Fig. 3: Derivation of robust, screen-ready, dFLASH clonal lines.

a Schematic for derivation and assessment of robustness in clonal lines, created in BioRender. Peet, D. (2025) https://BioRender.com/y10l388, of (b–e) HEK239T dFLASH-HIF (mcdFLASH-HIF), (h–k) T47D dFLASH-PGR (mcdFLASH-PGR) and (n–q) HEK293T dFLASH-synFIH (mcdFLASH-synFIH), analysed by flow cytometry, temporal HCI over 38 h and inter-plate robustness by mock multi-plate high throughput screening with HCI. b–e mcdFLASH-HIF was (b) treated with DMOG for 48 h and assessed for Tomato induction by flow cytometry relative to vehicle controls with fold change between populations stated and (c) treated with vehicle or 0.5 mM DMOG and imaged every 2 h for 38 h by HCI (mean ± sem, n = 8 biological replicates per group). d, e mcdFLASH-HIF was treated for 48 h with 1 mM DMOG or vehicle (6 biological replicates/plate, n = 10 plates) by HCI in a high throughput screening setting (HTS-HCI) for (d) normalised dFLASH expression and (e) Tomato MFI alone. f, g mcdFLASH-HIF was assessed by flow cytometry after a 24 h hypoxic (1% O₂) treatment, followed by a 4 h normoxic recovery period for (f) EGFP and Tomato and (g) Tomato alone, relative to a normoxic population (ctrl). h–k T47D mcdFLASH-PGR was (h) assessed after 48 h of treatment with 100 nM R5020 by flow cytometry for Tomato induction and (i) treated with 10 nM R5020, 35 nM E2, 10 nM DHT and vehicle then imaged every 2 h for 38 h by temporal HCI for normalised dFLASH expression (mean ± sem, n = 8 biological replicates per group). j, k T47D mcdFLASH-PGR was assessed by HTS-HCI at 48 h (48 biological replicates/plate, n = 5 plates) for (j) dFLASH normalised expression and (k) Tomato MFI alone. l T47D dFLASH-PGR cells were treated with increasing concentrations of R5020 (0.01-100 nM, n = 8 biological replicates per group, presented as mean ± sem) and imaged at 48 h to determine sensitivity to R5020. m Comparison of induction of the T47D mcdFLASH-PGR line to different steroids (10 nM R5020, 35 nM E2, 10 nM DHT, 10 nM Dex, 10 nM RA) by HCI after 48 h of treatment. m Are the mean ± sem of normalised Tomato/GFP (within each experiment) from n = 3 independent experiments (n = 8 biological replicates per group), except Dex and RA (n = 2 independent experiments (n = 8 biological replicates per group)). n HEK293T dFLASH-synFIH was assessed, with 200 ng/mL Dox +/- 1 mM DMOG by flow cytometry for Tomato induction. o mcdFLASH-synFIH was treated with 100 ng/mL Dox, 1 mM DMOG and relevant vehicle controls (0.1% water, 0.1% DMSO) and assessed for reporter induction by temporal HCI (mean ± sem, n = 4 biological replicates per group). p, q mcdFLASH-synFIH cells were treated with 200 ng/mL Dox (grey), 1 mM DMOG (red), vehicle (pink) or Dox and DMOG (orange) and assessed by HTS-HCI after 48 h (n = 8 biological replicates/plate, n = 3 plates) for (p) normalised dFLASH expression or (q) Tomato MFI induction between Dox only and Dox and DMOG treated populations. For (d, j, p) dashed lines represent 3 SD from relevant vehicle (+3 SD) or requisite ligand treated population (-3SD). Fold change for flow cytometry and HTS-HCI (FC) is displayed. Z’ was calculated from all HTS-HCI analysed plates. Z’ for all plates analysed was >0.5. e, k, q boxplots are presented as whiskers – 1.5 x IQR, box - 25th/75th percentile and median as the solid line. Source data are provided as a Source Data file.