Synthesis and Development of a Novel First-in-Class Cofilin Inhibitor for Neuroinflammation in Hemorrhagic Brain Injury

Abstract

Intracerebral hemorrhage (ICH) is devastating among stroke types with high mortality. To date, not a single therapeutic intervention has been successful. Cofilin plays a critical role in inflammation and cell death. In the current study, we embarked on designing and synthesizing a first-in-class small-molecule inhibitor of cofilin to target secondary complications of ICH, mainly neuroinflammation. A series of compounds were synthesized, and two lead compounds SZ-3 and SK-1-32 were selected for further studies. Neuronal and microglial viabilities were assessed by 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay using neuroblastoma (SHSY-5Y) and human microglial (HMC-3) cell lines, respectively. Lipopolysaccharide (LPS)-induced inflammation in HMC-3 cells was used for neurotoxicity assay. Other assays include nitric oxide (NO) by Griess reagent, cofilin inhibition by F-actin depolymerization, migration by scratch wound assay, tumor necrosis factor (TNF-α) by enzyme-linked immunosorbent assay (ELISA), protease-activated receptor-1 (PAR-1) by immunocytochemistry and Western blotting (WB), and protein expression levels of several proteins by WB. SK-1-32 increased neuronal/microglial survival, reduced NO, and prevented neurotoxicity. However, SZ-3 showed no effect on neuronal/microglial survival but prevented microglia from LPS-induced inflammation by decreasing NO and preventing neurotoxicity. Therefore, we selected SZ-3 for further molecular studies, as it showed potent anti-inflammatory activities. SZ-3 decreased cofilin severing activity, and its treatment of LPS-activated HMC-3 cells attenuated microglial activation and suppressed migration and proliferation. HMC-3 cells subjected to thrombin, as an in vitro model for hemorrhagic stroke, and treated with SZ-3 after 3 h showed significantly decreased NO and TNF-α, significantly increased protein expression of phosphocofilin, and decreased PAR-1. In addition, SZ-3-treated SHSY-5Y showed a significant increase in cell viability by significantly reducing nuclear factor-κ B (NF-κB), caspase-3, and high-temperature requirement (HtrA2). Together, our results support the novel idea of targeting cofilin to counter neuroinflammation during secondary injury following ICH.

Keywords: anti-inflammatory agent; brain injury; cofilin inhibitor; microglial activation; neuroinflammation.

Figure 1:

Ligand-binding site complex. (a) Three-dimensional view of cofilin structure (PDB ID 4BEX) shows the predictive potential ligand-binding sites. (b) Docked structure of lead compound into the C-terminal crystal structure of cofilin. (c) Docked structure of compound SZ-3 into the C-terminal crystal structure of cofilin shows better fit into the groove proposed in the study.

Figure 2:

Neuronal survival after treatment with synthesized compounds. Neuroblastoma (SHSY-5Y) cells were treated with different concentrations (0.5, 1, 5, 10, 20 and 50 μM) of (A) SK-1-32 and (B) SZ-3 for 24h and cell viability was measured by MTT-assay. SK-1-32 significantly increased proliferation of SHSY-5Y cells at 1, 5 and 10 μM whereas SZ-3 did not show any effect. Data expressed as mean±SEM, where the ^*/#p <0.05; **^/##p<0.01; ***^/###p<0.001 was considered significant. (n=3 independent experiments, one-way ANOVA followed by Dunnett post hoc comparisons). ^#decease compared to control, *increase compared to control.

Figure 3:

Microglial survival after treatment with synthesized compounds. HMC-3 cells were treated with different concentrations (0.5, 1, 5, 10, 20 and 50 μM) of (A) SK-1-32 and (B) SZ-3 for 24h and cell viability was measured by MTT-assay. SK-1-32 significantly increased survival of HMC-3at 10, 20 and 50 μM whereas SZ-3 increased only at 5 μM. Data expressed as mean±SEM, where the *^/#p<0.05; **^/##p<0.01; ***^/###p<0.001 was considered significant. (n=3 independent experiments, one-way ANOVA followed by Dunnett post hoc comparisons). ^#decease compared to control, *increase compared to control.

Figure 4:

Microglial survival after LPS-induced inflammation and treatment with synthesized compounds. HMC-3 cells were challenged with LPS (200 ng/mL) and treated with different concentrations (0.5, 1, 5, 10, 20 and 50 μM) of (A) SK-1-32 and (B) SZ-3 for 24h. Cell viability was measured by MTT-assay. SK-1-32 significantly reduced HMC-3 cell viability whereas SZ-3 significantly increased at 10 and 20 μM. Data expressed as mean±SEM, where the *^/#p< 0.05, **^/##p<0.01; ***^/###p<0.001 was considered significant (n=3 independent experiments, one-way ANOVA followed by Dunnett post hoc comparisons). ^#Difference between the control and LPS. *Difference within different concentrations relative to LPS.

Figure 5:

LPS-induced NO release and treatment with synthesized compounds. HMC-3 cells were challenged with LPS (200 ng/mL) and/or with LPS (200 ng/mL) and different concentrations (0.5, 1, 5, 10, 20 and 50 μM) of (A) SK-1-32 and (B) SZ-3 for 24h. Nitric oxide was analyzed using Griess reagent. SK-1-32 significantly reduced NO at 5, 10, 20 and 50 μM. SZ-3 significantly reduced the NO release from 05 to 50 μM concentrations. Data expressed as mean±SEM, where the *^/#p< 0.05, **^/##p<0.01; ***^/###p<0.001 was considered significant (n= triplicates of one experiment, one-way ANOVA followed by Dunnett post hoc comparisons). #Difference between the control and LPS. *Difference within different concentrations relative to LPS.

Figure 6:

LPS-induced neurotoxicity and effect of synthesized compounds. HMC-3 cells were challenged with LPS (200 ng/mL) and/or treated with different concentrations (0.5, 1, 5, 10, 20 and 50 μM) of (a) SK-1-32 and (b) SZ-3 for 24h. Next, the conditioned media from HMC-3-LPS challenged cells was transferred to the SHSY-5Y and incubated for 24 h. Cell viability was measured by MTT-assay. SK-1-32 significantly increased SHSY-5 survival at 0.5 to 20 μM and reduced at 50 μM. SZ-3 significantly increased from 0.5 to 5 μM concentrations. Data expressed as mean±SEM, where the (*^/#p< 0.05, **^/##p< 0.01, ***^/###p< 0.001) was considered significant (n= triplicates of one experiment, one-way ANOVA followed by Dunnett post hoc comparisons). #Difference between the control and LPS. *Difference within different concentrations relative to LPS.

Figure 7:

Distribution of lengths of actin filaments labeled with SiR-Actin probe and obtained from fluorescent microscope at 652 nm. (A) Polymerized F-actin divided into groups representing control (actin alone), vehicle (cofilin+ DMSO), SZ-3 5 μM (alone), SZ-3 10 μM (alone), Cofilin+ SZ-3 5 μM, and Cofilin+ SZ-3 10 μM. (B) Quantitative analysis shows no changed in the lengths of actin filaments with treatment of SZ-3 (5 or 10 μM) alone in comparison with control (actin alone) (n=3). (C) Cofilin caused severe disorganization of actin filaments in vehicle (DMSO). Actin filaments containing cofilin and treated with SZ-3 (5 or 10 μM) was significantly longer in comparison with vehicle. The length of SZ-3 5 μM was significantly longer than the 10 uM treated group (n=3). Data expressed as mean±SEM, where the Significant levels are presented as, *p<0.05 ^**** p<0.001

Figure 8:

Expression levels of NF-κB, cofilin and phosphocofilin in HMC3 microglia challenged with thrombin. (A) WB analysis shows a significant increase in NF-κB expression in vehicle (DMSO) exposed to thrombin in comparison to control. However, NF-κB expression level was significantly decreases with treatment of 5 uM and 10 μM of SZ-3 in comparison with vehicle (n=3). (B) Total cofilin level does not show any significant change in all groups (n=6). (C) P-cofilin level was significantly up-regulated with treatment of 5 μM and 10 μM of SZ-3 in comparison with vehicle (n=6). Data expressed as mean±SEM, where the significant levels are presented as *p<0.05, ^**p<0.01)

Figure 9:

PAR-1 protein expression in microglial after SZ-3 treatment. (A) HMC3 stained with DAPI (blue) and PAR-1 (red) (n=3). (B) PAR-1 were significantly increased in vehicle (thrombin+DMSO) 24 h after thrombin exposure as compared to control. PAR-1 protein expression in SZ-3 treated cultures (10 and 5μM) were significantly decreased in comparison to vehicle and no significant difference as compared control (n=5). Data expressed as mean±SEM, where the significant levels are presented as, * p<0.05.

Figure 10:

LPS-induced TNF-α and and SZ-3 antiinflammatory properties. (A) In ELISA assay, LPS challenged vehicle (DMSO) and increased SZ-3 treated cultures 10 and 5μM, significant increase TNF-α level as compared to control. TNF-α level was significantly decreased in SZ-3 treated cultures 5 μM (*as compared to vehicle (n=3). (B) Cells viability changes was not significant for all groups (n=6). Data expressed as mean±SEM, where the significant levels are presented as, *p<0.05 and ^***p<0.001

Figure 11:

Scratch wound assay to test the effect of SZ-3 on microglial migration. (a) Microglial cells migrated to the scratched area were imaged at 2 h intervals for 72 h. (b) The percentage of wound area occupied by migrated cells. SZ-3 ( 5 μM) suppresses microglial migration significantly at 12, 24, 46, 58 h as compared to vehicle (DMSO), and not significantly in comparison to control (n=3). Vehicle cell migration was increased significantly at 24, 46, 58 h as compared to control (n=3). Data expressed as mean±SEM, where the Significant levels are presented as, ^**p<0.01 and ^**** p<0.0001

Figure 12:

Neuroprotective properties of SZ-3. (A) SZ-3 effect on SH- SY5Y cell viability first tested alone by MTT assay, and the change in cell viability was not significant compared to DMSO- treated cultures (vehicle) (n=3). (B) In other MTT assay, the cells viability of thrombin challenged SH-SY5Y treated with DMSO (vehicle) was significantly reduced as compared to control (n=3). However, SZ-3 treated cultures with 20, 10 and 5 μM significantly increases the cell viability as compared to vehicle, but not significant in comparison to control (n=3). Data expressed as mean±SEM, where the significant levels are presented as, *p<0.05, ^**p<0.01 and ^**** p<0.0001

Figure 13:

Effects of SZ-3 on thrombin-induced apoptosis in SH-SY5Y cells. Western blot analysis of SH-SY5Y cells challenged with thrombin then treated after 3 h with vehicle (DMSO) or SZ-3 (10 and 5 μM). (A) NF-κB was significantly increased in vehicle as compared to control. (10 and 5 μM) was found to significantly decrease NF-κB as compared to vehicle, and not significant to control (n=5). (B) HtrA2 was significantly increased in vehicle as compared to control. (10 and 5 μM) was found to significantly decrease HtrA2 as compared to vehicle, and not significant to control (n=3). (C) Caspase 3 was significantly increased in vehicle as compared to control. (10 and 5 μM) was found to significantly decrease caspase 3 as compared to vehicle, and not significant to control (n=4). Cofilin level was not significantly changed in all groups (n=4). (E) However, the ratio p-cofilin/cofilin decreased significantly in vehicle as compared to control. 5μM was found to significantly decrease the ratio of p-cofilin/cofilin as compared to vehicle, and not significant to control (n=4). Data expressed as mean±SEM, where the Significant levels are presented as, *p<0.05, ^**p<0.01 and ^**** p<0.0001

Figure 14:

Molecular optimization strategy for SZ-3. The binding affinity of the compound was increased by adding a functional group in the para position of aminocyclohexane to anchor the molecule in the deep groove of the potential binding pocket and predicted Gibbs free energy (ΔG) of binding was kept as low as possible.

Figure 15:

The proposed hypothesis of SZ-3 inhibition of cofilin during ICH. SZ-3 binds and inhibits cofilin from binding to F-actin or its translocation to mitochondria and initiation of neuronal apoptosis, and thus, makes cofilin free in the cytosol for subsequent phosphorylation by LIMK. Additionally, SZ-3 inhibits cofilin binding to RelA/p65 and its translocation to the nucleus, attenuating microglial activation.

Scheme 1:

Synthesis of the designed Cofilin inhibitors