MAPK15 protects from oxidative stress-dependent cellular senescence by inducing the mitophagic process

Abstract

Mitochondria are the major source of reactive oxygen species (ROS), whose aberrant production by dysfunctional mitochondria leads to oxidative stress, thus contributing to aging as well as neurodegenerative disorders and cancer. Cells efficiently eliminate damaged mitochondria through a selective type of autophagy, named mitophagy. Here, we demonstrate the involvement of the atypical MAP kinase family member MAPK15 in cellular senescence, by preserving mitochondrial quality, thanks to its ability to control mitophagy and, therefore, prevent oxidative stress. We indeed demonstrate that reduced MAPK15 expression strongly decreases mitochondrial respiration and ATP production, while increasing mitochondrial ROS levels. We show that MAPK15 controls the mitophagic process by stimulating ULK1-dependent PRKN Ser¹⁰⁸ phosphorylation and inducing the recruitment of damaged mitochondria to autophagosomal and lysosomal compartments, thus leading to a reduction of their mass, but also by participating in the reorganization of the mitochondrial network that usually anticipates their disposal. Consequently, MAPK15-dependent mitophagy protects cells from accumulating nuclear DNA damage due to mitochondrial ROS and, consequently, from senescence deriving from this chronic DNA insult. Indeed, we ultimately demonstrate that MAPK15 protects primary human airway epithelial cells from senescence, establishing a new specific role for MAPK15 in controlling mitochondrial fitness by efficient disposal of old and damaged organelles and suggesting this kinase as a new potential therapeutic target in diverse age-associated human diseases.

Keywords: MAP kinases; Oxidative DNA damage; autophagy; cellular senescence; mitophagy; signal transduction.

FIGURE 1

MAPK15 regulates ATP production rates. In all experiments, HeLa cells were transfected with scrambled siRNA or two different siRNA against MAPK15 (#1 and #2). After 24 h, they were also transfected with MYC‐PRKN and after additional 48 h, we proceeded to analysis. (a) Analysis of ATP production. Comparison of mitochondrial ATP (mitoATP) production rate and glycolytic ATP (glycoATP) production rate at basal level. (b) Analysis of oxygen consumption rate (OCR) was performed upon subsequent additions of oligomycin, FCCP and the respiratory complex I and III inhibitors rotenone and antimycin A, as indicated. (c) Analysis of the extracellular acidification rate (ECAR) with glycolysis stress test. Subsequent additions of glucose, the ATP synthase inhibitor oligomycin, and the hexokinase inhibitor 2‐deoxy‐glucose (2‐DG) were carried out as indicated. One experiment, representative of 3 independent experiments, is shown

FIGURE 2

MAPK15 controls production of mt‐ROS. (a, c, e) Representative FACS histograms or (b, d, f) geometric mean fluorescent intensity (GeoMFI) of MitoSOX fluorescence, from HeLa cells transfected with scrambled siRNA or MAPK15 siRNA (#1) and, after 24 h, also transfected with MYC‐PRKN. After additional 48 h, samples underwent FACS analysis for MitoSOX red (5µM) fluorescence. Mitochondrial ROS were evaluated in basal conditions (a, b) and after Rotenone (4 h, 5 µM) (C, D), or FCCP (4 h, 30 µM) (e, f) insults. (g, i, k) Representative FACS histograms or corresponding (h, j, l) geometric mean fluorescent intensity (GeoMFI) Bars of MitoSOX fluorescence from HeLa cells transiently overexpressing MYC‐PRKN and empty vector (Ctrl) or MAPK15_WT or its mutant (AXXA, KD). Twenty‐four hours after transfection, samples underwent FACS analysis for MitoSOX red (5 µM) fluorescence. Mitochondrial ROS were evaluated in basal conditions (g, h) and after Rotenone (4 h, 5 µM) (I, J), or FCCP (4 h, 30 µM) (K, L) insults. Bars represents the standard deviation (SD) of 3 independent experiments (n = 3)

FIGURE 3

MAPK15 regulates mitochondrial volume and dynamics upon mitophagic stimuli. (a) HeLa cells were transfected with MYC‐PRKN and the empty vector (Ctrl) or MAPK15_WT or MAPK15_AXXA or MAPK15_KD. After 24 h, cells were treated with 30 µM FCCP or vehicle (8 h). Lysates were subjected to SDS‐PAGE followed by WB and analysed for indicated proteins. First two lanes do not express MYC‐PRKN. (b) HeLa cells were transfected with scrambled siRNA or MAPK15 siRNA (#1) and after 24 h, were transfected with MYC‐PRKN. After additional 48 h, samples were treated with 30 µM FCCP or vehicle (8 h). Lysates were then subjected to SDS‐PAGE followed by WB and analysed for indicated proteins. One experiment, representative of 3 independent experiments is shown. Densitometric analysis of bands is indicated. (c) HeLa cells were transfected with mCherry‐PRKN and Ctrl‐IRES‐GFP or MAPK15_WT‐IRES‐GFP or MAPK15_AXXA‐IRES‐GFP or MAPK15_KD‐IRES‐GFP. After 24 h, cells were incubated with 30 µM FCCP or vehicle (4 h). Cells were next fixed and subjected to immunofluorescence analysis. Scale bars correspond to 10 μm. (d) Mitochondria mean volume per cell, expressed in µm³. (e) HeLa cells were transfected with scrambled siRNA or MAPK15 siRNA (#1) and, after 24 h, further transfected with plasmid encoding for mCherry‐PRKN. After additional 48 h, cells were incubated 30 µM FCCP or vehicle (4 h). Cells were next fixed and subjected to immunofluorescence analysis. Scale bars correspond to 10 μm. (f) Mitochondria mean volume per cell, expressed in µm³. Bars represents the SD of 3 independent experiments (n = 3). (g) HeLa cells transiently overexpressing MYC‐PRKN and empty vector (Ctrl) or MAPK15_WT or MAPK15_AXXA or MAPK15_KD. Twenty‐four hours after transfection, samples were treated with 30 µM FCCP or vehicle (8 h). DNAs (10 ng) were subjected to qRT‐PCR for mitochondrially encoded NADH dehydrogenase 1 (MT‐ND1). The amount of PKM (pyruvate kinase M1/2), a nuclear‐encoded gene, was used for normalization purposes. (H) HeLa cells were transfected with scrambled siRNA or two MAPK15 siRNA (#1 or #2) and after 24 h, transfected with MYC‐PRKN. After additional 48 h, samples were treated with 30 µM FCCP or vehicle (8 h). DNAs (10 ng) were subjected a qRT‐PCR for MT‐ND1. The amount of PKM was used for normalization purposes. Bars represents average ratio ± SD between mitochondrial DNA and nuclear DNA (mt‐DNA:nDNA) of 3 independent experiments (n = 3)

FIGURE 4

MAPK15 regulates mitophagy by controlling the extent of PRKN activating phosphorylation. (a) HeLa cells stably expressing the empty vector (Ctrl) or MAPK15_WT were transfected with MYC‐PRKN and, after 24 h, were subjected to fractioning obtaining mitochondrial enrichment. Lysates were analysed for indicated proteins. Total, total lysate; Cyto, cytoplasmic fraction; Mito, mitochondrial enrichment. One experiment, representative of 3 independent experiments is shown (n = 3). Densitometric analysis of bands is shown. (b) HeLa cells stably expressing the empty vector (Ctrl), MAPK15_WT, MAPK15_AXXA or MAPK15_KD were transfected with MYC‐PRKN and, after 24 h, were treated with 30 µM FCCP (1 h), where indicated, and then subjected to fractioning obtaining mitochondrial enrichment. Lysates were subjected to SDS‐PAGE followed by WB for analysis for indicated proteins. (C) HeLa cells stably expressing pDsRed2‐Mito were transfected with MYC‐PRKN and empty vector (Ctrl) or HA‐MAPK15 WT. After 24 h, cells were treated 1 h with vehicle or 30 µM FCCP or with 100 nM BAF‐A1 or 30 µM FCCP plus 100 nM BAF‐A1. Cells were next fixed and subjected to immunofluorescence analysis. LC3B dots per cell on mitochondria were plotted as result of five representative fields. Representative images are given in Figure S5. (d) Same as in (c), but cells were stained for endogenous LAMP1. Number of mitochondria in acidic condition were plotted, as result of five representative fields. Representative images are given in Figure S6. Bars represents the SD of 3 independent experiments (n = 3). (e) HeLa cells were transfected with scrambled siRNA or MAPK15 siRNA. After 24 h, they were next transfected with MYC‐PRKN and mt‐Keima. After additional 48 h, mitophagy was analysed by confocal microscopy. Mitochondria in neutral condition (pH = 7) are shown in green, while mitochondria in acidic condition (pH = 4) are shown in magenta. Scale bars correspond to 10 μm. (f) Intensitometric analysis of normalized Magenta/Green mt‐Keima signal per cell ± SD of seven different fields from the experiment in (e). (g) HeLa cells were transfected with MYC‐PRKN and the empty vector (Ctrl) or MAPK15_WT and, after 24 h were collected and lysates were subjected to SDS‐PAGE followed by WB. One experiment, representative of 3 independent experiments, is shown. Densitometric analysis of bands is shown. (h) HeLa cells were transfected with MYC‐PRKN, the empty vector (Ctrl) or MAPK15_WT and, where indicated, with FLAG‐ULK1. Lysates were subjected to SDS‐PAGE followed by WB and analysed for indicated proteins. One experiment, representative of 3 independent experiments is shown. Densitometric analysis of bands is shown. (i) HeLa cells were transfected with scrambled siRNA or siRNA against MAPK15 (#1) and after 24 h, were transfected with MYC‐PRKN. After additional 48 h, lysates were subjected to SDS‐PAGE followed by WB. One experiment, representative of 3 independent experiments is shown. Densitometric analysis of bands is shown

FIGURE 5

MAPK15 prevents DNA damage induced by mt‐ROS and controls cellular senescence in HeLa and SH‐SY5Y cells. (a) HeLa cells were transfected with scrambled siRNA or two different MAPK15 siRNA (#1 or #2). After 24 h, they were transfected with mCherry‐PRKN. After additional 24 h, cells were treated with 100 µM mito‐TEMPO or vehicle (24 h). Cells were next fixed and subjected to immunofluorescence analysis. Scale bars correspond to 7,5 μm. Intensitometric analysis of nuclear γH2A.X fluorescence of five representative microscopy fields. (b) HeLa cells were transfected with scrambled siRNA or two different MAPK15 siRNA (#1 or #2) and after 24 h, they were transfected also with MYC‐PRKN. After additional 48 h, cells were fixed and incubated with a probe staining senescent cells (β‐galactosidase activity). Then, they were analysed by FACS. Bars represents the SD of 3 independent experiments (n = 3). (c) SH‐SY5Y cells were transfected with scrambled siRNA or two different MAPK15‐specific siRNA (#1 or #2). After 48 h, cells were treated with 100 µM mito‐TEMPO or vehicle (24 h). Cells were next fixed and subjected to immunofluorescence analysis. Scale bars correspond to 10 μm. (d) Intensitometric analysis of nuclear γH2A.X fluorescence from five representative microscopy fields from experiment in (c). (e) SH‐SY5Y cells were transfected with scrambled siRNA or two different MAPK15 siRNA (#1 or #2) and after 48 h, they were treated with 100 µM mito‐TEMPO or vehicle (24 h). After 72 h of transfection, cells were harvested and cell number was evaluated with Z2 Coulter Counter (Beckman Coulter). (f) SH‐SY5Y cells were transfected with scrambled siRNA or two different MAPK15 siRNA (#1 or #2) and after 48 h, they were treated with 100 µM mito‐TEMPO or vehicle (24 h). Then, they were analysed by FACS. Bars represents the SD of 3 independent experiments (n = 3). (g) SH‐SY5Y cells were transfected with scrambled siRNA or siRNA specific for MAPK15 (#1 or #2), followed by rescue with ectopically expressed wild‐type (WT), autophagy‐deficient (AXXA) or kinase‐dead (KD) MAPK15. Control sample was transfected with the empty vector. Total nuclear γH2A.X signal was quantified in five representative fields using Volocity software and the graph represents intensitometric analysis of nuclear γH2A.X fluorescence. Bars represent SD of five representative microscopy fields. (h) Same as in (g), but cells were analysed by FACS for β‐Gal activity. Bars represents the SD of 3 independent experiments (n = 3)

FIGURE 6

Reduced MAPK15 expression triggers cellular senescence in primary human Airway Epithelial Cells (hAEC). (a) hAEC cells were transfected with scrambled siRNA or two different MAPK15 siRNA (#1 or #2) and, after 72 h, they were harvested and cell number was evaluated with Z2 Coulter Counter (Beckman Coulter). (b) Same as in (a), but cells were lysed and subjected to WB analysis with indicated antibodies. (c) Same as in (a), but cells were fixed and incubated with a probe staining senescent cells (β‐galactosidase activity). Then, they were analyzed by FACS. (d) Same as in (a), but cells were fixed and subjected to immunofluorescence analysis with indicated antibody. Scale bars correspond to 10 μm. The accompanying graph shows intensitometric analysis of nuclear γH2A.X fluorescence from five representative microscopy fields. (e) Same as in (a), but cells were subjected to immuno‐FISH analysis. Representative images of DNA damage colocalizing with telomere (magenta, telomere; green, 53BP1; blue, nuclei) where white arrows indicate colocalization between telomeres and 53BP1. The accompanying graph shows the mean number of 53BP1 foci (red) and the colocalization between telomeres and 53BP1 (blue) per cell, >100 cells were analysed. Scale bars correspond to 5 μm. (f) Same as in (a), but cells were collected and subjected to qRT‐PCR to monitor mRNA expression of different cytokines associated to senescent‐associated secretory phenotype (SASP)