Synthetic mRNA splicing modulator compounds with in vivo antitumor activity

Abstract

We report our progress on the development of new synthetic anticancer lead compounds that modulate the splicing of mRNA. We also report the synthesis and evaluation of new biologically active ester and carbamate analogues. Further, we describe initial animal studies demonstrating the antitumor efficacy of compound 5 in vivo. Additionally, we report the enantioselective and diastereospecific synthesis of a new 1,3-dioxane series of active analogues. We confirm that compound 5 inhibits the splicing of mRNA in cell-free nuclear extracts and in a cell-based dual-reporter mRNA splicing assay. In summary, we have developed totally synthetic novel spliceosome modulators as therapeutic lead compounds for a number of highly aggressive cancers. Future efforts will be directed toward the more complete optimization of these compounds as potential human therapeutics.

Figure 1

The published structures of pladienolide B, 1 (FR901464), 2 (E7107) and spliceostatin.

Figure 2

An overview of some of the representative published data on SAR for 1 analogs., , Activity is reported relative to 1.

Figure 3

A 3D representation of an example of an overlay of the presumed key interaction groups in a low energy conformation of pladienolide B (shown in yellow, some atoms have been removed for clarity) and compound 3a, showing that the epoxy group and the carbonyloxy groups are the same distance in both molecules. This figure includes a calculated Van der Waal interaction surface with the following color coding; purple= hydrogen-bonding, green= hydrophobic, blue= mildly polar. This alignment represents the best S value (S=167.18) for molecules matching the hypothetical pharmacophore, and the second best overall of 69 alignments from 500 iterations. The alignment was prepared using the Molecular Operating System (MOE 2007.09, Chemical Computing Group, Inc.) using the Flexible Alignment function with both molecules, following a conformational minimization using MOE default settings.

Figure 4

Synthetic targets that matched a proposed spliceosome modulation pharmacophore.

Figure 5

Compound 5 inhibition of tumor growth in JeKo-1 tumor-bearing mice. JeKo-1 mantle cell lymphoma tumors were transplanted to NOD/SCID mice on day 1 and, beginning on day 4, the mice (4 per group) received IV injections of vehicle, 5, 10, 25 or 50 mg/kg of compound 5 daily for five consecutive days. Vehicle alone did not inhibit tumor growth as compared to saline-treated mice (data not shown). Arrows indicate the dosing schedule. The data are represented as mean ± SEM. Tumor volumes were modeled with treatment, linear and quadratic trends over time, and interaction of treatment with linear and quadratic trends over time in a repeated measure framework using autoregressive 1 AR(1) variance-covariance structure and empirical sandwich error estimates for fixed effect. Significance is reported as the comparison of tumor volumes in vehicle-treated mice to compound-treated animals. All analyses were performed using SAS software (SAS Institute, Cary, NC), Windows version 9.1. *, p = 0.01; **, p < 0.0001.

Figure 6

Compound 26 causes aggregation of the splicing factor SC35 and the SF3b protein SAP155. HeLa cells were incubated for 4 hours in culture media containing 10 µM compound 26 in DMSO (B and D) or an equal volume of DMSO (A and C) and stained with antibodies recognizing SC35 (panel A and B) or SAP155 (C and D). The cells were then stained with fluorescent secondary antibodies and examined with a fluorescence microscope using an objective with 60X magnification.

Figure 7

Compounds 5 and 26 selectively inhibit splicing. 293T cells were transiently transfected with a dual-reporter splicing construct (see Experimental Section for details) for 48 hours and subsequently treated with or without compound for 5 hours. The degree of cell death was negligible at this time at the above tested concentrations. The cells were lysed, and equalized amounts of total cellular proteins were assayed for luciferase and β-galactosidase activity using the Dual-Light System. The data are expressed as the relative ratio of luciferase to β-galactosidase activity compared to a control vehicle (DMSO)-treated sample, and are presented as the mean ± SEM from multiple independent experiments. Note that compound 27 (which lacks cytotoxic activity; data not shown) had no effect in this assay, and that compound 4 (which exhibits only minimal cytotoxic effects at these concentrations; see Table 1) altered splicing only very slightly at the highest concentration tested; thus, the cytotoxic effects of our compounds correlate closely with their effects on splicing.

Scheme 1

Synthesis of carbamate and ester analogs of compound 3a.b ^bReagents and conditions: (a) 4-nitrophenyl chloroformate, Et₃N, dichloromethane; (b) R₁-piperazine, 1,2-dichloroethane; c) isobutyryl chloride, Et₃N, DMAP, CH₂Cl₂; (d) 4-nitrophenyl chloroformate, pyridine, THF/CH₃CN; (e) (1-cycloheptyl)piperazine, 1,2-dichloroethane.

Scheme 2

Synthesis of new ester analogs. Reagents and conditions; a) For compound 12 the corresponding acid, EDC and DMAP were used, in all other cases the conditions were acid chloride/Et₃N/DMAP. In vitro cytotoxicity screening data as determined by XTT assays are shown for the compounds following a 72-hour exposure.

Scheme 3

The synthesis of 1,3-dioxane analogs.b ^bReagents, conditions and yields; a) 1H-imidazole-1-sulfonyl azide HCl, CuSO₄·5H₂O, K₂CO₃, MeOH; b) CF₃COOH/H₂O/CHCl₃; c) ylide (shown), benzene; d) 1) DIBAL-H, CH₂Cl₂; 64% overall; 2) 2-mercaptobenzothiazole, Ph₃P, DIAD, THF, 0 °C; 3) (NH₄)₆Mo₇O₂₄ tetrahydrate/ H₂O₂; 78% overall e) LiHMDS then aldehyde 24; 37% f) 1) Ph₃P, H₂O, benzene, 45 °C; 2) (S,Z)-4-(TBSO)pent-2-enoic acid, HBTU, iPr₂EtN; 69% overall 3) TBAF, THF; 47% 4) Isobutyric anhydride, Et₃N, DMAP; 96%; g) 1) Ph₃P, H₂O, benzene, 45 °C; 2)Acetic anhydride/pyridine.