(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



EGR1 drives cell proliferation by directly stimulating TFEB transcription in response to starvation [1]

['Marcella Cesana', 'Telethon Institute Of Genetics', 'Medicine', 'Tigem', 'Pozzuoli', 'Naples', 'Department Of Advanced Biomedical Sciences', 'Federico Ii University', 'Gennaro Tufano', 'Francesco Panariello']

Date: 2023-03 

The stress-responsive transcription factor EB (TFEB) is a master controller of lysosomal biogenesis and autophagy and plays a major role in several cancer-associated diseases. TFEB is regulated at the posttranslational level by the nutrient-sensitive kinase complex mTORC1. However, little is known about the regulation of TFEB transcription. Here, through integrative genomic approaches, we identify the immediate-early gene EGR1 as a positive transcriptional regulator of TFEB expression in human cells and demonstrate that, in the absence of EGR1, TFEB-mediated transcriptional response to starvation is impaired. Remarkably, both genetic and pharmacological inhibition of EGR1, using the MEK1/2 inhibitor Trametinib, significantly reduced the proliferation of 2D and 3D cultures of cells displaying constitutive activation of TFEB, including those from a patient with Birt-Hogg-Dubé (BHD) syndrome, a TFEB-driven inherited cancer condition. Overall, we uncover an additional layer of TFEB regulation consisting in modulating its transcription via EGR1 and propose that interfering with the EGR1-TFEB axis may represent a therapeutic strategy to counteract constitutive TFEB activation in cancer-associated conditions.

Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: A.B. is a cofounder of Casma Therapeutics and an advisory board member of Next Generation Diagnostic srl, Avilar Therapeutics and Coave Therapeutics. Davide Cacchiarelli is Co-Founder, Shareholder and Consultant of Next Generation Diagnostic srl.

Funding: We acknowledge the support of the following funding agencies: the Italian Telethon Foundation (to A.B.), the Associazione Italiana per la Ricerca sul Cancro (IG-22103 and 5x1000-21051 to A.B.), the Ministero dell'Università e della Ricerca (PRIN 2017E5L5P3 to A.B.), the European Research Council (H2020 AdG; LYSOSOMICS 694282 to A.B.), the Rita-Levi Montalcini program from MIUR to M.C, the FIRC-AIRC fellowship to S.A. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability: RNA sequencing and ChIP-sequencing data have been deposited on GEO (GSE209899). Quantitative data used to generate plots and histograms can be found in the spreadsheet S1 Data . A separate tab is associated with each Figure and Supporting Information . Uncropped images are found in S1 Raw Images .

Here, we surveyed transcription factors for their capacity to control TFEB expression and identified the immediate-early gene EGR1 as a positive transcriptional regulator of TFEB. Stress signals and secreted factors, including growth factors, tumor necrosis factors, inflammatory factors, ionizing radiation, reactive oxygen species, or others, are known to trigger EGR1 transcription via the MEK1/2/ERK signaling cascade [ 16 – 20 ]. Once activated, EGR1 initiates a transcriptional cascade that impacts a remarkable spectrum of, often opposing, cellular mechanisms, including survival and apoptosis, growth control and arrest, differentiation, and transformation [ 21 – 23 ]. Here, we showed that EGR1 controls TFEB transcription and contributes to TFEB-driven starvation response, significantly impacting the expression of several cell cycle regulators. Finally, we provided evidence that genetic and pharmacological inhibition of EGR1 using the FDA-approved compound Trametinib impairs TFEB-driven cell proliferation. Collectively, our data identify a novel layer in the regulation of TFEB, which may be targeted for therapeutic purposes.

TFEB activity is known to be regulated at the posttranslational level through phosphorylation mediated by the nutrient-sensing complex mTORC1. Upon nutrient depletion, TFEB translocates to the nucleus and induces a global transcriptional response to ensure adaptation to changes in cells’ energy demands [ 5 , 12 – 15 ]. Despite the increasing recognition of TFEB’s role in controlling vital metabolic processes and its involvement in several diseases, a knowledge gap remains on the mechanisms and players controlling its transcriptional regulation and their relevance in cellular adaptation to environmental cues.

An example is transcription factor EB (TFEB), a member of the microphthalmia family (MiT-TFE) of bHLH-leucine zipper transcription factors, which act as a global modulator of intracellular clearance and energy metabolism through the control of lysosomal biogenesis and autophagy [ 3 – 5 ]. Besides its role in cellular metabolism, TFEB is a crucial player in cancer biology [ 6 – 8 ]. Recently, we and others showed that constitutive activation of TFEB, through the inhibition of the noncanonical mTORC1 pathway driven by FLCN-RagC/D, induces renal tumorigenesis in mouse models of Birt-Hogg-Dubé (BHD) syndrome [ 9 , 10 ] and tuberous sclerosis [ 11 ].

Cells have evolved by improving their capacity to metabolically adapt to changes in substrate availability and energy requirements. This metabolic flexibility, essential to maintain energy homeostasis, is made possible by the coordinated interplay of diverse quality control mechanisms. In this scenario, transcriptional control of gene expression heavily impacts the homeostatic energy balance in both physiological and pathological conditions [ 1 ]. Importantly, acceleration of energy metabolism to fuel cell growth and division is a hallmark of cancer cells [ 2 ]. Indeed, cancer cells often exploit transcription factor-mediated catabolic programs to meet their requirements.

Results

Dissection of human TFEB locus revealed an elaborate structure with multiple regulatory regions To dissect the transcriptional architecture of the human TFEB gene, we interrogated expression profile datasets of a wide range of human tissues of embryonic origin [24,25]. We observed that TFEB displays a ubiquitous expression with a median fluctuation within one order of magnitude between different tissues (S1A Fig, left panel). This trend is also maintained across several ENCODE reference cell lines [26], with human embryonic stem cells (hESCs) displaying the highest levels of TFEB expression (S1A Fig, right panel). Despite a widespread expression of TFEB across tissues, its genomic locus results in a complex exon–intron structure and chromatin makeup. S1B Fig shows the human TFEB locus with mRNA species harboring from the negative strand, transcribed from right to left. The TFEB gene contains alternative 5′ noncoding exons and an approximately 40-kb first intron. Analysis of TFEB transcripts revealed multiple splicing isoforms with a prevalence of isoforms containing 8 coding exons and a common 3′ UTR region (red), encoding a 476 amino acid protein. The primary reference transcriptional isoform (underlined) displays the first exon with a common expression in most of the ENCODE reference cell lines [26] (Transcription—highlighted with a light gray bar, arrow #1). To gain more insights into the structure of the TFEB locus, we performed a computational analysis integrating existing genomic and epigenetic data. Specifically, we leveraged H3K4me1, H3K4me3, and DNase I profiles from the ENCODE and Epigenome Roadmap Project [27] to locate putative cis-regulatory elements (CREs), which we numbered from 1 to 7 (S1C Fig and see Methods). In particular, CREs 1, 2, and 3 are enriched for the promoter-specific modification H3K4me3 and represent the region where most alternative transcriptional start sites are located (Promoters—H3K4me3). In particular, the strongest promoter signal common to the major reference cell lines is localized upstream of the major TSS (arrow #2), right next to the strongest chromatin accessibility mark in the region (arrow #3), suggesting histone displacement, TF engaging, and assembly of the transcriptional machinery. The presence of an articulated, cell-specific H3K4me1 patterning in the TFEB locus suggests that distinct CREs are likely engaged in a cell-specific fashion (Enhancers—H3K4me1), as underlined by a different TFEB expression pattern across tissues (S1A Fig). Additionally, minor alternative TSS arising from CREs with a little promoter activity (i.e., 5 and 7) generate low abundant mRNAs without different protein-coding features compared to the major isoform. The engagement of differential CREs, labeled by enhancers, promoters, and/or accessibility marks, is further confirmed by chromatin folding data (Hi-C), which displays proximity contacts between the promoter regions and the other CREs enriched for the H3K4me1 enhancer mark (S1D Fig). Lastly, a long-distance interaction was observed between the promoter and the transcriptional termination region, with a concomitant “carry-over” of promoters and chromatin accessibility marks (arrow #4), as previously reported for actively transcribed loci [28]. In line with this observation, hESCs display the highest degree of enrichment of H3K4me1 and DNAseI accessibility in this region, thus resulting in a high level of TFEB expression (S1A Fig, right panel).

EGR1 regulates TFEB transcription To identify bona fide TFEB transcriptional regulators, we selected 78 TFs predicted to bind to multiple sites in the CREs 1 to 7 regions of the TFEB locus and whose expression showed higher correlation scores with TFEB expression (see Methods) (Fig 1A). We identified CRE1 as the core-promoter region of TFEB (TFEB PROM), as it resides immediately upstream of the major TSS (arrow #1) and exhibits the highest levels of histone displacement (arrow #2) and chromatin accessibility (arrow #3). Plasmids containing the selected TFs were cotransfected into HeLa cells together with a reporter construct in which the TFEB promoter region was fused with the firefly luciferase coding region. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. EGR1 positively modulates TFEB transcription. (A) Schematic representation of the filtering strategy to identify regulators of TFEB transcription. TFs were selected according to several criteria: (i) the number of predicted binding sites at the level of the identified CREs (Bumscore >3) and their correlation of expression with TFEB (Pearson correlation >0.3); (ii) TFs differentially expressed upon starvation (TFs DE STV); (iii) TFs with a Bumscore >2 up to 96 TFs. Selected TFs were cloned in the pLX304 vector and overexpressed in HeLa cells along with the TFEB promoter-reporter (TFEB PROM), in which the TFEB promoter region was cloned upstream of the firefly luciferase reporter gene (FLuc). (B) Survey of candidate TFs through luciferase-based promoter assay. (Left) Dot plot showing the luciferase activity relative to each TF measured as FLuc/RLuc ratio shown with respect to the empty control vector (CTRL). (Right) Table of the top 10 TFs with corresponding Log FC values (LFC) relative to the ratio FLuc/RLuc. (C) TFEB promoter activity assay. TFEB promoter (WT) and its mutated version for EGR1 binding sites (mut) were cloned upstream of the firefly luciferase coding region (FLuc). These constructs were cotransfected in HeLa cells alongside plasmids expressing EGR1 open reading frame or with a control empty vector (CTRL). A construct carrying the Renilla luciferase (RLuc) was transfected as a control. Bar plot showing the luciferase activity measured as FLuc/RLuc ratio shown with respect to the control vector. Mean ± SD values are shown. ANOVA was used; *p < 0.05, **p < 0.01, ***p < 0.001. (D) Normalized expression (CPM) of TFEB mRNA in HeLa cells upon CTRL and EGR1 overexpression measured by RNAseq (*** FDR < 0.001). (E) Representative image of the immunoblot analysis of TFEB and EGR1 levels upon EGR1 overexpression in HeLa cells (n = 3). GAPDH was used as a loading control. (F) (Upper) Line plot showing EGR1 and TFEB mRNA dynamics during a starvation time course. HeLa cells were treated with HBSS for the indicated time points. EGR1 and TFEB mRNA levels were quantified by RNAseq and shown as Log 2 FC with respect to their levels in FED conditions (** FDR < 0.01; *** FDR < 0.001). (Lower) Representative image of immunoblot analysis of EGR1 and TFEB levels in HeLa cells treated with HBSS for the indicated time points. GAPDH was used as a loading control. (G) ChIPseq analysis of HeLa cells undergoing starvation (HBSS) for 6 hours compared to cells in fed condition (FED). (Upper) Distribution plots of the average read coverage density of EGR1 binding signal within 2 kb from all TSSs in the genome. (Lower) Binding heatmaps displaying the individual read coverage density of the EGR1 binding signal. (H) Representative genome browser snapshots of TFEB promoter bound by EGR1 in starvation. Both reads distributions as line plots and peak intervals are displayed. H3k4me3, H3k4me1, and H3K27ac enrichments at the TFEB promoter during starvation are also displayed. The genomic localization of the TFEB promoter region (as described in S1 Fig) is reported (red line). Individual quantitative observations that underlie the data summarized here can be located under the Supporting information file as S1 Data. Uncropped images can be found in the Supporting information file as S1 Raw Images. ChIPseq, chromatin immunoprecipitation sequencing; CRE, cis-regulatory element; CTRL, control; EGR1, early growth response 1; FLuc, firefly luciferase; LFC, Log FC; RLuc, Renilla luciferase; TF, transcription factor; TFEB, transcription factor EB; TSS, transcriptional start site; WT, wild-type. https://doi.org/10.1371/journal.pbio.3002034.g001 This analysis revealed that 10 TFs could induce TFEB promoter activity above 5-fold compared to the control. Among them, the early growth response 1 (EGR1) TF displayed the strongest ability to increase the luciferase activity driven by the TFEB promoter and was chosen for further analyses (Fig 1B). EGR1 belongs to the immediate-early genes (IEGs) family and represents the earliest downstream nuclear target sensitive to changes in the extracellular environment, including nutrients and stress signals [29], which triggers its activation via the MEK1/2/ERK signaling pathway [21,30]. Therefore, based on the established role of TFEB as a nutrient and stress sensor in the cell, we hypothesized a functional correlation between EGR1 and TFEB. To validate EGR1 binding to the TFEB promoter, we cotransfected TFEB PROM and plasmids for EGR1 overexpression (EGR1) or control (CTRL) into HeLa cells. We observed a significant induction of luciferase activity relative to the wild-type (WT) promoter. Conversely, luciferase activity was significantly reduced in the presence of a TFEB PROM in which EGR1 binding sites were mutated (mut) (Fig 1C). To further confirm the finding that EGR1 positively controls TFEB expression, we overexpressed EGR1 in HeLa cells and performed RNAseq. We observed that TFEB mRNA and protein levels were significantly induced by EGR1 (Fig 1D and 1E). Gene ontology (GO) enrichment analysis of genes up-regulated upon EGR1 overexpression using both manually curated gene sets and KEGG pathways highlighted TFEB-regulated pathways, including “Autophagy,” “Lysosome” [5], and “mTOR signaling,” along with others, such as “TNF signaling,” “Epithelial-Mesenchymal Transition,” and “Hypoxia” (S2A Fig). In line with the GO results, we observed an induction of TFEB targets belonging to several pathways, including autophagy genes, lysosomal genes, and components of the mTORC1 signaling pathway (S2B Fig). Collectively, these results demonstrate that EGR1 regulates TFEB expression and its downstream transcriptional network.

EGR1 associates with TFEB promoter upon starvation Upon starvation, TFEB rapidly translocates to the nucleus and activates the transcription of its target genes [15]. As EGR1 is activated upon various stimuli, including nutrient depletion [21], we sought to investigate whether it contributes to the nutrient-sensing response by regulating TFEB transcription. Thus, we performed a starvation time course and analyzed the expression dynamics of both EGR1 and TFEB by RNAseq. EGR1 mRNA levels are low in basal conditions (FED) and significantly increase from 2 hours of starvation (2h) to remain high until 6 hours, after which they start decreasing but still remain higher than in the basal condition. Differently, TFEB expression starts to significantly increase from 6 hours of starvation (6h), reaching its maximum at 8 hours (8h) (Fig 1F, upper panel). Analysis of EGR1 and TFEB protein levels by western blot reflected their transcriptional dynamics (Fig 1F, lower panel). To test whether a functional correlation exists between EGR1 and TFEB during starvation, we evaluated endogenous EGR1 occupancy genome-wide by performing chromatin immunoprecipitation sequencing (ChIPseq) in HeLa cells in fed and starved conditions (HBSS). ChIPseq analysis revealed that, upon starvation, EGR1 associates with its target genes (Fig 1G) and is preferentially enriched at the TFEB promoter region, marked by the histone marker H3K4me3 (Fig 1H), supporting the hypothesis that EGR1 is responsible for TFEB mRNA induction during nutrient depletion. We also evaluated whether genes bound by EGR1 were up-regulated during starvation. Surprisingly, we observed that only 4.3% of them were up-regulated, whereas, for most of them, the expression was unchanged or reduced (S3A Fig). Globally, GO analysis of EGR1 targets up-regulated in starvation highlighted, among others, categories related to “TNF-alpha signaling,” “Hypoxia,” “Mitophagy,” and “G2-M checkpoint” (S3B Fig). Collectively, these data demonstrate that EGR1 association to the chromatin is triggered by starvation and that TFEB is among those few EGR1 targets to be up-regulated, further emphasizing a functional correlation between EGR1 and TFEB.

Modulation of EGR1 impacts TFEB-driven cell proliferation TFEB has recently emerged as a critical player in a wide array of cancer-associated conditions, in which it was found to promote tumorigenesis [6]. Indeed, in a previous study, we showed that renal tumorigenesis in BHD syndrome is caused by constitutive TFEB activation due to FLCN loss of function [9]. Thus, we sought to evaluate whether modulation of EGR1 expression influences the oncogenic properties of constitutive TFEB activation. We measured the clonogenic capacity of cells stably expressing short-hairpin RNAs against EGR1 (shEGR1) through a colony formation assay (CFU). HeLa WT cells treated with shEGR1 significantly reduced the number and size of the colonies compared to the control (shLUC) (Fig 3A, upper panel). This effect was significantly enhanced in HeLa cells depleted for FLCN (HeLa-FLCN KO), where TFEB is constantly nuclear, and acting on its expression levels might represent a valuable strategy to reduce its activity. Indeed, we observed a complete abolishment of colonies output in shEGR1-treated cells. Notably, in line with the oncogenic role of a constitutively active TFEB, we observed that HeLa-FLCN KO cells form more colonies than HeLa WT cells (Fig 3A, lower panel). Quantification of both colony number and size confirmed that shRNA-mediated down-regulation of EGR1 impacts the clonogenic capacity of HeLa WT and, more pronouncedly, HeLa-FLCN KO cells (Fig 3B). Measurement of mRNA levels by quantitative RT-PCR (Fig 3C, left panel) and immunoblot analysis (Fig 3C, right panel) confirmed a significant decrease of TFEB expression upon shEGR1 in HeLa-FLCN KO. To further investigate to which extent reduced EGR1 levels affect the oncogenic properties of HeLa-FLCN KO cells, we developed a 3D cellular system (3D spheroids) that better mimics the structural organization of a solid tumor (Fig 3D). Over 9 days, we monitored 3D spheroids’ growth in culture and quantified their size by High-Content Imaging. 3D spheroids derived from HeLa-FLCN KO infected with shEGR1 displayed a significant reduction in the kinetic of growth compared to control-treated cells (shLUC) (Fig 3E). To prove that the impairment of shEGR1-derived spheroids’ growth was due to EGR1-mediated down-regulation of TFEB, we transduced shEGR1-treated cells with a lentiviral construct for TFEB overexpression (Fig 3F). Quantification of spheroids’ size showed that TFEB overexpression could rescue the growth defects of shEGR1-derived spheroids (Fig 3G). Interestingly, in line with previous observations suggesting the involvement of TFEB in cell migration [31–33], we observed a significant increase in the migration capacity of HeLa-FLCN KO overexpressing TFEB, strengthening TFEB’s oncogenic role (S4A and S4B Fig). Collectively, these data indicate that HeLa-FLCN KO cells are sensitive to EGR1 down-regulation, which affects their oncogenic features by reducing TFEB expression.

Trametinib inhibits the EGR1-TFEB axis The mitogen-activated protein kinase (MAPK) signaling pathway is a well-established positive regulator of EGR1 expression [34]. Recently, the MEK1/2 inhibitor Trametinib, an FDA-approved compound, was shown to suppress inflammation in LPS-activated macrophages by inhibiting the MEK-ERK-Egr-1 pathway [35] (Fig 4A). Therefore, we test whether Trametinib affected TFEB transcription through inhibition of EGR1 and transfected the TFEB promoter construct (WT) in HeLa cells in the presence of Trametinib or DMSO as control. Luciferase assay showed that Trametinib reduced the activity of the TFEB promoter, proving that Trametinib negatively impacts TFEB transcription (Fig 4B). In accordance with the notion that EGR1 is a well-established target of the MEK-ERK pathway [30,35], we showed that the increase of EGR1 expression observed in starvation correlated with the activation of the MEK-ERK pathway, measured by the increase of the phosphorylation state of ERK by immunoblot analysis. Notably, Trametinib treatment completely abrogates pERK and, consequently, EGR1 expression, suggesting that EGR1 activation upon starvation depends on MERK-ERK pathway activation (S5A and S5B Fig). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Trametinib inhibits the EGR1-TFEB axis in BHD patient-derived cancer cells. (A) Schematic representation of the MEK1/2/ERK signaling cascade, which regulates EGR1 activation. (B) TFEB promoter activity assay upon Trametinib treatment. TFEB promoter (WT), along with a control vector (EMPTY), were transfected in HeLa cells for 48 hours, following starvation (HBSS) for 6 hours in the presence of DMSO or Trametinib. Bar plot showing the luciferase activity measured as FLuc/RLuc ratio. Mean ± SD values are shown. ANOVA was used; *p < 0.05, **p < 0.01, ***p < 0.001. (C) Trametinib dose–response curve. HeLa WT and HeLa-FLCN KO (KO) cells were treated with serial dilutions of Trametinib. The half maximal inhibitory concentration (IC50) was calculated by counting the nuclei of viable cells. (D) Bar plots showing relative quantification of EGR1 and TFEB mRNA levels measured by qPCR in HeLa-FLCN KO cells treated with DMSO or Trametinib. Values were normalized on the HPRT expression and displayed as a fold change relative to DMSO. Mean ± SD values are shown. ANOVA was used; *p < 0.05, **p < 0.01, ***p < 0.001. (E) Immunoblot analysis of EGR1 and TFEB expression in HeLa-FLCN KO cells treated with DMSO and Trametinib. GAPDH was used as a loading control. (F) Representative image of colony formation capacity of UOK-257 cells treated with DMSO and Trametinib. (G) Bar plots showing the quantification of colonies’ number and size (%Area) relative to indicated treatments. Mean ± SD values are shown. ANOVA was used; *p < 0.05, **p < 0.01, ***p < 0.001. (H) Representative image of colony formation capacity of UOK-257 cells transduced with control (CTRL), EGR1, and TFEB overexpression vectors and treated with Trametinib. (I) Bar plots showing the quantification of colonies’ number and size (%Area) relative to indicated treatments. Mean ± SD values are shown. ANOVA followed by Tukey’s multiple comparisons test was used; *p < 0.05, **p < 0.01, ***p < 0.001. (J) Bright-field and fluorescence representative images of 3D spheroids generated from UOK-257 cells transduced with control (CTRL) and TFEB overexpression vectors and treated with DMSO and Trametinib. (K) Dot plot showing the ratio of the intensity of the cell death dye over the size of each spheroid (Intensity/Area) derived from CTRL and TFEB-engineered UOK-257 cells upon indicated treatments. Mean ± SD values are shown. ANOVA was used; *p < 0.05, **p < 0.01, ***p < 0.001. Individual quantitative observations that underlie the data summarized here can be located under the Supporting information file as S1 Data. Uncropped images can be found in the Supporting information file as S1 Raw Images. BHD, Birt-Hogg-Dubé; CTRL, control; EGR1, early growth response 1; TFEB, transcription factor EB; WT, wild-type. https://doi.org/10.1371/journal.pbio.3002034.g004 In line with this, the efficacy of Trametinib treatment in reducing cell viability (IC50 of 16.18 µM in HeLa-FLCN KO versus 18.78 µM in WT cells) was more pronounced in HeLa FLCN-KO cells, where TFEB is active (Fig 4C). On a genome-wide level, Trametinib treatment in HeLa-FLCN KO cells negatively affected the expression of genes belonging to cell cycle–related pathways (i.e., “G2-M Checkpoint” and “Mitotic Spindle”) (S5C and S5D Fig). In agreement with the luciferase assay results, Trametinib significantly reduced endogenous TFEB and EGR1 at both RNA (Fig 4D) and protein levels (Fig 4E) in HeLa-FLCN KO cells.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002034

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/