Article Text

A1.22 NaÏve and memory CD4+T-cell DNA methylation profile in RA
  1. F Ponchel1,
  2. P Chambers1,
  3. A Droop2,
  4. R Parmar1,
  5. J Halstead-Rastrick1,
  6. P Emery1
  1. 1 NIHR-Leeds Musculoskeletal Biomedical Research Unit, Leeds & Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds, UK
  2. 2Leeds Institute of Cancer and Pathology, St James’s University Hospital, University of Leeds, UK


Introduction Several abnormal features of T-cell responses are observed in RA. These range from MHC disease association, repertoire distortion, compromised thymic activity, accelerated ageing, abnormal signalling, formation of secondary ectopic lymphoid structure. More recently the RA genetic predisposition has been shown to reside mostly in T-cell related genes. We investigated DNA methylation profiles in naïve and memory T-cells isolated from healthy controls and RA patients to identify additional levels of regulation in T-cell functionalities.

Methods 6 HC and 10 RA patients (5 ACPA+) were recruited. 30 ml of EDTA blood was collected and naïve and memory T-cells purified using cell sorting. DNA was extracted and treated for bi-sulfite conversion. A 450k micro-bead array (Illumina) was used to investigate genome-wide DNA methylation. QA/QC procedure included inspection of beta-value density plots generated using the “minfi” R package.

Results 30/32 samples passed QA/QC procedure. Signal from methylated and un-methylated probes was normalised and converted in to M values using the “minfi” package. M values were analysed using t-tests in the “gene filter” R package. Multidimentional scaling analysis clearly separated naïve from memory cells suggesting that the phenotype of the cells (notably between two CD4+ subsets) is essential for the detection of disease specific changes.

Memory cell analysis revealed 37050 differentially methylated positions between health and RA and 36750 for naïve cells (P < 0.05). We selected the top 1000 most significant differences for both subsets and compared them. Surprisingly only 180 positions were common to the 2 subsets but importantly included 5 positions for the IL-6 gene, 15 for histone deacethylation-4 (HDAC4), 26 in Wnt3/Frizzle8 and homeobox-1 pathway and 43 in various transcription factors (notably FoxP1) all suggestive of a strong disease effect. Memory cells showed additional differential methylation of positions near genes such SMAD 2/7, p21-CIP-1, FGFR-3, HSP70 and TNFR-member-13. Naïve cells showed similar differential methylation but for different members of these gene families (TNFR-I, TGFR-3, FGFR-1, HSP40) in addition to other specific genes such as interferon regulator 1, SUMO-1, several PAX transcription factors, P-selectin and importantly the IL-17R.

Conclusion These pilot data demonstrate that there are important differences in the methylation of DNA between RA patients and HC that can be identified in relevant pathways provided cells are separated into the same phenotype (naïve vs. memory) even if from the same lineage (both CD4+T-cells).

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.