- Email: [email protected]
Effect of Hyperphenylalaninemia Chemically Induced onin VitroIncorporation of32P into Cytoskeletal Proteins from Cerebral Cortex of Developing Rats
EXPERIMENTAL NEUROLOGY ARTICLE NO.
143, 188–195 (1997)
EN966351
Effect of Hyperphenylalaninemia Chemically Induced on in Vitro Incorporation of 32P into Cytoskeletal Proteins from Cerebral Cortex of Developing Rats MARTA S. DE FREITAS, ANGELA DE MATTOS-DUTRA, NADJA SCHRODER, CLO´ VIS M. D. WANNMACHER, AND REGINA PESSOA-PUREUR Departamento de Bioquı´mica, Instituto de Biocieˆncias, Universidade Federal do Rio Grande do Sul, Porto Alegre RS, Brazil
We studied the effect of hyperphenylalaninemia on in vitro incorporation of 32P into cytoskeletal proteins from cerebral cortex of rats by injecting L-phenylalanine plus a-methylphenylalanine subcutaneously from the 6th to the 14th day postpartum. Chronic hyperphenylalaninemia induced an increased in vitro phosphorylation of the 150-kDa neurofilament subunit and tubulins present in the cytoskeletal fraction at the end of the treatment and 3 days after treatment discontinuation. In addition, when in vitro phosphorylation of the cytoskeletal proteins from treated animals was performed in the presence of the drugs we observed a decreased in vitro incorporation of 32P into these proteins. Thus, the effect of L-phenylalanine plus a-methylphenylalanine on the endogenous protein kinase and phosphatase activities was examined and the results demonstrated that these drugs have an inhibitory effect on calcium/calmodulin-dependent protein kinase II and protein phosphatase type 1. r 1997 Academic Press
INTRODUCTION
The neuronal cytoskeleton comprises a protein network formed mainly of neurofilaments and microtubules interacting with one another and with a variety of associated proteins. Axons contain many longitudinally oriented neurofilaments and microtubules occupying most of the axoplasm (14). Neurofilaments are the intermediate neuronal filaments. They are composed of three subunits that exhibit apparent molecular masses of 200 (NF-H), 150 (NF-M), and 68 kDa (NF-L) by SDS–PAGE (12). Neurofilaments are believed to play an important role in the maintenance of neuronal shape and determination of axonal caliber (22, 13). 1 This work was supported by Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq), Fundac¸a˜o de Amparo a` Pesquisa do Estado do RGS (FAPERGS), and Pro´-Reitoria de Pesquisa e Po´s-Graduac¸a˜o da Universidade Federal do RGS (PROPESPUFRGS).
0014-4886/97 $25.00 Copyright r 1997 by Academic Press All rights of reproduction in any form reserved.
Microtubules are mainly constituted of a and b tubulin subunits. In the brain, they are involved in a variety of cellular functions including axonal transport in the mature neuron and synaptogenesis (29). Tubulins and neurofilaments undergo posttranslational modifications that modulate their physiological role in the neuronal cells. Tubulins and neurofilament subunits can be phosphorylated in their terminal regions by different protein kinases (30, 23). Phosphorylation regulates the ability of tubulin and neurofilament subunits to self-assemble into microtubules and neurofilaments, respectively. Moreover, their phosphorylation levels modulate the interaction between microtubules and neurofilaments and other cytoskeletal proteins (21). Phenylketonuria (PKU) is an inborn error of metabolism characterized by a severe deficiency or absence of hepatic phenylalanine hydroxylase (PAH) activity associated with brain dysfunction. Several studies have demonstrated the association of permanent brain damage with increased phenylalanine (Phe) levels during critical periods of brain development (15). Biochemical studies have shown a severe reduction in the amount of myelin in brains of untreated PKU patients and retarded synapse formation and dendritic arborization (26, 1). Animal models have been developed to study the pathophysiology of the biochemical and behavioral abnormalities found in human PKU. We have previously demonstrated that hyperphenylalaninemia induced in rats by treating the animals with daily administration of Phe plus a-methylphenylalanine (MePhe), a PAH inhibitor, reduces the NF-H content in cerebral cortex (25). We have also demonstrated that preincubation of tissue slices with Phe plus MePhe decreases in vitro 32P incorporation into NF-M and tubulins (5). In the present study we used the animal model to study the effects of hyperphenylalaninemia on in vitro 32P incorporation into cytoskeletal proteins. In addition, we report the inhibitory effect of Phe plus MePhe on the phosphorylating system associ-
188
HYPERPHENYLALANINEMIA AND CYTOSKELETON
ated with the cytoskeletal fraction from cerebral cortex of young rats. MATERIALS AND METHODS
Animals Wistar rats from our breeding stock were maintained on a 12-h light/12-h dark cycle in a constant temperature (22°C) colony room. On the day of birth the litter size was culled to eight pups. Litters with fewer than eight pups were not included in the experiments. Free water and a 20% (w/w) protein commercial chow were provided.
189
0.7 pM chymostatin (Sigma). The homogenate was centrifuged at 13,000g for 15 min at 4°C. The pellet was resuspended in 40 ml of the same buffer containing 0.85 M sucrose and centrifuged for 15 min at 13,000g. The pellet was dissolved in 50 mM 2(N-morpholino)ethanesulfonic acid (MES), pH 6.5, and 10 mM MgCl2 followed by protein determination (2). An aliquot containing 10 µg protein was removed for in vitro 32P incorporation assays. The remaining material was precipitated with 5% TCA, washed twice with ethanol and acetone, and dissolved in 1% SDS, and the protein concentration was determined (19). This fraction was analyzed by SDS– PAGE and used for quantitation as described below.
In Vivo Treatment
32P
Rats from the same litter were randomly divided into two groups. Rats from the experimental group received 3.0 µmol/g body weight of L-phenylalanine twice a day at 10-h intervals plus 1.6 µmol/g a-methylphenylalanine once a day. Drugs were dissolved in 0.9% saline and administered subcutaneously from the 6th to the 14th day postpartum in a volume of 20 µl/g body weight. Control animals received the same volume of saline. Rats were sacrificed on the 14th day or on the 17th day; the cerebral cortex was immediately removed and the cytoskeletal fraction was extracted as described below.
An enriched cytoskeletal fraction prepared as described above served as protein substrate and as a source of endogenous protein kinases and protein phosphatases. Each assay mixture contained 10 µg of protein. Phosphorylation was carried out in 60 µl of a buffer containing 50 mM MES, pH 6.5, and 10 mM MgCl2. In some assay mixtures, the 32P incorporation into the cytoskeletal fraction from chronically treated animals was performed by adding 1.0 mM L-phenylalanine plus 1.0 mM a-methylphenylalanine to the phosphorylation assay. The reaction was started by adding 2.0 µCi [g-32P]ATP (16.6 3 10210 Bq/mmol) (ICN Radiochemicals, Irvine, CA). After incubation for 5 min at 30°C, the reactions were stopped by adding Laemmli sample buffer and the samples were boiled for 3 min. Proteins were analyzed by SDS–PAGE, and phosphoprotein bands visualized by autoradiography were excised from the gel. The radioactivity incorporated was measured by liquid scintillation counting.
Preincubation of Tissue Slices This treatment was performed as described by de Freitas et al. (5). Cerebral cortex slices of 17-day-old rats were incubated in the absence or presence of 1.0 mM L-phenylalanine plus 1.0 mM a-methylphenylalanine. Incubation was carried out in a Dubnoff metabolic shaker for 1 h at 30°C in an atmosphere of 95% O2 and 5% CO2. Each flask contained approximately 600 mg of cerebral cortex in 2.0 ml Krebs–Ringer bicarbonate (KRB), pH 7.4, and 5 mM glucose containing all the protease inhibitors described below and 1 µM calpain inhibitor (Sigma. St. Louis, MO). Incubation was stopped by the addition of 20 vol ice-cold cytoskeletal extraction buffer and performed as described below. Preparation of a Triton-Insoluble Cytoskeletal Fraction The cytoskeletal fraction was obtained from cerebral cortex of animals submitted to the in vivo treatment or from tissue slices preincubated with the drugs. The cytoskeletal fraction was prepared as described by de Mattos et al. (7). Six hundred milligrams of tissue was homogenized 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 phenylmethylsulfonyl fluoride (PMSF), 1 mM benzamidine, 1 µM leupeptin, 0.7 µM antipain, 0.7 µM pepstatin, and
Incorporation Assay
Protein Kinase Inhibitor Assay Preincubated tissue slices were used to study the effect of the drugs on endogenous Ca21/calmodulindependent protein kinase II (CaMKII) and cAMPdependent protein kinase (PKA) using specific kinase inhibitors. The cytoskeletal fraction and the phosphorylation system were essentially as described above except that 32P incorporation was carried out in the absence or presence of 1.0 mM L-phenylalanine plus 1.0 mM a-methylphenylalanine into the assay mixture. Inhibition of PKA activity was assayed by adding 80 µM protein kinase A inhibitor fragment 6-22 amide (PKAI) (Sigma). Inhibition of CaMKII activity was performed in reaction mixtures containing 10 µM KN93 (Calbiochem Corporation, San Diego, CA) and the preparation was preincubated for 20 min at 30°C prior to in vitro phosphorylation. Results were expressed as percentage incorporation considering the control (absence of drugs or inhibitors) as 100%.
190
DE FREITAS ET AL.
Time-Course Dephosphorylation Assay Preincubated tissue slices were used to study the effect of the drugs on the phosphatase activity (PP1) associated with the cytoskeletal fraction. The cytoskeletal fraction and the phosphorylation system were obtained essentially as described above and in vitro 32P incorporation was performed in the absence or presence of 1.0 mM L-phenylalanine and 1.0 mM a-methylphenylalanine. In some assays, these drugs were added to the reaction mixtures after 5 min of incubation. In some assay mixtures, okadaic acid (Sigma), previously prepared as a 50 µM stock solution in 10% dimethyl sulfoxide (DMSO), was added to a final concentration of 0.5 µM. The reactions were stopped at specific intervals (5, 10, 20, and 30 min) by adding Laemmli sample buffer. SDS–PAGE Polyacrylamide gel electrophoresis was performed on 10% acrylamide according to the discontinuous system of Laemmli (16). Gels were stained with 0.25% (w/v) Coomassie blue R-250 (Sigma), 50% (v/v) methanol, and 10% (v/v) acetic acid and destained overnight in 50% methanol and 10% acetic acid. Quantitation Procedure Cytoskeletal proteins were quantified following SDS– PAGE as described by Rubin et al. (25). Briefly, destained, dried SDS–PAGE gels were scanned with a densitometer (Hoefer Scientific Instruments GS 300 Transmittance/Reflectance 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 percentage contribution to the total area. Therefore, cytoskeletal protein concentrations were calculated from percentage values considering that the total protein concentration of the cytoskeletal fraction measured by the Lowry method corresponds to 100% (total area of the densitometric scan). Statistical Analysis Data were analyzed by the Student t test, by the paired Student t test, or by one-way analysis of variance followed by the Duncan multiple range test when the F test was significant. All analyses were performed using the SPSS software program loaded on an IBM-PC compatible computer. RESULTS
The effect of chronic treatment with Phe plus MePhe on body and cerebral weights was estimated at the end
of treatment (14th day of life) and 3 days after treatment discontinuance (17th day of life). Body weight was not affected by the treatment (data not shown), but cerebral cortex weights were reduced at the end of treatment (544 mg 6 11; P , 0.01) as compared to control animals (624 mg 6 8). However, 3 days after treatment discontinuation, the cerebral cortex weight of treated animals (650 mg 6 12) did not differ from that of controls (669 mg 6 5). Under the experimental conditions used in this study, the NF-M subunit, a and b tubulin were good substrates for the endogenous phosphorylating system associated with the cytoskeletal fraction as previously described by us (6). Chronic administration of the drugs from 6th to 14th day induced an increased in vitro 32P incorporation into NF-M subunit, a and b tubulin, compared to control values both at the end of treatment (14th day of life) (Fig. 1A) and 3 days after treatment discontinuation (17th day of life) (Fig. 1B). We also investigated the effect of Phe plus MePhe on the in vitro cytoskeletal-associated phosphorylating system from cerebral cortex of 14-day-old chronically treated rats by adding the drugs to the phosphorylation assay, as described under Materials and Methods. The results showed a decreased level of in vitro 32P incorporation into the cytoskeletal proteins as a consequence of the addition of the drugs compared to treated animals (Fig. 1A) whose cytoskeletal fraction was phosphorylated in the absence of the drugs. In order to examine the effect of Phe plus MePhe specifically on the protein kinase activities associated with the cytoskeletal fraction, we carried out 32P incorporation experiments in the absence or presence of specific kinase inhibitors. The cytoskeletal fraction obtained after pretreatment of tissue slices with Phe plus MePhe was incubated with the same drugs in the absence or presence of PKA inhibitor (PKAI) or CaMKII inhibitor (KN93) and in vitro phosphorylation was carried out as described under Materials and Methods. Figure 2A shows that when PKAI or the drugs were added to the phosphorylation system, there was a decreased level of 32P incorporation into NF-M, a and b tubulin. However, an additive effect on these cytoskeletal proteins was observed by adding both PKAI and Phe plus MePhe, suggesting that although PKA is inhibited the effect of the drugs is not altered. When the incorporation assays were carried out in the presence of KN93 or Phe plus MePhe, there was a decreased level of 32P incorporation into NF-M, a and b tubulin (Fig. 2B). However, in the presence of both KN93 and the drugs we did not observe an additive inhibition. Conversely, in this case inhibition was similar to that obtained only with KN93, indicating that when CaMKII activity is inhibited the effect of the drugs is not detected. These findings suggest that the inhibitory effect of Phe plus MePhe on the in vitro 32P
HYPERPHENYLALANINEMIA AND CYTOSKELETON
191
incorporation into these cytoskeletal proteins may be due to CaMKII inhibition. The observation that Phe plus MePhe could affect kinase activity led us to investigate the effect of these drugs on the phosphatase associated with the cytoskeletal fraction. As previously demonstrated, only phosphatase type 1 (PP1) is associated with this cytoskeletal fraction (6). In Fig. 3, we can observe the in vitro dephosphorylation of NF-M, a and b tubulin obtained from tissue slices preincubated in the absence or presence of the drugs. Results demonstrate that the in vitro phosphorylation level of cytoskeletal proteins extracted
FIG. 2. Effect of Phe plus MePhe on cytoskeletal protein phosphorylation by PKA and CaMKII. (A) The cytoskeletal fraction obtained after preincubation of tissue slices with Phe plus MePhe was incubated in the presence of PKA inhibitor (PKAI), in the presence of Phe 1 MePhe, or in the presence of PKAI 1 Phe 1 MePhe. Controls were incubated without drug addition. (B) The cytoskeletal fraction obtained after preincubation of tissue slices in the presence of Phe plus MePhe was incubated in the presence of KN93, in the presence of Phe 1 MePhe, or in the presence of KN93 1 Phe 1 MePhe. Controls were incubated without drug addition. In vitro 32P incorporation was measured as described under Materials and Methods. Data are mean 6 SEM values for eight independent experiments. Statistically significant differences from controls (*P , 0.05; #P , 0.01) and from the other experimental groups (a, P , 0.05) are indicated.
FIG. 1. Effect of Phe plus MePhe treatment on in vitro incorporation of 32P into NF-M, a and b tubulin of the cytoskeletal fraction. (A) At the end of the treatment (14th day). (B) Three days after treatment discontinuation (17th day). The cytoskeletal fraction was extracted and in vitro 32P incorporation was measured as described under Materials and Methods. Data are mean 6 SEM values for nine independent experiments. Statistically significant differences from controls (*P , 0.05; #P , 0.01) and from treated animals (**P , 0.01) are indicated.
from control tissue slices is decreased up to 30 min of incubation. However, when tissue slices were preincubated with the drugs the in vitro phosphorylation level was decreased compared to controls. These results are in agreement with our previous experiments demonstrating that the drugs inhibit in vitro phosphorylation. In addition, the phosphorylation level remained constant between 5 and 30 min. This dephosphorylation pattern is consistent with that corresponding to inhibited phosphatase activity, as we can observe by comparison with the dephosphorylation pattern obtained in the presence of 0.5 µM okadaic acid, a concentration known
192
DE FREITAS ET AL.
FIG. 3. Time-course dephosphorylation. The cytoskeletal fraction obtained after preincubation of tissue slices with Phe plus MePhe was incubated in the presence of Phe 1 MePhe or in the presence of okadaic acid in a time-course dephosphorylation step carried out as described under Materials and Methods. Data are representative of three experiments.
to inhibit PP1. In addition, when these drugs were added to the phosphorylating system after 5 min of incubation, we can observe the same dephosphorylation pattern obtained by inhibition with okadaic acid of PP1 (Fig. 4). DISCUSSION
Hyperphenylalaninemia chemically induced by Phe plus MePhe administration has been used as an animal model of PKU (11). Subcutaneous injections of 5.2 µmol/g body weight Phe twice a day plus 2.4 µmol/g MePhe once a day beginning at 3 days postpartum induce a significant reduction in body and brain weight,
FIG. 4. Effect of Phe plus MePhe on cytoskeletal protein dephosphorylation by PP1. In vitro phosphorylation of cytoskeletal fraction was carried out and Phe 1 MePhe were added after 5 min of incubation. A time-course dephosphorylation step carried out as described under Material and Methods. Data are representative of three experiments.
HYPERPHENYLALANINEMIA AND CYTOSKELETON
observable at 30 days of age (17). Under this treatment, plasma Phe levels are higher than 3.0 mM. By injecting 3.0 µmol/g Phe twice a day plus 1.6 µmol/g MePhe once a day, beginning on the 6th day of age (when rat brain development is equivalent to the newborn human brain), as we did, plasma Phe levels are similar to those found in human phenylketonuria (between 1.2 and 3.0 mM ) and the animals present a persistent reduction of brain myelin content and a persistent deficit in learning and memory, measured by behavioral tasks (unpublished data). In the present study, the reduction of the cerebral cortex weight observed at the end of chronic treatment could be related to hypomyelination found in animal models. In the present work, we studied the effect of hyperphenylalaninemia on in vitro incorporation of 32P into cytoskeletal proteins by injecting the drugs subcutaneously from the 6th to the 14th day postpartum. The cytoskeletal fraction was prepared from the cerebral cortex of rats immediately after treatment (14 days old) or 3 days after treatment discontinuance (17 days old) and used for in vitro phosphorylation. The cytoskeletal proteins can be phosphorylated in vitro by associated endogenous protein kinases (27). We have previously described the phosphorylation/dephosphorylation system associated with the Triton-insoluble cytoskeletal fraction that phosphorylates/dephosphorylates in vitro NF-M, a and b tubulin from the cerebral cortex of young rats (6). This phosphorylation system comprises the cAMP-dependent and calcium/calmodulin-dependent protein kinases, as well as protein phosphatase type 1. Our results demonstrate that hyperphenylalaninemia induced an increased in vitro phosphorylation of both NF-M and tubulins present in the cytoskeletal fraction (Fig. 1A). Three days after treatment discontinuation, when plasma Phe levels were normal and MePhe was no longer detectable, in vitro incorporation of 32P into NF-M, a and b tubulin remained increased (Fig. 1B). We ascribe the greater in vitro incorporation of 32P into NF-M, a and b tubulin in treated animals to an increased number of vacant sites able to incorporate in vitro phosphate. This effect may be due to the inhibition of the endogenous phosphorylation or the stimulation of the endogenous dephosphorylation occurring during drug administration. We investigated the in vitro phosphorylation of the cytoskeletal proteins from treated animals in the presence of the drugs to verify whether their effect was directly on the phosphorylation system. Our results demonstrated that Phe plus MePhe added to the in vitro 32P assay system decreased the radioactivity incorporated into the proteins studied (Fig. 1A). These results are in agreement with our previous findings demonstrating that preincubation of tissue slices with different concentrations of
193
the drugs for 1 h, followed by measurement of in vitro phosphorylation in the presence of the same drugs, reduced in vitro 32P incorporation into NF-M and tubulins (5). The mechanisms underlying the effects of Phe and MePhe on the in vitro phosphorylating/ dephosphorylating system could be ascribed to the action of the drugs directly on protein kinases or protein phosphatases. To verify this possibility, we initially examined the effects of Phe and MePhe on the endogenous PKA and CaMK activities associated with the cytoskeletal fraction by using specific protein kinase inhibitors. The protein kinase A inhibitor fragment 6-22 amide was used for inhibition of PKA and KN93 for inhibition of CaMKII. These compounds have been shown to be effective and selective inhibitors of these protein kinases (10, 20). Our results have demonstrated that the inhibitory effect of PKAI and Phe plus MePhe on the in vitro 32P incorporation into NF-M, a and b tubulin (34–39%) represented the sum of the inhibition induced by PKAI (21–28%) or Phe plus MePhe (16–18%) estimated independently (Fig. 2A). However, the inhibitory effect of KN93 and Phe plus MePhe (26–33%) was similar to the inhibition by KN93 only (26–39%) (Fig. 2B). These findings suggest that the drugs induce their inhibitory effect on in vitro 32P incorporation by a mechanism independent of PKA activity. Otherwise, this mechanism seems to be related to CaMKII activity, since when this enzyme is blocked we do not detect the effect of the drugs. We also examined the effect of Phe plus MePhe on PP1 activity. For this study, we performed a timecourse dephosphorylation experiment. Results demonstrated that after experimental treatment with Phe plus MePhe dephosphorylation was inhibited, suggesting the action of these drugs on PP1 activity. The present study demonstrates that in vitro phosphorylation of cytoskeletal proteins was altered by chronic hyperphenylalaninemia induced from the 6th to the 14th day postpartum and this effect persisted 3 days after treatment discontinuation. Furthermore, in vitro experiments demonstrated that phenylalanine and a-methylphenylalanine have an inhibitory effect on calcium/calmodulin-dependent protein kinase II and protein phosphatase type 1 activities. We could consider that this inhibitory effect would be responsible for the persistent effect observed after treatment discontinuation. At present, we cannot exclude the effect of Phe plus MePhe on other protein kinases and phosphatases acting in vivo. In neurons, the dense and highly organized neurofilament and microtubule arrays serve to support nerve axons (14). Microtubules are major components of the neuronal cytoskeleton (4). It is known that not all a and
194
DE FREITAS ET AL.
b tubulin of brain cortex is found in neurons. Considering the ubuquitous distribution of tubulin, our results could be related to neuronal and glial tubulins. Phosphorylation of the cytoskeletal proteins is related to their capacity of association or interaction with other cytoskeletal proteins playing important roles in regulating the dynamic stability of the axonal cytoskeleton (14). Several studies have demonstrated the relationship between the phosphorylation of neurofilaments and tubulins and the interaction of these proteins with other axonal constituents (18, 28). Recent studies have shown that myelination or events associated with myelination have a critical influence on the changes in neurofilament dynamics and axon caliber expansion. Phosphorylation events at carboxylterminal tail domains of NF-H and NF-M promote caliber expansion, particularly prominent in myelinated axons. The interaction between myelinating glia and axon modulates phosphorylation levels of neurofilaments and possibly of other neuronal proteins (8, 24). According to these studies, myelin affects the phosphorylation of NF-H and NF-M subunits. The primary pathologic finding in the nervous system of PKU patients and in rats submitted to chronic hyperphenylalaninemia is hypomyelination. Thus, the reduced phosphorylation of cytoskeletal proteins, at least NF-M, could be related to hypomyelination in the animal model. However, myelin alterations could not explain in vitro effects of the drugs on partially purified cytoskeletal proteins. Taking into account the importance of the phosphorylation/dephosphorylation dynamics of cytoskeletal proteins in mature neurons as well as during brain development (9, 3) and considering that Phe plus MePhe addition causes CaMKII and PP1 inhibition, this effect could be related to the brain damage observed in experimental phenylketonuria. REFERENCES 1.
2.
3. 4.
5.
BAUMAN, M. L., AND T. L. KEMPER. 1982. Morphologic and histoanatomic observations of the brain in untreated phenylketonuria. Acta Neuropathol. 58: 55–63. BRADFORD, M. M. 1977. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248–254. BRADY, S. T. 1992. Axonal dynamics regeneration. In Neural Regeneration (A. Gorio, Ed.) pp. 7–36. Raven Press, New York. COHEN, E., S. BINET, AND V. MEININGER. 1987. In situ appearance of the cold-stable microtubules in the growing axons of the tectal plate of mouse investigated immunocytochemically after polyethyleneglycol (PEG) embedding. Dev. Brain Res. 36: 171–180. DE FREITAS, M. S., A. G. DE MATTOS, M. M. CAMARGO, C. M. D. WANNMACHER, AND R. PESSOA-PUREUR. 1995. Effect of phenylalanine and a-methylphenylalanine on in vitro incorporation of 32P into cytoskeletal cerebral proteins. Neurochem. Int. 26: 381– 385.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18.
19.
20.
21.
22.
DE FREITAS, M. S., A. G. DE MATTOS, M. M. CAMARGO, C. M. D. WANNMACHER, AND R. PESSOA-PUREUR. 1995. Cytoskeletalassociated protein kinase and phosphatase activities from cerebral cortex of young rats. Neurochem. Res. 20: 951–956. DE MATTOS, A. G., M. S. DE FREITAS, M. M. CAMARGO, AND R. PESSOA-PUREUR. 1994. Developmentally regulated in vitro phosphorylation of cytoskeletal proteins of cerebral cortex of normal and malnourished rats. Dev. Neurosci. 16: 38–43. DE WAEGH, S. M., V. M.-Y LEE, AND S. T. BRADY. 1992. Local modulation of neurofilament phosphorylation, axonal caliber, and slow axonal transport by myelinating Schwann cells. Cell 68: 451–463. D´IAZ-NIDO, J., L. SERRANO, M. A. HERNA´ NDEZ, AND J. AVILA. 1990. Phosphorylation of microtubule proteins in rat brain at different developmental stages: Comparison with that found in neuronal cultures. J. Neurochem. 54: 211–222. GLASS, D. B., L. J. LUNDOVIST, B. M. KATZ, AND D. A. WALSY. 1989. Protein kinase inhibitor-(6-22)-amide peptide analogs with standard and nonstandard amino acids substitutions for phenylalanine 10. Inhibition of cAMP-dependent protein kinase. J. Biol. Chem. 264: 14579–14584. GREENGARD, O., M. S. YOSS, AND J. A. DELVALLE. 1976. Alphamethylphenylalanine, a new inducer of chronic hyperphenylalaninemia in suckling rats. Science 192: 1007–1008. HOFFMAN, P. N., AND R. J. LASEK. 1975. The slow component of axonal transport. Identification of major structural polypeptides of the axon and their generality among mammalian neurons. J. Cell Biol. 66: 351–366. HOFFMAN, P. N., J. W. GRIFFIN, AND D. L. PRICE. 1984. Control of axonal caliber by neurofilament transport. J. Cell Biol. 99: 705–714. HOLLENBECK, P. J. 1989. The transport and assembly of the axonal cytoskeleton. J. Cell Biol. 108: 223–227. HOMMES, F. A. 1991. On the mechanism of permanent brain dysfunction in hyperphenylalaninemia. Biochem. Med. Metab Biol. 46: 277–287. LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277: 680–685. LANE, J. D., B. SCHONE, U. LANGENBECK, AND V. NEUHOFF. 1980. Characterization of experimental phenylketonuria. Augmentation of hyperphenylalaninemia with a-methylphenylalanine and p-chlorophenylalanine. Biochim. Biophys. Acta 627: 144– 156. LANNING, C. L., K. R. WILMARTH, AND M. B. ABOU-DONIA. 1994. In vitro binding of [14C]2,5-hexanedione to rat neuronal cytoskeletal proteins. Neurochem. Res. 19: 1165–1173. LOWRY, O. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265–267. MAMIYA, N., J. R. GOLDERING, Y. TSUNODA, I. M. MODLIN, K. YASUI, N. USUDA, T. ISHIKAWA, A. NATSUME, AND H. HIDAKA. 1993. Inhibition of acid secretion in gastric parietal cells by the Ca21/calmodulin-dependent protein kinase II inhibitor KN-93. Biochem. Biophys. Res. Commun. 195: 608–615. MIYASAKA, H., S. OKABE, K. ISHIGURO, T. UCHIDA, AND N. HIROKAWA. 1993. Interaction of the tail domain of high molecular weight subunits of neurofilaments with the COOH-terminal region of tubulin and its regulation by tau protein kinase II. J. Biol. Chem. 268: 22695–22702. MORRIS, J. R., AND R. J. LASEK. 1982. Stable polymers of the axonal cytoskeleton: The axoplasmic ghost. J. Cell. Biol. 92: 192–198.
HYPERPHENYLALANINEMIA AND CYTOSKELETON 23.
NIXON, R. A., AND R. K. SIHAG. 1991. Neurofilament phosphorylation: a new look at regulation and function. Trends Neurosci. 14: 501–506. 24. NIXON, R. A., P. A. PASKEVICH, R. K. SIHAG, AND C. Y. THAYER. 1994. 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: 1031–1046. 25. RUBIN, M. A., C. M. D. WANNMACHER, G. B. VALENTE, M. M. CAMARGO, AND R. P. PUREUR. 1992. Diminished concentration of the NF-H subunit of neurofilaments in cerebral cortex of rats chronically treated with proline, methylmalonate and phenylalanine plus a-methylphenylalanine. J. Inher. Met. Dis. 15: 252– 260. 26. SHAH, S. N., N. A. PETERSON, AND C. M. MCKEAN. 1972. Lipid
195
composition of human cerebral white matter and myelin in phenylketonuria. J. Neurochem. 19: 2369–2376. 27. SIHAG, R. K., AND R. A. NIXON. 1989. In vivo phosphorylation of distinct domains of the 70-kilodalton neurofilament subunit involves different protein kinases. J. Biol. Chem. 264: 457–464. 28. TASHIRO, T., R. IMAI, AND Y. KOMIYA. 1994. Early effects of b,b8-iminodipropionitrile on tubulin solubility and neurofilament phosphorylation in the axon. J. Neurochem. 63: 291–300. 29. VALE, R. D., B. J. SCHMAPP, T. MITCHISON, E. STEVER, T. S. REESE, AND M. P. SHEETZ. 1985. Different axoplasmic proteins generate movement in opposite directions along microtubules in vitro. Cell 43: 623–632. 30. WANDOSELL, F., L. SERRANO, AND J. AVILA. 1987. Phosphorylation of a tubulin carboxyl terminal tyrosine prevents its incorporation into microtubules. J. Biol. Chem. 262: 8268–8273.
143, 188–195 (1997)
EN966351
Effect of Hyperphenylalaninemia Chemically Induced on in Vitro Incorporation of 32P into Cytoskeletal Proteins from Cerebral Cortex of Developing Rats MARTA S. DE FREITAS, ANGELA DE MATTOS-DUTRA, NADJA SCHRODER, CLO´ VIS M. D. WANNMACHER, AND REGINA PESSOA-PUREUR Departamento de Bioquı´mica, Instituto de Biocieˆncias, Universidade Federal do Rio Grande do Sul, Porto Alegre RS, Brazil
We studied the effect of hyperphenylalaninemia on in vitro incorporation of 32P into cytoskeletal proteins from cerebral cortex of rats by injecting L-phenylalanine plus a-methylphenylalanine subcutaneously from the 6th to the 14th day postpartum. Chronic hyperphenylalaninemia induced an increased in vitro phosphorylation of the 150-kDa neurofilament subunit and tubulins present in the cytoskeletal fraction at the end of the treatment and 3 days after treatment discontinuation. In addition, when in vitro phosphorylation of the cytoskeletal proteins from treated animals was performed in the presence of the drugs we observed a decreased in vitro incorporation of 32P into these proteins. Thus, the effect of L-phenylalanine plus a-methylphenylalanine on the endogenous protein kinase and phosphatase activities was examined and the results demonstrated that these drugs have an inhibitory effect on calcium/calmodulin-dependent protein kinase II and protein phosphatase type 1. r 1997 Academic Press
INTRODUCTION
The neuronal cytoskeleton comprises a protein network formed mainly of neurofilaments and microtubules interacting with one another and with a variety of associated proteins. Axons contain many longitudinally oriented neurofilaments and microtubules occupying most of the axoplasm (14). Neurofilaments are the intermediate neuronal filaments. They are composed of three subunits that exhibit apparent molecular masses of 200 (NF-H), 150 (NF-M), and 68 kDa (NF-L) by SDS–PAGE (12). Neurofilaments are believed to play an important role in the maintenance of neuronal shape and determination of axonal caliber (22, 13). 1 This work was supported by Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq), Fundac¸a˜o de Amparo a` Pesquisa do Estado do RGS (FAPERGS), and Pro´-Reitoria de Pesquisa e Po´s-Graduac¸a˜o da Universidade Federal do RGS (PROPESPUFRGS).
0014-4886/97 $25.00 Copyright r 1997 by Academic Press All rights of reproduction in any form reserved.
Microtubules are mainly constituted of a and b tubulin subunits. In the brain, they are involved in a variety of cellular functions including axonal transport in the mature neuron and synaptogenesis (29). Tubulins and neurofilaments undergo posttranslational modifications that modulate their physiological role in the neuronal cells. Tubulins and neurofilament subunits can be phosphorylated in their terminal regions by different protein kinases (30, 23). Phosphorylation regulates the ability of tubulin and neurofilament subunits to self-assemble into microtubules and neurofilaments, respectively. Moreover, their phosphorylation levels modulate the interaction between microtubules and neurofilaments and other cytoskeletal proteins (21). Phenylketonuria (PKU) is an inborn error of metabolism characterized by a severe deficiency or absence of hepatic phenylalanine hydroxylase (PAH) activity associated with brain dysfunction. Several studies have demonstrated the association of permanent brain damage with increased phenylalanine (Phe) levels during critical periods of brain development (15). Biochemical studies have shown a severe reduction in the amount of myelin in brains of untreated PKU patients and retarded synapse formation and dendritic arborization (26, 1). Animal models have been developed to study the pathophysiology of the biochemical and behavioral abnormalities found in human PKU. We have previously demonstrated that hyperphenylalaninemia induced in rats by treating the animals with daily administration of Phe plus a-methylphenylalanine (MePhe), a PAH inhibitor, reduces the NF-H content in cerebral cortex (25). We have also demonstrated that preincubation of tissue slices with Phe plus MePhe decreases in vitro 32P incorporation into NF-M and tubulins (5). In the present study we used the animal model to study the effects of hyperphenylalaninemia on in vitro 32P incorporation into cytoskeletal proteins. In addition, we report the inhibitory effect of Phe plus MePhe on the phosphorylating system associ-
188
HYPERPHENYLALANINEMIA AND CYTOSKELETON
ated with the cytoskeletal fraction from cerebral cortex of young rats. MATERIALS AND METHODS
Animals Wistar rats from our breeding stock were maintained on a 12-h light/12-h dark cycle in a constant temperature (22°C) colony room. On the day of birth the litter size was culled to eight pups. Litters with fewer than eight pups were not included in the experiments. Free water and a 20% (w/w) protein commercial chow were provided.
189
0.7 pM chymostatin (Sigma). The homogenate was centrifuged at 13,000g for 15 min at 4°C. The pellet was resuspended in 40 ml of the same buffer containing 0.85 M sucrose and centrifuged for 15 min at 13,000g. The pellet was dissolved in 50 mM 2(N-morpholino)ethanesulfonic acid (MES), pH 6.5, and 10 mM MgCl2 followed by protein determination (2). An aliquot containing 10 µg protein was removed for in vitro 32P incorporation assays. The remaining material was precipitated with 5% TCA, washed twice with ethanol and acetone, and dissolved in 1% SDS, and the protein concentration was determined (19). This fraction was analyzed by SDS– PAGE and used for quantitation as described below.
In Vivo Treatment
32P
Rats from the same litter were randomly divided into two groups. Rats from the experimental group received 3.0 µmol/g body weight of L-phenylalanine twice a day at 10-h intervals plus 1.6 µmol/g a-methylphenylalanine once a day. Drugs were dissolved in 0.9% saline and administered subcutaneously from the 6th to the 14th day postpartum in a volume of 20 µl/g body weight. Control animals received the same volume of saline. Rats were sacrificed on the 14th day or on the 17th day; the cerebral cortex was immediately removed and the cytoskeletal fraction was extracted as described below.
An enriched cytoskeletal fraction prepared as described above served as protein substrate and as a source of endogenous protein kinases and protein phosphatases. Each assay mixture contained 10 µg of protein. Phosphorylation was carried out in 60 µl of a buffer containing 50 mM MES, pH 6.5, and 10 mM MgCl2. In some assay mixtures, the 32P incorporation into the cytoskeletal fraction from chronically treated animals was performed by adding 1.0 mM L-phenylalanine plus 1.0 mM a-methylphenylalanine to the phosphorylation assay. The reaction was started by adding 2.0 µCi [g-32P]ATP (16.6 3 10210 Bq/mmol) (ICN Radiochemicals, Irvine, CA). After incubation for 5 min at 30°C, the reactions were stopped by adding Laemmli sample buffer and the samples were boiled for 3 min. Proteins were analyzed by SDS–PAGE, and phosphoprotein bands visualized by autoradiography were excised from the gel. The radioactivity incorporated was measured by liquid scintillation counting.
Preincubation of Tissue Slices This treatment was performed as described by de Freitas et al. (5). Cerebral cortex slices of 17-day-old rats were incubated in the absence or presence of 1.0 mM L-phenylalanine plus 1.0 mM a-methylphenylalanine. Incubation was carried out in a Dubnoff metabolic shaker for 1 h at 30°C in an atmosphere of 95% O2 and 5% CO2. Each flask contained approximately 600 mg of cerebral cortex in 2.0 ml Krebs–Ringer bicarbonate (KRB), pH 7.4, and 5 mM glucose containing all the protease inhibitors described below and 1 µM calpain inhibitor (Sigma. St. Louis, MO). Incubation was stopped by the addition of 20 vol ice-cold cytoskeletal extraction buffer and performed as described below. Preparation of a Triton-Insoluble Cytoskeletal Fraction The cytoskeletal fraction was obtained from cerebral cortex of animals submitted to the in vivo treatment or from tissue slices preincubated with the drugs. The cytoskeletal fraction was prepared as described by de Mattos et al. (7). Six hundred milligrams of tissue was homogenized 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 phenylmethylsulfonyl fluoride (PMSF), 1 mM benzamidine, 1 µM leupeptin, 0.7 µM antipain, 0.7 µM pepstatin, and
Incorporation Assay
Protein Kinase Inhibitor Assay Preincubated tissue slices were used to study the effect of the drugs on endogenous Ca21/calmodulindependent protein kinase II (CaMKII) and cAMPdependent protein kinase (PKA) using specific kinase inhibitors. The cytoskeletal fraction and the phosphorylation system were essentially as described above except that 32P incorporation was carried out in the absence or presence of 1.0 mM L-phenylalanine plus 1.0 mM a-methylphenylalanine into the assay mixture. Inhibition of PKA activity was assayed by adding 80 µM protein kinase A inhibitor fragment 6-22 amide (PKAI) (Sigma). Inhibition of CaMKII activity was performed in reaction mixtures containing 10 µM KN93 (Calbiochem Corporation, San Diego, CA) and the preparation was preincubated for 20 min at 30°C prior to in vitro phosphorylation. Results were expressed as percentage incorporation considering the control (absence of drugs or inhibitors) as 100%.
190
DE FREITAS ET AL.
Time-Course Dephosphorylation Assay Preincubated tissue slices were used to study the effect of the drugs on the phosphatase activity (PP1) associated with the cytoskeletal fraction. The cytoskeletal fraction and the phosphorylation system were obtained essentially as described above and in vitro 32P incorporation was performed in the absence or presence of 1.0 mM L-phenylalanine and 1.0 mM a-methylphenylalanine. In some assays, these drugs were added to the reaction mixtures after 5 min of incubation. In some assay mixtures, okadaic acid (Sigma), previously prepared as a 50 µM stock solution in 10% dimethyl sulfoxide (DMSO), was added to a final concentration of 0.5 µM. The reactions were stopped at specific intervals (5, 10, 20, and 30 min) by adding Laemmli sample buffer. SDS–PAGE Polyacrylamide gel electrophoresis was performed on 10% acrylamide according to the discontinuous system of Laemmli (16). Gels were stained with 0.25% (w/v) Coomassie blue R-250 (Sigma), 50% (v/v) methanol, and 10% (v/v) acetic acid and destained overnight in 50% methanol and 10% acetic acid. Quantitation Procedure Cytoskeletal proteins were quantified following SDS– PAGE as described by Rubin et al. (25). Briefly, destained, dried SDS–PAGE gels were scanned with a densitometer (Hoefer Scientific Instruments GS 300 Transmittance/Reflectance 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 percentage contribution to the total area. Therefore, cytoskeletal protein concentrations were calculated from percentage values considering that the total protein concentration of the cytoskeletal fraction measured by the Lowry method corresponds to 100% (total area of the densitometric scan). Statistical Analysis Data were analyzed by the Student t test, by the paired Student t test, or by one-way analysis of variance followed by the Duncan multiple range test when the F test was significant. All analyses were performed using the SPSS software program loaded on an IBM-PC compatible computer. RESULTS
The effect of chronic treatment with Phe plus MePhe on body and cerebral weights was estimated at the end
of treatment (14th day of life) and 3 days after treatment discontinuance (17th day of life). Body weight was not affected by the treatment (data not shown), but cerebral cortex weights were reduced at the end of treatment (544 mg 6 11; P , 0.01) as compared to control animals (624 mg 6 8). However, 3 days after treatment discontinuation, the cerebral cortex weight of treated animals (650 mg 6 12) did not differ from that of controls (669 mg 6 5). Under the experimental conditions used in this study, the NF-M subunit, a and b tubulin were good substrates for the endogenous phosphorylating system associated with the cytoskeletal fraction as previously described by us (6). Chronic administration of the drugs from 6th to 14th day induced an increased in vitro 32P incorporation into NF-M subunit, a and b tubulin, compared to control values both at the end of treatment (14th day of life) (Fig. 1A) and 3 days after treatment discontinuation (17th day of life) (Fig. 1B). We also investigated the effect of Phe plus MePhe on the in vitro cytoskeletal-associated phosphorylating system from cerebral cortex of 14-day-old chronically treated rats by adding the drugs to the phosphorylation assay, as described under Materials and Methods. The results showed a decreased level of in vitro 32P incorporation into the cytoskeletal proteins as a consequence of the addition of the drugs compared to treated animals (Fig. 1A) whose cytoskeletal fraction was phosphorylated in the absence of the drugs. In order to examine the effect of Phe plus MePhe specifically on the protein kinase activities associated with the cytoskeletal fraction, we carried out 32P incorporation experiments in the absence or presence of specific kinase inhibitors. The cytoskeletal fraction obtained after pretreatment of tissue slices with Phe plus MePhe was incubated with the same drugs in the absence or presence of PKA inhibitor (PKAI) or CaMKII inhibitor (KN93) and in vitro phosphorylation was carried out as described under Materials and Methods. Figure 2A shows that when PKAI or the drugs were added to the phosphorylation system, there was a decreased level of 32P incorporation into NF-M, a and b tubulin. However, an additive effect on these cytoskeletal proteins was observed by adding both PKAI and Phe plus MePhe, suggesting that although PKA is inhibited the effect of the drugs is not altered. When the incorporation assays were carried out in the presence of KN93 or Phe plus MePhe, there was a decreased level of 32P incorporation into NF-M, a and b tubulin (Fig. 2B). However, in the presence of both KN93 and the drugs we did not observe an additive inhibition. Conversely, in this case inhibition was similar to that obtained only with KN93, indicating that when CaMKII activity is inhibited the effect of the drugs is not detected. These findings suggest that the inhibitory effect of Phe plus MePhe on the in vitro 32P
HYPERPHENYLALANINEMIA AND CYTOSKELETON
191
incorporation into these cytoskeletal proteins may be due to CaMKII inhibition. The observation that Phe plus MePhe could affect kinase activity led us to investigate the effect of these drugs on the phosphatase associated with the cytoskeletal fraction. As previously demonstrated, only phosphatase type 1 (PP1) is associated with this cytoskeletal fraction (6). In Fig. 3, we can observe the in vitro dephosphorylation of NF-M, a and b tubulin obtained from tissue slices preincubated in the absence or presence of the drugs. Results demonstrate that the in vitro phosphorylation level of cytoskeletal proteins extracted
FIG. 2. Effect of Phe plus MePhe on cytoskeletal protein phosphorylation by PKA and CaMKII. (A) The cytoskeletal fraction obtained after preincubation of tissue slices with Phe plus MePhe was incubated in the presence of PKA inhibitor (PKAI), in the presence of Phe 1 MePhe, or in the presence of PKAI 1 Phe 1 MePhe. Controls were incubated without drug addition. (B) The cytoskeletal fraction obtained after preincubation of tissue slices in the presence of Phe plus MePhe was incubated in the presence of KN93, in the presence of Phe 1 MePhe, or in the presence of KN93 1 Phe 1 MePhe. Controls were incubated without drug addition. In vitro 32P incorporation was measured as described under Materials and Methods. Data are mean 6 SEM values for eight independent experiments. Statistically significant differences from controls (*P , 0.05; #P , 0.01) and from the other experimental groups (a, P , 0.05) are indicated.
FIG. 1. Effect of Phe plus MePhe treatment on in vitro incorporation of 32P into NF-M, a and b tubulin of the cytoskeletal fraction. (A) At the end of the treatment (14th day). (B) Three days after treatment discontinuation (17th day). The cytoskeletal fraction was extracted and in vitro 32P incorporation was measured as described under Materials and Methods. Data are mean 6 SEM values for nine independent experiments. Statistically significant differences from controls (*P , 0.05; #P , 0.01) and from treated animals (**P , 0.01) are indicated.
from control tissue slices is decreased up to 30 min of incubation. However, when tissue slices were preincubated with the drugs the in vitro phosphorylation level was decreased compared to controls. These results are in agreement with our previous experiments demonstrating that the drugs inhibit in vitro phosphorylation. In addition, the phosphorylation level remained constant between 5 and 30 min. This dephosphorylation pattern is consistent with that corresponding to inhibited phosphatase activity, as we can observe by comparison with the dephosphorylation pattern obtained in the presence of 0.5 µM okadaic acid, a concentration known
192
DE FREITAS ET AL.
FIG. 3. Time-course dephosphorylation. The cytoskeletal fraction obtained after preincubation of tissue slices with Phe plus MePhe was incubated in the presence of Phe 1 MePhe or in the presence of okadaic acid in a time-course dephosphorylation step carried out as described under Materials and Methods. Data are representative of three experiments.
to inhibit PP1. In addition, when these drugs were added to the phosphorylating system after 5 min of incubation, we can observe the same dephosphorylation pattern obtained by inhibition with okadaic acid of PP1 (Fig. 4). DISCUSSION
Hyperphenylalaninemia chemically induced by Phe plus MePhe administration has been used as an animal model of PKU (11). Subcutaneous injections of 5.2 µmol/g body weight Phe twice a day plus 2.4 µmol/g MePhe once a day beginning at 3 days postpartum induce a significant reduction in body and brain weight,
FIG. 4. Effect of Phe plus MePhe on cytoskeletal protein dephosphorylation by PP1. In vitro phosphorylation of cytoskeletal fraction was carried out and Phe 1 MePhe were added after 5 min of incubation. A time-course dephosphorylation step carried out as described under Material and Methods. Data are representative of three experiments.
HYPERPHENYLALANINEMIA AND CYTOSKELETON
observable at 30 days of age (17). Under this treatment, plasma Phe levels are higher than 3.0 mM. By injecting 3.0 µmol/g Phe twice a day plus 1.6 µmol/g MePhe once a day, beginning on the 6th day of age (when rat brain development is equivalent to the newborn human brain), as we did, plasma Phe levels are similar to those found in human phenylketonuria (between 1.2 and 3.0 mM ) and the animals present a persistent reduction of brain myelin content and a persistent deficit in learning and memory, measured by behavioral tasks (unpublished data). In the present study, the reduction of the cerebral cortex weight observed at the end of chronic treatment could be related to hypomyelination found in animal models. In the present work, we studied the effect of hyperphenylalaninemia on in vitro incorporation of 32P into cytoskeletal proteins by injecting the drugs subcutaneously from the 6th to the 14th day postpartum. The cytoskeletal fraction was prepared from the cerebral cortex of rats immediately after treatment (14 days old) or 3 days after treatment discontinuance (17 days old) and used for in vitro phosphorylation. The cytoskeletal proteins can be phosphorylated in vitro by associated endogenous protein kinases (27). We have previously described the phosphorylation/dephosphorylation system associated with the Triton-insoluble cytoskeletal fraction that phosphorylates/dephosphorylates in vitro NF-M, a and b tubulin from the cerebral cortex of young rats (6). This phosphorylation system comprises the cAMP-dependent and calcium/calmodulin-dependent protein kinases, as well as protein phosphatase type 1. Our results demonstrate that hyperphenylalaninemia induced an increased in vitro phosphorylation of both NF-M and tubulins present in the cytoskeletal fraction (Fig. 1A). Three days after treatment discontinuation, when plasma Phe levels were normal and MePhe was no longer detectable, in vitro incorporation of 32P into NF-M, a and b tubulin remained increased (Fig. 1B). We ascribe the greater in vitro incorporation of 32P into NF-M, a and b tubulin in treated animals to an increased number of vacant sites able to incorporate in vitro phosphate. This effect may be due to the inhibition of the endogenous phosphorylation or the stimulation of the endogenous dephosphorylation occurring during drug administration. We investigated the in vitro phosphorylation of the cytoskeletal proteins from treated animals in the presence of the drugs to verify whether their effect was directly on the phosphorylation system. Our results demonstrated that Phe plus MePhe added to the in vitro 32P assay system decreased the radioactivity incorporated into the proteins studied (Fig. 1A). These results are in agreement with our previous findings demonstrating that preincubation of tissue slices with different concentrations of
193
the drugs for 1 h, followed by measurement of in vitro phosphorylation in the presence of the same drugs, reduced in vitro 32P incorporation into NF-M and tubulins (5). The mechanisms underlying the effects of Phe and MePhe on the in vitro phosphorylating/ dephosphorylating system could be ascribed to the action of the drugs directly on protein kinases or protein phosphatases. To verify this possibility, we initially examined the effects of Phe and MePhe on the endogenous PKA and CaMK activities associated with the cytoskeletal fraction by using specific protein kinase inhibitors. The protein kinase A inhibitor fragment 6-22 amide was used for inhibition of PKA and KN93 for inhibition of CaMKII. These compounds have been shown to be effective and selective inhibitors of these protein kinases (10, 20). Our results have demonstrated that the inhibitory effect of PKAI and Phe plus MePhe on the in vitro 32P incorporation into NF-M, a and b tubulin (34–39%) represented the sum of the inhibition induced by PKAI (21–28%) or Phe plus MePhe (16–18%) estimated independently (Fig. 2A). However, the inhibitory effect of KN93 and Phe plus MePhe (26–33%) was similar to the inhibition by KN93 only (26–39%) (Fig. 2B). These findings suggest that the drugs induce their inhibitory effect on in vitro 32P incorporation by a mechanism independent of PKA activity. Otherwise, this mechanism seems to be related to CaMKII activity, since when this enzyme is blocked we do not detect the effect of the drugs. We also examined the effect of Phe plus MePhe on PP1 activity. For this study, we performed a timecourse dephosphorylation experiment. Results demonstrated that after experimental treatment with Phe plus MePhe dephosphorylation was inhibited, suggesting the action of these drugs on PP1 activity. The present study demonstrates that in vitro phosphorylation of cytoskeletal proteins was altered by chronic hyperphenylalaninemia induced from the 6th to the 14th day postpartum and this effect persisted 3 days after treatment discontinuation. Furthermore, in vitro experiments demonstrated that phenylalanine and a-methylphenylalanine have an inhibitory effect on calcium/calmodulin-dependent protein kinase II and protein phosphatase type 1 activities. We could consider that this inhibitory effect would be responsible for the persistent effect observed after treatment discontinuation. At present, we cannot exclude the effect of Phe plus MePhe on other protein kinases and phosphatases acting in vivo. In neurons, the dense and highly organized neurofilament and microtubule arrays serve to support nerve axons (14). Microtubules are major components of the neuronal cytoskeleton (4). It is known that not all a and
194
DE FREITAS ET AL.
b tubulin of brain cortex is found in neurons. Considering the ubuquitous distribution of tubulin, our results could be related to neuronal and glial tubulins. Phosphorylation of the cytoskeletal proteins is related to their capacity of association or interaction with other cytoskeletal proteins playing important roles in regulating the dynamic stability of the axonal cytoskeleton (14). Several studies have demonstrated the relationship between the phosphorylation of neurofilaments and tubulins and the interaction of these proteins with other axonal constituents (18, 28). Recent studies have shown that myelination or events associated with myelination have a critical influence on the changes in neurofilament dynamics and axon caliber expansion. Phosphorylation events at carboxylterminal tail domains of NF-H and NF-M promote caliber expansion, particularly prominent in myelinated axons. The interaction between myelinating glia and axon modulates phosphorylation levels of neurofilaments and possibly of other neuronal proteins (8, 24). According to these studies, myelin affects the phosphorylation of NF-H and NF-M subunits. The primary pathologic finding in the nervous system of PKU patients and in rats submitted to chronic hyperphenylalaninemia is hypomyelination. Thus, the reduced phosphorylation of cytoskeletal proteins, at least NF-M, could be related to hypomyelination in the animal model. However, myelin alterations could not explain in vitro effects of the drugs on partially purified cytoskeletal proteins. Taking into account the importance of the phosphorylation/dephosphorylation dynamics of cytoskeletal proteins in mature neurons as well as during brain development (9, 3) and considering that Phe plus MePhe addition causes CaMKII and PP1 inhibition, this effect could be related to the brain damage observed in experimental phenylketonuria. REFERENCES 1.
2.
3. 4.
5.
BAUMAN, M. L., AND T. L. KEMPER. 1982. Morphologic and histoanatomic observations of the brain in untreated phenylketonuria. Acta Neuropathol. 58: 55–63. BRADFORD, M. M. 1977. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248–254. BRADY, S. T. 1992. Axonal dynamics regeneration. In Neural Regeneration (A. Gorio, Ed.) pp. 7–36. Raven Press, New York. COHEN, E., S. BINET, AND V. MEININGER. 1987. In situ appearance of the cold-stable microtubules in the growing axons of the tectal plate of mouse investigated immunocytochemically after polyethyleneglycol (PEG) embedding. Dev. Brain Res. 36: 171–180. DE FREITAS, M. S., A. G. DE MATTOS, M. M. CAMARGO, C. M. D. WANNMACHER, AND R. PESSOA-PUREUR. 1995. Effect of phenylalanine and a-methylphenylalanine on in vitro incorporation of 32P into cytoskeletal cerebral proteins. Neurochem. Int. 26: 381– 385.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18.
19.
20.
21.
22.
DE FREITAS, M. S., A. G. DE MATTOS, M. M. CAMARGO, C. M. D. WANNMACHER, AND R. PESSOA-PUREUR. 1995. Cytoskeletalassociated protein kinase and phosphatase activities from cerebral cortex of young rats. Neurochem. Res. 20: 951–956. DE MATTOS, A. G., M. S. DE FREITAS, M. M. CAMARGO, AND R. PESSOA-PUREUR. 1994. Developmentally regulated in vitro phosphorylation of cytoskeletal proteins of cerebral cortex of normal and malnourished rats. Dev. Neurosci. 16: 38–43. DE WAEGH, S. M., V. M.-Y LEE, AND S. T. BRADY. 1992. Local modulation of neurofilament phosphorylation, axonal caliber, and slow axonal transport by myelinating Schwann cells. Cell 68: 451–463. D´IAZ-NIDO, J., L. SERRANO, M. A. HERNA´ NDEZ, AND J. AVILA. 1990. Phosphorylation of microtubule proteins in rat brain at different developmental stages: Comparison with that found in neuronal cultures. J. Neurochem. 54: 211–222. GLASS, D. B., L. J. LUNDOVIST, B. M. KATZ, AND D. A. WALSY. 1989. Protein kinase inhibitor-(6-22)-amide peptide analogs with standard and nonstandard amino acids substitutions for phenylalanine 10. Inhibition of cAMP-dependent protein kinase. J. Biol. Chem. 264: 14579–14584. GREENGARD, O., M. S. YOSS, AND J. A. DELVALLE. 1976. Alphamethylphenylalanine, a new inducer of chronic hyperphenylalaninemia in suckling rats. Science 192: 1007–1008. HOFFMAN, P. N., AND R. J. LASEK. 1975. The slow component of axonal transport. Identification of major structural polypeptides of the axon and their generality among mammalian neurons. J. Cell Biol. 66: 351–366. HOFFMAN, P. N., J. W. GRIFFIN, AND D. L. PRICE. 1984. Control of axonal caliber by neurofilament transport. J. Cell Biol. 99: 705–714. HOLLENBECK, P. J. 1989. The transport and assembly of the axonal cytoskeleton. J. Cell Biol. 108: 223–227. HOMMES, F. A. 1991. On the mechanism of permanent brain dysfunction in hyperphenylalaninemia. Biochem. Med. Metab Biol. 46: 277–287. LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277: 680–685. LANE, J. D., B. SCHONE, U. LANGENBECK, AND V. NEUHOFF. 1980. Characterization of experimental phenylketonuria. Augmentation of hyperphenylalaninemia with a-methylphenylalanine and p-chlorophenylalanine. Biochim. Biophys. Acta 627: 144– 156. LANNING, C. L., K. R. WILMARTH, AND M. B. ABOU-DONIA. 1994. In vitro binding of [14C]2,5-hexanedione to rat neuronal cytoskeletal proteins. Neurochem. Res. 19: 1165–1173. LOWRY, O. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265–267. MAMIYA, N., J. R. GOLDERING, Y. TSUNODA, I. M. MODLIN, K. YASUI, N. USUDA, T. ISHIKAWA, A. NATSUME, AND H. HIDAKA. 1993. Inhibition of acid secretion in gastric parietal cells by the Ca21/calmodulin-dependent protein kinase II inhibitor KN-93. Biochem. Biophys. Res. Commun. 195: 608–615. MIYASAKA, H., S. OKABE, K. ISHIGURO, T. UCHIDA, AND N. HIROKAWA. 1993. Interaction of the tail domain of high molecular weight subunits of neurofilaments with the COOH-terminal region of tubulin and its regulation by tau protein kinase II. J. Biol. Chem. 268: 22695–22702. MORRIS, J. R., AND R. J. LASEK. 1982. Stable polymers of the axonal cytoskeleton: The axoplasmic ghost. J. Cell. Biol. 92: 192–198.
HYPERPHENYLALANINEMIA AND CYTOSKELETON 23.
NIXON, R. A., AND R. K. SIHAG. 1991. Neurofilament phosphorylation: a new look at regulation and function. Trends Neurosci. 14: 501–506. 24. NIXON, R. A., P. A. PASKEVICH, R. K. SIHAG, AND C. Y. THAYER. 1994. 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: 1031–1046. 25. RUBIN, M. A., C. M. D. WANNMACHER, G. B. VALENTE, M. M. CAMARGO, AND R. P. PUREUR. 1992. Diminished concentration of the NF-H subunit of neurofilaments in cerebral cortex of rats chronically treated with proline, methylmalonate and phenylalanine plus a-methylphenylalanine. J. Inher. Met. Dis. 15: 252– 260. 26. SHAH, S. N., N. A. PETERSON, AND C. M. MCKEAN. 1972. Lipid
195
composition of human cerebral white matter and myelin in phenylketonuria. J. Neurochem. 19: 2369–2376. 27. SIHAG, R. K., AND R. A. NIXON. 1989. In vivo phosphorylation of distinct domains of the 70-kilodalton neurofilament subunit involves different protein kinases. J. Biol. Chem. 264: 457–464. 28. TASHIRO, T., R. IMAI, AND Y. KOMIYA. 1994. Early effects of b,b8-iminodipropionitrile on tubulin solubility and neurofilament phosphorylation in the axon. J. Neurochem. 63: 291–300. 29. VALE, R. D., B. J. SCHMAPP, T. MITCHISON, E. STEVER, T. S. REESE, AND M. P. SHEETZ. 1985. Different axoplasmic proteins generate movement in opposite directions along microtubules in vitro. Cell 43: 623–632. 30. WANDOSELL, F., L. SERRANO, AND J. AVILA. 1987. Phosphorylation of a tubulin carboxyl terminal tyrosine prevents its incorporation into microtubules. J. Biol. Chem. 262: 8268–8273.