Loss of transcription factor EB dysregulates the G1/S transition and DNA replication in mammary epithelial cells

Abstract

Triple-negative breast cancer (TNBC) poses significant challenges for treatment given the lack of targeted therapies and increased probability of relapse. It is pertinent to identify vulnerabilities in TNBC and develop newer treatments. Our prior research demonstrated that transcription factor EB (TFEB) is necessary for TNBC survival by regulating DNA repair, apoptosis signaling, and the cell cycle. However, specific mechanisms by which TFEB targets DNA repair and cell cycle pathways are unclear, and whether these effects dictate TNBC survival is yet to be determined. Here, we show that TFEB knockdown decreased the expression of genes and proteins involved in DNA replication and cell cycle progression in MDA-MB-231 TNBC cells. DNA replication was decreased in cells lacking TFEB, as measured by EdU incorporation. TFEB silencing in MDA-MB-231 and noncancerous MCF10A cells impaired progression through the S-phase following G1/S synchronization; however, this proliferation defect could not be rescued by co-knockdown of suppressor RB1. Instead, TFEB knockdown reduced origin licensing in G1 and early S-phase MDA-MB-231 cells. TFEB silencing was associated with replication stress in MCF10A but not in TNBC cells. Lastly, we identified that TFEB knockdown renders TNBC cells more sensitive to inhibitors of Aurora Kinase A, a protein facilitating mitosis. Thus, inhibition of TFEB impairs cell cycle progress by decreasing origin licensing, leading to delayed entry into the S-phase, while rendering TNBC cells sensitive to Aurora kinase A inhibitors and decreasing cell viability. In contrast, TFEB silencing in noncancerous cells is associated with replication stress and leads to G1/S arrest.

Keywords: Aurora kinase A; DNA damage; DNA replication; MCF-10A; MDA-MB-231; RB1; RNA-Seq; Stathmin 1; TFEB; cell cycle; genome stability; origin licensing; transcription factor; triple negative breast cancer.

Figure 1

TFEB gene expression is elevated in TNBC patients.A and B, boxplots of TFEB RSEM normalized gene expression values as measured by RNA-Seq from breast tumor biopsies collected by the TCGA: breast cancer study, separated by either IHC subtype, or IHC estrogen receptor status. C, boxplots for PPP3R1, FLCN, FNIP1, and MAP4K3 normalized expression values from breast cancer patient tumor biopsies collected as part of the TCGA: breast cancer study, delineated by estrogen receptor status. Notches on boxplots indicate Tukey confidence intervals. ∗∗∗∗p < 0.0001, (A) one-way ANOVA, or (B and C) t test. TCGA, The Cancer Genome Atlas; TFEB, transcription factor EB; TNBC, triple-negative breast cancer.

Figure 2

Cell cycle genes are globally downregulated by TFEB knockdown.A, gene set enrichment analysis with RNA-Seq gene expression results from TFEB knockdown MDA-MB-231 cells (GSE139203), ordered by the normalized enrichment score. B, network analysis of significantly differentially expressed genes related to the cell cycle and associated GO and Reactome terms. TFEB, transcription factor EB.

Figure 3

Transcriptional activators of cell growth are downregulated by TFEB knockdown.A, gene expression of the indicated transcription factors as determined by RNA-Seq analysis of MDA-MB-231 cells with or without knockdown of TFEB, presented as DESeq2 normalized counts. B, genes significantly downregulated by TFEB knockdown were subjected to enrichment analysis against a database of ChIP-Seq results (ChEA) using Enrichr, and the significantly enriched chromatin factors displayed ordered by -log10 p-value of enrichment, with the color representing the ratio of enrichment. ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. TFEB, transcription factor EB.

Figure 4

TFEB knockdown reduces cell proliferation. A–C, EdU cell cycle analysis results depict the percentage of cells in the S-phase following 72 h of TFEB knockdown in the indicated cell lines. D–F, cell counts at the indicated time points following TFEB knockdown in MDA-MB-231, BT549, and MCF10A cells. G, percent of cells that were permeable 120 h after TFEB knockdown in the indicated cell lines. H–I, caspase 3/7 activity 96 h after TFEB knockdown in the indicated cell line, depicted as fluorescence intensity corrected to the protein content. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA (A–G), t test (H–I). TFEB, transcription factor EB.

Figure 5

TFEB silencing alters the level of G1/S regulatory proteins. A and B, immunoblots and quantification of the indicated proteins in MDA-MB-231 cells following 72 h of TFEB knockdown. C and D, immunoblots and quantification of the indicated proteins in MCF10A cells following 72 h of TFEB knockdown. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, t test. TFEB, transcription factor EB.

Figure 6

TFEB knockdown results in G1/S arrest. A and B, immunoblots and quantification from MCF10A cells with or without knockdown of TFEB at the indicated time points following synchronization at the G1/S transition through incubation for 18 h with 2 mM thymidine, an 8-h incubation in normal growth media, followed by a second incubation with 2 mM thymidine for 18 h ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, two-way ANOVA. C, EdU cell cycle analysis of TFEB knockdown MDA-MB-231 cells synchronized at the G1/S transition by 24 h of thymidine block and grown in the absence of thymidine for the indicated time points. D, the percentage of cells which entered the S-phase (EdU+) at the indicated times following thymidine block. TFEB, transcription factor EB.

Figure 7

Loss of RB1 function does not rescue G1/S arrest caused by TFEB knockdown.A and B, immunoblots and quantification of the indicated proteins from MDA-MB-231 cells treated with the indicated siRNAs for 72 h. C, percentage of cells in the S-phase quantified using EdU-DNA cell cycle analysis. D, percentage of dead cells determined by quantification of cell permeability. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA, or two-way ANOVA (D). TFEB, transcription factor EB.

Figure 8

Origin licensing and DNA replication genes are downregulated by TFEB knockdown.A and B, heatmap and enrichment plot for TFEB knockdown induced differential expression of genes involved in origin licensing and the replisome in MDA-MB-231 cells. C, ChIP-Seq peaks for Flag-TFEB (red) or wildtype control (gray) HeLa cells in the vicinity of the indicated genes, obtained from NCBI GEO series GSE1803222. TFEB, transcription factor EB.

Figure 9

TFEB silencing leads to origin under-licensing in MDA-MB-231 cells.A, imaging cytometry analysis of chromatin-bound MCM2 in MDA-MB-231 cells with or without TFEB knockdown, n = 6000 cells per treatment. B, gating strategy to assign cell cycle phases using EdU and DNA fluorescent intensities. C, smoothed density estimates for chromatin bound MCM2 levels by cell cycle gate as determined using EdU uptake and DNA content analysis. D, quantification of %MCM2 positive cells by cell cycle phase, n = 3 independent experiments. E and F, smoothed density estimate and MCM2 positivity by cell cycle phase from control or TFEB knockdown MDA-MB-231 cells released for 1 h following 24 h incubation with 2 mM thymidine, n = 3. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, (C) two-way ANOVA or (E) t test. TFEB, transcription factor EB.

Figure 10

TFEB knockdown elevates the rate of replication and decreases p21 expression.A and B, smoothed density estimate, quantification of total EdU fluorescence in S-phase cells, and the 95th percentile value for total EdU uptake, as a measure of maximal replication rate from control or TFEB knockdown MDA-MB-231 cells, n = 3. C, smoothed density estimates for p21 fluorescence intensity by cell cycle phase, as measured by imaging cytometry, for MDA-MB-231 cells treated with the indicated siRNA. D, quantification of the %p21 positive cells in G1 phase. ∗∗p < 0.01, ∗∗∗p < 0.001, t test. TFEB, transcription factor EB.

Figure 11

Kinase inhibitor screening identifies synthetic lethality with TFEB knockdown and Aurora kinase A inhibition.A, volcano plot for the results of the kinase inhibitor screen, depicting the statistical significance and change in cell viability between siCTRL and siTFEB#2 transfected MDA-MB-231 cells following treatment with the indicated inhibitor for 72 h at 10 μM. B and C, metabolic fractional viability of TFEB knockdown MDA-MB-231 and BT549 cells treated with the indicated concentration of phthalazinone pyrazole for 72 h ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA. TFEB, transcription factor EB.

Figure 12

Aurora kinase A inhibition significantly enhances TFEB knockdown induced cell death. A–D, colony formation assay and quantification of siCTRL or siTFEB#2 transfected (A and B) MDA-MB-231 cells or (C and D) BT549 cells, n = 6 treatments from two independent experiments. E and F, percent cell death as quantified by cell permeability in the indicated cell lines following 72 h of treatment with 4 μM phthalazinone pyrazole, n = 4 or 5. G, RNA-Seq quantification of STMN1 gene expression from TFEB silenced MDA-MB-231 cells. H and I, immunoblot quantification of STMN1 protein expression in MDA-MB-231 cells 72 h after treatment with scramble control shRNA/siRNA or TFEB knockdown shRNA/siRNA. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, t test (B and D), two-way ANOVA (E and F), one-way ANOVA (I). STMN1, Stathmin 1; TFEB, transcription factor EB.

Figure 13

Proposed model for TFEB mediated cell cycle regulation. In proliferating cells, TFEB promotes the expression of G1/S regulators, DNA replication machinery, and origin licensing to ensure progression through the S-phase while inhibiting cell death. In contrast, knockdown of TFEB causes G1/S arrest or delay in S-phase entry and elevates the rate of cell death in association with origin under-licensing. In addition, STMN1 upregulation in TFEB knockdown cells may cause increased sensitivity to AURKA inhibition. AURKA, Aurora kinase A; STMN1, Stathmin 1; TFEB, transcription factor EB; TNBC, triple-negative breast cancer.