Common innate pathways to autoimmune disease

Abstract

Until recently, autoimmune disease research has primarily been focused on elucidating the role of the adaptive immune system. In the past decade or so, the role of the innate immune system in the pathogenesis of autoimmunity has increasingly been realized. Recent findings have elucidated paradigm-shifting concepts, for example, the implications of “trained immunity” and a dysbiotic microbiome in the susceptibility of predisposed individuals to clinical autoimmunity. In addition, the application of modern technologies such as the quantum dot (Qdot) system and ‘Omics’ (e.g., genomics, proteomics, and metabolomics) data-processing tools has proven fruitful in revisiting mechanisms underlying autoimmune pathogenesis and in identifying novel therapeutic targets. This review highlights recent findings discussed at the American Autoimmune Related Disease Association (AARDA) 2019 colloquium. The findings covering autoimmune diseases and autoinflammatory diseases illustrate how new developments in common innate immune pathways can contribute to the better understanding and management of these immune-mediated disorders.

1. Introduction

At least 20 million Americans are affected by autoimmune disorders according to the American Autoimmune Related Disease Association (AARDA) [1]. It is anticipated that the burden of these diseases is likely to rise in the near future. The factors that contribute to aberrant autoimmunity are numerous and complex. As a result, uncovering of the underlying mechanisms in autoimmunity remains a challenge. However, cross-disciplinary and collaborative research continues to revolutionize our understanding of the underlying pathophysiology of these disorders. To maintain homeostasis, both arms of the immune system, the adaptive and innate, must work in concert and in balance. Autoimmune diseases are characterized by aberrant cellular and humoral immune responses to self antigens [2]. For this reason, autoimmune disease-related research has predominantly been focused on understanding the mechanistic role of the adaptive immune system. However, the innate immune system also plays a vital role in autoimmunity, including participation in certain effector functions regulated by lymphocytes [3,4]. Herein, we aim to elucidate the common innate pathway of autoimmunity. We describe findings pertaining to common autoimmune diseases such as systemic lupus erythematosus (SLE) (or lupus), rheumatoid arthritis (RA), and type 1 diabetes (T1D); to less common autoimmune diseases such as bullous pemphigoid (BP); and to autoinflammatory syndromes such as chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature syndrome (CANDLE) and stimulator of interferon genes (STING)-associated vasculopathy with onset in infancy (SAVI). Our objective is to highlight work that researchers discussed during the AARDA 2019 colloquium, “Common innate pathways to autoimmune disease.”

2. Innate Immune-mediated Inflammation and Autoreactive Responses

The application of novel technologies as well as the paradigm-shifting concepts regarding the role of innate immune system in auto-immunity research can lead to important medical advancements for effective management of these diseases. This includes both prevention and treatment of autoimmunity. Indeed, there has been success in the treatment of RA, MS, SLE, and autoinflammatory diseases with agents that block key cytokines and other mediators produced by innate cells. The initiation of an innate immune response often involves pattern-recognition receptors (PRRs), which have evolved to recognize stress and danger signals. These signals are received from our endogenous environment in the form of danger-associated molecular patterns (DAMPs) released from stressed and damaged cells, and from our exogenous environment in the form of pathogen-associated molecules patterns (PAMPs) from pathogens as well as various xenobiotic agents (e.g., mercury, cigarette smoke, and silica) (Fig. 1). In fact, previously described “Adjuvant effect” reflects in part the contribution of microbial and some other products to the autoimmune process [5]. The toll-like receptor (TLR) family is one of several PRR families that recognize these signals. TLR-mediated activation of myeloid cells (e.g., monocytes, macrophages, granulocytes, etc.), innate-lymphoid cells (ILCs), and γδ-T cells can contribute to the pathophysiology of RA [6], SLE [7], and numerous other autoimmune diseases [8,9], in part by inducing the expression of pro-inflammatory cytokines such as IL-1β, IL-6, IL-17, interferons (IFNs), and tumor-necrosis factor alpha (TNFα). The aberrant activation of these and other related inflammatory pathways contributes to diverse processes leading to tissue pathology and clinical manifestations. Furthermore, dysregulation of pro-inflammatory cytokines such as IL-1β can influence susceptibility/resistance to autoimmunity, e.g., arthritis [10]. However, there also are mechanisms through which the innate immune system may protect against autoimmunity. For example, phagocytosis of dead and dying cells by macrophages limits the amount of circulating PAMPs and DAMPs that activate TLRs and contribute to inflammation. Another example is mycobacterial adjuvant-induced effect on protection against T1D in mice, which has been attributed to modulation of not only adaptive immune response [11], but also innate immune response [12].

Fig. 1.

Contribution of innate immunity to disease pathogenesis. Besides the well-known contribution of adaptive immune system, the innate immune system also plays a vital role in autoimmunity. Initiation of the innate immune response involves pattern-recognition receptors (PRRs) that recognize stress and danger signals. The origin of these signals might be endogenous (e.g., danger- associated molecular patterns (DAMPs) released from stressed and damaged cells) or exogenous (e.g., pathogen-associated molecules patterns (PAMPs) from pathogens; and xenobiotic agents (e.g., mercury and cigarette smoke)). The hos’s microbiota and metabolic pathways can also influence innate responses, which can influence gender bias in autoimmune susceptibility and certain metabolic disorders. Similarly, altered cellular trafficking and recruitment, as well as excessive cell death and impaired clearance of cellular debris can also lead to immune pathology. (Abbreviations: BP, Bullous pemphigoid; CHB, congenital heart block; CVD, cardiovascular disease; RA, rheumatoid arthritis; SLE (or Lupus) systemic lupus erythematosus; T1D, type 1 diabetes; T2D, type 2 diabetes; and autoinflammatory syndromes such as CANDLE, chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature syndrome, and SAVI, stimulator of interferon genes (STING)-associated vasculopathy with onset in infancy.

3. Trained immunity

This paradigm-shifting concept, emphasizing that the innate immune system can develop resistance to reinfection, has been referred to as ‘innate immune memory’ [13], ‘trained immunity’ [14], and ‘systemic acquired resistance’ [15], among other terms [16]. Netea and colleagues proposed that mammalian innate immunity displays features of immunological memory of past encounters with pathogens or related stimuli, and referred to this phenomenon as “trained immunity [14]. Early evidence for a memory-like response of the innate immune system came from studies in organisms without an adaptive immune system, such as plants and invertebrates [17]. Evidence of the trained immune response has since been described in monocytes [18], natural killer (NK) cells [19] and innate lymphoid cells (ILCs) [20]. Classical immunological memory refers to the process by which lymphocytes of the adaptive immune system (i.e., T and B cells) undergo regulated genetic rearrangement, somatic gene mutation, and clonal expansion to produce and propagate antigen-specific lymphocytes. Adaptive memory allows the immune system to respond more rapidly and more effectively to clear an infection, if the same pathogen is encountered again. Early work on ‘trained immunity’ similarly focused on how an encounter with a pathogen caused fundamental and conserved changes within innate immune cells, thereby enhancing protection against a secondary encounter with the same or similar pathogen [21]. In addition to the fundamental role for adaptive memory in pathogen defense, adaptive memory is also necessary for self-tolerance, a process that is essential to prevent autoimmunity. In a similar way, cells of the innate immune system can either be primed or tolerized; however, unlike adaptive memory, which is antigen-specific, the response of an innate immune cell to one stimulus may be altered as a result of the so-called ‘training’ of that cell to a prior encounter with a seemingly unrelated stimulus (Fig. 2) [21]. Netea, Joosten, and colleagues showed that the process of ‘rained immunity’ involves epigenetic reprogramming [18,21,22].

Fig. 2.

A schematic view of “trained immunity”. A myeloid cell undergoes a ‘trained’ immune response to a primary stimulus, danger-associated molecule pattern (DAMP) or pathogen-associated molecular pattern (PAMP), which binds and signals through the Toll-like receptor (TLR)2 or TLR4 pathway. This stimulus alters the epigenetic and metabolic state of the cell in a manner that is conserved so as to position it for a hyper-inflammatory response to a secondary stimulus. Secondary stimuli can be immunologically unrelated but, as a result, the elevated expression of pro-inflammatory cytokines to this secondary stimulus can propagate or exacerbate an autoimmune response.

The activation of a TLR pathway can fundamentally change a cell’s metabolic and epigenetic wiring, indicative of ‘trained immunity’ [22]. This rewiring alters pathways that are critically important in autoimmunity, including inflammation and wound healing. The implication of this in autoimmunity is evident from work on monocytes exposed to DAMPs. The DAMPs that bind TLRs (i.e., TLR2 and TLR4) are released from stressed or damaged cells during chronic autoinflammation. As a result, they often accumulate in the tissues and are found at elevated levels within the circulation of autoimmune patients [23,24], for example, heat-shock proteins in the case of RA [25] and high mobility group box 1 protein (HMGB1) in the case of SLE [26]. Generally, negative feedback loops prevent aberrant inflammation, as is the case with lipopolysaccharide (LPS)-induced TLR-signaling, whereby nuclear-factor kappa beta (NF-κB) mediates a negative feedback loop to prevent excessive inflammation [27,28]. Similarly, monocytes are primed in a hyper-inflammatory state by exposure to relatively low concentrations of many DAMPs, but become tolerized when the concentration of these DAMPs is increased [29]. This concentration-specific switch was less prominent for the TLR4-ligand oS100A4, a type of DAMP. Monocytes exposed to oS100A4 became primed for an increased hyper-inflammatory response during a second stimulus, resembling their state after exposure to beta-glucan, a fungal pathogen-associated antigen. Additionally, monocytes exposed to oS100A4 recovered from this tolerized state that most other DAMPs induced at high concentrations only. This may in part explain the association between s100A4 and RA [30]. This finding also begs the question about how beta-glucan could contribute to autoimmunity. Beta-glucan is a component of the fungal cell wall that binds host receptor dectin-1, and as such is involved in fungal-mediated trained immune protection [31], but many foods also contain beta-glucan. Macrophages derived from beta-glucan-exposed monocytes are poised for a hyper-inflammatory state [18]. In this regard, the amount of beta-glucan consumed in diet and/or the fungal microbiome could theoretically prime an individual’s monocyte/macrophage population to be hyper-inflammatory. This could contribute to the initiation/propagation of autoimmunity.

Endogenous agents that do not bind TLRs may also contribute to ‘trained immunity’ of monocytes/macrophages. When monocytes were treated with aldosterone (a steroidal hormone associated with hypertension and studied extensively for its potential role in autoimmunity [32,33,34]) and later stimulated with a TLR2/ TLR4 agonist, they had a hyper-inflammatory signature [35]. Notably, this signature included increased expression of the pro-inflammatory cytokine interleukin-6 (IL-6). These effects were associated with a change in the expression of genes involved in fatty acid metabolism and in the cell’s epigenetic landscape. For an autoimmunity-prone individual, the implication then is that having hypertension or high aldosterone levels could poise monocyte-derived macrophages for an elevated pro-inflammatory response to DAMPs or pathogen-associated TLR2 ligands, thereby contributing to the onset and progression of autoimmunity.

These recent findings and related observations suggest that monocyte/macrophage populations, and possibly other innate immune cell populations, are susceptible to epigenetic and other changes upon priming to both endogenous and exogenous agents [36]. Such changes may persist long-term and poise such cells to exhibit hyper-inflammatory characteristics. For this reason, TLR-blockade therapies may prove efficacious for the prevention or treatment of autoimmunity. It is also reasonable to suggest that trained immunity may contribute to the epidemiological correlation that exists between environmental factors (e.g., infections, xenobiotics, and diet) and autoimmunity [37,38]. Furthermore, mechanistic understanding of ‘rained immunity ’ would open new avenues for enhancing the efficacy of vaccines and for developing better stratagies for the treatment of immunodeficiency, autoinflammatory, and autoimmune diseases.

4. Autoinflammatory syndromes

Autoinflammatory syndromes are a group of diseases characterized by recurrent flare ups of rashes, fever, and symptoms of other pathologies. These diseases are caused by mutations in genes that regulate the innate immune system. As a result of these mutations, there is aberrant signaling of inflammatory pathways, which normally function to prevent pathogenic infections (Fig. 1). Unlike, many autoimmune diseases in which an individual’s susceptibility to disease increases with age, the onset of autoinflammatory syndromes often occurs abruptly during adolescence. Autoinflammatory diseases include, but are not limited to, Familial Mediterranean Fever (FMF), TNF (Tumor Necrosis Factor) Receptor Associated Periodic Fever Syndrome (TRAPS), Mevalonate kinase deficiency (MKD), Pyrin-associated autoinflammation with neutrophilic dermatosis (PAAND), and interferonopathies. In regard to the underlying mechanisms, these diseases include those caused by IL-1-mediated NLRP3 inflammasome activation, IL-18-mediated diseases, and type-1 interferon (IFN)-mediated diseases.

Interferonopathies include CANDLE and SAVI. Interferonopathies are the result of monogenetic mutations that lead to an aberrant increase in type-1 IFN signaling [39,40,41]. These mutations can, for example, alter the homeostatic function of stromal interaction molecule (STIM), of stimulator of interferon genes (STING), or of a related-protein necessary for regulating the STIM-STING interaction. In addition to the importance for STIM to regulate adaptive immunity [42], recent findings suggest that STIM also plays an important role in regulating the induction of the type-I IFN pathway in macrophages. Indeed, individuals with loss-of-function mutations in STIM can develop autoimmunity [43]. In macrophages, STIM prevents aberrant expression of type-1 IFN genes by tethering STING to the endoplasmic reticulum [44]. Janus activated kinases (JAKs) are required to propagate pro-inflammatory cytokine (e.g. IL-6, and type 1 and type 2 IFNs) signaling; accordingly, JAK inhibitors have been used to treat autoimmune diseases, like RA [45]. Clinical studies conducted by Goldbach-Mansky, de Jesus, and colleagues sought to investigate the efficacy of using JAK inhibitors to treat autoinflammatory syndromes [46]. Baricitinib, a selective JAK1/JAK2 inhibitor, proved very successful in a clinical study involving 18 interferonopathy patients with either CANDLE or SAVI [47]. Of the 18 patients enrolled, 14 patients experienced significant improvement. In addition, the prescribed amount of prednisone was also markedly reduced, which was of advantage given the severe side-effects associated with the high dose and long-term use of corticosteroids. The success of baricitinib warrants further investigation of other inhibitors of JAKs or STATs in the treatment of these autoinflammatory diseases and certain autoimmune disorders.

5. Impact of environment on autoimmunity: xenobiotic agents, bacterial products, and the microbiome

The epidemiology of numerous autoimmune diseases and studies in animal models of autoimmunity have indicated that the environment strongly influences an individual’s risk of developing disease via multiple mechanisms [48,49,50,51]. This increased risk may involve an exposure to exogenous agents, often xenobiotics, such as mercury, silica [49,52], and cigarette smoke (which contains thousands of xenobiotic agents) [53]. In some cases, the influence of these xenobiotic agents may be directed through innate immune responses (Fig. 1). In addition to the association between xenobiotic agents and autoimmunity, there is growing evidence for the impact that our microbiota and related metabolic pathways can have on autoimmune processes [54,55,56,57].

5.1. Xenobiotic agents

5.1.1. Mercury

In the case of murine mercury-induced autoimmunity (mHgIA), susceptible mouse strains develop lupus-like disease following sub cutaneous injection of mercuric chloride (HgCl₂) [58]. Similarly, transgenic female mice that overexpressed IFNγ (type-2 IFN) in the epidermis develop spontaneous lupus-like disease [59]. Recent findings have elucidated how IFNγ can contribute to this disease and the mechanisms that might explain how exposure of the skin to mercury differentially drives local and systemic SLE pathologies. The significance of IFNγ in SLE has been demonstrated in clinical trials in which patients were treated with the IFNγ-blocking antibody AMG 811. This treatment reduced IFN-regulatory gene expression and serum levels of C-X-C Motif Chemokine Ligand 10 (CXCL10), which is responsible for neutrophil chemotaxis [60]. Interestingly, B10.S mice deficient in IFNγ still develop acute cutaneous local inflammation at the site of HgCl₂ administration, but do not develop systemic lupus-like disease [58]. In wild type mice, both NK cells and CD11⁺ cells that migrate to the injection site express IFNγ upon administration of HgCl₂ [61]. It was proposed that IFNγ and IL-6 expression in the skin following mercury exposure likely induces type-1 IFNs, which are necessary for systemic disease. Interestingly, Pollard and colleagues have characterized type-1-independent mHglA in certain murine models of lupus-like disease [62]. In the case of type-1 IFN-independent disease, ablation of TLR signaling in transgenic animals deficient in the adaptor protein 3 (AP-3) or solute carrier family 15 member 4 (Slcl5a4) effectively prevented lupus-like manifestations, whereas TLR agonists enhanced the disease. It is notable that an upregulated IL-6 signature is shared in both the type-1 IFN-dependent and type-1 IFN-independent mHglA models. Therefore, IL-6 signaling may be a crucial component of the crosstalk between the inflammatory response within the skin and systemic disease.

Many SLE patients are reported to have an elevated type-1 IFN signature, which is attributed in part to elevated expression of type-I IFN pathways within cells of their skin [63], it is important to note that this signature is not present in all SLE patients [64]. It has been proposed that in some SLE patients, an autocrine feedback loop of type-I IFN signaling in skin keratinocytes enhances inflammation [65]. It is possible that through this mechanism, exposure to mercury could enhance inflammation in patients with SLE and in those at risk of developing SLE.

5.1.2. Silica

Silica exposure is another risk factor in the development of SLE [66]. In the case of certain outbred mice, TLR7 appeared to play a role in the development of accelerated lupus-like disease [67]. This further supports the role that TLR signaling may have in the association of certain xenobiotic agents to SLE and other autoimmune diseases.

5.1.3. Cigarette smoke

The dangers of cigarette smoke have been known for decades, as illustrated by the fact that the first Surgeon General’s report: “Smoking and Heath: Report of the Advisory Committee of the Surgeon General of the Public Heath Service” was published in 1964. The percentage of total US adults that smoke has gradually been decreasing [68,69] but other forms of tobacco and smoking products have hit the US market in recent years. Hookah products have been particularly popular among adolescent and young adults and continue to gain popularity [70]. Additionally, the smoking of e-cigarettes or other vaping products (vaping is the inhaling of a vapor created by such devices) has become an epidemic among adolescent and young adults. The introduction and advertisement of these products has likely contributed to the fact that about 4.7 million Americans of high school-age smoke some form of tobacco, according to the Center for Disease Control and Prevention (CDC) [71]. Smoke, including cigarette smoke, contains thousands of chemicals [72], some of which may perturb the innate immune system and contribute to numerous autoimmune diseases [73]. In the case of nicotine, not in combination with other harmful cigarette chemicals, both immunosuppressive [74,75] and pro-inflammatory [76] effects have been reported on human dendritic cell (DC) populations. The differential effects that nicotine can have was further substantiated using the rat adjuvant-induced arthritis (AA) model of human RA by Moudgil and colleagues [77]. The pretreatment of rats with nicotine before the onset of AA resulted in more severe arthritis, whereas administered of nicotine after disease onset reduced disease severity. This change in arthritis severity showed a positive correlation with the levels of Thl/Thl7 cytokines, the levels of dendritic cell-produced cytokines, and the levels of antibodies to cyclic citrullinated protein/peptide (aCCP). Thus, nicotine modulated the key immune events involved in arthritis pathogenesis [77]. The impact of cigarette smoke on other innate immune cells has been previously reviewed [73]. The health implication of using alternative tobacco/nicotine smoke products remains largely understudied. Additionally, the impact on a person’s health from the individual chemicals contained within, and produced by, different modes of smoking is largely unknown. This is an important gap in information to be filled in the near future.

5.2. Activation of innate immunity for immunotherapy of autoimmunity

The pathogenesis of autoimmune diseases involves both genetic and environmental factors [50,51,54,78]. The gradually increasing incidence of autoimmune diseases has been explained in part by the “Hygiene hypothesis”. This hypothesis suggests the link between decreased exposure of the innate immune system to environmental immune stimulants (e.g., bacterial products) and its impact on the adaptive immune system, which leads to increased autoimmunity. Ridgway and colleagues have systematically examined the effect of deliberate activation of the innate immune system, specifically TLR-4 receptor activation, on experimental type 1 diabetes (T1D) in non-obese diabetic (NOD) mice [79,80]. Several previous studies have shown the prevention of T1D by treatment of NOD mice with bacterial LPS. However, there has not been much success in the treatment of T1D after the disease onset. Ridgway’s group has shown successful disease reversal by treatment of acute-onset T1D in NOD mice with an anti-TLR4/MD-2 agonistic monoclonal antibody. These antibodies targeted antigen-presenting cells (APCs) and rendered them tolerogenic [79,80]. These APCs in turn dampened pathogenic T cell responses involved in disease progression in T1D, but increased the number of Foxp3 + T regulatory (Treg) cells. The mechanism of tolerization of APCs was further evident from studies in NOD.scid mice lacking B cells, whereby T cell-driven disease in these mice was abrogated by treatment with an anti-TLR4/MD-2 antibody. These studies highlight a novel approach to control T1D when treatment aimed at activation of innate immune system is begun after disease onset.

5.3. Microbiome

5.3.1. Microbiome and gender-bias in autoimmunity

Sex-bias in susceptibility to autoimmunity has been well documented for several autoimmune diseases, whose incidence in females is 2–3 times higher than that in males. This outcome has been partially attributed to hormonal differences between males and females, among other factors, but differences in the microbiome may also be a risk factor. Studies by Chervonsky and colleagues have elaborated upon the effect of microbiome on the development of T1D in NOD mice, as well as the impact of the interplay between sex hormones and microbiome on T1D development in male vs. female NOD mice (Fig. 1) [81,82,83].

It was shown that the myeloid differentiation primary response 88 (MyD88)-deficient germ-free (GF) NOD mice develop T1D, whereas specific pathogen-free NOD mice are protected from this disease [84]. The former can be rendered resistant to T1D development by reconstitution with define microbial species present in the gut. Furthermore, MyD88 deficiency has a marked impact on the composition of gut flora. These results demonstrate the influence of the interaction between gut microbiota and innate immune system on T1D development. Additional investigations demonstrated a differential effect of TLR-4 vs. TLR-2 on T1D development such that interaction of certain microbiota with TLR-4/TRIF (Toll/Il-1 receptor (TIR)-domain containing adapter-inducing IFN beta) elicited tolerizing (protective) signals, while an interaction with TLR-2 induced pro-diabetic signals. This has been referred to as the ‘balanced signal hypothesis’ [85].

Chervonsky and colleagues also examined mechanisms underlying the observed differences in T1D in male vs. female NOD mice [86]. This difference in the incidence of T1D in the two sexes was observed in specific-pathogen free (SPF) NOD mice, but not in germ-free NOD mice [86]. This finding, along with the observation that microbiota differs in males vs. females and that reconstitution of microbial flora from male mice into female mice offered protection against T1D to the recipients, suggests the role of hormone-supported alteration in microbiota in gender bias in T1D. This was referred to as the ‘two-signal hypothesis’ for gender bias.

From the foregoing, it is clear that T1D, and likely other autoimmune diseases, are influenced by the intestinal microbiota. Therefore, correcting perturbation in the microbiota or utilizing its beneficial components may help prevent the onset of disease and reduce the prevalence of autoimmunity.

5.3.2. Microbiota and metabolic disease pathogenesis

An individual’s behavior and environment can shape the microbiome, and in doing so, contribute to health or disease (Fig. 1). Indeed, the ‘dysbiosis’ of the microbiome is associated with numerous autoimmune diseases [87,88]. Modern technologies can be applied to improve patient diagnostics in order to determine individualized treatment practices to best benefit the patient. For example, the analyses of patient genotyping and metagenomic data, along with patient plasma metabolomics data, revealed that the plasma levels of butyrate and propionate (both short-chain fatty acids (SCFAs)) were associated with diabetes risk and treatment efficacy, respectively [89]. (Collectively, ‘Omics’ refers to genomics, proteomics, metabolomics and other similar fields of study.) Gut commensals convert complex carbohydrates from the diet into SCFAs, which have previously been shown to have numerous health benefits [90,91]. The study by Joosten and colleagues goes a step further by identifying how an individual’s genome was associated with the bioavailability of butyrate. A higher concentration of plasma butyrate was correlated with a favorable insulin response; however, elevated plasma propionate level was associated with increased type-2-diabetes (T2D) risk [92]. In a separate study, obese males, at risk of diabetes and cardiovascular disease (CVD), were given 4 g of sodium butyrate for 4 weeks [89]. Monocytes from these individuals had a significantly reduced inflammatory response to oxidized low-density lipoprotein (oxLDL), a byproduct of oxidative stress and potential indicator of obesity and T2D, compared to monocytes from obese males who did not receive that supplement. This effect of butyrate supplementation was also tested on monocytes from lean males; however, no such effect was observed. Our microbiota produces metabolites which can serve as biomarkers of microbiome ‘dysbiosis’; however, recent findings indicate that the benefits associated with taking probiotics, prebiotics, or probiotic metabolites may not apply to all individuals. There has been substantial research on SCFAs and their effect on the immune system, but there are many other microbiota derived metabolites which require further investigation as to their role in health and autoimmunity.

6. Cell trafficking and recruitment

The recruitment of innate immune cells to inflamed tissues propagates chronic inflammation and tissue damage. This is evident, for example, within inflamed joints of RA and skin lesions of BP (Figs. 1, 3). The recruitment of cells to affected tissues involves a crosstalk between the adaptive and innate immune systems. This involves, among other factors, the activation and expression of adhesion molecules on endothelial cells and immune cells, as well as the release of chemokines that attract neutrophils and monocytes. Immune cells in circulation can then migrate to the site of inflammation and the surrounding tissue. Accordingly, blocking the recruitment of innate immune cells to inflamed tissues in autoimmunity might offer a promising new approach to disease management.

Fig. 3.

Cellular processes involved in the pathogenesis of autoimmunity. Tissue damage during autoimmunity contributes to, and can result from, a chronic forward-feeding pathophysiological network involving the innate immune system. The perturbed clearance of dead cells or improper regulation of apoptosis can be an underlying factor in autoimmune disease. The intracellular/nuclear debris released from dead cells may form immune complexes with autoantibodies. Free intracellular debris, including DAMPS (e.g., nucleic acids) is recognized by TLR-family receptors, whereas the Fc portion of autoantibodies in immune-complexes are recognized by Fc receptors (FcRs) on myeloid cells. This in turn induces the expression of pro-inflammatory cytokines (e.g., IFNs, IL-6, TNFα), which contribute to other pathophysiological processes, including enhanced tissue remodeling/damage, autoreactive adaptive immune response, and inflammatory response of other innate immune cells in the affected tissues.

6.1. Bullous Pemphigoid

BP presents as blistering of the subepidermis. Patients with BP typically present with elevated circulating autoantibodies against two hemidesmosomal proteins, BP180 and BP230, and high numbers of infiltrating innate inflammatory cells, including neutrophils, macrophage, mast cells, and eosinophils within the afflicted skin. Liu and colleagues have examined how these innate immune cells contribute to the pathophysiology of disease. There are several murine models of BP. Most of the work has been conducted using a murine model in which passive transfer of anti-murine IgG BP180 antibodies into the skin causes the development BP-like blistering. Antibodies appear to mediate cross-talk between the innate and adaptive immune systems. In this model, the binding of the pathogenic IgG F(ab) region with FcγRIII on skin-infiltrating neutrophils stimulates the release of neutrophil elastase (NE) and matrix-metalloprotease (MMP)-9, which contribute to blistering [93]. Although these pathogenic antibodies are necessary to induce disease, the adaptive arm of the immune system is not sufficient for blistering. The innate myeloid cells are necessary for the clinical manifestation of blistering. Indeed, deficiencies in one of the following – neutrophils, mast cells, or macrophages – protects mice from blistering. These findings suggest that tissue-residing and tissue-infiltrating innate immune cells contribute to the manifestation of blistering in this model, whereas mice lacking T and/or B cells still develop blistering [94]. These innate immune cells work in concert with one another to propagate inflammation to eventually cause blistering. Activated macrophages in the skin activate and recruit mast cells, which upon degranulation attract large numbers of neutrophils into the skin. In this model, mast cell degranulation releases histamine, cytokines, and proteases, which increase blood vessel permeability. The peak of mast cell degranulation coincides temporally with a significant increase in the rate of neutrophil recruitment to the injection site, supporting the idea that mast cell degranulation is required for pathogenic neutrophil recruitment. Mast cell protease-4 (mMCP-4), which has been linked to other autoimmune diseases besides BP, activates MMP9 in the skin and cleaves the hemidesmosomal transmembrane protein BP180, one of the two well-described pathogenic autoantibody targets in BP [95]. Perturbation of MMP activation contributes to the pathogenesis of numerous other autoimmune diseases such as RA [96,97]. Although, both macrophages and mast cells play an important role in BP, it is the neutrophils that cause most of the direct damage to the subepidermis. Regulation of neutrophil trafficking and recruitment has been reviewed previously [98]. Neutrophils are recruited in two seemingly semi-independent phases in the passive antibody transfer murine BP model. Early recruitment of neutrophils was dependent on LFA-1 p2 integrin, whereas later recruitment and accumulation of neutrophils within the skin was dependent on MAC-1 p2 integrin and FcγRIII [93,99]. Mice lacking MAC-1 showed higher numbers of neutrophils in the skin early on, but these cells failed to be retained as compared to the wild type controls, and therefore blistering did not occur. Additionally, the death of these neutrophils was mitigated when they were not retained within the skin, preventing blistering. Some of the reasons for this effect are discussed below (under Section 7).

6.2. Chemokine blockade therapy

The targeting of chemokine/ chemokine receptor systems for disease treatment can be difficult. For neutrophil-targeted therapies, such difficulties include - the redundancy in chemokine/ chemokine receptor function; the need to saturate the system with the inhibitor to interfere with the chemokine/ chemokine receptors function as circulating neutrophils are a relatively large portion of circulating cells; and the rapid neutrophil and chemokine/ chemokine receptor turn-over, both in circulation and on the cell surface, respectively. Luster and colleagues have applied advanced technologies such as intravital imaging (i.e., imaging of live animals at microscopic resolution; intravital microscopy) and the quantum dot (Qdot) fluorescence system to readdress known immune processes that regulate inflammation and autoimmunity, including antibody Fc receptors [100], chemokines, and complement pathways [101,102]. Application of novel concepts and intravital imaging to study cellular trafficking during disease will potentially help to overcome the failures of previous trials for the benefit of patients.

In RA, the success of chemokine blockade therapy has been met with complications, although targeted therapies in animal models have been very successful. The targeting of leukotriene B4 (LTB4) and its high affinity receptor BLT1 (LTB₄: BLT1) has been tested clinically but showed minimal success, although a similar targeting in preclinical studies of arthritis has shown promising results [103]. Using K/BXN model of autoantibody-induced arthritis, the role of BLT1 and chemokine receptors CCR1 and CXCR2 in neutrophil influx into the joints was examined. It was shown that BLT1 on neutrophils played an important role in chemotaxis of neutrophils, including those not expressing BLT1, into arthritic joints, which resulted in production of chemokines and arthritis induction [104]. Furthermore, blockade of BLT1 was effective in inhibiting arthritis progression. LTB₄ also signals through BLT2, which is expressed on synovial-like fibroblasts and regulates the expression of IL-1β and TNFα, two of the pro-inflammatory cytokines that are critically important in the pathophysiology of RA. In addition, LTB₄ has been shown to be a key mediator in cascades involving other mediators (e.g., complement, lipids, cytokines and chemokines) that regulate initial as well as sustained neutrophil influx into the joints leading to arthritis [103]. For example, complement C5a receptor (C5aR) and Fc gamma receptors (FcγRs) function in tandem, the former via secretion of LTB₄ and the latter via secretion of IL-1 [105]. Accordingly, the targeting of LTB₄ and its receptors continues to be a promising approach for developing novel therapeutics for arthritis and other diseases like BP and SLE. Additional new targets are being defined by results of intravital microscopy studies in live arthritic animals [101]. For example, C5aRl is involved in neutrophil arrest, while CXCR2 is involved in trans-endothelial migration of neutrophils.

For targeting chemokines for arthritis treatment purposes, the production of antibody cocktails and the development of broadly-specific antibodies to conserved regions within chemokine receptor families may prove to be more efficacious than the targeting of a single chemokine receptor with a single high affinity antibody. This approach was implemented in the design of broadly-recognizing antibodies against the ELR⁺ CXC family of chemokines [106]. Additionally, potentially therapeutic antibodies generated to specifically recognize non-glyco-saminoglycans (GAG)-bound CXCL10 effectively blocked neutrophil trafficking, whereas antibodies generated without specificity to the non-GAG-bound form of CXCL10 failed to show in vivo efficacy [107]. Thus, the binding-specificity of therapeutic antibodies to the intended target should be considered during the drug design process.

6.3. Platelets

Prematurely developing cardiovascular disease (CVD) is associated with chronic inflammation in several autoimmune diseases, including lupus and RA. Clancy, Buyon, and colleagues showed that platelets from SLE patients are hyperactive and contribute to the stimulation of endothelial cells [108]. SLE patients were also reported to have higher numbers of monocyte/leukocyte-platelet aggregates in their blood compared to healthy controls. The expression of genes involved in inflammation and immune cell recruitment during inflammation were highly upregulated in human umbilical cord vein endothelial cells (HUVECs) cultured with an SLE patien’s platelets as opposed to a healthy person’s platelets. This effect was IL-1β-dependent and its potential contribution to SLE is supported by previous findings indicating that megakaryocytes mature in response to IL-1β and that platelets can be activated by IL-1β [109]. In fact, platelet-mediated activation of endothelial cells can be inhibited by treatment of SLE patien’s platelets with an IL-1β-neutralizing antibody or pre-treatment of HUVECs with anti-IL-1 receptor antibodies [108].

Further genotyping and cellular mechanistic studies performed in SLE patients have revealed the key role of FcγRIIA, a low affinity receptor for human IgG, in driving the events associated with CVD. Specifically, polymorphism H131R (histidine to arginine at position 131) of FcγRIIA was associated with various features of CVD in SLE patients, but this genotype was not found in healthy controls [110]. This SLE-associated genotype, but not the wild type FcγRIIA, also showed a positive correlation with increased levels of soluble E-selectin. Taken together, above studies highlight the role of platelet-endothelial interactions, IL-1β, and FcγRIIA polymorphism in the development of premature CVD in SLE patients.

7. Cell death and clearance of cellular debris

In autoimmunity, excess cellular damage and cell death combined with inefficient clearance of the dead cell debris, including immune complexes, can directly contribute to disease pathology and disease exacerbation (Figs. 1, 3) [111], whereas impaired cell death pathway (e.g., Fas pathway) can afford protection in a subset of autoimmune diseases [112]. TLRs recognize nucleic acids that can be associated with immune complexes, as well as other DAMPs released from dying cells, thereby inducing inflammation. Cells of the innate immune system, especially a sub-population of macrophages, are tasked with clearing dying cells to prevent chronic inflammation. For SLE (Lupus) patients, there is often an accumulation of dead cells within their bone marrow. Increased numbers of syndecan-1 (CD138)-expressing plasma cells and elevated soluble-CD138 (sCD138) have been reported in SLE patients. For this reason, CD138 has been proposed as a therapeutic target to treat SLE [113,114]. Recent findings by Reeves and colleagues have revealed that CD138⁺ may not only be a marker of pathogenic plasma cells that contribute to disease, but also be a marker for a subset of macrophages that specialize in clearing apoptotic cells/debris. Mice with pristane-induced lupus generate excessive numbers of Ly6C^hi inflammatory peritoneal macrophages rather than small peritoneal macrophages (SPM) that are CD138⁺ [115]. These CD138⁺ macrophages express the scavenger receptor Marco and high levels of activated cyclic AMP-responsive element binding (CREB), which is a transcription factor implicated in generating alternative (M2) macrophages. It has been suggested that CD138⁺ macrophages inhibit inflammation by phagocytosis of dead and dying cells [116]. Activation of the liver X receptor (LXR) was previously shown to promote macrophage phagocytosis and immune tolerance [117,118]. Lupus mice treated with an LXR agonist develop less end-organ damage [116]. This reduction in end-organ damage was associated with an increase in CD138⁺ macrophages and fewer inflammatory monocytes. LXR agonists also decrease inflammatory cytokine production by human SLE monocytes, suggesting that LXR agonists may serve as promising therapeutics for SLE treatment. TLR4 signaling has been shown to inhibit LXR gene expression in macrophages [119]. This further suggests that TLRs may be crucial to the pathophysiology of certain autoimmune diseases, a common concept that was emphasized throughout the colloquium.

The role of the immune system in the development of lupus glomerulonephritis and autoantibody production has been extensively studied; however, relatively less is known about hematological manifestations of lupus affecting the bone marrow of SLE patients and how cell death could contribute to this pathology. Unlike the increase in the production of autoantibodies, the anemia and hypocellularity within the bone marrow of an SLE patient may be type-1 IFN-independent [120]. The bifurcation of these two lupus-associated pathologies may come down to a balance of the type-1 IFN production by monocytes/ macrophages vs. TNFα production by Lupus Erythematosus (LE) Cells (in these cells, nuclei are phagocytosed by mature polymorphonuclear leucocytes and digested). Necrotic cell components signal through TLR7 in both these processes, but the balance between the different cell types may dictate what pathologies manifest in an individual with SLE. Chronic inflammation and cell death-driven inflammation in SLE also pose a serious risk to the fetus during pregnancy and can result in neonatal lupus. The severity of neonatal lupus can vary, with cases of congenital heart block (CHB), the most severe form, leading to unsuccessful pregnancies as a result of fibrosis and calcification of the fetal heart.

CHB hearts have high numbers of infiltrating macrophages and fibroblasts, which likely contribute to the disease, as revealed by a series of studies by Buyon, Clancy, and colleagues [121,122]. In pregnant SLE patients, anti-SSA/Ro antibodies are transferred from the mother to the fetus and likely induce inflammation within the fetal heart. Exposure of macrophages to maternal anti-SSA/Ro antibody-containing immune complexes and self-RNA induces a type-I IFN signature and is likely to contribute to the fibrotic response of adjacent fibroblasts in the fetal heart [121]. There is an association between the elevated plasma type-I IFN signature in SLE patients that experience complications compared to pregnant SLE patients that do not have complications, indicating a likely role for type-1 IFN contributing to CHB [123,124]. Notable among the type-1 IFN-induced genes is sialic acid-binding Ig-like lectin 1 (Siglec-1), a receptor on monocytes/macrophages. The expression of Siglec-1 was also identified in the septal region of affected fetal hearts [124]. Accordingly, it has been proposed that Siglec-1⁺ macrophages, along with type 1-IFN-driven processes, may contribute to the pathology of CHB.

In addition to the type-1 IFN pathway, ssRNAs complexed within anti-SSA/Ro antibody-containing immune complexes can activate the TLR7/8 pathway of myeloid cells. As a result, the NF-κB- and STAT1-driven pathways is activated, which contributes to the inflammatory profile of these cells, along with epigenetic modifications [122]. Interestingly, TLR-7/8-ligation-induced epigenetic modifications may be blocked with hydroxychloroquine (HCQ), which may be useful for the prevention of organ damage, including CHB, in an anti-Ro offspring [122].

8. Conclusion

It is evident that common innate immune pathways contribute to the pathogenesis of numerous autoimmune disorders. It is important to further define the precise mechanisms involved in these processes. The application of modern, advanced technologies (e.g., intravital imaging, Qdot system) can help investigators to readdress old concepts (e.g., cell death and trafficking) in a new light as well as to unveil underpinnings of emerging newer concepts (e.g., ‘trained immunity’). This in turn would lead to better approaches for the prevention and treatment of autoimmune diseases. This is already evident from the success of newer therapeutic agents that target the mediators of innate immune response, such as the inhibitors of JAK/STAT pathway. Finally, increasing global industrialization will have consequences on human health. It is important to be both preemptive in our approach to mitigate the consequences of those changes and persuasive in making the general public more aware of the impact of such changes on health, particularly autoimmune diseases.