Dephosphorylation of tyrosine hydroxylase by brain protein phosphatases: a predominant role for type 2A

BRAIN RESEARCH ELSEVIER

Brain Research 637 (1994) 273-276

Research Report

Dephosphorylation of tyrosine hydroxylase by brain protein phosphatases: a predominant role for type 2A Ulrike Berresheim, Donald M. Kuhn * Cellular and Chntcal Neurobtology Program Department of Psychtatry Wayne State Unwerstty School of Medtcine Detroit, Mtchtgan, USA (Accepted 5 October 1993)

Abstract

Extracts from rat corpus striatum, or striatal proteins resolved by chromatography on DE-52, were tested for protein phosphatase activity using tyrosine hydroxylase, phosphorylated by cAMP-dependent protein kinase, as substrate. The predominant dephosphorylating activity was independent of divalent cations and was inhibited by low concentrations (100 nM) of okadaic acid, defining the phosphatase as type 2A. Phosphatase type 2C (Mg 2+ and Mn 2÷ stimulated) was evident in the presence of okadaic acid but at a level of approximately 10% of type 2A activity. Phosphatase 2B (Ca2+ and calmodulin dependent) mediated dephosphorylation of tyrosine hydroxylase was not apparent. The dephosphorylation of [32p]-tyrosine hydroxylase was not modulated by tetrahydrobiopterin, ATP, or GTP. These results indicate that tyrosine hydroxylase which has been phosphorylated by cAMP dependent protein kinase is dephosphorylated predominately by phosphatase type 2A in brain, and the activity of this phosphatase is not modulated by pteridines or nucleotides. Key words: Tyrosine hydroxylase; Protein phosphatase 2A; Protein kinase A 1. Introduction

Tyrosine hydroxylase (TH; EC 1.14.16.2) is the initial and rate limiting enzyme in the biosynthesis of the catecholamines in brain and adrenal medulla. TH is a substrate for a number of protein kinases and the phosphorylation of TH imparts regulatory control over its activity in neuronal and non-neuronal tissue alike [15]. Certain neurotransmitter receptors and second messengers are known to contribute to the regulation of TH through their influence on protein kinases and, in many cases, their influence on TH activity can be traced to phosphorylation of specific sites in TH [13,15]. Very little is known about the phosphatases which act on phospho-TH and the available information is somewhat conflicting. For example, Yamauchi and Fu-

* Corresponding author. Department of Psychiatry Wayne State Umversity School of Medicine Metropolitan Center for High Technology 2727 Second Avenue, Room 319 Detroit, MI 48201, USA. Fax: (1) (313) 963-4021. Abbreviations used: TH, Tyrosine hydroxylase; cAMP-PK, cAMPdependent protein kinase; EGTA, [ethylenebls (oxyethylenenltrilo)] tetraacetlc acid 0006-8993/94/$07.00 © 1994 Elsevier Science B.V All rights reserved SSD! 0006-8993(93)E1401-N

jisawa [14] demonstrated that bovine adrenal phosphoTH was dephosphorylated by a magnesium dependent process, suggesting the action of protein phosphatase 2C. On the other hand, it has been suggested that bovine adrenal [7,8] and striatal TH [8], and PC-12 TH [9] are dephosphorylated primarily by phosphatase type 2A. Nelson and Kaufman [11] described a heretofore unknown TH phosphatase in brain. These investigators demonstrated that a striatal phosphatase which dephosphorylated TH was dramatically stimulated by tetrahydrobiopterin (BH4) , the natural pteridine cofactor for TH. This phosphatase was also potently inhibited by GTP, the metabolic precursor of BH 4 in brain [11]. In view of this very novel regulatory mechanism for a phosphatase suggested by Nelson and Kaufman [11] and the importance it could have on TH function, the present experiments were designed to further characterize brain phosphatases which dephosphorylate [32p]TH, with emphasis on defining a role for BH 4. These initial studies concentrated on TH which was phosphorylated by cAMP dependent protein kinase (cAMP-PK) since this protein kinase is known to phosphorylate a single serine residue (serine 40) in TH [5,7-9].

274

U Berreshetm. D M. Kuhn ~Brain Research 637 (1994)273-276

2. Materials and methods 2 1 Materials The following chemicals and reagents (with suppher) were used in these studies' 32P-labelled A T P (3000 C l / m m o l , DuPont New England Nuclear); tetrahydroblopterm (Dr B Shirks, Jona, Switzerland), Sephadex G-25 (Pharmacia); DE-52 (Whatman); calmoduhn (Sigma Chemical C o , St. Louis, MO), and okadalc acid (Waco Chemicals USA, Inc, Richmond, VA) All other chemicals and reagents were of the highest purity commercially available. The catalytic subumt of c A M P - P K was purified from bovine heart as described by Beavo et al. [2] 2 2 Purification and phosphorylatton o f T H T H was purified to homogeneity from cultured PC12 cells as prewously described [10] Since rat T H is a single gene product [4], the PC12 enzyme ~s a statable substitute for the brain form in these studies. For phosphorylatlon, approximately 80/~g of T H was incubated with 2.0 /~g of the catalytic subumt of c A M P - P K in the presence of 0.01 m M A T P containing 150 p, Cl of [32p]ATP and 10 m M Mg 2+ in 10 m M Trls buffer pH 7 4. Reactions were carried out for 15 mln at 30°C in a reaction volume of 1 2 ml Phosphorylation reactions were terminated by cooling to 4°C, and samples were chromatographed on a 1.5 × 8 cm column of Sephadex G-25 equilibrated with 10 m M Trls buffer p H 7 4 containing 10% (v/v) glycerol, 0.1 m M E D T A , and 0.01% ( w / v ) sodium azlde, to remove unreacted [32p]ATP. De-salted phospho-TH (appearing in the column void volume) was used immediately for studies of phosphatase action. These phosphorylatlon conditions typically resulted in the incorporation of 0 4 - 0 6 moles of phosphate per 60,000 Da s u b u m t of TH, Protein was determined by the method of Bradford [3] using bovine serum albumin as a standard 2 3 Partial purification o f rat stnatal phosphatases Phosphatases from rat strlatum were resolved by chromatography on DE-52 as described by Nelson and K a u f m a n [11] Striatal tissue from 5 rats was homogenized fresh in 9 volumes of 0.25 M sucrose in a buffer containing 50 m M Tns-HC1, pH 7 0, 0.1 m M E G T A , 1 m g / l i t e r each of leupeptln, pepstatln A, and aprotinm, and 0 2 m M PMSF Homogenates were centrifuged at 100,000 × g for 20 m m and supernatants were ~mmediately used m phosphatase assays or applied to a DE-52 column. The column was eluted with a linear gradient of 0 - 1 M NaCI m 200 ml of equilibrating buffer, and fractions were used Immediately upon elution and were stored at 4°C for no more than 2 days as r e c o m m e n d e d by Nelson and K a u f m a n [11] 2 4 Phosphatase assays Phosphatase activity in strlatal extracts or column eluates was determined by the method described by Nelson and K a u f m a n [11] Reaction mtxtures contained 1 m M dithiothreltol, 0.1 M Trls-HCl, pH 80, 5 m M Mg2+CI2, 1 m M Mn2+C12, 20,000-30,000 cpm of [32p]TH, and 0.02-0.03 mg of phosphatase enzyme in a final volume of 0.3 ml The conditions for assessing the ability of a specific phosphatase to depbosphorylate T H were defined as follows [6]: phosphatase 2,4 - active in the absence of calcium (1.0 m M EGTA), Mg 2+ and Mn 2+ (1 0 m M E D T A ) and inhibited by okadaic acid (100 nM), phosphatase 2B - stimulated by Ca 2+ ( 2 0 mM) and calmoduhn (2.5/zg) in the presence of 1 0 m M E D T A and 100 nM okadaic acid, and inhibited by 10 m M E G T A or 0.1 m M calmidazohum, and, phosphatase 2C - stimulated by 5 m M Mg 2+ and 1 m M

Mn 2+ m the presence of 1 0 m M E G T A dnd 100 nM okadaic acid, and inhibited by 10 m M E D T A Phosphatase reactions were terminated with the addition of 0 1 ml of 5% a m m o n i u m molybdate m 4 N H2SO 4 and 0 1 ml of 10 m M sfl~cotungstlc acid, 0,5 m M K H 2 P O 4, and 10 m M H2SO 4 The ~Zp inorganic phosphate was extracted b) adding 0 6 ml of b e n z e n e / l - b u t a n o l (1-1), followed by vortexmg and centrffugatlon. An ahquot of (I 3 ml of the organic phase was removed and added to scintillation fluid and counted m a ~cmtdlat~on counter

3. Results

Rat striatal proteins were resolved by ion exchange chromatography on DE-52 and phosphatase activity was assayed using [32p]TH as the substrate. Phosphatase type 2A activity was eluted in a broad peak along with the eluting peak of protein. The shape of the eluted phosphatase activity peak was not altered if assays were conducted under conditions promoting either phosphatase type 2B or 2C activity (data not shown). Therefore, attempts to resolve phosphatase activities further were not made and peak fractions were pooled and used as the source of phosphatase enzyme in the remaining experiments. Fig. 1 presents the results of experiments where phosphatase type 2A, 2B, and 2C activities were selectively invoked as described in Experimental Procedures. The results demonstrate that almost all dephosphorylation of [32p]TH was catalyzed by type 2A phosphatase (i.e., independent of divalent cations and inhibited by 100 nM okadaic acid). Marginal type 2C activity (stimulated by Mg 2+ and Mn 2+ in the presence

~. o x

10

• Type 2A • Type 2B

8

• Type 2C

o

~" 2 n 0

0

i

i

J

i

15

30

45

60

Time of Incubation [ram)

Fig 1 Dephosphorylation of [32pITH by phosphatase types 2A, 2B, or 2C [3ZP]TH was incubated with parUa]ly purified phosphatase fractions from DE-52 chromatography under conditions which selectively invoked type 2A, 2B, or 2C phosphatase activities (see Materials and methods) Phosphatase reactions were incubated at 30°C for the indicated times and the lnorgamc 32p released from T H was extracted and quantified. Okadalc acid was added to incubations at a concentration of 100 nM. Results are the m e a n s of three separate experiments run in duplicate

U. Berresham, D.M. Kuhn ~Brain Research 637 (1994) 273-276 120I

.~

-

.

60 I

.

~.

~

H

i

i

L

• BH4 • dopam,ne

I

B. • ATP

120

• GTP

~

100

,

-

~

~

13_ 80 II

0

I

10 .7

I

I

I

10 -6 105 104 Concentrabon (M)

I

10-3

Fig. 2. Effects of BH 4 and dopamme (A), and GTP and ATP (B) on

dephosphorylatton of [32PITH by phosphatase 2A Partmlly purified phosphatase fractmns from DE-52 chromatographywere used as the source of protein phosphatase. All test compounds were added to phosphatase assays m the indicated concentrations and the dephosphorylatmn of TH was measured as described in Materials and methods The results are presented as % control (all components of the phosphatase 2A assay without the test compounds) and are the means 5:S E M. of 5-6 experiments run in duphcate.

of okadaic acid, and inhibited by E D T A ) could be observed at the longer incubation times. The levels of type 2C activity were approximately 10-12% of type 2A activity. T H was not dephosphorylated under conditions promoting type 2B phosphatase activity (Ca 2+ and calmodulin stimulated). The results were not changed if rat striatal extracts were used prior to chromatography (data not shown). The effects of BH 4 on the dephosphorylation of [32p]TH were investigated and the results are presented in Fig. 2A. These and all subsequent experiments were carried out under conditions favoring phosphatase type 2A since it was the predominant phosphatase. It can be seen in that concentrations of BH 4 from 10 -7 to 10-4M had little effect on the dephosphorylation of [32p]TH. At the highest concentration, BH 4 inhibited the dephosphorylation of T H by 20%. Since catecholamines suppress T H activity through a competitive interaction with BH 4 [1], the effect of dopamine on T H dephosphorylation was also studied. The data in Fig. 2A indicate that phosphatase activity was unaltered by dopamine over a concentration range which would significantly alter T H catalytic activity [1]. ATP and G T P were tested for their effects on the dephosphorylation of [32p]TH and the results are shown in Fig. 2B. It can be seen that neither of these nucleotides altered the dephosphorylation of T H over a broad concentration range.

4. Discussion

The results of the present experiments demonstrate that the predominant phosphatase in brain which de-

275

phophorylates TH, previously phosphorylated by cAMP-PK, is type 2A. This phosphatase is not dependent on divalent cations for activity and can be inhibited by low concentrations of okadaic acid. T H is also dephosphorylated to a minor extent in the presence of Mg z+ and sufficient okadaic acid to suppress 2A activity, suggesting the actions of phosphatase type 2C. Type 2B (Ca 2+ and calmodulin stimulated) activity was not apparent using T H as the phosphatase substrate. The observations that low concentrations of okadaic acid (100 nM) completely inhibited phosphatase activity rules out a role for type 1 phosphatase in dephosphorylating T H since much higher concentrations of okadaic acid are required to inhibit this phosphatase [8]. Taken together, the present results agree well with previous studies which demonstrated that the predominant phosphatase acting on phospho-TH in the bovine adrenal [7,8] and striatum [8] or in PC-12 cells [9] is type 2A. Nelson and Kaufman [11] presented evidence of a novel regulatory mechanism for a brain phosphatase which dephosphorylates TH. These investigators measured T H phosphatase activity in the presence of Mg 2+, Mn 2+, and E G T A which would favor the activity of phosphatase types 2A and 2C. Under these assay conditions, Nelson and Kaufman [11] demonstrated that BH 4 stimulated the dephosphorylation of [32p]TH by a factor of 8-fold. GTP, the metabolic precursor of BH 4 in brain, and to a lesser extent ATP, could substantially inhibit T H dephosphorylation. The magnitude of the effects of BH 4 and GTP on T H dephosphorylation were such that these agents would have profound influence over T H activity through their interactions with the phosphorylation/dephosphorylation cascade. Indeed, Nelson and Kaufman [11] concluded that the effect of BH 4 on T H dephosphorylation was maximal at a concentration 5 - 1 0 / ~ M , which is near the in vivo concentration of BH 4 in brain. Thus, under normal conditions, BH 4 could act to maintain T H in the dephosphorylated state. Based on the magnitude of the influence of BH 4 on T H dephosphorylation in the Nelson and Kaufman [11] study, and in view of the fact that BH 4 and G T P could also influence phosphatase action on substrates in addition to TH, it was important to investigate further the effects of BH4, ATP, and G T P on the dephosphorylation of TH. The present results with these agents stand in contrast to those of Nelson and Kaufman [11]. When BH 4 was added to dephosphorylation reactions over a wide concentration range, it did not alter phosphatase activity. Dopamine was also tested for its effects since this catechol interacts with T H near the BH 4 binding site and substantially reduces T H catalytic activity [1]. Like BH4, dopamine was without effects on T H dephosphorylation over a concentration range which would significantly alter T H activity.

276

U Berreshetm, D M Kuhn / Bram Research 637 (1994)273-276

GTP and ATP were also found to be without effect on the dephosphorylation of phospho-TH by brain phosphatase 2A. There was no evidence of a suppression of TH dephosphorylation by either substance. Thus, despite following the assay conditions of Nelson and Kaufman [12] as closely as possible, we were unable to alter the dephosphorylation of TH by any of these substances related to the reduced pteridine cofactor for TH. The stimulation of phosphatase activity toward phospho-TH by BH 4 and other reduced pteridines [11] does not easily fit into any of the well established classification schema for phosphatases [6]. The present results confirm that the predominant TH dephosphorylating activity in brain is protein phosphatase 2A in agreement with the results of Haavik et al. [8]. A specific phosphatase for TH or a link between the pteridine cofactors and phosphatases does not appear to exist. Further studies are underway to determine if phosphorylation of TH by calcium/ phospholipid-dependent or calcium/calmodulin dependent protein kinase II will determine which brain phosphatases dephosphorylate TH. Acknowledgements This research was supported in part by grants from the Umted Parkmson Foundation (D.M.K) and the M~chlgan Parkmson Foundation (U B.)

5. References [1] Almas, B, Le Bourdelles, B, Flatmark, T., Mallet, J and Haavik, J., Regulation of recombinant human tyrosme hydroxylase ~sozymes by catecholamine binding and phosphorylatlon Structure actwlty studies and mechanistic implications, Eur J Btochem, 209 (1992) 249-255 [2] Beavo, J A., Bechtel, P.J. and K.rebs, E.G., Preparation of homogenous eychc AMP dependent protein klnase(s) and its subumts from rabbit skeletal muscle, Methods Enzymol, 38 (1974) 299-308

[3] Bradford, M.M, A rapid and sensitive method for the quantltatlon of microgram quantlt~es of protein utd~zmg the principle ol protein-dye binding, Anal Btochem, 72 (1976) 248-254 [4] Brown, E R , Coker, G T and O'Malley, K.L, Orgamzatlon and evolution of the rat tyrosme hydroxylase gene, B:ochemt~trv, 26 (1987) 5208-5212 [5] Campbell, D G , Hardle, D.G and Vulhet, P R,, Identification of four phosphorylat~on s~tes m the N-terminal region of tyrosine hydroxylase, J Btol Chem, 261 (1986) 10489-10492 [6] Cohen, P, The structure and regulation of protein phosphatases, Annu Rel' Btochem, 58 (1989) 453-508 [7] George, R.J, Haycock, J W , Johnston, J P, Cravlso, G L and Waymlre, J C, In vitro phosphorylatlon of bovine adrenal chromaffm cell tyrosme hydroxylase by endogenous protein kmases, J Neurochem, 52 (1989) 274-284. [8] Haavlk, J., Schelhng, D L, Campbell, D G., Andersson, K.K, Flatmark, T and Cohen, P, Identification of protein phosphatase 2A as the major tyrosme hydroxylase phosphatase m adrenal medulla and corpus strmtum evidence from the effects of okadalc acld, FEBS Lett, 251 (1989) 36-42 [9l Haycock, J C., Phosphorylatlon of tyrosme hydroxylase m s~tu at serine 8, 19, 31, and 40, J Btol Chem, 265 (1990) 11682-11691 [10] Kuhn, D M and Bdhngsley, M.L, Tyroslne hydroxylase. Purification from PC12 cells, characterization, and production of antibodies, Neurochem Int, 11 (1987)463-475 [11] Nelson, T J and Kaufman, S, Activation of rat caudate tyroslne hydroxylase phosphatase by tetrahydroblopterm, J Blol Chem, 262 (1987) 16470-16475 [12] O'Farrell, P.H, High resolution two dimensional electrophoresis of proteins, J Btol Chem, 250 (1975) 4007-4021. [13] Waymlre, J.C, Johnston, J.P., Hummer-Llcktexg, K, Lloyd, A., Vlgny, A and Crawso, G L, Phosphorylatlon of bovine adrenal chromaffm cell tyrosine hydroxylase Temporal correlation of acetylcholine's effect on szte phosphorylatlon, enzyme actwat~on and catecholamme synthesis, J Btol Chem, 263 (1988) 1243912447 [14] Yamauchl, T and Fujmawa, H , Regulation of bowne adrenal tyrosme 3-monooxygenase by phosphorylatlon-dephosphorylatlon reaction, catalyzed by adenosme 3' 5'-monophosphate-dependent protein kmase and phosphoprotem phosphatase, J Btol Chem., 254 (1988) 6408-6413 [15] Zlgmond, R.E., Schwarzschdd, M.A and Rittenhouse, A . R , Acute regulation of tyroslne hydroxylase by nerve actwlty and by neurotransmltters wa phosphorylatlon, Annu Rec Neurosct, 12 (1989) 415-461