miR-211 is a prosurvival microRNA that regulates chop expression in a PERK-dependent manner

Abstract

MicroRNAs typically function at the level of posttranscriptional gene silencing within the cytoplasm; however, increasing evidence suggests that they may also function in nuclear, Argonaut-containing complexes, to directly repress target gene transcription. We have investigated the role of microRNAs in mediating endoplasmic reticulum (ER) stress responses. ER stress triggers the activation of three signaling molecules: Ire-1α/β, PERK, and ATF6, whose function is to facilitate adaption to the ensuing stress. We demonstrate that PERK induces miR-211, which in turn attenuates stress-dependent expression of the proapoptotic transcription factor chop/gadd153. MiR-211 directly targets the proximal chop/gadd153 promoter, where it increases histone methylation and represses chop expression. Maximal chop accumulation ultimately correlates with miR-211 downregulation. Our data suggest a model in which PERK-dependent miR-211 induction prevents premature chop accumulation and thereby provides a window of opportunity for the cell to re-establish homeostasis prior to apoptotic commitment.

Figure 1. PERK-dependent expression of miR-211

A) Heat map representation of ER-stress dependent micro-RNA expression. B) Q-PCR Analysis of PERK+/+ (WT), and PERK−/− (KO) treated with thapsigargin (500nM) or starved for glucose. Values are the average of at least 3-independent experiments and error bars indicate standard deviation between experimental replicates. C) Induction of CHOP and eIF2α phosphorylation in response to thapsigargin and glucose starvation. D–E) Q-PCR analysis of trpm1 or pri-211 in wild type or PERK−/− MEFs treated with thapsigargin (500nM) for the indicated intervals. Values are the average of at least 3-independent experiments and error bars indicate standard deviation between experimental replicates.

Figure 2. MiR-211 induction depends on eIF2α phosphorylation and expression of ATF4

A) Q-PCR analysis of PERK+/+ (WT), eIF2αA/A, and ATF4−/− (KO) MEFs treated with thapsigargin (500nM) for 5 hours. B) HeLa cells were either transfected with either empty, ATF4 or Nrf2 encoding vectors. Cells transfected with empty vector were treated with thapsigargin (500nM) as a positive control or not treated (negative control). After 24 hours, total RNA was isolated and miR-211 levels were assessed by Q-PCR. C) Cells in B were also assessed for chop expression by Q-PCR to confirm functional ATF4. All panels provide values that are the average of at least 3-independent experiments and error bars indicate standard deviation between experimental replicates.

Figure 3. MiR-211 regulates stress-dependent CHOP expression

A) The recognition site of miR-211 was cloned at the 3’UTR of luciferase gene (211-Luc); empty vector or 211-Luc were expressed in PERK+/+ or −/− MEFs and treated with thapsigargin. Error bars represent standard deviation for 3 independent experiments. B) Scrambled (control) or A211 were introduced into NIH3T3 cells, treated with thapsigargin for the indicated intervals and levels of CHOP and ATF4 were assessed by western. C) Scramble (control) or a miR-211 were introduced into NIH3T3, treated with thapsigargin for the indicated intervals and levels of phospho-PERK, CHOP and eIF4E were assessed by immunoblot. D) Q-PCR analysis of chop mRNA in NIH3T3 cells transfected with A211 and challenged with thapsigargin. The difference between untransfected and A211 transfected cells challenged with thapsigargin is significant (P=0.023). E) Q-PCR assessment of chop mRNA in cells challenged with thapsigargin +/− miR-211. Values are the average of at least 3-independent experiments and error bars indicate standard deviation.

Figure 4. MiR-211 regulates stress-dependent CHOP accumulation through binding sites in promoter

A) MicroRNA-211 pulls down Chop promoter RNA. Biotinylated wild type miR-211 or control scramble were introduced into NIH3T3 cells (mut-211). Cells were processed essentially for chromatin IP, except that RNA:RNA duplexes were collected on streptavidin beads. Precipitation of chop with miR-211 was analyzed by Q-PCR using primers scanning region −619 to −503. (Right) Equivalent amounts of biotinylated microRNAs were present in precipitates. B) NIH3T3 cells transfected with biotinylated microRNAs were stressed for 8 hours and then precipitated with streptavidin beads. Precipitates were analyzed for presence of RNA spanning first exon-intron junction (118 exonic-2820 intronic). Data represents the average of 3 experiments. C) MiR-211 regulates CHOP promoter activity through its recognition sites at nucleotide 121(site 1) and 141(site 2). NIH3T3 cells were transfected with either a promoter-less luciferase reporter, with the wild type or indicated mutant CHOP promoter (mut1=site1; mut2=site2). In addition, each transfected construct was supplemented with or without mimic miR-211. Error bars show values for 3 independent experiments. D) MiR-211 site deletion mutants in the CHOP promoter abrogates regulation by ER stress. Cells expressing the wild type chop-luc or the indicated mutants were challenged with thapsigargin for 8 hours. Values are the average of at least 3-independent experiments. E) NIH3T3 cells expressing chop-luc reporter constructs containing let-7 seed matches at either site 1 or 2 were challenged with thapsigargin following expression of a let-7 miR where indicated. The error bars provided in each panel indicate standard deviation.

Figure 5. MiR-211 induces histone modifications at the chop promoter

A) MiR-211 or A211 were introduced into NIH3T3 cells after which cells were stressed with thapsigargin for indicated intervals. Subsequently, association of RNA polymerase with the chop promoter was assessed by ChIP. B–C) H3K27me3 ChIP was performed to assess modification of either the chop (B) or neighboring Dctn2 and Gl1 promoters (C) following 5 hours of thapsigargin treatment. D) To determine Ago1 occupancy on the chop promoter, myc-tagged Ago1 construct along with either miR-211 or A211 oligos were introduced into HeLa cells and processed for ChIP using myc-beads. Error bars reflect standard deviation between 3 independent experiments.

Figure 6. Suppression of miR-211 accelerates apoptosis

A) NIH3T3 cells were transfected with control (−) or A211(+) for 48 hours and then treated with thapsigargin at indicated time-points. Cell lysates were probed for cleaved caspase 3, CHOP, ATF4 as indicated. B) Viability of NIH3T3 cells transfected with A211 or scrambled control and treated with 500nM thapsigargin for the indicated intervals. Error bars indicate values for 3-independent experiments. C) Wild-type or chop−/− MEFs were transfected with either scrambled or A211for 48 hours and then treated with thapsigargin at indicated time-points. Caspase-Glo reagent was added and caspase activation was assessed by luminescence. D) Following transfection of wild type or chop−/− MEFs with scramble or A211 and exposure to 50nM thapsigargin, viability was assessed by propidium iodide staining and cells containing less than 2N DNA content were quantified by FACS. Error bars indicate standard deviation between 3 independent experiments.

Figure 7. PERK-dependent regulation of miR-211 in tumors

Total RNA from tumors was purified and analyzed by Q-PCR. Sno202 served as an internal control for mouse samples and U6 for human samples. A) Analysis of miR-211 levels in MMTV-D1T286A mammary tumors. B) Accumulation of miR-211 and chop are inversely correlated in mammary carcinomas. C) MiR-211 expression was analyzed in MMTV-Neu tumors on either PERK+/+ or −/− background. D) Assessment of miR-211 levels in primary human lymphoma.