Cholesterol Biosynthesis

Why cholesterol biosynthesis matters

• You make most of your own cholesterol. Of the roughly 1 gram of cholesterol turned over each day, about two-thirds is synthesized de novo rather than absorbed from the diet — which is why cutting dietary cholesterol has a modest effect on blood levels and why a drug that blocks synthesis works. The liver and intestine are the largest producers, but essentially every nucleated cell can run the pathway.
• Statins are the best-selling drug class in medical history. Atorvastatin (Lipitor) alone grossed over 150 billion dollars, the most of any drug ever. Statins are competitive inhibitors of HMG-CoA reductase and lower LDL cholesterol by 30 to 55 percent, cutting major cardiovascular events by roughly a quarter per 1 mmol/L of LDL reduction in the Cholesterol Treatment Trialists meta-analyses.
• Membrane fluidity depends on it. Cholesterol is the master modulator of the plasma membrane: its planar four-ring core intercalates between phospholipid tails, stiffening membranes above their melting temperature and preventing tight crystalline packing below it. Animal plasma membranes can be up to 30 to 50 mol percent cholesterol, and it is the organizing lipid of ordered lipid rafts.
• It is the sole precursor of steroids, bile acids, and vitamin D. Every steroid hormone — cortisol, aldosterone, testosterone, estradiol, progesterone — is carved from cholesterol by side-chain cleavage (CYP11A1). Bile acids that emulsify dietary fat, and 7-dehydrocholesterol that UV light converts to vitamin D in skin, all originate here.
• The pathway feeds far more than cholesterol. The mevalonate route branches at farnesyl pyrophosphate to make dolichol (carrier for N-linked glycosylation), ubiquinone / coenzyme Q10 (electron transport chain), heme A, and the prenyl anchors that tether Ras, Rho, and Rab GTPases to membranes. This is why blocking the pathway has pleiotropic effects and why the branch enzymes are drug targets in their own right (bisphosphonates hit farnesyl pyrophosphate synthase in bone).
• Loss-of-function is lethal. Smith-Lemli-Opitz syndrome, caused by mutations in DHCR7 (the terminal enzyme), produces severe developmental malformations and intellectual disability from cholesterol starvation of the embryo — direct proof that the pathway is indispensable, not a mere hazard.
• It is a compartmentalized ATP and NADPH sink. The early mevalonate steps run in the cytosol; from squalene onward the reactions are membrane-bound in the endoplasmic reticulum. The ~16 NADPH per molecule links cholesterol output directly to the pentose phosphate pathway and cellular redox state.

How cholesterol biosynthesis works, step by step

The pathway divides naturally into four stages. Stage 1 — mevalonate. In the cytosol, two molecules of acetyl-CoA condense to acetoacetyl-CoA (thiolase), and a third acetyl-CoA is added by HMG-CoA synthase to form the six-carbon 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA). Then the committed, irreversible, rate-limiting reaction: HMG-CoA reductase, embedded in the ER membrane, reduces the thioester to a primary alcohol using two NADPH, releasing mevalonate and free CoA. Everything upstream is shared with ketone-body synthesis; this reduction is the true entry into sterol metabolism, which is precisely why the cell lavishes four layers of control on it and why statins block it.

Stage 2 — activated isoprene. Three sequential ATP-dependent steps phosphorylate mevalonate (mevalonate kinase, phosphomevalonate kinase) and then decarboxylate it (mevalonate diphosphate decarboxylase), producing the five-carbon isopentenyl pyrophosphate (IPP), the universal isoprene unit. An isomerase interconverts IPP with dimethylallyl pyrophosphate (DMAPP). Head-to-tail condensations by prenyltransferase (farnesyl pyrophosphate synthase) join DMAPP + IPP into the 10-carbon geranyl pyrophosphate, then add another IPP to make the 15-carbon farnesyl pyrophosphate (FPP) — the central branch point of all isoprenoid metabolism.

Stage 3 — squalene and the first sterol. Squalene synthase performs a distinctive tail-to-tail condensation of two FPP molecules (consuming NADPH) to make the symmetric 30-carbon linear hydrocarbon squalene, committing carbon irreversibly to sterols. Squalene monooxygenase (squalene epoxidase, an oxygen- and NADPH-dependent enzyme, and a secondary regulatory checkpoint) inserts an oxygen to form 2,3-oxidosqualene. Then the remarkable lanosterol synthase (oxidosqualene cyclase) folds the linear chain into a chair-boat-chair-boat conformation and, in a single concerted cyclization cascade, forms four fused rings and creates lanosterol — the first molecule with the sterol skeleton.

Stage 4 — polishing lanosterol into cholesterol. Lanosterol is a 30-carbon sterol; cholesterol is 27. Around 19 further enzymatic steps remove three methyl groups (as CO2 and formate), reduce the double bonds, and shift the remaining double bond into place. Two parallel routes exist — the Bloch pathway (via desmosterol) and the Kandutsch-Russell pathway (via 7-dehydrocholesterol), differing in the order of side-chain reduction. Both converge on the terminal step: DHCR7 (7-dehydrocholesterol reductase) reduces 7-dehydrocholesterol to cholesterol. That final intermediate, 7-dehydrocholesterol, is also the substrate UVB light photolyzes in skin to make vitamin D3.

Feedback regulation: the SREBP switch

Cholesterol synthesis is one of biology's cleanest examples of end-product feedback inhibition, and the sensor is elegant. The transcription factor SREBP-2 (sterol regulatory element-binding protein 2) is synthesized as an inactive precursor threaded through the ER membrane, held there by its escort protein SCAP. When ER cholesterol rises above roughly 5 mol percent of membrane lipid, cholesterol binds directly to SCAP (and oxysterols bind the retention protein INSIG), locking the SCAP–SREBP complex in the ER. When cholesterol drops, INSIG lets go; SCAP escorts SREBP-2 in COPII vesicles to the Golgi, where two proteases — Site-1 protease (S1P) and Site-2 protease (S2P) — clip it in a regulated intramembrane proteolysis. The liberated N-terminal transcription factor enters the nucleus and switches on more than 30 genes, including HMGCR, HMG-CoA synthase, squalene synthase, and the LDL receptor, ramping up both synthesis and uptake.

A second, faster loop operates on the HMG-CoA reductase protein itself. When sterols and lanosterol-derived intermediates accumulate, INSIG binds the membrane domain of HMG-CoA reductase and marks it for ubiquitination and ERAD (ER-associated degradation) — sterol-accelerated destruction that can halve the enzyme's half-life. On top of that, AMP-activated protein kinase (AMPK) phosphorylates and inactivates HMG-CoA reductase when cellular energy is low, coupling cholesterol synthesis to ATP status. Transcription, protein degradation, and phosphorylation together clamp cellular cholesterol into a remarkably narrow range. This same feedback explains a clinical paradox: statins, by lowering hepatic cholesterol, activate SREBP-2, which upregulates the LDL receptor and pulls more LDL out of blood — the real mechanism behind their cholesterol-lowering effect.

Where the carbons and cofactors go

Stage	Key enzyme(s)	Product	Carbons	Cofactor / feature
Condensation	Thiolase, HMG-CoA synthase	HMG-CoA	6	3 acetyl-CoA in; shared with ketogenesis
Committed step	HMG-CoA reductase	Mevalonate	6	2 NADPH; rate-limiting; statin target
Activation	Kinases + decarboxylase	Isopentenyl-PP (IPP)	5	3 ATP; loses 1 CO2
Prenyl chain	FPP synthase	Farnesyl-PP (FPP)	15	Branch point: dolichol, CoQ, prenylation
Commitment to sterol	Squalene synthase	Squalene	30	NADPH; irreversible tail-to-tail
Cyclization	Squalene epoxidase + lanosterol synthase	Lanosterol	30	O2; forms 4 rings in one cascade
Demethylation	~19 steps (Bloch / Kandutsch-Russell)	Cholesterol	27	Loses 3 methyls; DHCR7 terminal

De novo synthesis vs dietary uptake

Property	Endogenous synthesis	Dietary / LDL uptake
Source	Acetyl-CoA via mevalonate pathway	Absorbed via NPC1L1, delivered as LDL
Share of daily turnover	~2/3 (roughly 700–900 mg/day)	~1/3 (typical Western diet)
Main tissue	Liver, intestine (but nearly all cells)	Enterocytes, then whole-body via LDL receptor
Rate-limiting control	HMG-CoA reductase, SREBP-2	NPC1L1 (ezetimibe target), LDL receptor number
Drug that blocks it	Statins, bempedoic acid (ACL inhibitor)	Ezetimibe, PCSK9 inhibitors, bile-acid sequestrants
Feedback when the other rises	Synthesis is suppressed when uptake is high	Uptake receptors fall when cells are cholesterol-replete

Common misconceptions

• "Cholesterol comes mainly from food." The reverse is closer to true: roughly two-thirds of body cholesterol is synthesized de novo, and the body compensates for dietary intake by tuning synthesis down. This is why dietary cholesterol has a smaller effect on serum LDL than saturated fat or genetics, and why a synthesis inhibitor is effective.
• "Statins work by dissolving cholesterol in the blood." Statins never touch circulating cholesterol directly. They block HMG-CoA reductase inside liver cells; the resulting drop in hepatic cholesterol activates SREBP-2, which upregulates LDL receptors, and it is those extra receptors that clear LDL from blood. The clinical effect is downstream of a transcriptional feedback loop, not a solvent action.
• "Cholesterol is only harmful." Cholesterol is essential to life. It sets membrane fluidity, forms lipid rafts, and is the obligate precursor of all steroid hormones, bile acids, and vitamin D. A total block of the pathway (as in severe Smith-Lemli-Opitz syndrome) is developmentally catastrophic. Danger comes specifically from chronically elevated LDL particles oxidizing in artery walls, not from the molecule per se.
• "Lanosterol is basically cholesterol." Lanosterol is the first sterol but still has 30 carbons and three extra methyl groups; converting it to 27-carbon cholesterol takes about 19 more enzymatic steps. The intermediates matter: 7-dehydrocholesterol is the vitamin D precursor, and desmosterol accumulates when the terminal reductase is blocked (as by the failed anti-cholesterol drug triparanol, withdrawn in 1962 for causing cataracts).
• "The pathway only makes cholesterol." The mevalonate pathway is the sole animal source of all isoprenoids. Ubiquinone (coenzyme Q10), dolichol, heme A, and the prenyl anchors of small GTPases all branch off before the sterol commitment step. Some statin side effects (and the biology of bisphosphonate bone drugs) trace to these non-sterol branches.
• "Feedback is just product inhibition of one enzyme." Regulation is multilayered: transcriptional (SREBP-2 turning on 30+ genes), post-translational (sterol-accelerated ERAD degradation of HMG-CoA reductase), and energy-sensing (AMPK phosphorylation). It is a systems-level control circuit, not a single allosteric switch.

A famous history: mapping the pathway and finding statins

• Bloch's isotope tracing (1940s–50s). Trained in Rudolf Schoenheimer's pioneering isotope-labeling lab at Columbia, Konrad Bloch and David Rittenberg fed animals acetate labeled with deuterium and then carbon-14 and painstakingly traced where every carbon of cholesterol came from, proving that all 27 carbons derive from the two-carbon acetate unit. Feodor Lynen independently worked out the activated acetyl-CoA and mevalonate chemistry. The two shared the 1964 Nobel Prize in Physiology or Medicine "for their discoveries concerning the mechanism and regulation of the cholesterol and fatty acid metabolism."
• Cornforth and the stereochemistry (1975 Nobel in Chemistry). John Cornforth used chiral, isotopically labeled substrates to determine the exact three-dimensional stereochemistry of the squalene cyclization and other steps, revealing how a linear chain folds into the rigid four-ring sterol with defined chirality at every center.
• Endo's discovery of statins (1976). Akira Endo, hunting through thousands of fungal broths in Japan for HMG-CoA reductase inhibitors, isolated compactin (mevastatin) from Penicillium citrinum. It was the first statin. Merck's lovastatin, from Aspergillus terreus, became the first FDA-approved statin in 1987, launching a drug class that has since prevented millions of heart attacks.
• Brown and Goldstein and the LDL receptor (1985 Nobel). Michael Brown and Joseph Goldstein discovered the LDL receptor and the feedback logic linking uptake to synthesis, later identifying the SREBP–SCAP–INSIG sensor. Their work explained familial hypercholesterolemia (LDL-receptor mutations) and gave the mechanistic rationale for why statins work through receptor upregulation.
• Triparanol, the cautionary tale (1962). The first drug designed to inhibit the terminal enzyme (DHCR24, blocking desmosterol conversion) was pulled from the market for causing cataracts and skin damage from accumulated desmosterol — an early demonstration that intermediates of the pathway, not just the endpoint, are biologically potent.