Abstract
CD36 has been identified as a potential therapeutic target both in the leukemic cells and in tumor immune microenvironment. In acute myeloid leukemia, we recently found that APOC2 acts with CD36 to promote acute myeloid leukemia (AML) growth by activating the LYN-ERK signaling. CD36 also plays a role in lipid metabolism of cancer associated T-cells leading to impaired cytotoxic CD8+ T-cell and enhanced Treg cell function. CD36 is also present on various types of normal hematopoietic cells, including monocytes, macrophages, endothelial cells, and platelets and contribute to their normal functions. Whether targeting CD36 presents unwanted consequences on normal hematopoiesis remains unknown.
Here we analyzed mouse RNA-Seq data (BloodSpot, GSE60101) and found that Cd36 expression is low in the hematopoietic stem cells (HSC) and increases significantly as they differentiate to mature cells (115.4-fold in erythrocytes, 61.8-fold in NK cells, 32.9-fold in B cells, P < 0.001). We also observed differential expression patterns of CD36 RNA between human and mouse. The haemopedia human RNAseq data revealed high level of CD36 expression in dendritic cell lineage, macrophage/monocytes, and NK cells. While haemopedia mouse RNAseq data showed higher Cd36 expression in progenitor cells, erythrocytes, mast cells, basophils, and B cells. To investigate the effect of loss of Cd36 in normal hematopoiesis, we compared B6.129S1-Cd36tm1Mfe/J Cd36 knockout (KO) with C57BL/6J control (WT) mice (N=6 per group). The mice strains were confirmed by genotyping. In bone marrow (BM), spleen, and liver cells, the KO mice had reduced Cd36 mRNA level (~46-72%, P < 0.05), and decreased Cd36 surface expression (~20-40% Cd36+ population, P < 0.05) compared with WT mice. Differential blood count and hematological analysis revealed lower red blood cells count (5.85 vs 6.93 (x 106/uL), P = 0.0332), hemoglobin (10.03 vs 11.35 (gr/dL), P = 0.0109), and hematocrit (31.67 vs 35.48 (%), P = 0.02) in KO mice than WT mice, yet these values remained within the normal range. On the other hand, the count of white blood cells (1.47 vs 2.06 (x 103/uL), P = 0.2461), neutrophil (5.77 vs 6.10 (%), P = 0.8750), lymphocytes (81.55 vs 86.48 (%), P = 0.1509), monocytes (5.27 vs 3.23 (%), P = 0.0848), eosinophils (7.40 vs 4.08 (%), P = 0.3223), and platelets (230.67 vs 280.17 (x 1000/uL), P = 0.6779) were not significantly different between Cd36 KO and WT mice. Cd36 KO mice also exhibited similar T cell phenotype in the spleen compared with the WT mice: (% lymphocytes: 72.13% vs 67.97%, P = 0.7857; % CD3+: 29.80% vs 29.10%, P = 0.6982; % CD4+: 57.02% vs 54.52%, P = 0.0901; % CD8+: 32.12% vs 33.70%, P = 0.1246; and % CD25+: 12.03% vs 21.03%, P = 0.9920).
We also assessed the in vitro expansion of hematopoietic stem and progenitor cells (HSPC) in HSC expansion media containing SCF and TPO cytokines. The expansion of HSPC was not significantly different (P = 0.2461) between WT and KO mice (Day 10 to Day 5 fold change: 1.65-fold, P = 0.0187 in WT and 1.92-fold, P = 0.0049 in KO). Consistently, the MFI of CFSE labeled HSPC was decreased by 35% (P = 0.0047) in WT and 32% (P = 0.0006) in KO on day 9 relative to day 1 (P = 0.8292 between KO and WT). However, mouse colony forming cell assay using methylcellulose-based media showed that the HSPC of KO mice resulted in less colonies compared with WT mice (WT vs KO = 44 vs 26, P = 0.0108). We also assessed T cells proliferation in CD36 KO mice by in vitro culturing of mouse splenocytes with T cell expansion media containing IL2 and PHA. The number of viable T cells was 1.94-fold (P < 0.0001) in WT mice and 2.24-fold (P < 0.0001) in KO mice on day 10 relative to day 5 (P = 0.2519 comparing WT vs KO). In addition, the cell trace dye MFI was decreased by 67% (P = 0.0002) in WT splenocytes and 59% (P = 0.0030) in KO solenocytes (P = 0.2879 comparing WT vs KO). To assess the leukemia burden in KO mice, we injected 5 million FLT3-ITD/MLL-PTD mice leukemic cells via tail vein in either Cd36 KO mice (N = 5) or WT mice (N = 6) and compared the percentage of engraftment in their tissues. Cd36 KO mice exhibited similar leukemia engraftment in the BM (88.92% vs 86.38%, P = 0.0497), spleen (76.48% vs 68.10%, P = 0.1627), liver (16.12% vs 6.70%, P = 0.2521), and blood (16.33% vs 0.96%, P = 0.2504) compared with the WT group. Altogether, our preliminary data suggest that Cd36-KO has limited impact on normal hematopoietic cells.
Disclosures
Yaghmour:USC: Current Employment.
Author notes
Asterisk with author names denotes non-ASH members.