ADAM17-Mediated Reduction in CD14++CD16+ Monocytes ex vivo and Reduction in Intermediate Monocytes With Immune Paresis in Acute Pancreatitis and Acute Alcoholic Hepatitis

Abstract

Impaired immune responses and increased susceptibility to infection characterize acute inflammatory conditions such as pancreatitis and alcoholic hepatitis and are major causes of morbidity and mortality. However, the mechanisms that drive this apparent immune paresis remain poorly understood. Monocytes mediate host responses to damage and pathogens in health and disease, and three subsets of monocytes have been defined based on CD14 and CD16 expression. We sought to determine the changes in monocyte subsets in acute pancreatitis (AP) and acute alcoholic hepatitis (AAH), together with functional consequences and mechanisms that underlie this change. Peripheral blood mononuclear cells (PBMCs) from patients with AP or AAH were compared with healthy controls. Monocyte subsets were defined by HLA-DR, CD14, and CD16 expression. Changes in surface and intracellular protein expression and phosphorylation were determined by flow cytometry. Phenotype and function were assessed following stimulation with lipopolysaccharide (LPS) or other agonists in the presence of specific inhibitors of TNFα and a disintegrin and metalloproteinase 17 (ADAM17). Patients with AP and AAH had reduced CD14⁺⁺CD16⁺ intermediate monocytes compared to controls. Reduction of intermediate monocytes was recapitulated ex vivo by stimulating healthy control PBMCs with Toll-like receptor (TLR) agonists LPS, flagellin or polyinosilic:polycytidylic acid (poly I:C). Stimulation caused shedding of CD14 and CD16, which could be reversed using the ADAM17 inhibitor, TMI005 but not direct inhibitors of TNFα, a known ADAM17-target. Culturing PBMCs from healthy controls resulted in expansion of intermediate monocytes, which did not occur when LPS was in the culture medium. Cultured intermediate monocytes showed reduced expression of CX₃CR1, CCR2, TLR4, and TLR5. We found reduced migratory responses, intracellular signaling and pro-inflammatory cytokine production, and increased expression of IL-10. Stimulation with TLR agonists results in ADAM17-mediated shedding of phenotypic markers from CD16⁺ monocytes, leading to apparent "loss" of intermediate monocytes. Reduction in CD14⁺⁺CD16^- monocytes and increased CD14⁺⁺CD16⁺ is associated with altered responses in functional assays ex vivo. Patients with AP and AAH had reduced proportions of CD14⁺⁺CD16⁺ monocytes and reduced phosphorylation of NFκB and IL-6 production in response to bacterial LPS. Together, these processes may contribute to the susceptibility to infection observed in AP and AAH.

Keywords: ADAM17; acute alcoholic hepatitis; acute pancreatitis; infection; inflammation; monocytes.

Figure 1

CD16⁺ monocytes are reduced in acute pancreatitits and in response to LPS. Representative flow cytometry plots showing monocyte subsets in (A) healthy controls or (B) patients with acute pancreatitis (AP). Monocytes were identified by forward and side scatter, HLA-DR positivity and classified according to the expression of CD14 and CD16. (C) Proportion of intermediate monocytes was significantly reduced in patients with AP compared with controls (p = 0.008). (D) Proportion of non-classical monocytes was reduced in patients with AP compared to controls [3.4 vs. 6.3% (E)]. Patients with AP had reduced levels of NFκBp65 phosphorylation at S529 in monocytes stimulated with LPS compared with controls (p = 0.001, percentage increase over unstimulated baseline). (F) Patients with AP had reduced LPS-induced IL-6 production from PBMCs compared to controls (p = 0.006).

Figure 2

Apparent loss of intermediate monocytes in response to LPS stimulation ex vivo. (A) Representative flow cytometry plots showing monocyte subsets in healthy controls incubated for 3 h in medium (B) or LPS. (C) Intermediate monocytes as a proportion of all monocytes sampled from healthy controls were reduced after 3 h incubation with LPS compared with medium alone (p < 0.0001). (D) Viability of healthy control PBMCs sampled following 24 h incubation with medium and LPS by mammalian LIVE/DEAD™ viability/cytotoxicity kit (Invitrogen™), showing no increase in cell death with LPS (ns = not statistically significant). (E) No difference in proportion of intermediate monocytes from healthy controls following treatment with LPS when using standard tissue culture or ultra-low bind plates. (F) Intracellular staining for CD14 and (G) CD16 in monocytes was not increased following stimulation with LPS for 3 h. (H) Stimulation of magnetically-isolated healthy control monocytes with LPS for 3 h resulted in loss of cells from the intermediate monocyte gate (p = 0.031). (I) Intermediate monocytes were reduced in whole blood sampled from healthy donors incubated with LPS-stimulated for 3 h compared to medium alone (p = 0.004).

Figure 3

LPS-induced a reduction of phenotypic surface marker expression on intermediate monocytes. (A,B) Representative bivariate plots showing magnetically isolated, Carboxyfluorescein succinimidydl ester (CFSE) labeled, CD16+ monocytes incubated with or without LPS for 3 h. (A) Showing unchanged forward/side scatter and (B) marked reduction in CD16 and CD14 expression. (C) LPS induced a reduction in median intensity of (C) HLA-DR (p = 0.016), (D) CD14 (p = 0.016), and (E) CD16 (p = 0.016) on CFSE labeled CD16⁺ monocytes.

Figure 4

LPS-induced ADAM17-mediated CD14 and CD16 shedding from monocytes (A) Pre-treatment of PBMCs sampled from healthy controls with the ADAM17 inhibitor TMI005 (1 μg/mL) 45 min prior to incubation with LPS for 3 h prevented LPS-induced reduction of intermediate monocytes (p = 0.001). (B) Concentration of CD14 in supernatants from PBMCs increased following incubation with LPS for 3 h (p = 0.003) and was prevented by TMI005 (p = 0.004). (C) Concentration of CD16 in supernatants from PBMCs increased following incubation with LPS for 3 h (p = 0.0001) and was prevented by TMI005 (p = 0.016). Inhibition of TNFα with (D) Infliximab (50 μg/mL) or (E) SPD304 (1 μM) had no effect on LPS-induced apparent intermediate monocyte loss (ns = not statistically significant).

Figure 5

Ex vivo culture of monocytes leads to an altered phenotype. (A) Representative flow cytometry plots showing proportions of monocyte subsets from healthy controls (B) following incubation in culture for 24 h and (C) following incubation in culture in the presence of LPS for 24 h. (D) 24 h culture with LPS results in a smaller proportion of CD14⁺⁺CD16⁺ monocytes (p = 0.003) but this population is restored with prolonged incubation in culture (ns = not statistically significant). (E) Addition of TMI005 (1 μg/mL) had no effect on LPS-mediated inhibition of monocyte maturation ex vivo, however (F) addition of LPS to PBMCs following 24 h incubation in unstimulated medium led to a reduction of intermediate monocytes in an ADAM17-dependent manner.

Figure 6

Intermediate monocytes have an altered phenotype after 24 h culture ex vivo. Representative viSNE plots (Cytobank, Santa Clara, USA) showing monocyte subsets illustrating expression of CD14 (X axis) and CD16 (Y axis). Median fluorescence intensity for each fluorochrome-conjugated antibody is represented according to the color bar shown. PBMCs sampled from healthy controls were incubated for 0 and 24 h in culture medium. (A) 24 h culture did not reduce intermediate monocytes HLA-DR expression but resulted in a reduced trend of (B) CX₃CR1, (C) CCR2, (D) TLR4, (E) TLR5, and (F) CD36. Twenty-four hours culture increased intermediate monocyte expression of (G) CD11c, had no effect on (H) CD80 and a marginal reduction of (I) CD86. PBMCs sampled from 5 healthy controls were incubated for 0 and 24 h in culture medium and the expression for each fluorochrome-conjugated antibody on intermediate monocytes was illustrated in a dot plot.

Figure 7

Cultured monocytes have reduced function. (A) Migration toward MCP-1 is impaired when PBMCs sampled from healthy controls were first incubated in medium for 24 h (p = 0.002). Monocytes retrieved from the lower chamber expressed as a percentage of the total added to the upper chamber of a transwell plate. Culture of PBMCs ex vivo impairs (B) IFNα-mediated STAT1 phosphorylation (p = 0.016), (C) LPS-induced TNFα (p = 0.056) and (D) LPS-induced IL-6 production (p = 0.032) (E) but increases LPS-induced IL-10 production (p = 0.004) as determined by flow cytometry.