Transcriptional regulation of autophagy-lysosomal function in BRAF-driven melanoma progression and chemoresistance

Abstract

Autophagy maintains homeostasis and is induced upon stress. Yet, its mechanistic interaction with oncogenic signaling remains elusive. Here, we show that in BRAF^V600E-melanoma, autophagy is induced by BRAF inhibitor (BRAFi), as part of a transcriptional program coordinating lysosome biogenesis/function, mediated by the TFEB transcription factor. TFEB is phosphorylated and thus inactivated by BRAF^V600E via its downstream ERK independently of mTORC1. BRAFi disrupts TFEB phosphorylation, allowing its nuclear translocation, which is synergized by increased phosphorylation/inactivation of the ZKSCAN3 transcriptional repressor by JNK2/p38-MAPK. Blockade of BRAFi-induced transcriptional activation of autophagy-lysosomal function in melanoma xenografts causes enhanced tumor progression, EMT-transdifferentiation, metastatic dissemination, and chemoresistance, which is associated with elevated TGF-β levels and enhanced TGF-β signaling. Inhibition of TGF-β signaling restores tumor differentiation and drug responsiveness in melanoma cells. Thus, the "BRAF-TFEB-autophagy-lysosome" axis represents an intrinsic regulatory pathway in BRAF-mutant melanoma, coupling BRAF signaling with TGF-β signaling to drive tumor progression and chemoresistance.

Fig. 1

BRAF^V600E inhibitor triggers autophagy–lysosomal activation through TFEB. a, b Western blot analysis (a) and densitometric quantification (b) of the LC3-II/LC3-I and the p62/Actin ratios in A375 and MeWo cells treated with the indicated concentrations of PLX4720 for 12 h. Actin served as a loading control. n = 4 independent experiments. c Representative images of LysoTracker Red staining of A375 and MeWo cells treated for 12 h with DMSO or PLX4720 (1 μM). Quantification of relative fold induction of lysosomes by PLX4720 is shown in the right panel. n = 3 independent experiments. d Relative lysosome NAG activity in PLX4720-treated A375 cells. n = 3 independent experiments. e Expression analysis of the autophagy–lysosome relevant genes in PLX4720-treated A375 cells in the presence or absence of TFEB (shRNA). n = 3 independent experiments. f, g Western blot analysis (f) and densitometric quantification (g) of the LC3-II/LC3-I and p62/Actin ratios in PLX4720-treated A375 cells with shRNA-mediated depletion of the indicated genes. Expression of indicated proteins is also shown. Actin served as a loading control. n = 3 independent experiments. h, i Representative images (h) and quantification (i) of LysoTracker Red (red) and LAMP1 (green) immunostaining of PLX4720 (1 μM, 12 h-treated) A375 cells with depletion of the indicated genes. Note the reduced lysosome staining in PLX4720-treated cells upon TFEB depletion. n = 3 independent experiments. Scale bars, 10 μm. Data in a and f are from one experiment that is representative of three independent experiments. For all quantification, data represent the mean ± SD derived from the indicated number of independent experiments. Comparisons were made using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; n.s. not significant. See Supplementary Fig. 13 for uncropped data of a, f

Fig. 2

PLX4720 promotes TFEB activation through ERK inhibition. a Confocal analyses of the subcellular distribution of endogenous TFEB, TFE3, and MITF in A375 cells treated with DMSO or PLX4720 (1 μM, 12 h). Nuclei were stained with DAPI (blue). n = 3 independent experiments. b Quantification of nuclear translocation of TFEB, TFE3, and MITF in cells in (a). n = 150 cells, pooled from three independent experiments. c Immunoblots for TFEB, MITF, and TFEB3 in the cytoplasmic/nuclear fractions of A375 cells treated with PLX4720 (1 μM, 12 h). Lamin B1 is the control for the nuclear fractions, whereas LAMP1 and Tubulin are the controls for the cytoplasmic fractions. d Immunoblotting of endogenous TFEB and p-14-3-3-binding motif of TFEB from PLX4720 (1 μM, 12 h-treated) A375 cells. WCL whole-cell lysate. e PLX4720 disrupts TFEB interaction with 14-3-3. WCLs of A375 cells were immunoprecipitated (IP) with anti-TFEB, followed by immunoblotting (IB) with antibodies against 14-3-3 and TFEB. f Representative images (top) and quantification (middle) of nuclear translocation of TFEB in A375 cells stably expressing mTORC1 (E2419K), RagB GTPase (Q99L), or ERK (R67S/D321N), or DEPDC5-specific shRNA, with or without the treatment of PLX4720 (1 μM, 12 h), or ERK inhibitor FR180204 (ERKi, 10 μM, 24 h). IB showed protein expression as indicated with the corresponding mTORC1 activity (p-p70S6K and p-4E-BP1). n = 4 independent experiments. g Representative confocal images (top) and quantification (bottom) of nuclear localization of endogenous TFEB in A375 and MeWo cells treated with FR180204 or with ERK shRNA. n = 4 independent experiments. h Immunoblots for endogenous TFEB, TFE3, MITF, and ERK in cytoplasmic/nuclear fractions of A375 cells treated with DMSO or FR180204 (10 μM, 24 h) or with ERK shRNA. Scale bars, 10 μm. Data in c, d, e, f, and h are from one experiment that is representative of three independent experiments. For all quantification, data represent the mean ± SD derived from the indicated number of independent experiments. Comparisons were made using Student’s t-test. ***P < 0.001; n.s. not significant. See Supplementary Fig. 13 for uncropped data of c, e, f, h

Fig. 3

TFEB S142 phosphorylation in BRAF^V600E melanoma cells. a PLX4720 inhibits TFEB-ERK interaction and TFEB S142 phosphorylation. A375 and MeWo cells expressing TFEB-GFP were treated with PLX4720 (1 μM, 12 h), and WCLs were IP with anti-GFP-Trap beads, followed by IB with the indicated antibodies. n = 3. b PLX4720 releases ERK and TFEB from the lysosomes. n = 3. c Quantification of nuclear localization of TFEB-GFP in A375 cells expressing indicated TFEB mutants. n = 3. d In vitro phosphorylation of WT or TFEB mutants by ERK (CA). Phosphorylated TFEB were detected by autoradiography (top). The same gel was stained with Coomassie Blue to visualize total proteins (middle). Relative fold change in TFEB phosphorylation was quantified (bottom). e Representative images and quantification of cytoplasmic retention of WT, S142A, and S211A TFEB in A375 cells expressing ERK(CA) w/ or w/o treatment of PLX4720, Leptomycin B (LMB; 20 nM, 2 h), Raptor- or CRM1-shRNA. n = 3. f TFEB^S142E associates with the lysosomes regardless of the treatment of PLX4720 or Raptor-shRNA. n = 3. g S142A mutation reduces phosphorylation of the 14-3-3-binding motif in TFEB. WT or mutant TFEB-GFP was expressed in A375 cells and IP with GFP-Trap beads, followed by IB with the indicated antibody. n = 3. h The predominant role of S142 phosphorylation over S211 phosphorylation in TFEB interaction with 14-3-3. A375 cells as indicated were treated with PLX4720, followed by IP with GFP-Trap beads and IB for the indicated proteins. n = 3. i-k IB (i) and quantification of the LC3-II/LC3-I (j) and p62/Actin ratios (k) in A375 cells expressing WT/mutant TFEB w/ or w/o PLX4720 treatment. n = 3. Scale bars, 10 μm. Data in a, d, g, h, and i are from one that is representative of three independent experiments. For all quantification, data represent the mean ± SD derived from the indicated number of independent experiments. Comparisons were made using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001; n.s. not significant. See Supplementary Fig. 13 for uncropped data of a, e, h, i

Fig. 4

PLX4720 induces JNK2/p38 MAPK-dependent cytoplasmic translocation of ZKSCAN3. a Representative confocal images of subcellular translocation of endogenous TFEB (green) and ZKSCAN3 (red) in A375 cells treated with PLX4720 (1 μM, 12 h). n = 3 independent experiments. b Quantification of ZKSCAN3 cytoplasmic translocation in cells shown in (a). n = 150 cells per time point, pooled from three independent experiments. c Immunoblots for ZKSCAN3 in cytoplasmic and nuclear fractions of A375 cells treated with DMSO, PLX4720 (1 μM, 12 h), or Torin1 (1 μM, 3 h). Lamin B1 serves as the control for the nuclear fractions, whereas LAMP1 and Tubulin are the control of the cytoplasmic fractions. Note that PLX4720 induces cytoplasmic enrichment of ZKSCAN3, whereas Torin1 has no effect on its nucleocytoplasmic redistribution. d Effect of knockdown of p38, JNK1, JNK2, and PKCδ, or overexpression of the CA form of mTORC1 (E2419K) or ERK (R67S/D321N), on the subcellular translocation of endogenous TFEB (green) and ZKSCAN3 (red) in A375 cells treated with PLX4720 (1 μM, 12 h). Nuclei were stained with DAPI (blue). n = 3 independent experiments. e, f Quantification of cytoplasmic translocation of ZKSCAN3 (e) and immunoblot analyses (f) of JNK1/2 and p38 in cells shown in (d). n = 3 independent experiments. g Schematic representation of the dual role of PLX4720 in the transcriptional activation of autophagy and lysosome biogenesis/function: PLX4720 induces TFEB nuclear translocation by inhibition of ERK, whereas also triggering ZKSCAN3 cytoplasmic relocation by activation of JNK2/p38 MAPK, leading to synergistically prolonged activation of the autophagy–lysosomal pathway. See the main text for the details. Scale bars, 10 μm. Data in c and f are from one experiment that is representative of three independent experiments. For all quantification, data represent the mean ± SD derived from the indicated number of independent experiments. Comparisons were made using Student’s t-test. ***P < 0.001; n.s. not significant. See Supplementary Fig. 13 for uncropped data of c, f

Fig. 5

Effect of TFEB S142 phosphorylation on BRAF^V600E melanoma progression. a NOD/SCID mice bearing A375 xenografts tumors stably expressing vector, WT, TFEB^S142A or TFEB^S142E isolated on day 34 post-injection. b, c Tumor volume (b) and Tumor weight (g) upon autopsy at day 34 of A375 xenografts stably expressing the indicated TFEB. n = 5–6 mice per group per time point. d H&E and immunohistochemical (IHC) staining of Ki67, active Caspase 3, 4-HNE, and 8-oxo-dG in indicated A375 tumor genotypes. The levels of mitotic figures, necrosis, and multinucleated cells were quantified (right panels; n = 5–6 mice per group). N necrotic area, T tumor. Scale bars, 100 μm. e EM images of the indicated xenograft tumor genotypes. M mitochondria. Asterisks denote damaged mitochondria. Arrows denote autophagic vacuoles. Damaged mitochondria is quantified (bottom; n = 50). Scale bar, 200 nm. f IB of EMT-related factors and TGF-β/TGF-βR2/Smad2/3 signaling in indicated A375 xenograft tumor genotypes (two randomly chosen samples per group). Also shown (bottom) is the relative expression of indicated proteins in xenograft tumors. n = 3 independent experiments. g GSEA plot of A375 tumors showing the significantly changed autophagy–lysosome genes. The relevant complete GSEA data are in Supplementary Data 1a. FDR false discovery rate, NES normalized enrichment score. h Heat map depicting the expression of TGF-β target genes in indicated xenograft tumor genotypes. Each column represents a single replicate (n = 3). Also refer to Supplementary Data 1d. i IB of autophagy and EMT-related factors in A375 cells expressing TFEB^S142A or TFEB^S142E treated with TGF-β (10 ng/ml, 48 h), BafA1 (100 nM, 6 h), Beclin1 (BECN1)-shRNA, or the TGF-βRI inhibitor SB431542 (TGF-βRIi, 10 µM, 22 h). n = 3 independent experiments. Data in a, d, e are from one animal that is representative of 5–6 animals in each group. For all quantification, data represent the mean ± SD derived from the indicated number of independent experiments. Comparisons were made using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001. See Supplementary Fig. 13 for uncropped data of f, i

Fig. 6

TFEB S142 phosphorylation promotes the dissemination of BRAF^V600E melanoma cells. a Representative gross images (top row), H&E-stained section (second row), and immunohistochemical (IHC) staining of p62 and p-Smad3 in lung metastasis of B16-F10 melanoma cells stably expressing BRAF^V600E along with the indicated TFEB proteins. T metastatic melanoma tumors. b Quantification of the numbers of lung metastasis formed by B16-F10 melanoma cells as indicated (n = 5–6 mice per group; data represent mean ± SD). Comparisons were made (vs. Vector group) using Student’s t-test. ***P < 0.001. c Immunoblot analyses of E-cadherin, N-cadherin, Vimentin, active TGF-β, LC3 conversion, and p62 in indicated lung metastases (two randomly chosen samples per group; similar results were observed in all 10–12 samples per condition). Actin served as a loading control. n = 3 independent experiments. See Supplementary Fig. 13 for uncropped data. Scale bars, 500 μm. Data in c are from one experiment that is representative of three independent experiments. For all quantification, data represent the mean ± SD derived from the indicated number of independent experiments. Comparisons were made using Student’s t-test. ***P < 0.001

Fig. 7

Effect of TFEB S142 phosphorylation on PLX4720-resistant BRAF^V600E melanoma cells. a Colony formation assay of PLX4720-resistant A375^R melanoma cells stably expressing vector, WT TFEB, S142A, or S142E TFEB mutants as indicated. Bars are mean ± SD percentage of colonies for each group after 21 days. n = 3 independent experiments. b Western blot analysis of TFEB expression in cells shown in (a). Actin served as a loading control. Data are from one experiment that is representative of three independent experiments. c Tumor volume of xenografts formed after subcutaneous injection of NOD/SCID mice with A375^R melanoma cells stably expressing the indicated TFEB constructs. Results are the mean volume ± SD for 5–6 mice per group per time point. d Tumor weight from experiment shown in (c) upon autopsy at day 30. Representative images of tumor size of the indicated A375^R xenograft tumor genotypes are shown below. e Western blot analysis of autophagy (p62 and LC3-I/II), lysosome (CTSD), TGF-β/Smad2/3 signaling, EMT-related [E-cadherin (E-Cad.), N-cadherin (N-Cad.), and Vimentin, Twist1, ZEB1, Snail, and Slug], γ-H2AX, ERK and mTORC1 activation in the indicated xenografts (three randomly chosen samples per group; similar results observed in all 10–12 samples per condition). Actin served as a loading control. See Supplementary Fig. 13 for uncropped data. For all quantification, data represent the mean ± SD derived from indicated number of independent experiments. Comparisons were made using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 8

TFEB S142 phosphorylation confers BRAFi resistance in melanoma cells. a Representative images (left) and quantification (right) of the colonogenic survival of A375 cells stably expressing control shRNA (pGIPZ) or TFEB shRNA complemented with empty vector, WT TFEB, and S142A or S142E TFEB mutants that are shRNA-resistant, after treatment with DMSO or PLX4720 at the indicated concentrations. Endogenous and reconstituted TFEB expression was confirmed by immunoblotting (bottom right). n = 3 independent experiments. b Bioluminescence images (top) of tumor regression of the indicated A375^R xenograft tumor genotype in live NOD/SCID mice at the indicated time after inoculation. Radiant efficiency expressed as p/s/cm²/sr/(μW/cm²) was quantified (bottom). c Effect of PLX4720 on tumor response of xenografts formed by the indicated A375 cell lines. After tumor establishment (~ 500 mm³), mice bearing A375 xenografts were treated with PLX4720 (20 mg/kg, i.p.) daily for 21 days. Values are the mean tumor volume ± SD per time point for 6–7 mice per group. d, e Representative gross images (d) of livers with metastatic nodules (top), H&E-stained sections (second row) and IHC analysis of TGF-β and p-Smad3 of the indicated A375 xenograft genotypes 21 days after PLX4720 treatment (20 mg/kg, i.p., daily). Scale bars, 100 μm. The numbers of metastatic nodules in the liver in (d) was quantified in (e) (n = 5–6 mice per group). For all quantification, data represent the mean ± SD derived from indicated number of independent experiments. Comparisons were made using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001; n.s. not significant