Cytoplasmic mRNA decay represses RNA polymerase II transcription during early apoptosis

Figures

• 

  Figure 1.. mRNA decay during early apoptosis is accompanied by reduced synthesis of RNAPII transcripts. (A) Schematic representation of how the extrinsic apoptotic pathway accelerates mRNA decay. (B) Western blot of HCT116 lysates showing the depletion of full-length caspase 8 (CASP8) and caspase 3 (CASP3) over a time course of 100 ng/μL TRAIL treatment. Vinculin (VCL) serves as a loading control. Blot representative of those from four biological replicates. (C, D) RT-qPCR quantification of total (C) and nascent 4sU pulse-labeled (D) RNA at the indicated times post TRAIL treatment of HCT116 cells (n = 4). Also see Figure 1—figure supplement 1. No biotin control quantifies RNA not conjugated to biotin that is pulled down with strepdavidin selection beads. Fold changes were calculated from Ct values normalized to 18S rRNA in reference to mock treated cells. Graphs display mean ± SEM with individual biological replicates represented as dots. Statistically significant deviation from a null hypothesis of 1 was determined using one sample t test; *p<0.05, **p<0.01, ***p<0.001 (p values provided in Supplementary file 1A for all figures).

• 

  Figure 1—figure supplement 1.. mRNA decay during early apoptosis is accompanied by reduced synthesis of RNAPII transcripts. (A) RT-qPCR quantification of total RNA at the indicated times post 100 ng/μL TRAIL treatment of HCT116 cells (n = 4). (B) Stained agarose gel depicting a 200 nt RT-PCR product of 4sU-labeled 18S rRNA, extracted and isolated from an equal number of cells treated with 3 hr vehicle (‘mock’) or 100 ng/μL TRAIL. Gel representative of that from three biological replicates. (C) RT-qPCR quantification of 4sU-labeled RNA at the indicated times post 100 ng/μL TRAIL treatment of HCT116 cells (n = 4). Fold changes were calculated from Ct values normalized to 18S rRNA in reference to mock treated cells. Graphs display mean ± SEM with individual biological replicates represented as dots. Statistically significant deviation from a null hypothesis of 1 was determined using one sample t test; *p<0.05, **p<0.01, ***p<0.001.

• 

  Figure 2.. RNAPII transcription is globally repressed during early apoptosis. (A, B) RT-qPCR measurements of total (A) and nascent 4sU-labeled (B) RNA fold changes after 2 hr TRAIL treatment of HCT116 cells, including a 1 hr pre-treatment with either 40 μM zVAD or an equal volume of DMSO (‘mock’). Also see Figure 2—figure supplement 1A. RNA fold change values were calculated in reference to 18S rRNA. Bar graphs display mean ± SEM with individual biological replicates (n = 4) represented as dots. Statistically significant deviation from a null hypothesis of 1 was determined using one sample t test and indicated with asterisks directly above bars, while student’s t tests were performed to compare mean fold change values for mock inhibitor or scramble treated cells to those treated with inhibitor or a targeting siRNA and indicated with brackets. *p<0.05, **p<0.00.1, ***p<0.001. (C, D) rRNA-depleted cDNA sequencing libraries were reverse transcribed from 4sU-labeled RNA isolated from cells under the conditions described in (A, B). Transcripts that aligned to genes in the human genome are graphed with differential log2 fold change expression values (log2FC) on the y axis and fragments per…

• 

  Figure 2—figure supplement 1.. RNAPII transcription is globally repressed during early apoptosis. (A) Western blot of HCT116 lysates showing cleavage (or lack thereof) of the caspase 8 (CASP8) targets BID and caspase 3 (CASP3), as well as the CASP3 target PARP1 upon 2 hr 100 ng/μL TRAIL treatment in the presence or absence of 40 μM zVAD. Blot representative of that from three biological replicates. (B) Log2 fold change (log2FC) in abundance of the ACTB transcript upon 2 hr TRAIL treatment in the presence of either zVAD or DMSO vehicle (‘mock’). Fold changes from the same 4sU-labeled RNA samples were quantified by next-generation sequencing and differential expression analysis, as well as RT-qPCR normalized to 18S rRNA using both intronic and exonic gene-specific primers. Graph displays mean ± SEM with individual biological replicates (n = 2) represented as dots. Student’s t tests were performed to compare the log2FC values upon TRAIL treatment in the presence of DMSO vehicle or zVAD. **p<0.00.1 (C, D) Volcano plot depicting the –log10p values across biological duplicates for the fragment per kilobase per million read fold changes in the sequencing libraries depicted in Figure 2C,D. Each point represents a…

• 

  Figure 3.. Transcriptional repression during early apoptosis requires MOMP, but not necessarily caspase activity. (A) Western blots showing the efficacy of CASP3, CASP8, and caspase-activated DNase (CAD) protein depletion following nucleofection with the indicated siRNAs, with or without 2 hr TRAIL treatment. α-Tubulin (TUBA1A) serves as a loading control. Blot representative of those from four biological replicates. (B) RT-qPCR measurements of 4sU-labeled nascent transcripts with or without 2 hr TRAIL treatment in cells nucleofected with the indicated the siRNAs (n = 3). Also see Figure 3—figure supplement 1A–C. (C) Western blot detecting the indicated proteins in cells nucleofected with the indicated siRNA in the presence or absence of TRAIL. TUBA1A serves as a loading control. Blot representative of four biological replicates. The cleavage of CASP3 and CASP8 (as measured by the disappearance of the full-length form of each zymogen upon 2 hr TRAIL treatment, normalized to TUBA1A) is graphed below (n = 4). (D) 4sU-labeled RNA levels measured by RT-qPCR in HCT116 cells nucleofected with the indicated siRNAs, with or without 2 hr TRAIL treatment (n = 4). Also see Figure 3—figure supplement…

• 

  Figure 3—figure supplement 1.. Transcriptional repression during early apoptosis requires MOMP, but not necessarily caspase activity. (A) RT-qPCR quantification of the change in total RNA levels in HCT116 cells treated with the indicated siRNAs upon 2 hr TRAIL treatment. Fold change were calculated from Ct values normalized to 18S rRNA. (B) Quantification of BrdUTP-labeled HCT116 cells treated with TRAIL for the indicated times in a TUNEL assay (n = 4). Percent-positive cells calculated by flow cytometry using FloJo imaging software to count the number of cells with elevated FL2 fluorescence. (C) Western blot of HCT116 lysates from cells under conditions described in (A) depicting levels of the indicated RNAPII subunits and CASP3 activation. α-Tubulin (TUBA1A) serves as a loading control. Graph to the right quantifies the change in RNAPII subunit levels upon apoptosis induction by TUBA1A-normalized band density (n = 4). (D) Total RNA levels measured by RT-qPCR in HCT116 cells nucleofected with the indicated siRNAs, with or without 2 hr TRAIL treatment (n = 4). (E) Western blot of the cytoplasmic and mitochondrial fractions of HeLa cells treated with 10 μM raptinal or equal volume of DMSO, in the presence of 20…

• 

  Figure 4.. Apoptosis causes reduced RNAPII transcriptional output and promoter occupancy in an mRNA decay-dependent manner. (A) Western blots performed with lysates from HCT116 cells depleted of the indicated decay factors with and without 2 hr TRAIL treatment. Blot representative of three biological replicates. Apoptosis induction was confirmed by disappearance of the full-length CASP3 band, quantified in the graph below by measuring band intensity normalized to an α-tubulin (TUBA1A) loading control (n = 3). (B, C) Changes in total (B) and nascent 4sU-labeled (C) RNA upon 2 hr TRAIL treatment in cells nucleofected with the indicated siRNAs were quantified by RT-qPCR (n = 4). Fold changes were calculated from Ct values normalized to 18S rRNA. Also see Figure 4—figure supplement 1A–B. (D, E) Chromatin immunoprecipitation (ChIP)-qPCR was used to measure occupancy of the indicated promoters by hypophosphorylated RNAPII (D) or TBP (E) under cellular conditions described in (A). Rabbit and mouse IgG antibodies were included in parallel immunoprecipitation reactions with chromatin from scramble siRNA-treated non-apoptotic cells in lieu of TBP and RNAPII antibodies, respectively, as a control.…

• 

  Figure 4—figure supplement 1.. Apoptosis causes reduced RNAPII transcriptional output and promoter occupancy in an mRNA decay-dependent manner. (A, B) RT-qPCR quantification of the change in total (A) and 4sU-labeled (B) RNA levels in HCT116 cells treated with the indicated siRNAs upon 2 hr TRAIL treatment (n = 4). Fold changes were calculated from Ct values normalized to 18S rRNA. (C) Chromatin immunoprecipitation (ChIP)-qPCR was used to measure the change in hypophosorylated RNAPII occupancy at the indicated promoters in cells nucleofected with the indicated siRNAs (n = 4). Mouse IgG antibody was included in parallel immunoprecipitation reactions with chromatin from scramble siRNA-treated non-apoptotic cells in lieu of RNAPII antibody as a control. (D) Representative western blots on lysates of HCT116 cells depleted of either 3’ DFs (DIS3L2, EXOSC4, and PNPT1) or XRN1 before and after 2 hr TRAIL treatment, probing for the indicated protein or RPB1 C-terminal domain (CTD) phosphorylation state. Apoptosis induction was confirmed by observing CASP3 cleavage. A vinculin (VCL) loading control was imaged for the blots of each phosphorylation state, but a single representative panel is shown. Quantification of RPB1…

• 

  Figure 5.. Importin α/β transport links mRNA decay and transcription. (A) Western blots showing the indicated proteins in the nuclear and cytoplasmic fractions of apoptotic and non-apoptotic HCT116 cells nucleofected with the indicated siRNAs. The nuclear and cytoplasmic fractions of PABPC1 were imaged on the same membrane section but cropped and edited separately to visualize the low nuclear expression of this canonically cytoplasmic protein. Protein expression (right) was calculated by band intensity in reference to lamin B1 (LMNB1) or α-tubulin (TUBA1A) loading controls, for the nuclear and cytoplasmic fractions, respectively (n = 3). (B) Nuclear and cytoplasmic expression of the indicated proteins in apoptotic and non-apoptotic cells with a 1 hr pre-treatment with 25 μM ivermectin or an equal volume of EtOH (‘mock’). Nuclear levels of the importin α/β substrate nucleolin (NCL) were quantified by band intensity in reference to a LMNB1 loading control, while disappearance of full-length CASP3 was quantified in the cytoplasm in reference to TUBA1A (n = 3). Also see Figure 5—figure supplement 1A (C) Levels of nascent 4sU-labeled RNA (n = 3) were quantified by RT-qPCR under the cellular…

• 

  Figure 5—figure supplement 1.. Importin α/β transport links mRNA decay and transcription. (A) Western blot showing the indicated proteins in the nuclear and cytoplasmic fractions of HCT116 cells with or without 2 hr TRAIL treatment and/or 3 hr ivermectin. Lamin B1 (LMNB1) and TUBA1A served as nuclear and cytoplasmic loading controls, repectively. Blot representative of that from three biological replicates. (B) Changes in total RNA levels after 2 hr TRAIL treatment in HCT116 cells pretreated with either 25 μM ivermectin or an equal volume of ethanol (‘mock’) for 1 hr (n = 3). (C) Alternative analysis of data presented in Figure 5C, instead quantifying the fold change in nascent transcription in non-apoptotic cells in the presence or absence of 25 μM ivermectin (n = 3). (D) Western blot showing nuclear and cytoplasmic expression of the indicated proteins in HEK293T expressing the viral endonuclease muSOX with a 3 hr treatment of 25 μM ivermectin or equal volume of ethanol. Nuclear and cytoplasmic levels (right) of nucleolin (NCL) were quantified by band intensity in reference to a lamin B1 (LMNB1) or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) loading control, respectively. (E) RT-qPCR quantification of…

• 

  Figure 6.. Schematic representation of the cellular events connecting apoptosis induction with mRNA decay and RNAPII transcription.