Activation of Mitogen-activated Protein ( MAP ) Kinase by a MAP Kinase-Kinase *

Previously it has been shown that acute 12-0-tetradecanoylphorbol-13-acetate treatment of intact US37 cells results in activation of mitogen-activated protein (MAP) kinase and a MAP kinase activator. MAP kinase activator induces phosphorylation of MAP kinase on tyrosine and threonine residues, thereby activating MAP kinase. Here, experiments with the irreversible kinase inhibitor, 6’-p-fluorosulfonylbenzoyladenosine (FSBA), show that MAP kinase activator is in fact a MAP kinase-kinase. Treatment of MAP kinase activator with FSBA results in complete inactivation. This inactivation is prevented by a 10-fold excess of ATP. Inactivation of MAP kinase by FSBA does not affect the extent of threonineltyrosine phosphorylation induced by MAP kinase-kinase.

Reversible phosphorylation on serine, threonine, and tyrosine residues is a common mechanism by which the functions of proteins are regulated in mammalian cells (Ref. 1 and references therein). Recently, much work has centered on the regulation of the family of Eitogen-activated protein (MAP)' kinases (also known as extracellular-regulated kinases (ERKs)) (2-5). These kinases are activated by a wide range of mitogens and extracellular stimuli in diverse cell types. Activation is associated with numerous physiological responses (2).
Activity of MAP kinase has been shown to require phosphorylation on both tyrosine and threonine residues (6). Two key regulatory sites have been identified as the threonine and tyrosine in a conserved TEY motif (7,8). This initially led to the idea that MAP kinase acts as an integrator of signals from separate serine/threonine and tyrosine kinases (7). However, it has since been shown that a single activity is responsible for inducing the phosphorylation of MAP kinase on both tyrosine and threonine residues (9)(10)(11)16). This dual phosphorylation is accompanied by activation of MAP kinase.
There has been much debate as to the precise nature of the MAP kinase activating activity. A number of groups have shown that MAP kinase is capable of autophosphorylation on both tyrosine and threonine residues and that this is accompanied by limited autoactivation (12)(13)(14). Thus, it has been * The costs of publication of this article were defrayed in part by the payment of page charges, This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
proposed that the MAP kinase activator is not itself a kinase but simply interacts allosterically with MAP kinase and stimulates this autophosphorylation and autoactivation.
To address this question we have conducted experiments with the irreversible kinase inhibitor and ATP analogue FSBA. These experiments indicate that MAP kinase activator is in fact a dual tyrosine/threonine-specific kinase.

EXPERIMENTAL PROCEDURES
Materials-Radioisotopes and the ECL Western blotting kit were obtained from Amersham, UK. Other chemicals and biochemicals were obtained from Sigma, UK.
Partial Purification of Active and Inactive MAP Kinase-Both active and inactive MAP kinase were partially purified by phenyl-Sepharose chromatography. Active MAP kinase was prepared from cells treated with TPA as described previously (11). Inactive MAP kinase was prepared from untreated cells grown in 2% fetal calf serum for 16 h prior to harvesting. Extraction of cell pellets and phenyl-Sepharose chromatography were performed as described elsewhere (11). Active MAP kinase was detected by assaying kinase activity toward myelin basic protein (see "MAP Kinase and MAP Kinase Activator Assays"). Inactive MAP kinase was detected by Western blotting (see "Other Methods").The fractions containing active or inactive MAP kinase were pooled and stored in liquid nitrogen.
Partial Purification of MAP Kinase Activator-MAP kinase activator was partially purified from TPA-treated cells by Mono-Q and Mono-S ion-exchange chromatography essentially as described previously (11). However, the Mono-S column was equilibrated in and eluted with buffer containing 0.03% Brij 35 and 5% glycerol. Active fractions were pooled and stored in liquid nitrogen.
MAP Kinase and MAP Kinase Activator Assays-MAP kinase was assayed as previously (11) except that the concentration of myelin basic protein as substrate was 0.22 mg/ml. One unit of MAP kinase activity is that activity which incorporates 1 nmol of phosphate per min into substrate under these standard conditions.
MAP kinase activator was assayed as for MAP kinase except that the assay contained 0.02 unit/ml (basal activity) of inactive MAP kinase. Activation of this by MAP kinase activator was measured. One unit of MAP kinase activator is that amount of activity which increases the activity of MAP kinase by 1 unit under the conditions described.
Inactivation of MAP Kinase Activator by FSBA-MAP kinase activator was removed from storage buffer containing benzamidine, p-mercaptoethanol, and PMSF by rapid filtration through a Sephadex G-50 spun column at 4 "C. The column was preequilibrated in 20 mM P-glycerophosphate (pH 7.5), 20 mM NaF, 2 mM EDTA, 0.2 mM Na3V04, 0.03% Brij 35, 5% glycerol, and 0.1 mg/ml bovine serum albumin (Buffer A). MAP kinase activator was incubated overnight in the dark at 4 "C with a final concentration of 0.1 mM FSBA in 2% (v/v) Me&O or the Me2S0 vehicle; the final Me2S0 concentration was 2% v/v. Incubations were carried out in the presence or absence of 1 mM ATP (pH 7.5) as a competitor for FSBA. Excess FSBA and ATP were removed by rapid filtration through a Sephadex G-50 column at 4 "C. The column was equilibrated in buffer A (+0.3% v/v P-mercaptoethanol, 10 mM benzamidine, 50 pg/ml PMSF). The eluate was assayed for MAP kinase activator activity.
Inactivation of MAP Kinase by FSBA-MAP kinase was separated from benzamidine, P-mercaptoethanol, and PMSF by dialysis for 7 h against 2 X 500 ml of 20 mM P-glycerophosphate (pH 7.5), 20 mM NaF, 2 mM EDTA, 0.2 mM Na,VO,, 50% ethanediol at 4 "C. Bovine serum albumin to a final concentration of 0.1 mg/ml was added to the MAP kinase prior to dialysis. Following dialysis, MAP kinase was incubated overnight in the dark at 4 "C with a final concentration of 1.0 mM FSBA or the MezSO vehicle; the final Me2S0 concentration was 2% v/v. Excess FSBA was removed by dialysis for 6 h against 4 X 1000 ml of 20 mM P-glycerophosphate (pH 7.5), 20 mM NaF, 2 mM EDTA, 0.2 mM Na3V04, 0.3% v/v P-mercaptoethanol, 10 mM benzamidine, 50 pg/ml PMSF at 4 "C. Leupeptin was added to a final concentration of 25 pg/ml after dialysis.

13135
Activation of M A P Kinase by a MAP Kinase-Kinase Phosphorylation of MAP Kinase by MAP Kinase Actiuator-Inactive MAP kinase (treated with or without FSBA as above) was incubated with or without MAP kinase activator in 20 mM 8-glycerophosphate (pH 7.5), 20 mM NaF, 2 mM EDTA, 0.2 mM NasVOa, 10 mM MgC12, 0.13 mM [y3*P]ATP (1000 cpm/pmol). Incubations contained 0.01 unit of non-FSBA-treated inactive MAP kinase or the equivalent amount of protein from FSBA-treated inactive MAP kinase and 0.01 unit of MAP kinase activator. Samples were incubated for 10 min at 30 "C, and the reaction was stopped by addition of SDSsample buffer. Samples were separated by 10% SDS-polyacrylamide gel electrophoresis. The gels were dried and subjected to autoradiography.
Other Methods-Protein concentration was determined by the method of Bradford (17) using bovine y-globulin as a standard. Phosphoamino acid analysis was carried out following transfer of proteins to polyvinylidenedifluoride membranes as described previously (11). The antiserum for Western blotting was as described in Ref. 11. Western analyses were carriedout using an ECL kit according to the manufacturer's recommendation at an antiserum dilution of 1/5000. U937 cells were routinely grown in RPMI 1640 medium containing 5% fetal calf serum in 5% CO,. Cells were harvested as described previously (11).

RESULTS
In order to investigate the nature of the MAP kinase activator, initial studies were employed to titrate inactive MAP kinase with the MAP kinase activator. This was essential so that the MAP kinase activator could be employed within the linear range. Linearity was observed in the range of 0.05-0.5 units/ml MAP kinase activator under the standard assay conditions (not shown). Activation of MAP kinase by MAP kinase activator is accompanied by phosphorylation of MAP kinase on tyrosine and threonine residues (see later).
If the MAP kinase activator is a kinase then, like MAP kinase itself, it should be inactivated by FSBA. To determine this, MAP kinase and the MAP kinase activator were treated with FSBA and then assayed for their respective activities. It can be seen from Fig. 1 that FSBA treatment results in complete inactivation of both MAP kinase and MAP kinase activator under the conditions employed. Fig. 2 shows that the inhibition of the MAP kinase activator by FSBA is effectively competed out by a 10-fold excess (1 mM) of ATP. This is consistent with FSBA binding at a specific ATPbinding site as opposed to interaction at other nonspecific sites. Controls showed that the FSBA inhibition of the MAP  (1, 2) or MAP kinase activator (3, 4) were treated without (1,3) or with (2,4) FSBA. Excess FSBA was removed by dialysis (MAP kinase) or rapid filtration through Sephadex G-50 (MAP kinase activator); activities were then assayed as described under "Experimental Procedures." kinase or MAP kinase activator activity was not due to carryover of FSBA into the final assay; inclusion of buffer from a blank FSBA incubation resulted in only a 12% decrease in MAP kinase activity. Similarly there was no significant decrease in MAP kinase activator activity of the vehicle (Me2SO)-treated controls relative to material assayed directly from storage. These results suggest that MAP kinase activator is indeed a kinase.
If the MAP kinase activator is indeed a kinase which directly phosphorylates MAP kinase then inactivation of MAP kinase with FSBA should not affect the phosphorylation induced by the MAP kinase activator. In order to test this, MAP kinase was treated with or without FSBA and then treated with the MAP kinase activator. FSBA treatment resulted in 100% inactivation of MAP kinase activity as shown in Fig. 1. Fig. 3a shows that there is little difference in the extent of phosphorylation of FSBA-treated or vehicletreated MAP kinase when incubated with the MAP kinase activator. Quantitation of the amount of incorporated into each band by Cerenkov counting of the excised bands showed that there was 20% difference between the two; this is probably accounted for by the sequential manipulations required (see "Experimental Procedures"). Phosphoamino acid analysis showed that in both cases the phosphorylation was on tyrosine and threonine residues (Fig. 3b). DISCUSSION FSBA has been used to assess the nature of the MAP kinase activator. FSBA is an ATP analogue which binds to the ATPbinding site of kinases, alkylates the conserved lysine at this site, and thereby irreversibly inactivates the kinase (22). The evidence presented here indicates that the activator is in fact a dual specificity tyrosine/threonine kinase rather than an allosteric stimulator of MAP kinase autophosphorylation. First, the activator itself is completely inhibited by FSBA, and this inhibition is competed out by ATP suggesting that FSBA is binding at a specific ATP-binding site. Second, inactivation of MAP kinase by FSBA does not affect the phosphorylation obtained on incubation with the MAP kinase activator. These results are consistent with the recent report of Posada and Cooper (8) showing that Xenopus MAP kinase is phosphorylated in "trans" on tyrosine and threonine resi- dues by a kinase activity which is not detected by a MAP kinase antibody.
During the last year, a t least 11 dual specificity kinases which phosphorylate serine/threonine and tyrosine residues have been tentatively identified (15). However, for many such kinases it has been unclear as to whether the dual specificity is merely artifactual or physiologically important. MAP kinase-kinase is the first such kinase which has been shown to phosphorylate a physiological substrate on both tyrosine and threonine residues and regulate the function of that substrate.
It would appear then that MAP kinase is at the center of a kinase cascade that involves MAP kinase-kinase, MAP kinase itself, and probably insulin-stimulated protein kinase/p90 ribosomal S6 kinase (ISPK/p9OrYk) (18,23). Activation of this pathway is triggered by protein kinase C-dependent and independent pathways (19)(20)(21). Evidence to date suggests that MAP kinase-kinase is activated by phosphorylation on serine and/or threonine residues (10,16).' The upstream kinase does not appear to be a member of the PKC family'; the nature of the upstream components remains to be elucidated.