Chapter 26

Nitrogen Acquisition and Amino Acid Metabolism

Wheat harvest. Only plants and certain microorganisms are
able to transform the oxidized, inorganic forms of nitrogen
available in the inanimate environment into reduced, biologically
useful forms. ((c) Dick Durrance II/Woodfin Camp and Associates, Inc.)

Nitrogen is a vital macronutrient for all life, and in this chapter we begin our consideration of the pathways of nitrogen metabolism. We start with a presentation of the two principal routes for nitrogen acquisition from the inanimate environment: nitrate assimilation and nitrogen fixation. The reactions of ammonium assimilation follow. Glutamine synthetase merits particular attention because it conveys several important lessons in metabolic regulation. The pathways of amino acid biosynthesis and degradation are described; those involving the sulfur-containing amino acids provide an opportunity to introduce aspects of sulfur metabolism.

The Nitrogen Cycle

Nitrogen exists predominantly in an oxidized state in the inanimate environment, occurring principally as N₂ in the atmosphere or as nitrate ion (NO₃^-) in the soils and oceans. Its acquisition by biological systems is accompanied by its reduction to ammonium ion (NH₄⁺) and the incorporation of NH₄⁺ into organic linkage as amino or amido groups (Figure 26.1). The reduction of NO₃^- to NH₄⁺ occurs in green plants, various fungi, and certain bacteria in a two-step metabolic pathway known as nitrate assimilation. The formation of NH₄⁺ from N₂ gas is termed nitrogen fixation. N₂ fixation is an exclusively prokaryotic process, although bacteria in symbiotic association with certain green plants also carry out nitrogen fixation. No animals are capable of either nitrogen fixation or nitrate assimilation, so they are totally dependent on plants and microorganisms for the synthesis of organic nitrogenous compounds, such as amino acids and proteins, to satisfy their requirements for this essential element.
    Animals release excess nitrogen in a reduced form, either as NH₄⁺ or as organic nitrogenous compounds such as urea. The release of N occurs both during life and as a consequence of microbial decomposition following death. Various bacteria return the reduced forms of nitrogen back to the environment by oxidizing them. The oxidation of NH₄⁺ to NO₃^- by nitrifying bacteria, a group of chemoautotrophs, provides the sole source of chemical energy for the life of these microbes. Nitrate nitrogen also returns to the atmosphere as N₂ as a result of the metabolic activity of denitrifying bacteria. These bacteria are capable of using NO₃^- and similar oxidized inorganic forms of nitrogen as electron acceptors in place of O₂ in energy-producing pathways. The NO₃^- is reduced ultimately to dinitrogen (N₂). These bacteria thus deplete the levels of combined nitrogen,¹ important as a natural fertilizer, that might otherwise be available. However, such bacterial activity is being exploited in water treatment plants to reduce the load of combined nitrogen that might otherwise enter lakes, streams, and bays.

Figure 26.1 · The nitrogen cycle. Organic nitrogenous compounds are formed by the incorporation of NH₄⁺ into carbon skeletons. Ammonium can be formed from oxidized inorganic percursors by reductive reactions: nitrogen fixation reduces N₂ to NH₄⁺; nitrate assimilation reduces NO₃^- to NH₄⁺. Nitrifying bacteria can oxidize NH₄⁺ back to NO₃^- and obtain energy for growth in the process of nitrification. Denitrification is a form of bacterial respiration whereby nitrogen oxides serve as electron acceptors in the place of O₂ under anaerobic conditions.

¹N joined with other elements in chemical compounds.

26.1 · The Two Major Pathways of Biological N Acquisition

The Enzymology of Nitrate Assimilation

Nitrate assimilation occurs in two steps: the two-electron reduction of nitrate to nitrite, catalyzed by nitrate reductase (Equation 26.1), followed by the six-electron reduction of nitrite to ammonium, catalyzed by nitrite reductase (Equation 26.2).

 (1) NO₃^- + 2 H⁺ + 2 e^- ® NO₂^- + H₂O       (26.1)

 (2) NO₂^- + 8 H⁺ + 6 e^- ® NH₄⁺ + 2 H₂O   (26.2)

Nitrate assimilation is the predominant means by which green plants, algae, and many microorganisms acquire nitrogen. The pathway of nitrate assimilation accounts for over 99% of the inorganic nitrogen (nitrate or N₂) assimilated into organisms.

Figure 26.2 · The novel prosthetic groups of nitrate reductase and nitrite reductase. (a) The molybdenum cofactor of nitrate reductase. The molybdenum-free version of this compound is a pterin derivative called molybdopterin. (b) Siroheme, a uroporphyrin derivative, is a member of the isobacteriochlorin class of hemes, a group of porphyrins in which adjacent pyrrole rings are reduced. Siroheme is novel in having eight carboxylate-containing side chains. These carboxylate groups may act as H⁺ donors during the reduction of NO₂^- to NH₄⁺.

Nitrate Reductase

A pair of electrons is transferred from NADH via enzyme-associated sulfhydryl groups, FAD, cytochrome b₅₅₇, and MoCo (an essential molybdenum-containing cofactor) to nitrate, reducing it to nitrite. The brackets [] denote the protein-bound prosthetic groups that constitute an e^- transport chain between NADH and nitrate. Nitrate reductases typically are cytosolic 220-kD dimeric proteins. The structure of the molybdenum cofactor (MoCo) is shown in Figure 26.2a. Molybdenum cofactor is necessary for both nitrate reductase activity and the assembly of nitrate reductase subunits into the active holoenzyme dimer form. Molybdenum cofactor is also an essential cofactor for a variety of enzymes that catalyze hydroxylase-type reactions, including xanthine dehydrogenase, aldehyde oxidase, and sulfite oxidase.

Nitrite Reductase in Green Plants

Six electrons are required to reduce NO₂^- to NH₄⁺. Nitrite reductases in photosynthetic organisms obtain these electrons from six molecules of photosynthetically reduced ferredoxin (Fd_red). 

Photosynthetic nitrite reductases are 63-kD monomeric proteins having a tetranuclear iron-sulfur cluster and a novel heme, termed siroheme, as prosthetic groups. The [4Fe-4S] cluster and the siroheme act as a coupled e^- transfer center. Nitrite binds directly to siroheme, providing the sixth ligand, much as O₂ binds to the heme of hemoglobin. Nitrite is reduced to ammonium while liganded to siroheme. The structure of siroheme is shown in Figure 26.2b.
    In higher plants, nitrite reductase is found in chloroplasts, where it has ready access to its primary reductant, photosynthetically reduced ferredoxin. Microbial nitrite reductases are larger and more complex than plant nitrite reductases. Indeed, microbial nitrite reductases closely resemble nitrate reductases in having essential -SH groups and FAD prosthetic groups to couple enzyme-mediated NADPH oxidation to nitrite reduction (Figure 26.3).

Figure 26.3 · Domain organization within the enzymes of nitrate assimilation. The numbers denote residue number along the amino acid sequence of the proteins. The numbering for nitrate reductase is that from the green plant Arabidopsis thaliana; the plant nitrite reductase sequence shown here is spinach; the fungal nitrite reductase is Neurospora crassa. (Adapted in part from Campbell & Kinghorn, 1990. Trends in Biochemical Sciences 15:315-319.)

The Enzymology of Nitrogen Fixation

Nitrogen fixation involves the reduction of nitrogen gas (N₂) via an enzyme system found only in prokaryotic cells. The heart of the nitrogen fixation process is the enzyme known as nitrogenase, which catalyzes the reaction

 N₂ + 8 H⁺ + 8 e^- ® 2 NH₃ + H₂ (26.3)

Note that an obligatory reduction of two protons to hydrogen gas accompanies the biological reduction of N₂ to ammonia. Less than 1% of the inorganic N incorporated into organic compounds by organisms can be attributed to nitrogen fixation; however, this process provides the only direct biological access to the enormous reservoir of N₂ in the atmosphere.
    Although nitrogen fixation is exclusively prokaryotic, N₂-fixing bacteria may be either free-living or living as symbionts with higher plants. For example, Rhizobia are bacteria that fix nitrogen in symbiotic association with leguminous plants. Because nitrogen in a metabolically useful form is often the limiting nutrient for plant growth, such symbiotic associations can be an important factor in plant growth and agriculture.
    Despite the wide diversity of bacteria in which nitrogen fixation takes place, all N₂-fixing systems are nearly identical and all have four fundamental requirements: (1) the enzyme nitrogenase, (2) a strong reductant, such as reduced ferredoxin, (3) ATP, and (4) O₂-free conditions. In addition, several modes of regulation act to control nitrogen fixation.

THE NITROGENASE COMPLEX.  Two metalloproteins constitute the nitrogenase complex: the Fe-protein or nitrogenase reductase and the MoFe-protein, which is another name for nitrogenase. Nitrogenase reductase is a 60-kD homodimer possessing a single [4Fe-4S] cluster as a prosthetic group. Nitrogenase reductase is extremely O₂-sensitive. Nitrogenase reductase binds MgATP and hydrolyzes two ATP per electron transferred during nitrogen fixation. Because reduction of N₂ to 2 NH₃ + H₂ requires 8 electrons, 16 ATP are consumed per N₂ reduced.
     This ATP requirement seems paradoxical because the reduction of N₂ to NH₄⁺ is thermodynamically favorable (i.e., the free energy change for the reaction is negative). The solution to the paradox is found in the very strong bonding between the two N atoms in N₂ (Figure 26.4). Substantial energy input is needed to overcome this large activation energy and break the N-N triple bond. In this biological system, the energy is provided by ATP.

Figure 26.4 · The triple bond in N₂ must be broken during nitrogen fixation. A substantial energy input is needed to overcome this thermodynamic barrier, even though the overall free energy change, DG°', is negative.

     Nitrogenase, the MoFe protein, is a 240-kD a₂b₂-type heterotetramer. An ab dimer serves as the functional unit, and each ab dimer contains two types of metal centers, an unusual 8Fe-7S center known as the P-cluster (Figure 26.5a) and the novel 7Fe-1Mo-9S cluster known as the FeMo-cofactor (Figure 26.5b). Nitrogenase under unusual circumstances may contain an iron:vanadium cofactor instead of the molybdenum-containing one. Like nitrogenase reductase, nitrogenase is very oxygen-labile.

Figure 26.5 · Structures of the two types of metal clusters found in nitrogenase.
(a) The P-cluster. Two Fe₄S₃ clusters share a fourth S and are bridged by two thiol ligands from the protein (Cysa⁸⁸ and Cysb⁹⁵). (b) The FeMo-cofactor. This novel molybdenum-containing Fe-S complex contains 1 Mo, 7 Fe, and 9 S atoms; it is liganded to the protein via a Cysa²⁷⁵-S linkage to an Fe atom and a Hisa⁴⁴²-N linkage to the Mo atom. Homocitrate provides two oxo ligands to the Mo atom. (Adapted from Leigh, G. J., 1995. The mechanism of dinitrogen reduction by molybdenum nitrogenases. European Journal of Biochemistry 229:14-20.)

Figure 26.6 · The nitrogenase reaction. Depending on the bacterium, electrons for N₂ reduction may come from light, NADH, hydrogen gas, or pyruvate. The primary e^- donor for the nitrogenase system is reduced ferredoxin. Reduced ferredoxin passes electrons directly to nitrogenase reductase. A total of six electrons is required to reduce N₂ to 2 NH₃, and another two electrons are consumed in the obligatory reduction of 2 H⁺ to H₂. Nitrogenase reductase transfers e^- to nitrogenase one electron at a time. N₂ is bound at the critical FeMoCo prosthetic group of nitrogenase until all electrons and protons are added; no free intermediates such as HN=NH or H₂N-NH₂ are detectable.

THE NITROGENASE REACTION.  In the nitrogenase reaction (Figure 26.6), electrons from reduced ferredoxin pass to nitrogenase reductase, which serves as electron donor to nitrogenase, the enzyme that actually catalyzes N₂ fixation. Electron transfer from nitrogenase reductase to nitrogenase takes place through docking of nitrogenase reductase with an ab subunit pair of nitrogenase (Figure 27.7) and transfer of an electron according to the following sequence: Fe protein ® P-cluster ® FeMo-cofactor ® N₂. ATP hydrolysis is coupled to the transfer of an electron from the Fe protein to the P-cluster. ATP hydrolysis leads to conformational change in the nitrogenase reductase so that it no longer binds to nitrogenase. The ADP:oxidized nitrogenase reductase complex dissociates, making way for another ATP:reduced nitrogenase reductase complex to bind to nitrogenase. Nitrogenase is a rather slow enzyme: its optimal rate of e^- transfer is about 12 e^- pairs per second per enzyme molecule; that is, it reduces only three molecules of nitrogen gas per second. Because its activity is so weak, nitrogen-fixing cells maintain large amounts of nitrogenase so that their requirements for reduced N can be met. As much as 5% of the cellular protein may be nitrogenase. As indicated earlier, nitrogenase catalyzes the concomitant reduction of protons to hydrogen gas, its so-called hydrog-enase activity. This activity accompanies N₂ reduction in vivo and is energy-depleting. One equivalent of H₂ is formed for every 2 NH₃ (this unavoidable reaction leads to the stoichiometry given in Equation 26.3).

Figure 26.7 · (a) Ribbon diagram of nitrogenase reductase (the Fe protein). The ATP-binding site (here occupied by ADP) is at the left in this orientation, and the Fe₄S₄ cluster is at the right). (b) Model of the complex formed between nitrogenase reductase and a nitrogenase ab dimer. Within the nitrogenase reductase (shown in green and yellow here), the Fe₄S₄ center is closest to the nitrogenase ab dimer. The nitrogenase ab dimer (red and blue) is to the right of the nitrogenase reductase, with its FeMo-cofactor and P-cluster as space-filling models. (Adapted from Kim, J., and Rees, D. 1994. Nitrogenase and biological nitrogen fixation. Biochemistry 33:389-396.)

The Regulation of Nitrogen Fixation

To a first approximation, two regulatory controls are paramount (Figure 26.8): (a) ADP inhibits the activity of nitrogenase; thus, as the ATP/ADP ratio drops, nitrogen fixation is blocked. (b) NH₄⁺ represses the expression of the nif genes, the genes that encode the proteins of the nitrogen-fixing system. To date, some 20 nif genes have been identified with the nitrogen fixation process. Repression of nif gene expression by ammonium, the primary product of nitrogen fixation, is an efficient and effective way of shutting down N₂ fixation when its end product is not needed.

Figure 26.8 · Regulation of nitrogen fixation. (a) ADP inhibits nitrogenase activity.
(b) NH₄⁺ represses nif gene expression. (c) In some organisms, the nitrogenase complex is regulated by covalent modification. ADP-ribosylation of nitrogenase reductase leads to its inactivation. Nitrogenase reductase is a distant relative of the signal-transducing G-protein superfamily.

26.2 · The Fate of Ammonium

Ammonium enters organic linkage via three major reactions that are found in all cells. The enzymes mediating these reactions are (1) carbamoyl-phosphate synthetase I, (2) glutamate dehydrogenase, and (3) glutamine synthetase.
    Carbamoyl-phosphate synthetase I catalyzes one of the steps in the urea cycle. Two ATP are consumed, one in the activation of HCO₃^- for reaction with ammonium, and the other in the phosphorylation of the carbamate formed:

N-acetylglutamate is an essential allosteric activator for this enzyme.
    Glutamate dehydrogenase (GDH) catalyzes the reductive amination of a-ketoglutarate to yield glutamate. Reduced pyridine nucleotides (NADH or NADPH) provide the reducing power:

NH₄⁺ + a-ketoglutarate + NADPH + H⁺ ® glutamate + NADP⁺ + H₂O

This reaction provides an important interface between nitrogen metabolism and cellular pathways of carbon and energy metabolism because a-ketoglutarate is a citric acid cycle intermediate. In vertebrates, GDH is an a₆-type multimeric enzyme localized in the mitochondrial matrix that uses NADPH as electron donor when operating in the biosynthetic direction (the direction of glutamate synthesis) (Figure 26.9). In contrast, when GDH acts in the catabolic direction to generate a-ketoglutarate from glutamate, NAD⁺, not NADP⁺, is usually the electron acceptor. The catabolic activity is allosterically activated by ADP and inhibited by GTP. Some organisms (the fungus Neurospora crassa is one example) have two GDH isozymes, an NADP⁺-specific cytosolic enzyme that functions in the direction of glutamate synthesis and an NAD⁺-specific mitochondrial enzyme acting in the catabolic direction to convert excess glutamate into a-ketoglutarate for energy metabolism.

Figure 26.9 · The glutamate dehydrogenase reaction.

     Glutamine synthetase (GS) catalyzes the ATP-dependent amidation of the g-carboxyl group of glutamate to form glutamine. The reaction proceeds via a g-glutamyl-phosphate intermediate, and GS activity depends on the presence of divalent cations such as Mg²⁺ (Figure 26.10). Glutamine is a major N donor in the biosynthesis of many organic N compounds such as purines, pyrimidines, and other amino acids, and GS activity is tightly regulated, as we shall soon see. The amide-N of glutamine provides the nitrogen atom in these biosyntheses. In quantitative terms, GDH and GS are responsible for most of the ammonium assimilated into organic compounds.

Figure 26.10 · The enzymatic reaction catalyzed by glutamine synthetase.
(a) Activation of the g-carboxyl group of Glu by ATP precedes (b) amidation by NH₄⁺.

Pathways of Ammonium Assimilation

Figure 26.11 · The GDH/GS pathway of ammonium assimilatin. The sum of these reactions is the conversion of 1 a-ketoglutarate to 1 glutamine at the expense of 2 NH₄⁺, 1 ATP, and 1 NDAPH.

In organisms that enjoy environments rich in nitrogen, GDH and GS acting in sequence furnish the principal route of NH₄⁺ incorporation (Figure 26.11). However, GDH has a significantly higher K_m for NH₄⁺ than does GS. Consequently, in organisms such as green plants which grow under conditions where little NH₄⁺ is available, GDH is not effective and GS is the only NH₄⁺-assimilative reaction. Such a situation creates the need for an alternative mode of glutamate synthesis to replenish the glutamate consumed by the GS reaction. This need is filled by glutamate synthase (also known as GOGAT, the acronym for the other name of this enzyme - glutamate:oxo-glutarate amino-transferase). Glutamate synthase catalyzes the reductive amination of a-keto-glutarate using the amide-N of glutamine as the N donor:

 Reductant + a-KG + Gln ® 2 Glu + oxidized reductant

Two glutamates are formed¾one from amination of a-ketoglutarate and the other from deamidation of Gln (Figure 26.12). These glutamates can now serve as ammonium acceptors for glutamine synthesis by GS. Organisms variously use NADH, NADPH, or reduced ferredoxin as reductant. Glutamate synthases are typically large, complex proteins; in E. coli, GOGAT is an 800-kD flavoprotein containing both FMN and FAD as well as [4Fe-4S] clusters.

Figure 26.12 · The glutamate synthase reaction, showing the reductants exploited by different organisms in this reductive amination reaction.

     Together, GS and GOGAT constitute a second pathway of ammonium assimilation, in which GS is the only NH₄⁺-fixing step; the role of GOGAT is to regenerate glutamate (Figure 26.13). Note that this pathway consumes two equivalents of ATP and 1 NADPH (or similar reductant) per pair of N atoms introduced into Gln, in contrast to the GDH/GS pathway, in which only 1 ATP and 1 NADPH are used up per pair of NH₄⁺ fixed. Clearly, coping with a nitrogen-limited environment has its cost.

Figure 26.13 · The GS/GOGAT pathway of ammonium assimilation. The sum of these reactions results in the conversion of 1 a-ketoglutarate to 1 glutamine at the expense of 2 ATP and 1 NADPH.

26.3 · Escherichia coli Glutamine Synthetase: A Case Study in Enzyme Regulation

As indicated earlier, glutamine plays a pivotal role in nitrogen metabolism by donating its amide nitrogen to the biosynthesis of many important organic N compounds. Consistent with its metabolic importance, in enteric bacteria such as E. coli, GS is regulated at three different levels:

1. Its activity is regulated allosterically by feedback inhibition.

2. GS is interconverted between active and inactive forms by covalent modification.

3. Cellular amounts of GS are carefully controlled at the level of gene expression and protein synthesis.

Eukaryotic versions of glutamine synthetase show none of these regulatory features.
    Escherichia coli GS is a 600-kD dodecamer (a₁₂-type subunit organization) of identical 52-kD monomers (each monomer contains 468 amino acid residues). These monomers are arranged as a stack of two hexagons (Figure 26.14). The active sites are located at subunit interfaces within the hexagons; these active sites are recognizable in the X-ray crystallographic structure by the pair of divalent cations that occupy them. Adjacent subunits contribute to each active site, thus accounting for the fact that GS monomers are catalytically inactive.

Figure 26.14 · The subunit organization of bacterial glutamine synthetase.
(a) Diagram showing its dodecameric structure as a stack of two hexagons. (b) Molecular structure of glutamine synthetase from Salmonella typhimurium (a close relative of E.coli), as revealed by X-ray crystallographic analysis. (From Almassy, R. J., Janson, C. A., Hamlin, R., Xuong, N.-H., and Eisenberg, D., 1986. Nature 323:304. Photos courtesy of S.-H. Liaw and D. Eisenberg.)

ALLOSTERIC REGULATION OF GS.  Nine distinct feedback inhibitors (Gly, Ala, Ser, His, Trp, CTP, AMP, carbamoyl-P, and glucosamine-6-P) act on GS. Gly, Ala, and Ser are key indicators of amino acid metabolism in the cell; each of the other six compounds represents an end product of a biosynthetic pathway dependent on Gln (Figure 26.15). AMP competes with ATP for binding at the ATP substrate site. Gly, Ala, and Ser compete with Glu for binding at the active site. Carbamoyl-P binds at a site that overlaps both the Glu site and the site occupied by the g-PO₄ of ATP.

Figure 26.15 · The allosteric regulation of glutamine synthetase activity by feedback inhibition.

Figure 26.16 · Covalent modification of GS: adenylylation of Tyr³⁹⁷ in the glutamine synthetase polypeptide via an ATP-dependent reaction catalyzed by the converter enzyme adenylyl transferase (AT). From 1 through 12 GS monomers in the GS holoenzyme can be modified, with progressive inactivation as the ratio of [modified]/[unmodified] GS subunits increases.

COVALENT MODIFICATION OF GS.  Each GS subunit can be adenylylated at a specific tyrosine residue (Tyr³⁹⁷) in an ATP-dependent reaction (Figure 26.16). Adenylylation inactivates GS. If we define n as the average number of adenylyl groups per GS molecule, GS activity is inversely proportional to n. The number n varies from 0 (no adenylyl groups) to 12 (every subunit in each GS molecule is adenylylated). Adenylylation of GS is catalyzed by the converter enzyme ATP:GS:adenylyl transferase, or simply adenylyl transferase (AT). However, whether or not this covalent modification occurs is determined by a highly regulated cycle (Figure 26.17). AT not only catalyzes adenylylation of GS, it also catalyzes deadenylylation¾the phosphorolytic removal of the Tyr-linked adenylyl groups as ADP. The direction in which AT operates depends on the nature of a regulatory protein P_II associated with it. P_II is a 44-kD protein (tetramer of 11-kD subunits): The state of P_II controls the direction in which AT acts. If P_II is in its so-called P_IIA form, the AT:P_IIA complex acts to adenylylate GS. When P_II is in its so-called P_IID form, the AT:P_IID complex catalyzes the deadenylylation of GS. The active sites of AT:P_IIA and AT:P_IID are different, consistent with the difference in their catalytic roles. In addition, the AT:P_IIA and AT:P_IID complexes are allosterically regulated in a reciprocal fashion by the effectors a-KG and Gln. Gln activates AT:P_IIA activity and inhibits AT:P_IID activity; the effect of a-KG on the activities of these two complexes is diametrically opposite (Figure 26.17).
    Clearly, the determining factor regarding the degree of adenylylation, n, and hence the relative activity of GS, is the [Gln]/[a-KG] ratio. A high [Gln] level signals cellular nitrogen sufficiency, and GS becomes adenylylated and inactivated. In contrast, a high [a-KG] level is an indication of nitrogen limitation and a need for ammonium fixation by GS.

Figure 26.17 · The cyclic cascade system regulating the covalent modification of GS.

REGULATION OF GS THROUGH GENE EXPRESSION.  The gene that encodes the GS subunit in E. coli is designated GlnA. The GlnA gene is actively transcribed to yield GS mRNA for translation and synthesis of GS protein only if a specific transcriptional enhancer, NR_I, is in its phosphorylated form, NR_I-P. In turn, NR_I is phosphorylated in an ATP-dependent reaction catalyzed by NR_II, a protein kinase (Figure 26.18). However, if NR_II is complexed with P_IIA, it acts not as a kinase but as a phosphatase, and the transcriptionally active form of NR_I, namely NR_I-P, is converted back to NR_I with the result that GlnA transcription halts. Recall from the foregoing discussion that a high [Gln]/[a-KG] ratio favors P_IIA at the expense of P_IID. Under such conditions, GS gene expression is not necessary.

Figure 26.18 · Transcriptional regulation of GlnA expression through the reversible phosphorylation of NR_I, as controlled by NR_II and its association with P_IIA.

26.4 · Amino Acid Biosynthesis

Organisms show substantial differences in their capacity to synthesize the 20 amino acids common to proteins. Typically, plants and microorganisms can form all of their nitrogenous metabolites, including all of the amino acids, from inorganic forms of N such as NH₄⁺ and NO₃^-. In these organisms, the a-amino group for all amino acids is derived from glutamate, usually via transamination of the corresponding a-keto acid analog of the amino acid (Figure 26.19). In many cases, amino acid biosynthesis is thus a matter of synthesizing the appropriate a-keto acid carbon skeleton, followed by transamination with Glu. The amino acids can be classified according to the source of intermediates for the a-keto acid biosynthesis (Table 26.1). For example, the amino acids Glu, Gln, Pro, and Arg (and, in some instances, Lys) are all members of the a-ketoglutarate family because they are all derived from the citric acid cycle intermediate, a-ketoglutarate. We return to this classification scheme later when we discuss the individual biosynthetic pathways.
    Mammals can synthesize only 10 of the 20 common amino acids (Table 26.2); the others must be obtained in the diet. Those that can be synthesized are classified as nonessential, meaning it is not essential that these amino acids be part of the diet. In effect, mammals can synthesize the a-keto acid analogs of nonessential amino acids and form the amino acids by transamination. In contrast, mammals are incapable of constructing the carbon skeletons of essential amino acids, and so must rely on dietary sources for these essential metabolites. Excess dietary amino acids cannot be stored for future use, nor are they excreted unused. Instead, they are converted to common metabolic intermediates that can be either oxidized by the citric acid cycle or used to form glucose (see Section 26.5).

Figure 26.19 · Glutamate-dependent transamination of a-keto acid carbon skeletons is a primary mechanism for amino acid synthesis. The generic transamination - aminotransferase reaction involves the transfer of the a-amino group of glutamate to an a-keto acid acceptor (see Figure 14.22). The transamination of oxaloacetate by glutamate to yield aspartate and a-ketoglutarate is a prime example.

Transamination

Transamination involves transfer of an a-amino group from an amino acid to the a-keto position of an a-keto acid (Figure 26.19). In the process, the amino donor becomes an a-keto acid while the a-keto acid acceptor becomes an a-amino acid:

 Amino acid₁ + a-keto acid₂ ® a-keto acid₁ + amino acid₂

The predominant amino acid/a-keto acid pair in these reactions is glutamate/a-ketoglutarate, with the net effect that glutamate is the primary amino donor for the synthesis of amino acids. Transamination reactions are catalyzed by aminotransferases (the preferred name for enzymes formerly termed transaminases). Aminotransferases are named according to their amino acid substrates, as in glutamate-aspartate aminotransferase. Aminotransferases are prime examples of enzymes that catalyze double displacement (ping-pong) - type bisubstrate reactions (see Figure 14.22).

The Pathways of Amino Acid Biosynthesis

As indicated in Table 26.1, the amino acids can be grouped into families on the basis of the metabolic intermediates that serve as their precursors.

Figure 26.20 · The pathway of proline biosynthesis from glutamate. The enzymes are (1) g-glutamyl kinase, (2) glutamate-5-semialdehyde dehydrogenase, and (4) D¹-pyrroline-5-carboxylate reductase; reaction (3) occurs nonenzymatically.

The a-Ketoglutarate Family of Amino Acids

Amino acids derived from a-ketoglutarate include glutamate (Glu), glutamine (Gln), proline (Pro), arginine (Arg), and, in fungi and protoctists such as Euglena, lysine (Lys). We discussed the routes for Glu and Gln synthesis when we considered pathways of ammonium assimilation.
    Proline is derived from glutamate via a series of four reactions involving activation, then reduction, of the g-carboxyl group to an aldehyde (glutamate-5-semialdehyde), which spontaneously cyclizes to yield the internal Schiff base, D¹-pyrroline-5-carboxylate (Figure 26.20). NADPH-dependent reduction of the pyrroline double bond gives proline.

A Deeper Look
The Mechanism of the Aminotransferase (Transamination) Reaction
The aminotransferase (transamination) reaction is a workhorse in biological systems. It provides a general means for exchange of nitrogen between amino acids and a-keto acids. This vital reaction is catalyzed by pyridoxal phosphate (see Figures 18.26 and 18.27). The mechanism involves loss of the Ca proton, followed by an aldimine-ketimine tautomerization - literally a "flip-flop" of the Schiff	base double bond from the pyridoxal aldehyde carbon to the a-carbon of the amino acid substrate. This is followed by hydrolysis of the ketimine intermediate to yield the product a-keto acid. Left in the active site is a pyridoxamine phosphate intermediate, which combines with another (substrate) a-keto acid to form a second ketimine. Transaldiminization with a lysine at the active site completes the reaction.
The mechanism of PLP-catalyzed transamination reactions.

     Arginine biosynthesis involves enzymatic steps that are also part of the urea cycle, a metabolic pathway that functions in N excretion in certain animals. Net synthesis of arginine depends on the formation of ornithine. Interestingly, ornithine is derived from glutamate via a reaction pathway reminiscent of the proline biosynthetic pathway (Figure 26.21). Glutamate is first N-acetylated in an acetyl-CoA - dependent reaction to yield N-acetylglutamate (Figure 26.21). An ATP-dependent phosphorylation of N-acetylglutamate to give N-acetylglutamate-5-phosphate primes this substrate for a reduced pyridine nucleotide-dependent reduction to the semialdehyde. N-acetylglutamate-5-semialdehyde then is aminated by a glutamate-dependent aminotransferase, giving N-acetylornithine, which is deacylated to ornithine.

Figure 26.21 · (facing page) The bacterial pathway of ornithine biosynthesis from glutamate. The enzymes are (1) N-acetylglutamate synthase, (2) N-acetylglutamate kinase, (3) N-acetylglutamate-5-semialdehyde dehydrogenase, (4) N-acetylornithine d-aminotransferase, and (5) N-acetylornithine deacetylase. In mammals, ornithine is synthesized directly from glutamate-5-semialdehyde by a pathway that does not involve an N-acetyl block.

     Ornithine has three metabolic roles: (1) to serve as a precursor to arginine, (2) to function as an intermediate in the urea cycle, and (3) to act as an intermediate in Arg degradation. In any case, the d-NH₃⁺ of ornithine is carbamoylated in a reaction catalyzed by ornithine transcarbamoylase. The carbamoyl group is derived from carbamoyl-P synthesized by carbamoyl phosphate synthetase I (CPS-I). CPS-I is the mitochondrial CPS isozyme; it uses two ATPs in catalyzing the formation of carbamoyl-P from NH₃ and HCO₃^- (Figure 26.22). CPS-I represents the committed step in the urea cycle, and CPS-I is allosterically activated by N-acetylglutamate. Because N-acetylglutamate is both a precursor to ornithine synthesis and essential to the operation of the urea cycle, it serves to coordinate these related pathways.

Figure 26.22 · The mechanism of action of CPS-I, the NH₃-dependent mitochondrial CPS isozyme. (1) HCO₃^- is activated via an ATP-dependent phosphorylation. (2) Ammonia attacks the carbonyl carbon of carbonyl-P, displacing P_i to form carbamate. (3) Carbamate is phosphorylated via a second ATP to give carbamoyl-P.

    The product of the ornithine transcarbamoylase reaction is citrulline (Figure 26.23). Ornithine and citrulline are two a-amino acids of metabolic importance that nevertheless are not among the 20 a-amino acids commonly found in proteins. Like CPS-I, ornithine transcarbamoylase is a mitochondrial enzyme. The reactions of ornithine synthesis and the rest of the urea cycle enzymes occur in the cytosol.
    The pertinent feature of the citrulline side chain is the ureido group. In a complex reaction catalyzed by argininosuccinate synthetase, this ureido group is first activated by ATP to yield a citrullyl-AMP derivative, followed by displacement of AMP by aspartate to give argininosuccinate (Figure 26.23). The formation of arginine is then accomplished by argininosuccinase, which catalyzes the nonhydrolytic elimination of fumarate from argininosuccinate. This reaction completes the biosynthesis of Arg.

Figure 26.23 · The urea cycle series of reactions: Transfer of the carbamoyl group of carbamoyl-P to ornithine by ornithine transcarbamoylase (OTCase, reaction 1) yields citrulline. The citrulline ureido group is then activated by reaction with ATP to give a citrullyl-AMP intermediate (reaction 2a); AMP is then displaced by aspartate, which is linked to the carbon framework of citrulline via its a-amino group (reaction 2b). The course of reaction 2 was verified using ¹⁸O-labeled citrulline. The ¹⁸O label (indicated by the asterisk, *) was recovered in AMP. Citrulline and AMP are joined via the ureido *O atom. The product of this reaction is argininosuccinate; the enzyme catalyzing the two steps of reaction 2 is argininosuccinate synthetase. The next step (reaction 3) is carried out by argininosuccinase, which catalyzes the nonhydrolytic removal of fumarate from argininosuccinate to give arginine. Hydrolysis of Arg by arginase (reaction 4) yields urea and ornithine, completing the urea cycle.

The Urea Cycle¾Excretion of Excess-N Through Arg Breakdown

The carbon skeleton of arginine is derived principally from a-ketoglutarate, but the N and C atoms composing the guanidino group (Figure 26.23) of the Arg side chain come from NH₄⁺, HCO₃^- (as carbamoyl-P), and the a-NH₂ groups of glutamate and aspartate. The circle of the urea cycle is closed when ornithine is regenerated from Arg by the arginase-catalyzed hydrolysis of arginine. Urea is the other product of this reaction and lends its name to the cycle. In terrestrial vertebrates, urea synthesis is required to excrete excess nitrogen generated by increased amino acid catabolism - for example, following dietary consumption of more than adequate amounts of protein. Urea formation is basically confined to the liver. Increases in amino acid catabolism lead to elevated glutamate levels and a rise in N-acetylglutamate, the allosteric activator of CPS-I. Stimulation of CPS-I raises overall urea cycle activity because activities of the remaining enzymes of the cycle simply respond to increased substrate availability. Removal of potentially toxic NH₄⁺ by CPS-I is an important aspect of this regulation. The urea cycle is linked to the citric acid cycle through fumarate, a by-product of the action of argininosuccinase (Figure 26.23, reaction 3).

     Lysine biosynthesis in some fungi and in the protoctist Euglena also stems from a-ketoglutarate, making lysine a member of the a-ketoglutarate family of amino acids in these organisms. (As we shall see, the other organisms capable of lysine synthesis¾ namely, bacteria, other fungi, algae, and green plants - use aspartate as precursor.) To make lysine from a-ketoglutarate requires a lengthening of the carbon skeleton by one CH₂ unit to yield a-ketoadipate (Figure 26.24). This addition is accomplished by a series of reactions reminiscent of the initial stages of the citric acid cycle. First, a two-carbon acetyl-CoA unit is added to the a-carbon of a-ketoglutarate to form homocitrate. Then, in a reaction sequence like that catalyzed by aconitase, homoisocitrate is formed from homocitrate. Oxidative decarboxylation (as in isocitrate dehydrogenase) removes one carbon (the original a-carboxyl group of a-ketoglutarate), leaving a-ketoadipate. A glutamate-dependent aminotransferase enzyme then aminates a-ketoadipate to give a-aminoadipate. Next, the d-COO^- group is activated in an ATP-dependent adenylylation reaction, priming this d-COO^- group for reduction to an aldehyde by NADPH. a-Aminoadipic-6-semialdehyde is then reductively aminated by addition of glutamate to its aldehydic carbon in an NADPH-dependent reaction leading to the formation of saccharopine. Oxidative cleavage of saccharopine by way of an NAD⁺-dependent dehydrogenase activity yields a-ketoglutarate and lysine. This pathway is known as the a-aminoadipic acid pathway of lysine biosynthesis. Interestingly, lysine degradation in animals leads to formation of a-aminoadipate by a reverse series of reactions identical to those occurring along the last steps of this biosynthetic pathway.

Figure 26.24 · Lysine biosynthesis in certain fungi and Euglena: the a-aminoadipic acid pathway. Reactions 1 through 4 are reminiscent of the first four reactions in the citric acid cycle, except that the product a-ketoadipate has an additional -CH₂O unit. Reaction 5 is catalyzed by a glutamate-dependent aminotransferase; reaction 6 is the adenylylation of the d-carboxyl of a-aminoadipate to give the 6-adenylyl derivative. Reductive deadenylylation by an NADPH-dependent dehydrogenase in reaction 7 gives a-aminoadipic-6-semialdehyde, which in reaction 8 is coupled with glutamate via its amino group by a second NADPH-dependent dehydrogenase. Oxidative removal of the a-ketoglutarate moiety by NAD⁺-dependent saccharopine dehydrogenase in reaction 9 leaves this amino group as the e-NH₃⁺ of lysine.

The Aspartate Family of Amino Acids

The members of the aspartate family of amino acids include aspartate (Asp), asparagine (Asn), lysine (via the diaminopimelic acid pathway), methionine (Met), threonine (Thr), and isoleucine (Ile).

    Aspartate is formed from the citric acid cycle intermediate, oxaloacetate, by transfer of an amino group from glutamate via an

aminotransferase reaction (Figure 26.25). Like glutamate synthesis from a-ketoglutarate, aspartate synthesis is a drain on the citric acid cycle. As we already saw, the Asp amino group serves as the N donor in the conversion of citrulline to arginine. In Chapter 27, we shall see that this ¾NH₂ is also the source of one of the N atoms of the purine ring system during nucleotide biosynthesis, as well as the 6-amino-group of the major purine, adenine. In addition, the entire aspartate molecule is used in the biosynthesis of pyrimidine nucleotides.

Figure 26.25 · Aspartate biosynthesis via transamination of oxaloacetate by glutamate. The enzyme responsible is PLP-dependent glutamate;aspartate aminotransferase.

    Asparagine is formed by amidation of the b-carboxyl group of aspartate. In bacteria, in analogy with glutamine synthesis, the nitrogen added in this amidation comes directly from NH₄⁺. In other organisms, asparagine synthetase catalyzes the ATP-dependent transfer of the amido-N of glutamine to aspartate to yield glutamate and asparagine (Figure 26.26).

Figure 26.26 · Asparagine biosynthesis from Asp, Gln, and ATP. b-Aspartyladenylate is an enzyme-bound intermediate of asparagine synthetase; Asn, Glu, AMP, and PP_i are products. (Step A) Asp + ATP ® [b-aspartyladenylate] + PP_i. (Step B) [b-Aspartyladenylate] + Gln + H₂O ® Asn + Glu + AMP.

     Threonine, methionine, and lysine biosynthesis in bacteria proceeds from the common precursor, aspartate, which is converted first to aspartyl-b-phosphate and then to b-aspartyl-semialdehyde. The first reaction is an ATP-dependent phosphorylation catalyzed by aspartokinase (Figure 26.27). In E. coli, there are three isozymes of aspartokinase, designated aspartokinases I, II, and III. Each of these isozymes is uniquely controlled by one of the three end-product amino acids (Table 26.3). Thus, the biosynthesis of each of the three amino acids may be independently regulated through controls exerted on the formation or activity of a particular aspartokinase isozyme.

Figure 26.27 · (opposite) Biosynthesis of threonine, methionine, and lysine, members of the aspartate family of amino acids. b-Aspartyl-semialdehyde is a common precursor to all three. It is formed by aspartokinase (reaction 1) and b-aspartyl-semialdehyde dehydrogenase (reaction 2). From here, the pathways diverge. Reduction of b-aspartyl-semialdehyde by homoserine dehydrogenase (reaction 3) gives homoserine, a precursor to threonine and methionine but not lysine. The branch designated by reactions 4 and 5 (catalyzed by homoserine kinase and threonine synthase) gives rise to threonine. The other branch from homoserine (reactions 6 through 9) leads to methionine (the enzymes are, in order, homoserine acyltransferase, cystathionine synthase, cystathionine-b-lyase, and homocysteine methyltransferase). The route to lysine from b-aspartyl-semialdehyde is the so-called diaminopimelate pathway (reactions 10 through 16). Pyruvate is condensed with b-aspartyl-semialdehyde to yield 2,3-dihydropicolinate (reaction 10, dihydropicolinate synthase), which is then reduced by D¹-piperidine-2,6-dicarboxylate dehydrogenase (reaction 11). Succinylation (reaction 12, N-succinyl-2-amino-6-ketopimelate synthase) is accompanied by opening of the ring; amination ensues (reaction 13, succinyl-diaminopimelate aminotransferase), followed by desuccinylation (reaction 14, succinyl-diaminopimelate desuccinylase) to give l-l-a,e-diaminopimelate. Epimerization to the meso form (reaction 15, diaminopimelate epimerase), then decarboxylation (reaction 16, diaminopimelate decarboxylase), yields lysine.

     b-Aspartyl-semialdehyde is formed via NADPH-dependent reduction of aspartyl-b-phosphate in a reaction catalyzed by b-aspartyl-semialdehyde dehydrogenase (Figure 26.27). From here, the pathway of lysine synthesis diverges. The methyl carbon of pyruvate is condensed with b-aspartyl-semialdehyde, and H₂O is eliminated to yield the cyclic compound 2,3-dihydropicolinate (Figure 26.27). Lysine thus must be considered a member of both the aspartate and the pyruvate families of amino acids. Lysine is a feedback inhibitor of this branch-point enzyme. Dihydropicolinate is then reduced in an NADPH-dependent reaction to D¹-piperidine-2,6-dicarboxylate. A series of reactions, including a hydrolytic opening of the piperidine ring, a succinylation, a glutamate-dependent amination, and the hydrolytic removal of succinate, results in the formation of the symmetrical L,L-a,e-diaminopimelate. Epimerization of this intermediate to the meso form, followed by decarboxylation, yields the end product lysine. Because this pathway proceeds through the symmetrical L,L-a,e-diaminopimelate, one-half of the CO₂ evolved in the terminal decarboxylase step is derived from the carboxyl group of pyruvate, and one-half from the a-carboxyl of Asp.
     The other metabolic branch diverging from b-aspartyl-semialdehyde leads to threonine and methionine via homoserine, an analog of serine that is formed by the NADPH-dependent reduction of b-aspartyl-semialdehyde (Figure 26.27) catalyzed by homoserine dehydrogenase. From homoserine, the biosynthetic pathways leading to methionine and threonine separate. To form methionine, the ¾OH group of homoserine is first succinylated by homoserine acyltransferase. Methionine is a feedback inhibitor of this enzyme. The succinyl group of O-succinylhomoserine is then displaced by cysteine to yield cystathionine (Figure 26.27). The sulfur atom in methionine is contributed by a cysteine sulfhydryl. Cystathionine is then split to give pyruvate, NH₄⁺, and homocysteine, a nonprotein amino acid whose side chain is one ¾CH₂¾ group longer than Cys. Methylation of the homocysteine ¾SH via methyl transfer from the methyl donor, N⁵-methyl-THF (see chapters 18 and 27), gives methionine.

    In passing, it is important to note the role of methionine itself in methyl-ation reactions. The enzyme S-adenosylmethionine synthase catalyzes the reaction of methionine with ATP to form S-adenosylmethionine, or SAM (Figure 26.28). SAM serves as a methyl group donor in many methylation reactions, such as the formation of phosphatidylcholine from phosphatidylethanolamine (see Figure 8.5).

Figure 26.28 · The synthesis of S-adenosylmethionine (SAM) from methionine plus ATP, and the role of SAM as a substrate of methyltransferases in methyl donor reactions and in propylamine transfer reactions, as in the synthesis of polyamines.

    The remaining amino acids of the aspartate family are threonine and isoleucine. Threonine, like methionine, is synthesized from homoserine. Indeed, homoserine is the primary alcohol analog of the secondary alcohol Thr. To move this ¾OH from C-4 to C-3 requires activation of the hydroxyl through ATP-dependent phosphorylation by homoserine kinase. As the first reaction unique to Thr biosynthesis, homoserine kinase is feedback-inhibited by threonine. The last step is catalyzed by threonine synthase, a PLP-dependent enzyme (Figure 26.27).

    Isoleucine is included in the aspartate family of amino acids because four of its six carbons derive from Asp (via threonine) and only two come from pyruvate. Nevertheless, four of the five enzymes necessary for isoleucine synthesis are common to the pathway for biosynthesis of valine, so discussion of isoleucine synthesis is presented under the biosynthesis of the pyruvate family of amino acids.

The Pyruvate Family of Amino Acids

The pyruvate family of amino acids includes alanine (Ala), valine (Val), and leucine (Leu). Transamination of pyruvate, with glutamate as amino donor, gives alanine. Because these transamination reactions are readily reversible, alanine degradation occurs via thereverse route, with a-ketoglutarate serving as amino acceptor.

Figure 26.29 · Biosynthesis of valine and isoleucine. The enzymes are (1) threonine deaminase, (2) acetohydroxy acid synthase, (3) acetohydroxy acid isomeroreductase,
(4) dihydroxy acid dehydratase, and (5) glutamate-dependent aminotransferase. Feedback inhibition regulates this pathway: enzyme 1 is isoleucine-sensitive, and enzyme 2 is valine-sensitive.

    Transamination of pyruvate to alanine is a reaction found in virtually all organisms, but valine, leucine, and isoleucine are essential amino acids, and as such, they are not synthesized in mammals. The pathways of valine and isoleucine synthesis can be considered together because one set of four enzymes is common to the last four steps of both pathways (Figure 26.29). Both pathways begin with an a-keto acid. Isoleucine can be considered a structural analog of valine that has one extra ¾CH₂¾ unit, and its a-keto acid precursor, namely, a-ketobutyrate, is one carbon longer than the valine precursor, pyruvate. Interestingly, a-ketobutyrate is formed from threonine by the action of threonine deaminase, an enzyme that both deaminates and dehydrates Thr. Threonine deaminase, a PLP-dependent enzyme, is feedback-sensitive to isoleucine (this enzyme is also known as threonine dehydratase and serine dehydratase). So, part of the carbon skeleton for Ile comes from Asp by way of Thr. From here on, the Val and Ile pathways employ the same set of enzymes. The first reaction involves the generation of hydroxyethyl-thiamine pyrophosphate from pyruvate in a reaction analogous to those catalyzed by transketolase and the pyruvate dehydrogenase complex. The two-carbon hydroxyethyl group is transferred from TPP to the respective keto acid acceptor by acetohydroxy acid synthase (acetolactate synthase) to give a-acetolactate or a-aceto-a-hydroxybutyrate. NAD(P)H-dependent reduction of these a-keto hydroxy acids yields the dihydroxy acids a,b-dihydroxy-isovalerate and a,b-dihydroxy-b-methylvalerate. Dehydration of each of these dihydroxy acids by dihydroxy acid dehydratase gives the appropriate a-keto acid carbon skeletons a-ketoisovalerate and a-keto-b-methylvalerate. Transamination by the branched-chain amino acid aminotransferase yields Val or Ile, respectively (Figure 26.29).

Figure 26.30 · Biosynthesis of leucine. The enzymes are (1) a-isopropylmalate synthase, (2) a-isopropylmalate dehydratase, (3) isopropylmalate dehydrogenase, and (4) leucine aminotransferase. Enzyme 1 is feedback-inhibited by leucine.

     Leucine synthesis depends on these reactions as well, because a-ketoisovalerate is a precursor common to both Val and Leu (Figure 26.30). Although Val and Leu differ by only a single¾CH₂¾ in their respective side chains, the carboxyl group of a-ketoisovalerate first picks up two carbons from acetyl-CoA to give a-isopropylmalate in a reaction catalyzed by isopropylmalate synthase; the enzyme is sensitive to feedback inhibition by Leu (Figure 26.30). Isopropylmalate dehydratase converts the a-isomer to the b form, which undergoes an NAD⁺-dependent oxidative decarboxylation by isopropylmalate dehydrogenase, so that the carboxyl group of a-ketoisovalerate is lost as CO₂. Amination of a-ketoisocaproate by leucine aminotransferase gives Leu.

Figure 26.31 · Biosynthesis of serine from 3-phosphoglycerate. The enzymes are
(1) 3-phosphoglycerate dehydrogenase,
(2) 3-phosphoserine aminotransferase, and
(3) phosphoserine phosphatase.

The 3-Phosphoglycerate Family of Amino Acids

Serine, glycine, and cysteine are derived from the glycolytic intermediate 3-phosphoglycerate. The diversion of 3-PG from glycolysis is achieved via 3-phosphoglycerate dehydrogenase (Figure 26.31). This NAD⁺-dependent oxidation of 3-PG yields 3-phosphohydroxypyruvate¾which, as an a-keto acid, is a substrate for transamination by glutamate to give 3-phosphoserine. Serine phosphatase then generates serine. Serine inhibits the first enzyme, 3-PG dehydrogenase, and thereby feedback-regulates its own synthesis.
     Glycine is made from serine via two related enzymatic processes. In the first, serine hydroxymethyltransferase, a PLP-dependent enzyme, catalyzes the transfer of the serine b-carbon to tetrahydrofolate (THF), the principal agent of one-carbon metabolism (Figure 26.32a). Glycine and N⁵,N¹⁰-methylene-THF are the products. In addition, glycine can be synthesized by a reversal of the glycine oxidase reaction (Figure 26.32b). Here, glycine is formed when N⁵,N¹⁰-methylene-THF condenses with NH₄⁺ and CO₂. Via this route, the b-carbon of serine becomes part of glycine. The conversion of serine to glycine is a prominent means of generating one-carbon derivatives of THF, which are so important for the biosynthesis of purines and the C-5 methyl group of thymine (a pyrimidine), as well as the amino acid methionine. Glycine itself contributes to both purine and heme synthesis.

Figure 26.32 · Biosynthesis of glycine from serine (a) via serine hydroxymethyltransferase and (b) via glycine oxidase.

     Cysteine synthesis is accomplished by sulfhydryl transfer to serine. In some bacteria, H₂S condenses directly with serine via a PLP-dependent enzyme-catalyzed reaction, but in most microorganisms and green plants, the sulfhydrylation reaction requires an activated form of serine, O-acetylserine (Figure 26.33). O-acetylserine is made by serine acetyltransferase, with the transfer of an acetyl group from acetyl-CoA to the -OH of Ser. This enzyme is inhibited by Cys. O-Acetylserine then undergoes sulfhydrylation by H₂S with elimination of acetate; the enzyme is O-acetylserine sulfhydrylase.

Figure 26.33 · Cysteine biosynthesis. (a) Direct sulfhydrylation of serine by H₂S.
(b) H₂S-dependent sulfhydrylation of O-acetylserine.

SULFATE ASSIMILATION.  Given the prevailing oxidative nature of our environment and the reactivity and toxicity of H₂S, the source of sulfide for Cys synthesis merits discussion. In microorganisms and plants, sulfide is the product of sulfate assimilation. Sulfate is the common inorganic form of combined sulfur, and its assimilation involves several interesting ATP derivatives (Figure 26.34). 3'-Phosphoadenosine-5'-phosphosulfate (PAPS) is not only an intermediate in sulfate assimilation; it also serves as the substrate for synthesis of sulfate esters, such as the sulfated polysaccharides found in the glycocalyx of animal cells. The "activated sulfate" of PAPS is reduced to sulfite (SO₃^2-) in a thioredoxin-dependent reaction, and sulfite is then reduced to sulfide (S^2-).

Figure 26.34 · Sulfate assimilation and the generation of sulfide for synthesis of organic S compounds. In reaction 1, ATP sulfurylase catalyzes the formation of adenosine-5'-phosphosulfate (APS) + PP_i. In reaction 2, adenosine-5'-phosphosulfate 3'-phosphokinase catalyzes the reaction of adenosine 5'-phosphosulfate with a second ATP to form 3'-phosphoadenosine-5'-phosphosulfate (PAPS) + ADP. Both enzymes are Mg²⁺-dependent. In reaction 3, PAPS is reduced to sulfite (SO₃^2-) in a thioredoxin-dependent reaction. Thioredoxin is a small (12-kD) protein that functions in a number of biological reductions (see Chapter 27). In reaction 4, sulfite reductase catalyzes the six-electron reduction of sulfite to sulfide. NADPH is the electron donor. Sulfite reductase possesses siroheme as a prosthetic group, the same heme found in nitrite reductase (Figure 26.2), which also catalyzes a six-electron transfer reaction.

Figure 26.35 · Some of the aromatic compounds derived from chorismate.

Biosynthesis of the Aromatic Amino Acids

The aromatic amino acids, phenylalanine, tyrosine, and tryptophan, are derived from a shared pathway that has chorismic acid (Figure 26.35) as a key intermediate. Indeed, chorismate is common to the synthesis of cellular compounds having benzene rings, including these amino acids, the fat-soluble vitamins E and K, folic acid, and coenzyme Q and plastoquinone (the two quinones necessary to electron transport during respiration and photosynthesis, respectively). Lignin, a polymer of nine-carbon aromatic units, is also a derivative of chorismate. Lignin and related compounds can account for as much as 35% of the dry weight of higher plants; clearly, enormous amounts of carbon pass through the chorismate biosynthetic pathway.

Figure 26.36 · The shikimate pathway leading to the synthesis of chorismate. The starting substrates are phosphoenolpyruvate and erythrose-4-phosphate. The enzymes are (1) 2-keto-3-deoxy-d-arabino-heptulosonate-7-P synthase, (2) dehydroquinate synthase (note that the coenzyme NAD⁺ is not altered in this reaction), (3) 5-dehydroquinate dehydratase, (4) shikimate dehydrogenase, (5) shikimate kinase, (6) 3-enolpyruvylshikimate-5-phosphate synthase, and (7) chorismate synthase.

THE SHIKIMATE PATHWAY.  Chorismate biosynthesis occurs via the shikimate pathway (Figure 26.36). The precursors for this pathway are the common metabolic intermediates phosphoenolpyruvate and erythrose-4-phosphate. These intermediates are linked to form 3-deoxy-D-arabino-heptulosonate-7-phosphate (DAHP) by DAHP synthase. Although this reaction is remote from the ultimate aromatic amino acid end products, it is an important point for regulation of aromatic amino acid biosynthesis, as we shall see. In the next step on the way to chorismate, DAHP is cyclized to form a six-membered saturated ring compound, 3-dehydroquinate. A sequence of reactions ensues that introduces unsaturations into the ring, yielding shikimate, then chorismate. Note that the side chain of chorismate is derived from a second equivalent of phosphoenolpyruvate.

PHENYLALANINE AND TYROSINE.  At chorismate, the pathway separates into three branches, each leading specifically to one of the aromatic amino acids. The branches leading to phenylalanine and tyrosine both pass through prephenate (Figure 26.37).

Figure 26.37 · The biosynthesis of phenylalanine, tyrosine, and tryptophan from chorismate. The enzymes are (1) chorismate mutase, (2) prephenate dehydratase, (3) phenylalanine aminotransferase, (4) prephenate dehydrogenase, (5) tyrosine aminotransferase, (6) anthranilate synthase, (7) anthranilate-phosphoribosyl transferase, (8) N-(5'-phosphoribosyl)-anthranilate isomerase, (9) indole-3-glycerol phosphate synthase, (10) tryptophan synthase (a-subunit), and (11) tryptophan synthase (b-subunit).

In some organisms, such as E. coli, the branches are truly distinct because prephenate does not occur as a free intermediate, but remains bound to the bifunctional enzyme that catalyzes the first two reactions after chorismate. In any case, chorismate mutase is the first reaction leading to Phe or Tyr. In the Phe branch, the -OH group para to the prephenate carboxyl is removed by a dehydratase; in the Tyr branch, this -OH is retained and becomes the phenolic -OH of Tyr. Glutamate-dependent aminotransferases introduce the amino groups into the two a-keto acids, phenylpyruvate and 4-hydroxy-phenylpyruvate, to give Phe and Tyr, respectively. Some mammals can synthesize Tyr from Phe obtained in the diet via phenylalanine-4-monooxygenase, using O₂ and tetrahydrobiopterin, an analog of tetrahydrofolic acid, as co-substrates (Figure 26.38).

Figure 26.38 · The formation of tyrosine from phenylalanine. This reaction is normally the first step in phenylalanine degradation in most organisms; in mammals, however, it provides a route for the biosynthesis of Tyr from Phe. (Phenylalanine-4-monooxygenase is also known as phenylalanine hydroxylase.)

TRYPTOPHAN.  The pathway of tryptophan synthesis is perhaps the most thoroughly studied of any biosynthetic sequence, particularly in terms of its genetic organization and expression. The principal stalwart of this research is Charles Yanofsky of Stanford University, and his many original insights represent themes of general significance in metabolic regulation. Synthesis of Trp from chorismate requires six steps (Figure 26.37). In most microorganisms, the first enzyme, anthranilate synthase, is an a₂b₂-type protein, with the b-subunit acting in a glutamine-amidotransferase role to provide the -NH₂ group of anthranilate. Or, given high levels of NH₄⁺, the a-subunit can carry out the formation of anthranilate directly by a process in which the activity of the b-subunit is unnecessary. Furthermore, in certain enteric bacteria, such as E. coli and Salmonella typhimurium, the second reaction of the pathway, the phosphoribosyl-anthranilate transferase reaction, is an activity catalyzed by the a-subunit of anthranilate synthase. PRPP (5-phosphoribosyl-1-pyrophosphate), the substrate of this reaction, is also a precursor for purine biosynthesis (Chapter 27). Phosphoribosyl-anthranilate then undergoes a rearrangement wherein the ribose moiety is isomerized to the ribulosyl form in enol-1-(o-carboxyphenylamino)-1-deoxyribulose-5-phosphate (reaction 8). Decarboxylation and ring closure ensue to yield the indole nucleus as indole-3-glycerol phosphate (indole-3-glycerol phosphate synthase, reaction 9). In the last reaction, serine displaces glyceraldehyde-3-phosphate to give Trp. The enzyme tryptophan synthase is also an a₂b₂-type protein. The a-subunit cleaves indoleglycerol-3-phosphate to form indole and 3-glycerol phosphate. The indole is then channeled directly to the b-subunit, which adds serine in a PLP-dependent reaction.
    X-ray crystallographic analysis of the structure of tryptophan synthase from S. typhimurium reveals that the active sites of the a- and b-subunits of the enzyme, although separated from each other by 2.5 nm, are connected by a hydrophobic tunnel wide enough to accommodate the bound indole intermediate (Figure 26.39). Thus, indole can be transferred directly from one active site to the other without being lost from the enzyme complex and diluted in the surrounding milieu. This phenomenon of direct transfer of enzyme-bound metabolic intermediates, or channeling, increases the efficiency of the overall pathway by preventing loss and dilution of the intermediate. Channeling is a widespread mechanism for substrate transfer in metabolism, particularly among the enzymes of higher organisms.

Figure 26.39 · Tryptophan synthase is an example of a "channeling" multienzyme complex in which indole, the product of the a-reaction catalyzed by the a-subunit, passes intramolecularly to the b-subunit. In the b-subunit, the hydroxyl of the substrate l-serine is replaced with indole via a complicated pyridoxal phosphate - catalyzed reaction to produce the final product, l-tryptophan. The schematic figure shown here is a ribbon diagram of one a-subunit (blue) and neighboring b-subunit (the N-terminal domain of the b-subunit is in orange, C-terminal domain in red). The tunnel is outlined by the yellow dot surface and is shown with several indole molecules (green) packed in head-to-tail fashion. The labels "IPP" and "PLP" point to the active sites of the a- and the b-subunits, respectively, in which a competitive inhibitor (indole propanol phosphate, IPP) and the coenzyme PLP are bound. (Adapted from Hyde, C.C., et al., 1988. "Three-dimensional structure of the tryptophan synthase multienzyme complex from Salmonella typhimurium." Journal of Biological Chemistry 263:17857-17871.)

A Deeper Look
Amino Acid Biosynthesis Inhibitors as Herbicides
Unlike animals, plants can synthesize all 20 of the common amino acids. Inhibitors acting specifically on the plant enzymes that are capable of carrying out the biosynthesis of the "essential" amino acids (i.e., enzymes that animals lack) have been developed. These substances appear to be ideal for use as herbicides because they should show no effect on animals. Glyphosate, sold commercially as Roundup(r), is a PEP analog that acts as a specific inhibitor of 3-enolpyruvylshikimate-5-P synthase (Figure 26.36). Sulfmeturon methyl, a sulfonylurea herbicide that inhibits acetohydroxy acid synthase,	an enzyme common to Val, Leu, and Ile biosynthesis (Figure 26.29), is the active ingredient in Oust(r). Aminotriazole, sold as Amitrole(r), blocks His biosynthesis by inhibiting imidazole glycerol-P dehydratase (Figure 26.40). PPT (phosphinothricin) is a potent inhibitor of glutamine synthetase. Although Gln is a nonessential amino acid and glutamine synthetase is a ubiquitous enzyme, PPT is relatively safe for animals because it does not cross the blood - brain barrier and is rapidly cleared by the kidneys.

HISTDINE.  Like aromatic amino acid biosynthesis, histidine biosynthesis shares metabolic intermediates with the pathway of purine nucleotide synthesis. The pathway involves 10 separate steps, the first being an unusual reaction that links ATP and PRPP (Figure 26.40). Five carbon atoms from PRPP and one from ATP end up in histidine. Step 5 involves some novel chemistry: the substrate, phosphoribulosylformimino-5-aminoimidazole-4-carboxamide ribonucleotide, picks up an amino group (from the amide of glutamine) in a reaction accompanied by cleavage and ring closure to yield two imidazole compounds: the histidine precursor, imidazole glycerol phosphate, and a purine nucleotide precursor, 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR). Note that AICAR as a purine nucleotide precursor can ultimately replenish the ATP consumed in reaction 1. Nine enzymes act in histidine’s 10 synthetic steps. Reactions 9 and 10, the successive NAD⁺-dependent oxidations of an alcohol to an aldehyde and then to a carboxylic acid, are catalyzed by the same dehydrogenase.

Figure 26.40 · The pathway of histidine biosynthesis. The enzymes are (1) ATP-phosphoribosyl transferase, (2) pyrophosphohydrolase, (3) phosphoribosyl-AMP cyclohydrolase, (4) phosphoribosylformimino-5-aminoimidazole carboxamide ribonucleotide isomerase, (5) glutamine amidotransferase, (6) imidazole glycerol-P dehydratase, (7) l-histidinol phosphate aminotransferase, (8) histidinol phosphate phosphatase, and (9) histidinol dehydrogenase.

26.5 · Metabolic Degradation of Amino Acids

In normal human adults, close to 90% of the energy requirement is met by oxidation of carbohydrates and fats; the remainder comes from oxidation of the carbon skeletons of amino acids. The primary physiological purpose of amino acids is to serve as the building blocks for protein biosynthesis. The dietary amount of free amino acids is trivial under most circumstances. However, if excess protein is consumed in the diet or if the amount of amino acids released during normal turnover of cellular proteins exceeds the requirements for new protein synthesis, the amino acid surplus must be catabolized. Also, if carbohydrate intake is insufficient (as during fasting or starvation) or if carbohydrates cannot be appropriately metabolized due to disease (as in diabetes mellitus), body protein becomes an important fuel for metabolic energy.

Figure 26.41 · Metabolic degradation of the common amino acids. The 20 common amino acids can be classified according to their degradation products. Those that give rise to precursors for glucose synthesis, such as a-ketoglutarate, succinyl-CoA, fumarate, oxaloacetate, and pyruvate, are termed glucogenic (shown in pink). Those degraded to acetyl-CoA or acetoacetate are called ketogenic (shown in blue) because they can be converted to fatty acids or ketone bodies. Some amino acids are both glucogenic and ketogenic.

The 20 Common Amino Acids Are Degraded by 20 Different Pathways That Converge to Just 7 Metabolic Intermediates

Because the 20 common amino acids of proteins are distinctive in terms of their carbon skeletons, each amino acid requires its own unique degradative pathway. Because amino acid degradation normally supplies only 10% of the body’s energy, then, on average, degradation of any given amino acid will satisfy less than 1% of energy needs. Therefore, we will not discuss these pathways in detail. It so happens, however, that degradation of the carbon skeletons of the 20 common a-amino acids converges to just 7 metabolic intermediates: acetyl-CoA, succinyl-CoA, pyruvate, a-ketoglutarate, fumarate, oxaloacetate, and acetoacetate. Because succinyl-CoA, pyruvate, a-ketoglutarate, fumarate, and oxaloacetate can serve as precursors for glucose synthesis, amino acids giving rise to these intermediates are termed glucogenic. Those degraded to yield acetyl-CoA or acetoacetate are termed ketogenic, because these substances can be used to synthesize fatty acids or ketone bodies. Some amino acids are both glucogenic and ketogenic (Figure 26.41).

The C-3 Family of Amino Acids: Alanine, Serine, and Cysteine

The carbon skeletons of alanine, serine, and cysteine all converge to pyruvate (Figure 26.42). Transamination of alanine yields pyruvate:

 Alanine + a-ketoglutarate 34 pyruvate + glutamate

Deamination of serine by serine dehydratase also yields pyruvate. Cysteine is converted to pyruvate via a number of paths.

 The carbon skeletons of three other amino acids also become pyruvate. Glycine is convertible to serine and thus to pyruvate. The three carbon atoms of tryptophan that are not part of its indole ring appear as alanine (and, hence, pyruvate) upon Trp degradation. Threonine by one of its degradation routes is cleaved to glycine and acetaldehyde. The glycine is then converted to pyruvate via serine; the acetaldehyde is oxidized to acetyl-CoA (Figure 26.42).

Figure 26.42 · Formation of pyruvate from alanine, serine, cysteine, glycine, tryptophan, or threonine.