Inflammasomes Assemble: Cytoplasmic Guardians or Fallen Heroes?

Inflammation is a pivotal response of the innate immune system and yet also constitutes a fundamental pathophysiological driver of acute respiratory distress syndrome (ARDS). Pattern recognition receptors (PRRs) are the principal mechanisms by which innate immune cells detect pathogen- and damage-associated molecular patterns (PAMPs and DAMPs). Whilst membrane bound PRRs, such as toll-like receptors (TLRs), have been extensively studied in inflammatory diseases, the last two decades have seen an increased interest in role of the cytoplasmic PRRs called nucleotide-binding oligomerization domain-like receptors (NLRs). In a seminal study, Tschopp and colleagues first coined the term “inflammasome”, describing a multi-protein cytosolic complex that was assembled upon activation of NLRs and was responsible for caspase-1 activation, leading to cleavage of the pro-inflammatory cytokines interleukin-1 beta (IL-1β) and interleukin-18 (IL-18) to their active form.(1) Inflammasomes assemble in response to diverse molecular patterns, both endogenous and exogeneous, leading to downstream activation of pathways that participate in host defense and anti-microbial activity through immunomodulation and pyroptosis.(2)

Aberrant or dysregulated inflammasome activation is associated with several disease processes including autoimmune and metabolic disease and tumorigenesis.(2) In experimental models, dysregulated inflammasome activity has been associated with a wide array of lung diseases such as bacterial and viral pneumonia, chronic obstructive pulmonary disease, and lung fibrosis.(3) Inflammasomes have also been implicated in experimental models of acute lung injury (ALI). Mechanical stress has been associated with activation of the NLRP3 inflammasome in macrophages leading to ALI.(4) Likewise, inhibition of the NLRP3 inflammasome has been shown to be protective in lipopolysaccharide induced ALI.(5) While there is some data linking increased plasma levels of IL-18 to increased risk of mortality in humans with ARDS,(6) data directly implicating the inflammasome in human ARDS remains limited.

To further investigate the inflammasome in human ARDS, in this issue of the journal, Rogers and colleagues present a secondary analysis of plasma IL-18 levels in a large cohort of infection-associated ARDS patients enrolled in the SAILS trial testing rosuvastatin versus placebo (7). The main findings of the study were that higher levels of IL-18 at baseline were associated with adverse clinical outcomes, including mortality. Rising levels of IL-18 at day 3 were also predictive of adverse clinical outcomes, regardless of treatment randomization. Interestingly, the investigators found that randomization to rosuvastatin was associated with a higher proportion of patients with rising values of IL-18 compared to placebo.

The study is the largest to test the utility of plasma IL-18 in ARDS and the first, to our knowledge, in patients from a randomized controlled trial (RCT). The results are all the more compelling because the tested hypotheses were generated from prior experimental work by the investigators linking statins, the inflammasome, and lung injury. In one of these prior studies, the investigators identified inflammasomes as key mediators in a murine model of ventilator-induced lung injury and reported elevated plasma levels of IL-18 in a smaller human ARDS cohort.(6) Furthermore, in a separate study using a large cohort of smokers and former smokers, they found that statin therapy was associated with increased risk for interstitial lung abnormalities on computed tomography scanning.(8) When stratified by statin type, the adjusted increased risk of interstitial lung abnormality was only significant in patients treated with hydrophilic, and not lipophilic, statins. In the same paper, using a bleomycin mouse model, pre-treatment with pravastatin (a hydrophilic statin) enhanced lung injury, with higher levels of IL-1b and IL-18 in BAL and lung homogenate, further implicating the inflammasome. This experimental evidence supports that the association between rising IL-18 and rosuvastatin treatment observed in the present study may in fact be real.

If rosuvastatin does indeed potentiate inflammasome activation in ARDS, it is notable that clinical outcomes in the SAILS trial were similar between the rosuvastatin and placebo groups. It may be that the magnitude of inflammasome activation induced by brief treatment with rosuvastatin is not sufficient to cause adverse clinical outcomes. Alternatively, and perhaps more consistent with the observed findings in this paper, it may be that only a subset of patients treated with rosuvastatin develop inflammasome activation, as reflected by rising IL-18, and thus adverse clinical outcomes are not observed when the group is considered as a whole.

Connecting the dots, the main findings from the presented and prior studies suggest an association between inflammasome activity and clinical outcomes in an ARDS. Several questions, however, remain unanswered. Functionally, IL-18 bioactivity in circulation is limited by two factors, an inhibiting binding protein known to be upregulated in infection,(9) and levels of IL-18 receptor expression(10), neither of which were measured in the study. In patients with sepsis, Novick and colleagues found elevated levels of IL-18 binding protein leading to a decrease of free bioactive IL-18.(11) From the presented data, it is not possible to determine whether the measured levels of IL-18 are in a bound or free state. The kinetics of IL-18 binding protein and receptor expression in acute inflammation have been poorly studied, and a better understanding is needed before its role as a causative agent in dysregulated inflammation is fully conceived.

Other factors add to the uncertainty of whether the association between IL-18 and ARDS outcomes is causative or merely corollary. Inflammasome-driven caspase-1 is the main proteolytic cleaver of IL-18 to its active form, but inflammasome-independent enzymes can also increase free IL-18. Ligation of the transmembrane cell-death receptor Fas leads to caspase-8 activation, which is known to cleave IL-18 to its active state.(12) Proteinase-3, a proteolytic enzyme expressed by neutrophils and known to be highly expressed during systemic inflammation, can also perform this cleavage.(13, 14) Both enzymes are likely to be encountered in the milieu of severe inflammation and may contribute to elevated IL-18 levels in this setting. These inflammasome-independent pathways mandate a cautionary approach to interpreting the specificity of IL-18 as a marker of inflammasome activity.

IL-18 aside, several other pro-inflammatory markers have been associated with worse clinical outcomes in ARDS but have yet to change clinical practice.(15) The logical question to then ask is whether a single biomarker, in isolation, could ever sufficiently capture the complexity of information transmitted during the systemic inflammatory response such that its measurement could inform practice? Acute systemic inflammation is a series of complex networked responses that probably have non-linear interactions. Moreover, the multitude of highly conserved innate inflammatory pathways clearly have an important role to play in resolving the noxious insult. Indiscriminate abrogation of a single pathway in favour of another, regardless of severity of the response, may lead to failure or even harm. A recent secondary analysis by Meyer and colleagues found that recombinant IL-1 receptor antagonist therapy in sepsis, counterintuitively, seemed to benefit patients with higher endogenous plasma IL-1 receptor antagonist levels(16). This study emphasizes the limitations of using a single biomarker to capture the state of the inflammatory response. Better detection and differentiation of dysregulated inflammatory responses from protective ones is of the impending challenges of critical care research and will likely require a composite of biological variables.

The findings of this study which implicate rosuvastatin as a cause of increasing IL18 over time raise a further question as to whether the story of statins in sepsis and ARDS is over? After two recent ARDS RCTs (HARP-2 and SAILS) showed no benefit with statin therapy, it may have seemed so.(17, 18) A recent secondary analysis of the HARP-2 trial from our group, however, found a survival benefit with simvastatin therapy compared to placebo in a subgroup of patients with hyper-inflammatory characteristics.(19) Conversely, the same analytical process in SAILS found no benefit with rosuvastatin in the “hyper-inflammatory” subgroup.(20) Molecular differences in the type of statin used, namely, rosuvastatin being hydrophilic, and simvastatin being a lipophilic molecule, may have contributed to these findings. The putative anti-inflammatory benefits of statin therapy have been demonstrated primarily with lipophilic molecules.(20) Differential response to statin molecule-type has also been reported with inflammasome activation in peripheral blood mononuclear cells (PBMCs). Patients randomized to 8 months of treatment with a lipophilic statin (atorvastatin) showed diminished PBMC NLRP3 activation, reduced plasma IL-18 and IL-1β levels, whereas in patients randomized to a hydrophilic statin (rosuvastatin), these values remained unchanged.(21) These findings are in line with the prior findings of the investigators of the presented study.(8) Taken together with the findings of increased IL-18 with rosuvastatin therapy in the presented study, it would seem unlikely that hydrophilic statins warrant further evaluation in ARDS or sepsis.

Whilst simvastatin may someday offer potential as a therapeutic intervention in selected patients with ARDS, its continued pursuit as a meaningful intervention highlights the paucity of novel disease-modifying and targeted therapies currently available in critical care. It is here that the presented study by Rogers and colleagues may have its greatest impact. If, as suggested by these investigators, the inflammasome plays a role in the pathogenesis of ARDS, it could potentially provide fruitful sets of pathways and molecular targets to seek novel interventions. Future high-dimensional clinical and biological data will help us better understand the complexity of acute inflammation and to translate this line of research to clinical practice.