Measurement of the spin concentration of metalloprotein samples from saturation-magnetization data with particular...

A protocol for obtaining high-quality saturation-magnetization data from metalloprotein samples, employing a superconducting quantum interference device (SQUID) magnetometer, has previously been reported [E. P. Day, T. A. Kent, P. A. Lindahl, E. Münck, W. H. Orme-Johnson, H. Roder and A. Roy (1987) Biophys. J. 52, 837-853 and E. P. Day (1993) Methods Enzymol. 227, 437-463]. Following studies of several dozen different metalloprotein derivatives, the methodology has been further refined, particularly in the area of sample preparation. The details of the sample-handling procedures now in use are described, and moreover, the critical issue of verifying that contamination by paramagnetic impurities remains insignificant is considered. Importantly, it is shown that an independent determination of the quantity of paramagnetic sample present in the magnetometer is undesirable. Much more reliable parameters concerning the ground-state magnetic properties of the system under study are obtained if enough saturation-magnetization data are collected to enable the spin concentration to be determined during the subsequent fitting procedure. As proof of the validity of this method, the results of magnetization studies on ferricytochrome c, ferrocytochrome c and the benzohydroxamic acid adduct of horseradish peroxidase are presented. The ability of saturation-magnetization measurements to routinely determine spin concentration to within +/- 4% of accepted values is firmly established. In addition, a saturation-magnetization study has been performed on resting and fully reduced derivatives of cytochrome c oxidase. These results provide an illustration of the usefulness of the technique in probing some systems which have proved difficult to study by other methods. The increased difficulties inherent in obtaining meaningful data from these cytochrome c oxidase and other integer spin systems are delineated.