Journal: Cell systems

Article Title: Profiling the Non-genetic Origins of Cancer Drug Resistance with a Single-Cell Functional Genomics Approach Using Predictive Cell Dynamics.

doi: 10.1016/j.cels.2020.08.019

Figure Lengend Snippet: Figure 1. Constant Proportioning of Cell Response Heterogeneity Enables Stable Predictive Metrics of Drug Response for Phenotype- Coupled Same-Cell Profiling: Fate-Seq (A) Clonal expansion and drug selection reveal the plasticity of drug response phenotypes in isogenic cells. For clonal expansion: single-cell clones were obtained by limiting dilutions of the HeLa parental cells. For drug selection: parental HeLa cells were treated with TRAIL (25 ng/mL for 15 h). The surviving cells from the first treatments were allowed to regrow for 72 h. Once the surviving cells and the sub-clones had regrown into new cultures (72 h and 3 weeks, respectively), they were all treated with TRAIL at several doses, and their responses were compared with the parental cells. One representative dose response profile is shown for both clonal expansion and drug selection experiments. Bar graphs represent cell viability from at least three biological repeats of both clonal expansion and drug selection experiments. Cell viability was assessed 15 h post-treatment by CellTiter-Glo, and the absorbance is represented in percent of control relative to cell population treated with TRAIL-vehicle, data are represented as mean ± SD. (B) Single-cell trajectories of FRET ratio reporting caspase-8 activity. Live-cell microscopy experiments (> 10 h) were performed to establish the caspase-8 dynamics predicting the ultimate cell fate after treatment with TRAIL (25 ng/mL in HeLa cells stably expressing the FRET biosensor IC-RP; Albeck et al., 2008). One representative experiment of 143 tracked cells is presented. (C) The caspase-8 activation rate (time derivative of FRET ratio, dFRET ratio/dt; Roux et al., 2015) after TRAIL stimulation is calculated of for each single cell: we found a time after treatment (50 min) at which caspase-8 activation rate could accurately predict the ultimate cell response at the end of the experiment, using constant thresholds (colored lines; measured in at least three experiments). The star indicates the top 4 predicted cells of each group that can be processed in fate-seq (Figure 2). The cell fate prediction being done after 50-min TRAIL stimulation limits gene induction (Figure S1B) and allows to isolate the sensitive cells that would otherwise die. (D) Example live-cell microscopy experiment for caspase-8 early activation measurement in each identified cell. 50-min live-cell microscopy experiment was performed to collect caspase-8 dynamics for each cell cultured on a membrane ready for micro-dissection. About 20 fields of cells are monitored, and all cells are tracked for their FRET signal in time. (E) Single-cell trajectories of FRET ratio are reported for each cell tracked to perform the fate prediction based on their caspase-8 activation rate (dFRET ratio/dt) 50-min post-TRAIL treatment (25 ng/mL, see Figure 1C). One representative experiment of 158 tracked cells is presented. Red and blue trajectories with green dash lines were predicted sensitive and predicted resistant cells, respectively. Note that a single measure of the FRET ratio at 50 min would not be sufficient to

Article Snippet: HeLa cells (derived from human, female cervix adenocarcinoma) were obtained from the ATCC and cultured in DMEM Glutamax supplemented with 10% fetal bovine serum and penicillin/streptomycin (Thermo Fisher Scientific, France) or in phenol red-free DMEM Glutamax supplemented with 10% fetal bovine serum and penicillin / streptomycin (Thermo Fisher Scientific, France) during live-cell imaging.

Techniques: Selection, Clone Assay, Control, Activity Assay, Microscopy, Stable Transfection, Expressing, Activation Assay, Cell Culture, Membrane, Dissection

Journal: Cell systems

Article Title: Profiling the Non-genetic Origins of Cancer Drug Resistance with a Single-Cell Functional Genomics Approach Using Predictive Cell Dynamics.

doi: 10.1016/j.cels.2020.08.019

Figure Lengend Snippet: Figure 2. Fate-Seq Recovers a Drug-Efficacy Gene Signature from Unstructured Variability: Profiling TRAIL Response as a Proof of Principle Cells selected based on their predicted phenotype 50-min post-TRAIL treatment (25 ng/mL) were assayed by single-cell RNA sequencing. (A) Principal component analysis of the single-cell RNA-seq. Each dot represents a cell colored by its predicted phenotype. ‘‘High’’ refer to cells in (B). (B) TRAIL sensitivity signature of HeLa cells: hierarchical clustering of genes (FDR < 0.1 and |log2(FC)| > 2) from selected cells highlights the heterogeneity of cancer drug response information contained in isogenic cells treated together. ‘‘High’’ and ‘‘low’’ refer to the cell identifier in dataset. (C) Volcano-plot: x axis represents the difference in average gene expression between predicted sensitive and predicted resistant cells (log2 scale). y axis represents the FDR (log10 FDR). Red dots represent the gene candidates for functional validations. (D) MA-plot: x axis represents the average gene expression for each gene (log2 CPM + 1). y axis represents the difference in gene expression between predicted sensitive (high) versus predicted resistant cells (low), log2 scale. Red dots represent the gene candidates for functional validations.

Article Snippet: HeLa cells (derived from human, female cervix adenocarcinoma) were obtained from the ATCC and cultured in DMEM Glutamax supplemented with 10% fetal bovine serum and penicillin/streptomycin (Thermo Fisher Scientific, France) or in phenol red-free DMEM Glutamax supplemented with 10% fetal bovine serum and penicillin / streptomycin (Thermo Fisher Scientific, France) during live-cell imaging.

Techniques: RNA Sequencing, Gene Expression, Functional Assay