Figure 4. ERK5 directly phosphorylates KLF2 at multiple Ser/Thr-Pro motifs.

(A) KLF2 contains a predicted MAP kinase docking motif within its transcriptional activation domain (TAD), and two TPPXSP MAP kinase consensus phosphorylation motifs, one within its transcriptional repression domain (TRD), and one between the TRD and zinc finger (ZnF) domain. (B) Recombinant GST–KLF2 was incubated with [γ-³²P]-ATP and active recombinant His-ERK5 in the presence or absence of 10 µM of the selective ERK5 inhibitor XMD8-92. KLF2 and ERK5 phosphorylation were assessed by autoradiography and protein loading determined by coomassie blue staining. (C) Recombinant GST–KLF2 or GST-ATF2 (19–96) were incubated with [γ-³²P]-ATP and either unphosphorylated (ERK5) or phosphorylated active (ERK5*) recombinant His-ERK5. KLF2, ATF2 and ERK5 phosphorylation was assessed by autoradiography and protein loading determined by coomassie blue staining. (D) Recombinant GST–KLF2 was incubated with [γ-³²P]-ATP and active recombinant His-ERK5 in the presence or absence of 10 µM of the selective ERK5 inhibitor XMD8-92. KLF2 and ERK5 bands were excised from SDS–PAGE and stoichiometry of phosphorylation determined (n = 4). Data are represented as mean ± SEM. Statistical significance was determined by student t-test. (E) Active recombinant ERK5 was incubated with recombinant GST-KLF2 and [γ-³²P]-ATP and KLF2 trypsin digested before HPLC analysis. ³²P phosphorylated KLF2 peptides were identified and the sequence determined by mass spectrometry. (F) Active recombinant ERK5 was incubated with recombinant GST-KLF2 and [γ-³²P]-ATP. ³²P phosphorylated tryptic peptides were analysed by mass-spectrometry and Edman sequencing to identify phosphorylated residues. (G) Erk5/Mapk7^−/− mESCs were transfected with MEK5DD, 3xFLAG-KLF2, and either empty vector, wild-type ERK5 or kinase inactive (D200A) ERK5 (KD). FLAG-KLF2 was isolated by FLAG IP, and FLAG-KLF2 Ser175 phosphorylation and expression assessed by immunoblotting. (H) Normalised FLAG-KLF2 Ser175 phosphorylation levels were determined (n = 5). Data are represented as mean ± SEM. Statistical significance was determined by student t-test.