← Sage — Wipeouts · Sage hub

Sage — A Metabolic Hypothesis of Hepatology

Energy Flux, Nitric Oxide Matching, Metabolic Zonation, and the Crabtree-to-Warburg Progression in Liver Disease

Disclaimer. This essay presents a metabolic hypothesis intended for intellectual exploration and clinical discussion. It is not medical advice, diagnosis, or a treatment protocol. Mechanisms discussed are interpretive except where a specific experimental result is cited. Individual circumstances vary; any interventions should be undertaken only under appropriate medical supervision.

We are born with perfect biology. That is the first and most radical claim of the metabolic philosophy of medicine. Perfect for intermittent fuel availability, natural circadian entrainment by solar and seasonal cues, and the obligatory movement-strength hormesis that defined hominin life. What we label disease is, in the overwhelming majority of chronic cases, not a random error or a software bug in the genome. It is the predictable downstream consequence of a biology operating in chronic compensatory mode because its upstream ATP-generating capacity has been chronically compromised by mismatch.

Acute compensation is life-saving and productive. Chronic compensation becomes the pathology we name and attempt to suppress. Two related bioenergetic responses illustrate the distinction. The Crabtree effect is the reversible suppression of oxidative phosphorylation that occurs in some cells when glucose is abundant; substrate-level phosphorylation rises temporarily while mitochondria remain competent. The Warburg effect is the persistent reliance on aerobic glycolysis and reduced metabolic flexibility that appears when mitochondrial ATP-generating headroom has been eroded and the compensatory program locks in. Acute Crabtree-like shifts can still reverse when load falls or oxidative capacity is restored. Chronic Warburg-like shifts become architecture: fragmented mitochondria, suppressed fatty-acid oxidation, lipotoxicity, fibrosis, and the named syndromes of progressive liver disease.

The named syndromes of hepatology—metabolic dysfunction-associated steatotic liver disease (MASLD) and steatohepatitis (MASH), alcoholic liver disease, progressive fibrosis and cirrhosis, many forms of acute liver injury that fail to recover fully, and hepatocellular carcinoma—are visible clusters of locked compensatory responses in an organ whose principal work is the continuous management of whole-body energy flux. The underpowered periportal hepatocyte, the fragmented pericentral mitochondrion, the glycolytic activated stellate cell, and the rarefied sinusoidal microcirculation are not independent primary lesions; they are tissues whose ATP-generating headroom has been eroded and that have therefore shifted toward more primitive, less flexible ATP generation. Nitric oxide functions throughout as the short-range energy-flux manager that tries to keep delivery, consumption, and conduit integrity matched to that mitochondrial demand. When the manager fails, the mismatch persists and the compensatory programs become scars.

The Liver as Central Bioenergetic Hub and Energy Flux Manager

The liver receives approximately 25 % of cardiac output through dual portal and arterial supply yet constitutes only 2–3 % of body mass. Its energy demand ranks among the highest of solid organs because it continuously integrates nutrient fluxes from the gut, adipose tissue, and muscle while simultaneously synthesizing proteins, lipoproteins, bile acids, glucose, and ketones for the rest of the body. Hepatocytes contain among the densest mitochondrial populations in the organism; oxidative phosphorylation supplies the overwhelming majority of ATP under physiologic conditions. Preferred substrates shift with nutritional state: fatty-acid β-oxidation and ketogenesis dominate in the fasted state, while glycolysis and de novo lipogenesis rise in the fed state. A single palmitate molecule yields on the order of 100+ ATP versus ~30–36 from glucose, making fatty acids the preferred high-yield fuel for the energy-intensive biosynthetic and detoxification work of the hepatocyte.

Unlike the kidney’s proximal tubule, which performs largely unidirectional reabsorptive work, the liver operates as a bidirectional flux manager. It extracts excess substrate from portal blood, stores it (glycogen, triglyceride), converts it (gluconeogenesis, ketogenesis, urea synthesis), and releases it to systemic circulation according to systemic demand. This continuous matching of input, storage, conversion, and output is energy-intensive and depends on intact mitochondrial capacity and on local nitric-oxide signaling that keeps sinusoidal perfusion matched to hepatocyte demand. Liver sinusoidal endothelial cells (LSECs) constitutively express eNOS; shear stress and local hypoxia increase NO release, which relaxes adjacent hepatic stellate cells, maintains fenestrations, and modulates sinusoidal resistance. When these acts remain coupled, high energy flux is sustainable across the lobule. When they fail—through endothelial dysfunction, superoxide scavenging of NO, or loss of LSEC phenotype—the cell faces the same upstream deficit already identified in the metabolic hypotheses of aging, atherosclerosis, glaucoma, and nephrology: insufficient oxidative ATP relative to the cost of remaining ordered. Dashboard readings (elevated ALT/AST, steatosis, fibrosis scores, declining synthetic function, portal hypertension) follow.

Metabolic Zonation: Spatial Specialization of Energy Flux

The hepatic lobule is not metabolically homogeneous. The fall in oxygen and nutrients from portal (zone 1, periportal) to central vein (zone 3, pericentral) produces a division of labor that may itself be an ancestral energy-efficiency strategy. Periportal hepatocytes receive blood with higher oxygen tension (~60–65 mm Hg) (Kietzmann 2017) and the first pass of portal nutrients; they therefore specialize in energy-demanding oxidative pathways: gluconeogenesis, urea-cycle nitrogen disposal, fatty-acid β-oxidation, and cholesterol synthesis. They possess higher mitochondrial density, higher maximal respiratory capacity, and a preference for fatty-acid substrates. Pericentral hepatocytes receive blood with lower oxygen tension (~30–35 mm Hg) and specialize in glycolysis, lipogenesis, ketogenesis, glutamine synthesis, and xenobiotic phase I metabolism. Mid-lobular (zone 2) hepatocytes buffer the transitions.

This zonation is plastic and nutrient-sensing. AMPK and mTOR signaling, acting through phosphorylation of mitochondrial proteins, may shift mitochondrial phenotype and substrate preference according to local nutrient levels. Acute carbohydrate loads can produce a Crabtree-like rise in glycolysis in zone 3 while oxidative capacity in zone 1 remains intact. Chronic mismatch—persistent hyperinsulinemia, fructose overload, or loss of NO-mediated oxygen rationing—may erode the zonal specialization. Periportal mitochondria may lose fatty-acid oxidation capacity and begin to express pericentral lipogenic programs; pericentral mitochondria may fragment and generate excess ROS. The result is a loss of the spatial energy-efficiency architecture that normally allows the organ to handle large, intermittent substrate loads without locking into compensatory glycolysis.

Alcohol and Fructose as Disproportionate Stressors

Alcohol and fructose occupy a special place among hepatic stressors because both are metabolized almost exclusively by the liver and both bypass the regulatory checkpoints that normally limit glucose flux. Ethanol is oxidized by alcohol dehydrogenase and CYP2E1 to acetaldehyde, generating NADH and reactive oxygen species while directly damaging mitochondrial membranes and uncoupling respiration. Fructose is phosphorylated by ketohexokinase (fructokinase) to fructose-1-phosphate in a reaction that is not subject to the feedback inhibition that restrains glucokinase. The rapid ATP consumption produces a transient energy crisis, elevates uric acid, and drives de novo lipogenesis far more efficiently than an equicaloric load of glucose. On this hypothesis, both substrates therefore force an acute Crabtree-like shift: glycolytic and reductive pathways rise while residual oxidative capacity is still intact, yet the magnitude of the load rapidly erodes that capacity.

Excess simple carbohydrate produces the same directional shift through a different route. High portal glucose concentrations accelerate glycolysis, raise the NADH/NAD⁺ ratio, and inhibit pyruvate dehydrogenase and the TCA cycle—a Crabtree-like effect. When the carbohydrate is sucrose or high-fructose corn syrup, the fructose moiety amplifies the effect by simultaneously activating carbohydrate-response element-binding protein (ChREBP) and sterol regulatory element-binding protein-1c (SREBP-1c), transcriptionally up-regulating the entire lipogenic program while the energy shortfall impairs fatty-acid oxidation. The result is not merely caloric surplus but a preferential routing of carbon into triglyceride at the precise moment mitochondrial headroom is declining.

Historically, refined sugar was a rare commodity. Per-capita sugar intake in England was roughly four pounds per year in 1700; by the early 2000s, combined sugar and high-fructose corn syrup intake in the United States was almost 150 pounds per person per year, with a further rise after high-fructose corn syrup was introduced in the early 1970s (Johnson et al. 2007). The same period saw the emergence and subsequent epidemic of the diseases of civilization—obesity, type 2 diabetes, MASLD, hypertension, and cardiovascular disease—in populations that had previously tolerated high-starch diets without these complications. Alcohol, long recognized as a direct hepatotoxin, shares the identical metabolic logic: both are “first-pass” hepatic substrates whose chronic excess converts a reversible Crabtree-like adjustment into locked Warburg architecture. On this reading, the epidemiologic coincidence is not accidental; it may reflect the liver’s limited capacity to oxidize a continuous, unregulated substrate load that ancestral physiology never encountered.

Carbohydrate Metabolism and Glucose Homeostasis

Conventional model

The liver maintains euglycemia by storing glucose as glycogen in the fed state and producing glucose via glycogenolysis and gluconeogenesis in the fasted state. Insulin suppresses hepatic glucose output; glucagon and catecholamines stimulate it. Disease is framed as insulin resistance, excess gluconeogenesis, or glycogen storage defects.

Metabolic / energy-flux model

Gluconeogenesis is an ATP-consuming pathway (six ATP equivalents per glucose from lactate or alanine). It is localized to the well-oxygenated periportal zone where mitochondrial capacity is highest. An acute rise in portal glucose can elicit a reversible Crabtree-like suppression of oxidative phosphorylation in favor of glycolysis while residual mitochondrial competence remains. When mitochondrial headroom is chronically constrained—by substrate overload, lost NO matching, or impaired biogenesis—the hepatocyte may not suppress gluconeogenesis appropriately even in the fed state. Excess acetyl-CoA from accelerated β-oxidation or from de novo lipogenesis feeds the TCA cycle, elevating gluconeogenic flux as a compensatory attempt to dispose of reducing equivalents. The same energy shortfall impairs glycogen synthesis and storage. Fructose, which bypasses phosphofructokinase regulation and directly burdens the liver, accelerates this mismatch by driving rapid ATP depletion and uric acid generation while simultaneously promoting de novo lipogenesis. On this reading, the dashboard reading of fasting hyperglycemia is not a primary insulin-signaling failure but may be the downstream consequence of insufficient oxidative capacity to match the energetic cost of ordered glucose handling.

Lipid Metabolism, Fatty-Acid Oxidation, and MASLD/MASH

Conventional model

MASLD (formerly NAFLD) is driven by excess free fatty acid delivery from adipose tissue, de novo lipogenesis from carbohydrate, and impaired export as VLDL. Progression to MASH involves lipotoxicity, oxidative stress, inflammation, and fibrosis. Insulin resistance is both cause and consequence.

Metabolic / energy-flux model

Hepatocytes prefer fatty-acid β-oxidation for ATP generation. In the early stages of substrate overload, mitochondria adapt: β-oxidation, TCA-cycle flux, and OXPHOS capacity rise, protecting against steatosis. This is a reversible compensatory response analogous to the Crabtree effect—oxidative capacity remains competent while glycolytic and lipogenic pathways increase. Persistence of the load may erode headroom. Mitochondrial fragmentation increases (Drp1), fusion and biogenesis decline (Mfn2, PGC-1α), respiratory supercomplexes destabilize, and incomplete β-oxidation generates lipotoxic intermediates (ceramides, DAGs). ROS damage mtDNA and further suppress oxidative capacity. On this hypothesis, the shift locks into a Warburg-like phenotype: suppressed FAO, elevated aerobic glycolysis, lipid droplet accumulation, and a pro-inflammatory, pro-fibrotic secretome.

De novo lipogenesis is itself an energy-consuming process that further burdens already constrained mitochondria. Selective hepatic insulin resistance—impaired suppression of gluconeogenesis but preserved or enhanced lipogenesis—reflects the differential energetic cost and zonal location of these pathways. Endothelial NO loss at the sinusoids may be an early matching failure; reduced bioavailability permits leukocyte adhesion, loss of fenestrations, and activation of stellate cells. Experimental restoration of mitochondrial biogenesis or fatty-acid oxidation capacity can attenuate steatosis and fibrosis, consistent with the hypothesis that the underpowered hepatocyte, rather than the lipid itself, may be the primary lesion. Metabolic memory may be the persistence of these mitochondrial and NO-matching defects after caloric intake is reduced.

Protein and Amino-Acid Metabolism and the Urea Cycle

Conventional model

The liver synthesizes the majority of plasma proteins and disposes of nitrogen via the urea cycle. Impairment produces hypoalbuminemia, coagulopathy, and hyperammonemia.

Metabolic / energy-flux model

Protein synthesis accounts for a substantial fraction of cellular oxygen consumption: across mammalian tissues it is estimated at about 25–30 % of the oxygen consumption coupled to ATP synthesis, roughly a fifth of the total (Rolfe and Brown 1997). The urea cycle is likewise ATP-intensive and localized to the periportal zone. An acute amino-acid load raises ATP demand and can elicit transient compensatory adjustments while oxidative capacity remains intact. When mitochondrial headroom is constrained, the hepatocyte may prioritize essential ATP-generating pathways over protein synthesis and complete nitrogen disposal. Incomplete urea-cycle flux elevates ammonia, which itself impairs mitochondrial function in other organs (hepatic encephalopathy). On this reading, the underpowered epithelium cannot clear the nitrogen load cleanly and therefore initiates stress programs that favor senescence and fibrotic signaling. Restoration of oxidative capacity or reduction of nitrogen load may improve synthetic function even when histologic fibrosis is advanced, suggesting that the energy deficit, rather than irreversible architectural loss, may be rate-limiting in many cases.

Detoxification, Xenobiotic Metabolism, and Phase I/II Pathways

Conventional model

Phase I (primarily CYP450 oxidation) and Phase II (conjugation) convert lipophilic toxins and drugs into excretable forms. Toxicity arises from reactive intermediates or enzyme induction/inhibition.

Metabolic / energy-flux model

CYP450 reactions consume NADPH and molecular oxygen and generate reactive intermediates that must be rapidly conjugated. Both steps are energy-intensive and redox-sensitive. Phase I is preferentially localized to the lower-oxygen pericentral zone; Phase II capacity depends on adequate glutathione, sulfate, glucuronide, and methyl donors whose synthesis itself requires ATP and reducing equivalents. An acute toxin load can produce a Crabtree-like shift while residual oxidative capacity supports NADPH generation and conjugation. Chronic mitochondrial insufficiency collapses this matching: incomplete Phase I generates persistent reactive species, Phase II cofactors are depleted, and the hepatocyte activates stress and fibrotic programs. Acetaminophen toxicity is the prototype—excess NAPQI depletes glutathione and damages mitochondria, converting a manageable load into necrosis. Alcoholic liver disease may follow the same logic: ethanol metabolism via ADH and CYP2E1 generates acetaldehyde and ROS, directly damages mitochondria, and may lock the cell into a glycolytic, pro-fibrotic state. On this reading, the underpowered hepatocyte, not the toxin per se, may determine whether the insult remains reversible or becomes architecture.

Bile Acid Synthesis, Enterohepatic Circulation, and Cholesterol Flux

Conventional model

Bile acids are synthesized from cholesterol in the liver, conjugated, secreted into bile, and reabsorbed in the ileum. They emulsify dietary lipids and regulate metabolism via FXR and TGR5. Cholestasis produces toxicity from retained bile acids.

Metabolic / energy-flux model

Bile acid synthesis is a multi-step, energy-consuming pathway preferentially initiated in the pericentral zone and may be completed as intermediates flow toward the portal triad (a spatial “conveyor belt”). Conjugation with glycine or taurine and canalicular secretion are ATP-dependent. The enterohepatic circulation recycles the majority of the pool, reducing the energetic cost of de novo synthesis. An acute increase in cholesterol or bile-acid load can be handled by residual oxidative capacity. When mitochondrial headroom and NO matching are constrained, synthesis and secretion fail, leading to intracellular accumulation of toxic intermediates, FXR dysregulation, and further suppression of FAO. On this hypothesis, cholestatic injury represents a locked compensatory state in which the energetic cost of ordered bile-acid handling can no longer be met. Restoration of oxidative capacity or reduction of the cholesterol load may improve cholestatic markers even before histologic resolution, consistent with an energy-flux rather than purely architectural lesion.

Acute Liver Injury, Ischemia-Reperfusion, and Recovery

Conventional model

Acute liver injury is classified by etiology (ischemic, toxic, viral, autoimmune). Recovery depends on residual viable hepatocytes and the absence of ongoing insult. Many patients progress to chronic disease.

Metabolic / energy-flux model

Hepatocytes are vulnerable because their high biosynthetic and detoxification workload leaves limited glycolytic reserve. Ischemia or toxin immediately collapses mitochondrial ATP production, uncouples respiration, and fragments organelles—an acute bioenergetic crisis. Transient compensatory glycolysis is a Crabtree-like attempt that is usually insufficient to protect synthetic function or maintain mitochondrial quality control. Failure of mitophagy and biogenesis after the insult locks the epithelium into a persistent glycolytic, pro-fibrotic state—the Warburg transition from acute injury to chronic fibrosis. Interventions that support residual oxidative capacity, preserve sinusoidal NO tone, or reduce workload during the injury window may improve recovery; persistent bioenergetic failure converts the acute compensation into architecture. Regeneration itself is metabolically constrained: in mice, ETC-dysfunctional hepatocytes are suppressed during regeneration because they cannot generate sufficient acetyl-CoA from fatty acids (Wang et al. 2024), providing a quality-control mechanism that favors oxidative competence.

Chronic Liver Disease, Fibrosis, and Cirrhosis

Conventional model

Persistent injury activates hepatic stellate cells, which deposit extracellular matrix under TGF-β and other signals, producing progressive fibrosis and cirrhosis with portal hypertension and synthetic failure.

Metabolic / energy-flux model

Once a critical fraction of hepatocytes have completed the transition from reversible Crabtree-like glycolysis to persistent Warburg-like suppression of FAO and OXPHOS, the organ enters a self-reinforcing state of lipotoxicity, ROS amplification, impaired mitophagy, stellate-cell activation, and immune-cell metabolic reprogramming toward glycolysis. On this hypothesis, fibrosis is the histologic expression of locked compensatory metabolism. Loss of LSEC NO matching accelerates sinusoidal hypoxia and stellate-cell contraction, converting injury into fixed architecture and portal hypertension. Therapies that restore mitochondrial biogenesis, improve fatty-acid oxidation, preserve coupled endothelial NO, or reduce substrate load may interrupt the cycle at the engine level. Even advanced fibrosis may retain a reversible component when the upstream energy mismatch is corrected, as suggested by histologic improvement after bariatric surgery or sustained caloric restriction.

Hepatocellular Carcinoma

Conventional model

HCC is driven by chronic inflammation, cirrhosis, viral oncogenes, or mutational cascades (TP53, CTNNB1, etc.). Aerobic glycolysis (Warburg effect) is treated as a downstream consequence of oncogenic signaling.

Metabolic / energy-flux model

In the author's reading, the two-step sequence of the cancer hypothesis applies directly. First, chronic bioenergetic mismatch—substrate overload, sinusoidal hypoxia, impaired fatty-acid oxidation, or lost nitric-oxide oxygen rationing—forces hepatocytes into a Crabtree-like rise in glycolysis and reductive metabolism while residual oxidative capacity remains. When the mismatch persists, the same cells complete the transition to a locked Warburg phenotype: suppressed OXPHOS, mitochondrial fragmentation, lipid droplet accumulation, and HIF-driven transcriptional reprogramming. Second, natural selection within that niche favors clones that stably occupy the glycolytic state and that further disable residual mitochondrial checkpoints. On this reading, the tumor is not mutation-first; it may be a metabolic scar that has become a lineage. The same energy-flux and matching failures that drive interstitial fibrosis in MASLD may create the selective environment for this lineage. Dashboard mutations and angiogenic signaling are real and therapeutically actionable; on this reading they are downstream of, or permissive for, the bioenergetic lock rather than its origin. This reading is continuous with the cancer hypothesis in the series and predicts that interventions restoring oxidative headroom or sinusoidal matching will reduce the probability of the second, selective step even when they do not directly target specific oncogenes.

Systemic Effects and Cross-Organ Correlations

The liver’s central role in whole-body energy flux means that its bioenergetic failure is transmitted systemically. Hepatic insulin resistance and excess glucose production drive hyperglycemia and pancreatic β-cell stress. Impaired FAO and elevated circulating free fatty acids and ceramides may promote myocardial lipotoxicity and the same FAO-to-glycolysis transition described in the heart. Ammonia and other uremic-like toxins impair cerebral mitochondrial function (hepatic encephalopathy). The hepatorenal syndrome may reflect parallel failure of NO matching and oxidative capacity in the kidney’s energy-intensive tubules. Sarcopenia may be the skeletal-muscle expression of the same systemic bioenergetic collapse, amplified by reduced hepatic production of anabolic signals.

The eye has served as the accessible prototype of energy-intensive tissue. Retinal neurons, pigment epithelium, and laminar capillaries share the liver’s dependence on oxidative phosphorylation and on shear-timed nitric oxide for matching. Diabetic retinopathy and MASLD are parallel microvascular and parenchymal complications that may be driven, in part, by the same mitochondrial and NO programs. On this hypothesis, hypertension–glaucoma and hypertension–portal hypertension linkages are continuous: both may be tissues whose ATP headroom and NO matching have been eroded, after which modest hemodynamic load becomes pathogenic. The heart and kidney exhibit the same FAO preference and NO dependence. Atherosclerosis, neurodegeneration, and certain cancers follow the identical logic of locked compensatory metabolism once ATP-generating headroom and matching fail. Circulating mitochondrial DNA, succinate, FGF21, and GDF15 may provide bidirectional signaling that reinforces the multi-organ scar.

Relation to EnSurf

EnSurf organizes restoration of ancestral entrainment around three pillars that directly address the bioenergetic and matching deficits in the liver.

• Fuel. Animal fats and proteins, adequate cofactors, and periods of fasting or ketosis supply preferred substrates for hepatocyte FAO while limiting chronic carbohydrate-driven insulin signaling and de novo lipogenesis. Ketone bodies support redox balance and can be used by peripheral tissues, reducing hepatic gluconeogenic demand. Microbiome-accessible carbohydrates maintain short-chain fatty-acid signaling that supports hepatic metabolic flexibility. If you have cirrhosis or other liver disease, talk with your physician before changing protein intake or trying fasting or ketosis.

• Circadian. Solar light entrainment and sleep may protect mitochondrial quality-control programs and the nocturnal perfusion window in which sinusoidal and zonal matching are most vulnerable. Disruption of circadian rhythm may impair zonal specialization and promote steatosis.

• Strength. Resistance loading and hormetic stimuli drive PGC-1α-mediated biogenesis, improve systemic insulin sensitivity (thereby reducing portal substrate load), and write laminar shear that keeps LSEC eNOS coupled. Agents such as methylene blue, PQQ, berberine, or NR remain secondary supports for redox and biogenesis.

These interventions do not replace indicated pharmacologic therapy; they may address the upstream mismatch that makes those therapies necessary and reduce the probability that acute Crabtree-like compensations lock into Warburg architecture.

Conclusion

Hepatology’s major chronic syndromes are the tissue-specific expression of chronic compensatory metabolism in an organ whose defining work is the continuous management of whole-body energy flux. When mitochondrial ATP-generating capacity is chronically insufficient relative to load, and when nitric-oxide matching of delivery, rationing, and conduit protection fails, periportal and pericentral hepatocytes, sinusoidal endothelium, and stellate cells progress from reversible Crabtree-like shifts to locked Warburg-like glycolysis, lipid accumulation, ROS amplification, and fibrotic programs. The dashboard readings follow. The same engine and matching failure is visible in the retina, myocardium, kidney, and other high-demand tissues.

Restoring ancestral fuel oxidation, circadian entrainment, and hormetic loading may expand ATP-generating headroom and keep nitric-oxide matching coupled, thereby reducing the probability that acute compensation becomes the named diseases of hepatology. The hypothesis is testable, mechanistically continuous with the nitric-oxide and metabolic essays in this series, and clinically actionable under appropriate supervision.

Reference: Kietzmann T. Metabolic zonation of the liver: the oxygen gradient revisited. Redox Biology 2017;11:622–630. doi:10.1016/j.redox.2017.01.012

Reference: Johnson RJ, Segal MS, Sautin Y, Nakagawa T, Feig DI, Kang DH, Gersch MS, Benner S, Sánchez-Lozada LG. Potential role of sugar (fructose) in the epidemic of hypertension, obesity and the metabolic syndrome, diabetes, kidney disease, and cardiovascular disease. American Journal of Clinical Nutrition 2007;86(4):899–906. doi:10.1093/ajcn/86.4.899

Reference: Rolfe DF, Brown GC. Cellular energy utilization and molecular origin of standard metabolic rate in mammals. Physiological Reviews 1997;77(3):731–758. doi:10.1152/physrev.1997.77.3.731

Reference: Wang X, Menezes CJ, Jia Y, et al. Metabolic inflexibility promotes mitochondrial health during liver regeneration. Science 2024;384(6701):eadj4301. doi:10.1126/science.adj4301

Coach version: Coach — The Liver as an Energy Manager.

A working hypothesis, open to correction and revised as new evidence arrives. Last revised October 23, 2026.

Related: Sage — Nephrology · Sage — Nitric Oxide · Sage — Diabetes · Sage — GI · Sage — Fuel — Toxins · Sage — Cancer · Sage — The Good Doctor.

See the notice at the top of this page. Working hypothesis, not medical advice.