Mitochondrial DNA — your second genome and why it matters for aging

The story starts about 1.5 to 2 billion years ago.

Lynn Margulis, the biologist whose endosymbiotic theory spent decades being dismissed before becoming textbook consensus, proposed that mitochondria are not native to eukaryotic cells. They are the descendants of ancient proteobacteria that were engulfed by a larger cell, established a mutually beneficial relationship, and eventually became incorporated as a permanent resident. The evidence is overwhelming: mitochondria divide by binary fission the way bacteria do, not by the processes cells use for organelle division. They have double membranes, consistent with being engulfed. They're sensitive to antibiotics that target bacterial ribosomes. And they carry their own genome — circular, as bacterial chromosomes are, not packaged in histones the way nuclear DNA is.

The human mitochondrial genome is small by comparison to the nuclear genome, which contains about 3 billion base pairs across 46 chromosomes. mtDNA is a single circular molecule of approximately 16,569 base pairs encoding just 37 genes. Of those 37: 13 encode proteins, all of which are subunits of the electron transport chain, the molecular machinery through which mitochondria generate most of the cell's ATP. The remaining 24 genes encode the ribosomal RNAs and transfer RNAs that mitochondria use to translate those 13 proteins — a self-contained translation system, a vestige of the bacterial ancestor. Everything else the mitochondrion needs, which is most of what it needs, is encoded in nuclear DNA and imported into the organelle.

This arrangement creates an interesting evolutionary constraint. The mitochondrial genome is, by its position inside the organelle, directly adjacent to the electron transport chain — which is the primary source of reactive oxygen species in the cell. Every cell that generates ATP through oxidative phosphorylation produces some free radical byproduct. The nuclear genome is relatively well-protected by histones, nuclear architecture, and robust repair systems. The mitochondrial genome has none of this. It lacks histone packaging. Its repair capacity is limited. And it sits next to the most oxidatively active machinery in the cell. The result is a mutation rate estimated at ten to twenty times higher than nuclear DNA.

Over a lifetime, mtDNA accumulates mutations. This is not controversial. What matters for aging is the specific pattern of that accumulation, and a concept called heteroplasmy.

Each cell contains not one mitochondrion but hundreds to thousands. Each mitochondrion contains multiple copies of mtDNA. This means a cell might carry 2,000 to 10,000 mtDNA molecules simultaneously. Heteroplasmy refers to the state where, within a single cell, some copies of mtDNA carry a mutation and others don't. The proportion matters enormously. A mitochondrial mutation present in 5 percent of a cell's mtDNA copies will have very different functional consequences than the same mutation present in 80 percent. Below certain threshold proportions — which vary by mutation type and cell type — cells can maintain near-normal function because the intact copies compensate. Above the threshold, function declines. The disease expresses.

This threshold effect creates a puzzling clinical pattern in primary mitochondrial diseases. Conditions caused by mtDNA mutations — MELAS (mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes), Leber's hereditary optic neuropathy, MERRF (myoclonic epilepsy with ragged red fibers), and others — can present in remarkably variable ways even within families carrying the same mutation. A mother may carry a pathogenic mtDNA variant at 40 percent heteroplasmy with no symptoms. Her child may inherit it at 70 percent heteroplasmy, above the expression threshold, and develop significant disease. The inheritance of mtDNA is exclusively maternal — sperm mitochondria are eliminated after fertilization — and the bottleneck events during oogenesis cause heteroplasmy levels to shift dramatically and unpredictably between generations.

For aging in people without primary mitochondrial disease, the mtDNA story is subtler but still significant. The mutation-accumulation hypothesis of aging proposes that the progressive accumulation of somatic mtDNA mutations over decades contributes to declining mitochondrial function — reduced electron transport chain efficiency, lower ATP output, increased reactive oxygen species production — which in turn impairs the high-energy tissues that depend on mitochondrial capacity: heart muscle, skeletal muscle, neurons, the kidney. The clonal expansion of specific mtDNA mutations in aging tissues — where a single mutant copy replicates selectively within a cell until it dominates the population — is observed in aged human tissues, particularly muscle and brain. Whether this clonal expansion is a driver of aging or a consequence of it is still being worked out, but the association is documented.

mtDNA copy number is an emerging biomarker worth understanding. It refers to the number of mtDNA molecules per cell, which varies considerably across tissue types and declines in certain conditions. Lower mtDNA copy number in blood has been associated with increased mortality risk, diabetes, cardiovascular disease, and cognitive decline in epidemiological studies. It's imperfect as a biomarker — the relationship between blood mtDNA copy number and mitochondrial function in other tissues is indirect — but it's measurable and tracked in some functional medicine panels as a proxy for mitochondrial health.

Then there's a discovery that has significantly expanded what researchers think mtDNA is actually doing. The mitochondrial genome encodes more than just the electron transport chain. Embedded within it are short open reading frames that encode small peptides — mitochondrial-derived peptides, or MDPs — that appear to have signaling functions far beyond the mitochondrion. Humanin, discovered first in 2001, is encoded within the 16S rRNA gene of mtDNA. It has shown cytoprotective effects in a range of cellular and animal models, including apparent protection against amyloid-beta toxicity relevant to Alzheimer's disease. MOTS-c, discovered in 2015 by Chang Ahn and Pinchas Cohen, is encoded within the 12S rRNA gene and acts as a metabolic regulator — influencing insulin sensitivity, exercise metabolism, and aging biology in animal models. SHLP1 through SHLP6 (small humanin-like peptides) are additional MDPs with various emerging functions. The realization that the mitochondrial genome serves as a source of signaling peptides in addition to structural proteins has fundamentally changed how researchers think about its role. The mtDNA is not just a vestigial bacterial legacy — it's an active participant in systemic biology.

The release of damaged mtDNA into the cytoplasm — outside the mitochondria and outside the nucleus, where it is not supposed to be — creates another kind of signaling, and this one is inflammatory in a way with direct implications for aging. When mitochondria are damaged, their membranes can become permeable, releasing fragments of mtDNA into the cytosol. The cytosol has no tolerance for this: double-stranded DNA outside the nucleus is a pattern associated with infection, recognized by the innate immune sensor cGAS (cyclic GMP-AMP synthase). cGAS detects cytoplasmic mtDNA and activates the STING pathway, triggering type I interferon production and a broader inflammatory program. This mtDNA leak, amplified across aging tissues where mitochondrial damage accumulates, is now recognized as one of the contributors to inflammaging — the chronic low-grade inflammation that characterizes aged biology and underlies many age-associated diseases.

The intervention space is emerging. SS-31, also known as Elamipretide, is a tetrapeptide developed by Hazel Szeto that concentrates in the inner mitochondrial membrane, stabilizes cardiolipin (a lipid critical for electron transport chain function and membrane integrity), and has shown protection against mitochondrial dysfunction in multiple preclinical models and some early human trials, including for heart failure and age-associated muscle decline. SS-31 is currently in clinical development and is not FDA-approved for general use. MOTS-c is being investigated as an injectable peptide for metabolic and aging applications; it's in early human trial phases and available primarily as a research compound. NAD+ precursors — nicotinamide riboside and nicotinamide mononucleotide — work partly through sirtuin activation, and sirtuins including SIRT1 and SIRT3 regulate mitochondrial biogenesis and quality control, the processes by which cells generate new healthy mitochondria and eliminate damaged ones through mitophagy. Urolithin A works through a different angle on mitophagy — the selective autophagy that clears damaged mitochondria — and has shown some evidence of improved mitochondrial function in older adults in published human trials.

What the mtDNA story teaches about aging is structural. It reveals a genome that is, compared to nuclear DNA, less protected, faster-mutating, more directly exposed to the consequences of the metabolic work it supports — and that the accumulation of damage in this second genome has downstream effects not just on energy production but on inflammatory signaling, cellular stress responses, and intercellular communication through peptide signals that researchers are still characterizing. The mitochondrion, it turns out, is not simply a power plant. It is a sensor, a signaling hub, and an aging accelerant when things go wrong.

The same organelle that gives you ATP is also, over decades, one of the mechanisms by which your cells begin to lose their capacity to maintain themselves. That's not a design flaw. It's the consequence of sustaining aerobic metabolism in a biological system over a very long time. Understanding it is a prerequisite for thinking clearly about the interventions meant to address it.