CASK modulates the assembly and function of the Mint1/Munc18-1 complex to regulate insulin secretion

Calcium/calmodulin-dependent protein serine kinase (CASK) is a key player in vesicle transport and release in neurons. However, its precise role, particularly in nonneuronal systems, is incompletely understood. We report that CASK functions as an important regulator of insulin secretion. CASK depletion in mouse islets/β cells substantially reduces insulin secretion and vesicle docking/fusion. CASK forms a ternary complex with Mint1 and Munc18-1, and this event is regulated by glucose stimulation in β cells. The crystal structure of the CASK/Mint1 complex demonstrates that Mint1 exhibits a unique “whip”-like structure that wraps tightly around the CASK-CaMK domain, which contains dual hydrophobic interaction sites. When triggered by CASK binding, Mint1 modulates the assembly of the complex. Further investigation revealed that CASK-Mint1 binding is critical for ternary complex formation, thereby controlling Munc18-1 membrane localization and insulin secretion. Our work illustrates the distinctive molecular basis underlying CASK/Mint1/Munc18-1 complex formation and reveals the importance of the CASK-Mint1-Munc18 signaling axis in insulin secretion.


Islet size
Bright-field images of isolated islets were obtained using Leica TCS SP2 microscope (Leica, Germany). Islet area was analyzed using ImageJ software. CASK-floxed mice (flx+/+) were crossed with RIP-Cre transgene mice (RIP-Cre+) (Cre) to generate β-cell-specific CASK knockout mice (flx+/+;RIP-Cre+) . Islets were isolated from the pancreas of CASK-knockout and control mice (RIP-Cre transgene mice) (8-12 weeks) as described before (1). Hanks' balanced salt solution containing collagenase P (Roche) was injected into the pancreas via the common bile duct. The pancreas then was excised and was digested by incubation at 37 C for 15 min. Islets were semipurified by two centrifugation steps at 290 g for 1min at 4 °C, followed by centrifugation on a Ficoll discontinuous gradient (Sigma-Aldrich). Islets were manually harvested from this semipure preparation. Before insulin release experiments, islets were allowed to recover for 24 h in RPMI 1640 medium containing 10% fetal calf serum, 11 mM glucose, 100 IU penicillin, and 0.1 mg/ml streptomycin. To measure insulin release from islets in static incubations, islets of comparable size were preincubated for 30 min in KRBH buffer.

Insulin secretion assay
Released insulin was collected after incubating the islets for 60 min in KRBH with 2.5 mM glucose, 25 mM glucose or 2.5 mM glucose plus 30 mM KCl. Supernatants were collected and centrifuged as described for cells. Insulin levels in the supernatant were measured by ELISA (Ultra Sensitive Mouse Insulin ELISA Kit, Crystal Chem) following the manufacturer's instruction.
To analyze insulin release kinetics, a customized perfusion apparatus was used (BioRep). Islets of similar size in KRBH buffer were loaded into columns with Bio-Gel P-4Gell (Bio-Rad) and pre-incubated with 2.5 mM glucose in KRBH buffer for 45 min, with a constant flow of 100 μl/min. After the prerun, the islets were exposed sequentially to 2.5 mM glucose (10 min), 25 mM glucose (25 min), 2.5 mM glucose plus 30 mM KCl (5 min). The flow-through was collected in a 96 well plate (1 min per well). To measure total insulin content of the isolated islets, islets were sonicated to obtain islet homogenate, and then incubated with 75% ethanol containing 0.18 N HCl overnight at 4°C. Protein content in islet homogenate was measured using BCA Protein Assay Kit (Tiangen) according to the manufacture's instruction. Insulin levels were measured using Ultra Sensitive Mouse Insulin ELISA Kit (Crystal Chem) according to the manufacturer's instruction.

Immunofluorescence microscopy
Cells were fixed with cold methanol. Samples of mouse pancreas tissue were fixed with 4% paraformaldehdye, and cryoprotected with 30% sucrose in PBS. Tissues were frozen at -80°C and cut into 5-μm sections on a sliding microtome (Leica). Cover slips were mounted using Fluorescence Mounting Medium (DAKO). Nuclei were counterstained with DAPI. The following antibodies were used in immunofluorescent assay: monoclonal anti-CASK (prepared in-house,

Analytical gel-filtration chromatography
Analytical gel-filtration chromatography was carried out on an ÄKTA FPLC system (GE Healthcare). Purified CASK-CaMK, Mint1-MID-CID, Munc18-1 or the mixture of three proteins was concentrated to ~2.0 mg/ml (OD280) separately and 100µl sample was applied to an analytical gel-filtration Superdex-200 10/300 GL column (GE Healthcare) equilibrated with buffer (50 mM Tris, pH 8.0, 100 mM NaCl, 1 mM EDTA, and 1 mM DTT). The flow rate applied was 0.5 ml/min, and 0.5 ml fractions were collected. The column was standardized using Bio-Rad's gel-filtration standard mixture of Ferritin (Mr 440 kDa, with an elution volume of 11.3 ml), BSA ( Mr 134 and 67 kDa, with elution volume of 12.8 and 14.9 ml respectively), -lactoglobulin ( Mr 35 kDa, with an elution volume of 16.1 ml), and Cytochrome C (Mr 13.6 kDa, with a elution volume of 18 ml). The elution volumes of molecular weight markers were indicated at the top.

Isothermal titration calorimetry (ITC) assay
ITC was carried out on a MicroCalorimeter ITC200 (Microcal LLC) at 25°C. All proteins were dissolved in a buffer containing 50 mM Tris(pH 8.0), 100 mM NaCl,1 mM EDTA and 1 mM DTT. The titration processes were performed by injecting 20× 2 µl aliquots of protein sample in a syringe (concentration of ~250-350 µM) into the stirred protein sample in the calorimeter cell (concentration of ~25-40 µM) at time intervals of 120 s to ensure that the titration peak returned to baseline. Each experiment was repeated three times. The heat of dilution obtained by the titration of the syringe sample into the buffer was subtracted. The data were analyzed using ORIGIN 8.0 and fitted by the one-site-binding model.

Structural determination
The initial phase was determined by molecular replacement using the structure models of CASK-CaMK (PDB code: 3TAC) as the searching models with PHASER (2).An incomplete structure model was further manually built with COOT (3) and refined with PHENIX (4) against the 2.45 Å data set. The Mint1-CID was built manually according to the 2Fo-Fc and Fo-Fc electron density maps. In the final stage, an additional TLS refinement was performed in PHENIX. The overall quality of the final structural models of the CASK/Mint1 complex was assessed by PROCHECK (5). Sequence alignments were prepared using CLUSTAL-W (6) .The statistics for the data collection and structural refinement were summarized in Table S1. Atomic coordinates and structure factors have been deposited in the Protein Data Bank.