Nalley et al. reply

Abstract

Replying to: G. A. Collins, J. R. Lipford, R. J. Deshaies & W. P. Tansey Nature 461, 10.1038/nature08406 (2009)

Proteasome-mediated turnover of some^1,2, but clearly not all^3,4, transcriptional activators is important for their activity. To facilitate the analysis of activator–promoter complex lifetime in vivo, a parameter relevant to this issue, we developed a competition chromatin immunoprecipitation (ChIP) assay in which binding of a native transactivator to its cognate promoters is challenged by a ligand-activated competitor protein with the same DNA-binding specificity. We applied this technique to the yeast Gal4 system⁵ and concluded that under non-inducing conditions (raffinose media) Gal4–promoter complexes exchange rapidly, but under inducing conditions (galactose media) the activator–promoter complexes are long-lived. Collins et al.⁶ report that, surprisingly, the addition of oestradiol to yeast lacking Myc–G4–ER–VP16 increased the amount of DNA co-immunoprecipitated with native Gal4.

Main

This is a control we had not done, but have subsequently repeated and agree that this is the case (S.A.J. wishes to note that he had requested this control and it erroneously was not done). We thank Collins et al.⁶ for pointing out this omission. They go on to show that inducing competitor protein activity with 4-hydroxytamoxifen (4HT) results in a significant loss in the intensity of the ChIP signal owing to native Gal4, but that this ligand does not affect the intensity of these ChIP signals in the absence of competitor. They also show significant association of the competitor protein with the promoter, although with a different time course than Gal4 dissociation. We agree that these data indicate that a large fraction of Gal4–promoter complexes are kinetically labile in vivo under these (4HT-containing media) conditions. It is important to note that this odd effect of steroid is not a general problem in the application of this technology to the measurement of other activator–promoter half-lives⁴.

However, our ChIP data tracking association of the competitor protein do not support the conclusion of a rapidly exchanging Gal4–DNA complex in vivo in the presence of β-oestradiol rather than 4HT. There is no indication that these data are compromised by unanticipated effects of β-oestradiol. Under inducing conditions, much lower levels of association of the competitor protein with GAL promoters were observed when Gal4 was present than in Δgal4 cells when β-oestradiol was used to trigger the competition. These data argue for the presence of a stable, functional Gal4–promoter complex in the presence of galactose and under the particular conditions used in our study⁵. It may be that the stability of Gal4–promoter complexes are somehow affected by steroid receptor ligands, which would explain the different results observed by ourselves and Collins et al.⁶ for the association of the competitor protein in our respective experiments.