Detection of ERK1/2 activities using affinity reagents.

Abstract

The three major mammalian mitogen-activated protein kinase (MAPK) subfamilies include the extracellular signal-regulated kinases (ERK1/2), the c-Jun N-terminal kinases (JNKs), and the p38 kinases. These kinases are differentially activated in response to extracellular stimuli and show differential specificity toward their substrates, which is provided, in part, by the existence on the different substrates of specific kinase docking sites (reviewed in ref. ). Specific substrates of JNKs and p38s include the transcription factors c-Jun and ATF2, respectively, and GST-c-Jun and GST-ATF2 fusion proteins are widely used as suitable substrates to measure in vitro the kinase activity of these enzymes (, , , ). On the other hand, although a wide number of cytosolic, membrane-bound, and nuclear proteins are substrates of ERK1/2 (reviewed in refs. and ), measurement of ERK1/2 kinase activity in vitro is routinely performed using the nonphysiologic substrate myelin basic protein (MBP) (,). ERK1/2 and p38α bind to the protein tyrosine phosphatase PTP-SL through a kinase interaction motif (KIM) located outside of the protein tyrosine phosphatase (PTP) catalytic domain; on binding, PTP-SL is phosphorylated by these MAPKs, mainly at the Thr253 residue (,). Phosphorylation by active ERK1/2 and p38α of a GST-PTP-SL fusion protein containing the KIM and the MAPKs phosphorylation site (GST-PTP-SL 147–288) is illustrated in Fig. 1, in comparison with the phosphorylation of MBP. Open image in new window Fig. 1. Phosphorylation of GST-PTP-SL 147–288 and MBP by ERK1/2 and p38α. (A) Rat 1 cells were plated at semiconfluency and left for 6 h in the absence of fetal calf serum (FCS). Then, cells were stimulated with 10% FCS for 10 min (for ERK1/2; lanes 1 and 2) or with 0.5 M sorbitol for 20 min (for p38α, lanes 3 and 4) and lysed in lysis buffer. For each point, ERK1/2 and p38α were immunoprecipitated from cell lysates (0.75 mg of protein for ERK1/2, and 2.25 mg of protein for p38α) using specific anti-ERK1/2 or anti-p38α antibodies and protein A-Sepharose and subjected to immune complex in vitro kinase assays in the presence of equimolar amounts of GST-PTP-SL 147–288 (2 βg; lanes 1 and 3) or MBP (0.85 μg; lanes 2 and 4). (B) COS-7 cells were transfected with pCDNA3-HA-ERK2 or pECE-HA-MAPK (HA-p38α) in six-well plates using the DEAE-dextran method (1 μg of DNA/well). Cells were activated with 50 ng/mL of EGF for 5 min (for HA-ERK2; lanes 1 and 2) or with 0.5 M sorbitol for 20 min (for HA-p38α; lanes 3 and 4) and lysed. For each point, HA-ERK2 and HA-p38α were immunoprecipitated from cell lysates (from one well from a six-well plate for HA-ERK2, and from six wells from a six-well plate for HA-p38α) with the 12CA5 anti-HA monoclonal antibodies and protein A-Sepharose and subjected to in vitro kinase assays as in (A). For ERK1/2 and HA-ERK2, 2 μCi of [γ-32P]ATP/sample were used. For p38α and HA-p38α, 6 μCi of [γ-32P]ATP/sample were used. Note that autophosphorylation of HA-p38α is observed in these assays (asterisk with an arrow in [B], lane 4), which could mask the phosphorylation of GST-PTP-SL 147–288. No autophosphorylation of HA-ERK2 is observed. Autoradiographs of representative experiments are shown. Since the in vitro specific activity of ERK1/2 is much higher than that of p38α, different amounts of cell lysates and [γ-32P]ATP (see above), and different exposure times of the autoradiographs, were used for the assays with these two MAPKs. The exposure times were as follows: For (A), lanes 1 and 2 = 30 min; lanes 3 and 4=6 h. For (B), lanes 1 and 2 = 6 h; lanes 3 and 4 = 16 h. The relative expression of endogenous ERK1/2 and p38α in Rat 1 cells, and of recombinant HA-ERK2 and HA-p38α in COS-7 cells, was similar (not shown). All samples were analyzed by 15% SDS-PAGE under reducing conditions.