Abstract
Purpose: :
Pax-6 is a transcription factor controlling eye and brain development in human, mouse, zebrafish, and Drosophila. Pax-6 is phosphorylated by ERK, p38 and HIPK2 and phosphorylation greatly enhances its transactivity. In our previous study, we found that both 32kd and 46kd Pax-6 are detected in human lens epithelial cells, which form complex with PP -1 and display hyperphosphorylation when PP-1 is knocked down by RNAi, suggesting that PP-1 is the phosphatase to specifically dephosphorylate the two isoforms of Pax-6 (Yan et al., 2007. JBC. 282:13954-13965). However, as far what residues are targeted by PP-1 and the functional regulation of their downstream genes after dephosphorylational modulation remains largely unknown. Therefore, getting the answers to these questions becomes our major focus in the present investigation.
Methods: :
In vitro mutagenesis was used to create various mutants imitating constant phosphorylation and dephosphorylation, gel mobility shifting assay was used to test the binding affinity of both wild type and mutant isoforms of Pax-6. Luciferase reporter gene activity assay driven by the native alphaB-crystallin promoter (-426 to +44) was used to test the transactivity of different forms of Pax-6.
Results: :
We demonstrate that there are at least seven amino acid sites involved in the dephosphorylational regulation by PP-1, including TT303, 304, TS360, 361, S398, T373 and T281. The single mutants of both 32kd and 46kd Pax-6 imitating constant phosphorylation show higher transactivity than the corresponding dephosphorylation mutants. Among all the single mutants, the S398-D shows the highest transactivity while the S398-A displays the lowest. When compared among the single, double, triple and quadruple mutants, the quintuple phosphorylation mutants (5D) of both 32kd and 46kd Pax-6 have the highest transactivity, while the quintuple dephosphorylation mutants (5A) of both isoforms show the lowest, which further confirms that phosphorylation and dephosphorylation act as a significant mechanism regulating Pax-6 functions. In addition, gel mobility shift assay results demonstrate that the 32kd Pax-6 tends to bind the homeodomain-binding oligo (P3), while the 46kd Pax-6 can bind both the conservative PD-binding oligo (P6CON) and P3.
Conclusions: :
These data indicate that both 32kd and 46kd Pax-6 are negatively regulated by PP-1. The 32kd and 46kd Pax-6 may have different downstream target genes.
Keywords: transcription factors • gene/expression • signal transduction