Sequence alignments of the C-terminal residues of murine and human calreticulin (mCRT and hCRT), MPN-linked CRT mutants Ins5 and Del52.

Multiple sequence alignment using Clustal Omega was employed to compare C-domain sequences of mature mCRT, hCRT, Del52, and Ins5 proteins. Acidic residues (340 – 349) previously implicated in low affinity calcium binding by mCRT that are shared between wild-type and both CRTDel52 and CRTIns5 are highlighted in green. Additional acidic residues shared just between wild-type CRT and CRTIns5 are shown in red. There are additional acidic residues unique to wild-type CRT. Mature protein numbering is used for the CRT sequences.

Purification of recombinant human wild-type CRT, CRTDel52, and C-domain truncation mutants.

A) Representative chromatograms of recombinant GB1-and his-tagged human CRTWT (hCRTWT), CRTDel52, and the C-domain truncation mutants hCRT (1-351) and hCRT (1-339) purified by gel exclusion chromatography after expression in E. coli and initial purification using a nickel resin. B) Representative Coomassie blue-stained SDS-PAGE gels of purified proteins that were used for calcium-binding measurements.

Calcium binding to the low affinity sites of purified human CRT proteins.

Low affinity calcium binding by purified hCRTWT, CRTDel52, and the indicated truncation mutants was measured using isothermal titration calorimetry. A-D) Representative isotherm and fitting curves of calcium binding to the low affinity sites of the wild-type and mutant CRT proteins as indicated, measured using isothermal titration calorimetry (ITC). Prior to titration, the proteins were incubated with 50-100 µM CaCl2 to block the high affinity calcium binding sites. At 37°C, 10 mM CaCl2 was titrated against 100 µM protein. The heat change was measured for 25 injections, and the data were curve-fitted to calculate the dissociation constant, enthalpy, and binding stoichiometry using NanoAnalyzer software. Data replicates and statistical analyses are specified in Table 1.

Calorimetric data on measurement of calcium binding to the low affinity binding site of recombinant purified CRT.

ER calcium measurements in CRT-KO HEK293T cells and those reconstituted with wild-type CRT, CRTDel52, or CRTDel52-KDEL.

A) CRISPR-Cas9 editing of HEK cells using lentiviral constructs of calreticulin-targeting gRNA (CRT-KO) or empty vector (HEK-Vec). CRT-KO HEK cells were subsequently reconstituted with wild-type human CRT, CRTDel52, or CRTDel52 with a C-terminal KDEL sequence as indicated. Representative blots showing expression levels of CRT in the indicated HEK cells. Cell lysates were subjected to SDS-PAGE followed by immunoblotting. Wild-type CRT, mutant CRT, and GAPDH were detected with anti-CRT (N), anti-CRT C-(mut), and anti-GAPDH antibodies, respectively. B) Gating strategy for flow cytometry. HEK-Vec cells transiently transfected with the vector encoding the ratiometric GEM-CEPIA1er probe for 72 hours were used to assess ER calcium levels by flow cytometry based on the measurements of fluorescence emission signals for the Ca2+-bound or unbound probe at 452-455 nm and 498 nm, respectively following excitation with the 405 nm (violet) laser. Transfected cells were gated based on forward (FSC-A) and side scatter (SSC-A), then on 7AAD-negative (live) cells, and further on the cells positive for the calcium-bound or unbound probe signals. Representative dot plots of changes in fluorescence intensities in response to treatment with thapsigargin (Tg), a SERCA inhibitor, before (0 mins) or after (10 mins) treatment, are shown. C) Representative histograms showing changes to Ca2+-bound and unbound signals at different time points following Tg treatment of cells. Mean fluorescence intensities (MFI) for Ca2+-bound and unbound signals were used to calculate the bound/unbound ratios at different time points. D and E) The basal free ER calcium signals were measured as bound/unbound MFIs ratios from the corresponding histograms of CRT-KO or HEK-Vec cells (D, left panel) or CRT-KO cells reconstituted with the indicated CRT constructs (D, right panel). The fractional changes in ER calcium signals in response to Tg were measured at different time points in the indicated cells (E). The fractional change ratios were normalized to the baseline and calculated by subtracting the ratio at 10 min from the ratio at 0 mins (R0-R10/R0)x100. Statistical significance was determined using GraphPad Prism and based on paired t-tests (left panels of D and E) or one-way ANOVA analyses with Tukey’s test (right panels of D and E). ns, not significant; *** P value < 0.001. All measurements were undertaken in the presence of 2 mM extracellular Ca2+. Data shown in D and E are based on 7-15 independent experiments as indicated by the individual data points within the graphs. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Measurements of cytosolic calcium levels in megakaryoblastic (MEG-01) parental cells, CRT-KO cells, or those reconstituted with wild-type CRT or CRTDel52

Cells in 2 mM extracellular Ca2+ were loaded with the Fura-2/AM dye, and the ratio of the signals was calculated following excitation at 340 and 380 nm and emission at 510 nm. A) CRISPR-Cas9 editing of MEG-01 cells using lentiviral constructs encoding calreticulin-targeting gRNA (CRT-KO) or empty vector (Vec). CRT-KO MEG-01 cells were subsequently reconstituted with viral constructs encoding wild-type human CRT or CRTDel52 or transduced with empty vector (Vec) as indicated in the Figure panels. Representative immunoblots of CRT expression in the indicated cells detected by the anti-CRT(N) and anti-CRT(C-mut) antibodies. B) Representative calcium imaging data from indicated cells in the presence of 2 mM extracellular Ca2+ and following ER calcium depletion with SERCA inhibitor, cyclopiazonic acid (CPA). C and E) Basal cytosolic calcium levels were plotted as the means of ratios of emission signals following excitation at 340 and 380 nm. D and F) Fractional changes to cytosolic calcium levels in the indicated cell lines, measured following treatment with CPA and total calcium release determined as the area under the curve (AUC), are plotted relative to the baseline following CPA addition. Data shown in C-F are based on 4 or 5 independent measurements, each including 50-100 cells. A paired t-test was used to analyze the mean difference between two groups (C and D). Multiple comparisons among groups were analyzed using one-way ANOVA analyses with Tukey’s test (E and F). Statistical significance was determined based p-value ≤ 0.05 using GraphPad Prism. ns, not significant.

Measurements of cytosolic calcium levels and SOCE in MEG-01 parental cells, CRT-KO, or those reconstituted with wild-type CRT or CRTDel52.

Cells were loaded with Fura-2 dye, and the ratio of the 340 nm/380 nm signals was calculated following excitation at 340 and 380 nm and emission at 510 nm. A) Representative calcium imaging data from cells in the absence of extracellular calcium, followed by the addition of CPA, and subsequent addition of extracellular calcium with CPA as indicated. B and E) Compiled basal cytosolic calcium levels in the indicated cells in the absence of extracellular calcium plotted as the ratios of 340 nm/380 nm signals. C and F) Fractional changes to cytosolic calcium and total calcium release measured as the area under the curve (AUC) in the indicated cells following ER calcium depletion with the CPA. D and G) Fractional changes to cytosolic calcium following ER calcium depletion with the CPA and subsequent addition of extracellular calcium with CPA. Data are plotted as a fractional increase relative to baseline following CPA addition. SOCE in the indicated cells is plotted as the fractional increase in signal following the addition of 2 mM extracellular calcium in the presence of CPA. Data shown are based on 8-10 independent measurements, each with 20-100 cells. Comparisons were performed by paired t-test and one-way ANOVA analysis using GraphPad Prism. Statistical significance is based on a p-value ≤ 0.05. ns, not significant.

Transcriptional changes to the expression of genes encoding ER calcium binding/regulating proteins in CRT-KO MEG-01 cells.

A) RNA sequencing analysis was used to identify differentially expressed genes in Meg-01 CRT-KO cells in comparison to parental Meg-01 cells. Volcano plot shows significantly upregulated (blue) and downregulated genes (red) genes identified by RNA sequencing. Genes related to ER calcium binding and homeostasis have been highlighted and labeled. The differentially expressed genes were selected as those with a log2 fold-change of <-0.49 or >0.49 and p values <0.05. B) Heat map shows the differential expression of genes encoding ER resident calcium binding proteins or ER calcium regulators that were identified by RNA sequencing in CRT-KO Meg-01 cells compared to parental cells (n=3 biological replicates). C) Comparison of expression ratio of HSPA5, HSP90B1, PDIA3 and PDIA6 in CRT-KO Meg-01 cells (this study) or murine lung cancer cells (Tang et al., 2025) or heterozygous CRTDel52 knock-in (Del52-KI) cells (Fosselteder et al., 2023) normalized to their expression in respective control cell lines expressing wild type CRT. The normalized expression of various genes identified in each dataset was calculated as a ratio of expression in CRT-KO or CRTDel52 knock-in cells relative to the expression in wild-type CRT-expressing cells. Multiple unpaired t-tests were used to determine statistical significance with the Benjamini-Hochberg FDR correction (p-adjusted) <0.05. D-G) Left and middle panels: Relative expression of mRNA transcripts of HSPA5 (n=8-11), HSP90B1 (n=7), PDIA6 (n=6) and PDIA3 (n=6) genes in the indicated MEG-01 cells determined by RT-qPCR using specific gene primers. Relative gene expression was calculated by normalizing the target gene’s cycle threshold (Ct) to a single or multiple endogenous controls (ACTB, GAPDH and HPRT1) which was used for comparison across different cell lines as indicated. Data show 6 – 11 independent runs with duplicate measurements for each run. Student t-test was used to determine the statistical significance between two groups while multiple group comparison was performed by one-way ANOVA analysis using GraphPad Prism. Statistical significance is based on a p-value ≤ 0.05. ns, not significant. Right panels: Representative immunoblots of BiP, GRP94, PDIA6 and PDIA3 expression in the indicated cells detected by specific antibodies.