c-Myc Inhibits Macrophage Antimycobacterial Response in Mycobacterium tuberculosis Infection

Abstract

Background: Mycobacterium tuberculosis (MTB) remains a major cause of global mortality, yet natural immunity prevents disease in more than 90% of exposed individuals. Interferon gamma (IFN-γ) is a critical regulator of innate immunity and enhances macrophage antimicrobial responses.

Methods: Using in vitro systems approaches, we compared the effects of IFN-γ exposure before versus after infection. We manipulated c-Myc in primary macrophages with a tetracycline-inducible lentiviral system. c-Myc expression was also analyzed in tissues from murine tuberculosis models and human granulomas.

Results: Preinfection IFN-γ exposure primed macrophages for enhanced bacterial control, whereas postinfection exposure did not. We identified c-Myc signaling as a central determinant of macrophage antimycobacterial function. Inhibition of c-Myc via Omomyc enhanced bacterial control partly through mTORC1-dependent metabolic reprogramming and nitric oxide production. In vivo analyses, including murine models and human clinical histopathology, revealed strong associations between c-Myc expression, MTB persistence, and active tuberculosis.

Conclusion: c-Myc mediates immune privilege in MTB infection and represents a promising target for host-directed therapies to enhance macrophage function.

Keywords: Mycobacterium tuberculosis; c-Myc; granuloma; macrophage activation; metabolic reprogramming.

Figure 1.

Antimycobacterial properties of IFN-γ depend on the sequence of infection and IFN-γ exposure. A, Experimental workflow. BMDMs were isolated from C57BL/6J mice and differentiated for 6 d. Cells were either preactivated with IFN-γ (25 ng/mL) 24 h before infection or treated immediately after infection. Infection was performed with MTB H37Rv (MOI 5). B, Kinetics of intracellular bacterial burden. CFU/mL of intracellular MTB H37Rv were measured immediately after phagocytosis (day 0) and at days 3 and 5 postinfection. Data are mean ± SD of 5 independent experiments, normalized to day 0. C, Representative flow cytometry plot showing viable F4/80⁺CD11b⁺ BMDMs infected with MTB H37Rv live-dead strain. Infected (mCherry⁺) cells are subdivided into “permissive” or “restrictive” populations based on GFP induction (1 µg/mL doxycycline) after 24 h. D, Percentage of “permissive” macrophages (GFP⁺) at days 0, 3, and 5 postinfection. Statistical analysis: P values were determined by unpaired 2-tailed Student's t test (before vs after infection) and corrected for multiple testing using the 2-stage linear step-up method of Benjamini–Krieger–Yekutieli.

Figure 2.

Changes in c-Myc–associated transcriptional programs mirror MTB permissive versus controlling macrophages in vitro. A, Principal component analysis (PCA) of differently activated/infected BMDMs, 6 and 24 h after infection with MTB H37Rv (MOI 5). Each point represents an individual sample, with colors indicating the treatment. PC1 and PC2 are plotted with the percent variance explained indicated. B, Volcano plot of differential expression in BMDMs 24 h postinfection, comparing IFN-γ after versus IFN-γ before conditions. Genes are plotted by log₂ fold-change (x-axis) against −log₁₀(adjusted P) (y-axis). Red dots mark genes meeting |log₂FC| > 2 and adjusted P < .01; dashed lines denote these thresholds. C, GSEA dot plot of Hallmark pathways from BMDMs 24 h postinfection, comparing IFN-γ administered before infection versus IFN-γ added after infection. Dots represent normalized enrichment scores (adjusted P < .05), with positive values indicating enrichment in the before condition. The NES values were computed by fgsea on MSigDB Hallmarks (v2024.1.Mm). D, Connected-dot plot of GSEA comparing IFN-γ–activated BMDMs before and after relative to untreated, uninfected controls. Each point represents 1 gene set's NES: left circles mark sets significantly enriched in the before versus control comparison (adjusted P < .05), right circles mark sets significantly enriched in the after versus control comparison (adjusted P < .05). The y-axis shows NES values (positive = up-regulation; negative = down-regulation). Dashed lines connect the same gene set's NES across the 2 contrasts; the solid line highlights those gene sets that reverse their direction of regulation between pre- and postinfection. All data shown were derived from bulk RNA sequencing.

Figure 3.

c-Myc inhibition leads to a gain in anti-mycobacterial function and induces a pro-inflammatory phenotype in macrophages in vitro. A, Principal component analysis (PCA) of BMDMs expressing either c-Myc or Omomyc versus WT 24 h after Tet-On induction with doxycycline (100 ng/mL). Each point represents an individual sample, with colors indicating the experimental group. PC1 and PC2 are plotted with the percent variance explained on the axes. B, GSEA dot plot of Hallmark pathways from BMDMs expressing Omomyc compared to BMDMs expressing c-Myc 24 h after Tet-On induction with doxycycline. Dots represent normalized enrichment scores (adjusted P < .01), with positive values indicating enrichment in the Omomyc group. The NES values were computed by fgsea on MSigDB Hallmarks (v2024.1.Mm). C, Kinetics of intracellular bacterial burden in WT BMDMs, BMDMs treated with the chemical c-Myc inhibitor 10058-F4, and BMDMs expressing c-Myc or Omomyc. The CFU/mL of intracellular MTB H37Rv were measured immediately after phagocytosis (day 0) and at days 3 and 5 postinfection. Data are mean ± SD of 3 independent experiments, normalized to day 0. D, Quantification of the proportion of mCherry⁺ BMDMs expressing TNF-α (top) and iNOS (bottom) at days 0, 3, and 5 postinfection. The WT BMDMs, WT pretreated with IFN-γ, and BMDMs expressing c-Myc or Omomyc were infected and analyzed by flow cytometry. Data are mean ± SD of 3 independent experiments. Statistical analysis: P values were determined by unpaired 2-tailed Student's t test (Omomyc transduced vs WT untreated) and corrected for multiple testing using the 2-stage linear step-up method of Benjamini–Krieger–Yekutieli. A, and B, are bulk RNA sequencing data.

Figure 4.

c-Myc inhibition associated gain of antimycobacterial function is partly mediated by a shift toward mTORC1 signaling. A, Bar plot of Hallmark metabolic pathways significantly enriched by GSEA (adjusted P < .01). Bars show normalized enrichment scores (NES), positive NES denote enrichment in the Omomyc group. B, Heatmap of leading-edge genes from the MTORC1_SIGNALING Hallmark pathway. Rows are genes contributing most to the enrichment score, columns are individual Omomyc-expressing or WT BMDM samples. Values are log₂ fold-change relative to the average of WT samples. C, Representative flow cytometry plots showing the proportion of iNOS⁺ cells in mCherry⁺ BMDMs on day 3 after infection with mCherry-expressing MTB (MOI 5). The WT cells, Omomyc-expressing cells and Omomyc-expressing cells treated with rapamycin (10 µM) are displayed. All samples received 0.1% DMSO (solvent for rapamycin). D, Left: Proportion of infected (mCherry⁺) BMDMs expressing iNOS at days 0, 3, and 5 postinfection in WT, Omomyc-expressing, Omomyc-expressing + rapamycin, and WT + IFN-γ conditions. Data are mean ± SD (n = 3). Right: Kinetics of intracellular bacterial burden. The CFU/mL of MTB H37Rv were measured in WT BMDMs, Omomyc-expressing BMDMs, Omomyc-expressing + rapamycin, and WT + IFN-γ BMDMs immediately after phagocytosis (day 0) and at days 3 and 5 postinfection. Data represent mean relative changes (range) ± SD from 3 independent experiments, normalized to day 0. Statistical analysis: P values were determined by unpaired 2-tailed Student's t test (Omomyc transduced vs Omomyc transduced + rapamycin) and corrected for multiple testing using the 2-stage linear step-up method of Benjamini–Krieger–Yekutieli. A, and B, are bulk RNA sequencing data.

Figure 5.

c-Myc expression is associated with MTB infection in the contained MTB infection (CMTB) mouse model of tuberculosis and in the pulmonary infection model. A, Lymph nodes from control and CMTB mice were depleted of CD3⁺/CD20⁺cells and subjected to scRNA-seq (n = 5 per condition). Major clusters were annotated by ImmGen matching, with the monocyte/macrophage cluster circled. B, Left: Uniform Manifold Approximation and Projection (UMAP) of the circled monocyte/macrophage cluster showing clear segregation of infected versus noninfected cells. Right: Single-cell c-Myc gene scores in infected versus control cells, calculated using AddModuleScore (aggregated expression of control gene set subtracted). C, Representative lung pathology slides from aerosol infected mice with and without CMTB and stained with an anti c-Myc antibody (n = 3 per condition). Lesion size and number of c-Myc-expressing cells were analyzed using a semiautomated pipeline. D, Left: Boxplot showing the density of c-Myc-expressing cells measured as number c-Myc positive cells by lesion area (million pixels) in CMTB mice compared to control. Right: Correlation between number of c-Myc-expressing cells in a lesion and area of the lesion, in CMTB mice and in control mice. The Pearson correlation coefficients (r) are indicated on the graph. Data were analyzed using a 2-tailed Student's t test.

Figure 6.

c-Myc expression is associated with the immune privileged niche in the human granuloma during active TB. A, Clinical samples of human granulomas from patients with active TB were stained with hematoxylin and eosin (H&E) (i) and an immunohistochemical stain for c-Myc (ii–iii). Inner (black lines) and outer (yellow lines) areas of granulomas were manually segmented (iii) followed by automatic nuclei detection and classification of mean c-Myc staining intensity into negative (blue) or positive: weak (yellow), moderate (orange), or strong (red) (iv). The scale bar in all histological panels represents 200 µm. B, Boxplot showing the percentage of c-Myc–positive nuclei in granuloma regions. Mean values are 32.3% for inner granulomas and 23.1% for outer granulomas. The box spans the first and third quartiles (the 25th and 75th percentiles), and the whiskers extend an additional 1.5 * inter-quartile range (or distance between the first and third quartiles). The horizontal gray line shows the percentage of positive nuclei outside granulomas (17.1%). C, Individual nuclei were grouped by distance to the rim of the nearest inner granuloma into 100 µm bins and the percentages of nuclei with positive staining by intensity were calculated; negative distances indicate the nucleus is within the inner granuloma. The total number of nuclei within each bin is represented at the top. Data were analyzed using a 2-tailed Student's t test.