HDAC/σ1R Dual-Ligand as a Targeted Melanoma Therapeutic

Abstract

Background: In melanoma, multiligand drug strategies to disrupt cancer-associated epigenetic alterations and angiogenesis are particularly promising. Here, a novel dual-ligand with a single shared pharmacophore capable of simultaneously targeting histone deacetylases (HDACs) and sigma receptors (σRs) was synthesized and subjected to phenotypic in vitro screening. Methods: Tumor cell proliferation and spreading were investigated using immortalized human cancer and normal cell lines. Angiogenesis was also evaluated in mouse endothelial cells using a tube formation assay. Results: The dual-ligand compound exhibited superior potency in suppressing both uveal and cutaneous melanoma cell viability compared to other cancer cell types or normal cells. Melanoma selectivity reflected inhibition of the HDAC-dependent epigenetic regulation of tumor proliferative kinetics, without involvement of σR signaling. In contrast, the bifunctional compound inhibited the formation of capillary-like structures, formed by endothelial cells, and tumor cell spreading through the specific regulation of σ₁R signaling, but not HDAC activity. Conclusions: Together, the present findings suggest that dual-targeted HDAC/σ₁R ligands might efficiently and simultaneously disrupt tumor growth, dissemination and angiogenesis in melanoma, a strategy amenable to future clinical applications in precision cancer treatment.

Keywords: HDACi; angiogenesis; cancer proliferation; cell spreading; dual-ligands; melanoma; σ1 receptor.

Figure 1

Chemical scaffolds and molecular design for dual-function hybrid 5c.

Scheme 1

Synthesis of the dual-ligand compound 5c (5) and its precursor 3c (3). All reactions were performed at room temperature (RT). Principal reagents employed: (i) MeI, DMF, RT; (ii) AcOH, NaBH(OAc)₃, THF, RT; (iii) 1. LiOH 1 M, MeOH, RT; 2. ClCO₂Et, TEA, THF, RT; 3. NH₂OTHP, THF, RT; (iv) HCl 1.25 M.

Figure 2

Characterization of HDACi and the anticancer pharmacology by the dual-target hybrid 5c. (a) Dose–response curve of 5c on total HDAC activity in HCT116 cells. HDAC enzymatic activity was measured as described in Materials and Methods. (b) Antiproliferative dose–response curves of 5c in UM 92-1 (3, 10, 30, 100 and 300 nM) and MRC-5 (30, 100, 300, 1000 and 3000 nM) cells. (c) Antiproliferative dose–response curves of 5c (1, 3, 10, 30 and 100 nM) in various human cancer and normal cell lines. In (b,c), treatments were performed for 48 h, and proliferation was assessed using the crystal violet assay and spectrophotometry (with absorbance at 590 nm). Data were expressed as the percentage of the vehicle control (DMSO) and organized on a log scale. (d) Heat map representation of antiproliferative IC₅₀ values calculated for 5c in each of the five cancer cell lines investigated.

Figure 3

Assessment of the pharmacological antiproliferative target for 5c in human melanoma cells. (a) Proliferation of UM 92-1 cells exposed to 5c (30 nM) alone or in combination with selective σ₁R agonist pentazocine (PTZ, 2 μM) or σ₂R antagonist AC927 (2 μM), and 3c (employed at 30 and 100 nM), the 5c precursor compound lacking the HDACi hydroxamic moiety. Results were quantified by the crystal violet assay and spectrophotometry (absorbance, 590 nm). (b) Proliferation of CM A375 cells exposed to 5c (90 nM) alone or in the presence of PTZ (2 μM) or AC927 (2 μM). Hydroxamic-null 3c was employed at 90 and 200 nM. Results were quantified by acridine orange staining and fluorescence spectroscopy (excitation, 490 nm; emission, 520 nm). Data in (a,b) are expressed as percentages of the vehicle DMSO control (CTR; indicated as dashed blue lines). *, p < 0.05; **, p < 0.01; and ***, p < 0.001 vs. CTR by Student’s t-test.

Figure 4

Antiangiogenic effects of the HDAC/σ₁R dual-ligand 5c in mouse endothelial C166 cells. (a) Representative phase-contrast images (magnification, ×4) of the capillary tube-like structures formed upon incubation with VEGF-A (VEGF, 80 ng/mL) for 3 h. (b) Quantification of vascular network formations with the angiogenesis analyzer tool of the NIH-Image-J (software version 1.51k). Three parameters were measured: total length, total mesh area and number of nodes. Endothelial cells were treated with 30 nM or 5 µM of 5c and 3c, or (-)-1, the positive antiangiogenic control (5 µM). Data were normalized to the VEGF-A-stimulated effect over the vehicle (DMSO) control condition (CTR; indicated as dashed blue lines). **, p < 0.01 vs. VEGF by one-way ANOVA.

Figure 5

Antiangiogenic signaling by the HDAC/σ₁R dual-ligand 5c in endothelial C166 cells. (a) Representative phase-contrast images (magnification, ×4) of vasculature-like organizations promoted by VEGF-A (80 ng/mL; 3 h). (b) Angiogenesis was quantified as described in Figure 4b, followed by the combination of all three parameters evaluated (lengths, mesh area and nodes; shown in Figure S2) into a single score (reflecting mean ± SEM of the three distinct values, from three independent experiments). Results were expressed as angiogenesis inhibition activity (%), calculated as 100–[(A/B) × 100], where A is the value observed in the presence of 5c or (-)-1 (alone or plus pentazocine/AC927) and B is the correspondent value observed in the VEGF-A condition. Treatments included 5 µM of 5c or (-)-1, alone or in the presence of pentazocine (PTZ, 2 μM) or AC927 (AC, 2 μM). ***, p < 0.001 vs. 5c; ^#, p < 0.05 and ^##, p <0.01 vs. (-)-1 by one-way ANOVA.

Figure 6

Inhibition of membrane protrusion formation by the HDAC/σ₁R dual-ligand 5c. Cell spreading was quantified as the ratio of cells extending membrane protrusions (filopodia and lamellipodia) over the total cell number per microscopic field (magnification, ×20) on an inverted microscope, and expressed as the percent of spreading inhibition [42]. In MCF7 cells, dual-ligands were employed at 100 nM each. In all tumor cells, treatments were performed for 1 h. Percent inhibition of cell spreading was calculated with respect to the vehicle (DMSO) control condition (CTR; indicated as a dashed blue line). *, p < 0.05 and **, p < 0.01 vs. CTR by one-way ANOVA.