Types of Structural Regularities in Proteins

The third approach has been followed predominantly in many laboratories. A comparison of differences in the primary structures of insulins, corticotropins, and cytochromes c from different species (see Table I) may serve as typical examples of these efforts. The data from these studies provide evidence of the extensive structural relationship between functionally identical proteins from different species this relationship decreases with the increased difference in species. 

The most valuable confirmation of this view to date is, without doubt, to be found in the known structures of homologous proteins and peptide hormones, that is compounds of identical biological function isolated from different species. As is well known, the primary structures of the homologous insulins, corticotropins, hypertensins, posterior pituitary hormones, and heme peptide sequences from cytochromes c are closely similar and differ only at certain definite sites in the peptide chains. These can, in particular, serve as a useful point of departure in a search for more general principles governing protein structure, and in the comparison of different proteins. 

2 The specific differences in homologous proteins have been reviewed and discussed in several papers see e.g., Anfinsen and Redfield (1956), Vaughan and Steinberg (1959), Gibian (1961), Sorm et al. (1957), Yeas (1961), Sorm (1961b). 

Knichal, 1962). Both approaches indicate that there is no general preference for any particular dipeptide sequence or sequences. It also seems likely that there will be no forbidden tripeptide sequences since of the 8000 possible combinations of three amino acids as many as 1032 had already 

Certain indications for a selectivity of bonding are obtained by a comparison of the occurrence of peptide bonds of the types AB and BA in some individual proteins (Sorm et al, 1961). In a sufficiently long peptide chain the statistical probability of both these combinations should, of comse, be equal in the majority of cases, indeed, the experimental data are in agreement with statistical expectation, but in some instances there are deviations, as in the structure of ribonuclease, or in the a- and (3-chains of hemoglobin where there are differences in the incidence of several dipeptide combinations of the two sequences senses (Table III). 

---