Modeling the effects of proteins on pH in plasma.

Stewart's model of plasma acid-base balance (Can. J. Physiol. Pharmacol. 61: 1444-1461, 1983) has three weaknesses in the treatment of weak acids: 1) the combination of all weak acids into one entity, 2) inappropriate chemistry for the protein combination with H+, and 3) undocumented values for the dissociation parameters. The present study models serum albumin acid-base properties by fixed negative charges and the association of H+ with the imidazole side chain of histidine. This model has three parameters: 1) the net negative fixed charge (21 eq/mol), 2) the number of histidine residues (16/mol), and 3) the association constant for the imidazole side chain (1.77 x 10(-7) eq/l), all determined from published values. The model was compared with that of Figge, Mydosh, and Fencl (J. Lab. Clin. Med. 120: 713-719, 1992) and with the pH data of Figge, Rossing, and Fencl (J. Lab. Clin. Med. 117: 453-467, 1991). The predictions of pH were excellent, comparable to those found by Figge, Mydosh, and Fencl. The model has the advantages that its structure and parameter values are supported by the literature and that the acid-base effects of factors modifying protein can be investigated.