Protein hexapeptide folds

Fig. 7.2 Tlie crystal structure of mammalian Ser/Thr protein phosphatase-1, complexed with the toxin mycrocystin was determined at 2.1 A resolution. PPl has a single domain with a fold, distinct from that of the protein tyrosine phosphatases. The Ser/Thr protein phosphatase-1, is a metalloenzyme with two metal ions positioned at the active site with the help of a p-a-p-o-p scaffold. A dinuclear ion centre consisting of Mn2+ And Fe2+ g situated at the catalytic site that binds the phosphate moiety of the substrate. Ser/Thr phosphatases, PPl and PP2A, are inhibited by the membrane-permeable ocadaic acid and by cyclic hexapeptides, known as microcystins. The toxin molecule is depicted as a ball-and-stick structure. On the left and on the ri t, two different views of the same molecule are shown. Microcystin binds to three distinct regions of the phosphatase to the metaLbinding site, to a hydrophobic groove, and to the edge of a C-terminal groove in the vicinity of the active site. At the surface are binding sites for substrates and inhibitors. These ribbon models are reproduced vnth permission of the authors and Nature from ref. 9.

Protein segments adopt only a finite number of conformations in folded proteins.This limited dictionary of templates may be due to the inherent steric constraints of the backbone and other packing and electrostatic interactions in protein folding. This was most graphically demonstrated by Jones and Thirup. They showed that the majority of the polypeptide backbone of a new structure can be built up from component pieces of other structures, whether or not they are related. Unger et al. showed, in fact, that 76% of the main chain structure of an 82 protein data set can be constructed using only 100 unique hexapeptide templates. Amino acid side chains are also found in a restricted set of conformers. Recent work in our laboratory has shown that a set of four tetrapeptide templates is sufficient to define almost all loop conformations. 

The folding patterns of hexapeptide fragments have been determined from the atomic coordinates of the protein structures using the standard procedure of Kabsch and Sander [66]. Every residue is given a secondary structure descriptor. To reduce complexity, residues were classified as helical h (G, H or I in Kabsch-Sander nomenclature), extended strand s (E), P turn t (T) or remained unassigned n. With these four structural classes there are still 4096 possible folding patterns for hexapep-... 

The type of information that can be retrieved from this database is best illustrated by two characteristic examples shown in Ikble 17.2. The first example compiles hexapeptides related to the C-terminal part of the first and main a helix in the bacterial protein barnase (Ikble 17.2a), the structure of which is known from X-ray analysis [67], but is not contained in our database. Of over 30,000 related peptides generated, nine could be found in the database. They are moderately similar to the query peptide as reflected by rms distances greater than 0.1 and similarity indices below 0.8. Seven of the nine related peptides adopt full or partial helical folds in their respective protein structures, whereas only two less related peptides are found in partially extended conformations. This highly consistent picture is also borne out by the cumu-... 

Every hexapeptide in the database was selected in turn as a query peptide and its fold type predicted from sets of related peptides in the database, excluding the protein containing the query peptide. For over 90% of all hexapeptides at least one related peptide in another protein structure could be found in the database. With the highest cumulative score as fold type predictor, a sobering 40% of all predictions turn out to be correct, as is evident from Table 17.3. From the data shown in Table 17.3 a, a hidden bias towards prediction of helical and non-classified folds is apparent, whereas extended conformations and turns are underestimated. This was compensated for by modifying the pattern weights in the database by factors of 0.82, 1.06,... 

In each separate thread there are 10-nm-wide microfibrils bound into fibrillary bands of up to 2 000 nm in width. The protein chains are present in the microfibrils as folded chains. These so-called silk fibroins can be separated with chymotrypsin into an X-ray crystalline (60% by wt) and an amorphous part. The crystalline portion consists of uniform hexapeptide units (-ser-gly-ala-gly-ala-gly-). In fibroins from Bombyx mori L. there are ten of these hexapeptides. The 60 a-amino acid residues of the crystalline part are joined together with 33 amino acid units of the amorphous component in one peptide chain. The amino acids of the amorphous part vary widely in composition and sequence length. 

---