Thrombopoietin Receptor (MPL) Genotype Modifies the Myeloproliferative Phenotype in a JAK2 V617F Transgenic Mouse Model of Polycythemia Vera.

Abstract 964

The JAK2 V617F mutation is primarily associated with three chronic myeloproliferative disorders (MPD), polycythemia vera (PV), essential thrombocytosis (ET) and primary myelofibrosis (PMF) but how a single mutation could be responsible for three different disorders is still unresolved. A gene dosage effect was proposed based on the MPD phenotypes in mice with differential expression of a JAK2 V617F transgene, where low expression correlated with an ET phenotype and high expression with a PV phenotype. However, quantitative studies of JAK2 V617F expression in humans revealed significant overlap between PV and ET. Since JAK2 is the cognate tyrosine kinase for the erythropoietin (EPO) and thrombopoietin (TPO) receptors, and JAK2 V617F is expressed in pluripotent hematopoietic stem cells, PV is the ultimate clinical phenotype of the mutation. Furthermore, TPO but not EPO promotes the survival and proliferation of pluripotent hematopoietic stem cells, suggesting that the TPO receptor (Mpl) is essential not only for generating thrombocytosis, but also the stem cell expansion that is characteristic of PV. To examine the role of Mpl in the genesis of the JAK2 V617F MPD phenotype, we manipulated the MPL genotype in a transgenic mouse expressing 13 copies of JAK2 V617F (V617Ftg) (Blood 111:5109, 2009) by breeding these mice with MPL knockout mice (Science265:1445, 1994), which are hematologically normal except for profound thrombocytopenia, to create three genotypes: V617Ftg/MPL wild type (wt); V617Ftg/MPL heterozygote (het), and V617Ftg/MPL knockout (ko). We compared the blood counts, spleen weights, plasma TPO levels, and bone marrow and spleen histology of these three genotypes with each other and with MPL wt, MPL het and MPL ko mice over a 33 week period. Crossbreeding gave the expected genotypes, JAK2 V617F transgene expression was stable in all groups, platelet Mpl expression by immunoblotting correlated with MPL genotype, there was no unexpected mortality, and body weights were not different for any of the genotypes during the observation period. As expected, in V617Ftg/MPL wt mice there was a robust and persistent thrombocytosis (2087 +/− 641 × 10⁶/μL vs 1005 +/− 176 × 10⁶/μL, p<0.001), an erythrocytosis (hemoglobin, 18.3 +/− 1.1 gm % vs 14.9 +/− 0.72 gm %, p <0.001) that peaked at 14-16 weeks but then diminished, and a leukocytosis (16.3 +/− 5.1 × 10⁶/μL vs 12.9 +/−3.4 ×10⁶/μL, p = 0.043) as compared to MPL wt mice. By contrast, in V617Ftg/MPL ko mice, the PV phenotype was virtually abrogated in all cell types as compared to V617Ftg/MPL wt (hemoglobin, 16.1 +/− 0.87 vs 18.3 +/− 1.1, p< 0.001; leukocyte count, 11.3 +/− 2.8 vs 16. 3 +/− 5.1 , p= 0.003, and platelet count, 293 +/− 102 vs 2087 +/− 641, p< 0.001), and not different than their MPL ko counterparts except for a mild erythrocytosis (16.1 +/− 0.9 vs 14.9 +/−, p < 0.001), while in V617Ftg/MPL het mice, erythrocytosis was comparable to the V617Ftg/MPL wt mice and higher than in MPL het controls (17.9 +/− 1.4 gm% vs 14.9 +/− 0.9 gm % p <0.001), but there was only minimal thrombocytosis (1310 +/− 274 × 10⁶/μL vs 1021+/− 241 × 10⁶/μL, p< 0.001), and no leukocytosis (14.8 +/− 4.0 10⁶/μL vs 14.1 +/− 3.7 × 10⁶/μL, p=0.4 ) as compared to the MPL het mice. Marrow and spleen histology reflected the genotype and blood counts and spleen weight was increased equally in all three V617Ftg/MPL genotypes as compared to controls. Plasma TPO was elevated in MPL ko (5530 +/− 1334 pg/mL, p =0.006) and V617Ftg/MPL ko (4201 +/− 736 pg/mL, p = 0.001 ), but not in MPL het mice (723 +/− 720 pg/mL), compared to MPL wt mice (323 +/− 62 pg/mL), while in V617Ftg/MPL wt (163 +/− 52 pg/mL, p < 0.001) and V617Ftg/MPL het mice (176 +/− 56 pg/mL, p < 0.001) plasma TPO was lower than in MPL wt mice. Based on these data, we conclude that MPL genotype is an important modifier of the MPD phenotype in a JAK2 V617F transgenic mouse model of PV, not only for thrombopoiesis but, importantly, also for erythropoiesis and myelopoiesis. We also infer from these data that the impaired Mpl expression observed in human PV may also be a significant modifier of the JAK2 V617F phenotype, either by acting as a dominant-negative with respect to JAK2 V617F activity, or possibly through impaired plasma TPO regulation.