Fig 2: CD22 protein expression in B-cell malignancies is limited by exon 2 inclusion. A, Flow cytometric detection of exogenously expressed CD22 protein in CD22-deleted OCI-Ly8 cells using anti-CD22 antibodies directed toward either the extreme N-terminus (S-HCL-1) or the C-terminal region (RFB-4) of the extracellular domain. KO, knockout. B, Western blotting detection of CD22 in the same cell lines and parental controls using anti-CD22 antibodies directed toward the N-terminus (R&D Systems, MAB19681) and the C-terminus (Boster Bio, PB9691). C, Schematic annotating the CD22 exon splice junctions (gray arches) assayed using junction-spanning qPCR primers following treatment with Ex2In2 morpholino. Bottom right, morpholino sequence is shown in red, with complementary exon–intron junction sequence shown in blue/white. D, qRT-PCR detection of various CD22 mRNA isoforms in Reh B-ALL cells transfected for 48 hours with Ex2In2 (10 µmol/L and 100 µmol/L). Constitutive expression of ex13–14 junction was used as a measure of total CD22 expression (n = 6, 2 independent experiments with 3 technical replicates each). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. ns, not significant. E and F, Detection of CD22 protein in the morpholino-treated Reh cells by Western blotting and flow cytometric site density assay, respectively (n = 2 independent experiments). G, Viability assay performed on CD22-deleted OCI-Ly8 cells reconstituted with indicated CD22 isoforms and treated for 24 hours with indicated concentrations of inotuzumab (n = 3 technical replicates). IC50 95% confidence intervals (CI) were as follows: 54–119 ng/mL for CD22 KO, 25–85 ng/mL for CD22 ?ex2, and 5–11 ng/mL for CD22 ?ex5–6. H, Bar graph representing IC50 values from G. P values were determined using an unpaired t test. I, Viability assay performed on Ex2In2-treated (48 hours) and inotuzumab-treated (24 hours) Reh cells (n = 3 technical replicates). Cell viability was assessed using the WST-1 assay. IC50 95% CIs were 29–72 ng/mL for Ctrl treatment and 119–333 ng/mL for Ex2In2 treatment. J, Bar graph representing IC50 values from I on the log scale. P values were determined using an unpaired t test. Data in D, F, H, and J are presented as individual and mean values ± SD error bars. Data in G and I are presented as mean values ± SD error bars.

Fig 3: CD22 protein expression is limited by exon 2 inclusion in B-ALL cells. A, Correlation analysis of CD22 site density versus CD22 mRNA levels in pretreatment primary B-ALL bone marrow or peripheral blood specimens obtained from children enrolled on the COG AALL1621 phase II clinical trial. CD22 mRNA levels were measured by qRT-PCR using primers specific for the exon 13–14 (left) or exon 1–2 (right) junctions. CD22 expression was normalized to that of ß-actin. Regression coefficients and P values are shown for each comparison. B, Relative expression of CD22 exon 2–containing and exon 2–skipping splice variants within baseline AALL1621 B-ALL specimens. Each stack plot represents a single patient (designated by the COG unique specimen identifier). Yellow arrow highlights PAYYZW as a sample apparently devoid of protein-coding CD22 mRNA isoforms. The legend shows color-coded CD22 splice variants. C, Flow cytometric quantitation of CD22 molecules in paired pre– and post–inotuzumab treatment (pre-ino/post-ino) bone marrow specimens from AALL1621 patient PAWUXD with multiply relapsed B-ALL. D,CD22 mutational analysis of the paired PAWUXD samples. E,CD22 exon 2 splicing analysis of the paired PAWUXD samples. For color coding, refer to legend in B. F, Flow cytometric quantitation of CD22 molecules in paired pre- and posttreatment (pre-ino/post-ino) bone marrow specimens from AALL1621 patient PAVDRV with multiply relapsed B-ALL. G,CD22 mutational analysis of the paired PAVDRV samples. H,CD22 exon 2 splicing analysis of the paired PAVDRV samples. For color coding, refer to legend in B.

Fig 4: Biochemical and functional characterization of the CD22 ?ex5–6 isoform. A, RT-PCR detection of CD22 ?ex5–6 across 18 diagnostic de novo B-ALL samples driven by various genetic alterations. B, Western blotting using N-terminus–directed anti-CD22 antibody performed on CD22-deleted OCI-Ly8 cells engineered to express CD22 ?ex5–6 or FL isoforms. KO, knockout. C, Amino acid sequence of the peptide used for mAb production. D, RT-PCR detection of CD22 ?ex5–6 in OCI-Ly8 cell lines from B, human B-ALL cell lines, and B-ALL PDXs. E, Western blotting using the 11F11 mAb performed on cells from D. F, Flow cytometric detection of CD22 KO, CD22 FL, and CD22 ?ex5–6. FSC-A, forward scatter area. G, Western blotting demonstrating electrophoretic mobilities of CD22 isoforms treated with deglycosylating enzymes with or without denaturation. H, Western blotting detecting CD22 and total and phosphorylated (p) BLNK in derivatives from B treated with an anti-IgM antibody for indicated time intervals. I, Quantitation of pBLNK bands from H. The experiment was replicated twice with concordant results. J and K,In vitro killing assays performed on cells from B using HA22- and m971-based CD22 CAR T cells (CART22), respectively (n = 2 technical replicates). Data in both panels are represented as mean values ± SD error bars.