The Hippo pathway effectors YAP and TAZ promote cell growth by modulating amino acid signaling to mTORC1

Abstract

YAP and TAZ are transcriptional co-activators and function as the major effectors of the Hippo tumor suppressor pathway, which controls cell growth, tissue homeostasis, and organ size. Here we show that YAP/TAZ play an essential role in amino acid-induced mTORC1 activation, particularly under nutrient-limiting conditions. Mechanistically, YAP/TAZ act via the TEAD transcription factors to induce expression of the high-affinity leucine transporter LAT1, which is a heterodimeric complex of SLC7A5 and SLC3A2. Deletion of YAP/TAZ abolishes expression of LAT1 and reduces leucine uptake. Re-expression of SLC7A5 in YAP/TAZ knockout cells restores leucine uptake and mTORC1 activation. Moreover, SLC7A5 knockout cells phenocopies YAP/TAZ knockout cells which exhibit defective mTORC1 activation in response to amino acids. We further demonstrate that YAP/TAZ act through SLC7A5 to provide cells with a competitive growth advantage. Our study provides molecular insight into the mechanism of YAP/TAZ target genes in cell growth regulation.

Figure 1

YAP and TAZ modulate AA-induced mTORC1 activation. (A) Glutamine (Gln) potentiates mTORC1 activation by leucine (Leu). HEK293A cells were AA starved for 6 h, then Gln stimulated for 30 min followed by 10 min of Leu stimulation. Concentrations of 0.01×, 0.1×, and 1× denote 1%, 10%, and 100%, respectively, of the leucine concentration in normal DMEM medium. mTORC1 activity is determined by western blotting for the phosphorylation of S6K, S6, and 4E-BP1. L.e. and s.e. denote long exposure and short exposure, respectively. (B) YAP/TAZ are required for mTORC1 activation by AAs. Wild-type and YAP/ TAZ double knockout (Y/T dbKO) 293A cells were co-cultured, AA starved, and stimulated with Gln and Leu as indicated. Cells were AA starved (top row), or stimulated by 1× Gln (second row), or 1× Gln and 0.1× Leu (Gln/Leu, third row), or treated with rapamycin and Gln/Leu (bottom row). Some Y/T dbKO cells, as indicated by the lack of YAP/TAZ labeling, are marked with white arrows. Note, that upon Gln/Leu stimulation only wild-type (positive YAP/TAZ labeling) cells show positive pS6 signal, a marker of active mTORC1. See Supplementary information, Figure S1D for full size image. Scale bar, 20 μm. (C) AA-induced S6 phosphorylation depends on YAP/TAZ. The fraction of pS6-positive cells per image of each genotype was quantified in experiments similar to B. n >75 cells of each genotype, per treatment. #1 and #2 denotes two independent Y/T dbKO clones. (D) YAP and TAZ dictates glutamine-potentiated leucine stimulation of mTORC1. Western blots of cell lysates from 293A WT cells and Y/T dbKO cells. Cells were AA starved for 6 h and then stimulated with 1× Gln followed by 0.1× Leu, or only 0.1× Leu. The western blot was probed to assess mTORC1 activity. Western blots were also performed for expression of YAP, TAZ, and the LAT1 high-affinity leucine transporter (comprised of SLC7A5 and SLC3A2). Cyr61 is a known target gene of YAP/TAZ, whereas vinculin (Vinc) serves a loading control.

Figure 2

YAP and TAZ regulates leucine uptake via SLC7A5. (A) YAP and TAZ are essential for LAT1 expression. LAT1 is a hetero dimeric transporter complex composed of SLC7A5 and SLC3A2, which are linked via a disulfide bond. Cell lysates from different cell lines as depicted in the box to the right were prepared in either non-reducing buffer (no β-mercaptoethanol) or reducing buffer (containing β-mercaptoethanol). Note that both the SLC7A5 antibodies and the SLC3A2 antibodies detect the high molecular heterodimeric complex under non-reducing condition (left part). Under the reducing condition (right half), both SLC7A5 and SLC3A2 run at the expected molecular weights of individual monomers. Ectopic expression of YAP (sample #a) rescues protein expression of both SLC3A2 and SLC7A5 in Y/T dbKO cells. (B) YAP/TAZ knockout decreases leucine uptake. H³-leucine uptake assay was carried out in cell lines (denoted in the right panel of Figure 2A) and normalized to WT cells. Error bars are means ± SEM of triplicates from a representative experiment. (C) LATS1/2 regulate leucine uptake via SLC7A5. Stable cell lines (right panel in Figure 2A) were generated and uptake of H³-leucine is displayed as relative to WT cells. Error bars are means ± SEM of triplicates from a representative experiment. (D) YAP/TAZ do not affect the plasma membrane localization of SLC7A5. WT and Y/T db KO cells that both stably express V5 epitope tagged SLC7A5 were co-cultured. YAP/TAZ WT cells are marked by positive YAP signal and dbKO cells (no YAP signal) are highlighted with white arrows. Ectopic SLC7A5 is localized to the plasma membrane in both WT and YAP/TAZ dbKO cells. Scale bar, 10 μm. (E) Reintroduction of SLC7A5 rescues the expression of SLC3A2 in YAP/TAZ knockout cells. Western blots of cell lysates from different cell lines as depicted. Re-expression of V5-tagged SLC7A5 or the epithelial specific SLC7A7, which likewise complexes with SLC3A2, but is not endogenously expressed in 293A cells, rescues the protein expression of SLC3A2. See also Supplementary information, Figure S3B. (F) Reintroduction of SLC7A5 restores leucine uptake in YAP/TAZ knockout cells. H³-leucine uptake assay was carried out in the indicated cell lines. #1 and #2 denotes two independent Y/T dbKO clones. SLC7A5 was re-expressed in Y/T dbKO clone #1. Error bars are means ± SEM of triplicates from a representative experiment.

Figure 3

SLC7A5 is required for AA signaling to mTORC1. (A) SLC7A5 is involved in Gln/Leu-induced mTORC1 activation. Wild-type and SLC7A5 knockout (KO) 293A cells were AA starved for 6 h and Gln stimulated for 30 min followed by 10 min of Leu stimulation at different concentrations. SLC7A5-deficient cells, which are confirmed by the lack of SLC7A5 expression, are impaired in mTORC1 activation at low Leu concentrations. Similar results were observed in an additional cell line independently generated (data not shown). (B) SLC7A5 does not affect insulin-induced mTORC1 activation. Experiments were similar to A except that cells were serum starved and stimulated with 100 nM insulin for the times illustrated, with or without 1 h pretreatment of mTOR inhibitors Torin (Tor) or Rapamycin (Rapa). Note also the lack of SLC3A2 expression in SLC7A5 KO cells. Similar results were observed in an additional cell line independently generated with a separate sgRNA targeting SLC7A5 (data not shown). (C) Representative confocal image of Gln/Leu-stimulated 293A WT cells, labeled for pS6 and stained for DAPI. Scale bar, 20 μm. (D) Confocal image of Gln/Leu-stimulated SLC7A5 KO 293A cells, labeled for pS6 and stained for DAPI. Samples and images in C and D were processed in parallel and acquired with the same microscope settings. Scale bar, 20 μm. (E) Leucine-induced mTORC1 activation is diminished in SLC7A5 KO cells. Quantification of confocal images as in C and D, and plotted as fraction of pS6-positive cells per image. The treatments were AA starved, AA starved then 1× Gln stimulated, or Gln/Leu stimulated, or pretreated with rapamycin for 1 h and then stimulated with Gln/Leu. (F) SLC7A5 is essential for efficient leucine uptake. H³-leucine uptake assay was carried out for three independently generated SLC7A5 KO clones and normalized to WT. Error bars are means ± SEM of triplicates from a representative experiment.

Figure 4

SLC7A5 is a direct YAP/TAZ TEAD target gene to promote leucine uptake. (A) SLC7A5 expression is dependent on YAP/TAZ. qPCR of SLC7A5 mRNA levels in 293A WT cells compared to Y/T dbKO cells. (B) LATS1/2 dbKO increases SLC7A5 expression. qPCR of SLC7A5 mRNA levels from 293A WT cells compared to LATS1/2 dbKO cells. (C) The promoter of SLC7A5 contains several putative TEAD recognition motifs. Diagram depicting the upstream 2 kB of the SLC7A5 gene. TSS: transcription start site. ATG: start codon. Potential TEAD recognition motifs (CATTCC) are depicted with arrow heads. (D) YAP drives SLC7A5 promoter activity. The short (−1 075 to 0) or long (−2 000 to 0) fragment of SLC7A5 promoter region was fused to produce the luciferase reporters. These reporters were transfected into Y/T dbKO 293A cells in combination with YAP or vector control. Luciferase activity was measure and normalized to the co-transfected renilla control. Note that only the long form contains the predicted TEAD recognition motifs (upstream of −1 075). Data are means ± SD from triplicates in a representative experiment. (E) TEAD1 binds to the SLC7A5 promoter region. Chromatin immunoprecipitation (CHIP) with antibodies to endogenous TEAD1 (or control IgG) were carried out in 293A cells. The precipitated DNA was quantitated by real-time PCR analysis with primers specific for a promoter region (prom) or an intragenic control region (intra) of the indicated genes. CTGF (connective tissue growth factor) is a known direct target gene of the YAP-TEAD complex, and GAPDH (glyceraldehyde 3-phosphate dehydrogenase) serves as an additional negative control. Data are means ± SEM of triplicates from a representative experiment. (F) Ectopic expression of active YAP or TAZ increases SLC7A5 mRNA expression. qPCR from Y/T dbKO cells expressing doxycycline-inducible active YAP/TAZ as depicted. The cells were serum starved and doxycycline-induced for 16 h. Data are means ± SEM of triplicates from a representative experiment. (G) Ectopic expression of active YAP or TAZ increases SLC7A5 protein levels. Experiments are similar to F. ^* denotes antibody (sc-101199) that recognizes both YAP and TAZ. (H) SLC7A5 expression is dependent on YAP-TEAD interaction. Western blots of cell lysates from Y/T dbKO cell lines stably re-expressing YAP, or the TEAD binding-deficient YAP mutant (S94A). Note that the TEAD binding-deficient S94A YAP does not rescue the expression of SLC7A5. (I) Leucine uptake is dependent on YAP-mediated TEAD activation. H³-leucine uptake assay was carried out on Y/T dbKO cell lines with stable expression of vector control, YAP, and the TEAD binding-deficient YAP S94A mutant. Error bars are means ± SEM of triplicates from a representative experiment.

Figure 5

YAP/TAZ act through SLC7A5 to promote cell growth at low AA levels. (A) Ectopic SLC7A5 expression restores mTORC1 activation by AAs in Y/T dbKO KO cells. Y/T dbKO cells stably expressing vector or V5-tagged SLC7A5 were co-cultured. Cells were AA starved and then stimulated with Gln/Leu in the absence (upper panels) or presence (lower panels) of rapamycin. Note that SLC7A5 re-expression (marked by positive V5 staining, red) rescues pS6 stimulation upon Gln/Leu stimulation. See Supplementary information, Figure S4A and S4B for larger field of cells. Scale bar, 20 μm. (B) Ectopic SLC7A5 expression rescues mTORC1 activation by AAs. Experiments are similar to A. mTORC1 activation is determined by western blotting for phosphorylation of S6K and 4E-BP1. (C) Re-expression of SLC7A5 rescues S6K activation. Quantification of pS6K signals after 10 min of Leu stimulation from experiments as in B. Error bars are means ± SEM of triplicates. (D) SLC7A5 does not alter insulin-induced mTORC1 activation. Cells were serum starved and then stimulated with 100 nM insulin for the times indicated. Tor denotes 1-h pretreatment with Torin prior to insulin stimulation. (E) YAP/TAZ act through SLC7A5 to control cell size. The indicated cell lines were cultured for 36 h in normal DMEM (AA+), 10% concentration of all AAs, or normal concentration of all AAs except leucine, which was 1%. The relative cell size was determined by FACS and normalized to WT cells grown in full media. Data depicted is the average normalized cell size from four replicates. Error bars are means ± SEM from a representative experiment. (F) Y/T dbKO cells ectopically expressing SLC7A5 outgrow Y/T dbKO cells at low AA concentrations. Representative confocal images of cells labeled for V5 and stained with DAPI. Y/T dbKO cells expressing vector control or V5-tagged SLC7A5 were seeded at low density and grown for 144 h with medium change every 24 h, whereafter cells were seeded into LabTek chambers and processed for immunofluorescence on the following day. Note that SLC7A5 re-expressing cells outgrow Y/T dbKO control cells at low AA (middle row) and especially low leucine concentrations (bottom row), but not at full AA concentrations (top row). See Supplementary information, Figure S3D for pure cell cultures. Scale bar, 50 μm. (G) Expression of SLC7A5 confer cell growth advantages at low AA concentrations. Cell competition assay in co-culture of Y/T dbKO cells expressing either empty vector or V5-tagged SLC7A5 grown under different conditions for 144 h as in F. Medium was changed every 24 h to keep serum concentration within a similar range. The ratio of V5-expressing cells at the end of the experiment was quantified using FACS and data depicted are means ± SEM from a representative experiment.