Abstract
IFNs are well known as mediators of the antimicrobial response but also serve as important immunomodulatory cytokines in autoimmune and autoinflammatory diseases. An increasingly critical role for IFNs in evolution of skin inflammation in these patients has been recognized. IFNs are produced not only by infiltrating immune but also resident skin cells, with increased baseline IFN production priming for inflammatory cell activation, immune response amplification, and development of skin lesions. The IFN response differs by cell type and host factors and may be modified by other inflammatory pathway activation specific to individual diseases, leading to differing clinical phenotypes. Understanding the contribution of IFNs to skin and systemic disease pathogenesis is key to development of new therapeutics and improved patient outcomes. In this review, we summarize the immunomodulatory role of IFNs in skin, with a focus on type I, and provide insight into IFN dysregulation in autoimmune and autoinflammatory diseases.
Introduction
The skin comprises a critical physical and chemical barrier, and resident skin cells and resident and migratory immune cells secrete immunomodulatory proteins to protect us from colonization and invasion by foreign microorganisms. IFNs are one important class of signaling proteins secreted to combat potential infection. Although IFNs can serve a protective role, they also contribute to the pathogenesis of autoimmune and autoinflammatory diseases. Pathognomonic skin lesions frequently herald systemic autoimmune disease onset and represent key features that assist in diagnosis. Critically, an important role for type I IFNs in cutaneous and systemic disease pathogenesis has been recognized.
Types and role of IFNs
There are three main classes of IFNs, with type I IFNs representing the largest class. Type I IFNs in humans encompass 17 members, including 13 IFN-α subtypes, IFN-β, IFN-ω, IFN-ε, and IFN-κ (1). Although most cell types produce IFN-β, the primary producers of IFN-α are hematopoietic cells (2). There exists only a single type II IFN, IFN-γ, which is produced predominantly by T and NK cells, and assists in the antiviral immune response (3). The third class of IFNs, type III IFNs, is comprised of four IFN-λ subtypes (IFN-λ 1, 2, 3, or 4), with receptor expression mostly restricted to epithelial cells, myeloid cell subsets, and neuronal cells (4). IFN-λ is structurally similar to IL-10 family cytokines and has similar signaling effects to type I IFNs, exhibiting a role in the antimicrobial response of epithelial cells (5).
Type I IFNs serve an important immunomodulatory role in healthy skin, leading to promotion of Ag presentation and NK cell activation through shaping the innate immune response, activation and augmentation of the adaptive immune system, and induction of an antimicrobial state (1, 2). Type I IFNs also perform an antiproliferative and tumor immune surveillance role. Type II IFNs can specifically inhibit keratinocyte proliferation (6). Type III IFNs are induced by nucleic acid signaling, but their overall function in the skin requires additional study.
Downstream signaling of type I IFNs
All type I IFNs bind to the same heterodimeric transmembrane receptor, the IFN-α/βR (IFNAR), composed of IFNAR1 and 2; however, binding affinity and tissue-specific receptor expression can influence biological activity of the type I IFNs (7–10). The IFNAR is found on nearly all nucleated cells. Binding of type I IFNs to the IFNAR leads to activation of the JAK–STAT pathway (11). IFNAR1 is associated with tyrosine kinase 2 (TYK2), and IFNAR2 is associated with JAK1. Once JAK1 and TYK2 are activated, they phosphorylate tyrosine residues on cytoplasmic tails of the IFNAR, which serve as binding sites for the Src homology 2 (SH2) domain of STAT proteins 1–6 (12, 13). Relative STAT expression also influences specific STAT activation. Classically, a phosphorylated STAT1-and-STAT2 dimer translocates to the nucleus and associates with IFN regulatory factor (IRF) 9, resulting in formation of the IFN-stimulated gene factor 3 (ISGF3) complex (14). ISGF3 then binds to IFN-stimulated response elements (ISREs), resulting in activation of IFN-stimulated genes (ISGs) (Fig. 1).
Pathways of IFN dysregulation in skin lesions of patients with autoimmune and autoinflammatory diseases. CLE is shown as an example in (A). (B) represents pathways dysfunctional in AGS in red, CANDLE syndrome in gray, and SAVI in pink. IC, immune complex.
Localization of IFNs in skin
In healthy resident skin cells, keratinocytes produce IFNs at baseline, with minimal to no detectible production from fibroblasts or endothelial cells (15). This baseline IFN production is a result of chronic IFN-κ production, with no apparent contribution from other type I IFNs (15, 16). IFN-κ expression increases upon treatment with type I or II IFNs (16), and chronic elevation of baseline IFN-κ amplifies basal IFN responses (15). In inflammatory states, exposure to cytokines like TNF-α or antimicrobial peptides prime for additional type I IFN expression, particularly IFN-β (17).
Upon appropriate stimulation, many cell types are capable of type I IFN production. Immune cells, similar to keratinocytes, are poised to respond more rapidly to low levels of IFNs. Plasmacytoid dendritic cells (pDCs) are able to rapidly secrete large amounts of IFN-α and are known to accumulate in key tissues affected by inflammation in rheumatic disease. In autoimmune skin lesions, pDCs are recruited to the dermal–epidermal junction (18), contributing to interface dermatitis, a defining histopathologic feature in lupus and dermatomyositis (DM). Immune complexes in cutaneous lupus lesions can induce type I IFN production in pDCs (19). Langerhans cells, a specialized subset of dendritic cells residing in skin, also produce IFNs and release increased amounts of IFN-induced chemokines upon stimulation with TLR3 agonist polyinosinic–polycytidylic acid [poly(I:C)] as compared with monocyte-derived dendritic cells (20). Inflammatory monocytes have also been shown to be critical IFN producers in skin upon UVB stimulation (21). Interestingly, IFNAR knockout mice demonstrate increased skin inflammation, suggesting a protective role for type I IFNs after UVB irradiation in wild-type mice (21).
Triggers and regulation of IFN production in skin
Triggers for IFN production in skin can include UV irradiation, infection, injury and cell death, all of which generate damage-associated molecular patterns (DAMPs) or pathogen-associated molecular patterns (PAMPs). Upon sensing of DAMPs or PAMPs by pattern recognition receptors (PRRs), IFN production is induced. The IFN response can differ depending on the underlying trigger, responding receptor and cell type, immune response, and various host-modifying factors. Keratinocytes express a wide range of PRRs, including TLRs and cytoplasmic nucleic acid sensors, all with a unique ligand (PAMP or DAMP) preference (Table I). Altered TLR and increased cytosolic nucleic acid sensor expression is noted in autoimmune skin diseases (22, 23), suggesting a general disease mechanism by which an environment with chronically elevated IFNs may modify IFN response mechanisms. UV radiation exposure to the skin of healthy volunteers and mice has been shown to stimulate a striking cutaneous type I IFN response (21, 24). Which pathways sense and activate IFNs after UV exposure and how this differs in autoimmune disease are currently being investigated.
Multiple factors serve to modulate the cellular response to type I IFNs, including IFNAR downregulation, negative regulator and microRNA upregulation, differential STAT activation, cooperation of STATs with IFN regulatory factors, posttranslational modification, and chromatin remodeling (2). As an example of potential host-modifying factors, commensal microbial flora can serve as a rheostat of IFN responsiveness to viral infections in mice (25). Antibiotic-treated mice have been shown to demonstrate decreased IFN responsiveness after mucosal or systemic viral infection, and expression of IFN and ISGs is reduced in macrophages from antibiotic-treated mice (25). In human keratinocytes, treatment with IFNs in vitro leads to decreased barrier gene expression and increased Staphylococcus aureus adherence, which may induce further IFN production (26).
IFN effects and involvement in pathogenesis of autoimmune skin disease
Cutaneous lupus erythematosus.
Cutaneous disease in systemic lupus erythematosus (SLE) can be an isolated feature or associated with underlying systemic manifestations. Skin inflammation is present in the majority of patients and is often the first harbinger of disease onset or a disease flare, offering a crucial opportunity to potentially intervene even prior to onset of systemic inflammation. Multiple subtypes of cutaneous lupus exist, including acute cutaneous lupus erythematosus (CLE), subacute CLE, chronic CLE (including discoid lupus), and intermittent CLE (tumid lupus) (27). Cutaneous lupus lesions demonstrate a hallmark interface dermatitis, or inflammatory infiltrate bordering the dermoepidermal junction, characterized by apoptotic keratinocytes, vacuolar changes, CD8+ lymphocytes, and pDCs (28, 29). Even in SLE patients with no clinically apparent skin lesions, molecular signatures in nonlesional skin can still indicate an aberrant immune response. Both lesional and nonlesional skin from adults with lupus exhibit chronic upregulation of type I IFNs (15, 30), and comparison of isolated CLE versus systemic lupus associated CLE demonstrate similar gene expression profiles (31).
Although incompletely understood, the pathogenesis of CLE lesions is thought to be driven by IFNs (Fig. 1A). CLE patients exhibit an elevated IFN signature in peripheral blood that correlates with clinical cutaneous disease activity as assessed by the Cutaneous Lupus Erythematosus Disease Area and Severity Index (CLASI) (32, 33). IFN-ĸ production is increased at baseline in nonlesional SLE keratinocytes, leading to increased type I IFN responsiveness and UV light sensitivity (15, 34). Lupus keratinocytes also demonstrate a hypersensitive response to IFN stimulation, with a larger magnitude of change in ISG expression upon IFN treatment as compared with control keratinocytes (35). ISG expression, including myxovirus resistance gene A (MxA), is upregulated at baseline in both the epidermis and dermis of lesional CLE skin (36). IFN-inducible CXCL9, CXCL10, and CXCL11 are three of the five most upregulated chemokines in lesional CLE skin, and their receptor (CXCR3) is among the top three most differentially regulated chemokine receptors (37, 38). MxA, CXCL9, and CXCL10 cutaneous expression patterns also differ by CLE subtype, suggesting that IFNs may have a role in directing differing clinical phenotypes (39).
Key genetic risk variants involved in IFN signaling pathways have also been described in SLE, including IRF5 and STAT4 (40–42), but how each of these relates to skin disease has not been delineated. SLE patients with these genetic risk variants have also been noted to have differences in disease phenotype, which may in part be explained through altered IFN signaling. As an example, SLE patients with high risk IRF5 genotypes were demonstrated to have elevated serum IFN-α activity, with the highest levels observed in patients with anti-dsDNA or anti–RNA-binding protein autoantibodies (43). SLE patients with STAT4 risk alleles are diagnosed at a younger age and are more likely to have nephritis and anti-dsDNA autoantibodies (44, 45). Genetic risks for CLE have also been linked to IFN signaling, as polymorphisms in IFNK (46) are associated with skin disease in African American– and European-ancestry females with SLE and mutations in TREX1, a DNA exonuclease that when inhibited leads to accumulation of nucleic acids and increased IFN production, and result in familial chilblain lupus (47, 48).
It is well known that UV irradiation is a trigger for cutaneous inflammation and disease flare in SLE patients. UV irradiation is known to amplify the IFN response to nucleic acids in keratinocytes, and mice lacking TREX1 develop UV-induced skin lesions (49). ISGs are increased in lupus-prone mice and human patients versus healthy controls after UV irradiation (50, 51), and this coincides with enhanced CD123+ dendritic cell and CD68+ macrophage recruitment in SLE skin after UV irradiation (50). In C57BL/6J mice, UV irradiation induces not only a type I IFN response in skin but also a type I IFN response in peripheral blood and kidney tissue, suggesting a role for UV irradiation and cutaneous IFNs in the initiation of systemic inflammation (24). Interestingly, this type I IFN response is more pronounced in female versus male mice, lending insight into a potential mechanism by which females may be more susceptible to select autoimmune diseases such as SLE (24). In addition, IFNs repress UVB-mediated regulatory T cell induction in lupus-prone mice, which contributes to T cell activation (51). Importantly, persistence of IFN responses in CLE patients after UV exposure correlated with endothelial cell activation, likely contributing to leukocyte recruitment and development of clinical lesions (52). In pDCs, supernatant from UV-treated, apoptotic monocytes induces type I IFN production in combination with SLE total IgG (pooled from plasma of two patients), and both RNase and DNase treatment decrease type I IFN induction (53), suggesting that immune complexes predispose to inflammation following UVB.
Dermatomyositis.
DM is an idiopathic inflammatory myopathy characterized by pathognomonic rash, muscle weakness, and variable involvement of other organ systems, including the lungs, gastrointestinal tract, and heart. In children, skin inflammation is the most common presenting symptom and most classically manifests as scaly, erythematous, raised lesions over the knuckles or reddish-purple discoloration of the upper eyelids with associated edema (54).
Similar to CLE, skin inflammation can be an important indicator of ongoing disease activity, photosensitivity is common, and lesions exhibit an interface dermatitis (55, 56). However, the pathophysiology of DM skin lesions is not as well understood. Type I IFN signaling is upregulated in DM and juvenile DM (JDM) skin (57) as well as in muscle (58) and peripheral blood (59). The type I IFN signature in peripheral blood in DM and JDM has also been reported to correlate with disease activity (60, 61). Immunostaining of DM lesional skin demonstrates increased MxA staining in the epidermis, endothelial cells, and inflammatory cell infiltrate (62, 63). CXCL10 expression is also higher in DM lesional skin, predominantly in the upper dermis in the presence of lymphocytic infiltrate and in the epidermis near areas of interface dermatitis (62). In DM skin disease, similar to CLE, type I IFNs have been purported to lead to recruitment of CXCR3+ lymphocytes, with increased MxA staining correlating with higher numbers of CXCR3+ lymphocytes (62, 64). In anti–melanoma differentiation-associated 5 gene (MDA5) autoantibody-positive DM patients, MxA immunostaining in skin was distributed in blood vessels in the dermis, suggesting a role for IFNs in the vasculopathy that characterizes DM (65). Even nonlesional JDM skin has been described as altered, with increased numbers of pDCs and mast cells (66).
A recent analysis of two DM skin microarray datasets revealed both a type I and type II IFN signature (67). Similarly, DM muscle has been reported to have both a type I and II IFN signature, although the type I IFN signature may be somewhat more specific to DM versus other idiopathic inflammatory myopathies (68). IFN-β expression in the skin has been shown to correlate with ISGs (57), but whether IFN-κ also plays a role in DM skin remains to be determined.
Scleroderma.
Scleroderma is an autoimmune disorder with features of fibrosis, vasculopathy, and inflammation, contributing to pathogenesis at varying stages of disease (69–71). In systemic sclerosis (SSc), the extent of skin involvement associates with prognosis, with lower survival in patients with a higher baseline skin score and improved survival in those patients with improvement in skin thickening (72).
IFN treatment has been known to trigger both SSc and localized scleroderma (LSc), leading to speculation on the role of IFNs in scleroderma pathogenesis. LSc has been described at IFN-β injection sites (73). SSc has also been reported in multiple sclerosis patients after IFN-α and IFN-β treatment (74, 75). Immunostaining for MxA in lesional LSc biopsies shows expression that is most prominent in the deep dermis and subcutis near inflammatory infiltrates (76). CXCL10 staining is apparent in the dermal perivascular lymphoplasmacytic infiltrate (77). In pediatric LSc, bulk RNA sequencing of lesional skin confirms upregulation of IFN-γ and ISGs, including CXCL9, CXCL10, and CXCL11, and LSc patients with more active skin lesions had higher IFN scores (78). A SSc skin microarray study also revealed ISGs as the top upstream transcriptional regulators, with IFN-α and IFN-γ as the top upstream-activated cytokines (79). In this same study, ∼75% of patients had a fibroinflammatory signature, which included gene expression scores of ISGs, that was found to correlate with the modified Rodnan skin score (79). Cutaneous expression of ISGs IFI44 and SIGLEC-1 has also been demonstrated to correlate with the modified Rodnan skin score (80). Interestingly, IFN-κ was shown to be downregulated in SSc keratinocytes, suggesting that there may be different sources of type I IFNs based on cell type and individual autoimmune diseases (81).
Similar to gene expression studies in skin, an IFN signature in peripheral blood has been noted in both localized and systemic scleroderma patients (77, 82). As compared with the peripheral blood IFN signature in SLE, SSc patients were found to have upregulation of endothelial adhesion molecules, suggestive of the underlying vasculopathy that is central to the disease pathogenesis in SSc (82).
It has been suggested that IFN upregulation might play a role in the earlier stages of scleroderma pathogenesis. A study focused on patients with early SSc described that a type I IFN signature is still present in peripheral blood despite the absence of clinical evidence of fibrosis (83). In fact, in early and nonfibrotic compared with fibrotic SSc patients, the IFN score was higher (83). pDCs have been shown to produce IFN-α upon treatment with sera from SSc patients combined with necrotic material (84) in an FcγRII- and RNA-dependent manner, suggesting a role for immune complexes (85). Indeed, TLR8 overexpression in a murine model of disease exacerbates fibrosis (86), and expression of ISGs also increases in skin and fibroblasts from SSc patients upon TLR3 stimulation (87). In SSc patients treated via hematopoietic stem cell transplantation, there is a decrease in type I IFN expression in skin that correlates with decreased fibrosis and capillary regeneration (88).
Sjögren syndrome.
Sjögren syndrome (SS) is characterized by inflammation of the lacrimal and salivary glands, resulting in exocrine dysfunction, with clinical features of keratoconjunctivitis sicca/xeropthalmia and xerostomia. SS can be both a primary disease or secondary to/associated with another underlying rheumatic disease and is associated with hypergammaglobulinemia and production of the classic autoantibodies SSA/Ro and SSB/La. Cutaneous manifestations in SS occur in up to 50% of patients and can include xerosis, angular cheilitis, eyelid dermatitis, pruritis, cutaneous vasculitis, and skin lesions with histologic similarity to CLE (89, 90). Gene expression studies from both peripheral blood and salivary gland tissue highlight an IFN signature (91, 92), with a predominant type I IFN signature in peripheral blood and type II IFN signature in salivary gland tissue (92). Intriguingly, the type I IFN signature correlates with apoptotic gene expression (92), but whether this contributes to skin disease remains unknown. Monocytes from patients with primary SS also have a type I IFN signature in 55% of patients as compared with healthy controls (93). The importance of IFNs in SS is also reinforced by evidence in murine models, with SS mice that have a nonfunctional IFNR failing to develop clinical disease (94).
Psoriasis.
Type I IFN activation has also been described in psoriasis and psoriatic keratinocytes. Genetic polymorphisms that lead to activation of cytosolic signaling pathways and IFN production are risk factors for psoriasis (95); indeed, DDX58 (RIG-I) activation is required for IL-23 activation and psoriasis in murine models (96). Type I IFNs and ISGs are significantly elevated in psoriatic plaques (97–100). A phase I trial of MEDI-545, an anti–IFN-α mAb, was unable to show clinical benefit in patients with chronic psoriatic plaques, which may support the hypothesis that IFNs are involved in initiation of psoriasis, but not in chronic plaque formation (101). Further work to understand how IFNs contribute to psoriatic development is required.
IFN effects and involvement in pathogenesis of autoinflammatory skin disease
Interferonopathies.
The interferonopathies are autoinflammatory disorders characterized by overproduction of IFN due to mutations in genes involved in regulation of nucleic acid sensing. Through the study of interferonopathies, we have gained insight into the pathogenic role of IFNs and underlying disease mechanisms driven by IFNs. A spectrum of cutaneous manifestations are seen in the clinical presentation of interferonopathies, especially vasculopathy (chilblain-like rash, microangiopathic vasculopathy, and gangrene/ulcers/infarcts in acral areas) and skin eruptions of nodular erythema and violaceous plaques in cold-sensitive acral areas (102). Further, undifferentiated autoinflammatory disease patients with elevated IFN response gene scores more commonly had neutrophilic panniculitis (103). Further study has suggested that some disorders may favor NF-κB–driven pathology over that mediated by IFNs (103) but that IFN signature elevation is associated with erythematous, macular skin lesions, and Gottron’s papules (skin lesions common in patients with DM). Understanding the balance between IFN-mediated and other inflammatory activation is an important goal for future research.
Aicardi–Goutières syndrome.
Aicardi–Goutières syndrome (AGS) patients were first described with progressive encephalopathy, basal ganglia calcifications, white matter hypodensities, and persistent cerebrospinal fluid lymphocytosis (104). It was later noted that the most pathognomonic extra neurologic symptom of AGS was the cutaneous finding of chilblain-like lesions on the digits and that these patients also had elevated IFN-α in cerebrospinal fluid and serum (105). Chilblain-like lesions are reported in approximately half of AGS patients, most often on the fingers and toes, but also other acral surfaces, including the ears (106).
Mutations in genes encoding the cellular nucleases TREX1 (107), RNASEH2 complex (108), and SAMHD1 (109) among others have been discovered in AGS patients. Although these mutations lead to increased IFN generation, how these mutations directly lead to skin manifestations is not well understood. TREX1 encodes a 3′–5′ exonuclease that degrades ssDNA (110, 111), dsDNA (112), and ssRNA (113). Accumulation of nucleic acids causes a rise in IFN production in a cyclic GMP–AMP synthase (cGAS)– and stimulator of IFN genes (STING)–dependent manner, and deletion of TREX1 in keratinocytes raises ISG production in keratinocytes (114) (Fig. 1B). However, mice with a dysfunctional TREX1 do not get spontaneous skin lesions (112). This suggests that triggers are needed for phenotype. Indeed, mice with dysfunctional TREX1 exhibit increased ear swelling and inflammation when injected with DNA, independent of its oxidation status (wild-type mice develop lesions only from UV-oxidized DNA, which is resistant to TREX1 degradation) (115). Other mutations associated with TREX1 may also impact UVB sensitivity. Mutations in RNASEH2 can lead to defective repair of damaged RNA, which increases the propensity for UVB-mediated damage and type I IFN production in response (116). Case reports have linked AGS with photosensitivity (117), but how individual mutations contribute remains to be determined. In C57BL/6J mice exposed to UV radiation, both the type I IFN response in skin and peripheral blood is primarily dependent on the cGAS–STING pathway in the early response phase at 6 h postirradiation, lending insight into a potential role for cGAS–STING in the early type I IFN response and subsequent innate inflammatory cell recruitment (24).
Chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature syndrome.
Chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature (CANDLE) syndrome is categorized as a proteasome-associated autoinflammatory syndrome and is manifested by recurrent fevers, annular, purpuric rash, lipodystrophy, and multisystem inflammation. Skin biopsies from CANDLE syndrome patients demonstrate mononuclear cell and neutrophilic infiltrate with dermal collagen degeneration (118). Mutations in the PSMB8 gene were initially found in 8/9 patients from a CANDLE syndrome cohort, accompanied by elevated serum levels of CXCL10 and IFN signaling as a top dysregulated pathway on whole blood gene expression analysis (pathway gene list including both type I and type II IFN–induced genes) (118). Additional mutations in genes involved in proteasome activity have since been identified that result in a CANDLE syndrome phenotype, including PSMB4, PSMA3, PSMB9, and POMP, which encodes a proteasome maturation protein (119). In patients with proteasome alterations other than in PSMB8, skin biopsies demonstrated increased ubiquitin-positive keratinocytes and ubiquitin-rich inclusions in keratinocytes. CANDLE syndrome patient keratinocytes showed impairment in proteasome assembly, and small interfering RNA knockdown of patient proteasome mutations resulted in type I IFN induction (119). Indeed, in CANDLE syndrome, IFNs may participate in a feed-forward loop in which normal triggers of type I IFN production, such as UV light or infections, result in cellular stress and oxidized proteins that cannot be degraded, which results in further type I IFN production, upregulation of the immunoproteasome, and subsequent inflammation (Fig. 1B).
STING-associated vasculopathy with onset in infancy.
Another group of patients exhibiting lupus-like malar rash and vasculitic skin lesions in conjunction with interstitial lung disease have been described to harbor TMEM173 mutations, leading to gain-of-function in STING and subsequent IFN overproduction (120, 121) (Fig. 1B). At baseline, STING-associated vasculopathy with onset in infancy (SAVI) patients have maximal upregulation of type I IFN and ISGs with constitutive STAT1 phosphorylation (121). Lesional skin from SAVI patients is characterized by vascular inflammation of capillaries and microthrombosis, and dermal fibroblasts from SAVI patients are hypersensitive to treatment with even low-dose cyclic GMP–AMP, resulting in increased IRF3 phosphorylation and type I IFN transcription (121).
Murine models of SAVI-associated mutations also develop profoundly elevated ISG signatures. However, systemic disease is independent of the type I IFNR, suggesting other inflammatory pathways or other types of IFNs contribute to disease, at least in mice (122). Mice harboring SAVI-associated mutations have not been reported to develop skin disease, so how the type I IFN pathways participate in SAVI-associated skin manifestations is not yet known.
Insight into disease mechanisms by targeting the IFN pathway
Anifrolumab.
Anifrolumab is an mAb that binds to subunit 1 of the type I IFNR (IFNAR1), thereby blocking type I IFN activity. Trials of anifrolumab for treatment of SLE have shown promise for improvement in CLE disease activity. In a phase IIb, randomized, double-blind, placebo-controlled study of anifrolumab in adults with moderate-to-severe SLE, there was greater efficacy of anifrolumab in patients with a higher IFN signature, including improvement in skin disease activity, as assessed by the Systemic Lupus Erythematosus Disease Activity Index 2000, CLASI, and British Isles Lupus Assessment Group index (123, 124). Improvement in rash was only significantly improved as assessed by the British Isles Lupus Assessment Group in the low IFN signature subgroup (124). In the Treatment of Uncontrolled Lupus via the IFN Pathway (TULIP) trial II, there was ≥50% decrease in CLASI scores in half of the anifrolumab group compared with only 25% of the placebo group (p = 0.04) (125). In systemic scleroderma patients, anifrolumab treatment has also been shown to decrease type I ISG expression in patient skin biopsies collected 28 d after dosing with anifrolumab (126).
JAK inhibitors.
JAK inhibitors block one or multiple JAKs (JAK1, JAK2, JAK3, and TYK2), which are tyrosine kinases that bind to a wide variety of cytokine receptors (including all three types of IFNs) and thereby affect the immune response (127, 128). CLE lesions have been shown to exhibit high expression of phospho-JAK2, similar to CXCL10 and MxA, and treatment of keratinocytes and a three-dimensional epidermis model with ruxolitinib after poly(I:C) stimulation decreases type I ISG expression (129). Treatment of murine lupus with tofacitinib resulted in improvement of both systemic and cutaneous disease manifestations (130). In DM, skin disease has shown improvement after treatment with ruxolitinib, further supporting a role for IFNs in DM pathogenesis (131). Treatment of 18 interferonopathy patients with baricitinib led to a decrease in IFN scores and clinical symptoms, with improvement in cutaneous disease also reported, although not specifically scored (132). Similarly, treatment of cutaneous lesions in familial chilblain lupus with baricitinib leads to improvement in skin disease (133). Liu et al. (121) also demonstrated that treatment of SAVI patient T and B cells with JAK inhibitors blocks constitutive phosphorylation of STAT1.
Anti-BDCA2 Ab (BIIB059).
BIIB059 is a humanized mAb that binds blood dendritic cell Ag 2 (BDCA2), a C-type lectin and pDC-specific receptor. BIIB059 is believed to inhibit TLR-induced type I IFN and other inflammatory mediator production. In CLE, BIIB059 has been shown to reduce skin inflammation (134). In a randomized, double-blind, placebo-controlled trial of BIIB059 in SLE patients with active skin disease, BIIB059 decreased expression of MxA and IFITM3 and CD45+ cellular infiltrate in skin biopsies 4 wk after treatment and additionally improved CLASI scores (134). BIIB059 has also been described to reduce IFN-α production from pDCs of CLE patients after stimulation with TLR agonists, providing an additive therapeutic benefit to hydroxychloroquine (135).
Conclusions
Overproduction of type I IFNs is a unifying theme among many autoimmune and autoinflammatory patients with skin manifestations. In SLE/CLE, this contributes to inflammatory cell activation and photosensitivity, a mechanism that likely extends to other diseases, possibly the autoinflammatory diseases as well. Further research is needed to understand the ways in which IFNs drive disease and to identify which patients will benefit most from targeting of IFNs.
Disclosures
J.M.K. has served on advisory boards for AstraZeneca, Eli Lilly, Bristol Myers Squibb, Avion Pharma, Provention Bio, and Boehringer Ingleheim. The other authors have no financial conflicts of interest.
Footnotes
This work was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health Award R01-AR071384 (to J.M.K.), the A. Alfred Taubman Medical Research Institute (to J.M.K.), the Parfet Emerging Scholar Award (to J.M.K.), the Rheumatology Research Foundation (to J.M.K.), and the Doris Duke Charitable Foundation through a Physician Scientist Development award (to J.M.K.). J.L.T. was supported by a Eunice Kennedy Shriver National Institute of Child Health and Human Development K12 Child Health Research Center Career Development Award (K12 HD028820-28), a Michigan Institute for Clinical and Health Research Pathway to First Grant Award, and a Cure JM Foundation Research Grant. J.M.K. has received grant support from Celgene/Bristol Myers Squibb and Q32 Bio.
Abbreviations used in this article:
- AGS
- Aicardi–Goutières syndrome
- CANDLE
- chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature
- cGAS
- cyclic GMP–AMP synthase
- CLASI
- Cutaneous Lupus Erythematosus Disease Area and Severity Index
- CLE
- cutaneous lupus erythematosus
- DAMP
- damage-associated molecular pattern
- DM
- dermatomyositis
- IFNAR
- IFN-α/βR
- IRF
- IFN regulatory factor
- ISG
- IFN-stimulated gene
- JDM
- juvenile DM
- LSc
- localized scleroderma
- MxA
- myxovirus resistance gene A
- PAMP
- pathogen-associated molecular pattern
- pDC
- plasmacytoid dendritic cell
- poly(I:C)
- polyinosinic–polycytidylic acid
- PRR
- pattern recognition receptor
- SAVI
- STING-associated vasculopathy with onset in infancy
- SLE
- systemic lupus erythematosus
- SS
- Sjögren syndrome
- SSc
- systemic sclerosis
- STING
- stimulator of IFN genes
- TYK2
- tyrosine kinase 2.
- Received May 21, 2020.
- Accepted August 18, 2020.
- Copyright © 2020 by The American Association of Immunologists, Inc.
References
In this issue
