Nuclear export restricts Gdown1 to a mitotic function

Approximately half of purified mammalian RNA polymerase II (Pol II) is associated with a tightly interacting sub-stoichiometric subunit called Gdown1, encoded by the POLR2M gene. Previous in vitro and structural studies have established that Gdown1 potently inhibits transcription initiation through competitive interactions with general transcription factors and blocks the Pol II termination activity of transcription termination factor 2 (TTF2). However, the biological functions of Gdown1 remain poorly understood. Here, we utilized genetic, microscopic, and multi-omics approaches to functionally characterize Gdown1 in multiple human cell lines. Remarkably, acute depletion of Gdown1 led to minimal direct effects on gene expression and transcription. We show that Gdown1 resides predominantly in the cytoplasm of interphase cells, shuttles between the cytoplasm and nucleus, and is directly regulated by nuclear export. Gdown1 enters the nucleus at the onset of mitosis prior to complete breakdown of the nuclear envelope, whereupon association with Pol II becomes possible. Consistently, genetic ablation of Gdown1 is associated with partial de-repression of mitotic transcription, and Gdown1 KO cells present with a growth defect and evidence of aberrant mitoses coupled to p53 pathway activation. Evidence is presented demonstrating that Gdown1 association with Pol II elongation complexes in vitro modulates the combined functions of purified productive elongation factors PAF1C, RTF1, SPT6, DSIF, and P-TEFb. Collectively, our findings support a model wherein the Pol II-regulatory function of Gdown1 occurs during mitosis and is required for genome integrity. Overall design: Multiple different HeLa cells enabling the study of loss of function of Gdown1 were developed as described in the methods. To address changes in gene expression linked to Gdown1 loss, total RNA parental and Gdown1 KO #1 and 2 HeLa cell lines was isolated and prepared for sequencing as described in the paper. Samples were prepared in biological triplicate. To assay the effects of Gdown1 loss on nascent transcription, nuclei were isolated from asynchronous and mitotic Parental HeLa cells and Gdown1 KO #3 HeLa cells, and PRO-Seq was performed as described in the methods. PRO-Seq libraries were prepared in biological duplicate. To assay the direct effects of Gdown1 depletion on transcription, nuclei were isolated from Parental HeLa cells and HA-FKBP-Gdown1 HeLa cells treated for 4 h with DMSO or 400 nM dTAGV-1. Libraries were prepared in biological duplicate. To assay Gdown1 occupancy of chromatin, ChIP-Seq for Gdown1 was carried out as described in the text from HA-FKBP-Gdown1 HeLa cells, using an HA antibody. As a background control, the HA antibody was used for ChIP from unmodified Parental HeLa cells. ChIP-Seq was performed in biological duplicate. Gdown1 ChIP signals were compared to a single Pol II ChIP-Seq dataset, performed as described in the text with an antibody identifying POLR2A N-terminus. ChIP-Seq datasets for H3K4me3 and H3K4me1, which were utilized for the desgination of enhancers, were carried out as described. Previously published HeLa NasCap data were utilized in this study, and are accessible at NCBI GEO ####, SRA ####. Processed data are provided as normalized bw files and scaling factors are accessible in the supplemental data file.