Research ArticleHEALTH AND MEDICINE

Pharmacological modulation of mitochondrial calcium uniporter controls lung inflammation in cystic fibrosis

See allHide authors and affiliations

Science Advances  06 May 2020:
Vol. 6, no. 19, eaax9093
DOI: 10.1126/sciadv.aax9093
  • Fig. 1 The increase of ER-mitochondria tethering inhibits autophagy in CF bronchial cells during P. aeruginosa infection.

    (A) S9 (non-CF) and IB3-1 cells (CF) were infected with P. aeruginosa at an MOI of 100, and after 6 hours, proximity ligation assay (PLA) for IP3R3 and VDAC interactions was performed. Representative images with PLA signals (red) in the different cells are shown. The cell nuclei were stained with 4′,6-diamidino-2-phenylindole (blue). The bar chart shows quantification of PLA signals (%), respect to uninfected S9 cells (n = 25 to 30 independent visual field for each condition of three independent experiments). (B) (I) Immunoblots show VAPB and PTPIP51 expression in S9 (non-CF) and IB3-1 (CF) cells during P. aeruginosa infection. The cells were uninfected or infected for 3, 6, and 12 hours. The samples were probed using the antibodies indicated, where actin is used as loading control. Protein molecular mass markers are indicated in kilodalton. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 5). (C) S9 (non-CF) and IB3-1 cells (CF) were transfected with GFP-LC3–encoding plasmid then infected with P. aeruginosa, as indicated. Representative images of GFP-LC3–transfected non-CF and CF cells have been reported. The bars depict the percentage of cells showing the accumulation of GFP-LC3 in cluster (n = 10 to 20 independent visual field for each condition of three independent experiments). (D) Cells were infected with PAO1 at an MOI of 100 MOI for 6 hours and then treated with either vehicle or bafilomycin A1 (100 nM) as indicated. (I) Samples were probed on immunoblots for LC3 and β-tubulin as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 10). (E) VAPB or PTPIP51 overexpression increases the IP3R3-VDAC interactions in S9 cells. The bar chart shows quantification of PLA signals (%) respect to uninfected non-CF mock cells (n = 45 independent visual field for each condition of three independent experiments). (F) (I) VAPB or PTPIP51 overexpression was probed using Myc-tag and HA-tag antibodies, respectively, while glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is used as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 4). (G) LC3-clustering cell count in mock and VAPB- or PTPIP51-overexpressing S9 cells during pathogen infection was performed. The bars depict the percentage of cells showing the accumulation of GFP-LC3 in cluster (n = 40 independent visual field for each condition of three independent experiments). (H) (I) VAPB or PTPIP51 overexpression inhibits autophagic flux in S9 cells during P. aeruginosa infection. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 6). The reported data are means ± SE of at least three independent experiments. Student’s t test was used for indicated comparisons (*P < 0.05 and **P < 0.01).

  • Fig. 2 During P. aeruginosa infection the mitophagic and xenophagic response in CF bronchial cells is impaired.

    (A) Uninfected and P. aeruginosa–infected airways cells homogenates (hom) of IB3-1 (CF) and S9 (non-CF) cells were fractionated to obtain pure mitochondria (mit) and cytosol (cyt) fractions. Equal amount of proteins (10 μg) from each fraction were analyzed by Western blot using the indicated antibodies. VDAC and actin are, respectively, mitochondrial and cytosolic markers. (B) (I) Representative images of S9 and IB3-1 cells cotransfected with GFP-LC3 and mitochondrial RFP and then treated with PAO1 at an MOI of 100 for 6 hours. (II) Quantitative analysis of S9 and IB3-1 cells that contain mitochondria-localized LC3 puncta; the values are expressed as mean number of colocalized-puncta counts per cell (n = 15 to 20 independent visual field for each condition of three independent experiments). (C) (I) Representative images of mitochondrial redistribution of Parkin in S9 and IB3-1 cells after P. aeruginosa infection. Cells were cotransfected with Parkin cherry and mitochondrial GFP and then treated with PAO1 at an MOI of 100 for 6 hours. (II) The bars depict the percentage of cells showing the accumulation of Parkin in cluster. (III) Quantitative analysis of cells that contain mitochondria-localized Parkin cluster, the values are expressed as mean number of colocalized-cluster counts per cell (n = 20 to 22 independent visual field for each condition of at least three independent experiments). (D) (I) Infected S9 and IB3-1 cells were lysed after the addition of impermeable antibiotics and streaked on LB agar plates for the determination of intracellular colony-forming units (CFUs) (n = 8 of independent experiments). (II) Double bacteria labeling and confocal microscopy to quantify bacterial invasion in whole S9 and IB3-1 cells exposed for 6 hours to GFP–P. aeruginosa. The bars show the ratio GFP/RFP signal in non-CF and CF bronchial cells, as described in experimental procedure section. (E) (I) Representative images of intracellular P. aeruginosa internalization and recruitment of NDP52 to pathogen in IB3-1 cells. (II) The bar chart shows quantification of colocalization signals (%) between NDP52 and pathogen in S9 and IB3-1 cells (n = 8 to 13 independent visual field for each condition of at least three independent experiments). The reported data are means ± SE of at least three independent experiments. Paired data were analyzed by Student’s t test (*P < 0.05 and **P < 0.01).

  • Fig. 3 Abnormal P. aeruginosa–dependent UPRmt activation leads to worsening of autophagic defect and of inflammatory response in CF bronchial cells.

    (A) Representative confocal images of nuclear translocation of endogenous ATF5 (I) and ATF4 (II) in S9 (non-CF) and IB3-1 (CF) cells during pathogen infection. The graphs report the nuclear redistribution of ATF5 and ATF4 after PAO1 infection, expressed as percentage increase in fluorescent ratio signals (from cytosol to nucleus) with respect to uninfected condition (n = 6 to 10 independent visual field for each condition of three independent experiments). (B) Immunoblots of IB3-1 and S9 cells uninfected or infected for 3, 6, and 12 hours, as indicated. The samples were probed using the antibodies indicated, where actin is used as loading control. (C) (I) Autophagic flux in IB3-1 cells overexpressing ATF5 during P. aeruginosa infection. GFP- and ATF5GFP-transfected cells were infected with PAO1 at 100 MOI for 6 hours and then treated with bafilomycin A1, as indicated. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 6). (D) Immunoblots of cleaved caspase-1 and processed IL-1β from lysates (lys) or supernatants (sur) of IB3-1 and ATF5-overexpressed IB3-1 cells infected for 6 hours with P. aeruginosa. Actin was used as loading control. The quantification is expressed as the ratio of casp-1–p10/actin and cleaved IL-1β/actin. The bars are the means ± SE of four independent immunoblots. (E) IB3-1 and S9 cells were transfected with ATF5-GFP and after 6 hours of PAO1 infection, and the cultured cell supernatant were collected to quantify the levels of proinflammatory cytokines by enzyme-linked immunosorbent assay (ELISA) (n = 5 of independent experiments). The selective inhibitor of caspase-1, 20 μM Ac-YVAD-cmk, was added 30 min before infection. (F) Murine WT and NLRP3-null (NLRP3 KO)–derived embryonic fibroblasts (MEFs) were infected with PAO1 at an MOI of 100 for 6 hours and then treated with either vehicle or bafilomycin A1 as indicated. (I) Samples were probed on immunoblots for LC3 and actin as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 4). The reported data are means ± SE of at least three independent experiments. (G) Infected WT and NLRP3 KO MEFs were lysed after the addition of impermeable antibiotics and streaked on LB agar plates for the determination of intracellular CFUs (n = 3 of independent experiments). (H) WT and NLRP3 KO MEFs were transfected with ATF5-GFP and after 6 hours of PAO1 infection, and the cultured cell supernatant were collected to quantify the levels of proinflammatory cytokines by ELISA (n = 3 of independent experiments). Paired data were analyzed by Student’s t test (*P < 0.05 and **P < 0.01).

  • Fig. 4 MCU targeting abrogates the effects of VAPB and PTPIP51 on autophagy, restoring the mitochondrial physiology in CF bronchial cells exposed to pathogen.

    (A) Mitochondrial Ca2+ response in IB3-1 cells (CF cells) overexpressing VAPB or PTPIP51; representative traces are shown. Cells were cotransfected with mitochondrial-targeted aequorin with control empty vector (pcDNA3), VAPB, or PTPIP51 and stimulated with 100 μM histamine. (B) Stable MCU-silenced IB3-1 clone (IB3-1 shMCU) was transfected with pcDNA3, VAPB, or PTPIP51 and then infected with PAO1 at an MOI of 100 for 6 hours. PLA for IP3R3 and VDAC interactions was performed. The bar chart shows quantification of PLA signals (%), with respect to uninfected IB3-1 cells (mock) (n = 15 independent visual field for each condition of three independent experiments). (C) LC3-clusteing cell count in pcDNA3 and VAPB- or PTPIP51-overexpressing IB3-1 shMCU cells during pathogen infection was performed. The bars depict the percentage of cells showing the accumulation of GFP-LC3 in cluster (n = 25 independent visual field for each condition of at least three independent experiments). (D) Effects of VAPB and PTPIP51 on autophagic flux in IB3-1 shMCU cells during pathogen infection. (I) Samples were probed on immunoblots for LC3 and GAPDH as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 6). (E) The effects of MCU-inhibition by 1 μM KB-R7943 pretreatment on superoxide production in CF bronchial cells during PAO1 infection. The quantification of percentage change of cells positive for MitoSox staining compared with uninfected IB3-1 cells is shown (n = 8 independent experiments). (F) Levels of IL-1β and IL-18 in KB-R7943–treated mock and ATF5-overexpressing IB3-1 cells, collected after 6 hours of PAO1 infection at an MOI of 100 (n = 10 independent experiments). (G) Measurements of mitochondrial ΔΨ in KB-R7943–treated and untreated IB3-1 cells during bacterial infection. The bars show the change in the TMRM fluorescence level before and after treatment with P. aeruginosa strain, expressed as the percentage change with respect to untreated IB3-1 cells. FCCP was used to collapse the mitochondrial ΔΨ (n = 12 independent experiments). F.A.U., fluorescent arbitrary units. (H) Histamine-dependent mitochondrial Ca2+ responses of KB-R7943 pretreating IB3-1 cells at different time points before and after PAO1 infection. The histograms show all means ± SE of mitochondrial Ca2+ responses (n = 5 independent experiments). (I) S9 (non-CF) and IB3-1 cells (CF) were pretreated with KB-R7943 1 hour before the infection with PAO1 at an MOI of 100. PLA for IP3R3 and VDAC interactions was performed. The bar chart shows quantification of PLA signals (%), with respect to uninfected S9 cells (n = 15 independent visual field for each condition). The reported data are as means ± SE of at least three independent experiments. (J) Mitochondrial Ca2+ dynamics in S9- (non-CF), IB3-1– (CF), and KB-R7943–treated IB3-1 (CF-KB-R7943) cells exposed to PAO1 at an MOI of 100 were evaluated through ratiometric imaging of mitochondrial-targeted GCaMP6. Traces represent means ± SE of mitochondrial Ca2+ response from at least of 10 independent experiments. SE values are illustrated using gray shading. For statistical significance, the multiple Student’s t test has been used. The bar chart shows the quantification of the area under the curve. a.u., arbitrary unit. Student’s t test used for indicated comparisons (*P < 0.05 and **P < 0.01).

  • Fig. 5 KB-R7943 restores the autophagic response in CF bronchial cells.

    (A) (I) Immunoblot of untreated and KB-R7943–treated IB3-1 cells infected with PAO1 at an MOI of 100 for 6 hours. The samples were probed for LC3 and actin as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 5). (B) LC3-clustering cell count in untreated and KB-R7943–treating IB3-1 cells was performed. The bars depict the percentage of cells showing the accumulation of GFP-LC3 in cluster during P. aeruginosa infection (n = 20 independent visual field for each condition of at least three independent experiments). (C) Effects of KB-R7943 on autophagic flux in IB3-1 cells during pathogen infection. (I) Samples were probed on immunoblot for LC3 and GAPDH as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 6). (D) Immunoblots of mock and VAPB- and PTPIP51-overexpressing IB3-1 cells pretreated with KB-R7943 1 hour before P. aeruginosa infection, as indicated. (I) Samples were probed on immunoblot for LC3 and actin as a loading control. (II) Bar chart shows the ratio LC3-II/LC3-I following quantification of signals from immunoblots (n = 6). (E) LC3-clustering cell count was performed in mock and VAPB- or PTPIP51-overexpressing IB3-1 cells exposed to MCU-inhibitor, KB-R7943, and PAO1 at an MOI of 100, as indicated. The bars show the percentage of cells showing the accumulation of GFP-LC3 in cluster (n = 20 independent visual field for each condition of at least three independent experiments). (F) S9, KB-R7943–treating IB3-1 cells, and/or IB3-1 shMCU cells were infected with PAO1 at an MOI of 100 for 6 hours and then lysed after the addition of impermeable antibiotics and streaked on LB agar plates for the determination of intracellular CFUs (n = 10 of independent experiments). (G) Effect of KB-R7943 on IB3-1 cell apoptosis, during P. aeruginosa infection, using annexin V Alexa Fluor 488/propidium iodide staining. Cells were plated in 60-mm plates and treated for 1 hour with KB-R7943 before PAO1 infection, as indicated. Cell viability and death were evaluated using a Tali apoptosis kit and the Tali image-based cytometer. Bars show the percentage of IB3-1 (CF) and S9 (non-CF) cells that were annexin V–fluorescein isothiocyanate–positive (n = 6 independent experiments). The reported data are means ± SE of at least three independent experiments. Student’s t test was used for indicated comparisons (*P < 0.05 and **P < 0.01).

  • Fig. 6 KB-R7943 limits the P. aeruginosa–triggered inflammatory response in CF mice.

    C57BL/6 Cftrtm1UNCTgN(FABPCFTR)#Jaw mice (CFTR-KO) (n = 20 for each experimental condition) and their WT littermates (n = 20 for each experimental condition) were inoculated with 1.5 × 106 CFU of P. aeruginosa AA43 isolate embedded in agar beads. Mice were treated with KB-R7943 (300 μg/kg) or vehicle via aerosol administration by the Penn-Century MicroSprayer Aerosolizer every 12 hours starting 1 hour before infection. Every 12 hours, mice were monitored for the health status. Two days after infection, 1 hour after the last treatment, murine lungs and BALFs were collected and processed. (A) Kaplan-Meier survival curve for untreated and KB-R7943–treated CFTR-KO mice. The data were analyzed by log-rank (Mantel-Cox) test and Gehan-Breslow-Wilcoxon test (P = 0.0378; GraphPad Prism, USA) (n = 20 mice for each condition of three independent experiments). (B) Clearance and infection were determined on surviving mice. The data were analyzed by Fisher’s test (confidence intervals, 95%; *P < 0.05; GraphPad Prism, USA) (n = 20 mice for each condition of three independent experiments). (C) Bacterial burden in the lungs of KB-R7943–treated and untreated mice after 2 days from P. aeruginosa challenge is shown. Bar chart represent means ± SE of lung CFU in mice. Data were analyzed by two-way ANOVA and Tukey’s post hoc test (*P < 0.05, n = 15 mice analyzed of three independent experiments). (D) The images exemplify the lungs of WT and CFTR-KO mice stained with hematoxylin and eosin (H&E), pretreated 1 hour before infection with KB-R7943 and vehicle. Scale bars, 250 μm (I, IV, VII, and X) and 500 μm (II, III, V, VI, VIII, IX, XI, and XII). (E) Graphs summarize histological scoring of global inflammation grading based on H&E staining (n = 15 of independent experiments). Data were analyzed by Student’s t test used for indicated comparisons and by two-way ANOVA and Tukey’s post hoc test (**P < 0.01). (F) (I) Immunoblots show NLRP3 and LC3 expression in murine WT and CFTR-KO lung homogenates after P. aeruginosa infection. The samples were probed on immunoblots for NLRP3, LC3, and actin as a loading control. Bar chart shows the ratio NLRP3/actin (II) and ratio LC3-II/LC3-I (III) following quantification of signals from immunoblots (n =9 to 12). Student’s t test was used for indicated comparisons (*P < 0.05 and **P < 0.01). (G) The levels of released IL-1β (I) and IL-18 (II) were measured in BALF of WT and CFTR-KO mice exposed to P. aeruginosa infection, treated with KB-R7943 (300 μg/kg) and vehicle. The data were analyzed by two-way ANOVA and Tukey’s post hoc test (*P < 0.05 and **P < 0.01; n = 25 of independent experiments).

Supplementary Materials

  • Supplementary Materials

    Pharmacological modulation of mitochondrial calcium uniporter controls lung inflammation in cystic fibrosis

    Alessandro Rimessi, Chiara Pozzato, Lorenzo Carparelli, Alice Rossi, Serena Ranucci, Ida De Fino, Cristina Cigana, Anna Talarico, Mariusz R. Wieckowski, Carla M. P. Ribeiro, Claudio Trapella, Giacomo Rossi, Giulio Cabrini, Alessandra Bragonzi, Paolo Pinton

    Download Supplement

    This PDF file includes:

    • Figs. S1 to S9
    • Supplementary Materials and Methods
    • Graphical abstract
    • References

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article