Deletion of CD36 exhibits limited impact on normal hematopoiesis and the leukemia microenvironment

Background CD36 has been identified as a potential therapeutic target both in leukemic cells and in the tumor immune microenvironment. In acute myeloid leukemia (AML), we found that APOC2 acts with CD36 to promote leukemia 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. To establish CD36 as a viable therapeutic target in AML, we investigated whether targeting CD36 has any detrimental impact on normal hematopoietic cells. Methods Differential expression data of CD36 during human and mouse normal hematopoiesis were examined and compared. Cd36 knockout (Cd36-KO) mice were evaluated for blood analysis, hematopoietic stem cells and progenitors (HSPCs) function and phenotype analyses, and T cells in vitro expansion and phenotypes in comparison with wild type (WT) mice. In addition, MLL-PTD/FLT3-ITD leukemic cells were engrafted into Cd36-KO and WT mice, and leukemia burden was compared between groups. Results RNA-Seq data showed that Cd36 expression was low in HSPCs and increased as cells matured. Phenotypic analysis revealed limited changes in blood count except for a slight yet significantly lower red blood cell count and hemoglobin and hematocrit levels in Cd36-KO mice compared with WT mice (P < 0.05). In vitro cell proliferation assays of splenocytes and HSPCs from Cd36-KO mice showed a similar pattern of expansion to that of cells from WT mice. Characterization of HSPCs showed similar percentages of the different progenitor cell populations between Cd36-KO with WT mice. However, Cd36-KO mice exhibited ~ 40% reduction of the number of colonies developed from HSPCs cells compared with WT mice (P < 0.001). Cd36-KO and WT mice presented comparably healthy BM transplant in non-competitive models and developed similar leukemia burden. Conclusions Although the loss of Cd36 affects the hematopoietic stem cell and erythropoiesis, limited detrimental overall impact was observed on normal Hematopoietic and leukemic microenvironments. Altogether, considering the limited impact on normal hematopoiesis, therapeutic approaches to target CD36 in cancer are unlikely to result in toxicity to normal blood cells. Supplementary Information The online version contains supplementary material available at 10.1186/s11658-023-00455-8.

Figure S1. Cd36 differential gene expression patterns in normal hematopoiesis between mouse and human A B Figure S1. Cd36 differential gene expression patterns in normal hematopoiesis between mouse and human. (A-B) Log2 transformed gene expression level for Cd36 in mouse normal hematopoietic system obtained from GSE14833 and GSE6506 datasets and CD36 expression in normal human hematopoiesis obtained from datasets GSE17054, GSE19599, GSE11864, and E-MEXP-1242, which all datasets were downloaded from BloodSpot database. Data is presented as the mean of gene expression among each cell population and each colored dot represents the expression value of a single sample. Unpaired t-test analysis was used (***, P < 0.001; **, P < 0.01; *, P < 0.05).   (B-D) qPCR quantification of Cd36 knockout efficiency in healthy mouse bone marrow cells, spleen cells, and liver cells (n = 6 mice per group). The bar graph represents the mean of Cd36 mRNA level in per group with standard error of the mean. Welch's t-test was used to analyze the significant difference (**, P < 0.01).
(E-H) Cd36 knockout was confirmed using flow cytometry by comparing the cell surface Cd36 + population in bone marrow, spleen, liver, and blood tissues (n = 6 mice per group).
(I) The spleens and livers were collected from Cd36-KO and WT mice.
(J-K) The weight of spleen and liver organs were compared. The bar graph represents the mean of spleen and liver mass in Cd36-KO and WT mice with standard error of the mean. Each colored dot represents the mass value of every single mouse tissue (n = 6 mice per group). The differences between Cd36-KO and WT mice tissues were analyzed by unpaired t-test (**, P < 0.01; Abbreviation: ns, not significant).
(L-O) Cd36 knockout efficiency was confirmed using flow cytometry analysis by comparing the cell surface Cd36 + population and Cd36 MFI in BM, spleen, liver, and blood cells. Data is presented as the mean of Cd36 + surface expression or CD36 MFI and each colored dot represents the value of a single sample. (n = 6 mice per group). The differences between Cd36-KO and WT mice tissues were analyzed by Welch's t-test (*, P < 0.05; ***, P < 0.001; ****, P < 0.0001).

Figure S3. Hematological analysis reveals similar blood counts between Cd36-KO and WT mice (A-H)
Data is presented as the mean of blood count between Cd36-KO mice and WT mice for white blood cell, red blood cell, hemoglobin, hematocrit, neutrophil, lymphocyte, monocyte, and platelet. Each colored single dot represents the count for every single mouse (n = 6 mice per group). The differences between Cd36-KO and WT group were analyzed by unpaired t-test (*, P < 0.05; Abbreviation: ns, not significant).   (B-C) The weights of spleens and livers were measured, and data is presented as the mean of tissue mass from WT and KO group and each colored dot represents the mass of a single mouse tissue (n = 5 mice per group). The differences between Cd36-KO and WT groups were analyzed by Mann-Whitney test (Abbreviation: ns, not significant).     Figure S9. Cd36-KO mice exhibit similar AML engraftment with WT mice (A) Spleens and livers were collected from Cd36-KO and WT mice engrafted with 5 x 10 6 FLT3-ITD/MLL-PTD mouse leukemic cells (n = 5 mice in KO; n = 6 mice in WT), one CD45.1 allele bearing blank mice, and one CD45.2 allele bearing blank mice.
(B-C) The weights of spleens and livers were measured, and data is presented as the mean of tissue mass from WT and KO group and each colored dot represents the mass of a single mouse tissue (n = 6 mice in WT group; n = 5 mice in Cd36-KO group). The differences between Cd36-KO and WT groups were analyzed by Mann-Whitney test (Abbreviation: ns, not significant). (D-G) Representative flow cytometry result showing AML engraftment of FLT3-ITD/MLL-PTD mouse leukemic cells in the BM, spleen, liver, and blood cells of WT and Cd36-KO mice (n = 6 mice in WT group; n = 5 mice in Cd36-KO group).  (B-C) The weights of spleens and livers were measured, and data were presented as the mean of tissue mass from WT and KO group and each colored dot represents the mass of a single mouse tissue (n = 4 mice per group). The differences of tissue weight between Cd36-KO and WT groups were analyzed by Mann-Whitney test (Abbreviation: ns, not significant). (D-G) Representative flow cytometry result showing AML engraftment and quantification of engraftment of FLT3-ITD/MLL-PTD mouse leukemic cells in the BM, spleen, liver, and blood cells of WT and Cd36-KO mice (n = 4 mice). The difference between groups were analyzed using unpaired ttest (Abbreviation: ns, not significant).