATP/5 mmol/l Mg-ADP) or when the K_ATP channel activator diazoxide (200 μmol/l) was added to the intracellular solution. Interestingly, the inhibitory action of Mg-ADP on granule deprotonation was completely counteracted by tolbutamide (100 μmol/l). Admittedly, this assay measures H⁺ flux in the direction opposite to that expected under physiological conditions. A major advantage is the direct demonstration of the necessity of a counterconductance for H⁺ translocation over the granule membrane, apart from a favorable signal-noise ratio. However, it was also possible to detect the effects of Mg-ATP, Mg-ADP, and tolbutamide on H⁺ transport into the granule using a similar LysoSensor-based assay without CCCP (14).

The most exciting observation was that cytosolic conditions that permitted rapid H⁺transport over the granule membrane in either of the LysoSensor assays were associated with a high exocytotic capacity (Fig. 2C–D). When the cells were dialyzed with a control Ca²⁺-containing intracellular solution (free [Ca²⁺]_i ∼1.5 μmol/l), supplemented with Mg-ATP (3 mmol/l) and cAMP (0.1 mmol/l), exocytosis was rapidly initiated after establishing the standard whole-cell condition and averaged 32 ± 5 fF/s (n = 26). The addition of Mg-ADP (5 mmol/l) or diazoxide (200 μmol/l) reduced exocytotic rates by 75 and 47%, respectively. Of particular interest was the fact that tolbutamide (100 μmol/l) counteracted the Mg-ADP–induced inhibition of exocytosis, and in the simultaneous presence of Mg-ADP and tolbutamide, exocytosis averaged 95% of that observed under control conditions. Thus, the effects of tolbutamide on Cl⁻-transport/granule acidification mirror those on exocytosis.

In Fig. 3, we have summarized our current ideas on how SUs interfere with the β-cell exocytotic machinery. SU binding to the granular 65-kd receptor protein results in increased granular ClC-3 channel activity. In analogy with the situation in the K_ATP channel (20), it can be hypothesized that ATP directly interacts with the ClC-3 channel, whereas ADP binds to the granular SU receptor. In this scenario, SU binding would then relieve the ADP-mediated effects, and thereby accelerate acidification of the granule interior and priming for exocytosis. Future work is required to establish the exact identity of the granular SU receptor and its interactions with the ClC-3 channels.

We thank our colleagues in Lund and Manchester for sharing unpublished data. Financial support was obtained from the Swedish Medical Research Council (grants 8647, 9890, 12334, 12708, and 13147), the Swedish Diabetes Association, the Royal Physiographic Society, the Juvenile Diabetes Foundation International, the Knut and Alice Wallenbergs Stiftelse, the European Community, and the Novo Nordisk Foundation and DFG Th 345/6-1 to F.T.