Use of the direct linear plot to estimate, in an immunoglobulin solution, the proportion which is specific monoclonal antibody against an unknown cell surface antigen.

A Cornish-Bowden direct linear plot was used to assess the level of specific monoclonal antibody in a protein A preparation from mouse ascitic fluid. With this method, experimental observations are plotted directly as lines in parameter space and estimates of kinetic parameters are read directly from the plot without need for further calculation. This method is particularly well suited for the analysis of systems, such as the one outlined here, where conventional kinetic analysis is not possible because the preparation contains both specific and non-specific antibody. Results obtained with the direct linear plot showed that estimates of the level of specific immunoglobulin present in a given protein A preparation are consistent, not only within individual experiments, but also throughout a series of experiments using more than one labelled antibody preparation.

The influence of calcium ions on the conformation of fibrinogen fragment D: the use of chemical cross-linking agents.

Fibrinogen fragment D prepared in the presence of calcium ions (fragment D[Ca++]) shows qualitatively similar cross-linking patterns with dimethyl suberimidate, dimethyl adipimidate and tetranitromethane. Fragment D prepared in the presence of EDTA (fragment D[EDTA]) gives a consistently different pattern with these reagents. In the case of fragment D[EDTA] there is much more intermolecular cross-linking suggesting that the loss of the C-terminus of the gamma-chain remnant results in fragment D adopting a more open conformation. Neither the addition of 2M urea nor EDTA to fragment D[Ca++] alters its cross-linking pattern suggesting that the proposed conformational change follows cleavage of a plasmin susceptible bond which is normally protected by the presence of calcium ions.

Mapping epitope characteristics on carcinoembryonic antigen.

A method of epitope analysis is described in which the binding of one monoclonal antibody (MAb) to radiolabeled carcinoembryonic antigen (CEA) competes with the subsequent binding of an immobilised second MAb. From the degree of blocking obtained, we have identified both structurally related and independent epitopes on CEA. Using this technique to study fifteen MAbs, we have been able to recognise at least 6 unrelated epitopes of the CEA glycoprotein. Further characterisation of these epitopes was accomplished by means of immunohistochemistry. Of the fifteen MAbs, 6 were specific for CEA and reacted with at least 3 unrelated regions of the glycoprotein. Of the remaining 9 MAbs, 2 cross-reacted with erythrocytes, 5 with components of liver and 7 with polymorphonuclear neutrophils. Cross-reactions with liver were varied showing differential antibody specificity for bile canaliculi, Kupffer cells and bile duct epithelium. A high degree of correlation between epitope relatedness and immunohistochemical specificity was found. Two CEA-specific and 4 cross-reactive MAbs were also shown to react with ion-sensitive sites on the CEA glycoprotein.