Regulation of L-type Ca2+ Channel Activity and Insulin Secretion by Huntingtin-associated Protein 1*

Huntingtin-associated protein 1 (Hap1) was originally identified as a protein that binds to the Huntington disease protein, huntingtin. Growing evidence has shown that Hap1 participates in intracellular trafficking via its association with various microtubule-dependent transporters and organelles. Recent studies also revealed that Hap1 is involved in exocytosis such as insulin release from pancreatic β-cells. However, the mechanism underlying the action of Hap1 on insulin release remains to be investigated. We found that Hap1 knock-out mice had a lower plasma basal insulin level than control mice. Using cultured pancreatic β-cell lines, INS-1 cells, we confirmed that decreasing Hap1 reduces the number of secreted vesicles and inhibits vesicle exocytosis. Electrophysiology and imaging of intracellular Ca2+ measurements demonstrated that Hap1 depletion significantly reduces the influx of Ca2+ mediated by L-type Ca2+ channels (Cav). This decrease is not due to reduced expression of Cav1.2 channel mRNA but results from the decreased distribution of Cav1.2 on the plasma membrane of INS-1 cells. Fluorescence recovery after photobleaching showed a defective movement of Cav1.2 in Hap1 silencing INS-1 cells. Our findings suggest that Hap1 is important for insulin secretion of pancreatic β-cells via regulating the intracellular trafficking and plasma membrane localization of Cav1.2, providing new insight into the mechanisms that regulate insulin release from pancreatic β-cells.


Edited by Jeffrey Pessin
Huntingtin-associated protein 1 (Hap1) was originally identified as a protein that binds to the Huntington disease protein, huntingtin. Growing evidence has shown that Hap1 participates in intracellular trafficking via its association with various microtubule-dependent transporters and organelles. Recent studies also revealed that Hap1 is involved in exocytosis such as insulin release from pancreatic ␤-cells. However, the mechanism underlying the action of Hap1 on insulin release remains to be investigated. We found that Hap1 knock-out mice had a lower plasma basal insulin level than control mice. Using cultured pancreatic ␤-cell lines, INS-1 cells, we confirmed that decreasing Hap1 reduces the number of secreted vesicles and inhibits vesicle exocytosis. Electrophysiology and imaging of intracellular Ca 2؉ measurements demonstrated that Hap1 depletion significantly reduces the influx of Ca 2؉ mediated by L-type Ca 2؉ channels (Cav). This decrease is not due to reduced expression of Cav1.2 channel mRNA but results from the decreased distribution of Cav1.2 on the plasma membrane of INS-1 cells. Fluorescence recovery after photobleaching showed a defective movement of Cav1.2 in Hap1 silencing INS-1 cells. Our findings suggest that Hap1 is important for insulin secretion of pancreatic ␤-cells via regulating the intracellular trafficking and plasma membrane localization of Cav1.2, providing new insight into the mechanisms that regulate insulin release from pancreatic ␤-cells.
Huntingtin-associated protein 1 (Hap1) was the first huntingtin-interacting protein identified in yeast two-hybrid screens (1). Hap1 is expressed abundantly in the brain and is also present in some of the peripheral endocrine systems, such as the pituitary, thyroid, adrenal glands, and pancreas (2,3). In the pancreas, Hap1 is selectively expressed in ␤-cells that release insulin (4). A recent study showed that Hap1 is important for insulin release from ␤-cells, and when Hap1 expression is reduced, glucose-mediated insulin release is inhibited (5). Mutant mice with the selective Hap1 deficiency in pancreatic ␤-cells had impaired glucose tolerance and decreased insulin secretion in response to glucose stimulation (5). Because Hap1 binds the Huntington disease (HD) 4 protein huntingtin, Hap1 dysfunction may also be involved in metabolic abnormalities in HD, which is evident in HD mouse models and patients. For example, the R6/2 HD mice develop diabetes due to reduced ␤-cell mass and deficient insulin exocytosis (6). Mutant huntingtin forms aggregates in the cytoplasm and reduces insulin secretion from huntingtin-transfected pancreatic ␤-cell lines, NIT-1 cells, which could be reversed by overexpression of heat shock protein 40 (HSP40) (7). Patients with HD exhibit increased incidence of diabetes mellitus (8). Also, mutant huntingtin was found to interact with ␤-tubulin and disrupt vesicular transport and insulin secretion (9). Given these findings, it would be interesting to investigate how Hap1 is involved in insulin release in pancreatic ␤-cells.
Many proteins known to bind Hap1 are associated with endocytosis and vesicular trafficking. The binding of Hap1 with p150 Glued subunit of dynactin and huntingtin could enhance vesicular transport of brain-derived neurotrophic factor (10). Reducing Hap1 protein has been shown to disrupt kinesin light chain-dependent neurite outgrowth and transport of amyloid precursor protein vesicles (11). Phosphorylation at the C terminus of Hap1A, the A isoform of Hap1, plays an important role in the association of Hap1A with kinesin light chain and p150 Glued (12). Phosphorylation of Hap1A also enhances the binding of 14-3-3 protein involved in the assembly of protein complexes and the regulation of protein trafficking (13). In cultured primary neurons, lack of Hap1 decreased intracellular TrkA amounts and impaired neurite outgrowth (12). Reducing Hap1 level resulted in decrease of TrkB internalization and signaling transduction (14). Hap1 also associates with the inositol 1,4,5triphosphate receptor (InsP3R) (15), forming a complex required for proper Ca 2ϩ signaling (16). Consistently, Ca 2ϩ mobilization defects in ␤ cells are involved in the diabetes reported in HD patients (17).
Endocytosis and exocytosis function in many common ways in neurons and endocrine cells. Pancreatic ␤-cells, a major cellular component of the islets, are electrically excitable and exquisitely sensitive to glucose. In response to blood glucose, ␤-cells secrete insulin and play a unique role in glucose home-ostasis. Glucose homeostasis is critical to the health and survival of mammals. L-type Ca 2ϩ channels take center stage in the process of insulin release (18 -21). Neurotransmitter release is attributed to Ca 2ϩ influx through P/Q-and N-type Ca 2ϩ channels, and the L-type (Cav1.2 and Cav1.3) channels are considered as responsible for hormone secretion (22). Electrophysiological and molecular biological studies indicate that pancreatic ␤-cells express several subtypes of Cav channels. In particular, L-type Ca 2ϩ channels are considered crucial for ␤-cell function, given that dihydropyridines can potently suppress insulin secretion (23). Both Cav1.2 and Cav1.3 have been found in human and rodent islets, as well as in various ␤-cell lines (23,24). Up-regulation of ␤-cells' Ca 2ϩ channel activity and/or density results in enhanced insulin exocytosis and more efficient glucose homeostasis (25). Moreover, down-regulation of Cav channel activity and/or density causes less insulin secretion and glucose intolerance, which is associated with a group of type 2 diabetic patients (26 -29). Studies in mutant mice with Cav1.2 or Cav1.3 deficiency have confirmed that L-type Ca 2ϩ channels are crucial for the physiology of ␤-cells (30,31). Proper trafficking and distribution of L-type Ca 2ϩ channels are essential for insulin release.
Growing evidence indicates that Hap1 is likely involved in microtubule-dependent trafficking in neurons, such as vesicular transport, membrane receptor trafficking, and Ca 2ϩ release (32)(33)(34). Our previous study has also demonstrated that Hap1 may help to regulate transport of insulin-containing secretory granules along cortical actin filaments in pancreatic ␤-cells (35). Whether and how Hap1 regulates the trafficking or surface expression of L-type Ca 2ϩ channel in endocrine cells have not been reported. In this study, we provide both in vitro and in vivo data that Hap1 is required for insulin secretion from ␤-cells. Using patch clamp recordings, we demonstrated that Hap1 depletion significantly reduces the L-type Ca 2ϩ currents. Using biochemical and molecular biology techniques, we also found that Hap1 regulates the surface expression level and intracellular trafficking of L-type Ca 2ϩ channels Cav1.2 in INS-1 cells. These data suggest that Hap1 plays an important role in regulation of insulin secretion in ␤-cells and offer a new therapeutic target for ameliorating metabolic disorders due to defective insulin release from pancreatic ␤-cells.

Reducing Hap1 Expression Decreases the Release of Insulin-
To provide further evidence for the idea that Hap1 regulates insulin release from ␤-cells, we measured the plasma insulin level of Hap1 knock-out (KO) or Hap1 Ϫ/Ϫ mice and investigated the effect of Hap1 deficiency on secretion of cultured ␤-cell lines INS-1. Because Hap1 Ϫ/Ϫ mice have retarded growth and die 3-4 days after birth, and Hap1 ϩ/Ϫ heterozygous mice showed no obvious behavioral and body weight abnormalities to wild-type mice and lived as long as WT mice (36), we focused our studies on Hap1 Ϫ/Ϫ 3-4-day-old mice to investigate the role of Hap1 in insulin secretion. We collected the blood of Hap1 KO and WT pups and used the plasma to analyze insulin levels via radioimmunoassay (RIA). The results showed that the insulin level of KO mice was significantly lower than WT mice (Fig. 1A, upper panel), indicating that lack of Hap1 may impair insulin release in mice.
INS-1 has been widely used to investigate the mechanisms underlying insulin release. We used small interference RNA (siRNA) to reduce endogenous Hap1 expression in INS-1 cells. RIA detection found that the INS-1 cells transfected with Hap1-siRNA plasmid showed a lower level of insulin release compared with the control cells transfected with scramble plasmid (Fig. 1A, bottom panel). Immunofluorescent staining of the Hap1 signal in Hap1-siRNA-treated cells was markedly reduced compared with cells treated with scramble-siRNA (Fig.  1B). Western blotting also demonstrated that Hap1 expression is inhibited via Hap1-siRNA (Fig. 1C). We then performed membrane capacitance measurement experiments to examine the exocytosis of insulin-containing vesicles in INS-1 cells. To do so, we designed a voltage clamp depolarization protocol to distinguish vesicle release from the immediately releasable pool (IRP) and readily releasable pool (RRP). The stimulation protocol consisted of a train of six 30-ms depolarization (⌬Cm6) to deplete the IRP, followed by 10 100-ms depolarization (⌬Cm10) to elicit secretion from the RRP, and the holding potential from Ϫ80 to 0 mV ( Fig. 2A). In response to this stimulation protocol, there was a small capacitance increase after ⌬Cm6 and ⌬Cm10 corresponding to release from the IRP and RRP in Hap1-siRNA-treated cells. In contrast, the stimulated exocytosis in control cells was significantly greater for both IRP and RRP. The quantification results suggested that there was a decrease in the number of releasing vesicles in Hap1-silenced cells versus control cells (Fig. 2B). Taken together, these findings indicate that Hap1 deficiency affects insulin release from ␤-cells.
Dynamics of Insulin Granules Are Inhibited in Hap1-deficient ␤-Cells-Release of neurotransmitters and peptide hormones involves exocytotic fusion of secretory vesicles with the plasma membrane (37). Furthermore, we examined the dynamic parameters of exocytosis in INS-1 cells by imaging vesicle membrane-targeted fluorescent probes (VAMP2-pHluorin). Live cells were examined in a confocal microscope at the cell footprint to better visualize near PM fluorescent spots. We used fusion constructs of vesicle membrane protein synaptobrevin-2 (VAMP2) with a pH-sensitive green fluorescent protein PHluorin (VAMP2-pHluorin) to measure the vesicle release (38) by taking advantage of the drastically greater pH sensitivity of the probe. Because the lumen of the insulin vesicle is acidic (pH5.5-6.0), we captured a fluorescence image of VAMP2-pHluorin under the conditions in which pHluorin fluorescence is expected to be close to zero. Stimulation of Ca 2ϩ influx with 30 mM K ϩ caused insulin-containing spots to brighten and spread suddenly, which is consistent with release of the fluorescent peptide. As shown in Fig. 3, A and B, fluorescence intensity increased first in a central region, then spread into a surrounding annulus, and finally declined as dye diffused away. To determine whether there are any effects of loss of Hap1 on insulin release, we co-transfected both constructs siRNA and PH-VAMP2 simultaneously into the same cells and compared those transfected with Hap1-scramble. When cells were stimulated with 30 mM K ϩ for 300 s, the number of secreted vesicles in Hap1 knockdown cells was decreased compared with that from the control cells (Fig. 3D). Comparing the fluorescence intensity of each sample revealed that there is a significant reduction of insulin release events from the Hap1 deficiency cells (Fig. 3, C and E). Inhibiting Hap1 also diminished the duration of insulin release from secreted vesicles (Fig. 3F). Taken together, these findings indicate that Hap1 deficiency affects the dynamics of insulin release from ␤-cells.
Hap1 Deficiency Reduces L-type Ca 2ϩ Currents in INS-1 Cells-Cav channels play a critical role in insulin secretion as well as the survival of ␤-cells. Cav channels are ubiquitously expressed in various cell types throughout the body. There is considerable consensus that L-type Ca 2ϩ current is the major Ca 2ϩ current subtype in the ␤-cells from all the examined species (25,39,40). The ␤-cell Cav channel-mediated Ca 2ϩ entry directly stimulates secretory granule trafficking and triggers insulin exocytosis. This is the paramount way whereby Cav channels regulate insulin secretion (25,41). To determine the role of Hap1 in the Ca 2ϩ -dependent secretory pathway, we next compared the L-type Ca 2ϩ current in Hap1-siRNA-treated and control INS-1 cells using whole cell patch clamp measurements. TEA-containing extracellular solution and Cs ϩ were used to block the voltage-gated K ϩ channels and tetradotoxin to block the voltage-gated Na ϩ channels. From recording of whole cells with a holding potential of Ϫ70 mV, ␤-cells exhibited inward currents typical for Ca 2ϩ channels current in response to depolarizing voltage from a holding potential of Ϫ70 to ϩ50 mV in 10-mV increments (Fig. 4A, upper panel). Hap1-siRNA-treated cells showed distinctly small L-type Ca 2ϩ currents evoked by depolarizing step pulses (Fig. 4A, bottom  panel). Under the same condition, control cells exhibited larger Ca 2ϩ currents (Fig. 4A, middle panel). The mean current-voltage relationships (I-V) of L-type Ca 2ϩ channels were then obtained (Fig. 4B). As shown in Fig. 4C, the averaged Ba 2ϩ current density (pA/pF) at 10 mV in control INS-1 cells expressing scramble-siRNA was much higher than that of Hap1-siRNA cells.  DECEMBER 16, 2016 • VOLUME 291 • NUMBER 51

Hap1 and Insulin Secretion
Inhibiting Hap1 Expression Lowers the Ca 2ϩ Influx of INS-1 Cells-The low levels of L-type Ca 2ϩ current and current density in Hap1 deficiency cells led us to speculate that the knockdown of Hap1 expression may decrease the Ca 2ϩ influx. To verify this hypothesis, we used Ca 2ϩ microfluorimetry to measure Ca 2ϩ entry evoked by extracellular high K ϩ stimulation in INS-1 cells, which expressed either Hap1-siRNA or scramble-siRNA and were loaded with Fura-2-AM. We found the increase in [Ca 2ϩ ] i in Hap1-siRNA and control cells in response to 30 mM K ϩ stimulation. Deficiency of Hap1 led to a lower level of high K ϩ -stimulated [Ca 2ϩ ] i rise than control cells (Fig. 5, A and B). The averaged changes in 340/380 intensity ratio are plotted in Fig. 5C. We found that INS-1 cells expressed endogenous L-type voltage Ca 2ϩ channels predominantly, as the intracellular Ca 2ϩ level was rapidly and significantly inhibited by nifedipine, a typical L-type Ca 2ϩ channel blocker (Fig.  5D). These findings indicate that Hap1 is likely to functionally regulate the activity of L-type Ca 2ϩ channels so that loss of Hap1 reduces the L-type Ca 2ϩ currents and membrane depolarization-induced Ca 2ϩ influx.
Loss of Hap1 Down-regulates the Surface Expression of Cav1.2 in INS-1 Cells-To rule out the possibility that Hap1 may regulate L-type Ca 2ϩ channel transcription, we performed RT-PCR using INS-1 cells treated with Hap1-siRNA or scramble-siRNA for 48 h. Our data show that Hap1 deficiency has no effect on Cav1.2 mRNA expression. Hap1 is involved in the intracellular transport of several membrane receptors (12,16). To reveal whether the reduction of L-type Ca 2ϩ currents and decrease of Ca 2ϩ influx in Hap1 deficiency INS-1 cells resulted from diminishment of transportation of Cav channel to the cell surface, we checked the effect of Hap1 loss on the cell surface expression of Cav1.2 in INS-1 cells. The plasmid of YFP-Cav1.2-HA was constructed to express YFP at the intracellular N terminus and a HA epitope in the extracellular S5-H5 loop of domain II (YH-Cav1.2). The YFP and HA tags had no effect on the channel but can help us to measure the total number of Cav1.2 channels expressed on the cell surface (42). To verify the effect of Hap1 on Cav1.2 channel localization on the plasma membrane, YH-Cav1.2, ␤1b, ␣2␦, and Hap1-siRNA or scramble-siRNA were co-transfected in INS-1 cells, and 48 h later we measured the average YFP intensity expressed on the PM. The YFP fluorescence at the edge of the cell membrane in control cells was much stronger than Hap1-silencing cells (Fig.  6A). As a result, there was a significant decrease of Cav1.2 expression level on membrane in Hap1-deficient cells compared with control cells (Fig. 6, A and B). We then analyzed the ratio of Cav1.2 mean intensity in the PM versus the cytosol in cells by fixing them and staining with anti-Cav1.2 antibodies (Fig. 6C). There was a small decrease of this ratio in the Hap1deficient cells, compared with control cells (Fig. 6, C and D). We also investigated the membrane localization of YH-Cav1.2 channels by fixing the cells and staining with anti-HA antibodies without membrane permeabilization. The non-permeabilized cells had YFP fluorescence throughout the whole cell, but HA staining was seen only at the edge of the cell membrane. There was a significant decrease of Cav1.2 expression levels on the membrane in Hap1 deficiency cells compared with control cells (Fig. 6, E and F). Evidence for the effect of Hap1 on Cav1.2 surface expression was also obtained by immunoblotting of subcellular fractions samples (Fig. 6G). Furthermore, using immunoprecipitation with antibodies to Cav1.2 or Hap1, we found that Cav1.2 interacts with Hap1 (Fig. 7, A and B), also supporting the idea that Hap1 is important for the localization of Cav1.2 to the plasma membrane in INS-1 cells.
Trafficking of Cav1.2 Is Deficient in Hap1-silencing ␤-Cells-Because the surface level of Cav1.2 was significantly affected by Hap1 deficiency, we then investigated whether Hap1 is involved in the regulation of Cav1.2 intracellular trafficking. To detect the intracellular movement of Cav1.2, we examined Cav1.2 trafficking in INS-1 cells using fluorescence recovery after photobleaching (FRAP). We transiently co-transfected INS-1 cells with Hap1-siRNA ϩ Cav1.2-YFP or scramble-siRNA ϩ Cav1.2-YFP vectors, and we measured the movement of Cav1.2-YFP in these groups. We obtained a series of images for Cav1.2-YFP recovery during 0 to 130 s after photobleaching. Hap1-siRNA cells showed obviously slower recovery after bleaching Cav1.2-YFP (Fig. 8A, bottom panel). Under the same conditions, scramble-siRNA cells exhibited a faster recovery (Fig. 8A, upper panel). Fig. 8B shows the fluorescence recovery data were fit by single exponential curve equation to obtain recovery time constants () (see under "Experimental Procedures"). When measuring the recovery kinetics, Hap1-siRNA-treated cells took a longer time (31.2 Ϯ 5.8 s) than con-trol cells (20.2 Ϯ 7.3 s) with a significant difference (Fig. 8C). It was evident that Hap1 silencing decreases the speed of Cav1.2 intracellular trafficking and also prolongs the recovery time of Cav1.2, implying that Hap1 is able to assist Cav1.2 transport in ␤-cells.

Discussion
Although mainly expressed in the brain, Hap1 is also expressed in endocrine systems that were involved in hormone release (1,3,43,44). Immunofluorescent double staining of mouse pancreas showed that Hap1 is selectively expressed in the insulin-releasing ␤-cells but not in ␣-cells and ␦-cells that produce glucagon and somatostatin (4). This specific expression in pancreatic ␤-cells implies that Hap1 is involved in the function of insulin release. Recent studies discovered that mutant mice with selective Hap1 elimination in pancreatic ␤-cells displayed a significant inhibition of insulin release and glucose tolerance. The same phenomenon was also found in cultured pancreatic ␤-cell lines Min6 and NIT cells (5). In our studies, we provided new evidence that loss of Hap1 decreased the plasma insulin level in Hap1 knock-out mice and INS-1 cells. Moreover, our mechanistic studies revealed that Hap1 deficiency inhibited the kinetics of insulin exocytosis by reducing the translocation of Cav1.2 to the plasma membrane.  Pancreatic ␤-cells are electrically excitable and exquisitely sensitive to glucose. The secretory machinery of insulin release from islet ␤-cells shares many properties with the release of neurotransmitters in neurons, including synaptic vesicle docking, fusion, and release (45,46). It is known that insulin is produced in pancreatic ␤-cells, transported by the secretory vesicle, and released from the ␤-cells by exocytosis. This process requires the fusion of the secretory vesicle membrane with the plasma membrane, followed by release of insulin into the extracellular space to reach the blood circulation (47). In contrast to the exocytosis of neurotransmitters in neurons, insulin release is triggered by Ca 2ϩ influx through L-type Ca 2ϩ channels in pancreatic ␤-cells (48,49). L-type Ca 2ϩ channels have been reported to co-localize with insulin-containing vesicles (50). L-type Ca 2ϩ current is the major Ca 2ϩ current subtype in the ␤-cells (25,39,40). All these results indicate that L-type Ca 2ϩ channels play an important role in insulin secretion from pancreatic ␤-cells. Our studies suggest that Hap1 regulates insulin release through the intracellular trafficking of L-type Ca 2ϩ channels.
Growing evidence suggests that Hap1 may maintain neuronal transmission and neurotrophic functions by regulating intracellular trafficking, recycling, and stabilization of receptors. Hap1 may increase the receptors at the plasma surface via inhibition of lysosomal degradation and enhancing endocytic recycling. For example, Hap1 overexpression increases the activity and surface level of GABA A receptor in the hypothalamus (51). Hap1 modulates the normal membrane level of TrkA by preventing the degradation of internalized TrkA (12). Hap1 can also increase the TrkB level by interacting with Ahi1 (52). Overexpression of Hap1 prevents the trafficking of internalized EGF receptor from early endosomes to lysosomes, resulting in the suppression of EGF receptor degradation (36). InsP3R1, an intracellular Ca 2ϩ release channel, is another membrane receptor that also binds to Hap1 (16). The binding of Htt and Hap1 can influence Ca 2ϩ signaling mediated by InsP3R1 in neurons (15). Our studies found that loss of Hap1 could impair L-type Ca 2ϩ currents and Ca 2ϩ influx. Our data from FRAP and immunofluorescence showed that knockdown of Hap1 reduced  the intracellular trafficking and surface expression of Cav1.2. We also demonstrated that there was an interaction between Hap1 and Cav1.2 via immunoprecipitation, which suggested that Cav1.2 could be a new target for the future therapeutic regulation of insulin release. INS-1 cells and even mouse islet ␤-cells express both Cav1.2 and Cav1.3 channels (25,53,54). It remains to be determined whether Hap1 protein interacts with Cav1.3 channels.
Ca 2ϩ is a triggering signal for neurotransmitter and hormone release. Alteration in the density and function of Ca 2ϩ channels in ␤-cells occurs under pathophysiological conditions. The activity of the ␤-cell Ca 2ϩ channel can be up-or down-regulated in different animal models of diabetes (27,55). Ca 2ϩ channel mutation results in ␤-cell dysfunction that is possibly associated with type 2 diabetes (56,57). How Cav channel dysfunction contributes to the pathological alterations in ␤-cells remains to be fully investigated. Further analysis of the Ca 2ϩ -dependent exocytosis and endocytosis in pancreatic ␤-cells from a genetically manipulated mouse model will be necessary for elucidating the underlying mechanism. Both HD patients and mouse models showed abnormal glucose tolerance and insulin secretion (6,9,58,59). It would be interesting to investigate whether impaired Hap1 function is involved in HD pancreatic dysfunction and type 2 diabetes and whether the addition of Hap1 could improve the insulin secretion in ␤-cells. In addition, there is impaired secretion of various hormones in HD patients and hypothalamic and neuroendocrine changes were seen in HD (60 -64). Hap1 dysfunction could also be involved in the endocrine alteration and metabolic dysfunction seen in HD. Our finding that Hap1 regulates Cav channel trafficking to influence insulin release from ␤-cells provides a new direction to investigate the mechanisms underlying impaired insulin release in a variety of pathological conditions.

Experimental Procedures
Animals-Hap1 knock-out mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals were kept under a standardized breeding environment with free water and food. Postnatal day 1 Hap1 ϩ/ϩ (wild type, WT) and Hap1 Ϫ/Ϫ (knock-out, KO) pups were obtained from mice crossed between Hap1 ϩ/Ϫ .
Radioimmunoassay of Insulin Secretion-Mice were sacrificed by decapitation; the blood was collected and centrifuged to collect plasma. All steps are operating on the ice. The plasma was diluted with PBS, and the basal insulin level was measured using RIA with the DFM 96-type multitube radioimmunoassay counter. For measuring insulin release from INS-1 cells, Hap1-siRNA or Hap1-scramble-treated cells were plated in 6-well plates and maintained at 37°C in 5% CO 2 , 95% air-humidified atmosphere for 48 h. The medium in 6-well plates was collected for measuring released insulin using RIA.
Cell Culture and Transfection-The rat insulin-producing INS-1 cells (ATCC) were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS), 1 mM sodium pyruvate, 10 mM HEPES, 50 M ␤-mercaptoethanol, 50 units/ml penicillin, and 50 mg/ml streptomycin. Cell cultures were maintained at 37°C in 5% CO 2 , 95% air-humidified atmosphere. Cells were transfected using Lipofectamine 2000 (Invit-rogen) according to the manufacturer's instructions. Briefly, cells were plated onto 30-mm round glass coverslips in a 6-well plate. The following day, cells were transfected with various plasmids. Six hours later, the medium was replaced with complete DMEM, and the transfected cells were maintained in culture overnight. After 48 h, the transfected cells were plated onto 30-mm round glass coverslips and mounted in a Teflon chamber. YFP-HA-Cav1.2, ␤1b, and ␣2␦1 plasmids were obtained from the laboratory of Ricardo E. Dolmetsch, Stanford University.
For siRNA knockdown experiments, the following 19-mer oligonucleotide sequences were selected as described previously (12), 5Ј-GAAGTATGTCCTCCAGCAA-3Ј. Short hairpin oligonucleotides were designed and inserted into Pgenesil-1 vector (Invitrogen) between BamHI and HindIII sites to construct two plasmids, mCherry-Hap1-siRNA and EGFP-Hap1-siRNA. This shRNA was effective at reducing the expression of Hap1.
Electrophysiology-Whole cell voltage clamp recordings were applied to INS-1 cells with standard techniques. For knockdown studies, INS-1 cells were transfected with 2 g of scramble or the corresponding rat Hap1-siRNA plasmid also expressed GFP. GFP-positive cells were recorded 48 h after transfection. Whole cell currents were recorded using EPC 10 amplifier (HEKA) with the following external solution (in mM): 20 BaCl 2 , 120 TEA-Cl, 1 MgCl 2 , 10 glucose, 10 HEPES, and 0.001 tetradotoxin, pH adjusted to 7.4 with TEA-OH. Pipette solution contained (in mM) the following: 135 CsMeSO 3 , 5 CsCl, 0.5 EGTA, 1 MgCl 2 , 4 Mg-ATP, 0.5 Na-GTP, and 10 HEPES, and pH was adjusted to 7.4 with CsOH. Data were filtered at 2 kHz and digitized at 5 kHz. All experiments were performed at room temperature.
Intracellular Ca 2ϩ Measurements-INS-1 cells were loaded with 2 M ratiometric dye fura-2/AM (Invitrogen) for 30 min at 37°C in Ca 2ϩ -free solution (140 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 10 mM glucose, 10 mM HEPES). Cells were then washed with Ca 2ϩ -free solution, and fura-2 loaded into cells was allowed to de-esterify for 30 min at 25°C. Intracellular Ca 2ϩ was measured using a dual-wavelength excitation (340/380 nm) photometry system on an inverted microscope (TE2000, Nikon, Japan) equipped with a polychromatic xenon light source (TILL photonics, Germany). Briefly, fura-2 was excited at 340 and 380 nm alternatively at a rate of 1 Hz, and the fluorescence emission was first passed through a 415-nm long pass dichroic filter and then monitored at 515 nm with a photodiode controlled by the TILL photometry system and X-Chart extension of Pulse software (HEKA, Lambrecht, Germany). Intracellular Ca 2ϩ concentrations are represented by the ratio of fluorescence emission at the two excitation wavelengths.
Immunofluorescent Labeling-INS-1 cells transfected with YFP-Cav1.2-HA were washed three times with PBS and fixed with 4% paraformaldehyde for 20 min (for non-permeabilization, this step was omitted) and then incubated with primary antibodies to HA antibody (Sigma) at 1:1000 dilution overnight. Cy3-conjugated secondary antibodies (Jackson Immuno-Research) were used at a dilution of 1:500 at room temperature under dark conditions for 2 h. Photographs were taken with a laser scanning confocal microscope (Olympus FV1000; Tokyo, Japan) using a-plan Apochromat ϫ60/NA1.42 oil immersion objective and analyzed using Fluoview 5 software (Olympus).
Subcellular Fractionation and Immunoprecipitation-Proteins in cytosol and plasma membrane were prepared using Plasma membrane and cytoplasmic protein extraction kit (Beyotime, Jiangsu, China). Briefly, cells were harvested and then resuspended in 1 ml of buffer A. The nucleus and cell debris were removed from the homogenate by centrifugation at 700 ϫ g for 10 min at 4°C. The resulting supernatant was centrifuged at 14,000 ϫ g for 30 min at 4°C. The membrane pellet was solubilized in buffer B for 30 min at 4°C. Insoluble material was removed by centrifugation at 14,000 ϫ g for 10 min at 4°C. For immunoblotting, 60 g of proteins were resolved by 10% SDS-polyacrylamide gels and transferred onto nitrocellulose membranes. The membrane was incubated with polyclonal anti-Cav1.2 antibody (1:500, Sigma) overnight at 4°C, followed by incubation with horseradish peroxidase-conjugated secondary antibody for 2 h. The immunoreactive bands were visualized using an enhanced chemiluminescence kit (Pierce Thermo Scientific; Rockford, IL).
For immunoprecipitation (IP), INS-1 cells were washed twice with PBS and then lysed in chilled Nonidet P-40 buffer (1% Nonidet P-40, 150 mM NaCl, 50 mM Tris-HCl, pH 8.0) containing 100 M phenylmethylsulfonyl fluoride and Sigma protease inhibitor mixture, followed by incubation for 1 h at 4°C and subsequent centrifugation (12,000 rpm, 10 min at 4°C). The supernatant was then subjected to immunoprecipitation with anti-Cav1.2 or anti-Hap1 antibodies overnight at 4°C, and the immunocomplex was recovered by incubating with 20 l of 50% protein G-Sepharose beads for 1 h at 4°C followed by brief centrifugation. The immunoprecipitates were washed three times with lysis buffer. After the final wash, beads were resuspended in gel-loading buffer and boiled for 10 min. The lysates and immunoprecipitates were immunoblotted as described above.
Confocal Microscopy Imaging-Experiments were performed using the Olympus FV1000 laser scanning confocal microscopy system (Olympus, Tokyo, Japan). Coverslips containing INS-1 cells co-expressing VAMP2-pHluorin (PH-VAMP2) and mCherry-Hap1-scramble (mCh-Hap1-scramble) or mCherry-Hap1-siRNA (mCh-Hap1-siRNA) were placed into a perfusion chamber on a stage of an inverted Olympus IX71 microscope. Structural data acquisition was performed in the sequential line mode for the best spatiotemporal reliability. Live cells were examined in a confocal microscope at the cell footprint to better visualize near PM fluorescent spots. The sample area is scanned at resolution of 512 ϫ 512 pixels. pHluorin and mCherry were excited at 488 and 543 nm, respectively. Fluoview imaging software (Olympus, Tokyo) and Matlab software (The MathWorks, Inc.) were used for imaging data analysis. All experiments were performed at 22-25°C.
FRAP in Live Cells-INS-1 cells were seeded on coverslips and co-transfected with Hap1-siRNA or scramble-siRNA together with Cav1.2-YFP. Forty eight hours later, we mounted coverslips in the experimental chamber at 25°C. The culture medium was then replaced with standard bathing solution medium. Briefly, the fluorescence of YFP was assessed before and after photobleaching in a timed series. To bleach YFP, the laser was set to 100% for 488 nm, and 5 bleaches were taken. A total of 60 scans was performed at 2-s intervals after bleaching. Photobleaching was performed by full argon laser outputs and focused on the regions of interest. We collected images before and after bleaching and analyzed the recovered fluorescent intensity data. Fluorescence recovery traces were fitted with an exponential curve as shown in Equation 1, where A i is the amplitude of each component; t is time; and i is the time constant of each component.
Statistics Analysis-Statistical analyses were carried out using the SigmaPlot 11 software (Systat Software Inc.). All of the data indicate the mean Ϯ S.D., with sample number (n) referring to either coverslip or bouton number. Statistical significance was tested using either Student's t test or analysis of variance. Differences were considered statistically significant with a p value Ͻ0.05.
Author Contributions-J. P., J. H., and S. Y. conducted most of the experiments and analyzed the results. C. S. and X. L. conducted immunofluorescence experiments and Western blottings. T. Y. conducted radioimmunoassay for insulin secretion of Hap1 knockout mice. J. H. and H. L. conceived the idea for the project, designed the experiments, analyzed and interpreted the results, and wrote the paper with J. P.