Transcription factor motif enrichment in whole transcriptome analysis identifies STAT4 and BCL6 as the most prominent binding motif in systemic juvenile idiopathic arthritis

Transcription factor motif enrichment in whole transcriptome analysis identifies STAT4 and BCL6... Background: The term systemic juvenile idiopathic arthritis (sJIA) describes an autoinflammatory condition characterized by arthritis and severe systemic inflammation, which in later stages can transform into interleukin (IL)- 17-driven autoimmune arthritis. IL-1 antagonists have been used with good efficacy in the early stages of sJIA. Methods: A whole transcriptome analysis of peripheral blood RNA samples was performed in six patients with sJIA and active systemic disease, before initiating treatment with the IL-1β receptor antagonist anakinra, and after induction of inactive disease, compared with a single-sample control cohort of 21 patients in several clinical stages of sJIA activity. Whole transcriptomes were compared longitudinally and interindividually including gene ontology and motif enrichment analysis of differentially expressed genes. Results: There were 741 transcripts were identified using a threshold with a p value <0.01 and a fold change > 2. HLADRB1 and CD74 were identified as the most strongly upregulated genes in inactive compared to active disease; CD177 expression was significantly enhanced in active disease compared to inactive disease. Motif enrichment analysis revealed STAT4, BCL6, and STAT3 as the most prominent transcription factors that were present during active disease. In addition, strong upregulation of the major histocompatability complex II (MHCII) ligand CD74 was found in both active and inactive sJIA compared to healthy controls. Conclusion: Using transcription factor motif enrichment, this study identifies novel putative pathways in sJIA (STAT4, BCL6) implicating B cell activation at an earlier stage than predicted in refractory disease. The implication of BCL-6 dependent pathways argues for occurrence of autoimmunity early within the process of sJIA chronification. Transcriptional regulation of HLA-DRB1, a recently described independent genetic risk factor, in combination with its cooperating partner CD74 in patients where sJIA is confirmed, supports pathogenic involvement in alterations in antigen presentation during sJIA. Keywords: Juvenile systemic arthritis, Juvenile idiopathic arthritis, RNA expression, HLA-DRB1, CD74, CD177 * Correspondence: huegle.boris@rheuma-kinderklinik.de German Center for Pediatric and Adolescent Rheumatology, Gehfeldstrasse 24, 82467 Garmisch-Partenkirchen, Germany Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 2 of 11 Background while taking non-steroidal antirheumatic drugs and/or corti- Systemic juvenile idiopathic arthritis (sJIA) is an inflam- costeroids, (3) achievement of subsequent inactive disease, matory disease with severe systemic inflammation lead- and (4) available samples from at least two time points, i.e. at ing to marked morbidity and mortality. Children with initial manifestation prior to treatment with IL-1 antagonists, sJIA usually present with fever, arthritis, and a typical and at a point of inactive disease on treatment with IL-1 an- rash and can have a highly variable outcome [1]. Despite tagonists. Additional samples from recurring disease flares sJIA being listed as a subtype of juvenile idiopathic arth- and subsequent periods of inactivity were also acquired, ritis, it is currently considered to be of autoinflammatory where available. rather than autoimmune origin [2, 3]. At least two phe- A second cohort, cohort II, was set up for verification notypes of sJIA can be identified by their clinical course: purposes. These were patients with only a single sample patients with monocyclic sJIA with only a single phase available, who had active systemic disease, active polyarti- of severe systemic inflammation and concomitant arth- cular disease without systemic activity or inactive disease. ritis, and patients developing chronic disease with recur- Inactive disease was defined according to the American rent flares of autoinflammation and severe arthritis [1]. College of Rheumatology provisional criteria [8]. Active Peripheral blood mononuclear cells (PBMC) and mono- systemic disease was defined as presence of fever, rash, cytic cells from patients with active sJIA have been ana- serositis, splenomegaly, or generalized lymphadenopathy lyzed in gene expression studies and have been shown to attributable to JIA, with or without concurrent arthritis at have a variable RNA expression pattern in patients with that time. Active polyarticular disease was defined as pres- sJIA compared to healthy controls, patients with other ence of arthritis without fever, rash, serositis, splenomeg- autoimmune diseases, and patients with other subtypes of aly, or generalized lymphadenopathy. JIA, with uneven results [4, 5]. With the exception of a re- cent study looking at RNA expression in patient samples Data collection from two clinical trials of canakinumab, these studies have A retrospective chart survey was used to extract demo- so far only used a cross-sectional approach, and patients graphic data, including date of first manifestation, date of with sJIA were not stratified according to their disease ac- diagnosis, total joint count at diagnosis, laboratory param- tivity [6]. eters at diagnosis including C-reactive protein, ferritin and The objective of this study was to perform longitudinal thrombocyte count and the initial dose of prednisolone whole transcriptome analysis of children with sJIA re- and anakinra. fractory to conventional, non-biological therapy before treatment with IL-1 antagonists and after achieving in- Sample preparation active disease, to identify transcriptional patterns and Whole blood was drawn during active disease before possible novel markers involved in the pathogenesis of start of treatment with anakinra, after achievement of in- the disease. active disease and on subsequent visits. Peripheral blood was collected using PAXgene Blood RNA tubes (Qiagen, Methods Valencia, CA, USA). RNA was extracted at the collection Patients site using RNeasy columns (Qiagen, Valencia, CA, USA), Clinical data and patient samples were acquired from the then stored at − 20 °C. RNA quantity and quality was AID-Net database, a German registry and biobank that pro- assessed using a Pico100 Picodrop ul Spectrophotometer spectively collects information and biomaterials from pa- (Picodrop, Saffron Walden, UK). tients with autoinflammatory syndromes including periodic fever syndromes and sJIA. Written informed consent was Expression analysis obtained from the patients prior to inclusion. Healthy aged- Genome-wide transcriptome analyses were performed in co- matched controls undergoing elective surgery were recruited hort I using Gene Chip® Human HTA 2.0 arrays (Affymetrix, from the pediatric department of the university hospital Santa Clara, CA, USA). Prior to analysis, RNA quality was RWTH Aachen. Further information on data safety and assessed using the RNA 6000 Nano Assay with the 2100 pseudonymization has been published previously [7]. A Bioanalyzer (Agilent, Santa Clara, CA, USA). Samples for the single-center sample of all patients with sJIA at the German HTA 2.0 arrays were prepared and hybridized to the arrays Center for Pediatric and Adolescent Rheumatology was ob- according to the Affymetrix WT Plus Kit manual. Briefly, for tained between January 2010 and March 2015. All patients each sample, 100 ng of total RNA was reverse transcribed who met the following criteria were included in cohort I, into complementary DNA (cDNA) using a random hexamer which was set up for longitudinal analysis: (1) confirmed oligonucleotide tagged with a T7 promoter sequence. After diagnosis of sJIA according to International League of Asso- second-strand synthesis, double-strand cDNA was used as a ciations for Rheumatology (ILAR) criteria [3], (2) initial treat- template for amplification with T7 RNA polymerase to ob- ment with IL-1 antagonists after persistence of symptoms tain antisense cRNA. Random hexamers and deoxyribose Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 3 of 11 adenine triphosphates (dNTPs) spiked out with deoxyuridine Statistical analysis triphosphate (dUTP) were then used to reverse transcribe Clinical data were analyzed using descriptive statistics. the cRNA into single-stranded sense strand cDNA. The Statistical analysis was performed using SPSS version 21.0 cDNA was then fragmented with uracil DNA glycosylase (SPSS Inc., Chicago, USA). Microarray data were imported and apurinic/apyrimidic endonuclease 1. Fragment size was into GeneSpring GX 7.3.1 software (Agilent Technologies, checked using the 2100 Bioanalyzer and ranged from 50 to Santa Clara, USA) and preprocessed using robust multi- 200 bp. Fragmented sense cDNA was biotin-end-labeled chip analysis (RMA), followed by normalization of each with terminal deoxynucleotidyl transferase (TdT) and probes probe to the median of all samples. Distance-weighted dis- were hybridized to the Gene 2.0 arrays at 45 °C for 16 h with crimination was used to align the centroids of predefined 60 rpms.Hybridizedarrayswerewashedand stainedona groups (12–16) to control for batch-to-batch variation. Fluidics Station 450 (program FS450 0002) and scanned on a Gene Ontology (GO)-based analysis of biological GeneChip® Scanner 3000 7G (both Affymetrix). Raw process was performed using AltAnalyze 2.1.0 software image data were analyzed with Affymetrix® Expression (altanalyze.org); significance values were between an ad- ConsoleTM Software (Affymetrix, USA), and gene ex- justed p-value of 3.17e-08 and 0.0005. pression intensities were normalized and summarized with arobustmultiarrayaverage algorithm[9]. Transcripts Motif enrichment analysis that were expressed differently more than twofold Transcription factor motif enrichment on upregulated/ with a raw p value <0.01 between the sample groups downregulated genes was performed using MotifMatch were categorized as regulated. Enrichment analysis for (www.regulatory-genomics.org). In short, this software Wiki pathways was performed using WebGestalt [10]. searches for binding sites on the promoter region of all For the enrichment analysis only genes that changed candidate genes (1 kbp upstream). Motifs were ob- at least twofold with a p value <0.01 between patients tained from the Jaspar database. It then performs a with active disease and those with inactive disease Fisher exact test to evaluate if the proportion of bind- were taken into consideration. ing sites in the gene sets is higher than expected by chance. The p-values were adjusted for multiple test- Reverse transcription-polymerase chain reaction (RT-PCR) ingusing theBenjamini-Hochbergmethod. For verification purposes, RT-PCR for several genes was performed in cohort I and II. The genes selected were Results chosen both due to the results of the expression analysis Study population and previous descriptions in the literature [6, 11]. cDNA Cohort I included longitudinal samples of six children was generated from RNA using RevertAid H Minus First with sJIA, with all patients having at least a sample pair Strand cDNA Synthesis Kit (Thermo Fisher Scientific, prior to treatment with anakinra and with inactive dis- USA) according to the manufacturer’s instructions. ease on anakinra; samples were also available from two Standard real-time PCR was carried out on TaqMan patients during a flare after withdrawal of anakinra. with the ABI prism 7300 real-time PCR systems (Ap- Cohort II consisted of single samples from eight patients plied Biosystems by Life Technologies, Germany) using with systemically active sJIA, five patients with sJIA with the DNA intercalating dye SYBR Green Kit (Eurogentec, a polyarticular flare but no clinical signs of systemic ac- Germany). The housekeeping gene used was ribosomal tivity, and eight patients with inactive sJIA. The clinical protein L (RPL). The following primer sequences were and demographic data at time of diagnosis in both co- used: for HLA-DRB1, TTC TTC AAT GGG ACG GAG horts are given in Table 1. All patients were of Caucasian CG (forward) and TTC CAG TAC TCA GCG TCA GG origin and initially showed a typical clinical picture of (reverse); for CD74, TTA TCT CCA ACA ATG AGC sJIA with rash, fever, and arthritis, and typical bloodwork AAC T (forward) and ACA GGA AGT AGG CGG TGG with elevated inflammatory markers. All patients in co- T (reverse); for CD177, CAT GTG TGG AAG GTG hort I reacted rapidly to treatment with IL-1 antagonists, TCC GA (forward) and CTT GGG GTC CGC TCT achieving an inactive disease state within days to weeks. CAA TG (reverse); and for RPL, AGGTATGCTGCCC CACAAAAC (forward), TGTAGGCTTCAGACGCAC Patients with sJIA and inactive disease have differences in GAC (reverse). RNA expression profiles compared to patients with active The relative quantification method was applied and disease and disease flares delta cycle threshold (ΔCt) values were determined by Using a p value <0.01 and fold change > 2, 741 tran- subtracting the Ct of the housekeeping gene (RPL) from scripts encoding for 481 known genes were identified the Ct of the target gene for each sample, respectively. (Additional file 1: Table S1) that were significantly differ- Fold change was compared in active disease and inactive ently expressed in inactive disease compared to active disease in the same individual using the ΔCt method. disease (both on initial presentation and during disease Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 4 of 11 Table 1 Demographic, clinical and laboratory characteristics of the study cohorts Patient Cohort I (n = 6) Cohort II (n =8) Cohort II (n =5) Cohort II (n = 8) Active disease, systemic Active disease, polyarticular Inactive disease Gender 5 male, 1 female 5 male, 3 female 2 male, 3 female 5 male, 3 female Age at diagnosis, years (median, range) 5.3 (1.8–12.9) 6.8 (0.4–17.4) 6.0 (0.5–15.3) 3.5 (0.4–9.7) Time since first symptoms (median, range) 71 days (47–107 days) 18.6 months (1.6–109.7 months) 86.9 months (7.4–198.3 months) 105.1 months (52.6–195.8 months) Number of active joints (median, range) 2 (1–18) 3.5 (0–10) 3 (2–4) 0 (0–0) 3 3 3 3 3 3 3 3 Platelet count (median, range) 520,500/mm (474,000–557,000/mm ) 352,000/mm (151,000–649,000/mm ) 276,000/mm (195,000–395,000/mm ) 306,500/mm (208,000–394,000/mm ) Rheumatoid factor, negative 6/6 (100%) 8/8 (100%) 5/5 (100%) 8/8 (100%) Ferritin, μg/l (median, range) 414 (32–1785 ) 754 (234–9980 μg/l) n.d. n.d. C-reactive protein, mg/dl (median, range) 11.84 (4.87–25.6) 3.65 (1.10–26.4) 0.07 (0.03–6.57) 0.11 (0.03–0.63) Initial prednisolone dose, mg/kg (median, 1.6 (0–2.08) n.a. n.a. n.a. range) Initial anakinra dose, mg/kg (median, 1.63 (1.02–2.5) n.a. n.a. n.a. range) n.a. not applicable, n.d. not determined Pat6 id Pat5 id Pat4 id Pat3 id Pat2 id Pat1 id Pat6 ad Pat5 ad Pat4 ad Pat3 ad Pat2 ad Pat1 ad Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 5 of 11 Table 2 Ontology-based analysis of the most significantly flare), of which most were associated with immune- me- regulated genes diated processes (Table 2, Figs. 1 and 2). Of these, genes, Inflammatory response (GO:0006954) 239 were downregulated while 242 were upregulated in active disease. Using fold change > 3 as a more stringent Regulation of T cell differentiation (GO:0045580) criterion, more than 100 genes still remained. Gene Acute inflammatory response (GO:0002526) Ontology (GO)-based analysis favored pathways of the Positive regulation of T cell differentiation in thymus (GO:0033089) innate immune response as the most significantly repre- Response to molecule of bacterial origin (GO:0002237) sented pathways in active disease (Table 2). Some of the Regulation of lymphocyte differentiation (GO:0045619) highly regulated genes (HLA-DRB1, CD74, CD177) were Regulation of T cell activation (GO:0050863) confirmed using RT-PCR, as described below. Additional data on ANXA3/annexin A 3, a gene locus where a SNP Regulation of syncytium formation by plasma membrane fusion (GO:0060142) within the gene has been identified as a risk factor in rheumatoid arthritis, and IL-1 receptor associated kinase Positive regulation of leukocyte activation (GO:0002696) 3 (IRAK3), are presented in Additional file 2: Figures S1 Positive regulation of cell activation (GO:0050867) and S2 [12]. Activation of innate immune response (GO:0002218) Of note, patient 5 had a markedly different expression Response to lipopolysaccharide (GO:0032496) pattern in active disease, and patients 2 and 5 had a dif- Detection of external biotic stimulus (GO:0098581) ferent expression pattern in inactive disease (Fig. 1). Pa- Activation of immune response (GO:0002253) tient 5 differs to the other patients in being the only female patient. The marked difference in gene expres- Negative regulation of immune response (GO:0050777) sion patterns is, however, more likely due to the fact that Shown are the 15 biological processes that are the most stringent according to the p value in the Gene Ontology (GO) analysis patient 5 received three methylprednisolone pulses prior to the RNA sample being drawn. Patient 2 also received Tenbrock 171005 heatMap sorted single values −2.4 −1.8 −1.2 −0.6 0.0 0.6 1.2 1.8 2.4 Differential Expression (log2) PROK2 IRAK3 DSC2 ORM2 ORM1 S100A8 WDFY3-AS1 CMTM2 BASP1 NFE2 NABP1 NCF4 SLC22A15 TET2 ITGAM FCGR2A PADI4 NLRC4 MGAM2 CR1 DYSF PYGL FAM129A SLC2A3 GAS7 LIMK2 SLC22A4 PFKFB3 PFKFB3 HCAR3 NAIP HRH2 ADM PGS1 SIRPD RNF24 FAM157B GLT1D1 TECPR2 SIGLEC9 BCL6 ADGRG3 NAIP NAIP IL1R1 ADAM9 PLXNC1 PLXNC1 GPR141 MAPK14 MCTP2 TLR1 SRPK1 CR1L SLC25A37 GCA VNN3 CPEB4 VNN1 KREMEN1 MGAM IRAK3 TLR5 HP STEAP4 CREB5 WDFY3 GK ACSL1 ANXA3 CD177 MCEMP1 ALPL CD177P1 SYTL2 YME1L1 CD2 TRGV3 ESYT1 PRKCH TESPA1 CD74 SKAP1 TRAJ1 TRGC2 TARP KLRG1 FAM117B MKL2 SIRPG IL7R CYSLTR2 CDKAL1 CCDC6 CLSTN1 SPOCK2 ARID5B AES HLA-DRB1 HLA-DRB1 Fig. 1 Heatmap with hierarchical clustering of patients (Pat) with active disease (ad) and inactive disease (id). Shown are all genes with fold change > 3 Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 6 of 11 Fig. 2 Regulated genes in patients with active disease (ad) versus inactive disease (id) using fold change (FC) > 2 and a p value <0.01. Red dots include genes with fold change > 2 in active disease, blue dots with a fold change <− 2. Names of highly regulated and significant genes are depicted in the graph using different colors one pulse of methylprednisolone, which the others did binding in the promoter of upregulated or downregu- not. lated genes. The most prominent binding motif in up- As for the changes in inactive disease, patients 1 and 5 regulated genes in active sJIA was STAT4, followed by required ongoing IL-1 blockade and reacted with systemic BCL6 and STAT3 (adjusted p value <0.005) (Fig. 3). signs to withdrawal, while the other patients were able to Moreover BCL6 was also upregulated on the array in ac- taper and discontinue the IL-1 antagonists (both anakinra tive disease (fold change 5.15, p = 0.002, Fig. 2), while and canakinumab) over the course of the next year. STAT4 and STAT3 were not. Motif enrichment analysis identifies STAT4, BCL6 and HLA-DRB1 is upregulated in inactive compared to active STAT3 as the most prominent binding sites during active disease disease Initially we examined transcripts shown to be regulated In order to identify pathways that might be of relevance during sJIA, to validate that the patients under study exhib- in sJIA, we performed transcription factor (TF) motif ited the features described previously for this condition. For enrichment analysis to evaluate which TFs are preferably example, there was a fold change of 4.76 (p =0.006) in Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 7 of 11 Fig. 3 Motifs of the most regulated transcription factor binding sites in the whole transcriptome of patients with active systemic juvenile idiopathic arthritis (sJIA) compared to inactive sJIA, analyzed using MotifMatch S100A8 in our patients, as expected [13, 14]. Our analysis change 4.73, p=0.036, Fig. 2). However, and more strikingly, revealed HLA-DRB1 as the most strongly upregulated gene in cohort II CD74 was much higher expressed in patients in inactive compared to activedisease (foldchange6.8, p = with systemic activity as well as inactive disease compared to 0.003, Fig. 2). This observation was confirmed using RT- age-matched healthy controls using RT-PCR (Fig. 5). The PCR in the longitudinal per-patient sample analysis, while protein encoded by this gene is the invariant chain of the we did not find upregulation in cohort II compared to HLA-DR complex, which is associated with class II major healthy controls due to high individual expression differ- histocompatibility complex (MHC) including HLADRB1 ences (Fig. 4). Other HLA class II genes, for example HLA- and is an important chaperone that regulates antigen presen- DRB4 (fold change − 2.92, p = 0.008, Fig. 2), HLA-DRB3 tation for immune response. and HLA-DRB6, were found to be downregulated. CD177 is strongly upregulated in active disease CD74 is regulated in sJIA irrespective of disease activity CD177 was upregulated in patients with active disease CD74 was also upregulated in inactive compared to compared to those with inactive disease or healthy con- active disease on an individual per patient base (fold trols (fold change 37, p =0.007, Fig. 2), which has recently Fig. 4 Expression of HLA-DRB1 in active (AD) versus inactive (ID) systemic juvenile idiopathic arthritis using reverse transcription PCR. Left graph, fold change (n-fold) of longitudinal samples before and after treatment with anakinra, with inactive disease set to level 1 (*p < 0.05). Right graph, delta cycle threshold (ΔCt) values (relative expression values related to ribosomal protein L (RPL)) in active systemic disease (cohorts I and II), active polyarticular disease, and inactive disease/remission, and controls (all cohort II) Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 8 of 11 refractory to conventional therapy, we were able to show differential regulation of a variety of genes, comparing inactive disease on IL-1 antagonist treatment with active disease within the same patient before treatment. Regulation of specific genes has been confirmed in an independent cohort of single samples from patients with sJIA with either active or inactive disease. Previously, a specific sJIA signature of RNA expres- sion has been described in 44 patients with sJIA in various stages of the disease [4]. In this study, 12 patients with systemic and arthritic symptoms dis- played a specific expression pattern compared to healthy controls and patients with polyarticular (non- systemic) flare. Similar results were found by Barnes et al. in a cohort of 21 patients with sJIA, which also demonstrated a specific pattern of sJIA different from other subtypes of JIA [5]. A recent study by Brachat Fig. 5 Expression of CD74 using reverse transcription PCR: delta et al. on sJIA RNA samples from two studies using cycle threshold (ΔCt) values (relative expression values related to canakinumab also demonstrated a large number of ribosomal protein L) in controls, active systemic disease, active differentially expressed genes in patients prior to and polyarticular, and inactive disease (all ***p < 0.001) after initiation of anti-IL-1 treatment [6]. Similar to these preceding studies, the overall specific sJIA been described in the study by Brachat et al. and confirms signature consists of regulation of the inflammatory their findings using RT-PCR (Fig. 6)[6]. This gene en- response and innate immune pathways including the codes a glycosyl-phosphatidylinositol-linked cell surface IL-1 pathway, IL-6 and toll-like receptor 1, but also modu- glycoprotein that plays a role in neutrophil activation. The lation of lymphocyte differentiation and response, with an protein can bind platelet endothelial cell adhesion emphasis on T cells (Table 2). A recent large genome-wide molecule-1 and function in neutrophil transmigration. association study also demonstrated a framework of pathophysiological pathways that appears to be specific Discussion for sJIA [15]. By using comparative array analysis of RNA expression Transcription factor motif enrichment analysis in in a cohort of sJIA patients in different stages of disease patients with active and inactive disease identified STAT4, BCL6, and STAT3 as the most prominent motifs within the regulated genes. STAT4 is mainly expressed in myeloid cells and is the transcription factor downstream of IL-12, which has been identified as a potential biomarker in sJIA, but has not been discussed as a potential therapeutic target so far [16– 18]. We have previously shown that the likelihood of the occurrence of a polymorphism enhancing IL-12 expression was somewhat higher in patients with sJIA compared to other forms of JIA, which points toward IL-12 being more prominent in sJIA than previously thought [19]. In our array analysis, IL-12 was not dif- ferentially expressed; however, the IL-23 receptor, which is involved in IL-12 downstream signaling was downregulated in active disease (fold change 2.5, p = 0.012, Fig. 2), which suggests a physiological reaction towards the proinflammatory state. The identification of BCL6 as the second most prom- Fig. 6 Expression of CD177 using reverse transcription PCR: delta inent binding motif is more striking: BCL6 is a lineage cycle threshold (ΔCt) values (relative expression values related to transcription factor for follicular T helper (Tfh) cells, a ribosomal protein L) in controls, active disease, and inactive cell type that is important for B cellular responses. This disease (***p < 0.001) cell type is especially important for autoimmune arthritis Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 9 of 11 mediated by gut bacteria [20]. BCL6 expression has been [29]. However, RNA expression patterns of the shared shown to be sustained by IL-6 signaling in patients with epitope are variable, and apparently dependent on the rheumatoid arthritis (RA), and specific targeting of IL-6 copy number [27]. using tocilizumab in patients with RA results in a signifi- CD74 is the invariant chain of HLA-DR and therefore cant reduction of circulating Tfh cells; IL-21 production a cooperation partner on the cell surface, where it critic- by Tfh cells was also correlated with reduced expression ally regulates antigen presentation. It also serves as a cell of antibody-producing plasmablasts [21]. BCL6 was also surface receptor for the cytokine macrophage migration directly upregulated on the array, which supports the hy- inhibitory factor (MIF) which, when bound to the pothesis that this TF is of major importance (fold encoded protein, initiates survival pathways and cell pro- change 5.15, p = 0.002, Fig. 2), and would also argue for liferation. In cooperation with IL-12, MIF is important even earlier involvement of autoimmunity with B cell re- for survival in children with malaria and trypanosoma sponses in the course of the disease [22]. There is some infection [30, 31]. CD74 also interacts with amyloid pre- evidence of B cell-driven autoimmunity in the later cursor protein and suppresses the production of beta phases of sJIA, in which autoantibody production has amyloid [32]. The high expression of CD74 even in an been demonstrated [23]. Our findings here indicate that inactive disease state could serve as a possible candidate B cell activation might already be present during the first for disease markers for sJIA, and possibly - in the con- phase of disease where autoinflammation is the most text of MIF - even for therapeutic intervention. prominent feature. This might potentially determine An additional gene that was highly upregulated in the whether the patients develop a polyarticular course later. active stages of sJIA was CD177, confirming recent re- In contrast, finding STAT3 as the third most prominent sults [6]. CD177 codes for human neutrophil antigen 2 transcription factor is unsurprising, since STAT3 signals (HNA-2), also called NB1, a cell surface glycoprotein downstream of IL-6, a well-known therapeutic target [33]. HNA-2 expression is highly variable in humans and using tocilizumab in sJIA [24, 25]. In the study by Bra- is lacking in approximately 3–5% of the North American chat et al., where canakinumab was used for IL-1ß population [34]. HNA-2 plays an important role in mye- blockade, IL-6 declined by day 3 and remained sup- loid cell proliferation and function of neutrophils, in- pressed over time [6]. STAT3 was not differentially cluding transendothelial migration [35]. It has also been expressed on the array; however, STAT3 is usually acti- associated with anti-neutrophil-cytoplasmic antibody vated by phosphorylation and not by transcriptional (ANCA)-associated vasculitides, where CD177 has been regulation. Nevertheless, IL-6 signals via JAK3, which proposed as a receptor of mPR3 on the neutrophil sur- then activates STAT3 and is downstream of the IL-6 re- face [36]. Interestingly, CD177 has also been observed as ceptor, and it was found to be enhanced (fold change 2. the most upregulated parameter in a microarray study of 5, p = 0.006). Vice versa, the antagonist of STAT3 signal- purified neutrophils from patients with septic shock. ing is the suppressor of cytokine signaling 3 (SOCS3), However, since we performed our analysis in whole which was highly induced in active disease (fold blood, the numbers of neutrophils could also have had a change 8.5, p = 0.012, Fig. 2), probably to counterbalance considerable effect on our findings. There was, however, the overwhelming immune stimulation. no difference in CD177 transcription between active sys- HLA-DRB1 and its cooperation partner CD74 were temic and active polyarticular disease, while neutrophil upregulated in the active disease stage, and CD74 was counts were significantly lower in polyarticular disease upregulated in the inactive disease stage in patients with than in systemic flares (data not shown). Nevertheless, sJIA in this study. Certain genotypes of HLA-DRB1, in light of the large variations in CD177 expression in most notably HLA-DRB1*11, have been found to be controls and throughout the patient cohorts, these data strongly associated with sJIA in a large recent study of have to be interpreted with caution. 982 patients across nine different populations, and also This is a small study with a limited number of patients in a fine-mapping study of the HLA locus comparing it examined longitudinally, even if the results are con- to other forms of JIA [11, 15, 26]. However, no expres- firmed in a second cohort. As RNA was extracted from sion studies of HLA-DRB1 in sJIA have been performed whole blood rather than sorted cells due to the con- to date. HLA-DRB1 has also previously been demon- straints of a biobank, and given that a number of genes strated to be strongly associated with early, severe RA discussed, especially CD177, are expressed by neutro- [27]. RA-associated HLA-DRB1 alleles have conserved phils, changes in neutrophil numbers could have signifi- amino acid sequences in position 70–74 of the molecule. cantly impacted the results. However, not all genes This molecular structure is termed the shared epitope, described here are predominantly expressed in neutro- with a variety of hypotheses explaining its function [28]. phils and CD177 was also consistently upregulated in An association with the shared epitope has also been patients with polyarticular flares and low neutrophil found in children with rheumatoid-factor positive JIA numbers. As variable cell numbers are a valid point of Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 10 of 11 criticism of preceding studies as well, confirmation using and BV participated in the design and coordination of the study and helped draft the manuscript. IGG performed the genetic and statistical analysis. All sorted cells is a logical next step in researching gene ex- authors read and approved the final manuscript. pression patterns in sJIA [4–6]. The strength of this study is the longitudinal examination of patients who Ethics approval and consent to participate Clinical data and patient samples were acquired from the AID-Net database, have undergone a consistent institutional treatment a German registry and biobank that prospectively collects information and protocol with IL-1 agonists in different stages of their biomaterials from patients with autoinflammatory syndromes including peri- disease. odic fever syndromes and sJIA. Written informed consent was obtained from the patients prior to inclusion. Conclusions Competing interests By using this longitudinal analysis, our study identifies The authors declare that they have no competing interests. novel pathways (STAT4 and BCL6) that might be of relevance in sJIA and indicates strong upregulation of Publisher’sNote HLA-DRB1 in cooperation with CD74 in patients with Springer Nature remains neutral with regard to jurisdictional claims in sJIA in inactive disease upon treatment with IL-1 antag- published maps and institutional affiliations. onists. This provides the first functional confirmation of Author details a previous study, which identified HLA-DRB1 as a risk 1 German Center for Pediatric and Adolescent Rheumatology, Gehfeldstrasse factor in sJIA. Additionally CD177 was confirmed as a 24, 82467 Garmisch-Partenkirchen, Germany. Department of Pediatrics, Universitätsklinikum Aachen, Aachen, Germany. IZKF Research Group new marker in sJIA. Studies with larger patient cohorts Bioinformatics, RWTH Aachen Medical Faculty, Aachen, Germany. University using flow cytometry for protein expression are neces- Medical Center Utrecht, Utrecht, Netherlands. sary to confirm these results. Received: 4 January 2018 Accepted: 23 April 2018 Additional files References Additional file 1: Table S1. List of regulated genes with fold change 1. Singh-Grewal D, Schneider R, Bayer N, Feldman BM. Predictors of disease (FC) > 2 in patients with active disease (ad) versus inactive disease (id) course and remission in systemic juvenile idiopathic arthritis: significance of showing signal intensity in both disease states, gene symbol and early clinical and laboratory features. Arthritis Rheum. 2006;54:1595–601. description. (XLSX 60 kb) 2. Mellins ED, Macaubas C, Grom AA. Pathogenesis of systemic juvenile idiopathic arthritis: some answers, more questions. Nat Rev Rheumatol. Additional file 2: Figure S1. Expression of ANXA3 using RT-PCR: delta 2011;7:416–26. Ct values (relative expression values related to RPL) in controls, active 3. Petty RE, Southwood TR, Manners P, Baum J, Glass DN, Goldenberg J, He X, systemic or polyarticular disease and inactive disease (***p < 0.001, Maldonado-Cocco J, Orozco-Alcala J, Prieur AM, et al. International League ****p < 0.0001). Figure S2. Expression of IRAK3 using RT-PCR: delta Ct of Associations for Rheumatology classification of juvenile idiopathic values (relative expression values related to RPL) in controls, active arthritis: second revision, Edmonton, 2001. J Rheumatol. 2004;31:390–2. systemic or polyarticular disease and inactive disease (**p < 0.01). 4. Allantaz F, Chaussabel D, Stichweh D, Bennett L, Allman W, Mejias A, Ardura (DOCX 619 kb) M, Chung W, Smith E, Wise C, et al. Blood leukocyte microarrays to diagnose systemic onset juvenile idiopathic arthritis and follow the Abbreviations response to IL-1 blockade. J Exp Med. 2007;204:2131–44. bp: Base pairs; cDNA: Complementary DNA; Ct: Cycle threshold; HNA- 5. Barnes MG, Grom AA, Thompson SD, Griffin TA, Pavlidis P, Itert L, Fall N, 2: Human neutrophil antigen 2; GO: Gene Ontology; IRAK: IL-1 receptor Sowders DP, Hinze CH, Aronow BJ, et al. Subtype-specific peripheral blood associated kinase; IL: Interleukin; ILAR: International League of Associations gene expression profiles in recent-onset juvenile idiopathic arthritis. Arthritis for Rheumatology; MHC: Major histocompatibility complex; MIF: Macrophage Rheum. 2009;60:2102–12. migration inhibitory factor; PBMC: Peripheral blood mononuclear cells; 6. Brachat AH, Grom AA, Wulffraat N, Brunner HI, Quartier P, Brik R, McCann L, RA: Rheumatoid arthritis; RPL: Ribosomal protein L; RT-PCR: Reverse Ozdogan H, Rutkowska-Sak L, Schneider R, et al. Early changes in gene transcription PCR; sJIA: Systemic juvenile idiopathic arthritis; TF: Transcription expression and inflammatory proteins in systemic juvenile idiopathic factor; Tfh: Follicular T helper (cells) arthritis patients on canakinumab therapy. Arthritis Res Ther. 2017;19:13. 7. Lainka E, Bielak M, Hilger V, Basu O, Neudorf U, Wittkowski H, Holzinger D, Acknowledgements Roth J, Niehues T, Foell D. Translational research network and patient We thank Nienke Ter Haar for careful review of the manuscript. registry for auto-inflammatory diseases. Rheumatology (Oxford). 2011; 50:237–42. Funding 8. Wallace CA, Giannini EH, Huang B, Itert L, Ruperto N, Childhood Arthritis This study was supported by the Interdisciplinary Center for Clinical Research Rheumatology Research A, Pediatric Rheumatology Collaborative Study G, (IZKF) Aachen and UCAN-AC (Understanding Childhood Arthritis Network Paediatric Rheumatology International Trials O. American College of Aachen) and an unrestricted grant from Novartis AG to the German Center Rheumatology provisional criteria for defining clinical inactive disease in for Pediatric and Adolescent Rheumatology. The AID-Net database is select categories of juvenile idiopathic arthritis. Arthritis Care Res (Hoboken). supported by the Federal Ministry of Education and Research (BMBF project 2011;63:929–36. 01GM08104, 01GM1112D). 9. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density Availability of data and materials oligonucleotide array probe level data. Biostatistics. 2003;4:249–64. The datasets used and/or analyzed during the current study are available 10. Wang J, Duncan D, Shi Z, Zhang B. WEB-based gene set analysis toolkit from the corresponding author on reasonable request. (WebGestalt): update 2013. Nucleic Acids Res. 2013;41:W77–83. 11. Ombrello MJ, Remmers EF, Tachmazidou I, Grom A, Foell D, Haas JP, Martini Authors’ contributions A, Gattorno M, Ozen S, Prahalad S, et al. HLA-DRB1*11 and variants of the BH, JPH, and KT conceived of the study and drafted the manuscript. AS and MHC class II locus are strong risk factors for systemic juvenile idiopathic NF participated in data collection and reviewed the manuscript. KO, BD, FT, arthritis. Proc Natl Acad Sci U S A. 2015;112:15970–5. Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 11 of 11 12. Okada Y, Terao C, Ikari K, Kochi Y, Ohmura K, Suzuki A, Kawaguchi T, factor (MIF) protein and blood mononuclear cell MIF transcripts in children Stahl EA, Kurreeman FA, Nishida N, et al. Meta-analysis identifies nine with Plasmodium falciparum malaria. Clin Immunol. 2006;119:219–25. new loci associated with rheumatoid arthritis in the Japanese 31. Terrazas CA, Huitron E, Vazquez A, Juarez I, Camacho GM, Calleja EA, population. Nat Genet. 2012;44:511–6. Rodriguez-Sosa M. MIF synergizes with Trypanosoma cruzi antigens to 13. Gerss J, Roth J, Holzinger D, Ruperto N, Wittkowski H, Frosch M, Wulffraat N, promote efficient dendritic cell maturation and IL-12 production via p38 Wedderburn L, Stanevicha V, Mihaylova D, et al. Phagocyte-specific S100 MAPK. Int J Biol Sci. 2011;7:1298–310. proteins and high-sensitivity C reactive protein as biomarkers for a risk- 32. Matsuda S, Matsuda Y, D'Adamio L. CD74 interacts with APP and suppresses adapted treatment to maintain remission in juvenile idiopathic arthritis: a the production of Abeta. Mol Neurodegener. 2009;4:41. comparative study. Ann Rheum Dis. 2012;71:1991–7. 33. Stroncek DF, Caruccio L, Bettinotti M. CD177: a member of the Ly-6 gene superfamily involved with neutrophil proliferation and polycythemia vera. J 14. Holzinger D, Frosch M, Kastrup A, Prince FH, Otten MH, Van Suijlekom-Smit Transl Med. 2004;2:8. LW, ten Cate R, Hoppenreijs EP, Hansmann S, Moncrieffe H, et al. The toll- 34. Matsuo K, Lin A, Procter JL, Clement L, Stroncek D. Variations in the like receptor 4 agonist MRP8/14 protein complex is a sensitive indicator for expression of granulocyte antigen NB1. Transfusion. 2000;40:654–62. disease activity and predicts relapses in systemic-onset juvenile idiopathic 35. Bayat B, Werth S, Sachs UJ, Newman DK, Newman PJ, Santoso S. Neutrophil arthritis. Ann Rheum Dis. 2012;71:974–80. transmigration mediated by the neutrophil-specific antigen CD177 is 15. Ombrello MJ, Arthur VL, Remmers EF, Hinks A, Tachmazidou I, Grom AA, influenced by the endothelial S536N dimorphism of platelet endothelial cell Foell D, Martini A, Gattorno M, Ozen S, et al. Genetic architecture adhesion molecule-1. J Immunol. 2010;184:3889–96. distinguishes systemic juvenile idiopathic arthritis from other forms of 36. Hu N, Mora-Jensen H, Theilgaard-Monch K, Doornbos-van der Meer B, juvenile idiopathic arthritis: clinical and therapeutic implications. Ann Huitema MG, Stegeman CA, Heeringa P, Kallenberg CG, Westra J. Differential Rheum Dis. 2017;76:906–13. expression of granulopoiesis related genes in neutrophil subsets 16. de Jager W, Hoppenreijs EP, Wulffraat NM, Wedderburn LR, Kuis W, Prakken distinguished by membrane expression of CD177. PLoS One. 2014;9:e99671. BJ. Blood and synovial fluid cytokine signatures in patients with juvenile idiopathic arthritis: a cross-sectional study. Ann Rheum Dis. 2007;66:589–98. 17. Gohar F, Kessel C, Lavric M, Holzinger D, Foell D. Review of biomarkers in systemic juvenile idiopathic arthritis: helpful tools or just playing tricks? Arthritis Res Ther. 2016;18:163. 18. Yilmaz M, Kendirli SG, Altintas D, Bingol G, Antmen B. Cytokine levels in serum of patients with juvenile rheumatoid arthritis. Clin Rheumatol. 2001; 20:30–5. 19. Eberhardt CS, Haas JP, Girschick H, Schwarz T, Morbach H, Rosen-Wolff A, Foell D, Dannecker G, Schepp C, Ganser G, et al. No association of IL-12p40 pro1.1 polymorphism with juvenile idiopathic arthritis. Pediatr Rheumatol Online J. 2015;13:61. 20. Teng F, Klinger CN, Felix KM, Bradley CP, Wu E, Tran NL, Umesaki Y, Wu HJ. Gut microbiota drive autoimmune arthritis by promoting differentiation and migration of Peyer's patch T follicular helper cells. Immunity. 2016;44:875–88. 21. Chavele KM, Merry E, Ehrenstein MR. Cutting edge: circulating plasmablasts induce the differentiation of human T follicular helper cells via IL-6 production. J Immunol. 2015;194:2482–5. 22. Nigrovic PA, Mannion M, Prince FH, Zeft A, Rabinovich CE, van Rossum MA, Cortis E, Pardeo M, Miettunen PM, Janow G, et al. Anakinra as first-line disease-modifying therapy in systemic juvenile idiopathic arthritis: report of forty-six patients from an international multicenter series. Arthritis Rheum. 2011;63:545–55. 23. Hugle B, Hinze C, Lainka E, Fischer N, Haas JP. Development of positive antinuclear antibodies and rheumatoid factor in systemic juvenile idiopathic arthritis points toward an autoimmune phenotype later in the disease course. Pediatr Rheumatol Online J. 2014;12:28. 24. Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, Schaper F. Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J. 2003;374:1–20. 25. Yokota S, Imagawa T, Mori M, Miyamae T, Aihara Y, Takei S, Iwata N, Umebayashi H, Murata T, Miyoshi M, et al. Efficacy and safety of tocilizumab in patients with systemic-onset juvenile idiopathic arthritis: a randomised, double-blind, placebo-controlled, withdrawal phase III trial. Lancet. 2008;371:998–1006. 26. Hinks A, Bowes J, Cobb J, Ainsworth HC, Marion MC, Comeau ME, Sudman M, Han B, Juvenile Arthritis Consortium for I, Becker ML, et al. Fine-mapping the MHC locus in juvenile idiopathic arthritis (JIA) reveals genetic heterogeneity corresponding to distinct adult inflammatory arthritic diseases. Ann Rheum Dis. 2017;76:765–72. 27. Kerlan-Candon S, Combe B, Vincent R, Clot J, Pinet V, Eliaou JF. HLA-DRB1 gene transcripts in rheumatoid arthritis. Clin Exp Immunol. 2001;124:142–9. 28. Holoshitz J. The rheumatoid arthritis HLA-DRB1 shared epitope. Curr Opin Rheumatol. 2010;22:293–8. 29. Prahalad S, Thompson SD, Conneely KN, Jiang Y, Leong T, Prozonic J, Brown MR, Ponder LA, Angeles-Han ST, Vogler LB, et al. Hierarchy of risk of childhood-onset rheumatoid arthritis conferred by HLA-DRB1 alleles encoding the shared epitope. Arthritis Rheum. 2012;64:925–30. 30. Awandare GA, Hittner JB, Kremsner PG, Ochiel DO, Keller CC, Weinberg JB, Clark IA, Perkins DJ. Decreased circulating macrophage migration inhibitory http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Arthritis Research & Therapy Springer Journals

Transcription factor motif enrichment in whole transcriptome analysis identifies STAT4 and BCL6 as the most prominent binding motif in systemic juvenile idiopathic arthritis

Free
11 pages

Loading next page...
 
/lp/springer_journal/transcription-factor-motif-enrichment-in-whole-transcriptome-analysis-gN9GFiYlsA
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Rheumatology; Orthopedics
eISSN
1478-6362
D.O.I.
10.1186/s13075-018-1603-2
Publisher site
See Article on Publisher Site

Abstract

Background: The term systemic juvenile idiopathic arthritis (sJIA) describes an autoinflammatory condition characterized by arthritis and severe systemic inflammation, which in later stages can transform into interleukin (IL)- 17-driven autoimmune arthritis. IL-1 antagonists have been used with good efficacy in the early stages of sJIA. Methods: A whole transcriptome analysis of peripheral blood RNA samples was performed in six patients with sJIA and active systemic disease, before initiating treatment with the IL-1β receptor antagonist anakinra, and after induction of inactive disease, compared with a single-sample control cohort of 21 patients in several clinical stages of sJIA activity. Whole transcriptomes were compared longitudinally and interindividually including gene ontology and motif enrichment analysis of differentially expressed genes. Results: There were 741 transcripts were identified using a threshold with a p value <0.01 and a fold change > 2. HLADRB1 and CD74 were identified as the most strongly upregulated genes in inactive compared to active disease; CD177 expression was significantly enhanced in active disease compared to inactive disease. Motif enrichment analysis revealed STAT4, BCL6, and STAT3 as the most prominent transcription factors that were present during active disease. In addition, strong upregulation of the major histocompatability complex II (MHCII) ligand CD74 was found in both active and inactive sJIA compared to healthy controls. Conclusion: Using transcription factor motif enrichment, this study identifies novel putative pathways in sJIA (STAT4, BCL6) implicating B cell activation at an earlier stage than predicted in refractory disease. The implication of BCL-6 dependent pathways argues for occurrence of autoimmunity early within the process of sJIA chronification. Transcriptional regulation of HLA-DRB1, a recently described independent genetic risk factor, in combination with its cooperating partner CD74 in patients where sJIA is confirmed, supports pathogenic involvement in alterations in antigen presentation during sJIA. Keywords: Juvenile systemic arthritis, Juvenile idiopathic arthritis, RNA expression, HLA-DRB1, CD74, CD177 * Correspondence: huegle.boris@rheuma-kinderklinik.de German Center for Pediatric and Adolescent Rheumatology, Gehfeldstrasse 24, 82467 Garmisch-Partenkirchen, Germany Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 2 of 11 Background while taking non-steroidal antirheumatic drugs and/or corti- Systemic juvenile idiopathic arthritis (sJIA) is an inflam- costeroids, (3) achievement of subsequent inactive disease, matory disease with severe systemic inflammation lead- and (4) available samples from at least two time points, i.e. at ing to marked morbidity and mortality. Children with initial manifestation prior to treatment with IL-1 antagonists, sJIA usually present with fever, arthritis, and a typical and at a point of inactive disease on treatment with IL-1 an- rash and can have a highly variable outcome [1]. Despite tagonists. Additional samples from recurring disease flares sJIA being listed as a subtype of juvenile idiopathic arth- and subsequent periods of inactivity were also acquired, ritis, it is currently considered to be of autoinflammatory where available. rather than autoimmune origin [2, 3]. At least two phe- A second cohort, cohort II, was set up for verification notypes of sJIA can be identified by their clinical course: purposes. These were patients with only a single sample patients with monocyclic sJIA with only a single phase available, who had active systemic disease, active polyarti- of severe systemic inflammation and concomitant arth- cular disease without systemic activity or inactive disease. ritis, and patients developing chronic disease with recur- Inactive disease was defined according to the American rent flares of autoinflammation and severe arthritis [1]. College of Rheumatology provisional criteria [8]. Active Peripheral blood mononuclear cells (PBMC) and mono- systemic disease was defined as presence of fever, rash, cytic cells from patients with active sJIA have been ana- serositis, splenomegaly, or generalized lymphadenopathy lyzed in gene expression studies and have been shown to attributable to JIA, with or without concurrent arthritis at have a variable RNA expression pattern in patients with that time. Active polyarticular disease was defined as pres- sJIA compared to healthy controls, patients with other ence of arthritis without fever, rash, serositis, splenomeg- autoimmune diseases, and patients with other subtypes of aly, or generalized lymphadenopathy. JIA, with uneven results [4, 5]. With the exception of a re- cent study looking at RNA expression in patient samples Data collection from two clinical trials of canakinumab, these studies have A retrospective chart survey was used to extract demo- so far only used a cross-sectional approach, and patients graphic data, including date of first manifestation, date of with sJIA were not stratified according to their disease ac- diagnosis, total joint count at diagnosis, laboratory param- tivity [6]. eters at diagnosis including C-reactive protein, ferritin and The objective of this study was to perform longitudinal thrombocyte count and the initial dose of prednisolone whole transcriptome analysis of children with sJIA re- and anakinra. fractory to conventional, non-biological therapy before treatment with IL-1 antagonists and after achieving in- Sample preparation active disease, to identify transcriptional patterns and Whole blood was drawn during active disease before possible novel markers involved in the pathogenesis of start of treatment with anakinra, after achievement of in- the disease. active disease and on subsequent visits. Peripheral blood was collected using PAXgene Blood RNA tubes (Qiagen, Methods Valencia, CA, USA). RNA was extracted at the collection Patients site using RNeasy columns (Qiagen, Valencia, CA, USA), Clinical data and patient samples were acquired from the then stored at − 20 °C. RNA quantity and quality was AID-Net database, a German registry and biobank that pro- assessed using a Pico100 Picodrop ul Spectrophotometer spectively collects information and biomaterials from pa- (Picodrop, Saffron Walden, UK). tients with autoinflammatory syndromes including periodic fever syndromes and sJIA. Written informed consent was Expression analysis obtained from the patients prior to inclusion. Healthy aged- Genome-wide transcriptome analyses were performed in co- matched controls undergoing elective surgery were recruited hort I using Gene Chip® Human HTA 2.0 arrays (Affymetrix, from the pediatric department of the university hospital Santa Clara, CA, USA). Prior to analysis, RNA quality was RWTH Aachen. Further information on data safety and assessed using the RNA 6000 Nano Assay with the 2100 pseudonymization has been published previously [7]. A Bioanalyzer (Agilent, Santa Clara, CA, USA). Samples for the single-center sample of all patients with sJIA at the German HTA 2.0 arrays were prepared and hybridized to the arrays Center for Pediatric and Adolescent Rheumatology was ob- according to the Affymetrix WT Plus Kit manual. Briefly, for tained between January 2010 and March 2015. All patients each sample, 100 ng of total RNA was reverse transcribed who met the following criteria were included in cohort I, into complementary DNA (cDNA) using a random hexamer which was set up for longitudinal analysis: (1) confirmed oligonucleotide tagged with a T7 promoter sequence. After diagnosis of sJIA according to International League of Asso- second-strand synthesis, double-strand cDNA was used as a ciations for Rheumatology (ILAR) criteria [3], (2) initial treat- template for amplification with T7 RNA polymerase to ob- ment with IL-1 antagonists after persistence of symptoms tain antisense cRNA. Random hexamers and deoxyribose Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 3 of 11 adenine triphosphates (dNTPs) spiked out with deoxyuridine Statistical analysis triphosphate (dUTP) were then used to reverse transcribe Clinical data were analyzed using descriptive statistics. the cRNA into single-stranded sense strand cDNA. The Statistical analysis was performed using SPSS version 21.0 cDNA was then fragmented with uracil DNA glycosylase (SPSS Inc., Chicago, USA). Microarray data were imported and apurinic/apyrimidic endonuclease 1. Fragment size was into GeneSpring GX 7.3.1 software (Agilent Technologies, checked using the 2100 Bioanalyzer and ranged from 50 to Santa Clara, USA) and preprocessed using robust multi- 200 bp. Fragmented sense cDNA was biotin-end-labeled chip analysis (RMA), followed by normalization of each with terminal deoxynucleotidyl transferase (TdT) and probes probe to the median of all samples. Distance-weighted dis- were hybridized to the Gene 2.0 arrays at 45 °C for 16 h with crimination was used to align the centroids of predefined 60 rpms.Hybridizedarrayswerewashedand stainedona groups (12–16) to control for batch-to-batch variation. Fluidics Station 450 (program FS450 0002) and scanned on a Gene Ontology (GO)-based analysis of biological GeneChip® Scanner 3000 7G (both Affymetrix). Raw process was performed using AltAnalyze 2.1.0 software image data were analyzed with Affymetrix® Expression (altanalyze.org); significance values were between an ad- ConsoleTM Software (Affymetrix, USA), and gene ex- justed p-value of 3.17e-08 and 0.0005. pression intensities were normalized and summarized with arobustmultiarrayaverage algorithm[9]. Transcripts Motif enrichment analysis that were expressed differently more than twofold Transcription factor motif enrichment on upregulated/ with a raw p value <0.01 between the sample groups downregulated genes was performed using MotifMatch were categorized as regulated. Enrichment analysis for (www.regulatory-genomics.org). In short, this software Wiki pathways was performed using WebGestalt [10]. searches for binding sites on the promoter region of all For the enrichment analysis only genes that changed candidate genes (1 kbp upstream). Motifs were ob- at least twofold with a p value <0.01 between patients tained from the Jaspar database. It then performs a with active disease and those with inactive disease Fisher exact test to evaluate if the proportion of bind- were taken into consideration. ing sites in the gene sets is higher than expected by chance. The p-values were adjusted for multiple test- Reverse transcription-polymerase chain reaction (RT-PCR) ingusing theBenjamini-Hochbergmethod. For verification purposes, RT-PCR for several genes was performed in cohort I and II. The genes selected were Results chosen both due to the results of the expression analysis Study population and previous descriptions in the literature [6, 11]. cDNA Cohort I included longitudinal samples of six children was generated from RNA using RevertAid H Minus First with sJIA, with all patients having at least a sample pair Strand cDNA Synthesis Kit (Thermo Fisher Scientific, prior to treatment with anakinra and with inactive dis- USA) according to the manufacturer’s instructions. ease on anakinra; samples were also available from two Standard real-time PCR was carried out on TaqMan patients during a flare after withdrawal of anakinra. with the ABI prism 7300 real-time PCR systems (Ap- Cohort II consisted of single samples from eight patients plied Biosystems by Life Technologies, Germany) using with systemically active sJIA, five patients with sJIA with the DNA intercalating dye SYBR Green Kit (Eurogentec, a polyarticular flare but no clinical signs of systemic ac- Germany). The housekeeping gene used was ribosomal tivity, and eight patients with inactive sJIA. The clinical protein L (RPL). The following primer sequences were and demographic data at time of diagnosis in both co- used: for HLA-DRB1, TTC TTC AAT GGG ACG GAG horts are given in Table 1. All patients were of Caucasian CG (forward) and TTC CAG TAC TCA GCG TCA GG origin and initially showed a typical clinical picture of (reverse); for CD74, TTA TCT CCA ACA ATG AGC sJIA with rash, fever, and arthritis, and typical bloodwork AAC T (forward) and ACA GGA AGT AGG CGG TGG with elevated inflammatory markers. All patients in co- T (reverse); for CD177, CAT GTG TGG AAG GTG hort I reacted rapidly to treatment with IL-1 antagonists, TCC GA (forward) and CTT GGG GTC CGC TCT achieving an inactive disease state within days to weeks. CAA TG (reverse); and for RPL, AGGTATGCTGCCC CACAAAAC (forward), TGTAGGCTTCAGACGCAC Patients with sJIA and inactive disease have differences in GAC (reverse). RNA expression profiles compared to patients with active The relative quantification method was applied and disease and disease flares delta cycle threshold (ΔCt) values were determined by Using a p value <0.01 and fold change > 2, 741 tran- subtracting the Ct of the housekeeping gene (RPL) from scripts encoding for 481 known genes were identified the Ct of the target gene for each sample, respectively. (Additional file 1: Table S1) that were significantly differ- Fold change was compared in active disease and inactive ently expressed in inactive disease compared to active disease in the same individual using the ΔCt method. disease (both on initial presentation and during disease Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 4 of 11 Table 1 Demographic, clinical and laboratory characteristics of the study cohorts Patient Cohort I (n = 6) Cohort II (n =8) Cohort II (n =5) Cohort II (n = 8) Active disease, systemic Active disease, polyarticular Inactive disease Gender 5 male, 1 female 5 male, 3 female 2 male, 3 female 5 male, 3 female Age at diagnosis, years (median, range) 5.3 (1.8–12.9) 6.8 (0.4–17.4) 6.0 (0.5–15.3) 3.5 (0.4–9.7) Time since first symptoms (median, range) 71 days (47–107 days) 18.6 months (1.6–109.7 months) 86.9 months (7.4–198.3 months) 105.1 months (52.6–195.8 months) Number of active joints (median, range) 2 (1–18) 3.5 (0–10) 3 (2–4) 0 (0–0) 3 3 3 3 3 3 3 3 Platelet count (median, range) 520,500/mm (474,000–557,000/mm ) 352,000/mm (151,000–649,000/mm ) 276,000/mm (195,000–395,000/mm ) 306,500/mm (208,000–394,000/mm ) Rheumatoid factor, negative 6/6 (100%) 8/8 (100%) 5/5 (100%) 8/8 (100%) Ferritin, μg/l (median, range) 414 (32–1785 ) 754 (234–9980 μg/l) n.d. n.d. C-reactive protein, mg/dl (median, range) 11.84 (4.87–25.6) 3.65 (1.10–26.4) 0.07 (0.03–6.57) 0.11 (0.03–0.63) Initial prednisolone dose, mg/kg (median, 1.6 (0–2.08) n.a. n.a. n.a. range) Initial anakinra dose, mg/kg (median, 1.63 (1.02–2.5) n.a. n.a. n.a. range) n.a. not applicable, n.d. not determined Pat6 id Pat5 id Pat4 id Pat3 id Pat2 id Pat1 id Pat6 ad Pat5 ad Pat4 ad Pat3 ad Pat2 ad Pat1 ad Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 5 of 11 Table 2 Ontology-based analysis of the most significantly flare), of which most were associated with immune- me- regulated genes diated processes (Table 2, Figs. 1 and 2). Of these, genes, Inflammatory response (GO:0006954) 239 were downregulated while 242 were upregulated in active disease. Using fold change > 3 as a more stringent Regulation of T cell differentiation (GO:0045580) criterion, more than 100 genes still remained. Gene Acute inflammatory response (GO:0002526) Ontology (GO)-based analysis favored pathways of the Positive regulation of T cell differentiation in thymus (GO:0033089) innate immune response as the most significantly repre- Response to molecule of bacterial origin (GO:0002237) sented pathways in active disease (Table 2). Some of the Regulation of lymphocyte differentiation (GO:0045619) highly regulated genes (HLA-DRB1, CD74, CD177) were Regulation of T cell activation (GO:0050863) confirmed using RT-PCR, as described below. Additional data on ANXA3/annexin A 3, a gene locus where a SNP Regulation of syncytium formation by plasma membrane fusion (GO:0060142) within the gene has been identified as a risk factor in rheumatoid arthritis, and IL-1 receptor associated kinase Positive regulation of leukocyte activation (GO:0002696) 3 (IRAK3), are presented in Additional file 2: Figures S1 Positive regulation of cell activation (GO:0050867) and S2 [12]. Activation of innate immune response (GO:0002218) Of note, patient 5 had a markedly different expression Response to lipopolysaccharide (GO:0032496) pattern in active disease, and patients 2 and 5 had a dif- Detection of external biotic stimulus (GO:0098581) ferent expression pattern in inactive disease (Fig. 1). Pa- Activation of immune response (GO:0002253) tient 5 differs to the other patients in being the only female patient. The marked difference in gene expres- Negative regulation of immune response (GO:0050777) sion patterns is, however, more likely due to the fact that Shown are the 15 biological processes that are the most stringent according to the p value in the Gene Ontology (GO) analysis patient 5 received three methylprednisolone pulses prior to the RNA sample being drawn. Patient 2 also received Tenbrock 171005 heatMap sorted single values −2.4 −1.8 −1.2 −0.6 0.0 0.6 1.2 1.8 2.4 Differential Expression (log2) PROK2 IRAK3 DSC2 ORM2 ORM1 S100A8 WDFY3-AS1 CMTM2 BASP1 NFE2 NABP1 NCF4 SLC22A15 TET2 ITGAM FCGR2A PADI4 NLRC4 MGAM2 CR1 DYSF PYGL FAM129A SLC2A3 GAS7 LIMK2 SLC22A4 PFKFB3 PFKFB3 HCAR3 NAIP HRH2 ADM PGS1 SIRPD RNF24 FAM157B GLT1D1 TECPR2 SIGLEC9 BCL6 ADGRG3 NAIP NAIP IL1R1 ADAM9 PLXNC1 PLXNC1 GPR141 MAPK14 MCTP2 TLR1 SRPK1 CR1L SLC25A37 GCA VNN3 CPEB4 VNN1 KREMEN1 MGAM IRAK3 TLR5 HP STEAP4 CREB5 WDFY3 GK ACSL1 ANXA3 CD177 MCEMP1 ALPL CD177P1 SYTL2 YME1L1 CD2 TRGV3 ESYT1 PRKCH TESPA1 CD74 SKAP1 TRAJ1 TRGC2 TARP KLRG1 FAM117B MKL2 SIRPG IL7R CYSLTR2 CDKAL1 CCDC6 CLSTN1 SPOCK2 ARID5B AES HLA-DRB1 HLA-DRB1 Fig. 1 Heatmap with hierarchical clustering of patients (Pat) with active disease (ad) and inactive disease (id). Shown are all genes with fold change > 3 Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 6 of 11 Fig. 2 Regulated genes in patients with active disease (ad) versus inactive disease (id) using fold change (FC) > 2 and a p value <0.01. Red dots include genes with fold change > 2 in active disease, blue dots with a fold change <− 2. Names of highly regulated and significant genes are depicted in the graph using different colors one pulse of methylprednisolone, which the others did binding in the promoter of upregulated or downregu- not. lated genes. The most prominent binding motif in up- As for the changes in inactive disease, patients 1 and 5 regulated genes in active sJIA was STAT4, followed by required ongoing IL-1 blockade and reacted with systemic BCL6 and STAT3 (adjusted p value <0.005) (Fig. 3). signs to withdrawal, while the other patients were able to Moreover BCL6 was also upregulated on the array in ac- taper and discontinue the IL-1 antagonists (both anakinra tive disease (fold change 5.15, p = 0.002, Fig. 2), while and canakinumab) over the course of the next year. STAT4 and STAT3 were not. Motif enrichment analysis identifies STAT4, BCL6 and HLA-DRB1 is upregulated in inactive compared to active STAT3 as the most prominent binding sites during active disease disease Initially we examined transcripts shown to be regulated In order to identify pathways that might be of relevance during sJIA, to validate that the patients under study exhib- in sJIA, we performed transcription factor (TF) motif ited the features described previously for this condition. For enrichment analysis to evaluate which TFs are preferably example, there was a fold change of 4.76 (p =0.006) in Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 7 of 11 Fig. 3 Motifs of the most regulated transcription factor binding sites in the whole transcriptome of patients with active systemic juvenile idiopathic arthritis (sJIA) compared to inactive sJIA, analyzed using MotifMatch S100A8 in our patients, as expected [13, 14]. Our analysis change 4.73, p=0.036, Fig. 2). However, and more strikingly, revealed HLA-DRB1 as the most strongly upregulated gene in cohort II CD74 was much higher expressed in patients in inactive compared to activedisease (foldchange6.8, p = with systemic activity as well as inactive disease compared to 0.003, Fig. 2). This observation was confirmed using RT- age-matched healthy controls using RT-PCR (Fig. 5). The PCR in the longitudinal per-patient sample analysis, while protein encoded by this gene is the invariant chain of the we did not find upregulation in cohort II compared to HLA-DR complex, which is associated with class II major healthy controls due to high individual expression differ- histocompatibility complex (MHC) including HLADRB1 ences (Fig. 4). Other HLA class II genes, for example HLA- and is an important chaperone that regulates antigen presen- DRB4 (fold change − 2.92, p = 0.008, Fig. 2), HLA-DRB3 tation for immune response. and HLA-DRB6, were found to be downregulated. CD177 is strongly upregulated in active disease CD74 is regulated in sJIA irrespective of disease activity CD177 was upregulated in patients with active disease CD74 was also upregulated in inactive compared to compared to those with inactive disease or healthy con- active disease on an individual per patient base (fold trols (fold change 37, p =0.007, Fig. 2), which has recently Fig. 4 Expression of HLA-DRB1 in active (AD) versus inactive (ID) systemic juvenile idiopathic arthritis using reverse transcription PCR. Left graph, fold change (n-fold) of longitudinal samples before and after treatment with anakinra, with inactive disease set to level 1 (*p < 0.05). Right graph, delta cycle threshold (ΔCt) values (relative expression values related to ribosomal protein L (RPL)) in active systemic disease (cohorts I and II), active polyarticular disease, and inactive disease/remission, and controls (all cohort II) Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 8 of 11 refractory to conventional therapy, we were able to show differential regulation of a variety of genes, comparing inactive disease on IL-1 antagonist treatment with active disease within the same patient before treatment. Regulation of specific genes has been confirmed in an independent cohort of single samples from patients with sJIA with either active or inactive disease. Previously, a specific sJIA signature of RNA expres- sion has been described in 44 patients with sJIA in various stages of the disease [4]. In this study, 12 patients with systemic and arthritic symptoms dis- played a specific expression pattern compared to healthy controls and patients with polyarticular (non- systemic) flare. Similar results were found by Barnes et al. in a cohort of 21 patients with sJIA, which also demonstrated a specific pattern of sJIA different from other subtypes of JIA [5]. A recent study by Brachat Fig. 5 Expression of CD74 using reverse transcription PCR: delta et al. on sJIA RNA samples from two studies using cycle threshold (ΔCt) values (relative expression values related to canakinumab also demonstrated a large number of ribosomal protein L) in controls, active systemic disease, active differentially expressed genes in patients prior to and polyarticular, and inactive disease (all ***p < 0.001) after initiation of anti-IL-1 treatment [6]. Similar to these preceding studies, the overall specific sJIA been described in the study by Brachat et al. and confirms signature consists of regulation of the inflammatory their findings using RT-PCR (Fig. 6)[6]. This gene en- response and innate immune pathways including the codes a glycosyl-phosphatidylinositol-linked cell surface IL-1 pathway, IL-6 and toll-like receptor 1, but also modu- glycoprotein that plays a role in neutrophil activation. The lation of lymphocyte differentiation and response, with an protein can bind platelet endothelial cell adhesion emphasis on T cells (Table 2). A recent large genome-wide molecule-1 and function in neutrophil transmigration. association study also demonstrated a framework of pathophysiological pathways that appears to be specific Discussion for sJIA [15]. By using comparative array analysis of RNA expression Transcription factor motif enrichment analysis in in a cohort of sJIA patients in different stages of disease patients with active and inactive disease identified STAT4, BCL6, and STAT3 as the most prominent motifs within the regulated genes. STAT4 is mainly expressed in myeloid cells and is the transcription factor downstream of IL-12, which has been identified as a potential biomarker in sJIA, but has not been discussed as a potential therapeutic target so far [16– 18]. We have previously shown that the likelihood of the occurrence of a polymorphism enhancing IL-12 expression was somewhat higher in patients with sJIA compared to other forms of JIA, which points toward IL-12 being more prominent in sJIA than previously thought [19]. In our array analysis, IL-12 was not dif- ferentially expressed; however, the IL-23 receptor, which is involved in IL-12 downstream signaling was downregulated in active disease (fold change 2.5, p = 0.012, Fig. 2), which suggests a physiological reaction towards the proinflammatory state. The identification of BCL6 as the second most prom- Fig. 6 Expression of CD177 using reverse transcription PCR: delta inent binding motif is more striking: BCL6 is a lineage cycle threshold (ΔCt) values (relative expression values related to transcription factor for follicular T helper (Tfh) cells, a ribosomal protein L) in controls, active disease, and inactive cell type that is important for B cellular responses. This disease (***p < 0.001) cell type is especially important for autoimmune arthritis Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 9 of 11 mediated by gut bacteria [20]. BCL6 expression has been [29]. However, RNA expression patterns of the shared shown to be sustained by IL-6 signaling in patients with epitope are variable, and apparently dependent on the rheumatoid arthritis (RA), and specific targeting of IL-6 copy number [27]. using tocilizumab in patients with RA results in a signifi- CD74 is the invariant chain of HLA-DR and therefore cant reduction of circulating Tfh cells; IL-21 production a cooperation partner on the cell surface, where it critic- by Tfh cells was also correlated with reduced expression ally regulates antigen presentation. It also serves as a cell of antibody-producing plasmablasts [21]. BCL6 was also surface receptor for the cytokine macrophage migration directly upregulated on the array, which supports the hy- inhibitory factor (MIF) which, when bound to the pothesis that this TF is of major importance (fold encoded protein, initiates survival pathways and cell pro- change 5.15, p = 0.002, Fig. 2), and would also argue for liferation. In cooperation with IL-12, MIF is important even earlier involvement of autoimmunity with B cell re- for survival in children with malaria and trypanosoma sponses in the course of the disease [22]. There is some infection [30, 31]. CD74 also interacts with amyloid pre- evidence of B cell-driven autoimmunity in the later cursor protein and suppresses the production of beta phases of sJIA, in which autoantibody production has amyloid [32]. The high expression of CD74 even in an been demonstrated [23]. Our findings here indicate that inactive disease state could serve as a possible candidate B cell activation might already be present during the first for disease markers for sJIA, and possibly - in the con- phase of disease where autoinflammation is the most text of MIF - even for therapeutic intervention. prominent feature. This might potentially determine An additional gene that was highly upregulated in the whether the patients develop a polyarticular course later. active stages of sJIA was CD177, confirming recent re- In contrast, finding STAT3 as the third most prominent sults [6]. CD177 codes for human neutrophil antigen 2 transcription factor is unsurprising, since STAT3 signals (HNA-2), also called NB1, a cell surface glycoprotein downstream of IL-6, a well-known therapeutic target [33]. HNA-2 expression is highly variable in humans and using tocilizumab in sJIA [24, 25]. In the study by Bra- is lacking in approximately 3–5% of the North American chat et al., where canakinumab was used for IL-1ß population [34]. HNA-2 plays an important role in mye- blockade, IL-6 declined by day 3 and remained sup- loid cell proliferation and function of neutrophils, in- pressed over time [6]. STAT3 was not differentially cluding transendothelial migration [35]. It has also been expressed on the array; however, STAT3 is usually acti- associated with anti-neutrophil-cytoplasmic antibody vated by phosphorylation and not by transcriptional (ANCA)-associated vasculitides, where CD177 has been regulation. Nevertheless, IL-6 signals via JAK3, which proposed as a receptor of mPR3 on the neutrophil sur- then activates STAT3 and is downstream of the IL-6 re- face [36]. Interestingly, CD177 has also been observed as ceptor, and it was found to be enhanced (fold change 2. the most upregulated parameter in a microarray study of 5, p = 0.006). Vice versa, the antagonist of STAT3 signal- purified neutrophils from patients with septic shock. ing is the suppressor of cytokine signaling 3 (SOCS3), However, since we performed our analysis in whole which was highly induced in active disease (fold blood, the numbers of neutrophils could also have had a change 8.5, p = 0.012, Fig. 2), probably to counterbalance considerable effect on our findings. There was, however, the overwhelming immune stimulation. no difference in CD177 transcription between active sys- HLA-DRB1 and its cooperation partner CD74 were temic and active polyarticular disease, while neutrophil upregulated in the active disease stage, and CD74 was counts were significantly lower in polyarticular disease upregulated in the inactive disease stage in patients with than in systemic flares (data not shown). Nevertheless, sJIA in this study. Certain genotypes of HLA-DRB1, in light of the large variations in CD177 expression in most notably HLA-DRB1*11, have been found to be controls and throughout the patient cohorts, these data strongly associated with sJIA in a large recent study of have to be interpreted with caution. 982 patients across nine different populations, and also This is a small study with a limited number of patients in a fine-mapping study of the HLA locus comparing it examined longitudinally, even if the results are con- to other forms of JIA [11, 15, 26]. However, no expres- firmed in a second cohort. As RNA was extracted from sion studies of HLA-DRB1 in sJIA have been performed whole blood rather than sorted cells due to the con- to date. HLA-DRB1 has also previously been demon- straints of a biobank, and given that a number of genes strated to be strongly associated with early, severe RA discussed, especially CD177, are expressed by neutro- [27]. RA-associated HLA-DRB1 alleles have conserved phils, changes in neutrophil numbers could have signifi- amino acid sequences in position 70–74 of the molecule. cantly impacted the results. However, not all genes This molecular structure is termed the shared epitope, described here are predominantly expressed in neutro- with a variety of hypotheses explaining its function [28]. phils and CD177 was also consistently upregulated in An association with the shared epitope has also been patients with polyarticular flares and low neutrophil found in children with rheumatoid-factor positive JIA numbers. As variable cell numbers are a valid point of Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 10 of 11 criticism of preceding studies as well, confirmation using and BV participated in the design and coordination of the study and helped draft the manuscript. IGG performed the genetic and statistical analysis. All sorted cells is a logical next step in researching gene ex- authors read and approved the final manuscript. pression patterns in sJIA [4–6]. The strength of this study is the longitudinal examination of patients who Ethics approval and consent to participate Clinical data and patient samples were acquired from the AID-Net database, have undergone a consistent institutional treatment a German registry and biobank that prospectively collects information and protocol with IL-1 agonists in different stages of their biomaterials from patients with autoinflammatory syndromes including peri- disease. odic fever syndromes and sJIA. Written informed consent was obtained from the patients prior to inclusion. Conclusions Competing interests By using this longitudinal analysis, our study identifies The authors declare that they have no competing interests. novel pathways (STAT4 and BCL6) that might be of relevance in sJIA and indicates strong upregulation of Publisher’sNote HLA-DRB1 in cooperation with CD74 in patients with Springer Nature remains neutral with regard to jurisdictional claims in sJIA in inactive disease upon treatment with IL-1 antag- published maps and institutional affiliations. onists. This provides the first functional confirmation of Author details a previous study, which identified HLA-DRB1 as a risk 1 German Center for Pediatric and Adolescent Rheumatology, Gehfeldstrasse factor in sJIA. Additionally CD177 was confirmed as a 24, 82467 Garmisch-Partenkirchen, Germany. Department of Pediatrics, Universitätsklinikum Aachen, Aachen, Germany. IZKF Research Group new marker in sJIA. Studies with larger patient cohorts Bioinformatics, RWTH Aachen Medical Faculty, Aachen, Germany. University using flow cytometry for protein expression are neces- Medical Center Utrecht, Utrecht, Netherlands. sary to confirm these results. Received: 4 January 2018 Accepted: 23 April 2018 Additional files References Additional file 1: Table S1. List of regulated genes with fold change 1. Singh-Grewal D, Schneider R, Bayer N, Feldman BM. Predictors of disease (FC) > 2 in patients with active disease (ad) versus inactive disease (id) course and remission in systemic juvenile idiopathic arthritis: significance of showing signal intensity in both disease states, gene symbol and early clinical and laboratory features. Arthritis Rheum. 2006;54:1595–601. description. (XLSX 60 kb) 2. Mellins ED, Macaubas C, Grom AA. Pathogenesis of systemic juvenile idiopathic arthritis: some answers, more questions. Nat Rev Rheumatol. Additional file 2: Figure S1. Expression of ANXA3 using RT-PCR: delta 2011;7:416–26. Ct values (relative expression values related to RPL) in controls, active 3. Petty RE, Southwood TR, Manners P, Baum J, Glass DN, Goldenberg J, He X, systemic or polyarticular disease and inactive disease (***p < 0.001, Maldonado-Cocco J, Orozco-Alcala J, Prieur AM, et al. International League ****p < 0.0001). Figure S2. Expression of IRAK3 using RT-PCR: delta Ct of Associations for Rheumatology classification of juvenile idiopathic values (relative expression values related to RPL) in controls, active arthritis: second revision, Edmonton, 2001. J Rheumatol. 2004;31:390–2. systemic or polyarticular disease and inactive disease (**p < 0.01). 4. Allantaz F, Chaussabel D, Stichweh D, Bennett L, Allman W, Mejias A, Ardura (DOCX 619 kb) M, Chung W, Smith E, Wise C, et al. Blood leukocyte microarrays to diagnose systemic onset juvenile idiopathic arthritis and follow the Abbreviations response to IL-1 blockade. J Exp Med. 2007;204:2131–44. bp: Base pairs; cDNA: Complementary DNA; Ct: Cycle threshold; HNA- 5. Barnes MG, Grom AA, Thompson SD, Griffin TA, Pavlidis P, Itert L, Fall N, 2: Human neutrophil antigen 2; GO: Gene Ontology; IRAK: IL-1 receptor Sowders DP, Hinze CH, Aronow BJ, et al. Subtype-specific peripheral blood associated kinase; IL: Interleukin; ILAR: International League of Associations gene expression profiles in recent-onset juvenile idiopathic arthritis. Arthritis for Rheumatology; MHC: Major histocompatibility complex; MIF: Macrophage Rheum. 2009;60:2102–12. migration inhibitory factor; PBMC: Peripheral blood mononuclear cells; 6. Brachat AH, Grom AA, Wulffraat N, Brunner HI, Quartier P, Brik R, McCann L, RA: Rheumatoid arthritis; RPL: Ribosomal protein L; RT-PCR: Reverse Ozdogan H, Rutkowska-Sak L, Schneider R, et al. Early changes in gene transcription PCR; sJIA: Systemic juvenile idiopathic arthritis; TF: Transcription expression and inflammatory proteins in systemic juvenile idiopathic factor; Tfh: Follicular T helper (cells) arthritis patients on canakinumab therapy. Arthritis Res Ther. 2017;19:13. 7. Lainka E, Bielak M, Hilger V, Basu O, Neudorf U, Wittkowski H, Holzinger D, Acknowledgements Roth J, Niehues T, Foell D. Translational research network and patient We thank Nienke Ter Haar for careful review of the manuscript. registry for auto-inflammatory diseases. Rheumatology (Oxford). 2011; 50:237–42. Funding 8. Wallace CA, Giannini EH, Huang B, Itert L, Ruperto N, Childhood Arthritis This study was supported by the Interdisciplinary Center for Clinical Research Rheumatology Research A, Pediatric Rheumatology Collaborative Study G, (IZKF) Aachen and UCAN-AC (Understanding Childhood Arthritis Network Paediatric Rheumatology International Trials O. American College of Aachen) and an unrestricted grant from Novartis AG to the German Center Rheumatology provisional criteria for defining clinical inactive disease in for Pediatric and Adolescent Rheumatology. The AID-Net database is select categories of juvenile idiopathic arthritis. Arthritis Care Res (Hoboken). supported by the Federal Ministry of Education and Research (BMBF project 2011;63:929–36. 01GM08104, 01GM1112D). 9. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density Availability of data and materials oligonucleotide array probe level data. Biostatistics. 2003;4:249–64. The datasets used and/or analyzed during the current study are available 10. Wang J, Duncan D, Shi Z, Zhang B. WEB-based gene set analysis toolkit from the corresponding author on reasonable request. (WebGestalt): update 2013. Nucleic Acids Res. 2013;41:W77–83. 11. Ombrello MJ, Remmers EF, Tachmazidou I, Grom A, Foell D, Haas JP, Martini Authors’ contributions A, Gattorno M, Ozen S, Prahalad S, et al. HLA-DRB1*11 and variants of the BH, JPH, and KT conceived of the study and drafted the manuscript. AS and MHC class II locus are strong risk factors for systemic juvenile idiopathic NF participated in data collection and reviewed the manuscript. KO, BD, FT, arthritis. Proc Natl Acad Sci U S A. 2015;112:15970–5. Hügle et al. Arthritis Research & Therapy (2018) 20:98 Page 11 of 11 12. Okada Y, Terao C, Ikari K, Kochi Y, Ohmura K, Suzuki A, Kawaguchi T, factor (MIF) protein and blood mononuclear cell MIF transcripts in children Stahl EA, Kurreeman FA, Nishida N, et al. Meta-analysis identifies nine with Plasmodium falciparum malaria. Clin Immunol. 2006;119:219–25. new loci associated with rheumatoid arthritis in the Japanese 31. Terrazas CA, Huitron E, Vazquez A, Juarez I, Camacho GM, Calleja EA, population. Nat Genet. 2012;44:511–6. Rodriguez-Sosa M. MIF synergizes with Trypanosoma cruzi antigens to 13. Gerss J, Roth J, Holzinger D, Ruperto N, Wittkowski H, Frosch M, Wulffraat N, promote efficient dendritic cell maturation and IL-12 production via p38 Wedderburn L, Stanevicha V, Mihaylova D, et al. Phagocyte-specific S100 MAPK. Int J Biol Sci. 2011;7:1298–310. proteins and high-sensitivity C reactive protein as biomarkers for a risk- 32. Matsuda S, Matsuda Y, D'Adamio L. CD74 interacts with APP and suppresses adapted treatment to maintain remission in juvenile idiopathic arthritis: a the production of Abeta. Mol Neurodegener. 2009;4:41. comparative study. Ann Rheum Dis. 2012;71:1991–7. 33. Stroncek DF, Caruccio L, Bettinotti M. CD177: a member of the Ly-6 gene superfamily involved with neutrophil proliferation and polycythemia vera. J 14. Holzinger D, Frosch M, Kastrup A, Prince FH, Otten MH, Van Suijlekom-Smit Transl Med. 2004;2:8. LW, ten Cate R, Hoppenreijs EP, Hansmann S, Moncrieffe H, et al. The toll- 34. Matsuo K, Lin A, Procter JL, Clement L, Stroncek D. Variations in the like receptor 4 agonist MRP8/14 protein complex is a sensitive indicator for expression of granulocyte antigen NB1. Transfusion. 2000;40:654–62. disease activity and predicts relapses in systemic-onset juvenile idiopathic 35. Bayat B, Werth S, Sachs UJ, Newman DK, Newman PJ, Santoso S. Neutrophil arthritis. Ann Rheum Dis. 2012;71:974–80. transmigration mediated by the neutrophil-specific antigen CD177 is 15. Ombrello MJ, Arthur VL, Remmers EF, Hinks A, Tachmazidou I, Grom AA, influenced by the endothelial S536N dimorphism of platelet endothelial cell Foell D, Martini A, Gattorno M, Ozen S, et al. Genetic architecture adhesion molecule-1. J Immunol. 2010;184:3889–96. distinguishes systemic juvenile idiopathic arthritis from other forms of 36. Hu N, Mora-Jensen H, Theilgaard-Monch K, Doornbos-van der Meer B, juvenile idiopathic arthritis: clinical and therapeutic implications. Ann Huitema MG, Stegeman CA, Heeringa P, Kallenberg CG, Westra J. Differential Rheum Dis. 2017;76:906–13. expression of granulopoiesis related genes in neutrophil subsets 16. de Jager W, Hoppenreijs EP, Wulffraat NM, Wedderburn LR, Kuis W, Prakken distinguished by membrane expression of CD177. PLoS One. 2014;9:e99671. BJ. Blood and synovial fluid cytokine signatures in patients with juvenile idiopathic arthritis: a cross-sectional study. Ann Rheum Dis. 2007;66:589–98. 17. Gohar F, Kessel C, Lavric M, Holzinger D, Foell D. Review of biomarkers in systemic juvenile idiopathic arthritis: helpful tools or just playing tricks? Arthritis Res Ther. 2016;18:163. 18. Yilmaz M, Kendirli SG, Altintas D, Bingol G, Antmen B. Cytokine levels in serum of patients with juvenile rheumatoid arthritis. Clin Rheumatol. 2001; 20:30–5. 19. Eberhardt CS, Haas JP, Girschick H, Schwarz T, Morbach H, Rosen-Wolff A, Foell D, Dannecker G, Schepp C, Ganser G, et al. No association of IL-12p40 pro1.1 polymorphism with juvenile idiopathic arthritis. Pediatr Rheumatol Online J. 2015;13:61. 20. Teng F, Klinger CN, Felix KM, Bradley CP, Wu E, Tran NL, Umesaki Y, Wu HJ. Gut microbiota drive autoimmune arthritis by promoting differentiation and migration of Peyer's patch T follicular helper cells. Immunity. 2016;44:875–88. 21. Chavele KM, Merry E, Ehrenstein MR. Cutting edge: circulating plasmablasts induce the differentiation of human T follicular helper cells via IL-6 production. J Immunol. 2015;194:2482–5. 22. Nigrovic PA, Mannion M, Prince FH, Zeft A, Rabinovich CE, van Rossum MA, Cortis E, Pardeo M, Miettunen PM, Janow G, et al. Anakinra as first-line disease-modifying therapy in systemic juvenile idiopathic arthritis: report of forty-six patients from an international multicenter series. Arthritis Rheum. 2011;63:545–55. 23. Hugle B, Hinze C, Lainka E, Fischer N, Haas JP. Development of positive antinuclear antibodies and rheumatoid factor in systemic juvenile idiopathic arthritis points toward an autoimmune phenotype later in the disease course. Pediatr Rheumatol Online J. 2014;12:28. 24. Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, Schaper F. Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J. 2003;374:1–20. 25. Yokota S, Imagawa T, Mori M, Miyamae T, Aihara Y, Takei S, Iwata N, Umebayashi H, Murata T, Miyoshi M, et al. Efficacy and safety of tocilizumab in patients with systemic-onset juvenile idiopathic arthritis: a randomised, double-blind, placebo-controlled, withdrawal phase III trial. Lancet. 2008;371:998–1006. 26. Hinks A, Bowes J, Cobb J, Ainsworth HC, Marion MC, Comeau ME, Sudman M, Han B, Juvenile Arthritis Consortium for I, Becker ML, et al. Fine-mapping the MHC locus in juvenile idiopathic arthritis (JIA) reveals genetic heterogeneity corresponding to distinct adult inflammatory arthritic diseases. Ann Rheum Dis. 2017;76:765–72. 27. Kerlan-Candon S, Combe B, Vincent R, Clot J, Pinet V, Eliaou JF. HLA-DRB1 gene transcripts in rheumatoid arthritis. Clin Exp Immunol. 2001;124:142–9. 28. Holoshitz J. The rheumatoid arthritis HLA-DRB1 shared epitope. Curr Opin Rheumatol. 2010;22:293–8. 29. Prahalad S, Thompson SD, Conneely KN, Jiang Y, Leong T, Prozonic J, Brown MR, Ponder LA, Angeles-Han ST, Vogler LB, et al. Hierarchy of risk of childhood-onset rheumatoid arthritis conferred by HLA-DRB1 alleles encoding the shared epitope. Arthritis Rheum. 2012;64:925–30. 30. Awandare GA, Hittner JB, Kremsner PG, Ochiel DO, Keller CC, Weinberg JB, Clark IA, Perkins DJ. Decreased circulating macrophage migration inhibitory

Journal

Arthritis Research & TherapySpringer Journals

Published: May 30, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off