Methylmalonic acid reduces the in vitro phosphorylation of cytoskeletal proteins in the cerebral cortex of rats

Brain Research 763 Ž1997. 221–231

Research report

Methylmalonic acid reduces the in vitro phosphorylation of cytoskeletal proteins in the cerebral cortex of rats A. De Mattos-Dutra, M.S. De Freitas, N. Schroder, A.C. Zilles, M. Wajner, R. Pessoa-Pureur ¨

)

Departamento de Bioquımica, Instituto de Biociencias, UniÕersidade Federal do Rio Grande do Sul, 90046-900 Porto Alegre RS, Brazil ´ ˆ Accepted 25 March 1997

Abstract The present work was undertaken to determine the action of methylmalonic acid ŽMMA., a metabolite, which accumulates in high amounts in methylmalonic acidemia, on the endogenous phosphorylating system associated with the cytoskeletal fraction proteins of cerebral cortex of young rats. We demonstrated that pre-treatment of cerebral cortex slices of young rats with 2.5 mM buffered methylmalonic acid ŽMMA. is effective in decreasing in vitro incorporation of w 32 PxATP into neurofilament subunits ŽNF-M and NF-L. and a- and b-tubulins. Based on the fact that this system contains cAMP-dependent protein kinase ŽPKA., Ca2qrcalmodulin-dependent protein kinase II ŽCaMKII. and protein phosphatase 1 ŽPP1., we first tested the effect of MMA on the kinase activities by using the specific activators cAMP and Ca2qrcalmodulin or the inhibitors PKAI or KN-93 for PKA and CaMKII, respectively. We observed that MMA totally inhibited the stimulatory effect of cAMP and interfered with the inhibitory effect of PKAI. In addition, the metabolite partially prevented the stimulatory effect of Ca2qrcalmodulin and interfered with the effect of KN-93. Furthermore, in vitro dephosphorylation of neurofilament subunits and tubulins was totally inhibited in brain slices pre-treated with MMA. Taken together, these results suggest that MMA, at the same concentrations found in tissues of methylmalonic acidemic children, inhibits the in vitro activities of PKA, CaMKII and PP1 associated with the cytoskeletal fraction of the cerebral cortex of rats, a fact that may be involved with the pathogenesis of the neurological dysfunction characteristic of methylmalonic acidemia. q 1997 Elsevier Science B.V. Keywords: Methylmalonic acid; Organic acidemias; Neurotoxicity; Cytoskeletal protein; Phosphorylation

1. Introduction Tissue accumulation of methylmalonic acid ŽMMA. is the biochemical hallmark of a group of inherited metabolic disorders called methylmalonic acidemias ŽMcKusick 251000., caused by a severe deficiency or absence of activity of the enzyme L-methylmalonyl-CoA mutase ŽEC 5.4.99.2. w17x. Patients with methylmalonic acidemia present neurological dysfunction as the most characteristic finding of the disorder, and this seems to be related not only to the transfer of MMA from blood to brain through the blood–brain barrier but also to the active MMA production within the brain w30x. The disease is severe enough to cause a fatal outcome in a significant number of patients if not diagnosed and treated promptly. Those who survive present with a variable degree of mental retardation w17,43x. Various metabolic effects have been attributed to MMA ) Corresponding author. Universidade Federal do Rio Grande do Sul, Instituto de Biociencias, Departamento de Bioquımica, Rua Sarmento ˆ ´ Leite 500, 90046-900 Porto Alegre RS, Brazil. Fax: q55 Ž51. 227-1343.

0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 4 1 5 - 0

and methylmalonyl-CoA w42,24x. We have previously described important effects of MMA on brain metabolism in vitro and in vivo, such as decrease in N-acetylneuraminic acid content in rat cerebellum w56x, diminished oxidation of glucose and ketone bodies by brain prisms of fasting rats, probably due to inhibition of succinate dehydrogenase and b-hydroxybutyrate dehydrogenase activities, respectively w14–16x, as well as an augmented release of lactate and a diminished CO 2 production in brain w57x. We also demonstrated that intrastriatal administration of MMA provokes convulsions in rats through glutamatergic mechanisms w38x and that chronic administration of MMA to young animals results in a diminished concentration of the high molecular mass neurofilament subunit ŽNF-H. in the cerebral cortex of rats w46x. The cytoskeleton of all cells is formed by intermediate filaments, microtubules and microfilaments, which are distinct elements made of fibrous polymer structures. Neurofilament ŽNF., a major intermediate filament ŽIF. expressed in neurons, is composed of three subunits with apparent molecular masses of 200 ŽNF-H., 160 ŽNF-M.

222

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

and 70 ŽNF-L. kDa on SDS-PAGE w27x. NFs are responsible, at least in part, for the control of axonal caliber w28,39x. Microtubules are mainly composed of a- and b-tubulins with molecular masses of 56 and 54 kDa, respectively w55x. Microtubules, which are assembled from a- and b-tubulins, together with microtubule-associated proteins, play an important role during neuronal process extension w7x and also provide the substrate for axonal organelle transport in the mature neuron w40x. Protein phosphorylation has a fundamental role in biological regulation, particularly in the regulation of neuronal function, such as neurotransmitter biosynthesis, axoplasmic transport, neurotransmitter release, generation of post-synaptic potentials, ion channel conductance, neuronal shape and motility, elaboration of dendritic and axonal processes, development and maintenance of differentiated properties of neurons and gene expression w51x. Many extracellular signals exert their physiological action by regulating the extent of phosphorylation of specific phosphoproteins in their target cells w51x. Alteration of the phosphorylation level of a protein is attained by increases or decreases of protein kinase or protein phosphatase activities. Activation of specific kinases is the most studied mechanism through which extracellular signals regulate protein phosphorylation in neural tissue. Many types of CNS proteins are regulated by phosphorylation and this is also the case for cytoskeletal proteins. Neurofilament subunits and tubulins can be phosphorylated in their terminal regions, both in vivo and in vitro by different protein kinases w47,58x. It has been postulated that the phosphorylation of NF tails may provide an electrostatic repulsion to maintain maximal distance between filaments, assuring the axonal caliber w49,4x. Several other roles for the phosphorylation of NFs have been postulated. These include regulating binding to microtubules w26x, acting as Ca2q sink w36x and protecting NFs from degradation w44x. Tubulin phosphorylation may contribute to the polymerization andror maintenance of the microtubular networks in neurons and the protein kinases that control microtubular phosphorylation may play a prominent role in neuronal differentiation w6x. Phosphorylation of cytoskeletal proteins is involved in the regulation of structural and dynamic aspects of the cell cycle, being particularly important in the regulation of neuronal function. There is growing evidence that alterations in cytoskeletal protein phosphorylation are related to a number of neurodegenerative diseases, like Alzheimer’s disease ŽAD., characterized by neuronal loss and the formation of neurofibrillary tangles ŽNFT. and senile plaques w52x. NFTs consist of paired helical filaments in which the microtubule-associated t protein is abnormally phosphorylated w23x. In addition, several neurotoxicants, such as aliphatic hexacarbons w34x, b , b X-iminodipropionitrile ŽIDPN. w22x, carbon disulfide w32x and acrylamide w20x, produce a central-peripheral neuropathy resulting in the accumulation of neurofilaments proximal to the nodes of

Ranvier w35x. Although the mechanisms of these neuropathies are still unknown, changes in the phosphorylation state of cytoskeletal proteins are though to be involved w34x. We have recently reported that, in chemically induced hyperphenylalaninemia, pre-treatment of rat cerebral cortex slices with phenylalanineq a-methylphenylalanine decreased the in vitro phosphorylation of both the 160-kDa neurofilament subunit and a-tubulin w9x. In the present report, we describe the inhibitory effect of MMA on the in vitro phosphorylation of NF-M, NF-L, a- and b-tubulins obtained from the Triton-insoluble cytoskeletal fraction of cerebral cortex of rats. Our results suggest that the inhibitory effect of MMA is mediated by protein kinase A ŽPKA., Ca2qrcalmodulin-dependent protein kinase II ŽCaMKII. and protein phosphatase 1 ŽPP1..

2. Materials and methods 2.1. Animals Wistar rats were obtained from our breeding stock. Rats were maintained on a 12-h lightr12-h dark cycle in a constant temperature Ž228C. colony room. Free water and a 20% Žwrw. protein commercial chow were provided. On the day of birth, the litter size was culled to 8 pups. Litters of - 8 pups were not included in the experiments. 2.2. Pre-treatment of tissue slices 400-m m-thick slices of cerebral cortex from 17-day-old rats were incubated in the presence or absence of 2.5 mM buffered methylmalonic acid ŽMMA. ŽSigma, St. Louis, MO. pH 7.2–7.4. Incubation was carried out in a Dubnoff metabolic shaker for 1 h at 308C in an atmosphere of 95% O 2r5% CO 2 . It has been previously reported that, in these experimental conditions, brain slices are metabolically active w13x. Each flask contained f 600 mg of cerebral cortex slices in 2 ml Krebs-Ringer bicarbonate buffer pH 7.4 Ž112.4 mM NaCl, 4.74 mM KCl, 2.54 mM CaCl 2 , 1.18 mM KH 2 PO4 , 1.18 mM MgSO4 , 3.58 mM NaHCO 3 , 2.31 mM sodium phosphate buffer. and 5 mM glucose, containing all the protease inhibitors described below and supplemented with 1 m M calpastatin ŽSigma.. Incubations were stopped by the addition of 20 vols. ice-cold cytoskeletal extraction buffer and performed as described below. 2.3. Immunoblotting Cytoskeletal fractions Ž100 m g. were separated by SDS-PAGE and transferred by electrophoresis to nitrocellulose according to the method of Towbin w54x. The blot was then washed 1 h in Tris-buffered saline ŽTBS. Ž0.5 M NaCl pH 7.5, 20 mM Tris-HCl, 5%. followed by a 2-h incubation in blocking solution ŽTBS q defatted dry milk..

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

After incubation, the blot was washed for 5 min with blocking solutionq 0.05% Tween-20 ŽT-TBS. and then incubated in blocking solution containing one of the following monoclonal antibodies: anti-NF-150 ŽNN-18. diluted 1 : 100, anti-NF-68 ŽNR-4. diluted 1 : 100, anti-atubulin Žclone DM1A. diluted 1 : 500, anti-b-tubulin III Žclone SDL3D10. diluted 1 : 200 ŽSigma.. The blot was then washed 5 = Ž10 min each. with TBS, washed 30 min in blocking solution and incubated for 2 h in antibody solution containing peroxidase-conjugated rabbit antimouse IgG diluted 1 : 4000. The blot was again washed 5 = with TBS. Development was carried out for NF-M, NF-L and a-tubulin with 4-chloro-1-naphtol at 0.6 mgrml in TBS pH 7.2, containing 0.01% Žvrv. H 2 O 2 30% as described by Hawkes and colleagues w25x. For b-tubulin, the development was carried out with kit ECL ŽRPN 2109. ŽAmersham.. 2.4. Triton-insoluble cytoskeletal preparation from cerebral cortex The cytoskeletal fraction was prepared as decribed by de Mattos et al. w11x. Cerebral cortex Ž600 mg. was homogeneized in 40 ml of ice-cold buffer containing 50 mM Tris-HCl pH 6.8, 5 mM EGTA, 1% Triton X-100 and the following protease inhibitors: 1 mM phenylmethylsulphonyl fluoride ŽPMSF., 1 mM benzamidine, 10 = 10 -1 m M leupeptin, 7 = 10 -1 m M antipain, 7 = 10y1 m M pepstatin and 7 = 10y7 m M chymostatin ŽSigma.. The homogenate was centrifuged at 13 000 = g for 15 min, 48C, in a Sorvall centrifuge, RC-2B, SS-34 rotor ŽDu Pont Instruments, Newtown, CT.. The insoluble pellet was resuspended in 40 ml of the same buffer containing 0.85 M sucrose and centrifuged under the same conditions. The pellet was dissolved in 50 mM MES Ž 2 w Nmorpholinoxethanesulphonic acid. pH 6.5 and 10 mM MgCl 2 and protein concentration was determined by the method of Bradford w2x. 2.5. In Õitro

32

P incorporation assays

An enriched cytoskeletal fraction prepared as described above served as a protein substrate and a source of endogenous protein kinases and phosphatases. Each assay mixture contained 10 m g of protein. Phosphorylation was carried out in 60-m l assay mixtures at pH 6.5 in buffer containing 50 mM MES, 10 mM MgCl 2 , 15 m M ATP and 1 m M calpastatin w9x. The reaction was started by adding 5 pM w g- 32 PxATP Ž5.5 = 10y1 0 Bqrmmol. ŽICN Radiochemicals, Irvine, CA.. After 5 min of incubation at 308C, the reaction was stopped by adding Laemmli sample buffer and the samples were boiled for 3 min. Proteins were analyzed by SDS-PAGE and autoradiograms were obtained from the gels. 32 P incorporation into cytoskeletal proteins was measured by liquid scintillator counting in 5.06 M toluene, 2.05 M ethanol, 12.4 mM PPO Ž2,5-diphenyloxazole. and 0.6 mM POPOP w1,4-bisŽ5-phenyl-2-

223

oxazol. benzene; 2,2-p-phenylene-bisŽ5-phenyloxazole.x ŽSigma. and 34.5% Triton X-100. The endogenous kinase activities were also assayed using the synthetic peptide substrates syntide 2 and kemptide ŽSigma., respectively. The assay mixture was essentially as described above, except that it contained 40 m M syntide 2 or 30 m M kemptide in the absence or presence of 2.5 mM MMA. After 5 min of incubation at 308C, reactions were stopped by spotting 20 m l of the assay mixture on phosphocellulose filters P-81 ŽGibco-BRL.. Filters were washed with 75 mM phosphoric acid and the radioactivity was measured in a liquid scintillator counter. 2.5.1. CaM KII and PKA actiÕity assays The standard assay mixture was the same as described above. Assays for Caq2 rcalmodulin-dependent protein kinase ŽCaMKII. activity contained 1 m M calmodulin ŽSigma. and 1 mM CaCl 2 in the absence or presence of buffered 2.5 mM MMA pH 7.2–7.4. cAMP-dependent protein kinase ŽPKA. activity was determined in reaction mixtures containing 20 m M cAMP ŽSigma., 2 mM EGTA and 1 mM MgCl 2 in the absence or presence of MMA. The endogenous CaMKII and PKA activities were also assayed in the presence of the exogenous substrates syntide 2 Ž40 m M. and kemptide Ž30 m M. using the same conditions described above. The effect of inhibitors on the kinase activity were determined by incubation of the assay mixtures with KN93 ŽCalbiochem, San Diego, CA. or protein kinase A inhibitor fragment 6–22 amide ŽPKAI. ŽSigma. which are the specific inhibitors of CaMKII and PKA activity, respectively. The CaMKII inhibitory assay contained 10 m M KN93 w18x in the absence or presence of buffered MMA and the PKA inhibitory assay contained 80 m M PKAI w19x, also in the absence or presence of methylmalonic acid. After 5 min of incubation at 308C, the reactions were stopped as described above and submitted to SDS-PAGE. 2.5.2. Phosphatase actiÕity assay The standard assay system was essentially as described above. The effect of MMA on the phosphatase activity was carried out by adding the buffered 2.5 mM MMA pH 7.2–7.4 to the reaction mixtures at 0 or 5 min of incubation. The assays that received MMA at 0 min were incubated for 5, 10, 20 or 30 min. The assays that received the acid at 5 min were incubated for 10, 20 or 30 min. Okadaic acid ŽSigma. previously prepared as a 50 m M stock solution in 10% dimethyl sulfoxide ŽDMSO. was added to some reaction mixtures to give final concentration of 0.5 m M, the concentration known to inhibit PP1 activity w9x. The reactions were stopped after 5, 10, 20 or 30 min and submitted to SDS-PAGE. 2.6. Polyacrylamide gel electrophoresis (SDS-PAGE) SDS-PAGE was performed on 10% acrylamide according to the discontinuous system of Laemmli w33x. Gels

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

224

Fig. 2. Effect of pre-treatment of brain slices with MMA on the 32 P in vitro incorporation into the Triton-insoluble cytoskeletal fraction proteins. 150-kDa neurofilament subunit ŽNF-M., 68-kDa neurofilament subunit ŽNF-L., a- and b-tubulins Ž a- and b-tub.. Tissue slices were incubated for 1 h at 308C with 2.5 mM MMA. The cytoskeletal fraction was incubated in the presence of the acid and in vitro 32 P incorporation measured as described in Section 2. Data are mean"S.E.M. values of 5–6 experiments. Statistically significant differences from controls, as determined by Student’s t-test, are indicated: a P - 0.01; ) P - 0.05.

Fig. 1. A: SDS-PAGE and autoradiography of the cytoskeletal fraction. Ža. Autoradiography obtained after in vitro incorporation of 32 P into cytoskeletal fraction. The positions of the 150- and 68-kDa neurofilament subunits, a- and b-tubulins Žfrom top to bottom. are marked by arrows to the left of lane a. Žb. SDS-PAGE of the cytoskeletal fraction stained with Coomassie blue. B: immunoblotting of neurofilament Ža,b. and tubulin Žc,d. subunits from Triton-insoluble cytoskeletal fraction. Samples were analyzed by SDS-PAGE, transferred to nitrocellulose and incubated with monoclonal antibodies: Ža. NN18 ŽNF-M., Žb. NR4 ŽNF-L., Žc. Clone DM1A Ž a-tub. and Žd. Clone SDL3D10 Ž b-tub., as described in Section 2. Development of a–c was caried out with 4-chloro-naphtol and of d with chemiluminescence.

After SDS-PAGE, cytoskeletal proteins were quantified as described by Rubin et al. w46x. Briefly, destained, dried SDS-PAGE gels were scanned in a densitometer ŽHoefer Scientific Instruments GS 300 TransmittancerReflectance Scanning Densitometer, San Francisco, CA. equipped with a chart recorder. The relative distribution of cytoskeletal proteins was calculated by cutting out and weighing the area under each peak of the densitometric scan and calculating its percent contribution to the total area. Therefore, cytoskeletal protein concentrations were calculated from

were stained with 0.25% Žwrv. Coomassie blue R-250 ŽSigma., 50% Žvrv. methanol and 10% Žvrv. acetic acid and destained overnight in 50% methanol and 10% acetic acid. Table 1 Effect of cAMP and Caq2 rcalmodulin on the in vitro incorporation of Exogenous substrates

32

Syntide 2 q EGTA Syntide 2 q EGTAq cAMP Syntide 2 Syntide 2 q Caq2 q calmodulin Kemptideq EGTA Kemptideq EGTA q cAMP Kemptide Kemptideq Caq2 q calmodulin

13466 26555 19991 35349 8532 74854 19225 19713

32

P into the exogenous substrates syntide 2 and kemptide

P incorporation Žcpm.

Percentage of incorporation 100 197 100 177 100 877 100 102

Values correspond to a representative experiment. Cytoskeletal fractions were incubated in the presence of syntide 2 and kemptide. Table 2 Effect of methylmalonate ŽMMA. on the

32

P in vitro incorporation into the cytoskeletal fraction and exogenous substrates

Treatment

Substrates

Control

Cytoskeletal fraction 48988" 403 Cytoskeletal fractionq MMA 33690" 465 a

MMA

Cytoskeletal fractionq syntide 2 65843" 2285 a Cytoskeletal fractionq syntide 2 q MMA 36595" 1547 a,a

Cytoskeletal fractionq kemptide 62983 " 530 a Cytoskeletal fractionq kemptideq MMA 39331 " b,a

Data represent 32 P incorporation Žcpm. into substrates and are expressed as mean " S.E.M. values of 3–6 experiments. The cytoskeletal fraction was prepared as indicated in Section 2 and 2.5 mM methylmalonate was added at the beginning of the experiments. Comparison between means was calculated by Student’s t-test. a Statistically significant differences from control for P - 0.01. a,b Statistically significant differences from cytoskeletal fractionq syntide 2 or kemptide, respectively, for P - 0.01.

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

percentage values considering that the total protein concentration of the cytoskeletal fraction measured by the method of Lowry et al. w37x corresponded to 100% Žtotal area of the densitometric scan..

2.7. Statistical analysis Data were analyzed by one-way analysis of variance ŽANOVA., followed by Duncan’s multiple range test when the F-test was significant and by Student’s t-test. All analyses were performed using the SPSS software program on an IBM-PC compatible computer.

225

3. Results Fig. 1A shows that, under our experimental conditions, cytoskeletal proteins from rat cerebral cortex are phosphorylated when incubated with w 32 PxATP. Fig. 1B shows an immunoblotting of the cytoskeletal fraction demonstrating that this fraction contained the neurofilament subunits ŽNF-M and NF-L. along with a- and b-tubulins as identified by monoclonal antibodies. The autoradiograph clearly indicates that the 150- ŽNF-M. and 68-kDa ŽNF-L. neurofilament subunits as well as a- and b-tubulins are good substrates for the endogenous phosphorylation system. The exogenous substrates syntide 2 and kemptide are also good

Fig. 3. Effect of pre-treatment of brain slices with 2.5 mM MMA on cAMP-dependent in vitro incorporation of 32 P into cerebral cortex cytoskeletal proteins. 150-kDa neurofilament subunit ŽNF-M.; 68-kDa neurofilament subunit ŽNF-L.; a-tubulin Žalpha-tub.; b-tubulin Žbeta-tub.. Tissue slices were incubated for 1 h at 308C with 2.5 mM MMA. Controls did not contain the acid. The cytoskeletal fraction obtained was incubated in 60 m l assay mixtures containing 50 mM MES pH 6.5, 1 mM MgCl 2 , 2 mM EGTA in the presence of 20 m M cAMP or 2.5 mM MMA or both ŽMMAq cAMP.. The in vitro 32 P incorporation was measured as described in Section 2. Data are mean " S.E.M. values of 4 experiments. Statistically significant differences between the means for the various groups were determined by ANOVA followed by Duncan’s multiple range test: P - 0.05: a / b / c.

226

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

substrates for the endogenous phosphorylation system ŽTable 1.. It is also observed in the table that, when assays contained the protein kinase activators, cAMP and Ca2qrcalmodulin, syntide 2 acted as good substrate for both PKA and CaMKII. However, calmodulin did not stimulate 32 P incorporation into kemptide. The effect of 2.5 mM methylmalonic acid ŽMMA. on the in vitro 32 P incorporation into the cytoskeletal fraction is displayed in Table 2. It is observed that MMA inhibits this incorporation by f 31%. The same effect of the acid was also evidenced when the exogenous substrates syntide 2 Ž44.5%. and kemptide Ž37.6%. were added to the cytoskeletal fraction. In order to identify which proteins were affected by the metabolite, we pre-incubated brain slices in the presence of 2.5 mM MMA and studied the in vitro 32 P incorporation

into the various proteins of the cytoskeletal fraction. We observed that phosphorylation was significantly blocked in the neurofilament subunits M ŽNF-M. and L ŽNF-L., aand b-tubulins ŽFig. 2.. To determine more precisely the inhibitory effect of the metabolite on the phosphorylation system associated with the cytoskeletal fraction, the effect of MMA on cAMP-dependent protein kinase ŽPKA. and Ca2qrcalmodulin-dependent protein kinase II ŽCaMKII. was assayed by adding the specific protein kinase activators, cAMP and Ca2qrcalmodulin, respectively, to the incubation system in the presence or absence of the acid. We first verified that cAMP enhances the in vitro 32 P incorporation into NF-M, NF-L and tubulins ŽFig. 3.. It is also clear in this figure that 2.5 mM MMA inhibits in vitro 32 P incorporation into these proteins and also prevents the activating

Fig. 4. Effect of pre-treatment of tissue slices with MMA on Ca2qrcalmodulin-dependent in vitro incorporation of 32 P into cerebral cortex cytoskeletal proteins. 150-kDa neurofilament subunit ŽNF-M.; 68-kDa neurofilament subunit ŽNF-L.; a-tubulin Žalpha-tub.; b-tubulin Žbeta-tub.. Tissue slices were incubated for 1 h at 308C with 2.5 mM MMA. the cytoskeletal fraction obtained was incubated in 60 m l assay mixtures containing 50 mM MES pH 6.5, 10 mM MgCl 2 , in the presence of 1 m m calmodulin, 1 mM CaCl 2 ŽCa2qq CaM. or 2.5 mM MMA or both ŽMMAq Ca2qq CaM.. The in vitro 32 P incorporation was measured as described in Section 2. Data are mean" S.E.M. values from 4 experiments. Statistically significant differences between the means for the various groups were determined by ANOVA followed by Duncan’s multiple range test: P - 0.05: a / b / c / d.

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

effect of cAMP. These results suggest that the inhibition of 32 P incorporation induced by MMA might be mediated by cAMP-dependent protein kinase. Next, we tested the effect of MMA on CaMKII activity ŽFig. 4.. We first certified that Ca2qrcalmodulin activates 32 P incorporation into the cytoskeletal proteins. When MMA was added to the incubation system in the presence of Ca2qrcalmodulin, we observed a differential effect on the various proteins. The in vitro 32 P incorporation into NF-M, NF-L Žsubunit. and a-tubulin subunit was partially blocked by the metabolite in the presence of Ca2qrcalmodulin ŽFig. 4A–C.. However, when MMA and Ca2qrcalmodulin were present in the incubation system, MMA did not prevent the effect of the activators on the in vitro 32 P incorporation into b-tubulin ŽFig. 4D.. These results suggest that CaMKII is somehow involved in the inhibitory effect of in vitro 32 P incorporation induced by MMA in some cytoskeletal fraction proteins ŽNF-M, NF-L and a-tubulin. but not in b-tubulin. We also investigated the influence of PKAI ŽFig. 5A. and KN93 ŽFig. 5B., inhibitors of PKA and CaMKII, respectively, on the effect of MMA on 32 P incorporation into cytoskeletal proteins. For these experiments, brain slices were first pre-incubated with MMA, the cytoskeletal fraction was obtained as usual and the incorporation assays were carried out in the presence of MMA and PKAI or KN93. Controls were identical but with no added acid. Fig. 5A shows that both PKAI and MMA added separately to the incubation system decrease in vitro incorporation of 32 P into the cytoskeletal proteins. However, when PKAI was added to the incubation assay after pre-incubation of brain slices with MMA, we observed that the effects were not additive, suggesting that the mechanisms of action of MMA and PKAI are somehow related. Similar results were obtained when the incorporation assays were undertaken in the presence of KN93, a specific CaMKII inhibitor, following MMA pre-treatment. Under these conditions, inhibition wasŽis. similar to that obtained only with KN93, suggesting that the inhibitions caused by KN93 and MMA are also related ŽFig. 5B.. These findings, taken together, suggest that the inhibitory effects of MMA on in vitro 32 P incorporation into the cytoskeletal proteins may be mediated by PKA and CaMKII activities. The effect of MMA on the phosphatase activity associated with the cytoskeletal fraction was also determined ŽFig. 6.. The figure shows the in vitro dephosphorylation of NF-M, NF-L, a- and b-tubulin proteins obtained from brain slices pre-incubated in the absence or presence of MMA. Our results demonstrate that the in vitro phosphorylation level of all studied cytoskeletal proteins extracted from control tissue slices decreased during incubation Ž30 min. in the absence of okadaic acid ŽOA.. We also observed a constant rate of phosphorylation over a period of 30 min when the phosphatase inhibitor okadaic acid was added to the system. Next, we tested the action of MMA

227

Fig. 5. Comparison of the effects of methylmalonate ŽMMA. and of the inhibitors of PKA ŽPKA1. and CaMKII ŽKN93. on the 32 P in vitro incorporation into the cytoskeletal proteins. 150-kDa neurofilament subunit ŽNF-M.; 68-kDa neurofilament subunit ŽNF-L.; a-tubulin Žalpha tub. and b-tubulin Žbeta tub.. A: cytoskeletal fractions were incubated with w 32 PxATP Ž2 m Cir10 m g of protein. in the presence of the protein kinase A inhibitor PKAI, or 2.5 mM MMA separately or in the presence of both drugs. B: cytoskeletal fractions were incubated with w 32 PxATP in the presence of the Ca2qrcalmodulin kinase II inhibitor KN93, or 2.5 mM MMA alone or with both drugs in the assay. The in vitro 32 P incorporation was measured as described in Section 2. Data Žmean"S.E.M. values from 4 experiments. are expressed as percentage of control Ž100%. assays without MMA or inhibitors. Statistically significant differences between the means for the various groups were determined by ANOVA followed by the Duncan multiple range test: P - 0.05: a s b.

added at time zero ŽMMA. or after 5 min ŽMMA 5. after P incubation. When tissue slices were pre-incubated with the drug at the beginning of incubation ŽMMA., the in vitro phosphorylation level was decreased compared to controls and remained so throughout the entire experiment. These results are in agreement with our previous experiments demonstrating that the drug inhibits in vitro 32 P incorporation. When MMA was added after 5 min of 32 P incubation, the phosphorylation level was higher and remained the same throughout incubation, indicating an inhibited phosphatase activity as we can observe by comparison with the dephosphorylation pattern obtained in the 32

228

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

Fig. 6. Effect of pre-treatment of tissue slices with MMA on the time course of dephosphorylation of cytoskeletal fraction proteins from the cerebral cortex. 150-kDa neurofilament subunit ŽNF-M. ŽA.; 68-kDa neurofilament subunit ŽNF-L. ŽB.; a-tubulin Žalpha-tub. ŽC. and b-tubulin Žbeta-tub. ŽD.. Cytoskeletal fraction obtained from control or MMA pre-treated tissue slices were incubated with w 32 PxATP Ž2 m Cir10 m g of protein. and the reactions stopped at specific intervals Ž5, 10, 20 and 30 min. by adding Laemmli sample buffer Žcontrols.. 2.5 mM MMA was added to the incubation mixture at 0 X min ŽMMA. or after 5 min ŽMMA 5 .. In order to inhibit the endogenous phosphatase activity okadaic acid was added to a final concentration of 0.5 m m in some assays. Data are expressed as mean " S.D. values of 3 experiments. Statistically significant differences from controls at 5 min as determined by Student’s t-test are indicated: a P - 0.01.

presence of 0.5 m M okadaic acid, a drug known to inhibit protein phosphatase w1x. Considering that we have previously reported that protein phosphatase type 1 ŽPP1. is the only phosphatase activity associated with the Triton-insoluble cytoskeletal fraction from cerebral cortex w8x, the present results indicate an additional effect of MMA inhibiting PP1.

4. Discussion We first demonstrated that the endogenous phosphorylating system present in our cytoskeletal fraction is active

for NFs and tubulins as well as for the exogenous substratres. Furthermore, we showed that pre-treatment of brain slices with 2.5 mM buffered MMA for 1 h inhibits the endogenous phosphorylation system associated with the Triton-insoluble cytoskeletal fraction obtained from cerebral cortex of young rats. Our results show that the drug is effective in inhibiting 32 P incorporation into endogenous NF subunits and tubulins ŽTable 2.. This effect was also observed by using the exogenous substrates syntide 2 and kemptide. However, kemptide is more selective for PKA. These findings are in agreement with previous reports using syntide 2 as a substrate for CaMKII, CaMKIV, PKA and PKC w31,59x. One possible explanation of the in-

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

hibitory effect of MMA on the in vitro 32 P incorporation into endogenous and exogenous substrates is that the acid acts by inhibiting protein kinase activities associated with the cytoskeletal fraction. We have previously reported that at least two distinct protein kinase activities are associated with the Triton-insoluble cytoskeletal fraction from cerebral cortex of rats: a calciumrcalmodulin-dependent protein kinase II ŽCaMKII. and a cAMP-dependent protein kinase ŽPKA. w9x. We, therefore, carried out a more detailed study on the effect of the drug on the endogenous kinases associated with the cytoskeletal fraction. We used two distinct approaches: Ž1. the specific kinase activators cAMP and Ca2qrcalmodulin and Ž2. the kinase inhibitors PKAI and KN93 in the absence or presence of MMA. PKAI is a competitive inhibitor of the catalytic subunit of cAMP-dependent protein kinase. Its action occurs through specific hydrophobic andror aromatic ŽPhe10 . binding to a hydrophobic pocket or cleft near the active site of the enzyme w19x whereas KN93, a CaMKII inhibitor, binds to CaMKII and competitively interferes with calmodulin-binding to the enzyme w53x. We observed that pre-treatment of tissue slices with 2.5 mM MMA totally prevented the activation of PKA activity by exogenous cAMP into NF-M, NF-L a- and b-tubulins ŽFig. 3.. In addition, when PKAI was added to the incubation system, the inhibitory activity on 32 P uptake was similar to that observed when the acid was tested alone or combined with the inhibitor ŽFig. 5A.. Since there was no additive inhibitory effect of both drugs on this kinase, it is possible that MMA might act, like PKAI, on the phosphorylation system. On the other hand, results obtained from the experiments in which Ca2qrcalmodulin or KN93 was added to the phosphorylation system of pre-treated slices suggest that CaMKII may also be involved in the inhibitory effect induced by MMA. As shown in Fig. 4, the activating effect of Ca2qrcalmodulin on CaMKII activity was partially blocked in NF-M, NF-L and a-tubulin ŽA–C.. This effect did not seem to result from a decreased CaMKII activity since the 32 P incorporation into b-tubulin was apparently not affected by the acid ŽFig. 4D.. The reasons for this apparent substrate dependence of the inhibitory action of MMA are unknown. Nonetheless, we could ascribe the 32 P incorporation into b-tubulin in the presence of Ca2qrcalmodulin and MMA to the presence of other Ca2q-dependent specific enzymes phosphorylating this protein. We should cite a previous report showing an inhibitory effect of another organic acid, arachidonic acid, on CaMK II activity by reversibly binding to the regulatory region of the kinase w45x. Furthermore, when pretreated brain slices were incubated with the protein kinase inhibitor KN93, we did not observe additive inhibitory effects in the presence of MMA as would be expected if the inhibitory mechanisms were independent ŽFig. 5B.. These findings, therefore, indicate superposed mechanisms of action as suggested for PKAI. Taken together, these

229

results suggest that PKA and CaMKII activities are somehow involved in the in vitro effect of MMA on 32 P incorporation into the cytoskeletal proteins. We also clearly demonstrated that MMA is able to inhibit the phosphatase activities present in the phosphorylation system since the presence of the acid in the incubation system consistently prevented dephosphorylation of neurofilament subunits and tubulins ŽFig. 6.. The precise mechanisms underlying the inhibitory effect of MMA on these enzymatic activities remain to be clarified. It is possible that MMA exert its inhibitory effect by directly or indirectly interacting with the enzymes controlling phosphorylation and dephosphorylation. Regarding kinase activities, MMA might partially block the active site or alternatively bind to other sites on the enzymes, producing an indirect action on the catalytic site. It is also possible that MMA may act by inducing conformational changes in the enzyme, leading to a decreased second messenger or substrate binding to the enzyme and, therefore, to a diminished enzyme activity. Evidence for this hypothesis appears in the experiments using cAMP as PKA activator ŽFig. 3.. Nonetheless, we cannot exclude the possibility that the drug binds directly to second messengers or to the inhibitor moleculeŽs. preventing their action. Otherwise, considering the differential effects of MMA on the stimulatory actions of Ca2qrcalmodulin on CaMKII in distinct cytoskeletal phosphoproteins, it is tempting to speculate that the metabolite has different affinity for the distinct enzyme substrate complexes. Several neurotoxic agents, such as metal, have been reported to alter the phosphorylation level of cytoskeletal proteins by different mechanisms. In particular, aluminium is related to the abnormal appearance within perikarya of phosphorylated neurofilament protein epitopes w50x. Amongst the several possible mechanisms related to this accumulation, an aluminium-induced inhibition of endogenous phosphatase activities has been proposed w60x. Lithium chloride inhibited the incorporation of 32 P into NF-M as well as NF-H whereas incorporation into low molecular mass neurofilament protein and b-tubulin was unaffected. The authors suggest that this effect was probably due to the interference by Liq itself with the phosphorylation of NF-M and NF-H by specific neurofilament kinaseŽs. w1x. In addition, we have recently reported that chronic hyperphenylalaninemia induced in rats decreased the in vitro 32 P incorporation into cytoskeletal proteins, probably by inhibiting calciumrcalmodulin-dependent protein kinase II and protein phosphatase type 1 activities w10x. Several studies have demonstrated the relationship between the phosphorylation of neurofilaments and the axonal caliber w5x. Hypomyelination in mutant mice is correlated with reduction in phosphorylation of neurofilaments and inhibition of radial growth w12x. Within this context, since hypo or demyelination is characteristically a primary pathologic finding in the central nervous system of rats submitted to chronic hyperphenylalaninemia and in pa-

230

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

tients with phenylketonuria and methylmalonic acidemia w3,29,48x, our results obtained with brain slices might lead to a disturbed myelogenesis. In conclusion, we demonstrate in the present study that MMA, at doses found in blood, brain and other tissues of patients with methylmalonic acidemia, markedly affects the phosphorylation system associated with cytoskeletal proteins in rat cerebral cortex. It is suggested that the decreased neurofilament phosphorylation observed in vitro in the present study may occur in vivo. The presence of poorly phosphorylated neurofilaments and other cytoskeletal phosphoproteins may affect the transport of neurofilaments, their interaction with other cytoskeletal proteins and their turnover due to resistance to proteolysis w41,21x. All of these factors may contribute to MMA neurotoxicity. However, more research is necessary to clarify the mechanisms underlying these effects but inhibition of phosphatases and, possibly, protein kinases by the metabolite occurs as shown by our data. Even though it is difficult to extrapolate our results to the human condition, if that is the case, we presume that intereference with this system would probably lead to the deleterious action of MMA on the brain, a fact that might explain at least in part the pathogenesis of the severe neurological dysfunction of methylmalonic acidemic patients.

w9x

w10x

w11x

w12x

w13x

w14x

w15x

w16x

Acknowledgements

w17x

This work was supported by grants from CNPq, FINEP, FAPERGS and PROPESPrUFRGS.

w18x

References w1x G.S. Bennett, D. Laskowska, C. Di Lullo, Lithium chloride inhibits the phosphorylation of newly synthesized neurofilament protein, NF-M, in cultured chicken sensory neurons, J. Neurochem. 57 Ž1991. 120–129. w2x M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 Ž1977. 248–254. w3x J. Brisman, P.T. Ozand, CT and MR of the brain in the diagnosis of organic acidemias, Experiences from 107 patients, Brain Dev. 16 Ž1994. 102–124. w4x M.J. Carden, J.Q. Trojanowski, W.W. Schlaepfer, V.M.-Y. Lee, Two-stage expression of neurofilament polypeptides during rat neurogenesis with early establishment of adult phosphorylation patterns, J. Neurosci. 7 Ž1987. 3489–3504. w5x D.W. Cleveland, M.J. Monteiro, P.C. Wong, S.R. Gill, J.D. Gearhart, P.N. Hoffman, Involvment of neurofilaments in the radial growth of axons, J. Cell Sci. 15 Ž1991. 85–95. w6x B.E. Crute, R.G. Van Buskirk, A casein kinase-like kinase phosphorylates b-tubulin and may be a microtubule-associated protein, J. Neurochem. 59 Ž1992. 2017–2023. w7x M.P. Daniels, Colchicine inhibition of nerve fiber formation in vitro, J. Cell Biol. 53 Ž1972. 164–176. w8x M.S. De Freitas, A.G. De Mattos, M.M. Camargo, C.M.D. Wannmacher, R. Pessoa-Pureur, Cytoskeletal-associated protein kinase

w19x

w20x

w21x

w22x

w23x

w24x

w25x

w26x

and phosphatase activities from cerebral cortex of young rats, Neurochem. Res. 20 Ž1995. 951–956. M.S. De Freitas, A.G. De Mattos, M.M. Camargo, C.M.D. Wannmacher, R. Pessoa-Pureur, Effect of phenylalanine and alpha-methylphenylalanine on in vitro incorporation of 32 P into cytoskeletal cerebral proteins, Neurochem. Int. 26 Ž1995. 381–385. M.S. De Freitas, A.G. De Mattos, N. Schroder, C.M.D. Wannmacher and R. Pessoa-Pureur, Effect of hyperphenylalaninemia chemically induced on in vitro incorporation of 32 P into cytoskeletal proteins from cerebral cortex of developing rats, Exp. Neurol. Ž1996, in press.. A.G. De Mattos, M.S. De Freitas, M.M. Camargo, Pessoa-Pureur, Developmentally regulated in vitro phosphorylation of cytoskeletal proteins of the cerebral cortex of normal and malnourished rats, Dev. Neurosci. 16 Ž1994. 38–43. S.M. De Waegh, V.M.-Y. Lee, S.T. Brady, Local modulation of neurofilament phosphorylation, axonal caliber, and slow axonal transport by myelinating Schwann cells, Cell 68 Ž1992. 451–463. C.S. Dutra-Filho, M. Wajner, E. Gassen, R. Candiago, A. Wilhelms, H. Malfussi, C.M.D. Wannmacher, Effect of organic acids on in vitro glucose oxidation by cerebral cortex of young rats, Med. Sci. Res. 23 Ž1995. 25–26. J.C. Dutra, M. Wajner, C.F. Wannmacher, C.M.D. Wannmacher, Methylmalonate increases glucose uptake in vitro by brain of young rats, Biochem. Soc. Trans. 17 Ž1989. S72–S73. J.C. Dutra, M. Wajner, C.F. Wannmacher, C.S. Dutra-Filho, C.M.D. Wannmacher, Effects of methylmalonate and propionate on uptake of glucose and ketone bodies in vitro by brain of developing rats, Biochem. Med. Metab. Biol. 45 Ž1991. 56–64. J.C. Dutra, M. Wajner, C.M.D. Wannmacher, C.S. Dutra-Filho, E. Rache, Inhibition of succinate dehydrogenase and beta-hydroxybutyrate dehydrogenase activities by methylmalonate in brain and liver of developing rats, J. Inher. Metab. Dis. 16 Ž1993. 147–153. W.A. Fenton and L.E. Rosenberg, Disorders of propionate and methylmalonate metabolism, in: C.R. Scriver, A.L. Beadet, W.S. Sly, D. Valle ŽEds.., The Metabolic Basis of Inherited Disease, 7th edn., McGraw-Hill, New York, NY, 1995, pp. 1423–1449. K. Gietzen, A. Wuthrich, H. Bader, A new powerful inhibitor of red ´ blood cell Ca2q-transport ATPase and of calmodulin-regulated functions, Biochem. Biophys. Res. Commun. 101 Ž1981. 418–425. D.B. Glass, L.J. Lundquist, B.M. Katz, D.A. Walsh, Protein kinase inhibitor-Ž6–22.-amide peptide analogs with standard and nonstandard amino acid substitutions for phenylalanine 10, J. Biol. Chem. 264 Ž1989. 14579–14584. B.G. Gold, J.W. Griffin, D.L. Price, Slow axonal transport in acrylamide neuropathy: different abnormalities produced by a single dose and continuous administration, J. Neurosci. 5 Ž1985. 1755– 1768. J.A. Greenwood, J.C. Troncoso, A.C. Costello, Phosphorylation modulates calpain-mediated proteolysis and calmodulin binding of the 200-kDa and 160-kDa neurofilament proteins, J. Neurochem. 61 Ž1993. 191–199. J.W. Griffin, P.N. Hoffman, A.W. Clark, P.T. Carroll, D.L. Price, Slow axonal transport of neurofilament proteins: impairment by b , b X-iminodipropionitrile administration, Science 202 Ž1978. 633– 635. I. Grundke-Iqbal, K. Iqbal, Y.-C. Tung, M. Quinlan, H.M. Wisniewski, L.I. Binder, Abnormal phosphorylation of the microtubule associated protein tau in Alzheimer cytoskeletal pathology, Proc. Natl. Acad. Sci. USA 83 Ž1986. 4913–4917. M.L. Halperin, C.M. Schiller, I.B. Fritz, The inhibition by methylmalonic acid of malate transport by the dicarboxilate carrier in rat liver mitochondria, J. Clin. Invest. 50 Ž1971. 2276–2282. R. Hawkes, E. Niday, J. Gordon, A dot immnobinding assay for monoclonal and other antibodies, Anal. Biochem. 119 Ž1982. 142– 147. S. Hisanaga, N. Hirokawa, Dephosphorylation induced interactions

A. De Mattos-Dutra et al.r Brain Research 763 (1997) 221–231

w27x

w28x

w29x

w30x

w31x

w32x

w33x w34x

w35x

w36x

w37x

w38x

w39x

w40x

w41x

w42x

w43x

w44x

of neurofilaments with microtubules, J. Biol. Chem. 265 Ž1990. 21852–21858. S. Hisanaga, M. Kusubata, E. Okumura, T. Kishimoto, Phosphorylation of neurofilament H subunit at the tail domain by CDC2 kinase dissociates the association to microtubules, J. Biol. Chem. 266 Ž32. Ž1991. 21789–21803. P.N. Hoffman, D. Cleveland, J. Griffin, P. Landes, N. Cowan, D. Price, Neurofilament gene expression; a major determinant of axonal calber, Proc. Natl. Acad. Sci. USA 84 Ž1987. 3472–3476. G.F. Hoffmann, K.M. Gibson, F.K. Treks, W.L. Nihan, H.J. Bremer, D Rating, Neurological manifestations of organic acid disorders, Eur. J. Pediatr. 153 Ž1994. 94–100. G.F. Hoffmann, W. Meier-Augenstein, S. Stockler, R. Surtees, D. ¨ Rating, W.L. Nyhan, Physiology and pathophysiology of organic acids in cerebrospinal fluid, J. Inher. Metab. Dis. 16 Ž1993. 648–669. A. Ishida, I. Kameshita, S. Okuno, T. Kitane, H. Fujisawa, A novel highly specific and potent inhibitor of calmodulin-dependent preotein kinase II, Biochem. Biophys. Acta 212 Ž3. Ž1995. 806–812. J. Juntunen, M. Haltia, I. Linnvila, Histochemically demonstrateable non-specific cholinesterase as an indicator of peripheral nerve lesion in carbon disulfide-induced polyneuropathy, Acta Neuropathol. ŽBerlin. 29 Ž1974. 361–366. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 277 Ž1970. 680–685. C.L. Lanning, R. Wilmarth, M.B. Abou-Donia, In vitro binding of w 14 Cx2,5-hexanedione to rat neuronal cytoskeletal proteins, Neurochem. Res. 19 Ž9. Ž1994. 1165–1173. D.M. Lapadula, R.D. Irwin, E. Suwita, M.B. Abou-Donia, Cross-linking of neurofilament proteins of rat spinal cord in vivo after administration of 2,5 hexanedione, J. Neurochem. 46 Ž1986. 1843– 1850. S. Lefebre, W.E. Mushynski, Calcium binding to untreated and dephosphorylated porcine neurofilaments, Biochem. Biophys. Res. Commun. 145 Ž1987. 1006–1011. O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193 Ž1951. 265–267. C.F. Mello, J. Gegnini, R.F. Jimenez-Bernal, M. Rubin, J. Bastian, E. Costa, M. Wajner, Intrasticial methylmalonic administration induces rotational behaviour and convulsion through glutamatergic mechanisms, Brain Res. 721 Ž1996. 120–125. S. Monaco, L. Atilio-Gambetti, R.J. Lasek, M. Katz, P. Gambetti, Experimental increase of neurofilament transport rate: decreases in neurofilament number and in axonal diameter, J. Neuropathol. Exp. Neurol. 48 Ž1989. 23–32. R.L. Morris, P.J. Hollenbeck, Axonal transport of mitochondria along microtubules and F-actin in living vertebrate neurons, J. Cell Biol. 131 Ž1995. 1315–1326. R.A. Nixon, P.A. Paskevich, R.K. Sihag, C.Y. Thayer, Phosphorylation on carboxyl terminus domains of neurofilament proteins in retinal ganglion cell neurons in vivo: influences on regional neurofilament accumulation, interneurofilament spacing, and axon caliber, J. Cell Biol. 126 Ž1994. 1031–1046. V.G. Oberholzer, B. Levin, E.A. Burgess, W.F. Young, Methylmalonic aciduria. An inborn error of metabolism leading to chronic metabolic acidosis, Arch. Dis. Child. 42 Ž1967. 492–504. H. Ogier, C. Charpentier and J.H. Saudubray, Organic acidemias, in: J. Fernandes, J.M. Saudubray, K. Tada ŽEds.., Inborn Metabolic Diseases, 1st edn., Springer, Berlin, Germany, 1990, pp. 271–299. H.C. Pant, Dephosphorylation of neurofilament proteins enhances their susceptibility to degradation by calpain, Biochem. J. 256 Ž1988. 665–668.

231

w45x D. Piomelli, J.K.T. Wang, T.S. Sihra, A.C. Nairn, A.J. Czernik, P. Greengard, Inhibition of Ca2qrcalmodulin-dependent protein kinase II by arachidonic acid and its metabolites, Proc. Natl. Acad Sci. USA 86 Ž1989. 8550–8554. w46x M.A. Rubin, C.M.D. Wannmacher, G.B. Valente, M.M. Camargo, R. Pessoa-Pureur, Diminished concentration of the NF-H subunit of neurofilament in cerebral cortex of rats chronically treated with proline, methylmalonate and phenylalanine plus alpha-methylphenylalanine, J. Inher. Metab. Dis. 15 Ž1992. 252–260. w47x T. Saitoh, H. Shima, Y. Osawa, M. Nagao, B.A. Hemmings, T. Kishimoto, S. Hisanaga, Neurofilament-associated protein phosphatase 2A: its possible in preserving neurofilaments in filamentous states, Biochemistry 34 Ž22. Ž1995. 7376–7384. w48x C.R. Scriver, S. Kaufman, R.C. Eisensmith and S.L.C. Woo, The phenylalaninemias, in: C.R. Scriver, A.L. Beadet, W.S. Sly, D. Valle ŽEds.., The Metabolic Basis of Inherited Disease, 7th edn., McGraw-Hill, New York, NY, 1995, pp. 1015–1075. w49x G. Shaw, Neurofilaments: abundant but mysterious neuronaltructures, Bioassays 4 Ž1986. 9548–9552. w50x T.B. Shea, J.F. Clarke, T.R. Wheelock, P. Paskevich, R.A. Nixon, Aluminium salts induce the accumulation of neurofilaments in perikarya of NB2ard1 neuroblastoma, Brain Res. 492 Ž1989. 53–64. w51x G.J. Siegel, B.W. Agranoff, R.W. Albers and P.B. Molinoff ŽEds.., Basic Neurochemistry, Protein Phosphorylation and Neuronal Function, Raven, New York, NY, 1993, 449 pp. w52x N.A. Simonian, T. Elvhage, A.J. Czernik, P. Greengard, B.T. Hyman, Calciumrcalmodulin-dependent protein kinase II immunostaining is preserved in Alzheimer’s disease hippocampal neurons, Brain Res. 657 Ž1994. 294–299. w53x H. Tokumitsu, T. Chijiwa, M. Hagiwara, A. Mizutani, M. Terasawa, H. Hidaka, KN-62, 1-N,O-bis-Ž5-isoquinolinesulfonyl.-N-methyltyrosyl-4-phenylpiperazine, a specific inhibitor of Ca2qrcalmodulin-dependent protein kinase II, J. Biol. Chem. 265 Ž1990. 5315– 5320. w54x H. Towbin, T. Staehelin, J. Gordon, Electrophoresis transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications, Proc. Natl. Acad. Sci. USA 76 Ž1979. 4350–4354. w55x G.B. Valente, A.G. Mattos, M.A. Rubin, M.M. Paz, C.I. Tasca, R. Pessoa-Pureur, Malnutrition increases insoluble-to-soluble tubulin ratio and in vitro incorporation of w 32 PxATP in rat cerebral cortex, Neurochem. Int. 21 Ž1992. 595–603. w56x M. Wajner, E.C. Brites, J.C. Dutra, M.S. Buchater, A.H. Pons, R.F. Pires, L.E. Wannmacher, A. Rosa-Junior, V.M. Trindade, C.M.D. ´ Wannmacher, Diminished concentrations of ganglioside-N-acetyl neuroaminic acid ŽNANA. in cerebellum of young rats receiving chronic administration of methylmalonic acid, J. Neurol. Sci. 85 Ž1988. 233–238. w57x M. Wajner, J.C. Dutra, S.E. Cardoso, C.M.D. Wannmacher, E.R. Motta, Effect of methylmalonate on in vitro lactate release and carbon dioxide production by brain of suckling rats, J. Inher. Metab. Dis. 15 Ž1982. 92–96. w58x F. Wandosell, L. Serrano, J. Avila, Phosphorylation of a-tubulin carboxyl terminal tyrosine prevents its incorporation into microtubules, J. Biol. Chem. 262 Ž1987. 8268–8283. w59x H. Yamamoto, K. Fukunaga, K. Lee, T.R. Soderling, Ischemia-induced loss of brain calciumrcalmodulin-dependent protein kinase II, J. Neurochem. 58 Ž1992. 1110–1117. w60x H. Yamamoto, Y. Saitoh, S. Yasugawa, E. Miyamoto, Dephosphorylation of s factor by protein phosphatase 2A in synaptosomal cytosol fractions, and inhibition by aluminium, J. Neurochem. 55 Ž1990. 683–690.