• cirrhosis
• inflammation
• cytokines
• ascites

We read with great interest the study of Monteiro and colleagues1 reporting the diagnostic significance and the prognostic relevance of elevated serum concentrations of interleukin (IL)-1α in compensated cirrhosis and IL-1β in recompensated cirrhosis. While we do agree that systemic inflammation is a major driver for the progression of liver disease and acute-on-chronic liver failure, caution is advised when interpreting serum IL-1β concentrations as evidence for underlying inflammasome activation causing complications of cirrhosis. Studies1–4 have been inconsistent about stage-dependent concentrations of circulating IL-1β in cirrhosis and acute-on-chronic liver failure (ACLF). Although some of these discrepancies can be attributed to differences in patients, cohort sizes, causes and severity of liver damage, the short serum half-life of cleaved IL-1β, and the sensitivity and specificity of the assays used, one major factor presumably affecting serum concentration is the cellular source of IL-1β and its mode of release.

The release of bioactive IL-1β from murine macrophages has first been described as a two-step activation dogma. While a first (priming) signal would be provided by pathogen-associated molecular patterns (PAMPs) or inflammatory cytokines in order to upregulate the inflammasome components, a second signal would be necessary for inflammasome assembly and activation. Possible second signals are diverse and comprise microbial motifs, endogenous danger-associated molecular patterns (DAMPs) and environmental irritants.5 6 Human monocytes, which constitute the main source of IL-1β within peripheral blood mononuclear cells (PBMCs), are able to release bioactive IL-1β in response to extracellular lipopolysaccharide (LPS) alone in a process called alternative inflammasome activation,7 which is independent of potassium efflux, pyroptosome formation and pyroptosis. In addition, IL-1β can also be released by the direct binding of intracellular LPS to inflammatory caspases 4/5.8 In contrast to IL-1β, the release of the alarmin IL-1α is often the result of inflammasome-independent epithelial cell death, acknowledging some evidence for cleavage by inflammatory caspases and its release by inflammasome-mediated calcium fluxes.9 10

In order to investigate whether circulating monocytes represent a relevant source of circulating IL-1β in human cirrhosis, we isolated CD14+ monocytes from patients with acutely decompensated cirrhosis and healthy controls and determined the gene expression of inflammasome components and the inflammasome activation by PAMPs and DAMPs in vitro. Monocytes from patients with decompensated cirrhosis and ascites had markedly higher expression levels of IL1B as compared with healthy controls, whereas the expression of CASP1, PYCARD (ASC) and NLRP3 was not significantly increased (figure 1A). Consistent with monocyte priming by soluble factors, the incubation of monocytes from healthy subjects with serum from patients with decompensated cirrhosis resulted in an increased release of IL-1β in response to TLR4 ligation (figure 1B). There were no significant differences in the IL-1β release after NLRP3 inflammasome activation by ATP or nigericin (figure 1B). These data underline that primed circulating monocytes may indeed represent a significant source of elevated circulating IL-1β levels in patients with decompensated cirrhosis, which can be unrelated to canonical inflammasome activation. However, our data do not exclude canonical inflammasome activation as a source of IL-1β in patients with recompensated cirrhosis.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Monocyte-derived IL-1β in decompensated cirrhosis. (A) Gene expression of inflammasome components in monocytes from patients with decompensated cirrhosis. PBMCs were isolated from whole blood, and CD14+ monocytes were immunomagnetically enriched from patients with AD of cirrhosis (n=23), patients with AD and ACLF (n=4), and healthy controls (n=4) with a purity of >95%. Relative gene expression in unstimulated monocytes was determined by quantitative reverse transcription PCR relative to ACTB expression and displayed as violin plots. *p<0.05, **p<0.01, ***p<0.001 vs healthy controls in Mann-Whitney U test. (B) Release of IL-1β from monocytes after priming with human serum. CD14+ monocytes were immunomagnetically enriched from healthy controls (left panel) and patients with decompensated cirrhosis and ascites (right panel), seeded at a density of 100 000 per well and incubated in the presence of 20% (v/v) serum from patients with cirrhosis or controls for 18 hours. Subsequently, cells were stimulated with 10 ng/mL LPS for 24 hours or 10 ng/mL LPS for 3 hours followed by 3 mM ATP over 3 hours, 1 µM nigericin over 3 hours or 5 µg/mL poly(dA:dT) over 21 hours. Cell culture supernatants were analysed for IL-1β by ELISA (KHC0011, Thermo Fisher Scientific). Means with SEM and individual data are shown. *p<0.05, **p<0.01; in Wilcoxon’s signed-rank test. ACLF, acute-on-chronic liver failure; AD, acute decompensation; IL, interleukin; LPS, lipopolysaccharide; PBMCs, peripheral blood mononuclear cells.

Monteiro and colleagues1 did not find evidence for circulating IL-1β as a prognostic biomarker in compensated, preascitic cirrhosis stages. Whether IL-1β provides diagnostic benefit over other inflammatory biomarkers, such as IL-6, IL-10 or C reactive protein after a decompensating event remains to be further evaluated. Despite a reported 18% increased hazard of fatal ACLF per IL-1β increase by 100 pg/mL in patients with recompensated cirrhosis,1 the facts that (1) almost two-thirds had undetectable serum levels and (2) incorporating continuous serum levels did not significantly improve the diagnostic accuracy of the composite CLIF Consortium Acute Decompensation score do not yet support the use of serum IL-1β for risk stratification in clinical practice.