Fig 1.
Expression of CD302 on leukemic blasts, CD34+CD38- LSC and HSC.
(A) Gating strategy used to identify CD45loSSClo AML blasts and CD34+CD38- LSC. (B) Scatter dot plots showing the CD33 and CD302 expression on AML blasts (n = 33) and LSC fraction (n = 20). Samples were stained with MMRI-20 and CD33 mAb. Populations with a geoMFI ratio ≥3, shown above the dotted line, were considered to be positive. (C) Relationship between mean CD33 and CD302 expression on AML blasts and on AML LSC from patients. The solid lines were generated by linear regression. (D) Six parameter data (including CD117, CD34, CD33, CD38, CD45 and CD302) from five concatenated AML patient samples was converted into two t-SNE dimensions and overlayed with heatmaps of the indicated marker’s MFI. (E) Summary of CD302 expression by AML blasts across FAB subtypes based on their morphology and immunophenotypic characteristics, as outlined in the 2008 WHO classification. (F) CD302 and CD33 expression on AML samples with monocytic differentiation, FAB subtype M4 and M5, were compared to other subtypes of AML.
Fig 2.
Antibodies targeting CD302 are able to be internalised and mediate ADCC of target cells.
(A) Flow cytometry histograms showing the surface expression of CD302 on leukemic target cell lines as determined by staining with the MMRI-20 compared to a mouse IgG1 isotype control. (B) CD302 internalisation by HL-60 cells determined by flow cytometry. Total staining determined by MMRI-20-FITC 37°C incubation for the indicated times after which residual surface CD302 was measured with anti-mouse IgG-PE at 4°C. MFI of antibody staining is reported relative to pre-incubation levels. (C) MMRI-20 induced ADCC against HL-60 target cells. Calcein-AM labelled HL-60 were incubated for 18h with mouse spleen effectors at a 1:10 ratio, together with1000U IL-2 and the indicated concentrations of MMRI-20 or isotype control mAb. Target cell killing was measured as 7-AAD+ Calcein-AM+ cells by flow cytometry and presented relative to death in target alone (0%) or with 2% Triton X solution (100%). **** p<0.0001, two-way ANOVA. (D) ADCC elicited against HL-60 (CD302hi) and U937 (CD302lo) leukemic targets using 20μg/ml MMRI-20 or isotype mAb control. Experiments representative of three experiments. Differences tested by two-way ANOVA. (E-F) HL-60 cells were incubated with either MMRI-20 or isotype control mAb for 30 mins at 37°C and tested for their ability to migrate across 5 μm transwells coated with (E) fibronectin or (F) HS-5 cells towards 160 ng/ml CXCL12 or media alone. Cells in bottom chamber were enumerated after 4h by using flow cytometry and migration presented as the chemotaxis index. Circles connected by lines represent individual paired experiments. No significant difference (NS) between MMRI-20 and isotype group (paired t-test).
Fig 3.
Ex vivo MMRI-20 binding reduces engraftment of leukemic cell lines HL-60 in NOD/SCID mice.
(A) Bar graphs showing absolute frequency of HL-60 cells coated with MMRI-20 or isotype control mAb in BM, spleen and blood of NOD/SCID mice 21 days after iv injection (n = 6/group). HL-60 cells were identified as human CD33+, human CD45+, mouse CD45- cells in tissue cell suspensions by flow cytometry (see panel B of S1 Fig). (B) Disease scores and (C) survival curves of five NOD/SCID mice injected with MMRI-20 or isotype control mAb coated HL-60 cells.
Fig 4.
CD302 is expressed on healthy BM and cord blood.
(A) Representative flow cytometry histograms showing the expression of CD302 on healthy BM (n = 3) and CB (n = 5) HSC, MPP and CD34+CD38- populations. Gating strategy shown in panel A of S1 Fig. (B) geoMFI ratios of CD302 expression on HSC in all BM and CB samples. Dotted line indicates a geoMFI of 3 above which was considered positive. (C) Graphs showing CFU of multi-lineage, myeloid and erythroid lineages counted 12–14 days after seeding wells with equal numbers of Lin-CD302+ and CD302- sorted CB populations (n = 6). ND, no detection of CFU colonies.
Fig 5.
PBD delivery through CD302 mAb mediates killing of leukemic but not hepatic cell lines.
(A) Comparison of HepG2 or HL-60 killing after 96 h culture with the indicated concentrations of MMRI-20 or isotype control mAb together with equimolar concentrations of a GAM IgG secondary antibody conjugated with PBD toxin. Cell viability measured with the CellTiter-Glo luminescent assay and compared as a % to untreated controls. One of three representative experiments shown. ***p<0.001, 2-way ANOVA. (B) PBMC were cultured in quadruplicate for 96h with 1nM MMRI-20 or isotype control mAb together with equimolar concentrations of a GAM IgG secondary antibody conjugated with PBD toxin. Flow cytometry was used to count viable (DAPI-) dendritic cells (Lin- HLA-DR+CD11c+) and monocyte (SSChiLineage+HLA-DR+CD11c+) in wells after culture and compared as a % to untreated controls. One of two representative experiments shown.