Incorporation of L-Tyrosine, L-Phenylalanine and L-3,4-Dihydroxyphenylalanine as Single Units into Rat Brain Tubulin

Transcription

1 Eur. I. Biochem. 59, (1975) Incorporation of L-Tyrosine, L-Phenylalanine and L-3,4-Dihydroxyphenylalanine as Single Units into Rat Brain Tubulin Carlos A. ARCE, Julio A. RODRIGUEZ, Hector S. BARRA. and Ranwel CAPUTTO Departainento de Quiinica Biologica, Facultad de Ciencias Quhicas, Universidad Nacional de Cordoba (Received June 3/July 22, 1975) The product of the incorporation of [14C]tyrosine as single unit into a protein of the soluble fraction of rat brain homogenate was purified by following a procedure used to purify tubulin. Sodium dodecylsulphate - polyacrylamide gel electrophoresis of the purified material showed a single protein band containing all the radioactivity. Purification data indicate that this protein accounts for 10.2 of the total protein of the supernatant fraction. This is in good agreement with the amount found for tubulin by the [3H]colchicine-binding method (10.5 % of the total protein). The incorporated ['"CItyrosine was found in the a-subunit of tubulin. Protein labelled with [3H]colchicine and [14C]tyrosine was precipitated with vinblastine sulphate and the radioactivity of 3H and that of 14C were quantitatively recovered in the precipitate (98 %). Sodium dodecylsulphate - polyacrylamide gel electrophoresis of the vinblastine precipitate showed that the I4C radioactivity moved with the tubulin band. Results obtained in experiments with phenylalanine and 3,4-dihydroxyphenylalanine were identical to those obtained for tyrosine. Binding of colchicine did not interfere with the incorporation of tyrosine. About 30% of tubulin from rat brain supernatant fraction can incorporate tyrosine as single unit. In previous reports [l- 31 we described the presence in the soluble fraction of rat brain homogenate of activities by which L-tyrosine, L-phenylalanine and ~-3,4-dihydroxyphenylalanine are incorporated as single units into the C-terminal position of an endogenous protein. Besides the protein and the involved amino acid the incorporating system requires ATP, Mg2+ and K+. Evidence that the three amino acids are incorporated into the same protein was provided by experiments that showed that the incorporation of any of them excluded the incorporation of the others in either simultaneous or successive incubations. An unexpected development in these studies was the finding [4] that the protein that incorporates these amino acids has properties similar to those commonly used to define tubulin, the major protein component of microtubules. Both proteins have the capability of being in widely different states of aggregation, the lowest of which is a dimer of molecular weight ; the dimer is formed by two subunits of molecular weight 54000; after labelling the soluble protein of the homogenate with [3H]colchicine and [14C]tyrosine the markers move identically in chromatographic experiments, and finally, the protein labelled with ['4C]tyrosine is precipitated with vinblastine sulphate as does the [3H]colchicine-tubulin complex. In the present paper besides providing further evidence that the acceptor protein is tubulin it will be shown that the tubulin subunit that contains the incorporated amino acid (tyrosine, phenylalanine or 3,4-dihydroxyphenylalanine) is the a-subunit. MATERIALS AND METHODS Chemicals ~-[U-"C]Tyrosine, ~-[U-'~C]phenylalanine, L- [3H]phenylalanine, ~-3,4-dihydroxy[U-'~C]phenylalanine and [meth~xy-~h]colchicine were purchased from New England Nuclear Corp. (Boston, Mass., U.S.A.). Sephadex G-25 (20-80 p) was supplied by Sigma Chemical Co. (St. Louis, Mo., U.S.A.). Vinblastine sulphate was a generous gift of Dr Robert Furman from Eli Lilly Company (U.S.A.). Preparation and Purification of the Labelled Protein Brains from day-old rats were homogenized in 1 vol of 10mM sodium phosphate buffer

2 146 Incorporation of Amino Acids as Single Units into Tubulin (ph 7.0) containing 5 mm MgCl,, 0.1 mm GTP and 0.24 M sucrose (phosphate-mg-gtp-sucrose buffer) and centrifuged at x g for 1 h. The supernatant fraction from this centrifugation contains the incorporating protein. The protein was labelled by incubating, in 1 ml of total volume, 0.9ml brain preparation, 2.5 pmol ATP, 30 pmol KCI, 12.5 pmol MgC12 and 0.1 pmol (1.8 pci) ['"Cltyrosine or 3.0 nmol (1.4 pci) ['"Clphenylalanine or 1.0 nmol (10 pci) [3H]phenylalanine or 4.4 nmol (1.7 pci) 3,4-dihydroxy['"C]phenylalanine. The ph was 7.0. In the case of the incorporation of 3,4-dihydroxy- ['"Clphenylalanine, 12 mm ascorbic acid was also added to the incubation mixture. The incubation was at 37 "C for 20 min. When ['"Clphenylalanine or [3H]phenylalanine or 3,4-dihydroxy['"C]phenylalanine were used to obtain the labelled protein, the soluble brain fraction was previously passed through a column (1 x 25 cm) of Sephadex G-25 equilibrated with phosphate-mg-gtp-sucrose buffer, to eliminate free amino acids. The labelled protein was purified by following the method of Weisenberg and Timasheff [5] for the purification of tubulin but omitting the MgCI, precipitation step. Protein was determined by the method of Lowry et al. [6] with crystalline bovine albumin as standard. Dry weight determination was used for the purified protein. The material, exhaustively dialyzed against water, was placed into a tared vial and dried at 110 "C in a drying pistol as described by Cardenas et al. [7]. The radioactivity of the labelled protein was measured in the material insoluble in hot trichloroacetic acid as previously described [8] but using 10 ml of a solution containing 2 g of 2,5-diphenyloxazole and 100 g of naphthalene per liter of dioxane as scintillation fluid. Preparation of Microtubular Protein hv Vinblastine Sulphate Precipitation 1 ml of the protein material was incubated with 2 mm vinblastine sulphate at 0 "C for 1 h, and centrifuged at x g for 30 min. The vinblastine precipitate and the supernatant solution were subjected to gel electrophoresis. Binding of [3H]Colchicine Tubulin concentration was determined according to Weisenberg et al. [9]. A 0.2-ml sample of the material to be assayed was incubated at 37 "C with 0.04 pmol (2 pci) [3H]colchicine during 1 h. After incubation the mixture was cooled and applied to a column (1 x 30 cm) of Sephadex G-25 equilibrated and eluted with phosphate-mg-gtp-sucrose buffer; fractions of 0.5 ml were collected and radioactivity counted. Simultaneous r3 HJColchicine Binding and ['"C] Tyrosine Incorporation The incubation mixture was identical to that described for ['"C]tyrosine incorporation but containing 0.05 M [3H]colchicine (specific activity 20 pci/ pmol). The final volume was 2 ml. After incubation at 37 "C for 1 h the solution was passed through a column (1 x 30 cm) of Sephadex G-25 to eliminate unbound [3H]colchicine and ['"Cltyrosine. Polyacrylamide Gel Electrophoresis Sodium dodecylsulphate - polyacrylamide gel electrophoresis (10 "/, acrylamide, 0.13 % bisacrylamide) was performed as described by Weber and Osborn [lo] except that gels were 0.8 x 8 cm in size. The protein samples were reduced by incubating at 90 "C for 3 min with 1 % 2-mercaptoethanol in the presence of 1 % sodium dodecylsulphate and 0.01 M sodium phosphate buffer (ph 7.0). Dodecylsulphate-urea gels (PH 8.7) were made according to Davis [ll] as described by Eipper [12], but with the inclusion of 0.1 %sodium dodecylsulphate and 8 M urea in all of the solutions used to prepare the gels. The protein samples were incubated during 3 min at 90 "C in 0.01 M Tris-HC1 (ph 7.3) containing 8 M urea, 0.5% sodium dodecylsulphate and 0.5% 2-mercaptoethanol. The running buffer was M Tris M glycine (ph 8.3) containing 0.1 %,sodium dodecylsulphate. Electrophoresis was carried out at room temperature at 3 ma per gel during 5 h. Bromophenol blue was used as the tracker dye. Gels were stained with 0.25 % Coomassie brilliant blue in methanol/acetic acid/water (45/9/46, by vol.) and destained with the same solution but omitting the dye. Scanning of the gels was carried out at 590 nm in a Canalco (model G) spectrophotometer equipped with linear transport. Radioactivity from '"C-labelled protein was determined in gel slices (1 mm thick) as described before [4]. For 3H-labelled protein the gel slices were oxidized in a Carbon Hydrogen Nitrogen analyzer (model 185, Hewlett Packard) and the 'H,O was collected into vials containing 20 ml of the scintillation fluid. RESULTS Determination of Tubulin and the Amino-Acid-Acceptor Protein in Brain Homogenate Since it is generally accepted that about 95 % pure tubulin can be obtained by the method of Weisenberg and Timasheff [5], we decided to apply it for the purification of proteinyl-['"cltyrosine in an attempt to test quantitatively the hypothesis on the identity

3 C. A. Arce, J. A. Rodriguez, H. S. Barra, and R. Caputto 147 Table 1. Purficalion ~fprorein~l-['~c]iyrosine The purification of the labelled protein was carried out by following the method of Weisenberg and Timasheff [5] for tubulin purification. The starting material was 30 g of rat brain. For details see Materials and Methods Fraction Volume lo6 x Total Total [14C]Tyrosine Yield Purification radioactivity protein incorporated ml counts/min mg pmol/mg protein % -fold Soluble supernatant (NH4)2S04 (32-43%) DEAE-Sephadex eluate Fig. 1. Dodecylsulphare-pol~acr)~lamidegel electrophoresis qfproieinj/-[ "CJrjrosine. The conditions for electrophoresis and the radioactivity measurement were as described in Materials and Methods. (A) Purified pr~teinyl-['~c]tyrosine (130 pg) (see Table 1, DEAE-Sephadex eluate). (B) I, protein and radioactivity of the vinblastine sulphate precipitate (100 pg) from a soluble brain preparation after [14C]tyrosine incorporation ; 11, protein of the supernatant solution (100 pg) after vinbkdstine sulphate precipitation of both proteins. The supernatant fraction of rat brain homogenate was obtained as described in Materials and Methods. The content of tubulin was determined in a sample of the supernatant fraction by the ['H]colchicine-binding method. Results showed that 960 pmol (average of three determinations that ranged from 840 to 1060 pmol) of tubulin was present per mg of protein, that is to say, that 10.5% of the protein in the soluble fraction was tubulin (molecular weight ). Another sample of the same supernatant fraction was used to obtain pr~teinyl-[~~c]- tyrosine which was then purified by using the method of Weisenberg and Timasheff [5] for the purification of tubulin. The obtained protein (DEAE-Sephadex eluate, Table 1) was subjected to dodecylsulphate - polyacrylamide gel electrophoresis ; by scanning at 590 nm more than 95 % of the protein was found in a single band; in this band it was also found practically all the radioactivity (Fig. 1A). At this step of the purification pr~teinyl-['~c]tyrosine had a specific radioactivity 9.8-fold higher than in the original soluble fraction. This result showed that there is good agreement between the value found for the concentration of tubulin by the method of [3H]colchicine binding (10.5%) and that calculated from the purification of pr~teinyl-['~c]tyrosine (10.2 %). Protein labelled with ['4C]phenylalanine was purified by the same method applied to pr~teinyl-['~c]tyrosine; it was found that it also migrated identically to tubulin in dodecylsulphate - polyacrylamide gel electrophoresis. Complete agreement was also found after precipitation with vinblastine sulphate of the protein labelled simultaneously with ['H]colchicine and [14C]tyrosine. The 'H and 14C radioactivity in the supernatant solution was measured before and after vinblastine sulphate precipitation. Results showed that tubulin precipitated by vinblastine sulphate carried 98 % of 3H and 98% of I4C radioactivity bound to protein. When the vinblavine precipitate was subjected to dodecylsulphate - polyacrylamide gel electrophoresis more than 95 % of the 14C radioactivity migrated with the band of tubulin (Fig. 1B). Protein labelled with

4 148 Incorporation of Amino Acids as Single Units into Tubulin [3H]phenylalanine or 3,4-dihydro~y['~C]phenylaIanine but omitting colchicine were also precipitated with vinblastine sulphate. Again, the labelled protein found in the vinblastine precipitate migrated as tubulin in dodecylsulphate - polyacrylamide gel electrophoresis. Effect of Colchicine on Tyrosine Incorporation The influence of colchicine on maximal incorporation of tyrosine was studied by measuring the incorporation of ['4C]tyrosine in preparations preincubated with and without [3H]colchicine. Results showed that the amount of ['4C]tyrosine incorporated was similar in both cases (300 and 281 pmol/mg protein, respectively; data are average of three determinations that ranged from 270 to 318 pmol/mg protein for the former and 265 to 305 pmol/mg protein for the later case). The incorporation of ['4C]tyrosine after colchicine binding was carried out without elimination of free [3H]colchicine. The incubation mixture was as given in Materials and Methods, except for the concentration of tyrosine (0.05 mm; specific activity 6 pci/pmol) and for the brain preparation which was previously passed through a column (1 x 25 cm) of Sephadex G-25 equilibrated with phosphate-mg-gtp-sucrose buffer ; the volume of the incubation mixture was 0.2 ml. Similarly, the incorporation of tyrosine had no effect on the [3H]colchicine binding (950 and 948 pmol/mg protein in preparations preincubated with and without tyrosine, respectively). It was previously shown [3] that the activity of the incorporating system decreases considerably when the brain preparation is kept at 37 'C in sucrose/2-mercaptoethanol/tris/hcl buffer (ph 7.4). In this same buffer the amount of tyrosine that can be incorporated decreases with the time in a minor degree when colchicine is present. Brain preparations preincubated for 1 h without colchicine incorporated about % of the original amount, whereas preincubated with colchicine the incorporation was :d of the original. Identification of' the Labelled Tubulin Subunit The a and p subunits of tubulin labelled with [14C]tyrosine or ['4C]phenylalanine were separated by subjecting the purified preparation (DEAE-Sephadex eluate, Table 1) to dodecylsulphate- urea- polyacrylamide gel electrophoresis. The gels were stained, cut in slices (1-mm thick) and counted. In both cases practically all of the radioactivity was found in the a-subunit. Tubulin preparation labelled with 3,4-dihydr~xy['~C]phenylalanine and purified by precipitation with vinblastine sulphate was also subjected to electrophoresis. In this case too the radioactivity coincided with the a-subunit of tubulin. Total versus Amino-Acid-Acceptor Tuhulin Bound tyrosine can be exchanged with free tyrosine in a crude supernatant fraction [13]. Tyrosine incorporation can be obtained in conditions in which the exchange reaction practically does not occur by using a brain preparation that has been previously passed through a column of Sephadex G-25. In these conditions approximately 300 pmol of tyrosine per mg of protein was incorporated. Comparison of this incorporation with the amount of [3H]colchicine bound to protein (950 pmol/mg protein) indicates that about 30 % of tubulin in rat brain is in such conditions that can accept tyrosine (or phenylalanine or 3,4-dihydroxyphenylalanine) as single unit. DISCUSSION The hypothesis on the identity of tubulin with the acceptor protein of tyrosine or phenylalanine or 3,4-dihydroxyphenylalanine [4] receives strong support from the quantitative analysis and electrophoretic studies on polyacrylamide gels presented in this paper. By using the [3H]colchicine-binding method it was found that about 10% of the soluble protein in rat brain cytosol is tubulin. According to this result, the labelled protein should be purified 10-fold when pure tubulin is obtained. This was what occurred: by following a method of tubulin preparation [5], the labelled protein was purified 9.8-fold (Table 1). The possibility that the labelled protein was present in the starting material in low concentration and coincidentally purified 9.8-fold is discarded because from data in Table 1 and on the basis that one mol of tyrosine was incorporated per mol (molecular weight ) of acceptor protein [4,14], it can be calculated that pr~teinyl-['~c]tyrosine formed in vitro accounted for 33 % of the purified material. Results obtained by following a completely different method of purification of tubulin confirmed the conclusion about its identity with the tyrosine-labelled protein. After treating with vinblastine sulphate the soluble protein labelled with [3H]colchicine and ['4C]- tyrosine the produced precipitate contained 98 % of both the 3H and 14C originally bound to the protein. Dodecylsulphate - polyacrylamide gel electrophoresis showed that the radioactivity of prot~inyl-['~c]- tyrosine which had been purified up to the step of DEAE-Sephadex eluate by the method of Weisenberg and Timasheff [5] coincided with the band of tubulin (Fig. 1A). The same agreement was obtained when tubulin was purified by precipitation with vinblastine sulphate (Fig. 1B). When the a and j? subunits of tubulin were separated on dodecylsulphate - urea gel electrophoresis the radioactivity migrated with the cc-subunit. Similar results were obtained with proteinyl-

5 C. A. Arce. J. A. Rodriguez. H. S. Barra, and R. Caputto 149 [14C]phenylalanine and proteiny1-3,4-dihydro~y['~c]- phenylalanine. It was previously reported that tyrosine, phenylalanine and 3,4-dihydroxyphenylalanine are attached to the C-terminal position of the protein [2,3], and the binding amino acid was identified as glutamic acid or glutamine [14]. Bound tyrosine can be exchanged with free tyrosine [13], but in conditions in which this exchange practically does not occur it was found that about 3076 of the tubulin is the acceptor protein. According to these results and assuming that the amino-acid-acceptor protein and the product are in the same subunit we expect that the a-subunit of tubulin from the soluble fraction of rat brain homogenate is formed by a heterogeneous population in which about 30% of the C-terminal amino acid is glutamic acid or glutamine and the remaining 70% contains tyrosine, phenylalanine or dopa as C-terminal amino acid. From the K,,, values and the concentration of these amino acids in normal rat brain [4] we expect tyrosine to be the quantitatively main amino acid in the C-terminal position of the a-subunit of tubulin. This work was supported in part by Grants from thc Conseju Nacional cle 1nve.srigacionr.s Cienr;/icos Tkcnicas, Argentina and the Academia Nucional de Cicwcias. Cordoba, Argentina. C. A. Arce was supported h) a fellowship from the Consejo Nacional de 1nwsrigacione.s Cicni(\rcus y Ti;cnicas, Argentina. REFERENCES 1. Barra, H. S., Rodi-igucz, J. A,. Arce, C. A. & Caputto, R. (1973) J. Neurochem. 20, Barra, H. S.. Arce, C. A., Rodriguez, J. A. & Caputto, R. (1973) J. Neurochem. 21, Rodi-iguez. J. A,. Barra. H. S., Arce, C. A,. Hallak. M. E. & Caputto, R. (1975) Biochem. J. 149, Barra, H. S., Arce, C. A., Rodriguez, J. A. & Caputto, R. (1974) Biochrm. Biophys. Res. Commun. 60, Weisenberg, R. C. & Timasheff, S. N. (1970) Biochemistrj., 9, Lowry, 0. H., Rosebrough, N. J., Farr. A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, Cardenas. J. M., Dyson, R. D. & Strandholm, J. J. (1973) J. Biol. Chem. 248, Barra, H. S., Ufiates, L. E., Sayavedra. M. S. & Caputto, R. (1972) J. Neurochem. IY, Weisenberg, R. C., Borisy, G. G. & Taylor, E. W. (1968) Biochemistrj,, 7, Weber, K.&Osborn, M.(1969) J. Biol. Chem. 244, Davis, B. J. (1964) Ann. N. Y. Acad. Sci Eippcr. B. A. (1974) J. Biol. Chem. 249, Rodriguez, J. A,, Arce, C. A,. Barra, H. S. & Caputto, R. (1973) Biochem. Biophys. Re.\. Commun. 54, Arce, C. A., Barra, H. S., Rodriguez. J. A. & Caputto, R. (1975) FEBS Leu. 50, 5-7. C. A. Arce, J. A. RO~I-~~LICZ, H. S. Barra, and R. Caputto, Departamento de Quimica Biologica, Facultad de Ciencias Quhcas, Ciuddd Universitaria, Cordoba, Argentina