Novel Quinazoline Derivatives Inhibit Splicing of Fungal Group II Introns

Abstract

We report the discovery of small molecules that target the RNA tertiary structure of self-splicing group II introns and display potent antifungal activity against yeasts, including the major public health threat Candida parapsilosis. High-throughput screening efforts against a yeast group II intron resulted in an inhibitor class which was then synthetically optimized for enhanced inhibitory activity and antifungal efficacy. The most highly refined compounds in this series display strong, gene-specific antifungal activity against C. parapsilosis. This work demonstrates the utility of combining advanced RNA screening methodologies with medicinal chemistry pipelines to identify high-affinity ligands targeting RNA tertiary structures with important roles in human health and disease.

Figure 1

Experimental pipeline for discovery of new group II intron inhibitors.

Figure 2

General synthesis of quinazoline analogues (a) NaOMe, MeOH, 65 °C, 16 h, 78%. (b) DMF, SOCl₂, 75 °C, 2 h, 91%. Purple hexagon indicates aryl and benzylamines.

Figure 3

Summary of the SAR data. The effect of substituents on the K_i for the in vitro splicing of the ai5γ intron and minimum inhibitory concentration (MIC) for C. parapsilosis. Three functional regions of parental scaffolds 1 and 3 [4-aminoquinazoline, phenyl (or benzyl), and pyridine moieties] are highlighted in green, blue, and orange, respectively. Highlighted regions of the compound analogues are modified relative to the parental compounds.

Figure 4

Inhibition of the ai5γ group II intron splicing by new inhibitors in vitro. (A) Schematic of the ai5γ intron splicing. (B) PAGE analysis of representative time courses of the ai5γ intron splicing in the absence (left) and in the presence (right) of compound 17. (C) Representative K_i curves were obtained by plotting the rate constants (k_obs values) vs concentration of inhibitors. Data represent average of n = 2 independent experiments. Error bars are s.e.m. For comparison, the K_i values for the previously published active inhibitors intronistat A and intronistat B are 2.1 ± 0.2 and 0.36 ± 0.02 μM, respectively.

Figure 5

Inhibition of the ai5γ intron splicing by 17 at 20 μM (red) and after dilution from 20 μM to 1 μM (black). Data represent average of n = 3 independent experiments, error bars are s.e.m.

Figure 6

C. parapsilosis COX1 exhibits a splicing defect in the presence of targeted intron inhibitors 16 and 17. C. parapsilosis was grown and treated with DMSO vehicle only, inactive compound (19), or active compound (16, 17). Relative levels of total and unspliced C. parapsilosis COX1 transcripts are indicated by RT-qPCR quantification of amplicons covering the exon (total) or intron-exon junction from the group IIB intron (unspliced). Mean values and s.e.m. from n = 3 independent experiments are shown with C. parapsilosis PGK1 as a standard. For comparison, in the presence of intronistat B, the levels of unspliced C. parapsilosis COX1 transcripts are 1.5-fold higher than those of total C. parapsilosis COX1.