H3 lysine 4 di- and tri-methylation deposited by cryptic transcription attenuates promoter activation

Abstract

Set1-dependent H3K4 di- and tri-methylation (H3K4me2/3) have been associated with active transcription. Recent data indicate that the H3K4me2/3 also plays a poorly characterized RNA-dependent repressive role. Here, we show that GAL1 promoter is attenuated by the H3K4me2/3 deposited by cryptic transcription. The H3K4me2/3 delay the recruitment of RNA polymerase II (RNAPII) and TBP on GAL1 promoter. Inactivation of RNA decay components revealed the existence of the RNAPII-dependent unstable RNAs, initiating upstream of GAL1 (GAL1ucut). GAL1ucut RNAs are synthesized in glucose and require the Reb1 transcription factor. Consistent with a regulatory function of the cryptic transcription, Reb1 depletion leads to a decrease of H3K4me3 on GAL10-GAL1 locus in glucose and to an acceleration of GAL1 induction. A candidate approach shows that the RPD3 histone deacetylase attenuates GAL1 induction and is tethered at the GAL10-GAL1 locus by H3K4me2/3 upon repression. Strikingly, Set1-dependent Rpd3 recruitment represses also the usage of a hidden promoter within SUC2, suggesting a general function for H3K4me2/3 in promoter fidelity. Our data support a model wherein certain promoters are embedded in a repressive chromatin controlled by cryptic transcription.

Figure 1

Set1-dependent histone H3K4me2/3 attenuate GAL1 activation. Northern blot experiments with total RNA extracted from (A). WT (YAM908), set1Δ (YAM249), set1(463–1080) (YAM912), (B) WT (YAM92), set1Δ (YAM249), spp1Δ (YAM804), sdc1Δ (YAM800) and set2Δ (YAM678) and set1Δset2Δ (YAM623) and (C) with WT (YAM212), H3K4A (YAM216) and H3K36A (YAM215) strains. scR1 RNA is a loading control and GAL1 has been probed with P1 probe (see Figure 2A). Time of induction is indicated in minutes after a shift from glucose (G) to raffinose (R) and galactose (gal) containing media. (D) Quantification of GAL1/scR1 levels at 60 min of induction. GAL1/scR1 levels of WT have been arbitrary set to 1. Error bars represent the standard deviations of at least three independent experiments.

Figure 2

Set1-dependent H3K4me2/3 control GAL1 transcription initiation. (A) Schematic view of the GAL10-GAL1 locus with positions of probes (P1, P2 and P3) and amplicons (A, B, C, D, E and F) for northern blot and real-time PCR, respectively. (B) Set1 and Sdc1 affect RNAPII occupancy on GAL1 gene. Chromatin immunoprecipitation (ChIP) experiments were carried out with an anti-RNAPII antibody in WT (YAM1242), set1Δ (YAM1243) and sdc1Δ (YAM1237) strains after transfer in galactose media (times in minutes). Amplicons correspond to E and F probes. Results are presented as percentages of input normalized with the 3′ end of RPO21 gene. (C) Set1 and Sdc1 regulate TBP-HA occupancy on GAL1 promoter. Same experiment as in (B), but with an anti-HA antibody. Amplicon corresponds to the D probe. Percentages of input were normalized with a tRNA region (tf(GAA)P2 in ChrXVI).

Figure 3

Reb1-dependent cryptic transcription generates GAL1 upstream transcripts mainly destabilized by Xrn1. (A) Xrn1 exoribonuclease destabilizes GAL1ucut RNAs. Northern blot experiments with total RNA extracted from WT (YAM92), xrn1Δ (YAM97) and trf4Δ (YAM456) grown in glucose or galactose (gal) for 3 h. scR1 RNA is a loading control and GAL1ucut, GAL1and GAL10 have been probed with P2 probe (see Figure 2A). The three species of GAL1ucut (A, B and C) are labelled with arrows. (B) Schematic representation of the different RNAs detected at the GAL10-GAL1 locus and determination of the main starting site of the GAL1ucut (+572 from GAL10 stop codon). Classic and putative TATA boxes are represented by white boxes and the Reb1 binding site (Reb1BS) is indicated. (C) GAL1ucut A, B and C are controlled by Reb1. Same as in (A), but with WT (YAM1), xrn1Δ (YAM6), reb1-1 (YAM1591) and xrn1Δreb1-1 (YAM1650), grown at 30°C in glucose and transferred to 37°C during 3 h. Ratios of the different GAL1ucut and RTL ncRNA are indicated (Berretta et al, 2008). (D) GAL1ucut RNA transcription attenuates GAL1 and GAL10 induction. Same as in Figure 1A, but RNA extracted from WT (YAM1) and reb1-1 (YAM1591) strains. Cells were grown at 30°C in glucose, diluted in raffinose containing media for 2 h at 37°C then transferred in galactose containing media at 37°C and RNA analyzed at the indicated time (min). The GAL1 and GAL10 RNAs were detected with the P2 probe.

Figure 4

GAL1 upstream transcripts are not Set1 dependent. (A) The GAL1ucut RNAs do not control GAL1 induction. Same as in Figure 1A, but with WT (YAM92), xrn1Δ (YAM97) and xrn1Δtrf4Δ (YAM458). GAL1ucut are labelled with an arrow and the GAL1 read through RNA (GAL1RT) with an asterisk. (B) Quantification of GAL1/scR1 levels at 60 min of induction. Same as in Figure 1C. (C) GAL1ucut RNAs are synthesised upon repressive conditions, destabilized by Xrn1 but not controlled by Set1. Reverse transcriptions were performed with the GAL1UAS primer P3 (position in Figure 2A) and amplified by PCR with amplicon D of GAL1 UAS region. Data were normalized with scR1 RNA. WT (YAM92), xrn1Δ (YAM97), set1Δ (YAM249) and xrn1Δset1Δ (YAM448) strains were grown in glucose, transferred during 2 h in raffinose and shifted to galactose containing media. The samples were extracted at the indicated time (min).

Figure 5

RNAPII- and Reb1-dependent cryptic transcription control residual H3K4me3 and H3K4me2 at the GAL10-GAL1 locus. (A) Schematic view of the GAL10-GAL1 locus with positions of amplicons for real-time PCR. (B) Cryptic transcription deposits H3K4me3 on GAL10-GAL1 locus in repressive condition. Chromatin immunoprecipitation (ChIP) experiments were performed with an anti-H3K4me3 antibody in WT (YAM15) and rpb1-1 (YAM268) and set1Δ (YAM1494) strains grown in glucose-containing media at 28°C, then shifted to 37°C in 1 h. H3K4me3 signals were normalized with H3 signals performed with an anti-H3 antibody on the same chromatin. Amplicons correspond to A, B, C, D, E and F shown in (A). Results are presented as relative levels of H3K4me3/H3 to those measured in set1Δ strain (WT/set1Δ and rpb1-1/set1Δ). (C) Cryptic transcription deposits H3K4me2 on GAL10-GAL1 locus in repressive condition. Same as in (B), but with an anti-H3K4me2 antibody. (D) Reb1 controls H3K4me3 on the GAL10-GAL1 locus. Same as in (B), but with WT (YAM1) and reb1-1 (YAM1591) strains. Cells were grown in glucose containing media at 28°C then shifted to 37°C in 3 h.

Figure 6

RPD3S complex attenuates GAL1 induction. (A) RPD3S attenuates the GAL1 induction. Same experiment as in Figure 1A, but with WT (YAM1), rpd3Δ (YAM4) and eaf3Δ (YAM5) and WT (YAM1483), rco1Δ (YAM1489) and sds3Δ (YAM1488) strains. (B) Rco1-plant homeobox domain (PHD) is required to attenuate GAL1 induction. Same as in Figure 1A, but with WT (YAM118) and rco1PHDΔ (YAM1465). (C) Quantification of GAL1/scR1 at 60 min of induction for strains used in (A) and (B) but, also for pho23Δ (YAM1370), bye1Δ (YAM1371), yng1Δ (YAM1369) and set3Δ (YAM1302) strains. (D) H3K4me2/3 control Rpd3 occupancy at the GAL10-GAL1 locus upon repressive conditions. Chromatin immunoprecipitant (ChIP) experiments were performed as in (A), but with an anti-Myc antibody in WT (YAM1477), H3K4A (YAM1481) strains grown in glucose. Results are presented as percentage of input normalized with the promoter of INO1.

Figure 7

H3K4me2/3 recruit Rpd3 to control SUC2 promoter fidelity. (A) Schematic view of the SUC2 transcripts with positions of the probes for northern blot and qPCR. (B) Set1-dependent H3K4me2/3 repress a SUC2 spurious transcript (suc2S). Northern blot with total RNAs extracted from WT (YAM1 and 92), rpd3Δ (YAM4), set1Δ (YAM249) spp1Δ (YAM804) and sdc1Δ (YAM800) strains. WT cells were grown in glucose (G) and transferred in raffinose (R) containing media. Mutants were grown in glucose containing media (G) only. scR1 is a loading control and SUC2 has been probed according to Figure 7A. (C) Rpd3-Myc is targeted to SUC2, IMD2 and PHO5 promoters and is dependent on H3K4me. Chromatin immunoprecipitant (ChIP) experiments were performed with an anti-Myc antibody in WT (YAM1477) and H3K4A (YAM1481) strains in glucose. Amplicons correspond to SUC2, PHO5, IMD2, IME2 and INO1 promoters. Results are presented as percentage of input normalized with the promoter of INO1. (D) Model of promoter attenuation by H3K4me2/3. Cryptic transcription activates the deposition of H3K4me2/3 on promoter proximal regions and triggers the recruitment of the RPD3S complex inhibiting the pre-initiation complex (PIC) formation and promoter activity.