CD38 Overexpressed Monocytes Play a Critical Role in the Fibrogenesis in Myeloproliferative Neoplasms and Can be Used As a Therapeutic Target to Ameliorate Fibrosis

Background: Several studies have shown monocytes were the major cellular source of proinflammatory cytokines and monocyte-derived fibrocytes, playing an important role in myelofibrosis (MF). We aim to investigate the role of monocytes in MF acceleration and explore the mechanism.

Method: We analyzed the clinical data from 387 MF patients and generated inducible Nras^G12D and Jak2^V617F (NJ) knock-in mice to explore the mechanism.

Results: In 387 PMF patients, there were 63 cases with monocytosis (≥1x10⁹/L). Patients with monocytosis had a higher MF grade (Grade2-3: 88.5% vs 74.2%, p=0.02), inferior OS (37 months vs NOT reached, p=0.004). Patients with monocytosis were more likely to have mutated ASXL1 (49.2% vs 25.2%; p<0.001), SRSF2 (12.7% vs 5.5%; p=0.03), NRAS (11.1% vs 3.7%; p=0.02) and RUNX1 (7.9% vs 2.2%; p=0.02).

In mouse model, peripheral blood (PB) count showed an obvious monocytosis accompanied with a significant increase in the percentage of CD11b⁺CD115⁺ monocytes in NJ recipient mice at 8 weeks after transplantation. At 28 weeks after transplantation, most of NJ revealed a decrease in hemoglobin, splenomegaly, reticulin infiltration and BM failure (mean ckit⁺ cells per Femur: Jak2^V617F (J) 2.52x10⁷(n=5) vs NJ 0.99x10⁷(n=6), p=0.0238), but not in other age-matched mice with Nras^G12D(N) or Jak2^V617F. Both at 20 weeks (pre-fibrotic stage) and 28 weeks (fibrotic stage), NJ presented a significantly higher percentage of CD11b⁺CD115⁺ monocytes in BM. Relative to the fibrotic stage, the increase in the CD11b⁺CD115⁺ Ly6c^high monocyte was more obvious in the pre-fibrotic stage. In vitro, the differentiation assay revealed that monocyte-derived fibrocytes mainly differentiated from the Ly6c^high subpopulation and significantly increased in N and NJ. Although Ly6c^high monocytes from both N and NJ were more likely to differentiate to fibrocytes in vitro, only NJ presented MF phenotype. Therefore, we further explored the gene expression file in Ly6c^high and Ly6c^low monocytes. Compared with N, monocytes of NJ showed significantly elevated expression of CD38. Relative to J, monocyte from Asxl1^-/- Jak2^V617Fmouse (AJ, we published in Haematologica,2023) also showed significantly higher expression of CD38, which is consistent with more severe fibrosis in AJ.

In patients, we investigated CD38 expression in PB CD14⁺ monocytes from healthy control (HC, n=7), CMML(n=11), PV(n=27), ET(n=26), and PMF(n=65). CD14⁺ monocytes from PV and PMF showed significantly higher expression of CD38, compared with HC, CMML, and ET via RT-qPCR. Monocyte from PV patients with grade 1 fibrosis had higher CD38 expression than those without fibrosis.

CD38, a nicotinamide adenine dinucleotide (NAD)+ hydrolase, is a type II plasma membrane protein that may regulate cell recruitment, cytokine release, and NAD⁺ level. NAD⁺ level was significantly lower in NJ Ly6c^high monocytes. GSEA analysis revealed that pathways related to glycolysis, oxidative phosphorylation, and calcium concentration were positively enriched in NJ monocytes compared with N. Increased expression of CD38 may be the reason of these alterations. WT and N mice were intraperitoneally injected with 3 mg/kg Lipopolysaccharide (LPS,CD38 inducer) twice a week for a month. After the LPS exposure, N mice developed severe bone marrow fibrosis (BM cells per femur: PBS 10.45x10⁷ vs LPS 0.04x10⁷; p=0.0002, n=2 for each group) with elevated expression of CD38 in monocyte and splenomegaly. After the LPS exposure, WT mice also had slight myelofibrosis. These data indicated CD38 may be a crucial factor for accelerating fibrosis.

In vitro, both pharmacological suppression of CD38 by 78c and supplementation with the NAD+ precursor nicotinamide mononucleotide (NMN), markedly inhibited the formation of fibrocytes with boosting NAD⁺ levels in CD115⁺ monocytes from both mice and patients. In vivo, the NJ mice were divided into two groups that received 4-week solvent (n=6) or 78c (n=5) treatment separately. After 4-week follow-up, our data indicated that the 78c delayed the onset of fibrosis and anemia (mean HGB: control 108.2 g/L vs 78c 189.4 g/L; p=0.036) and finally extend the survival. Data on 78c in combination with ruxolitinib in vivo will be presented at the meeting.

Conclusion: our study illustrated the crucial role of CD38 in monocyte differentiation and fibrogenesis in MPN. Targeting CD38 might represent a novel therapy to ameliorate myelofibrosis.