Kv2.1 Clustering Contributes to Insulin Exocytosis and Rescues Human β-Cell Dysfunction
Article Figures & Tables
Figures
- Figure 1
Kv2.1, but not Kv2.2, controls exocytosis in human β-cells. A: Expression of mRNA encoding Kv2.1 (KCNB1) and Kv2.2 (KCNB2) in islets from human donors assessed by quantitative PCR (n = 11 donors). B: Knockdown of KCNB1 and KCNB2 expression in human islet cells, assessed by quantitative PCR, following transfection with control siRNA duplexes (si-Scrambled) or siRNAs targeting Kv2.1 (si-Kv2.1) or Kv2.2 (si-Kv2.2) (n = 6 donors). Representative traces (C) and averaged current–voltage relationships (D) of Kv currents recorded from human β-cells following transfection with si-Scrambled (gray squares), si-Kv2.1 (black circles), si-Kv2.2 (gray circles), or both (black squares) (n = 20, 13, 21, and 17 cells from 4 donors, respectively). Representative capacitance traces (E) and averaged cumulative exocytotic responses (F) of human β-cells to a series of membrane depolarizations following transfection with si-Scrambled, si-Kv2.1, or si-Kv2.2 (n = 23, 32, and 31 cells from 5 donors, respectively). *P < 0.05; **P < 0.01; ***P < 0.001 compared with the Kv2.1 group (A) or scrambled control (B–F).
- Figure 2
Kv2.1 forms clusters in insulin-secreting cells. Human β-cells (A) or INS 832/13 cells (B) were immunostained with anti-Kv2.1 (red) and anti-insulin (green) antibodies and visualized by TIRF microscopy (representative of 23 cells from 3 donors and 26 cells in 3 experiments, respectively). C: INS 832/13 cell expressing mCherry-Kv2.1-WT (red) and NPY-Venus (green; 33 cells in 3 experiments). Scale bars, 10 μm.
- Figure 3
Superresolution imaging of Kv2.1 clusters. A: PALM images of HEK 293 and INS 832/13 cells expressing PAmCherry-Kv2.1-WT (red). Inset zoom-in area corresponds to the region of interest (ROI). Scale bars, 5 μm. B: PAmCherry-Kv2.1-WT molecule coordinate plots for ROIs in A, with cluster regions determined by spatial pattern analysis, highlighted in red in the background. The percentage of PAmCherry-Kv2.1-WT molecules found within clustered domains (C) and the average number of molecules per cluster (D) in HEK 293 cells (n = 22) and INS 832/13 cells (n = 29). E: Extracellular cross-linking with DTSSP followed by blotting of protein lysates with anti-Kv2.1 antibody demonstrates that the tagged (PAmCherry-Kv2.1-WT) channels form large-molecular-weight complexes at the cell surface, consistent with channel clustering. Breakdown of cross-links with 2-ME reveals the expression of channel monomers (representative of n = 6 experiments).
- Figure 4
A clustering-deficient Kv2.1 reduces secretory granule recruitment to the plasma membrane. A: mCherry and Myc-tagged clustering-deficient Kv2.1 channels were generated by truncating the final 318 amino acids (mCherry/Myc-Kv2.1-ΔC318) of the rat sequence. B–D: When expressed in HEK 293 (B; n = 6 experiments) or INS 832/13 cells (C; n = 10 experiments), the Myc-Kv2.1-ΔC318 formed fewer high-molecular-weight clusters, which could be broken down to channel monomers by 2-ME. Although Myc-Kv2.1-ΔC318 retained some ability to form clusters in INS 832/13 cells, likely because of combination with native Kv2.1 (Supplementary Fig. 5), quantification (D; n = 6 and 10) revealed that the majority of the signal remains tetrameric (i.e., single-channel rather than cluster). Additional assessment of the clustering of these constructs is presented in Supplementary Fig. 4. E and F: When expressed in INS 832/13 cells and visualized by TIRF microscopy, mCherry-Kv2.1-ΔC318 forms fewer clusters than mCherry-Kv2.1-WT and results in fewer membrane-resident secretory granules, marked by NPY-Venus (n = 37; 62 cells from 5 experiments). Scale bars, 10 μm. ***P < 0.001 compared with mCherry-Kv2.1-WT.
- Figure 5
Clustering-deficient Kv2.1 retains electrical activity and syntaxin 1A binding, but does not facilitate exocytosis. Representative Kv currents (A) and quantified current-voltage relationships (B) from INS 832/13 cells expressing GFP alone, Myc-Kv2.1-WT, or Myc-Kv2.1-ΔC318 (n = 17, 19, and 18 cells, respectively). C and D: Coimmunoprecipitation (IP) of syntaxin 1A (Syn1A) pulled down both Myc-Kv2.1-WT and Myc-Kv2.1-ΔC318 equally well (n = 7 experiments). Representative capacitance traces (E) and cumulative exocytotic responses (F) of INS 832/13 cells expressing GFP alone, Myc-Kv2.1-WT, or Myc-Kv2.1-ΔC318 (n = 15, 22, and 17 cells, respectively). *P < 0.01 as indicated; ***P < 0.001 versus GFP control.
- Figure 6
Kv2.1 and a role for the channel in granule recruitment. A: Knockdown of Kv2.1 protein in Ad-shKv2.1–infected INS 832/13 cells compared with control subjects (Ad-shScrambled; n = 4 experiments). B and C: Knockdown of Kv2.1 in INS 832/13 cells resulted in a decreased density of membrane-resident secretory granules marked with NPY-Venus and visualized by TIRF microscopy at both 1 and 16.7 mmol/L glucose. Representative images (B) and quantified data (C) are shown (n = 30, 30, 27, 30, 28, and 30 cells in 3 experiments). Syn1A, syntaxin 1A. Scale bar, 10 μm. **P < 0.01; ***P < 0.001 compared with 1 mmol/L glucose control or as indicated.
- Figure 7
Kv2.1 clusters promote secretory granule recruitment. Representative images (A and B) and quantified data (C and D) of INS 832/13 cells expressing mCherry-Kv2.1-WT (red) or mCherry-Kv2.1-ΔC318 (red) and NPY-Venus (green) at 1 mmol/L glucose and after 15 or 30 min of 16.7 mmol/L glucose, visualized by TIRF microscopy. Glucose stimulation increased the density of membrane-resident secretory granules (C) in the mCherry-Kv2.1-WT group (black bars), but not the mCherry-Kv2.1-ΔC318 group (gray bars), the latter of which had fewer channel clusters (D). Scale bars, 10 μm. *P < 0.05; **P < 0.01 compared with the 1 mmol/L glucose condition, or as indicated.
- Figure 8
Upregulation of Kv2.1 in human T2D β-cells improves exocytotic function. Knockdown of Kv2.1 (A; KCNB1; n = 13 and 13 cells from 2 donors) or Kv2.2 (B; KCNB2; n = 19 and 23 cells from 3 donors) expression failed to reduce voltage-dependent K+ currents in β-cells from donors with T2D. C: Expression of Kv2.1 (KCNB1) mRNA in islets from ND donors or donors with T2D (n = 11 and 5 donors, respectively). Representative capacitance responses at 10 mmol/L glucose from untransfected (D) β-cells of ND donors and donors with T2D and of transfected β-cells (E) of donors with T2D expressing GFP alone or together with Kv2.1-WT. F: The cumulative exocytotic responses from D and E (n = 30 cells from 3 ND donors and n = 30, 32 and 34 cells from 3–5 donors with T2D). G: Images from live-cell TIRF of ND and T2D β-cells expressing mCherry, mCherry-Kv2.1-WT, or mCherry-Kv2.1-ΔC318 and the granule marker NPY-EGFP (greyscale). Red dots indicate exocytotic events occurring over 2 min upon increasing glucose from 2.8 to 5 mmol/L. Scale bars, 5 μm. Average frequency of exocytotic events in ND (H; n = 17, 20, and 21 cells from 2 donors) and T2D (I; n = 24, 34, and 30 cells from 2 donors with T2D) β-cells. J and K: Insulin secretory profiles of islets from a donor with T2D transduced with control adenovirus (Ad-GFP; circles) or an electrically silent full-length Kv2.1 (Ad-Kv2.1W365C/Y380T; squares). The first phase secretory response is shown on expanded time scale (K) (one donor; experiment run in duplicate). L: Averaged and individual (circles) area under the curve (AUC) values for baseline, first-phase, and second-phase responses of islets from donors with T2D transduced as in J and K (n = 3 donors in duplicate). *P < 0.05 versus ND or as indicated; **P < 0.01 versus mCherry; ***P < 0.001 as indicated.
In this Issue
