Recent developments in SLE pathogenesis and applications for therapy

Abstract

Purpose of Review

Systemic lupus erythematosus (SLE) pathogenesis is complex. Aberrancies of immune function that previously were described but not well understood are now becoming better characterized, in part through recognition of monogenic cases of lupus-like disease.

Recent Findings

We highlight here recent descriptions of metabolic dysfunction, cytokine dysregulation, signaling defects, and DNA damage pathways in SLE. Specifically, we review the effects of signaling abnormalities in mammalian target of rapamycin (mTOR), Rho kinase, Bruton’s tyrosine kinase (BTK), and Ras pathways. The importance of DNA damage sensing and repair pathways, and their influence on the overproduction of type I interferon in SLE are also reviewed.

Summary

Recent findings in SLE pathogenesis expand on previous understandings of broad immune dysfunction. These findings have clinical applications, as the dysregulated pathways described here can be targeted by existing and preclinical therapies.

Introduction

Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease defined by the presence of autoantibodies, particularly antibodies directed against nuclear antigens. These autoantibodies, which unify the many different clinical presentations of SLE, reflect a breach of central tolerance. Defects in the clearance of apoptotic debris and aberrant presentation of self-antigens are major mechanisms that contribute to this breach. Excessive plasmacytoid dendritic cell activation and interferon production amplify the inflammatory response in SLE. Finally, end organ tissue damage is mediated by immune complexes and abnormal activation of T lymphocytes and other immune cells. These mechanisms are all known to be influenced by genetic, environmental, and hormonal factors.

The last few years have seen some interesting developments in our understanding of SLE pathogenesis. The spectrum of abnormalities that have been characterized continues to expand, currrently including metabolic derangements, signaling and biochemical defects in immune cells, and impaired sensing and repair of DNA damage (Figure). Correlation of these dysregulated pathways with specific clinical and pathophysiologic aspects of SLE has been aided by the study of monogenic forms of lupus-like disease. Further, characterization of these pathway defects in SLE has allowed identification of new targets for therapeutic intervention.

Figure.

Schematic of immune abnormalities known to contribute to SLE pathophysiology. This diagram greatly underestimates the complexity of interactions that are dysregulated in SLE in order to highlight recent findings described here.

Metabolic defects

Signaling through the T cell receptor (TCR) is dysregulated in SLE in multiple ways, including aggregation of lipid rafts around TCR clusters, downregulation of the CD3ζ chain, and decreased upregulation of IL-2 in response to TCR activation [1].

The CD3ζ in SLE T cells is downregulated and its function is substituted by FcεRIγ, which signals through SYK rather than ZAP70, resulting in a stronger signal of the TCR [1]. The lysosomal degradation of CD3ζ is aggravated by the increased activity of the mammalian target of rapamycin (mTOR) in these cells, resulting in upregulated endosomal trafficking and turnover of cell surface markers [2]. mTOR is a sensor of mitochondrial polarization which coordinates multiple cellular pathways as a component of mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2). T cells from SLE patients treated with rapamycin, which inhibits mTOR activity, showed restored levels of CD3ζ and normalized signaling through the TCR [2].

Mitochondrial hyperpolarization and increased mTORC1 activity likely influences the T cell phenotype in SLE in many other ways, including altered follicular helper and regulatory T cell profiles [3,4]. CD3+CD4-CD8- double negative (DN) T cells are thought to play a pathogenic role in SLE through the secretion of IL-17. Treatment of SLE T cells with rapamycin decreases in vitro IL-17 production and promotes the development of regulatory T cells (Tregs) [3]. The expression of IL-17 and other pro-inflammatory cytokines in DN T cells appears to be especially dependent on mTORC1 activity [3]. In Tregs, IL-2 signaling induces mTORC1 activity, which in turn influences cholesterol metabolism and upregulates inhibitory pathways important for Treg function [5]. This mTOR activation pathway in Tregs is reliant on activation of the protein phosphatase PP2A [6]; abnormalities of PP2A expression and activation have been described in SLE patients [7].

Clinical evidence for the role of mTOR activity in SLE is further supported by the coexistence of SLE and tuberous sclerosis, described in recent case reports [8–10]. Tuberous sclerosis is a rare neurologic condition associated with benign tumor growths due to mutations in either TSC1 (hamartin) or TSC2 (tuberin). Hamartin and tuberin form a complex that inhibits mTORC1; immune profiling of one of the described tuberous sclerosis patients with SLE demonstrated significant mitochondrial hyperpolarization and increased mTOR activity in vitro [10].

Cholesterol homestasis

The glycosphingolipid profile within lipid rafts is altered in SLE, with increased expression of lactosylceramide and other species of glycosphingolipids when compared to T cells from healthy controls [11]. This increase is associated with increased TCR activation and appears to be due to upregulation of liver X receptor β (LXRβ), a nuclear regulator of glycosphingolipid homeostasis. LXRα polymorphisms have been associated with SLE [12], and mice deficient in LXRα and LXRβ develop lupus-like disease [13]. These LXRs influence immune cell function in multiple ways. LXR activity promotes cholesterol efflux through upregulation of the ATP binding cassette transporters ABCA1 and ABCG1. In murine T cells, deficiency of ABCG1 results in intracellular cholesterol accumulation with consequent T cell activation and proliferation [14]. Notably, in Tregs, intracellular accumulation of ceramide increases activity of PP2A, linking cholesterol pathways again back to T cell activation [6].

However, a recent study in mice suggests that it is impairment of cholesterol efflux in dendritic cells, but not T cells, that contributes to lupus-like immune activation. Dendritic cells from mice with double deficiency of ABCA1 and ABCG1 showed marked cholesterol accumulation and also NLRP3 inflammasome activation with increased secretion of IL-1β and IL-18 [15]. Selective deficiency of the ABCA1/ABCG1 transporters in dendritic cells was sufficient to induce a lupus-like phenotype with lymphadenopathy and glomerulonephritis [15]. It is not clear how intracellular cholesterol accumulation leads to inflammasome activation. One proposed mechanism is that cholesterol increases stability of Toll-like receptors (TLR) in lipid raft clusters, enhancing the TLR signal response [16].

Regulatory T cells and low-dose IL-2

T cell production of IL-2 is impaired in SLE due to abnormal TCR signaling responses as well as repressed IL-2 transcription [17]. IL-2 is generally a pro-inflammatory cytokine, but is also critical for the development and function of Tregs [18]. Deficiency of IL-2 likely contributes to the Treg abnormalities observed in SLE [19]. In mouse models of lupus, treatment with IL-2 has resulted in variable levels of improvement [20,21].

In humans, low-dose IL-2 therapy was first trialed with good success in two other conditions characterized by Treg dysfunction, graft-versus-host disease and hepatitis C-induced vasculitis [22,23]. There have now been several reports of low-dose IL-2 therapy in SLE. In one case report, an SLE patient experienced remarkable improvement in skin rash and myositis after a 2 month treatment course with recombinant IL-2 [24]. The same researchers then described five patients with active SLE treated with daily subcutaneous injections of IL-2 administered over 5 consecutive days [25]. Treatment with just this single course resulted in significant increases in Treg numbers as well as in CD25 expression on Tregs [25].

In a larger uncontrolled study, recombinant IL-2 administered over a 12 week treatment period resulted in increased number and function of Tregs, while follicular helper T cells (Tfh) and DN T cell populations declined [26]. Clinically, 90% (34/38) of patients showed a 4 point drop in their SLE disease activity index (SLEDAI) score over the 12 week treatment period [26]. These reports claiming impressive therapeutic efficacy should await the results of controlled studies. Recombinant IL-2 is currently approved for the treatment of select malignancies, and its efficacy in autoimmunity remains under investigation.

Interferon (IFN)

SLE patients characteristically show increased serum IFN-α levels and a pattern of increased expression of type I IFN-stimulated genes in peripheral immune cells, known as the IFN signature [27]. This is in part related to expanded numbers of plasmacytoid dendritic cells (pDCs), the primary producers of IFN-α in response to nucliec acid. IFN-α, in turn, has a number of effects that drive lupus pathophysiology, including increased expression of BAFF, IL-6, and other cytokines, as well as increased autoantibody production [27–29].

Monogenic cases of lupus-like disease have also emphasized the importance of type I IFN in effecting these immune abnormalities. Familial chilblain lupus and Aicardi-Goutieres syndrome are both syndromes characterized by autoantibodies and systemic inflammation, and are both due to mutations in TREX1 [30,31]. Lack of the TREX1 exonuclease allows accumulation of nucleic acid fragments during the DNA damage response; these fragments ultimately stimulate type I IFN production [32]. Mutations in other components of the DNA damage sensing and repair pathways also result in upregulation of type I IFN and a lupus-like phenotype. Examples of these “interferonopathies” include RNASEH2A/B/C, SAMHD1, ADAR, and IFIH1 [33]. Outside of these monogenic cases, polymorphisms in these genes may also contribute to the IFN signature and to SLE disease risk. As an example, TREX1 polymorphisms were found at a frequency of 0.5% in one SLE cohort.

Type I IFN may play a particular role in the central nervous system (CNS) manifestations of lupus, as suggested by the severe neurologic phenotype of Aicardi-Goutieres. In one case report, TREX1 mutation was found to underlie the development of CNS lupus in a young child, while in a very large cohort of SLE patients, a TREX1 haplotype was associated with risk for neurologic involvement [34,35]. IFN-α was shown recently to activate microglial cells which then engulf neuronal structures, pruning synapses [36]. IFN-mediated synapse loss probably represents a mechanism responsible for CNS manifestations in patients with SLE.

These findings, among others, suggest the type I IFN signaling pathway as an appropriate therapeutic target. Anifrolumab is a monoclonal antibody directed against the IFNα receptor subunit 1 (IFNAR1). The first clinical trial of anifrolumab was recently published [37]. In a phase IIb trial, 305 patients were randomized to receive placebo, low dose (300 mg), or high dose (1000 mg) infusions every 4 weeks. The primary endpoint was a composite of the SLE responder index and reduction in corticosteroid use. Patients who received anifrolumab had significantly higher rates of reaching the primary endpoint, with the greatest response seen in patients with a high IFN signature [37].

Type II IFN (IFN-γ) may also play a role in SLE pathogenesis. IFN-γ is secreted by both T cells as well as by macrophages and other innate immune cells, and induces upregulation of a distinct but overlapping set of genes as compared to type I IFN. A recent study examined IFN activity, autoantibodies, and cytokine levels in serum samples collected longitudinally before and after diagnosis of SLE [38]. In this cohort, 75% of patients showed elevated serum IFN-α activity prior to diagnosis of SLE. However, in all patients, development of autoantibodies preceded significant IFN-α activity. Interestingly, serum levels of IFN-γ and the IFN-γ-induced protein 10 (IP-10, also known as CXCL10) were elevated in SLE patients even before the development of autoantibodies [38]. This finding suggests that IFN-γ, more so than IFN-α, may influence the initial loss of tolerance and development of autoantibodies that characterizes early lupus pathogenesis. T cells from patients with SLE have previously been noted to produce more IFN-γ in response to stimulation, and this in turn promotes higher secretion of B cell activating factor (BAFF) by monocytes [39].

Both type I and type II IFN signaling pathways can be targeted with Jak inhibitors. Studies of tofacitinib in murine lupus models have been promising, and there are at least two phase I clinical trials in progress [40,41]. Jak inhibitors have also been used in the treatment of monogenic interferonopathies with reported success [42,43].

DNA damage repair and chromatin remodeling

The report of defects in DNA damage repair genes causing lupus-like disease has led to increased scrutiny of this pathway. Repair of DNA damage in response to oxidative stress is impaired in SLE neutrophils, while lymphoblastoid cells lines from SLE patients show defective repair of radiation-induced DNA damage [44,45]. A recent study similarly showed impairment of both nucleotide excision repair and double-strand break repair in SLE mononuclear cells treated with an alkylating agent [46]. Even in cells taken from SLE patients with quiescent disease, DNA damage repair was decreased compared to healthy controls, and the degree of impairment correlated with higher apoptotic susceptibility [46]. Interestingly, treatment with vorinostat, a histone deacetylase (HDAC) inhibitor, improved the efficiency of DNA damage repair; the authors hypothesize that vorinostat reverses the abnormal chromatin compaction in SLE that impedes access to sites of DNA damage [46].

Histone deacetylases may be therapeutic targets for other reasons. Epigenetic dysregulation is well documented in SLE, particularly global hypomethylation in T cells [47]. HDAC inhibitors ameliorate disease in multiple mouse models of lupus [48,49]. In vitro, HDAC inhibitors directly inhibit B cell proliferative responses to T cell activation and TLR4 stimulation [50]. Specific inhibitors targeting different classes of HDACs have variable effects on B cell differentiation and antibody production [50]. Treatment of lupus-prone mice with a non-specific HDAC inhibitor, panobinostat, dramatically reduced circulating naïve B and plasma cell numbers as well as autoantibody levels [50]. Interestingly, treatment of immunized mice with panobinostat did not affect the memory B cell compartment, suggesting that humoral autoimmunity might be treated with HDAC inhibitors while preserving B cell immunocompetence [50].

Rho kinases (ROCKs)

ROCK1 and ROCK2 are serine/threonine kinases activated by the GTPase RhoA. The ROCK kinases participate in multiple signaling pathways in both hematopoietic and non-hematopoietic cell types [51]. In T cells, ROCK2 can directly activate IRF4, a transcription factor necessary for Th17 differentiation [52]. Lupus-prone mice show increased ROCK2 activation, and treatment of these mice with fasudil, a ROCK inhibitor, ameliorated both Th17 dysregulation and their lupus-like disease [52]. ROCK activity in T cells is augmented by PP2A, which as discussed above is upregulated in SLE [7,53]. A recent study demonstrated that IL-17A and IL-21 production due increased ROCK activity in SLE T cells can be blocked using ROCK inhibitors [54]. The authors found that selective and non-selective ROCK inhibitors have overlapping but similar effects, concluding that selective ROCK2 inhibition may be sufficient to modulate the Th17-axis dysfunction seen in SLE [54]. KD025, a selective ROCK2 inhibitor, is in early phase clinical trials for several different immune-related conditions.

Bruton’s tyrosine kinase (BTK)

BTK, a critical factor in the B cell receptor signaling cascade, has long been an attractive therapeutic target in SLE. BTK signaling is necessary for B cell activation, and congenital deficiency of BTK results in agammaglobulinemia and immunodeficiency. As SLE is often thought of as an autoantibody-driven disease, it is perhaps not surprising that BTK inhibition decreases autoantibody production and ameliorates nephritis in multiple different murine models of lupus [55–58]. However, BTK also mediates signaling through the Fcγ receptor and toll-like receptors (TLR). The production of IFN-α by plasmacytoid DCs in response to nucleic acid is dependent on BTK signaling through TLR9 [59]. BTK inhibition may thus interrupt SLE pathophysiology in multiple ways. In a recent murine study of IFN-α-accelerated lupus nephritis, a BTK inhibitor was more effective than BAFF or Syk inhibition at decreasing autoantibody levels, reducing plasma cell numbers, and improving survival [60]. There is currently one BTK inhibitor approved for treatment of B cell malignancy; in autoimmunity, a phase 2 study of the BTK inhibitor LY3337641 is currently in progress for treatment of rheumatoid arthritis (clinicaltrials.gov, NCT02628028).

RAS

The role of RAS/MAP kinase signaling in SLE T cell dysfunction has been highlighted by several clinical associations. Ras is a small GTPase that exists in three isoforms – K-ras, N-ras, and H-ras. These enzymes are involved in many different signaling and cell cycle pathways, including mTOR and MAPK. Noonan syndrome is a neurodevelopmental disorder associated with various gene mutations in the RAS/MAPK pathway; some of these genes have also been implicated in SLE genome-wide association studies (GWAS), and there have been several case reports of SLE in Noonan syndrome patients [61]. Somatic gain-of-function mutations in RAS genes have been implicated in both malignancies as well as non-malignant lymphoproliferative syndromes. More recently, an unusual presentation of SLE in a young boy was also associated with a somatic gain-of-function mutation in KRAS [62].

Due to their role as oncogenes, the RAS GTPases have been an intensive area of interest in cancer therapy. However, targeted inhibition of these enzymes has proved challenging, and there are currently no approved RAS inhibitors on the market [63].

Conclusion

SLE is a heterogeneous disease and the underlying mechanisms may vary among patients, further increasing the complexity in understanding pathogenesis (Figure) [64]. With this complexity in mind, this review is not intended to be a comprehensive list of new findings in SLE. Recognition that monogenic forms of lupus may yield important insights in pathophysiology suggest that there are likely to be many more advances in understanding in the years to come, allowing further identification of potential targets for therapy.

Bullet Points.

• Defects in multiple metabolic pathways contribute to immune dysregulation in SLE

• Impaired DNA damage repair leads to lupus-like disease, in part by inducing type I IFN overproduction

• IFN signaling may be an effective therapeutic target in SLE

• Monogenic cases of lupus-like disease may inform further understanding of pathologic mechanisms in SLE.