Interferon Inhibits Netosis in Neutrophils Derived from Patients with Myeloproliferative Neoplasms in a Neutrophil Subpopulation-Specific Manner

Introduction

Several studies have demonstrated increased neutrophil extracellular traps formation (NETosis) in neutrophils derived from patients with myeloproliferative neoplasms (MPNs) that may contribute to the thrombotic tendency in these diseases. Interferon-α (IFNα) is an effective cytoreductive treatment for patients with polycythemia vera (PV) and essential thrombocytosis (ET). In addition, IFNα reduces the rate of thrombotic complications in these diseases. The mechanism by which IFNα acts in MPNs is not fully understood.

Recently, evidence has suggested that neutrophils are heterogeneous with regards to morphology and function. High density neutrophils (HDNs, also referred to as normal-density neutrophils) and low-density neutrophils (LDNs) are 2 suggested subpopulation.

Aim

To investigate the characteristics of neutrophil extracellular traps (NET) formation in HDNs and LDNs and the effect of IFNα on NETosis in MPNs.

Methods

We enrolled untreated MPNs patients (with either PV or ET) and controls at a tertiary medical center. Neutrophils were harvested from blood samples and subpopulations (HDNs/LDNs) were isolated using Percoll. NETosis was assessed ex-vivo at baseline (following PMA stimulation) and after ex-vivo exposure to IFNα by quantifying NET-bound neutrophil elastase (NE) activity and cell-free DNA (cfDNA) and by co-localization of DNA and histone H3 using ELISA, immunofluorescence and confocal microscopy.

Results

NET formation is higher in HDNs compared to LDNs in MPNs

Following PMA induction, higher levels of NETosis markers were found in HDNs compared to LDNs derived from the same MPNs patients (n=9). Median level of NET-bound NE activity was 36.04mU/mL (IQR 22.09-41.18mU/mL) in HDNs compared to 4.87mU/mL (1.17-25.94mU/mL) in LDNs (p=0.01). In parallel, the median amount of cfDNA released by HDNs was 1.12 (0.57-1.61) compared to 0.49 (0.37-0.68) in LDNs (p=0.003). Confocal microscopy images of neutrophils stained for DNA and histone H3 confirmed these findings. This phenomenon was unique for patients as no significant difference in NETosis markers was demonstrated between HDNs and LDNs derived from gender matched controls (n=6).

IFNα effectively inhibits NETosis in MPNs

In healthy controls (n=10) there was no difference in the level of NE activity between samples that were incubated with IFNα and samples that were not (27.32mU/mL (5.46-37.34 mU/mL) vs 26.78mU/mL (4.62-36.7 mU/mL), respectively, p=0.12). In patients with MPNs (n=17), NET-bound NE activity was significantly lower in all samples that were incubated with IFNα compared to samples from the same patients that were not (21.76mU/mL (5.46–31.56mU/mL) vs 30.73mU/mL (6.92–36.09mU/mL), respectively (p=0.007)).

The inhibitory effect of IFNα on NETosis is more prominent in HDNs

In HDNs derived from MPNs patients, there was a significant reduction in released cfDNA in samples that were incubated with IFNα compared to samples that were not (0.72 (0.24-1.13) vs 1.12 (0.57-1.61), p=0.006). Conversely, no significant difference was noted in cfDNA released from LDNs in samples that were treated or untreated with IFNα (0.73 (0.35-0.93) vs 0.49 (0.37-0.68), p=0.079). Moreover, confocal microscopy images revealed that HDNs treated with INFα show a significant decrease in histone H3 whereas LDNs did not. Healthy controls showed no difference in NETosis markers in both HDNs and LDNs following incubation with IFNα.

Conclusion We demonstrate for the first time differential kinetics of NETosis and response to IFNα exposure in specific neutrophil subpopulations derived from MPNs patients. In this analysis, HDNs drove excess NETosis and were more sensitive to IFNα inhibition.