Many aspects of the structure of chymosin are quite unique even though structure comparisons indicate a high degree of structural homology with other eukaryotic aspartic proteinases. The structural homology is shown to be directly related to the sequence homology which varies from 30 to 60%. The recent structures of pepsin (Abad-Zapatero et al., 1990; Sielecki et al., 1990; Cooper et al., 1990) have allowed the first preliminary comparisons of two different gastric enzymes. These structures are quite similar, even more so than the structures of the fungal proteinases. However, unlike chymosin, the position of Tyr77 in the flap of pepsin is similar to that found in the fungal aspartic proteinases despite the fact that pepsin is more similar in the flap sequence and the S1 binding site to chymosin than to the fungal proteinases. Attempts at obtaining crystals complexed with substrate analogs which are suitable for diffraction studies have been unsuccessful. Therefore, substrate binding has been examined by model building substrates and substrate analogs into the active site cleft of the structure determined from X-ray studies. The model complexes have been compared with the structures of inhibitor-aspartic proteinase complexes have been previously reported. The results reported here indicate that there are valid reasons why the natural substrate, kappa-casein, binds and is cleaved between positions 105-106. The positively charged histidine residues (98, 100, and 102) of kappa-casein, which are located prior to the cleavage site, appear to be able to interact with negatively charged residues of chymosin which are quite distant from the active site. These residues include Glu288, Asp279, and Glu280 of chymosin. The latter two residues are approximately 20 and 25 A from the center of the active site. These studies also suggest that the difference in activities of the A and B isozymes of chymosin may be due to the increased binding affinity of the substrate as a result of strong electrostatic interactions with Asp244 of chymosin and positively charged His102 of the substrate. An examination of the charged amino acid residues of the chymosin structure has produced two interesting observations. First, there is an asymmetric distribution of charged residues; the N-terminal domain has a smaller net negative charge than the C-terminal domain. This is due to a patch of positive charges on the surface located in the region from residues 48 to 62. Electrostatic calculations in which overall dipole moments were estimated for each of the eukaryotic aspartic proteinases have been performed.(ABSTRACT TRUNCATED AT 400 WORDS)