Reversible linkage of two distinct small molecule inhibitors of Myc generates a dimeric inhibitor with improved potency that is active in myc over-expressing cancer cell lines

Abstract

We describe the successful application of a novel approach for generating dimeric Myc inhibitors by modifying and reversibly linking two previously described small molecules. We synthesized two directed libraries of monomers, each comprised of a ligand, a connector, and a bioorthogonal linker element, to identify the optimal dimer configuration required to inhibit Myc. We identified combinations of monomers, termed self-assembling dimeric inhibitors, which displayed synergistic inhibition of Myc-dependent cell growth. We confirmed that these dimeric inhibitors directly bind to Myc blocking its interaction with Max and affect transcription of MYC dependent genes. Control combinations that are unable to form a dimer do not show any synergistic effects in these assays. Collectively, these data validate our new approach to generate more potent and selective inhibitors of Myc by self-assembly from smaller, lower affinity components. This approach provides an opportunity for developing novel therapeutics against Myc and other challenging protein:protein interaction (PPI) target classes.

Fig 1. Overview of the basis for generating self-assembling dimeric inhibitors of the Myc transcription factor.

A) Schematic representation of the self-assembling dimer approach. Individual monomers (Blue and Green) composed of ligand, connector and a paired bioorthoganol linker are delivered to the cells, cross the plasma membrane and react to form an active dimeric inhibitor in the cells. Dimer assembly may occur in the cellular milieu or on the target of interest. B) Schematic representation of the boronic acid/diol equilibria utilized during formation of dimer. Trigonal planar, neutral species are in equilibrium with the charged chiral tetrahedral species. For a given diol, in the cellular milieu at pH 7.4 the equilibria are determined by the pKas of the boronic acids employed and by the pKas of the boronate esters formed. Racemization of the chiral charged species occurs very rapidly and the biological target will select for the most preferred dimer. C) Summary of library design: Structures of the two parent molecules C01 (left) and C02 (right) and attachment positions; connectors are either alkyl chains or PEG-units; R and R’ are linked to the connectors via amide or carbon bonds; synthetic details of selected library members are provided in the supplementary experimental procedures.

Fig 2. Select combinations of monomers have synergistic activity in a cell proliferation assay.

(A and B) Dose-response curves for two different combinations, E07+N12 (A) and E08+N11 (B)tested in the cell proliferation screen. In each case the dose-response curve for each individual monomer is plotted. The dose-response curves for the predicted additive response (Bliss) and the combination experimental data are plotted with an increasing concentration of E07 orE08 in the presence of N11 or N12 (30 μM). The data is plotted as a mean ± SEM from 3 independent experiments.

Fig 3. The dimeric inhibitors directly bind to Myc and block its interaction with Max.

A) Inhibitors show saturating binding of Myc in SPR experiments. Equilibrium Response Units (RU), normalized to maximal saturated values in individual experiments, are plotted (mean ± SEM) as a function of inhibitor concentration. B) Dose response curves for the inhibition of Myc:Max interaction as determined by ELISA. The data are represented as a fraction of activity compared to a DMSO treated control sample and are plotted as a mean of 2–5 experiments ± SD. The X-axis refers to the concentration of each monomer used.

Fig 4. The dimeric inhibitors block Myc:Max but not Max:Max binding to DNA.

Gel mobility shift assay showing the effects on Myc:Max DNA complex formation by the dimeric inhibitor E07+N12 (A) and the non-dimerizable control combination E07+C12 (B). The bands that represent protein-DNA complex or naked DNA are shown on the right hand side of each panel. The concentrations indicated are in μM.

Fig 5. Dimeric inhibitors of Myc drive anti-proliferative effects in Myc over-expressing cell lines that are correlated with a decrease in Myc protein levels.

A) Daudi cells were treated with the indicated compounds or combinations for 72 hours and cell viability measured (left panel, * p< 0.05, ** p<0.001, ns not significant). In a parallel experiment Daudi cells were treated with E08+N11 or E08+C11 combinations for the indicated times and protein lysates probed with the indicated antibodies (right panel). E08 was used at 10μM and N11 or C11 were used at 30μM. (B) Raji and (C) K562 cells were treated and analyzed as detailed in (A).

Fig 6. Self-assembling dimeric inhibitor E08+N11, but not the non-dimerizable control E08+C11, inhibits Myc-dependent gene expression.

Daudi (A) and Raji (B) cells were treated with E08+N11 or E08+C11 for 24 hours and gene expression levels were analyzed using a human Myc-target PCR array. The 50 genes that change expression the most relative to a DMSO control from each cell line are shown.