Exportin-1 functions as an adaptor for transcription factor-mediated docking of chromatin at the nuclear pore complex

Description

Nuclear pore proteins (Nups) physically interact with hundreds of chromosomal sites, impacting transcription. In yeast, transcription factors mediate interactions between Nups and enhancers and promoters. To define the molecular basis of this mechanism, we exploited a separation-of-function mutation in the Gcn4 transcription factor that blocks its interaction with the nuclear pore complex (NPC). This mutation reduces the interaction of Gcn4 with the highly conserved nuclear export factor Crm1/Xpo1. Crm1 and Nups co-occupy enhancers and Crm1 inhibition blocks interaction of the nuclear pore protein Nup2 with the genome. In vivo, Crm1 interacts stably with the NPC. In vitro, Crm1 binds both Gcn4 and Nup2 directly. Importantly, the interaction between Crm1 and Gcn4 requires neither Ran-GTP nor the nuclear export sequence binding site. Finally, Crm1 and Ran-GTP stimulate DNA binding by Gcn4, suggesting that allosteric coupling between Crm1-Ran-GTP binding and DNA binding facilitates docking of transcription factor-bound enhancers at the NPC. This dataset contains processed aligned reads (.bam files) from four main questions we addressed in our paper: 1) Where do Crm1, Nups, and RNA Pol II bind in the genome? [ChEC-seq2] 2) What is the relationship between Crm1 and Nup2? Are they interdependent or does the binding of one factor depend on the other? [ChEC-seq2] 3) Does loss of Crm1 or Nups (Nup1/Nup2) affect nascent transcription? If so, how? [SLAM-seq] 4) How does Leptomycin B (LMB, inhibitor of Crm1) affect binding of Crm1 and transcription factor Gcn4 to the genome? [ChEC-seq2] ChEC-seq2 data was obtained and processed as described in VanBelzen et al. 2024, with reads trimmed with Trimmomatic and mapped to the Saccharomyces cerevisiae genome (sacCer3) using Bowtie2. For peak calling analyses, we processed the bam files by trimming them to the first base pair and called peaks using DoubleChEC (VanBelzen et al. 2024). SLAM-seq was performed as described in Herzog et al. 2017 and Alalam et al., 2022, using 0.2 mM 4-thiouracil (4sU) for 6 min at 30°C prior to harvesting, and processed using SLAM-DUNK (Herzog et al. 2017), which generated the read counts (.tsv files). Differential expression was analyzed using DESeq2 (Love et al., 2014). All sample information (BioProject, Biosample, SRA, GEO) is included in the Sample information spreadsheet.

Steps to reproduce

Yeast cells were grown in synthetic complete medium containing glucose (SDC) before treatments with LMB, 3-IAA, or estradiol + 5-Ph-IAA. Crm1 was inhibited by treating LMB-sensitive strains (which have the T539C mutation) with 185 nM LMB (Cayman Chemical Company, cat#10004976) for 30 min at 30°C or degraded by treating grAID strains with 1 µM estradiol (E2; Sigma Aldrich, cat#E8875) for 30 minutes at 30°C followed by 1 µM of 5-Ph-IAA (Fisher Scientific, cat#NC1957890) for 2 hours at 30°C. Nups were degraded by treating Nup1-AID and Nup2-AID strains with 500 µM Indole-3-Acetic Acid (3-IAA; AID Gold Biotechnology, cat#I-110-100) for 1h at 30°C. All treatments were performed prior to harvesting. ChEC-seq2 and its subsequent data analysis in R were performed as described in VanBelzen et al. 2024. MNase cleavage of various proteins of interest were carried out in SDC + digitonin (BioSynth, cat# XD175329) + 2mM calcium, with digestion times optimized for each fusion protein of interest. All experiments had at least three biological replicates. Libraries were made using the fragments (>5kb) from the partial MNase digestion that were purified, end-repaired and ligated to a P5-compatible Y-adapter, tagmented with a Tn5 loaded with adapters, and amplified for 15 PCR cycles using the KAPA library amplification kit with Nextera XT primers. The indexed libraries were then pooled and sequenced using Illumina HiSeq 4000 or NovaSeq X Plus for 50 bp single-end reads. Sequencing reads were assessed and processed using FastQC, Trimmomatic, Bowtie2, and Samtools as described in VanBelzen et al. 2024; mapped reads were trimmed to the first base pair adjacent to the cleaved DNA and peaks called using DoubleChEC. SLAM-seq was performed as described in Herzog et al. 2017 and Alalam et al. 2022. Yeast cultures were switched into SDC-Ura supplemented with 0.2 mM 4- thiouracil (4sU; Fisher Scientific, cat#AAH6191903) after drug treatments for 6 min at 30°C and then pelleted and snap frozen. Total RNA was extracted using Phenol:Chloroform:IAA (25:24:1) and precipitated. 5 ug total RNA was treated with 10 mM iodoacetamide (Sigma-Aldrich, cat# I1149), for 15 min at 50°C. Sequencing libraries were made using the QuantSeq 3’mRNA-seq Library Prep Kit for Illumina (FWD; Lexogen) following manufacturer's recommendations and sequenced on Illumina HiSeq 4000 or Element AVITI for 50 bp single-end reads. All experiments had at least three biological replicates. SLAM-seq data was analyzed by SLAM-DUNK v0.4.3 (Herzog et al. 2017) with default parameters and a BED-file containing coordinates for 3’ UTRs generated using standard genomic coordinates for genes with 325bp 3’UTRs. For reads with more than one T>C conversion (in nascent RNA reads), DESeq2 v1.46.0 (Love et al., 2014) was used to identify mRNAs and nascent transcripts that were differentially expressed between treatments. For all gel and microscopy images, please refer to figure legends and methods in our publication.

Institutions

Categories

Funding

Related Links

Licence