A growing body of evidence implicates chemotherapy-resistant and quiescent leukemia stem cells (LSC) in the relapse of Acute Myeloid Leukemia (AML). Similar to hematopoietic stem cells, these LSCs reside in the bone marrow microenvironment in niches comprised of other cells or extracellular matrix (ECM). There are various components of the LSC niche that ensures LSC quiescence through expression of cell adhesion molecules. Many in vitro models of AML do not incorporate features of the bone marrow microenvironment or LSC niche, and thus, are less representative of the in vivo leukemic state. In our published work and that of others, AML cells co-cultured in a stromal-based model demonstrated increased resistance to chemotherapy as compared to conventional suspension cultures. In our prior experience, we decellularized Wharton’s jelly, the gelatinous material in the umbilical cord, to obtain a matrix material devoid of cells. This Wharton’s jelly matrix is composed of collagen, fibronectin, and hyaluronic acid with its CD44 receptor, which are components of bone marrow extracellular matrix. Therefore, we hypothesized that decellularized Wharton’s jelly matrix (DWJM) could serve as an ECM-based model to study acute myeloid leukemia behavior and chemotherapy sensitivity. We hypothesized that leukemia cells cultured in DWJM will be more resistant to chemotherapy compared to leukemic cells grown in suspension.
We used DWJM as a scaffolding material to culture leukemia cells from AML cell lines (Kasumi-1, HL-60, and MV411). We evaluated the effect of DWJM on leukemia cell growth including viability by trypan blue exclusion, proliferation by alamarBlue, and differentiation by flow cytometry. We also examined the expression pattern of the adhesion molecule CD44 by Western blot and immunofluorescence while leukemia cells are growing in DWJM. Cell cycle was analyzed by flow cytometry following cell staining with propidium iodide. Furthermore, we compared the different apoptotic responses, by flow cytometry, of leukemia cells cultured in suspension or DWJM to doxorubicin treatment as well as to the combination of doxorubicin and CD44 neutralizing antibody.
Leukemic cell lines demonstrated successive decrease in their proliferation over time. This was associated with decreased viability, albeit to a lower extent. Cell cycle analysis showed that 86% of MV411 cells and 70% of Kasumi-1 cells cultured on DWJM were arrested in G0/G1 state consistent with quiescence. Additionally, no significant differentiation in cell lines occurred during culture in DWJM. In Kasumi I cells, the adhesion molecule CD44 was highly immunoexpressed when cultured in DWJM, while no CD44 was expressed by cell in suspension as measured by Western blot. Interestingly, leukemia cells seeded on DWJM changed their morphology, forming spindle-shaped cells associated with the expected round cell morphology. In terms of response to doxorubicin treatment measured by apoptosis flow cytometry, MV411 cells cultured in DWJM were significantly more resistant to doxorubicin compared to cells in suspension, while Kasumi I cells were significantly more sensitive to doxorubicin when grown in DWJM. Interestingly, when doxorubicin was combined with a CD44 neutralizing antibody given in similar doses, MV411 demonstrated significant increase in cell death while Kasumi I demonstrated significant decrease in cell death.
Leukemia cell lines can be cultured in DWJM without undergoing differentiation. Leukemic cell growth in this ECM model was characterized by decreased proliferation, dormancy, changes in morphology by acquiring a spindle-shaped appearance, increased expression of the cell adhesion molecule CD44, and by acquiring drug-resistance characteristics in MV411 cell line. CD44 neutralizing antibody did overcome chemotherapy resistance in MV411 but not in Kasumi-1. This data suggests an important role for CD44 in mediating chemotherapy sensitivity in an ECM-based model of AML.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.