SAGA and a novel Drosophila export complex anchor efficient transcription and mRNA export to NPC

Abstract

SAGA/TFTC-type multiprotein complexes play important roles in the regulation of transcription. We have investigated the importance of the nuclear positioning of a gene, its transcription and the consequent export of the nascent mRNA. We show that E(y)2 is a subunit of the SAGA/TFTC-type histone acetyl transferase complex in Drosophila and that E(y)2 concentrates at the nuclear periphery. We demonstrate an interaction between E(y)2 and the nuclear pore complex (NPC) and show that SAGA/TFTC also contacts the NPC at the nuclear periphery. E(y)2 forms also a complex with X-linked male sterile 2 (Xmas-2) to regulate mRNA transport both in normal conditions and after heat shock. Importantly, E(y)2 and Xmas-2 knockdown decreases the contact between the heat-shock protein 70 (hsp70) gene loci and the nuclear envelope before and after activation and interferes with transcription. Thus, E(y)2 and Xmas-2 together with SAGA/TFTC function in the anchoring of a subset of transcription sites to the NPCs to achieve efficient transcription and mRNA export.

Figure 1

E(y)2 is a bona fide subunit of the Drosophila SAGA/TFTC complex. (A) The Drosophila embryo nuclear extract was immunoprecipitated with antibodies raised against GCN5, E(y)2 or a rabbit preimmune serum (PI). Input nuclear fraction (7.5%) (Input), and protein A-Sepharose-antibody-bound proteins (15%) (IP), were resolved by SDS–PAGE. Blots were revealed using antibodies raised against TRRAP, GCN5, ADA2b, TAF10 and E(y)2. (B) E(y)2 colocalizes with GCN5 at many sites on polytene chromosomes of Drosophila. Whole chromosomes, enlarged fragments as well as the merged images are shown.

Figure 2

E(y)2 is localized to nuclear periphery, is in contact with the NPC and is also present in dots in the nucleus. (A) The immunostaining of Drosophila S2 cells with antibodies against E(y)2, GCN5, ADA2b or TAF1 (in red), an antibody against lamin (in green) (magnification, × 1000). (B) Immunofluorescent staining of Drosophila S2 cells with antibodies against E(y)2, GCN5, ADA2b or TAF1 (in red), an antibody against NPC (Mab414) (in green) (magnification, × 1000). (C) Drosophila embryo nuclear extract was immunoprecipitated with antibodies raised against E(y)2, GCN5 or with preimmune serum (PI). The input nuclear fraction (7.5%) (Input) and the protein A-Sepharose-antibody-bound proteins (15%) (IP) were resolved by SDS–PAGE. Blots were revealed with Mab414 with identifies nucleoporin p62 on western blots. (D) Gallery of electron micrographs showing ultrathin cryo sections of Drosophila S2 cells labelled with purified E(y)2-specific antibody and revealed with Protein A gold. Gold particles, which can be distinguished as small black dots, are preferentially located in the vicinity of the nuclear membrane (nm) and appear often associated with the NPC, shown by white arrow heads. The nuclear and cytoplasmic sides in the cells are labelled (n) and (c), respectively. Scale bar: 100 nm.

Figure 3

E(y)2 interacts with Xmas-2 and both proteins colocalize with NPC on the nuclear periphery. (A) The results of interaction of E(y)2 fused to the C-terminus of LexA with Xmas-2 (amino acids 755–1370) fused GAL4 activation domain (AD) in yeast two-hybrid assay. The activation of HIS3 and LacZ reporter genes were measured. β-galactosidase activity (U) was determined using the following formula: U=1000 × OD₅₇₈/(t × 0.5 × OD₆₀₀), where t is the incubation time (in min). (B) Endogenous E(y)2 interacts with Xmas-2 in vivo. The Drosophila embryo nuclear extract was immunoprecipitated with antibodies raised against E(y)2 or Xmas-2, or preimmune serum (PI). A 7.5% portion of the input nuclear fraction (Input), 7.5% of the supernatant of the IPs (SN) and 20% of the protein A-Sepharose-bound proteins (IP), were resolved by SDS–PAGE. Blots were revealed using antibodies raised against E(y)2 and Xmas-2. (C) E(y)2 colocalizes with Xmas-2 at many sites on polytene chromosomes. Whole chromosomes, enlarged fragments as well as the merged images are shown. (D) Drosophila embryo nuclear extract was immunoprecipitated with polyclonal antibodies against Xmas-2, or with preimmune serum (PI). Input nuclear fraction (7.5%) (Input) and protein A-Sepharose-antibody-bound proteins (15%) (IP) were resolved by SDS–PAGE. Blots were revealed with Mab414. (E) The co-immunostaining of Drosophila S2 cells with antibodies raised against Xmas-2, E(y)2, lamin and Mab414 (magnification, × 1000).

Figure 4

E(y)2 and Xmas-2 are required for mRNA export from the nucleus. (A) The efficiency of E(y)2 or Xmas-2 knockdown was tested by western blot analysis. E(y)2 and Xmas-2 protein expression present in wild-type S2 cells (0) or in S2 cells after 5 days of RNAi treatment (5) were analyzed. Tubulin was used as a loading control. (B) E(y)2 and Xmas-2 are required for poly(A)+ RNA export from the nucleus. RNAi was performed using either dsRNA corresponding to a fragment of pSK II vector as a control or the E(y)2 or the Xmas-2 cDNAs. RNA FISH was carried out using a Cy3-labelled oligo(dT) probe to identify poly(A)+ RNA. Nuclear envelope is stained with lamin. Representative examples of cells are shown (magnification, × 1000).

Figure 5

The knockdown of E(y)2 or Xmas-2 expression in S2 cells inhibits the transport of individual transcripts. (A) The distribution of hsp70 transcript was detected by RNA FISH in control (pSK II), E(y)2 or Xmas-2 RNAi-treated S2 cells. The Cy3-labelled probe corresponding to hsp70 fragment was used to identify hsp70 transcripts (red). Nuclei are stained with DAPI. Representative examples of cells are shown (magnification, × 1000). (B) Quantification of the levels of hsp70, actin and trf2 transcripts in the nuclei of control (pSK), E(y)2 or Xmas-2 RNAi-treated S2 cells without HS (NHS) or after HS. Nuclei were isolated, RNA prepared and the individual transcript levels were measured by RT–qPCR using oligonucleotides amplifying the indicated transcripts. In each experiment, the level of transcript present in the control RNAi was taken as one (black bars) and the changes due to the E(y)2 RNAi (light gray bars) or the Xmas-2 RNAi (dark gray bars) are represented as fold changes compared to the control. The error bars represent the variations between three independent experiments.

Figure 6

E(y)2 and Xmas-2 participate in the anchoring of the hsp70 loci to the nuclear envelope and in the regulation of transcription of the hsp70 gene. (A) Localization of the hsp70 (panels a and b) or the trf2 (panels c and d) loci in wild-type S2 cells and after E(y)2, Xmas-2 or control RNAi before (panels a and c) and after (panels b and d) HS was revealed by DNA FISH (red). Nuclei are stained with DAPI (blue). The six copies of the hsp70 genes present at different chromosomal locations (87A and 87B) are grouped as detected by DNA FISH (magnification, × 1000). (B) The quantification of the presence of the hsp70 and the trf2 loci at the nuclear periphery in wild-type S2 and RNAi-treated cells as indicated in (A) before and after HS. For each cell, the confocal single-plane image was used to determine the position of the signal relative to the nuclear envelope (see Materials and methods). The nuclear space was divided into three concentric zones of equal surface as described by Taddei et al (2006). Bar graphs represent the percent of spots in the zone closest to the nuclear envelope without HS (black bars) and after HS induction (gray bars). Each bar represents the mean value from three different RNAi experiments±s.e.m. (C) The level of the hsp70 transcript was measured by RT–PCR in S2 cells after control RNAi (black bars) and after RNAi of E(y)2 (gray bars) and Xmas-2 (empty bars). The values of the hsp70 transcript without HS are indicated above the bars (the standard deviation is less than 10%). Expression levels in total RNA preparations were normalized against rRNA. The level of hsp70 transcript in wild-type cells without HS was set to one. (D) Northern blot analysis of individual transcripts after RNAi treatment. Total RNA was prepared from control, E(y)2 or Xmas-2 RNAi-treated S2 cells. Probes specific for individual genes (as indicated) were used for hybridization. A 10 μg portion of total RNA was loaded per lane. The 18s rRNA was used as a loading control (data not shown).