The hydroxyl radical produced in the apoplast has been demonstrated to facilitate cell wall loosening during cell elongation. Cell wall-bound peroxidases (PODs) have been implicated in hydroxyl radical formation. For this mechanism, the apoplast or cell walls should contain the electron donors for (i) H(2)O(2) formation from dioxygen; and (ii) the POD-catalyzed reduction of H(2)O(2) to the hydroxyl radical. The aim of the work was to identify the electron donors in these reactions. In this report, hydroxyl radical (.OH) generation in the cell wall isolated from pea roots was detected in the absence of any exogenous reductants, suggesting that the plant cell wall possesses the capacity to generate .OH in situ. Distinct POD and Mn-superoxide dismutase (Mn-SOD) isoforms different from other cellular isoforms were shown by native gel electropho-resis to be preferably bound to the cell walls. Electron paramagnetic resonance (EPR) spectroscopy of cell wall isolates containing the spin-trapping reagent, 5-diethoxyphosphoryl-5-methyl-1-pyrroline-N-oxide (DEPMPO), was used for detection of and differentiation between .OH and the superoxide radical (O(2)(-).). The data obtained using POD inhibitors confirmed that tightly bound cell wall PODs are involved in DEPMPO/OH adduct formation. A decrease in DEPMPO/OH adduct formation in the presence of H(2)O(2) scavengers demonstrated that this hydroxyl radical was derived from H(2)O(2). During the generation of .OH, the concentration of quinhydrone structures (as detected by EPR spectroscopy) increased, suggesting that the H(2)O(2) required for the formation of .OH in isolated cell walls is produced during the reduction of O(2) by hydroxycinnamic acids. Cell wall isolates in which the proteins have been denaturated (including the endogenous POD and SOD) did not produce .OH. Addition of exogenous H(2)O(2) again induced the production of .OH, and these were shown to originate from the Fenton reaction with tightly bound metal ions. However, the appearance of the DEPMPO/OOH adduct could also be observed, due to the production of O(2)(-). when endogenous SOD has been inactivated. Also, O(2)(-). was converted to .OH in an in vitro horseradish peroxidase (HRP)/H(2)O(2) system to which exogenous SOD has been added. Taken together with the discovery of the cell wall-bound Mn-SOD isoform, these results support the role of such a cell wall-bound SOD in the formation of .OH jointly with the cell wall-bound POD. According to the above findings, it seems that the hydroxycinnamic acids from the cell wall, acting as reductants, contribute to the formation of H(2)O(2) in the presence of O(2) in an autocatalytic manner, and that POD and Mn-SOD coupled together generate .OH from such H(2)O(2).