sign in sign up
<<
Home , Guanine, Guanine deaminase

J. Bioche,n. 91, 167-176 (1982)

Guanine Deaminase from Rat Brain. Purification, Characteristics,

and Contribution to Ammoniagenesis in the Brain

Syuzo MIYAMOTO, Hirofumi OGAWA,1 Hiroshi SHIRAKI, and Hachiro NAKAGAWA2

Division of , Institute for Protein Research, Osaka University, Suita, Osaka 565

Received for publication, June 18, 1981

1. deaminase [EC 3.5.4.3] was purified to a homogeneous state from rat brain by a procedure involving ammonium sulfate fractionation, DE-52 column chromatography, hydroxylapatite column chromatography, gel filtration on ACA-34 and isoelectric focusing. Homogeneity was shown by polyacrylamide gel electro phoresis in the presence and absence of SDS. 2. The molecular weight of the was determined by gel filtration (105,000), and that of its subunit by SDS-polyacrylamide gel electrophoresis (52,000). From these findings, we concluded that the native enzyme consisted of two identical sub- units. 3. The Km values for guanine and 8-azaguanine were calculated to be 0.17 mM and 0.67 mM, respectively. This enzyme was markedly inhibited by 5-amino 4-

imidazolecarboxamide (AICA), a precursor of synthesis, with a Ki value of 82 ƒÊM. 4. was purified from rat liver by the procedure used for purifi cation of the brain enzyme and anti-liver enzyme serum was raised in rabbits. This antiserum cross-reacted with the brain enzyme without spur formation in the Ouch terlony double diffusion test. 5. The ƒÁ-globulin fraction of the anti-guanine deaminase serum and AICA inhibited more than 50% of the ammoniagenesis in the brain system in which the purine nucleotide cycle operates. It was also shown that guanine were degraded via and guanine liberating in the same brain system when sub strates for the purine nucleotide cycle were omitted. On the basis of these findings it is suggested that guanine deaminase contributes to ammoniagenesis in the brain.

1 Present address: Department of , Medical School, Toyama Medical and Pharmaceutical University, Sugitani, Toyama 930-01. 2 To whom reprint requests should be sent. Abbreviations: SDS, sodium dodecyl sulfate; AICA, 5-amino 4-imidazolecarboxamide; AICAR, 5-amino 4- imidazolecarboxamide .

Vol. 91, No. 1, 1982 167 168 S. MIYAMOTO, H. OGAWA, H. SHIRAKI, and H. NAKAGAWA

The activity of guanine deaminase [guanine amino adenylate kinase (rabbit muscle) [EC 2.7.4.3], and , EC 3.5.4.3], catalyzing the conversion cytochrome c (horse heart) were from Boehringer, of guanine to and ammonia, is especially W. Germany. DEAE-cellulose (DE-52) was from high in the brain and liver, although it is also Whatman Ltd., England. Hydroxylapatite was detected in various other organs of animals (1). from Clarkson Chemical Co., U.S.A. Ultrogel However, the physiological role of this enzyme is ACA-34 and Ampholines (pH range, 4-6 and 3- unknown. 10) were from LKB, Sweden. Freund's complete It was found that the ammonia concentration and incomplete adjuvants were products of Difco, in the brain increases on deprivation of paradoxical U.S.A. TSK-GEL G3000SW silica gel prepacked sleep (2), and decreases during sleep. These find columns for high pressure liquid chromatography ings suggested that the ammonia concentration were obtained from Toyo Soda. Other chemicals reflects the state of functional activity of the brain. were of analytical reagent grade. Many investigators have examined the sources of Enzyme Assays-Guanine deaminase was as ammonia in the brain to elucidate its correlation sayed by the method of Roush and Norris (4), with brain function. Recently, Schultz and spectrophotometric determination of decrease in

Lowenstein suggested that the purine nucleotide guanine concentration. The reaction mixture con cycle accounts for at least 50% of the ammo tained enzyme solution, 0.16 mM guanine and 83 niagenesis occurring endogenously in brain slices mM Tris-HCI buffer, pH 8.0, in a final volume of (3). This raises the questions of what are the 1.2 ml. After preincubation for 10 min at 37•Ž, other sources of ammonia and of whether some the reaction was started by adding guanine, and ammonia formed in the cycle is derived from after incubation for 20 min it was stopped by guanine nucleotide, because GTP, a substrate of adding 0.3 ml of 15% HClO4. The absorbance of the cycle, is degraded to GDP by adenylosuccinate the supernatant obtained by centrifugation was synthetase [EC 6.3.4.4] during the operation of the measured at 248 nm. When nucleotides and cycle and a series of converting the latter related compounds that interfere with the spectro

to guanine is thought to be present in the brain. photometric assay were present in the reaction We thought that some specific inhibitor(s) of the mixture, the activity was measured by determina enzyme, such as anti-guanine deaminase serum or tion of ammonia by the enzymatic method of Kun

other chemical compounds, would be useful for (5). The sample (deproteinized and neutralized) solving these problems. Therefore, we attempted was added to the reaction medium consisting of to purify the enzyme from rat brain. 0.3 M Tris-HCl buffer, pH 8.0, 30 mM ƒ¿-ketoglu This paper reports the purification and tarate, 0.36 mm NADH, and 12 units of glutamate characteristics of guanine deaminase of rat brain. dehydrogenase in a final volume of 1.5 ml. The The inhibitory effects of AICA, a specific inhibitor reaction was started by adding glutamate dehydro

of guanine deaminase, and anti-guanine deaminase genase and allowed to proceed at 25•Ž for 90 min. serum on ammoniagenesis in a rat brain system The decrease in absorbance at 340 nm due to are also reported. oxidation of NADH by ammonia was measured. One unit of enzyme activity was defined as the amount catalyzing the formation of 1 ƒÊmol of MATERIALS AND METHODS ammonia per min under the standard assay con Materials-Male Wistar strain rats, weighing ditions; this is equivalent to a 1.17 decrease in 200 to 300 g, were killed by decapitation. The absorbance at 248 nm per min. brain and liver were removed and stored at -70•Ž Protein concentration was determined by the until use. method of Lowry et al. (6) with bovine serum NADH, AICA, AICAR, guanine, and dithio albumin as a standard. When the enzyme prepa threitol were purchased from Sigma Chemical Co., ration contained ampholine, which is nondialyzable U.S.A. Pyruvate kinase (rabbit muscle) [EC and interferes with protein determination, protein 2.7.1.40], lactate dehydrogenase (hog muscle) [EC was precipitated by adding HClO4 to a final con 1.1.1.27], glutamate dehydrogenase (beef liver) centration of 5%, and the precipitate was washed [EC 1.4.1.3], hexokinase (yeast) [EC 2.7.1.1], once with 5 % HClO4, dissolved in 0.1 N NaOH

J. Biochem. GUANINE DEAMINASE IN RAT BRAIN 169 and subjected to quantitative determination. Step 1. Preparation of crude extract: Rat Gel Electrophoreses-Disc and SDS-poly brain (50 g) was homogenized with 200 ml of 0.2 acrylamide gel electrophoreses were performed by M KPB, pH 7.0, containing 0.1 mM dithiothreitol. the methods described by Williams and Reisfeld All solutions used in the following experiments (7) and Weber and Osborn (8), respectively. Pro contained 0.1 mM dithiothreitol. The homogenate tein was stained with 0.25% Coomassie brilliant was centrifuged at 20,000 x g for 30 min and the blue R-250. supernatant was used for further purification. Determination of Molecular Weight-Molec Step 2. AmMonium sulfate fractionation: ular weight was determined by gel filtration on a The supernatant was brought to 40% saturation TSK-GEL G3000SW column with a Hitachi 633 with amMonium sulfate slowly by adding the Liquid Chromatograph. The column was eluted powdered salt. The mixture was stood for 30 with 10 mM potassium phosphate buffer (KPB), min at 0°C and then centrifuged at 10,000 x g for pH 7.0, containing 0.3 M KCl at a flow rate of 20 min. The supernatant was brought to 70 0.5 ml per min. The molecular weight of sub- saturation with amMonium sulfate by adding the units of the enzyme was calculated from the powdered salt. The mixture was stood for 30 mobility on SDS-polyacrylamide gel in comparison min and then centrifuged at 10,000 x g for 20 min. with those of standard . The resulting precipitate was dissolved in 100 ml Preparation of Anti-Rat Liver Guanine De of 50 mM KPB, pH 6.0, and dialyzed against 3 aminase Serum-Purified liver enzyme (1 mg of changes of 2 liters of the same buffer. protein in 2 ml) was emulsified with 2 ml of Step 3. DE-52 column chromatography: The Freund's adjuvant and injected subcutaneously dialysate was applied to a DE-52 column (1.5 x into various regions of two male rabbits, weighing 15 cm) equilibrated with 50 mM KPB, pH 6.0. The about 3 kg. Two weeks later, the same dose was column was washed with 100 ml of the same injected subcutaneously into other regions of the buffer, and then the enzyme was eluted with 200 same rabbits, and two weeks after this, 0.5 mg of ml of a linear gradient of 0-0.3 M KCl in the same enzyme protein emulsified with Freund's incom buffer. Fractions showing high guanine deaminase

plete adjuvant was injected. One week after the activity were pooled and dialyzed against 3 changes third injection, blood was collected from the ear of 2 liters of 10 mM KPB, pH 7.0. vein and serum was separated by centrifugation. Step 4. Hvdroaylapatite column chromatog A double diffusion test was carried out by the raphy: The dialysate was applied to a hydrox method of Ouchterlony (9). The ƒÁ-globulin frac ylapatite column (1.6 x 10 cm) equilibrated with tion of antiserum was obtained by pre the same buffer. The column was washed with

cipitation by the method of Nichol and Deutsh 100 ml of the same buffer and the enzyme was (10). eluted with 200 ml of a linear gradient of 10-300 Preparation of Brain Extract-Rats were mM KPB, pH 7.0. Fractions showing high enzyme killed by decapitation, and the brain was removed, activity were combined and concentrated to 3 ml soaked in saline and homogenized in 3 volumes of with the aid of a collodion bag, SM 132000 20 mM -HCl buffer, pH 7.0, containing (Satorius, W. Germany). 150 mM KCl and 10 mM MgCl2. The homogenate Step 5. Gel filtration on Ultrogel ACA-34: was centrifuged for 60 min at 105,000 x g and the The preparation from Step 4 was applied to supernatant was applied to a Sephadex G-50 Ultrogel ACA-34 (2 x 90 cm) equilibrated with column to remove low molecular weight substrates. 10 mM KPB, pH 7.0, and the enzyme was eluted Fractions of high protein concentration were with the same buffer. Fractions with high enzyme collected and used for experiments on ammo activity were combined and stored at 0°C. niagenesis. Step 6. Isoelectric focusing-The preparation from Step 5 was subjected to isoelectric focusing by the method of Matsuo et al. (11). RESULTS Table ‡T sumMarizes the results of the puri Purification of Guanine Deaminase from Rat fication of guanine deaminase from rat brain. Brain-All procedures were performed at 4°C. As shown in this table, the enzyme was purified

vol. 91, No. 1, 1982 170 S. MIYAMOTO, H. OGAWA, H. SHIRAKI, and H. NAKAGAWA

TABLE ‡T. Summary of purification of guanine deaminase from rat brain. Fifty g of rat brain was used for puri fication of the enzyme.

Fig. 1. Polyacrylamide gel electrophoresis of the puri Fig. 2. Molecular weight of guanine deaminase in rat fied enzyme from rat brain. Right: Gel electrophoresis brain. The molecular weight was calculated from the of the enzyme preparation at the final step of purifica elution volume of guanine deaminase from a TSK-GEL tion in a 7% running gel. The buffer used was diethyl G3000SW column in comparison with those of pyruvate -Tris, pH 8.0. Left: Gel electrophoresis of kinase (MW=237,000), lactate dehydrogenase (MW= the same sample in SDS-polyacrylamide gel (10% 140,000), creatine kinase (MW=81,000), and adenylate acrylamide and 0.1% SDS). Protein was stained with kinase (MW=21,000). 0.25% R 250. that the native enzyme exists as a dimer composed 2,800-fold in an 8 % yield from the crude extract. of two identical subunits. Figure 1 shows the results of disc- and SDS- pI: On isoelectric focusing the enzyme polyacrylamide gel electrophoreses of the enzyme showed an isoelectric point (pI) of pH 4.8. preparation at the final purification step. The Optimum pH: The optimum pH of the puri enzyme gave a single protein band with both fied enzyme was examined with sodium citrate methods. (pH 5.0-6.2), potassium phosphate (pH 5.8-7.8), Characteristics of Brain Guanine Deaminase- Tris-HCl (pH 7.4-9.0), and -NaOH (pH Molecular weight: The molecular weight of the 8.6-10.2) buffers. Activity was maximal between native enzyme was determined by gel filtration as pH 7.4-9.4. described in " MATERIALS AND METHODS." Stability: The purified enzyme could be From the slope shown in Fig. 2, the molecular stored at 0•Ž for more than 2 weeks without weight was calculated to be 105,000. appreciable loss of activity in 10 mm KPB, pH On SDS-gel electrophoresis, however, it ap 7.0. containing 0.1 mm dithiothreitol. peared to be 52,000 (Fig. 3). These results suggest Km value for substrate: The brain enzyme

J. Biochem. GUANINE DEAMINASE IN RAT BRAIN 171

TABLE ‡U. Effects of nucleotides and related com

pounds on brain guanine deaminase activity. The reaction mixture consisted of enzyme solution (protein, 1.4 ƒÊg), 100 mM Tris-HCl buffer, pH 8.0, 0.16 mM guanine, and various compounds (0.4 mM each) in a final volume of 1.2 ml. The reaction mixture was pre incubated for 5 min at 37•Ž and then the reaction was started by adding guanine. After incubation for 20 min at 37•Ž, the reaction was terminated by adding HClO4, to a final concentration of 3%. The super- natant obtained by centrifugation was neutralized by adding 0.5 ml of 1.8 N KOH and the amMonia concen tration was measured as described in " MATERIALS AND METHODS."

Fig. 3. Molecular weight of the guanine deaminase subunit estimated by SDS-polyacrylamide gel electro phoresis. Protein samples were incubated in 10 mM sodium phosphate buffer, pH 7.2, containing 1 % SDS and 0.1 mM dithiothreitol at 50•Ž for 2 h. Electro

phoresis was carried out at a constant current of 8 mA per tube for 5 h using bromphenol blue as a running marker. Gels were stained with 0.25% Coomassie brilliant blue R-250.

a 5-Phosphoribosyl 1-pyrophosphate.

Fig. 4. Double reciprocal plots of initial velocities with various guanine and 8-azaguanine concentrations. From the double reciprocal plots of the activity Purified enzyme preparation was used for these experi against the concentrations of substrates shown in ments. The enzyme activity was measured by the Fig. 4, the apparent Km values for guanine and method using glutamate dehydrogenase as described in 8-azaguanine were calculated to be 0.17 mM and " MATERIALS AND METHODS ." 0.67 mM, respectively. Inhibitors: Lewis reported that AICA inhib catalyzed the of guanine and 8- its guanine deaminase from rabbit liver (12). We azaguanine, but did not catalyze that of other compared the effects of various nucleotides and guanine derivatives such as GTP, GDP, GMP, their precursors, including AICA, on the rat brain and guanosine or derivatives such as enzyme. As shown in Table ‡U, only AICA AMP, and adenine. So the effects of inhibited the enzyme strongly. GTP was slightly these two substrates on its activity were examined. inhibitory, but this was probably because it is an

Vol. 91, No. 1, 1982 172 S. MIYAMOTO, H. OGAWA, H. SHIRAKI, and H. NAKAGAWA allosteric inhibitor of the glutamate dehydrogenase pI value was determined to be 4.8 by isoelectric (13) used in the amMonia assay. focusing. These values are almost identical with Figure 5 shows that AICA is a competitive those of the brain enzyme. inhibitor with a Ki value of 82 ƒÊm. Cross-reactivity of anti-liver enzyme with brain Divalent cations, such as Ca2+, Ba2+, and enzyme: The liver enzyme was injected into Mg2+ (50 mM), and EDTA (5 mM) did not affect rabbits as described in " MATERIALS AND the enzyme activity. METHODS" to raise antibody. The reactions of Purification of Liver Guanine Deaminase and the resulting antiserum with the liver and brain Comparison of It with the Brain Enzyme-Purifica enzymes were examined by Ouchterlony's gel tion of liver enzyme: Bergstrom and Bieber (14) double diffusion test. As shown in Fig. 6, the reported that the properties of the enzyme from antiserum formed a single precipitin line with rat brain are similar to those of the enzyme from both enzymes under the present experimental rabbit liver. This prompted us to examine whether conditions and these two precipitin lines fused rat liver guanine deaminase is imMunologically without spur formation. The antiserum also identical with the brain enzyme, because if so, the completely inhibited the activities as of both the liver should be a better source than the brain for brain and liver enzyme (data not shown). These obtaining enzyme for raising antiserum, since the findings suggest that the brain enzyme is imMu total activity in the liver is about 4 times that in nochemically identical with the liver enzyme. the brain, although its specific activity is slightly Liberation of amMonia from GTP by brain less. extract and its pathway: We examined whether The liver enzyme was purified from 100 g of amMonia was liberated from GTP, GDP, GMP, rat liver by the procedure used for purification of and guanosine by brain extract prepared as de the brain enzyme. About 200-fold purification scribed in " MATERIALS AND METHODS" and was achieved and the final preparation appeared whether amMonia formation was inhibited by at least 90% pure on disc gel electrophoresis. anti-guanine deaminase serum. Addition of a The molecular weight of the liver enzyme was GTP-generating system, such as creatine kinase calculated to be 108,000 by gel filtration and its plus creatine phosphate, almost completely inhib ited the liberation of amMonia from GTP. ATP, a potent inhibitor of 5•Œ-nucleotidase [EC 3.1.3.5]

(15), also suppressed the liberation of amMonia from GMP. AmMonia was liberated from gua nosine, even though the brain extract had been

Fig. 5. Competitive inhibition by AICA. Double Fig. 6. Uuchterlony gel double dtttuston test. Well I reciprocal plots were obtained with various guanine contained 100 ƒÊl of the antiserum, well 2, 100 ƒÊl of concentrations and fixed concentrations of AICA, purified liver enzyme containing 49 munits of guanine ranging from 0 to 0.2 mM. The enzyme assay was deaminase, and well 3, 100 ƒÊl of purified brain enzyme carried out as described in the legend to Fig. 4. containing 49 munits of guanine deaminase.

J. Biochem. GUANINE DEAMINASE IN RAT BRAIN 173 passed through Sephadex G-50 to remove low amined the problem of how much amMoniagenesis molecular weight compounds, but the addition of was dependent on the degradation of GTP in the inorganic phosphate, a substrate of guanosine system used by Schultz and Lowenstein (3) for phosphorylase [EC 2.4.2.15], markedly increased investigations on the purine nucleotide cycle in the yield of amMonia from the (Table the brain. As shown in Fig. 7A, amMoniagenesis

‡V). AmMonia liberation from GMP, guanosine took place linearly after addition of hexokinase or guanine was completely inhibited by pretreat and glucose to trigger the reaction by supplying ment of the brain extract with sufficient ƒÁ-globulin AMP, the substrate of AMP deaminase [EC fraction from the anti-guanine deaminase serum 3.5.4.6], via ADP and by releasing the inhibition

(Table ‡W). These findings indicate that GTP is of 5•Œ-nucleotidase by ATP. The treatment with converted to guanine via GDP, GMP, and gua ƒÁ-globulin fraction of the anti-rat liver guanine

nosine, and consequently that amMonia produc deaminase serum inhibited amMoniagenesis by tion from GTP is mediated by guanine deaminase. 60% in this system during incubation for 60 min, Effects of anti-guanine deaminase serum and in comparison with that with ƒÁ-globulin fraction

AICA on amMoniagenesis in the brain system: of the normal rabbit serum (Fig. 7A). As described above, brain extract degraded GTP AICA (1 mM) was also tested in the same via guanine to liberate amMonia. Next we ex- system, because it was shown to be a potent inhibitor of guanine deaminase. As shown in Fig. 7B, this compound caused about 50% inhibi TABLE ‡V. Liberation of amMonia from various tion of amMoniagenesis. guanine compounds in brain extract. Extract was pre pared from 5 rat brains and chromatographed on a Sephadex G-50 column as described in " MATERIALS AND METHODS." The reaction mixture consisted TABLE ‡W. Effects of anti-guanine deaminase serum of brain extract (protein, 2.85 mg), 20 mM imidazole on liberation of amMonia from GMP and guanosine. HCl buffer, pH 7.0, and 0.5 mM substrate (guanine was Extract was prepared from 3 rat brains and desalted on added to a final concentration of 0.2 mM due to its low a Sephadex G-50 column as described in " MATERI solubility) in a final volume of 0.6 ml. When indicated, ALS AND METHODS." Brain extract (6.4 ml; pro a purine -generating system tein, 19.4 mg) was treated with 1.2 ml of the ƒÁ-globulin (PTNGS*, 5 mM creatine phosphate and 12 units/ml of fraction of antiserum (anti-guanine deaminase) or creatine kinase) or 10 mM KPB (P1**), pH 7.0, was normal rabbit serum (control), prepared as described added. The reaction was started by adding extract and in " MATERIALS AND METHODS," at 37•Ž for terminated by adding HClO4 to a final concentration 30 min and allowed to stand at 0•Ž for I h. Aliquots of 4.3% after incubation for 40 min at 37`C. AmMonia of 0.4 ml were used for the experiments on amMonia produced endogenously was less than 4 nmol/mg pro genesis from GMP and guanosine, and 0.4 ml of this tein. extract diluted 20-fold with the medium was used for measurement of amMoniagenesis from guanine. The

reaction mixture consisted of 0.5 mM substrate (guanine

was added to a final concentration of 0.2 mM) in a final

volume of 0.6 ml. The reaction was started by adding

extract and terminated after 20-min incubation at 37•Ž

by adding HClO, to a final concentration of 4.3

a The reaction was terminated after incubation for 5 min. b The reaction mixture contained 71 Erg protein of brain extract as enzyme source. a P1, 0.67 mM.

Vol. 91, No. 1, 1982 174 S. MIYAMOTO, H. OGAWA, H. SHIRAKI, and H. NAKAGAWA

Fig. 7. Effects of anti-guanine deaminase serum and AICA on amMonia genesis in rat brain extract. (A) Brain extract (3.6 ml; protein, 14.8 mg) was incubated at 37•Ž for 30 min with 1.2 ml of ƒÁ-globulin fraction prepared from the antiserum. As a control, the extract was treated with 1.2 ml of y-globulin fraction prepared from normal rabbit serum. The supernatant obtained by centrifugation was used as enzyme source. The amount of antibody used was 10-fold that required to inhibit completely guanine deaminase activity included in the system after incubation for 30 min at 37•Ž. The complete reaction mix ture in a final volume of 6 ml contained 0.5 mM GTP, 1 mM IMP, 2 mM aspartate, 5 mM creatine phosphate, 12 units/ml creatine kinase and the brain extract treated with y-globulin fraction from anti-guanine deaminase serum (ƒ¢) or with that from control serum (•ü). At 20 min after addition of aspartate, hexokinase (HK; 72 units) and glucose (glc; 8 mM) were added to the reaction mixture at 37•Ž. Aliquots of 0.5 ml were taken every 20 min and imMediately mixed with 0.1 ml of cold 30% HClO4 to stop the reaction. After centrifugation, the super natant was neutralized with KOH. The supernatant obtained by centrifugation was used for quantitative determination of amMonia. (B) The reaction mixture was as described in (A) except that un-treated brain extract (protein, 24.6 mg) was used and incubation was carried out in the presence (ƒ¢) or absence (•ü) of 1 mM AICA.

Kumar et al. (17) reported that rat brain

DISCUSSION guanine deaminase was present in the light mito chondrial fraction and supernatant, and that the

Rossi et al. (16) purified guanine deaminase from enzyme in the latter fraction could be separated

pig brain to a homogeneous state. They reported into two types, enzymes A and B, with maximum that the native enzyme was composed of two specific activities of 77.5 and 590 units/mg protein,

[identical subunits with molecular weights of 50,000, respectively, by DEAE-cellulose column chro and that its isoelectric point was about pH 5.0. matography. The maximum specific activity of These values are similar to those for our purified their enzyme B was about twice that of our puri

preparation of the rat brain enzyme. However, fied enzyme preparation, but they did not show the Km value for guanine of the pig brain enzyme their enzyme B to be homogeneous. Accordingly

(11 ƒÊM) was one order lower than that of the rat our purified preparation is the first rat brain brain enzyme (0.17 mM). enzyme shown to be homogeneous.

J. Biochem. GUANINE DEAMINASE IN RAT BRAIN 175

We employed a DE-52 column at the second purine nucleotide cycle were added at physiolog step of purification, but did not obtain the two ical concentrations. However, we demonstrated forms of enzyme reported by Kumar et al. (17). that the y-globulin fraction of anti-guanine de Furthermore, we did not detect any enzyme aminase serum and AICA (1 mM) inhibited more activity in the light mitochondrial fraction or its than 50% of this ammoniagenesis. These findings inhibitor in the heavy mitochondrial fraction. indicate that deamination of guanine nucleotide Hindman and Knox (18) also reported that there by way of guanine deaminase is more active than was no evidence for the presence of a specific ammonia production by the purine nucleotide cycle inhibitor of guanine deaminase in rat brain. in the brain extract. Polymeric forms of guanine deaminase have been Davies and Taylor (21) compared the activ obtained from the liver (14, 19). Rossi et al. ities of guanine deaminase and 3•Œ,5•Œ-cyclic GMP

(16) suggested that thiol reagents contribute to phosphodiesterase and suggested that the two dissociation of polymeric forms of the liver en enzymes were closely correlated. This indicates zyme. On the basis of this suggestion, the two another important role for guanine deaminase, forms of guanine deaminase in the supernatant namely, regulation of the 3•Œ,5•Œ-cyclic GMP level fraction of rat brain reported by Kumar et al. needed for some neural functions (22). To make (17) could be due to polymerization of native the physiological role of guanine deaminase clearer, enzyme. Stepwise elution of the enzyme from a however, the problems of whether guanine de

DEAE-cellulose column may be another reason aminase actually contributes to ammonia produc for the separation of two forms. tion from endogenous substrates in intact brain As shown in Table ‡V, rat brain extract cata cells and of whether the change in the 3•Œ,5•Œ-cyclic

lyzed the formation of ammonia not only from GMP level is really reciprocal with that of guanine

guanine, but also from GTP, GDP, AMP, and deaminase in brain cells must be solved. guanosine. It was also shown that ATP, a potent inhibitor of 5•Œ-nucleotidase (15), inhibited ammo REFERENCES niagenesis from GMP, and that addition of phos . Galanti, B., Russo, M., Nardiello, S., & Giusti, G. phate markedly enhanced ammoniagenesis from (1975) Enzyme 20, 90-97 guanosine. Moreover, liberation of ammonia 2. Haulica, 1., Ababei, L., Teodorescu, C., Rosca, V., from GMP and guanosine was completely sup Haulica, A., Moisiu, M., & Haller, C. (1970) J. pressed by treatment with the ƒÁ-globulin fraction Neurochem.17,823-826 of anti-guanine deaminase serum. These findings 3. Schultz, V. & Lowenstein, J.M. (1976) J. Biol. Chem. indicate that ammonia formation from guanine 251,485 492 nucleotides was catalyzed by guanine deaminase 4. Roush, A. & Norris, E.R. (1950) Arch. Biochem. in cooperation with 5•Œ-nucleotidase and guanosine 29, 124-129 5. Kun, E. & Kearney, E.B. (1974) in Methods of phosphorylase. Enzymatic Anahvsis (Bergmeyer, H.U., ed.) Vol. 4, Jordan et al. (20) reported that guanosine pp. 1802-1806, Verlag Chenue Weinheirn Academic deaminase is present in the brain. However, Press, Inc., New York and London Table ‡W shows that ammoniagenesis from GMP 6. Lowry, O.H., Rosebrough, N.J., Farr, AL., & or guanosine was completely inhibited by the Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 ƒÁ-globulin fraction of anti-rat liver guanine de 7. Williams, D.E. & Reisfeld, R.A. (1964) ,Ann. N.Y aminase serum. These findings suggest that neither Acad. Sci. 121, 373-381 GMP deaminase nor is 8. Weber, K. & Osborn, M. (1969) J. Biol. Chem. 244,

present in brain extract. 4406-4412 Schultz and Lowenstein (3) reported that side 9. Ouchterlony, O. (1948) ,Arkir Kemi. Mineral. Geol. 26B, No. 14, 1 reactions catalyzed by nucleoside phosphorylase 10. Nichol, J.C. & Deutsh, H.F. (1948) J. Am. Cheat. and guanine deaminase complicate interpretation Soc. 70, 80-8311 of results, but that the purine nucleotide cycle is . Matsuo, Y., Horiuchi, Y., & Nakamura, T. (1969) active in the brain. We confirmed that ammo Kagaku no Ryvoiki (in Japanese) 88, 164-186 niagenesis took place in a brain system to which 12. Lewis, A.S. & Glantz, M.D. (1974) J. Biol. Chem. various substrates needed for operation of the 249, 3862 -3866

Vol. 91, No. 1, 1982

1 176 S. MIYAMOTO, H. OGAWA, H. SHIRAKI, and H. NAKAGAWA

13. Tomkins, G.M., Yielding, K.L., Curran, J.F., 18. Hindman, H.B. & Knox, W.E. (1978) Enzyme 23, Summers, M.R., & Bitensky, M.W. (1965) J. Biol. 395-403 Chem. 240, 3793-3798 19. Fogle, P.J. & Bieber, A.L. (1975) Prep. Biochem. 14. Bergstrom, J.D. & Bieber, A.L. (1979) Arch. Bio 5, 59-77 chem. Biophys. 194,107-116 20. Jordan, W.K., March, R., Houchin, O.B., & Popp, 15. Ipta, P.L. (1968) Biochemistry 7, 507-515 E. (1954) J. Neurochem. 4, 170-174 16. Rossi, C.A., Hakin, G., & Solaini, G. (1978) Bio 21. Davies, L.P. & Taylor, K.M. (1979) J. Neurochem. chim. Biophys. Acta 526, 235-246 23, 951-952 17. Kumar, S., Tewari, K.K., & Krishnan, P.S. (1965) 22. Lee, T.P., Kuo, J.F., & Greengard, P. (1972) Proc. Biochem. J. 95, 797-802 Natl. Acad. Sci. U.S. 69, 3287-3291

J. Biochem