# Fatty Acid Synthesis: Building the Chain Two Carbons at a Time

> Fatty acid synthesis turns a fed cell's spare acetyl CoA into long-chain fat, and it is deliberately not beta-oxidation run backward. Three differences make that possible: the work happens in the cytoplasm on a dedicated carrier (ACP), it spends NADPH instead of banking NADH, and one committed, tightly regulated enzyme (acetyl CoA carboxylase) gates whether new fatty acids get made. Trace the carbon from citrate to palmitate and the whole pathway resolves into one stoichiometric ledger: 8 acetyl CoA, 7 ATP, and 14 NADPH per C16 chain.

Course: Biochemistry (Fuels, Fat & Nitrogen). Level: university elective. About 25 minutes.
Lesson page (link this one when citing): https://stemhop.com/lessons/fatty-acid-synthesis/

Builds on: Fatty Acid Degradation: The β-Oxidation Spiral · The Citric Acid Cycle: Harvesting Electrons · The Pentose Phosphate Pathway: NADPH and Five-Carbon Sugars

## What you'll learn

- Explain the three ways fatty acid synthesis differs from beta-oxidation (cytoplasm vs mitochondria, ACP/phosphopantetheine vs CoA, NADPH consumed vs NADH/FADH2 produced) and why that separation permits independent regulation and prevents a futile cycle.
- Trace carbon and reducing power through the citrate shuttle, accounting for how citrate delivers both acetyl groups and, via malic enzyme, up to 8 of the 14 NADPH, with the rest from the pentose phosphate pathway and other routes.
- Identify acetyl CoA carboxylase as the committed, biotin-dependent step and explain why spending ATP to add CO2 (making malonyl CoA) pulls condensation forward even though CO2 is released again.
- Walk the four-step CRDR cycle on fatty acid synthase (condensation, reduction, dehydration, reduction) and reproduce the full stoichiometry: 8 acetyl CoA + 7 ATP + 14 NADPH + 7 HCO3⁻ yield palmitate + 7 CO2 + 8 CoA + 14 NADP⁺ + 7 ADP + 7 Pᵢ + 6 H2O.
- Describe the three layers of ACC regulation (AMPK/PP2A phosphorylation, citrate/palmitoyl-CoA allostery, insulin/glucagon hormones) and the reciprocal block of CPT-I by malonyl CoA, then apply it to metformin and to alcohol-induced fatty liver.

**One cookie. One slice of pizza. If you don’t burn it, your body stores it.** Most of the fat in it is stored as it is. The sugar and starch refill glycogen first, and some of the surplus is turned into new fat by this pathway, running mainly in your liver and fat cells. That carbon becomes acetyl CoA, and in the fed state acetyl CoA is stitched into long fatty acid chains two carbons at a time.

This lesson answers two questions. What does it cost to build one fatty acid? And how does a cell that is building fat avoid burning it at the same time? The film above follows the ATP-for-CO₂ trick through one turn, then all seven turns to palmitate; the sections below add the chemical detail, full ledger, delivery bill and carnitine palmitoyltransferase I (CPT-I) control.

The film’s rule: after the acetyl primer, every two carbons arrive as one malonyl CoA, for one ATP at acetyl CoA carboxylase (ACC) and two NADPH at fatty acid synthase (FAS). The rule covers saturated chains built from acetyl CoA already in the cytoplasm. Delivering the acetyl groups and finding the NADPH are separate bills, paid later. Odd-chain fats start from a different primer (Going deeper, “Where the rules bend”), and double bonds are added after the chain is built.

## Why it is not beta-oxidation in reverse

You have met beta-oxidation, the pathway that *takes fat apart*. It is tempting to assume synthesis runs it backward, but three differences set the two apart.

**Location.** Degradation happens mainly in the mitochondrion. Synthesis happens mainly in the cytoplasm. The cell physically separates the factory from the demolition crew.

**Carrier.** Beta-oxidation carries its intermediates on coenzyme A. Synthesis carries them on a dedicated acyl carrier protein, ACP, using a long phosphopantetheine arm that swings the growing chain from one active site to the next without ever letting it float free.

**Reducing power.** Degradation *produces* NADH and FADH2. Synthesis *consumes* NADPH. The extra phosphate on NADPH is a tag that enzymes read, so NAD and NADP form two separate pools kept in opposite states. In the cytoplasm NAD is kept mostly oxidized, as NAD⁺, ready to accept electrons from burning fuel; NADP is kept mostly reduced, as NADPH, ready to hand electrons to building. The two pools do not mix on their own: moving electrons from NADH to NADPH costs the cell energy (Paying the bill).

**Predict.** Suppose synthesis were the exact reverse of beta-oxidation: in the mitochondrial matrix, on CoA, using NADH. What could the cell no longer do?

> **Check your prediction.** Switch one direction on while holding the other off. With one place, one carrier and one electron pool, new chains would sit beside the enzymes that burn them: a futile cycle, building and tearing down the same molecule at once and paying for both. Separation lets the cell regulate each direction independently.

*Figure (AI-generated illustration).* Synthesis is not beta-oxidation run backward. Burning happens in the mitochondrion with coenzyme A as carrier and makes NADH and FADH2; building happens in the cytoplasm with ACP as carrier and consumes NADPH. Image: two halves compare burning fat in a mitochondrion with CoA carrying the chain and NADH and FADH2 produced, against building fat in the cytoplasm with ACP carrying the chain and NADPH consumed.

## Stage 1: Transport. Citrate carries the acetyl group out

Acetyl CoA is made in the mitochondrion, but it cannot cross the inner mitochondrial membrane. So the cell sends the acetyl group across in disguise.

Citrate is the disguise. The first step of the tricarboxylic acid (TCA) cycle joins acetyl CoA to oxaloacetate to make citrate, and when fuel is plentiful, citrate builds up and is exported to the cytoplasm. There, **ATP-citrate lyase** spends an ATP to cleave it back into acetyl CoA and oxaloacetate. Think of citrate as an Uber for acetyl groups.

The leftover oxaloacetate helps pay for the build; we return to it once we know the bill.

## Stage 2: Activation. The committed step

The film above shows **acetyl CoA carboxylase (ACC)** spending one ATP to attach a carboxyl group, supplied as bicarbonate, to cytoplasmic acetyl CoA, making **malonyl CoA**.

ACC carries biotin, a vitamin tethered to a lysine. Biotin is a swinging arm, a CO₂ taxi in the informal picture: it picks up the carboxyl group at one active site and drops it onto acetyl CoA at another. If you met pyruvate carboxylase in gluconeogenesis, this is the same trick.

More precisely, the malonyl CoA made by ACC1, the form that feeds synthesis, has essentially one fate: addition to a fatty acid. That is why ACC1 is the committed step and why the cell’s fast controls act here rather than later.

## Stage 3: Elongation, one turn at a time

The film above animates **fatty acid synthase (FAS)** repeating **C**ondensation, **R**eduction, **D**ehydration, **R**eduction, or **CRDR**. In mammals FAS is one enormous protein, a pair of identical chains (a homodimer), and each chain carries every tool the cycle needs.

| FAS domain | What it does |
| --- | --- |
| Malonyl/acetyl transferase (MAT) | loads the acetyl primer and each malonyl unit |
| Ketoacyl synthase (KS) | **C**: joins the chain to the malonyl unit |
| Ketoreductase (KR) | **R**: C=O to CH–OH, using NADPH |
| Dehydratase (DH) | **D**: removes water, leaving a C=C |
| Enoyl reductase (ER) | **R**: C=C to CH₂–CH₂, using NADPH |
| Thioesterase (TE) | releases the finished chain |

ACP holds the chain by a thioester, the same sulfur link CoA uses.

### Turn one, drawn out

Start with the two pieces loaded. The **primer** is an acetyl group, parked on a cysteine of the KS domain. The **extender** is a malonyl group on ACP: an acetyl group carrying the extra carboxyl that acetyl CoA carboxylase (ACC) added.

> CH₃–CO–S–KS (acetyl, the primer)
>
> ⁻OOC–CH₂–CO–S–ACP (malonyl, the extender)

**Condensation.** KS joins them. The malonyl unit loses its extra carboxyl as CO₂, and the carbon that was next to it attacks the primer’s C=O. The result is a four-carbon chain on ACP, **acetoacetyl-ACP**:

> CH₃–CO–CH₂–CO–S–ACP + CO₂

**Predict.** A researcher gives a liver extract bicarbonate labeled with carbon-14, so the carbon that ACC adds is radioactive. After palmitate is made, where is the label?

> **Check your prediction.** Little to none of it is in the palmitate. The carbon ACC added leaves as CO₂ at condensation, in every turn: the classic proof that the CO₂ is a leaving group, not a building block.

Why pay for a carbon that leaves again? Joining two acetyl groups directly is uphill; losing CO₂ is strongly downhill, and coupling the two makes each join effectively one-way. The enzyme thiolase can make the direct join (cholesterol synthesis starts that way), but the equilibrium lies far toward the separate pieces, so each join would be easily undone and the chain would grow only as fast as later steps could pull it along.

The film’s two-NADPH bill now gets a chemical meaning. Condensation left a C=O on the **β-carbon**, two carbons along from the thioester carbon. Turning it into the CH₂ of a fatty acid takes four electrons: the carbon’s oxidation state falls from +2 to −2. Each NADPH delivers two electrons as a hydride, in two reductions separated by dehydration.

> **Reduction (KR, 1 NADPH):** CH₃–CH(OH)–CH₂–CO–S–ACP. The C=O becomes an alcohol.
>
> **Dehydration (DH):** CH₃–CH=CH–CO–S–ACP + H₂O. Water leaves, leaving a double bond.
>
> **Reduction (ER, 1 NADPH):** CH₃–CH₂–CH₂–CO–S–ACP. The double bond is saturated.

That is butyryl-ACP, four carbons and fully saturated, for one malonyl CoA (so one ATP, spent at ACC) and two NADPH. It moves back to the KS cysteine, a new malonyl unit loads onto ACP, and CRDR runs again. Each turn adds two carbons at the end nearest the sulfur, so the primer’s two carbons end up at the far, methyl end of the finished fatty acid.

### The bill for any chain

Because every turn has the same price, the whole cost follows from one count: how many turns?

> **Equation card: What a saturated fatty acid of n carbons costs to build, two carbons at a time, starting from acetyl CoA already in the cytoplasm. Fatty acid synthase itself builds up to 16 carbons.**
>
>     t = (n − 2) / 2, ATP = t, NADPH = 2 × t, acetyl CoA = t + 1
>
> In words: Subtract the primer's two carbons and halve what is left to get the number of turns. Each turn uses one malonyl CoA, which cost one ATP, and two NADPH. The primer adds one acetyl CoA that costs no ATP.
>
> - n: carbons in the finished saturated chain (an even number)
> - t: turns of the CRDR cycle, one per malonyl CoA added (count)
>
> Sanity check: For n = 4, one turn: 2 acetyl CoA, 1 ATP and 2 NADPH, matching turn one above.
> Try one number Stearate has 18 carbons. FAS stops at 16 and releases free palmitate. The cell reactivates it as palmitoyl CoA, for one more ATP (split to AMP), and an elongase on the endoplasmic reticulum adds the last two carbons at the same per-turn price. Turns: t = (18 − 2) / 2 = 8. ATP at ACC = 8, plus the one ATP to reactivate palmitate. NADPH = 2 × 8 = 16. Acetyl CoA = 8 + 1 = 9: one primer and eight that became malonyl CoA.

**Predict.** Use the box: how many acetyl CoA, ATP and NADPH does one palmitate (C16) take?

> **Check your prediction.** Seven turns: 8 acetyl CoA, 7 ATP, 14 NADPH. If you got 8 ATP, you charged the primer; it is used as acetyl CoA and never goes through ACC.

Everything in and out gives the full ledger:

> **8 acetyl CoA + 7 ATP + 14 NADPH + 7 HCO₃⁻ → palmitate + 7 CO₂ + 8 CoA + 14 NADP⁺ + 7 ADP + 7 Pᵢ + 6 H₂O** (protons omitted)

Every term is the rule again. Seven bicarbonates go in at ACC and seven CO₂ come out at condensation. Seven dehydrations release seven waters, and the thioesterase uses one to cut palmitate free, leaving six.

The map below lays out the route in eight steps. Its running total counts one pass through the cycle, −1 ATP, −2 NADPH and 0 CO₂ (the carboxyl goes in and comes back out), and leaves the delivery step out. Open every step and it multiplies by seven turns for palmitate.

*Interactive on the lesson page:* Fatty acid synthesis · delivery, commitment, and one CRDR turn tap a step to reveal it Running total 0 ATP0 NADPH0 CO₂ F

## Paying the bill: NADPH and delivery

The oxaloacetate left over from Stage 1 supplies part of the 14 NADPH. It is reduced to malate, which spends an NADH, and **malic enzyme** decarboxylates the malate to pyruvate, making NADPH. The pyruvate returns to the mitochondrion, where pyruvate carboxylase spends an ATP to make oxaloacetate again. In effect the shuttle turns an NADH into an NADPH for the price of an ATP. It runs once per acetyl group delivered, so it can supply up to 8 of the 14 NADPH. The pentose phosphate pathway and other routes supply the rest, and the real split varies by tissue.

*Figure (AI-generated illustration).* Citrate carries acetyl groups across the membrane; ATP-citrate lyase releases acetyl CoA in the cytoplasm, and the leftover oxaloacetate is reduced to malate, which malic enzyme decarboxylates to pyruvate, making NADPH on the way. Up to eight of the fourteen NADPH per palmitate can come from this shuttle. Image: citrate crosses the inner mitochondrial membrane into the cytoplasm, where ATP-citrate lyase releases acetyl CoA for fat synthesis and the leftover oxaloacetate is reduced to malate, which malic enzyme decarboxylates to pyruvate, making NADPH; the pyruvate returns to the matrix.

Delivery has an ATP price too. The ledger’s 7 ATP count only the building; delivering 8 acetyl groups as citrate costs 8 more at ATP-citrate lyase, and the rest depends on where the oxaloacetate goes:

| Oxaloacetate returns as | ATP-citrate lyase | Pyruvate carboxylase | ACC | Total ATP | NADPH from the shuttle |
| --- | --- | --- | --- | --- | --- |
| malate (no NADPH made) | 8 | 0 | 7 | 15 | 0 |
| pyruvate, via malic enzyme | 8 | 8 | 7 | 23 | 8 (for 8 NADH), leaving 6 for the pentose phosphate pathway |

## Regulation: malonyl CoA’s second job

Back to the rule: Acetyl CoA carboxylase (ACC) is where the cell decides whether to build. It is controlled three ways.

| Layer | Turns ACC on | Turns ACC off |
| --- | --- | --- |
| Phosphorylation | PP2A removes the phosphate | AMPK, the cell’s low-fuel sensor, phosphorylates ACC |
| Allostery | citrate (abundant fuel) makes ACC polymerize into active filaments | palmitoyl CoA (the end product) depolymerizes them, a clean feedback brake |
| Hormones | insulin | glucagon and epinephrine |

Over hours to days, insulin and glucose also switch on the ACC and fatty acid synthase (FAS) genes, through the transcription factors SREBP-1c and ChREBP.

Exercise switches AMP-activated protein kinase (AMPK) on, and so, in experiments, may the diabetes drug metformin; the mechanism is debated (Going deeper).

The second job is the elegant part: malonyl CoA blocks carnitine palmitoyltransferase I (CPT-I). Much of the malonyl CoA that does this blocking is made by a second form of the enzyme, ACC2, which sits on the mitochondrion; ACC1 mostly feeds synthesis. So when the cell is building fat, it also tends to close the door on oxidation. ACC1’s malonyl CoA is mainly a substrate; ACC2’s is mainly a signal.

*Figure (AI-generated illustration).* Malonyl CoA has two jobs. ACC1's malonyl CoA mostly feeds fatty acid synthase as a building block; much of the malonyl CoA that closes CPT-I is made by ACC2 on the mitochondrion, so a cell that is building fat tends not to burn it. Image: in the cytoplasm ACC1 makes malonyl CoA that feeds fatty acid synthase, while ACC2 on the mitochondrial outer membrane makes malonyl CoA that blocks the CPT-I gate, keeping long-chain fats out. Insulin and citrate turn ACC on; AMPK and glucagon turn it off. An inset contrasts the fed state (much malonyl CoA, gate closed) with fasting (little malonyl CoA, gate open, fat entering the mitochondrion).

Try the see-saw below: toggle fed and fasting, or drag ACC activity, and watch malonyl CoA turn synthesis on while it closes the CPT-I gate.

*Interactive on the lesson page:* Build or burn — malonyl CoA tips the balance

**Predict.** Glucagon arrives at a fasting liver. Follow it to ACC, then malonyl CoA, then CPT-I. What happens to fat burning?

> **Check your prediction.** It rises. Glucagon inhibits ACC, so less malonyl CoA is made. Synthesis slows for lack of its building block, and the block on CPT-I lifts, so fat enters the mitochondrion and is burned.

**Predict.** Now a harder case. Metabolizing ethanol floods the liver with NADH, turns AMPK down and turns the fat-making genes up. Predict what happens to ACC, malonyl CoA, CPT-I and beta-oxidation.

> **Check your prediction.** ACC runs harder, malonyl CoA rises and CPT-I closes. Beta-oxidation is hit twice: the gate is shut, and the high NADH-to-NAD⁺ ratio slows beta-oxidation directly. Less burning plus more building and storing produces the fatty liver of heavy drinking.

## After palmitate

Fatty acid synthase (FAS) mostly stops at sixteen carbons. Longer chains and double bonds are added afterward, mainly on the endoplasmic reticulum. Elongases extend the chain from malonyl CoA at the same per-turn price, as in the stearate example. Desaturases install double bonds by a different chemistry, using O₂, NADH and cytochrome b5. Mammals cannot put a double bond near the methyl end of the chain, so **linoleate and alpha-linolenate are essential fatty acids** you must eat. Downstream, arachidonate becomes the eicosanoids, the signals that aspirin blocks.

Next, in **chapter 29**, acetyl CoA and the fatty acids you just built go on to become storage fat, membrane phospholipids, and cholesterol, and cholesterol gets its own committed enzyme and its own famous drug.

## Going deeper

The main line does not depend on these details, except that the metformin section fills in the hedge on a drug named in the objectives.

### Biotin and the 1927 egg-white rats

Biotin, ACC’s CO₂ taxi, was found the hard way. In 1927, Margaret Boas in London fed rats raw egg white and watched them sicken. The culprit was avidin, an egg-white protein that binds biotin so tightly (a dissociation constant near 10 to the minus 15 molar) that it sequesters the vitamin and starves the enzymes that need it. That observation started the hunt that later identified biotin as the missing vitamin.

### Why two fats are essential

Counting from the carboxyl carbon, mammalian desaturases can place a double bond as far out as carbon 9, and no farther. The omega-6 and omega-3 positions are counted from the other, methyl end; in an 18-carbon chain they sit at carbons 12 and 15 from the carboxyl end, beyond carbon 9, so we cannot put double bonds there. Burr and Burr showed the consequence between 1929 and 1932: rats raised on fat-free diets sickened, linoleate cured them, and alpha-linolenate proved essential too.

### Aspirin and the platelet

Downstream of the essential fats, arachidonate becomes the eicosanoids: cyclooxygenase turns it into prostaglandins and thromboxanes. Aspirin works by permanently acetylating that enzyme. Platelets have no nucleus, so they make little or no replacement enzyme. That is why the effect of a single dose fades only gradually, over about a week, as new platelets replace the old ones.

### Metformin and AMPK

This is one proposed way metformin helps in diabetes: in experiments it inhibits Complex I, the AMP-to-ATP ratio climbs, and AMP-activated protein kinase (AMPK) switches on and phosphorylates acetyl CoA carboxylase (ACC). Malonyl CoA falls, so synthesis slows and carnitine palmitoyltransferase I (CPT-I) opens, and oxidation rises. How metformin lowers blood sugar is still debated, and much of that effect does not need AMPK. (Metformin descends from French lilac, also called goat’s rue, a folk remedy for high blood sugar.)

### One more tell: D and L

Beyond location, carrier and reducing power, a fourth difference: the first reduction (KR) makes D-3-hydroxyacyl intermediates, while mitochondrial beta-oxidation runs through the L-isomers. Same atoms, opposite stereochemistry: one more sign that synthesis and degradation are different machines.

### The ledger’s protons

The ledger is printed with protons omitted. Every carbon, nitrogen, oxygen, phosphorus and sulfur in it balances, and so does the water, but the protons are left out. Textbooks print different proton counts because they use different conventions for how charged each molecule is. The water count works because the oxygen that ends up on the phosphate comes from bicarbonate, not from water.

### Where the rules bend

- “Mainly” in the location rule matters. Very-long-chain fatty acids start their oxidation in peroxisomes, and a separate set of small enzymes inside the mitochondrion (mitochondrial fatty acid synthesis) makes an octanoyl group, the starting point for lipoic acid.
- Odd-chain fatty acids start from a three-carbon primer, propionyl CoA, instead of acetyl CoA.
- Not all malonyl CoA becomes fat: besides ACC2’s signalling pool, malonyl CoA decarboxylase can break it down.

---

From STEMhop (https://stemhop.com), free lessons by Christopher Deppmann. Lesson text © the authors. Practice questions: https://stemhop.com/lessons/fatty-acid-synthesis/
