Macrophage anti-bacterial activity is controlled by adenylate kinase 4-mediated mitochondrial DNA synthesis

Abstract

Macrophage antibacterial activity requires mtROS production. The specific gene(s) that participates in the mtROS-mediated antibacterial process remains unclear. We showed that Listeria and Salmonella infections in human and mouse macrophages increased mtDNA copy number with which dictates antibacterial activity. Interestingly, adenylate kinase 4 (Ak4) expression was upregulated in macrophages after infection. Ak4 KO mice as well as macrophage-specific Ak4 KO mice became highly susceptible to bacterial infections. Ak4 is critical for the increase of mtDNA synthesis and mitochondrial mass in macrophages after bacterial infection. Biochemically, Ak4 transfers a phosphate group from ATP/GTP to (d)AMP for (d)ADP formation, and the K18A and G89S/A166D mutations abolished this function. Our results suggest that induction of Ak4 after infection produces more dADP, whose conversion to dATP in mitochondria supports mtDNA synthesis and the subsequent increase of mtROS production. Loss of this metabolic coupling in Ak4 KO macrophages diminishes antibacterial activity. Our findings highlight the vital role of Ak4 in macrophage defense against pathogenic bacteria.

Graphical abstract

Figure 1.

mtDNA synthesis is crucial for macrophage antibacterial activity. TG-pMacs or human THP-1 macrophages were infected with Listeria at a MOI of 5 or Salmonella at an MOI of 10 for 0, 2, 6, 12, and 24 h. (A) The mtDNA copy number in Listeria-infected TG-pMacs was determined by qPCR (n = 4). (B) Mean fluorescence intensity (MFI) of MitoTracker Green in Listeria-infected TG-pMacs was analyzed by flow cytometry (n = 4). (C) Pgc-1α and Tfam protein in Listeria-infected TG-pMacs were analyzed by western blotting. (D) mtDNA copy number in dNs-, Gem-, or ddC-treated TG-pMac with indicated concentration after Listeria infection was determined by qPCR (n = 4). (E and F) Intracellular bacterial loads in dNs-, Gem-, or ddC-treated TG-pMacs after infection with Listeria at an MOI of 5 (E) or Salmonella at an MOI of 10 (F) were assessed by plating cell lysates onto TSA plates and counting CFUs at 24 h after plating (n = 4). (G) THP-1 macrophages were infected with Listeria for the indicated time. mtDNA copy number in Listeria-infected THP-1 macrophages was determined by qPCR (n = 4). (H and I) Intracellular bacterial load in dNs-, Gem-, or ddC-treated THP-1 macrophages after infection with Listeria at an MOI of 5 (H) or Salmonella at an MOI of 10 (I) was assessed by plating cell lysates onto TSA plates and counting CFUs at 24 h after plating (n = 4). mtDNA copy number was normalized to nuclear DNA (nDNA). Protein expression levels were normalized to β-actin. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. Source data are available for this figure: SourceData F1.

Figure S1.

mtDNA copy number increases during Salmonella infection, and the expression of iNOS, IL-6, and TNF-α correlates with mtDNA copy number following Listeria infection. TG-pMacs or BMDMs were infected with Listeria at a MOI of 5 or Salmonella at an MOI of 10 for the indicated time. (A) mtDNA copy number in Salmonella-infected TG-pMacs was determined by qPCR (n = 4). (B) Mean fluorescence intensity (MFI) of MitoTracker Green in Salmonella-infected TG-pMacs was analyzed by flow cytometry (n = 4). (C) Pgc-1α and Tfam protein levels in Listeria-infected or 20 ng/ml of IL-4/IL-13–treated BMDMs were analyzed by western blotting. (D) iNos protein expressions in 0, 1, and 10 μM dNs-treated TG-pMacs with Listeria infection were analyzed by western blotting. (E) IL-6 and TNFα productions in 0, 1, and 10 μM dNs-treated TG-pMacs with Listeria infection were measured by ELISA (n = 4). (F) mtDNA copy number in dNs-, Gem-, or ddC-treated TG-pMac with indicated concentration after Salmonella infection was determined by qPCR (n = 4). (G) mtDNA copy number in Salmonella-infected THP-1 macrophages was determined by qPCR (n = 4). mtDNA copy number was normalized to nuclear DNA (nDNA). Protein expression levels were normalized to α-tubulin. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. Source data are available for this figure: SourceData FS1.

Figure 2.

Ak4 is essential for macrophage mitochondrial biogenesis after Listeria infection. (A) Heatmap illustrates differential expression of the Ak gene family in human monocyte-derived macrophages 24 h after Listeria infection (GSE34103). (B) RT-qPCR analysis of Ak family mRNA expression in TG-pMacs with or without Listeria infection (n = 4). (C) RT-qPCR analysis of Ak family mRNA expression in Listeria-infected WT and Ak4 KO TG-pMacs (n = 4). (D) WT and Ak4 KO pMacs were isolated 3 days after Listeria infection. Intracellular ATP levels (left) and ADP/ATP ratio (right) in WT (n = 6) and Ak4 KO (n = 7) pMacs were measured by bioluminescence. (E) Ak4 protein was determined in TG-pMacs after Listeria infection for the indicated time. (F) mtDNA copy number in Listeria-infected WT and Ak4 KO TG-pMacs was measured by qPCR at 0, 2, 6, 12, and 24 h after gentamicin treatment (n = 4). #, comparison with uninfected WT; $, comparison with uninfected Ak4 KO. (G) Mitochondria-encoded gene expressions in Listeria-infected WT and Ak4 KO TG-pMacs were analyzed by RT-qPCR at 24 h after gentamicin treatment (n = 3). (H) Representative images of Listeria-infected WT (n = 15) and Ak4 KO (n = 18) BMDMs were quantified at 24 h after infection using cryo-EM. Mitochondria were indicated by the red star. Scale bar, 1 μm or 200 nm. (I and J) MFI of MitoTracker Green (I) and MitoTracker Deep Red (J) measured mitochondrial mass and mitochondrial membrane potential, respectively, in WT and Ak4 KO pMacs, were analyzed by flow cytometry (n = 4). (K) OCR in Listeria-infected WT and Ak4 KO pMacs was measured using the Seahorse XF-96 analyzer (n = 3). mRNA expression levels were normalized to Actb, and protein levels were normalized to β-actin. mtDNA copy number was normalized to nDNA. Statistical significance was determined by an unpaired two-tailed Student’s t test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. nDNA, nuclear DNA; MFI, mean fluorescence intensity. Data are presented as mean ± SD. Source data are available for this figure: SourceData F2.

Figure S2.

Ak4 is significantly increased after Salmonella infection. Heatmap illustrates differential expression of the Ak gene family in mouse splenic macrophages 24 h after Salmonella infection (GSE183728), Statistical significance was determined by an unpaired two-tailed Student’s t test. **P < 0.01.

Figure S3.

Ak4 is essential for maintaining mitochondrial number and function after Listeria infection. (A) Schematic representation of the generation of conventional Ak4 KO mice using the CRISPR/Cas9 system. (B and C) Ak4 mRNA (B) and protein (C) levels in LPS/IFNγ-stimulated WT and Ak4 KO BMDMs were analyzed by RT-qPCR and western blotting, respectively (n = 3). (D–H) WT and Ak4 KO BMDMs were infected with or without Listeria for 6 h. (D and E) MFI of MitoTracker Green (D) and MitoTracker Deep Red (E) were analyzed by flow cytometry (n = 4). (F and G) OCR levels were measured by Seahorse analysis (n = 3). (H) MFI of MitoSox was analyzed by flow cytometry (n = 4). mRNA levels were normalized to Actb. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA (B and D–H). ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. MFI, mean fluorescence intensity. Source data are available for this figure: SourceData FS3.

Figure S4.

Ak4 controls mtDNA synthesis through the deoxynucleotide metabolism in macrophages after bacterial infection. (A) Pgc-1α and Tfam protein levels in WT and Ak4 KO BMDMs with or without Listeria infection were determined by western blotting. (B) mtDNA copy number in scramble and siTfam-treated BMDMs with or without Ak4 overexpression was measured by qPCR (n = 4). (C–G) WT and Ak4 KO TG-pMacs were pretreated with DMSO, 10 μM dNs, or 5 μM Gem for 1 h, followed by infection with Salmonella at an MOI of 10 for 1 h. Cells were then treated with 250 μg/ml gentamicin, washed with PBS, and maintained in 50 μg/ml gentamicin for 24 h (C) or 6 h (D–G) prior to analysis. (C) mtDNA copy number in WT and Ak4 KO TG-pMacs was measured by qPCR (n = 4). (D–G) MFI of MitoTracker Green (D and E) and MitoTracker Deep Red (F and G) in Salmonella-infected WT and Ak4 KO TG-pMacs was analyzed by flow cytometry (n = 4). Protein expression levels were normalized to α-tubulin. mtDNA copy number was normalized to nDNA. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA. *P < 0.05; **P < 0.01; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. nDNA, nuclear DNA; MFI, mean fluorescence intensity. Source data are available for this figure: SourceData FS4.

Figure 3.

Ak4 controls mtDNA synthesis through the deoxynucleotide metabolism in macrophages after Listeria infection. (A–H) WT and Ak4 KO TG-pMacs were pretreated with DMSO, 10 μM dNs (A, C, and F), 5 μM Gem (B, D, and G), or 10 μM ddC (B, E, and H) for 1 h, followed by infection with Listeria at an MOI of 5 for 1 h. Cells were then treated with 250 μg/ml gentamicin, washed with PBS, and maintained in 50 μg/ml gentamicin for 24 h (A and B) or 6 h (C–H) prior to analysis. (A and B) mtDNA copy number in WT and Ak4 KO TG-pMacs was measured by qPCR (n = 4). (C–H) MFI of MitoTracker Green (C–E) and MitoTracker Deep Red (F–H) in Listeria-infected WT and Ak4 KO TG-pMacs was analyzed by flow cytometry (n = 4). mtDNA copy number was normalized to nDNA. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. nDNA, nuclear DNA; MFI, mean fluorescence intensity.

Figure 4.

Ak4 kinase activity is required for the regulation of mtDNA synthesis. Mock, Ak4 WT, or kinase-dead Ak4 mutants were transduced into Ak4 KO BMDMs or TG-pMacs using lentiviral vectors. Mock-transduced WT macrophages served as controls. Cells were infected with Listeria at an MOI of 5 for 1 h, treated with 250 μg/ml gentamicin, washed with PBS, and maintained in 50 μg/ml gentamicin for either 24 h (C) or 6 h (D–F). (A) A ribbon diagram of the Ak4 model in complex with AMP and ATP is shown. Conserved regions within the NMP-binding domain are highlighted in green, the LID domain in red, and the P-loop in yellow. The mutated residues K18, G89, R122, N137, A166, and T199 are labeled accordingly. (B) Intracellular ATP levels were measured in lentiviral-transduced WT (mock) and Ak4 KO (mock, Ak4 WT, or kinase-dead mutant) Listeria-infected BMDMs (n = 6). ATP levels were normalized to those in Ak4 WT-transduced Ak4 KO BMDMs. Ak4 protein expression was assessed by western blotting (lower panel). (C) mtDNA copy number in transduced WT and Ak4 KO TG-pMacs was quantified by qPCR (n = 3). (D and E) MFI of MitoTracker Green (D) and MitoTracker Deep Red (E) in transduced TG-pMacs was measured by flow cytometry (n = 4). (F) OCR was assessed in transduced cells using an XF-96 analyzer (n = 3). Protein expression levels were normalized to β-actin. mtDNA copy number was normalized to nDNA. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of three independent experiments, and each point represents data from one mouse with two technical repeats. nDNA, nuclear DNA; MFI, mean fluorescence intensity. Source data are available for this figure: SourceData F4.

Figure S5.

Evolutionary conservation of Ak4 and effects of kinase-dead mutations in M0 macrophages. (A) Multiple sequence alignment of Ak4 orthologs from representative vertebrates is shown. The query Ak4 sequence was obtained via NCBI BLAST using house mouse (M, musculus) as the reference, with selected matches including human (H. sapiens, 90% sequence identity), striped hyena (H. hyaena, 90%), blue whale (B. musculus, 90%), domestic cattle (Bos taurus, 88%), Goode’s thornscrub tortoise (Gopherus evgoodei, 81%), American alligator (Alligator mississippiensis, 81%), red crowned crane (Grus japonensis, 82%), tiger rattlesnake (Crotalus tigris, 80%), and coelacanth (L. chalumnae, 74%). The secondary-structure annotation is displayed in the top row. Conserved regions within the NMP-binding domain are highlighted in green box, the LID domain in red box, and the P loop in yellow box. Mutated residues K18, G89, R122, N137, A166, and T199 are marked with triangles. (B) Mock, Ak4 WT, or kinase-dead Ak4 mutants were transduced into Ak4 KO BMDMs using lentiviral vectors. Mock-transduced WT BMDMs served as controls. Intracellular ATP levels were measured in lentiviral-transduced WT (mock) and Ak4 KO (mock, Ak4 WT, or kinase-dead mutant) M0 BMDMs (n = 3). Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats.

Figure 5.

Ak4 is essential to macrophage-mediated resistance to bacterial infection. WT and Ak4 KO TG-pMacs were infected with Listeria at an MOI of 5 for 1 h, followed by treatment with 250 μg/ml gentamicin. Cells were washed with PBS and maintained in 50 μg/ml gentamicin until harvest. (A) Concentrations of IL-1β, IL-6, TNFα, and IL-12 in the supernatant of Listeria-infected WT and Ak4 KO TG-pMacs at 24 h after infection were measured by ELISA (n = 4–7). (B) Transcript levels of Ccl2, Cxcl1, and Cxcl3 in Listeria-infected WT and Ak4 KO TG-pMacs were assessed by RT-qPCR at 2 h after gentamicin treatment (n = 3). (C and D) Bacterial burden in Listeria-infected WT and Ak4 KO TG-pMacs was quantified by CFU assays at 0 (C), 2, and 6 h (D) after Gem treatment (n = 4 per time point). (E) Enumeration of intracellular Listeria in WT (n = 19) and Ak4 KO (n = 13) BMDMs at 24 h after infection was measured using cryo-EM. Mitochondria were indicated by the red arrow. Scale bar, 2 μm. (F–K) Female and male mice were i.p. infected with 5 × 10⁴ and 5 × 10⁵Listeria, respectively. Survival was monitored daily for 14 days. On day 3 after infection, spleens and livers were harvested, homogenized, and plated on TSA plates to assess bacterial load. (F and G) Survival rate (n = 19 for WT and Ak4 KO) (F) and bacterial burden (n = 7 for WT and Ak4 KO) (G) in the liver and spleen of female mice. (H and I) Survival rate (n = 14 Ak4^f/f, n = 20 Ak4^f/fLysM^Cre) (H) and bacterial load (n = 5 for WT and Ak4 KO) (I) in the liver and spleen of female mice. (J and K) Survival rate (n = 18 for Ak4^f/f, n = 14 for Ak4^f/fCD4^Cre) (J) and bacterial load (n = 9 for Ak4^f/f, n = 5 for Ak4^f/fCD4^Cre) (K) in the liver and spleen of male mice. Data are presented as mean ± SD. Statistical significance was determined by unpaired two-tailed Student’s t test (A–E, G, I, and K) or log-rank test (F, H, and J). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of three independent experiments, and each point represents data from one mouse with two technical repeats.

Figure 6.

Ak4 enhances macrophage antibacterial activities by regulating mtDNA synthesis to boost mtROS production. WT and Ak4 KO TG-pMacs were pretreated with DMSO, dNs, Gem, or ddC at the indicated concentration for 1 h, followed by infection with Listeria at an MOI of 5 for 1 h. Cells were then treated with 250 μg/ml gentamicin, washed with PBS, and maintained in 50 μg/ml gentamicin for 6 h prior to analysis. (A and B) MFI of MitoSox and H₂DCFDA in WT and Ak4 KO pMacs from Listeria-infected mice for 3 days was analyzed by flow cytometry (n = 6–10). (C–E) WT and Ak4 KO TG-pMacs were pretreated with DMSO, 1 μM MitoPQ (a mitochondria-targeted redox cycling compound [C]), 200 μM MitoTempo (a mitochondria-targeted antioxidant [D]), or 20 mM NAC (a general cellular antioxidant [E]) for 1 h prior to Listeria infection. Intracellular bacterial loads were assessed by plating cell lysates onto TSA plates and counting CFUs at 24 h after plating (n = 3–4). (F) MFI of MitoSox in dNs-, Gem-, and ddC-treated WT TG-pMacs after Listeria infection was analyzed by flow cytometry (n = 4). (G–I) MFI of MitoSox in dNs- (G), Gem- (H), or ddC- (I) treated WT and Ak4 KO TG-pMacs after Listeria infection was analyzed by flow cytometry (n = 4). (J and K) MFI of MitoSox in dNs- (J) or Gem- (K) treated WT and Ak4 KO TG-pMacs after Salmonella infection was analyzed by flow cytometry (n = 4). (L) Mock, Ak4 WT, or kinase-dead Ak4 mutants were transduced into Ak4 KO TG-pMacs using lentiviral vectors. Mock-transduced WT TG-pMacs served as controls. MFI of MitoSox from transduced cells after Listeria infection was measured by flow cytometry (n = 4). Data are presented as mean ± SD. Statistical significance was determined by unpaired two-tailed Student’s t test (A and B) or one-way ANOVA (C–L). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats. MFI, mean fluorescence intensity.

Figure 7.

Ak4 regulates mtDNA synthesis through its kinase activity to enhance host defense against bacterial infection. WT and Ak4 KO TG-pMacs were pretreated with DMSO, 10 μM dNs, 5 μM Gem, or 10 μM ddC for 1 h, followed by infection with Listeria at an MOI of 5 or Salmonella at an MOI of 10 for 1 h. Cells were then treated with 250 μg/ml gentamicin, washed with PBS, and maintained in 50 μg/ml gentamicin for 6 h prior to analysis. (A–E) Intracellular bacterial loads of Listeria or Salmonella in dNs- (A and B), Gem- (C and D), or ddC-treated (E) WT and Ak4 KO TG-pMacs were assessed by plating cell lysates onto TSA plates and quantified by CFU assay at 24 h after plating (n = 4). (F) Mock, Ak4 WT, or kinase-dead Ak4 mutants were transduced into Ak4 KO TG-pMacs using lentiviral vectors. Mock-transduced WT TG-pMacs served as controls. Intracellular bacterial load was assessed by plating cell lysates onto TSA plates and quantified by CFU assay at 24 h after plating (n = 4). (G and H) Female mice were i.p. infected with 5 × 10⁴Listeria or 1 × 10⁴Salmonella. Survival was monitored daily for 14 days. On day 3 after infection, spleens and livers were harvested, homogenized, and plated on TSA plates to assess bacterial load. Bacterial burden (G) and survival rate (H) in WT (n = 12), Ak4 KO (n = 14), Ak4^K18 (n = 15), and Ak4^K18A (n = 16) mice after i.p. infection with 5 × 10⁴Listeria. (I) Survival rate in WT (n = 7), Ak4 KO (n = 8), Ak4^K18 (n = 8), and Ak4^K18A (n = 6) mice after i.p. infection with 1 × 10⁴Salmonella. Data are presented as mean ± SD. Statistical significance was determined by one-way ANOVA (A–G) or log-rank test (H–I). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are representative of two independent experiments, and each point represents data from one mouse with two technical repeats.