A full-length cDNA encoding phenylalanine ammonia-lyase (PAL) from Zea mays L. was isolated and the coding region was expressed in Escherichia coli as a C-terminal fusion to glutathione S-transferase. After purification by glutathione-Sepharose chromatography, the glutathione S-transferase moiety was cleaved off and the resulting PAL enzyme analyzed. In contrast to PAL from dicots, this maize PAL isozyme catalyzed the deamination of both i-phenylalanine (PAL activity) and i-tyrosine (tyrosine ammonialyase activity). These results provide unequivocal proof that PAL and tyrosine ammonia-lyase activities reside in the same polypeptide. In spite of large differences in the Michaelis constant and turnover number of the two activities, their catalytic efficiencies are very similar. Also, both activities have the same p H and temperature optima. These results imply that maize can produce p-coumaric acid from both phenylalanine and tyrosine.PAL (EC 4.3.1.5) catalyzes the elimination of ammonia from L-Phe to yield (E)-cinnamic acid ( Fig. 1) (Koukol and Com, 1961), the first step in the phenylpropanoid pathway (for reviews, see Hahlbrock and Scheel, 1989;Dixon and Paiva, 1995;Whetten and Sederoff, 1995;Campbell and Sederoff, 1996). PAL activity has been found in some fungi and in a11 higher plants analyzed but not in animals. Because of the central function of PAL at a branch point of metabolism, this enzyme is one of the best studied in plants; the same is true for its corresponding gene(s).In a11 plants analyzed, PAL has been shown to occur as a tetramer, and multiple forms have often been isolated (Bolwell et al., 1985). PAL multigene families, such as those in tomato (Lycopersicon esculentum; Lee et al., 1992), parsley (Petroselinum crispum; Appert et al., 1994), or Arabidopsis thaliana (Wanner et al., 1995), explain at least in part these different forms. The occurrence of heteromers is likely but has as yet not been proven. On the other hand, apparent charge isoforms of a poplar PAL were recovered from insect cell cultures producing the protein from a baculovirus expression vector containing the respective cDNA (McKegney et al., 1996).Results from severa1 studies (Neish, 1961; Young et al., 1966;Havir et al., 1971;Jangaard, 1974) indicated that the enzyme from monocots utilizes Tyr (TAL activity) in addition to Phe, whereas the enzyme from dicots utilizes only Phe efficiently. The four homotetrameric parsley PAL (Havir et al., 1971), indicating that PAL from maize also has TAL activity. However, the uncertainty remains that a protein with TAL activity copurifies with PAL activity. Referring to the heterologous expression of parsley PAL cDNAs (Schulz et al., 1989; Appert et al., 1994), Whetten and Sederoff (1995) stated: "The same experiment could be performed with maize cDNA to test for PAL and TAL activities in the same polypeptide." This is exactly the experiment we report here. Expression of a PAL-specific cDNA from maize in Escherichia coli provided the ultimate proof that PAL and TAL activities reside in ...

The present structure reflects the open conformation of the enzyme which is probably stabilized through two residues, a lysine and an arginine, located in the cleft between the domains. Binding of the negatively charged UDPGlcNAc to these residues could neutralize the repulsive force between the two domains, thereby allowing the movement of a catalytically active cysteine residue towards the cleft.

The induced-fit mechanism in Enterobacter cloacae MurA has been investigated by kinetic studies and X-ray crystallography. The antibiotic fosfomycin, an irreversible inhibitor of MurA, induced a structural change in UDP-N-acetylglucosamine (UDPGlcNAc)-liganded enzyme with a time dependence similar to that observed for the inactivation progress. The mechanism of action of fosfomycin on MurA appeared to be of the bimolecular type, the overall rate constants of inactivation and structural change being = 104 M(-1) s(-1) and = 85 M(-1) s(-1), respectively. Fosfomycin as well as the second MurA substrate, phosphoenolpyruvate (PEP), are known to interact with the side chain of Cys115. Like wild-type MurA, the catalytically inactive single-site mutant protein Cys115Ser structurally interacted with UDPGlcNAc in a rapidly reversible reaction. However, in contrast to wild-type enzyme, binding of PEP to mutant protein induced a rate-limited, biphasic structural change. Fosfomycin did not affect the structure of the mutant protein. The crystal structure of unliganded Cys115Ser MurA at 1.9 A resolution revealed that the overall conformation of the loop comprising residues 112-121 is not influenced by the mutation. However, other than Cys115 in wild-type MurA, Ser115 exhibits two distinct side-chain conformations. A detailed view on the loop revealed the existence of an elaborate hydrogen-bonding network mainly supplied by water molecules, presumably stabilizing its conformation in the unliganded state. The comparison between the known crystal structures of MurA, together with the kinetic data obtained, suggest intermediate conformational states in the MurA reaction, in which the loop undergoes multiple structural changes upon ligand binding.

UDP-N-acetylglucosamine enolpyruvyl transferase (MurA) and 5-enolpyruvylshikimate 3-phosphate synthase (EPSPS) have both a unique three-dimensional topology and overall reaction mechanism in common. In the case of MurA, the substrate-free, unliganded protein exhibits an "open" conformation. Upon binding of substrates, the protein forms a much more tightly packed so-called "closed" form following an induced fit mechanism. In this closed form, the substrates are properly positioned for catalysis. On the basis of the structural and mechanistic similarities of MurA and EPSPS, a similar conformational change is likely to occur in EPSPS to generate a catalytically competent active site. However, there is currently little experimental evidence available to support the occurrence of such a conformational change in EPSPS. Using limited tryptic digestion of MurA,(1) it could be shown that formation of the "closed" conformation of MurA is accompanied by a marked increase of stability toward proteolytic degradation. Formation of the closed conformation was achieved by addition of either an excess of both substrates or the sugar nucleotide substrate in conjunction with the antibiotic fosfomycin. Analysis of the MurA tryptic fragments by MALDI-TOF mass spectrometry demonstrates that the protection of the protein in either case is caused by (1) a specific shielding of regions thereby becoming less accessible as a result of the conformational change, and (2) an unspecific overall protection of the whole protein due to an apparently reduced flexibility of the peptide backbone in the binary and ternary complexes. The establishment of methods to describe the effects of tryptic digestion on MurA under various conditions was then extended to EPSPS. Although EPSPS was found to be much more stable toward proteolysis than MurA, the presence of shikimate 3-phosphate (S3P) and the inhibitor glyphosate led to a pronounced suppression of proteolytic degradation. When unliganded EPSPS was treated with trypsin, three of the peptide fragments obtained could be identified by mass spectrometry. Two of these are located in a region corresponding to the "catalytic" loop in MurA which participates in the conformational change. This indicates a conformational change in EPSPS, similar to the one observed in MurA, leading to the protection mentioned above. Corroborating evidence was obtained using a conformational sensitive monoclonal antibody against EPSPS which showed a 20-fold reduced affinity toward the protein complexed with S3P and glyphosate as compared to the unliganded enzyme.