(R)-Roscovitine and CFTR modulators enhance killing of multi-drug resistant Burkholderia cenocepacia by cystic fibrosis macrophages

Abstract

Cystic fibrosis (CF) is characterized by chronic bacterial infections and heightened inflammation. Widespread ineffective antibiotic use has led to increased isolation of drug resistant bacterial strains from respiratory samples. (R)-roscovitine (Seliciclib) is a unique drug that has many benefits in CF studies. We sought to determine roscovitine's impact on macrophage function and killing of multi-drug resistant bacteria. Human blood monocytes were isolated from CF (F508del/F508del) and non-CF persons and derived into macrophages (MDMs). MDMs were infected with CF clinical isolates of B. cenocepacia and P. aeruginosa. MDMs were treated with (R)-roscovitine or its main hepatic metabolite (M3). Macrophage responses to infection and subsequent treatment were determined. (R)-roscovitine and M3 significantly increased killing of B. cenocepacia and P. aeruginosa in CF MDMs in a dose-dependent manner. (R)-roscovitine-mediated effects were partially dependent on CFTR and the TRPC6 channel. (R)-roscovitine-mediated killing of B. cenocepacia was enhanced by combination with the CFTR modulator tezacaftor/ivacaftor and/or the alternative CFTR modulator cysteamine. (R)-roscovitine also increased MDM CFTR function compared to tezacaftor/ivacaftor treatment alone. (R)-roscovitine increases CF macrophage-mediated killing of antibiotic-resistant bacteria. (R)-roscovitine also enhances other macrophage functions including CFTR-mediated ion efflux. Effects of (R)-roscovitine are greatest when combined with CFTR modulators or cysteamine, justifying further clinical testing of (R)-roscovitine or optimized derivatives.

Figure 1

Structure of the 2,6,9-tri-substituted purine (R)-roscovitine and its main metabolite, M3. Drugability of both compounds according to the Lipinski rule of five. * P is the octanol–water partition coefficient.

Figure 2

Roscovitine and M3 metabolite increase CF macrophage-mediated killing of multi-drug resistant bacteria. CFU assay for CF MDMs infected with multi-drug resistant clinical isolates of B. cenocepacia (Bc, A,B) and P. aeruginosa (Pa, C,D). CF MDMs were grouped according to the presence or absence of treatment with increasing concentrations of roscovitine (A,C), M3 (B,D), or cysteamine (A–D). Data represented as violin plots to show distribution of data. Dose ranges of roscovitine and M3 shown only for CF. n = 3–8/group, one-way ANOVA with post-hoc Tukey. Group ANOVA p value < 0.0001 for B. cenocepacia group. Group ANOVA p value 0.6832 for P. aeruginosa group. Individual p values shown for comparisons between non-CF columns with non-CF Bc or non-CF Pa as reference, and between CF columns with CF Bc or CF Pa as reference.

Figure 3

Roscovitine’s impact on killing of bacteria is partially TRPC6 and CFTR-dependent. (A) CFUs for 24 h bacterial killing assay of roscovitine or M3 against multi-drug resistant B. cenocepacia (Bc) in media devoid of human cells. n = 4, significance via t-test. B) CFU assay for non-CF MDMs infected with B. cenocepacia in the presence (CFTRinh) or absence (non-CF) of a CFTR inhibitor and increasing concentrations of roscovitine. n = 5, significance via t-test with group comparison via one-way ANOVA. Group comparisons to Bc alone. Group ANOVA p value 0.004. Data represented as violin plots to show distribution of data (C) CFU assay for CF and non-CF MDMs infected with B. cenocepacia in the absence (NT) or presence of 10 μM TRPC6 inhibitor for 4 h, with or without 50 μM roscovitine or 50 μM M3. n = 5. Significance via t-test with group comparison via one-way ANOVA. Group ANOVA p value 0.0019. Group comparisons to non-CF NT and CF NT. Data represented as violin plots to show distribution of data. D) Summed CF MDM % phagocytosis of RFP-expressing B. cenocepacia normalized to non-CF MDMs, with or without treatment with tezacaftor/ivacaftor (T/I), roscovitine, M3 or combinations. n = 4, MOI 50, one-way ANOVA with post-hoc Tukey p value < 0.0001.

Figure 4

Roscovitine enhances CFTR modulator-induced functional changes in macrophages. Data represented as violin plots to show distribution of data. All this figure experiments utilized 50 μM roscovitine or M3. (A) Halide efflux assay for CFTR function in non-CF and CF MDMs without (NT) or with infection with B. cenocepacia (Bc). CF MDMs were additionally treated with tezacaftor/ivacaftor (T/I), roscovitine (Rosc), M3, or T/I combined with roscovitine or M3. Results are presented as summary of mean CFTR function at peak of forskolin stimulation (10 min) as well as in a corresponding chloride efflux time course. n = 6–9, one-way ANOVA with post-hoc Tukey. Group ANOVA p value < 0.0001. (B) Representative western blot for CFTR in CF and non-CF MDMs infected with B. cenocepacia (Bc) and treated with tezacaftor/ivacaftor (T/I) roscovitine, (Rosc), M3, or T/I combined with roscovitine or M3. (C) Densitometric analysis of (B) normalized to loading control GAPDH, n = 3, significance determined relative to non-CF NT. * = p value < 0.05, ** = p value < 0.01., *** = p value < 0.001. (D) CFU assay for CF MDMs infected with B. cenocepacia (Bc) and treated with tezacaftor/ivacaftor (T/I) or T/I combined with roscovitine or M3. n = 5. Group ANOVA p value 0.0084. (E) CFU assay for CF MDMs infected with B. cenocepacia (Bc) and treated with tezacaftor/ivacaftor (T/I) and cysteamine, or T/I and cysteamine combined with roscovitine or M3. n = 5. Group ANOVA p value 0.0015.