Tumor Necrosis Factor-alpha (TNF) is elevated in myeloproliferative neoplasm (MPN) and plays a key role in expansion of the JAK2V617F neoplastic clone. A high TNF environment, as is the case in MPN patients, gives TNF resistant JAK2V617F mutant cells a selective advantage over their TNF sensitive non-mutant counterparts resulting in expansion of the neoplastic clone. To efficiently target TNF production therapeutically it is necessary to identify the mechanism driving this excessive TNF production. TNF is classically produced by monocytes after stimulation through Toll-like receptors (TLR), crucial pattern recognition receptors for microbial products. Because TLR signaling plays an integral role in inflammation and TNF production, we hypothesized that exaggerated signaling of the TLR pathway is the mechanism by which TNF is overproduced in MPN.


We compared the response of peripheral blood monocytes from MPN versus normal controls to the TLR agonists R848 (TLR7/8), LPS (TLR4), or zymosan (TLR2). After stimulation with each of these TLR agonists for 24 hours, CD14+ monocytes from MPN patients (n=18) produced increased amounts of TNF (measured by ELISA) as compared to normal controls (n=10) at all concentrations tested (p<0.05). The increased TNF production in MPN monocytes could not be explained by a higher fraction of inflammatory (CD16+ CD14+) monocytes in MPN. We next used phosflow to detect whether MPN patients have abnormal activation of downstream signaling molecules following stimulation with TLR agonists. At early time points (15min) following stimulation, MPN patients (n=6) and normal controls (n=6) phosphorylated p38 and ERK1/2 equally. At later time points (2hrs) MPN patients maintained phosphorylation of p38 and ERK1/2, whereas in normal controls phosphorylation of p38 and ERK1/2 returned to baseline (p<0.05).

This suggested that the negative feedback regulation of TLR signaling may be defective in MPN patients resulting in prolonged TNF production after stimulation. To test this, we compared the tempo of TNF production following LPS stimulation in MPN versus normal monocytes using intracellular flow cytometry as well as ELISA. Monocytes from normal controls (n=12) cease production of TNF by 9 hours post LPS stimulation, whereas MPN monocytes (n=12) continue to produce TNF at 18 hours (p<0.05).

LPS stimulation induces a negative feedback loop culminating in the production of IL-10, an anti-inflammatory cytokine which serves to dampen TNF production. We found that MPN monocytes produce abundant IL-10 in response to LPS stimulation. Next we measured the ability of IL-10 to dampen TNF production in LPS stimulated monocytes comparing MPN patients versus normal controls. Addition of low concentrations of rhIL-10 (1ng/ml) to LPS stimulated monocytes reduced TNF production by 50% in normal controls (n=8), whereas it reduced TNF production by only 10-20% in MPN (n=5). However, higher concentrations of IL-10 (>10ng/ml) was able to dampen TNF production in LPS stimulated MPN monocytes by 50%.


MPN monocytes produce excessive amounts of TNF in response to TLR ligation due to a defect in the negative regulatory feedback loop which normally serves to dampen TNF production following TLR ligation. We have localized this defect to a blunted response to the anti-inflammatory cytokine IL-10. MPN monocytes are less responsive to the anti-inflammatory actions of IL-10 at low concentrations but these effects can be restored by increasing the concentration of IL-10. Together, these data suggest that administration of IL-10 may reduce excessive inflammatory cytokine production in MPN.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.