Article Text

Download PDFPDF

  1. Cai L Davies1,2,
  2. Malin C. Erlandsson1,3,
  3. Sofia Töyrä Silfverswärd1,
  4. Karin Me Andersson1,
  5. Maria I Bokarewa1,3
  1. 1University of Gothenburg, Department of Rheumatology and Inflammation Research, Gothenburg, Sweden
  2. 2Keele University, Stoke-On-Trent, United Kingdom
  3. 3Sahlgrenska University Hospital, Gothenburg, Sweden


Background Rheumatoid arthritis (RA) is more prevalent in females. It is reported to be alleviated during pregnancy, and increase in severity during menopause, which implies estrogen as an important contributor in RA pathogenesis. Micro-RNA (miR) are short, non-coding RNAs, that act within a RNA-induced silencing complex. miRs have recently emerged as important epigenetic controls of leukocyte maturation and function.

Objectives To study the epigenetic effect of estrogen on the transcription of microRNA and their bio-processor enzymes in RA patients.

Methods The leukocytes of 145 female RA patients split for estrogen receptor alpha (ERα) (dCT 9.57) were analyzed for the expression of Dicer, Drosha and DGCR8 mRNA by RT-PCR and serum levels of TGFb-dependent cytokines IL4 and IL9. Micro-RNA transcription array was performed by 3D-Gene™ microarray measuring >2560 miRs (TATAA Biocenter, Gothenburg) in human primary leukocytes, fat tissue and plasma. The samples were split by expression of estrogen receptor alpha (ERα, dCT 9.57), a proxy of an active estrogen signaling, and used to identify miRs of interest. Bioinformatic analysis was performed using DIANA, miRDB, miRTarBase and Ensembl, using microRNA nucleotide sequences obtained from miRBase, to predict signaling pathways and gene targets of miR. To confirm estrogens effect on miR processing, leukocyte cultures of RA patients were exposed to estrogen and subjected to miR, gene and protein analysis.

Results Comparing leukocytes with different ERα, we identified 214 miRs with high and 7 miRs with low expression when ERα was high. Cross-analysis in the fat and serum sample miR array identified most of those miRs. Bioinformatic analysis of the upregulated miRs confirmed that 16miRs were involved in the estrogen signaling pathway (p=0.0063) and 15 TGFb signaling pathway (p=<0.0001), where these miRs had 61 common predicted gene targets.

To study if transcription of these predicted targets was dependent on estrogen signaling, we took advantage of the clinical RA material. Patients with high ERα were significantly younger (51y vs 61y, p<0.00001), which confirmed active estrogen signaling. High ERα had significantly higher expression of miR bioprocessing enzymes Dicer (p=0.0197), Drosha (p=0.0454), DGCR8 (P=0.0192) and lower disease activity (DAS28, p=0.0023; ESR p=0.005, IL6, p=0.0129).

Exposure of leukocytes to estrogen in culture induced a significant reduction in IL9 (p=0.0078), which suggests a suppression of TGFb signaling.

Conclusion High ERα expression can be considered a significant regulator of miR transcription in leukocytes. miR and bio-processors regulated by estrogen could be of importance in the pathogenesis on RA by targeting genes responsible for regulation of inflammation and the non-protective elements of RA development; explaining the ameliorating effects of estrogen.

Disclosure of Interests None declared

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.