Fig. 1: Experimental approach and primary dataset overview. | Nature Structural & Molecular Biology

Fig. 1: Experimental approach and primary dataset overview.

From: A small molecule stabilizer rescues the surface expression of nearly all missense variants in a GPCR

Fig. 1

a, A library of V2R variants was generated using SUNi mutagenesis and barcoded with random DNA barcodes. PacBio long-read sequencing was used to link variants with barcodes. Then, the library was integrated into HEK293T LLP-iCasp9-Blast landing-pad cells. Fluorescent antibody staining, followed by FACS, was used to fractionate cells by expression level, then DNA was sequenced with Illumina short reads to identify the variant abundances in each bin. Neg, negative control; Lib, library. b, Histogram of variant effect scores. Premature truncating variants are shown in red, synonymous wild-type variants are shown in green and missense variants are shown in gray. Dashed lines indicate the thresholds between well expressed, moderately expressed and poorly expressed variants. The number of missense variants in each category is reported above the graph. c, Heatmap representation of the data. V2R receptor positions are on the x axis, and mutant amino acid identities are on the y axis. Blue cells indicate surface expression below the wild-type level; gray indicates that surface expression is the same as the wild-type level; and red indicates that surface expression is above the wild-type level. d, Comparison of variant effects at positions in TM helices, or in extra- or intra-cellular loops. n = 2,620; 1,421; and 2,171 for TM, extracellular and intracellular, respectively. The center value (white circle) of each violin plot is the median; the box limits represent the first and third quartiles; and the whiskers extend to either the most extreme value, or interquartile range × 1.5, whichever is smaller. Mann–Whitney U test, two-sided, P = 3.16 × 10−135, P = 8.54 × 10−153. e, Comparison of variant effects at positions in TM helices 1–7. Violin plot features are the same as in d. f, Predicted ΔG of membrane insertion (ΔG prediction server v1.0; full protein scan; helix length, 21). g, Top, median position surface expression score displayed in color on the AlphaFold2-predicted V2R structure. Bottom, density plot of median position scores for the whole receptor (gray) or in TM helices (green). h, Top, per-position correlation of surface expression scores with Kyte–Doolittle hydrophobicity shown in color on the AlphaFold2-predicted V2R structure. Bottom, density plot of per-position correlations for the whole receptor (gray) or within TM helices (green). i, Comparison of ten variant scores measured individually or en masse (DMS measurements). The s.e.m. is shown for DMS measurements. The linear fit is shown in red. Measurement of isogenic cell lines was done once but encompasses measurements from >10,000 cells for each genotype. a.u., arbitrary units. j, Effects of substitutions with cysteine in the extra- or intra-cellular portions of the receptor. n = 69, 116 for extracellular and intracellular, respectively. The violin features are the same as in d. Mann–Whitney U test, two-sided, P = 7.3 × 10−10.

Source data

Back to article page