A role for IL-1 inhibitors in the treatment of non-alcoholic fatty liver disease (NAFLD)?

KEYWORDS:

• Canakinumab
• diabetes
• drug therapy
• metabolic inflammation
• nonalcoholic fatty liver disease

1. Introduction

Interleukin-1 (IL-1) was one of the first characterized cytokines with a receptor repertoire in most organs of the human body [1]. IL-1 type cytokines and IL-1 receptor (IL-1R) members have grown rapidly in the last decades including mainly pro- but also anti-inflammatory mediators such as IL-1 alpha (IL-1α), IL-1 beta (IL-1β), IL-1 receptor antagonist (IL-1Ra), IL-18, IL-33, IL-36, IL-37, IL-38, and others (Table 1). IL-1 type cytokines are expressed by immune cells (e.g. macrophages, dendritic cells, B lymphocytes, NK cells) and epithelial cells in a variety of organs (e.g. the liver, adipose tissue). IL-1 type cytokines activate target cells/structures via specific IL-1Rs, now referred to as IL-1Rs 1–10 [1]. Whereas IL-1α is able to be biologically active as a precursor molecule (also named ‘alarmin’), IL-1β requires processing specifically by caspase-1, which is activated by a certain inflammasome. This inflammasome contains three components: a protein-nucleotide-binding domain and leucine-rich repeat pyrin-containing protein-3 (NLRP3) or cryopyrin, an apoptosis-associated speck-like protein containing CARD (ASC) and the pro-caspase protease caspase-1 [2]. Both IL-1α and active IL-1β induce many other cytokines including chemokines and play a fundamental role in an inflammatory response. Pleiotropic functions of IL-1 are implicated in autoinflammatory and autoimmune disease, similar to inflammation consequent to infection or sterile triggers [3]. Sterile inflammation prototypically underlies nonalcoholic steatohepatitis (NASH), the inflammatory phenotype of nonalcoholic fatty liver disease (NAFLD) [4]. Many pathologies in liver diseases are cytokine-driven and various pro-inflammatory cytokines including IL-1α, IL-1β, tumor necrosis factor-alpha (TNFα) and IL-6 are involved in inflammation, steatosis, fibrosis and cancer development. Besides NAFLD, studies from the past years revealed that various IL-1 type cytokines affect insulin resistance, adipose tissue inflammation and atherosclerosis [5].

Table 1. IL-1 type cytokine members as potential targets for IL-1 interfering therapies.

Cytokine	Receptor	Biological activity
IL-1 subfamily
IL-1α	sIL-1R1, IL-1R1/3	Pro-inflammatory
IL-1β	sIL-1R2, IL-1R2/3	Pro-inflammatory
IL-1Ra	IL-1R1	Anti-inflammatory
IL-33	IL-1R3/4	Pro-inflammatory
IL-18 subfamily
IL-18	IL-1R5/7; IL-18BP	Pro-inflammatory
IL-37	IL-1R5/8	Anti-inflammatory
IL-36 subfamily
IL-36Ra	IL-1R3/6	Anti-inflammatory
IL-36α	IL-1R3/6	Pro-inflammatory
IL-36β	IL-1R3/6	Pro-inflammatory
IL-36γ	IL-1R3/6	Pro-inflammatory
IL-38	IL-1R1/3/6	Anti-inflammatory

Legend: IL-18bp, IL-18 binding protein: binds to receptor and blocks IL-18 activity.

2. The role of IL-1α/β in steatosis and steatohepatitis and potential therapeutic approaches

NAFLD is currently the major cause of liver disease worldwide [4]. It is now anticipated that NAFLD is the hepatic manifestation of the metabolic syndrome and NAFLD refers to a large spectrum of liver diseases ranging from rather benign hepatic steatosis to steatohepatitis (i.e. NASH), cirrhosis and hepatocellular carcinoma. Its current pathophysiology suggests a complex interplay of diverse metabolic, genetic and environmental factors, including a yet incompletely defined interaction between the gut microbiota and innate immunity. The final net result, i.e. liver inflammation and fibrosis, is caused by several ‘hits’ which include lipotoxicity, gut-derived signals such as endotoxin and other metabolites that activate toll-like receptors (TLRs) and pro-inflammatory cytokines, oxidative stress. Moreover, inflammatory signals derived from the adipose tissue fuel metabolic inflammation [6,7]. However, clinicians still lack valid biomarkers to define advanced stages of the disease, for example, IL-1 type-cytokines such as IL-1Ra may serve as a potential biomarker for the assessment of inflammatory activity in NAFLD in the future [8].

The role of IL-1α/β in steatosis and steatohepatitis has been investigated in the past years [9]. Kamari and colleagues established that diet-induced NASH induced hepatic expression of IL-1α/β in mice. Despite less hepatic inflammation in both IL-1α and IL-1β-deficient mice after 18 weeks of an atherogenic diet, only IL-1α^−/- mice had increased steatosis and hepatic/plasma cholesterol levels after 10 weeks of such a diet. Therefore, this study implicates that IL-1α contributes to both steatosis and inflammation, whereas IL-1β is mainly relevant for the inflammatory phenotype in this experimental model. This study suggests that both IL-1 type cytokines play a crucial but somewhat different roles in diet-induced liver steatosis and inflammation. Furthermore, hepatic and not bone-marrow-derived-IL-1α/β deficiency was protective against diet-induced inflammation supporting a key role for liver-derived IL-1 type-cytokines in NASH pathologies. In accordance with Kumari´s study, caspase-1 KO mice were also protected from high-fat diet (HFD)-induced hepatic steatosis, inflammation and fibrogenesis [10]. The role of other IL-1 type cytokines in NAFLD models remains unresolved. One study suggested that IL-33 attenuated liver steatosis but worsened fibrosis [11] whereas another preclinical study failed to demonstrate any role for endogenous IL-33 in the progression toward fibrosis in HFD-induced steatohepatitis [12]. A potential role for IL-1 type cytokines in the progression of NAFLD has also been suggested by preclinical studies. The transcriptomic signature in the natural course of murine NAFLD progressing from steatosis toward steatohepatitis is characterized by activation of IL-1 type cytokines [13]. The role of IL-18 in NAFLD remains similarly unclear. One study observed that IL-18^−/- mice develop hypercholesterolemia and finally NAFLD while administration of recombinant IL-18 improved dyslipidemia [14]. All these preclinical studies support the notion that IL-1 pathways including its processing inflammasome are crucially involved in obesity-related liver inflammation.

Insulin resistance has been suggested as one of the key pathophysiological features in human NAFLD [4]. Stienstra and colleagues demonstrated that caspase-1 and IL-1β activity increase in adipose tissue of genetically obese mice as well as after exposure to an HFD [15]. Mice deficient in caspase-1 were more insulin-sensitive and treatment of obese mice with a caspase-1 inhibitor improved insulin sensitivity. Several pro-inflammatory cytokines including IL-1 type cytokines and TNFα have been proposed to play a role in the regulation of inflammation-associated insulin resistance. The IL-1β expression is increased in the adipose tissue of obese insulin-resistant mice and adipocyte-derived IL-1β is able to influence insulin sensitivity in the liver. IL-1β affects insulin signaling by reducing insulin receptor substrate 1 expression through an ERK-dependent mechanism [16]. We recently observed that IL-37 reflects a key IL-1 type cytokine ameliorating obesity-induced inflammation and insulin resistance [17]. In this study, mice transgenic for human IL-37 (IL-37tg) demonstrated reduced numbers of adipose tissue cells (mainly macrophages), increased circulating levels of adiponectin and preserved glucose tolerance and insulin sensitivity after 16 weeks of HFD. Importantly, in humans, increased steady-state IL-37 adipose tissue mRNA levels were positively correlated with insulin sensitivity. This is in accordance with data published by our group showing that massive weight loss after bariatric surgery results in a substantial increase in adipose IL-37 expression [18]. In these human studies, we identified especially the adipose tissue as a major source of IL-1 type cytokines [17,18]. The most relevant data that IL-1 type cytokines contribute to insulin resistance come from a human study demonstrating that neutralization of IL-1 by IL-1Ra improved hyperglycemia and glycemia [19]. In hemodialysis patients, it was observed that administration of IL-1Ra subcutaneously over 4 weeks was accompanied by an increase in circulating adiponectin levels whereas homeostatic model assessment of insulin resistance (HOMA) was not affected [20]. Further studies targeting pro-inflammatory cytokines such as IL-1β or TNFα in type 2 diabetes are eagerly awaited.

Ablation of NLRP3 in mice prevents obesity-induced inflammasome activation in adipose and liver tissue, reduces IL-18 adipose tissue expression and improves metabolic functions [21]. Certain NLRP3 inhibitors such as MCC950 have improved NAFLD pathology including liver histology in diabetic mice [22]. In this trial, treatment with MCC950 was performed over 24 weeks and this treatment lowered circulating concentrations of IL-1β, IL-6, and MCP-1. In addition, NLRP3 deletion was able to abrogate liver pathology including the amount of inflammation after a challenge with palmitic acid [23]. These studies suggest that targeting NLRP3 by small molecules could reflect an attractive treatment strategy for NAFLD also in humans.

3. Expert opinion

In many diseases including NAFLD, and especially in the case of NASH multiple cytokines contribute to disease pathogenesis, as investigated in disease models in mice. Over the last three decades, members of the IL-1 cytokine family are increasingly recognized as key players in acute and chronic inflammatory conditions including metabolic inflammation [3]. IL-1 type cytokines mediate key features of NAFLD including hepatic steatosis, inflammatory infiltrates, fibrosis, insulin resistance and adipose tissue inflammation [5]. Interestingly, IL-1β also exhibits a physiologic role together with insulin in the regulation of glucose metabolism and immunity [24]. In this study, IL-1β and insulin promoted each other´s synthesis suggesting a role for insulin in the postprandial secretion of IL-1β. It is well established in the cytokine field that low concentrations of these mediators, e.g. also TNFα are physiologically relevant and important, but still raises a word of caution that blockade of IL-1β could have also a negative impact on glucose control [24]. It has also to be acknowledged that in the largest so far reported study blocking IL-1β (see below) there appeared no evidence of hypoglycemia and/or glucose-lowering effects. Several preclinical studies convincingly showed a crucial role for IL-1 type cytokines in NAFLD, similar to blockade of regulatory signaling hubs that process IL-1β (e.g. the NLRP3 inflammasome). These experimental studies indicate that blocking IL-1β has beneficial effects on certain aspects of metabolic diseases. Moreover, IL-1-type cytokines such as IL-1α and IL-36 might be involved in metabolic inflammation and importantly IL-1R3 has been recently shown to be the co-receptor in various signaling pathways involving six cytokines of the IL-1 family (IL-1α, IL-1β, IL-33, IL-36α, IL-36β, IL-36γ) [25]. In this elegant study, neutralization of IL-1R3 by a monoclonal antibody was able to block various pathologies including peritonitis, allergic inflammation and psoriasis suggesting that such a treatment strategy could efficiently target various IL-1-type cytokines.

In humans, the most compelling evidences regarding a major role for IL-1β in metabolic dysbalance and inflammation came from the recently published CANTOs trial [26]. In this randomized, double-blind study, 10,061 patients with high cardiovascular risk (i.e. with previous myocardial infarction and elevated high-sensitivity C-reactive protein) either received a placebo or canakinumab, a monoclonal antibody targeting IL1-β, and patients were followed-up for ~3.7 years. Canakinumab improved metabolic inflammation and protected against non-fatal myocardial infarction and stroke but had no effect on serum low density and high-density lipoprotein levels. Whether this approach is effective in patients with less advanced cardiovascular disease is currently unclear. It is notable that canakinumab treatment increased the risk for fatal infections, likely because IL-1β is required for an appropriate immune response. Interestingly, the rate of new-onset type 2 diabetes was not reduced by canakinumab which, however, does not rule out a critical involvement of IL-1β in pancreatic disease processes [27]. Unfortunately, no data on hepatic steatosis is available from this trial, although it would be highly informative whether canakinumab impacts NAFLD in humans. Such data could provide the basis for IL-1β studies in human NAFLD which would be strongly supported by experimental data in mice. Clinical studies targeting the role of IL-1 cytokines by various strategies (e.g. monoclonal antibodies, small molecules) are eagerly awaited for human inflammatory liver diseases.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.