Chapter 25

Lipid Biosynthesis

Southern elephant seal, Mirounga leonina. (Gerald Lacz/Peter Arnold, Inc.)

We turn now to the biosynthesis of lipid structures. We begin with a discussion of the biosynthesis of fatty acids, stressing the basic pathways, additional means of elongation, mechanisms for the introduction of double bonds, and regulation of fatty acid synthesis. Sections then follow on the biosynthesis of glycerophospholipids, sphingolipids, eicosanoids, and cholesterol. The transport of lipids through the body in lipoprotein complexes is described, and the chapter closes with discussions of the biosynthesis of bile salts and steroid hormones.

25.1 × The Fatty Acid Biosynthesis and Degradation Pathways Are Different

We have already seen several cases in which the synthesis of a class of biomolecules is conducted differently from degradation (glycolysis versus gluconeogenesis and glycogen or starch breakdown versus polysaccharide synthesis, for example). Likewise, the synthesis of fatty acids and other lipid components is different from their degradation. Fatty acid synthesis involves a set of reactions that follow a strategy different in several ways from the corresponding degradative process:

1. Intermediates in fatty acid synthesis are linked covalently to the sulfhydryl groups of special proteins, the acyl carrier proteins. In contrast, fatty acid breakdown intermediates are bound to the -SH group of coenzyme A.

2. Fatty acid synthesis occurs in the cytosol, whereas fatty acid degradation takes place in mitochondria.

3. In animals, the enzymes of fatty acid synthesis are components of one long polypeptide chain, the fatty acid synthase, whereas no similar association exists for the degradative enzymes. (Plants and bacteria employ separate enzymes to carry out the biosynthetic reactions.)

4. The coenzyme for the oxidation-reduction reactions of fatty acid synthesis is NADP⁺/NADPH, whereas degradation involves the NAD⁺/NADH couple.

Formation of Malonyl-CoA Activates Acetate Units for Fatty Acid Synthesis

The design strategy for fatty acid synthesis is this:

a. Fatty acid chains are constructed by the addition of two-carbon units derived from acetyl-CoA.

b. The acetate units are activated by formation of malonyl-CoA (at the expense of ATP).

c. The addition of two-carbon units to the growing chain is driven by decarboxylation of malonyl-CoA.

d. The elongation reactions are repeated until the growing chain reaches 16 carbons in length (palmitic acid).

e. Other enzymes then add double bonds and additional carbon units to the chain.

Fatty Acid Biosynthesis Depends on the Reductive Power of NADPH

The net reaction for the formation of palmitate from acetyl-CoA is

Acetyl-CoA + 7 malonyl-CoA² + 14 NADPH + 14 H⁺ ®
palmitoyl-CoA + 7 HCO₃² + 7 CoASH + 14 NADP⁺  (25.1)

(Levels of free fatty acids are very low in the typical cell. The palmitate made in this process is rapidly converted to CoA esters in preparation for the formation of triacylglycerols and phospholipids.)

Providing Cytosolic Acetyl-CoA and Reducing Power for Fatty Acid Synthesis

Eukaryotic cells face a dilemma in providing suitable amounts of substrate for fatty acid synthesis. Sufficient quantities of acetyl-CoA, malonyl-CoA, and NADPH must be generated in the cytosol for fatty acid synthesis. Malonyl-CoA is made by carboxylation of acetyl-CoA, so the problem reduces to generating sufficient acetyl-CoA and NADPH.

 There are three principal sources of acetyl-CoA (Figure 25.1):

1. Amino acid degradation produces cytosolic acetyl-CoA.

2. Fatty acid oxidation produces mitochondrial acetyl-CoA.

3. Glycolysis yields cytosolic pyruvate, which (after transport into the mitochondria) is converted to acetyl-CoA by pyruvate dehydrogenase.

The acetyl-CoA derived from amino acid degradation is normally insufficient for fatty acid biosynthesis, and the acetyl-CoA produced by pyruvate dehydrogenase and by fatty acid oxidation cannot cross the mitochondrial membrane to participate directly in fatty acid synthesis. Instead, acetyl-CoA is linked with oxaloacetate to form citrate, which is transported from the mitochondrial matrix to the cytosol (Figure 25.1). Here it can be converted back into acetyl- CoA and oxaloacetate by ATP-citrate lyase. In this manner, mitochondrial acetyl-CoA becomes the substrate for cytosolic fatty acid synthesis. (Oxaloacetate returns to the mitochondria in the form of either pyruvate or malate, which is then reconverted to acetyl-CoA and oxaloacetate, respectively.)

Figure 25.1 · The citrate-malate-pyruvate shuttle provides cytosolic acetate units and reducing equivalents (electrons) for fatty acid synthesis. The shuttle collects carbon substrates, primarily from glycolysis but also from fatty acid oxidation and amino acid catabolism. Most of the reducing equivalents are glycolytic in origin. Pathways that provide carbon for fatty acid synthesis are shown in blue; pathways that supply electrons for fatty acid synthesis are shown in red.

    NADPH can be produced in the pentose phosphate pathway as well as by malic enzyme (Figure 25.1). Reducing equivalents (electrons) derived from glycolysis in the form of NADH can be transformed into NADPH by the combined action of malate dehydrogenase and malic enzyme:

 Oxaloacetate + NADH + H⁺ ® malate + NAD⁺

 Malate + NADP⁺ ® pyruvate + CO₂ + NADPH + H⁺

     How many of the 14 NADPH needed to form one palmitate (Eq. 25.1) can be made in this way? The answer depends on the status of malate. Every citrate entering the cytosol produces one acetyl-CoA and one malate (Figure 25.1). Every malate oxidized by malic enzyme produces one NADPH, at the expense of a decarboxylation to pyruvate. Thus, when malate is oxidized, one NADPH is produced for every acetyl-CoA. Conversion of 8 acetyl-CoA units to one palmitate would then be accompanied by production of 8 NADPH. (The other 6 NADPH required [Eq. 25.1] would be provided by the pentose phosphate pathway.) On the other hand, for every malate returned to the mitochondria, one NADPH fewer is produced.

Acetate Units Are Committed to Fatty Acid Synthesis by Formation of Malonyl-CoA

Rittenberg and Bloch showed in the late 1940s that acetate units are the building blocks of fatty acids. Their work, together with the discovery by Salih Wakil that bicarbonate is required for fatty acid biosynthesis, eventually made clear that this pathway involves synthesis of malonyl-CoA. The carboxylation of acetyl-CoA to form malonyl-CoA is essentially irreversible and is the committed step in the synthesis of fatty acids (Figure 25.2). The reaction is catalyzed by acetyl-CoA carboxylase, which contains a biotin prosthetic group. This carboxylase is the only enzyme of fatty acid synthesis in animals that is not part of the multi-enzyme complex called fatty acid synthase.

Figure 25.2 · (a) The acetyl-CoA carboxylase reaction produces malonyl-CoA for fatty acid synthesis. (b) A mechanism for the acetyl-CoA carboxylase reaction. Bicarbonate is activated for carboxylation reactions by formation of N-carboxybiotin. ATP drives the reaction forward, with transient formation of a carbonylphosphate intermediate (Step 1). In a typical biotin-dependent reaction, nucleophilic attack by the acetyl- CoA carbanion on the carboxyl carbon of N-carboxybiotin - a transcarboxylation - yields the carboxylated product (Step 2).

Acetyl-CoA Carboxylase Is Biotin-Dependent and Displays Ping-Pong Kinetics

The biotin prosthetic group of acetyl-CoA carboxylase is covalently linked to the e-amino group of an active-site lysine in a manner similar to pyruvate carboxylase (Figure 23.3). The reaction mechanism is also analogous to that of pyruvate carboxylase (Figure 23.4): ATP-driven carboxylation of biotin is followed by transfer of the activated CO₂ to acetyl-CoA to form malonyl-CoA. The enzyme from Escherichia coli has three subunits: (1) a biotin carboxyl carrier protein (a dimer of 22.5-kD subunits); (2) biotin carboxylase (a dimer of 51-kD subunits), which adds CO₂ to the prosthetic group; and (3) transcarboxylase (an a₂b₂ tetramer with 30-kD and 35-kD subunits), which transfers the activated CO₂ unit to acetyl-CoA. The long, flexible biotin-lysine chain (biocytin) enables the activated carboxyl group to be carried between the biotin carboxylase and the transcarboxylase (Figure 25.3).

Figure 25.3 · In the acetyl-CoA carboxylase reaction, the biotin ring, on its flexible tether, acquires carboxyl groups from carbonylphosphate on the carboxylase subunit and transfers them to acyl-CoA molecules on the transcarboxylase subunits.

Acetyl-CoA Carboxylase in Animals Is a Multifunctional Protein

In animals, acetyl-CoA carboxylase (ACC) is a filamentous polymer (4 to 8 x 10⁶ D) composed of 230-kD protomers. Each of these subunits contains the biotin carboxyl carrier moiety, biotin carboxylase, and transcarboxylase activities, as well as allosteric regulatory sites. Animal ACC is thus a multifunctional protein. The polymeric form is active, but the 230-kD protomers are inactive. The activity of ACC is thus dependent upon the position of the equilibrium between these two forms:

Inactive protomers 88zy88 active polymer

 Because this enzyme catalyzes the committed step in fatty acid biosynthesis, it is carefully regulated. Palmitoyl-CoA, the final product of fatty acid biosynthesis, shifts the equilibrium toward the inactive protomers, whereas citrate, an important allosteric activator of this enzyme, shifts the equilibrium toward the active polymeric form of the enzyme. Acetyl-CoA carboxylase shows the kinetic behavior of a Monod-Wyman-Changeux V-system allosteric enzyme (Chapter 15).

Figure 25.4 · Models of the acetyl-CoA carboxylase polypeptide, with phosphorylation sites indicated, along with the protein kinases responsible. Phosphorylation at Ser¹²⁰⁰ is primarily responsible for decreasing the affinity for citrate.

Phosphorylation of ACC Modulates Activation by Citrate and Inhibition by Palmitoyl-CoA

The regulatory effects of citrate and palmitoyl-CoA are dependent on the phosphor-ylation state of acetyl-CoA carboxylase. The animal enzyme is phosphorylated at 8 to 10 sites on each enzyme subunit (Figure 25.4). Some of these sites are regulatory, whereas others are "silent" and have no effect on enzyme activity. Unphosphorylated acetyl-CoA carboxylase binds citrate with high affinity and thus is active at very low citrate concentrations (Figure 25.5). Phosphorylation of the regulatory sites decreases the affinity of the enzyme for citrate, and in this case high levels of citrate are required to activate the carboxylase. The inhibition by fatty acyl-CoAs operates in a imilar but opposite manner. Thus, low levels of fatty acyl-CoA inhibit the phosphorylated carboxylase, but the dephosphoenzyme is inhibited only by high levels of fatty acyl-CoA. Specific phosphatases act to dephosphorylate ACC, thereby increasing the sensitivity to citrate.

Figure 25.5 · The activity of acetyl-CoA carboxylase is modulated by phosphorylation and dephosphorylation. The dephospho form of the enzyme is activated by low [citrate] and inhibited only by high levels of fatty acyl-CoA. In contrast, the phosphorylated form of the enzyme is activated only by high levels of citrate, but is very sensitive to inhibition by fatty acyl-CoA.

Acyl Carrier Proteins Carry the Intermediates in Fatty Acid Synthesis

The basic building blocks of fatty acid synthesis are acetyl and malonyl groups, but they are not transferred directly from CoA to the growing fatty acid chain. Rather, they are first passed to acyl carrier protein (or simply ACP), discovered by P. Roy Vagelos. This protein consists (in E. coli) of a single polypeptide chain of 77 residues to which is attached (on a serine residue) a phosphopante-theine group, the same group that forms the "business end" of coenzyme A. Thus, acyl carrier protein is a somewhat larger version of coenzyme A, specialized for use in fatty acid biosynthesis (Figure 25.6).

Figure 25.6 · Fatty acids are conjugated both to coenzyme A and to acyl carrier protein through the sulfhydryl of phosphopantetheine prosthetic groups.

 The enzymes that catalyze formation of acetyl-ACP and malonyl-ACP and the subsequent reactions of fatty acid synthesis are organized quite differently in different organisms. We first discuss fatty acid biosynthesis in bacteria and plants, where the various reactions are catalyzed by separate, independent proteins. Then we discuss the animal version of fatty acid biosynthesis, which involves a single multienzyme complex called fatty acid synthase.

Fatty Acid Synthesis in Bacteria and Plants

The individual steps in the elongation of the fatty acid chain are quite similar in bacteria, fungi, plants, and animals. The ease of purification of the separate enzymes from bacteria and plants made it possible in the beginning to sort out each step in the pathway, and then by extension to see the pattern of biosynthesis in animals. The reactions are summarized in Figure 25.7. The elongation reactions begin with the formation of acetyl-ACP and malonyl-ACP, which are formed by acetyl transacylase (acetyl transferase) and malonyl transacylase (malonyl transferase), respectively. The acetyl transacylase enzyme is not highly specific — it can transfer other acyl groups, such as the propionyl group, but at much lower rates. (Fatty acids with odd numbers of carbons are made beginning with a propionyl group transfer by this enzyme.) Malonyl transacylase, on the other hand, is highly specific.

Figure 25.7 · The pathway of palmitate synthesis from acetyl-CoA and malonyl-CoA. Acetyl and malonyl building blocks are introduced as acyl carrier protein conjugates. Decarboxylation drives the b-ketoacyl-ACP synthase and results in the addition of two-carbon units to the growing chain. Concentrations of free fatty acids are extremely low in most cells, and newly synthesized fatty acids exist primarily as acyl-CoA esters.

Decarboxylation Drives the Condensation of Acetyl-CoA and Malonyl-CoA

Another transacylase reaction transfers the acetyl group from ACP to b-keto-acyl-ACP synthase (KSase), also known as acyl-malonyl-ACP condensing enzyme. The first actual elongation reaction involves the condensation of acetyl-ACP and malonyl-ACP by the b-ketoacyl-ACP synthase to form acetoacetyl-ACP (Figure 25.7). One might ask at this point: Why is the three-carbon malonyl group used here as a two-carbon donor? The answer is that this is yet another example of a decarboxylation driving a desired but otherwise thermodynamically unfavorable reaction. The decarboxylation that accompanies the reaction with malonyl-ACP drives the synthesis of acetoacetyl-ACP. Note that hydrolysis of ATP drove the carboxylation of acetyl-CoA to form malonyl-ACP, so, indirectly, ATP is responsible for the condensation reaction to form acetoacetyl-ACP. Malonyl-CoA can be viewed as a form of stored energy for driving fatty acid synthesis.

 It is also worth noting that the carbon of the carboxyl group that was added to drive this reaction is the one removed by the condensing enzyme. Thus, all the carbons of acetoacetyl-ACP (and of the fatty acids to be made) are derived from acetate units of acetyl-CoA.

Reduction of the b-Carbonyl Group Follows a Now-Familiar Route

The next three steps reduction of the b-carbonyl group to form a b-alcohol, followed by dehydration and reduction to saturate the chain (Figure 25.7) look very similar to the fatty acid degradation pathway in reverse. However, there are two crucial differences between fatty acid biosynthesis and fatty acid oxidation (besides the fact that different enzymes are involved): First, the alcohol formed in the first step has the d configuration rather than the l form seen in catabolism, and, second, the reducing coenzyme is NADPH, although NAD⁺ and FAD are the oxidants in the catabolic pathway.

 The net result of this biosynthetic cycle is the synthesis of a four-carbon unit, a butyryl group, from two smaller building blocks. In the next cycle of the process, this butyryl-ACP condenses with another malonyl-ACP to make a six-carbon b-ketoacyl-ACP and CO₂. Subsequent reduction to a b-alcohol, dehydration, and another reduction yield a six-carbon saturated acyl-ACP. This cycle continues with the net addition of a two-carbon unit in each turn until the chain is 16 carbons long (Figure 25.7). The b-ketoacyl-ACP synthase cannot accommodate larger substrates, so the reaction cycle ends with a 16-carbon chain. Hydrolysis of the C₁₆-acyl-ACP yields a palmitic acid and the free ACP.

 In the end, seven malonyl-CoA molecules and one acetyl-CoA yield a palmitate (shown here as palmitoyl-CoA):

Acetyl-CoA + 7 malonyl-CoA² + 14 NADPH + 14 H⁺ ®
palmitoyl-CoA + 7 HCO₃² + 14 NADP⁺ + 7 CoASH

The formation of seven malonyl-CoA molecules requires

7 Acetyl-CoA + 7 HCO₃² + 7 ATP⁴² ®
7 malonyl-CoA² + 7 ADP³² + 7 P_i²² + 7 H¹

Thus, the overall reaction of acetyl-CoA to yield palmitic acid is

palmitoyl-CoA + 14 NADP⁺ + 7 CoASH + 7 ADP³² + 7P_i²²

Note: These equations are stoichiometric and are charge balanced. See Problem 1 at the end of the chapter for practice in balancing these equations.

Figure 25.8 · In yeast, the functional groups and enzyme activities required for fatty acid synthesis are distributed between a and Gr-beta subunits.

Fatty Acid Synthesis in Eukaryotes Occurs on a Multienzyme Complex

Figure 25.9 · Fatty acid synthase in animals contains all the functional groups and enzyme activities on a single multifunctional subunit. The active enzyme is a head-to-tail dimer of identical subunits. (Adapted from Wakil, S. J., Stoops, J. K., and Joshi, V. C., 1983. Annual Review of Biochemistry 52:556.)

In contrast to bacterial and plant systems, the reactions of fatty acid synthesis beyond the acetyl-CoA carboxylase in animal systems are carried out by a special multienzyme complex called fatty acid synthase (FAS). In yeast, this 2.4 x 10⁶ D complex contains two different peptide chains, an a subunit of 213 kD and a b subunit of 203 kD, arranged in an a₆b₆ dodecamer. The separate enzyme activities associated with each chain are shown in Figure 25.8. In animal systems, FAS is a dimer of identical 250-kD multifunctional polypeptides. Studies of the action of proteolytic enzymes on this polypeptide have led to a model involving three separate domains joined by flexible connecting sequences (Figure 25.9). The first domain is responsible for the binding of acetyl and malonyl building blocks and for the condensation of these units. This domain includes the acetyl transferase, the malonyl transferase, and the acyl-malonyl-ACP condensing enzyme (the b-ketoacyl synthase). The second domain is primarily responsible for the reduction of the intermediate synthesized in domain 1, and contains the acyl carrier protein, the b-ketoacyl reductase, the dehydratase, and the enoyl-ACP reductase. The third domain contains the thioesterase that liberates the product palmitate when the growing acyl chain reaches its limit length of 16 carbons. The close association of activities in this complex permits efficient exposure of intermediates to one active site and then the next. The presence of all these activities on a single polypeptide ensures that the cell will simultaneously synthesize all the enzymes needed for fatty acid synthesis.

The Mechanism of Fatty Acid Synthase

Figure 25.10 · Acetyl units are covalently linked to a serine residue at the active site of the acetyl transferase in eukaryotes. A similar reaction links malonyl units to the malonyl transferase.

The first domain of one subunit of the fatty acid synthase interacts with the second and third domains of the other subunit; that is, the subunits are arranged in a head-to-tail fashion (Figure 25.9). The first step in the fatty acid synthase reaction is the formation of an acetyl-O-enzyme intermediate between the acetyl group of an acetyl-CoA and an active-site serine of the acetyl transferase (Figure 25.10). In a similar manner, a malonyl-O-enzyme intermediate is formed between malonyl-CoA and a serine residue of the malonyl transferase. The acetyl group on the acetyl transferase is then transferred to the -SH group of the acyl carrier protein, as shown in Figure 25.11. The next step is the transfer of the acetyl group to the b-ketoacyl-ACP synthase, or condensing enzyme. This frees the acyl carrier protein to acquire the malonyl group from the malonyl transferase. The next step is the condensation reaction, in which decarboxylation facilitates the concerted attack of the remaining two-carbon unit of the acyl carrier protein at the carbonyl carbon of the acetate group on the condensing enzyme. Note that decarboxylation forms a transient, highly nucleo-philic carbanion which can attack the acetate group.

Figure 25.11 · The mechanism of the fatty acyl synthase reaction in eukaryotes. (1) Acetyl and malonyl groups are loaded onto acetyl transferase and malonyl transferase, respectively. (2) The acetate unit that forms the base of the nascent chain is transferred first to the acyl carrier protein domain and (3) then to the b-ketoacyl synthase. (4) Attack by ACP on the carbonyl carbon of a malonyl unit on malonyl transferase forms malonyl-ACP. (5) Decarboxylation leaves a reactive, transient carbanion that can attack the carbonyl carbon of the acetyl group on the b-ketoacyl synthase. (6) Reduction of the keto group, dehydration, and saturation of the resulting double bond follow, leaving an acyl group on ACP, and steps 3 through 6 repeat to lengthen the nascent chain.
    The next three steps reduction of the carbonyl to an alcohol, dehydration to yield a trans-a,b double bond, and reduction to yield a saturated chain — are identical to those occurring in bacteria and plants (Figure 25.7) and resemble the reverse of the reactions of fatty acid oxidation (and the conversion of succinate to oxaloacetate in the TCA cycle). This synthetic cycle now repeats until the growing chain is 16 carbons long. It is then released by the thioesterase domain on the synthase. The amino acid sequence of the thioesterase domain is homologous with serine proteases; the enzyme has an active-site serine that carries out nucleophilic attack on the carbonyl carbon of the fatty acyl thioester to be cleaved.

Further Processing of C₁₆ Fatty Acids

Additional Elongation

As seen already, palmitate is the primary product of the fatty acid synthase. Cells synthesize many other fatty acids. Shorter chains are easily made if the chain is released before reaching 16 carbons in length. Longer chains are made through special elongation reactions, which occur both in the mitochondria and at the surface of the endoplasmic reticulum. The ER reactions are actually quite similar to those we have just discussed: addition of two-carbon units at the carboxyl end of the chain by means of oxidative decarboxylations involving malonyl-CoA. As was the case for the fatty acid synthase, this decarboxylation provides the thermodynamic driving force for the condensation reaction. The mitochondrial reactions involve addition (and subsequent reduction) of acetyl units. These reactions (Figure 25.12) are essentially a reversal of fatty acid oxidation, with the exception that NADPH is utilized in the saturation of the double bond, instead of FADH₂.

Figure 25.12 · Elongation of fatty acids in mitochondria is initiated by the thiolase reaction. The b-ketoacyl intermediate thus formed undergoes the same three reactions (in reverse order) that are the basis of b-oxidation of fatty acids. Reduction of the b-keto group is followed by dehydration to form a double bond. Reduction of the double bond yields a fatty acyl-CoA that is elongated by two carbons. Note that the reducing coenzyme for the second step is NADH, whereas the reductant for the fourth step is NADPH.

Introduction of a Single cis Double Bond

Both prokaryotes and eukaryotes are capable of introducing a single cis double bond in a newly synthesized fatty acid. Bacteria such as E. coli carry out this process in an O₂-independent pathway, whereas eukaryotes have adopted an O₂-dependent pathway. There is a fundamental chemical difference between the two. The O₂-dependent reaction can occur anywhere in the fatty acid chain, with no (additional) need to activate the desired bond toward dehydrogenation. However, in the absence of O₂, some other means must be found to activate the bond in question. Thus, in the bacterial reaction, dehydrogenation occurs while the bond of interest is still near the b-carbonyl or b-hydroxy group and the thioester group at the end of the chain.

    In E. coli, the biosynthesis of a monounsaturated fatty acid begins with four normal cycles of elongation to form a 10-carbon intermediate, b-hydroxydecanoyl-ACP (Figure 25.13). At this point, b-hydroxydecanoyl thioester dehydrase forms a double bond b,g to the thioester and in the cis configuration. This is followed by three rounds of the normal elongation reactions to form palmitoleoyl-ACP. Elongation may terminate at this point or may be followed by additional biosynthetic events. The principal unsaturated fatty acid in E. coli, cis-vaccenic acid, is formed by an additional elongation step, using palmitoleoyl-ACP as a substrate.

Figure 25.13 · Double bonds are introduced into the growing fatty acid chain in E. coli by specific dehydrases. Palmitoleoyl-ACP is synthesized by a sequence of reactions involving four rounds of chain elongation, followed by double bond insertion by b-hydroxydecanoyl thioester dehydrase and three additional elongation steps. Another elongation cycle produces cis-vaccenic acid.

Unsaturation Reactions Occur in Eukaryotes in the Middle of an Aliphatic Chain

The addition of double bonds to fatty acids in eukaryotes does not occur until the fatty acyl chain has reached its full length (usually 16 to 18 carbons). Dehydrogenation of stearoyl-CoA occurs in the middle of the chain despite the absence of any useful functional group on the chain to facilitate activation:

 CH₃O(CH₂)₁₆CO-SCoA ® CH₃O(CH₂)₇CH = CH(CH₂)₇CO-SCoA

This impressive reaction is catalyzed by stearoyl-CoA desaturase, a 53-kD enzyme containing a nonheme iron center. NADH and oxygen (O₂) are required, as are two other proteins: cytochrome b₅ reductase (a 43-kD flavoprotein) and cytochrome b₅ (16.7 kD). All three proteins are associated with the endoplasmic reticulum membrane. Cytochrome b₅ reductase transfers a pair of electrons from NADH through FAD to cytochrome b₅ (Figure 25.14). Oxidation of reduced cytochrome b₅ is coupled to reduction of nonheme Fe³⁺ to Fe²⁺ in the desaturase. The Fe³⁺ accepts a pair of electrons (one at a time in a cycle) from cytochrome b₅ and creates a cis double bond at the 9,10-position of the stearoyl-CoA substrate. O₂ is the terminal electron acceptor in this fatty acyl desaturation cycle. Note that two water molecules are made, which means that four electrons are transferred overall. Two of these come through the reaction sequence from NADH, and two come from the fatty acyl substrate that is being dehydrogenated.

Figure 25.14 · The conversion of stearoyl-CoA to oleoyl-CoA in eukaryotes is catalyzed by stearoyl-CoA desaturase in a reaction sequence that also involves cytochrome b₅ and cytochrome b₅ reductase. Two electrons are passed from NADH through the chain of reactions as shown, and two electrons are also derived from the fatty acyl substrate. linoleic acid in eukaryotes. This is the only means by which animals can synthesize fatty acids with double bonds at positions beyond C-9.

The Unsaturation Reaction May Be Followed by Chain Elongation

Additional chain elongation can occur following this single desaturation reaction. The oleoyl-CoA produced can be elongated by two carbons to form a 20:1 cis-D¹¹ fatty acyl-CoA. If the starting fatty acid is palmitate, reactions similar to the preceding scheme yield palmitoleoyl-CoA (16:1 cis-D⁹), which subsequently can be elongated to yield cis-vaccenic acid (18:1 cis-D¹¹). Similarly, C₁₆ and C₁₈ fatty acids can be elongated to yield C₂₂ and C₂₄ fatty acids, such as are often found in sphingolipids.

Biosynthesis of Polyunsaturated Fatty Acids

Organisms differ with respect to formation, processing, and utilization of polyunsaturated fatty acids. E. coli, for example, does not have any polyunsaturated fatty acids. Eukaryotes do synthesize a variety of polyunsaturated fatty acids, certain organisms more than others. For example, plants manufacture double bonds between the D⁹ and the methyl end of the chain, but mammals cannot. Plants readily desaturate oleic acid at the 12-position (to give linoleic acid) or at both the 12- and 15-positions (producing linolenic acid). Mammals require polyunsaturated fatty acids, but must acquire them in their diet. As such, they are referred to as essential fatty acids. On the other hand, mammals can introduce double bonds between the double bond at the 8- or 9-position and the carboxyl group. Enzyme complexes in the endoplasmic reticulum desaturate the 5-position, provided a double bond exists at the 8-position, and form a double bond at the 6-position if one already exists at the 9-position. Thus, oleate can be unsaturated at the 6,7-position to give an 18:2 cis-D⁶,D⁹ fatty acid.

Arachidonic Acid Is Synthesized from Linoleic Acid by Mammals

Mammals can add additional double bonds to unsaturated fatty acids in their diets. Their ability to make arachidonic acid from linoleic acid is one example (Figure 25.15). This fatty acid is the precursor for prostaglandins and other biologically active derivatives such as leukotrienes. Synthesis involves formation of a linoleoyl ester of CoA from dietary linoleic acid, followed by introduction of a double bond at the 6-position. The triply unsaturated product is then elongated (by malonyl-CoA with a decarboxylation step) to yield a 20-carbon fatty acid with double bonds at the 8-, 11-, and 14-positions. A second desaturation reaction at the 5-position followed by an acyl-CoA synthetase reaction (Chapter 24) liberates the product, a 20-carbon fatty acid with double bonds at the 5-, 8-, 11-, and 14-positions.

Regulatory Control of Fatty Acid Metabolism An Interplay of Allosteric Modifiers and Phosphorylation - Dephosphorylation Cycles

The control and regulation of fatty acid synthesis is intimately related to regulation of fatty acid breakdown, glycolysis, and the TCA cycle. Acetyl-CoA is an important intermediate metabolite in all these processes. In these terms, it is easy to appreciate the interlocking relationships in Figure 25.16. Malonyl-CoA can act to prevent fatty acyl-CoA derivatives from entering the mitochondria by inhibiting the carnitine acyltransferase that is responsible for this transport. In this way, when fatty acid synthesis is turned on (as signaled by higher levels of malonyl-CoA), b-oxidation is inhibited. As we pointed out earlier, citrate is an important allosteric activator of acetyl-CoA carboxylase, and fatty acyl-CoAs are inhibitors. The degree of inhibition is proportional to the chain length of the fatty acyl-CoA; longer chains show a higher affinity for the allosteric inhibition site on acetyl-CoA carboxylase. Palmitoyl-CoA, stearoyl-CoA, and arachidyl-CoA are the most potent inhibitors of the carboxylase.

Figure 25.16 · Regulation of fatty acid synthesis and fatty acid oxidation are coupled as shown. Malonyl-CoA, produced during fatty acid synthesis, inhibits the uptake of fatty acylcarnitine (and thus fatty acid oxidation) by mitochondria. When fatty acyl CoA levels rise, fatty acid synthesis is inhibited and fatty acid oxidation activity increases. Rising citrate levels (which reflect an abundance of acetyl-CoA) similarly signal the initiation of fatty acid synthesis.

Hormonal Signals Regulate ACC and Fatty Acid Biosynthesis

As described earlier, citrate activation and palmitoyl-CoA inhibition of acetyl-CoA carboxylase are strongly dependent on the phosphorylation state of the enzyme. This provides a crucial connection to hormonal regulation. Many of the enzymes that act to phosphorylate acetyl-CoA carboxylase (Figure 25.4) are controlled by hormonal signals. Glucagon is a good example (Figure 25.17). As noted in Chapter 23, glucagon binding to membrane receptors activates an intracellular cascade involving activation of adenylyl cyclase. Cyclic AMP produced by the cyclase activates a protein kinase, which then phosphorylates acetyl-CoA carboxylase. Unless citrate levels are high, phosphorylation causes inhibition of fatty acid biosynthesis. The carboxylase (and fatty acid synthesis) can be reactivated by a specific phosphatase, which dephosphorylates the carboxylase. Also indicated in Figure 25.17 is the simultaneous activation by glucagon of triacylglycerol lipases, which hydrolyze triacylglycerols, releasing fatty acids for b-oxidation. Both the inactivation of acetyl-CoA carboxylase and the activation of triacylglycerol lipase are counteracted by insulin, whose receptor acts to stimulate a phosphodiesterase that converts cAMP to AMP.

Figure 25.17 · Hormonal signals regulate fatty acid synthesis, primarily through actions on acetyl-CoA carboxylase. Availability of fatty acids also depends upon hormonal activation of triacylglycerol lipase.

25.2 × Biosynthesis of Complex Lipids

Complex lipids consist of backbone structures to which fatty acids are covalently bound. Principal classes include the glycerolipids, for which glycerol is the backbone, and sphingolipids, which are built on a sphingosine backbone. The two major classes of glycerolipids are glycerophospholipids and triacylglycerols. The phospholipids, which include both glycerophospholipids and sphingomyelins, are crucial components of membrane structure. They are also precursors of hormones such as the eicosanoids (e.g., prostaglandins) and signal molecules, such as the breakdown products of phosphatidylinositol.
    Different organisms possess greatly different complements of lipids and therefore invoke somewhat different lipid biosynthetic pathways. For example, sphingolipids and triacylglycerols are produced only in eukaryotes. In contrast, bacteria usually have rather simple lipid compositions. Phosphatidylethanol-amine accounts for at least 75% of the phospholipids in E. coli, with phosphatidyl-glycerol and cardiolipin accounting for most of the rest. E. coli membranes possess no phosphatidylcholine, phosphatidylinositol, sphingolipids, or choles-terol. On the other hand, some bacteria (such as Pseudomonas) can synthesize phosphatidylcholine, for example. In this section and the one following, we consider some of the pathways for the synthesis of glycerolipids, sphingolipids, and the eicosanoids, which are derived from phospholipids.

Glycerolipid Biosynthesis

A common pathway operates in nearly all organisms for the synthesis of phosphatidic acid, the precursor to other glycerolipids. Glycerokinase catalyzes the phosphorylation of glycerol to form glycerol-3-phosphate, which is then acyl-ated at both the 1- and 2-positions to yield phosphatidic acid (Figure 25.18). The first acylation, at position 1, is catalyzed by glycerol-3-phosphate acyltransferase, an enzyme that in most organisms is specific for saturated fatty acyl groups. Eukaryotic systems can also utilize dihydroxyacetone phosphate as a starting point for synthesis of phosphatidic acid (Figure 25.18). Again a specific acyltransferase adds the first acyl chain, followed by reduction of the backbone keto group by acyldihydroxyacetone phosphate reductase, using NADPH as the reductant. Alternatively, dihydroxyacetone phosphate can be reduced to glycerol-3-phosphate by glycerol-3-phosphate dehydrogenase.

Figure 25.18 · Synthesis of glycerolipids in eukaryotes begins with the formation of phosphatidic acid, which may be formed from dihydroxyacetone phosphate or glycerol as shown.

Figure 25.19 · Diacylglycerol and CDP-diacylglycerol are the principal precursors of glycerolipids in eukaryotes. Phosphatidylethanolamine and phosphatidylcholine are formed by reaction of diacylglycerol with CDP-ethanolamine or CDP-choline, respectively.

Eukaryotes Synthesize Glycerolipids from CDP-Diacylglycerol or Diacylglycerol

In eukaryotes, phosphatidic acid is converted directly either to diacylglycerol or to cytidine diphosphodiacylglycerol (or simply CDP-diacylglycerol; Figure 25.19). From these two precursors, all other glycerophospholipids in eukaryotes are derived. Diacylglycerol is a precursor for synthesis of triacylglycerol, phosphatidylethanolamine, and phosphatidylcholine. Triacylglycerol is synthesized mainly in adipose tissue, liver, and intestines and serves as the principal energy storage molecule in eukaryotes. Triacylglycerol biosynthesis in liver and adipose tissue occurs via diacylglycerol acyltransferase, an enzyme bound to the cytoplasmic face of the endoplasmic reticulum. A different route is used, however, in intestines. Recall (Figure 24.3) that triacylglycerols from the diet are broken down to 2-monoacylglycerols by specific lipases. Acyltransferases then acylate 2-monoacylglycerol to produce new triacylglycerols (Figure 25.20).

Figure 25.20 · Triacylglycerols are formed primarily by the action of acyltransferases on mono- and diacylglycerol. Acyltransferase in E. coli is an integral membrane protein (83 kD) and can utilize either fatty acyl-CoAs or acylated acyl carrier proteins as substrates. It shows a particular preference for palmitoyl groups. Eukaryotic acyltransferases use only fatty acyl-CoA molecules as substrates.

Phosphatidylethanolamine Is Synthesized from Diacylglycerol and CDP-Ethanolamine

Phosphatidylethanolamine synthesis begins with phosphorylation of ethanol-amine to form phosphoethanolamine (Figure 25.19). The next reaction involves transfer of a cytidylyl group from CTP to form CDP-ethanolamine and pyrophosphate. As always, PP_i hydrolysis drives this reaction forward. A specific phosphoethanolamine transferase then links phosphoethanolamine to the di-acylglycerol backbone. Biosynthesis of phosphatidylcholine is entirely analogous because animals synthesize it directly. All of the choline utilized in this pathway must be acquired from the diet. Yeast, certain bacteria, and animal livers, however, can convert phosphatidylethanolamine to phosphatidylcholine by methylation reactions involving S-adenosylmethionine (see Chapter 26).

Exchange of Ethanolamine for Serine Converts Phosphatidylethanolamine to Phosphatidylserine

Mammals synthesize phosphatidylserine (PS) in a calcium ion-dependent reaction involving aminoalcohol exchange (Figure 25.21). The enzyme catalyzing this reaction is associated with the endoplasmic reticulum and will accept phosphatidylethanolamine (PE) and other phospholipid substrates. A mitochondrial PS decarboxylase can subsequently convert PS to PE. No other pathway converting serine to ethanolamine has been found.

Figure 25.21 · The interconversion of phosphatidylethanolamine and phosphatidylserine in mammals.

Eukaryotes Synthesize Other Phospholipids via CDP-Diacylglycerol

Eukaryotes also use CDP-diacylglycerol, derived from phosphatidic acid, as a precursor for several other important phospholipids, including phosphatidyl-inositol (PI), phosphatidylglycerol (PG), and cardiolipin (Figure 25.22). PI accounts for only about 2 to 8% of the lipids in most animal membranes, but breakdown products of PI, including inositol-1,4,5-trisphosphate and diacylglycerol, are second messengers in a vast array of cellular signaling processes.

Figure 25.22 · CDP-diacylglycerol is a precursor of phosphatidylinositol, phosphatidylglycerol, and cardiolipin in eukaryotes.

Dihydroxyacetone Phosphate Is a Precursor to the Plasmalogens

Certain glycerophospholipids possess alkyl or alkenyl ether groups at the 1-position in place of an acyl ester group. These glyceroether phospholipids are synthesized from dihydroxyacetone phosphate (Figure 25.23). Acylation of dihydroxyacetone phosphate (DHAP) is followed by an exchange reaction, in which the acyl group is removed as a carboxylic acid and a long-chain alcohol adds to the 1-position. This long-chain alcohol is derived from the corresponding acyl-CoA by means of an acyl-CoA reductase reaction involving oxidation of two molecules of NADH. The 2-keto group of the DHAP backbone is then reduced to an alcohol, followed by acylation. The resulting 1-alkyl-2-acylglycero-3-phosphate can react in a manner similar to phosphatidic acid to produce ether analogs of phosphatidylcholine, phosphatidylethanolamine, and so forth (Figure 25.23). In addition, specific desaturase enzymes associated with the endoplasmic reticulum can desaturate the alkyl ether chains of these lipids as shown. The products, which contain a,b-unsaturated ether-linked chains at the C-1 position, are plasmalogens; they are abundant in cardiac tissue and in the central nervous system. The desaturases catalyzing these reactions are distinct from but similar to those that introduce unsaturations in fatty acyl-CoAs. These enzymes use cytochrome b₅ as a cofactor, NADH as a reductant, and O₂ as a terminal electron acceptor.

Figure 25.23 · Biosynthesis of plasmalogens in animals. Acylation at C-1 is followed by exchange of the acyl group for a long-chain alcohol. Reduction of the keto group at C-2 is followed by transferase reactions, which add an acyl group at C-2 and a polar head-group moiety, and a desaturase reaction that forms a double bond in the alkyl chain. The first two enzymes are of cytoplasmic origin, and the last transferase is located at the endoplasmic reticulum.

Platelet Activating Factor

A particularly interesting ether phospholipid with unusual physiological properties has recently been characterized. As shown in Figure 25.24, 1-alkyl-2-acetylglycerophosphocholine, also known as platelet activating factor, possesses an alkyl ether at C-1 and an acetyl group at C-2. The very short chain at C-2 makes this molecule much more water-soluble than typical glycerolipids. Platelet activating factor displays a dramatic ability to dilate blood vessels (and thus reduce blood pressure in hypertensive animals) and to aggregate platelets.

Figure 25.24 · Platelet activating factor, formed from 1-alkyl-2-lysophosphatidylcholine by acetylation at C-2, is degraded by the action of acetylhydrolase.

Sphingolipid Biosynthesis

Sphingolipids, ubiquitous components of eukaryotic cell membranes, are pres-ent at high levels in neural tissues. The myelin sheath that insulates nerve axons is particularly rich in sphingomyelin and other related lipids. Prokaryotic organisms normally do not contain sphingolipids. Sphingolipids are built upon sphingosine backbones rather than glycerol. The initial reaction, which involves condensation of serine and palmitoyl-CoA with release of bicarbonate, is catalyzed by 3-ketosphinganine synthase, a PLP-dependent enzyme (Figure 25.25). Reduction of the ketone product to form sphinganine is catalyzed by 3-keto-sphinganine reductase, with NADPH as a reactant. In the next step, sphinganine is acylated to form N-acyl sphinganine, which is then desaturated to form ceramide. Sphingosine itself does not appear to be an intermediate in this pathway in mammals.

Figure 25.25 · Biosynthesis of sphingolipids in animals begins with the 3-ketosphinganine synthase reaction, a PLP-dependent condensation of palmitoyl-CoA and serine. Subsequent reduction of the keto group, acylation, and desaturation (via reduction of an electron acceptor, X) form ceramide, the precursor of other sphingolipids.

Ceramide Is the Precursor for Other Sphingolipids and Cerebrosides

Ceramide is the building block for all other sphingolipids. Sphingomyelin, for example, is produced by transfer of phosphocholine from phosphatidylcholine (Figure 25.26). Glycosylation of ceramide by sugar nucleotides yields cerebrosides, such as galactosylceramide, which makes up about 15% of the lipids of myelin sheath structures. Cerebrosides that contain one or more sialic acid (N-acetylneuraminic acid) moieties are called gangliosides. Several dozen gangliosides have been characterized, and the general form of the biosynthetic pathway is illustrated for the case of ganglioside GM₂ (Figure 25.26). Sugar units are added to the developing ganglioside from nucleotide derivatives, including UDP-N-acetylglucosamine, UDP-galactose, and UDP-glucose.

Figure 25.26 · Glycosylceramides (such as galactosylceramide), gangliosides, and sphingomyelins are synthesized from ceramide in animals.

25.3 × Eicosanoid Biosynthesis and Function

Eicosanoids, so named because they are all derived from 20-carbon fatty acids, are ubiquitous breakdown products of phospholipids. In response to appropriate stimuli, cells activate the breakdown of selected phospholipids (Figure 25.27). Phospholipase A₂ (Chapter 8) selectively cleaves fatty acids from the C-2 position of phospholipids. Often these are unsaturated fatty acids, among which is arachidonic acid. Arachidonic acid may also be released from phospholipids by the combined actions of phospholipase C (which yields diacylglycerols) and diacylglycerol lipase (which releases fatty acids).

Figure 25.27 · Arachidonic acid, derived from breakdown of phospholipids (PL), is the precursor of prostaglandins, thromboxanes, and leukotrienes. The letters used to name the prostaglandins are assigned on the basis of similarities in structure and physical properties. The class denoted PGE, for example, consists of b-hydroxyketones that are soluble in ether, whereas PGF denotes 1,3-diols that are soluble in phosphate buffer. PGA denotes prostaglan-dins possessing a, b-unsaturated ketones. The number following the letters refers to the number of carbon - carbon double bonds. Thus, PGE₂ contains two double bonds.

Eicosanoids Are Local Hormones

Animal cells can modify arachidonic acid and other polyunsaturated fatty acids, in processes often involving cyclization and oxygenation, to produce so-called local hormones that (1) exert their effects at very low concentrations and (2) usually act near their sites of synthesis. These substances include the prostaglandins (PG) (Figure 25.27) as well as thromboxanes (Tx), leukotrienes, and other hydroxyeicosanoic acids. Thromboxanes, discovered in blood platelets (thrombocytes), are cyclic ethers (TxB₂ is actually a hemiacetal; see Figure 25.27) with a hydroxyl group at C-15.

Prostaglandins Are Formed from Arachidonate by Oxidation and Cyclization

All prostaglandins are cyclopentanoic acids derived from arachidonic acid. The biosynthesis of prostaglandins is initiated by an enzyme associated with the endoplasmic reticulum, called prostaglandin endoperoxide synthase, also known as cyclooxygenase. The enzyme catalyzes simultaneous oxidation and cyclization of arachidonic acid. The enzyme is viewed as having two distinct activities, cyclooxygenase and peroxidase, as shown in Figure 25.28.

Figure 25.28 · Prostaglandin endoperoxide synthase, the enzyme that converts arachidonic acid to prostaglandin PGH₂, possesses two distinct activities: cyclooxygenase (steps 1 and 2) and glutathione (GSSG) - dependent hydroperoxidase (step 3). Cyclooxygenase is the site of action of aspirin and many other analgesic agents.

A Variety of Stimuli Trigger Arachidonate Release and Eicosanoid Synthesis

The release of arachidonate and the synthesis or interconversion of eicosanoids can be initiated by a variety of stimuli, including histamine, hormones such as epinephrine and bradykinin, proteases such as thrombin, and even serum albumin. An important mechanism of arachidonate release and eicosanoid synthesis involves tissue injury and inflammation. When tissue damage or injury occurs, special inflammatory cells, monocytes and neutrophils, invade the injured tissue and interact with the resident cells (e.g., smooth muscle cells and fibroblasts). This interaction typically leads to arachidonate release and eicosanoid synthesis. Examples of tissue injury in which eicosanoid synthesis has been characterized include heart attack (myocardial infarction), rheumatoid arthritis, and ulcerative colitis.

"Take Two Aspirin and " Inhibit Your Prostaglandin Synthesis

In 1971, biochemist John Vane was the first to show that aspirin (acetylsalicylate; Figure 25.29) exerts most of its effects by inhibiting the biosynthesis of prostaglandins. Its site of action is prostaglandin endoperoxide synthase. Cyclooxygenase activity is destroyed when aspirin O-acetylates Ser⁵³⁰ on the enzyme. From this you may begin to infer something about how prostaglandins (and aspirin) function. Prostaglandins are known to enhance inflammation in animal tissues. Aspirin exerts its powerful anti-inflammatory effect by inhibiting this first step in their synthesis. Aspirin does not have any measurable effect on the peroxidase activity of the synthase. Other nonsteroidal anti-inflammatory agents, such as ibuprofen (Figure 25.29) and phenylbutazone, inhibit the cyclooxygenase by competing at the active site with arachidonate or with the peroxyacid intermediate (PGG₂, Figure 25.28). See A Deeper Look, page 834.

Figure 25.29 · (a) The structures of several common analgesic agents. Acetaminophen is marketed under the tradename Tylenol^(r). Ibuprofen is sold as Motrin^(r), Nuprin^(r), and Advil^(r). (b) Acetylsalicylate (aspirin) inhibits the cyclooxygenase activity of endoperoxide synthase via acetylation (covalent modification) of Ser⁵³⁰.

25.4 × Cholesterol Biosynthesis

The most prevalent steroid in animal cells is cholesterol (Figure 25.30). Plants contain no cholesterol, but they do contain other steroids very similar to cholesterol in structure (see page 256). Cholesterol serves as a crucial component of cell membranes and as a precursor to bile acids (e.g., cholate, glycocholate, taurocholate) and steroid hormones (e.g., testosterone, estradiol, progesterone). Also, vitamin D₃ is derived from 7-dehydrocholesterol, the immediate precursor of cholesterol. Liver is the primary site of cholesterol biosynthesis.

Figure 25.30 · The structure of cholesterol, drawn (a) in the traditional planar motif and (b) in a form that more accurately describes the conformation of the ring system.

Figure 25.31 · The biosynthesis of 3R-meva-lonate from acetyl-CoA.

Mevalonate Is Synthesized from Acetyl-CoA via HMG-CoA Synthase

The cholesterol biosynthetic pathway begins in the cytosol with the synthesis of mevalonate from acetyl-CoA (Figure 25.31). The first step is the b-ketothi-olase-catalyzed Claisen condensation of two molecules of acetyl-CoA to form acetoacetyl-CoA. In the next reaction, acetyl-CoA and acetoacetyl-CoA join to form 3-hydroxy-3-methylglutaryl-CoA, which is abbreviated HMG-CoA. The reactiona second Claisen condensationis catalyzed by HMG-CoA synthase. The third step in the pathway is the rate-limiting step in cholesterol biosynthesis. Here, HMG-CoA undergoes two NADPH-dependent reductions to produce 3R-mevalonate (Figure 25.32). The reaction is catalyzed by HMG-CoA reductase, a 97-kD glycoprotein that traverses the endoplasmic reticulum membrane with its active site facing the cytosol. As the rate-limiting step, HMG-CoA reductase is the principal site of regulation in cholesterol synthesis.

Figure 25.32 · A reaction mechanism for HMG-CoA reductase. Two successive NADPH-dependent reductions convert the thioester, HMG-CoA, to a primary alcohol.

 Three different regulatory mechanisms are involved:

1. Phosphorylation by cAMP-dependent protein kinases inactivates the reductase. This inactivation can be reversed by two specific phosphatases (Figure 25.33).

2. Degradation of HMG-CoA reductase. This enzyme has a half-life of only three hours, and the half-life itself depends on cholesterol levels: high [cholesterol] means a short half-life for HMG-CoA reductase.

3. Gene expression cholesterol levels control the amount of mRNA. If [cholesterol] is high, levels of mRNA coding for the reductase are reduced. If [cholesterol] is low, more mRNA is made. (Regulation of gene expression is discussed in Chapter 31.)

Figure 25.33 · HMG-CoA reductase activity is modulated by a cycle of phosphorylation and dephosphorylation.

A Thiolase Brainteaser

If acetate units can be condensed by the thiolase reaction to yield acetoacetate in the first step of cholesterol synthesis, why couldn’t this same reaction also be used in fatty acid synthesis, avoiding all the complexity of the fatty acyl synthase? The answer is that the thiolase reaction is more or less reversible but slightly favors the cleavage reaction. In the cholesterol synthesis pathway, subsequent reactions, including HMG-CoA reductase and the following kinase reactions, pull the thiolase-catalyzed condensation forward. However, in the case of fatty acid synthesis, a succession of eight thiolase condensations would be distinctly unfavorable from an energetic perspective. Given the necessity of repeated reactions in fatty acid synthesis, it makes better energetic sense to use a reaction that is favorable in the desired direction.

Squalene Is Synthesized from Mevalonate

The biosynthesis of squalene involves conversion of mevalonate to two key 5-carbon intermediates, isopentenyl pyrophosphate and dimethylallyl pyrophosphate, which join to yield farnesyl pyrophosphate and then squalene. A series of four reactions converts mevalonate to isopentenyl pyrophosphate and then to dimethylallyl pyrophosphate (Figure 25.34). The first three steps each consume an ATP, two for the purpose of forming a pyrophosphate at the 5-position and the third to drive the decarboxylation and double bond formation in the third step. Pyrophosphomevalonate decarboxylase phosphorylates the 3-hydroxyl group, and this is followed by trans elimination of the phosphate and carboxyl groups to form the double bond in isopentenyl pyrophosphate. Isomerization of the double bond yields the dimethylallyl pyrophosphate. Condensation of these two 5-carbon intermediates produces geranyl pyrophosphate; addition of another 5-carbon isopentenyl group gives farnesyl pyrophosphate. Both steps in the production of farnesyl pyrophosphate occur with release of pyrophosphate, hydrolysis of which drives these reactions forward. Note too that the linkage of isoprene units to form farnesyl pyrophosphate occurs in a head-to-tail fashion. This is the general rule in biosynthesis of molecules involving isoprene linkages. The next step the joining of two farnesyl pyrophosphates to produce squalene is a "tail-to-tail" condensation and represents an important exception to the general rule.

Figure 25.34 · The conversion of mevalonate to squalene.

Critical Developments in Biochemistry
The Long Search for the Route of Cholesterol Biosynthesis
Heilbron, Kamm, and Owens suggested as early as 1926 that squalene is a precursor of cholesterol. That same year, H. J. Channon demonstrated that animals fed squalene from shark oil produced more cholesterol in their tissues. Bloch and Rittenberg showed in the 1940s that a significant amount of the carbon in the tetracyclic moiety and in the aliphatic side chain of cholesterol was derived from acetate. In 1934, Sir Robert Robinson suggested a scheme for the cyclization of squalene to form cholesterol before the biosynthetic link between acetate and squalene was understood. Squalene is actually a polymer of isoprene units, and Bonner and Arreguin suggested in 1949 that three acetate units could join to form 5-carbon isoprene units (see figure, part a).     In 1952, Konrad Bloch and Robert Langdon showed conclusively that labeled squalene is synthesized rapidly from labeled acetate and also that cholesterol is derived from squalene. Langdon, a graduate student of Bloch’s, performed the critical experiments in Bloch’s laboratory at the University of Chicago, while Bloch spent the summer in Bermuda attempting to demonstrate that radioactively labeled squalene would be converted to cholesterol in shark livers. As Bloch himself admitted, "All I was able to learn was that sharks of manageable length are very difficult to catch and their oily livers impossible to slice" (Bloch, 1987).	In 1953, Bloch, together with the eminent organic chemist R. B. Woodward, proposed a new scheme (see figure, part b) for the cyclization of squalene. (Together with Fyodor Lynen, Bloch received the Nobel Prize in medicine or physiology in 1964 for his work.) The picture was nearly complete, but one crucial question remained:How could isoprene be the intermediate in the transformation of acetate into squalene? In 1956, Karl Folkers and his colleagues at Merck, Sharpe and Dohme isolated mevalonic acid and also showed that mevalonate was the precursor of isoprene units. The search for the remaining details (described in the text) made the biosynthesis of cholesterol one of the most enduring and challenging bioorganic problems of the forties, fifties, and sixties. Even today, several of the enzyme mechanisms remain poorly understood.
(a) An isoprene unit and a scheme for head-to-tail linking of isoprene units. (b) The cyclization of squalene to form lanosterol, as proposed by Bloch and Woodward.

    Squalene monooxygenase, an enzyme bound to the endoplasmic reticulum, converts squalene to squalene-2,3-epoxide (Figure 25.35). This reaction employs FAD and NADPH as coenzymes and requires O₂ as well as a cytosolic protein called soluble protein activator. A second ER membrane enzyme, 2,3-oxidosqualene lanosterol cyclase, catalyzes the second reaction, which involves a succession of 1,2 shifts of hydride ions and methyl groups.

Figure 25.35 · Cholesterol is synthesized from squalene via lanosterol. The primary route from lanosterol involves 20 steps, the last of which converts 7-dehydrocholesterol to cholesterol. An alternative route produces desmosterol as the penultimate intermediate.

Human Biochemistry
Lovastatin Lowers Serum Cholesterol Levels
Chemists and biochemists have long sought a means of reducing serum cholesterol levels to reduce the risk of heart attack and cardiovascular disease. Because HMG-CoA reductase is the rate-limiting step in cholesterol biosynthesis, this enzyme is a likely drug target. Mevinolin, also known as lovastatin (see figure), was isolated from a strain of Aspergillus terreus and developed at Merck, Sharpe and Dohme for this purpose. It is now a widely prescribed cholesterol-lowering drug. Dramatic reductions of serum cholesterol are observed at doses of 20 to 80 mg per day.	Lovastatin is administered as an inactive lactone. After oral ingestion, it is hydrolyzed to the active mevinolinic acid, a competitive inhibitor of the reductase with a KI of 0.6 nM. Mevinolinic acid is thought to behave as a transition-state analog (Chapter 16) of the tetrahedral intermediate formed in the HMG-CoA reductase reaction (see figure). Derivatives of lovastatin have been found to be even more potent in cholesterol-lowering trials. Synvinolin lowers serum cholesterol levels at much lower doses than lovastatin.
The structures of (inactive) lovastatin, (active) mevinolinic acid, mevalonate, and synvinolin.

Conversion of Lanosterol to Cholesterol Requires 20 Additional Steps

Although lanosterol may appear similar to cholesterol in structure, another 20 steps are required to convert lanosterol to cholesterol (Figure 25.35). The enzymes responsible for this are all associated with the endoplasmic reticulum. The primary pathway involves 7-dehydrocholesterol as the penultimate intermediate. An alternative pathway, also composed of many steps, produces the intermediate desmosterol. Reduction of the double bond at C-24 yields cholesterol. Cholesterol esters a principal form of circulating cholesterol are synthesized by acyl-CoA:cholesterol acyltransferases (ACAT) on the cytoplasmic face of the endoplasmic reticulum.

25.5 × Transport of Many Lipids Occurs via Lipoprotein Complexes

When most lipids circulate in the body, they do so in the form of lipoprotein complexes. Simple, unesterified fatty acids are merely bound to serum albumin and other proteins in blood plasma, but phospholipids, triacylglycerols, cholesterol, and cholesterol esters are all transported in the form of lipoproteins. At various sites in the body, lipoproteins interact with specific receptors and enzymes that transfer or modify their lipid cargoes. It is now customary to classify lipoproteins according to their densities (Table 25.1). The densities are related to the relative amounts of lipid and protein in the complexes. Because most proteins have densities of about 1.3 to 1.4 g/mL, and lipid aggregates usually possess densities of about 0.8 g/mL, the more protein and the less lipid in a complex, the denser the lipoprotein. Thus, there are high-density lipoproteins (HDL), low-density lipoproteins (LDL), intermediate-density lipoproteins (IDL), very low density lipoproteins (VLDL), and also chylomicrons. Chylo-microns have the lowest protein-to-lipid ratio and thus are the lowest-density lipoproteins. They are also the largest.

Figure 25.36 · Photograph of an arterial plaque. (Science Photo Library/Photo Researchers, Inc.)

The Structure and Synthesis of the Lipoproteins

HDL and VLDL are assembled primarily in the endoplasmic reticulum of the liver (with smaller amounts produced in the intestine), whereas chylomicrons form in the intestine. LDL is not synthesized directly, but is made from VLDL. LDL appears to be the major circulatory complex for cholesterol and cholesterol esters. The primary task of chylomicrons is to transport triacylglycerols. Despite all this, it is extremely important to note that each of these lipoprotein classes contains some of each type of lipid. The relative amounts of HDL and LDL are important in the disposition of cholesterol in the body and in the development of arterial plaques (Figure 25.36). The structures of the various lipoproteins are approximately similar, and they consist of a core of mobile triacylglycerols or cholesterol esters surrounded by a single layer of phospholipid, into which is inserted a mixture of cholesterol and proteins (Figure 25.37). Note that the phospholipids are oriented with their polar head groups facing outward to interact with solvent water, and that the phospholipids thus shield the hydrophobic lipids inside from the solvent water outside. The proteins also function as recognition sites for the various lipoprotein receptors throughout the body. A number of different apoproteins have been identified in lipoproteins (Table 25.2), and others may exist as well. The apoproteins are abundant in hydrophobic amino acid residues, as is appropriate for interactions with lipids. A cholesterol ester transfer protein also associates with lipoproteins.

Figure 25.37 · A model for the structure of a typical lipoprotein. (a) A core of cholesterol and cholesteryl esters is surrounded by a phospholipid (monolayer) membrane. Apolipopro-tein A-I is modeled here as a long amphipathic a-helix, with the nonpolar face of the helix embedded in the hydrophobic core of the lipid particle and the polar face of the helix exposed to solvent. (b) A ribbon diagram of apolipoprotein A-I. (Adapted from Borhani, D. W., Rogers, D. P., Engler, J. A., and Brouillette, C. G., 1997. Crystal structure of truncated human apolipoprotein A-I suggests a lipid-bound conformation. Proceedings of the National Academy of Sciences 94:12291-12296.)

Lipoproteins in Circulation Are Progressively Degraded by Lipoprotein Lipase

Figure 25.38 · Lipoprotein components are synthesized predominantly in the ER of liver cells. Following assembly of lipoprotein particles (red dots) in the ER and processing in the Golgi, lipoproteins are packaged in secretory vesicles for export from the cell (via exocytosis) and released into the circulatory system.nd degradation of lipoprotein particles. (ACAT is acyl-CoA cholesterol acyltransferase.)

The livers and intestines of animals are the primary sources of circulating lipids. Chylomicrons carry triacylglycerol and cholesterol esters from the intestines to other tissues, and VLDLs carry lipid from liver, as shown in Figure 25.38. At various target sites, particularly in the capillaries of muscle and adipose cells, these particles are degraded by lipoprotein lipase, which hydrolyzes triacylglycerols. Lipase action causes progressive loss of triacylglycerol (and apoprotein) and makes the lipoproteins smaller. This process gradually converts VLDL particles to IDL and then LDL particles, which are either returned to the liver for reprocessing or redirected to adipose tissues and adrenal glands. Every 24 hours, nearly half of all circulating LDL is removed from circulation in this way. The LDL binds to specific LDL receptors, which cluster in domains of the plasma membrane known as coated pits (discussed in subsequent paragraphs). These domains eventually invaginate to form coated vesicles (Figure 25.39). Within the cell, these vesicles fuse with lysosomes, and the LDLs are degraded by lysosomal acid lipases.

Figure 25.39 · Endocytosis and degradation of lipoprotein particles. (ACAT is acyl-CoA cholesterol acyltransferase.)

    High-density lipoproteins (HDL) have much longer life spans in the body (5 to 6 days) than other lipoproteins. Newly formed HDL contains virtually no cholesterol ester. However, over time, cholesterol esters are accumulated through the action of lecithin:cholesterol acyltransferase (LCAT), a 59-kD glycoprotein associated with HDLs. Another associated protein, cholesterol ester transfer protein, transfers some of these esters to VLDL and LDL. Alternatively, HDLs function to return cholesterol and cholesterol esters to the liver. This latter process apparently explains the correlation between high HDL levels and reduced risk of cardiovascular disease. (High LDL levels, on the other hand, are correlated with an increased risk of coronary artery and cardiovascular disease.)

Structure of the LDL Receptor

The LDL receptor in plasma membranes (Figure 25.40) consists of 839 amino acid residues and is composed of five domains. These domains include an LDL-binding domain of 292 residues, a segment of about 350 to 400 residues containing N-linked oligosaccharides, a 58-residue segment of O-linked oligosaccharides, a 22-residue membrane-spanning segment, and a 50-residue segment extending into the cytosol. The clustering of receptors prior to the formation of coated vesicles requires the presence of this cytosolic segment.

Figure 25.40 · The structure of the LDL receptor. The amino-terminal binding domain is responsible for recognition and binding of LDL apoprotein. The O-linked oligosaccharide-rich domain may act as a molecular spacer, raising the binding domain above the glycocalyx. The cytosolic domain is required for aggregation of LDL receptors during endocytosis.

Defects in Lipoprotein Metabolism Can Lead to Elevated Serum Cholesterol

The mechanism of LDL metabolism and the various defects that can occur therein have been studied extensively by Michael Brown and Joseph Goldstein, who received the Nobel Prize in medicine or physiology in 1985. Familial hypercholesterolemia is the term given to a variety of inherited metabolic defects that lead to greatly elevated levels of serum cholesterol much of it in the form of LDL particles. The general genetic defect responsible for familial hyper-cholesterolemia is the absence or dysfunction of LDL receptors in the body. Only about half the normal level of LDL receptors is found in heterozygous individuals (persons carrying one normal gene and one defective gene). Homozygotes (with two copies of the defective gene) have few if any functional LDL receptors. In such cases, LDLs (and cholesterol) cannot be absorbed, and plasma levels of LDL (and cholesterol) are very high. Typical heterozygotes display serum cholesterol levels of 300 to 400 mg/dL, but homozygotes carry serum cholesterol levels of 600 to 800 mg/dL or even higher. There are two possible causes of an absence of LDL receptors. Either receptor synthesis does not occur at all, or the newly synthesized protein does not successfully reach the plasma membrane, due to faulty processing in the Golgi or faulty transport to the plasma membrane. Even when LDL receptors are made and reach the plasma membrane, they may fail to function for two reasons. They may be unable to form clusters competent in coated pit formation because of folding or sequence anomalies in the carboxy-terminal domain, or they may be unable to bind LDL because of sequence or folding anomalies in the LDL-binding domain.

25.6 × Biosynthesis of Bile Acids

Bile acids, which exist mainly as bile salts, are polar carboxylic acid derivatives of cholesterol that are important in the digestion of food, especially the solubilization of ingested fats. The Na⁺ and K⁺ salts of glycocholic acid and taurocholic acid are the principal bile salts (Figure 25.41). Glycocholate and taurocholate are conjugates of cholic acid with glycine and taurine, respectively. Because they contain both nonpolar and polar domains, these bile salt conjugates are highly effective as detergents. These substances are made in the liver, stored in the gallbladder, and secreted as needed into the intestines.

Figure 25.41 · Cholic acid, a bile salt, is synthesized from cholesterol via 7a-hydroxy-cholesterol. Conjugation with taurine or glycine produces taurocholic acid and glycocholic acid, respectively. Taurocholate and glycocholate are freely water-soluble and are highly effective detergents.

    The formation of bile salts represents the major pathway for cholesterol degradation. The first step involves hydroxylation at C-7 (Figure 25.41). 7a-Hydroxylase, which catalyzes the reaction, is a mixed-function oxidase involving cytochrome P450. Mixed-function oxidases use O₂ as substrate. One oxygen atom goes to hydroxylate the substrate, while the other is reduced to water (Figure 25.42). The function of cytochrome P450 is to activate O₂ for the hydroxylation reaction. Such hydroxylations are quite common in the synthetic routes for cholesterol, bile acids, and steroid hormones and also in detoxification pathways for aromatic compounds. Several of these are considered in the next section. 7a-Hydroxycholesterol is the precursor for cholic acid.

Figure 25.42 · The mixed-function oxidase activity of 7a-hydroxylase.

25.7 ×