Genes and genetics in human SLE

Introduction

Many technological advances in the past two decades have led to an explosion of genome wide association studies (GWAS) and subsequent meta-analyses identifying close to 150 novel SLE risk loci across multiple ancestries and in some cases spanning multiple autoimmune diseases. A recent study of 27,574 individuals spanning European, African American and Hispanic Amerindian (HA) ancestries fine-mapped established SLE risk loci, characterized risk loci for ancestry dependent or independent contribution, and identified 24 novel risk loci. The pathogenesis of SLE is complex and the exact mechanisms remain undefined; however, pathways within the innate and adaptive immune responses have critical roles in disease development. In this chapter SLE susceptibility genes that surpass the GWAS significance level ( P < 5 × 10 ⁻⁸ ) will be discussed based on their role in these immunological pathways ( Fig. 11.1 ).

While the number of risk loci continues to grow, many loci identify large haplotypes containing multiple genes in strong linkage disequilibrium (LD), which makes it important to localize the underlying casual variants that confer disease risk and determine functional consequences of these gene variants. Therefore fine-mapping studies and functional analyses are imperative and help to determine variants that confer SLE disease risk, provide insight into how these genes affect phenotypic manifestations of disease, elucidate the pathogenic mechanisms of SLE, and identify possible targets for disease treatment.

A majority of SLE risk variants have been mapped to noncoding regions of the DNA, indicating a vital role for the regulation of gene expression through transactivating factors, post-translational modifications, or epigenetic changes to chromatin architecture. A large number of SLE risk genes encode transcription factors ( Fig. 11.1 ) that may affect expression levels of SLE risk gene products, resulting in the modulation of various pathways. Therefore SLE-associated genes that encode transcription factors will be specifically reviewed.

Finally, we discuss SLE risk genes that have become novel therapeutic targets and the utilization drug repositioning to identify alternative treatments for SLE. Innovative ideas on the molecular stratification of patients, personalized immunomonitoring and the precision of genomic medicine for the study and treatment of SLE are also discussed.

Transcription factors

An abundance of SLE risk genes encoding transcription factors have been identified in recent genetic studies that are expressed in both innate and adaptive cell types ( Fig. 11.1 ). One study identified 16 transcription factors encoding genes associated with SLE risk, highlighting the importance of immune cell regulation and implying that future studies focused on trans expression quantitative-trait loci (eQTLs) may be critical to understanding SLE pathogenesis. Yet another study identified SLE associated genes in or proximally located to 21 transcription factors that achieved GWAS significance of ( P < 5 × 10 ⁻⁸ ). It has been demonstrated that activation of transcription factors plays a role in disease development across multiple autoimmune diseases. Here we detail the genetic risk for some of the more well-known transcription factors and their roles in SLE pathogenesis.

Interferon regulatory factor genes are essential for activating transcription of type I IFN genes, and SLE associated genetic variants have been identified across multiple ancestries in or near IRF5 , IRF7 , and IRF8 . Two distinct genetic effects within SLE attributed to the IRF5 locus have been characterized by computational modeling. The first, by four variants within the IRF5 promoter was present across five ancestries and the second an 85.5 kb haplotype present only in Europeans spanning IRF5 and transportin 3 ( TPO3 ) genes. Increased binding of the ZBTB3 transcription factor to the risk allele in one of the variants of the IRF5 promoter may lead to increased IRF5 expression, suggesting it may be a causal variant for increased SLE risk in patients. Another region within the IRF5 promoter containing 3–4 (CGGGG) repeats confers increased susceptibility to apoptosis in monocytes, but not B cells, this appears to be due to increased transcription of IRF5 as a result of an additional binding site (the fourth CGGGG repeat) for the SP1 transcription factor. Integrating genetic data with eQTLs led to the identification of a single nucleotide polymorphism (SNP) located within the IRF7 haplotype, which confers not only a cis -eQTL effect on IRF7 expression, but also a trans -eQTL effect, regulating type I IFN responses in activated dendritic cells.

The relationship between genetic variants and the epigenetic effects of DNA methylation led to the discovery of methylation QTLs (meQTLs) associated with seven SLE GWAS risk loci ( PTPRC , MHC Class III, UHRF1BP1 , UBE2L3 , and transcription factors IRF5 , IRF7 , and IKZF3 ). It was determined that the downstream SLE associated SNP within UBE2L3 functions as a meQTL to a differentially methylated CpG site within the promoter of UBE2L3 , indicating that the epigenetic modulation of regulatory regions appears to play a role in SLE phenotype. Recently defined histone quantitative trait loci (hQTLs), three dimensional (3D), epigenetic post-translational modifications that effect genetic variants through the induction of allele-specific imbalances, are another epigenetic mechanism that affects gene regulation and is described further in the context of HLA genes and T cell signaling.

The STAT4 transcription factor plays a role in the IFN pathway, regulates T helper cell differentiation, mediates the IL-12 response in lymphocytes, and is associated with the presence of anti-dsDNA antibodies and renal disorders in SLE. The STAT1 - STAT4 locus is one of the more well studied loci in lupus. Fine mapping of the locus identified one variant (rs11889341) associated with increased SLE-associated risk across all major ancestries. Enhanced binding of the HMGA1 transcription factor to the causal variant located within the third intron of STAT4 was shown to decrease repressor activity leading to increased expression of STAT1 in Epstein–Barr virus transformed lymphoblastoid cell lines (LCLs) from lupus patients.

The Ets family of transcription factors is a large family of transcription factors that play a role in the immune response, bind a short DNA consensus motif, and are often functionally redundant. Three Ets family members have reported SLE risk associations ETS1 , FLI1 , and ELK1 , with ETS1 and FLI1 having been implicated in the pathogenesis of SLE and also implicated in genetic risk of other autoimmune diseases. Increased binding of the activated form of STAT1 (pSTAT1) to an SLE risk variant within ETS1 was observed, and analysis of eQTL loci indicate a resultant decrease in ETS1 expression within the Han Chinese ancestral cohort.

Additionally, three members of the Ikaros transcription factor family have been identified as SLE risk loci IKZF1 , IKZF2 , and IKZF3 . IKZF1 regulates lymphocyte differentiation, proliferation, and BCR signaling. The SLE risk variant of IKZF1 cis regulates IKZF1 levels, and trans regulates the expression of C1QB and five type I IFN response genes further illustrating that trans -eQTL analysis can generate insight into downstream effects of disease associated variants and the pathogenesis of SLE.

The SLE risk allele (T > C) associated with decreased transcription factor BLIMP1 expression in female SLE patients is located within the intergenic region between ATG5 and PRDM1 and creates a KLF4 transcription factor binding site, which together with the recruitment of histone deacetylases, suppresses the expression of BLIMP1 in monocyte-derived dendritic cells. In female mice, with a dendritic cell specific deletion of Prdm1 , IL-6 expression increases resulting in enhanced numbers of germinal center B cells and T follicular helper cells, leading to lupus-like disease. BLIMP1 is a transcriptional repressor of the cysteine protease Cathepsin S, which regulates the loading of peptides onto MHC class II molecules. In dendritic cells of female Prdm1 deficient mice, increased Cathepsin S expression led to altered antigen presentation, likely activating CD4+ T cells to differentiate into T follicular helper cells. In addition, Prdm1 deficient mice exhibit a more diverse Vβ repertoire in T follicular helper cells. A reduction in lupus-like disease was observed in Prdm1 deficient mice treated with a Cathepsin S inhibitor. The effects of BLIMP1 modulation have been observed primarily in female patients and mice and may contribute to the female sex bias in SLE.

The complex interaction between genes and the environment contribute to the heterogeneity of SLE. One mechanism by which this interaction may be facilitated is through the interaction of virus protein products and SLE risk loci. Epstein–Barr virus protein product EBNA2 was recently shown to interact with 26 out of the 50 SLE risk loci examined (including the transcription factors IRF5 , IRF8 , ELF1 , IKZF1 , IKZF3 , and JAZF1 ), colocalizing with several additional transcription factors including many of the NFκB subunits. EBNA2 binding to risk variants in loci such as CD44 , resulted in increased expression and a subsequent activation and migration of B cells. In addition, the HLA-B*08:01 allele identified through a fine-mapping study as potentially causal binds to EBNA2 and complexes with several transcription factors.

Collectively, the considerable number of transcription factors encoded by SLE risk associated genes and their extensive involvement in the regulation of gene expression networks across multiple ancestries and innate and adaptive immune response highlight the importance of understanding their role in the pathogenesis of SLE.

Clearance of apoptotic cells and immune complexes

Inefficient clearance of apoptotic cells and immune complexes (ICs) can lead to the accumulation of self-antigens and organ deposition during the development of SLE, may initiate the autoimmune response, and promote chronic inflammation. Genetic defects in the classical complement pathway, FCγRs, and other genes associated with apoptosis contribute to both monogenic and polygenic forms of SLE. Associations with highly penetrant single-gene mutations result in the complete deficiency of classical complement genes, such as C1Q and C4 , have been shown in early-onset juvenile SLE. Decreased copy number variations (CNVs) of FCGR3B and missense mutations of FCGR2A/FCGR3A also play a role in SLE. Neutrophil cytosolic factor 1 ( NCF1 ), which is a subunit of phagocyte NADPH oxidase 2 (NOX2), is involved in reactive oxygen species (ROS) production, phagocytosis of apoptotic cells, and autophagy, also exhibits a drastic CNV effect. Whereby having multiple copies of NCF1 (≥3) was protective against developing SLE and possessing only one copy of NCF1 increased the risk of SLE in Koreans, Chinese, and European Americans, suggesting that decreased production of ROS production is an SLE risk factor. NOX2-derived ROS have been recently shown to play a key role in the phagosomal disposal of apoptotic cells to limit crosspresentation of autoantigens.

Several studies have demonstrated increased risk of lupus nephritis (LN) associated with patients carrying ITGAM risk alleles. ITGAM , encodes CD11b, which combines with CD18 to form complement receptor 3 and is responsible for the phagocytosis of complement coated particles and ICs and regulates leukocyte adhesion, migration, and apoptosis through interactions with another SLE associated gene, ICAM1 . One of the SNPs within ITGAM results in a missense amino acid change (Arg to His R77H) that causes impaired phagocytosis of complement opsonized targets in monocytes, neutrophils, and macrophages, which may result in tissue damage due to altered IC clearance and deposition.

Autophagy

Dysfunction in the lysosomal degradation of intracellular components via distinct methods macroautophagy, microautophagy, chaperone mediated autophagy, and a noncanonical form LC3-associated phagocytosis (LAP) may play a role in the development of many diseases, including SLE. Processes associated with autophagy can affect the immune response through innate immune signaling (cytokine secretion and clearance of microbes), removal of damaged mitochondria producing ROS, antigen presentation, and leukocyte development. Several SLE risk genes are also associated with the various autophagy pathways including ATG5 , ATG16L2 , WDFY4 , HIP1 , CDKN1B , DRAM1 , CLEC16A , PRKCD , NCF1 , and NCF2 . Several polymorphisms associated with SLE susceptibility have been identified within ATG5 , which encodes a critical component of both macroautophagy and LAP pathways. Mice deficient in Atg5 and other components of the LAP pathway exhibit significantly increased inflammatory cytokine levels (IL-1β, IL-6, IP-10, and MCP-1) in addition to other autoimmune characteristics and develop lupus-like disease. ATG16L2 is a risk locus identified in Asians, and its encoded protein ATG16L2 has little known function, but its homolog ATG16L1 physically interacts with ATG5 . Multiple GWAS studies have previously associated WDFY4 with SLE susceptibility in Europeans and Asians, but until recently no immune function had been defined. Gene silencing of the SLE risk gene WDFY4 , resulted in an increase in the autophagy biomarkers LC3-II, PIK3C3, and p62 and a larger number of autophagosomes in B cells, suggesting a decrease in autophagy. Furthermore, a decrease in most B cell populations, an impaired antibody response to antigens and milder disease in Wdfy4 knockout mice, suggests that WDFY4 may modulate B cell fate through the LAP pathway. The likely causative variants of two other SLE risk genes which modulate autophagy and have striking effect sizes (OR > 2.5), NCF1 (Arg90His) and NCF2 (His389Gln) and (Arg395Trp) encode two separate regulatory subunits of the phagocytic NOX2 complex, which is essential to LAP. In the case of NCF2, multiple other variants may confer changes associated with disease. A decrease in required ROS production, in some cases NOX2-derived, which prevents a LAP associated autoinflammatory lupus-like response in dying cells, has been associated with both NCF1 and NCF2 risk alleles. Continued research into the specific mechanisms these SLE risk loci are utilizing to affect autophagy in SLE patients is needed to further understand how they contribute to the pathogenesis of SLE.