participation of a stress-activated protein kinase cascade in the activation of tyrosine hydroxylase...

Eur. J. Biochem. 247, 1180-1189 (1997) 0 FEBS 1997

Participation of a stress-activated protein kinase cascade in the activation of tyrosine hydroxylase in chromaffin cells Gareth THOMAS', Jan HAAVIK' and Philip COHEN'

' MRC Protein Phosphorylation Unit, Department of Biochemistry, University of Dundee, Scotland Department of Biochemistry and Molecular Biology, University of Bergen, Norway

(Received 18 FebruaryR May 1997) ~ EJB 97 0258/1

Sodium arsenite and osmotic shock both stimulated stress-activated protein kinase-2 (SAPK2, also termed RK, p38, CSBP and Mxi2) and its downstream target mitogen-activated protein kinase (MAP kinase)-activated protein kinase-2 (MAPKAP-K2) in bovine adrenal chromaffin and rat PC12 cells. The same stimuli also increased tyrosine hydroxylase activity 2- 3-fold and induced its phosphorylation at Serl9, a residue phosphorylated by MAPKAP-K2 in vitro. The arsenite-induced activation of tyrosine hydroxylase and its phosphorylation at Serl9 were prevented by SB 203580 at concentrations similar to those that inhibited SAPK2 in vitro. These results indicate that MAPKAP-K2 mediates the stress-induced activation of tyrosine hydroxylase. SB 203580 had no effect on the phosphorylation or activation of tyrosine hydroxylase induced by nerve growth factor or forskolin, which trigger the phosphorylation of Ser3 1 and Ser40, respectively. Stimulation of bovine adrenal chromaffin cells with acetylcholine activated SAPK2 and MAPKAP-K2, as well as p42/p44 MAP kinases and their downstream target MAPKAP-K1 . The half-times for activation of MAPKAP-K1 and MAPKAP-K2 (1 min) were similar. In contrast, the activation of tyrosine hydroxylase by acetylcholine peaked within 1 min and gradually declined thereafter. Neither SB 203580 (which blocked the activation of MAPKAP-K2 by acetylcholine) nor PD 98059 (which prevented the activation of p42/p44 MAP kinases by acetylcholine) affected tyrosine hydroxylase activation after 1 min, but these compounds inhibited activation by 40-50% after 5 min. PD 98059 prevented the acetylcholine-induced phosphorylation of tyrosine hydroxylase at Ser31, the residue targetted by p42Ip44 MAP kinases in vitro, but did not inhibit the phosphorylation of Ser40 (which is phosphorylated by MAPKAP-K1 in vitro). Our results establish that p42/p44 MAP kinases mediate the acetylcholine-induced phosphorylation of tyrosine hydroxylase at Ser3 1. SB 203580 did not suppress the phosphorylation of Serl9 or Ser40 by acetylcholine but, like PD 98059, this drug decreased the phosphorylation of Ser3 1. SAPK2 may therefore contribute to the acetylcholine-induced activation of tyrosine hydroxylase by facilitating (in an unknown way) its phosphorylation by MAP kinases.

Keywords: tyrosine hydroxylase ; stress ; chromaffin cell ; PC12 cell ; mitogen-activated protein kinase.

Tyrosine hydroxylase, the rate-limiting enzyme in catecholamine synthesis, is found predominantly in the adrenal medulla and central and sympathetic nervous systems. In these cells tyrosine hydroxylase activity is increased by agonists that stimulate catcholamine secretion, providing a mechanism for replenishing the stores of catecholamines that have been lost via secretion. The acute activation of tyrosine hydroxylase is mediated by phosphorylation, and four serine residues (Ser8, Serl9, Ser3 1 and Ser40) are phosphorylated in vivo [I-41. The protein kinases known to phosphorylate these sites in vitro are shown in Fig. 1. The phosphorylation of Ser31 or Ser40 activates tyrosine

Correspondence to G. Thomas, MRC Protein Phosphorylation Unit, Department of Biochemistry, University of Dundee, Dundee DD1 4HN, UK

Fux: +44 1382 223778. E-mud: [email protected] URL: http://www.dundee.ac.uk/biochemistry/ Abbreviations. CaM-KII, calciudcalmodulin-dependent protein ki-

nase-ll; MAP kinase, mitogen-activated protein kinase; MAPKAP-K, MAP-kinase-activated protein kinase; 6Me-H4Pte, 6R,S-6-methyl- 5,6,7,8-tetrahydropterin; NGF, nerve growth factor; PKA, CAMP-dependent protein kinase; PKC, protein kinase-C; PC12 cells, phaeochromocytoma 12 cells; SAPK, stress-activated protein kinase.

hydroxylase in vitro, although the extent of activation reported in the literature is variable (reviewed in [S]). Phosphorylation of Ser8 and Serl9 does not activate tyrosine hydroxylase directly, but Serl9 phosphorylation is reported to make tyrosine hydroxylase sensitive to activation by members of the 14.3.3 family of proteins [6].

In rat brain [7], adrenal medulla [S] and phaeochromocytoma (PC) 12 cells (an adrenal medullary cell line) [2, 31 the phosphorylation of Serl9, Ser31 and Ser40 (but not Ser8) is increased by agonists that stimulate tyrosine hydroxylase activity and catecholamine synthesis. Ser31 is the major residue whose phosphorylation increases when PC12 cells are stimulated with nerve growth factor (NGF), a neurotrophic factor which induces sus- tained activation of p42 and p44 mitogen-activated protein kinases (MAP kinases), suggesting that these enzymes may mediate the phosphorylation of tyrosine hydroxylase by NGF [3]. Tumour-promoting phorbol esters, which activate protein kinase-C (PKC), also trigger the phosphorylation of Ser31 predominantly [3], suggesting that PKC does not phosphorylate tyrosine hydroxylase at Ser4O in vivo, but has the potential to activate tyrosine hydroxylase indirectly by triggering activation of the MAP kinase cascade. Exposure of perfused adrenal glands to nicotine for a few minutes induces a sixfold increase in the

Thomas et al. (ELI): J . Biochem. 247) 1181

CDK MAPKAP-KZ MAPK PKA, PKC MAPKAP-Kl 1 car 1 MAPKAP-K2 Cay-KII

S i r 8 Sir19 Sei31 Ser40

-C N Fig. 1. Protein kinases known to phosphorylate tyrosine hydroxylase in vitro. The residues numbers refer to rat tyrosine hydroxylase. Ser8 is phosphorylated by a cyclin-dependent protein kinase (CDK) [34], while Serl9 is phosphorylated by CaM-KII [I] and by MAPKAP-K2 [5]. Ser31 is phosphorylated by the p42 and p44 isoforms of MAP kinase [4] and Ser40 by PKA [I], CaM-KII [I, 351 PKC 19, 361, MAPKAP- K1 (a downstream target of p42 and p44 MAP kinases) and by MAP-

- _ _ _ _

KAP-K2 [5].

phosphorylation of Serl9 and much smaller increases in the phosphorylation of Ser31 and Ser40, while acetylcholine causes large increases in the phosphorylation of Serl9 and Ser40 within 30 s and in Ser31 phosphorylation after 5 min [9]. The activation of tyrosine hydroxylase induced by acetylcholine or K' depolar-

70 r I

h . % 3 8 v

h

2 c) u m

c .m

.m

E lo t 0-

0 10 20 30 40 50 60

time (min)

60

50 h

40

E h 30 v

Y .m .- P Y p 20

10

0

Arsenite SB 203580

C

+ + + +

isation is prevented by inhibitors of calciudcalmodulin-dependent protein kinase-I1 (CaM-KII), KN62 and KN93 [lo, 11 1. However, KN04 and KN92, which are structural isomers of KN62 and KN93, respectively, that do not inhibit CaM-KII, are equally effective in preventing the nicotine or depolarisation- induced activation of tyrosine hydroxylase [ 121. Thus, further evidence is needed to establish whether CaM-KII mediates tyrosine hydroxylase phosphorylation at Serl9 and Ser40 in vivo. Dibutyryl CAMP, which elevates the intracellular concentration of CAMP, increases the phosphorylation of tyrosine hydroxylase at Ser40, an effect presumed to be mediated by CAMP-dependent protein kinase (PKA) [3, 4, 91. However, it is unclear whether PKA mediates the increased phosphorylation of Ser40 triggered by physiological agonists.

MAP-kinase-activated protein kinase-2 (MAPKAP-K2) ic activated in vivo by stress-activated protein kinase-2 (SAPK2), a MAP kinase homologue,which is activated by adverse stimuli, such as heat and osmotic shock, ultraviolet-C irradiation, and inhibitors of protein synthesis. Catecholamines are released from the adrenal medulla in response to anxiety or fear (which could be regarded as other types of stress). We therefore wondered

250

h

% 5 200 a

p 150

v

h c .m

.13 U

P 9 100

2 50 3 E

0

250

h M

$ m

j

2 50 2

E 0 150 v

.m

a loo

0

Arsenite SB 203580

0 10 20 30 40 50 60

time (min)

+ + - + +

Fig.2. Activation of tyrosine hydroxylase (TH) and MAPKAP-K2 by arsenite in bovine adrenal chromaffin cells and PC12 cells. (A, B) Primary chromaffin cells were stimulated for the times indicated in the presence (0) or absence (0) of 0.5 mM sodium arsenite, and tyrosine hydroxylase (A) and MAPKAP-K2 (B) activities were measured after cell lysis. A, cells were incubated for 1 h in the presence of 20 pM SB 203580 prior to stimulation with arsenite. (C, D) PC12 cells were incubated for 1 h in the presence or absence of 20 pM SB 203580, then for 30 min in the continued presence or absence of SB 203580. Tyrosine hydroxylase (C) and MAPKAP-K2 (D) activities were measured as in (A, B). The results in (A-D) are presented as means2SEM for six determinations from two separate dishes. Similar results were obtained in several separate experiments (A, B) and another separate experiment (C, D).

1182 Thomas et al. ( E m J. Biocheni. 247)

3 / A SAPKZ 2 n

whether the SAPK2/MAPKAP-K2 cascade might play a role in the activation of tyrosine hydroxylase in vivo. The availability of a specific inhibitor of SAPK2, which prevents the activation of MAPKAP-K2 in vivo (SB203580) [13], provided an opportu- nity to examine this possibility. In this paper we find that three stimuli that trigger the activation of MAPKAP-K2 in chromaffin cells (sodium arsenite, osmotic shock and acetylcholine) activate tyrosine hydroxylase. However, while SB 203580 completely suppresses the activation of tyrosine hydroxylase and phosphorylation of Serl9 induced by sodium arsenite or osmotic shock at concentrations similar to those which prevent the activation of MAPKAP-K2, this drug does not prevent the phosphorylation of Serl9 induced by acetylcholine. Acetylcholine induces the activation of p42 and p44 MAP kinases in chromaffin cells, and PD 98059, a drug that prevents the activation of these enzymes [ 141, partially suppresses the acetylcholine-induced activation of tyrosine hydroxylase by preventing its phosphorylation at Ser31.

MATERIALS AND METHODS

Materials. Sodium arsenite was from Fluka, forskolin and acetylcholine from Sigma, the dihydrochloride derivative of (6R,S)-6-methyl-5,6,7,8-tetrahydropterin (6Me-H4Pte) from Schircks Laboratories, NGF, microcystin-LR and media for cell culture from GIBCO/BRL, protein-G-Sepharose from Phar- macia, and modified trypsin from The Boehringer Corp. SB 203580 was a generous gift from Drs J. Lee and P. Young (SmithKline Beecham, King of Prussia PA, USA), while PD 98059 was kindly provided by Dr A. Saltiel (Parke-Davis, MI, USA). Antiserum against the human tyrosine hydroxylase-I isoform was raised in rabbits at the University of Bergen. Antisera raised against a peptide near the C-terminus of MAPKAP-K2 [I51 and against a peptide comprising residues 605-627 of mouse Rsk-2 [I41 were raised in sheep at Scottish Antibody Production Unit and purified by Dr S. Dale in Dundee. Antise- rum raised against the C-terminal 14 residues of the Xenopus homologue of SAPK2 [I61 was a generous gift from Dr A. Ne- breda (EMBL, Heidelberg). Human tyrosine hydroxylase-I was a generous gift from Dr J . Mallet (LGNKNRS, Hopital de la Pite Salpetriere, Paris) expressed in Escherichia coli and purified as described [17]. The specific peptide inhibitor of CAMP- dependent protein kinase, the MAPKAP-K2 substrate peptide KKLNRTLSVA and a 32-residue peptide closely related to the C-terminus of ribosomal protein S6, [G245,G246]S6-(218-249) [I81 were synthesised by Mr F. B. Caudwell at Dundee.

Cell culture, stimulation and lysis. Rat PC12 cells were cultured on 6-cm diameter collagen-coated dishes in a 95% air/ 5 % CO, atmosphere as described [19]. Bovine adrenal chromaffin cells were prepared from fresh adrenals [20], plated on 3.5-cm diameter dishes at 2X106 cells/dish, and used 3-7 days after plating. Prior to stimulation, cells were incubated for 30 rnin in fresh medium and then for 1 h with or without inhibitors before stimulation with agonists. Following stimulation, the medium was removed and the dishes placed on ice, washed with 1 ml ice-cold 137 mM NaCI, 2.7 mM KCI, 8.1 mM disodium hydrogen phosphate, 1.5 mM potassium dihydrogen phosphate and lysed in 200 pl 20 mM Tridacetate pH 7.0, 0.27 M sucrose, 1 mM EDTA, 1 mM EGTA, 1 % (by mass) Triton X-100,lO mM sodium sn-glycerol 2-phosphate, 1 mM sodium orthovanadate, 1 mM benzamidine, 4 pg/ml leupeptin, 0.1 % (by vol.) 2-mer- captoethanol and 1 pM microcystin-LR. Lysates were centrifuged for 5 rnin at 13000Xg. The supernatants were removed and either used immediately or frozen in liquid nitrogen and stored at -80°C.

1

0.5

0

0.6

0.4 E

0.2

0 0 10 20 30

fraction number

I\ i Oe2

0 10 20 30 fraction number

Fig.3. Activation of SAPKZ and MAPK in bovine adrenal chromaffin cells by arsenite and acetylcholine. (A) Cells were stimulated for 30 min in the presence (0) or absence (0) of 0.5 mM sodium arsenite, lysed and chromatographed on Mono Q using a SMART system [22]. The flow rate was 0.2 mumin and fractions (0.1 ml) were assayed for their ability to activate MAPKAP-K2 (MAPKAP-K2 activator). The NaCl gradient is shown by the broken line. (B) Same as (A), except that the cells were stimulated for 5 min in the presence (0) or absence (0) of 100 pM acetylcholine. A, cells were incubated for 1 h with 50 pM PD 98059 prior to stimulation with acetylcholine. The positions of p42 and p44 MAP kinases (MAPK) and SAPK2 are marked. Although p42/ p44 MAP kinases activate MAPKAP-K2 in v i m , they do not activate MAPKAP-K2 in vivo for the reasons discussed in the text.

Immunoprecipitation of 3ZP-labelled tyrosine hydroxylase from rat PC12 cells and bovine adrenal chromaffin cells. 32P- labelling of PC12 cells and cell lysis was carried out as described in [21]. 'ZP-labelling of chromaffin cells was carried out in an identical manner except that the cells were incubated with [3ZP]orthophosphate for 2 h rather than 4 h prior to lysis, and were not washed prior to lysis. The lysates were centrifuged at 4°C for 10 rnin at 13OOOXg and the supernatants incubated for 30 min on a shaking platform with 5 p1 protein-G-Sephar- ose coupled to 20 p1 rabbit pre-immune serum. The suspensions were centrifuged for 2 min at 13000Xg and the supernatants incubated for 60 rnin with 5 p1 protein-G-Sepharose coupled to 20 p1 rabbit anti-(human tyrosine hydroxylase-I) serum. The protein-G- Sepharose . antibody . tyrosine hydroxylase complex was washed four times with 1 .O ml lysis buffer containing 0.5 M NaCl and twice with lysis buffer without 0.5 M NaCI.

Tryptic digestion of immunoprecipitated, "P-labelled tyrosine hydroxylase. Tyrosine hydroxylase immunoprecipitated from rat PC12 cells or bovine adrenal chromaffin cells was denatured in 10 pl 1 % SDS at 100°C and, after cooling to ambi- ent temperature, was electrophoresed on a 10 % polyacrylamide gel. After autoradiography, the 60-kDa band corresponding to tyrosine hydroxylase was excised and the protein eluted, precipi- tated with trichloroacetic acid, and oxidised by incubation for 1 h on ice with 0.1 ml performic acid. The solution was diluted

Thomas et al. (ELN J . Biuchem. 24T, 1183

2.5

2

n a I g 1.5

8 .- c) m

Y 0 m ' 0.5

.e 1

a 0.1 1 10

SB 203580 (pM)

4 n

3

.I 8 3

9 2

Y d b. .I Y

2 0

3 1

z n

B

0.1 1 i n

SB 203580 (FM)

Fig. 4. The activation of tyrosine hydroxylase (TH) MAPKAP-K2 by arsenite in bovine chromaffin cells is inhibited by similar concentrations of SB 203580. Cells were incubated for 60 min in the presence of the indicated concentrations of SB 203580, then for 30 min in the presence (0) or absence (0) of 0.5 mM sodium arsenite in the continued presence or absence of SB 203580. The lysates were assayed for tyrosine hydroxylase (A) and MAPKAP-K2 (B) activities. The results are presented as means t SEM for six determinations from two separate dishes (A) or for three determinations using lysate from a single dish of cells (B). Similar results were obtained in three (A) or one (B) further experiment.

with 4 vol. water, frozen and dried under vacuum, before resus- pension and digestion with trypsin as described previously WI.

Immunoprecipitation of protein kinases. Protein-G - Sepharose . antibody complexes were prepared as described above except that 2 pg anti-(MAPKAP-K2) antibody or 5 pg anti-(MAPKAP-K1) antibody or 5 p1 anti-SAPK2 serum were added to 5 pl protein-G-Sepharose. The lysates were incubated and the immunoprecipitates were washed and assayed as described previously [22]. Since the anti-SAPK2 antibodies were inhibitory, immunoprecipitation of this enzyme was assessed by measuring the extent to which the activity could be depleted from the solution.

Protein kinase assays. MAPKAP-K2 was assayed routinely by the phosphorylation of the peptide KKLNRTLSVA [23] and MAPKAP-K1 by the phosphorylation of [G245,G246]S6- (218-249) [24] and 1 U activity was that amount which catalysed the incorporation of 1 nmol phosphate into substrate in 1 min. The final concentration of each peptide in the assays was 30 pM, and their phosphorylation was limited to less than 10% to ensure that initial rate conditions were met. MAPKAP-K2 activators were assayed by their ability to activate MAPKAP- K2 as described 1221, and 1 U SAPK2 activity was that amount which increased the activity of MAPKAP-K2 by 1 U in 1 min.

Assay of tyrosine hydroxylase. These were carried out essentially as described [25]. The incubations (0.1 ml) containing 100 mM Hepes pH 7.0, 25 pM ['Hltyrosine (200 mCi/ mmol), 0.5 mg/ml catalase, 0.5 mM 6Me-H4Pte, 5mM dithio- threitol, 2.5 mM HCl and cell lysate (20 pg). The reactions were carried out at 30°C, started by the addition of 6Me-H4Pte and terminated after 10 min by adding 1 nil 7.5% charcoal in 1 M HCI. The suspension was centrifuged for 3 min at 13OOOXg and 200 p1 supernatant was added to 1 ml scintillant and the radioactivity measured. Control incubations were carried in which either cell lysate or 6Me-H4Pte were omitted and these reaction blanks subtracted from the values obtained in the presence of these substances.

RESULTS

Stressful stimuli activate tyrosine hydroxylase via the SAPK2 pathway. To examine whether catecholamine synthesis is triggered by the stress-activated kinase cascade that leads to the activation of MAPKAP-K2 we initially measured tyrosine hydroxylase activity after exposing bovine adrenal chromaffin cells (Fig. 2A) and rat PC12 cells (Fig. 2C) to sodium arsenite. This compound is known to mimic many of the effects of heat shock and has been shown to activate the SAPK2/MAPKAP-K2 pathway in other cells 113, 261. These studies showed that arsenite stimulated tyrosine hydroxylase activity by 2-3-fold after 30-60 min, the half-time for activation in bovine chromaffin cells being 15 min (Fig. 2A). Moreover, the effect of arsenite was completely suppressed by SB 203580 (Fig. 2A and 2C), a specific inhibitor of SAPK2 (see introduction). Consistent with this observation, arsenite induced the appearance of an activator of MAPKAP-K2, which eluted from Mono Q at a similar concentration of NaCl (0.35 M) to SAPK2 from other mammalian cells (Fig. 3A). This activator was inhibited by SB 203580 (data not shown), the IC,, (0.1 pM) being slightly lower than that obtained previously for SAPK2 from other vertebrate sources [I 31. The MAPKAP-K2 activator was immunoprecipitated essentially quantitatively by anti-SAPK2 antibodies, and immunoprecipitation was prevented by pre-incubation of the antibody with the specific peptide immunogen (data not shown).

Arsenite induced the activation of MAPKAP kinase-2 in bovine adrenal chromaffin cells and PC12 cells as expected (Fig. 2B and D), the half-time for activation of MAPKAP-K2 being similar to that observed for the activation of tyrosine hydroxylase (Fig. 2 B). The arsenite-induced activations of MAP- KAP-K2 (Fig. 2B) and tyrosine hydroxylase (Fig. 2A) were almost completely suppressed by SB 203580, the IC,, values (=1 pM) being similar for both enzymes (Fig. 4).

Osmotic shock is known to activate the SAPK2 pathway in other cells [13]. Exposure of primary chromaffin cells to 0.5 M sorbitol activated MAPKAP-K2 (fivefold) and tyrosine hydroxylase (twofold), the activation of each enzyme being completely

1184

Table 1. SB203580 does not prevent the activation of tyrosine hydroxylase by forskolin or NGF. Cells were incubated for 1 h in the absence (-1 or presence (+) of 20 pM SB 203580, then stimulated for 30 inin with agonist (0.5 mM sodium arsenite, 50 nglml NGF or 20 pM forskolin). Tyrosine hydroxylase (TH) activity was measured after cell lysis and the results are presented as the means f SEM for six determinations using two dishes of PC12 cells and for triplicate determinations using a single dish of bovine chromaffin cells. Similar results were obtained i n three separate experiments. The specific activity of tyrosine hydroxylase in extracts from unstimulated PC12 cells was 26.3 mU/mg and in unstimulated bovine chromaffin cells 25.7 mU/mg.

Thomas et al. ( E m J. Biochem. 247)

Cell line Agonist SB 203580 Relative TH activity

PC12 none none arsenite arsenite NGF NGF forskolin forskol in

Bovine chromaffin cells none none arsenite arsenite forskolin forskol i n

I .O 0.88 +- 0.03 2.1250.38 0.96 i 0.28 1.54 2 0.08 1.51 2 0.03 5.27 2 0.17 4.37 2 0.40

1 .o 0.88 2 0.04 2.46 i 0.07 1.16 -C 0.05 4.92 ? 0.34 5.08 2 0.44

Table 2. Effect of arsenite and SB 203580 on the 32P-labelling of tyrosine hydroxylase at Serl9, Ser31 and Ser40 in PC12 cells. ?*P- labelled tyrosine hydroxylase (Fig. 5A) was digested with trypsin and chromatographed on the C,, column as in Fig. 5B. The "P-radioactivity associated with the peptides containing Serl9 (S19a, S19b), Ser31 (S31) and Ser40 (S40) was quantitated. The results are presented as the means 5 SEM for three separate experiments.

Agonist SB 203580 "P-radioactivity associated with (20 PM)

Ser19 Ser31 Ser40

1.0 1 .o 1 .o - None Arsenite - 13.024.6 1.220.1 1 . 7 2 1.1 Arsenite + 2.2 2 0.9 0.7 2 0.3 0.9 5 0.3

Table 3. Effect of acetylcholine on the 3ZP-labelling of tyrosine hydroxylase at Ser8, Serl9, Ser31 and Ser40 in bovine adrenal chromaffin cells. "P-labelled tyrosine hydroxylase from Fig. 7 was digested with trypsin and chromatographed on the C,, column as in Fig. 5B. The "P-radioactivity associated with the peptides containing Ser8, Ser19, Ser3 1 and Ser4O was quantitated. Results are presented as means 5 SEM for the number of experiments indicated in parentheses. Where indicated, cclls were incubated for 1 h in the presence of 50 pM PD 98059 (PD) or 20 pM SB 203580 (SB) prior to stimulation.

Agonist Inhibitor "P-radioactivity associated with

Ser8 Serl9 Ser31 Ser40

1 .o 1 .o 1 .o 1 .o None (7) -

Acetylcholine (7) - 1.650.7 3.0?1.3 8.552.7 3 .121 .3 Acetylcholine (3) PD 1.0+0.3 2.020.3 2.320.3 2 .951 .2 Acetylcholine (3) SB 1 . 1 2 0 . 3 2 .250 .2 2.321.3 2 . 0 2 1 . 3

Fig. 5. Effect of arsenite and NGF on the phosphorylation of tyrosine hydroxylase and its tryptic peptides in PC12 cells. (A) PC12 cells were incubated for 4 h with ['2P]orthophosphate as in [21] with SB 203580 being included at 20 pM for the final hour, where indicated. The cells were stimulated for 15 min with 0.5 mM sodium arsenite or 50 ngl ml NGF and lysed. Tyrosine hydroxylase was immunoprecipitated from the lysate of one dish of cells, denatured in SDS, electrophoresed on a 10 YC polyacrylamide gel and autoradiographed to reveal the 60-kDa band corresponding to rat tyrosine hydroxylase. (B) Tyrosine hydroxylase immunoprecipitated from cell lysates was digested with trypsin and applied to a Vydac C,, column (Separations Group) equilibrated in 0.1 % (by vol.) trifluoroacetic acid. The column was developed with a linear acetonitrile gradient (broken line) at a flow rate of 0.8 mllmin and fractions of 0.8 ml were collected. 0, cells stimulated for 15 min with 0.5 mM sodium arsenite; 0 unstimulated cells. The peptides containing Ser8 (S8), Serl9 (S 19), Ser31 (S31) and Ser40 (S40) are marked. The two peptides marked S19a and S19b correspond to residues 17-23 and 16-23, respectively, while the peptides S31 and S40 correspond to residues 25-33 and 38-46, respectively. The identities of these peptides have been established by Haycock [3j and the location of the phosphorylation site within each "P-labelled peptide was confirmed in the present work by solid-phase sequence analysis 1261. A single burst of radioactivity was obtained after the third (Slga), fourth (S19b), seventh (S31) and third (S40) cycle of Edman degradation (data not shown). The minor peptide whose labelling was unaffected by arsenite was not analysed in the present work, but is reported to comprise residues 2-12 and to be phosphorylated at Ser8 [3] . (C) Same as (B), except that the cells were stimulated for 15 inin with NGF (0) or not stimulated (0). Solid-phase sequence analysis established that Ser31 phosphorylation was stimulated specifically by NGF.

Thomas et al. ( E m J. Biochem. 247) 1185

200 h M E

E 150 5 W

h

* a

Y .m

.m s 100

2

2 2 0

4 50

M

A T

100

n

\ z a 75 E

.% 50

a

W

h Y .- Y 0 m

b 25

0

I

0 10 20 30 time (min)

C

0 10 20 30 time (min)

n

\ z 2oo 2 E W

150 h

* .c, u m 100

Y .m .-

7 4

M c 50 M E4 4

d

100

n z 5 75

E

'5 50

W

h Y

.m Y u m

3 25

0

T B

I

I

0 10 20 30

time (min)

D

0 10 20 30

time (min) Fig. 6. Activation of MAPKAP-K2, MAPKAP-K1 and tyrosine hydroxylase by acetylcholine in bovine adrenal chromaffin cells. Cells were stimulated for the times indicated in the presence (0) or absence (0) of 0.1 mM acetylcholine, and MAPKAP-K2 (A), MAPKAP-K1 (B) and tyrosine hydroxylase (C, D) activities were measured after cell lysis. (A and C) A, cells were incubated for 1 h in the presence of 20 pM SB 203580 prior to stimulation with acetylcholine. (D) A, cells were incubated for 1 h in the presence of 50 pM PD 98059 prior to stimulation with acetylcholine. The results are presented as means 2 SEM for six determinations from two separate dishes of cells (A, B, D). Similar results were obtained in at least three (A) or one (B, D) further experiment. For the experiment in (C), results are presented as meansiSEM for triplicate determinations from the indicated number of dishes.

suppressed by SB 203580. Sorbitol induced the activation of tyrosine hydroxylase and MAPKAP-K2 with a half-time of 5 min, i.e. more rapidly than the arsenite-induced activation of these enzymes. However, the activation was transient, returning to near basal levels after 30 rnin (data not shown).

Stimulation of PC12 cells with NGF, or exposure of either rat PC12 or bovine adrenal chromaffin cells to the CAMP-elevat- ing agent forskolin, did not activate MAPKAP-K2 (data not shown), and the activation of tyrosine hydroxylase by these agonists was not affected significantly by SB 203580 (Table 1).

Identification of Serl9 as the major residue in tyrosine hydroxylase that becomes phosphorylated in response to arse-

nite in rat PC12 cells. Consistent with the activation of tyrosine hydroxylase by arsenite, this agonist increased the phosphorylation of rat tyrosine hydroxylase in PC12 cells, an effect that was prevented by SB 203580 (Fig. 5A). In contrast, the NGF- stimulated phosphorylation of rat tyrosine hydroxylase and the level of tyrosine hydroxylase phosphorylation in unstimulated cells were unaffected by SB 203580 (Fig. 5A).

3ZP-labelled tyrosine hydroxylase was immunoprecipitated from PC12 cell lysates and digested with trypsin, and subsequent chromatography on a C,, column resolved five tryptic phosphopeptides. The identitites of these peptides were established by Haycock [3] and confirmed in the present work (Fig. 5B). The

1186 Thomas et al. ( E m J. Biochem. 247)

K

.n 0 4 - e n b c1 0

m d

2- .CI

Y

first two phosphopeptides that eluted from the column were phosphorylated at Serl9 and the "P-radioactivity associated with both of these peaks increased more than tenfold after stimulation with arsenite (Fig. 5B). The increased 32P-Iabeling of the peaks was prevented by SB 203580 (Table 2) . In contrast, NGF only stimulated the phosphorylation of the peptide containing Ser31 (Fig. 5C).

The effect of arsenite on the phosphorylation of tyrosine hydroxylase was examined in bovine adrenal chromaffin cells with very similar results. Arsenite increased the 12P-labelling of tyrosine hydroxylase, and tryptic digestion followed by C,, chromatography and sequence analysis revealed a specific increase in phosphorylation of the peptides containing Serl9. The arsenite- induced phosphorylation of Serl9 was prevented by prior incubation of the cells with SB 203580 (data not shown).

-20 s

-10 5

k

f 0

c1 .I

S31

4

0

Acetylcholine triggers the activation of SAPK2IMAPKAP- K2 and MAP kinaseMAPKAP-K1 in bovine chromaffin cells. The binding of acetylcholine to nicotinic receptors on adrenal medullary cells triggers the release of catecholamines and their resynthesis in vivo. Acetylcholine was found to activate MAPKAP-K2, the half-time for activation (about 1 min; Fig. 6A) being 20-fold faster than the half-time for activation by arsenite (Fig. 2B). Activation of MAPKAP-K2 was prevented by SB 203580 (Fig. 6A). Mono Q chromatography of lysates from acetylcholine-stimulated cells revealed three peaks capable of activating MAPKAP-K2 (Fig. 3B). One of these eluted at the same NaCl concentration as SAPK2 from arsenite- treated cells (Fig. 3 A), was inmunoprecipitated by anti-SAPK2 antibodies and inhibited by SB 203580 with an IC,, of 0.1 pM (data not shown). These findings establish that this enzyme is SAPK2 and that it mediates the acetylcholine-induced activation

The other two activators of MAPKAP-K2 (Fig. 3B) eluted at the same NaCl concentrations as p42 and p44 MAP kinases from other mammalian cells and, like these MAP kinases, had high activity towards myelin basic protein (data not shown). Moreover, their appearance was prevented by PD 98059 (Fig. 3 B), a drug that inhibits activation of the MAP kinase cascade [14]. However, although p42 and p44 MAP kinases can activate MAPKAP-K2 in vitro ([26], Fig. 3B), they are not rate limiting for its activation in vivo because the activation of MAP- KAP-K2 by acetylcholine is completely inhibited by SB 203580 (Fig. 6A) and unaffected by PD 98059 (data not shown). The finding that PD 98059 slightly decreased the activation of SAPK2 (Fig. 3B) suggests that the p42/p44 MAP kinases may play an unidentified role in enhancing the activation of SAPK2 by acetylcholine.

Consistent with the activation of p42 and p44 MAP kinases, acetylcholine also activated MAPKAP-K1 (a physiological substrate of p42/p44 MAP kinases) with similar kinetics to the activation of MAPKAP-K2 (Fig. 6A and B).

of MAPJSAP-K2.

Effect of PD 98059 on the acetylcholine-induced activation and phosphorylation of tyrosine hydroxylase in bovine adrenal chromaffin cells. Acetylcholine induced a greater activation of tyrosine hydroxylase (Fig. 6C and D) than arsenite (Fig. 2A), which peaked after 1 min and decreased thereafter, returning to near basal levels after 30 min. PD 98059 had almost no effect on tyrosine hydroxylase activation after 1 min, but suppressed activity by 40% after 5 min (Fig. 6D). Stimulation with acetylcholine increased the '2P-labelling of four tryptic peptides derived from tyrosine hydroxylase, whose identitites were established by Haycock [9] and confirmed in the present work (Fig. 7). Amino-acid-sequence differences explain why the tryptic peptides from bovine tyrosine hydroxylase (Fig. 7) elute at

h

time (min) Fig. 7. Effect of acetylcholine on the phosphorylation of tyrosine- hydroxylase-derived tryptic peptides in bovine adrenal chromaffin cells. Cells were incubated for 2 h with [32P]orthophosphate, then stimulated for 5 inin in the absence or presence of 0.1 mM acetylcholine. Tyrosine hydroxylase was immunoprecipitated from the cell lysates, digested with trypsin and chromatographed on a C,, column as described in the legend to Fig. 5, except that fractions of 0.2 ml were collected. The full lines show the results from cells stimulated witti acetylcholine and the broken line the results from unstimulated cells. The elution positions of the five major tryptic phosphopeptides containing Ser8 (S8), Serl9 (S19) Ser31 (S31), and Ser40 (S40) coincide with those observed in an earlier study [9], except that only the oxidised form of the S31 peptide was observed because the protein was subjected to performic acid oxidation prior to tryptic digestion. The phosphopeptides correspond to the same residue numbers as for rat tyrosine hydroxylase (Fig. 51, but they elute at different positions due to amino-acid-sequence differences between rat and bovine tyrosine hydroxylases. The location of the phosphorylation site within each '*P-labelled peptide was established by solid-phase sequence analysis [26]. A single burst of radioactivity was obtained after the seventh (%), third (S19a), fourth (S19b), seventh (S31) and third (S40) cycle of Edman degradation (data not shown), as expected.

different acetonitrile concentrations from their rat counterparts (Fig. 5B). PD 98059 suppressed the phosphorylation of Ser31 without affecting the phosphorylation of Serl9 or Ser40 significantly (Table 3).

Effect of SB 203580 on the activation and phosphorylation of tyrosine hydroxylase in bovine adrenal chromaffin cells induced by acetylcholine. SB 203580 (20 pM) had almost no effect on tyrosine hydroxylase activation by acetylcholine after 1 min, but like PD 98059, it partially suppressed activation after 5 min (Fig. 6 C). In contrast to the activation of tyrosine hydroxylase by arsenite, however, SB 203580 did not suppress the acetylcholine-induced phosphorylation of tyrosine hydroxylase at Ser19 (or Ser40) to a significant extent, but suppressed the phosphorylation of Ser31 by 80% (Table 3). Consistent with this observation, the effects of PD 98059 and SB 203580 on tyrosine hydroxylase activation by acetylcholine were not additive, and the presence of both compounds did not inhibit activation more than either inhibitor alone (data not shown). The effect of SB 203580 appeared to be specific, however, since SKF 105809 (a closely related analogue which does not inhibit SAPK2) [13] had no effect on the activation of tyrosine hydroxylase or MAP- KAP-K2 by acetylcholine (data not shown). Moreover, SB 203580 had no effect on the acetylcholine-induced activation of p42/p44 MAP kinases and did not affect the phosphorylation of tyrosine hydroxylase by p42 MAP kinase in vitro. Furthermore, SAPK2 could not phosphorylate purified tyrosine hydroxylase in vitro (data not shown).

Thomas et al. (Eur: J . Biochem. 247) 1187

Stress Acetylcholine

1

?

4 MKK6

SAP= 1 . f- SB 203580

c MAPKAP-K2

1 TH (Serl9)

1 catecholamine

Synthesis

I C ~ M Kinase II? I I I I

c MKKl I MKK6 c I

I 1 SB 203580%

c ~ 4 2 1 ~ 4 4 MAPK

MAPKAP-K1 I I 7

TH (Ser31) TH (Serl9, Ser40)

Catecholamine Synthesis

c MAPKAP-K2

Fig. 8. Molecular mechanisms by which stressful stimuli and acetylcholine trigger the phosphorylation of tyrosine hydroxylase at Serl9 and Ser31, respectively. PD 98059, which binds specifically to MAP kinase kinase-I (MKK1) and prevents its activation, suppressed the activation of p42/p44 MAP kinases by acetylcholine. This indicates that MKKl is the upstream activator of p42/p44 MAP kinases in bovine chromaffin cells. The identity of the MKKl activator in chromaffin cells remains to be established, but in PC12 cells it is an isoform of Raf. PD 98059 prevented the acetylcholine-induced phosphorylation of tyrosine hydroxylase at Ser3 1, indicating that the phosphorylation of this residue is mediated by p42/ p44 MAP kinases. PD 98059 did not suppress the acetylcholine-induced 3ZP-labelling of tyrosine hydroxylase at Ser40, indicating that MAPKAP- K1 is not a rate-limiting enzyme for this phosphorylation. SB 203580, a specific inhibitor of SAPK2, prevents the activation of MAPKAP-K2 and hence the phosphorylation of tyrosine hydroxylase at Serl9 induced by stressful stimuli. However SB 203580 does not inhibit the 32P-Iahelling at Serl9 or Ser40 induced by acetylcholine at a time (1 min) when activation of tyrosine hydroxylase is maximal, indicating that MAPKAP-K2 activity is not rate-limiting for the phosphorylation of these two residues. The phosphorylation of Serl9 and Ser4O induced by acetylcholine is dependent on the presence of calcium [29] and may be mediated by CaM-KII, but the evidence is not definitive, for reasons discussed in the text. Mono S chromatography of extracts from arsenite-stimulated or acetylcholine-stimulated chromaffin cells revealed two major peaks that activated SAPK2. The activator retained by the column eluted with MKK6 and was immunoprecipitated quantitatively by anti-MKK6 antibodies. The activator not retained by the column was not immunoprecipitated by anti-MKK3, anti-MKK4 or anti-MKK6 antibodies and its identity remains to be established (Thomas, G., unpublished results).

DISCUSSION In this paper we demonstrate that exposure of chromaffin

cells to stressful stimuli (sodium arsenite and osmotic shock) activates tyrosine hydroxylase 2- 3-fold and that this correlates with its specific phosphorylation at Serl9. The activation of tyrosine hydroxylase, and the phosphorylation of Serl9, is prevented by SB 203580 at a concentration similar to that which prevents the activation of MAPKAP-K2. Serl9 lies in a consen- sus sequence for phosphorylation by MAPKAP-K2 (Phe-Xaa- Arg-Xaa-Xaa-Ser-) [23] and is phosphorylated in vitro at a similar rate to heat-shock protein 27 [ 5 ] , a physiological substrate for MAPKAP-K2 [13]. Moreover, when PC12 cell extracts were chromatographed on Mono Q, only one peak of protein kinase activity was found to be stimulated by arsenite and to phosphorylate the peptide KGFRRAVSEQDAK (corresponding to residues 12-24 of rat tyrosine hydroxylase; data not shown). This enzyme eluted with MAPKAP-K2, was immunoprecipitated by anti-(MAPKAP-K2) antibodies, and its appearance was prevented by SB 203580 (Thomas, G., unpublished results). The only other protein kinase known to phosphorylate tyrosine hydroxylase at Serl9 is CaM-KII, whose activity is unaffected by SB 203580 1221. Adverse stimuli therefore appear to activate tyrosine hydroxylase via the mechanism depicted in Fig. 8.

Although our results suggest that MAPKAP-K2 mediates the stress-induced activation of tyrosine hydroxylase in chromaffin

cells, it should be noted that MAPKAP-K2 is not specific for Serl9 in vitro but also phosphorylates Ser4O [ 5 ] . The discrep- ancy between the in vitro and in vivo data suggests that the dephosphorylation of Ser40 (probably catalysed by protein phos- phatase 2A [27]) is much faster than its rate of phosphorylation by MAPKAP-K2. Alternatively, phosphorylation of Ser40 may be suppressed (or its dephosphorylation enhanced) under stressful conditions by the interaction of tyrosine hydroxylase with another protein.

Although our results establish that stressful stimuli activate tyrosine hydroxylase in chromaffin cells via the SAPK2hlAP- KAP-K2 pathway, further work is needed to establish whether this is purely a pharmacologicallpathological effect or whether physiological stimuli exist that activate tyrosine hydroxylase via this route in either chromaffin cells or the central nervous system. In other cells bacterial lipopolysaccharide, tumour necrosis factor and interleukin-1 are potent activators of MAPKAP-K2 [13, 151 but we have been unable to detect any activation of tyrosine hydroxylase or MAPKAP-K2 by these agonists in bovine adrenal chromaffin cells (Thomas, G., unpublished results).

In chromaffin cells, acetylcholine is thought to mediate the activation of tyrosine hydroxylase that results from stimulation of the splanchnic nerve. Acetylcholine opens calcium channels in the plasma membrane triggering a rise in the cytosolic concentration of calcium ions (reviewed in [28]) and the activation

1188 Thomas et al. (Eua J. Biochem. 247)

of calcium/calmodulin-dependent protein kinases, such as CaM- KII. Serl9 and, to a lesser extent, Ser40 are residues whose phosphorylation is induced up to 1 min after exposure to acetylcholine [91 when the activation of tyrosine hydroxylase is maximal (Fig. 6C, [29]). CaM-KII phosphorylates Serl9 preferen- tially and Ser40 more slowly in vitro [5], raising the possibility that this enzyme mediates the acetylcholine-induced phosphorylation of these two sites (Fig. 8). However, although the nicotine-induced activation of tyrosine hydroxylase is prevented by KN62, an inhibitor of CaM-KII [lo, 111, more recent studies have revealed that tyrosine hydroxylase activation is also prevented by KN04, a structural analogue of KN62 which does not inhibit CaM-KII [ 121. Definitive evidence that CaM-KII mediates the phosphorylation of Serl9 (or Ser4O) in vivo is therefore still lacking.

In this paper we have shown that acetylcholine induces a rapid activation of the MAPMAPKAP-K1 and SAPK2MAP- KAP-K2 pathways (Fig. 6). The mechanism of activation is un- certain, but may involve the protein tyrosine kinase Pyk2 which, in PC12 cells, becomes activated within seconds in response to calcium influx, and which (in transfection-based experiments) activates both MAPK and SAPKl (also known as c-dun N-terminal kinase), another MAPK homologue that is activated by the same stimuli as SAPK2 [30, 311. The MAPWMAPKAP-K1 and SAPK2/MAPKAP-K2 pathways do not contribute significantly to the initial activation of tyrosine hydroxylase by acetylcholine because specific inhibitors of these pathways (PD 98059 and SB 20358; Fig. 8) have no effect up to 1 rnin (Fig. 6C, D). However, after 5 min, PD 98059 prevents the phosphorylation of tyrosine hydroxylase at Ser3 1 and partially suppresses its activation. These findings are consistent with those of Haycock, who found no increase in Ser31 phosphorylation up to 2 min after stimulation with acetylcholine and reported that Ser3 1 phosphorylation only became significant after 4 inin [9]. Since p42 and p44 MAP kinases are the only enzymes known to phosphorylate Ser31 in vitro, and their activation by acetylcholine is prevented by PD 98059 (Fig. 3B), these findings establish that p42/p44 MAP kinases mediate the acetylcholine-induced phosphorylation of Ser31 (Fig. 8). In contrast, PD 98059 does not prevent the acetylcholine-induced phosphorylation of Ser40 (Ta- ble 3) indicating that, although MAPKAP-K1 is activated by acetylcholine in vivo and phosphorylates tyrosine hydroxylase at Ser4O in vitro, it is not rate limiting for Ser4O phosphorylation under these conditions.

SB 203580 prevented the activation of MAPKAP-K2 by acetylcholine (Fig. 6A) and, like PD 98059, it partially inhibited tyrosine hydroxylase activation after 5 min (Fig. 6C). The effect of SB 203580 appeared to be specific because SKF 105809, a closely related analogue which does not inhibit SAPK2 [13], did not prevent tyrosine hydroxylase activation by acetylcholine or arsenite (Thomas, G., unpublished results). However, in contrast to its effect on the activation of tyrosine hydroxylase by arsenite, SB 203580 did not decrease the phosphorylation of Serl9 (or Ser4O) significantly, and its major effect was to suppress the phosphorylation of Ser31 (Table 3). As SAPK2 does not phosphorylate tyrosine hydroxylase at Ser31 (or any other residue) in vitro, the effect of SB 203580 must be indirect. SB 203580 does not prevent the phosphorylation of tyrosine hydroxylase by p42 MAP kinase in vitro, nor does it affect the activation of p42/ p44 MAP kinases by acetylcholine (Thomas, G., unpublished results). These observations suggest that the inhibition of SAPK2 prevents ~ 4 2 1 ~ 4 4 MAP kinases from phosphorylating tyrosine hydroxylase in vivo in an unknown way. Possible mechanisms would include inhibition of the translocation of MAP kinases from the cytosol (where tyrosine hydroxylase is located)

to the nucleus, or stimulation of the interaction of tyrosine hydroxylase with an unknown activator(s).

In summary, MAPKAP-K2 mediates the activation of tyrosine hydroxylase and its phosphorylation at Serl9 in response to stressful stimuli, while p42/p44 MAP kinases contribute to the activation of tyrosine hydroxylase by acetylcholine by phosphorylating Ser3 1. CaM-KII may mediate the acetylcholine-induced phosphorylation of Serl9 and Ser40 (Fig. 8), but the evidence is not conclusive. It is therefore possible that the acetylcholine-induced phosphorylation of tyrosine hydroxylase at Serl9 andlor Ser40 is mediated by other stress-activated protein kinases that are unaffected by either SB 203580 or PD 98059, such as SAPK1, SAPK3 [32] or SAPK4 [33]. Since SAPKs phosphorylate serine and threonine residues that are followed by proline, these enzymes could not phosphorylate tyrosine hydroxylase at Serl9 or Ser40 directly, but could exert their effects indirectly by activating other protein kinases.

We thank the Wellcome Trust for a postgraduate studentship (G. T.), and the UK Medical Research Council (P. C.), the Royal Society (P. C.) and the Research Council of Norway (J. H.) for financial support.

REFERENCES 1. Campbell, D. G., Hardie, D. G. & Vulliet, P. R. (1986) J. Biol.

2. Mitchell, J. P., Hardie, D. G. & Vulliet, P. R. (1990) J. Biol. Chem.

3. Haycock, J. W. (1990) J. Biol. Chem. 265, 11682-11691. 4. Haycock, J. W., Ahn, N. G., Cobb, M. H. & Krebs, E. G. (1992)

Proc. Natl Acad. Sci. USA 89, 2365 -2369. 5 . Sutherland, C., Alterio, J. , Campbell, D. G., Le Bourdelles, B., Mal-

let, J., Haavik, J. & Cohen, P. (1993) ELM J. Biochem. 217, 715- 722.

6. Ichimura, T., Isobe, T., Okuyama, T., Yamauchi, T. & Fujisawa, H. (1987) FEBS Lett. 219, 79-82.

7. Haycock, J. W. & Haycock, D. A. (1991) J. Biot. Chem. 266,5650- 5657.

8. Haycock, J. W. & Wakade, A. R. (1992) J. Neurochem. 58,57-64. 9. Haycock, J. W. (1993) Neurochem. Res. 18, 15-26.

10. Tokumitsu, H., Chijiwa, T., Hagiwara, M., Nagatsu, T. & Hidaka, H. (1991) Biochem. Biophys. Re.r. Commun. 181, 968-975.

11. Ishikawa, N., Hashiba, Y. & Hidaka, H. (1991) J. Pharmacol. Exp. Ther 254, 598-602.

12. Marley, P. D. & Thomson, K. A. (1996) Biochem. Biophys. Res. Commun. 221, 15-18.

13. Cuenda, A., Rouse, J., Doza, Y. N., Meier, R., Cohen, P., Gallagher, T. F., Young, P. R. & Lee, 3. C. (1995) FEBS Lett. 364, 229- 233.

14. Alessi, D. R., Cuenda, A,, Cohen, P., Dudley, D. T. & Saltiel, A. R. (1995) J. Biol. Chem. 270, 27489-27494.

15. Beyaert, R., Cuenda, A., Vanden Berghe, W., Plaisance, S., Lee, J. C., Haegeman, G., Cohen, P. & Fiers, W. (1996) EMBO J. 15,

16. Rouse, J., Cohen, P., Trigon, S . , Morange, M., Alonso-Llamazares, A,, Zamanillo, D., Hunt, T. & Nebreda, A. (1994) Cell 78, 1027- 1037.

17. Haavik, J., LeBourdelles, B., Martinez, A,, Flatmark, T. & Mallet, J. (1991) Eur: J . Biochern. 199, 371-378.

18. Price, D. J., Nemenoff, R. A. & Avruch, J. (1989) J. Biol. Chem. 264, 13 825-13 833.

19. Gomez, N., Tonks, N. K., Morrison, T. & Cohen, P. (1990) FEBS Lett. 271, 1 19 - 122.

20. Livett, B. G. (1984). Physiol. Rev. 64, 1130-1161. 21. Ben-Levy, R., Leighton, I. A,, Doza, Y. N., Attwood, P., Monice,

N., Marshall, C. J. & Cohen, P. (1995) EMBO J. 14, 5820-5930. 22. Tan, Y., Rouse, J., Zhang, A., Cariati, S., Cohen, P. & Comb, M. J.

23. Stokoe, D., Caudwell, F. B., Cohen, P. T. W. & Cohen, P. (1993)

Chem. 261, 10489-10492.

265, 22358-22364.

1914- 1923.

(1996) EMBO J . 15, 4629-4642.

Biochem. J. 296, 843-849.

Thomas et al. (Eul: J. Biochem. 247) 1189

24. Lavoinne, A,, Erikson, E., Maller, J . L., Price, D. J., Avruch, J. &

25. Reinhard J. F. Jr, Smith, G. K. & Nichol, C. A. (1986) Llfe Sci. 39,

26. Stokoe, D., Campbell, D. G., Nakielny, S., Hidaka, H., Leevers, S. J., Marshall, C. & Cohen, P. (1992) EMBO J. 11, 3985-3999.

27. Haavik, J., Schelling, D. L., Campbell, D. G., Andersson, K. K., Flatmark, T. & Cohen, P. (1989) FEBS Lett. 251, 36-42.

28. Burgoyne, R. D. (1991) Biochim. Biophys. Acta 1071, 174-202. 29. Waymire, J. C., Johnston, J. P., Hummer-Lickteig, K., Lloyd, A,,

Vigny, A. & Craviso, G. L. (1988) J. B id . Chem. 263, 12439- 12447.

30. Lev, S., Moreno, H., Martinez, R., Canoll, P., Peles, E., Musaccio, J. M., Plowman, G. D., Rudy, B. & Schlessinger, J. (1995) Nature

Cohen, P. (1991) Eur: J . Biochem. 199, 723-728.

2185 -2189.

376, 737-745.

31. Tokiwa, G., Dikic, I., Lev, S. & Schlessinger, J. (1996) Science 273,

32. Cuenda, A,, Cohen, P., Buee-Scherrer, V. & Goedert, M. (1997)

33. Goedert, M., Cuenda, A,, Craxton, M., Jakes, R. & Cohen, P. (1997)

34. Hall, F. L. & Vulliet, P. R. (1991) Curl: Opin. Cell B id . 3, 176-

35. Vulliet, P. R., Woodgett, J. R. & Cohen, P. (1984) J. Bid. Chem.

36. Albert, K. A,, Helmer-Matyjek, E., Nairn, A. C., Muller, T. H., Hay- cock, J. W., Green, L. A,, Goldstein, M. & Greengard, P. (1984) Proc. Natl Acad. Sci. USA 81. 7713-7717.

792-794.

EMRO J. 16, 295-305.

EMBO J. 16, 3563-3571.

184.

259, 13680-13683.

participation of a stress-activated protein kinase cascade in the activation of tyrosine hydroxylase...

Documents