Do we need to synthesise fatty acids?

It is obvious that bicarbonate (or carbon dioxide) is essential for fatty acid synthesis from acetyl CoA.

In the next set of experiments adipocytes were incubated with [14C]bicarbonate in the buffer and non-radioactive acetyl CoA, again together with insulin. The results were as follows (figures show dpm ± standard deviation for 5 x replicate experiments):

Although bicarbonate is required for fatty acid synthesis from acetyl CoA, no label from bicarbonate is incorporated into the fatty acids formed. This suggests that there is an intermediate that is carboxylated, but then loses the carbon dioxide again. The overall reaction can be drawn as:

Two enzymes that are required for fatty acid synthesis were isolated from adipocytes. When both together were incubated with [14C]acetyl CoA in a bicarbonate buffer no radioactive palmitate (C16:0) was formed unless NADPH was added to the incubation mixture. Further studies showed that 14 mol of NADPH were oxidised per mol of palmitate formed.

Looking at the overview of fatty acid synthesis shown above it is obvious that there must be reduction reactions if acetyl CoA is to form a saturated hydrocarbon chain. Presumably one mol of acetyl CoA forms the carboxyl group of palmitate, but the remaining 7 mol have to be reduced. The stoichiometry (14 mol of NADPH consumed per mol of palmitate formed) suggests that there are two reduction reactions for each acetyl CoA added.

The obvious experiment would be to incubate the enzymes with acetyl CoA and bicarbonate using [3H]NADPH, and measure the incorporation of label into the palmitate formed.

Neither enzyme alone catalyses any synthesis of palmitate or any other fatty acid. Studies of the smaller enzyme showed that it catalyses the carboxylation of acetyl CoA to malonyl CoA.

Further studies of this enzyme showed that it contains the vitamin biotin, bound to a lysine residue at the active site as biocytin, and that the first step in the reaction is formation of carboxybiocytin. This then acts as the donor of the carboxyl group to form malonyl CoA.

This was the first demonstration of a metabolic function of biotin; it is now know to be required for a small number of other carboxylation reactions, including the carboxylation of pyruvate to oxaloacetate in gluconeogenesis (see the exercise on breathless after sprinting).

The incubations with [14C-3]malonyl CoA confirm the earlier result that although bicarbonate is required to form malonyl CoA, the carbon that was added (carbon-3 of malonyl CoA) is lost again and is not incorporated into the palmitate formed.

Enzyme 2 obviously requires both acetyl CoA and malonyl CoA to synthesise palmitate. However, only about one seventh as much acetyl CoA as malonyl CoA is incorporated into palmitate. This means that we can draw up the overall reaction of fatty acid synthesis as:

The palmitate formed in the incubations with [14C-U]acetyl CoA + non-radioactive malonyl CoA and [14C-1,2]malonyl CoA + non-radioactive acetyl CoA was subjected to chemical degradation to identify which carbon atoms were labelled in each set of incubations. The results were as shown on the right.

It seems likely that carbons 14 and 15 of palmitate arise from acetyl CoA, while the other carbons arise from malonyl CoA, but lose the carbon that was added from bicarbonate during the reaction.

If none of the intermediate products can be isolated, this suggests that fatty acid synthase is either a multi-enzyme complex, or a single large protein with multiple active sites, and that intermediates are chanelled directly from one active site to the next, until the final product, palmitate, is released.

In bacteria and plants a number of intermediates of fatty acid synthesis can be isolated, and separate enzymes catalysing each step of the pathway can be identified. The steps involved in fatty acid synthesis were therefore investigated mainly using the avocado, since this fruit synthesises and contains a considerable amount of triacylglycerol.

The first observation was that although the starting substrates are acetyl CoA and malonyl CoA, both the acetyl and malonyl moieties are transferred onto acyl carrier proteins (ACP). These are flexible proteins that have a prosthetic group derived from the vitamin pantothenic acid and the amine cysteamine, derived from the amino acid cysteine.

Like CoA, this prosthetic group forms a thio-ester with fatty acids.

The first reaction of fatty acid synthesis is condensation between the malonyl moiety of malonyl ACP and the acetyl moiety of acetyl ACP, with loss of carbon dioxide:

We know that there are 14 reduction reactions in the synthesis of palmitate from acetyl CoA, and this suggests that there are two reduction reactions for each acetyl moiety added from malonyl CoA. Therefore, the most likely next reaction is reduction of the oxo-group of acetoacetyl CoA to a hydroxyl group.

Incubation of acetoacetyl ACP with [3H]NADPH will show incorporation of the [3H] label into hydroxybutyrate.

Incubations using [18O]malonyl CoA showed the release of [18O]labelled water. Incubations using [3H]NADPH showed the release of [3H]water.

An obvious source of water would be dehydration, removing the hydroxyl group on carbon 3 formed by reduction in the previous reaction and hydrogen from carbon 2, forming a carbon-carbon double bond:

We know there are two reduction reactions involved, so it is likely that the next reaction is reduction of the carbon-carbon double bond to yield a saturated fatty acyl ACP - in this case butyryl ACP.

We have now synthesised a 4-carbon saturated fatty acid esterified to acyl carrier protein. This can now condense with another molecule of malonyl CoA and undergo the same sequence of reactions. This continues until the product is palmitoyl ACP, when the palmitoyl moiety is cleaved off to form palmitoyl CoA, leaving the acyl carrier protein free to bind acetyl group to start over again.

Click here to download a printable version of the pathway

The structure of the mammalian fatty acid synthase has been determined. It consists of two protein chains, each of which contains all of the enzymes for palmitate synthesis from malonyl CoA (structure from Mayer et al Science 311: 1258-62 2006)

The growing fatty acid chain is carried by the acyl carrier protein (the red star in the diagram) to each of the active sites in turn.

Because the growing fatty acyl ACP has to travel to each of the active sites of the enzyme in turn, it is not possible to "skip" a reaction. The enoyl ACP must be reduced to the fatty acyl ACP before the sequence of reactions can begin again with the addition of another malonyl ACP.

Fatty acid synthase only forms palmitate - how are longer chain fatty acids and unsaturated fatty acids synthesised?

We have seen in that the only product of the cytosolic fatty acid synthase is palmitate (C16:0), and even though there is an unsaturated fatty acyl intermediate in fatty acid synthesis, all the reactions of the enzyme must be completed; it is not possible to "skip" one reaction to form an unsaturated fatty acid. This raises the question of how longer chain fatty acids are formed, and how unsaturated fatty acids can be synthesised.

The smooth endoplasmic reticulum also has a fatty acid synthase, which can accept palmitoyl CoA and, following the same sequence of reactions as the cytosolic enzyme, elongate it to produce longer chain fatty acids.

Unsaturated fatty acids are formed by desaturases in the endoplasmic reticulum.

Oleyl CoA (C18:1 n-9) is formed from stearyl CoA (C18:0), which itself is synthesised from palmitoyl CoA by one cycle of reactions catalysed by the endoplasmic reticulum fatty acid synthase. Reduced cytochrome b5 reduces oxygen to water by removing two hydrogens from the fatty acyl CoA to form a carbon-carbon double bond. The oxidised cytochrome is reduced by FADH2, which in turn is reduced by NADH.

We saw in the exercise fats and oils - are all fats the same? that human beings have enzymes that can introduce carbon-carbon double bonds between an existing double bond and the carboxyl group of a fatty acid, but not between an existing double bond and the methyl group. The only exception is the delta-9 desaturase shown above, which can introduce a double bond at position n-9 into the saturated fatty acid stearate. This means that there is a requirement for a source of n-3 and n-6 fatty acids in the diet, although as long as one of each is available, longer chain and more unsaturated fatty acids can be synthesised.

Oleic acid (C18:1 n-3) can be synthesised from stearic acid, as shown above; it can then undergo elongation and further desaturation to form C20:3 n-9. Further elongation and desaturation is possible, but does not occur to any significant extent.

Linoleic acid (C18:3 n-6) is the precursor of the n-6 family of fatty acids; it is a dietary essential.

Alpha-linolenic acid (C18:3 n-3) is the precursor of the n-3 family of fatty acids; it is also a dietary essential.

Both linoleic and alpha-linolenic acids undergo elongation and desaturation; C20:3 n-6 and C20:4 n-3 are the precursors for synthesis of prostaglandins and other eicosanoids (so-called because they are derived from C20 fatty acids - eicosa means 20 in Greek). Further elongation and desaturation leads to the synthesis of very long chain poly-unsaturated fatty acids that have other functions, especially in membranes.

Synthesis of C22:5 n-3 and n-6 requires elongation of C22:4 to C24:4, followed by desaturation to C24:5 then beta oxidation to reduce the chain length to C22, because there is no delta-5 desaturase.

Because the same enzymes act to elongate and desaturate both the n-3 and n-6 families of fatty acids, they compete with each other, and the balance between dietary intakes of n-3 and n-6 fatty acids is important. The eicosanoids formed from C20:4 n-3 and C20:4 n-6 have different, and in some cases opposing, physiological actions.

Key points from this exercise:

• The first committed step for fatty acid biosynthesis is carboxylation of acetyl CoA to malonyl CoA; of the 16 carbons in palmitate, only carbons 14 and 15 arise from acetyl CoA; the remainder come from malonyl CoA.
• The coenzyme for acetyl CoA carboxylase is the vitamin biotin.
• The carboxyl group that was added to acetyl CoA to form malonyl CoA is lost as carbon dioxide in the next reaction of fatty acid synthesis.
• Mammalian fatty acid synthase is a large protein with multiple active sites; the growing fatty acid chain is chanelled from one active site to the next by the flexible acyl carrier protein. The prosthetic group of the acyl carrier protein is derived from the vitamin pantothenic acid and the amine cysteamine. Like CoA, this prosthetic group forms a thio-ester with the carboxyl group of fatty acids and derivatives.
• The first reaction of fatty acid synthetase is condensation between malonyl ACP and acetyl CoA, to form acetoacetyl ACP.
• Acetoacetyl ACP is reduced to hydroxybutyryl ACP, with NADPH as the reductant.
• Hydroxybutyryl CoA undergoes dehydration to yield a carbon-carbon double bond between carbons 2-3.
• The unsaturated fatty acyl ACP is reduced to a saturated fatty acyl ACP (butyryl ACP in the first cycle of reactions), again with NADPH as the reductant.
• The reaction cycle continues with the addition of a further malonyl ACP to the growing fatty acyl ACP until the final product is palmitoyl ACP (C16:0), which is transferred onto CoA, forming palmitoyl CoA and leaving the ACP free to bind another acetyl group and begin the sequence of reactions again.
• Although there is intermediate formation of a 2-3 unsaturated fatty acyl ACP intermediate, the complete sequence of reactions must occur, so unsaturated fatty acids cannot by synthesised by the cytosolic fatty acid synthase.
• Palmitoyl CoA formed by the cytosolic fatty acid synthase can undergo chain elongation and desaturation in the endoplasmic reticulum to synthesise long-chain saturated and polyunsaturated fatty acids.
• Mammalian enzymes can introduce a carbon-carbon double bond between an existing double bond and the carboxyl group of the fatty acid, but not between an existing double bond and the methyl group. This means that there is a requirement for a dietary source of n-3 and n-6 polyunsaturated fatty acids. There is a mammalian delta-9 desaturase that can synthesise oleate (C18:1 n-9) from stearate (C18:0).
• The same desaturases and elongases act on n-3, n-6 and n-9 fatty acids, so there is competition between them. This means that the balance of n-3 : n-6 polyunsaturated fatty acids in the diet is important.
• The C20 n-3 and n-6 polyunsaturated fatty acids are precursors for the prostaglandins and other eicosanoids. The two families (n-3 and n-6) have different physiological actions, so that again the balance of n-3 : n-6 polyunsaturated fatty acids in the diet is important.