The interleukin-1 (IL-1) superfamily encompasses eleven structurally and evolutionarily related cytokines that mainly act by binding to specific receptors (1). The IL-1 receptor (IL-1R) complexes are formed by a ligand-binding chain and an accessory chain necessary for signal transduction. The IL-1R family comprises ten related transmembrane molecules with extracellular Ig-like domains and intracellular TIR (Toll-Like Receptor/IL-1 Receptor) domains (2). Most IL-1 and IL-1R molecules are involved in inflammation/immunostimulation and its regulation. It is notable that, while IL-1R are derived from invertebrate TIR-containing receptors involved in innate immunity, IL-1 cytokines only appeared in vertebrates (likely linked to the development of adaptive immunity) and the majority of them is only present in mammals (3, 4).

In mammals, eleven cytokines of the IL-1 family have been identified, the inflammatory IL-1α, IL-1β, IL-18, IL-36α, IL-36β, IL-36γ, and the regulatory/anti-inflammatory IL-1Ra, IL-33, IL-36Ra, IL-37 and IL-38 (overview in 1). IL-1 family cytokines mainly act by binding to receptor complexes, composed by a ligand binding chain and an accessory signal transducing chain. Signalling is initiate by the pairing of the intracellular TIR domains of the two receptor chains. Receptors of the IL-1R family are ten structurally related transmembrane proteins, which may also give rise to soluble receptors upon gene splicing or proteolysis of the membrane chains (overview in 2). IL-1R1 binds the agonist ligands IL-1α and IL-1β, and its signalling depends on the engagement of the accessory chain IL-1R3. IL-1R1 also binds IL-1Ra but in this case fails to recruit the IL-1R3, and no activation occurs. Binding of IL-38 has also been reported. IL-1R2 is similar to IL-1R1 in its ligand binding capacity and IL-1R3 engagement, but it lacks the intracellular TIR domain and its therefore an inhibitory receptor. IL-1R4 is the receptor for IL-33, an uses the same IL-1R3 as accessory chain for signalling. IL-1R5 is the receptor for IL-18, and uses its specific accessory chain IL-1R7. It was reported that IL-37 also binds to IL-1R5 but is unable to recruit IL-1R7. IL-1R6 is the ligand binding chain for all the IL-36 isoforms (IL-36α, IL-36β, IL-36γ) and uses the promiscuous IL-1R3 as accessory chain. IL-1R6 also binds the receptor antagonist IL-36Ra and fails to engage the accessory chain. Binding of IL-38 has also been reported. IL-1R8 is an inhibitory receptor, which can interact with several IL-1R complexes and interfere with their signalling and may also act as inhibitory accessory chain for IL-37 bound to IL-1R5 and IL-38 bound to IL-1R6. IL-1R9 is an orphan receptor, although binding to IL-38 has been reported, predominantly expressed in neurons and likely involve in IL-1-independent neuronal functions. Likewise, IL-1R10 is abundantly expressed in the brain and has no IL-1-dependent functions (for complete references, please see the reference lists of 1, 2).

IL-1 and IL-1R molecules expressed in brain have as their main role the modulation of neuronal plasticity and function, in addition to mediating immune protective functions, in both physiologic and pathologic condition (5–8). IL-1, originally described as endogenous pyrogen or leukocytic pyrogen, is mainly produced in the hypothalamus in response to inflammatory agents and can induce fever by upregulating cyclooxygenase-2, the enzyme responsible for the synthesis of the vasoactive and inflammatory prostaglandin E₂ (9, 10). A central nervous system (CNS)-restricted expression of splice variants of different IL-1 and IL-1R family members, as truncated forms with regulatory functions, have been identified for both IL-1 (11) and IL-18 (12–14). In the case of IL-37, only one of its five splice variants (IL-37a) is present in the brain (15). Some IL-1R members are only present in the brain, including a particular variant of the accessory receptor chain IL-1R3 (IL-1R3b), expressed in neurons and involved in IL-1-mediated neuroprotection. Notably, the receptor complex for IL-1α and β in the brain can be the canonical one (IL-1R1 plus IL-1R3, present everywhere in the body) that mediates inflammatory effects, or the alternative one (IL-1R1 plus IL-1R3b, specific for the brain) that mediates neuroprotective effects (2, 16–18). Other brain-specific receptors are the two orphan receptors IL-1R9 and IL-1R10, which are mostly expressed in neurons. IL-1R9 is likely involved in memory and learning capacity, as its mutations are linked to X-linked mental retardation and autism (2, 19, 20). The functions of IL-18 in the brain are also peculiar, since the cytokine is active (inflammation and homeostasis) even in the absence of its receptor IL-1R5, while IL-1R5 seems to be involved in the pathogenesis of experimental allergic encephalomyelitis (EAE, the mouse model of multiple sclerosis) in the absence of its IL-18 ligand (21, 22). This suggests that a CNS-specific network of IL-1 family cytokines and receptors is crucial in maintaining brain homeostasis, thus highlighting the limitation of a simplistic inflammatory paradigm to explain the function of these cytokines in neuroinflammatory diseases.

In this perspective, we examine the available experimental and clinical evidence of the involvement of these molecules in two major neurodegenerative diseases, Alzheimer’s disease (AD) and multiple sclerosis (MS), and propose a more global picture of the pathogenic and pathological role of the IL-1/IL-1R system.

The NOD-like receptor 3 (NLRP3) inflammasome, a multiprotein complex that regulates production and release of inflammatory cytokines, in the brain is mainly expressed in microglia but also in astrocytes and neurons. Besides being involved in protection from bacterial, fungal and viral agents, NLRP3 plays a role in neurodevelopment and neuroprotection, but its aberrant activation is crucial in neuroinflammation (113).

Different pathways, including impaired autophagy and dysregulated cell death mechanisms, upregulate NLRP3 activity in AD (114). Aβ and tau can directly activate NLRP3 in microglia (115–117), and several compounds that ameliorate AD-associated pathology reduce NLRP3 activity.

Strong evidence points to a critical role of inflammasome activation in MS. Caspase-1 is more expressed in peripheral blood mononuclear cells (PBMC) of MS patients than in controls (118). Its serum levels can be considered a biomarker of MS, even if they do not identify more active or more severe patients (101). In untreated patients, upregulation of NLRP3 gene expression and higher caspase 1 activity were detected in circulating monocytes (119). NLRP3 upregulation characterized patients with PPMS and was associated with hyperexpression of several inflammatory chemokines and cytokines. IL-1β was one of the most expressed genes, correlated with faster disease progression in PPMS (119). There is a marked increase in the number of IL-1β, caspase-1, and gasdermin D (GSDMD)-positive cells in white matter of MS versus non-MS patients. GSDMD, activated by inflammatory caspases, forms membrane pores responsible for pyroptosis an inflammasome-driven programmed cell death (120). In MS lesions, inflammasome activation and pyroptosis are detected in microglia and oligodendrocytes, suggesting a role in demyelination. Inhibition of inflammasome activation and caspase-1-dependent IL-1β production have already been exploited in MS therapy: IFNβ treatment reduces NLRP3 activity and induces IL-10, thus decreasing the production of pro-IL-1α, pro-IL-1β and mature cytokines (121).

In experimental autoimmune encephalomyelitis (EAE), the mouse MS model, blockade of caspase-1 expression or activity attenuates neuroinflammation (122). Caspase-1 inhibition also reduces GSDMD activation and pyroptosis, limiting oligodendrocyte death and demyelination (120). Similarly, NLRP3 KO or IL-18 KO mice have a less severe disease course, with a marked reduction of CNS inflammatory infiltrate (123, 124). Microglia-specific deletion of the gene encoding the NFκB regulatory protein A20 leads to microglial cell proliferation, susceptibility to LPS-induced inflammation and increased EAE severity, reversed by NLRP3 or caspase-1 genetic deletion (125), indicating a critical role of microglia in MS, mediated by inflammasome activation. The inflammasome involvement in neuroinflammation is also suggested by data in a cohort of MS patients, in which low penetrance mutations of the NLRP3 gene were detected in 16% of the cases (126). These mutations are associated with mild symptoms (suggestive of autoinflammatory disease) and increased production of IL-1β and IL-18 (127).

Inflammation is involved in other neurological diseases, both neurodegenerative (see below) and of traumatic or other origin (trauma, stroke, autism, schizophrenia, epilepsy). Focusing on neurodegenerative disorders, these are classified in four main groups: amyloidoses, tauopathies, alpha-synucleinopathies, and TDP (transactivation response DNA binding protein)-43 proteinopathies, all characterised by abnormal protein conformations and the formation of protein aggregates in neuronal cells in different brain areas (128, 129). AD is the most common and prevalent of both amyloidoses and tauopathies, along with frontotemporal dementia with parkinsonism and some forms of prion diseases. Main α-synuclein-related diseases include Parkinson’s disease (PD), Dementia with Lewy Bodies, and Multiple System Atrophy, while Amyotrophic Lateral Sclerosis (ALS) is mainly associated with TDP-43 pathology. The Huntington disease (HD) is associated with mutations in the huntingtin protein, which clumps within neurons provoking cell damage and death (130).

Although it is well known and documented that all neurodegenerative diseases share the chronic aberrant inflammation (131), less is known on the role of IL-1 family cytokines and receptors in specific neurodegenerative diseases (132). In addition to AD and MS, reviewed in the previous sections, here we will briefly examine the involvement of IL-1 family molecules in PD, HD, and ALS.

PD is characterised by the death of dopaminergic neurons in the substantia nigra and by the cytoplasmatic aggregation of fibrillar α-synuclein in Lewy bodies (133). Increase of α-synuclein released outside the cells drives the activation of microglia, and the microglia-dependent production of TNF-α, NO, and IL-1β sustains the neuroinflammatory process in PD (134). Recent results in the 6-OHDA mouse model showed that the increased levels of IL-1β and TNF-α have a different kinetics, with TNF-α appearing in the advanced disease stage, while increased level of IL-1β in serum were already evident with a moderate degree of lesion in substantia nigra, suggesting a prognostic value for this inflammatory cytokine (135). Although the results on the level and role of IL-1β in PD patients are conflicting (136), the activation of the NLRP3 inflammasome seems to be determinant for the development of PD at least in vitro. Indeed, when activated, the inflammasome can cause the aggregation of α-synuclein in a neuronal cell model of PD, as caspase-1 directly cleaves α-synuclein generating a truncated protein highly prone to aggregation and able to provoke cell death (137). The involvement of the NLRP3 inflammasome activation in PD also includes two additional mechanisms, the interaction with the parkin protein, whose mutations are responsible of autosomal recessive familial and sporadic early-onset PD (138, 139); and the association with mitochondrial dysfunction that affects neuron performance and favours neuronal degeneration (140). Although IL-1β is a major product of the NLRP3 inflammasome activation, its role in causing/sustaining inflammation in PD patients is still debated (141).

Increasing evidence suggest that inflammation contributes to the development of HD, but the involvement of IL-1 family is still poorly known. The role of IL-1β has been investigated in mice expressing a mutant and pathogenic huntingtin crossed with animals knocked out for IL-1R1 (142). These mice showed more severe neurological symptoms compared to control mice expressing IL-1R1, suggesting a protective role for IL-1R1 signalling in preventing the HD neuropathology. Whether IL-1β is the molecule that activates IL-1R1 signalling is not known. Although IL-1β has been identified as one of the inflammation-related markers of the disease in the porcine HD model (143), its level does not increase in plasma of HD patients, at variance with IL-6 or other inflammatory factors (144, 145). An involvement of the NLRP3 inflammasome has been also suggested for HD, although evidence is still poor and mostly from pre-clinical studies (146).

ALS is a neuromuscular disorder characterized by the progressive loss of anterior-lateral horn spinal cord motoneurons (147). Sporadic ALS affects the 90-95% of patients, while the familiar form affects the remaining 5-10% of patients. The two most relevant genetic mutations associated with ALS are a defect in the hexanucleotide repeat expansion (HRE) in intron 1 of the C9orf72 gene (which makes a protein that regulates actin dynamics and endosomal recycling of GluR1 at the synapse) (148), and in the free radical scavenging enzyme Cu,Zn-superoxide dismutase (SOD1) gene (149). The involvement of inflammation in the etiopathogenesis of ALS is starting to be investigated, but studies are still limited (150, 151). In a recent study, the authors suggest hyperinflammation and immunodeficiency as primary triggers of motoneuron death in ALS, and underline the complex link between the development ALS and T cell and monocyte profiles of patients, as well as polymorphisms in cytokine and chemokine receptors, suggesting a need for personalised therapies based on immunophenotyping (152). In this regard, recently new polymorphisms in the gene encoding IL-1β and in genes involved in oxidative stress (e.g., SOD2) have been identified as modifiers of ALS progression (153). IL-1β is one of the main biomarkers of inflammation in ALS (151, 152). Even if it is not the best marker of ALS severity (154), its plasma levels negatively correlate with survival in ALS patients with genetic variant C9orf72HRE (155). Among the IL-1 family cytokines and receptors, it has been observed that only IL-18 is associated with sporadic ALS, whereas the serum levels of the cytokines IL-33 and IL-36 and the soluble receptors sIL-1R2 and sIL-1R4 were comparable between ALS patients and healthy controls, and IL-1β, IL-1Ra and IL-37 were below detection (156). The possible pathogenic role of IL-18 in ASL has been also recently underlined by a Whole Genome Sequencing study, which identified a genetic variant in a noncoding region of the gene encoding the IL-18 receptor accessory protein (IL18RAP) able to reduce mRNA stability and motor neuron neurotoxicity, thereby decreasing 5x the risk of developing ALS (157). A protective role has been suggested for another member of IL-1 family, IL-33 (158). In the ALS mouse model transgenic for G93A-superoxide dismutase 1 (SOD1-G93A), long-term IL-33 administration delays disease onset in females but not males, probably through peripheral Th2 response (159).

The role of IL-1 family cytokines and receptors in the etiology and pathogenesis of neurodegenerative diseases is still poorly understood. However, there is strong evidence of their involvement in the initiation and regulation of inflammation associated to most CNS diseases, and their participation to neuroinflammation has been widely described in AD an MS. This is evident despite the limitations due to the current lack of reliable animal models that realistically recapitulate the pathological features of the human diseases, such as in the case of sporadic AD (160). In humans, longitudinal molecular monitoring of inflammation is also hampered by the reduced accessibility of human brain and the limited availability of clinical studies with extensive follow-ups (161, 162).

From the combined in vivo and in vitro functional and molecular evidence, it is clear that these factors play a homeostatic and neuroprotective role in healthy conditions. It is notable that the IL-1 family system in the CNS includes several molecules and functions that are CNS-restricted, implying tissue-specific needs and consequent functional adaptation. Disease conditions imply an imbalance of the IL-1 system network ( Table 1 ).

From the available experimental and clinical evidence, we can draw the following conclusions and suggestions.

DB, PM, PI and PB wrote the manuscript and critically revised it. DB and PI contributed to figures and tables. All authors contributed to the article and approved the submitted version.