Mitochondrial-derived peptides in energy metabolism

Abstract

Mitochondrial-derived peptides (MDPs) are small bioactive peptides encoded by short open-reading frames (sORF) in mitochondrial DNA that do not necessarily have traditional hallmarks of protein-coding genes. To date, eight MDPs have been identified, all of which have been shown to have various cyto- or metaboloprotective properties. The 12S ribosomal RNA (MT-RNR1) gene harbors the sequence for MOTS-c, whereas the other seven MDPs [humanin and small humanin-like peptides (SHLP) 1–6] are encoded by the 16S ribosomal RNA gene. Here, we review the evidence that endogenous MDPs are sensitive to changes in metabolism, showing that metabolic conditions like obesity, diabetes, and aging are associated with lower circulating MDPs, whereas in humans muscle MDP expression is upregulated in response to stress that perturbs the mitochondria like exercise, some mtDNA mutation-associated diseases, and healthy aging, which potentially suggests a tissue-specific response aimed at restoring cellular or mitochondrial homeostasis. Consistent with this, treatment of rodents with humanin, MOTS-c, and SHLP2 can enhance insulin sensitivity and offer protection against a range of age-associated metabolic disorders. Furthermore, assessing how mtDNA variants alter the functions of MDPs is beginning to provide evidence that MDPs are metabolic signal transducers in humans. Taken together, MDPs appear to form an important aspect of a retrograde signaling network that communicates mitochondrial status with the wider cell and to distal tissues to modulate adaptative responses to metabolic stress. It remains to be fully determined whether the metaboloprotective properties of MDPs can be harnessed into therapies for metabolic disease.

INTRODUCTION

Traditionally, the identification of protein-coding open-reading frames (ORFs) within genomes has focused on sequences that are initiated with an AUG start codon, have conserved homologous amino acid sequences, and are >100 codons in length (2). However, there is growing evidence that short ORFs that lack these classical hallmarks of protein-coding genes can encode biologically active small peptides or micropeptides (<100 amino acids) and can be found in transcripts annotated for larger proteins or long noncoding RNAs (lncRNAs) (5). Although efforts are increasing to systematically identify functionally relevant nuclear short ORFs, several small regulatory peptides encoded by mitochondrial genome (mtDNA) short ORFs (collectively termed mitochondrial derived peptides) have been shown to have broad cellular cyto- and metaboloprotective properties (6, 13, 31).

The mitochondria’s involvement in maintaining cellular homeostasis extends beyond that of energy production to include a regulatory role in a range of processes, including immune/inflammatory responses, proteostasis, adaptive stress responses, and apoptosis (43). To achieve this, the mitochondria have developed extensive retrograde signaling networks to communicate with the nuclear genome, other intracellular organelles, and potentially neighboring cells or organs (43), of which mitochondrial-derived peptides (MDPs) appear to form a critical aspect. The first described MDP, humanin, is encoded within the 16S ribosomal RNA gene (MT-RNR2) (15, 30), and more recently, the MT-RNR2 gene has been shown to harbor sequences for several additional small-humanin-like peptides (SHLP1–6) (6). Despite the mtDNA containing tens to hundreds of potential peptide-encoding short ORFs, the only other MDP to have been described is the mitochondrial open-reading frame of the 12S rRNA-c (MOTS-c), a 16-amino acid peptide transcribed from the 12S ribosomal RNA (MT-RNR1) gene that appears to have potent metabolic modulation properties (31). In this review, we argue that MDPs are metabolically active peptides by summarizing the evidence that they are endogenously responsive to metabolic stress and can promote an adaptive response to metabolic stressors, focusing on metabolic disease, aging, and exercise (Fig. 1).

Fig. 1.

Summary of metabolic stressors that modulate mitochondrial-derived peptide (MDP) expression and in vivo metabolic effects of MDP treatment in rodents.

MITOCHONDRIAL-DERIVED PEPTIDES IN METABOLIC DISEASE

Mitochondrial dysfunction is a key player in the pathophysiology of metabolic diseases, including obesity, insulin resistance, type 2 diabetes, and nonalcoholic fatty liver disease (NAFLD) (3). As such, it is not surprising that there is a growing number of studies that have assessed the effect of administering exogenous MDPs (or their analogs) on glucose and lipid metabolism of rodents under metabolically challenging conditions (Table 1). Of the known native MDPs, MOTS-c has most consistently been reported to have metaboloprotective properties in multiple models of metabolic dysfunction (Table 1). Initial reports by Lee et al. (31) showed that the stable overexpression of MOTS-c in cultured cells promotes glucose clearance and lactate accumulation in an AMP-activated protein kinase (AMPK) and sirtuin (SIRT) 1-dependent manner. Both of these proteins are nutrient sensors that have the ability to control cellular substrate utilization and energy metabolism in response to metabolic perturbations, and consistent with this, daily administration of MOTS-c increased glucose tolerance and insulin sensitivity of aged and diet-induced obese mice (31). Whether the enhanced skeletal muscle insulin sensitivity in mice was the direct result of MOTS-c-induced muscle AMPK activation and/or increased energy expenditure, in part, by increased lipid oxidation is not clear (31); two studies suggest that MOTS-c may increase the thermogenic capacity of white and brown fat in an AMPK-dependent manner, leading to weight loss (35, 36).

Table 1.

Metabolic outcomes of exogenous MDP treatment in vivo

Reference(s)	Model	MDP	Dose	Response
Aging
Kim et al. (23)	Aged mice	HNG	5 mg·kg−1·day−1	↑Hippocampus Akt and ERK phosphorylation
Qin et al. (47)	Aged mice	HNG	4 mg·kg−1·2× wk−1	↓Reduced myocardial fibrosis
Yen et al. (56)	Aged mice	HNG	4 mg·kg−1·2× wk−1	↑Metabolic healthspan, ↔lifespan
Lee et al. (31)	Aged mice	MOTS-c	5 mg·kg−1·day−1	↑Insulin sensitivity
Reynolds et al. (50)	Aged mice	MOTS-c	15 mg·kg·3× wk	↑Lifespan, ↓aging markers
Exercise
Reynolds et al. (50)	Treadmill run (mice)	MOTS-c	5–15 mg·kg−1·day−1	↑Performance
Cardiovascular
Oh et al. (45), Zhang et al. (59)	apoE-deficient mice	HNGF6A	0.4 mg·kg−1·day−1	↑Aortic function, ↓atherogenesis
Wei et al. (52)	Rat; vitamin D3 + nicotine	MOTS-c	5 mg·kg−1·day−1	↓Vascular calcification
Metabolic
Han et al. (14)	APP/PS1 mice	HNG	50–100 μg·kg−1·day−1	↓IRS-1, ↑Akt phosphorylation in brain
Lu et al. (35)	Cold exposure (mice)	MOTS-c	5 mg·kg−1·day−1	↑Cold adaptation (browning WAT and BAT response)
Li et al. (32)	d-Galactose-treated mice	MOTS-c	10 mg·kg−1·day−1	↓Hepatic lipid accumulation
Gong et al. (12)	DIO mice	HNG	2 mg·kg−1·day−1	↓Fat mass and tissue lipid, ↑glucose homeostasis
Mehta et al. (38), Kim et al. (24)	DIO mice	HNG, SHLP2, MOTS-c	2.5 mg·kg−1·day−1	↓Metabolic disease metabolite signatures in blood
Lee et al. (31)	DIO mice	MOTS-c	0.5–5 mg·kg−1·day−1	↑Insulin sensitivity, ↓fat mass and tissue lipid
Kuliawat et al. (27)	Rat	HNGF6A	0.07 mg·kg−1·h−1	↑GSIS
Cobb et al. (6)	Rats and mice	SHLP2, SHLP3	icv 0.16 μg·kg−1·min−1, 2 mg·kg−1·BID−1	SHLP2 ↑insulin sensitivity, SHLP3 ↑IL-6 and MCP-1
Muzumdar et al. (44)	Rat: icv and iv	HNGF6A	20 μg icv, 0.05 mg·kg−1·h iv−1	↑Insulin sensitivity
Muzumdar et al. (44)	Diabetic rat	HNGF6A	100 ug	↓Blood glucose
Lu et al. (36)	Ovariectomy mice	MOTS-c	5 mg·kg−1·day−1	↓Fat mass, ↓lipid, ↑glucose homeostasis

APP/PS1, amyloid precursor protein/presenilin-1; BAT, brown adipose tissue; DIO, diet induced obesity; GSIS, glucose stimulated insulin secretion; HNG/HNGF6A, humanin analogs; icv, intracerebroventricular; iv, intravenous; MDP, mitochondrial-derived peptide WAT, white adipose tissue. ↑Increase; ↓decrease.

Systemic MOTS-c treatment has beneficial effects in multiple rodent models of metabolic stress, including attenuating ovariectomy-induced fat accumulation, insulin resistance (36), and bone loss (42), reducing d-galactose-induced peripheral lipid accumulation and mitochondrial dysfunction (32), and downregulating circulating metabolite profiles that are associated with type 2 diabetes and obesity (24). Consistent with these metaboloprotective effects of MOTS-c in rodents, circulating levels in humans have been reported to be reduced with obesity (9), insulin resistance (4), type 2 diabetes (48), chronic kidney disease (33), and endothelial function (Table 2) (47). Therefore, it will be interesting to determine whether restoring circulating MOTS-c levels in patients with metabolic dysfunction can improve clinical outcomes. Indeed, clinical trials on MOTS-c and a MOTS-c analog are underway, with indications for coronary artery disease in patients with type 2 diabetes (NCT04027712) and nonalcoholic hepatic steatosis and obesity (NCT03998514).

Table 2.

Endogenous MDP response to metabolic stressors in vivo

Reference(s)	Metabolic Stress	MDP	Tissue	Effect
Aging
Muzumdar et al. (44), Bachar et al. (1), D’Souza et al. (8)	Aging (human)	Humanin, MOTS-c	Blood	↓
Conte et al. (7)	Aging (human)	Humanin	Blood	↑
D’Souza et al. (8)	Aging (human)	MOTS-c	Muscle	↑
Muzumdar et al. (44), Cobb et al. (6), Lee et al. (31)	Aging (rodent)	Humanin, MOTS-c, SHLP2	Blood, muscle, hypothalamus	↓
Exercise
Woodhead et al. (54)	AEx (human)	Humanin	Muscle, blood	↑,↑
Woodhead et al. (54)	TRx (human)		Muscle, blood	↔, ↓
Woodhead et al. (54)	AEx, TRx (human)	SHLP2	Blood	↔
Woodhead et al. (54)	AEx, TRx (human)	SHLP6	Blood	↑, ↓
Reynolds et al. (50)	AEx (human)	MOTS-c	Muscle, blood	↑,↑
Gidlund et al. (11)	TRx (human)	Humanin	Muscle, blood	↑↔,↔
Ramanjenaya et al. (49)	TRx (human)	MOTS-c	Blood	↔
Cardiovascular
Widmer et al. (53), Zhloba et al. (60), Qin et al. (47)	Heart disease (human)	Humanin	Blood	↓
Mangkhang et al. (37)	Mitral valve disease (canine)	Humanin	Blood	↓
Metabolic disorders
Ramanjaneya et al. (48)	Type 2 diabetes (human)	Humanin, MOTS-c	Blood	↓
Cataldo et al. (4)	IR (human)	MOTS-c	Blood	↑
Cataldo et al. (4)	Obesity (human)	MOTS-c	Blood	↔
Du et al. (9)	Obesity (human)	MOTS-c	Blood	↓
Liu et al. (33)	Kidney disease (human)	MOTS-c, Humanin	Muscle, blood Muscle, blood Muscle	↓ ↓,↑ ↑#
Other
Lu et al. (35)	Cold exposure (rodent)	MOTS-c	Blood	↓
Ramanjaneya et al. (49)	Intralipid infusion (human)	MOTS-c	Blood	↑
Kariya et al. (19), Kin et al. (25)	mtDNA-related diseases (human)	Humanin	Muscle	↑

AEx, acute exercise bout; blood, serum or plasma; IR, insulin resistance; MDP, mitochondrial-derived peptide; TRx, exercise training. ↑, increase; ↓, decrease; ↔, no change;

^#Relative to mtDNA.

Humanin was first identified in a cDNA library screen derived from a surviving brain fraction of an individual with Alzheimer’s disease. Consistently, humanin has best been known for its neuroprotective effects, in part, by regulating pro-apoptotic pathways including Bax-related proteins (13) and insulin-like growth factor binding protein-3 (IGFBP-3) (17). Circulating levels of humanin have been shown to be reduced in several metabolic disorders (Table 2), including cardiovascular disease (53, 60) and diabetes (48). Interestingly, however, muscles of patients with the mitochondrial mutations that lead to MELAS (mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes) and CPEO (chronic progressive external ophthalmoplegia) have elevated humanin expression (19, 25), which could be a tissue-specific stress response aimed at restoring/repairing mitochondrial homeostasis, as humanin can enhance mitochondrial metabolism under metabolic stress (20, 51), and the expression of humanin in the muscle can increase in response to acute exercise stress (54).

Several humanin analogs have been manufactured with improved potency and stability, including HNG, which has a glycine-to-serine substitution at position 14, F6AHN with an alanine-to-phenylalanine substitution at position 6 (abrogates IGFBP-3 binding), and HNGF6A, which contains both substitutions. The infusion of humanin, its analogs, or SHLP2 centrally (intracerebroventricularly) enhances insulin sensitivity in rodents, as determined by euglycemic-hyperinsulinemic clamp studies (44). Similar effects were observed when F6AHN and HNGF6A were infused peripherally (44), whereas intraperitoneal injections attenuate high-fat diet-induced increases in fat mass and lipid accumulation, particularly in the liver (12). Humanin may partially act via the activation of hypothalamic STAT3 to promote the suppression of hepatic glucose production. However, this appears to occur in concert with improved peripheral tissue glucose uptake (44). Indeed, under cellular stress, humanin can activate the insulin-stimulated glucose transport pathway, including insulin receptor substrate 1 and Akt (14, 55).

The observation that a single dose of humanin can lower blood glucose in Zucker diabetic fatty (ZDF) rats by maintaining high insulin levels has led to the hypothesis that humanin may also be insulinotropic (44). Indeed, HNGF6A infusion can increase insulin levels during hyperglycaemic clamps in mice and promote glucose-stimulated insulin secretion of murine β-cells by enhancing the sensitivity of the β-cells to glucose (27). Furthermore, humanin treatment can delay the onset of type 1 diabetes in nonobese diabetic (NOD) mouse (16). Taken together, this suggests that the MDPs MOTS-c, humanin, and SHLP2 are metabolically active peptides that respond to metabolic stress (Table 2) and have the potential to modulate insulin sensitivity, secretion, and energy utilization pathways. The precise molecular networks that underlie these functions are currently active topics of investigation.

MITOCHONDRIAL-DERIVED PEPTIDES IN AGING

Aging is associated with a progressive loss of cellular homeostasis and resilience, increasing susceptibility to multiple chronic diseases, and is at least partially dependent on metabolism at multiple levels, as demonstrated by dietary (e.g., dietary restriction), genetic (e.g., insulin/insulin-like signaling), and pharmacological (e.g., rapamycin, metformin) interventions that extend a healthy lifespan (34). Mitochondrial-nuclear communication is considered key to cellular fitness and organismal healthspan (43), and although traditionally thought to be primarily mediated by nuclear-encoded proteins, transient molecules, and mitochondrial metabolites, mitochondrial-encoded factors are now also emerging as key players. Notably, under stress conditions, MOTS-c can translocate to the nucleus and regulate adaptive gene expression through interactions with stress-responsive transcription factors and chromatin binding (22). When delivered via intraperitoneal injection in rodents, MOTS-c and other MDPs can exert tissue specific effects, suggesting that they can also cross the extracellular space (Ref. 31 and Table 1).

The levels of MDPs have been reported to be age related (Table 2). In cross-sectional studies, plasma humanin levels were lower in aged mice (2 mo vs. 13 mo) and humans [45–65 vs. 65–80 vs. 81–110 yr (44); 39 vs. 60 yr (1)]. Furthermore, humanin and SHLP2 systemic and tissue levels were lower in older rodents compared with young (6, 44). In contrast, however, within a large cohort of 693 individuals of varying health status and age (21–113 yr), Conte et al. (7) reported a strong positive association between age and plasma humanin levels. These differences may be attributed to the larger sample size, greater age range, and differing individual characteristics; however, the latter cannot be assessed due to limited participant data being reported in the earlier studies (1, 44). Because humanin can extend metabolic healthspan (56), an increase in plasma humanin was interpreted as a hormetic response aimed at improving the ability of cells/tissues to cope with stress (7) and postulated to have beneficial or detrimental effects, depending on the levels of response (39). This is compatible with the finding that patients with chronic kidney disease have higher circulating humanin levels and lower skeletal muscle levels compared with healthy controls (33). As for MOTS-c, in cross-sectional studies, circulating levels were found to be reduced in aged mice (4 mo vs. 32 mo) (31) and in humans [18–30 vs. 45–55 vs. 70–81 yr (8); significant negative correlation (48)]. However, whereas MOTS-c levels in older skeletal muscle from mice are lower (4 mo vs. 32 mo) (31), levels are increased in aged men (18–30 vs. 45–55 vs. 70–81 yr) (8). This discrepancy may be because 32-mo-old mice from the National Institute on Aging (NIA) aged rodent colony represent an end-of-life morbid condition, whereas the human tissue donors were still active and healthy.

Humanin is heavily involved in the growth hormone (GH)/IGF-1 axis, one of the most prominent endocrine regulators of aging. GH-deficient Ames mice are long-lived and show higher circulating humanin levels, whereas short-lived GH-transgenic mice had lower humanin levels compared with their wild-type mice (29). Furthermore, intermittent MOTS-c treatment initiated later in life reversed age-dependent loss of physical capacity and improved aging metabolism in mice (50). Because aging is linked with a progressive decline in mitochondrial function and disruption of metabolic homeostasis (28), MDPs may play a key role in slowing down the rate of aging and delaying the onset of age-related diseases. Therefore, we suggest that an age-dependent decline in MDP expression and/or function may dampen mitochondrial communication and consequently reduce the cellular capacity to dynamically adapt to insults and the ever-shifting conditions.

MITOCHONDRIAL-DERIVED PEPTIDES IN EXERCISE

Regular exercise is a therapeutic and preventative measure for most metabolic diseases and increases the activity of the mitochondria to provide energy for the sustained contractile activity of muscle. Mitochondrial-derived peptides, particularly MOTS-c, activate similar signaling pathways to exercise and when administered exogenously promote exercise-like adaptations (22, 31), leading to speculation that MOTS-c may be an exercise mitokine. Investigations into exercise and MDPs have been relatively scarce. In a pre- versus posttraining study design, Gidlund et al. (11) reported that 12 wk of resistance training in middle-aged prediabetic men resulted in an increase in intramuscular humanin expression; however, similar responses were not seen following Nordic walking (11), 8 wk of aerobic training (plasma measures only) (49), or 2 wk of high-intensity interval training (54) in participants that ranged from young, healthy males (54) to middle-aged individuals with indications of metabolic disease (11, 49). Although it is difficult to discern the reasons for the inconsistent effect of exercise training on humanin levels, it is possible that muscle humanin is more responsive to resistance training (11).

More recently, MOTS-c and humanin have been observed to increase in muscle (11.9-fold) and plasma (1.5-fold) following acute high-intensity cycling exercise in healthy young men (50, 54), whereas plasma SHLP6 but not SHLP2 concentration also responds to exercise. In support of the hypothesis that skeletal muscle is a source of circulating MDPs during exercise, contraction of isolated mouse muscle rapidly (within 10 min) increases intramuscular humanin and MOTS-c expression (Ref. 54, and Woodhead JST and Merry TL, unpublished observations). Although it remains to be determined what intracellular signals regulate MDPs during exercise, the rapid increase in levels following the onset of exercise/contraction suggests a suppression of MDP degradation rather than an upregulation of transcription (54). Reactive oxygen species (ROS) increase during exercise and regulate adaptive responses to exercise training (39), and oxidative stress in cell culture promotes MOTS-c translocation to the nucleus and expression (22). The thiols in both humanin and MOTS-c sequences provide a potential mechanism through which oxidative stress may directly alter the stability of the peptides. However, the effects of oxidative and other metabolic stresses (metformin and serum/glucose restriction) on MOTS-c levels appear to be AMPK dependent (22). Because AMPK is a multifunctional and exercise-sensitive cellular energy sensor, it is possible that MDPs may be responsive to multiple signals associated with the change in energy status that occurs with exercise/contraction. Although the role and molecular targets of endogenously produced MDPs during exercise are yet to be identified, higher doses [15 mg·kg⁻¹·day⁻¹ as opposed to 5 mg·kg⁻¹·day⁻¹ used in obesity interventions (31)] of MOTS-c for as little as 2 wk can increase running capacity of both young and old mice (50). Mechanistically, MOTS-c treatment upregulated glycolytic and protein metabolism markers following exercise and led to an enrichment of genes associated with protein regulation/metabolism, cellular metabolism, and oxidative stress response, which are largely under the control of heat shock proteins (50). Therefore, it is tempting to speculate that MOTS-c may regulate adaptive responses to exercise-related stress conditions by forming an integral part of the mitochondrial retrograde signaling network activated during exercise (40). In support for MOTS-c potentially acting to facilitate nuclear genomic interactions that lead to enhanced metabolic flexibility (defined by the overall adaptive capacity to a shift in metabolic supply-demand equilibrium in response to a perturbation) (22, 50) as part of the exercise training response, a loss-of-function MOTS-c polymorphism (K14Q, MT-1382A>C) has been linked to increased type 2 diabetes susceptibility in those with low physical activity levels (58).

GENETIC VARIATIONS IN MITOCHONDRIAL DNA AND THEIR FUNCTIONAL IMPLICATION FOR MDPs

The human mitochondrial genome encodes 37 genes that are involved in oxidative phosphorylation, ribosomes, and translation within the mitochondria. As such, alteration of mtDNA copy number and mtDNA polymorphisms can affect mitochondrial function and cell metabolism to modify metabolic disease risk (58). Because mtDNA sequences are more varied by ethnicity compared with nuclear DNA sequences (41), studying mtDNA polymorphisms provides a unique insight into ethnicity-specific disease risk and functional significance of mtDNA regions. Indeed, mitochondrial genomic association studies are beginning to reveal the influence of mtDNA on metabolic disease, with mitochondrial variants MT-16320 being associated with blood glucose levels, MT-8706 and MT-8898 with waist/hip ratio (26), and MT-8414 and MT-16189 increasing the risk of type 2 diabetes in an ethnicity-specific manner (18, 46); however, the functional effects of the mtDNA polymorphisms remain relatively underexplored.

Sequence variation in mtDNA may result in differences in the function of classically recognized mtDNA-encoded proteins leading to alterations in mitochondrial respiration, reactive oxygen species (ROS) production, mitochondrial matrix pH, and intracellular calcium levels (18, 21). Equally, mtDNA polymorphisms could also impact the function of MDPs. Consistent with this, the MT-2706 variant in the humanin coding short ORF is associated with a decrease in circulating humanin levels and is associated with accelerated cognitive aging (57). More recently it has been recognized that 5–10% of people with East Asian ancestry have a variant (MT-1382) in the MOTS-c coding region that causes a K14Q amino acid replacement in the MOTS-c peptide (10). A meta-analysis of three independent Japanese cohorts (n = 27,527) demonstrated that the C allele of MT-1382 variant is associated with an increased risk of type 2 diabetes in males (when adjusted by for age and BMI), but not in females, and that this effect was dependent on physical activity levels (58). Importantly, treating high-fat-fed mice with K14Q MOTS-c does not confer the metabolic benefits associated with native MOTS-c administration (31, 58), suggesting that the MT-1382 variant results in inactive endogenous MOTS-c, which contributes to increased metabolic disease risk. Therefore, MDP treatment could potentially be a means of preventing metabolic dysfunction associated with this mtDNA variant that inactivates endogenous MDPs.

CONCLUSIONS AND FUTURE PERSPECTIVES

There is growing evidence that many MDPs are responsive to changes in metabolism in a tissue- and stress-specific manner and that exogenous humanin and MOTS-c protect against metabolic dysfunction associated with aging and energy imbalance through improving insulin sensitivity and activating energy-consuming pathways. However, the underlying mechanisms driving these responses are just beginning to be investigated, and there are still many open questions, including how MDP transcription is regulated and how transcripts from the mitochondria are exported and translated. Whereas CNTFR/GP130/WSX-1 and FPRL-1 are putative receptors for humanin (30), establishing whether other MDPs also have extracellular receptors or primarily exert their effects on energy metabolism by intracellular interactions (22) will increase our understanding of the biological significance of these peptides. The difficulty of editing the mitochondrial genome has slowed progress in this field; however, naturally occurring genetic variants within the mitochondrial genome (mitochondrial haplotypes) will likely be able to provide causative mechanistic insight into currently identified MDPs and perhaps guide which of the many additional short open-reading frames of mtDNA also encode biologically active MDPs. In addition, antagonists such as blocking antibodies to be developed against MDPs will further aid to define their endogenous role in health and disease. Understanding these aspects of MDP regulation will help elucidate how these new players in the mitochondrial retrograde signaling network modulate adaptative responses to metabolic stress and whether they can be harnessed to treat metabolic disease.

GRANTS

This work was funded by a Marsden Fast-start grant (T.L.M.). T.L.M. was supported by a Rutherford Discovery Fellowship. C.L. was supported by the NIA (R01AG052258), Ellison Medical Foundation (EMF), AFAR, and the Hanson-Thorell Family.

DISCLOSURES

C.L. is a consultant and shareholder of Cohbar, Inc. None of the other authors has any conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

T.L.M. and J.S.T.W. prepared figures; T.L.M., A.C., J.S.T.W., J.C.R., H.K., S.K., and C.L. drafted manuscript; T.L.M., A.C., J.S.T.W., J.C.R., H.K., S.K., and C.L. edited and revised manuscript; T.L.M., A.C., J.S.T.W., J.C.R., H.K., S.K., and C.L. approved final version of manuscript.