Fig 2: PIPKIα and IPMK control the expression of YAP/TAZ target genes by physical association.(A) The indicated proteins were knocked down in MDA-MB-231 cells by transfecting siRNAs for 72 h and protein levels were analyzed by immunoblotting. Representative immunoblots of n = 3 independent experiments are shown. Similar to YAP knockdown, knockdown of either PIPKIα or IPMK strongly reduced the expression of the YAP targets CTGF and CYR61 at the protein level while p53 knockdown had no effect. (B) MDA-MB-231 cells were transfected with non-targeting siRNA (siControl) vs. siRNA against human PIPKIα. The mRNA levels of the indicated YAP/TAZ target genes were analyzed by RT-qPCR. The graph is shown as mean ± s.d. of n = 3 independent experiments. *P < 0.05; **P < 0.01, and n.s.; not significant in Student’s t test. (C) PIPKIα and IPMK were knocked down in MDA-MB-231 cells and the mRNA levels of CTGF and CYR61 were analyzed by RT-qPCR. The graph is shown as mean ± s.d. of n = 3 independent experiments. Knockdown of either PIPKIα or IPMK reduced the expression of CTGF and CYR61 at the mRNA level. *P < 0.05; **P < 0.01, and n.s.; not significant in Student’s t test. (D, E) Vector control (Mock) or Flag-tagged WT or mtWW YAP were co-transfected with HA-tagged PIPKIα (D) or Myc-tagged IPMK (E) in HEK293 cells for 48 h. PIPKIα and IPMK were immunoprecipitated with the indicated antibodies and the associated YAP was analyzed by immunoblotting. Representative immunoblot images of n = 3 independent experiments are shown. Inactivating the WW domains of YAP prevented it from binding PIPKIα and IPMK. (F, G) WT HA-PIPKIα (F) or WT Myc-IPMK (G) or the LPXAA mutants were co-transfected with Flag-YAP in HEK293 cells for 48 h. YAP was immunoprecipitated with an anti-Flag antibody and the associated PIPKIα and IPMK were analyzed by immunoblotting. Representative immunoblot images of n = 3 independent experiments are shown. Mutating the LPXAA motifs in PIPKIα and IPMK strongly reduced their association with YAP. Source data are available online for this figure.

Fig 3: The association of TAZ with TEAD1 is facilitated by phosphoinositides, and the binding of phosphoinositides does not alter the nuclear localization of YAP nor its binding to PIPKIα and IPMK.(A) The schematic diagrams show the polybasic motifs (PBM) located within the TEAD-binding domains (TBD) of YAP and TAZ that may mediate their binding to phosphoinositides. DBD, DNA-binding domain; TAD, transactivation domain; Y/TBD, YAP/TAZ-binding domain in the TEADs. (B) A Coomassie-stained gel of the three recombinant proteins used in the study is shown. The positions of the full-length proteins are indicated by arrows. (C) A detailed protocol used for the in vitro binding assays with lipids is presented. Note that the lipids and Triton X-100 were used at concentrations below their critical micelle concentrations. (D) 0.02 μM His6-TEAD1 was incubated with 0.05 μM GST-TAZ in the absence or presence of the indicated lipids (0.4 μM). TAZ was pulled down with glutathione beads and the associated TEAD1 was analyzed by immunoblotting. The graph shows the mean ± s.d. of n = 3 independent experiments. *P < 0.05; **P < 0.01, and n.s.; not significant in Student’s t test. (E) The chemical structures of the lipids used in the study are shown. (F) Flag-tagged WT YAP and the 2Q and 3Q mutants were transiently expressed in MDA-MB-231 cells. Cells were fixed and exogenous YAP proteins were visualized by immunostaining with an anti-Flag antibody. The images were obtained by widefield epifluorescence microscopy. DAPI staining was used to distinguish the nucleus from the cytoplasm. Representative immunostaining images of n = 2 independent experiments are shown. Scale bar, 75 μm. All the forms of YAP localize to the nucleus. (G, H) 0.1 μM GST alone, GST-WT, GST-2Q, and GST-3Q YAP recombinant proteins were incubated with 0.05 μM His6-tagged recombinant PIPKIα (B) or IPMK (C). The YAP proteins were pulled down with glutathione beads and the associated PIPKIα and IPMK were analyzed with immunoblotting. Representative immunoblot images of n = 3 independent experiments are shown. Both PIPKIα and IPMK bound equally well to all the forms of YAP.

Fig 4: The binding of phosphoinositides to YAP controls its association with the TEADs and the expression of the YAP target gene CTGF.(A) The schematic diagram shows the location of the PBM in YAP. The amino acid sequences of the WT PBM and the phosphoinositide-binding-defective mutants 1Q, 2Q, and 3Q are shown. (B) In all, 0.5 μM WT GST-YAP and the 1Q, 2Q, and 3Q mutants were incubated with 10 volume % of the indicated PIP Beads (Echelon Biosciences). After washing, the amount of YAP bound to the phosphoinositide beads and remained unbound in supernatant were analyzed by immunoblotting. Representative immunoblot images of n = 3 independent experiments are shown (bottom). A representative Coomassie-stained image of the recombinant proteins used is shown (top). The binding of PI(4,5)P2 and PI(3,4,5)P3 to 1Q and 2Q mutants was partially reduced, while the 3Q binding was completed abrogated compared to WT YAP. (C) Overall, 0.02 μM His6-TEAD1 was incubated with 0.05 μM GST-YAP in the absence or presence of the indicated synthetic lipids (0.4 μM). YAP was pulled down with glutathione beads and the associated TEAD1 was analyzed by immunoblotting. The graph shows mean±s.d. of n = 3 independent experiments. In the presence of PI(4,5)P2 and PI(3,4,5)P3, TEAD1 associated with WT YAP but not the mutants. *P < 0.05; **P < 0.01, and n.s.; not significant in Student’s t test. (D) Flag-tagged WT YAP and the 2Q and 3Q mutants were co-transfected with Myc-TEAD4 in HEK293 cells. Exogenous TEAD4 was immunoprecipitated and the associated YAP proteins were analyzed by immunoblotting with an anti-Flag antibody. Representative immunoblot images of n = 3 independent experiments are shown. Only WT YAP co-immunoprecipitated with TEAD4. (E) Flag-tagged WT YAP and the 2Q and 3Q mutants were co-transfected with HA-p73 in HEK293 cells. Exogenous p73 was immunoprecipitated and the associated YAP proteins were analyzed by immunoblotting with an anti-Flag antibody. Representative immunoblot images of n = 3 independent experiments are shown. All the forms of YAP associated with p73. (F) Flag-tagged WT YAP and the 2Q and 3Q mutants were transfected in HEK293 cells. Expression of endogenous CTGF was measured by immunoblotting. Representative immunoblot images of n = 3 independent experiments are shown. The 2Q and 3Q mutants are impaired in their ability to promote the expression of CTGF. Source data are available online for this figure.

Fig 5: The binding of phosphoinositides to YAP controls breast cancer cell motility.(A–C) Endogenous YAP was depleted in MDA-MB-231 by siRNAs. After 24 h WT YAP and the 2Q and 3Q mutants were then ectopically expressed for an additional 48 h. The expression of CTGF was determined by immunoblotting. Only WT YAP fully rescued the expression of CTGF. Serum, laminin, and vitronectin-induced cell migration was measured using Transwell inserts (A). Only WT YAP fully restored the migration rates. The relative apoptosis of the cells shown in (C) was measured using the Caspase-Glo assay (Promega). As an assay control, 1 μM staurosporine was used (B). None of the constructs promoted apoptosis. The relative proliferation of the cells was measured using the Cell-Titer Glo assay. As an assay control, 1 μM staurosporine was used (C). None of the constructs significantly altered the proliferation rate nor apoptosis. The graphs show the mean±s.d. of n = 3 independent experiments. *P < 0.05; **P < 0.01, and n.s.; not significant in Student’s t test. (D) Normal and breast cancer biopsy tissues were stained with anti-PIPKIα and anti-IPMK antibodies. The intensity of staining was histologically scored using a scale of 0, 1, 2, and 3 (Appendix Fig. S3A,S3B). The breast cancers were divided into subtypes ER+, HER2+, PR+, and triple-negative. The graph shows the distribution of the scores. The expression of both PIPKIα and IPMK was higher in all the subtypes of breast cancer. (E) A schematic representation of how PIPKIα and IPMK regulate the YAP/TAZ–TEAD pathway in breast cancer. In response to stimulation via integrins (αβ), growth factor receptors (GF), and G-protein coupled receptors (GPCR), YAP/TAZ translocate into the nucleus. In the nucleus, PIPKIa and IPMK form a complex with YAP/TAZ. In the complex, phosphoinositides generated by PIPKIα and IPMK bind to and in turn facilitate the association of YAP/TAZ with the TEADs. The nuclear phosphoinositides stimulate the expression of YAP/TAZ target genes, promoting breast cancer cell motility. Source data are available online for this figure.