Mitochondrial peptides modulate mitochondrial function during cellular senescence

Mitochondrial biogenesis and energetics are altered during doxorubicin-induced senescence

To understand changes in mitochondrial biogenesis and energetics during cellular senescence, we investigated mitochondrial number, ATP production, and oxidative respiration in cellular senescence models. We used doxorubicin- and hydrogen peroxide-induced senescence of human primary adult dermal fibroblasts (HDFa) as DNA damage response models. The quiescent cells were used as a control for the mitochondrial biogenesis and energetics studies. Doxorubicin-induced senescent cells had higher mitochondria copy numbers compared to non-senescent cells (Fig. 1A). Immunofluorescence of mitochondria using mitochondrial outer membrane protein, mitochondrial import receptor subunit TOM20 antibodies showed that the area of TOM20 staining is more, and mitochondria appeared to be connected in the senescent cells (Fig. 1B). To determine whether the increased mitochondria mass increases energy production in senescence cells, we measured cellular ATP levels, and observed that senescent cells produced more ATP compared to non-senescent cells (Fig. 1C). To understand whether higher ATP production was due to mitochondrial respiration from more mitochondria or increased glycolysis, we measured the real-time oxygen consumption rate (OCR) and extracellular acidification rate (ECAR). Senescent cells showed higher basal OCR, spare respiratory capacity, and greater ATP production compared to non-senescent proliferating cells (Fig. 1D). We also examined a different cell line and model of senescence. Hydrogen peroxide-induced senescence in primary dermal fibroblasts (HDFa) also demonstrated higher basal OCR (Fig. S1). Moreover, doxorubicin-induced senescence in HCA2 (primary foreskin) fibroblasts demonstrated increased basal OCR: non-senescent cells consumed 37.77 ± 2.94 pmoles/min; senescent cells consumed 63.3 ± 8.22 pmoles/min. These results demonstrate that elevated mitochondrial respiration may occur in senescent cells. Glycolysis, glycolytic capacity and glycolytic reserve were not altered in senescent cells. However, there was a trend for ECAR to increase in senescent cells (Fig. 1E). Both spare respiratory capacity and glycolytic capacity seem to be increased in senescent cells, suggesting that senescent cells are more resistant to metabolic challenges. These results suggest that mitochondrial number and energy production play important roles in energy production to support the energy demands of senescent cells.

Mitochondria depend more on glucose and fatty acid utilization during doxorubicin-induced cellular senescence

Mitochondria use three different fuels for energy production: glucose, glutamine, and fatty acids. We therefore determined fuel utilization and mitochondrial respiration in senescent cells. Senescent cells showed a trend toward increased glucose and fatty acid utilization and decreased glutamine utilization (Fig. S2). Further, glucose uptake increased in senescent cells (Fig. 2A). These results suggest that the glucose utilized more in the mitochondrial respiration. In agreement with higher fatty acid utilization, the expression of carnitine palmitoyltransferase I (CPT1), which facilitates translocation of long chain fatty acids into mitochondria for beta oxidation, increased in senescent cells compared to non-senescent cells (Fig. 2B). We also examined protein levels of glutaminase (GLS1), which converts glutamine to glutamate for subsequent conversion to alpha-ketoglutarate and oxidation by the TCA cycle. GLS1 protein levels remained unaltered in senescent cells (Fig. 2C).

Glycolysis but not mitochondrial respiration increased during replicative senescence

We next measured OCR and ECAR to determine the cellular energetics of replicatively senescent primary human dermal fibroblasts. As opposed to doxorubicin-induced senescent cells, replicatively senescent cells showed levels of mitochondrial respiration similar to non-senescent cells, although basal respiration was slightly higher than in non-senescent cells (Fig. 3A). Consistent with other studies, replicatively senescent primary dermal fibroblasts showed increased glycolysis and, expectedly, dramatically increased glucose uptake (Figs. 3B and 3C). Senescent cells have high energy demand to increase the size of the cells and produce SASPs. Replicatively senescent cells have been shown to utilize glycolysis to supply the energy for those cellular processes [42,48,49]. On the other hand, doxorubicin-induced senescent cells show an increase in mitochondrial mass and utilized mitochondrial respiration. During DNA damage-induced senescence, mitochondria play important roles not only for respiration and ROS production, but also for DNA damage response signaling [50]. Thus, the expansion of the mitochondria and mitochondrial function are vital for doxorubicin-induced senescent phenotypes. The increased mitochondrial mass and enhanced mitochondrial function found in DNA damage signaling-mediated senescent such as doxorubicin, may not occur in replicatively senescent cells and, therefore, the enhanced glycolysis may be the main energy supply for replicatively senescent cells.

Several studies showed that doxorubicin per se alters mitochondrial respiration and glycolysis. Doxorubicin increased several glycolytic enzymes and increased glycolytic activity, and decreased TCA cycle and fatty acid oxidation in rat heart after 6 weekly injections of doxorubicin [51,52]. Doxorubicin increased the mRNA of TCA cycle relevant enzymes in rat perfused heart [53]. Although the effects of doxorubicin on TCA cycle are contradictory, the increased mitochondrial respiration in doxorubicin-induced senescent cells can be driven by the gene expression by doxorubicin as well as the increase of mitochondrial mass.

Mitochondrial DNA methylation regulates mitochondrial-encoded genes expression

Mammalian mitochondrial DNA is methylated at CpG sites, although the importance of this phenomenon is unknown (Fig. 4A). We recently reported that several mtDNA CpG sites were methylated during senescence [54]. Here, we examined the methylation status of CpG sites 1, 2 and 4 in doxorubicin-induced senescence. CpG1 showed a higher level of methylation, whereas CpG4 showed a lower level of methylation in senescent cells compared to non-senescent cells (Figs. 4B and 4C). The methylation level at the CpG2 site was not changed (not shown). Increased DNA methylation typically correlates with decreased gene expression. However, methylation of CpG4, which is present in the MT-CO1 gene, decreased, and COX1 expression increased in senescent cells (Fig. 4D). The increased level of COX1 is consistent with increased mitochondria mass and mitochondrial respiration in senescent cells.

Mitochondrial-derived peptides are altered during senescence

To determine whether MDP levels varied during senescence, we measured humanin, SHLP2 and SHLP6, which are derived from the 16S rRNA region, and MOTS-c, which is derived from the 12S rRNA region, by ELISA. The levels of humanin and MOTS-c increased (Fig. 5A), but SHLP2 and SHLP6 levels remained unchanged during doxorubicin-induced senescence (Figs. S3A-S3B). Humanin levels were not changed in replicative senescence, whereas MOTS-c levels slightly decreased (Fig. 5B). As these four MDPs were differentially regulated, we speculate that the increased mitochondria mass itself is not responsible for elevating humanin and MOTS-c levels in doxorubicin-induced senescence.

Humanin and MOTS-c regulate SASP production

Humanin and MOTS-c administration showed beneficial effects on age-related diseases in mice [10,55]. We hypothesized that humanin and MOTS-c can kill senescent cells and tested the potential role of these peptides as senolytics. To understand the effect of humanin and MOTS-c in doxorubicin-induced senescent cells, we examined cell death, mitochondrial respiration and SASP production following treatment with these peptides. We treated quiescent and doxorubicin-induced senescent cells with 1 and 10μM humanin and MOTS-c peptides for 3 days and examined the number of cells and cell death. Neither peptide altered the number of cells after treatment (data not shown). In addition, humanin did not induce caspase-3 dependent apoptosis in senescent cells (Fig. S4A), and MOTS-c treatment had no effect on caspase-3 dependent apoptosis in non-senescent and doxorubicin-induced senescent cells (data not shown). We used Annexin V and PI staining and MTT assay to examine the cell death following humanin treatment. We did not see the cell death by humanin in both doxorubicin-induced and replicatively senescent cells (Figs. S4B-4C).

Humanin and MOTS-c can modulate mitochondrial energetics. Since MOTS-c reduced mitochondrial respiration in HEK293 cells, we hypothesized that MOTS-c would decrease the elevated mitochondrial respiration in senescent cells and reduce the SASP production.

In non-senescent cells, humanin increases whereas MOTS-c decreases mitochondrial respiration. Unexpectedly, humanin and MOTS-c both increased mitochondrial respiration in senescent cells (Fig. 6A). Additionally, we showed that senescent cells depend more on glucose and fatty acid utilization to support the elevated mitochondrial respiration. Because humanin and MOTS-c treatment increased mitochondrial respiration in senescent cells, we examined whether these peptides increase fatty acid utilization in these cells. Specifically, we examined the level of CPT1 in the presence or absence of these peptides.

MOTS-c, but not humanin, increased CPT1 levels in senescent, but not non-senescent, cells (Fig. 6B). This finding suggests that MOTS-c increases fatty acid utilization in senescent cells to further increase mitochondrial respiration.

Next, we examined the levels of a small subset of SASP factors in non-senescent (quiescent) and senescent cells following humanin and MOTS-c treatment. Exposure to these peptides increased production of the SASP factors IL-6, IL-1β, IL-8, IL-10 and TNFα (Table 1). We also examined the effects of humanin and MOTS-c in replicatively senescent cells. Humanin and MOTS-c treatment increased the level of these SASP factors in replicatively senescent cells (Table 2). Because trifluoroacetic acid (TFA) is a common contaminant in purification of peptide, trace amounts of TFA may affect our assays. We examined whether TFA alters the SASP of senescent cells and found that 40μM TFA did not alter SASP production such as IL-6, IL-1β, IL-8, IL-10 and TNFα (data not shown), indicating that humanin and MOTS-c increase SASP factors independent of TFA. Humanin activates JAK/STAT pathways through gp130 receptor complexes [15]. We treated cells with humanin in the presence or absence of JAK inhibitor to understand whether JAK pathways are involved in SASP production by humanin. JAK inhibitor treatment diminished the IL-6 production induced by humanin (Table 3). However, IL-10 and TNF-α levels were not altered by the JAK inhibitor. These results suggest that other signaling pathways may be involved in SASP production, and deciphering these signaling pathways will be informative. Humanin and MOTS-c increased a subset of SASP factors in both non-senescent and senescent cells, which raises the question of whether humanin and MOTS-c can increase cytokine production in general. To understand whether these peptides specifically upregulate SASP, we measured cytokines which have not been demonstrated to be induced during cellular senescence including eotaxin, eotaxin-3, IFN-γ, and I-TAC, in doxorubicin-induced and replicatively senescent dermal fibroblast cells. The levels of eotaxin, IFN-γ, and I-TAC were too low to be detected in the non-senescent and senescent cells, and eotaxin-3 levels were not altered by humanin and MOTS-c (table 4). These results suggest that humanin and MOTS-c specifically increases a small subset of SASP factors during senescence.

Reagents and antibodies

HNG (a potent analogue of humanin with a glycine substitution, S14G) and MOTS-c were synthesized by Genscript (Piscataway, NJ, USA). HNG and MOTS-c were initially dissolved in Milli-Q water. Doxorubicin (Sigma, St. Louis, MO, USA) and hydrogen peroxide (Sigma) were used for inducing senescence status in culture. JAK inhibitor I (Millipore Sigma, Burlington, Massachusetts, USA) were used to inhibit JAK pathway. The following antibodies were used in this study: anti-GAPDH antibody (Cat. #5174S), anti-Lamin B1 antibody (Cat. #12586), Cleaved Caspase-3 (Asp175) antibody (Cat. #9661S), CPT1A antibody (Cat. #12252S), anti-rabbit IgG, HRP-linked antibody (Cat. #7074), and anti-mouse IgG, HRP-linked antibody (Cat. #7076). These antibodies are supplied by Cell Signaling Technology (Danvers, MA, USA). Anti-Tom20 antibody (Cat. #SC-17764; Santa Cruz Biotechnology, Dallas, Texas, USA), anti-glutaminase antibody (GLS; Cat. #701965; ThermoFisher Scientific, Wilmington, DE, USA), and anti-β-actin antibody (Cat. A5316; Sigma) were also used.

Cell culture and treatment

Primary dermal fibroblasts from normal human adults (HDFa) were purchased from ATCC (Cat#. PCS-201-012; Manassas, VA, USA) and cultured in Fibroblast Basal Medium supplemented with Fibroblast Growth Kit-Low serum (ATCC) at 37° C in 5% CO₂. HCA2 primary human fibroblasts from neonatal foreskin, from O. Pereira-Smith were cultured in in high glucose Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Omega Scientific, Tarzana, CA, USA) at 37° C in 5% CO₂ and 5% O₂. For replicative senescence, primary dermal fibroblast cells were passaged every 3 days until they reach 20-25 population doublings (PDs) and were compared to cells at 5-6 PDs. Primary dermal fibroblasts were also induced to senesce by treating with doxorubicin or hydrogen peroxide [58,59]. Briefly, cells were treated with 250 nM doxorubicin for 24 h; the medium was replaced with complete medium, and then again, every 3 days for 10-14 days. Cells were treated with 200 μM hydrogen peroxide for 2 h; the medium was replaced with complete medium, and then cells were re-treated with hydrogen peroxide every 3 days for 7-10 days. To obtain the quiescent status of non-senescent cells, cells were switched to Fibroblast Basal Medium supplemented with Fibroblast Growth Kit- serum-free (ATCC) for 24h. For peptide treatments in serum free media, cells were switched to Fibroblast Basal Medium supplemented with Fibroblast Growth Kit- serum-free (ATCC) for 2 h followed by addition of the peptide for 1-3 days.

Western blot analysis

Cells were lysed with RIPA Lysis and Extraction Buffer (ThermoFisher Scientific) plus the Halt protease & phosphatase inhibitor cocktail (ThermoFisher Scientific). The lysates were incubated on ice for 10 min, then disrupted using a sonicator, and the supernatants collected by centrifugation at 15,000 x g for 15 min at 4° C. The Pierce™ BCA Protein Assay Kit (ThermoFisher Scientific) was used to quantify protein content in the lysates. Proteins (10-40 μg) were separated on NuPAGE™ 4-12% Bis-Tris Protein Gels and blotted onto PVDF membranes (ThermoFisher Scientific). Membranes were incubated with primary antibody at 4° C overnight according to the manufacturer’s instructions. After several washes with Tris-buffered saline containing 0.1% Tween-20, membranes were incubated with the appropriate HRP-conjugated secondary antibody at room temperature (RT) for 1 h. Enhanced chemiluminescence was used to detect specific bands. Membranes were imaged on a Bio-Rad ChemiDoc XRS⁺ imager. Relative intensities of the bands were measured using Image J, a free program provided by National Institute of Health (Bethesda, Maryland, USA).

Immunocytochemistry

Primary dermal fibroblasts plated on coverslips were fixed with 4% paraformaldehyde for 10 min at RT. After fixation, cells were permeabilized with 0.2% Triton X-100 in phosphate-buffered saline (PBS) for 10 min at RT and blocked in PBS containing 0.2% Triton X-100 and 1% bovine serum albumin (BSA) for 1-h at RT. Cells were then incubated with mouse anti-Tom20 antibody (1:100; Santa Cruz Biotechnology) in PBS containing 0.2% Triton X-100 and 1% BSA at 4° C overnight. After three washes with PBS, cells were incubated with Alexa Fluor 488-conjugated donkey anti-mouse IgG (1:200; Invitrogen) and Alexa Fluor 568-conjugated donkey anti-rabbit IgG (1:200; Invitrogen) in PBS containing 0.2% Triton X-100 and 1% BSA for 1-h at RT in the dark. Nuclei were stained for 5 min at RT in PBS containing Hoechst 33258 (2mg/ml; Invitrogen). Coverslips were mounted with ProLong Gold antifade reagent (Invitrogen) and observed under an LSM780 confocal microscope (Carl Zeiss, Germany).

DNA extraction and mitochondrial DNA copy number measurement

Genomic DNA from 300,000 cells was extracted with DNeasy Blood & Tissue Kits (Qiagen, Valencia, CA, USA). The mitochondrial copy number was estimated by real-time PCR (CFX Connect Real-Time System, Biorad) using two mtDNA targets (ND1, CYB) and two nuclear DNA targets (β-actin, 36B4) (IDT, CA, USA). Real-time PCR was performed by using SsoAdvancedTM Universal SYBR® Green Supermix, following the protocol provided by the manufacturer. The ratio of mtDNA to nuclear DNA was calculated by averaging the differences in amplification efficiency between ND1 and β-actin (∆Ct) and between CYB and 36B4 (∆Ct). The primers used for amplification are: ND1 Forward primer (5’gaagtcaccctagccatcattc3’) and Reverse primer (5’gcaggagtaatcagaggtgttc3’); CYB Forward primer (5’ ctaggcgacccagacaattatac3’) and Reverse primer (5’ ttagggacggatcggagaat3’); β actin Forward primer (5’ actcttccagccttccttcc3’) and Reverse primer (5’ ggcaggacttagcttccaca3’); 36B4 Forward primer (5’ ggaatgtgggctttgtgttc3’) and Reverse primer (5’ cccaattgtccccttacctt3’).

Cellular ATP measurement

Intracellular ATP levels were determined using an ATP Colorimetric/Fluorometric Assay Kit (BioVision, Inc., Milpitas, CA, USA) and carried out according to the manufacturer’s protocol. Briefly, 1x 10⁶ cells were lysed in ATP Assay Buffer, then homogenized using Amicon 10kDa Ultra-0.5 mL Centrifugal Filters (Millipore Sigma, Billerica, MA, USA). Lysates were transferred to a 96- well plate, and incubated with Reaction Buffer for 30 min at RT. Absorbance from standard curve and samples was measured on a plate spectrophotometer (Molecular Designs, Sunnyvale, CA) at 579 nm.

Glucose uptake assay

Cellular glucose uptake was determined using a Glucose Uptake Cell-Based Assay Kit (Cayman chemical, Ann Arbor, MI, USA) according to the manufacturer’s protocol. The kit uses 2-NBDG, a fluorescently-labeled deoxyglucose analog, as a probe for detecting glucose taken up by cells. 2-NBDG was measured on a plate spectrophotometer (Molecular Designs) with fluorescent filters (excitation/emission = 485/535 nm).

Measurement of mitochondrial energetics (OCR, ECAR, Mito Fuel Usage)

Cellular bioenergetics were determined by measuring the oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of the cells with the XF-96 Flux Analyzer (Seahorse Biosciences, North Billerica, MA, USA) at the Gerontology Seahorse Core Facility, USC. Seahorse XF Cell Mito Stress Test, Glycolysis Stress Test, and Mito Fuel Flex Test kits were used to examine OCR, ECAR, and mitochondrial fuel usage, respectively, according to the manufacturer’s instructions. A seeding density of 20,000 cells per well was used. OCR, basal respiration, mitochondrial respiration associated with cellular ATP production, and spare respiration capacity were estimated by challenging cells with oligomycin, FCCP (carbonyl cyanide 4-[trifluoromethoxy]phenylhydrazone), rotenone, and antimycin A. Briefly, mitochondrial ATP production is correlated with a decrease in OCR following exposure to oligomycin and spare respiratory capacity was estimated from the difference between FCCP-stimulated OCR and basal OCR. Regarding ECAR, glycolysis, glycolytic capacity and glycolytic reserve were estimated by challenging cells with glucose, oligomycin, and 2-deoxy-glucose (2-DG). Briefly, glycolysis was determined by glucose-induced response, glycolytic capacity was determined by the shift following oligomycin treatment, and glycolytic reserve was defined by the difference between glycolytic capacity and glycolysis rate. Mitochondrial fuel usage was determined by the rate of oxidation of each fuel (glucose, glutamine, and long-chain fatty acid) by measuring OCR of cells in the presence or absence of fuel pathway inhibitors. UK5099, BPTES, and Etomoxir were used as inhibitors of glucose, glutamine, and long-chain fatty acids oxidation pathways, respectively. All readings were normalized to total DNA content.

Annexin V and PI staining

Muse Annexin V and Dead Cell Assay kit (Millipore Sigma) were used to stain cells for quantifying the live, early and late apoptosis, and cell death [60]. All procedures are done according to the manufacturer’s instructions. The Muse Cell Analyzer are used to quantify 5,000 cells.

MTT assay

The viability of cells after humanin treatment were determined using the tetrazolium dye MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay. All the treatments were done using 5,000 cells in 96 well plate. The purple insoluble formazan was read on a microplate reader at a wavelength of 570nm.

Measurement of mtDNA methylation

mtDNA methylation was measured as described [54]. Briefly, mtDNA-containing cellular DNAs were treated with sodium bisulfite using an EZ DNA Methylation Kit (Zymo, CA), following the protocol provided by the manufacturer. The bisulfite-treated DNA was amplified by PCR in a 6 𝜇l reaction mixture containing 2 𝜇l 3× Klen-TaqI mix, 2 𝜇l template DNA, and 2 𝜇l 2.5 𝜇M primers. After initial denaturing at 95∘ C for 5 min, the DNA was amplified for 38 cycles at 95∘ C for 20 s, 62∘ C for 20 s of annealing, and 72∘ C for 20 s of extension and finally with extension at 72∘ C for 2 min. The PCR primers used to amplify the bisulfite-treated mtDNA are: CpG1 forward primer (5’ACGGAATAAGTTATTTTAGGGATAAT3’) and reverse primer (5’ACGAACCTTTAATAACGACTACACCA3’); CpG2 forward primer (5’CGAATTAAACCAAAAAATTAATTAATAC3’) and reverse primer (5’AGGAGTTTAAATTTTTTTATTTTTAGGAT3’); CpG4 forward primer (5’AAACCTACAAATAATAAAATATTTCATA3’) and reverse primer (5’TTTATTTTTAGGTTATATTTTAGATTAA3’).

The status of mtDNA methylation was determined by restriction enzymes that distinguish methylated and unmethylated DNAs. Specifically, DNAs were digested with Taq I at 65∘ C for 2h or HpyCH4IV at 37∘ C for 2h and separated on 3% agarose gel. Taq I recognizes methylated TCGA sites and HpyCH4IV digests methylated ACGT sites. After treatment with sodium bisulfate, unmethylated cytosines were converted to uracils and were converted to TTGA and ATGT after PCR, which are not digested by Taq I and HpyCH4IV, respectively. The methylated and unmethylated bands were scanned for quantitation [54].

MDP measurements

The levels of MDPs including humanin, the small humanin-like peptides (SHLPs) and MOTS-c in conditioned medium were measured by an in-house sandwich ELISA as described [7,8,61]. Conditioned medium was generated by adding fibroblast Basal Medium supplemented with Fibroblast Growth Kit-Low serum (ATCC) for 24 h to cells. 100 μl of conditioned medium was used for the assay.

SASP measurements

Conditioned medium was generated as described above. IL-6, IL-1β, IL-8, IL-10, and TNFα, Eotaxin, Eotaxin-3, IFN-γ, and I-TAC levels in the conditioned medium were determined using the customized V-PLEX pro-inflammatory panel 1 (MESO SCALE DISCOVERY, Rockville, MD, USA) according to the manufacturer’s protocols. For normalization, the concentrations obtained with this assay were divided by the number of cells in each sample.

Statistical analysis

Data are presented as mean ± S.E.M. Significant differences were determined by Student’s t-tests, one-way ANOVA followed by Tukey’s post hoc test using GraphPad Prism 5 software. Values of *<0.05, **<0.01, ***<0.001 were considered statistically significant.