ON THE NATURE OF FREE RADICALS DETECTED BY PARAMAGNETIC RESONANCE IN BIOLOGICAL SUBSTANCES

Observations of weak, relatively sharp electron-spin resonance signals in a variety of lyophilized biological materials have been reported by Commoner, Townsend. and Pake.1 Studies at Duke University2 have shown that similar signals can be produced in numerous organic chemicals, including biological substances, by X-irradiation in oxygen or by the admission of oxygen to samples which had previously been irradiated in a vacuum. These signals produced by oxygen addition have no nuclear hyperfine structure but have a small asymmetry in the g factor of the type expected for an electron localized on a bound oxygen molecule. The signals thus produced are long-lived, usually much more so than the signals formed by irradiation without oxygen. Because similar, long-lived signals can be produced by addition of oxygen to irradiated biological materials, it seems possible that signals like those observed by Commoner, Townsend, and Pakel in unirradiated biological substances might arise from peroxide radicals, ROO, or from semibonded, trapped molecular oxygen. Indeed, this possibility had been suggested earlier by Shields, Ard, and Gordy,' who observed similar signals along with signals from paramagnetic metallic ions in plant materials which had not been lyophilized or irradiated. Peroxide radicals might be built up to detectable proportions by combination of oxygen with radicals which themselves never reach detectable proportions or which have resonances too broad for detection with the usual spectrometers. Detectable paramagnetic oxygen may also be trapped by certain substances or charged groups which are not themselves radicals and do not have an unpaired spin. Resonance of gaseous, untrapped oxygen is difficult to detect under atmospheric conditions because the signals are spread out by interaction of the electronic spin with the molecular rotation and by molecular collisions. Gaseous oxygen would not be observable in our experiments. In the present experiments we have sought to determine whether the signals of the type observed by Commoner, Townsend, and Pakel arise from peroxide radicals or trapped oxygen. We have been able to demonstrate that such signals can indeed be produced by the admission of molecular oxygen to lyophilized-but unirradiated-biological materials such as rat spleen, liver, kidney, or heart. The results strongly suggest that the greater part, if not all, of the natural electron-spin resonance signal strength in such materials results from absorbed oxygen, probably weakly bonded to some as yet unknown element or group in the material. For most measurements, tissue was removed from adult male Sprague-Dawley rats immediately after cervical dislocation and severing the jugular veins, sliced quickly by hand on a glass surface cooled to 00, and put into liquid nitrogen. This