Induction of folate receptor type |[beta]| in a bone marrow engraftment model of acute myelogenous leukemia

The glycosyl-phosphatidylinositol-anchored folate receptor (FR) has been widely recognized as a potential target for the selective delivery of therapeutic agents to pathological tissues.1 This is because of the narrow tissue specificity of FR expression in normal tissues, its restricted expression at the luminal surface of epithelial cells that is inaccessible to the circulation, the ability of the receptor to bind and internalize both small and particulate chemical conjugates of folic acid and the ability of the receptor to mediate various types of immunotherapy. The subtype of FR has served as a therapeutic target in many promising pre-clinical and clinical studies of carcinomas, whereas FR expressed in activated macrophage has proven to be a promising target in the treatment of inflammatory diseases.2 We have previously reported that the expression of FR is limited to the myelomonocytic lineage and that its expression is increased during neutrophil maturation and during the activation of macrophage;3, 4 however, among normal hematopoietic cells, only the receptor expressed in activated macrophage is functional in having the ability to bind folate.4, 5 FR is also expressed in a functional form in approximately 70% of acute myelogenous leukemia (AML) in which it is frequently coexpressed with CD34.6 These observations suggest that FR is a potential target for drug delivery to AML cells, but prospects of its clinical utility are limited by the variable and the frequently low expression of the receptor in the leukemic cells of AML patients.6 Our previous in vitro studies in both AML cell lines and primary AML cells have demonstrated that all-trans retinoic acid (ATRA) acts directly on the FR gene to selectively upregulate its expression6, 7 and that this effect is potentiated by innocuous histone deacetylase (HDAC) inhibitors, including valproic acid (VPA).8 The receptor induction sensitized AML cells to FR-targeted drugs including folate-coated liposomal doxorubicin5 and the antifolate, dideazatetrahydrofolate.8 Treatment with ATRA also increased the survival of mice that were injected KG1 AML cells in their peritoneal cavity upon treatment with FR-targeted liposomal doxorubicin.5 Combination treatment with innocuous agents that upregulate FR in AML cells is, therefore, an attractive means of addressing a major limitation of experimental FR-targeted therapies in AML. Two important considerations must precede the design of clinical studies to test whether treatment with ATRA, either alone or in combination with an HDAC inhibitor, will up-regulate FR in the leukemic cells in the bone marrow of AML patients. First, since maintenance of AML is dependent on a smaller population of leukemic stem cells that possess extensive proliferative capacity, it is important to test whether FR-expressing AML cells from patients are capable of proliferation in the marrow. Second, induction of FR by ATRA alone or in combination with an HDAC inhibitor must be demonstrated in an appropriate animal model of AML. These questions are addressed in this study using the non-obese diabetic mouse with severe combined immunodeficiency (NOD-SCID mouse) model of AML. Leukemic stem cells in peripheral blood or bone marrow as well as in several AML cell lines can engraft with high efficiency in the bone marrow of NOD-SCID mice, which have been exposed to a sublethal dose of radiation.9 These mice not only lack B- and T-cell functions, but also have compromised natural killer cell and macrophage activities. The engrafting leukemia stem cells are termed SCID leukemia-initiating cells (SL-ICs). SL-ICs from AMLs (M1, M2, M4 and M5) were largely CD34+/CD38- and phenotypically similar to normal hematopoietic SCID-repopulating cells.10, 11 These cells could persist through serial transplantations and were able to proliferate and differentiate into leukemoblasts in the cascading process of SL-IC clonogenic progenitors (CFUs) blast cells, producing a disease in the engrafted animals identical to that in the donor.10, 11 Since a quantitative method to distinguish between human and mouse FR protein is unavailable, FR expression in the engrafted human cells in NOD-SCID mice was distinguished from that of the host marrow cells by using real-time reverse transcription-PCR assays that were specific for the mRNAs of the human receptor and for human GAPDH (internal control). RNA was isolated from pre- and post-engraftment samples of primary AML cells from three patients for identification and quantification of human FR mRNA and normalized in each sample to the mRNA for human GAPDH (Table 1). The results show that FR+ AML cells could engraft and be enriched in the marrow and could represent a proliferating population of the leukemic cells. To test the ability of ATRA to upregulate FR in AML cells engrafted in mouse bone marrow, MV4-11 cells were chosen rather than primary patient AML cells to have a sufficient amount of model cells for the experiment and to allow reproducibility. Since prolonged exposure to ATRA may induce apoptosis in AML cells, ATRA-resistant (MV4-11R) and ATRA-sensitive (MV4-11S) sublines of MV4-11 cells were first selected by soft agar cloning. Exposure to 0.1 M ATRA in vitro induced a relatively high degree of apoptosis in the MV4-11S subline compared with the MV4-11R subline, as measured by annexin-V staining using the Guava PC apoptosis assay (Guava Technologies, Hayward, CA, USA) (Figure 1). In contrast to the MV4-11S cells, the MV4-1R subline showed net proliferation in the presence of ATRA (data not shown). ATRA induced hFR mRNA in both sublines within 48 h of treatment during which apoptosis was minimal, although the relative mRNA levels were lower in the MV4-11S cells (Figure 2). NOD-SCID mice engrafted with 1 107 cells from either of the two MV4-11 sublines were administered ATRA (10 mg/kg) and VPA (300 mg/kg) intraperitoneally either alone or in combination with the vehicle (phosphate-buffered saline) control for 5 consecutive days. From flow cytometry of the bone marrow harvested from the treated mice, it was determined that the MV4-11 cells engrafted in all of the mice with efficiencies in the range 2–14%, as determined by the proportion of CD45-positive cells (data not shown). The expression of human FR in the bone marrow cells of the mice engrafted with the MV4-11R subline increased approximately 10-fold in the mice treated with ATRA (Figure 3). Further, VPA potentiated this induction of FR by ATRA by several-fold (Figure 3). As expected from the high level of ATRA-induced apoptosis in the MV4-11S subline, hFR was undetectable in the bone marrow cells of mice engrafted with the MV4-11S subline (data not shown). The results of this study validate previous in vitro studies of the induction of FR by ATRA and HDAC inhibitors in an appropriate in vivo model of ATRA-resistant AML, setting the stage for clinical studies of FR induction therapy in AML patients. This work was supported by NIH RO1 Grant CA080183 to MR and Grant CA095673 to RJL. MR also received support from a Harold and Helen McMaster endowment.