Autophagy proteins LC3B, ATG5 and ATG12 participate in quality control after mitochondrial damage and influence lifespan - PubMed (original) (raw)

Autophagy proteins LC3B, ATG5 and ATG12 participate in quality control after mitochondrial damage and influence lifespan

Sören Mai et al. Autophagy. 2012 Jan.

Abstract

Mitochondrial health is maintained by the quality control mechanisms of mitochondrial dynamics (fission and fusion) and mitophagy. Decline of these processes is thought to contribute to aging and neurodegenerative diseases. To investigate the role of mitochondrial quality control in aging on the cellular level, human umbilical vein endothelial cells (HUVEC) were subjected to mitochondria-targeted damage by combining staining of mitochondria and irradiation. This treatment induced a short boost of reactive oxygen species, which resulted in transient fragmentation of mitochondria followed by mitophagy, while mitochondrial dynamics were impaired. Furthermore, targeted mitochondrial damage upregulated autophagy factors LC3B, ATG5 and ATG12. Consequently these proteins were overexpressed in HUVEC as an in vitro aging model, which significantly enhanced the replicative life span up to 150% and the number of population doublings up to 200%, whereas overexpression of LAMP-1 did not alter the life span. Overexpression of LC3B, ATG5 and ATG12 resulted in an improved mitochondrial membrane potential, enhanced ATP production and generated anti-apoptotic effects, while ROS levels remained unchanged and the amount of oxidized proteins increased. Taken together, these data relate LC3B, ATG5 and ATG12 to mitochondrial quality control after oxidative damage, and to cellular longevity.

PubMed Disclaimer

Figures

Figure 1

Mitochondrial dynamics does not act in early quality control after ROS-induced mitochondrial damage. (A) Young HUVEC were treated for 10 min with hydrogen peroxide (final concentration 13.2 mM) or were stained with MTR and irradiated for 15 or 30 min. Immediately after treatment the amount of oxidized proteins in mitochondrial and whole cell lysates was analyzed by oxyblotting. After irradiation, mitochondrial lysates contained a significantly higher amount of oxidized proteins compared with whole cell lysates. In contrast, hydrogen peroxide-treated cells showed a much higher content of oxidized proteins in the whole cell lysate compared with the mitochondrial fraction; n = 3; p < 0.05. (B) Mitochondria in HeLa cells transfected with mtCFP were stained with MTR and either irradiated or not treated, subsequently co-cultured with nonirradiated mtGFP transfected HeLa cells and after 4 h fused by PEG. Cycloheximide (CHX) was added 30 min before PEG fusion to prevent the influence of ongoing protein synthesis. Samples were fixed at the indicated time points after PEG treatment and depicted by CLSM. While in controls complete mixing of the fluorescent proteins was observed, no fusion of fragmented mitochondria in irradiated cells took place. Color coding: green, GFP; blue, CFP; red, MTR; magenta, GFP, CFP and MTR; white, MTR plus CFP; yellow, MTR and GFP. (C) The PEG fusion experiments were quantified by Pearson correlation. A Pearson correlation factor of 1 means 100% colocalization, while low and negative values stand for lack of colocalization and strict separation, respectively. In control cells, an almost complete colocalization of mtCEF/MTR and mtGFP was observed while no colocalization of mtCEF/MTR and mtGFP was detected in the irradiated samples; n = 5−11; p < 0.05.

Figure 2

Mitophagy is the primary QC mechanism after ROS-induced mitochondrial damage. (A) HUVEC were transiently transfected with GFP-Parkin, irradiated after staining with MTR, and analyzed by CLSM at the indicated time points. In nontreated (nt) cells GFP-Parkin never associated with mitochondria at any time. In contrast, GFP-Parkin translocated in irradiated cells 2 h after irradiation to fragmented mitochondria, while 8 h after irradiation GFP-Parkin was again almost completely situated in the cytosol. (B) LC3-II, LC3-I and actin protein levels were determined in irradiated and control cells by western blotting at the indicated time points. (C) LC3-II western blots as shown representatively in (B) were normalized to actin protein content. The LC3-II/actin ratio of nonirradiated control cells was set as 1 for each time point and the LC3-II/actin ratio of irradiated cells was related to the respective control at the same time point. A time-dependent increase of the LC3-II/actin ratio after irradiation compared with nonirradiated control cells occurred; n = 6; p < 0.05. (D and E) Starting from 24 h after irradiation MTR-stained mitochondrial clusters appeared that could not be re-stained with MitoTracker Green, indicating loss of membrane potential (D, representative picture taken 72 h post irradiation). Quantification of the area of mitochondrial clusters demonstrated their increase until 72 h after irradiation (E); at least 100 cells/sample; n = 9−52; p < 0.05. (F and G) The lysosomal area was visualized by LAMP-2 staining in nonirradiated control cells and irradiated cells as shown in the representative pictures in (F) and related to the cell area (G). The relative lysosomal area of nonirradiated cells was set to 100%. An increase of the lysosomal area 72 h after irradiation compared with nonirradiated cells and irradiated cells at 0 h is discernable; n = 10−22; p < 0.05. (H and I) The amount of LAMP-2 protein was determined at the indicated time points by western blotting as shown by a representative blot (H) and normalized to actin protein levels. The quantification of the lysosomal protein showed a significant phasal upregulation after irradiation compared with nonirradiated controls (I); n = 3−4; p < 0.05.

Figure 2

Figure 3

Overexpression of ATG5, LC3B and ATG12 increased life span in two cell models. (A) Young HUVEC were stained with MTR and irradiated (irrad) or nonirradiated (control). At the indicated time points after irradiation, mRNA levels were quantified by semiquantitative RT-PCR and normalized to the nonirradiated control. PGC1α, ATG5, ATG12 and LC3B were upregulated 8 h and 72 h after irradiation; n = 3−6; p < 0.05. (B) MTR-stained HeLa cells were transfected with plasmids expressing GFP-LC3B, GFP-ATG5 and GFP as control. 48 h after transfection the MTR fluorescence was quantified. The MTR fluorescence of GFP-LC3B and GFP-ATG5 was significantly decreased compared with GFP-expressing cells; n = 3; p < 0.05. (C) HUVEC stably transfected with GFP, GFP-ATG5, GFP-LC3B, GFP-ATG12 and LAMP-1-GFP were stained with MTR and analyzed by CLSM. In LC3B-, ATG5-and to a lesser amount ATG12-overexpressing senescent cells, a cytosolic GFP fluorescence with fluorescent autophagosomes was discernible; in LAMP-1 transfected cells the lysosomes are clearly labeled. (D) Young HUVEC (PD13) that were either nontransfected or stably transfected with GFP, GFP-ATG5, GFP-LC3B GFP-ATG12 and LAMP-1-GFP were cultivated in triplicate until they reached replicative senescence. All autophagy protein-expressing cells exhibited a significantly extended life span and higher PD numbers compared with control cells (non-transfected and GFP-overexpressing cells). In contrast LAMP-1-overexpressing HUVEC reached senescence after a similar cultivation time as the control cells, indicating that enhanced expression of autophagy proteins prolonged the replicative life span of HUVEC; p < 0.05. (E) Young CEF in passage 8 were infected with virus coding either for GFP or GFP-gLC3B (see inset) and cultivated in triplicate until they reached replicative senescence. Overexpression of GFP-gLC3B resulted in a prolonged replicative life span and significantly increased PD number in comparison to GFP-transfected cells, in accordance with the results obtained for HUVEC presented in Figure 3D;p < 0.005. (F) Pre-senescent HUVEC (PD 22) were either nontransfected or stably transfected with plasmids expressing GFP, GFP-ATG5, GFP-LC3B GFP-ATG12 or LAMP-1-GFP. Cells were cultivated in triplicate until they reached replicative senescence. In contrast to young HUVEC (Fig. 3D), no expansion of the life span or the PD number became apparent, indicating the absence of a rejuvenating effect.

Figure 4

Overexpression of LC3B, ATG5, ATG12 and LAMP-1 mediates cell protective effects. (A and B) Young HUVEC expressing the indicated GFP fusion proteins (A) or young CEF overexpressing GFP and GFP-gLC3B (B) were treated with increasing doses of hydrogen peroxide for 10 min and the relative cell number of surviving adherent cells was determined and normalized to the untreated cells 24 h after treatment. Overexpression of the autophagy genes or LAMP-1 protected against hydrogen peroxide-induced cell death; (A) n = 4−6; p < 0.05; * = significant to GFP transfected cells, ** = significant to nontransfected cells; (B) n = 8; p < 0.05. (C and D) Young HUVEC stably overexpressing the indicated proteins (C) and young CEF overexpressing GFP and GFP-gLC3B (D) were stained with the mitochondrial membrane potential indicator TMRE and depicted by microscopy with constant settings. The TMRE fluorescence intensity was normalizedtothe fluorescence intensity of nontransfected HUVEC, and GFP transfected CEF, respectively. Overexpression of the autophagy genes and to a lesser amount LAMP-1 enhanced the mitochondrial membrane potential in both cell models; (C) n = 4−9; p < 0.05, (D) n = 7; p < 0.05. (E) Young HUVEC were stably transfected with GFP, GFP-ATG5, GFP-ATG12, GFP-LC3B and LAMP-1-GFP. Their ATP content was normalized to the GFP-transfected cells. Overexpression of the autophagy genes but not of LAMP-1 enhanced the ATP content; n = 3−4; p < 0.05. (F) A large fragment of the mtDNA (8.9 kb) was amplified by PCR and normalized on an amplified small fragment of the mtDNA (0.2 kb). The relative amplification of the 8.9-kb fragment acts as parameter for mtDNA integrity. The quantification in nontransfected and GFP-LC3B-overexpressing HUVEC at different time points of their growth curve shows an increased relative amplification of the 8.9-kb fragment after LC3B overexpression, indicating enhanced protection against mtDNA damage; 10–15 d (young cells): n = 3; 50–60 d (pre-senescent): n = 6; p < 0.05.

Figure 5

Overexpression of LC3B, ATG5 and ATG12 results in accumulation of oxidized material. (A and B) To determine the ROS content, young HUVEC stably overexpressing the indicated GFP-fusion proteins (A) or young CEF overexpressing GFP and GFP-gLC3B (B) were stained with DHE and the relative ethidium fluorescence was normalized to nontransfected HUVEC (A), or GFP transfected CEF (B). Overexpression of autophagy proteins or LAMP-1 did not alter the ROS levels in young HUVEC and CEF; n = 3−6. (C and D) The amount of oxidatively damaged proteins was analyzed by oxyblot in young HUVEC (C) and CEF (D) expressing the indicated GFP fusion proteins and normalized to actin (C) or total protein content (D). The amount of oxidized proteins of nontransfected cells or GFP-transfected (D) cells was set to 100%. A significant increase of carbonylated proteins was measured in cells overexpressing autophagy proteins in both cell types; (C) n = 4−7; p < 0.05, (D) n = 3; p < 0.05. (E and F) The amount of the active form of LC3, LC3-II, was determined by western blotting of nontransfected HUVEC and HUVEC overexpressing the indicated fusion proteins. A representative western blot is shown in (E). The quantification (F) shows after normalization to actin that overexpression of the autophagy proteins but not of LAMP-1 increases significantly the LC3-II/actin ratio compared with GFP and nontransfected cells; n = 6−8; p < 0.005. (G and H) Measurement of the lysosomal activity in nontransfected or overexpressing HUVEC (G), or CEF (H), respectively, was performed by quantifying the acid phosphatase activity. Only cells overexpressing LAMP-1-GFP demonstrated a significantly enhanced acid phosphatase activity; (G) n = 3−5; p < 0.05, (H) n = 5. (I) HUVEC were transiently transfected with either GFP and GFP-ATG5 or scrambled siRNA and a siRNA against ATG5. After 72 h the amount of oxidized proteins was determined by oxyblotting, normalized to actin protein expression and compared with the respective controls; transient overexpression of GFP-ATG5 increased significantly the amount of oxidized proteins compared with GFP-transfected cells; n = 3; p < 0.05.

Figure 5

References

1. Tatsuta T, Langer T. Quality control of mitochondria: protection against neurodegeneration and ageing. EMBO J. 2008;27:306–314. doi: 10.1038/sj.emboj.7601972. - DOI - PMC - PubMed
1. Bereiter-Hahn J, Jendrach M. Mitochondrial dynamics. Int Rev Cell Mol Biol. 2010;284:1–65. doi: 10.1016/S1937-6448(10)84001-8. - DOI - PubMed
1. Ono T, Isobe K, Nakada K, Hayashi JI. Human cells are protected from mitochondrial dysfunction by complementation of DNA products in fused mitochondria. Nat Genet. 2001;28:272–275. doi: 10.1038/90116. - DOI - PubMed
1. Nakada K, Inoue K, Ono T, Isobe K, Ogura A, Goto YI, et al. Intermitochondrial complementation: Mitochondria-specific system preventing mice from expression of disease phenotypes by mutant mtDNA. Nat Med. 2001;7:934–940. doi: 10.1038/90976. - DOI - PubMed
1. Sato A, Nakada K, Shitara H, Yonekawa H, Hayashi J. In vivo interaction between mitochondria carrying mtDNAs from different mouse species. Genetics. 2004;167:1855–1861. doi: 10.1534/genetics.103.021287. - DOI - PMC - PubMed

Autophagy proteins LC3B, ATG5 and ATG12 participate in quality control after mitochondrial damage and influence lifespan - PubMed (original) (raw)