Correction of the F508del-CFTR protein processing defect in vitro by the investigational drug VX-809

Abstract

Cystic fibrosis (CF) is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene that impair the function of CFTR, an epithelial chloride channel required for proper function of the lung, pancreas, and other organs. Most patients with CF carry the F508del CFTR mutation, which causes defective CFTR protein folding and processing in the endoplasmic reticulum, resulting in minimal amounts of CFTR at the cell surface. One strategy to treat these patients is to correct the processing of F508del-CFTR with small molecules. Here we describe the in vitro pharmacology of VX-809, a CFTR corrector that was advanced into clinical development for the treatment of CF. In cultured human bronchial epithelial cells isolated from patients with CF homozygous for F508del, VX-809 improved F508del-CFTR processing in the endoplasmic reticulum and enhanced chloride secretion to approximately 14% of non-CF human bronchial epithelial cells (EC(50), 81 ± 19 nM), a level associated with mild CF in patients with less disruptive CFTR mutations. F508del-CFTR corrected by VX-809 exhibited biochemical and functional characteristics similar to normal CFTR, including biochemical susceptibility to proteolysis, residence time in the plasma membrane, and single-channel open probability. VX-809 was more efficacious and selective for CFTR than previously reported CFTR correctors. VX-809 represents a class of CFTR corrector that specifically addresses the underlying processing defect in F508del-CFTR.

Fig. 1.

VRT-768 and VX-809 increased F508del-CFTR maturation and chloride transport in FRT cells. (A) Chemical structures of VRT-768 and VX-809. (B) F508del-CFTR maturation, expressed as a ratio of mature CFTR to total CFTR (mature and immature), in FRT cells treated for 48 h with the indicated VRT-768 (•) or VX-809 (○) concentrations (n = 4). (C) F508del-CFTR function in FRT cells preincubated for 48 h with the indicated VRT-768 (•) or VX-809 (○) concentrations (n = 3). Forskolin was added to activate CFTR through the cAMP/PKA pathway, and CFTR function was measured as transepithelial currents in the presence of basolateral to apical chloride gradient. Asterisks indicate significant difference vs. VRT-768 (P < 0.05; t test).

Fig. 2.

Biochemical and functional data suggest that VX-809 acted at the level of the ER to allow a fraction of the F508del-CFTR pool to adopt a properly folded form. (A) Representative gels from a pulse-chase experiment shows the conversion of immature to mature CFTR in HEK-293 cells expressing CFTR (□) or F508del-CFTR following 24-h incubation with vehicle (○) or 3 μM VX-809 (•). Arrows indicate mature CFTR. (B) Quantification of mature F508del-CFTR shown in A (n = 4). (C) Quantification of immature CFTR during the 180-min chase in cells pretreated with vehicle (open symbols) or 3 μM VX-809 (filled symbols) in the presence (circles) and absence of BFA (squares). (D) Immunoblot of trypsin-digested microsomes from cells expressing CFTR or F508del-CFTR pretreated with vehicle or 3 μM VX-809 for 24 h. Trypsin concentrations were 0, 15, 30, 60, 120, 240, 480, and 960 μg/mL (labeled 1–8, respectively). Arrows indicate mature CFTR. (E) Quantification of the data in D. The trypsin concentration value required to eliminate 50% of the full-length (mature and immature) CFTR protein (C_try50%) is indicated by the dotted lines. (F) Quantification of the P_o of CFTR channels in excised membranes from NIH 3T3 cells expressing CFTR or F508del-CFTR. Cells expressing F508del-CFTR were pretreated for 48 h with vehicle or 3 μM VX-809 at 37 °C or vehicle at 27 °C. To activate CFTR, 1 mM ATP and 100 U/mL PKA were added to the bath. Acute VX-770 addition further increased the P_o of F508del-CFTR following VX-809 treatment. Asterisks indicate significant difference vs. CFTR (dotted line).

Fig. 3.

VX-809 increased CFTR maturation and chloride secretion in cultured F508del-HBE. (A) Glycosylation pattern of CFTR (Upper) and F508del-CFTR pretreated for 48 h with VX-809 at the indicated concentrations (Lower). (B) Quantification of the data in A (n = 3) expressed as a percentage of the mature/total CFTR in the absence of VX-809 (as percent of control). (C) Representative recording of the forskolin (10 μM)-stimulated I_T in F508del-HBE pretreated for 48 h with VX-809 at the indicated concentrations. Before adding forskolin, amiloride was added to block the epithelial Na⁺ channel. A basolateral-to-apical chloride gradient was used for Ussing chamber experiments. (D) Quantification of the forskolin-stimulated I_T in F508del-HBE isolated from seven patients with CF homozygous for the F508del-CFTR mutation (left y axis). Right y axis shows the I_T normalized to the 10 μM forskolin-stimulated I_T in non-CF HBE. (E) The onset of VX-809 action was determined by measuring the CFTR-mediated I_T in F508del-HBE pretreated with 3 μM VX-809 for the indicated times (n = 6; data from single donor lung). (F) Cell surface turnover of F508del-CFTR was determined by first incubating F508del-HBE for 48 h with 3 μM VX-809 and then measuring the forskolin-stimulated I_T at the indicated times after VX-809 washout (data from single donor lung; n = 6). (G) Concentration response curve for VX-809 in the absence (•) and presence of 1 μM VX-770 (○) in F508del-HBE from a single donor. (H) Mean (±SEM) forskolin-stimulated I_T values in F508del-HBE pretreated for 48 h with VX-809 (3 μM), 4-phenylbutyrate (4-PB; 1,500 μM), Corr-4a (10 μM), and VRT-325 (6.7 μM) for 8 d with 1 μM SAHA or 2 to 4 h with 100 μM miglustat. The concentration and treatment duration for each compound were based on the maximally effective experimental conditions published for the previously described CFTR corrector (18, 30, 43, 44). Asterisks indicate significant (P < 0.05; paired t test) increase in I_T vs. untreated levels.

Fig. 4.

VX-809 did not improve the processing of non-CFTR misfolded proteins. The effects of 48 h treatment with 10 μM VX-809, 6.7 μM VRT-325 (18), and 10 μM Corr-4a (19) on the maturation of normal and mutant hERG (G601S-hERG) (A) and P-gp (G268V-P-gp) (B) transiently expressed in HEK-293-cells. (C) Summary of the effects of VX-809, Corr-4a, and VRT-325 on the cellular processing of normal (i.e., WT) and mutant (mut) CFTR, P-gp, and hERG (Fig. S7). Dash indicates no significant difference vs. vehicle-treated controls. A single plus sign indicates a significant difference compared with vehicle-treated controls, whereas a double plus sign indicates a significant difference between vehicle-treated controls and the other compound treatments (ANOVA followed by Tukey multiple-comparison test; n = 3–5). (D and E) Additive effects of VX-809 and VRT-325 (D) or Corr-4a (E) at the indicated concentrations on CFTR-mediated chloride transport in cultured F508del-HBE isolated from a single donor bronchi (n = 4).