Identification of conserved nucleomodulins in mycobacteria through functional screening.

(A) Schematic representation of the bioinformatic pipeline for identifying conserved nucleomodulins in Mycobacterium species. Genomic sequences from M. tuberculosis H37Rv, M. tuberculosis H37Ra, M. avium, M. marinum, and M. bovis BCG were analyzed. Signal peptides were predicted using SignalP 5.0 (D-score ≥ 0.5), non-classical secretion signals were identified using SecretomeP 2.0 (NN score ≥ 0.9), and nuclear localization signals (NLSs) were predicted using cNLS Mapper (score ≥ 2.0). (B) Classification of conserved cellular nucleomodulins. Secreted proteins were categorized based on the presence or absence of predicted NLS motifs. (C) Subcellular localization of EGFP-tagged candidate nucleomodulins. (left) Confocal microscopy images of seven Tat/SPI proteins fused to EGFP (green), nuclei were stained with DAPI (blue). (right) Quantification of nuclear EGFP fluorescence intensity. Scale bar, 10 µm. Data are presented as mean ± SD (n = 25 cells/group).

Hypothetical protein MgdE enters the host nucleus via dual NLS.

(A) Domain architecture of MgdE. Schematic representation of MgdE with the annotated functional domains, including a Tat signal peptide (1–37 aa, twin-arginine translocation motif), a GDSL-like lipase/acylhydrolase catalytic domain (74–258 aa), and two nuclear localization signals (NLS1: 108–111 aa, NLS2: 300–305 aa). (B) Phylogenetic and structural conservation of MgdE. (left) Neighbor–joining phylogenetic tree of MgdE homologs across Mycobacterium species (1,000 bootstrap replicates, values ≥50% shown). (right) Clustal Omega sequence alignment highlighting conserved residues (≥90% identity, red). Species abbreviations: M. tuberculosis (Mtu), M. bovis BCG (Mbb), M. tuberculosis H37Ra (Mra), M. africanum (Maf), M. tuberculosis CDC1551 (Mtc), M. tuberculosis GM041182 (Tbm), M. shinjukuense (Msh), M. marinum (Mma), M. lupini (Mli), M. avium (Mav), M. manitobense (Mman), M. intracellulare (Min). (C) Subcellular localization of EGFP-tagged wild-type MgdE and its NLS-deletion mutants (MgdEΔNLS1, MgdEΔNLS2, and MgdEΔNLS1-2). (left) Schematic representation of EGFP-tagged constructs. (right) Representative confocal microscopy images of HEK293T cells transfected with the indicated constructs for 36 h. EGFP fluorescence (green) and nuclear staining with DAPI (blue) were visualized using FLUOVIEW software (v5.0). Scale bar, 10 μm. Images were acquired using a 100× oil immersion objective (NA = 1.4). (D) Quantification of nuclear EGFP intensity in cells expressing wild-type or mutant MgdE constructs. Data are presented as mean ± SD (n = 15 cells). Statistical significance was assessed by two-tailed unpaired Student’s t-tests (***P < 0.001). (E) Western blot analysis of nuclear and cytoplasmic fractions from HEK293T cells transfected with wild-type MgdE and its NLS-deletion mutants. The empty EGFP vector was used as a negative control. EGFP and MgdE-EGFP fusion proteins were detected using an anti-GFP antibody. Histone H3 and β-actin served as nuclear and cytoplasmic markers, respectively. (F) Nuclear localization of MgdE during infection. THP-1 macrophages were infected with recombinant M. bovis BCG strains expressing Flag-tagged wild-type MgdE or NLS-deletion mutants (MgdEΔNLS1, MgdEΔNLS2, and MgdEΔNLS1-2) for 24 hours. Subcellular fractionation was performed, and cytoplasmic and nuclear fractions were analyzed by immunoblotting using anti-Flag antibodies. Histone H3 and β-actin were used as nuclear and cytoplasmic markers, respectively.

Nuclear localization of MgdE facilitates mycobacterial intracellular survival in macrophages.

(A–B) Intracellular survival of M. bovis BCG strains. THP-1 human macrophages (A) and RAW264.7 murine macrophages (B) were infected (MOI = 10) with wild-type BCG (WT), MgdE deletion mutant (ΔMgdE), complemented strain (Comp-MgdE), or NLS-deficient complement (Comp-MgdEΔNLS1-2). Bacterial survival was assessed by CFU enumeration at 2, 24, 48, and 72 h post-infection. (C–H) Cytokine expression in the infected THP-1 cells. qRT-PCR analysis of cytokine mRNA levels in THP-1 cells infected with WT or ΔMgdE strains for 4–24 h. Target genes include IL1B (C), IL6 (D), IL10 (E), CSF1 (F), CSF2 (G), and CSF3 (H). Data are presented as mean ±SD of three biologically independent experiments, analyzed using two-tailed unpaired Student’s t-tests (*P < 0.05, **P < 0.01, and ***P < 0.001).

MgdE directly interacts with ASH2L and WDR5, core components of the host COMPASS complex.

(A) Yeast cells were co-transformed with bait (pGBKT7) and prey (pGADT7) plasmids expressing wild-type MgdE and human COMPASS components (ASH2L, WDR5, RbBP5, and DPY30). Growth was monitored on non-selective (-Leu/-Trp, left) and selective (-Leu/-Trp/-Ade/-His + 200 ng/μL aureobasidin A, right) media. Controls: CK− (pGBKT7-lam + pGADT7-T, negative), CK+ (pGBKT7-p53 + pGADT7-T, positive). (B–C) Cells were co-transfected with Flag-MgdE and HA-tagged ASH2L, WDR5, or RbBP5 (1:1 molar ratio). At 36 h post-transfection, the lysates were immunoprecipitated using (B) anti-HA or (C) anti-Flag antibodies, followed by immunoblotting with anti-HA, anti-Flag, and anti-GAPDH (loading control). The input lanes represent 5% of the total lysate.

The conserved residues D224 and H247 mediate the binding ability of MgdE to WDR5.

(A) Y2H assay identifying interactions between MgdE mutants and COMPASS complex subunits. Yeast cells were co-transformed with bait (pGBKT7) and prey (pGADT7) plasmids expressing wild-type or mutant MgdE and human COMPASS subunits (ASH2L, WDR5, RbBP5, and DPY30). Growth was assessed on non-selective (-Leu/-Trp, left) and selective (-Leu/-Trp/-Ade/-His + 200 ng/μL aureobasidin A, right) media. Controls: CK− (pGBKT7-lam + pGADT7-T, negative) and CK+ (pGBKT7-p53 + pGADT7-T, positive). (B) Co-IP analysis of MgdE mutants with WDR5. HEK293T cells were co-transfected with Flag-tagged MgdE mutants and HA-tagged WDR5 (1:1 molar ratio). Complexes were immunoprecipitated using anti-HA antibody and protein A/G beads, followed by immunoblotting with anti-Flag and anti-HA antibodies. (C) Nuclear distribution of wild-type and mutants MgdE. Confocal microscopy of HEK293T cells expressing wild-type or D224A/H247A MgdE-EGFP at 12 and 24 h post-transfection (hpt). Nuclear foci were visualized by EGFP (green) and DAPI (blue) staining. Scale bar, 10 µm. Images were acquired with a ×100 oil immersion objective (NA = 1.4). (D) Immunoblot analysis of H3K4me3 levels. HEK293T cells expressing wild-type or D224A/H247A mutant MgdE were analyzed for changes in H3K4me3 levels over 0–24 h post-transfection. Histone H3 was used as a loading control. The data represent three independent biological replicates.

MgdE suppresses host inflammatory responses probably by inhibition of COMPASS complex-mediated H3K4 methylation.

(A) Volcano plot of DEGs. DEGs between MgdE-deleted strain (ΔMgdE) and wild-type BCG (WT) were visualized in a volcano plot. Genes with |log2-fold change| ≥ 1 and P < 0.05 were considered significant. The x-axis represents log2fold change, and the y-axis shows -log10(P-value). (B) GO enrichment analysis of DEGs. GO analysis revealed the significant enrichment of immune and inflammatory processes in ΔMgdE-infected macrophages compared to that in WT-infected cells, including positive regulation of response to external stimulus (GO:0032103), response to molecule of bacterial origin (GO:0002237), response to virus (GO:0009615), response to lipopolysaccharide (GO:0032496), leukocyte migration (GO:0050900), viral process (GO:0016032), regulation of response to biotic stimulus (GO:0002831), T cell activation (GO:0042110), and mononuclear cell differentiation (GO:1903131). (C) KEGG pathway enrichment analysis of DEGs. Chord diagram of KEGG pathway enrichment analysis showing signaling pathways that are significantly enriched in ΔMgdE. (D) Heatmap of the inflammatory gene expression. Heatmap depicting log₂-foldchange levels of inflammatory genes involved in the JAK–STAT and cytokine signaling pathways. Upregulated and downregulated genes in ΔMgdE are shown in green and red, respectively. The data were Z-score normalized. (E-G) qRT-PCR analysis of cytokine mRNA levels in THP-1 cells infected with WT, ΔMgdE, MgdE-complemented (Comp-MgdE), and NLS-deleted complement (Comp-MgdEΔNLS1-2) at 24 h post-infection. Cytokines analyzed include IL1B (E), IL6 (F), and IL10 (G). Data are presented as mean ± SD of three biologically independent experiments, analyzed using two-tailed unpaired Student’s t-tests (*P < 0.05, **P < 0.01, and ***P < 0.001).

Nuclear localization of MgdE is essential for mycobacterial survival in mice.

(A) Bacterial burden in the lungs of the infected mice. C57BL/6 mice (n = 6/group) maintained under SPF conditions were intratracheally infected with 1.0 × 107 colony-forming units (CFU) of M. bovis BCG strains, including wild-type (WT), MgdE-deleted (ΔMgdE), MgdE-complemented (Comp-MgdE), and NLS-deleted complement (Comp-MgdEΔNLS1-2). Lung bacterial loads were quantified using CFU assays at 0, 14, 21, 28, and 56 d post-infection. (B) H&E-stained lung sections from infected mice (as in A) revealed granulomatous inflammation. Scale bars: 200 μm. (C–D) Pro-inflammatory cytokine expression in mice spleen. qRT-PCR analysis of cytokine mRNA levels of Il1b (C) and Il6 (D) in spleen tissues from infected mice (n = 6/group) at 2 and 28 days post-infection. Data are presented as mean ± SD from six biologically independent experiments. Statistical significance was determined by a two-tailed unpaired Student’s t-test (*P < 0.05, **P < 0.01, and ***P < 0.001). (E) Mechanistic model showing how mycobacterial nucleomodulin MgdE hijacks the COMPASS complex to suppress H3K4me3 and promote immune evasion. Upon M. bovis BCG infection, the nucleomodulin MgdE is delivered into the host nucleus via its nuclear localization signal (NLS) and directly binds to the COMPASS complex subunits, ASH2L or WDR5. This interaction disrupts H3K4 trimethylation (H3K4me3) deposition, leading to the epigenetic suppression of pro-inflammatory cytokine transcription (e.g., IL6), thereby facilitating the intracellular survival of the pathogen.

Comparative analysis of classical and non-classical secreted proteins in mycobacterial species.

(A) Venn diagram showing the distribution of predicted classical secreted proteins in four Mycobacterium species: M. tuberculosis H37Rv (Mtu) and M. tuberculosis H37Ra (Mra), M. bovis BCG (Mbb), M. marinum (Mmar), and M. avium (Mav). Proteins were predicted using SignalP 5.0, with a signal peptide score (D-score) ≥ 0.5. A total of 125 proteins were conserved across all species. (B) Venn diagram showing the distribution of predicted non-classical secreted proteins across the same Mycobacterium species, predicted using SecretomeP 2.0 with a neural network (NN) score ≥ 0.9. Ten proteins were conserved across all species. Bar graphs (left: classical, right: non-classical) summarize the total number of predicted secreted proteins per species.

Subnuclear localization of MgdE-EGFP.

(A) Immunoblot analysis of bacterial lysates and culture supernatants from M. bovis BCG strains expressing C-terminally Flag-tagged MgdE. Endogenous Ag85B and GlpX were detected using anti-Ag85B and anti-GlpX antibodies, serving as positive and negative controls for protein secretion, respectively. MgdE-Flag was detected using an anti-Flag antibody. (B) Confocal microscopy was used to assess the nuclear distribution of MgdE-EGFP at various time points post-transfection. Nuclei were stained with DAPI (blue), MgdE-EGFP is shown in green. Scale bar: 10 μm. Images were acquired using a 100× oil immersion objective (NA = 1.4). (C) Quantification of nuclear EGFP intensity in cells expressing wild-type or mutant MgdE constructs. Data are presented as mean ± SD (n = 12 cells). (D) Quantitative RT-PCR analysis of mgdE mRNA expression in HEK293T cells at different time points post-transfection (4–48 h). Data represent the transcriptional level of mgdE relative to HPRT. (E) Western blot analysis of nuclear fractions at different time points post-transfection (4–48 h), showing time-dependent nuclear accumulation of MgdE-EGFP. Histone H3 was used as a loading control for nuclear proteins. The lower panel shows quantification of nuclear MgdE-EGFP levels normalized to Histone H3. Data are presented as mean ± SD (n = 3). (F) Quantification of the nuclear and cytoplasmic distribution of EGFP, wild-type MgdE, and its NLS-deletion mutants based on the Western blot results shown in (Figure 2E).

Deletion of the nuclear localization signal of MgdE does not affect the growth of M. bovis BCG strains.

(A-B) Construction and validation of the MgdE-deleted strain of M. bovis BCG. (A) Schematic diagram of the homologous recombination strategy used to delete mgdE from the M. bovis BCG genome. (B) Wild-type and mutant strains were used as templates to amplify the mgdE gene (600 bp upstream–600 bp downstream) by PCR. Lanes 1 and 3: wild-type genomic DNA, lanes 2 and 4: ΔmgdE genomic DNA. (C) Growth curve analysis of M. bovis BCG strains. Growth of BCG strains, including wild-type BCG (WT), MgdE-deleted (ΔMgdE), MgdE-complemented (Comp-MgdE), and NLS-deleted complemented (Comp-MgdEΔNLS1-2) strains, was measured in 7H9 medium.

MgdE interacts with COMPASS complex subunits.

(A) Predicted binding affinities between MgdE and COMPASS core subunits. The predicted local distance difference test (pLDDT) scores calculated using AlphaFold 2.2.0 for the interactions between MgdE and the COMPASS subunits were as follows: ASH2L (pLDDT = 0.47), RbBP5 (pLDDT = 0.30), WDR5 (pLDDT = 0.77), and DPY30 (pLDDT = 0.62). Confidence levels are categorized as follows: High confidence: pLDDT ≥ 0.7 (strong predicted binding), Medium confidence: 0.5 ≤ pLDDT < 0.7 (moderate binding), Low confidence: pLDDT < 0.5 (weak predicted binding). (B) Structural modeling of MgdE-COMPASS interactions. AlphaFold generated models of the simulated binding interfaces are shown, with MgdE highlighted in red and COMPASS subunits (ASH2L, WDR5, RbBP5, and DPY30) shown in gray. (C) Co-IP analysis of MgdE mutants with ASH2L. HEK293T cells were co-transfected with Flag-tagged MgdE mutants and HA-tagged ASH2L (1:1 molar ratio). Co-IP was performed using anti-HA antibody-bound protein A/G beads, and immunoprecipitated complexes were analyzed by immunoblotting with anti-Flag and anti-HA antibodies.

MgdE suppresses cellular inflammatory responses during M. bovis BCG infection.

(A) KEGG pathway enrichment analysis. KEGG pathway analysis identified significantly enriched pathways in THP-1 cells infected with the ΔMgdE strain compared to those infected with wild-type BCG (WT). DEGs were predominantly associated with immune and signaling pathways, including thyroid hormone synthesis (hsa04918), AMPK signaling pathway (hsa04152), PPAR signaling pathway (hsa03320), apelin signaling pathway (hsa04371), Cytokine–cytokine receptor interaction (hsa04060), JAK–STAT signaling pathway (hsa04630), and hematopoietic cell lineage (hsa04640). (B) Elevated inflammatory gene expression in ΔMgdE-infected cells. Key upregulated inflammatory genes in ΔMgdE-vs. WT-infected cells are highlighted. (C) Functional enrichment analysis of upregulated genes. Analysis using STRING and Cytoscape revealed regulation of inflammatory responses as a top enriched biological process (P < 0.05).

MgdE facilitates bacterial colonization in the spleens of infected mice.

Bacterial colonization was assessed in splenic homogenates from infected mice (as described in Figure 7A) by quantifying bacterial DNA using quantitative PCR at 2, 14, 21, 28, and 56 days post-infection. Data are presented as mean ± SD (n = 3). Statistical significance was determined using a two-tailed unpaired Student’s t-test (P < 0.05, **P < 0.01, and ***P < 0.001).