SP6616 as a new Kv2.1 channel inhibitor efficiently promotes β-cell survival involving both PKC/Erk1/2 and CaM/PI3K/Akt signaling pathways

Kv2.1 as a voltage-gated potassium (Kv) channel subunit has a pivotal role in the regulation of glucose-stimulated insulin secretion (GSIS) and pancreatic β-cell apoptosis, and is believed to be a promising target for anti-diabetic drug discovery, although the mechanism underlying the Kv2.1-mediated β-cell apoptosis is obscure. Here, the small molecular compound, ethyl 5-(3-ethoxy-4-methoxyphenyl)-2-(4-hydroxy-3-methoxybenzylidene)-7-methyl-3-oxo-2,3-dihydro-5H-[1,3]thiazolo[3,2–a]pyrimidine-6-carboxylate (SP6616) was discovered to be a new Kv2.1 inhibitor. It was effective in both promoting GSIS and protecting β cells from apoptosis. Evaluation of SP6616 on either high-fat diet combined with streptozocin-induced type 2 diabetic mice or db/db mice further verified its efficacy in the amelioration of β-cell dysfunction and glucose homeostasis. SP6616 treatment efficiently increased serum insulin level, restored β-cell mass, decreased fasting blood glucose and glycated hemoglobin levels, and improved oral glucose tolerance. Mechanism study indicated that the promotion of SP6616 on β-cell survival was tightly linked to its regulation against both protein kinases C (PKC)/extracellular-regulated protein kinases 1/2 (Erk1/2) and calmodulin(CaM)/phosphatidylinositol 3-kinase(PI3K)/serine/threonine-specific protein kinase (Akt) signaling pathways. To our knowledge, this may be the first report on the underlying pathway responsible for the Kv2.1-mediated β-cell protection. In addition, our study has also highlighted the potential of SP6616 in the treatment of type 2 diabetes.

Type 2 diabetes mellitus (T2DM) is a chronic, complex and multifactorial metabolic disorder mainly characterized by hyperglycemia with insulin resistance and deficiency. T2DM has become a serious global health problem bringing heavy burdens to societies. 1 Currently, a series of anti-T2DM drugs are being clinically used, but their existing side effects are still triggering the urgent need for novel agents in the treatment of this disease. 1 Recently, accumulating evidence has revealed that pancreatic β-cell dysfunctions including glucose-stimulated insulin secretion (GSIS) defect and β-cell mass loss are major determinants for the progression from prediabetes with normoglycemia to diabetes with hyperglycemia, and the result that insulin resistance in prediabetes needs compensatory insulin hypersecretion likely leads to a progressive decline in islet β-cell function. 2 Therefore, an ideal strategy for T2DM treatment is to improve pancreatic β-cell function. 1,2 Numerous electrical signaling systems including K + , Na + , Ca 2+ and Cl À fluxes across β-cell membranes have been determined to participate in the function and/or survival of pancreatic β cells. 3 There are three major potassium fluxes in β cells, including K + effluxes regulated by voltage-gated K + (Kvs) or ATP-sensitive K + (K ATP ) channel and calciumactivated potassium channel (KCa 2+ ). 3 Kv2.1 as a voltagegated potassium (Kv) family member accounts for the majority of Kv currents in both rodent and human and negatively regulates GSIS. 4 In β cells, the ATP derived from glucose metabolism efficiently depolarizes β cells leading to the opening of voltage-gated ion channels. 3 Activated K + currents produce the repolarization of β-cell action potential resulting in the shutdown of voltage-dependent Ca 2+ channels (VDCCs), abolishment of VDCC-mediated Ca 2+ influx and blockage of insulin secretion. 3,5 Reports have demonstrated that Kv2.1 signaling regulation is involved in the apoptosis processes of neuron and β cells. 6,7 For example, Kv2.1 overexpression activates the mitochondrial or ER stress-induced apoptosis 6 and elevates the sensitivity of cells to apoptotic factors, 7 whereas transient expression of Kv2.1 function-deficient mutant avoids neuronal apoptosis. 7 Therefore, Kv2.1 channel is crucial to insulin secretion and/or β-cell apoptosis, and Kv2.1 inhibitors function potently in the promotion of insulin secretion and/or β-cell protection, [8][9][10] although the mechanisms underlying the regulation of β-cell protection still remain unclear.

Results
SP6616 is a Kv2.1 inhibitor SP6616 inhibited membrane potential in CHO-Kv2.1 cells: Given that the membrane potential-sensitive fluorescent dye is powerful for screening regulators of ion channels, 11 the membrane potential (FLIPR membrane potential assay kit) based platform by FlexStationII384 was at first applied to screen Kv2.1 inhibitor candidates against the lab compound library. As shown in Figure 1b, Kv2.1 inhibitor ScTx-1 (stromatoxin-1,100 nM) 12 obviously inhibited the membrane potential in CHO-Kv2.1 cells, indicating the efficacy of this platform in screening Kv2.1 inhibitor candidates.
Accordingly, SP6616 was discovered to be active in inhibiting membrane potential in CHO-Kv2.1 cells (Figure 1b) by IC 50 at 2.58 μM (Figure 1d). Moreover, the result that neither SP6616 (20 μM) nor ScTx-1 (100 nM) inhibited membrane potential in normal CHO cells further confirmed the inhibition of SP6616 against Kv2.1 channel (Figure 1c). In addition, SP6616 was also found to inhibit Kv2.2 channel by IC 50 (Figures 1f and g), in which ScTx-1 (100 nM) was used as a positive control (Figure 1e). Therefore, all results have determined that SP6616 was a Kv2 inhibitor with slight selectivity against Kv2.1 over Kv2.2.
SP6616 promoted GSIS: GSIS assay was conducted relating to the effect of SP6616 on insulin secretion. As shown in Figure 2a (ScTx-1 and glibenclamide as positive controls), SP6616 dosedependently activated insulin secretion in response to high concentration of glucose (16.8 mM) stimulation.
It is noted that the published reports indicated that Kv2.1N transfection in rat islet reduced approximately 60%-outward K + currents, 9,14 while in the current work, the effects of SP6616 were almost fully abolished in Kv2.1N-transfected cells. Such a discrepancy may be caused by the signal transduction from current blockage to insulin secretion or antiapoptosis in cells. Similarly, such a non-linear relationship between current blockage and insulin secretion has been also reported elsewhere. 9,10 Taken together, SP6616 was a new Kv2.1 inhibitor with dual effects on both insulin secretion promotion and β-cell protection.
Potentiation of SP6616 on GSIS links to glucosestimulated Ca 2+ influx. Considering that Kv channel activation can induce membrane repolarization and VDCCs closure further reducing insulin secretion and K V channel inhibition heightens intracellular Ca 2+ level and stimulates insulin secretion, 3,5 we next detected intracellular Ca 2+ level mediated by SP6616 in INS-832/13 cells. As shown in Figure 2h, either ScTx-1(100 nM) or SP6616 (10 μM) increased intracellular Ca 2+ level in the presence of 16.8 mM glucose. And such an intracellular Ca 2+ increase was blocked by depleting extracellular calcium in Hank's balanced salt solution (HBSS) buffer or by nifedipine (L-VDCC blocker) 15 (Figures 2i and j). These results thereby revealed that SP6616-stimulated Ca 2+ influx in response to high glucose, similar to the published K V channel inhibitionmediated GSIS event. 16 Ca 2+ influx/PKC/Erk1/2 and Ca 2+ influx/CaM/PI3K/Akt pathways are responsible for SP6616-mediated β-cell survival. Apoptosis is the process of programmed cell death, and regulated by a variety of extrinsic factors. 17 Although the signaling pathways in apoptosis are complicated, signaling of Erk1/2, p38, JNK, Akt or NFκB is determined to be vital in apoptosis and proliferation. 17,18 Therefore, we examined whether SP6616-mediated β-cell survival was implicated in any of those five signaling pathways in INS-832/13 cells. As demonstrated in Figures 3a and b, SP6616 reversed the STZ-induced decrease of either Erk1/2 or Akt phosphorylation, but rendered no effects on p38, JNK or NFκB phosphorylation (Supplementary Figure 2). Accordingly, we next investigated SP6616 protection against β cells by focusing on Erk1/2 and Akt signaling.    Figure 3), whereas the published results indicated that GFX exhibited no effects on p-Erk1/2 in the cells with treatment of glucose or IGF-1. 21 We here tentatively supposed that such a discrepancy may be due to the different experimental conditions. Taken together, Ca 2+ influx/PKC/Erk1/2 pathway was determined to be involved in the protection of SP6616 against STZ-induced β-cell apoptosis.
SP6616 regulated Akt and its downstream effectors Forkhead box protein O1 (FoxO1), X-linked inhibitor of apoptosis protein (XIAP) and Bad: Next, we investigated the regulation of SP6616 against Akt signaling in INS-832/13 cells. As shown in Figures 4a and b, SP6616 had no effects on Akt phosphorylation (p-Akt) but reversed the STZ-induced decrease in p-Akt. Notably, incubation of wortmannin (PI3K inhibitor) 22 in the cells caused the inactivity of SP6616 in recovering the STZ-reduced Akt phosphorylation (Figures 4c  and d). These results thereby implied the regulation of SP6616 against Akt signaling. In addition, to determine whether the SP6616-increased p-Akt was dependent on Kv2.1 regulation, the relevant assays in Kv2.1N-transfected INS-832/13 cells were performed. As illustrated in Figures 4e and f, the ability of SP6616 in reversing the STZ-decreased p-Akt weakened, this result thereby indicated that SP6616-stimulated Akt phosphorylation in a Kv2.1-dependent manner.
Given that the effectors involved in Akt-mediated antiapoptotic pathways mainly include FoxO1, Bad and XIAP in β cells, 23 we next examined the potential regulation of SP6616 against these three downstream proteins. As shown in Figures 4g-j, SP6616 reversed the STZ-induced decreases in phosphorylated FoxO1 (p-Ser256)/Bad (p-Ser136) and protein level of XIAP. Moreover, western blot results (Figures  4k-n) showed that wortmannin treatment could block all above SP6616-induced effects, thus addressing the dependence of the regulation against Akt in the signaling. Therefore, all results showed that both Ca 2+ influx/PKC/ Erk1/2 and Ca 2+ influx/CaM/PI3K/Akt signaling pathways were involved in SP6616-mediated β-cell protection.
PKC/Erk1/2 and CaM/PI3K/Akt pathways were required in parallel for SP6616 protection against β cells: As either PKC/Erk1/2 or CaM/PI3K/Akt pathway has been determined to be involved in the protection of SP6616 against β-cell apoptosis, we next examined whether these two signaling pathways were required for the SP6616-induced protection against the cells. MTT assay was at first carried out. As indicated in Figures 5a-c, treatment with either U0126 (Figure 5a) or wortmannin (Figure 5b) in the cells failed to deprive SP6616 of its capability in protecting cell viability against the STZ-induced apoptosis. However, co-incubation of both U0126 and wortmannin (Figure 5c) in the cells almost blocked such SP6616-induced protection. Moreover, the results in quantitative evaluation of apoptosis by Annexin V-FITC staining further confirmed that SP6616 attenuated STZ-induced apoptosis and co-incubation of both U0126 and wortmannin in the cells could block this attenuation (Figures 5d and e).
Therefore, all results implied that PKC/Erk1/2 and CaM/ PI3K/Akt pathways were required in parallel for the SP6616induced β-cell survival promotion as summarized in Figure 8e.
SP6616 ameliorates hyperglycemia in type 2 diabetic model mice. As SP6616 has been determined to promote GSIS and β-cell survival, we next examined its activity in amelioration of hyperglycemia on type 2 diabetic model mice.
In the assay, the model mice HFD/STZ and db/db were applied, and the male mice were administered with SP6616 (50 mg/kg/day) or vehicle by i.p. injection for 5 weeks. The results showed that SP6616 administration lowered the fasting blood glucose and glycated hemoglobin (HbA1c) levels (Figures 6a-d), and improved the glucose tolerance (Figures 6e-h) and insulin secretion during oral glucose tolerance test (OGTT; Figures 6i and j) in both models. SP6616 promotes insulin secretion and β-cell mass in type 2 diabetic model mice. Considering that SP6616 improved β-cell dysfunction by promoting insulin secretion and protecting β cell from apoptosis, we next evaluated the potential of this agent in stimulating plasma insulin content and insulin-positive islet mass in both the two diabetic mice. As expected, SP6616-treated groups possessed higher serum insulin levels (Figures 7a and b) and more insulinpositive islets compared with vehicle groups (Figures 7c-f).
SP6616 regulates Erk1/2 and Akt signaling in vivo. In view of the cell-based result that PKC/Erk1/2 and CaM/PI3K/ Akt pathways were responsible for SP6616-mediated β-cell New Kv2.1 channel inhibitor TT Zhou et al protection, we next evaluated SP6616 regulation against these two pathways in vivo. In the assay, the pancreatic tissues of both diabetic model mice were assayed by western blot against the key proteins involved in the pathways. As shown in Figures 8a-d, SP6616 administration in either model caused the increases in phosphorylated PKC, Erk1/2, Akt, FoxO1 and Bad and protein level of XIAP, totally consistent with the cell-based results, confirming the involvements of both Erk1/2 and Akt signaling in the protection of SP6616 against pancreatic β cells.

Discussion
Kv2.1 channel is widely expressed in mammalian tissues including cardiomyocytes, muscles, brain and pancreatic β cells. 27 As a major Kv family member, Kv2.1 channel contributes to 65-80% of the total Kv currents in human and rodent β cells. It has a crucial role in pancreatic β-cell membrane repolarization and its function. 3,4,9 The fact that Kv2.1 inhibition promotes insulin secretion in response to high glucose implies the possibility in avoiding the side effect of hypoglycemia. 28,29 Besides, Kv2.1 also functions potently in the regulation of cell apoptosis although the underlying mechanisms have not yet been unveiled. 7 Currently, several kinds of Kv2.1 inhibitors have been discovered. For example, peptide-type inhibitors include hanatoxin, guangxitoxin-1E, heteroscordratoxins, ScTx-1, SGTx1, syntaxin-1A, SsmTx-I and plasma gelsolin. [30][31][32][33][34][35][36] Small molecular inhibitors galantamine and isoliquiritigenin block Kv2.1 currents with little data on GSIS or cell apoptosis; 37,38 RY796 and C-1 enhance GSIS; 8,9 donepezil and 48F10 abolish neuronal apoptosis. 39,40 Previously, we reported natural product vindoline functioned in promotion of both insulin secretion and β-cell protection. 10 SP6616 is a new kind of small molecular Kv2 inhibitor with slight selectivity against Kv2.1 over Kv2.2 sharing totally different structure with the published inhibitors. Structurally, SP6616 possesses (1,3) thiazolo(3,2-a)pyrimidine scaffold whose derivatives are known to exhibit varied biological activities, including antiviral, anti-neoplastic, anti-bacterial and anti-inflammatory. 41 Our current work has further expanded the pharmacological applications of this kind of compound. To our knowledge, SP6616 and vindoline may be the only two small molecular Kv2.1 inhibitors able to both promote insulin secretion and survival. Moreover, SP6616 as a new Kv2.1 inhibitor effectively ameliorates β-cell dysfunction and improves glucose homeostasis in vivo. All these results have highlighted the potential of SP6616 in the treatment of type 2 diabetes.
It is accepted that activation of Kv channel can inhibit insulin secretion by inducing membrane repolarization and closure of VDCCs, and K V inhibition stimulates insulin secretion. 3,5 Here, we found that SP6616 as a Kv2.1 inhibitor effectively stimulated GSIS by following this underlying mechanism.
Ca 2+ is a ubiquitous cellular signaling molecule controlling a variety of cellular processes including cell survival. 42 PKC isoform as a downstream transducer of Ca 2+ participates in multifarious signaling pathways of biological processes including survival, proliferation, tumorigenesis and angiogenesis. 43 Erk1/2 is an important member of the MAPK family and has a critical role in pancreatic β cells, particularly in the regulation of proliferation and survival. 44,45 An increase of intracellular Ca 2+ can evoke PKC activation in triggering Erk1/2 stimulation. 19 CaM, a loop-helix-loop Ca 2+ -binding protein as another downstream transducer of Ca 2+ potently regulates multiple processes in eukaryotic cells, like proliferation and growth. 46 Besides, increase of intracellular-free Ca 2+ activates PI3K/Akt signaling via CaM in different cell lines. 24,25 PI3K/Akt pathway is known to promote survival of many cell lines, the anti-apoptotic targets of Akt signaling mainly include FoxO1, Bad and XIAP in β cells. 23 FoxO1 is a transcription factor regulating cellular processes like glucose metabolism, apoptosis, cell cycle regulation and DNA damage repair. 47 Phosphorylation of FoxO1 regulated by Akt promotes its nuclear exclusion and inhibits its pro-apoptosis function. 48 Besides, Akt inactivates the pro-apoptotic activity of Bad by mediating the phosphorylation at Ser136. 23 Akt has also been shown to promote cell survival by enhancing the stability of XIAP, 49 which is one of the conserved family of IAP that suppresses apoptosis by directly binding and inhibiting caspases activity. 50 Here, we have well determined the regulation of SP6616 against the STZ-reduced intracellular Ca 2+ and phosphorylation levels or protein levels of the related effectors such as PKC, Erk1/2, Akt, FoxO1, Bad and XIAP both in vitro and in vivo. All results have clearly expounded the potential mechanisms underlying SP6616 protection against β cells. To our knowledge, PKC/Erk1/2 and CaM/PI3K/Akt may be the first reported pathways linked to the regulation of Kv2.1-mediated β-cell protection. Interestingly, Bcl-2 has a central role in eukaryotic cell survival by inhibiting cell death, but Bcl-2 regulation is here probably not involved in the  Figure 4), which may be due to the insensitivity of Bcl-2 against this apoptotic event. 51 Given that Kv2.1 channel is also highly expressed in mammalian cardiomyocytes 27 and cardiotoxicity evaluation is vital for drug development, the potential effect of SP6616 on cardiac function in normal mice was also examined in the current work. As indicated in electrocardiography assay (Supplementary Figure 5), acute administration of SP6616 slightly prolonged QT intervals without affecting heart rates, which is consistent with the report that QT intervals are obviously prolonged without effect on heart rates in mice New Kv2.1 channel inhibitor TT Zhou et al expressing a dominant-negative Kv2 α subunit. 52 Our results imply that anti-diabetic drug development targeting SP6616 as a lead compound needs further investigation containing pharmacokinetics, pharmaceutics, drug toxicology and even structural modification.
In conclusion, we identified that small molecule SP6616 as a new Kv2.1 inhibitor effectively enhanced insulin secretion and protected β cells from apoptosis. It is determined that PKC/Erk1/2 and CaM/PI3K/Akt pathways are required in parallel for Kv2.1-mediated β-cell protection (Figure 8e).  Membrane potential assay. Cellular membrane potential was detected by FLIPR membrane potential assay kit (Molecular Devices, Sunnyvale, CA, USA) according to the instruction manual in CHO-Kv2.1 or CHO cells. Briefly, cells were plated into 96-well microplates and incubated overnight. After treating with compounds and membrane potential dye for 30 min, the plates were loaded into FlexStationII384 (Molecular Devices), followed by injecting 20 μM compounds and 100 mM KCl into the wells to generate the change of cellular membrane potential. The signals in each well were acquired for 120 s containing 20 s pre-injection basal reading at excitation wavelength of 530 nm and emission wavelength of 565 nm. These data were analyzed and shown as the area under the curve (AUC).
Electrophysiological recording assay. The whole-cell patch clamp recordings were performed using cultured CHO-Kv2.1 cells at room temperature with Axopatch-200B amplifier (Molecular Devices) as described previously. 53 The electrodes were pulled from borosilicate glass capillaries (1B150F-4; World Precision Instruments, Sarasota, FL, USA) by using Flaming/Brown type micropipette puller (P-97; Sutter Instrument, Novato, CA, USA). Pipettes had resistances of 3-7 MΩ when filled with a solution as following composition: 140 mM KCl, 2 mM MgCl 2 , 10 mM EGTA, 1 mM CaCl 2 , 10 mM HEPES (pH7.3). Cells were bath perfused with a solution of the following composition: 150 mM NaCl, 5 mM KCl, 0.5 mM CaCl 2 , 1.2 mM MgCl 2 , 10 mM HEPES (pH7.3). The signals were filtered at 1 kHz, digitized using a DigiData 1440 A (Molecular Devices), and analyzed with the software of pClamp 10.2 (Molecular Devices). Whole-cell currents were recorded using the protocol as follows: the holding potential was set at -80mV, and stepwise depolarized from − 80 to 120 mV in 20 mV increments and then repolarized to − 60 mV. Initially, we made compensation to get rid of pipette resistance during whole-cell patching, and then we made 60-80% compensation of both series resistance and capacitance of cell bodies to avoid voltage deviations (for the detailed procedures, see also Supplementary Information), especially when the current amplitude is large. For large cells, we further perform the correction offline by recording much of empty control CHO cells. 54 Liquid junction potentials were o2 mV, which were calculated using JPCalc software. 55 GSIS assay. GSIS assay was carried out according to the published approach. 13 INS-832/13 cells were plated into 24-well plates. After pre-incubated  MTT assay. MTT assay was performed according to previously described. 13 INS-832/13 cells were plated into 48-well plates and incubated with different concentrations of SP6616 and STZ (0.4 mM) for 24 h (unless indication, STZ concentration was 0.4 mM and incubation time was fixed at 24 h throughout this current work).
Flow cytometry assay. INS-832/13 cells were plated into six-well plates and incubated with corresponding compounds as indicated before collecting. Western blot and immunohistochemistry assays. Western blot assays were performed as previously described. 56 Cell or tissue lysate was separated by SDS-PAGE and transferred to nitrocellulose membrane (GE Healthcare, Madison, WI, USA). After incubation with the corresponding antibodies, membranes were visualized using the West-Dura detection system (Thermo Scientific, Waltham, MA, USA). The signal was collected by ImageQuant LAS 4000 mini (GE Health, USA). Immunohistochemistry assay of pancreas was performed as previously described. 10 Animal experiments. All animals received humane care and were raised at a relative humidity of 50% with a 12-h light-dark cycle at 20-25°C and given ad libitum access to water and food. The animal-relevant protocols were approved by the Institutional Animal Care and Use Committees at Shanghai Institute of Materia Medica. HFD/STZ-induced type 2 diabetic mice were constructed as described. 13,57 Briefly, 6-week-old C57/BL6 male mice were intraperitoneally injected with STZ (25 mg/kg/ day) continuously for 5 days after feeding with HFD containing 58% fat for 4 weeks. To select diabetic mice, 6-h fasting plasma glucose was measured in the STZ-injected mice after 3 days. db/db male mice (BKS.Cg-Dock7 m +/+ Lepr db /J) were from Jackson Laboratory (Sacramento, CA, USA). Both diabetic mice were assigned randomly to two groups by glucose level and body weight (n = 8). Vehicle (2% DMSO and 8% Tween 80 dissolved in saline) or SP6616 (50 mg/kg/day) was administrated daily by intraperitoneal injection for 5 weeks. Fasting blood glucose level from 6-h fasted mice was measured weekly. OGTT (1.5 g/kg) was carried out on diabetic mice after fasted overnight at the fourth week. Glucose level was measured from tail blood at 0, 15, 30, 45, 60, 90 and 120 min. Meanwhile, the insulin release during OGTT was also detected. Blood sample was obtained from tail veins and serum insulin concentration was determined by AlphaLISA insulin kit (PerkinElmer). At the termination of the study, mice were killed and tissues were analyzed. Data analysis. Data were shown as means ± S.E.M. Two-tailed unpaired t-test was performed for comparison of two groups and one-way ANOVA analysis for 42 groups by GraphPad Prism 5 software (GraphPad Software, La Jolla, CA, USA). Significant differences were shown as *Po0.05; **Po0.01; ***Po0.001.