October 19, 2006, Third Annual PhysChem Symposium, Obernai, France

Predicting the Isoelectric Point of Modified Proteins

Leif Nørskov-Lauritsen

Abstract

The isoelectric point (pI) of a protein can be defined as the pH at which the net charge of the molecule is zero because the positively and negatively charged amino acids have been titrated to balance each other. Calculating the isoelectric point is fairly straightforward provided one knows the pK_a values of the involved amino acids. Traditionally, one has done this by assigning a standard pK_a value to each ionisable amino acid[1]. This simplification ignores the fact that the pK_a value of one amino acid depends on the protonation state of its neighbours. The present work does not tackle that problem. Instead, it addresses the presence of non-natural amino acids or other modifications to a protein for which one does not have readily available pK_a values.

Basically, the approach we have developed is to dissect each ionisable group from the protein and calculate its pK_a using ACD/pK_a DB. With a small program, all the pK_a values are then combined to derive a titration curve (net charge vs. pH) from which the pI is finally estimated.

This procedure offers a simple way of handling anything from natural modifications of proteins, such as glycosylation or a disulphide bridge formation to complicated organic chemistry linkers. The talk will also illustrate some of the limitations of the method.

Download the presentation in MS PowerPoint (1.12 Mb ZIP file) or Adobe Acrobat format (354 Kb PDF file).