Reveal the Neutrophil: Elucidating the Role of a Neutrophil-Specific JAK2-V617F Mutation

Introduction: Myeloproliferative neoplasms (MPN) are characterized by monoclonal proliferation of myeloid cells. The clinical presentation of classic BCR-ABL negative MPN includes blood count changes, splenomegaly, chronic inflammation and, prognostically relevant, thrombosis. Within polycythemia vera, essential thrombocythemia and primary myelofibrosis, the most frequent genetic alteration is the point mutation V617F (VF) in the Jak2 gene. Due to an activation of leucocyte β1 and β2 integrins, in particular on neutrophils, JAK2-VF is driving thrombus formation (Edelmann et al., JCI, 2018). Thus, understanding the role of neutrophils within JAK2-VF-positive MPN is of special interest. We developed a murine JAK2-VF Catchup model of MPN that is characterized by the neutrophil-specific expression of JAK2-VF in combination with the fluorescent protein tdTomato allowing in-vivo two-photon microscopy. Neutrophil-specificity was achieved through a knock-in allele of Cre recombinase and tdTomato replacing exon 1 of Ly6G (Hasenberg et al., Nat Methods, 2015).

Methods: The murine JAK2-VF Catchup model was developed by crossing JAK2^{+/loxP-VF-loxP} mice and Ly6G^{+/Cre-tdTomato} Catchup^IVM-red mice. Experiments were performed in 12 weeks old JAK2^{+/loxP-VF-loxP} Ly6G^{+/Cre-tdTomato} Catchup^IVM-red (JAK2^+/VF Catchup) mice (n=5) and JAK2^+/+ Ly6G^{+/Cre-tdTomato} Catchup^IVM-red (JAK2^+/+ Catchup) mice (n=5) as wildtype (WT) controls. Whole blood cell counts and spleen weight were determined. Bone marrow (BM) cells were harvested from hind limbs followed by isolation of granulocytes using negative selection. In order to determine hematopoietic stem and progenitor cells (HSPCs), BM and spleen cell suspensions were analyzed by flow cytometry (FC). Using ICAM-1, VCAM-1, E-selectin and P-selectin, each with and without Fc-tag, adhesion properties of the isolated granulocytes were examined by either FC or microplate reader analysis. The expression of β1 integrin was further analyzed by FC.

Results and Discussion: JAK2^+/VF Catchup mice showed a slight but significant increase in platelet counts [1544±58.02 Gpt/l; WT: 1190±90.50 Gpt/l; p=0.011], with the mean value still remaining within the normal range. Hematocrit, red blood cell, white blood cell and neutrophil count as well as spleen weight were also within the normal range showing no differences from their WT controls. Interestingly, neutrophil and platelet counts of JAK2^+/VF Catchup mice significantly correlated with one another [p=0.0251; R²=0.8529]. MPN-typical changes in number and composition of c-Kit⁺ HSPCs in either BM or spleen could not be found in JAK2^+/VF Catchup mice. Directly compared to reliably MPN-developing Vav-1-Cre JAK2^+/VF mice, JAK2^+/VF Catchup mice showed significantly less megakaryocyte-erythroid progenitor (MEP), Lin^- Sca1⁺ c-Kit⁺ (LSK) and multipotent progenitor cells (MPP) in the BM. In spleen of JAK2^+/VF Catchup mice, the fractions of Lin^- c-Kit⁺ (LK) and granulocyte-macrophage progenitor cells (GMP) were significantly lower compared to Vav-1-Cre JAK2^+/VF mice. Thus, the neutrophil-specific JAK2-VF mutation does not lead to apparent numerical or compositional changes of HSPCs. These results show that a neutrophil-specific JAK2-VF mutation is not sufficient to induce a characteristic phenotype of JAK2-VF-positive MPN.

The expression of β1 integrin on granulocytes was significantly increased in JAK2^+/VF Catchup mice [fold change vs. control; 1.126±0.041; WT: 1.0±0.011; p=0.0194]. Unexpectedly, despite increased β1 integrin expression, the binding of granulocytes from JAK2^+/VF Catchup mice to either ICAM-1, VCAM-1, E-selectin or P-selectin corresponded to that of their WT controls. Static adhesion and soluble ligand binding assays showed similar results. Since adhesion of granulocytes appeared unchanged in-vitro, we hypothesize that expression of JAK2-VF in neutrophils is not sufficient to overactivate integrins but secondary extrinsic signals (e.g. cytokines) may play a dominant role. Further experimentation is under way to elucidate this molecular mechanism.

Conclusions: In order to dissect the properties of JAK2-VF-positive neutrophils, the JAK2-VF Catchup model represents a promising murine model of MPN. Herewith, the initiation of thrombosis and the development of other granulocyte-driven disease manifestations of JAK2-VF induced MPN can be thoroughly examined.