Protein crystallization peptide bonds

An analysis of metal binding to peptide carbonyl groups (Chakrabarti, 1990), mainly calcium ions in protein crystal structures, shows that the cations tend to lie in the peptide plane near the C=0 bond direction. Generally, this binding occurs in turns in proteins or in regions with no regular secondary structures. Ca---0 distances range from 2.2 to 2.5 A, and metal ions do not deviate by more than 35° from the peptide plane. Thus, metal ions in proteins do not, Chakrabarti observed, bind in lone-pair directions. 

The enzyme consists of a single polypeptide chain of Mr 13 680 and 124 amino acid residues.187,188 The bond between Ala-20 and Ser-21 may be cleaved by subtilisin. Interestingly, the peptide remains attached to the rest of the protein by noncovalent bonds. The modified protein, called ribonuclease S, and the native protein, now termed ribonuclease A, have identical catalytic activities. Because of its small size, its availability, and its ruggedness, ribonuclease is very amenable to physical and chemical study. It was the first enzyme to be sequenced.187 The crystal structures of both forms of the enzyme were solved at 2.0-A resolution several years ago.189,190 Subsequently, crystal structures of many complexes of the enzyme with substrate and transition analogues and products have been solved at very high resolution.191 Further, because the catalytic activity depends on the ionizations of two histidine residues, the enzyme has been extensively studied by NMR (the imidazole rings of histidines are easily studied by this method—see Chapter 5). 

The structures of the basic building blocks of the architecture of proteins were determined by Linus Pauling and R. B. Corey many years before the solution of the structures of globular proteins.13 They solved the structures of crystalline small peptides to find the dimensions and geometry of the peptide bond. Then, by constructing very precise models, they found structures that could fit the x-ray diffraction patterns of fibrous proteins. The diffraction patterns of fibers do not consist of the lattice of points found from crystals, but a series of lines corresponding to the repeat distances between constantly recurring elements of structure. 

The x-ray diffraction studies of the crystals of small peptides showed that the peptide bond is planar and trans (anti) (Figure 1.3). The same structure has been found for all peptide bonds in proteins, with a few rare exceptions. This planarity results from a considerable delocalization of the lone pair of electrons of the nitrogen onto the carbonyl oxygen. The C—N bond is consequently shortened, and it has double-bond character (equation 1.2). Twisting of the bond breaks it and loses the 75 to 90 kJ/mol (18 to 21 kcal/mol) of delocalization energy. 

Much uncertainty reigned over the nature of proteins, the best known of which were hemoglobin, the digestive enzymes, and later, insulin. Properties of individual amino acids and the peptide bond were studied early in this century, but it was not until urease was crystallized by Sumner1 in 1926, followed by the isolation of other pure enzymes, that it was finally accepted in the 1930s that enzymes were proteins and that their catalytic properties were not the function of some adsorbed low molecular weight entity. Somewhat later, towards the end of the 1930s, coenzymes were isolated and their roles established. 

The serine proteases are a dass of proteolytic enzyme (they catalyze the hydrolysis of either ester or peptide bonds in proteins) that require an active site residue for covalent catalysis. The active site residue, the catalytic Ser-195, is particularly activated by hydrogen-bonding interactions with His-57 and Asp-102. Crystal structures show that Ser-195, His-57, and Asp-102 are dose in space. Together these three residues, which are located in the substrate binding (SI) pocket, form the famed catalytic triad of the serine proteases. In humans and mammals serine proteases perform many important functions, especially the digestion of dietary protein, in the blood-dotting cascade, and in the complement system ... 

Recently, the crystal structure of S-protein complexed with the model peptide has been solved to moderate resolution (3 A) (Taylor et al., 1985). Most of the structural features envisioned in the design of the model peptide were indeed observed in the structure of the complex. The peptide is in a helical conformation, the histidine is held in a reasonable orientation for catalysis, and the complex is stabilized by nonbonded interactions between the hydrophobic cleft of S-protein and the side chains of Phe-8 and Met-13 of the peptide. There were also a number of subtle differences between the structures of the native and the model S-protein S-peptide complexes. Most notably, the N terminus of the peptide has undergone a major reorientation that prevents Glu-2 from forming a hydrogen bond with Arg-10. Further, the 8-nitrogen of the active-site... 

Figure 10.1 Basic polypeptide geometry. The upper panel shows a short peptide sequence of three amino acids joined by two peptide bonds. A relatively rigid planar structure, indicated by dashed lines, is formed by each peptide bond. The relative positions of two adjacent peptide bond planes is determined by the rotational dihedral angles

In the peptide bond, y can have positive and negative values because the Ca atoms distinguish between each side of the peptide plane. This is not so in the small molecule survey where a number of different types of molecules were considered. Therefore, absolute values should be compared, 18(9)° in proteins and 30° to 80° in small molecule crystal structures. 

Proteolytic enzymes, such as the serine proteases, are among the best characterized of all enzymes.They are important in digestive processes because they break down proteins. They each catalyze the same type of reaction, that is. the breaking of peptide bonds by hydrolysis. The crystal structures of several serine proteases have been determined, and the mechanism of hydrolysis is similar for each. The specificity of each enzyme is, however, different and is dictated by the nature of the side chains flanking the scissile peptide bond (the bond that is broken in catalytic mechanism. Chymotrypsin is one of the best characterized of these serine proteases. The preferred substrates of chymotrypsin have bulky aromatic side chains. The crystal structure determination of the active site of chymotrypsin, illustrated in Figure 18.12, has provided much of the information used to elucidate a plausible mechanism of action of the enzyme. In the first step of any catalyzed reaction, the enzyme and substrate form a complex, ES, the Michaelis complex. The hydrolysis of the peptide bond by chymotrypsin involves three amino acid residues,... 

---