DEV007120 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Immunohistochemistry staining of the E16 eye tissues to detect the possible downstream targets of the MEKK1-JNK pathway. Immunohistochemistry was performed on the sections of M1^+/ΔKDJ2^−/− and M1^+/ΔKD J1^−/− fetuses. No obvious differences were observed at the levels of JunD, Fos, C/EBPα, C/EBPβ and phospho-Smad2 in the developing eyelid epithelium (arrowheads) of M1^+/ΔKD J^−/−andM1^+/ΔKD J1^−/−fetuses.

• Supplemental Figure S2 -

  Fig. S2. Relative levels of p-JNK/total JNK and p-c-Jun/total c-Jun in M1^+/ΔKD J2^−/− and M1^+/ΔKD J1^−/− keratinocytes. Following western blot analyses (Fig. 3A), the relative levels of p-JNK/total JNK and p-c-Jun/total c-Jun in the M1^+/ΔKD J2^−/− and M1^+/ΔKD J1^−/− keratinocytes treated by various growth factors were quantified by chemiluminescence imaging analyses and represented graphically. The results are the average of at least three independent experiments.

• Supplemental Figure S3 -

  Fig. S3. JNK1 and JNK2 have subtle structural differences but the same activity in c-Jun phosphorylation. (A) SABLE analysis revealed a putative structural difference between JNK1 and JNK2 in the variable region (blue square), in which an extra β-strand (green arrow) is predicted in JNK2, but not in JNK1. In addition, the adjacent strands are predicted with higher confidence levels, suggesting a more ordered conformation of the loop in JNK2. The amino acid sequence, the predicted secondary structures (with alpha-helices represented as red zigzags and loops as blue lines), the confidence levels for secondary structures represented as red bars, the predicted relative solvent accessibilities (with fully buried residues represented as black boxes) and the confidence for relative solvent accessibility predictions are shown. (B) The JNK isoforms have the same affinity for the interaction with c-Jun. We tested the possibility that JNK1 and JNK2 had different efficiencies towards c-Jun phosphorylation, which might be responsible for the marked reduction of phospho-c-Jun in Mekk1^+/ΔKD Jnk1^−/− cells. GST-c-Jun-conjugated glutathione beads were used to pull down JNKs from lysates of wild-type, M1^+/ΔKD J2^−/− and M1^+/ΔKD J1^−/− MEFs. The pulled down and total input JNKs were detected by western blotting. The pulled down/total JNK ratios were quantified following chemiluminescence imaging analyses and the quantification from two independent experiments are shown graphically. Although, in agreement with previous observations (Sabapathy et al., 2004), c-Jun seemed to interact with JNK2 more efficiently than with JNK1 in wild-type MEFs, it showed the same binding affinity for JNK1 and JNK2 in the Mekk1^+/ΔKD Jnk2^−/− and Mekk1^+/ΔKD Jnk1^−/− MEFs, respectively. Hence, the differences in JNK1 and JNK2 activities are not caused by the JNK isoforms differentially interacting with downstream effector c-Jun. (C) JNK1 and JNK2 have the same activity in c-Jun phosphorylation. In vitro kinase assay using various amounts of active JNK1 or JNK2 proteins as kinases and the purified c-Jun as a substrate. The ratio of phospho-c-Jun/total c-Jun over various amount of phospho-JNK was quantified following chemiluminescence imaging analyses and the data were represented graphically. Active JNK1 and JNK2 have the same efficiency in phosphorylating c-Jun.