Extended Data Fig. 1: UTX undergoes cIDR-dependent phase separation. | Nature

Extended Data Fig. 1: UTX undergoes cIDR-dependent phase separation.

From: UTX condensation underlies its tumour-suppressive activity

Extended Data Fig. 1

a. Relative expression of UTX 555* mutant in 10 cancer patient samples that harbor this mutation and for which the mRNA level data are available. From cBioportal. Each bar is a patient sample. z-score indicates the number of standard deviations away from the mean of expression in the reference. It shows that 3 samples showed increase in mRNA level, and 4 samples showed reduction within 1 standard deviation of expression in reference samples, suggesting that this nonsense mutation does not usually lead to transcript decay and should express a truncated UTX (1–554). b. Top, a schematic of the full-length UTX. Bottom, a plot for number of non-synonymous mutations (type indicated in legend) within each 100 amino acids-long region on UTX, from 394 patients in TCGA. The plot is aligned with the top schematic. c. Top, a schematic showing division of the full-length UTX into 5 regions that were individually purified as fusion with Maltose-binding protein (MBP, to enhance solubility). See Constructs in Methods for rationale for region division. Bottom, phase contrast images of 50 µM MBP-UTX fragments corresponding to each indicated UTX region in 30 mM NaCl after treatment with TEV protease for 1 hr to remove the MBP tag. d. Coomassie blue staining image of purified EGFP and EGFP-cIDR. e. Fluorescence (left) and DIC (right) microscopy images of 30 µM EGFP-cIDR at increasing concentrations of NaCl at indicated times after dilution in the condensation buffer. Note the increase of droplet size and the occasional appearance of irregular shape at 45 min, suggesting growth and aging phenomena that are typical for phase separation. f. Fluorescence (left) and DIC (right) images showing wetting of 30 µM EGFP-cIDR at the surface of the coverslip immediately after dilution. g. Coomassie blue staining image of purified EGFP-UTX (419-848). h. Fluorescence (up) and DIC (bottom) images of EGFP-UTX (419–848) at indicated concentrations. i. Coomassie blue staining image of purified mEGFP and mEGFP-UTX (419–848). j. Fluorescence (up) and DIC (bottom) images of mEGFP and mEGFP-UTX (419–848) at indicated concentrations and time points after dilution in condensation buffer. k. Coomassie blue staining image of purified mCherry and mCherry-UTX (419–848). Note that, throughout this manuscript, a band of unknown identity, indicated by the asterisk, always appeared at ~15kD smaller than the purified mCherry or mCherry-tagged protein of interest, regardless of the identity of the protein of interest. We thus speculate that it is either a degradation product of mCherry or an unrelated protein tightly associated with mCherry. l. Fluorescence (up) and DIC (bottom) images of 40 µM mCherry and mCherry-UTX (419–848) at indicated time points after dilution. m. Immunoblotting by indicated antibody of total lysates from untransfected control or 293T cells transfected with indicated constructs that are fused to EGFP. n. Quantification of nuclear UTX concentration by anti-UTX immunoblotting. Total lysates from indicated number of 293T and mouse ES cells were loaded, along with indicated ng of purified EGFP-UTX (1–848). We determined that the nuclear concentration of endogenous UTX to be ~153 nM in 293T cancer cells and ~383 nM in mouse ESCs. o. Transfected UTX forms nuclear foci at a concentration that is on the same order of magnitude as the endogenous level. Left, immunoblotting (top, by anti-UTX antibody) and Ponceau S staining (bottom) of total lysates from untransfected control or 293T cells transfected with a very low dose of plasmid for UTX-EGFP, with a transfection efficiency of ~30% as examined under microscope. Three different amounts (1:2 serial dilution) of the lysates and purified EGFP-UTX (1-848) were loaded as indicated. UTX-EGFP amount was roughly 2 times that of the endogenous UTX, and estimated to be < 1 µM in the transfected cells, assuming that the negative cell under microscope had little contribution to the UTX-EGFP immunoblotting signal. The fluorescence image of the corresponding cells is shown on the right.

Back to article page