ObjectiveSystemic lupus erythematosus (SLE), an autoimmune disorder, has been associated with nearly 100 susceptibility loci. Nevertheless, these loci only partially explain SLE heritability and their putative causal variants are rarely prioritised, which make challenging to elucidate disease biology. To detect new SLE loci and causal variants, we performed the largest genome-wide meta-analysis for SLE in East Asian populations.MethodsWe newly genotyped 10 029 SLE cases and 180 167 controls and subsequently meta-analysed them jointly with 3348 SLE cases and 14 826 controls from published studies in East Asians. We further applied a Bayesian statistical approach to localise the putative causal variants for SLE associations.ResultsWe identified 113 genetic regions including 46 novel loci at genome-wide significance (p<5×10−8). Conditional analysis detected 233 association signals within these loci, which suggest widespread allelic heterogeneity. We detected genome-wide associations at six new missense variants. Bayesian statistical fine-mapping analysis prioritised the putative causal variants to a small set of variants (95% credible set size ≤10) for 28 association signals. We identified 110 putative causal variants with posterior probabilities ≥0.1 for 57 SLE loci, among which we prioritised 10 most likely putative causal variants (posterior probability ≥0.8). Linkage disequilibrium score regression detected genetic correlations for SLE with albumin/globulin ratio (rg=−0.242) and non-albumin protein (rg=0.238).ConclusionThis study reiterates the power of large-scale genome-wide meta-analysis for novel genetic discovery. These findings shed light on genetic and biological understandings of SLE.

Systemic lupus erythematosus (SLE), an autoimmune disorder, has been associated with nearly 100 susceptibility loci1-8. Nevertheless, these loci only partially explain SLE heritability and provide limited biological insight. We report the largest study of SLE in East Asians (13,377 cases and 194,993 controls), identifying 233 association signals within 113 (46 novel) genetic loci. We detect six new lead missense variants and prioritize ten most likely putative causal variants, one of which we demonstrate exhibits allele-specific regulatory effect on ACAP1 in vitro. We suggest 677 effector genes with potential for drug repurposing, and provide evidence that two distinct association signals at a single locus act on different genes (NCF2 and SMG7). We demonstrate that SLE-risk variants overlap with cell-specific active regulatory elements, notably EBNA2-mediated super-enhancers in Epstein-Barr Virus-transformed B cells, and implicate the role for non-immune cells in SLE biology. These findings shed light on genetic and biological understandings of SLE.

No abstract

SLE affects millions with increasing diagnostic prevalence. Genetic studies have identified &gt;600 mostly regulatory variant associations. EBV has been nominated as a causal factor for SLE from immunochemistry and epidemiologic studies. EBNA2 concentrates at SLE risk loci (Nat Genet 50:699 2018) suggesting mechanisms. Additional evidence supporting EBV as an SLE origin would advance this hypothesis of etiology. Previous simulation using our RELI algorithm revealed an astonishing association with Epstein-Barr virus nuclear antigen 2 (EBNA2) (OR=5.96, Pc=E-25) (Nat Genet 50:699, 2018). The &gt;100 risk alleles associated at p&lt;5E-8 for SLE were curated and reduced to 83 loci by linkage disequilibrium pruning. We evaluated the available 53 virally encoded and 13,051 human TF ChIP-seq (chromatin immunoprecipitation with DNA sequencing) datasets. In addition to EBNA2 (now OR=3.7, Pc=1.96E-18 after Bonferroni correction), two of the 53 viral TF data sets, EBNA3C and EBNALP, each of which are EBV Latency III gene products, were also strongly associated with the SLE risk loci (OR=6.0, Pc=4.19E-18; and OR=3.3, Pc=5.53E-18). Moreover, among the human TF datasets strongly associated with SLE risk loci (Pc&lt;E-6), those from EBV infected and transformed B cells are highly concentrated (OR≈56, P&lt;E-100) and, in addition, the human TFs that are known to form super-enhancers upon B cell transformation by EBV (e.g., NFκB subunits) also tend to be concentrated at the SLE risk loci (OR&gt;30, p&lt;E-25). Meanwhile, genetic associations for depression, anxiety, and schizophrenia show no such relationships. These new results further the possible validity of the hypothesis that SLE is originally caused by the Latency III expression program of EBV.

Objective IBD is a chronic inflammatory disorder of the GI tract with complex etiology that involves both genetic variants and environmental factors. Several viruses and bacteria have been suggested as potential causes of the disease. The clinical course of IBD appears to be influenced by EBV tissue infection. With a prevalence of &gt;90% in the adult human population, EBV infection is nearly ubiquitous. We reported that the EBV transcription co-factor EBNA2 was concentrated in the risk loci of IBD relative to the remainder of the genome (Pc=1.24E-13, (RR) = 3.33, Nature Genetics 50:699, 2018). The objective of this study was to determine whether new data and studies made available since the data used for our 2018 study further supported the possibility that EBV was related to some IBD as a potential etiologic agent. Methods Genomic approaches offer previously unavailable perspectives for understanding disease mechanisms. The previous analysis was based upon the 9 viral transcription factor and co-factor (TF) datasets available in human cells in 2015.Of the 52 now available there are 29 from EBV. The number of established IBD loci qualifying for analysis has expanded from 112 in 2015 to 324 available now. We applied simulation analysis to assess statistical significance of TF associations with IBD risk loci using RELI (Regulatory Element Locus Intersection) (Nat Genet 50:699, 2018). RELI tests the probability of the observed association by the count of intersections of the same variant from two sources. We extracted 700 SNPs in European ancestry from 93 published GWAS and 350 additional candidate gene studies for IBD. All variants with linkage disequilibrium r2&gt;0.8 of the best variant were included, while loci with r2&gt;0.2 with a more highly significant locus were excluded. Results Among the 52 viral TF datasets, 139 of the 324 IBD loci were occupied by EBNALP (RR=2.08793, Pc=3.7321E-23), 113 by EBNA3C (RR=2.35113, Pc=2.84678E-22), 88 loci by EBNA2 (RR=3.2, Pc=7.8E-32) and 28 loci by EBNA3ABC, which used an antibody that did not distinguish between EBNA3 subtypes. Analysis of 11,535 human TF ChIP-seq datasets produced 1578 with strong associations (p&lt;E-6) with IBD risk loci. The locus intersections of the three EBV gene products optimally clustered at 150 IBD loci (46%) with a set of 26 human TFs (p&lt;2E-308), thereby nominating these loci as potentially having EBV-dependent mechanisms that alter genetic risk for IBD. The human TFs were enriched for (ChIP-seq) data obtained from EBV-infected B lymphocyte cell line (OR = 10.12425329, Chi-squared = 241.2925605) Moreover, the human TFs that participate in the super-enhancers that form in the latency III EBV infected and transformed B cell are also concentrated among the statistically significant associations (p&lt;E-6) with the IBD risk loci (OR) = 2.9448, P = 10E-7). Conclusion These preliminary findings nominate EBV for a role in the pathogenesis and etiology of IBD by mechanisms operating in transformed B cells through the latency III EBV program of viral gene expression.