All animal procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and all protocols were approved by the Animal Experimentation Ethics Committees of Ningxia Medical University (Yinchuan, China). Original breeding pairs of PICK1^−/− mice were obtained from Professor Jun Xia (Hong Kong University of Science and Technology, Hong Kong, China) and maintained in the Experimental Animal Center of Ningxia Medical University. Mice were kept in temperature-controlled conditions under 12/12 h light/dark cycles with food and water available ad libitum. Wild type ICR mice were obtained from the Experimental Animal Center of Ningxia Medical University (Ningxia, China). A total of 24 male mice (age, 8–10 weeks; weight, 25–35 g) were used throughout the study. All drugs were from Sigma-Aldrich (Merck Millipore, Darmstadt, Germany), Tocris Cookson (Bristol, UK), or Ascent Scientific Ltd. (Bristol, UK) unless stated otherwise.

Pancreatic slices were prepared according to previous studies (4,17). The abdominal cavity of mice was opened and warm (37°C) low-gelling agarose (1.9% wt/vol; Seaplaque GTG agarose; BioWhittaker Molecular Applications, Inc., Rockland, ME, USA) was injected into the distally clamped bile duct. The whole pancreas was immediately cooled with ice-cold extracellular solution (ECS, in mM): 125 NaCl, 2.5 KCl, 26 NaHCO₃, 1.25 NaH₂PO₄, 2 Na-pyruvate, 0.5 ascorbic acid, 3 myo-inositol, 6 lactic acid, 1 MgCl₂ and 2 CaCl₂, adjusted to pH 7.3 and oxygenated with 95% O₂/5% CO₂, 300±10 mOsm/kg. The pancreas was hardened agarose was extracted, placed in a small dish filled with warm agarose, and cooled rapidly on ice. Four small cubes were cut from the agarose-embedded pancreatic tissue and glued onto the sample plate of a vibratome (VT1000S; Leica Microsystems GmbH, Wetzlar, Germany). Slices (140 µm) were sectioned at a speed of 0.05 mm/s and 70 Hz in ice-cold ECS bubbled with 95% O₂/5% CO₂. Slices were kept in ice-cold ECS for at least 1 h prior to use.

Slices were placed in a submerged chamber and perfused with ECS (30°C) at 1.5 ml/min. Cells were visualized under an upright microscope (Zeiss Axioskop 2 FS; Carl Zeiss AG, Oberkochen, Germany) and a mounted with a Cohu CCD camera (Cohu, Inc., Poway, CA, USA) with 5X digital amplification. β cells from second or third layers in selected islets were used for whole-cell recordings. The recording pipettes were pulled on an electrode puller (P-97; Sutter Instrument Co., Novato, CA, USA) and had resistances of 2–4 MΩ. To measure the glutamate currents, electrodes were filled with a solution containing 125 mM CsCl, 40 mM HEPES, 2 mM MgCl₂ and 20 mM tetraethylammonium-Cl (pH 7.2 with CsOH). To measure voltage oscillation and K_ATP conductance, electrodes were filled with a solution containing: 150 mM KCl, 10 mM HEPES, 2 mM MgCl₂, 0.05 mM EGTA (pH 7.2 with KOH). Currents were filtered at 3 kHz and digitized at 10 kHz using an EPC10 amplifier (HEKA Elektronik Dr. Schulze GmbH, Lambrecht/Pfalz, Germany). MPS-2 multichannel microperfusion (MPS-2; INBIO, Shanghai, China) was used to locally puff solutions onto cells. Each injector delivered solution at a flow rate of 0.25 ml/min. The amplitude of the equilibrium response was the mean value of data points measured the final second prior to the termination of drug application.

The capacitance change induced by depolarization pulses (18–20) was used to measure the exocytosis of β cells. A 20 mV peak-to-peak 800 Hz sine wave was added to the holding potential and 10 cycles were averaged for each data point.

K_ATP channel conductance. K_ATP channel conductance (G_KATP) was measured according to a previous study (4). Whole-cell patch-clamp with K+-based 0 ATP pipette solution was made to allow dialysis of intracellular ATP. After the whole-cell configuration, ramp stimuli (−100 mV to-40 mV at 0.6 mV/msec steps) were applied at 1 Hz. The dominant current component between −100 and −40 mV ran through tolbutamide-sensitive K_ATP channels. At membrane potentials exceeding-30 mV, voltage-dependent K+ channels were activated (21). Mean amplitudes of current in the first and last 10 msec were acquired in an individual ramp. The increments of current between these two durations (ΔI) were calculated as current values of the ramp. The increment of voltage between first and last 10 msec (ΔV) was measured as 54 mV. K_ATP channel conductance was thereby calculated by Ohm's law (G_KATP=ΔI/54 mV) and depicted as one data point.

The statistical significance of the differences between groups was determined using one-way analysis of variance followed by a post hoc Tukey's test for multiple comparisons. P<0.05 was considered to indicate a statistically significant difference. Data in the text and figures are presented as the mean ± standard error. Statistical analysis was performed using SPSS software version 16.0 (SPSS, Inc., Chicago, IL, USA).

It was previously demonstrated that a 500-msec depolarizing pulse leads to a calcium influx and stimulates exocytosis and AMPARs participate in the exocytosis of β cells in the bath solution containing 3 mM glucose (4). Using the same method, the effect of PICK1 on β cell exocytosis was examined. To achieve this, a fusion protein, MBP-PICK1, was synthesized and dialyzed into β cells through patch pipettes. β cells were voltage-clamped at −80 mV and depolarized to 0 mV for 500 msec in order to induce exocytosis. Capacitance changes were defined as the difference between capacitances prior and subsequent to depolarization. Responses during depolarization were omitted (dashed line) because of irregular spikes (22). The averaged values of increased normalized capacitance were 12.1±1.5 fF/pF (wild type, n=15), 11.9±1.6 fF/pF (PICK1, n=16), 16.5±1.4 fF/pF (glutamate, n=15), 5.0±1.8 fF/pF (glutamate + PICK1, n=17) (Fig. 1A). The effect of EVKI (100 µM), a peptide that interferes with the interaction between GluR2 and PICK1 (23), was observed on the basal and depolarization-induced exocytosis. Increased normalized capacitance was 10.5±1.2 fF/pF (n=16) in the basal group and 18.6±2.0 fF/pF (n=16) in the EVKI + glutamate group (Fig. 1B), indicating that EVKI increases glutamate-evoked exocytosis. Subsequently, the effect of the application of PICK1 + EVKI or PICK1 + EVKI + glutamate were examined and no significantly different change in glutamate-induced exocytosis was observed between these two groups (PICK1 + EVKI, 10.1±0.9 fF/pF, n=17; PICK1 + EVKI + glutamate, 10.3±1.0 fF/pF, n=17) (Fig. 1C). Together, the results demonstrated that PICK1 decreased but EVKI increased the exocytosis of β cells (Fig. 1D), indicating that PICK1 prevents AMPAR-mediated β cell exocytosis.

Whether exocytosis is changed in β cells from PICK1^−/− mice was further examined. It was identified that normalized capacitance was increased to 19.9±2.4 fF/pF (n=17) in PICK1^−/− mice (Fig. 2A) compared with wild type mice, indicating that glutamate-induced exocytosis is increased in PICK1^−/− mouse β cells.

A previous study indicated that AMPAR activation increases cytosolic cyclic guanosine monophosphate (cGMP) and inhibits K_ATP channels (4). In addition, inhibition of K_ATP increases [Ca²⁺]_i and promotes the docking of insulin-containing granules (4). Therefore, whether PICK1 also affects K_ATP was investigated. A K⁺-based ATP-free pipette solution was used in whole-cell recordings to measure the conductance of K_ATP (4,24,25). Subsequent to establishing whole-cell configuration, K_ATP conductance rapidly increased to a peak (Fig. 3A) within 2 min. The mean maximum value was 3.9±0.2 nS/pF (Fig. 3B; n=12). PICK1 treatment increased the peak amplitude to 4.6±0.2 nS/pF (n=11, P<0.05) while peak amplitude declined to 3.1±0.2 nS/pF in PICK1^−/− β cells (n=11, P<0.05). These data suggest that PICK1 modulates β cell exocytosis through increasing K_ATP channel conductance.

Closure of K_ATP by cytosolic ATP induces a typical spontaneous burst-like membrane voltage oscillation (21), which is a determining factor for insulin secretion (21,26). The effect of PICK1 on K_ATP-dependent voltage oscillation was therefore examined. β cells were perfused with a background solution containing 11 mM glucose, which elicited stable bursts that commonly appeared within 30 min subsequent to the formation of the whole-cell clamp, and reliably lasted as long as the whole-cell recording continued. Following the baseline recordings, PICK1 was allowed to perfuse into β cells through patch pipettes to assess its effect on the oscillation (Fig. 4A). The average amplitude and duration of spikes were 18.5±0.9 mV and 13.1±0.5 sec (n=15) in the control, while PICK1 attenuated burst amplitude to 12.5±0.8 mV (n=16) and burst duration to 7.8±0.4 sec (n=16) (Fig. 4B). In agreement with Fig. 2, these data clearly indicated suggest that PICK1 modulates K_ATP channel.