Grøntved, Lars4; Bugge, Anne K.4; Roeder, Robert G.3; Mandrup, Susanne4
1 Faculty of Science, SDU2 Department of Biochemistry and Molecular Biology, Faculty of Science, SDU3 Laboratory of Biochemistry and Molecular Biology, The Rockefeller University, 1230 York Avenue, New York, NY100214 Department of Biochemistry and Molecular Biology, Faculty of Science, SDU
The mediator complex serves as a transcriptional co-activator complex by acting as a bridge between promoter-bound transcription factors and the preinitiation complex. Genetic and biochemical studies indicate that nuclear receptors recruit the mediator complex through direct interaction with the LXXLL motifs of the mediator subunit MED1/TRAP220/DRIP205/PBP. We and others have shown that the peroxisome proliferator activated receptor (PPAR) subtypes have different biochemical properties and highly distinct abilities to activate target genes. As biochemical and genetic data suggest that the different subunits of the mediator complex are required in diverse transcriptional pathways, we wanted to investigate whether composition of the mediator complex was different depending on the PPAR subtype. We therefore implied a strategy to knock down different mediator subunits and subsequently analyzed the ability of PPARgamma and PPARalpha to activate target genes. To our surprise we found that knock down of MED1 neither affected PPARalpha nor PPARgamma-mediated activation of target genes. Recently it was reported that PPARgamma is able activate target genes in proliferating but not in confluent MEFs derived from TRAP220 KO mice. Interestingly, rescue experiments in confluent TRAP220 KO MEFs with different versions of MED1 indicate that the LXXLL motif is not necessary for PPARgamma mediated gene activation (Ge et al, MCB published online ahead of print 2007). By analogy, we show here that MED1 is dispensable for PPARalpha transcriptional activity in proliferating but is necessary in confluent AML-12 cells and TRAP220 KO MEFs. Collectively this indicates that the PPARs might have adopted an alternative mediator recruitment mechanism that is dispensable of direct interaction with MED1 on endogenous target promoters. Using a truncated version of PPARalpha that lacks the AF1 domain we show that alternative mediator recruitment is likely to be through the AF2 domain or RXR. A functional screen for LXXLL motifs present in the different mediator subunits may reveal potential candidates.