Page 94 - Haematologica3

N. Landberg et al.
ADCC assays, we found specific and dose-dependent cell
killing of CD36 expressing KU812 cells (Figure 5A). By con- trast, the polyclonal CD36 antibody was not associated with direct toxic effects in the absence of human NK effec- tor cells (Online Supplementary Figure S11). Notably, we also found that CD36 can be used as a target for ADCC-medi- ated cell killing of primary CD34+ CML cells (n=3, Figure 5B), whereas minimal cell death was observed in corre- sponding NBM cells (n=2, Figure 5C). We conclude that CD36 targeting antibodies can direct human NK cells to specifically eliminate CML cells by ADCC.
Discussion
CML is propagated by leukemic stem cells which are par- tially insensitive to TKI-based therapies, and therefore believed to be responsible for disease relapse upon with- drawal of treatment.6-9,23 The identification of cell surface markers upregulated on primitive CML cells may provide important biological insights, allow for their prospective isolation and characterization, and provide novel means for therapeutic targeting of the treatment resistant cells.
Herein, we used RNA sequencing of CD34+CD38low CML cells to identify genes encoding cell surface proteins upregulated on primitive CML cells and with low or absent expression on corresponding normal cells. In total, we identified upregulation of 32 candidate cell surface markers, of which 16 were further evaluated for protein expression by flow cytometry. Previous studies aimed at identifying differentially expressed genes encoding cell surface markers on candidate CML stem cells mainly used microarray-based approaches.11,12,24,25 Of the identi- fied markers, some have previously been described as being upregulated at the transcriptional level, but apart from IL2RA (CD25), DPP4 (CD26) and IL1RAP, their bio- logical roles have not been functionally studied in the context of chronic phase CML.
We focused in particular on novel cell surface molecules present on primitive CML cells and with low to absent expression on corresponding normal cells, since this may reveal new therapeutic markers on primitive CML cells that can be selectively targeted.26 A similar approach recently allowed us to identify IL1RAP as a therapeutic target on CML stem cells.11,27 Of the 32 upregulated transcripts, we validated the cell surface protein expression of four previ- ously well-studied molecules: IL1RAP, IL2RA, DPP4, and NCAM1.11-13,19,28 In addition, we identified seven novel cell surface markers expressed on primitive CML cells: CD36, LEPR (CD295), TFRC (CD71), ITGB3 (CD61), CD7, FCGR2A (CD32), and GP6. Of these markers, ITGB3 has been shown to be upregulated and functionally important for the growth and homing of AML cells,29 and TFRC has been described to be expressed at various levels in AML.30 However, we found that only CD36 and LEPR were specif- ically upregulated on primitive CML cells when compared to corresponding cells from healthy BM.
The CD36 molecule is a heavily glycosylated transmem- brane protein and scavenger receptor expressed in adipose tissue as well as on thrombocytes, monocytes and macrophages with a role in phagocytosis.31 It has been sug- gested to have a role in the formation of atherosclerotic plaques and the associated inflammation.32 Interestingly, the expression of CD36 was recently shown to demark a meta- bolically distinct and treatment refractory subgroup of
leukemic stem cells in AML and blast crisis CML.33 Moreover, CD36 has been shown to be essential in the metastatic spread of several forms of cancer.34 CD36 anti- bodies were found to block metastasis in xenograft models of oral carcinoma, malignant melanoma and breast cancer, possibly by interfering with the metabolic use of fatty acids.34 However, the expression of CD36 and its putative therapeutic importance in chronic phase CML has not been addressed thus far. We found that CD36 is upregulated on CD34+CD38low CML cells. Interestingly, we also discovered that the CD36 expressing subpopulation within the CD34+CD38lowIL1RAP+ CML fraction was less sensitive to imatinib treatment, and that these cells could be specifically killed by ADCC using CD36 antibodies. IL1RAP targeting has been shown to induce killing of CML cells in a similar fashion,11,27 but herein we describe that CD36 antibodies specifically target a cell population within the IL1RAP expressing population that is less sensitive to imatinib. This could provide new means with which to target the cells responsible for relapse after imatinib cessation. However, whether targeting IL1RAP or CD36 would be the best approach remains unclear; it is possible that it could be beneficial to target both markers for a potentially additive effect.
Many CML patients are treated with imatinib, and we therefore sought to determine the effect of imatinib on CD36 expression. However, CD36 expression is distinctly reduced during in vitro culture even in the absence of ima- tinib. Instead, a more direct approach was used, and we show that CD36 expression within the primitive CD34+CD38low population is drastically decreased during the first three months of therapy. It remains of interest, but is unclear whether the repeated measure of CD36 expres- sion could act as a surrogate for response during imatinib treatment.
In addition to CD36, we also found LEPR to be upregu- lated on primitive CML cells. LEPR is the receptor for the well-studied peptide hormone leptin, which is mainly pro- duced by adipocytes and is known to be involved in the regulation of bodyweight, BM microenvironment, normal hematopoiesis, and proliferation of AML cells.35-41 We observed no growth promoting effects in vitro on primitive CML cells following stimulation by the ligand leptin, but the mechanism of action in CML could be different. Indeed, we note with interest that both CD36 and LEPR are involved in adipose tissue homeostasis, and a potential interplay between the adipocyte containing BM microenvi- ronment and CML stem cells could provide important growth or survival signals for the neoplastic stem cells.
In conclusion, we herein identify upregulation of several novel cell surface markers, including CD36 and LEPR, on primitive CML cells that may provide novel ways to study and target CML stem cells. In addition, we define CD36 as a marker of cells within the primitive CML cell population in chronic phase CML with decreased sensitivity to ima- tinib that are vulnerable to antibody-based therapeutic tar- geting.
Acknowledgments
The authors would like to thank the Nordic CML Study Group for their help in obtaining and distributing the CML samples between the Nordic countries, in particular we wish to thank Kimmo Porkka, Jesper Stentoft, Bjørn Tore Gjertsen, Jeroen Janssen, Kourosh Lotfi, Leif Stenke, and Ulla Strömberg. The NordCML006 and the BFORE studies were supported by
454
haematologica | 2018; 103(3)