Gradient elution chromatography is frequently used for the preparative separation of peptides and proteins. Separations at high loadings are often avoided because peaks become asymmetrical and saturate the detector. However, non-linear interactions which become important at high loadings may actually improve the separation with greater concentrations of the product being extracted. In this study, the separation of a mixture of two chemotactic peptides N-formyl-met-phe (X~phe) and N-formyl-met-trp (X~trp) was considered using reversed-phase (RP) chromatography. These runs were limited by the solubility of the peptides. The effects of solubility and feed conditions on the chromatograms were examined. Because of the poor solubility of the peptides in low organic solvent concentration, runs where feed conditions are different from the column inlet conditions were examined. Since these two components exhibit low selectivity on the RP column, separation is difficult. Productivities were obtained and compared for the preparative purification of these compounds using isocratic, stepwise and gradient elution with acetonitrile/buffer and methanol/water mobile phase conditions. Selectivity reversal was present in methanol/water conditions. Selectivity reversal implies that the order of retention of the feed compound changes as the mobile phase composition varies. Curved gradient shapes were used in an attempt to improve the separation under selectivity reversal limitation. Using a combination of solubility methods and non-linear interactions at high loadings, simultaneous concentration and purification was achieved. / Graduation date: 1998

Fujinami sarcoma virus encodes a 140/000 m.w. polypeptide (P140) which has been correlated as the agent of transformation in host chicken fibroblasts and mammalian fibroblasts. To conclusively identify the role of P140 in the transformation process it will be necessary to obtain intact/ purified P140. The availability of an antibody monoclonally specific to the N-terminal gag encoded portion of P140 suggested a one-step immunoaffinity purification of P140. After purification of the antibody out of mouse ascites fluid, by 50% ammonium sulfate fractionation and ion exchange chromatography, antibody was linked to a Sepharose 4-B matrix activated with cyanogen bromide. The anti-pl9 affinity matrix bound intact P140 as a doublet relative to a polyclonal anti-pl9. Chaotropic agents, high pH and low pH treatments all failed to elute the bound P140 from the affinity matrix. Failing the purification of intact P140 a method of partial proteolysis was used to produce varying sized fragments of P140/ with the goal of purifying these fragments for further work on the role of P140. Trypsin alone in a limited proteolysis produced small, unstable peptides too close in size distribution to be effectively purified. Chymotrypsin alone produced a broad range of more stable peptides, with a predominance of a 45,000 m.w. peptide. Chymotrypsin-trypsin consecutive proteolysis produced a very stable 35,000 m.w. peptide. Gel filtration of the chymotryptic peptides was ineffective as the peptides coraplexed and were not fractionated. Ion exchange chromatography fractionated the complexing chymotryptic peptides, making possible the purification of these peptides. The stable 45,000 m.w. peptide retained some kinase activity, as it phosphorylated the substrate enolase, similar to but less intense than intact P140. A 30,000 m.w. peptide only phosphorylating after ion exchange did not phosphorylate enolase. / Science, Faculty of / Microbiology and Immunology, Department of / Graduate