Homogenates of brain and kidney from several mammalian species and of rat liver and spleen were prepared in ice-cold potassium phosphate buffer (0-026 M, #>H 68) with a Potter-Elvehjem glass homogenizer. Aliquots were deproteinized with 1 per cent picric acid (w/v) or 3 per cent sulphosalicylic acid (w/v) and analysed for GABA by partition chromatography or high voltage electrophoresis on filter paper, and by elution chromatography with a Beckman-Spinco a.mino-acid analyser6. Fresh homogenate (100 mg) was also incubated in conditions slightly modified from those described by Roberts and Simonsen7, using L-glutamic acid-U-14C (specific activity, 186 mCi/mmole, New England Nuclear Corp.) or L-glutamic acid-l-14C (20 mCi/mmole, California Biochemical Corp.), both of confirmed radiochemical purity. GAD activity was assayed by measuring 14CO2 released from L-(1-14C)-glutamate. Table 1. GABA CONCENTRATION AND GAD ACTIVITY IN MAMMALIANKIDNEY SpeciesGABAf (moles g) GAD activity t (^moles CO2/100 mg homogenate/120 min) Relative GAD activity Kidney xlOO BrainMan (1) 0-440 0-076 __ Rat (10) 0-037 0-144 17-1Mouse (4) 0-033 0-096 7-4 Guinea-pig (2) 0-078 0-056 5-3Numbers in parentheses indicate the number of kidneys examined; results are the mean. The human tissue was obtained at necropsy 10 h after death; its GABA concentration is comparable with values cited in refs. 4 and 5. t Measured in deproteinized homogenate by elution chromatography on 22 cm column of Beckman-Spinco amino-acid analyser6.J Conditions of reaction. Concentration of materials in final volume: L-12C-glutamate, 3-1 mM; L-(l-14C)-glutamate, 0-00065 mM; pyridoxal-5'-phosphate, 0-236 mM, 0-1 M potassium phosphate buffer pH 6-8. Reaction started by adding 0-6 ml. of kidney homogenate in phosphate buffer, equivalent to about 50 mg wet weight, to 0-2 ml. of reaction mixture. Incubation performed at 37 C for 120 min in cross-arm flasks7 and terminated by addition of 0-2 ml. 8 N H2SO4. 14CO2 was collected in a scintillation flask containing 0-4 ml. hyamine. After addition of scintillation mixture (0-1 g POPOP, 4-0 g PPO in 1 1. of toluene), the radioactivity was measured in a Nuclear Chicago Unilux II counter operating at 80 per cent efficiency. Values are the mean of at least three observations. GABA is present in appreciable amounts in homogenates of human, rat, guinea-pig and mouse kidney (Table 1). The identity of GABA was confirmed by partition and elution chromatography and by two electrophoretic methods8. Kidney homogenates incubated with uniformly 14C-labelled L-glutamic acid formed a radioactive ninhy-drin-positive compound which had the chromatographic properties of GABA8. Homogenates incubated with 1-14C labelled glutamate produced 0-153moles of 14CO2/100 mg homogenate/120 min from 2-97 (moles of endogenous and exogenous substrate; the net synthesis of GABA in paired samples, analysed by ion exchange chromatography, was 0-168 pimoles/100 mg homogenate/120 min. The GAD activity in kidney relative to brain in each of the species. studied is shown in Table 1. The apparent GAD activity in rat liver and spleen relative to brain was also evaluated and found to be 16 per cent and 2 per cent respectively. We investigated some of the characteristics of GAD in rat kidney homogenates. Activity is directly proportional to the amount of homogenate used in the reaction (Fig. la). First-order kinetics are observed when the substrate concentration is varied between 0-38 mM and 3-1 mM; the first approximation of the Km obtained from the Lineweaver-Burk plot is 4 mM (Fig. 1&), a value which is close to 3-0 mM, cited for GAD in acetone extracts of mouse brain and measured by the same assay technique9. Additional characteristics of GAD in brain and kidney of the rat were compared. The pfL optimum is 6-8 in both tissues, a value corresponding to earlier work in rat brain2. The temperature optimum is 37 C in both tissues; however, activity at 37 C compared with 27 C is proportionately higher in brain than in kidney. Removal of pyridoxal-5-phosphate from the incubation medium inhibits brain GAD 81 per cent and kidney GAD 48 per cent; the difference in tissue response may reflect either variation in the amount of endogenous coenzyme present in the two types of homogenate, or possibly an isozymic characteristic for the apoenzyme.GAD activity is more or less confined to the renal cortex; the activity, in relation to wet weight, is less than 1 per cent in renal papilla compared with cortical activity. Activity in the cell is found predominantly in the supernatant fraction in brain and kidney after centrifugation of the homogenate at 100,0000 for 1 h at 0 C. Fig. 1. Rat kidney homogenates were incubated in conditions for GAD assay (see Table 1). a, Activity is directly proportional to amount of homogenate (wet weight) added to reaction mixture, b, Lineweaver-Burk plot depicting Michaelis-Menten kinetics for GAD activity. The apparent Km is 4 mM.Mammalian brain synthesizes GAB A by decarboxyla-tion of glutamic acid, rather than by reverse transamina-tion of succinic acid semialdehyde10. The quantities of GAB A in kidney and the conditions in which it is formed thus indicate that GAD activity is important also in this organ. Because GAD is not found in the renal papilla where anaerobic metabolism prevails, it is likely that the GABA pathway in kidney functions as an oxidative shunt in cortical metabolism11; there is no neuroinhibitory role for GABA in this organ. The GABA pathway could provide an efficient route for glutamate oxidation in conditions of renal acid-base metabolism, where the amide group of glutamine is the principal source of urinary ammonia12. The presence of GAD in kidney is also relevant to an autosomal recessive convulsive disorder in man known as vitamin B6 dependency8'13; the trait is believed to result from a mutation which impairs the binding of pyridoxal-5-phosphate by the GAD apoenzyme. The presence of GAD in kidney may offer an opportunity, through renal biopsy material, to investigate this hypothesis.This work was supported in part by a grant from the Medical Research Council of Canada. D. T. W. is a Queen Elizabeth II Canadian Fund research fellow.