Genes and Immunity (2013) 14, 133–146 & 2013 Macmillan Publishers Limited All rights reserved 1466-4879/13 www.nature.com/gene

REVIEW The applied basic research of systemic lupus erythematosus based on the biological omics

W Sui1,3, X Hou1,2,3, W Che1, M Yang1 and Y Dai1

Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterized by the production of autoantibodies directed against nuclear self-antigens and circulating immune complexes. This results in damages to various organs or systems, including skin, joints, kidneys and the central nervous system. Clinical manifestations of SLE could be diverse, including glomerulonephritis, dermatitis, thrombosis, vasculitis, seizures and arthritis. The complicated pathogenesis and varied clinical symptoms of SLE pose great challenges in the diagnosis and monitoring of this disease. Unfortunately, the etiological factors and pathogenesis of SLE are still not completely understood. It is noteworthy that recent advances in our understanding of the biological omics and emerging technologies have been providing new tools in the analyses of SLE, such as genomics, epigenomics, transcriptomics, proteomics, metabolomics and so on. In this article, we summarize our current knowledge in this field for a better understanding of the pathogenesis, diagnosis and treatment for SLE.

Genes and Immunity (2013) 14, 133–146; doi:10.1038/gene.2013.3; published online 28 February 2013 Keywords: systemic lupus erythematosus; biological omics; pathogenesis

INTRODUCTION executor of physiological function. Posttranslational modifications Systemic lupus erythematosus (SLE) is an autoimmune systemic (PTMs) alter many aspects of protein chemistry, including primary disease, and 90% of the cases of SLE occur in women. It is and tertiary structure, biological (and/or enzymatic) functions and characterized by the presence of large quantities of autoantibo- proteolytic degradation that may be critical in generating dies to components of the cell nucleus. These autoantibodies and immunogenic or tolerogenic self-peptides, which also play an some self-reactive T cells contribute directly to the pathologic important role in the initiation and propagation of SLE. changes in SLE. The generation of autoantibodies is the result of a In summary, the pathogenesis of SLE probably involves multiple cascade of immune dysregulation, including inappropriate regula- factors, and each of these factors acts both independently and tion of hyperreactive B and T cells, loss of immune tolerance and concurrently, and may differ between individuals. Thus, a large defective clearance of apoptotic cells and immune complexes in body of work has been done, but the mechanisms of SLE the presence of an appropriate genetic predisposition. The role of pathogenesis are still not completely understood and the cause of genetic variation has been extensively investigated in the lupus, the diversity of symptoms is not full clear. Until now, renal biopsy revealing polymorphisms throughout the genome correlated with remains the ‘gold standard’ for assessing disease severity, but disease. Although studies investigating common genetic variants multiple biopsies to gauge treatment efficacy are not feasible in SIE have revealed a number of susceptibility loci, not all carriers because of their invasive nature. Development of biomarkers for a develop clinical disease, and patients with the same or similar deeper understanding of SLE pathogenesis and improving alleles might present with different disease course. As a complex diagnosis, prognosis, and treatment is still one of the main goals human disease, it may broadly reside in epigenetic mechanisms. and challenges in SLE research. Alternative biomarker discovery Epigenomics describes changes in gene expression not linked approaches include proteomics, metabolomics and so on that may to alterations in the underlying genomic sequence, whereas help to find new biomarkers that can be used to more reliably and environmental factors, such as medications, hormones, stress, securely confirm SLE or monitor their progression with high infections, cigarette exposure, UV exposure and toxicants, can sensitivity and specificity. In this paper, our goal is to provide a modify epigenetic marks, influencing disease onset and progres- brief overview of current research of SLE and to introduce some of sion.1–4 Moreover, transcription factors (TFs) play a central role in the key discussion points in terms of understanding of lupus regulating gene expression and transcription. Indeed, there is pathogenesis and providing new targets for therapy. growing evidence showing that there is a significant correlation between SLE and TFs. In addition, RNA editing and alternative splicing can therefore correct, extend or diversify the information GENOME RESEARCH OF SLE encoded within the corresponding genomic sequence, and During the past decades, linkage studies and candidate gene translational control is an important regulatory mechanism for studies have assessed many genes for potential roles in growth-related genes and cytokine expression. Protein is the predisposing to SLE. Until now, the number of confirmed genes

1Nephrology Department, Guilin 181st Hospital, Guangxi Key Laboratory of Metabolic Diseases Research, Guilin, Guangxi, People’s Republic of China and 2College of Life Science, Guangxi Normal University, Guilin, Guangxi, People’s Republic of China. Correspondence: Professor Y Dai, Clinical Medical Research Center, Second Clinical Medical College of Jinan University (Shenzhen People’s Hospital), Shenzhen, Guangdong, 518020, People’s Republic of China. E-mail: [email protected] 3These authors contributed equally to the work. Received 1 December 2012; revised 16 January 2013; accepted 17 January 2013; published online 28 February 2013 Applied basic research of SLE W Sui et al 134 predisposing to SLE is over 30 (Table 1), and many candidate HLA association in non-Caucasian populations is less well genes have been identified based on their location or possible established. pathogenic roles. Specific characteristics of the human leukocyte antigen (HLA) region, as well as complement factor deficiencies, HLA class III and the complement system. In the class III gene, may facilitate nuclear antigen presentation, thereby triggering mutS homolog 5 (MSH5) gene has been associated with SLE, and autoantibody production. The genetic polymorphism of cytokines the strongest association was confirmed in a genome-wide may conduce to deregulate lymphocyte activity. Furthermore, the association study by Harley et al.6 Superkiller viralicidic activity polymorphism of the Fc receptors of immunoglobulins may affect 2-like (SKIV2L), is another class III gene for potential roles in SLE, immune complex clearance, thereby promoting tissue damage. and a study of 314 trios from the United Kingdom implicated this locus independent of class II variants.7 The integrin a M (ITGAM) gene has also been associated with SLE in a number of studies.6,8 Major histocompatibility complex Additionally, it is notable that the strongest single genetic risk The extended major histocompatibility complex is a gene-dense, factors of SLE are complement defects. For instance, 90% of transcriptionally active, 7.6-Mb interval on chromosome 6p21.3. It homozygous C1q-deficient individuals develop SLE.9 C1q comprises the classical HLA class I (HLA-A, -B and -C) and class II participates in the clearance of apoptotic cells, and thus has a regions (HLA-DR, -DQ and -DP) that encode the genes involved in role in the maintenance of immune tolerance. A hierarchy of antigen presentation. This region is highly polymorphic and, not susceptibility among the absence of classical complement surprisingly, has been associated with most of the autoimmune, component (C1q4C44C2, in decreasing order of risk for SLE inflammatory and infectious diseases. susceptibility) has been suggested.10 The C4 fraction of complement is encoded by two class II genes, C4A and C4B. HLA class II. Many studies, the earliest of which were done more Complete C4 deficiency is associated with a 70% risk of SLE than 20 years ago, have shown that SLE is associated with HLA development. Lower copy number of C4 is a risk factor for SLE, and higher gene copy number of C4 is a protective factor against class II haplotypes involving the HLA-DRB1 and HLA-DQB1 loci; in 11 particular, haplotypes bearing the DRB1*1501/DQB1*0602 (DR2) SLE disease susceptibility. and DRB1*0301/DQB1*0201 (DR3) alleles have been associated with SLE in Caucasian populations, and studies suggest that these IgG Fc receptor (FCGR) alleles confer an overall two- to threefold increased risk for SLE.5 Fc g-receptors (FcgRs) are members of the immunoglobulin Class II HLA specificities seem associated mainly with specific superfamily that recognize and bind the constant (Fc) portion of autoantibody profiles. In general, compared with Caucasians, the specific monomeric immunoglobulin G (IgG) and IgG-containing

Table 1. List of selected candidate genes

Candidate genes Full name Region Variation Reference

C1q Complement component 1, q subcomponent 1p36.12 Deletion 9,10 FCGR2A Fc gamma receptors class IIA 1q23.3 rs1801274 13 FCGR3A Fc g receptors class III A 1q23.3 F176V 12,21 FCGR3B Fc g receptors class III B 1q23.3 Copy-number variation (CNV) 22 TNFS4 TNF superfamily, member 4 1q25.1 rs2205960 46,47 PTPN22 Protein tyrosine phosphatase non-receptor type 22 1p13.2 rs2476601 66–68 CRP C-reactive protein 1q32.2 rs3093061 58,59 IL10 Interleukin 10 1q31.1 rs3024505 47 RASGRP3 RAS guanyl-releasing protein 3 2p22.3 rs13385731 47 STAT4 Signal transducer and activator of transcription 4 2q32.2 rs7574865 6,46,47 PDCD1 Programmed cell death 1 2q37.3 PD1.3G/A 50,51 PXK PX domain containing serine/threonine kinase 3p14.3 rs2176082 6,47 BANK1 B-cell scaffold protein with ankyrin repeats 1 4q24 rs10516487 47 TNIP1 TNFAIP3 interacting protein 1 5q33.1 rs7708392 46,47 PTTG1 Pituitary tumor-transforming 1 5q33.3 rs2431099 6,47 HLA-DRB1 MHC, class II, DRb1 6p21.3 Haplotypes 5,6,47 ATG5 Autophagy-related 5 homolog 6q21 rs2245124 6,46,47 PRDM1 PR domain containing 1, with ZNF domain 6q21 rs6568431 6,46,47 C2 Complement component 2 6p21.32 Deletion 10 C4 Complement component 4 6p21.32 CNV 10,11 TNF-a Tumor necrosis factor-a 6p21.31 rs361525 38,39 UHRF1BP1 UHRF1-binding protein 1 6p21.31 rs11755393 47 TNFAIP3 TNF-a-induced protein 3 6q23.3 rs6920220 44–46 IKZF1 IKAROS family zinc-finger 1 7p12.2 rs4917014 46,47 JAZF1 Juxtaposed with another zinc-finger gene 1 7p15.1 rs849142 47 IRF5 Interferon regulatory factor 5 7q32.1 rs10954213 29 BLK B-lymphoid tyrosine kinase 8p23.1 rs2736340 6,46,47 WDFY4 WDFY family member 4 10q11.22 rs1913517 46 IRF7 Interferon regulatory factor 7 11p15.5 rs4963128 6,33,47 DDX6 DEAD box polypeptide 6 11q23.3 rs503425 46,47 ETS1 V-ets erythroblastosis virus E26 oncogene homolog 1 11q24.3 rs6590330 46 SLC15A4 Solute carrier family 15, member 4 12q24.32 rs1385374 46,47 ITGAM Integrin a M 16p11.2 rs1143679 6,8,47 UBE2L3 Ubiquitin-conjugating enzyme E2L 3 22q11.2 rs5754217 6,46,47 MECP2 Methyl CpG-binding protein 2 Xq28 rs2269368 47 CSK C-src tyrosine kinase 15q24.1 rs34933034 71

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited Applied basic research of SLE W Sui et al 135 immune complexes. In humans, the FcgR genes are clustered on role in the development of autoimmune processes. More than 32 the long arm of chromosome 1q21.1–24,12 and the classical FcgR years ago, Hooks et al.23 found IFN-g in the sera of patients with family is divided into three receptor families, that is, FcgRI (CD64), SLE and showed a good correlation between immune IFN-k titers FcgRII (CD32) and FcgRIII (CD16), based on structural homology. and disease activity. It is likely that excessive induction of IFN-g gene expression upregulates IgG production by mononuclear cells 24 FcgRlls.FcgRlls play an important role in the clearance of in patients with SLE, which may enhance disease development. immune complexes, whereas their function in SLE may be It is also noteworthy that many IFN-regulatory factors are also impaired, evidence that nonsynonymous G-to-A variant in the strongly associated risk factors for SLE because they can induce 25 FCGR2A gene (rs1801274) results in a single amino-acid difference transcription of IFN-a mRNA. Graham et al. identified that the at position 131 (R131 and H131) in the second extracellular Ig-like rs2004640 SNP of IRF5 is a risk factor for SLE. Further support for domain of the FcgRIIa protein.13 This allelic difference alters this finding comes from a genome-wide association study in 26–28 recognition of ligand. The FcgRIIA-H131 (histidine residue at individuals of European, African and Asian ancestry. Lo¨fgren 29 position 131) allele is able to bind IgG2 effectively, whereas the et al. recently genotyped IRF5 using the CGGGG insertion/ R131 (arginine residue at position 131) binds less efficiently to deletion and the rs2004640, rs2070197 and rs10954213 SNPs in IgG2 and might delay clearance of IgG2-containing immune 1488 SLE patients and 1466 healthy controls. They showed that complexes.12 Moreover, a single-nucleotide polymorphism (SNP) rs10954213 is the main SNP responsible for altered IRF5 corresponding to isoleucine (I) to threonine (T) at the FcgRIIb expression in the peripheral blood mononuclear cells (PBMCs). 30 residue 232 (also known as I/T 187 excluding the signal peptide) in Hikami et al. found that a functional polymorphism in the the transmembrane domain that may alter the B-cell receptor 30-untranslated region (UTR) of SPI1, which is known to regulate 31,32 signaling has been associated with SLE in the Chinese, Japanese expression of IRF2, IRF4 and IRF8, plays a potential role in and Thai populations,14–16 but not in the Blacks and Whites in the predisposing to SLE. The IRF7 gene can also stimulate induction of 33 6 United States and the Swedish Whites.17,18 Therefore, the FCGR2B IFN-a RNA. Harley et al. reported that there was a significant I/T 232 may be a risk factor for SLE in Asians but not in other association between SLE risk and locus between IRF7 and the studied populations. In 2004, a study by Su et al.19 showed a PHRF1 gene in a European population. Moreover, recent studies promoter haplotype that alters FcgRIIb promoter activity. The less have shown that the risk allele of STAT4 was significantly frequent promoter haplotype ( À 386C À 120A) showed increased associated with increased sensitivity to IFN-a signaling in lupus 34 promoter activity and drove higher receptor expression in both patients. SLE patients who carry the STAT4 risk variant have transfected cell lines and cells ex vivo from genotyped donors increased expression of downstream IFN-I-regulated genes in vivo 35 than the more frequent haplotype ( À 386G À 120T).19,20 The less as compared with patients who do not carry the allele. It was frequent and more active promoter haplotype was associated with also reported that rs7574865, a variant allele of STAT4, is strongly SLE in a Caucasian population with an odds ratio of 1.6.19 associated with SLE characterized by double-stranded DNA autoantibodies.34 FcgRIIIs.FcgRIIIa (CD16) is expressed on cell surfaces of natural Tumor necrosis factor-a (TNF-a) gene is located on chromo- some 6 (6p21.31), within the class III region of major histocompat- killer cells, monocytes and macrophages, and it binds to both IgG1 36 and IgG3 subclasses. A T–G polymorphism results in phenylalanine ibility complex. TNF-a is a ubiquitous cytokine that plays an (F)–valine (V) at amino-acid 176.12 Individuals homozygous for F–F important role in various physiologic as well as pathologic processes such as inflammation, immunoregulation, proliferation bind IgG1 and IgG3 less efficiently than those with V–V genotypes, 37 suggesting less efficient clearance of IgG1- or IgG3-containing and apoptosis. It has two differing actions in SLE. On one hand, immune complexes.12 A recent meta-analysis of 41000 subjects TNF-a could be an immunosuppressive mediator of autoantibody in each of the three categories (lupus nephritis (LN), SLE without synthesis, and on the other hand, TNF-a might be a proinflammatory factor acutely released in the local tissues. renal involvement and non-SLE controls) has concluded that the 38 176 F allele confers a 1.2-fold increased risk for developing LN in Several SNPs have been identified in the TNF-a promoter. Among these, a G-to A-substitution at position À 238 (TNF-a patients of European, African and Asian descent but not for SLE 39 susceptibility per se in the absence of nephritis.21 Furthermore, À 238 G4A, rs361525) has been associated with SLE, especially three different allotypic variants of FCGR3B, NA1, NA2 and SH, in the Caucasian population. Another SNP of TNF-a promoter is have been identified through serological studies. The six SNP the G-to-A substitution at position À 3008 (TNF-a À 308 G4A, differences underlying these three serologic allotypes include five rs1800629). A recent meta-analysis of TNF-a promoter À 308A/G polymorphism has concluded that the A allele contributed to nonsynonymous SNPs and one synonymous SNP. The five amino- 40 acid changes are all in the first extracellular domain of FcgRIIIb. susceptibility to SLE in Caucasians but not in Asians. It is also Among these, the amino-acid 65 change results in a loss of a noteworthy that TNF-a is a pleiotropic inflammatory cytokine 22 whose effects are mediated through two distinct cell surface glycosylation site in the NA2 allele. The NA2/NA2 variant has 41 been shown to confer reduced phagocytic capacity of neutrophils receptors, TNF-RI (TNFRSF1A) and TNF-RII (TNFRSF1B). as compared with the NA1/NA1 genotype and is possibly A polymorphism at position 196 (a methionine-to-arginine substitution; M196R) of TNFRSF1B was significantly associated associated with thrombocytopenia in SLE. More recently, a copy 42 43 number variation in FcgRIIIb has also been associated with SLE, it with an increased risk of SLE. Moreover, Graham et al. utilized was also reported that increased copy number (CN) was protective family-based and case–control approaches to identify a risk allele and decreased CN was a risk factor. upstream of TNFS4, which is responsible for increased TNFSF4 expression and predisposes to SLE. Moreover, a number of polymorphisms in TNF a-induced protein 3 (TNFAIP3) have also T-helper cell (Th) cytokines been associated with increased susceptibility to SLE.44,45 TNFAIP3- Interferon-a (IFN-a) is a pleiotropic type I IFN with the potential to interacting protein 1 (TNIP1) is also identified by genome-wide break immunologic self-tolerance by activating antigen-present- association study as significantly correlated with SLE in the ing cells after uptake of self-material. Moreover, IFN-a promotes European and Chinese populations.26,46 long-term antibody production, class switching and immunologi- Moreover, IL10 is also an attractive positional candidate gene as cal memory. Several sets of compelling data suggest an important it maps in 1q32 because it is a potent stimulator of B cells and has pathogenic role for IFN-a in SLE, and papers published as early as a direct effect on B-cell survival and autoantibody production. 1979 have described increased serum levels of IFN-a in patients B cells and monocytes of SLE patients produce an increased with SLE, particularly those with active disease. IFN-g plays a key amount of interleukin-10 (IL-10) compared with nonaffected

& 2013 Macmillan Publishers Limited Genes and Immunity (2013) 133 – 146 Applied basic research of SLE W Sui et al 136 individuals, and many lines of evidence suggest interleukin-10 loci.6,46,47 These include genes encoding proteins important for production level to the SNP haplotypes. SNPs located more distally the biological role of Toll-like receptor/IFN-I and lymphocyte in the 50 flanking region were tested for association with SLE. A signaling in the pathogenesis of SLE (UBE2L3, IKZF1, MECP2, ATG5, genome-wide association replication study by Gateva et al.47 PRDM1 and ETS1) and proteins with roles in lymphocyte confirmed that rs3024505, a variant allele of IL10, is strongly development and signaling (BLK, RASGRP3, BANK1), and proteins associated with SLE in individuals of European ancestry participating in other/not known immune function (JAZF1, PXK, (P ¼ 3.95 Â 10 À 8). Herein, it would also be necessary to show PTTG1, LRRC18-WDFY4, DDX6, SLC15A4, UHRF1BP1). In summary, that some other genes encoding IL protein have also been the rapidly growing catalog of loci associated with SLE represents associated with SLE, including IL12 and IL18,48,49 which are a key early step in our understanding of this disease; however, the involved in the onset and progression of SLE. vast majority of the genetic burden of this disease has yet to be discovered.

Other susceptibility genes The programmed cell death 1 (PDCD1) gene located within 2q37 EPIGENOMICS RESEARCH OF SLE encodes an inhibitory immunoreceptor of the CD28/CTLA4/ICOS 50 Regardless of a growing number of known predisposing alleles, family that has a pivotal role in peripheral tolerance. Pdcd1 À / À not all carriers develop clinical disease. Patients with the same or mice have been shown to develop arthritis and lupus-like similar alleles might present with different disease course, and the glomerulonephritis. A recent study of 2500 individuals has shown prevalence of SLE in monozygotic twins is 25–45%.72 Thus, association between an intronic SNP in PDCD1 and SLE environmental factors appear to influence disease onset, susceptibility, conferring a 2.6-fold increased risk to Europeans 73 51 progression and outcome. Epigenetic modifications can and a 3.5-fold increased risk to Mexicans. Indeed, PDCD1 has in influence gene expression, thereby altering cellular function many studies been shown to be associated not only with SLE, but 52–55 without modifying the genomic sequence in response to a also rheumatoid arthritis, type I diabetes and multiple sclerosis. variety of environmental stimuli. By defining the set of genes that C-reactive protein (CRP), a pentraxin, is an innate immune are competent for expression and the set of genes that are modulator that facilitates the clearance and handling of cellular 56 repressed, the character of the cell is determined largely by the debris and apoptotic bodies. It is an important liver-derived epigenomics.73 Three main epigenetic processes are in the center acute-phase protein that can increase up to 1000-fold in serum as of epigenetic control: DNA methylation, nucleosome repositioning a response to diverse stimuli such as infection or injury. Some by histone modifications and microRNAs (miRNAs).74 studies have suggested that SLE is characterized by lower CRP levels than would be predicted.57 Further studies by Edberg and other investigators58,59 have shown that CRP levels are influenced DNA methylation by genetic variation in the CRP promoter. In multiple independent The most widely researched epigenetic modification to date is study populations, including both African-Americans and DNA methylation. Cytosine methylation of the regulatory Caucasians, the variation in the CRP promoter at CRP-707 sequences of DNA is associated with transcriptional inactivation (rs3093061) strongly and reproducibly associates with the SLE of genes, whereas hypomethylation contributes to the activation phenotype. of transcription. Accumulating evidence indicates that lympho- PTPN22 gene encodes the cytoplasmic lymphoid-specific cytes of SLE patients exhibit globally hypomethylated DNA.75,76 phosphatase (Lyp). A functional PTPN22 1858C4T (R620 W) DNA fragments isolated from the plasma of SLE patients contain polymorphism (rs2476601) confers significant risk for autoimmune B1% of methylcytosine, whereas genomic DNA of healthy disease. Underlying this association, there are several different individuals contains 2.5–4.5% of methylcytosine. Variable mechanisms with different evidences.60–65 Some groups have methylation of ribosomal genes was reported whereby 18S and established the possibility that the R620W amino-acid change 28S ribosomal subunit genes are hypomethylated in lupus disrupts binding of Lyp to Csk,60,61 thereby disturbing the individuals of discordant pairs and hypomethylated in both regulation of the T-cell receptor (TCR)-signaling kinases, Lck, Fyn individuals of concordant lupus twin pairs examined. and ZAP-70. Studies by Menard et al.64 revealed that the Remarkably, the PBMCs of SLE patients display a lower transcript association strength of PTPN22 for autoimmunity may be due to level and enzymatic activity of DNMT1 (DNA (cytosine-5)- not only impaired removal of autoreactive B cells, but also methyltransferase 1) than PBMCs isolated from the blood of upregulation of genes such as CD40, TRAF1 and IRF5. Moreover, healthy individuals, which is further verified by the report that Zikherman et al.65 demonstrated that PTPN22 deficiency demethylating agents, including 5-azacytidine, procainamide and cooperates with the CD45 E613R genetic background to hydralazine, can induce lupus-like features in both animal models provoke autoantibody production and a lupus-like autoimmune and human T lymphocytes.74,77,78 disease. Genetic studies with different populations revealed a A generally hypomethylated state of T- and B-lymphocyte different frequency of the disease-associated 1858T allele in genes has been reported in SLE patients. It is thought that the Europe,66–68 but not associated with SLE in the Norwegian and overexpression of these methylation-sensitive genes leads to Turkish populations.69 In contrast to the R620W SNP, a rare T-cell autoreactivity and subsequent induction of autoreactive missense substitution (R263Q) in PTPN22 was shown to reduce B-cell immunoglobulin production. It is noteworthy that Jeffries phosphatase activity and was associated with protection from et al.79 utilized high-throughput methylation arrays to analyze the SLE.70 In addition, Manjarrez-Orduno et al.71 recently reported an methylation status of over 27 000 CpG sites in promoter regions of association of CSK with SLE and refined its location to the intronic nearly 15 000 genes for 12 female lupus cases and 12 controls. polymorphism rs34933034 (odds ratio ¼ 1.32; P ¼ 1.04 Â 10 À 9). They found hypomethylation of 236 CpG sites and hypermethylation The risk allele at this SNP is associated with increased CSK of another 105 sites. The hypo-methylated genes, including expression and augments inhibitory phosphorylation of Lyn. In ADAMTS1, ALX4, CD9, ESRRA, FGF8, HOXA13, HOXD11, MMP9, MSX1, carriers of the risk allele, there is increased B-cell receptor- PDGFRA and SOX5, are involved in connective tissue development, mediated activation of mature B cells, as well as higher whereas the hyper-methylated genes are involved in folate concentrations of plasma IgM relative to individuals with the biosynthesis. nonrisk haplotype. Although the mechanisms for the occurrence of DNA hypo- Moreover, it is noteworthy that recent genome-wide association methylation in patients with active lupus are not fully addressed, and candidate gene studies have identified many additional risk there are three possible mechanisms account for this. One

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited Applied basic research of SLE W Sui et al 137 possibility is that GADD45a is suspected to promote demethyla- thereby regulating access of the transcription complex. In recent tion by interacting with MBD-4 and AID78,80 (Figure 1). A second years, increasing evidence has suggested that histone possibility is that the mitogen-activated protein kinase/extracel- modifications are related to the pathogenesis of SLE. Among lular signal-regulated kinase pathway can regulate DNMT these, histone lysine methylation is one of the most important levels.74,78 In addition, it is important to note that the TF RFX1 epigenetic modifications. Aberrant alterations in histone lysine plays an important role in the molecular mechanisms. RFX1 binds methylation patterns that change chromatin structure could lead to promoter regions of target genes and recruits the corepressors to dysregulated gene transcription and disease progression. DNMT1 and histone deacetylase 1, consequently suppressing Dai et al.83 adopted chromatin immunoprecipitation linked to target genes. However, in lupus T cells, decreased levels of RFX1 at microarrays (ChIP-chip) technology to profile and compare the CD11a and CD70 promoters leads to decreased methylation and variations in H3K4me3 at the genome-wide level of PBMCs from increased expression of both of these genes in lupus CD4 þ T cells SLE patients and healthy controls. They found that two key via reduction in DNMT1 and histone deacetylase 1 recruitment, relevant genes PTPN22 and LRP1B showed a significant decrease which results in lupus-like T- and B-cell hyperactivity.81 in histone H3K4me3 in active and inactive SLE patients as compared with healthy subjects. In addition, histone acetylation is associated with competence for transcription and histone Histone modifications acetyl transferases are recruited by many TFs. H4 acetylation, for Histone complexes possess flexible N-terminal tails that are example, participates in the transcriptional regulation of inducible accessible to posttranslational modifications and strongly affect genes.84 Zhang et al.85 found that H4 acetylation was significantly the functional capacities of nucleosomes.74 They include not only increased in monocytes from patients with SLE as compared with acetylation and methylation of the protruding histone tails but controls. also phosphorylation, ubiquitylation, sumoylation, citrullination, It appears that increased histone modification is a double- ADP ribosylation and proline isomerization.82 Each histone tail edged sword. For instance, acetylation of histone tails leads to an modification occurs on specific residues and acts globally to open conformation of chromatin that is permissive to gene tighten or relax the chromatin surrounding a particular gene, expression. Mice with a B cell-specific knock-in mutant version of p300 that lacks only histone acetyltransferase activity develop a lupus-like autoimmune disease characterized by splenomegaly, nephritis and vasculitis.86 On the other hand, triacetylated histone H4 is capable of modulating disease in MRLlpr mice, enhancing mortality, proteinuria, skin lesions and glomerular deposition of antibody–antigen complexes, whereas the unmodified protein had no such effects.87 Moreover, it is possible that the apoptosis- induced chromatin modifications are an initial event that results in disruption of the immune tolerance to self and formation of pathogenic chromatin autoantibodies in SLE.

MiRNAs The miRNAs are endogenous 21–25 nucleotide noncoding RNA molecules that are potent negative regulators of gene activity.88 Recent studies have shown that miRNAs regulate the function of both the innate and adaptive immune system, which are involved in various immune pathways,89 and could potentially serve as disease biomarkers and therapeutic targets.90 Altered miRNA expression has been reported in human autoimmune diseases including SLE.91 Dai et al.92 comprehensively analyzed miRNAs in the PBMCs of SLE patients by using miRNA microarray. They identified 16 SLE-related miRNAs, with 7 miRNAs downregulated and 9 miRNAs up-regulated in SLE PBMC. Subsequently, they further compare the miRNA profile of kidney biopsies between N patients and controls.93 They found 66 miRNAs in renal samples, of which 30 miRNAs were downregulated and 36 miRNAs upregulated in LN biopsies. In this regard, many miRNAs have been identified based on their possible pathogenic roles. For example, miR-21 regulates aberrant T-cell responses through regulation of PDCD4 expression in SLE.94 miR- 125a contributes to elevated levels of the inflammatory chemokine RANTES (regulated and normal T cell expressed and secreted) in SLE, via targeting of KLF13.95 In addition, the knocking down of miR-126 in SLE CD4 þ T cells decreased their autoimmune activity and their stimulatory effect on IgG Figure 1. Methylated CpG (CmpG) demethylation by GADD45a. DNA production in the co-cultured B cells.96 MiR-146a is a negative demethylation may occur through a two-step enzymatic process, regulator of the IFN pathway, and underexpression of miR-146a promoted by GADD45a. The first step involves deamination of conduces to alterations in the type I IFN pathway in lupus patients CmpG nucleotides by activation-induced deaminase (AID), generat- 90 ing a thymine product and a G:T mismatch. The second step by targeting the key signaling proteins (Toll-like receptor 7). involves thymine base removal by methyl-CpG-binding domain 4 Moreover, it is important to note that miRNAs have a close (MBD-4). In the following, the missing base is replaced by a relationship and complex interplay with DNA methylation and nonmethylated cytosine group, resulting in demethylation of CmpG histone modifications. On one hand, histone modifications and dinucleotides.80 DNA methylation can regulate the expression of certain miRNAs.

& 2013 Macmillan Publishers Limited Genes and Immunity (2013) 133 – 146 Applied basic research of SLE W Sui et al 138 Ding et al.97 found that miR-142 À 3p/ À 5p is downregulated in was a significant correlation in mRNA expression of T-bet with IFN- SLE CD4 þ T cells because of a deregulation of histone g, and of GATA-3 with IL-4. These findings were further confirmed methylation (H3K27me3) and DNA methylation in its promoter. by Chan et al.107 Moreover, recent studies have shown that a Subsequently, the decreased miR-142 À 3p or miR-142 À 5p can risk haplotype of STAT-4 for SLE is overexpressed. Several increase CD84 and IL-10 or SAP (stress-activated protein) protein TFs, including AP-1 and NF-kB, have been found to increase levels, respectively, which contributes to aberrant T-cell responses STAT-4 expression by interacting with its promoter.108 STAT4 plays and B-cell hyperresponsiveness in SLE. On the other hand, miRNAs a key role in type I IFN receptor signaling that results in type I IFN can affect DNA methylation and histone modifications by over-expression and contributes to the disease pathogenesis of targeting methylase and acetylases enzymes directly. Zhao SLE. The TF IRF5 is also a major mediator of Toll-like receptor- et al.96 report that miR-126 correlates with decreased DNMT1 triggered expression of proinflammatory cytokines such as type I expression by interaction with the 30-UTR region of DNMT1 in SLE IFN and TNF-a.100 IRF5 could also promote expression of the CD4 þ T cells. STAT4 gene in the same way; thus, both STAT4 and IRF5, which are functionally linked, contain polymorphisms that confer increased risk for SLE. TRANSCRIPTOME RESEARCH OF SLE In addition, NF-ATc2 regulates CD154 and IL-2 gene transcrip- As a group of crucial components in the interconnecting tion by binding to the promoters. Juang et al.109 present evidence regulatory networks, TFs play a central role in regulating gene that NF-ATc2 is increased in the nuclei of activated SLE T cells, transcription and expression. Analysis of the human genome which results in increased CD154 and decreased IL-2 transcription sequence predicts there may be as many as 1850 human TFs.98 in patients with SLE. The explanations for these discrepancies TFs interact with specific DNA-binding elements presented in the may be that the transcriptional repressor CREM (cyclic AMP- promoters of certain genes to promote transcriptional initiation. response element modulator)-mediated AP-1 (c-fos/c-jun Defects in TF regulation, structure and function play critical roles heterodimer) deficiency in SLE T cells leads to the null effect of in the development and function of the immune system.99,100 upregulated NF-AT on the transcription of IL-2. The decisive Accumulating evidence has suggested that the TFs with elements for the transcription of the IL-2 lie within 300 bp differences in expressions and activity may be potential novel upstream of the start codon of the IL-2 gene promoter, including biomarkers or help to illuminate pathogenic mechanisms of AP-1 sites, NFAT sites, NF-kB site and CREM/CREB site and complex genetic diseases such as SLE. others. PP2A is the primary enzyme that leads to dephosphorylation of pCREB in T lymphocytes,110 and PP2A- Aberrant expression of TFs mediated dephosphorylation at SP-1 serine 59 is critical for CREM P1 promoter activity. In SLE T cells, the overexpressed PP2A In order to reveal the relationship between SLE and TFs, Sui contributes to decreased levels of pCREB and increased levels of et al.101 compared the levels of TF expression by microarray dephosphorylated SP-1, and the dephosphorylated SP-1 drives technology in the PBMCs of SLE patients with normal controls. In increased expression of CREM through the CREM P1 promoter.111 the study, a total of 345 TFs were identified, with 92 of them It is conceivable that the decreased pCREB and increased CREM differentially expressed, in which 78 TFs were upregulated may permit CREM to occupy their common À 180 binding site on and 14 TFs downregulated in SLE as compared with the control 112 the IL-2 promoter and directly suppress IL-2 production. group. They further indicated that TFs such as activator protein-1 Moreover, CREM binding to the c-fos promoter in unstimulated (AP-1), Pbx1 and myocyte enhancer factor-2 (MEF-2) were live T cells from SLE patients is significantly increased. Increased potential diagnosis biomarkers and probable factors involved in binding results in decreased transcription of c-fos gene, decreased the pathogenesis of SLE. Considering that the important role of production of c-fos protein and AP-1 (c-jun/c-fos dimmers) FOXOs in controlling lymphocyte activation and immune binding in the nuclei of SLE T cells. In this regard, a lack of NF- functions. Kuo et al.102 evaluated the transcript profiles of kB activity, particularly the function provided by the p65 subunit, FOXO1, FOXO3a and FOXO4 genes in PBMCs from SLE and RA also provides a mechanism for a defect in IL-2 expression. Taken patients as compared with healthy controls. Their results showed together, decreased expression of the transcriptional enhancers that transcript levels of FOXO1, but not of FOXO3a or FOXO4, were NF-kB, pCREB and AP-1, as well as increased expression of the significantly downregulated compared with normal controls and transcriptional repressor CREM, leads to decreased production of correlated with lupus disease activity. It suggests that the IL-2 in SLE T cells. The above information suggests that the transcriptional dysregulation in FOXO1 is possibly linked to the interaction and coexpression between TFs and cytokine provide pathogenesis of SLE and rheumatoid arthritis. Additionally, positive and negative feedback regulation for Th1 and Th2 Burgos et al.103 found that significantly increased levels of c-Rel development. (a member of nuclear factor (NF)-kB/Rel family of TFs) mRNA were found in PBMCs from SLE patients. It is noteworthy that estrogen induces serine protease-mediated proteolysis of signal transducer 104 and activator of transcription 1 (STAT-1) and NF-kB. In this POSTTRANSCRIPTIONAL REGULATION OF SLE way, estrogen enhances STAT-1 DNA binding activity without Recently, a widespread incidence of A-to-I editing of Alu-contain- increasing the expression of phosphorylated and total STAT-1 ing and intronic regions of pre-mRNAs in the human transcrip- protein. Subsequently, the modified STAT-1 and NF-kB proteins in tome has been identified in patients with SLE.113 Messenger RNA splenocytes from estrogen-treated mice have differential roles in editing induced by ADARs (adenosine deaminases acting on RNA) regulating different inflammatory molecules. and other editing enzymes is a process defined as the co- or posttranscriptional modification of mRNA through nucleotide Abnormal gene transcription substitutions, insertions and deletions, which play a pivotal role in Imbalance in the cytokines produced by the two subsets of the regulation of gene expression and in pathophysiology.114,115 T-helper cells, Th1 and Th2, probably plays an important role in In addition to RNA editing, alternative splicing is also a powerful the pathogenesis of SLE.105 T-bet and GATA-3 are the principal TFs mechanism of gene regulation. In this way, a single human gene for the differentiation of Th1 Th2 lymphocytes, respectively. Lit can produce a variety of protein products with related yet distinct et al.106 found significantly increased mRNA levels of T-bet and functions that may allow cells of different types or at different IFN-g and decreased mRNA levels of GATA-3 and IL-4 in SLE developmental stages to perform different tasks. However, patients, which correlated with lupus disease activity, and there aberrant mRNA editing and splicing could cause the cells to

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited Applied basic research of SLE W Sui et al 139 synthesize dysfunctional proteins or to make an otherwise useful Tsokos127 and other investigators identify ASF/SF2 (alternative protein at the wrong time. splicing factor/splicing factor 2) as a novel factor that binds to the 30-UTR of CD3z in the regulation of alternative splicing and protein Messenger RNA editing expression in SLE patient T cells. The CD3z mRNA is a 1.472-kb spliced product of the eight exons with a coding region of 492 bp Laxminarayana et al.116–118 and Orlowski et al.119 demonstrated and an exon VIII-encoded long downstream 906-bp 30-UTR. The the occurrence of altered mRNA editing profile of PDE8A1, ADAR2 reverse transcriptase-PCR analysis of the TCRz chain 30-UTR and a regulatory subunit of protein kinase A (RIa) gene transcripts showed an alternatively spliced (AS) 344-bp product with both in SLE T cells, which has been confirmed by the absence of such splicing donor and acceptor sites. It was suggested that activation mutations in genomic DNA. They observed A-to-I hypoediting at of alternative splicing within the 30-UTR due to usage of two known sites of PDE8A1, specifically in T cells derived from SLE internal splice sites (50 and 30) leads to splice deletion of 562 bases patients and in normal T cells activated with type I IFNs. Similarly, (nucleotides 672–1233) of the TCRz chain mRNA, which results in altered editing at known sites and novel editing sites in the ADAR2 the generation of a 344-bp AS variant. Therefore, SLE T cells gene products was uncovered, including decreased editing of the exhibit variably increased levels of the AS isoform,128 whereas 23 and 24 sites from A to I and increased editing of 6 and þ þ À T cells from healthy individuals express mainly the wild-type 30 from U to C in SLE T lymphocytes. Notably, the results of þ isoform. Although this alternative splicing just affects the UTR and sequenced cDNA of RIa and corresponding genomic DNA of the not the coding region, the 344-bp AS CD3z isoform lacks a coding region showed that there was 1.22 10 À 3/bp transcript  translation regulatory sequence and two key regulatory mutation in SLE T cells—a frequency 7.5 times higher than that in adenosine/uridine-rich elements. Thus, the transcript stability control T cells. Furthermore, Toyabe et al.115 found that there are and translation of AS isoform are significantly lower than that of increased nucleotide substitutions and deletions of DAP12 mRNA the 906-bp wild-type CD3z isoform. Consequently, the relative in natural killer cells from SLE patients, which results in decreased amount of CD3z protein generated by the AS isoform is expression of DAP12. Wu et al.120 found that a novel FAS mRNA significantly lower than that from the wild-type isoform,129 mutation is significantly increased in SLE; this FAS mRNA mutation which can explain why the TCR-associated CD3z chain is is caused by a site-specific editing process, in which an adenine is decreased in SLE patients. In addition to CD3z gene, abnormal inserted at a defined location that leads to the production of a expression of AS isoforms of other genes is frequently observed in novel defective FAS receptor (edFAS) (Figure 2). It was interesting patients with SLE, including the adhesion molecule CD44 and the to note that the increased editing was accompanied by an TF CREM.130,131 increased expression of ADAR1; ADAR1 mRNA content 3.5 times higher in SLE T cells than in control T cells has been reported. An explanation for the abnormal mRNA editing is that high circulating levels of IFN-a upregulate IFN-inducible 150 kDa ADAR1 transcript PROTEOMICS OF SLE levels in SLE patients, which results in increased ADAR1 enzymatic Protein is the executor of physiological function. As we enter the 132 activity,121 and suppression of ADAR2-catalyzed transcript editing, postgenomics era, proteomics has been implanted into almost thus leading to increased or decreased editing of gene transcripts every field of life sciences and medicine. The feasibility of induced by different ADARs. These pathological RNA editing of proteomics technology with multiple samples is ideally suited genes account for aberrant gene expression and signal for investigating the complicate alteration of proteomics in transduction and, ultimately, impaired immunity in SLE patho- patients with SLE, which may help to find new biomarkers that genesis. For instance, mutant RIa transcripts are patho- can be used to reliably confirm SLE or monitor their progression with physiolgically significant, for they can encode diverse and high sensitivity and specificity. Using this method, we can better aberrant RIa isoforms, including truncated and dominant- understand the pathogenesis, diagnosis and treatment for SLE. negative subunits, which result in deficient type 1 protein kinase A (PKA-I) activity. As we know, PKA isozymes phosphorylate Proteins in PBMCs multiple intracellular substrates, including plasma membrane Dai et al.133,134 identified an overall pattern of differentially proteins, intracellular receptors and TFs.122–124 We propose that expressed proteins profiles in PBMCs of patients with SLE as deficient PKA-I isozyme activity contributes to the pathogenesis of compared with healthy controls. In 2008, they found that 20 SLE by hindering effective signal transduction and impairing T-cell proteins are differentially expressed in the PBMCs of patients effector functions. Similarly, aberrant FAS mRNA editing results in with SLE.133 Among these, 11 protein spots were upregulated the production of a defective FAS receptor, which is unable to and 9 protein spots were downregulated, and 5 differentially mediate apoptosis signal.125 expressed proteins were identified as immunoglobulin J chain, apolipoprotein A-IV precursor, calprotectin L1H, zinc-finger protein Alternative splicing subfamily 1A (all upregulated) and glutathione S-transferase (GST; Alternative splicing is also a powerful mechanism of gene downregulated). As described here, it is worthwhile to mention regulation that conduces to the generation of numerous that GST is involved in the detoxification of reactive oxygen transcripts and proteins from a single gene.126 Moulton and species, and genetic polymorphisms of GSTM1, GSTT1 and GSTP1

Figure 2. Sequence tracings show an adenine insertion in FAS complementary DNA (cDNA). The upper panel shows that the wild-type FAS (wtFAS) cDNA contains a stretch of six adenines (underlined), and the lower panel shows edited FAS (edFAS) cDNA containing seven adenines (underlined) in the same region.120

& 2013 Macmillan Publishers Limited Genes and Immunity (2013) 133 – 146 Applied basic research of SLE W Sui et al 140 are associated with altered enzyme activity and susceptibility to returned to baseline by 4 months post flare, suggesting it may be SLE. In order to further investigate the altered proteins profiles in a biomarker for following response to therapy. SLE, a total of 452 proteins involving different biological functions 134 and cellular locations were identified by them in 2010. Their Tissues proteins experimental results revealed that 67 unique proteins were Cytokeratins are major structural proteins specifically expressed in observed with twofold or more alteration in levels across 141 epithelial cells, in which they form intermediate filaments. groups. Comparative proteome analysis differentiating active Cytokeratins are encoded by a large multigene family, and are from inactive SLE revealed 17 upregulated and 13 down- classified as type I keratin 9 (K9 À K20) or type II (K1–K8). It is of regulated proteins. Notably, STRAP (serine/threonine kinase importance that an alternative pattern is open to keratinocytes in receptor-associated protein) was found to reduce more than response to epidermal injury or certain skin disorders.142,143 Fang threefold in active SLE patients in contrast to the healthy controls, et al.144 used a proteomics approach to clarify the abnormal and was inversely correlated to the SLE disease activity. The expression of cytokeratins in lesions from SLE patients. Their result possible molecular mechanisms of action of STRAP are related to showed 16 protein spots that were upregulated and 9 that were its interaction with the TGF-b receptor or apoptosis signal- downregulated in SLE lesions in contrast to the normal epidermis. regulating kinase 1.135 Among them, K1 and K10 were downregulated in SLE lesions whereas K2e, K5, K6, K14, K16 and involucrin were obviously Serum proteins upregulated. Especially, K6 was only expressed in SLE lesions but Human serum contains thousands of proteolytically derived not in normal skin. In addition, the distribution of protein also peptides that may provide a robust correlation of the physiolo- presented an altered pattern between SLE and control subjects. gical and pathological processes in the entire body. Measuring For instance, K5 and K14 were not only limited to basal cells but to panels of peptidome markers might be more sensitive and specific suprabasal cells, whereas K5/K14 were limited to basal cells in than conventional biomarker approaches. Previous studies from normal epithelia. Similarly, Sui et al.145 utilized iTRAQ (isobaric tags the laboratories of Huang et al.136 and Dai et al.137 demonstrated for relative and absolute quantitation) technology to analyze the differentially expressed disease-related and condition-specific total proteins of renal tissue in patient with LN. In their study, a peptides in sera from patient with SLE to develop and validate total of 490 distinct proteins involving different biological the peptide classification model for the diagnosis of SLE. The functions and cellular locations were identified. Of these, 113 experimental results of Huang et al.136 exhibited 60 differential proteins were upregulated or downregulated at onefold or more proteins peaks (Po0.05) between SLE and control subjects, 28 of alteration in levels as compared with normal controls. Four which were upregulated and 32 downregulated in SLE patients. proteins showed significant deviation, and they are identified as Notably, a cluster of four proteins at m/z 3376.02, 4070.09, 7770.45 argininosuccinate synthetase (ASS), aldolase (all downregulated), and 28045.10 segregate SLE from non-SLE with a sensitivity of heterogeneous nuclear ribonucleoprotein (hnRNP-) and annexins 91% and a specificity of 92%. Similarly, the experimental results of (all upregulated). Dai et al.137 show that individuals with active SLE, stable SLE and rheumatoid arthritis shared 23 significantly differentiated peptides compared with normal controls, including 10 downregulated and PROTEIN MODIFICATION OF SLE 13 upregulated peptides. Among them, four peptide ion Classical biochemistry tells us that 20 amino acids make up most signatures (m/z 3192, 3263, 4268 and 5754) achieved a of the proteins in nature. However, it is noteworthy that between recognition capacity of 95.74% and a cross-validation of 89.82% 50 and 90% of the proteins in the human body acquire PTM. The to discriminate active SLE from healthy control samples, and a most frequently observed PTMs include phosphorylation, glyco- group of peptides at m/z 2661, 2863, 3263 and 5754 separated sylation, acetylation, ubiquitination and citrullination. PTMs alter stable SLE effectively from normal controls with a recognition various aspects of the protein chemistry, including primary and capacity of 96.30% and a cross-validation of 78.08%. Based on this tertiary structure, biological (and/or enzymatic) functions and information, we can infer that proteomics is effective in facilitating proteolytic degradation that may be decisive in generating the construction of a sensitive and specific diagnostic model. immunogenic or tolerogenic self-peptides. Accumulating evi- dence suggests the involvement of PTMs in the pathophysiology Urine proteins of SLE. LN is a serious and frequent complication of SLE. Urine proteins that are differentially expressed during renal flare may be the PTMs with aberrant signaling novel and predictive biomarkers of LN. Several SLE nephritis Phosphorylation of proteins on tyrosine, serine or threonine by studies examined the urine proteome by surface-enhanced laser protein kinases is a key biochemical process in intracellular desorption/ionization (SELDI) platform and produced encouraging signaling. Taher et al.146 applied the kinome array technique to results.135–137 One reported two protein ions at m/z 3340 and 3980 generate a kinase profile of B lymphocytes from SLE patients and that together segregated active nephritis from inactive nephritis healthy controls. They found that there were changes in the with sensitivity and specificity of 92%, but the actual peptides phosphorylation pattern in B lymphocytes from a majority of the were not identified.138 Another group of investigators found eight SLE patients, with an increase in the phosphorylation of 27 target protein ions, which correlated with renal disease activity; most of substrates and a decrease in the phosphorylation of 34 target these were 420 kDa and mostly unidentified, but one seemed to substrates relative to controls. Their results further indicate that be albumin.139 Zhang et al.140 used SELDI time-of-flight mass the activity of phosphatidylinositol 3-kinase, MAP, Rho, FAK, ASK-1, spectrometry to examine the low-molecular-weight proteome of CDK-1, PKC, PKA, Tec and the cell cycle kinase ATR was altered in serial lupus urine samples. They found that there are 27 protein SLE patients. Among them, phosphatidylinositol 3-kinase activity ions differentially expressed between specific flare intervals. was increased, which regulates survival and differentiation. The Altered protein ions were identified as 20/25 amino-acid activity of ATR, which regulates the cell cycle, was decreased in isoforms of hepcidin, fragments of a1-antitrypsin (A1AT) and SLE patients in contrast to controls. albumin. Hepcidin 20 (m/z 2198) increased 4 months before flare Additionally, T cells from patients with SLE are characterized by and then gradually returned to baseline by 4 months post flare, decreased expression of CD3z chain and increased expression of implying it may be a useful marker of impending renal flare, FcRg chain, which contributes to aberrant signaling.147 Abnormal whereas hepcidin 25 (m/z 2432) decreased at renal flare and then PTMs of TFs can account for this deregulation. The TF EIf-1 is

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited Applied basic research of SLE W Sui et al 141 produced as a 68-kDa protein, and sequence analysis of EIf-1 Moreover, several studies have suggested that PTMs and revealed the existence of multiple PKC phosphorylation and especially phosphorylation and citrullination can influence auto- glycosylation sites. It undergoes partial phosphorylation and antibody binding onto different autoantigens in patients with SLE. O-linked glycosylation to convert to an 80-kDa protein. This Terzoglou et al.158 showed that phosphorylation of the linear form stays in cytoplasm, where it can be further phosphorylated B-cell epitope 349–368aa of La autoantigen/Sjogren syndrome and O-glycosylated to convert to a 98-kDa form, which travels to antigen B (La-SSB) autoantigen enhances its relative avidity and the nucleus. In the nucleus, TF Elf-1 binds the CD3 z and FcRg autoantibody binding. They also demonstrated Ro60 kDa auto- promoters at the GGAA element.148,149 However, in SLE T cells, the antigen contains two linear B-cell epitopes, spanning the 80-kDa EIf-1 protein does not undergo proper posttranslational sequences 169–190aa and 211–232aa.159,160 The epitope 169– modification, which results in decreased levels or absence of the 190aa was recognized mainly from sera of SLE patients, and this 98-kDa form. Elf-1 is a TF that can serve as a transcription activator sequence contains two arginine and one serine amino-acid or suppressor.150 Decreased blinding of EIf-1 to the CD3z and FcRg residues, which can potentially be modified posttranslationally promoters is associated with decreased expression of CD3z and by citrullination or phosphorylation, respectively. Their experi- increased expression of FcRg in SLE T cells. mental results showed that PTMs of the B cell epitope 169–190aa of the autoantigen Ro60 kDa can decrease recognition of this epitope by autoantibodies found in the sera from patients with PTMs with self-antigens autoimmune diseases including SLE. In this regard, other It is clear from the literature that PTM of self-proteins is an investigators discovered that the aberrant N-glycosylation on the important explanation of how immune tolerance is lost to self- kidney proteins of a SLE model mouse.161 The mice lacking the antigens (Figure 3).151–153 Several sets of compelling data have gene for a-mannosidase II, which regulates hybrid to complex suggested that modified self-antigen influences immune branching patterns of extracellular asparagine (N)-linked recognition, whereas lack of a normally occurring PTM within a oligosaccharides chains (N-glycans), develop a lupus-like self-antigen may also induce autoimmunity. For instance, nephritis characterized by autoantibodies to histones, Sm snRNP the sera of SLE patients contain autoantibodies to both (small nuclear ribonuclear protein) antigen, and DNA. It is thought nonphosphorylated and phosphorylated SR proteins (pre-mRNA that the lack of complex N-glycan branching may affect self- splicing factors).154 In this regard, Utz et al.152 and other antigen recognition. Although the mechanisms are not fully investigators155,156 have shown that cells undergoing apoptosis addressed, these observations can be explained in three ways. are uniquely suited to present modified self-proteins to the First, the modified residues may conduce directly to antibody immune system in such a way that it overcomes normal binding and a possible modification of one or more of these mechanisms of peripheral tolerance. It was also revealed that residues can influence the affinity of the antibodies. Second, the proteins modified during apoptosis may also be preferred targets tested PTMs may have affected the secondary structure of the for autoantibody production.154,157 For instance, SAP kinase in peptides. Third, the existence of these modifications alter the phosphorylation of autoantigens during apoptosis connects a overall electrical charge of the peptides.160 Taking the La/SSB common protein modification with autoantibody production in protein as an example, the 349–368aa La/SSB epitope is highly SLE patients. positively charged (isoelectric point ¼ 10.4), whereas a phosphate

Figure 3. Posttranslationally modified proteins initiate autoimmune responses. Self-proteins can be modified during cellular stress, such as apoptosis, inflammation or aging. These PTM self-proteins are released from apoptotic or necrotic cells, where they are phagocytosed by macrophages and other antigen-presenting cells (APCs); once inside the APC, the existence of these modifications can change how proteases cleave the self-antigen, thus generating new epitopes. These modified peptides are then presented in the context of MHC class II molecules to T and B cells that have escaped the thymus or bone marrow during negative selection because the modified peptide is not present in those organs. Finally, autoreactive T cells and B cells infiltrate host tissue and induce autoimmune pathology.153

& 2013 Macmillan Publishers Limited Genes and Immunity (2013) 133 – 146 Applied basic research of SLE W Sui et al 142 group has a negative charge. The incorporation of a negatively content was observed, indicative of possible edema. There exists a charged group into a positively charged protein region would be close correlation between energy metabolism and cell volume expected to alter the local tertiary structure via ion–ion interac- regulation in cultured astrocytes. Increased concentrations of tions,162 eventually affecting anti-La/SSB antibody binding. different metabolites such as glutamine and glutamate can cause changes in ionic strength and thereby lead to a change in water transport and content. As the brain is in a defined space, waterflux METABOLOMICS OF SLE and cell volume are critical and intracerebral swelling can be fatal. Alternative biomarker discovery approaches include metabolo- Thus, the altered glucose metabolism correlates with brain lesions mics, that is, the systematic study of unique chemical fingerprints and behavior in SLE patients.168 Additionally, a significant increase or small-molecule metabolite profiles that are the result of specific in lactate concentrations in lupus brains was a result of de novo cellular processes. Nuclear magnetic resonance (NMR) spectro- synthesis from glucose other than pyruvate synthesized from scopy in combination with pattern recognition analytical techni- precursors. An increase in lactate has several pathological ques has grown in popularity for metabolic profiling. Until now, implications because severe lactic acidosis can result in accumulated evidence has suggested that the metabolite profiling complete infarction of the tissue and, given its vasodilating is altered in patients with SLE and association with disease activity. effect on cerebral vessels, even at neutral pH levels it may contribute to changes in cerebral blood flow.169 Thus, an increase Blood metabolites in lactate is indicative of brain dysfunction, and the selective changes in brain metabolites and osmolytes may be the Recently, Ouyang et al.163 utilized 1H NMR spectroscopy-based underlying cause of neurological disturbances and brain edema metabolomics approach to identify the characteristics of that occur in SLE. metabolite profiling from serum from patients with SLE. In their study, 600 Hz NMR spectroscopy was applied to detect serum of 64 SLE patients and 35 healthy controls. They found that the SLE Urinary metabolites serum samples were characterized by reduced concentrations of In the study of Rosendale et al.170 they showed that two urinary valine, tyrosine, phenylalanine, lysine, isoleucine, histidine, metabolites, taurine and citrate, were found to accurately glutamine, alanine, citrate, creatinine, creatine, pyruvate, distinguish between proliferative LN (class III/IV) and pure high-density lipoprotein, cholesterol, glycerol, formate and membranous LN (class V ) patients. Class III/IV LN patients had increased concentrations of N-acetyl glycoprotein, very-low- 410-fold lower levels of urinary taurine in contrast to class V density lipoprotein and low-density lipoprotein in comparison patients, who had mostly normal levels (Po0.01). Urinary citrate with the control population. Among these, the decreased citrate levels were eightfold lower in class V LN in comparison with class and pyruvate in serum possibly resulted from meeting the energy III/IV patients, who had normal levels of urinary citrate (Po0.05). charge in the Krebs cycle, which might reflect an increased energy Moreover, urinary hippurate levels accurately distinguished demand relative to the decreased energy availability under between class V patients, who had normal levels of urinary inflammatory conditions. The elevated levels of very-low-density hippurate, compared with focal segmental glomerulosclerosis lipoprotein and low-density lipoprotein, and lower levels of high- patients, who completely lacked urinary hippurate (Po0.001). density lipoprotein, which have been identified as the ‘lupus In this regard, previous studies have suggested that taurine and pattern’, might be related to the inflammatory process because citrate are both measures of tubular cell function.171 The kidneys poly-unsaturated fatty acids (for example, leukotrienes or filter and reabsorb metabolites to maintain a metabolic prostaglandins) are precursors of inflammatory mediators. In this equilibrium, and the presence of renal pathologies can impair regard, several investigators have further observed that lipid the filtration of small metabolites such as taurine and citrate peroxidative damage was significantly increased in SLE,164,165 through the glomerulus and their subsequent re-absorption in the which is a risk factor for cardiovascular disease. Based on these renal tubules, leading to alterations in metabolic profiles.172 findings, it is likely that SLE mainly influenced energy, amino acid, lipid, and purine nucleotide metabolism. Furthermore, principal Bone metabolites component analysis and partial least square discriminant analysis More recently, increasing interest has been focused on bone models were established by Dai et al.,163 which was capable of metabolism in patients with rheumatic disorders.173,174 A large distinguishing SLE patients and healthy controls with a specificity body of work has demonstrated that SLE is associated with a and sensitivity of 97.1% and 60.9%, respectively. decreased bone mineral density (BMD), which is related to the severity of the disease rather than to the glucocorticoid Brains metabolites treatment.175,176 Kalla et al.177 described that SLE is associated Involvement of the brain is one of the most important with significant cortical bone loss, and the BMD measured at the complications of SLE. It is clear from the literature that brain lumbar spine or the total femur is reduced.175 However, other function is closely correlated with glucose utilization because the studies have presented inconsistent results. Dhillon et al.178 did brain is dependent exclusively on glucose as its sole source of not find a decreased lumbar bone mineral density as assessed by energy. Komatsu et al.166 evaluated 12 active SLE patients with or dual-energy X ray absorptiometry. There are several possible without psychiatric symptoms. Their study revealed that explanations for this discordance, such as small sample size and psychiatric patients had significantly decreased regional cerebral ethnic background. In the study of Hansen et al.,179 BMD and metabolic rates for glucose in the prefrontal, inferior parietal and biochemical markers of bone turnover was examined and anterior cingulate regions. In order to obtain an insight into the followed for up to 2 years in a cohort of 36 patients with SLE. dynamics of changes in glucose metabolism, Jessy et al.167 used They found that the BMD was decreased in the hand and neck but 1H and 13C NMR spectroscopy to determine the rates of was normal in the spine and distal forearm. Moreover, the serum incorporation of glucose into amino acids and lactate via cell- concentration of the carboxyterminal crosslinked telopeptide of specific pathways in mice with lupus. They found that the MRL/lpr type I collagen was increased in almost all patients. Teichmann mice had a significant increase of 30–80% (Po0.001) in total brain et al.180 found that the reduction of BMD in female patients with glutamine, glutamate and lactate concentrations, whereas alanine, SLE was related to the significantly increased excretion of urinary aspartate, N-acetyl aspartate and g-aminobutyric acid remained pyridinoline, to hypoparathyroidism, and to the decrease in serum unchanged as compared with the congenic MRL þ / þ control osteocalcin. Moreover, Redlich et al.181 showed that albumin mice. Furthermore, a significant increase in total brain water corrected serum calcium concentrations were slightly but

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited Applied basic research of SLE W Sui et al 143 significantly increased in the SLE group. Phosphorus values 5 Tsao BP, Wu H. The genetics of human lupus. In: Wallace DJ, Hahn BH (eds) were found to be significantly lower in SLE patients, but there Dubois’ lupus erythematosus. 7th edn Lippincott Williams & Wilkins: Philadelphia, were no significant differences in parathyroid hormone and 2007, pp 54–81. 25-hydroxyvitamin vitamin D levels. 6 Harley JB, Alarco´n-Riquelme ME, Criswell LA. Genome-wide association scan in women with systemic lupus erythematosus identifies susceptibility variants in ITGAM, PXK, KIAA1542 and other loci. Nat Genet 2008; 40: 204–210. CONCLUSIONS 7 Fernando MM, Stevens CR, Sabeti PC, Walsh EC, McWhinnie AJ, Shah A et al. Identification of two independent risk factors for lupus within the MHC in United Taken together, the past decades have seen momentous Kingdom families. Plos Genet 2007; 3: e192. progress in the field of SLE, with a better understanding of the 8 Yang W, Shen N, Ye DQ. Genome-wide association study in Asian populations pathophysiology of the disease and the identification of many identifies variants in ETS1 and WDFY4 associated with systemic lupus erythe- novel biomarker candidates for disease monitoring and treatment matosus. PLoS Genet 2010; 6: e1000841. evaluation as well as developing novel strategies for therapy. As 9 Slingsby JH, Norsworthy P, Pearce G. Homozygous hereditary C1q deficiency and described here, we clearly show that multiple susceptibility genes systemic lupus erythematosus. A new family and the molecular basis of C1q contribute to the development of SLE. Many of the SLE deficiency in three families. Arthritis Rheum 1996; 39: 663–670. susceptibility genes have known immune functions and can be 10 Barilla-LaBarca ML, Atkinson JP. Rheumatic syndromes associated with com- placed into several focal pathways. These pathways highlight the plement deficiency. Curr Opin Rheumatol 2003; 15: 55–60. importance of immune complex clearance (complement and 11 Wu YL. Phenotypes, genotypes and disease susceptibility associated with gene copy number variations: complement C4 CNVs in European American healthy phagocytosis), lymphocyte signaling (T- and B-cell signaling) and subjects and those with systemic lupus erythematosus. Cytogenet Genome Res the innate immune response (IFN and NF-kB signaling) in SLE 2009; 123: 131–141. predisposition. Current follow-up efforts are now focused on 12 Salmon JE, Pricop L. Human receptors for immunoglobulin G: key elements in precisely identifying the causative genetic variants and their the pathogenesis of rheumatic disease. Arthritis Rheum 2001; 44: 739–750. effects, and the biological mechanisms through which they 13 Salmon JE, Millard S, Schachter LA, Arnett FC, Ginzler EM, Gourley MF et al. predispose to SLE, which will provide a major advance in our FcgammaRIIA alleles are heritable risk factors for lupus nephritis in African ability to translate genetic associations into diagnostic applica- Americans. J Clin Invest 1996; 97: 1348–1354. tions. In addition, reduced DNA methylation, histone hypoacetyla- 14 Chu ZT, Tsuchiya N, Kyogoku C. Association of Fcgamma receptorIIb poly- tion and hyperacetylation and the overexpression of certain morphism with susceptibility to systemic lupus erythematosus in Chinese: a miRNAs, resulting in immune imbalance, have been shown to be common susceptibility gene in the Asian population. Tissue Antigens 2004; 63: associated with the onset and progression of SLE. Future work 21–27. 15 Kyogoku C, Dijstelbloem HM, Tsuchiya N, Hatta Y, Kato1 H, Yamaguchi1 A et al. should determine how epigenomics connects the environment to Fcgamma receptor gene polymorphisms in Japanese patients with systemic cell biology and autoreactivity. Unlike genetic alterations, which lupus erythematosus: contribution of FCGR2B to gene susceptibility. Arthritis are permanent, epigenetic alterations are reversible. This opens Rheum 2002; 46: 1242–1254. the possibility of introducing preventive measures and using 16 Siriboonrit U, Tsuchiya N, Sirikong M. Association of Fcgamma receptor IIb and epigenetic drugs to reverse the pattern of epigenetic alterations to IIIb polymorphisms with susceptibility to systemic lupus erythematosus in Thais. relieve the phenotype. Moreover, the aberrant expressions and Tissue Antigens 2003; 61: 374–383. activity of TFs, altered transcript editing and dysfunctional PTMs 17 Li X, Wu J, Carter RH. A novel polymorphism in the Fcgamma receptor IB(CD32B) also play important roles in the pathogenesis and development of transmembrane region alters receptor signaling. Arthritis Rheum 2003; 48: SLE. On the other hand, the protein and metabolic profiles provide 3242–3252. high sensitivity and specificity to predict the diagnosis of SLE. 18 Magnusson V, Zunec R, Odeberg J, Sturfelt G, Truedsson L, Gunnarsson I et al. Continued work will be required to confirm the quantification of Polymorphisms of the Fcgamma receptor type Ib gene are not associated with systemic lupus erythematosus in the Swedish population. Arthritis Rheum 2004; some interested biomarker candidates in a larger and different 50: 1348–1350. population of SLE patients. Overall, much remains to be done 19 Su K, Wu J, Edberg JC. A promoter haplotype of the immunoreceptor before the etiology of SLE and LN becomes fully understood. tyrosine-based inhibitory motif-bearing Fcgamma RIIb alters receptor expression and associates with autoimmunity. I. Regulatory FCGR2B polymorphisms and their association with systemic lupus erythematosus. J Immunol 2004; 172: CONFLICT OF INTEREST 7186–7191. The authors declare no conflict of interest. 20 Su K, Li X, Edberg JC. A promoter haplotype of the immunoreceptor tyrosine- based inhibitory motif-bearing FcgammaRIIb alters receptor expression and associates with autoimmunity. II. Differential binding of GATA4 and Yin-Yang1 ACKNOWLEDGEMENTS transcription factors and correlated receptor expression and function. J Immunol 2004; 172: 7192–7199. We thank Minglin Ou for helpful discussions. This work was supported by funds 21 Karassa FB. The Fcg RIIIA-F158 allele is a risk factor for the development of lupus received from Guangxi Natural Science Foundation (no. 2012GXNSFDA053017), nephritis: a meta-analysis. Kidney Int 2003; 63: 1475–1482. Guangxi Key Laboratory of Metabolic Diseases Research (no. 11-031-33 and 22 Salmon JE, Edberg JC, Kimberly RP. Fcgamma receptor III on human neutrophils 12-071-32) and Guilin Scientific Research and Technology Development Program Allelic variants have functionally distinct capacities. J Clin Invest 1990; 85: (no. 20110119-8-1). 1287–1295. 23 Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med REFERENCES 1979; 301: 5–8. 1 Bengtsson AA, Rylander L, Hagmar L, Nived O, Sturfelt G. Risk factors for 24 Takahashi S, Fossati L, Iwamoto M, Iwamoto R, Motta R, Kobayakawa T et al. developing systemic lupus erythematosus: a case-control study in Southern Imbalance towards Th1 predominance is associated with acceleration of lupus- Sweden. Rheumatology 2002; 41: 563–571. like autoimmune syndrome in MRL mice. J Clin Invest 1996; 97: 1597–1604. 2 Cheong HS, Lee SO, Choi CB, Sung YK, Shin HD, Bae1 SC. MERTK polymorphisms 25 Graham RR, Kozyrev SV, Baechler EC, Reddy MV, Plenge1 RM, Bauer JW et al. A associated with risk of haematological disorders among Korean SLE patients. common haplotype of interferon regulatory factor 5 (IRF5) regulates splicing and Rheumatology 2007; 46: 209–214. expression and is associated with increased risk of systemic lupus erythemato- 3 Lopez-Nevot MA, Ramal L, Jimenez-Alonso J, Martı´n J. The inducible nitric oxide sus. Nat Genet 2006; 38: 550–555. synthase promoter polymorphism does not confer susceptibility to systemic 26 Graham RR, Kyogoku C, Sigurdsson S. Three functional variants of IFN regulatory lupus erythematosus. Rheumatology 2003; 42: 113–116. factor 5 (IRF5) define risk and protective haplotypes for human lupus. Proc Natl 4 Villarreal J, Crosdale D, Ollier W. Mannose binding lectin and Fc gamma RIIa Acad Sci USA 2007; 104: 6758–6763. (CD32) polymorphism in Spanish systemic lupus erythematosus patients. 27 Siu HO, Yang W, Lau CS. Association of a haplotype of IRF5 gene with systemic Rheumatology 2001; 40: 1009–1012. lupus erythematosus in chinese. J Rheumatol 2008; 35: 360–362.

& 2013 Macmillan Publishers Limited Genes and Immunity (2013) 133 – 146 Applied basic research of SLE W Sui et al 144 28 Kelly JA, Kelley JM, Kaufman KM, Kilpatrick J, Bruner1 GR, Merrill JT et al. Inter- 54 Nielsen C, Hansen D, Husby S, Jacobsen BB, Lillevang ST. Association of a feron regulatory factor-5 is genetically associated with systemic lupus erythe- putative regulatory polymorphism in the PD-1 gene with susceptibility to type 1 matosus in African Americans. Genes Immun 2008; 9: 187–194. diabetes. Tissue Antigens 2003; 62: 492–497. 29 Lo¨fgren SE, Yin H, Delgado-Vega AM. Promoter insertion/deletion in the IRF5 55 Kroner A, Mehling M, Hemmer B, Rieckmann P, Toyka KV, Maurer M et al. A PD-1 gene is highly associated with susceptibility to systemic lupus erythematosus in polymorphism is associated with disease progression in multiple sclerosis. distinct populations, but exerts a modest effect on gene expression in peripheral Ann Neurol 2005; 58: 50–57. blood mononuclear cells. J Rheumatol 2010; 37: 574–578. 56 Carroll M. The complement system in regulation of adaptive immunity. 30 Hikami K, Kawasaki A, Ito I. Association of a functional polymorphism in the Nat Immunol 2004; 5: 981–986. 3’-untranslated region of SPI1 with systemic lupus erythematosus. Arthritis 57 Marnell L, Mold C, Du Clos T. C-reactive protein: ligands, receptors and role in Rheum 2011; 63: 755–763. inflammation. Clin Immunol 2005; 117: 104–111. 31 Yee AA, Yin P, Siderovski DP, Mak1 TW, Litchfield DW, Arrowsmith CH. 58 Crawford D, Sanders C, Qin X, Smith J, Shephard C, Wong M et al. Genetic Cooperative interaction between the DNA-binding domains of PU.1 and IRF4. variation is associated with C-reactive protein levels in the Third National Health J Mol Biol 1998; 279: 1075–1083. and Nutrition Examination Survey. Circulation 2006; 114: 2458–2465. 32 Huang W, Horvath E, Eklund EA. PU.1, interferon regulatory factor (IRF) 2, and the 59 Carlson C, Aldred S, Lee P, Tracy R, Schwartz S, Rieder M et al. Polymorphisms interferon consensus sequence-binding protein (ICSBP/IRF8) cooperate to activate within the C-reactive protein (CRP) promoter region are associated with plasma NF1 transcription in differentiating myeloid cells. J Biol Chem 2007; 282: 6629–6643. CRP levels. Am J Hum Genet 2005; 77: 64–77. 33 Zhang L, Pagano JS. Structure and function of IRF-7. J Interferon Cytokine Res 60 Gregersen PK. Gaining insight into PTPN22 and autoimmunity. Nat Genet 2005; 2002; 22: 95–101. 37: 1300–1302. 34 Taylor KE, Remmers EF, Lee AT, Ortmann WA, Plenge RM, Tian C et al. Specificity 61 Bottini N, Musumeci L, Alonso A, Rahmouni S, Nika K, Rostamkhani M et al. A of the STAT4 genetic association for severe disease manifestations of systemic functional variant of lymphoid tyrosine phosphatase is associated with type I lupus erythematosus. PLoS Genet 2008; 4: e1000084. diabetes. Nat Genet 2004; 36: 337–338. 35 Kariuki SN, Kirou KA, MacDermott EJ. Cutting edge: autoimmune disease risk 62 Zhang JY, Zahir N, Jiang QH, Miliotis H, Heyraud S, Meng XW et al. The auto- variant of STAT4 confers increased sensitivity to IFN-a in lupus patients in vivo. immune disease-associated PTPN22 variant promotes calpain-mediated Lyp/Pep J Immunol 2009; 182: 34–38. degradation associated with lymphocyte and dendritic cell hyperresponsive- 36 Dunham I, Sargent CA, Trowsdale J, Campbell RD. Molecular mapping of the ness. Nat Genet 2011; 43: 902–907. human major histocompatibility complex by pulsed-field gel electrophoresis. 63 Arechiga AF, Habib T, He YT, Zhang X, Zhang ZY, Funk A et al. Cutting edge: The Proc Natl Acad Sci USA 1987; 84: 7237–7241. PTPN22 allelic variant associated with autoimmunity impairs B cell signaling. 37 Vassalli P. The pathophysiology of tumor necrosis factors. Annu Rev Immunol J Immunol 2009; 182: 3343–3347. 1992; 10: 411–452. 64 Menard L, Saadoun D, Isnardi I, Ng YS, Meyers G, Massad C et al. The PTPN22 38 Allen RD. Polymorphism of the human TNF-alpha promoter-random variation or allele encoding an R620W variant interferes with the removal of developing functional diversity. Mol Immunol 1999; 36: 1017–1027. autoreactive B cells in humans. J Clin Invest 2011; 121: 3635–3644. 39 Bouma G, Crusius JB, Oudkerk Pool M, Kolkman JJ, von Blomberg BM, 65 Zikherman JL, Hermiston M, Steiner D, Hasegawa K, Chan A, Weiss A et al. Kostense PJ et al. Secretion of tumour necrosis factor alpha and lymphotoxin PTPN22 deficiency cooperates with the CD45 E613R allele to break tolerance on alpha in relation to polymorphisms in the TNF genes and HLA-DR alleles. Rele- a non-autoimmune background. J Immunol 2009; 182: 4093–4106. vance for inflammatory bowel disease. Scand J Immunol 1996; 43: 456–463. 66 Mori M, Yamada R, Kobayashi K, Kawaida R, Yamamoto K. Ethnic differences in 40 Lee YH, Harley JB, Nath SK. Meta-analysis of TNF-alpha promoter-308A/G poly- allele frequency of autoimmune-disease-associated SNPs. J Hum Genet 2005; 50: morphism and SLE susceptibility. Eur J Hum Genet 2006; 14: 364–371. 264–266. 41 Lo´pez P, Gutie´rrez C, Sua´re A. IL-10 and TNFa genotypes in SLE. J Biomed Bio- 67 Gregersen PK, Lee HS, Batliwalla F, Begovich AB. PTPN22: setting thresholds for technol 2010; 2010: 838390. autoimmunity. Semin Immunol 2006; 18: 214–223. 42 Yap DYH, Lai KN. Cytokines and their roles in the pathogenesis of systemic lupus 68 Pradhan V, Borse V, Ghosh K. PTPN22 gene polymorphisms in autoimmune erythematosus: from basics to recent advances. J Biomed Biotechnol 2010; 2010: diseases with special reference to systemic lupus erythematosus disease sus- 365083. ceptibility. J Postgrad Med 2010; 56: 239–242. 43 Graham DSC, Graham RR, Manku H, Wong AK, Whittaker JC, Gaffney PM et al. 69 Aksoy R, Duman T, Keskin O, Du¨zgu¨n N. No association of PTPN22 R620W gene Polymorphism at the TNF superfamily gene TNFSF4 confers susceptibility to polymorphism with rheumatic heart disease and systemic lupus erythematosus. systemic lupus erythematosus. Nat Genet 2008; 40: 83–89. 44 Bates JS, Lessard CJ, Leon JM, Nguyen T, Battiest LJ, Rodgers J et al. Meta-analysis Mol Biol Rep 2011; 38: 5393–5396. and imputation identifies a 109kb risk haplotype spanning TNFAIP3 associated 70 Orru V, Tsai SJ, Rueda B, Fiorillo E, Stanford SM, Dasgupta J et al. A loss-of- with lupus nephritis and hematologic manifestations. Genes Immun 2009; 10: function variant of PTPN22 is associated with reduced risk of systemic lupus 470–477. erythematosus. Hum Mol Genet 2009; 18: 569–579. 45 Musone SL, Taylor KE, Lu TT, Nititham J, Ferreira RC, Ortmann W et al. Multiple 71 Manjarrez-Orduno N, Marasco E, Chung SA, Katz MS, Kiridly JF, Simpfendorfer KR polymorphisms in the TNFAIP3 region are independently associated with sys- et al. CSK regulatory polymorphism is associated with systemic lupus erythe- temic lupus erythematosus. Nat Genet 2008; 40: 1062–1064. matosus and influences B-cell signaling and activation. Nat Genet 2012; 44: 46 Han JW, Zheng HF, Cui Y, Nititham J, Ferreira RC, Ortmann W et al. Genome-wide 1227–1230. association study in a Chinese Han population identifies nine new susceptibility 72 Suggs MJ, Majithiab V, Lewisa RE, Cruse JM. HLA DRB*1503 allelic haplotype loci for systemic lupus erythematosus. Nat Genet 2009; 41: 1234–1237. predominance and associated immunodysregulation in systemic lupus erythe- 47 Gateva V, Sandling JK, Hom G, Taylor KE, Chung SA, Sun X et al. A large-scale matosus. Exp Mol Pathol 2011; 91: 548–562. replication study identifies TNIP1, PRDM1, JAZF1, UHRF1BP1 and IL10 as risk loci 73 Hedrich CM. Genetic variation and epigenetic patterns in autoimmunity. J Genet for systemic lupus erythematosus. Nat Genet 2009; 41: 1228–1233. Syndr Gene Ther. 2:0e2. doi:10.4172/2157-7412.10000e2. 48 Tokano Y, Morimoto S, Kaneko H, Amano H, Nozawa K, Takasaki Y et al. Levels of 74 Renaudineau Y, Youinou P. Epigenetics and autoimmunity with special emphasis IL-12 in the sera of patients with systemic lupus erythematosus (SLE)-relation to on methylation. Keio J Med 2011; 60: 10–16. Th1- and Th2-derived cytokines. Clin Exp Immunol 1999; 116: 169–173. 75 Richardson B, Scheinbart L, Strahler J, Gross L, Hanash S, Johnson M. Evidence for 49 Calvani N, Richards HB, Tucci M, Pannarale G, Silvestris F. Up-regulation of IL-18 impaired T cell DNA methylation in systemic lupus erythematosus and rheu- and predominance of a Th1 immune response is a hallmark of lupus nephritis. matoid arthritis. Arthritis Rheum 1990; 33: 1665–1673. Clin Exp Immunol 2004; 138: 171–178. 76 Corvetta A, Bitta RD, Luchetti MM, Pomponio G. 5-Methylcytosine content of 50 Nishimura H, Honjo T. PD-1: an inhibitory immunoreceptor involved in peri- DNA in blood, synovial mononuclear cells and synovial tissue from patients pheral tolerance. Trends Immunol 2001; 22: 265–268. affected by autoimmune rheumatic diseases. J Chromatogr 1991; 566: 481–491. 51 Prokunina L, Castillejo-Lo´pez C, Oberg F, Gunnarsson I, Berg L, Magnusson V 77 Hedrich CM, Tsokos GC. Epigenetic mechanisms in systemic lupus erythemato- et al. A regulatory polymorphism in PDCD1 is associated with susceptibility to sus and other autoimmune diseases. Trends Mol Med 2011; 12: 714–724. systemic lupus erythematosus in humans. Nat Genet 2002; 32: 666–669. 78 Zoua li M. Epigenetics in lupus. Ann NY Acad Sci 2011; 1217: 154–165. 52 Hiromine Y, Ikegami H, Fujisawa T, Nojima K, Kawabata Y, Nosoa S et al. 79 Jeffries MA, Dozmorov M, Tang Y, Merrill JT, Wren JD, Sawalha AH. Genomewide Trinucleotide repeats of programmed cell death-1 gene are associated with DNA methylation patterns in CD4 þ T cells from patients with systemic lupus susceptibility to type 1 diabetes mellitus. Metabolism 2007; 56: 905–909. erythematosus. Epigenetics 2011; 6: 593–601. 53 Sanghera DK, Manzi S, Bontempo F. Role of an intronic polymorphism in the 80 Rai K, Huggins IJ, James SR, Karpf AR, Jones DA, Cairns BR. DNA demethylation in PDCD1 gene with the risk of sporadic systemic lupus erythematosus and the zebrafish involves the coupling of a deaminase, a glycosylase, and gadd45. Cell occurrence of antiphospholipid antibodies. Hum Genet 2004; 115: 393–398. 2008; 135: 1201–1212.

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited Applied basic research of SLE W Sui et al 145 81 Zhao M, Sun YM, Gao F, Wu XY, Tang JL, Yin H et al. Epigenetics and SLE: RFX1 108 Remoli ME, Ragimbeau J, Giacomini E, Gafa V, Severa M, Lande R et al. downregulation causes CD11a and CD70 overexpression by altering epigenetic NF-{kappa}B is required for STAT-4 expression during dendritic cell maturation. modifications in lupus CD4 þ T cells. J Autoimmun 2010; 35: 58–69. J Leukoc Biol 2007; 81: 355–363. 82 Dieker J, Muller S. Epigenetic histone code and autoimmunity. Clin Rev Allergy 109 Kyttaris VC, Wang Y, Juang YT, Weinstein A, Tsokos GC. Increased Levels of Immunol 2010; 39: 78–84. NF-ATc2 differentially regulate CD154 and IL-2 genes in T cells from patients 83 Dai Y, Zhang L, Hu C, Zhang Y. Genome-wide analysis of histone H3 lysine 4 with systemic lupus erythematosus. J Immunol 2007; 178: 1960–1966. trimethylation by ChIP-chip in peripheral blood mononuclear cells of systemic 110 Westphal RS, Anderson KA, Means AR, Wadzinski BE. A signaling complex of lupus erythematosus patients. Clin Exp Rheumatol 2010; 28: 158–168. Ca2 þ calmodulin-dependent protein kinase IV and protein phosphatase 2A. 84 Clayton AL, Hazzalin CA, Mahadevan LC. Enhanced histone acetylation and Science 1998; 280: 1258–1261. transcription: a dynamic perspective. Mol Cell 2006; 23: 289–296. 111 Juang YT, Rauen T, Wang Y, Ichinose K, Benedyk K, Tenbrock K et al. 85 Zhang Z, Song L, Maurer K, Petri MA, Sullivan KE. Global H4 acetylation analysis Transcriptional activation of the cAMP-responsive modulator promoter in by ChIP-chip in systemic lupus erythematosus monocytes. Genes Immun 2010; human T cells is regulated by protein phosphatase 2A-mediated 11: 124–133. dephosphorylation of SP-1 and reflects disease activity in patients with systemic 86 Forster N, Gallinat S, Jablonska J, Weiss S, Elsa¨sser HP, Lutz W. p300 protein lupus erythematosus. J Biol Chem 2011; 286: 1795–1801. acetyltransferase activity suppresses systemic lupus erythematosus-like auto- 112 Tenbrock K, Juang YT, Tolnay M, Tsokos GC. The cyclic adenosine 5’-mono- immune disease in mice. J Immunol 2007; 178: 6941–6948. phosphate response element modulator suppresses IL-2 production in stimu- 87 Dieker JW, Fransen JH, van Bavel CC, Briand JP, Jacobs CW, Muller S et al. lated T cells by a chromatin-dependent mechanism. J Immunol 2003; 170: Apoptosis-induced acetylation of histones is pathogenic in systemic lupus ery- 2971–2976. thematosus. Arthritis Rheum 2007; 56: 1921–1933. 113 Levanon E, Eisenberg E, Yelin R, Nemzer S, Hallegger M, Shemesh R et al. Sys- 88 Flynt AS, Lai EC. Biological principles of microRNA-mediated regulation: shared tematic identification of abundant A-to-I editing sites in the human tran- themes amid diversity. Nat Rev Genet 2008; 9: 831–842. scriptome. Nat Biotechnol 2004; 22: 1001–1005. 89 Baltimore D, Boldin MP, O’Connell RM, Rao DS, Taganov KD. MicroRNAs: new 114 Laxminarayana D, Khan IU, Kammer GM. Transcript mutations of the alpha regulators of immune cell development and function. Nat Immunol 2008; 9: regulatory subunit of protein kinase A and up-regu-lation of the RNA editing 839–845. gene transcripts in lupus T lympho-cytes. Lancet 2002; 360: 842–849. 90 Tang Y, Luo X, Cui H, Ni X, Yuan M, Guo Y et al. MicroRNA-146A contributes to 115 Toyabe S, Kaneko U, Uchiyama M. Decreased DAP12 expression in natural killer abnormal activation of the type I interferon pathway in human lupus by tar- lymphocytes from patients with systemic lupus erythematosus is associated with geting the key signaling proteins. Arthritis Rheum 2009; 60: 1065–1075. increased transcript mutations. J Autoimmun 2004; 23: 371–378. 91 Te JL, Dozmorov IM, Guthridge JM, Nguyen KL, Cavett JW, Kelly JA et al. Iden- 116 Laxminarayana D, Khan IU, Kammer MG. Transcript mutations of the regulatory tification of unique microRNA signature associated with lupus nephritis. PLoS subunit of protein kinase A and up-regulation of the RNA-editing gene transcript One 2010; 5: e10344. in lupus T lymphocytes. Lancet 2002; 360: 842–849. 92 Dai Y, Huang YS, Tang M, Lv TY, Hu CX, Tan YH et al. Microarray analysis of 117 Laxminarayana D, Kammer MG. mRNA mutations of type 1 protein kinase A microRNA expression in peripheral blood cells of systemic lupus erythematosus regulatory subunitasubject with systemic lupus erythematosus. Int Immunol patients. Lupus 2007; 16: 939–946. 2000; 12: 1521–1529. 93 Dai Y, Sui WG, Lan HJ, Yan Q, Huang H, Huang YS. Comprehensive analysis of 118 Laxminarayana D, O’Rourke KS, Maas S, Olorenshaw I. Altered editing in RNA microRNA expression patterns in renal biopsies of lupus nephritis patients. editing adenosine deaminase ADAR2 gene transcripts of systemic lupus ery- Rheumatol Int 2009; 29: 749–754. thematosus T lymphocytes. Immunology 2007; 121: 359–369. 94 Elias Stagakis GB, Panayotis V. Identification of novel microRNA signatures linked 119 Orlowski RJ, O’Rourke KS, Olorenshaw I, Hawkins GA, Maas S, Laxminarayana D. to human lupus disease activity and pathogenesis: miR-21 regulates aberrant Altered editing in cyclic nucleotide phosphodiesterase 8A1 gene transcripts of T cell responses through regulation of PDCD4 expression. Ann Rheum Dis 2011; systemic lupus erythematosus T lymphocytes. Immunology 2008; 125: 408–419. 70: 1496–1506. 120 Wu JM, Xie FL, Qian K, Gibson AW, Edberg JC, Kimberly RP. FAS mRNA editing in 95 Zhao X, Tang Y, Qu B, Cui H, Wang S, Wang L et al. MicroRNA-125a contributes to human systemic lupus erythematosus. Hum Mutat 2011; 32: 1268–1277. elevated inflammatory chemokine RANTES levels via targeting KLF13 in systemic 121 Laxminarayana D, Khan IU, O’Rourke KS, Giri B. Induction of 150 kDa adenosine lupus erythematosus. Arthritis Rheum 2010; 62: 3425–3435. deaminase that acts on RNA (ADAR1) gene expression in normal T lymphocytes 96 Zhao S, Wang Y, Liang Y, Zhao M, Long H, Ding S et al. MicroRNA-126 regulates by physiological activation. Immunology 2007; 122: 623–633. DNA methylation in CD4 þ T cells and contributes to systemic lupus erythe- 122 Khan IU, Wallin R, Gupta RS, Kammer GM. Protein kinase A-catalyzed its matosus by targeting DNA methyltransferase 1. Arthritis Rheum 2011; 63: attachment to histone 2B in the T lymphocyte plasma membrane. Proc Natl Acad 1376–1386. 97 Ding S, Liang YS, Zhao M, Liang GP, Long H, Zhao S et al. Decreased miR-142-3p/5p Sci USA 1998; 95: 10425–10430. 123 Chen DS, Pace PE, Coombes RC, Ali S. Phosphorylation of human estrogen expression causes CD4 þ T cell activation and B cell hyperstimulation in systemic lupus erythematosus. Arthritis Rheum 2012; 64: 2953–2963. receptor alpha by protein kinase A regulation dimerization. Mol Cell Biol 1999; 98 Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG et al. The sequence 19: 1002–1015. of the human genome. Science 2001; 291: 1304–1352. 124 Yamamoto KK, Gonzalez GA, Biggs WH, Montminy MR. Phosphorylation-induced 99 Matthias P, Rolink AG. Transcriptional networks in developing and mature B cells. binding and transcriptional efficacy of nuclear factor CREB. Nature 1998; 334: Nat Rev Immunol 2005; 5: 497–508. 494–498. 100 Kawai T, Akira S. TLR signaling. Cell Death Differ 2006; 13: 816–825. 125 Brunner T, Mogil RJ, LaFace D, Yoo NJ, Mahboubi A, Echeverri F et al. Cell- 101 Sui WG, Lin H, Chen JJ, Ou ML, Dai Y. Comprehensive analysis of transcription autonomous FAS (CD95)/FAS-ligand interaction mediates activation-induced factor expression patterns in peripheral blood mononuclear cell of systemic apoptosis in T-cell hybridomas. Nature 1995; 373: 441–444. lupus erythematosus. Int J Rheum Dis 2012; 15: 212–219. 126 Johnson JM, Castle J, Garrett-Engele P, Kan Z, Loerch PM, Armour CD et al. 102 Kuo CC, Lin SC. Altered FOXO1 transcript levels in peripheral blood mononuclear Genome-wide survey of human alternative pre-mRNA splicing with exon junc- cells of systemic lupus erythematosus and rheumatoid arthritis patients. Mol Med tion microarrays. Science 2003; 302: 2141–2144. 2007; 13: 561–566. 127 Moulton VR, Tsokos GC. Alternative splicing factor/splicing factor 2 regulates the 103 Burgos P, Metz C, Bull P, Pincheira R, Massardo L, Erra´zuriz C et al. Increased expression of the z subunit of the human T cell receptor-associated CD3 com- expression of c-rel, from the NF-KB/Rel family, in T cells from patients with plex. J Biol Chem 2010; 285: 12490–12496. systemic lupus erythematosus. J Rheumatol 2000; 27: 116–127. 128 Nambiar MP, Enyedy EJ, Warke VG, Krishnan S, Dennis G, Kammer GM et al. 104 Dai RJ, Phillips RA, Karpuzoglu E. Estrogen regulates transcription factors STAT-1 Polymorphisms/mutations of TCR-zeta-chain promoter and 3’ untranslated and NF-kB to promote inducible nitric oxide synthase (iNOS) and inflammatory region and selective expression of TCR zeta-chain with an alternatively spliced 3’ responses. J Immunol 2009; 183: 6998–7005. untranslated region in patients with systemic lupus erythematosus. J Autoimmun 105 Foster MH, Kelley VR. Lupus nephritis: update on pathogenesis and disease 2001; 16: 133–142. mechanisms. Semin Nephrol 1999; 19: 173–181. 129 Chowdhury B, Tsokos CG, Krishnan S, Robertson J, Fisher CU, Warke RG et al. 106 Lit LC, Wong CK, Li EK, Tam LS, Lam CW, Lo YM. Elevated gene expression of Decreased stability and translation of T cell receptor zeta mRNA with an alter- Th1/Th2 associated transcription factors is correlated with disease activity in natively spliced 3’-untranslated region contribute to zeta chain down-regulation patients with systemic lupus erythematosus. J Rheumatol 2007; 34: 89–96. in patients with systemic lupus erythematosus. J Biol Chem 2005; 280: 107 Chan RW, Lai FM, Li EK, Tan LS, Chow KM, Li PK et al. Imbalance of Th1/Th2 18959–18966. transcription factors in patients with lupus nephritis. Rheumatology 2006; 45: 130 Tenbrock K, Juang YT, Gourley MF, Nambiar MP, Tsokos GC. Antisense cyclic 951–957. adenosine 5’-monophosphate response element modulator up-regulates IL-2 in

& 2013 Macmillan Publishers Limited Genes and Immunity (2013) 133 – 146 Applied basic research of SLE W Sui et al 146 T cells from patients with systemic lupus erythematosus. J Immunol 2002; 169: 157 Casciola-Rosen LA, Anhalt GJ, Rosen A. DNA-dependent protein kinase is one of 4147–4152. a subset of autoantigens specifically cleaved early during apoptosis. J Exp Med 131 Li Y, Harada T, Juang YT, Kyttaris VC, Wang Y, Zidanic M et al. Phosphorylated 1995; 182: 1625–1634. ERM is responsible for increased T cell polarization, adhesion, and migration in 158 Terzoglou AG, Routsias JG, Avrameas S, Moutsopoulos HM, Tzioufas AG. Pre- patients with systemic lupus erythematosus. J Immunol 2007; 178: 1938–1947. ferential recognition of the phosphorylated major linear B-cell epitope of La/SSB 132 Chen ZC. Advances in cancer proteomics study. Ai Zheng 2004; 23: 113–117. 349–368aa by anti-La/SSB autoantibodies from patients with systemic auto- 133 Dai Y, Hu C, Huang Y, Huang HY, Liu J, Lv T. A proteomic study of peripheral immune diseases. Clin Exp Immunol 2006; 144: 432–439. blood mononuclear cells in systemic lupus erythematosus. Lupus 2008; 17: 159 Terzoglou AG, Routsias JG, Moutsopoulos HM, Tzioufas AG. Post-translational 799–804. modifications of the major linear epitope 169–190aa of Ro60 kDa autoantigen 134 Wang LQ, Dai Y, Qi SW, Sun B, Wen JL, Zhang L et al. Comparative proteome alter the autoantibody binding. Clin Exp Immunol 2006; 146: 60–65. analysis of peripheral blood mononuclear cells in systemic lupus erythematosus 160 Tzioufas AG, Wassmuth R, Dafni UG, Guialis A, Haga HJ, Isenberg DA et al. with iTRAQ quantitative proteomics. Rheumatol Int 2012; 32: 585–593. Clinical, immunological, and immunogenetic aspects of autoantibody produc- 135 Wang L, Dai Y, Peng W, Qi S, Ouyang X, Tu Z. Differential expression of serine- tion against Ro/SSA, La/SSB and their linear epitopes in primary Sjogren’s syn- threonine kinase receptor-associated protein in patients with systemiclupus drome (pSS): a European multicentre study. Ann Rheum Dis 2002; 61: 398–404. erythematosus. Lupus 2011; 20: 921–927. 161 Hashii N, Kawasaki N, Itoh S, Nakajima Y, Kawanishi T, Yamaguchi T. Alteration of 136 Huang ZC, Shi YY, Cai B, Wang LL, Wu YK, Ying BW et al. Promising N-glycosylation in the kidney in a mouse model of systemic lupus erythema- diagnostic model for systemic lupus erythematosus using proteomic fingerprint tosus: relative quantification of N-glycans using an isotope-tagging method. technology. J Sichuan Univ (Med Sci Edi) V 2009; 40: 499–503. Immunology 2009; 126: 336–345. 137 Dai Y, Hu C, Wang L, zhang L, xiao X, Tan Y. Serum peptidome patients of 162 Gelsthorpe M, Pulumati M, McCallum C, Dang-Vu K, Tsubota SI. The putative cell human systemic lupus erythematosus based on magnetic bead separation cycle gene, enhancer of rudimentary, encodes a highly conserved protein found and MALDI-TOF mass spectrometry analysis. Scand J Rheumatol 2010; 39: in plants and animals. Gene 1997; 186: 189–195. 240–246. 163 Ouyang X, Dai Y, Wen JL, Wang LX. 1H NMR-based metabolomic study of 138 Mosley K, Tam FWK, Edwards RJ, Crozier J, Pusey CD, Lightstone L. Urinary metabolic profiling for systemic lupus erythematosus. Lupus 2011; 20: 1411–1420. proteomic profiles distinguish between active and inactive SLE nephritis. 164 Michel P, Eggert W, Albrecht-Nebe H, Grune T. Increased lipid peroxidation in Rheumatology 2006; 45: 1497–1504. children with autoimmune disease. Acta Paediat 1997; 86: 609–612. 139 Suzuki M, Ross GF, Wiers K, Nelson S, Bennett M, Passo MH et al. Identification of 165 Serban MG, Tanaseanu S. Lipid peroxidation in autoimmune systemic vasculi- a urinary proteomic signature for lupus nephritis in children. Pediatr Nephrol tides. Effect of corticoid treatment on lipid peroxidation. Antioxidant protection 2007; 22: 2047–2057. with vitamin E. Rom J Intern Med 1994; 32: 137–142. 140 Zhang XL, Jin M, Wu HF, Nadasdy T, Nadasdy G, Harris N et al. Biomarkers of 166 Komatsu N, Kodama K, Yamanouchi N, Okada S, Noda S, Nawata Y et al. Meta- lupus nephritis determined by serial urine proteomics. Kidney Int 2008; 74: bolic rate for glucose in systemic lupus erythematosus patients with psychiatric 799–807. symptoms. Eur Neurol 1999; 42: 41–48. 141 Anderton BH. Intermediate filaments: a family of homologous structures. 167 Alexander JJ, Zwingmann C, Quigg R. MRL/lpr mice have alterations in brain J Muscle Res Cell Motil 1981; 2: 141–166. metabolism as shown with [1H–13C] NMR spectroscopy. Neurochem Int 2005; 47: 142 Watanabe S, Wagatsuma K, Ichikawa E, Takahashi H. Abnormal distribution of 143–151. epidermal protein antigens in psoriatic epidermis. J Dermatol 1991; 18: 143–151. 168 Weiner SM, Otte A, Schumacher M, Brink I, Juengling FD, Sobanksi T et al. 143 Watanabe S, Wagatsuma K, Takahashi H. Immunohistochemical localization of Alterations of cerebral glucose metabolism indicate progress to severe mor- cytokeratins and involucrin in calcifying epithelioma: comparative studies with phological brain lesions in neuropsychiatric systemic lupus erythematosus. normal skin. Br J Dermatol 1994; 131: 506–513. Lupus 2000; 9: 386–389. 144 Fang S, Zeng FQ, Guo Q. Comparative proteomics analysis of cytokeratin 169 Laptook AR, Peterson J, Michael Porter A. Effects of lactic acid infusions and pH on and involucrin expression in lesions from patients with systemic lupus cerebral blood flow and metabolism. J Cereb Blood Flow Metabol 1988; 8: 193–200. erythematosus. Acta Biochim Biophys Sin 2008; 40: 989–995. 170 Romick-Rosendale LE, Brunner HI, Bennett MR, Mina R, Nelson S, Petri M et al. 145 Sui WG, Tang DE, Zou GM, Chen JJ, Ou ML, Zhang Y et al. Differential proteomic Identification of urinary metabolites that distinguish membranous lupus analysis of renal tissue in lupus nephritis using iTRAQ reagent technology. nephritis from proliferative lupus nephritis and focal segmental glomerulo- Rheumatol Int 2012; 32: 3537–3543. 146 Taher TE, Parikh K, Flores-Borja F, Mletzko S, Isenberg DA, Peppelenbosch MP sclerosis. Arthritis Res Ther 2011; 13: R199. ¨ et al. Protein phosphorylation and kinome profiling reveal altered regulation of 171 Christians U, Schmitz V, Schoning W, Bendrick-Peart J, Klawitter J, Haschke M multiple signaling pathways in B lymphocytes from patients with systemic lupus et al. Toxicodynamic therapeutic drug monitoring of immunosuppressants: erythematosus. Arthritis & Rheumatism 2010; 62: 2412–2423. promises, reality and challenges. Ther Drug Monit 2008; 30: 151–158. 147 Juang YT, Wang Y, Jiang G, Peng HB, Ergin S, Finnell M et al. PP2A dephos- 172 Quaggin SE, Kreidberg JA. Development of the renal glomerulus: good neigh- phorylates Elf-1 and determines the expression of CD3z and FcRg in human bors and good fences. Development 2008; 135: 609–620. systemic lupus erythematosus T cells. J Immunol 2008; 181: 3658–3664. 173 Hahn TJ, Hahn BH. Osteoporosis in patients with rheumatic diseases: principles 148 Juang YT, Solomou E, Rellahan B, Tsokos GC. Phosphorylation and O-linked of diagnosis and therapy. Semin Arthritis Rheum 1976; 6: 165–188. glycosylation of Elf-1 leads to its translocation to the nucleus and binding to the 174 Dykman TR, Gluck OS, Murphy WA, Hahn TJ, Hahn BH. Evaluation of factors promoter of the T cell receptor z chain. J Immunol 2002; 168: 2865–2871. associated with glucocorticoid–induced osteopenia in patients with rheumatic 149 Rellahan BL, Jensen JP, Weissman AM. Transcriptional regulation of the T cell diseases. Arthritis Rheum 1985; 28: 361–368. antigen receptor zeta subunit: identification of a tissue-restricted promoter. J Exp 175 Kalla AA, Fataar AF, Jessop SJ, Bewerunge L. Loss of trabecular bone mineral Med 1994; 180: 1529–1534. density in systemic lupus erythematosus. Arthritis Rheum 1993; 36: 1726–1734. 150 Honda S, Kobayashi T, Kajino K, Urakami S, Igawa M, Hino O. Ets protein Elf-1 176 Sels F, Dequeker J, Verwilghen J, Mbuyi-Muamba JM. SLE and osteoporosis: bidirectionally suppresses transcriptional activities of the tumor suppressor Tsc2 dependence and/or independence on glucocorticoids. Lupus 1996; 5: 89–92. gene and the repair-related Nth1 gene. Mol Carcinog 2003; 37: 122–129. 177 Kalla AA, Meyers OL, Parkyn ND, Kotze TJ. Osteoporosis screening-radio- 151 Doyle HA, Mamula MJ. Post-translational protein modifications in antigen grammetry revisited. Br J Rheumatol 1989; 28: 511–517. recognition and autoimmunity. Trends Immunol 2001; 22: 443–449. 178 Dhillon VB, Davies MC, Hall ML, Round JM, Ell PJ, Jacobs HS et al. Assessment of 152 Utz PJ, Hottelet M, Schur PH, Anderson P. Proteins phosphorylated during stress- the effect of oral corticosteroids on bone mineral density in systemic lupus induced apoptosis are common targets for autoantibody production in patients erythematosus: a preliminary study with dual energy x ray absorptiometry. Ann with systemic lupus erythematosus. J Exp Med 1997; 185: 843–854. Rheum Dis 1990; 49: 624–626. 153 Doyle HA, Mamula MJ. Autoantigenesis: the evolution of protein modifications in 179 Hansen M, Halberg P, Kollerup G, Pedersen-Zbinden B, Hørslev-Petersen K, autoimmune disease. Curr Opin Immunol 2012; 24: 112–118. Hyldstrup L et al. Bone metabolism in patients with systemic lupus erythema- 154 Golan TD, Elkon KB, Gharavi AE, Krueger JG. Enhanced membrane binding of tosus: effect of disease activity and glucocorticoid treatment. Scand J Rheumatol autoantibodies to cultured keratinocytes of systemic lupus erythematosus 1998; 27: 197–206. patients after ultraviolet B/ultraviolet A irradiation. J Clin Invest 1992; 90: 1067–1076. 180 Teichmann J, Lange U, Stracke H, Federlin K, Bretzel RG. Bone metabolism and 155 Zhang C, Ao A, Seth A, Schlossman SF. A mitochondrial membrane protein bone mineral density of systemic lupus erythematosus at the time of diagnosis. defined by a novel monoclonal antibody is preferentially detected in apoptotic Rheumatol Int 1999; 18: 137–140. cells. J Immunol 1996; 157: 3980–3987. 181 Redlich K, Ziegler S, Kiener HP, Spitzauer S, Stohlawetz P, Bernecker P et al. Bone 156 Mevorach D, Zhou J, Elkon K. Immunization of mice with apoptotic cells induces mineral density and biochemical parameters of bone metabolism in female low levels of autoantibodies. Arthritis Rheum 1996; 39: 143s. patients with systemic lupus erythematosus. Ann Rheum Dis 2000; 59: 308–310.

Genes and Immunity (2013) 133 – 146 & 2013 Macmillan Publishers Limited