Background: Inflammation is an important phenomenon in hematologic malignancies, though its drivers are not well defined. Within the hematopoietic system, the complement cascade plays a complex role in regulating hematopoietic stem and progenitor cell (hSPC) homing and exit to systemic circulation (Lenkiewicz et al, Frontiers in Immunology 2019). The role the complement system plays in both normal and malignant hSPC egress from the bone marrow has been recognized, though its role in pathogenesis of hematologic malignancies is not well established. Complement cascade contributions to inflammation-mediated pathologies expanded the traditional view of the cascade from an acute process at the site of injury or pathogen invasion, to a mediator of chronic inflammation (Markiewskiet al, American Journal of Pathology 2007). We previously uncovered a rare single nucleotide polymorphism in an inhibitor of the complement cascade Complement Factor I - CFI (CFIG119R). This loss-of-function mutation was detected using whole genome sequencing in 20% of patients with primary myelofibrosis (PMF). In a separate PMF patient cohort, we observed an increase in global complement activity through elevated CH50 levels and decrease in C3 levels (Sadler et al, Blood 2021). There remains a gap in the understanding of how the complement cascade in general and CFI specifically contribute to development and progression of hematologic malignancies. Therefore, we established a murine model of CFI deficiency to determine mechanistic details of complement cascade contribution to both systemic and local inflammation and development of JAK2V617F-dependent pathologies.

Methods: To determine the effect of CFI deficiency on homeostasis of hematopoietic system a CFI KO mouse model was generated by floxing exon 2 of the Cfi gene and removing it using CRISPR/Cas9-driven cre recombination. Inactivation of Cfi was confirmed by genotyping. Frequencies of granulocytes, B cells, plasmacytes, T cells, macrophages/monocytes, platelets, erythroid cells were assessed by FACS within the fraction of bone marrow mononuclear cells in 4-week-old CFIWT, HET (n=8) and KO (n=8) mice using established profiling protocols (Chorzalska et al, Blood 2018). Complete blood counts (CBCs) were performed on peripheral blood of 4-week-old Cfi WT (n=20), HET (n=26) and KO (n=10) mice. To determine the effect of CFI deficiency on JAK2V617F-driven phenotypes we performed colony forming assay using CFI WT (n=4, 2F, 2M)) or HET (n=4, 2F, 2M) bone marrows that were transduced with JAK2 V617F encoding retrovirus.

Results: CFI KO and HET animals were viable and fertile. CBC analyses of CFI KO, HET vs. WT 4-week-old mice indicated no significant changes with an exception of a significant increase in red blood cell width distribution (RDW) in KO vs. WT and KO vs. HET peripheral blood samples suggestive of a mild anemia. Trends towards granulocytosis and monocytosis as well as leukopenia, erythropenia and decrease in hemoglobin levels were noted. Changes within the red blood cell compartment were confirmed by FSCS. In addition, flow cytometry data showed a significant decrease in CD19+ B-cells in CFI HET vs. WT animals, suggestive of dysregulation of B-cell maturation. Finally, colony forming unit assays showed 59% increase in number of colonies formed by CFI HET/JAK2 V617F+ vs. CFI WT/JAK2 V617F+ bone marrow cells indicative that loss of one copy of Cfi is sufficient to exacerbate JAK2V617F-dependent cellular phenotype.

Conclusions: Early characterization of our newly generated CFI KO mouse model shows a phenotype indicative of CFI-loss induced changes within the homeostasis of hematopoietic system affecting red blood cell and B-cell compartments. Our data also indicate that loss of one copy of Cfi may accelerate cell growth and proliferation characteristics of JAK2 V617F-dependent myeloproliferative neoplasms. Detailed analysis of the CFI KO model may uncover novel therapeutically targetable mechanisms that link the pro-inflammatory and myeloproliferative phenotypes of some stem cell disorders, including JAK2 V617F-dependent MPNs.

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution