Background/Purpose: Once considered non-functional, the vast regions of the human genome that do not contain protein-coding genes are now known to contain important regulatory elements.  These regulatory elements include both enhancer regions and regions encoding regulatory RNA transcripts such as long non-coding RNA (lncRNA).  These studies were undertaken to examine in detail the transcriptome of  neutrophils in children with polyarticular juvenile idiopathic arthritis (JIA). 

Methods: We isolated RNA from neutrophils of children with 3 different phenotypes: children with active, untreated disease (AD, n=3), children on medication who fit criteria for remission by the Wallace criteria (CRM, n=3), and children with cystic fibrosis (CF, n=3), a control group designed to distinguish those transcriptome features specific to JIA and those common to other forms of chronic, soft tissue inflammation.  RNA samples were prepared for sequencing using the Illumina TruSeq RNA prep kit. Sequencing was performed using the Illumina HiSeq 2000. The pass filter reads were mapped to the genome (NCBI/build 37.2 ) using TopHat (version 2.0.4). Probable transcripts were assembled using Cufflinks (version 2.0.2), and differential expressed transcripts were determined using DESeq. Fold change (FC) calculations were obtained using the log2(FPKM) ratio, where FPKM is the fragments per kilobase of exon model per million mapped fragments.

Results: For all phenotypes, we detected between 8,000-9,000 expressed genes, with children in CRM showing the highest number of expressed genes (n=8,668).  Of these, 7,734 were common to all 3 phenotypes.  Multiple isoforms of known genes were expressed in the 3 phenotypes, most of which were common to all 3. Of the protein-coding regions, there are 62 differentially expressed genes (29 down-regulated, 33 up-regulated) in fold change 1.2 or greater when AD compared to CRM, 64 DE genes (6 down-regulated, 58 up-regulated) in the AD vs CF comparison and 90 DE (40 down-regulated, 50 up-regulated,) in CRM vs CF. Principle component analysis (PCA) and hierarchical cluster analysis of DE genes showed clear separation between AD, CRM and CF. The functional enrichment analysis using DAVID for the DE genes shows they are associated with immune response (IFIH1, C4A, IFITM3, TNFSF15, APOBEC3G, OAS1, C4BPA, TRIM22, NOD2, FCGR2C, P2RY14, HLA-DRB5, ODZ1, CFD), innate immune response (IFIH1, NOD2, C4A, APOBEC3G, C4BPA, CFD), and inflammatory response (ORM1, TNFAIP6, C4A, IDO1, C4BPA, CFD, SPP1), etc. Finally, novel, previously uncharacterized transcripts were seen in each of the 3 phenotypes (n=5 for AD, n=11 for CRM, and n=9 for CF).  These transcripts may represent previously-unknown lncRNA transcripts.

Conclusion: The sensitivity and dynamic range of RNAseq allow a much more detailed view of the neutrophil transcriptome.  The finding of state-specific transcripts is expected to lead to new insights into JIA pathogenesis, response to therapy, and the development of diagnostic and prognostic biomarkers.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/detection-of-state-specific-rna-transcripts-in-juvenile-idiopathic-arthritis-neutrophils-using-rna-sequencing/