C-terminal truncation and knockout (KO) CALR rescue apoptosis induced by hematoxylin in CALR-mutated cells. (A) Annexin V/PI staining of Ba/F3-MPL cells after hematoxylin treatment for 24 hours. Representative fluorescence-activated cell sorting plots of annexin V/PI staining in both CALR WT and mutant cell lines; 10 and 50 ng/mL of murine IL-3 were used, and 10 ng/mL of human TPO (hTPO) was used. (B) Quantification of apoptotic cells (annexin V⁺; n = 3). Data are shown as mean ± standard deviation. The statistical test was conducted by 1-way analysis of variance followed by Bonferroni’s multiple comparison tests. WT cells were tested in the presence of 10 and 50 ng/mL of murine IL-3 or 10 ng/mL of hTPO. Mutant cell lines were tested in either absence of cytokine or presence of 10 and 50 ng/mL of murine IL-3 or 10 ng/mL of hTPO. (C) KO of CALR in Ba/F3-MPL cells shown by western blot (top). Dose response of Ba/F3-MPL CALR WT and CALR KO cell lines (bottom). CALR KO cell lines were generated by CRISPR-Cas9 targeting exon 3 of CALR gene from Ba/F3-MPL cell lines; 3 individual KO clones and 2 WT clones were used in the experiment (n = 5). (D) Sequence alignment of the C-terminal sequences of WT CALR, disease frameshift mutant CALR (CALR del79 and CALR del37 as 2 examples), and C-terminal–truncated CALR (CALR del2 and CALR ins1 as 2 examples) from Ba/F3 cells. Nonmutated sequences (green), disease frameshift mutant–specific sequences (red), WT-specific sequences (blue), and nondisease frameshift mutant sequences (brown) are indicated (top). Dose response of Ba/F3-MPL cell lines carrying WT CALR, disease frameshift mutant CALR, and C-terminal–truncated CALR (bottom; n = 3). ****P < .0001.