Fig. 2: Anti-apoptotic MCL-1 is required for tumour growth and survival in an orthotopic GBM PDX model.

A Representative images of neurosphere growth from G7 GSC vector^CRISPR (upper panel), MCL-1.1^CRISPR (middle panel) and MCL-1.2^CRISPR (lower panel). B Effects on self-renewal in vitro were assessed by an extreme limiting dilution assay (ELDA) for neurosphere formation frequency by G7 GSC vector^CRISPR vs. MCL-1.1^CRISPR or MCL-1.2^CRISPR. Error bars represent 95% confidence interval of n = 3 independent experiments (nonsignificant (ns) p = 0.908, *p = 0.00795, calculated using the ELDA website http://bioinf.wehi.edu.au/software/elda/ [69]). C Proliferation assay of indicated cell lines using IncuCyte live cell imaging (percentage cell density over 6 days). Error bars represent mean ± SD from n = 3 independent experiments. D Representative images of brain MRI scans (tumour indicated by red dashed line) next to corresponding pseudocolour representations of iRFP signal of mice bearing iRFP tagged G7 GSC vector^CRISPR (upper panel) and MCL-1.2^CRISPR (lower panel) xenografts, respectively. iRFP signal was detected by PEARL scans (700 nm channel) (1) at week 8 and (2) at week 20 (vector^CRISPR) or week 36 (MCL-1^CRISPR) post injection. E Quantification of time to 20% iRFP signal increase of G7 vector^CRISPR n = 13 vs. G7 MCL-1^CRISPR tumours n = 19, compared to four weeks post injection (baseline signal). Error bars represent mean ± SD (**p = 0.0013) Mann–Whitney. F Kaplan–Meier survival graph of mice with orthotopic xenografts of G7 GSC iRFP vector^CRISPR n = 9 (median survival 87 days) vs. MCL-1^CRISPR tumours n = 10 (median survival not definable, nd) post tumour cell implantation (****p < 0.0001) Log-rank (Mantel–Cox) test.