Correspondence to Dr Guojun Hou; [email protected] ; Professor Huihua Ding; [email protected]
WHAT IS ALREADY KNOWN ON THIS TOPIC
Systemic lupus erythematosus (SLE) is a highly heterogeneous disease with fluctuating disease course and the identification of a reliable biomarker is of imperative need.
B cell is considered to have a crucial pathogenic role in SLE as it is causative to the excessive generation of autoantibodies, which characterised lupus.
WHAT THIS STUDY ADDS
We found an SLE gene module in B cells that was significantly correlated with SLE clinical traits in and enriched in interferon pathway.
We identified lncRNA RP11-273G15.2 as a biologically significant biomarker as it coexpressed with genes critical to SLE in B cells and it outperformed IFN scores to reflect disease activity.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our study provides a potential biomarker for SLE by combining bioinformatic analysis with clinical validation, and it might act as an entry point for us to identify therapeutic target and explore disease pathogenesis.
Introduction
Systemic lupus erythematosus (SLE) is a prototypical autoimmune disease. Both genetic predisposition and environmental factors contribute to its pathogenesis. In patients with SLE, the defect of immune tolerance leads to autoreactive innate and adaptive immune responses. Particularly, the B cell compartment is considered to have an essential pathogenic role in SLE as they are responsible for the overproduction of autoantibodies.1 The B cell-targeted therapy, such as belimemab, could effectively attenuate the symptom of SLE, which further demonstrates the pivotal role of B cells in lupus.2 Thus, the advance of B cell study can facilitate the development of effective diagnosis, monitoring and therapeutic strategies of SLE. Meanwhile, the identification and validation of a reliable biomarker related to B cell pathophysiology in patients with SLE is of urgent need.
Long non-coding RNAs (lncRNAs) are non-protein-coding RNAs that are >200 nucleotides in length,3 some of which are dysregulated in autoimmune diseases and could be causes or consequences of these diseases.4 LncRNAs can function as key regulators of gene expression at epigenetic, transcriptional, post-transcriptional and post-translational levels.4 For instance, lnc-DC is exclusively expressed in conventional DC cells, and it controls DC differentiation through binding to STAT3.5 Recent findings have shown that multiple lncRNAs, such as RP11-2B6,2, MALAT-1 and NEAT1 were engaged in the pathogenesis of autoimmune diseases.6–8 Moreover, previous studies indicate that some lncRNAs have the potential to work as biomarkers to reflect SLE activity.9 10 However, there are few studies focusing on the lncRNA profiles in B cells of patients with SLE, and B cell expressing lncRNA’s value as biomarker deserve to be evaluated.
The discovery and confirmation of a viable biomarker in SLE investigations remain a major challenge.11 Strategies like transcriptional profiling, mass spectrometry-based proteomic profiling, peptidomic profiling, etc, have been adopted to discover disease biomarkers. Traditional methods usually identify individual molecules as biomarkers. In contrast, a bioinformatic strategy named weighted gene coexpression network analysis (WGCNA) can relate gene modules rather than individual genes to external traits, which can reduce the multiple testing problem and is more biologically significant and reliable.12 WGCNA constructed gene coexpression networks based on their expression pattern, and it links gene modules to clinical traits, which can help us identify gene modules associated with phenotypes and get insights into the underlying pathogenic processes. Currently, few studies adopted this method to screen lncRNA as a potential biomarker in SLE.
In this study, we integrated WGCNA method, bulk and single-cell RNA-sequencing data to discover and validate a B cell highly expressing lncRNA as potential biomarker, which is pathologically associated with SLE and has the potential to advance the diagnosis, monitoring and therapeutic interventions of SLE.
Materials and methods
Study design and study subjects
In this study, we first employed public B cell RNA-seq data (GSE164457) to perform a WGCNA analysis, from which we discovered a gene module named Saddlebrown as the most prominent module associated with clinical traits, and genes in this module were highly enriched in interferon (IFN) pathway (figure 1). Afterward, we confirmed a candidate lncRNA in this module, RP11-273G15.2, was differentially expressed in 5 B cell subpopulations of patients with SLE compared with healthy controls (HCs) (E-GEAD-397). We selected it as our candidate, and subsequently evaluated its distribution and association with IFN pathway at a single-cell level (GSE163121). Finally, we validated RP11-273G15.2’s potential as a biomarker for SLE in an independent cohort collected in our medical centre (figure 1).
Enlarge this image.Figure 1. Overall schematics for screening of candidate lncRNAs. (1) WGCNA was applied to high-throughput SLE B-cell transcriptome data (GSE164457) to construction networks with co-expressed genes in the same module. Gene modules were linked to disease phenotypes and the most clinically relevant module was selected to do an enrichment analysis. (2) B cell RNA-seq data (E-GEAD-397) were analysed. LncRNA RP11-273G15.2 was in the critical module and differentially expressed between HCs and SLE. (3) Distribution of RP11-273G15.2 was investigated at a single-cell level (GSE163121). (4) Real-time PCR was used to validate the potential of RP11-273G15.2 as a biomarker in an independent cohort. HCs, healthy control; SLE, systemic lupus erythematosus; WGCNA, weighted gene coexpression network analysis.
In the validation phase, 51 patients with SLE and 31 age-matched and gender-matched healthy controls were recruited between 2017 and 2018 by the Department of Rheumatology at RenJi Hospital, Shanghai Jiao Tong University School of Medicine. All patients fulfilled the ACR 1997 revised criteria or 2012 SLICC criteria for SLE.13 14 All participants were Han Chinese people. Medical records’ review, physician assessment and laboratory test data were retrospectively collected. Disease activity was evaluated according to the SLE Disease Activity Index 2000 (SLEDAI-2K).15 The IFN scores were calculated as previously described.16 Briefly, we used three representative interferon-stimulated genes (ISGs) to calculate IFN scores: IFIT1, MX1 and IFIT3. To standardise the gene expression level, the mean expression level of IFIT3 detected by RT-qPCR in the healthy control group was calculated and subtracted from the expression level of IFIT3 of each sample in the healthy control group and SLE group, and the remainder was divided by the SD value of IFIT3 in the healthy control group. IFIT1 and MX1 standardised expression levels were computed in the same manner. Finally, the three values were added up to yield the IFN score for each sample. The demographic and clinical information of all patients with SLE is listed in online supplemental table 1.
Sample collection and B cell isolation
Whole blood from participants was collected into EDTA-coated vacutainer tubes. Peripheral blood mononuclear cells were isolated from whole blood by Ficoll density-gradient centrifugation. B cells were further isolated using anti-CD19 magnetic beads (Miltenyi Biotec) and stored in Trizol reagent until use. Total RNA was extracted from B cells.
RT-qPCR
Quantity and quality of RNA were determined by NanoDrop-2000 spectrophotometer (NanoDrop Technologies, Wilmington, Delaware). RNA was reverse transcribed into cDNA using PrimeScript RT reagent (Vazyme R333-01) in a total volume of 10 ul. The following protocol was used: 50 °C for 15 min, 85 °C for 5 s. qPCR was performed using SYBR qPCR Master Mix (Vazyme Q712-02) by QuantStudio 7 Flex Real-Time PCR System. RPL13A was used as the reference gene. All the primers used are as follows: RP11-273G15.2-F:
Weighted gene coexpression network analysis
R package WGCNA was used to construct gene coexpression networks. CD19+B cell RNA-seq data of 120 SLE patients were obtained from GEO database through the accession number GSE164457.17 Genes with detectable expression levels in more than 20 CD19+B cell samples were included in the WGCNA analysis.12 First, samples were clustered according to the Euclidean distance. By inspection, eight outliers were identified and removed by setting the height cut of 3e+06. Next, we choose an empirical value of 12 as the soft threshold to cluster genes into different modules and construct the scale-free gene network. Finally, gene modules were related to clinical features of the subjects and the phenotypic essential module was identified.
Differentially expressed lncRNA analysis
RNA sequencing data were downloaded from NBDC Human Database through the accession number E-GEAD-397. The counts of lncRNAs were analysed by R package DEseq2. After removing lncRNAs with 0 count in more than 30% of samples, lncRNAs with fold change >2 or <0.5 and FDR <0.05 between healthy controls (HCs) and SLE were selected as differentially expressed lncRNAs. RNA-seq data from 10 SLE patients with low expression of IFN stimulated genes (IFNneg), 10 patients with high expression of IFN-stimulated genes (IFNpos), and 12 HCs were also used to identify differentially expressed lncRNAs from GEO database through the accession number GSE149050.18
Enrichment analysis
Pathway enrichment was performed by Cytoscape V.3.8.219 using the GO ImmuneSystemProcess analysis. The bubble plot was drawn by the R package ggplot2.
Single-cell analysis
For peripheral B cells, single-cell RNA-seq data were downloaded from GEO database with the accession number GSE16312120 and analysed by Seurat V.4.0.6.21 Cells with less than 50 genes or more than 12.5% mitochondrial genes detected were removed. Cells with myeloid cell, T cell or NK cell origin were removed. B cell clusters were annotated manually by critical B cell subtype markers like IGHD, IGHG1, CD27, etc.
Statistical analysis
GraphPad Prism V.9 was used for all statistical analyses and figure plotting. The normality of data was established using Shapiro-Wilk tests. For non-normally distributed data, non-parametric Mann-Whitney U-test was used to make a comparison between the two groups. For the comparison of RP11-273G15.2 expressing cells between HCs and patients with SLE, an unpaired t-test was applied. For correlation analysis, Spearman’s method was employed. The diagnostic and disease activity monitoring performance of the biomarker was assessed using receiver operating characteristic (ROC) curve analysis. Statistical significance was defined as a two-tailed value of p<0.05.
Results
Genes in the disease-critical module were enriched in type I IFN pathway
To find the B cell critical gene modules that were associated with SLE, we first performed a WGCNA on RNA sequencing data of B cells from 120 patients with SLE (GSE164457).17 Genes with detectable expression levels in B cells were applied to WGCNA. In the process of network construction, coexpression similarity between gene pairs was transformed into connection strengths and they formed an adjacency matrix, which was used to define network modules.22 In our analysis, all filtered genes in B cells were clustered into 29 modules (figure 2A and online supplemental table 2) with densely connected genes in the same gene module, and the genes that could not be classified in the grey module.
Enlarge this image.Figure 2. Clustering of genes in B cells of patients with SLE by WGCNA. (A) Dendrogram of clustered gene modules. (B) Heatmap of association between gene modules and phenotypes. Colours reflected correlation coefficient. Blue and red colours represent negatively correlated and positively correlated, respectively. (C) Bubble plot of pathway enrichment of genes in Saddlebrown module. GO pathway terms were displayed in Y axis and X axis displayed gene ratio. Colours reflected p values, whereas bubble size represented gene counts. SLE, systemic lupus erythematosus; WGCNA, weighted gene coexpression network analysis.
To identify which module was critical to SLE, a correlation analysis was performed between these modules and clinical manifestations (figure 2B). Notably, the module named Saddlebrown was the one most significantly correlated with SLE traits, including the presence of malar rash (p=0.037), positivity of anti-dsDNA antibodies (p=0.016), positivity of anti-Smith antibodies (p=1.824E-6) and renal involvement (p=0.022) (online supplemental table 3). There were 71 coexpressed genes in this module, including 4 non-coding RNAs and 67 protein-coding genes. Based on the strong correlation between this module and clinical manifestations, we concentrated on the Saddlebrown module with interest in the following study.
The Saddlebrown module was significantly associated with clinical phenotypes, indicating it might relate to essentially biological processes in B cells critical to SLE. To further explore the immunological functions of this module, we performed an enrichment analysis using the genes in this module. As depicted in figure 2C, genes in this module were highly enriched in type I IFN pathway. Based on these observations, we named Saddlebrown module as interferon-associated module. To sum, our rediscovery offered validity of the importance of type I IFN to disease performance. Furthermore, lncRNAs clustered in this module might be related with type I IFN pathway, and their potential as biomarkers deserve to be further validated.
RP11-273G15.2 was differentially expressed in various B cell subtypes
In order to identify lncRNA biomarkers in SLE patients’ B cells, we further analysed lncRNA transcriptomic landscapes in five different B cell subpopulations from NBDC Human Database (E-GEAD-397).23 As shown in figure 3A, there were 108, 69, 19, 42 and 114 differentially expressed lncRNAs in double negative B cells, naïve B cells, plasmablasts, switched memory B cells and unswitched memory B cells, respectively (padj <0.05, fold change >2 or <0.5) (online supplemental table 4). Notably, there was only one differentially expressed lncRNA, RP11-273G15.2, which was also present in the interferon-associated module, overlapped among these 5 B cell subpopulations (figure 3B,C).
Enlarge this image.Figure 3. RP11-273G15.2 was differentially expressed in various B cell subsets. (A) Volcano plots indicate lncRNA expression in different B cell subsets. Blue and red colours represent downregulation and upregulation, respectively. (B) Venn diagram indicates the differentially expressed lncRNAs between HCs and SLE. (C) Expression of RP11-273G15.2 in different B cell subpopulations of healthy controls and patients with SLE. Mann-Whitney test was used. ****p<0.0001. HCs, healthy control; SLE, systemic lupus erythematosus.
We also investigated other two autoimmune diseases: rheumatoid arthritis and systemic sclerosis (SSc). As demonstrated in the left panel of online supplemental figure 1A–E, the levels of RP11-273G15.2 were significantly increased in double-negative B cells, naïve B cells, unswitched memory B cells and plasmablasts of patients with SLE compared with those in RA and SSc patients, with RP11-273G15.2 between SLE and RA patients in switched memory B cells as an exception (online supplemental figure 1C). By ROC analysis, RP11-273G15.2 could discriminate patients with SLE from RA and patients with SScin all subsets of B cells except RA patients in switched memory B cells. The greatest area under the curves (AUCs) were in naïve B cell for both of RA (AUC=0.7924, p<0.0001) and SSc (AUC=0.7854, p<0.0001) (right panel of online supplemental figure 1A–E). It suggested that B cell RP11-273G15.2 could be a potential diagnostic biomarker for SLE with its discriminative ability in SLE versus RA, SSc and HC.
As mentioned above, coexpressed genes were clustered into the same module. Coding genes in the interferon associated module were enriched in type I IFN pathway (figure 2C), which indicated RP11-273G15.2 was coexpressed with ISGs in B cells. To further confirm its relationship with type I IFN pathway, we analysed RP11-273G15.2 expression from patients with SLE with low expression of ISGs (IFNneg), SLE patients with high expression of ISGs (IFNpos), and HCs (GSE149050).18 Consistent with the results of WGCNA analysis, RP11-273G15.2 was highly expressed in B cells from IFN positive patients while there was no significant difference between HCs and IFN negative patients (online supplemental figure 2A, B), and this result was further confirmed in our independent cohort (online supplemental figure 2C). Consequently, we chose RP11-273G15.2 as our candidate biomarker.
Analysis of RP11-273G15.2 distribution at a single-cell level
As B cells are divided into various subtypes, transcriptome data at single-cell resolution can yield the expression pattern of RP11-273G15.2 in different subpopulations of B cells and provide insights into the pathogenic processes of lupus. Single-cell RNA-sequencing data of SLE patients’ and HCs’ B cells (GSE163121) were analysed by Seurat4.0.6. Peripheral B cells were clustered into 11 groups (figure 4A,B, online supplemental figures 3 and 4). As shown in figure 4C, more SLE peripheral B cells expressed RP11-273G15.2 than HC B cells, and in general, there was a higher percentage of SLE B cells that expressed RP11-273G15.2 than HC B cells (figure 4D). Moreover, in accordance with bulk RNA-seq data, the expression levels of RP11-273G15.2 were positively correlated with ISGs in single cell level of SLE B cells (figure 4E).
Enlarge this image.Figure 4. RP11-273G15.2 expression at a single-cell level. (A) UMAP displaying different B cell subsets. (B) Percentages of different subpopulations of B cells were shown. (C) Feature plot of RP11-273G15.2 expression in HC and SLE B cells. (D) The distribution of RP11-273G15.2 expression across different B cell clusters in patients with HC and SLE. Analysis was done by unpaired t-test. (E) Correlation of RP11-273G15.2 levels with different ISGs on a single cell. Correlation analysis was done by Spearman’s method. *p<0.05; **p<0.01; ****p<0.0001. DN2, double-negative 2; HCs, healthy control; IFN, interferon; IFN+DN2, double-negative two with interferon signature; IFN+Naïve, naïve with interferon signature; ISGs, interferon-stimulated genes; PB, plasmablasts; SLE, systemic lupus erythematosus; UMAP, uniform manifold approximation and projection.
RP11-273G15.2 is associated with disease activity and clinical traits
To validate whether RP11-273G15.2 could act as a biomarker, we evaluated its performance in an independent cohort of n=82 (HC=31, SLE=51). In consistent with previous findings, RP11-273G15.2 was highly expressed in SLE patients’ B cells compared with HCs (figure 5A). Its performance as a diagnostic biomarker was assessed by the receiver operating curve (ROC) analysis. As shown in figure 5B, AUC value was 0.7577 (95% CI 0.6544 to 0.8611; p<0.0001) for SLE versus HCs. For the purpose to evaluate its ability to reflect the activation status of IFN signalling pathway. Expression levels of 3 ISGs, IFIT1, MX1 and IFIT3 in B cells were detected by RT-qPCR. The results showed that RP11-273G15.2 expression levels were positively correlated with these 3 ISGs (online supplemental figure 5A–C). Then individual IFN-scores were calculated based on these 3 ISGs. We found that there was a statistically significant correlation between RP11-273G15.2 expression levels and IFN-scores (p<0.0001, figure 5C). To further explore the value of RP11-273G15.2 to indicate disease activity, we performed a correlation analysis between RP11-273G15.2 levels and SLEDAI scores. As demonstrated in figure 5D, RP11-273G15.2 levels were highly correlated with SLEDAI scores, supporting its role as an indicator of disease activity. Remarkably, RP11-273G15.2 outperformed IFN score as a biomarker to reflect disease activity, as AUC value was 0.8696 (95% CI 0.7635 to 0.9757; p =0.0071) for RP11-273G15.2 and 0.7696 (95% CI 0.5002 to 1.000; p = 0.0495) for IFN score (figure 5E). Besides, IFN scores and SLEDAI scores were also positively correlated (online supplemental figure 5D), in accordance with previous findings.16
Enlarge this image.Figure 5. Correlation of RP11-273G15.2 expression in B cells with IFN scores and disease activity. (A) The expression level of RP11-273G15.2 in B cells of healthy controls (HCs) and SLE patients. Mann-Whitney test was used. (B) ROC curve of RP11-273G15.2 level in discriminating SLE from HC. (C-D) The expression level of RP11-273G15.2 in SLE patients’ B cell was positively correlated with (C) IFN scores and (D) SLEDAI scores. (E) ROC curve of RP11-273G15.2 expression level and IFN score in separating SLE patients with SLEDAI-2K scores<6 from those with SLEDAI-2K scores>=6. Spearman’s method was used for correlation analysis. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. AUC, area under the curve; IFN, interferon; ROC, receiver operating characteristic; SLE, systemic lupus erythematosus; SLEDAI, SLE Disease Activity Index.
We also compared the expression level of RP11-273G15.2 in SLE patients with or without a specific clinical feature. As summarised in table 1, RP11-273G15.2 levels were associated with multiple clinical features and could be used to indicate the presence of these features, including vasculitis (p=0.004), thrombocytopenia (p=0.010), lymphocytopenia (p=0.007) and leukocytopenia (p=0.006). Moreover, RP11-273G15.2 expression level was correlated with the presence of some autoantibodies, including anti-Sm (p=0.008), anti-U1RNP (p=0.009), anti-ribosomal P protein (p=0.004) antibodies, but no significant correlation was detected for complements, erythrocyte sedimentation rate and autoantibodies including ANA, anti-dsDNA, anti-SSA-Ro52, anti-SSA-Ro60, anti-SS-B (La), anti-histone and anti-nucleosome. To note, both of anti-Sm and anti-ribosomal P protein antibodies are highly specific diagnostic biomarkers of lupus and are currently used in clinical practice.11 Taken together, RP11-273G15.2 could reflect multiple aspects of patients with SLE, including IFN signalling status, disease activity, symptoms and autoantibody positivity.
Table 1RP11-273G15.2 expression levels in patients with SLE with or without clinical features
| SLE | Clinical features present | SLE | Clinical features absent | P value | |
| N | Median (IQR, 10−3) | N | Median (IQR, 10−3) | ||
| Clinical features at sample collection | |||||
| Vasculitis | 3/51 (5.88%) | 2.81 (2.62, 2.98) | 48/51 (94.12%) | 0.73 (0.38, 1.27) | 0.004** |
| Malar rash | 17/51 (33.33%) | 1.16 (0.72, 1.87) | 34/51 (66.67%) | 0.67 (0.35, 1.21) | 0.050 |
| Discoid rash | 9/51 (17.65%) | 0.59 (0.34, 1.46) | 42/51 (82.35%) | 0.76 (0.50, 1.32) | 0.952 |
| Oral ulcer | 1/51 (1.96%) | 3.28 (3.28, 3.28) | 50/51 (98.04%) | 0.74 (0.44, 1.32) | 0.078 |
| Thrombocytopenia | 8/51 (15.69%) | 1.55 (0.95, 2.93) | 43/51 (84.31%) | 0.72 (0.35, 1.27) | 0.010* |
| Lymphocytopenia | 22/51 (43.14%) | 1.18 (0.76, 1.93) | 29/51 (56.86%) | 0.59 (0.37, 0.99) | 0.007** |
| Anaemia | 32/51 (62.75%) | 0.83 (0.54, 1.79) | 19/51 (37.25%) | 0.60 (0.36, 1.32) | 0.250 |
| Leukocytopenia | 11/51 (21.57%) | 1.81 (1.18, 2.20) | 40/51 (78.43%) | 0.73 (0.35, 1.16) | 0.006** |
| Renal involvement | 24/51 (47.06%) | 0.73 (0.44, 1.14) | 27/51 (52.94%) | 0.99 (0.45, 1.58) | 0.533 |
| Arthritis | 8/51 (15.69%) | 1.21 (0.59, 1.38) | 43/51 (84.31%) | 0.73 (0.38, 1.51) | 0.452 |
| Serositis | 9/51 (17.65%) | 0.73 (0.52, 1.90) | 42/51 (82.35%) | 0.76 (0.44, 1.27) | 0.584 |
| Neuropsychiatric | 3/51 (5.88%) | 0.78 (0.65, 0.92) | 48/51 (94.12%) | 0.74 (0.38, 1.58) | 0.865 |
| Photosensitivity | 4/51 (7.84%) | 1.15 (0.64, 1.54) | 47/51 (92.16%) | 0.73 (0.44, 1.39) | 0.853 |
| Laboratory test | |||||
| C3<0.9 g/L | 13/50 (26.00%) | 0.74 (0.57, 1.11) | 37/50 (74.00%) | 0.73 (0.39, 1.46) | 0.799 |
| C4<0.1 g/L | 31/50 (62.00%) | 0.73 (0.43, 1.13) | 19/50 (38.00%) | 1.20 (0.44, 1.85) | 0.219 |
| ANA | 51/51 (100.00%) | 0.74 (0.44, 1.39) | 0/51 (0.00%) | – | – |
| Anti-dsDNA | 38/51 (74.51%) | 0.94 (0.58, 1.51) | 13/51 (25.49%) | 0.52 (0.39, 1.20) | 0.300 |
| Anti-Sm | 12/50 (24.00%) | 1.38 (1.07, 2.20) | 38/50 (76.00%) | 0.73 (0.35, 1.18) | 0.008** |
| Anti-U1RNP | 23/50 (46.00%) | 1.20 (0.60, 1.84) | 27/50 (54.00%) | 0.72 (0.29, 1.08) | 0.009** |
| Anti-SSA-Ro52 | 28/50 (56.00%) | 0.76 (0.44, 1.26) | 22/50 (44.00%) | 0.91 (0.58, 1.75) | 0.584 |
| Anti-SSA-Ro60 | 27/50 (54.00%) | 0.99 (0.50, 1.38) | 23/50 (46.00%) | 0.72 (0.45, 1.51) | 0.514 |
| Anti-SS-B (La) | 8/50 (16.00%) | 0.76 (0.36, 1.02) | 42/50 (84.00%) | 0.81 (0.52, 1.58) | 0.427 |
| Anti-ribosomal P protein | 16/51 (31.37%) | 1.56 (0.76, 2.62) | 35/51 (68.63%) | 0.61 (0.35, 1.10) | 0.004** |
| Anti-histone | 13/51 (25.49%) | 1.09 (0.33, 1.33) | 38/51 (74.51%) | 0.74 (0.50, 1.38) | 0.931 |
| Anti-nucleosome | 33/51 (64.71%) | 0.73 (0.39, 1.33) | 18/51 (35.29%) | 0.94 (0.50, 1.38) | 0.783 |
| ESR>20 mm/h(woman), >15 mm/h (man) | 43/51 (84.31%) | 0.74 (0.44, 1.58) | 8/51 (15.69%) | 0.67 (0.40, 1.25) | 0.533 |
P values were for comparing RP11-273G15.2 levels in patients with or without clinical features by Mann-Whitney Wilcoxon U test, *p<0.05; **p<0.01;***p<0.001; ****p<0.0001.
ANA, antinuclear antibody; C3, complement 3; C4, complement 4; ESR, erythrocyte sedimentation rate; SLEDAI, Systemic Lupus Erythematosus Disease Activity Index; U1RNP, U1 ribonucleoprotein.
Discussion
In this study, we integrated high-throughput B-cell transcriptome data of patients with SLE, WGCNA analysis, single-cell RNA-sequencing data of SLE patients’ B cells to discover potential lncRNA biomarkers and performed validation in an independent cohort. We found that lncRNA RP11-273G15.2, dysregulated in SLE patients, was highly correlated with IFN scores and SLEDAI scores, which might help SLE diagnosis, disease activity monitoring and contribute to the development of clinical management of patients with SLE.
Traditional methods that discover SLE biomarkers usually rely on differential gene expression analysis. However, it would generate many candidates, and it is still a challenge to detect a biologically relevant and reliable biomarker. We applied WGCNA to high-throughput B-cell transcriptome data of patients with SLE, and rather than a single gene, a cluster of genes was found to be substantially related to SLE clinical manifestations. These genes were highly enriched in type I IFN pathway. Actually, type I IFN can impact B-cell functions involving in SLE in multiple ways, including effects on receptor engagement, toll-like receptor expression, cell migration, antigen presentation, cytokine responsiveness, cytokine production, survival, differentiation and class-switch recombination.24–29 In the following investigation, we focused on lncRNAs in this gene cluster, which were more pathogenically significant and have greater potential to be reliable biomarkers of SLE. Resultantly, the correlation between SLE phenotypes and gene expression patterns in B cells was established, and a phenotypic essential gene module was detected.
The disease course of SLE is unpredictable and fluctuating. Due to the occurrence of flares could lead to organ damage and poor disease prognosis in lupus patients, the monitoring of disease activity and timely interventions are of great importance. In reality, the progression of the disease is determined by disease-critical cell subpopulations, and the dysregulation of key genes in these subpopulations could reflect disease activity. Our results demonstrated that lncRNA RP11-273G15.2 was coexpressed with coding genes critical to SLE and highly expressed in B cell of patients with SLE compared with healthy controls. Furthermore, the data of another cohort illustrated RP11-273G15.2 expression levels were positively correlated with SLEDAI-2K scores, which were taken as the standard measurement of disease activity in clinical practice. This process would accelerate the progress of biomarker discovery and lupus study. However, establishment of the predictive capacity of RP11-273G15.2 as a biomarker still needs to be validated in the larger SLE cohort.
Our findings illuminate that RP11-273G15.2 coexpresses with ISGs; however, the function and mechanism of RP11-273G15.2-mediating SLE pathogenesis are unclear. The bulk of evidence has demonstrated that lncRNA could disrupt the immune homeostasis involving autoimmune disease pathogenesis by affecting disease critical gene expression. Investigating the function and mechanism of RP11-273G15.2 dysfunction would definitely help to dissect the mechanism driving SLE pathogenesis. Using bioinformatic analysis, we preliminary explored the function of RP11-273G15.2. The online platform ENCORI30 predicts that it could interact with miR-19a-3p and miR-19b-3p. Overexpression of miR-19a and miR-19b was shown to downregulate the expression of IFN-related genes.31 Consequently, RP11-273G15.2 might engage in IFN pathway by acting as a competitive endogenous RNA (ceRNA) to bind miR-19 and prevent its regulatory function. However, these are very preliminary exploration. Future studies are not only tasked with validating whether this lncRNA could act as a biomarker but also with establishing a comprehensive understanding of how this lncRNA drives the SLE process.
Taken together, our study suggested that expression level of RP11-273G15.2 was highly correlated with IFN scores and disease activity in patients with SLE and it might work as a biomarker for SLE diagnosis and disease activity monitoring.
We acknowledged the participants in this study.
Data availability statement
Data are available in a public, open access repository. All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
This study involves human participants and was approved by Ethics Committee of Renji Hospital, Shanghai Jiao Tong University School of Medicine. Participants gave informed consent to participate in the study before taking part.
XZ, YC and ZYin contributed equally.
Correction notice This article has been corrected since it was published. The data of SLE B cells in online supplemental figure 1B-D were incorrect. The supplemental figure 1 and its description in the Results section has been corrected.
Contributors GH, XZ, HD and NS contributed to the conception and design of the study. XZ, YC and ZYin contributed to the bioinformatic analysis and the acquisition of data. YZ, YS, DD, XL, L-HZ and ZYe contributed to the collection of samples and data analysis. All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version of the manuscript for publication. GH was responsible for the overall content of the manuscript and is the guarantor of the work.
Funding This study was supported by grants from the National Natural Science Foundation of China (31930037, 82302024), Shanghai Science and Technology Innovation Plan (21Y31900200) and Sanming Project of Medicine in Shenzhen (SZSM201602087). The funding agencies did not have any role in study design and the collection, analysis and interpretation of data; the writing or the approval of the manuscript.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer-reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
1 Lipsky PE. Systemic lupus erythematosus: an autoimmune disease of B cell hyperactivity. Nat Immunol 2001; 2: 764–6. doi:10.1038/ni0901-764
2 Navarra SV, Guzmán RM, Gallacher AE, et al. Efficacy and safety of belimumab in patients with active systemic lupus erythematosus: a randomised, placebo-controlled, phase 3 trial. Lancet 2011; 377: 721–31. doi:10.1016/S0140-6736(10)61354-2
3 Quinn JJ, Chang HY. Unique features of long non-coding RNA biogenesis and function. Nat Rev Genet 2016; 17: 47–62. doi:10.1038/nrg.2015.10
4 Tang Y, Zhou T, Yu X, et al. The role of long non-coding RNAs in rheumatic diseases. Nat Rev Rheumatol 2017; 13: 657–69. doi:10.1038/nrrheum.2017.162
5 Wang P, Xue Y, Han Y, et al. The STAT3-binding long noncoding RNA LNC-DC controls human dendritic cell differentiation. Science 2014; 344: 310–3. doi:10.1126/science.1251456
6 Yang H, Liang N, Wang M, et al. Long noncoding RNA MALAT-1 is a novel inflammatory regulator in human systemic lupus erythematosus. Oncotarget 2017; 8: 77400–6. doi:10.18632/oncotarget.20490
7 Zhang F, Wu L, Qian J, et al. Identification of the long noncoding RNA NEAT1 as a novel inflammatory regulator acting through MAPK pathway in human lupus. J Autoimmun 2016; 75: 96–104. doi:10.1016/j.jaut.2016.07.012
8 Liao Z, Ye Z, Xue Z, et al. Identification of renal long non-coding RNA RP11-2B6.2 as a positive regulator of type I interferon signaling pathway in lupus nephritis. Front Immunol 2019; 10. doi:10.3389/fimmu.2019.00975
9 Wu Y, Zhang F, Ma J, et al. Association of large intergenic noncoding RNA expression with disease activity and organ damage in systemic lupus erythematosus. Arthritis Res Ther 2015; 17: 131. doi:10.1186/s13075-015-0632-3
10 Cai B, Cai J, Yin Z, et al. Long non-coding RNA expression profiles in neutrophils revealed potential biomarker for prediction of renal involvement in SLE patients. Rheumatology (Oxford) 2021; 60: 1734–46. doi:10.1093/rheumatology/keaa575
11 González LA, Ugarte-Gil MF, Alarcón GS. Systemic lupus erythematosus: the search for the ideal biomarker. Lupus 2021; 30: 181–203. doi:10.1177/0961203320979051
12 Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008; 9: 559. doi:10.1186/1471-2105-9-559
13 Hochberg MC. Updating the American college of rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997; 40: 1725. doi:10.1002/art.1780400928
14 Petri M, Orbai A-M, Alarcón GS, et al. Derivation and validation of the systemic lupus international collaborating clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum 2012; 64: 2677–86. doi:10.1002/art.34473
15 Gladman DD, Ibañez D, Urowitz MB. Systemic lupus erythematosus disease activity index 2000. J Rheumatol 2002; 29: 288–91.
16 Feng X, Wu H, Grossman JM, et al. Association of increased interferon-inducible gene expression with disease activity and lupus nephritis in patients with systemic lupus erythematosus. Arthritis Rheum 2006; 54: 2951–62. doi:10.1002/art.22044
17 Andreoletti G, Lanata CM, Trupin L, et al. Transcriptomic analysis of immune cells in a multi-ethnic cohort of systemic lupus erythematosus patients identifies ethnicity- and disease-specific expression signatures. Commun Biol 2021; 4: 488. doi:10.1038/s42003-021-02000-9
18 Panwar B, Schmiedel BJ, Liang S, et al. Multi-cell type gene coexpression network analysis reveals coordinated interferon response and cross-cell type correlations in systemic lupus erythematosus. Genome Res 2021; 31: 659–76. doi:10.1101/gr.265249.120
19 Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 2003; 13: 2498–504. doi:10.1101/gr.1239303
20 Bhamidipati K, Silberstein JL, Chaichian Y, et al. CD52 is elevated on B cells of SLE patients and regulates B cell function. Front Immunol 2020; 11: 626820. doi:10.3389/fimmu.2020.626820
21 Hao Y, Hao S, Andersen-Nissen E, et al. Integrated analysis of multimodal single-cell data. Cell 2021; 184: 3573–87. doi:10.1016/j.cell.2021.04.048
22 Zhang B, Horvath S. A general framework for weighted gene co-expression network analysis. Stat Appl Genet Mol Biol 2005; 4: Article17. doi:10.2202/1544-6115.1128
23 Ota M, Nagafuchi Y, Hatano H, et al. Dynamic landscape of immune cell-specific gene regulation in immune-mediated diseases. Cell 2021; 184: 3006–21. doi:10.1016/j.cell.2021.03.056
24 Kiefer K, Oropallo MA, Cancro MP, et al. Role of type I Interferons in the activation of autoreactive B cells. Immunol Cell Biol 2012; 90: 498–504. doi:10.1038/icb.2012.10
25 Kirou KA, Lee C, George S, et al. Activation of the interferon-alpha pathway identifies a subgroup of systemic lupus erythematosus patients with distinct serologic features and active disease. Arthritis Rheum 2005; 52: 1491–503. doi:10.1002/art.21031
26 Palanichamy A, Bauer JW, Yalavarthi S, et al. Neutrophil-mediated IFN activation in the bone marrow alters B cell development in human and murine systemic lupus erythematosus. J Immunol 2014; 192: 906–18. doi:10.4049/jimmunol.1302112
27 Liu M, Guo Q, Wu C, et al. Type I interferons promote the survival and proinflammatory properties of transitional B cells in systemic lupus erythematosus patients. Cell Mol Immunol 2019; 16: 367–79. doi:10.1038/s41423-018-0010-6
28 Uccellini MB, Busconi L, Green NM, et al. Autoreactive B cells discriminate CpG-rich and CpG-poor DNA and this response is modulated by IFN-alpha. J Immunol 2008; 181: 5875–84. doi:10.4049/jimmunol.181.9.5875
29 Bekeredjian-Ding IB, Wagner M, Hornung V, et al. Plasmacytoid dendritic cells control TLR7 sensitivity of naive B cells via type I IFN. J Immunol 2005; 174: 4043–50. doi:10.4049/jimmunol.174.7.4043
30 Li J-H, Liu S, Zhou H, et al. starBase V2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data. Nucleic Acids Res 2014; 42: D92–7. doi:10.1093/nar/gkt1248
31 Li J, Lin T-Y, Chen L, et al. miR-19 regulates the expression of interferon-induced genes and MHC class I genes in human cancer cells. Int J Med Sci 2020; 17: 953–64. doi:10.7150/ijms.44377
© 2024 Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY. Published by BMJ. https://creativecommons.org/licenses/by/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/ . Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Ye, Zhizhong 2 ; Ding, Huihua 1 ; Hou, Guojun 1
1 Shanghai Institute of Rheumatology, Renji Hospital, Shanghai Jiao Tong University School of Medicine (SJTUSM), Shanghai, China
2 Shenzhen Futian Hospital for Rheumatic Diseases, Shenzhen, Guangdong, China