FACT is recruited to the +1 nucleosome of transcribed genes and spreads in a Chd1-dependent manner [ChIP-chip]

The histone chaperone FACT occupies transcribed regions where it plays prominent roles in maintaining chromatin integrity and preserving epigenetic information. How it is targeted to transcribed regions, however, remains unclear. Proposed models for how FACT finds its way to transcriptionally active chromatin include docking on the RNA polymerase II (RNAPII) C-terminal domain (CTD), recruitment by elongation factors, recognition of modified histone tails and binding partially disassembled nucleosomes. Here, we systematically tested these and other scenarios in Saccharomyces cerevisiae and found that FACT binds transcribed chromatin, not RNAPII. Through a combination of experimental and mathematical modeling evidence, we propose that FACT recognizes the +1 nucleosome, as it is partially unwrapped by the engaging RNAPII, and spreads to downstream nucleosomes aided by the chromatin remodeler Chd1. Our work clarifies how FACT interacts with genes, suggests a processive mechanism for FACT function, and provides a framework to further dissect the molecular mechanisms of transcription-coupled histone chaperoning. To investigate how FACT is recruited to chromatin in yeast Saccharomyces cerevisiae, we performed chromatin immunoprecipitation (ChIP), followed by high-resolution tiling arrays (ChIP-chip), of FACT subunits (Spt16 and Pob3) and RNAPII (Rpb3) in wild type cells. FACT and RNAPII occupancies were also measured in a temperature-sensitive mutant for RNAPII (rpb1-1) allowing to shutting down transcription after heat-shocking cells for 80 min at 37°C. To determine the contribution of CTD phosphorylation to FACT recruitment, FACT and RNAPII ChIP-chip experiments were performed in various CTD kinase mutants including kin28as, ctk1Δ, cdk8Δ, bur2Δ and the double kin28as/bur2Δ. We also profiled FACT and RNAPII using RNAPII CTD mutants where all serines at positions 2, 5 or 7 were replaced with alanines. In addition, FACT and RNAPII occupancies were measured in PAF1C mutants (paf1Δ, cdc73Δ, rtf1Δ, ctr9Δ, and leo1Δ), in cells where the N-terminal domain of Cet1 is truncated (cet1ΔN), in a temperature-sensitive mutant for the histone chaperone Spt6 (spt6-1004) and in ATP-dependent chromatin remodeler mutants (chd1Δ, isw1Δ, isw2Δ). In order to test if transcription generates a specific histone modification pattern that may be recognized by FACT, we performed FACT and RNAPII ChIP-chip experiments in sas3Δ (the catalytic subunit of NuA3), gcn5Δ, set2Δ, bre1Δ (the E3 ligase for H2B ubiquitylation), ubp8Δ (the main deubiquitylase for H2B) and ubp10Δ (a FACT-dependent H2B deubiquitylase) cells. Finally, FACT, RNAPII and Chd1 (3HA–tagged) occupancies were determined in a Chd1 catalytic dead mutant (chd1-K407R). All ChIP samples were labeled with Cy5 and hybridized against Input DNA or NoTag ChIP samples (labeled with Cy3). The microarrays used were described before (Jeronimo and Robert, 2014). The ChIP-chip data were normalized using the Limma Loess method, and replicates were combined as described previously (Ren et al. 2000, Science 290(5500):2306-9). The data were subjected to one round of smoothing using a Gaussian sliding window with a standard deviation of 100 bp to generate data points in 10-bp intervals (Guillemette et al. 2005, PLoS Biol 3(12):e384). Each experiment was done at least from two independent biological replicates.