Mitochondria: Structure and Function

A mitochondrion is a double-membrane-bound organelle, typically 0.5–1 μm in diameter and 1–10 μm long, though its shape is dynamic — mitochondria constantly fuse and divide, sometimes forming long branching networks rather than isolated "bean" shapes. The outer membrane is smooth and relatively permeable, studded with porin channels (VDAC, voltage-dependent anion channels) that allow ions and small molecules up to about 5,000 daltons to pass freely. The inner membrane is dramatically different: it is far less permeable, rich in a unique phospholipid called cardiolipin, and folded into invaginations called cristae that vastly increase its surface area. Between the two membranes lies a narrow intermembrane space, chemically similar to the cytosol in small-molecule composition; enclosed by the inner membrane is the matrix, a dense, protein-rich compartment containing the enzymes of the citric acid cycle, mitochondrial DNA, ribosomes, and tRNAs. This four-compartment layout (outer membrane, intermembrane space, inner membrane, matrix) is the physical basis of everything mitochondria do. The matrix is where pyruvate and fatty acids are oxidized and where the citric acid cycle strips electrons from carbon fuels, loading them onto the carriers NADH and FADH₂. The inner membrane is where those electrons are actually used: a series of protein complexes accepts the electrons, passes them downhill in energy, and uses the released energy to pump protons out of the matrix. The cristae are where this pumping is concentrated — their surface area in a single liver mitochondrion has been estimated to hold roughly five times more inner-membrane area than a smooth sphere of the same size would allow, packing in more electron transport chains and more ATP synthase motors per organelle. The key conceptual move, due to biochemist Peter Mitchell, is that oxidative phosphorylation is not one chemical reaction handing energy to another. It is two separate machines linked by a gradient. One machine (the electron transport chain) spends the energy of electron transfer to pump protons uphill, building an electrochemical potential across the inner membrane called the proton-motive force. The other machine (ATP synthase) lets protons run back downhill through it, and captures that flow mechanically, using it to force ADP and inorganic phosphate together into ATP. This is chemiosmotic coupling, and it is why an intact, sealed inner membrane is not incidental to mitochondrial function — it is the entire point.

Step 1 — Fuel delivery to the matrix. Pyruvate (from cytosolic glycolysis) and fatty acids are transported across both mitochondrial membranes into the matrix. Pyruvate is converted to acetyl-CoA by pyruvate dehydrogenase, releasing CO₂. Fatty acids are broken down by β-oxidation into acetyl-CoA units directly in the matrix.

Step 2 — The citric acid cycle. Acetyl-CoA enters the citric acid cycle (Krebs cycle), a ring of eight enzymatic steps in the matrix. Each turn of the cycle oxidizes one acetyl group fully to two CO₂ molecules, while transferring high-energy electrons onto carrier molecules:

• 3 NAD⁺ → 3 NADH
• 1 FAD → 1 FADH₂
• 1 GDP (or ADP) → 1 GTP (or ATP), a substrate-level phosphorylation

Step 3 — Electron transport chain (inner membrane, embedded in cristae). NADH and FADH₂ deliver their electrons to a sequence of four large protein complexes (Complex I, II, III, IV) plus two small mobile carriers (ubiquinone/coenzyme Q and cytochrome c):

Electrons flow from carrier to carrier because each successive complex has a higher affinity for electrons (a more positive reduction potential), releasing free energy at each transfer. At Complex IV, electrons finally reduce molecular oxygen to water — this is why mitochondria consume the O₂ we breathe and why oxidative phosphorylation fails catastrophically without oxygen. The energy released at Complexes I, III, and IV is used to actively pump H⁺ ions from the matrix into the intermembrane space, against their concentration gradient.

Step 4 — Building the proton-motive force. Because the inner membrane is impermeable to protons, this pumping creates two things simultaneously: a pH gradient (matrix becomes more alkaline, roughly pH 8 versus pH 7 in the intermembrane space) and a voltage across the membrane (matrix side negative, typically around −150 to −180 mV). Together these constitute the proton-motive force, on the order of 200 mV — the actual stored form of the energy extracted from food, held as electrochemical potential rather than chemical bond energy.

Step 5 — ATP synthase and rotary catalysis. ATP synthase is a two-part rotary motor embedded in the inner membrane. Its Fo portion forms a proton channel through the membrane; its F1 portion, projecting into the matrix, contains the catalytic sites. Protons flowing back down their gradient through Fo cause a ring of c-subunits to spin; this rotation is mechanically transmitted to a central stalk (the γ subunit) that turns inside the F1 head. As the γ subunit rotates, it forces conformational changes in three catalytic β subunits in sequence — an "open," "loose," and "tight" state — driving the binding of ADP and phosphate, their condensation into ATP, and finally the release of ATP. This is Paul Boyer's binding-change mechanism, confirmed by the crystal structure work of John Walker. Roughly three protons pass through the motor per ATP synthesized, and about 10 protons for a full 360° rotation producing 3 ATP.

Step 6 — Export. Newly made ATP is exchanged for cytosolic ADP by a dedicated inner-membrane antiporter (the ATP/ADP translocase), so ATP exits to power the rest of the cell while ADP re-enters for another round.

Before Mitchell, the dominant model of oxidative phosphorylation was "chemical coupling": biochemists searched for a high-energy chemical intermediate — analogous to the 1,3-bisphosphoglycerate intermediate in glycolytic substrate-level phosphorylation — that would directly transfer energy from electron transport to ADP phosphorylation. Despite two decades of intensive effort in the 1950s and early 1960s, no one could isolate this hypothetical intermediate. In 1961, Peter Mitchell proposed instead that the coupling was not chemical but electrochemical: electron transport pumps protons across a sealed membrane, and the resulting gradient itself is the energy currency that drives ATP synthesis. This was radical because it required no diffusible high-energy chemical intermediate at all — the membrane itself, as a compartment boundary, was the energy-storing device. Mitchell's hypothesis made specific testable predictions: 1. An intact, closed, proton-impermeable membrane is essential — punch holes in it and ATP synthesis stops even if electron transport continues. 2. Applying an artificial pH gradient or an electrical potential across the membrane, with no electron transport at all, should be sufficient to drive ATP synthesis. 3. Uncouplers — small lipid-soluble molecules that carry protons across membranes (like 2,4-dinitrophenol) — should abolish ATP synthesis without blocking electron transport, because they short-circuit the gradient. This had actually been observed for years before Mitchell's theory explained why. The decisive confirmation came from a landmark reconstitution experiment by Efraim Racker and Walther Stoeckenius in 1974. They took two entirely unrelated proteins — bacteriorhodopsin, a light-driven proton pump from the archaeon Halobacterium halobium, and purified mammalian ATP synthase (F1-Fo) — and co-reconstituted them into synthetic phospholipid vesicles (liposomes), with no other mitochondrial or bacterial proteins present. Bacteriorhodopsin has no evolutionary or mechanistic relationship to the mitochondrial electron transport chain; its only relevant property is that shining light on it pumps protons across a membrane. When Racker and Stoeckenius illuminated these vesicles, bacteriorhodopsin pumped protons into the vesicle, building a pH gradient and voltage, and ATP synthase — with no electron transport chain anywhere in the system — synthesized ATP from ADP and phosphate. This showed unambiguously that a proton gradient alone, generated by any pump whatsoever, is sufficient to drive ATP synthesis; no chemical intermediate specific to respiration is needed. It was one of the cleanest "sufficiency" experiments in cell biology. Independent support came from André Jagendorf's earlier (1966) experiment in chloroplasts: isolated thylakoids soaked in acid (pH 4) and then rapidly shifted into an alkaline buffer (pH 8) — an artificially imposed pH jump with no light, no electron flow — synthesized a burst of ATP in the dark. This "acid-bath" experiment showed a pH gradient by itself, imposed by the experimenter, could drive ATP synthesis, directly confirming prediction 2 in a related photosynthetic membrane system (foreshadowing the shared logic of Chloroplast Structure). Mitchell received the Nobel Prize in Chemistry in 1978, largely on the strength of these convergent lines of evidence. The chemical-coupling hypothesis was abandoned; chemiosmosis is now textbook consensus, though the precise stoichiometry of protons pumped per electron pair, and per ATP synthesized, is still refined by ongoing structural and biophysical work.

Example 1 — Calculating ATP yield and where the "P/O ratio" comes from. Consider the complete oxidation of one glucose molecule via glycolysis, pyruvate oxidation, and the citric acid cycle. Per glucose (accounting for two pyruvates):

Each NADH, upon donating electrons at Complex I, contributes enough proton pumping to yield about 2.5 ATP via ATP synthase; each FADH₂, entering at Complex II (after Complex I), yields only about 1.5 ATP, because it bypasses Complex I's pumping entirely. So:

• 10 NADH × 2.5 = 25 ATP
• 2 FADH₂ × 1.5 = 3 ATP
• + 4 substrate-level ATP
• Total ≈ 32 ATP per glucose

(Older textbooks cite 36–38 ATP using idealized integer P/O ratios of 3 and 2; the modern estimate of ~32 accounts for the cost of shuttling cytosolic NADH's electrons across the inner membrane and non-integer stoichiometry measured experimentally.) The key lesson: FADH₂ is intrinsically "worth less" than NADH because it enters the chain downstream of one proton-pumping step — a direct, quantitative consequence of the electron transport chain's physical architecture in the inner membrane.

Example 2 — A clinical reasoning case: uncouplers and fever without infection. A patient presents with high fever, rapid heart rate, profuse sweating, and weight loss, but no evidence of infection or inflammation. History reveals they have been taking an unregulated "fat-burning" supplement containing 2,4-dinitrophenol (DNP), an industrial chemical historically (and illegally) sold as a weight-loss drug in the 1930s and occasionally still trafficked today. Reasoning through the mechanism: DNP is a lipid-soluble weak acid that can pick up a proton in the intermembrane space, diffuse across the inner mitochondrial membrane, and release the proton in the matrix — bypassing ATP synthase entirely. This collapses the proton-motive force without producing any ATP. Because the electrons still need somewhere to go, the electron transport chain runs faster to compensate, consuming more oxygen and more fuel — but the energy that would normally be captured as ATP is instead dissipated as heat. This explains every symptom: fever (uncoupled respiration releases heat instead of storing it in ATP), rapid metabolic rate and weight loss (cells burn through fuel stores trying to maintain ATP levels), and potential fatal hyperthermia at high doses, since there is no negative feedback stopping the chain from simply running hotter and hotter. This is precisely why DNP was banned by the FDA in 1938 after a wave of poisonings — a direct, cautionary application of the chemiosmotic mechanism to a real toxicological emergency. Interestingly, mammals exploit a controlled version of the same principle physiologically: brown adipose tissue expresses uncoupling protein 1 (UCP1), which deliberately leaks protons to generate heat for non-shivering thermogenesis in infants and hibernating animals — the same mechanism, but regulated rather than uncontrolled.

Peter Dennis Mitchell was a British biochemist who spent much of his career outside conventional academic institutions, working from a private, self-funded research institute — Glynn House, later Glynn Research Laboratories — that he established in Cornwall using proceeds from selling his family's farm. This unconventional path let him pursue an idea that mainstream biochemistry, focused on finding a chemical intermediate, was reluctant to entertain: that a membrane's geometry, not a molecule's chemistry, was the missing piece of oxidative phosphorylation. His chemiosmotic hypothesis was met initially with considerable skepticism from established biochemists, including several who had spent their careers hunting for the "high-energy intermediate" chemiosmosis rendered unnecessary. The reconstitution experiments of the 1970s, especially Racker and Stoeckenius's bacteriorhodopsin work, turned the tide decisively in his favor.

Mitochondrial disease and diagnosis. Mutations in mitochondrial DNA or nuclear genes encoding ETC subunits cause diseases such as MELAS syndrome and Leber's hereditary optic neuropathy, which preferentially affect high-energy tissues (muscle, brain, eye) because these tissues depend most heavily on oxidative phosphorylation. Muscle biopsies showing "ragged red fibers" under histochemical staining directly visualize abnormal mitochondrial proliferation as cells try to compensate for defective ATP output.

Cancer metabolism. Many tumor cells favor aerobic glycolysis over oxidative phosphorylation even when oxygen is plentiful (the Warburg effect), and drugs targeting mitochondrial metabolism (such as metformin, which mildly inhibits Complex I) are being investigated as adjunct cancer therapies.

Pharmacology and toxicology. Understanding the electron transport chain explains cyanide poisoning (blocks Complex IV, halting all downstream ATP synthesis) and guides the design and monitoring of drugs with mitochondrial side effects, including some antiretroviral and antibiotic classes that inadvertently inhibit mitochondrial ribosomes because of their bacterial ancestry.

Aging and neurodegeneration. Accumulated mitochondrial DNA mutations and declining mitophagy (autophagic clearance of damaged mitochondria) are implicated in Parkinson's disease, where mutations in PINK1 and Parkin — proteins that tag damaged mitochondria for autophagic destruction — link mitochondrial quality control directly to neuronal survival.

Biotechnology and forensics. Because mitochondrial DNA is maternally inherited and present in high copy number per cell, it is used in forensic identification from degraded samples and in tracing maternal lineages in population genetics and human evolutionary history.

Agriculture and livestock. Selective breeding and mitochondrial haplotype analysis inform livestock productivity traits linked to metabolic efficiency, and engineered crop mitochondrial genomes are studied for cytoplasmic male sterility, a trait exploited in hybrid seed production.

Ask the tutor: How does the number and packing of cristae actually get measured, and why do some diseased mitochondria show visibly abnormal cristae shapes under electron microscopy?