The SAGA core module is critical during Drosophila oogenesis and is broadly recruited to promoters

Abstract

The Spt/Ada-Gcn5 Acetyltransferase (SAGA) coactivator complex has multiple modules with different enzymatic and non-enzymatic functions. How each module contributes to gene expression is not well understood. During Drosophila oogenesis, the enzymatic functions are not equally required, which may indicate that different genes require different enzymatic functions. An analogy for this phenomenon is the handyman principle: while a handyman has many tools, which tool he uses depends on what requires maintenance. Here we analyzed the role of the non-enzymatic core module during Drosophila oogenesis, which interacts with TBP. We show that depletion of SAGA-specific core subunits blocked egg chamber development at earlier stages than depletion of enzymatic subunits. These results, as well as additional genetic analyses, point to an interaction with TBP and suggest a differential role of SAGA modules at different promoter types. However, SAGA subunits co-occupied all promoter types of active genes in ChIP-seq and ChIP-nexus experiments, and the complex was not specifically associated with distinct promoter types in the ovary. The high-resolution genomic binding profiles were congruent with SAGA recruitment by activators upstream of the start site, and retention on chromatin by interactions with modified histones downstream of the start site. Our data illustrate that a distinct genetic requirement for specific components may conceal the fact that the entire complex is physically present and suggests that the biological context defines which module functions are critical.

Fig 1. Different requirements of the SAGA modules during oogenesis.

(A) Cartoon of the SAGA complex. (B-G) Differential contrast images of GLC ovarioles and controls. Loss of the enzymatic functions leads to no or a late defect in oogenesis, because DUB subunit Nonstop is dispensable, whereas the ada2b[1] GLC ovaries display a stage 12 defect. (B) Nonstop GLC ovariole. (C) Ada2b[1] GLC ovariole. (D) wda[11] ovariole. (E) wda[EMS] ovariole. (F) OreR wild type control ovariole. (G) ovoD control ovariole. Scale bar: 50 μM.

Fig 2. The core module of the SAGA complex is required for mid-oogenesis.

(A) Overview of SAGA core subunits. (B) saf6[CDSdel] ovariole. (C) SAF6 RNAi ovariole. (D) WDA RNAi ovariole. (E) TAF10b RNAi ovariole. (F) Spt3 RNAi ovariole. Scale bar: 50 μM.

Fig 3. TFIID subunit depletion blocks oogenesis earlier than SAGA subunit depletion.

(A) TBP RNAi ovary pair (B) TAF4 RNAi ovary pair. (C) TAF1 RNAi ovary pair. (D) OvoD control ovary pair. Scale bar: 50 uM.

Fig 4. Genome-wide occupancy of SAGA subunits in ovary tissue shows that SAGA broadly binds to active promoters.

(A) Normalized ChIP signal in a 1001 bp window centered at the TSS of active genes sorted by WDA levels. (B) Average binding of SAGA subunits and Pol II at all active genes. (C) Active genes were divided into ten quantiles based on CAGE expression levels (decreasing from left to right) and each box plot panel displays the SAGA subunit occupancy per quantile (log2 ratio over input). This shows that SAGA occupancy correlates moderately with transcriptional activity.

Fig 5. SAGA footprint at different promoter types in cells.

(A) Average footprints of SAGA subunits in reads per million (RPM) (left to right TATA, DPE TCT, HK). From top to bottom TBP and TRF2, SAGA subunits WDA, Spt3, SAF6 and Ada2b. (B) SAGA levels are overall similar across promoter types.

Fig 6. SAGA and affinity-tagged TRF2 isoforms do not stably interact in S2 cells.

(A) Reciprocal coimmunoprecipitation after transient transfection with pAcTRF2-S-Flag-HA. Using rat anti-HA to immunoprecipitated TRF2-S-Flag-HA, SAGA did not coimmunoprecipitate. (B) Reciprocal coimmunoprecipitation after transient transfection with pAcTRF2-L-Flag-HA. Using rat anti-HA to immunoprecipitated TRF2-L-Flag-HA, SAGA did not coimmunoprecipitate. (C) Coimmunoprecipitation after transient transfection with pAcTRF2-S-Flag-HA. Using Gcn5 as a bait, affinity-tagged TRF2-S does not coimmunoprecipitate. (D) Coimmunoprecipitation after transient transfection with pAcTRF2-L-Flag-HA. Using Gcn5 as a bait, affinity-tagged TRF2-L does not coimmunoprecipitate. In: unbut U: unbout E: eluate.

Fig 7. Cartoon of SAGA core module functions during oogenesis.

The SAGA complex binds upstream and downstream of the TSS at active genes, without discriminating by promoter type. The upstream peak is likely due to interactions of the core module and Nipped A, which interact with activators. The complex is also present at the +1 nucleosome, where it can read H2bUb, H3K4me3 and H3K9/4ac and is retained. After its recruitment to the +1 nucleosome, the enzymatic DUB and HAT activities of SAGA retain the complex, carryout histone modifications and may help maintain the core promoter region in an open state, while the core module may contribute to deposition of TBP onto the core promoter. We speculate that through its capacity to interact with TBP, the SAGA core module may help to increase the concentration of TBP near promoter chromatin.