Treated carbon fiber electrodes were used with differential normal pulse voltammetry (DNPV) for in vivo determination of the relative participation of uric acid (UA) to peak 3 derived between 250-300 mV in the dorsal horn of the spinal cord of anesthetized rats. In vitro, treated carbon fiber electrodes respond linearly over a large range of concentrations of UA (oxidation potential around 250 mV) and 5-hydroxyindoleacetic acid (5-HIAA, oxidation potential around 280-290 mV), but are 3 to 4 times more sensitive to 5-HIAA than to UA. In vivo the question remains as to the exact nature of peak 3 because the difference between oxidation potentials of UA and 5-HIAA is not great enough to permit a separate monitoring of the two compounds. In normal rats, administration of the xanthine oxidase inhibitor allopurinol, produced a progressive decrease of the signal, which reached 64.3% of controls at 120 min (35.6% diminution) after injection, and then plateaued around this value for up to 2 h. The administration of the monoamine oxidase inhibitor (MAOI) clorgyline, produced a classical decay in the voltammograms due to a diminution of endogenous 5-HIAA; however, allopurinol injected 3 h after MAOI gave an additional decrease of peak 3 of about 28%. Finally, in rats pretreated with parachlorophenylalanine (pCPA), the residual peak (32.48% as compared to peak 3 of normal rats taken as 100%), the potential of which is shifted to near that of UA, could be decreased by allopurinol to a level of 9.6% of the peak in control animals.(ABSTRACT TRUNCATED AT 250 WORDS)