The Acute Promyelocytic Leukemia-Associated Fusion Protein PML/RAR Blocks t-RA-Induced Differentiation in a Subset of Cells with Stem Cell Potential.

Acute promyelocytic leukemia (APL) is distinguished from other AMLs by cytogenetic, clinical, as well as biological characteristics. The hallmark of APL is the t(15;17) which leads to the expression of the PML/RAR fusion protein. PML/RAR is the central leukemia-inducing lesion in APL and is directly targeted by all trans retinoic acid (t-RA). Patients suffering from APL undergo complete hematologic but not molecular remission upon treatment with t-RA. Virtually all patients treated with t-RA-monotherapy had a rapid relapse within few months. But in the combination with an anthracycline, such as doxorubicin or idarubicin, t-RA improved the long term outcome of APL-patients dramatically. Nothing is known about why t-RA-monotherapy is unable to eradicate completely the leukemic population and how it increases the response to chemotherapy.

In vitro, the exposure of early hemopoietic stem cells (HSCs) to t-RA does not induce differentiation but selects immature progenitors. Moreover, mice lacking the t-RA-specific receptor RARalpha do not exhibit an impairment of granulopoiesis or hemopoiesis. The indication, that t-RA may be involved in the hemopoietic differentiation, is given by the HL-60 cell line which undergoes granulocytic differentiation at the pharmacological dosages (10⁻⁶M) of t-RA. Furthermore vitamin A-deficient mice or mice treated with a antagonist of t-RA accumulate more immature granulocytes in the bone marrow. PML/RAR mediates the response of APL blasts to t-RA, but it is completely unclear, which effect t-RA exerts on the PML/RAR-positive leukemic stem cells which maintains the blast population and represents the source of relapse. Therefore we investigated the effect of t-RA on a cell population with stem cell capacity expressing PML/RAR isolated from the APL cell line NB4 as well as from CD34+/CD38- KG-1 cells transfected with PML/RAR.

Here we report that i) the NB4 cells engrafted in NOD/SCID mice indicating the presence of a subpopulation with stem cell capacity in NB4 cells; ii) NB4 had a Hoechst 3342 excluding side population (SP) representing about 1% of the whole cell population; iii) t-RA reduced but did not deplete the side population in NB4 cells; iv) the expression of PML/RAR increased CD34+/CD38- population in KG-1 cells from 75% to over 95%; v) t-RA reduced the CD34+/CD38- population from 75% to 3,5% in mock transfected KG-1 confirming its capacity to induce differentiation, whereas in PML/RAR-positive KG-1 cells it led only to a reduction from 98% to a 25%, which still maintain the capacity to engraft in NOD-SCID mice; vi) also the expression of other fusion proteins, such as AML-1/ETO or PLZF/RAR, associated with t-RA-resistant AML-subtypes, increased the percentage of CD34+/CD38- KG-1 cells over 90%, which was reduced by t-RA only to 35% and 19%, respectively.

Taken together these data suggest that a subset of early HSC expressing PML/RAR exhibit the same t-RA-resistant phenotype as HSC expressing fusion proteins associated with AML-subtypes which, in contrast to APL, do not respond to t-RA. These data may give an explanation, why APL-patients do not achieve complete molecular remission upon t-RA monotherapy and undergo early relapse.