Peroxisomal ROS control cytosolic Mycobacterium tuberculosis replication in human macrophages

Abstract

Peroxisomes are organelles involved in many metabolic processes including lipid metabolism, reactive oxygen species (ROS) turnover, and antimicrobial immune responses. However, the cellular mechanisms by which peroxisomes contribute to bacterial elimination in macrophages remain elusive. Here, we investigated peroxisome function in iPSC-derived human macrophages (iPSDM) during infection with Mycobacterium tuberculosis (Mtb). We discovered that Mtb-triggered peroxisome biogenesis requires the ESX-1 type 7 secretion system, critical for cytosolic access. iPSDM lacking peroxisomes were permissive to Mtb wild-type (WT) replication but were able to restrict an Mtb mutant missing functional ESX-1, suggesting a role for peroxisomes in the control of cytosolic but not phagosomal Mtb. Using genetically encoded localization-dependent ROS probes, we found peroxisomes increased ROS levels during Mtb WT infection. Thus, human macrophages respond to the infection by increasing peroxisomes that generate ROS primarily to restrict cytosolic Mtb. Our data uncover a peroxisome-controlled, ROS-mediated mechanism that contributes to the restriction of cytosolic bacteria.

Graphical abstract

Figure 1.

Mtb infection triggers peroxisome biogenesis in human macrophages. (A) Heatmap of differentially expressed genes from RNA sequencing analysis of iPSDM infected with either Mtb WT or Mtb ΔRD1 for 2 or 48 h. Data normalized to the uninfected control. Left: RNA expression of peroxisomal protein associated with β-oxidation, α-oxidation, transport of lipid, and antioxidant activity. Top right: PPAR genes regulation during infection. Bottom right: RNA expression of peroxisomal protein associated with membrane assembly, cargo, and division of peroxisomes. (B) Top: Schematic representation of the CMV:EGFP-PTS1 allele in the AAVS1 locus. Bottom: Confocal images of iPSDM expressing EGFP-PTS1 (green) stained for PEX14 (red) and nuclear staining (blue). Scale bars: 10 µm, zoom: 5 µm. (C) Confocal images of iPSDM after 24 h of infection (Uninfected, Mtb WT and ΔRD1). Nuclear staining (blue), PEX14 staining (orange), and Mtb-E2-Crimson (red). Scale bars: 10 µm. (D) Pipeline for masking PEX14 (red) and EGFP-PTS1 (green) to perform colocalization study. Arrowheads point at elongated peroxisomes. Scale bars: 5 µm. (E) Analysis of peroxisome number during Mtb infection. Violin plot displaying the distribution of PEX14 puncta; quantification from N > 80 cells analyzed at each time point. Data were collected from three independent experiments. Quantiles representing the distribution are shown. (F) Quantification of EGFP-PTS1 puncta per cells. N > 80 cells were quantified per each time point from three independent experiments. (G) Pearson correlation between Mtb area and PEX14 puncta in infected iPSDM with Mtb WT and ΔRD1, at 24 hpi (light violet) and 48 hpi (dark violet). Results are from 50 cells obtained from three independent experiments. (H) Quantification of PEX14 volume (µm³) per cell (UI = uninfected, M. WT = Mtb WT, M. ΔRD1 = Mtb ΔRD1.) N > 80 cells were quantified from three independent experiments. (I) Western blot analysis of peroxisomal protein PEX14, ACOX1, CAT, and HSD17B4 (HSD) of iPSDM at 24 hpi of infection (Uninfected, Mtb WT, and ΔRD1). (J) Analysis of colocalization of PEX14 and EGFP_PTS1 peroxisomes. Left: Confocal images of iPSDM at 24 hpi infected with Mtb WT and ΔRD1. Nuclear staining (blue), PEX14 staining (orange), EGFP_PTS1 (green), and Mtb-E2-Crimson (magenta). Arrowheads point at PEX14⁺/PTS1⁻ peroxisomes. Scale bars: 10 µm. Right: Quantification of PEX14/PTS1 colocalization per cells. Error bars indicate SD. Data representative from one out of three independent experiments. N = 60 cells were quantified. Significance was determined by two-way ANOVA with Tukey’s multiple comparison post-test (C, E, F, and J). P value 0.002 (**), <0.0001 (***). Source data are available for this figure: SourceData F1.

Figure S1.

Generation and characterization of PEX3^−/− iPSC and iPSDM clones. (A and B) Selection of PEX3^−/− clones. (A) PCR genotyping of the expanded clones, D2 and E3. (B) Top: Schematic representation of the PEX3 KO CRISPR strategy. PCR primers (blue), sgRNA (orange). Bottom: Sanger sequencing of the upper band for the D2 clones (*). (C) Immunofluorescence of iPSC PEX3^−/− for pluripotent markers (OCT3/4, TRA-1-60, and TRA-1-81). Scale bars: 100 µm. (D) Flow cytometry characterization of PEX3^+/+ and PEX3^−/− monocytes and macrophages. Names of the markers are indicated on the graph graphs. Black, negative sample; red, isotype control; blue, marker.

Figure 2.

Human macrophages lacking peroxisomes are unable to restrict Mtb WT replication. (A) Confocal images of iPSDM PEX3^+/+ and PEX3^−/− for GFP-PTS1 (green), PEX14 (green), TOM20 (red), CAT (red), and PMP70 (green). Nuclear staining (blue). Scale bars: 10 µm. (B) Analysis of Mtb WT growth in iPSDM lacking peroxisomes. Left: Confocal images of iPSDM PEX3^+/+ and PEX3^−/− (clone 1 and 2) at 24 and 72 hpi infected with Mtb WT. Nuclear staining (blue) and Mtb-E2-Crimson (red). Scale bars: 100 µm, zoom: 10 µm. Right: Growth index of Mtb WT in iPSDM control (PEX3^+/+) or KO for PEX3 (PEX3^−/− clones 1 and 2). Data representative of one out of three independent experiments (n = 4 independent wells). (C) Analysis of Mtb ΔRD1 growth in iPSDM lacking peroxisomes. Left: Confocal images of iPSDM PEX3^+/+ and PEX3^−/− (clone 1 and 2) at 24 and 72 hpi infected with Mtb ΔRD1. Nuclear staining (blue) and Mtb-E2-Crimson (red). Scale bars: 100 µm, zoom 10 µm. Right: Growth index of Mtb WT in iPSDM control (PEX3^+/+) or KO for PEX3 (PEX3^−/− clones 1 and 2). Data are representative of one out of three independent experiments (n = 4 technical replicate per each condition). Significance was determined by two-way ANOVA with Tukey’s multiple comparison post-test (B and C). P value 0.033 (*), 0.002 (**), <0.0001 (***).

Figure S2.

Characterization of PEX3^−/− iPSDM clones and uptake and growth of Mtb in PEX3^+/+ and PEX3^−/− iPSDM. (A) Rescue experiment with PEX3_turbo for 24 h. Snapshot of iPSDM PEX3^+/+ and PEX3^−/−. Nuclear staining (blue), PEX3_turbo (orange), and EGFP-PTS1 (green). Scale bars: 10 µm. (B) Western blot of peroxisomal related protein expressed in iPSDM PEX3^+/+ and PEX3^−/− at the steady state. (C) Quantification of CAT expression from B normalized with actin. (D) Analysis of Mtb growth in iPSDM PEX3^+/+ and PEX3^−/−. Violin plot representation of Mtb area (px2) per cells over time (2, 24, 48, and 72 hpi) in PEX3^+/+ and PEX3^−/− (clone 1 and 2) during infection with Mtb WT and Mtb ∆RD1. Significance was determined for the 2 hpi time point by two-way ANOVA with Tukey’s multiple comparison post-test. Source data are available for this figure: SourceData FS2.

Figure S3.

HyPer reporter: Monitoring peroxisomal (Pexo_Hyper), cytosolic (Cyto_Hyper), and endosomal (Endo_Hyper) H₂O₂. Related to Figs. 2 and 3. Pharmacological modulation of peroxisomes: minimal inhibitory concentration (MIC) and in vitro characterization of the drugs. (A) Schematic representation of the Cyto_Hyper reporter (top) and live snapshot of iPSDM expressing the reporter and treated with H₂O₂ as a positive control (middle). Quantification of the Cyto_Hyper ratio with and without H₂O₂ (bottom). (B) Schematic representation of the Pexo_Hyper reporter (top) and live snapshot of iPSDM expressing the reporter and treated with H₂O₂ as a positive control (middle). Quantification of the Pexo_Hyper ratio with and without H₂O₂ (bottom). (C) Schematic representation of the Endo_Hyper reporter (top) and live snapshot of iPSDM expressing the reporter and treated with H₂O₂ as a positive control (middle). Quantification of the Endo_Hyper ratio with and without H₂O₂ (bottom). Significance was determined by unpaired t test. P value (APA) 0.033 (*), <0.0001 (***). Scale bars: 10 µm. (D–F) MIC of 4-PBA (D), 3-AT (E), and GW (F) for Mtb WT and ∆RD1. Mtb growth measured as OD₆₀₀ at 6, 9, and 13 d. (G) Heatmap of differentially expressed genes from qPCR analysis of iPSDM treated with either GW or 3-AT for 48 h normalized to the untreated control.

Figure 3.

Peroxisomal H₂O₂ increases after infection with Mtb WT and it is important to control cytosolic replication in macrophages. (A–C) Analysis of Pexo_Hyper ratio in iPSDM during infection (Uninfected, Mtb WT, and ΔRD1). (A) Snapshot of Pexo_Hyper reporter during Mtb infection. Confocal images of iPSDM at 24 hpi of infection. GFP_UV (green), GFP (red), and Mtb-E2-Crimson (cyan) and a ratiometric imaging of the Pexo_Hyper reporter. Scale bars: 10 µm. (B) Each point represents the ratio of cells infected over 30 h along with a trend line and standard deviation. A total of more N > 300 cells were quantified at each time point. Significance was determined for the last two time points (27 and 30 hpi) by two-way ANOVA with Tukey’s post-doc test. P value (APA) 0.033 (*), 0.002 (**), <0.0001 (***). (C) The left graph shows the Pexo_Hyper ratio of iPSDM infected and bystander in the Mtb WT infected well. The right graph shows the ratio of iPSDM infected and bystander in the Mtb ΔRD1 infected well. N > 300 cells were quantified per each time point. (D) Analysis of peroxisome’s number with peroxisome modulators. Left: Confocal imaging of iPSDM treated for 48 h with GW or 3-AT. Nuclear staining (blue) and PEX14 (red). Scale bars: 10 µm, zoom: 5 µm. Right: Quantification of PEX14 puncta per iPSDM control (CTRL) or treated with the drug (GW and 3-AT). Data representative from one out of two independent experiments. N = 30 cells were quantified. (E) Quantification of PEX14 volume (µm³) per cells. Data representative from one out of two independent experiments. N = 20 cells were quantified. (F) Quantification of Pexo_Hyper ratio in iPSDM untreated (CTRL) or treated with the drug (GW). Data are representative of one out of two independent experiments (n = 3 independent wells per replicate), N > 200 cells were quantified. (G) Analysis of Mtb growth with peroxisome modulators. Left: Confocal images of iPSDM at 72 hpi infected with Mtb WT (top) and ΔRD1 (bottom). Nuclear staining (blue) and Mtb-E2-Crimson (red). Scale bars: 100 µm. Right: Growth index of Mtb WT (top) and Mtb ΔRD1 (bottom), for iPSDM control (CTRL) or treated with the drug (GW and 3-AT). Data representative from one out of three independent experiments (bars represent SD of n = 3 independent wells per replicate). Significance was determined using unpaired t test (E), one-way ANOVA with Dunnett’s multiple comparison post-test (D and F), and by two-way ANOVA with Dunnett’s multiple comparison post-test (G) and Tukey’s multiple comparisons test (B). P value 0.033 (*), 0.002 (**), and <0.0001 (***).

Figure 4.

Peroxisome-dependent restriction of Mtb is associated with higher levels of ROS in the cytosol. (A–C) Cyto_Hyper reporter during Mtb infection. (A) Confocal images of iPSDM at 24 hpi of infection (Uninfected, Mtb WT, and ΔRD1). The left graph shows Cyto_Hyper in iPSDM PEX3^+/+ and right graph shows Cyto_Hyper in PEX3^−/− iPSDM. Top: Merge of GFP_UV (green), GFP (red), and Mtb-E2-Crimson (cyan). Bottom: Ratiometric imaging of the Cyto_Hyper reporter. Scale bars: 10 µm. (B) Quantification of Cyto_Hyper ratio in iPSDM PEX3^+/+ and PEX3^−/− during infection at 20 and 40 hpi. Data representative from one out of two independent experiments (n = 4 independent wells per replicate). Significance was determined using unpaired by two-way ANOVA with Šídák’s multiple comparisons post-test. P value (APA) 0.033 (*), 0.002 (**). (C) 3D surface (left) and line plot (right) of Cyto_Hyper reporter in iPSDM PEX3^+/+ (1 box) and PEX3^−/− (2 box) infected with Mtb WT. (D and E) Endo_Hyper reporter during Mtb infection. (D) Snapshot of Endo_Hyper reporter in iPSDM PEX3^+/+ and PEX3^−/− during infection over 24 hpi. Merge of GFP_UV (green), GFP (red), and Mtb-E2-Crimson (cyan). Scale bars: 10 µm. (E) Analysis of Endo_Hyper ratio around Mtb (area 0.5 µm) in iPSDM PEX3^+/+ and PEX3^−/− infected with Mtb WT and ΔRD1 at 24 hpi. The red line represents the median of Mtb-Endo_Hyper ratio in PEX3^+/+ and the light blue line represents the median of Mtb-Endo_Hyper ratio in PEX3^−/− iPSDM. N > 500 Mtb regions of interest were quantified per each condition.

Figure 5.

Peroxisome-dependent restriction of Mtb is spatial-temporally separated from the NADPH oxidase activity in human macrophages. (A) Workflow for nucleofection (nf) approach to KO CYBB (CYBB^nf) and PEX3 (PEX3^nf) genes in HMDM. (B) Western blot of HMDM CTRL, PEX3^nf, and CYBB^nf for gp91-phox protein (CYBB). (C) Left: Immunofluorescence of HMDM CTRL and PEX3^nf. Nuclear staining (blue), PMP70 (orange), and CAT (green). Scale bars: 20 µm. Right: Quantification of CAT and PMP70 puncta area per area of cells. Significance was determined by two-way ANOVA with Šidák’s test. P value <0.0001 (***). (D) Mtb growth in HMDM CTRL, PEX3^nf, and CYBB^nf. Left: Confocal images of HMDM CTRL, PEX3^nf, and CYBB^nf at 60 hpi infected with Mtb WT (top) and ΔRD1 (bottom). Nuclear staining (blue) and Mtb-E2-Crimson (red). Scale bars: 100 µm. Right: Fold change of Mtb growth in HMDM CTRL (green), PEX3^nf (orange), and CYBB^nf (light blue) over 98 h of infection. The top graph shows the fold change of Mtb WT and the bottom graph the fold change of Mtb ΔRD1. Data representative from one out of two independent experiments (n = 3 independent wells per replicate). Significance was determined only for the last time point (98 hpi) by two-way ANOVA with Dunnett’s post-doc test. P value 0.033 (*).