“Abnormalities of the Type I Interferon signaling pathway in Lupus autoimmunity”.

Abstract

Type I interferons (IFNs), mostly IFNα and IFNβ, and the type I IFN Signature are important in the pathogenesis of Systemic Lupus Erythematosus (SLE), an autoimmune chronic condition linked to inflammation. Both IFNα and IFNβ trigger a signaling cascade that, through the activation of JAK1, TYK2, STAT1 and STAT2, initiates gene transcription of IFN stimulated genes (ISGs). Noteworthy, other STAT family members and IFN Responsive Factors (IRFs) can also contribute to the activation of the IFN response. Aberrant type I IFN signaling, therefore, can exacerbate SLE by deregulated homeostasis leading to unnecessary persistence of the biological effects of type I IFNs.

The etiopathogenesis of SLE is partially known and considered multifactorial. Family-based and genome wide association studies (GWAS) have identified genetic and transcriptional abnormalities in key molecules directly involved in the type I IFN signaling pathway, namely TYK2, STAT1 and STAT4, and IRF5. Gain-of-function mutations that heighten IFNα/β production, which in turn maintain type I IFN signaling, are found in other pathologies like the interferonopathies. However, the distinctive characteristics have yet to be determined. Signaling molecules activated in response to type I IFNs are upregulated in immune cell subsets and affected tissues of SLE patients.

Moreover, Type I IFNs induce chromatin remodeling leading to a state permissive to transcription and SLE patients have increased global and gene-specific epigenetic modifications such as hypomethylation of DNA and histone acetylation. Epigenome wide association studies (EWAS) highlight important differences between SLE patients and healthy controls in Interferon Stimulated Genes (ISGs). The combination of environmental and genetic factors may stimulate type I IFN signaling transiently and produce long-lasting detrimental effects through epigenetic alterations.

Substantial evidence for the pathogenic role of type I IFNs in SLE advocates the clinical use of neutralizing anti-type I IFN receptor antibodies as a therapeutic strategy, with clinical studies already showing promising results. Current and future clinical trials will determine whether targeting molecules of the type I IFN signaling pathway, like non-selective JAK inhibitors or specific TYK2 inhibitors, may benefit people living with lupus.

Introduction

Systemic Lupus Erythematosus (SLE) is a prototype autoimmune disease characterized by the production of autoantibodies against nuclear and cytoplasmic antigens, including nucleic acids and ribonucleoproteins [1–3]. The role of type I interferons (IFNs) [4] in SLE has been investigated for over four decades [5,6]. Type I IFNs are a family of cytokines first discovered as soluble factors that interfered with viral replication [7] and later found to be pivotal in the activation of innate [8] and adaptive immune responses [9–11]. In the human genome, type I IFNs are comprised of 13 IFNα subtypes, along with IFNβ, IFNε, IFNκ,and IFNω in single form. All of them bind to the same interferon alpha and beta receptor (IFNAR), and the IFNα members and IFNβ are the ones most extensively studied [12–14]. Forty years ago, early studies informed that SLE patients had high serum levels of IFNα or type I IFN activity [15–17]. Twenty years later, subsequent studies revealed that SLE patients with no clinical evidence of infections highly expressed a large group of genes identified as upregulated by type I IFNs, referred thereafter as Interferon Stimulated Genes or ISGs [18–20]. These genes, measured at first as mRNA transcripts in peripheral blood cells, were referred to as the “IFN Signature”, which is detected in the earliest and most acute phases of the disease [21,22]. This distinctive IFN signature is frequently seen in pediatric patients [19] and has been proposed as a marker of specific morbidities in SLE [20,23–25]. Determining protein levels of type I IFNs (IFNα subtypes or IFNβ) in the serum of patients with SLE, other diseases or in healthy controls remains challenging. Most studies report the IFN Signature at the transcriptional level as a measure of type I IFN activity. They hint at a response to IFNα without completely discriminating the specific patterns induced by single members of the type I IFN family; particularly as new data emerges that implicates multiple type I Interferons as inducers of the IFN Signature in lupus, including IFNβ [26]. Recent studies document increased levels of serum IFNβ in SLE patients [27]. Furthermore, new technologies have now confirmed the RNA data of the IFN Signature and the original IFNα measurements [15–17] at the protein level, both measuring serum protein levels and activity of serum IFNα [24,28]. Adding to that, use of epigenetic CpG methylation data can be useful to adequately assess type I IFN activity. In doing so, the use of a DNA methylation IFN score has been found to be a potential surrogate marker of the RNA-based IFN-signature [29].

The IFN Signature and the associated higher response to type I IFNs [20,30,31] are seemingly linked to many genetic and environmental factors, which may be different in distinct subsets of SLE patients [32]. Genetic factors are widely accepted to play an active role and considered necessary but not sufficient for disease development [33,34]. Environmental triggers, like viral and bacterial infections, as well as alterations in the microbiome, are actively being studied to better assemble a complex and more accurate depiction of lupus pathogenesis [3,35–40]. Endogenous inducers of type I IFNs are released during cell death and tissue damage, often fueling a vicious cycle in which immune complexes containing endogenous DNA or ribonucleoproteins and autoantibodies cause tissue damage and propel the release of more autoantigens [41,42]. Moreover, immune complexes formed by endogenous DNA and autoantibodies are known stimulators of IFNα responses, as well as modified nucleic acids per se can induce IFNα [43–46] and IFNβ [46,47]. Nucleic acids can be released extracellularly and recognized by endosomal Toll Like Receptors (TLRs), or they can be overrepresented in the cytoplasm and recognized by cytoplasmic Pattern Recognition Receptors (PRRs) [3,48,49]. Plasmacytoid dendritic cells are the main cellular source of IFNα [50,51]. They are considered pivotal in SLE pathogenesis [43,52–54], whereas other innate immune cells such as monocytes and conventional dendritic cells can also contribute to the type I IFN response producing mostly IFNβ [31,47,55–57]. Similarly, neutrophils can trigger in other cells the innate production of type I IFNs by extruding DNA through the process of Netting, that allows DNA to activate IFNα and/or IFNβ synthesis through TLR9 or STING-dependent pathways [46,58–60]. Neutrophils can also respond to chromatin by secreting IFNβ themselves to amplify the stimulation of the IFN Signature [61].

Elucidation of the signaling pathways downstream of IFNAR have been the focus of intense research in cell biology, pathology, virology and cancer. Results from these disciplines have provided the scaffold to build on the contribution of the genetic, epigenetic and environmental factors that influence the role of type I IFNs in lupus pathogenesis. It’s noteworthy that the IFN Signature in SLE is not exclusive as other autoimmune diseases, ranging from Rheumatoid Arthritis to Sjogren Syndrome to autoimmune myositis, have shown similar abnormal responses [62]. In each of these conditions, common and specific causes may play a role. In this review, we will focus our attention on the signaling pathways mediating the response to type I IFNs; how they affect the activation of the IFN Signature, and the corresponding specific abnormalities in SLE (Figure 1).

Figure 1. Abnormalities in the signaling pathway that exacerbate type I IFN responses.

Highlighted are identified components of the signaling pathway downstream of the Interferon-alpha/beta receptor (IFNAR) and Toll-like receptor pathways that, when deregulated, contribute to the pathogenesis of SLE. Polymorphisms associated with SLE are indicated in bold and underlined. Genes that have upregulated/downregulated expressions are indicated with green/red arrows, respectively.

Signaling downstream of Type I IFNs

Type I IFNs is a family of cytokines which includes 13 IFNα subtytpes, IFNβ, IFNε, IFNκ, and IFNω [12,13]. All type I IFNs initiate a signaling cascade upon binding to their cognate receptor complex composed of IFNAR1 and IFNAR2 [63]. Assembly of this ternary complex induces a conformational change in the receptor structure enabling their activation by tyrosine kinases JAK1 and TYK2; each of which is found pre-associated with IFNAR2 and IFNAR1, respectively, and become activated themselves by transphosphorylation. The Janus kinase (JAK) family belongs to a group of ten recognized families of non-receptor tyrosine kinases and consists of four enzymes, JAK1–3 and Tyrosine kinase 2 (TYK2) [64]. The canonical view is that tyrosine phosphorylation of the IFNAR subunits by JAK1 and TYK2 creates docking sites for the recruitment of pre-existing STAT1 and STAT2. JAKs activate STAT1 and STAT2, which in turn heterodimerize and associate with the transcription factor IRF9 to form the heterotrimeric complex ISGF3; the latter is translocated to the nucleus and initiates gene transcription by recognizing IFN-stimulated response elements (ISRE) found in the promoters of ISGs. Type I IFNs also induce the formation of STAT1 homodimers that bind the gamma-activated sites (GAS) and stimulate the transcription of additional ISGs. In recent years, evidence has been accumulating of an ISGF3-like transcriptional complex consisting of STAT2 homodimers joined with IRF9 that directs a prolonged ISG transcriptional program and overlaps with a subset of genes induced by ISGF3 [65], thus highlighting the pivotal role of STAT2 in type I IFN signaling. It is important to mention that while IRF9 binds the core ISRE sequence, STAT2 as a homodimer or as a STAT2/STAT1 heterodimer is unable to directly bind DNA, but it contributes its transactivation domain for gene transcription [66]. Recent attention has been given to constitutive unphosphorylated pools of STAT2/IRF9 complexes that shuttle between the cytoplasm and the nucleus in the absence of type I IFN stimulation [67] for their role in maintaining basal ISG expression. STAT2 chromatin immunoprecipitation (ChIP)-chip assays previously revealed unphosphorylated STAT2 occupancy of several ISG promoters [68]. Subsequent studies have reported that the initial and robust ISGF3 response to type I IFNs is followed by an increase in the expression of ISGF3 components that assemble without tyrosine phosphorylation to amplify type I IFN responses [69]. We previously reported that the constitutive expression of ISGs, including Cxcl10, Isg15, and Irf7 genes, was markedly reduced by 70% in murine conventional dendritic cells (cDCs) from Stat2^−/− mice when compared with wild type cDCs [70], thus supporting a role for unphosphorylated STAT2 in the maintenance of basal ISG production or a constitutive low production of type I IFNs. The concept of tonic ISGF3 activation downstream of type I IFN signaling driven by the production of low quantities of type I IFNs was proposed several years ago, to maintain adequate levels of ISGF3 components and prepare cells for quick and robust ISGF3-mediated biological responses [71]. However, this view was recently revisited, in which basal expression of many ISGs is stimulated by a preformed STAT2/IRF9 complex in the absence of IFNAR engagement [72]. Most notable was the finding that resting cells responding to type I IFNs rapidly switched signaling complexes from STAT2/IRF9 to ISGF3 assembled in the nucleus. Hence, alterations in type I IFN signaling that disturb homeostasis and unnecessarily prolong the biological effects of IFNs could elicit uncontrolled destructive effects.

Altered Signaling of Type I IFN responses in SLE

The response to type I IFNs, either or both IFNα and IFNβ, and specific steps in the signaling pathway mediating their functions have been shown to be dysregulated in SLE patients as detected in whole blood and in individual immune cell populations [73,74]. Moreover, specific tissues targeted by lupus autoimmunity show an abnormal type I IFN response [26,75–77]. JAK1 and STAT2 were shown to be constitutively phosphorylated in peripheral blood mononuclear cells (PBMCs) of SLE patients, and stimulation with IFN-β induced 3–4 folds higher STAT2 phosphorylation, which remained sustained for a prolonged time when compared to control cells from healthy subjects. This hyperactivation induced higher MxA mRNA levels (an ISG) in SLE patients than in healthy controls, possibly with the contribution of a deficient up-regulation of regulatory SOCS1 protein [78]. A recent study has found that when assessing expression and phosphorylation of STAT1 after IFNα and IFN γ stimulation, SLE patients showed significantly elevated expression of STAT1 in B cells than healthy individuals. This abnormality was associated with increasing disease activity and Siglec-1 expression on monocytes; an indicator that the increased expression of STAT1 in B cells is associated with the IFN Signature [79]. Furthermore, a study of single-cell gene expression in SLE monocytes revealed increased expression of most of the signaling molecules directly involved in IFNAR signaling, including JAK1, STAT1 and STAT2, in both classical and nonclassical monocytes from patients with high disease activity [80].

Several studies have shown how responses to damage-associated molecular patterns (DAMPs) such as mitochondrial DNA and necrotic/apoptotic cells can induce ISGs [81–85], as well as responses to pathogen-associated molecular patterns (PAMPs) triggered by bacterial and viral infections, through the activation of TLR and other PRRs [35–37,86,87]. The higher levels of type I IFNs and STATs found in SLE patients may be due to chronic exposure to DAMPs and PAMPs because of subclinical infections [3,88], or may have genetic causes in some patients, aggravated by environmental triggers that can leave long-lasting effects through epigenetic alterations.

Genetic variants altering type I IFN Signaling in Lupus

The discovery of the IFN Signature in SLE has ignited the search of the causal factors prompting the exacerbation of type I IFN responses [6,89]. Family-based and genome wide association studies (GWAS), together with gene profiling of large cohorts have identified genetic, epigenetic and transcriptional abnormalities in molecules directly involved in the signaling pathways mediating the response to type I IFNs, as well as those involved in its production [90,91]. As described in the previous paragraphs, the activation of the response to type I IFNs and the induction of the IFN Signature is complex and can be influenced by a network of interactions with other critical signaling pathways, from MAPKs to NFκB, and by the positive and negative feedbacks elicited by the ISGs themselves [4].

A significant body of literature addresses the alterations of these molecular networks in SLE. Genetic studies have afforded significant advances in the knowledge of SLE risk, by enabling the discovery of potential genetic factors that are involved in the regulation of the type I IFN signaling pathway [92–95]. The use of GWAS has identified more than 100 SLE susceptibility loci along with documentation of genetic variants associated with the pathogenesis of SLE [33,96]. Many of these genetic variants implicate aberrant regulation of both innate and adaptive immune responses that, in particular, involve transcription factors [97,98]. We focus here on molecules directly contributing to signaling triggered by IFNα and IFNβ. Little information is available regarding specific alterations involving signaling induced by the other type I IFNs, namely IFNε, IFNκ,and IFNω [99].

IFNARs.

High serum levels of IFNα and IFNβ have been reported in SLE patients throughout the history of the investigations of IFNs in SLE [15–17,27]. To date, no polymorphisms in type I IFNs loci have been associated with higher risk of lupus in the human population; the same applies to subunits of the type I IFN receptor [97,100–102]. In murine models of lupus, the genes for IFNα and IFNβ are located in the Sle2 susceptibility locus that is derived from the lupus-prone New Zealand strain of mice and was introgressed into the triple congenic Sle1,2,3 mice [103,104]. Deletion of Ifnar in lupus prone mice causes delayed and milder lupus autoimmunity in spontaneous and induced models of lupus [105–110]. However, the role of type I IFN genes in the Sle2 allele and in the susceptibility of Sle2 mice to lupus remains ambiguous because Sle2 seems to cause a decrease in IFN responses [104]. Since high serum IFN-α activity is a human heritable trait, as shown by clustering in specific families in both SLE patients and their healthy first-degree relatives [30], the lack of risk-associated polymorphisms in type I IFN genes and IFNARs suggests that genetic variations affecting the type I IFN response must be occurring at other levels of the IFN signaling pathway. Alternatively, variations may be present in non-coding regions of the DNA, and therefore, difficult to identify because these may differ in SLE patients.

JAK1 and TYK2.

TYK2 plays an essential role in triggering the signaling of important cytokines, including type I IFNs. Single nucleotide polymorphisms (SNPs) of TYK2 were identified in families with SLE [111], and these results were extended by GWAS in many cohorts [97,100,101]. Through a meta-analysis, two TYK2 polymorphisms, rs2304256 and rs280519, were found associated with SLE in patients of different ancestry [112]. The specific mechanism of the rs2304256 polymorphism is not entirely clear, but what is known is that it causes an amino acid substitution within a region of TYK2 that is essential for its interaction with IFNAR1 and may cause an aberrant IFN response [112]. GWAS have revealed another SNP (rs34536443) in the TYK2 gene that encodes a rare protective variant known as TYK2 P1104A (TYK2^p). This genetic variant is responsible for reducing type I IFN signaling in healthy individuals and altering autoimmune pathogenesis through decreasing IL-12, IL-23, and type I IFN signaling in T cells [113]. Moreover, of the known long non-coding RNAs (lncRNA), elevated lupus nephritis (LN)-associated lncRNA RP₁₁-2b6.2 has been observed in kidney biopsies of LN patients. Curiously, decreased LncRNA Rp₁₁-2b6.2 serves as a positive regulator, given that when its levels are low, the phosphorylation of JAK1, TYK2 and STAT1 in the type I IFN pathway is inhibited [114].

No specific polymorphisms of JAK1 have been found to be associated with the risk of developing SLE, but its activation and regulation may play a role in supporting the IFN Signature.

STAT1/STAT2.

STAT1 and STAT2 are essential transcription factors downstream of type I IFN signaling. STAT1 was identified to have SLE-associated risk alleles [115]. Since the STAT1 locus is adjacent to the STAT4 locus, and the latter was also found to have variants associated with higher risk of developing SLE [111,116,117], the role of STAT1 polymorphisms in SLE pathogenesis has been difficult to discriminate from the role of STAT4. It is worth noting that researchers who identified the same STAT4 risk allele in Japanese populations that is associated with SLE in Caucasians (see below), did not find evidence for a specific role of STAT1 in the genetic susceptibility to SLE [118].

To date, no polymorphisms of STAT2 influencing the risk of developing SLE have been identified. While STAT2 expression and function are augmented in SLE (see below), an indirect effect on STAT2 caused by genetic variants of other genes remains to be determined. Similarly, STAT1 was found to be upregulated in SLE patients and highly expressed in various immune cell subsets, including dendritic cells in human [119] and mice [31]. A transcriptome profiling study of SLE patients identified the overexpression of the long non-coding RNA (lncRNA) linc00513 associated with two alleles. This elevated linc00513 was a positive regulator of the type I IFN signaling pathway by promoting increased phosphorylation of STAT1 and STAT2 [120]. The increased expression of linc00513 correlated with the measure of ISGs in SLE patients, thus potentially contributing to the IFN Signature [120].

The function of STAT1 or STAT2 may be also affected by polymorphisms in other transcription factors. For example, ETS1 is a transcription factor important in regulating immune cell proliferation and differentiation [121]. Many GWAS have found genetic variants of ETS1 associated with susceptibility to SLE in Asian populations [122,123]. Ets1-deficient mice develop a lupus-like disease characterized by autoantibodies and immune-complex deposition in the kidney [124]. PBMCs from patients carrying lupus risk ETS1 alleles express significantly lower levels of ETS1 mRNA than control subjects, suggesting that ETS1 risk variants contribute to SLE through reduced ETS1 expression. The lupus risk ETS1 variant rs6590330 was shown to enhance phosphorylated STAT1 binding to DNA [125]. This observation lends to the hypothesis that one of the mechanisms underlying the effects of ETS1 variants on SLE is that less ETS1 leads to increased STAT1 binding to other promoters, possibly ISGs. Since STAT1 is shared with type II IFN (IFN-γ) signaling, alterations in STAT1 expression may also affect the functions induced by IFN-γ and its role in lupus pathogenesis, as elegantly shown in lupus prone mice [126,127].

STAT4.

STAT4 can be activated by type I IFNs and participates in amplifying its positive feedback loop [128]. Identified as a risk allele in the first GWAS, STAT4 is associated with specific clinical manifestations including early disease onset, ischemic cerebrovascular disease, severe nephritis and renal insufficiency [111,116,117]. Recent SLE human studies have investigated the effect of the STAT4 risk allele in T cells of carriers. SLE patients show elevated levels of STAT4 protein, leading to both increased phosphorylation of STAT4 in response to IL-12 and IFNα, as well as IL-12 induced IFN-γ production [129]. However, healthy individuals who are carriers of the risk allele display the opposite effect; they show decreased phosphorylation of STAT4 in activated T cells [130]. It was further determined that IFNα is an environmental modifier of the STAT4 risk allele [130].

IRF5 and other IRFs.

The analysis of SNPs in genes affecting the upstream events of the type I IFN pathway has revealed a number of genetic risk polymorphisms in the IRF family of transcription factors that are associated with high type I IFN levels in lupus patients [84,131]. IRF1, IRF5, IRF7 and IRF8 are considered primary inducers of type I IFN production, mediating the signaling pathways of major pattern recognition receptors (PRRs) like TLRs and RIG-I-like receptors (RLRs), and SLE-associated SNPs have been reported in these transcription factors [93,102,132]. Some genetic variations of IRF3 have been found in SLE patients but it remains to be clarified whether these polymorphisms increase the risk or confer protection from developing lupus [133–135]. Although almost one hundred SLE-associated SNPs have been found within the IRF5 locus, mostly in non-coding regions, five genetic variants have been associated with abnormalities in the expression or function of IRF5, which may affect lupus pathology [94,102,111,136]. Studies have revealed increased type I IFN activity in the sera of SLE patients with IRF5 risk polymorphisms, which correlates with elevated IRF5 expression levels and positivity for either anti-RNA binding protein (anti-RBP) or anti-double-stranded DNA (anti-dsDNA) autoantibodies [84,131]. Indeed, many SNPs in the SLE-associated risk variants of IRF5, as well as STAT4, were found to correlate with the levels of anti-dsDNA autoantibodies, therefore suggesting that these genes, and the augmented type I IFN response, may increase the risk of developing SLE by promoting autoantibody production [137]. The same risk alleles for IRF5, when present in healthy donors, are associated with increased activation of innate immune cells like pDCs and neutrophils, detectable autoantibodies and type I IFN pathway enrichment, therefore behaving as pre-symptomatic SLE [136].

Epigenetic Regulation of type I IFN signaling and disease activity in lupus

Many reports indicate the SLE patients have global and gene-specific DNA methylation changes [138]. DNA methylation is an important epigenetic modification that, especially if occurring in promoters, suppresses the expression of relevant genes. Specifically, epigenetic modifications are important in restricting or sustaining the response to type I IFNs for the induction of ISGs [139]. Type I IFNs induce chromatin remodeling leading to a state permissive to transcription. Evidence indicates that abnormal DNA hypomethylation in T cells is an important epigenetic hallmark in SLE [140]. Significant hypomethylation of ISGs in naïve T cells from lupus patients, and among those ISGs specifically STAT1, suggests an epigenetic increase in accessibility for transcription in these genetic loci [141]. This epigenetic change can help explain the increased levels of STAT1 and other ISGs in SLE cells. In an independent report, the performance of epigenome wide association study (EWAS) of whole blood highlighted that the most evident differences in methylation between SLE patients and healthy controls were in ISGs, which included IRF5 and IRF7, and showed decreased methylation in patients with SLE. [142] In other cohorts, IRF7 gene was also found hypomethylated throughout T-cells, B-cells and CD14+ monocytes at CpG islands, as well as monocyte-specific hypomethylation within the gene body [143]. Furthermore, a functional gene ontology analysis of ISGs for DNA hypomethylation revealed an overrepresentation of genes involved in induction and regulation of apoptotic processes, as well as NFκB activation, suggesting that epigenetic changes in genes other than ISGs, like those within the Fas/FasL-mediated apoptotic pathway, also contribute to SLE [144]. The modification of DNA methylation levels at regulatory regions of ISGs ultimately was associated with various phenotypic manifestations of SLE, such as nephritis and skin rash [142]. Genetic variants influence methylation marks, as well as environmental stressors [145]. It is unknown how methylation marks change over time in SLE patients.

Moreover, type I IFNs induce an increase in tri-methylation of H3K4 (H3K4me3), a histone mark that promotes transcription. Sullivan’s group implemented an innovative approach to study the regulation of ISGs in SLE, which consists in cross-referencing the GWAS results with those from the transcriptome and epigenome. They found that epigenetic modifications, like H3K4me3 histone modifications, are increased at sites highlighted by GWAS as associated with SLE. Surprisingly, the RNA expression of these GWAS-associated genes was decreased in three main immune cell populations, monocytes, T and B cells, in SLE patients. STAT1 was one of the genes found to be under-expressed in SLE patients. The authors recognized that their focus on mild cases of SLE may justify these results that remain a sobering memento of the complexity of studies bridging from genetics to functionality [146]. Other reports have documented increased expression of STAT1/2 in SLE patients. The STAT1 locus is adjacent to the STAT4 locus, and it has been shown in B cell lines that SLE B cells carrying the SLE STAT4 polymorphisms express higher levels of STAT1 protein [147]. The analysis of the transcriptional profile and epigenetic landscape of Low Density Granulocytes (LDG), an abnormal population of neutrophils present in SLE, showed that these cells are more transcriptionally active with a specific up-regulation of genes involved in histone acetylation. ATACseq showed increased chromatin accessibility, with a number of “open peaks” twenty time higher than in normal neutrophils, in parallel with increased ISG expression [148]. Altogether, these findings indicate that the upregulated expression of type I IFN signaling molecules observed in SLE may be due to classic genetic causes in some cases, while other cases can be explained by environment-induced epigenetic modifications.

Defects of IFN-induced signaling pathways in other pathologies.

Several mechanisms are in place to dynamically restrict type I IFN signaling that, if left unrestrained, can be detrimental to the host. These tightly regulated controls include among others internalization and degradation of cell surface IFNAR1 [149], induction of inhibitory proteins SOCS1 [150] and USP18 [151] that, when combined, trigger a negative feedback loop and suppresses JAK activity and further activation of STAT1 and STAT2. Importantly, STAT2 plays dual roles in the activation and inhibition of type I IFN signaling. In the latter, STAT2 functions as an adaptor molecule for the recruitment of USP18 to IFNAR2 that displaces JAK1 from the IFN receptor [152]. Impaired activation of the type I IFN signaling pathway has been implicated in multiple immune related diseases. In humans, several germline STAT1 mutations have been identified that confer either a loss-of-function or gain-of-function phenotype with an outcome of susceptibility to an array of viral infections and predominantly fungal (chronic mucocutaneous candidiasis) and bacterial (mycobacterial) infections [153–155]. Similar phenotypes have also been described in patients with inborn errors of IFNAR1 [156], IFNAR2 [157], JAK1 [158], TYK2 [159], IRF9 [160] and STAT2 [161,162]. Most unexpected was the clinical phenotype observed in patients with IFNAR2 and STAT2 deficiencies who became severely ill after receiving the measles, mumps and rubella vaccine.

On the opposite spectrum, dysregulation of type I IFN signaling is strongly associated with inflammatory conditions. A collection of rare Mendelian autoinflammatory disorders termed type I interferonopathies is characterized by exacerbated type I IFN signaling activity and elevated ISG signature [163]. Type I interferonopathies are monogenic disorders and display diverse clinical phenotypes that can be daunting to diagnose. Examples of these conditions are Aicardi-Goutières syndrome and spondyloenchondrodysplasia with pathological phenotypes that partially overlap with SLE [164,165]. Since 2006, mutations in several genes have been identified in patients as the cause of type I interferonopathies. Loss-of-function and gain-of-function mutations, which heighten type I IFN production and sustain type I IFN signaling, drive the clinical phenotypes by distinct mechanisms. One mechanism involves the accumulation of single or double stranded DNA and RNA pools caused by loss-of-function mutations in genes involved in editing (ADAR1) and degradation (TREX1, RNASEH2A, −2B, −2C, SAMHD1) of nucleic acids [166]. Another mechanism is persistent nucleic acid sensing caused by gain-of-function mutations in cytosolic sensors MDA5 [167], STING [168], and RIG-I [169] that become hypersensitive to cytosolic nucleic ligands. Impaired regulation at the receptor level is a third mechanism. Loss-of-function mutations in USP18, an ISG and key negative regulator of type I IFN signaling, is also linked to type I interferonopathies [170].

The most recent addition to the list of genes linked to type I interferonopathies is STAT2 [171,172]. Patients with homozygous STAT2 mutations display hypersensitivity to type I IFN due to its inability to recruit USP18 to IFNAR2. Nonetheless, mutations in USP18 and STAT2 were fatal and underscore the non-redundant immunopathogenic and inflammatory nature of type I IFNs. Of note, STAT2 has been implicated in other inflammatory conditions that display dysregulated STAT2 activity such as psoriasis [173], and specific STAT2 polymorphisms have been associated with asthma susceptibility [174].

It is interesting that only a subset of patients with interferonopathies have clear overlap with a classic SLE autoimmunity mediated by T and B cells [175]. The evidence suggests that lupus autoimmunity requires much more than an exacerbated type I IFN response. For example, the importance of checkpoint defects that modify the threshold of tolerance of T and B lymphocytes is supported by data from murine models [176]. Alternatively, we can hypothesize that a moderate increase in type I IFN response, like the one existing in lupus, promotes autoimmunity, whereas the stronger response prevailing in interferonopathies may paralyze the adaptive immune response and cause defects in cell development and tissue damage rather than triggering autoimmunity.

Therapeutic effects of selective inhibitors of the type I IFN signaling pathway

SLE is a chronic disease with no cure. The current treatment protocols include steroidal and non-steroidal anti-inflammatory drugs, the antimalarial hydroxychloroquine and immunosuppressive drugs, like inhibitors of DNA replication cyclophosphamide and mycophenolate mofetil acid, or biologics interfering with B cell development [177]. The pathogenic role of type I IFNs in SLE prompted clinical trials testing their inhibition as new therapeutic strategy in SLE. After the disappointing results with blocking IFNα [178,179], in a recent phase 3 trial, administration of anifrolumab, a neutralizing antibody against IFNAR, has proven surprisingly safe and efficacious in SLE patients [180]. Inhibition of mediators of type I IFN signaling may confer the same or potentially enhance these effects. JAK1–3 inhibitors like baricitinib and tofacitinib have been shown to be effective when tested in mouse models of lupus and provided promising results in the first clinical trial in SLE patients [74,181,182]. Selective inhibitors of TYK2 have been recently proposed as an alternative treatment to specifically block the signaling downstream of type I IFNs without interfering with the function of other cytokines [182,183]. Future studies will very likely extend the list of therapeutic strategies able to block type I IFNs in SLE.

Conclusions.

Much has been learned about the involvement of type I IFNs and their signaling pathway in the pathogenesis of SLE. Family-based studies and GWAS have revealed genetic components influencing the IFN Signature. Additionally, EWAS are beginning to expose the rules of engagements of short-lived and possibly long-lasting effects of infections or pollutants on its regulation and on the immune system at large. Many outstanding questions remain, like the role of unphosphorylated STAT1 and STAT2 in supporting the threshold of activation of transcription in SLE, or whether a deficient suppressive function of STAT2 has a role in unleashing the IFN Signature in SLE. Current and future clinical trials will indicate whether non-selective JAK inhibitors rather than specific TYK2 inhibitors may be more beneficial in suppressing type I IFN responses in lupus and whether such inhibition will provide clinical benefits to people living with lupus.

List of abbreviations:

ADAR1

Adenosine deaminases acting on RNA

Anti-RBP

anti-RNA binding protein

Anti-dsDNA

anti-double-stranded DNA

cDC

Conventional Dendritic Cells

ChIP

Chromatin immunoprecipitation

DAMPs

Damage-associated molecular patterns

DNA

deoxyribonucleic acid

EWAS

Epigenome-Wide Association Studies

GAS

gamma-activated sites

GWAS

Genome Wide Association Study

H3K4me3

Histone H3 lysine K4 (H3K4)

IFN

Interferon

IFNAR

Interferon-alpha/beta receptor

IRF

Interferon Responsive Factor

ISGs

Interferon stimulated genes

ISGF3

Interferon-stimulated gene factor 3

IL-12

Interleukin 12

IL-23

Interleukin 23

ISRE

IFN-stimulated response elements

JAK

Janus Kinase

LncRNA

Long non-coding RNA

LN

Lupus Nephritis

MAPK

Mitogen-activated protein kinase

MDA5

Melanoma differentiation-associated protein 5

mRNA

Messenger ribonucleic acid

NF-kB

Nuclear Factor Kappa B

PAMPs

Pathogen-associated molecular patterns

PBMCs

Peripheral blood mononuclear cells

PRRs

Pattern recognition receptors

RIG-I

retinoic acid inducible gene I

RLRs

RIG-I-like receptors

RNA

Ribonucleic Acid

SAMHD1

Sterile Alpha Motif (SAM) domain- Histidine- Aspartic (HD) domain-containing protein 1

SNP

Single Nucleotide Polymorphism

SOCS1

Suppressor of cytokine signaling 1

STAT

Signal transducer and activator of transcription

STING

Stimulator of interferon genes

SLE

Systemic Lupus Erythematosus

TLRs

Toll-like receptors

TREX1

Three prime repair exonuclease

TYK2

Tyrosine Kinase 2

USP18

Ubiquitin specific peptidase 18