Lysates have been immunoprecipitated with M2 and immunoblotted with anti-His. Enter was immunoblotted with possibly M2 or anti-GAPDH antibodies. C-D) Cells have been incubated with MG132 for 2 h prior to lysis and NEM included to lysis, 
IP and WB buffers. The knowledge offered correspond to one particular agent experiment of at minimum 6 conducted (n ! six) (B) and to a single consultant experiment of at least two carried out (n ! 2) (C, D). E) Cells had been co-transfected with pGL3-S1Skp1-Luc reporter vector, both WT or the K308 construct collectively with pSG5-His-SUMO-1 or pcDNA-SENP2-SV5. Relative luciferase buy 194785-18-7 actions ended up also assayed in Ubc9 invalidated cells. HEK-293 cells have been reverse transfected for 24 h making use of siRNAs focusing on Ubc9 (siUbc9) or a scrambled management sequence (siCtrl) ahead of becoming transfected as explained above to assess the relative transcriptional activities of Lf and the K308 mutant. Relative luciferase actions are expressed as explained in Resources and Methods (n!5 p < 0.05 (), p < 0.01 ())expression of SUMO-1 justified the conclusion drawn here that the SUMOylation of Lf is part of a regulatory event that governs its activity. The small amount of Lf-SUMO forms present in cells could not account for the 2.5 fold increment observed in the transcriptional activity induced by the M5S mutant compared to WT. This has been already described for numerous transcription factors and suggests that SUMOylation is required to initiate transcriptional repression but not to maintain it [19,48]. We then compared the activity of mutants in which only one SUMOylation site was preserved to that of M5S in order to evaluate the impact of adding only one regulatory site at a time. The K391 mutant, which is poorly modified, showed transcriptional activity nearly comparable to that of M5S (Fig 4A) suggesting that the presence of SUMO on this site does not crucially regulate Lf transcriptional activity. 18434589In contrast, the transcriptional activities of the K308, K361 and K379 mutants were strongly inhibited, by 10-fold for K308 and by nearly 6 fold for the other sites compared to M5S, and by 4 fold for K308 and by around 2.5 fold for the other two sites compared to WT, suggesting that these three sites are important for regulation. K361, which is poorly SUMOylated, is nevertheless strongly involved in the repression process. The transcriptional activity of the K13 mutant also Fig 5. A SUMOylation/acetylation switch at K13 controls Lf transcriptional activity. (A) K13 is the main acetylation site. Cells were co-transfected by WT, the mutant constructs or the null vector and then lysed 24 h later. Lysates were immunoprecipitated with anti-acetyllysine antibodies and immunoblotted with M2. Input was immunoblotted with either M2 or anti-GAPDH antibodies and used as loading control (n = 3). B) Relative transcriptional activity of K13 and K379 mutants compared to WT. Cells were co-transfected with pGL3-S1Skp1-Luc reporter vector and WT, K13 or K379. His-SUMO-1 expression vector and/or the deacetylase inhibitor Trichostatin A (TSA, 15 ng/mL) were used to modulate the acetylation/SUMOylation ratio.
