Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 11 (1), 42

Tissue-infiltrating Macrophages Mediate an Exosome-Based Metabolic Reprogramming Upon DNA Damage

Affiliations

Tissue-infiltrating Macrophages Mediate an Exosome-Based Metabolic Reprogramming Upon DNA Damage

Evi Goulielmaki et al. Nat Commun.

Abstract

DNA damage and metabolic disorders are intimately linked with premature disease onset but the underlying mechanisms remain poorly understood. Here, we show that persistent DNA damage accumulation in tissue-infiltrating macrophages carrying an ERCC1-XPF DNA repair defect (Er1F/-) triggers Golgi dispersal, dilation of endoplasmic reticulum, autophagy and exosome biogenesis leading to the secretion of extracellular vesicles (EVs) in vivo and ex vivo. Macrophage-derived EVs accumulate in Er1F/- animal sera and are secreted in macrophage media after DNA damage. The Er1F/- EV cargo is taken up by recipient cells leading to an increase in insulin-independent glucose transporter levels, enhanced cellular glucose uptake, higher cellular oxygen consumption rate and greater tolerance to glucose challenge in mice. We find that high glucose in EV-targeted cells triggers pro-inflammatory stimuli via mTOR activation. This, in turn, establishes chronic inflammation and tissue pathology in mice with important ramifications for DNA repair-deficient, progeroid syndromes and aging.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. DNA damage accumulation in Er1F/− macrophages.
a Lys-Cre-driven Rosa-YFP expression in thioglycolate-elicited peritoneal macrophages (TEMs; n > 500 cells counted per genotype). The numbers indicate the average percentage of GFP (+) cells ± SEM), b Lys-Cre-driven Rosa-YFP expression in hepatocytes and primary pancreatic cells (PPCs) shown by confocal microscopy (n > 100 cells counted per genotype; the numbers indicate the average percentage of GFP (+) cells ± SEM) and c western blotting. d Immunofluorescence staining of Lys-Cre-driven Rosa-YFP expression in the Er1F/− pancreas and the white adipose tissue (WAT) that are infiltrated with MAC1-positive macrophages (indicated by the arrowheads). e Western blotting of ERCC1 protein in whole-cell (w) cytoplasmic (c) and nuclear (n) extracts. Tubulin (TUB), and Fibrillarin (FIB) were used as loading controls (as indicated). The graph represents the fold change (F.C.) of ERCC1 protein levels in Er1F/− samples compared to corresponding Er1F/+ controls (n = 3). f Cell type-specific ablation of ERCC1 (indicated by the arrowhead) in bone marrow-derived (BMDMs) and TEMs expressing the macrophage-specific antigen MAC1. The numbers indicate the average percentage of ERCC1 (+) nuclei ± SEM in Er1F/+and Er1F/− BMDMs and TEMs (n > 150 cells were counted per genotype). g Immunofluorescence detection of γ-H2AX in Er1F/− and Er1F/+ BMDMs and h TEMs. i Immunofluorescence detection of FANCI, pATM and RAD51 in Er1F/− and Er1F/+ BMDMs (in each case n > 200 cells were counted per genotype). j Immunofluorescence detection of pATM in Er1F/− and Er1F+ TEMs (n > 150 cells were counted per genotype). k Immunofluorescence detection of Caspase 3 (CASP3) (n > 300 cells were counted per genotype) and l GL13 (indicated by the arrowhead), commercially available SenTraGor®, in Er1F/− and Er1F+ BMDMs. Fluorescence intensity was calculated in n > 50 cells per genotype. Gray line is set at 5 μm scale, unless otherwise indicated. Error bars indicate S.E.M. among replicates (n ≥ 3). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test).
Fig. 2
Fig. 2. Abrogation of ERCC1 triggers cytoplasmic stress responses in Er1F/− macrophages.
a Immunofluorescence detection of GRP78 (n > 250 cells counted per genotype) marking the dilation of ER, LC3 (n > 40 cells counted per genotype) for autophagy, P62 for autophagic activity (n > 15 optical fields per genotype, ~150 cells per field) and b Gm130 for Golgi dispersal (~500 cells per genotype; arrowhead) in Er1F/− and Er1F/+ BMDMs. For GRP78 and LC3, the colored numbers indicate the average percentage of (+) stained cells ± SEM for the indicated, color-matched protein. For p62, the colored numbers indicate the average mean fluorescence intensity ± SEM of p62 signal. For Gm130, the green-colored numbers indicate the average percentage of (+) stained cells ± SEM showing Golgi dispersal. c Western blot levels of GRP78, P62, LC3, and Tubulin in Er1F/− and Er1F+ BMDMs. The graph shows the fold change of indicated protein levels in Er1F/− BMDMs compared to Er1F+ corresponding controls (n = 3 per group). d Immunofluorescence detection of GRP78 (for ER stress) (n > 500 cells counted per genotype), LC3 (for autophagy) (n > 750 cells counted per genotype) and e Gm130 (for Golgi dispersal) in MMC-treated BMDMs (n > 500 cells counted per genotype). Numbers indicate the average percentage of (+) stained cells ± SEM for the indicated, color-matched protein. For Gm130, the green-colored numbers indicate the average percentage of (+) stained cells ± SEM showing Golgi dispersal. f Immunofluorescence detection of LC3 (for autophagy) (~250 cells counted per genotype) and Gm130 (for Golgi dispersal) (~950 cells counted per genotype) in LPS-stimulated BMDMs. g Immunofluorescence detection of Gm130 (for Golgi dispersal) in MMC-treated and control BMDMs exposed to ATM (ATMi) (~200 cells counted) or ATR (ATRi) (500 cells counted) inhibitor (as indicated). The green-colored numbers indicate the average percentage of (+) stained cells ± SEM showing Golgi dispersal. h Western blot levels of GRP78 and LC3 in MMC-treated and control (ctrl) macrophages exposed to ATM (ATMi) or ATR (ATRi) inhibitor (as indicated; Tubl.: tubulin, unt: untreated). The graph represents the fold change in indicated protein levels in MMC-treated macrophages exposed to ATM (ATMi) or ATR (ATRi) inhibitor compared to corresponding controls (n = 4 per group) (ij). Representative transmission electron micrographs of Er1F+ (i) and Er1F/− (jl) BMDMs. Arrowheads depict the presence of intracellular vesicles (j left panel), organized in larger vacuolar structures (j right panel), the appearance of cytoplasm-filled projections (k left panel), the convoluted network of pseudopodia-like structures (k right panel) containing vesicles (l left panel) and pseudopodia-associated extracellular vesicles (l right panel). Scale bars are shown separately for each micrograph. The significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test). Gray line is set at 5 μm scale.
Fig. 3
Fig. 3. The ERCC-XPF defect in macrophages triggers metabolic changes in Er1F/− mice.
a Weights curves of 2-months-old Er1F/− and Er1F/+ animals (n = 8) over a period of 22 weeks. b Glucose tolerance test (GTT) graphs of 2-months-old Er1F/ and Er1F/+ mice fed on a normal diet for a period of 2-, 4-, and 6-months (M), as indicated. c GTT graphs of 2-months-old Er1F/− and Er1F+ mice (n = 10) fed on a high-fat diet for a period of 4 months (M). d Steady-state glucose serum levels of 4- and 6-months (M) old Er1F/− and Er1F/+ mice after 2 h of fasting (n = 8). e Insulin serum levels of 8-months (M) old Er1F/− (n = 10) and Er1F/+ (n = 7) mice (f) Insulin tolerance test (ITT) graphs of 8-months (M) old Er1F/− and Er1F/+ mice (n = 8). g Representative periodic acid–Schiff (PAS) staining and quantification (3 optical fields per animal) of glycogen in Er1F/+ and Er1F/− livers (n = 4 animals per genotype). h Gyg1, Gys, Pygl, and Gsk3 mRNA levels in Er1F/+ (red dotted line) and Er1F/− livers. i Representative Red-oil staining and quantification (three optical fields per animal) of triglycerides in the liver of Er1F+ and Er1F/− mice (n = 3) fed on normal or high fat diet (as indicated); arrowhead indicates the decrease in fat deposition in the livers of Er1F/− animals fed on normal or high fat diet (as indicated). Error bars indicate S.E.M. among replicates (n ≥ 3). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test). Gray line is set at 5 μm scale.
Fig. 4
Fig. 4. Systemic inflammation and gene expression changes in Er1F/− mice.
a Infiltration of foamy cells (macrophages) in a region of lipogranuloma in Er1F/− perirenal fat (as indicated) and inflammatory infiltration of lymphocytes and monocytes in Er1F/− kidneys. b Inflammatory infiltration of lymphocytes and monocytes in Er1F/− lungs and livers. See also Supplementary Fig. 3E, F for magnified inlays. The graph indicates the percentage of inflammatory (inflam.) infiltrates per tissue area in Er1F/− (n = 9) and Er1F/+ mice (n = 7). c Detection of GL13 (+) macrophages in perirenal fat of Er1F/− (n = 5) and Er1F/+ (n = 7) mice. d Immunofluorescence detection of PECAM-1, ICAM-1 and VCAM-1 along with CD45 and MAC1 (shown by the arrows) in the liver (n = 4; 4–6 optical fields per animal). e pancreas (n = 3; 3 optical fields per animal) and f the white adipose tissue (WAT) (n = 3; two optical fields per animal) of Er1F/− and Er1F+ mice indicating the expression of cell adhesion molecules and the presence of monocytic/lymphocytic infiltrates in Er1F/− tissues. Colored numbers indicate the average mean fluorescence intensity ± SEM for the color-matched protein (as indicated). g Heat-map representation of significant gene expression changes (n = 1756 genes) in Er1F/− BMDMs compared to corresponding control cells. h Over-represented GO biological processes and i pathways (Reactome) of Er1F/− BMDMs compared to corresponding control cells; p: −log of p-value which is calculated by Fisher’s exact test right-tailed, R: ratio of number of genes in the indicated pathway divided by the total number of genes that make up that pathway. j Heat-map representation of gene expression changes associated with significantly over-represented biological processes in Er1F/− BMDMs compared to corresponding control cells (as indicated). k Interleukin and chemokine mRNA levels in Er1F/− compared to Er1F/+ (red dotted line) BMDMs. Error bars indicate S.E.M. among replicates (n ≥ 3). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 and ***≤0.005 (two-tailed Student’s t-test), “+”: one-tailed-t-test. F.E.: fold enrichment, W: (Er1F/+), E: Er1F/−. See also supplementary data 1. Gray line is set at 10 μm scale.
Fig. 5
Fig. 5. DNA damage promotes the generation and secretion of extracellular vesicles (EVs) in Er1F/− macrophages.
a Schematic representation of the high-throughput MS analysis in Er1F/− compared to Er1F/+ BMDMs media. b Venn’s diagram of proteins identified in Er1F/− media from two independent biological replicates. c List of significantly over-represented GO terms associated with Cellular Component. d Number of observed (obs.) and expected (exp.) known protein interactions within the core 211 shared proteins set. e Schematic representation of the major protein complex identified in BMDM media. f Western blot analysis of CD9, ALIX, RAB10, and RAC1 proteins levels in the EV fraction of Er1F/− and Er1F/+ sera (n = 6; see also Supplementary Fig. 5A; left panel). g Transmission electron microscopy of EVs marking the presence of exosomes with a size 30–80 nm in Er1F/− TEM media. h Western blot analysis of CD9, ALIX, RAB10, RAC2, and RAC1 proteins levels in Er1F/− compared to Er1F/+ EV fraction of BMDM media (n = 5). A graph showing the fold change and statistical significance of the indicated protein levels is shown in Supplementary Fig. 5A; right panel. i Western blot analysis of CD9, ALIX, RAB10, RAC2, and RAC1 proteins levels in the EV fraction of media derived from the MMC-treated and control BMDMs exposed to ATM (ATMi) or ATR (ATRi) inhibitors (as indicated; n = 3). A graph showing the fold change and statistical significance of the indicated protein levels is shown in Supplementary Fig. 5C. j IL8 and IL6 protein levels in Er1F/− and Er1F/+ sera and BMDM media (as indicated). k Immunofluorescence detection of RAC1 (~500 cells per genotype), RAB10 (~150 cells per genotype) and RAC2 (~150 cells per genotype) in Er1F/− and Er1F/+ PPCs (n > 400 cells per genotype) (see also Supplementary Fig. 5e for RHOA), l hepatocytes (n > 100 cells per genotype) and m thioglycolate-elicited macrophages (TEMs) (n > 500 cells per genotype). Colored numbers indicate the average percentage of positively stained cells ± SEM for the indicated, color-matched protein. Error bars indicate S.E.M. among replicates (n ≥ 3). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test). (nd): not detected. Gray line is set at 5 μm scale.
Fig. 6
Fig. 6. Er1F/− EVs stimulate the glucose uptake in EV-recipient cells.
a Immunofluorescence detection of the pKH67-labelled EVs in PPCs. Numbers indicate the average percentage of positive cells ± SEM (n > 800 cells counted in three independent experiments). b Immunofluorescence detection of RAB10-GFP and c RAC1-GFP in PPCs treated with EVs from Er1F/− BMDMs transfected with RAB10-GFP or RAC1-GFP. Numbers indicate the average percentage of positively stained cells ± SEM for the indicated, color-matched protein. (n > 700 cells counted in three independent experiments). d Immunofluorescence detection of fluorescent tracer 2-NBDG for the monitoring of glucose uptake in PPCs (~2500 cells per genotype) and hepatocytes (~250 cells per genotype) exposed to Er1F/− and Er1F/+-derived EVs (as indicated). e Immunofluorescence detection of fluorescent tracer 2-NBDG for the monitoring of glucose uptake in PPCs exposed to Er1F/− and Er1F/+ EVs derived from animal sera (ser.; n > 700 cells per genotype). f Immunofluorescence detection of fluorescent tracer 2-NBDG for the monitoring of glucose uptake in PPCs exposed to MMC EVs (as indicated, n > 2000 cells per treatment). g Western blotting of GLUT1 and GLUT3 protein levels in PPCs exposed to Er1F/− and Er1F/+-derived EVs (n = 4). A graph showing the fold change and statistical significance of GLUT1 and GLUT3 protein levels is shown in Supplementary Fig. 7E. h Glut1 mRNA levels (in fold change; fc) in PPCs exposed to EVs derived from Er1F/− (Er1F/− EVs) or MMC-treated (MMC-EVs) macrophages compared to untreated wt. macrophages (red dotted line) for 24 h (see also Supplementary Fig. 7B–D). i Immunofluorescence detection of GLUT1 and RAB10 in PPCs exposed to Er1F/− and Er1F+-derived EVs (see also Supplementary Fig. 7G, H). Numbers indicate the average percenatge of GLUT1, RAB10 double positive cells in each experimental condition (n > 200 cells counted in three independent experiments). j Oxygen consumption rate (OCR) of PPCs maintained in the presence of 15 mmol glucose that were exposed to Er1F/+, Er1F/− or MMC-treated, macrophage-derived EVs (upper panel) and to Er1F/+ or Er1F/− serum-derived EVs (lower panel); numbers indicate the % increase in OCR compared to corresponding controls. k Western blotting of GLUT1 and GLUT3 protein levels in Er1F/− and Er1F/+ liver and pancreata (n = 3). A graph showing the fold change and statistical significance of GLUT1 and GLUT3 protein levels is shown in Supplementary Fig. 8A. l Immunofluorescence detection of GLUT1 and GLUT3 in the pancreas of Er1F/− and Er1F/+ mice (n = 3, 2–3 optical fields per animal; see also Supplementary Fig. 8B). Numbers indicate the average mean fluorescence intensity. Grey line is set at 10 μm scale. Error bars indicate S.E.M. among replicates (n ≥ 3). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test). Gray line is set at 5 μm scale.
Fig. 7
Fig. 7. Glucose uptake activates mTOR and pro-inflammatory responses in EV-recipient cells.
a Immunofluorescence detection of iNOS accumulation (indicated by the arrowhead) in PPCs exposed to culture media (CM) and the EVs derived from MMC-treated and untreated control macrophages (n = 3, >600 cells per treatment) or b Er1F/− and Er1F/+ macrophages (n = 3, > 1200 cells/treatment) (see also Supplementary Fig. 8D). c Immunofluorescence detection of NF-kβ nuclear translocation (indicated by the arrowhead) in PPCs exposed to CM and the EVs derived from MMC-treated and untreated control macrophages or (n = 3, >700 cells per treatment). d Er1F/− and Er1F/+ macrophages (see also Supplementary Fig. 8E). (n = 3, >500 cells per treatment). e Immunofluorescence detection of NF-kβ (shown by the arrowhead) in PPCs exposed to CM supplemented with EVs from MMC-treated and control macrophages upon low (5 mmol) or high (15 mmol) glucose concentration (n = 4, >750 cells per treatment; see also Supplementary Fig. 9A–C). f Western blotting of phosphorylated pS6K, phosphorylated p4EBP1, phosphorylated pAKT1, phosphorylated pAKT2 and phosphorylated pAKT protein levels in Er1F/− and Er1F/+ pancreata (n = 4). The graph represents the fold change (F.C) of indicated protein levels in Er1F/− pancreata to wt. controls. g Western blotting of pS6K, p4EBP1 and iNOS protein levels in MMC-treated macrophages exposed to rapamycin (n = 3, Rap/cin; as indicated). The graph represents the fold change of protein levels in MMC-treated macrophages exposed to rapamycin compared to MMC-treated macrophage control (ctrl.) cells. h Immunofluorescence detection of NF-kβ and 2-NBDG in PPCs (indicated by the arrowheads) exposed to CM supplemented with EVs derived from Er1F/− and Er1F/+ BMDMs in the presence or absence of rapamycin (Rap/cin). (n = 6, > 1000 cells counted per treatment). The graph shows the percenatge of positively stained cells. Error bars indicate S.E.M. among replicates (n ≥ 3). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test). Gray line is set at 5 μm scale.
Fig. 8
Fig. 8. Exogenous delivery of Er1F/− exosomes promote GLUT1 accumulation, glucose uptake and inflammation in vivo.
a Immunofluorescence detection of GLUT1 in the liver of 4-weeks-old C57BL/6 animals intravenously injected with EVs derived from either Er1−/− or wt. MEFs and b Er1F/+ or Er1F/− macrophages (as indicated). Numbers show average mean fluorescence intensity ± SEM, n = 4 animals per genotype, >3 optical fields/animal. c Western blotting of GLUT1 in the muscle and liver of C57BL/6 animals intravenously injected with EVs derived from either Er1−/− or wt. MEFs and d Er1F/+ or Er1F/− macrophages (as indicated). The graph represents the fold change (F. C) of GLUT1 protein levels in the liver or muscle tissues of animals treated with Er1−/− MEF or Er1F/− macrophage-derived EVs as compared to corresponding tissues of animals treated with Er1+/+ MEFs or Er1F/+ macrophage-derived EVs. e Glucose tolerance test (GTT) graphs of C57BL/6 mice injected intraperitoneally every 24 h for a period of 10 days with EVs derived from either Er1−/− or wt. MEFs (n = 6) and Er1F/+ or Er1F/− macrophages (n = 8) (as indicated). f The accumulation of irreparable DNA lesions in tissue-infiltrating macrophages activates the secretion of EVs in vivo and ex vivo. Er1F/− macrophage-derived EVs are targeted to recipient tissues and cells triggering the expression and translocation of insulin-independent glucose transporters GLUT1 and 3 in cell membrane. This leads to an increase in cellular glucose uptake in cells and greater tolerance to glucose challenge in higher cellular oxygen consumption rate and greater tolerance to glucose challenge in Er1F/− mice. In turn, high glucose levels activate pro-inflammatory responses in an mTOR-dependent manner leading to chronic inflammation in Er1F/− animals. Error bars indicate S.E.M. among replicates (n is indicated in each panel). Asterisk indicates the significance set at p-value: *≤0.05, **≤0.01 (two-tailed Student’s t-test). Gray line is set at 10 μm scale.

Similar articles

See all similar articles

References

    1. Harper JW, Elledge SJ. The DNA damage response: ten years after. Mol. Cell. 2007;28:739–745. doi: 10.1016/j.molcel.2007.11.015. - DOI - PubMed
    1. Hoeijmakers JH. Genome maintenance mechanisms for preventing cancer. Nature. 2001;411:366–374. doi: 10.1038/35077232. - DOI - PubMed
    1. Gregg SQ, Robinson AR, Niedernhofer LJ. Physiological consequences of defects in ERCC1-XPF DNA repair endonuclease. DNA Repair (Amst.) 2011;10:781–791. doi: 10.1016/j.dnarep.2011.04.026. - DOI - PMC - PubMed
    1. Apostolou Z, Chatzinikolaou G, Stratigi K, Garinis GA. Nucleotide excision repair and transcription-associated genome instability. Bioessays. 2019;41:e1800201. doi: 10.1002/bies.201800201. - DOI - PubMed
    1. Kamileri I, Karakasilioti I, Garinis GA. Nucleotide excision repair: new tricks with old bricks. Trends Genet.: TIG. 2012;28:566–573. doi: 10.1016/j.tig.2012.06.004. - DOI - PubMed

Publication types

Feedback