Article Text

Download PDFPDF

507 RNAseq gene expression confirms the importance of GWAS associated risk genes in lupus nephritis
  1. Mikhail Olferiev,
  2. Dina Greenman,
  3. Jeffrey Zhang-Sun,
  4. David Fernandez,
  5. Kyriakos A Kirou and
  6. Mary K Crow
  1. Hospital for Special Surgery, New York, USA


Background The era of GWAS studies identified over 90 SLE risk loci, while the genetic risk factors associated with lupus nephritis (LN) require further study. The absence of strong associations might reflect uncertainty about future LN status or insufficient coverage within the array. To highlight the importance of previously identified risk genes and pathways predisposing to LN, we investigated PBMC RNAseq data from a cohort of SLE patients followed for several years with and without biopsy-proven LN.

Methods PBMC RNAseq data, including some longitudinal data, from 46 LN samples (30 patients) and 44 samples from SLE patients without LN (28 patients) were studied. The analysis of differentially expressed genes was performed using the limma R package. The reported LN genetic loci were collected from published data. The regularized logistic regression was used to select the most important genes.

Results Comparing LN and non-LN samples, 109 genes were differentially expressed between groups (logFC 1.5, 5% FDR). The functional analysis identified genes related to glomerular membrane formation (COL4A3, COL9A3, MMP9), WNT signaling (WNT1, WNT7A, TPBG), and cell adhesion (FBLN7, ADAM23, TNFAIP6). To find the most informative genes to distinguish LN patients, we used 3 models based on: (1) differentially expressed genes, (2) GWAS reported genes, (3) a combination of the above. All models were based on the same 67 (70%) training and 23 (30%) testing sample sets and efficiently segregated LN patients (AUC 0.9, 0.8, and 0.9). Despite the equal efficiency of the first and the third model, the inclusion of IKZF1 and PRPF18 genes reduced the number of required predictor genes.

Conclusion LN might be an initial presentation of SLE disease or can have a late-onset. The analysis of longitudinal samples helps classify SLE patients correctly and may be of predictive value. Polymorphism in IKZF1 was reported in association with several autoimmune diseases and found to relate to Th and dendritic cell activation. The PRPF18 gene encodes a pre-mRNA splicing factor. Interestingly, another polymorphism at that locus is highly associated with darker skin color in the African-American population, a group with high risk for severe lupus. As a limitation of our study is the relatively small number of participating SLE patients, additional patient data will be needed to confirm these results.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.