The Thrombopoietin Receptor (MPL) Is a Selective Marker for the Identification of Hematopoietic Stem Cells from Myelofibrosis Patients

Recently, it has been shown that both JAK2V617F and mutated calreticulin require MPL for oncogenic transformation (Sangkhae et al., Blood 2014; Elf et al., Cancer Discov. 2016). We have reported that treatment of myelofibrosis (MF) CD34⁺ cells with a thrombopoietin (TPO) receptor antagonist, LCP4, selectively depletes MF hematopoietic stem cell (HSC) and hematopoietic progenitor cells (HPC) (Wang et al., Blood 2016). Since MF is thought to originate at the level of the HSC, we hypothesize that MF HSCs expressing MPL might differ in function from those lacking MPL and might be associated with disease initiation and progression. To test this hypothesis, we have studied the nature of these two types of MF HSCs and HPCs.

Initially, using flow cytometry we showed that MPL+ cells accounted for 45.0% and 17.7% of CD34+ cells isolated from the spleens of 2 JAK2V617F+MF patients (anti-MPL mAb, Clone REA250, Miltenyi Biotec). CD34+MPL+ or CD34+MPL- cells were then isolated and assayed for HPCs. As expected, in the presence of TPO, megakaryocyte (MK) progenitors were generated by MF CD34+MPL+ cells (24-96/1250 cells), but rarely from CD34+MPL- cells (1-4.5/1250 cells). Greater numbers of common myeloid (4.3-fold) and myeloid progenitors (2.2-fold), but smaller numbers of erythroid progenitors (4.5-fold less) were generated from CD34+MPL+ cells as compared with CD34+MPL- cells. These data indicate that MF CD34+MPL+ cells primarily give rise to common myeloid, MK and myeloid progenitors whereas MF CD34+MPL- cells contain erythroid progenitors.

The marrow repopulating potential of MF CD34+MPL+ or CD34+MPL- cells was then assessed by transplanting these cell populations into NSG mice. Four months after transplantation, as few as 1×10³ CD34+MPL+ cells were able to engraft recipient mice to a similar degree as 1×10⁵ CD34+MPL- cells. Moreover, there was a clear relationship between the numbers of CD34+MPL+ cells but not CD34+MPL- cells infused and the degree of human cell chimerism achieved in the bone marrow and spleen of these mice. When 1 × 10⁴ and 1× 10⁵ CD34+MPL+ cells were transplanted, significant numbers of human cells belonging to various hematopoietic lineages were detected in recipient mice. Furthermore, only CD34+MPL+ cells were able to engraft secondary recipients. The frequency of SCID repopulating cells within CD34+MPL+ cells was 1,400-fold greater than that calculated for CD34+MPL- cells. These findings indicate that long-term MF HSCs are primarily present in CD34+ cells expressing MPL.

The influence of TPO/MPL signaling on the proliferation of MF HSCs and HPCs was then examined by culturing single MF CD34+MPL+ or CD34+MPL- cells. Single CD34+MPL+ cells generated greater numbers of clones of various sizes (<50: 1.7-fold; 50-200:0.8-fold; >200 cells: 2.1-fold) than an equal number of single CD34+MPL- cells. Next, the number of progenitors generated after replating of the primary clones was assessed. Clones derived from single CD34+MPL+ cells were able to generate 55±20 secondary CFU-GM, while clones from CD34+MPL- cells were incapable of generating secondary HPC. These data show that the proliferation of MF HSCs expressing and lacking MPL differs markedly in their contribution to the clonal expansion of MF HSCs/HPCs.

Finally, the inhibitory effect of the TPO receptor antagonist, LCP4, on the proliferation of MF MPL+ or MPL- CD34+ cells was evaluated. The production of CFU-GEMM was eliminated and a 43.8-50% reduction in the number of CFU-MK was achieved following the treatment of CD34+MPL+ cells by LCP4. Surprisingly, the treatment of CD34+MPL- cells from one patient with LCP4 also resulted in a 100% and 48.7% reduction in the number of CFU-GEMM and CFU-MK, respectively. Moreover, LCP4 treatment reduced the percentage of total JAK2V617F+ and homozygous JAK2V617F+ CFU-GM generated from CD34+MPL+ cells. The number of clones derived from single MF CD34+MPL+ cells was reduced by 61.5-66.7% in the presence of LCP4, but LCP4 did not alter the proliferation of MF CD34+ cells lacking MPL. These data indicate that the depletion of malignant HSCs/HPCs by LCP4 is achieved by interrupting TPO stimulation of MPL expressing HSC/HPC.

We conclude that CD34+MPL+ cells represent a subset of malignant HSCs that are responsible for the origin of MF, and that MPL antagonist therapy represents a promising therapeutic strategy which substantially alters the natural history of MF.