Unraveling the Genetics of Shared Clinical and Serological Manifestations in Patients With Systemic Inflammatory Autoimmune Diseases

Abstract

Objective: Systemic inflammatory autoimmune diseases (SIADs) such as systemic lupus erythematosus (SLE), primary Sjögren disease (pSS), and idiopathic inflammatory myopathies (myositis) are complex conditions characterized by shared circulating autoantibodies and clinical manifestations, including skin rashes, among others. This study was aimed at elucidating the genetics underlying these common features.

Methods: We performed targeted DNA sequencing of coding and regulatory regions from approximately 1,900 immune-related genes in a large cohort of 2,292 well-characterized Scandinavian patients with SIADs with SLE, pSS, and myositis as well as 1,252 controls. A gene-based functionally weighted genetic score for aggregate testing of all genetic variants, including rare variants, was complemented by in silico functional analyses and in vitro reporter experiments.

Results: Case-control association analysis detected known and potentially novel genetic loci in agreement with previous genetic and transcriptomics findings linked to the SIAD autoimmune background. Intriguingly, case-case comparisons between patient subgroups with and without specific autoantibodies revealed that the subgroups defined by antinuclear antibodies and anti-double-stranded DNA antibodies have unique genetic profiles reflecting their heterogeneity. When focusing on clinical features, we overall showed that dual-specificity phosphatase 1 (DUSP1) protective genetic variants lead to increased gene expression and potentially to anti-inflammatory effects on the SIAD-associated skin phenotype. This is consistent with recent genetic findings on eczema and with the previously reported down-regulation of the MAPK signaling-related gene DUSP1 in other skin disorders.

Conclusion: Together, this suggests common molecular mechanisms potentially underlying overlapping clinical manifestations shared among different disorders and informs clinical heterogeneity, which could be translated to improve disease diagnostic and treatment, also in more generalized disease frameworks.

Figure 1

Bar plot showing the stratification of the different serological and clinical patient subgroups defined by the occurrence of autoantibodies and clinical features shared among patients with SIADs. Every subgroup is further categorized based on the prevalence of patients with SLE, pSS, or myositis. For additional numerical information, see Table 1 and Supplementary Table 1. ANA, antinuclear antibodies; dsDNA, double‐stranded DNA; pSS, primary Sjögren disease; RF, rheumatoid factor; SIAD, systemic inflammatory autoimmune disease; SLE, systemic lupus erythematosus; SM, Smith.

Figure 2

Heatmap showing the results of the genetic score–based aggregate case–control and case–case association testing for the subgroups of patients with SIAD defined based on the presence of shared autoantibodies. For each autoantibody indicated at the top, the results of the case–control analysis derive from the contrast between the patients with SIADs who were autoantibody positive and the control individuals, whereas the results of the case–case analysis derive from the contrast between the patients with SIADs who were autoantibody positive and those who were autoantibody negative (patients with missing data were not considered in the analysis). The number of samples of each group contrasted is shown under the autoantibody at the top. The results are further categorized into those pertaining genes located in the MHC locus and those located outside this region. The Bonferroni‐corrected P values were obtained after correcting the association raw P values from the logistic regression model for the 1,795 genes tested. ANA, antinuclear antibody; dsDNA, double‐stranded DNA; MHC, major histocompatibility complex; RF, rheumatoid factor; SIAD, systemic inflammatory autoimmune disease; SM, Smith.

Figure 3

Heatmap showing the results of the genetic score–based aggregate case–control and case–case association testing for the subgroups of patients with SIAD defined based on the presence of shared clinical manifestations and comorbidities. For each clinical feature indicated at the top, the results of the case–control analysis derive from the contrast between the patients with SIADs who were positive for the feature and the control individuals, whereas the results of the case–case analysis derive from the contrast between the patients with SIADs who were positive and negative for the feature (patients with missing data are not considered in the analysis). The number of samples of each group contrasted is shown under the feature at the top. The results are further categorized into those pertaining genes located in the MHC locus and those located outside this region. The Bonferroni‐corrected P values were obtained after correcting the association raw P values from the logistic regression model for the 1,795 genes tested. MHC, major histocompatibility complex; SIAD, systemic inflammatory autoimmune disease.

Figure 4

Genetic effects, in silico annotations, and reporter assay results for each DUSP1 variant detected in our study. (A) Zoonomia Project evolutionary constraint metric. (B) Delta allele frequency obtained as the minor (ie, effect) allele frequency difference between the patients with systemic inflammatory autoimmune diseases with and without skin involvement. (C) “Gene annotation”: regional annotation (pale green, upstream/downstream; green, intronic; dark green, exonic); “MAF group”: brown, common; yellow, rare; and “ORegAnno” and “cCREs”: gray, overlap with ORegAnno/ENCODE cCREs; white, no overlap. (D) The DUSP1 variants included in each of 18 enhancer regions tested with reporter assay are depicted with horizontal black lines and connected with dashed arrows to the corresponding region. (E) Expression levels of all 49 reporter allelic constructs constituting the 18 total tested enhancer regions. Reporter constructs are shown on the x‐axis and identified as SNV ID (only numerical) and allele for single‐variant construct or multiple SNV IDs (only numerical) and alleles for haplotype constructs. Alleles and construct orientation refer to the negative strand and are relative to the transcription start site, respectively. Noneffect and effect alleles refer to the major and minor alleles, respectively. For details, see Supplementary Materials and Methods. Unpaired Student's t‐test and one‐way analysis of variance were used for single‐variant and haplotype constructs, respectively. *P < 0.05, **P < 0.01, ***P < 0.001. cCRE, candidate cis‐regulatory element; DUSP1, dual‐specificity phosphatase 1; ENCODE, Encyclopedia of DNA Elements; ID, identifier; MAF, minor allele frequency; ns, nonsignificant; ORegAnno, Open Regulatory Annotation database; SNV, single‐nucleotide variant.

Figure 5

UCSC Genome Browser view (Hg38) of the DUSP1 locus. DUSP1 variants were detected and tested in this study (“DUSP1 variants detected in this study”). Nonsynonymous and synonymous variants are indicated in red and blue, respectively. The same color labels apply for the DUSP1 variants detected in the study by Grosche et al (“DUSP1 variants detected in Grosche study”). “Conserved TFBS (from Hg19)” are conserved TFBSs lifted over to GRCh38. “TFBSs (Andrews et al)” are TFBSs sourced from the Zoonomia project²¹. CTCF TFBSs are colored in red. “JASPAR 2022 TFBSs” are experimentally defined TFBSs for eukaryotes. “ENCODE cCREs” are ENCODE Registry of cCREs in the human genome. “ORegAnno” is the open regulatory region associated with active gene expression. “GeneHancer (double elite) interactions” are highly filtered interactions between regulatory elements and genes. The light blue interactions are associations between DUSP1 and cis‐regulatory elements. “GeneHancer (double elite) regulatory elements” are highly filtered regulatory elements. “In situ Hi‐C chromatin structure on HFFc6” is a zoomed‐in heatmap of chromatin folding data from in situ HFFc6 cell lines. “Zoonomia mammalian constraint” consists of mammalian evolutionary constraint phyloP scores (Christmas et al ²¹ ). cCRE, candidate cis‐regulatory element; e; ENCODE, Encyclopedia of DNA Elements; ORegAnno, Open Regulatory Annotation database; SNV, single‐nucleotide variant; TFBS, transcription factor binding site.