Mitochondrial-Derived Peptides Exacerbate Senescence

Abstract

Mitochondrial-derived peptides (MDPs), encoded by mitochondrial DNA, play a cytoprotective role by helping preserve mitochondrial function and cell viability under stressful conditions. Humanin and its homologs and MOTS-c are two of several MDPs hypothesized to have antiaging activity based on correlative studies. For example, humanin plasma levels are inversely correlated with growth hormone and insulin-like growth factor 1 expression, which may promote accelerated aging. Humanin has been shown to protect cells from beta amyloid toxicity and preserve endothelial cell function in a mouse model of atherosclerosis. Furthermore, both humanin and MOTS-c improve insulin sensitivity in mouse models of type 2 diabetes. Recently it was reported that a potent analogue of humanin blocks cardiac fibrosis in aging mice. Although it has been hypothesized that MDPs might have senolytic activity, in a recent report humanin and MOTS-c both exacerbate the senescence-associated-secretory-phenotype (SASP) in senescent cells by stimulating the secretion of IL-6, IL-1β, IL-8, IL-10 and tumor necrosis factor α. It appears that the cytoprotective activity of the MDPs may be permissive for increased expression of a set of proinflammatory cytokines. Given the potential benefits of MDPs in many of the same diseases associated with the presence of senescent cells, a combination of senolytic and MDP-based treatments may be additive or synergistic. The MDPs would protect normal cells, whereas senescent cells would be eliminated by the senolytic therapy. It is even possible that MDPs by increasing the SASP phenotype would make the senescent cells more apt to be cleared by the immune system or more sensitive to senolytics. In contrast, if the MDPs actually cytoprotect the senescent cells, then the treatment can be performed serially with the senolytic used first.

Introduction

Mitochondrial-derived peptides (MDPs) are proteins encoded by mitochondrial DNA that play a cytoprotective role by helping preserve mitochondrial function and cell viability under stressful conditions. The first characterized MDP was humanim,¹ a 21 or 24 amino acid peptide encoded by the mitochondrial 16S ribosomal RNA gene that can be translated in the mitochondria (21 aa) or the cytoplasm (24 aa).² Humanin and other MDPs are found both intracellularly and in the plasma and are thought to be released through a poorly understood secretory mechanism.³ Cell entry is even less well understood. Humanin was discovered in a screen for neurotrophic factors that inhibited apoptosis induced by a pathological beta amyloid mutant protein. Humanin and its 1000 × more potent analogue mutant humanin (S14G) known as HNG was found to block memory deficits associated with neurotoxic scopolamine,^1,4
–6 Aβ, and 3-quinuclidinyl benzilate. Interestingly, humanin/HNG reduced Aβ levels and memory deficits in the triple transgenic APPswe, taup310 L, PS-1 M146 V mouse model of Alzheimer's disease (AD).⁷ In double transgenic (APPswe, PS-1dE9) mice with accumulated plaques, HNG reduced the number of plaques over time and improved cognitive dysfunction.⁸ Humanin has been reported to have cytoprotective activity in stroke,⁹ and in an SOD1 familial ALS model when combined with activity-dependent neurotrophic factor in a hybrid peptide.¹⁰ Interestingly, humanin has been reported to be less capable of benefiting other neurodegenerative diseases such as prion disease and Huntington's disease in mouse models.^1,4,11

Humanin and various analogues have also been reported to confer protection in diabetes,¹² atherosclerosis,^8,13 and myocardial infarction and reperfusion injury.¹⁴ With respect to atherosclerosis, humanin maintains endothelial cell function and blocks progression of atherosclerotic plaque in the hypercholesterolemic ApoE mouse model. Humanin is protective against various cellular stresses including oxidative stress,¹⁵ hypoxia,¹⁶ and serum starvation,¹⁷ although the evidence for these indications is preliminary.

Humanin has multiple interactors, not all of which are likely known. For example, Humanin interacts with BAX and Bid to block apoptosis in the mitochondria. Humanin also interacts with insulin-like growth factor-binding protein 3 (IGFBP3),^18,19 to inhibit IGFBP3-induced apoptosis in cultured glial cells. Humanin directly interacts with the formyl-methionine-leucyl-phenylalanine (f-MLP) receptors FPRL1, FPRL2, and with gp130 that is part of a complex with WSX1, CHTF receptor to effect STAT signaling.^20
–22 Hashimoto et al. also found that humanin interaction with this complex is necessary to protect neurons from Aβ and is STAT dependent.²² Humanin and SHLP2, one of six other more recently described humanin-like peptides (SHLP1-6) also encoded by the 16S RNA gene,²³ exhibit chaperone activity that may partially block misfolding of islet amyloid polypeptide and thus play a role in type 2 diabetes (T2D).²⁴ Humanin is also known to stimulate chaperone-mediated autophagy (CMA), which helps maintain quality control by directing oxidized proteins to the lysosome.²⁵ CMA may be a critical component of the ability of humanin to confer cytoprotection because this activity is lost in CMA-deficient cells.

Humanin also has been reported to inhibit age-dependent cardiac fibrosis, a process believed to play a major role in age-related ventricular stiffness and heart failure. Eighteen-month-old female mice were injected with humanin analogue HNG twice weekly for 6 weeks. HNG treatment increased the ratio of cardiomyocytes to fibroblasts to levels similar to those of young 5-month-old mice. The high collagen levels associated with aging were also reduced to near normal levels by such treatment. As might be expected from its effects on other cell types, apoptosis of cardiomyocytes was inhibited as was fibroblast proliferation and expression of profibrotic factors such as TGF-β1, FGF-2, and MMP-2.²⁶ HNG upregulated AKT/GSK-3β that likely accounts for some of its antifibrotic activity in maintaining cardiomyocyte viability and reducing fibroblast-mediated repair linked to fibrosis.²⁶

Because of their involvement in energy production and free radical generation, mitochondria likely play a major role in aging and age-related diseases.^6
–8 In fact, improvement of mitochondrial function has been shown to ameliorate age-related memory loss in aged mice.⁹ Contrasting studies have shown that humanin levels both decrease and increase with age,^27,28 results that need to be clarified. However, either result suggests that humanin could play a role in aging and age-related diseases, such as AD, atherosclerosis, and diabetes. Along with lower humanin levels in the hypothalamus, skeletal muscle, and cortex of older rodents, the circulating levels of humanin were found to decline with age in both humans and mice.²⁷ Notably, circulating humanin levels were found to be (1) significantly higher in long-lived Ames dwarf mice but lower in short-lived growth hormone (GH) transgenic mice, (2) significantly higher in a GH-deficient cohort of patients with Laron syndrome, and (3) reduced in mice and humans treated with GH or IGF-1.²⁹

A second MDP was much more recently discovered: MOTS-c, which is a 16-amino acid peptide encoded by the mitochondrial 12S rRNA gene.³⁰ The primary target of MOTS-c is thought to be the skeletal muscle. MOTS-c inhibits the folate cycle and de novo purine biosynthesis, which results in adenosine monophosphate-activated protein kinase (AMPK) activation and the accumulation of exercise mimetic 5-aminoimidazole-4-carboxamide ribonucleotide, a known activator of AMPK. Treatment of mice with MOTS-c prevented both high-fat-diet-induced age-dependent insulin resistance and diet-mediated obesity.³⁰ MOTS-c stimulates glucose uptake and is reported to suppress oxidative respiration. This is consistent with the “Crabtree effect” whereby glucose can inhibit cellular respiration and oxidative phosphorylation.³¹ Notably, the glucose taken up in response to MOTS-c is sent to the pentose phosphate pathway, which provides intermediates for purine synthesis and is not metabolized through glycolysis.

MOTS-c has been identified as the first in a class of mitochondrial peptides that can regulate gene expression in the nucleus by interaction with transcription factors, resulting in retrograde signaling.³² Metabolic stress induced by glucose restriction, serum deprivation, and tert-butyl hydrogen peroxide-induced oxidative stress activates AMPK, which is required for translocation of MOTS-c to the nucleus. MOTS-c is likely carried by a protein that localizes to the nucleus under metabolic stress. In the nucleus, MOTS-c interacts with antioxidant responsive element-regulating transcription factors such as NRF1 and NRF2 to stimulate transcription of target genes involved in mitochondrial protection. Overexpression of MOTS-c significantly potentiates NRF2 signaling.³² Cytoprotection by MOTS-c is likely to result from this newly described mechanism, and there would be distinctive possibilities for collaboration between MOTS-c and humanin in protecting cells from stress by different but complementary mechanisms.

Both humanin and MOTS-c have been linked to longevity. It should be relatively obvious that factors that maintain cell function and prevent cell death are likely to be helpful in maintaining homeostasis with the possible caveat of increased cancer rates, which so far have not been correlated with MDPs.

The data supporting a role of humanin and MOTS-c in longevity are circumstantial. Levels of MOTS-c³⁰ and humanin²⁷ were initially reported to decline with age in mice and humans. Moreover, there is a negative correlation with GH/IGF-1 status. Mice with GH/IGF-1 defects such as long-lived GH-deficient Ames mice express low levels of humanin, and treatment of both mice and humans with GH or IGF-1 lowers circulating levels of humanin. Circulating humanin levels are negatively correlated with circulating IGF-1 levels in rodents, and positively correlated with longevity²⁹; humanin expression also declines with age in rodents and humans.²⁷ The question arises, does this make sense. Wouldn't one expect mito-protective peptides to increase with age, as more cells undergo stress? New data have been reported that directly contradict the initial reports. Conte et al. report that humanin and aging-related cytoprotective factors GDF15 and FGF-21 all are positively correlated with age when examining a human cohort of 693 subjects ranging in age from 21 to 113 years.²⁸ Moreover, high levels of these proteins, especially GDF15, are correlated with risk of mortality. Does this mean that these factors are detrimental? Probably not, rather increased cell stress associated with aging-related pathology induces higher levels of the protective factors. Indeed overexpression of GDF15 significantly extends lifespan in female mice.³³ The differences in the reported data could be due to using antibodies with different specificities and should be further investigated, but we suspect the data of Conte et al.²⁸ to be accurate. The mitochondrial m.1382A>C polymorphism, which results in a mutant form (Lys14Gln) of MOTS-c, is found in a haplogroup associated with exceptional longevity in the Japanese population.³⁴ Mechanistically, one way that MDPs might maintain cell vigor is exemplified by data showing that MOTS-c peptide increases NAD⁺ levels, which have been reported to decline with aging in a functionally important way.³⁵

MDPs Increase Senescence-Associated-Secretory-Phenotype in Senescent Fibroblasts

Because MDPs such as humanin and MOTS-c have potential beneficial activity in preventing and ameliorating age-related diseases that are coincidentally associated with cell senescence, and their expression levels decline with increasing age, Kim et al. hypothesized that MDPs may act to attenuate the senescence phenotype or even as function as senolytic agents themselves.³⁶ However, Kim et al. were surprised to find that expression of humanin and MOTS-c increases in DNA damage-induced senescent cells and exacerbates the senescence-associated-secretory-phenotype (SASP) phenotype in cell culture experiments. This immediately raises important questions about the interaction of MDPs and cell senescence in aging and tissue repair.

Kim et al. (2018) characterized the metabolic phenotypes of DNA damage-induced senescence by hydrogen peroxide and doxorubicin treatment of primary human adult dermal fibroblasts and replicative senescence by passaging to near the Hayflick limit. Doxorubicin significantly increased mitochondrial DNA copy numbers by real-time polymerase chain reaction, mitochondrial mass by immunostainng for mitochondrial outer membrane protein TOM20, Cox1 production by Western analysis, adenosine triphosphate (ATP) production by an ATP quantitation kit, and oxygen consumption rate (OCR) by a flux analyzer. Similar increases in OCR were seen for hydrogen peroxide treatment as well. Although the rate of glycolysis was unchanged, doxorubicin-treated senescent cells used increased glucose and fatty acids and decreased glutamine for energy production. Consistent with the increased fatty acid utilization is an increase in expression levels of carnitine palmitoyltransferase I (CPT1A), which catalyzes the rate limiting step in the beta-oxidation of long fatty acids.³⁶

By contrast, Kim et al. found that glycolysis but not OCR increased when fibroblasts were passaged to replicative senescence with significantly increased glucose uptake. Given that DNA damage at telomeres has been hypothesized to play a role in replicative senescence, it is surprising that overt DNA damage associated with hydrogen peroxide and doxorubicin results in such a fundamental metabolic difference with replicative senescence. Interestingly, doxorubicin does induce increased glycolysis in rat hearts after six weekly inductions, suggesting that its effects on senescent fibroblasts are either cell-type-specific or senescence-specific.³⁶ It would have been informative to include oncogene-induced and radiation-induced senescence in these experiments to better understand the subtle differences in senescent cell states.

Kim et al. observed that doxorubicin-induced senescence increased levels of humanin and MOTS-c peptides as measured by ELISA, but that expression levels of MDPs SHLP2 and SHLP6 were unaltered. Since SHLP2 and SHLP6 are expressed from the same mitochondrial 16S RNA gene as humanin, the authors suggest that increased expression of humanin and MOTS-c is due to differential regulation and not just due to increased number of mitochondria,³⁶ but we believe that an important role for increased mtDNA copy number cannot be ruled out for humanin or MOTS-c. By contrast in replicative senescence, in which the number of mitochondria are unaltered, humanin levels were unchanged and MOTS-c is only slightly decreased.

Kim et al. hypothesized that humanin and MOTS-c might be senolytic given their benefit to age-associated diseases demonstrated in previous studies. They compared quiescent fibroblasts with doxorubicin-induced cells. Number of cells was not altered, there was no change in apoptosis as assessed by caspase 3 activity, and annexin V staining and cell viability were also unchanged as assessed by uptake of the dye propidium iodide and MTT, assays that measure the activity of NAD(P)H-dependent cellular oxidoreductase enzymes, which are lost upon cell death.³⁶

In quiescent cells, humanin increased, but MOTS-c decreased mitochondrial respiration. MOTS-c is known to reduce mitochondrial respiration in other cell types.³⁰ However, both humanin and MOTS-c increased mitochondrial respiration in doxorubicin-induced senescent cells. Interestingly MOTC-c but not humanin increased CPT1A expression, suggesting that MOTS-c increased mitochondrial respiration by increasing the rate of fatty acid catabolism.³⁶

Interrogating a limited panel of proinflammatory cytokines, Kim et al. observed that DNA damage-induced senescence increased levels of all cytokines studied: IL-6, IL-1β, IL-8, IL-10 and tumor necrosis factor α (TNFα), whereas replicative senescence resulted in a somewhat more modest increase in IL-1β, IL-8, IL-10, and TNFα.³⁶ It does not appear possible to compare cytokine levels directly between the replicative senescence and doxorubicin-induced senescence data, most likely due to differences in number of cells and culture conditions. Thus, the effects of humanin and MOTS-c on proinflammatory cytokines appear inconsistent for normal quiescent cells as described in tables 1 and 2 in Kim et al.³⁶ Humanin binds to gp130 receptor complexes to activate JAK/STAT signaling pathways, A JAK inhibitor attenuated humanin-induced IL-6 expression, but not IL-10 and TNFα, suggesting that JAK/STAT is only one of several signaling pathways involved. Hence, more work remains to be done to characterize the relevant biochemical pathways. Interestingly, Kim et al. report that treatment with neither humanin nor MOTC-c induces senescence as assayed by beta-galactosidase induction.³⁶

These data, although very preliminary, [i.e., in need of (1) replication, (2) confirmation in vivo, and (3) more careful controls for the induction of quiescence, e.g., the use of serum-free synthetic culture medium, making comparison with senescent cells in normal medium difficult], are nonetheless potentially very significant. It appears that senescent cells induced by significant DNA damage express increased levels of MDPs, which, in turn, increase at least a subset of the SASP. How does this make sense biologically? Perhaps those MDPs that can stimulate SASP, and it is quite possible that humanin and MOTS-c are not the only ones, play a constructive role in tissue repair by maintaining the health of the tissue surrounding the wound or damaged area. These factors may then cooperate with senescent cells, which are known to play a key role in wound healing and regeneration.³⁷ Senescent cells stimulate inflammatory and fibrotic processes needed for successful regeneration. A testable hypothesis is that MDP stimulation of SASP may speed up tissue repair and wound healing.

Medical Implications

The key medical implications of this work lie at the intersection of the therapeutic potential of MDPs and antisenescent cell therapeutics. MDPs and their engineered analogues by themselves appear to have therapeutic potential in T2D, AD, cardiac fibrosis, and cardiovascular disease among others. Senolytics and drugs that attenuate SASP have potential in atherosclerosis, fibrotic diseases such as idiopathic pulmonary fibrosis, osteoarthritis, and cardiac fibrosis, and possibly many other diseases. However, these two classes of potential therapeutics appear to work by decidedly different mechanisms that may interfere with each other. However, the combined effect of targeting both senescent cells with senolytics and normal cells with MDPs may be possible.

But first it would be useful to understand something more about the basic biology, Would increasing the SASP phenotype with MDPs or other agents actually make senescent cells more likely to be recognized and cleared by the immune system? Or would MDPs given together with senolytics block the action of senolytics because they prevent cytotoxicity in normal cells? Or both?

The most sensible combination therapeutic regimen might be to administer a senolytic drug first to eliminate senescent cells and follow with MDPs or other therapeutics that are potentially beneficial and protect normal cells. In that way, any problems with the potentiation of senescent cells by MDPs could be avoided.

One interesting observation is that MDPs target many of the same cell types: endothelial cells, cardiomyocytes, and neurons as other putative beneficial factors such as GDF15³⁸ or even GDF11.³⁹ Might there be a biological interaction in which one induces the other, or do they involve separate pathways that could synergize?

Since damage or cell stress stimulates MDP expression, agents that result in mitohormesis may stimulate MDP expression. However, basic research is necessary to investigate this possibility.

Given the relatively strong data from preclinical animal models supporting development of MDPs, it is not surprising that there has been at least one company, Cohbar, Inc., that seeks to create age-related disease therapies based on MDPs (www.cohbar.com/science/pipeline). Their most advanced preclinical drug is based on an analogue of MOTS-c for nonalcoholic steatohepatitis and obesity.

Conclusion

MDPs, encoded by mitochondrial DNA, play a cytoprotective role by helping preserve mitochondrial function and cell viability under stressful conditions and are potentially drug candidates for a variety of aging-related diseases. Given that the set of diseases in which MDPs may provide benefit intersect the set of diseases associated with the presence of senescent cells, a combination of senolytic and MDP-based treatments may prove additive or synergistic. Serial treatment with a senolytic followed by an MDP may prove a fruitful avenue for future study and possible development.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Hashimoto

, Niikura

, Tajima

, et al. A rescue factor abolishing neuronal cell death by a wide spectrum of familial Alzheimer's disease genes and Aβ. Proc Natl Acad Sci U S A, 2001; 98:6336–6341.

Yen

, Lee

, Mehta

, Cohen

. The emerging role of the mitochondrial-derived peptide humanin in stress resistance. J Mol Endocrinol, 2013; 50:R11–R19.

Yamagishi

, Hashimoto

, Niikura

, Nishimoto

. Identification of essential amino acids in Humanin, a neuroprotective factor against Alzheimer's disease-relevant insults. Peptides, 2003; 24:585–595.

Hashimoto

, Niikura

, Ito

, et al. Detailed characterization of neuroprotection by a rescue factor humanin against various Alzheimer's disease-relevant insults. J Neurosci, 2001; 21:9235–9245.

Krejcova

, Patocka

, Slaninova

. Effect of humanin analogues on experimentally induced impairment of spatial memory in rats. J Pept Sci, 2004; 10:636–639.

Tajima

, Kawasumi

, Chiba

, et al. A humanin derivative, S14G-HN, prevents amyloid-β-induced memory impairment in mice. J Neurosci Res, 2005; 79:714–723.

Niikura

, Sidahmed

, Hirata-Fukae

, Aisen

, Matsuoka

. A humanin derivative reduces amyloid beta accumulation and ameliorates memory deficit in triple transgenic mice. PLoS One, 2011; 6:e16259.

Zhang

, Zhang

, Li

, et al. S14G-humanin improves cognitive deficits and reduces amyloid pathology in the middle-aged APPswe/PS1dE9 mice. Pharmacol Biochem Behav, 2012; 100:361–369.

, Chua

, Gao

, Hamdy

, Chua

BHL

. Humanin is a novel neuroprotective agent against stroke. Stroke, 2006; 37:2613–2619.

10.

Chiba

, Yamada

, Sasabe

, et al. Colivelin prolongs survival of an ALS model mouse. Biochem Biophys Res Commun, 2006; 343:793–798.

11.

Friedlander

RM.

Apoptosis and caspases in neurodegenerative diseases. N Engl J Med, 2003; 348:1365–1375.

12.

Hoang

, Park

, Cobb

, et al. The neurosurvival factor Humanin inhibits beta cell apoptosis via Stat3 activation and delays and ameliorates diabetes in NOD mice. Metabolism, 2010; 59:343–349.

13.

, Bachar

, Zacharias

, et al. Humanin preserves endothelial function and prevents atherosclerotic plaque progression in hypercholesterolemic ApoE deficient mice. Atherosclerosis, 2011; 219:65–73.

14.

Muzumdar

, Huffman

, Calvert

, et al. Acute humanin therapy attenuates myocardial ischemia and reperfusion injury in mice. Arterioscler Thromb Vasc Biol, 2010; 30:1940–1948.

15.

Bachar

, Scheffer

, Schroeder

, et al. Humanin is expressed in human vascular walls and has a cytoprotective effect against oxidized LDL-induced oxidative stress. Cardiovasc Res, 2010; 88:360–366.

16.

Men

, Zhang

, Yang

, Gao

. An AD-related neuroprotector rescues transformed rat retinal ganglion cells from CoCl₂-induced apoptosis. J Mol Neurosci MN, 2012; 47:144–149.

17.

Kariya

, Takahashi

, Hirano

, Ueno

. Humanin improves impaired metabolic activity and prolongs survival of serum-deprived human lymphocytes. Mol Cell Biochem, 2003; 254:83–89.

18.

Guo

, Zhai

, Cabezas

, et al. Humanin peptide suppresses apoptosis by interfering with Bax activation. Nature, 2003; 423:456–461.

19.

Ikonen

, Liu

, Hashimoto

, et al. Interaction between the Alzheimer's survival peptide humanin and insulin-like growth factor-binding protein 3 regulates cell survival and apoptosis. Proc Natl Acad Sci U S A, 2003; 100:13042–13047.

20.

Ying

, Iribarren

, Zhou

, et al. Humanin, a newly identified neuroprotective factor, uses the G protein-coupled formylpeptide receptor-like-1 as a functional receptor. J Immunol, 2004; 172:7078–7085.

21.

Harada

, Habata

, Hosoya

, et al. N-Formylated humanin activates both formyl peptide receptor-like 1 and 2. Biochem Biophys Res Commun, 2004; 324:255–261.

22.

Hashimoto

, Kurita

, Aiso

, Nishimoto

, Matsuoka

. Humanin inhibits neuronal cell death by interacting with a cytokine receptor complex or complexes involving CNTF receptor α/WSX-1/gp130. Mol Biol Cell, 2009; 20:2864–2873.

23.

Cobb

, Lee

, Xiao

, et al. Naturally occurring mitochondrial-derived peptides are age-dependent regulators of apoptosis, insulin sensitivity, and inflammatory markers. Aging, 2016; 8:796–808.

24.

Okada

, Teranishi

, Lobo

, et al. The mitochondrial-derived peptides, humaninS14G and small humanin-like peptide 2, exhibit chaperone-like activity. Sci Rep, 2017; 7:7802.

25.

Gong

, Tasset

. Humanin enhances the cellular response to stress by activation of chaperone-mediated autophagy. Oncotarget, 2018; 9:10832–10833.

26.

Qin

, Mehta

, Yen

, et al. Chronic treatment with the mitochondrial peptide humanin prevents age-related myocardial fibrosis in mice. Am J Physiol Heart Circ Physiol, 2018 [Epub ahead of print]; DOI:10.1152/ajpheart.00685.2017.

27.

Muzumdar

, Huffman

, Atzmon

, et al. Humanin: A novel central regulator of peripheral insulin action. PLoS One, 2009; 4:e6334.

28.

Conte

, Ostan

, Fabbri

, et al. Human aging and longevity are characterized by high levels of mitokines. J Gerontol A Biol Sci Med Sci, 2018 [Epub ahead of print]; DOI:10.1093/gerona/gly153.

29.

Lee

, Wan

, Miyazaki

, et al. IGF-I regulates the age-dependent signaling peptide humanin. Aging Cell, 2014; 13:958–961.

30.

Lee

, Zeng

, Drew

, et al. The Mitochondrial-derived peptide MOTS-c promotes metabolic homeostasis and reduces obesity and insulin resistance. Cell Metab, 2015; 21:443–454.

31.

Ibsen

KH.

The Crabtree effect: A review. Cancer Res, 1961; 21:829–841.

32.

Kim

, Son

, Benayoun

, Lee

The mitochondrial-encoded peptide MOTS-c translocates to the nucleus to regulate nuclear gene expression in response to metabolic stress. Cell Metab, 2018 [Epub ahead of print]; DOI:10.1016/j.cmet.2018.06.008.

33.

Wang

, Chrysovergis

, Kosak

, et al. hNAG-1 increases lifespan by regulating energy metabolism and insulin/IGF-1/mTOR signaling. Aging, 2014; 6:690–704.

34.

Fuku

, Pareja-Galeano

, Zempo

, et al. The mitochondrial-derived peptide MOTS-c: A player in exceptional longevity?. Aging Cell, 2015; 14:921–923.

35.

Mendelsohn

, Larrick

. The NAD+/PARP1/SIRT1 axis in aging. Rejuvenation Res, 2017; 20:244–247.

36.

Kim

S-J

, Mehta

, Wan

, et al. Mitochondrial peptides modulate mitochondrial function during cellular senescence. Aging, 2018; 10:1239–1256.

37.

Yun

MH.

Cellular senescence in tissue repair: Every cloud has a silver lining. Int J Dev Biol, 2018; 62:591–604.

38.

Fujita

, Taniguchi

, Shinkai

, Tanaka

, Ito

. Secreted growth differentiation factor 15 as a potential biomarker for mitochondrial dysfunctions in aging and age-related disorders. Geriatr Gerontol Int, 2016; 16 Suppl 1:17–29.

39.

Fan

, Gaur

, Sun

, Yang

. The growth differentiation factor 11 (GDF11) and myostatin (MSTN) in tissue specific aging. Mech Ageing Dev, 2017; 164:108–112.