Inhibition of serum and glucocorticoid regulated kinase-1 as novel therapy for cardiac arrhythmia disorders

Figures

• 

  Figure 1. Computer-Aided Drug Discovery for SGK1 inhibitors. (A) Virtual model for SGK1 used for CADD. (B) SGK inhibitors were R-substituted-methylene-2-thioxodihydropyrimidine-4,6(1H,5H)-dione (left) derivatives with the two leading candidate structures shown (right). (C) Dose response curve for lead inhibitor 5377051 (in presence of 1 ng of recombinant SGK1). Dilutions of the inhibitor were added to recombinant SGK1 and SGK1 activity was assayed using a fluorescence polarization assay in triplicates.

• 

  Figure 2. Inhibition of SGK1 in cultured cardiomyocytes by lead compounds. (A) Inhibition of SGK1 activity assessed by GSK3β phosphorylation in CMs infected by Ad.SGK1-CA. (B) Different concentrations of 5377051 were assessed at 48 hours after treatment with immunoblotting (left) and quantitated (right). (C). Effect of SGK1 inhibitors on Akt-induced phosphorylation of GSK3-beta in CMs infected by Ad.-myr-Akt was assessed by immunoblotting. (D) Comparison of potency of 10 micromolar compounds 5377051, 6410136 with the published inhibitor EMD63863. Immunoblots were cut to facilitate the incubation the same samples with multiple antibodies and cropped for clarity. All samples were loaded contiguously as shown. Quantified data are the composite of three independent experiments, each with three technical replicates. Each experimental set was conducted with the same batch of adenovirus.

• 

  Figure 3. Validation of lead SGK1 inhibitor compound in a prostate cancer cell line. (A) Western blot demonstrating inhibition of SGK1 by compound 5377051 blocked the phosphorylation of SGK1 target NDRG-1 in response to the androgen analog R1881. (B) Quantification of NDRG-1 phorsphorylation. (C) 5377051 inhibited the proliferation of LNCaP prostate cancer cell line induced by the androgen analog R1881 as assayed by Cy-quant assay. Immunoblots were cut to facilitate the incubation the same samples with multiple antibodies and cropped for clarity. All samples were loaded contiguously as shown.

• 

  Figure 4. SGK1 regulates NaV1.5 channel function. (A) Expression of SGK1-CA in HEK-NaV1.5 cells leads to increase in INa density. Inset scale = vertical: 50 pA/pF, horizontal: 2 ms. (B) SGK1 activation leads to a hyperpolarizing shift of steady state activation and inactivation curves for INa. (C) Peak sodium currents are increased with expression of SGK1-CA while inhibition of SGK1 by the dominant negative kinase dead SGK-1 (SGK1-DN) leads to decrease in INa density (*p-value < 0.005 by one way ANOVA, n = 5–11 for each condition). (D) Mutation of a novel SGK1 putative site T1590A leads to abrogation of the increase in INa with SGK1 activation. (E) Quantification of peak sodium currents (NS = non-significant change, n = 3–5 for each condition).

• 

  Figure 5. Inhibition with lead compound 5377051 modulates INa by inhibiting SGK1. (A) Addition of 5377051 at 10 uM significantly decreases INa even with co-expression of SGK1-CA. (B) Quantification of changes in peak sodium current with addition of inhibitor 2 (5377051) with and without co-expression of SGK1-CA (n = 4–6 per condition, *p-value < 0.05). (C) Representative time course of inhibition by compound 5377051 as compared to addition of DMSO. (D) Average time for inhibition demonstrating a half-time of 200 s (n = 4).

• 

  Figure 6. Inhibition of SGK1 in cardiomyocytes selectively reduces the late sodium current INaL. (A) Representative INa from NRVMs at baseline and after acute treatment with compound 5377051 demonstrates preferential reduction of late current. (B) Quantification of the peak current does not demonstrate a significant reduction with administration of the inhibitor (Peak current density: Pre = −560.1 ± 180 pA/pF versus Post = −396.1 ± 230.4 pA/pF). (C) Selective inhibition of native late sodium current with incubation of the lead compound (Late current density: Pre = 5.12 ± 2.4 pA/pF versus Post = −2.3 ± 1.5 pA/pF). (D) Relative decrease in peak to late current ratio. (*p-value < 0.05, †p-value = non-significant, n = 6 for all experiments).

• 

  Figure 7. Treatment with SGK1 inhibitor 5377051 restores prolonged APD in a cellular model of LQT3. (A) Representative fluorescent tracings of spontaneous action-potentials from wild-type hiPSC-CMs expressing ArcLight with or without the SGK1 inhibitor 5377051 quantified by measuring the action potential duration at 80% repolarization (APD80) (WT + DMSO, 244 ± 9 ms, n = 15 versus WT + 5377051, 235 ± 8 ms, n = 11) (B) Fluorescent action potential recordings from LQT3 hiPSC-CMs with 5377051 demonstrate significant shorting of the APD80 (LQT3 + DMSO, 278 ± 7 ms, n = 22 versus LQT3 + 5377051, 218 ± 10 ms, n = 16). (C) Cumulative APD data from WT and LQT3 hiPSC-CMs with and without SGK1 inhibitor. Statistics was performed by applying a one-way ANOVA with Bonferroni’s post hoc correction: *P < 0.05, **p < 0.01 and ***P < 0.001, versus LQT3 + DMSO.

• 

  Figure 8. SGK1 inhibition rescues a zebrafish model of long QT. (A) Still frame image of zebrafish heart with chamber labels (A = atrium) (V = ventricle). Representative tracings of chamber wall position demonstrating 2:1 AV block in mutant bkd −/− zebrafish (upper panel) and with rescue of the phenotype (lower panel). (B) Knock-down with SGK1 morpholino or inhibition of SGK1 (500 μM SGK1-DN mRNA) in the breakdance zebrafish mutants with long QT at 1–2 cell stage leads to significant rescue of 2:1 AV block at 99 hpf stage (5% spontaneous reversion at this stage); *p-value < 0.05 by Z-test. (C) Compound 5377051 added at 48 hours post fertilization (hpf) and assessed at 72 hpf for rescue of phenotype 2 with superimposed rescue (solid line) and toxicity (dashed lines) curves shown on left (C) and representative bar graph on right (D). *p < 0.05 by Z-test.