Single-Cell RNA SequencingBioinformatic Pipeline Optimized for Non-Coding GenesIdentified LINC01503 and MIR193-BHG as Potential Pathogenic Effectors in Systemic Sclerosis Fibroblasts

Sophie Wagner¹, Elena Pachera², Gino Andrea Bonazza³, Celina Geiss⁴, Astrid Hofman⁴, Laura Much⁴, Lumeng Li⁴, Pietro Bearzi³, Anna-Maria Hoffmann-Vold⁵ and Oliver Distler⁶, ¹University of Zurich, Schlieren, Switzerland, ²University Hospital Zurich, Zurich, Switzerland, ³Center of Experimental Rheumatology, Department of Rheumatology, University Hospital Zurich, University of Zurich, Schlieren, Switzerland, ⁴Center of Experimental Rheumatology, Department of Rheumatology, University Hospital Zurich, University of Zurich, Zurich, Switzerland, ⁵Oslo University Hospital, Oslo, Norway, ⁶Department of Rheumatology, University Hospital Zurich, University of Zurich, Zurich, Switzerland, Zurich, Switzerland

Meeting: ACR Convergence 2024

Keywords: Fibroblasts, Dermal, Non-coding RNA, skin, Systemic sclerosis

Session Information

Date: Monday, November 18, 2024

Title: Systemic Sclerosis & Related Disorders – Basic Science Poster II

Session Type: Poster Session C

Session Time: 10:30AM-12:30PM

Background/Purpose: Long non-coding RNAs (lncRNA) are a class of transcripts that regulate gene expression. Currently, only few lncRNAs have been linked to SSc pathogenesis. As lncRNAs often evade large screens due to their high cell specificity and low copy numbers, distinct analysis methods are needed to identify differentially expressed lncRNAs. Here, we aimed to identify candidate lncRNAs that may play a role in SSc using a bioinformatic pipeline optimized for non-coding genes.

Methods: Single cell RNA sequencing (scRNA-seq) data from sex matched SSc and healthy control (HC) skin biopsies were obtained with the 10X Genomics workflow. To analyze lncRNAs expression, we optimized a scRNA-seq bioinformatic pipeline, which included STARsolo alignment and differential gene expression analysis (DGEA) with DESeq2. Human dermal fibroblasts were isolated from skin biopsies of SSc patients and HC. LncRNA expression was analyzed by RT-PCR. Cells were treated with 10 ng/mL TGF-β1. LncRNAs MIR193-BHG and LINC01503 were silenced in SSc dermal fibroblasts using locked nucleic acid antisense oligonucleotides (LNA GapmeRs), followed by bulk RNA sequencing (bulk RNA-seq) and subsequent DGEA as well as Gene Set Enrichment Analysis (GSEA).

Results: Our optimized bioinformatic pipeline enabled us to account for the low expression levels and cell specificity of lncRNAs. With a focus on fibroblasts, DGEA revealed 23 differentially expressed lncRNAs from SSc patients (n=7) as compared to HCs (n=6). Pseudo bulk analysis provided the five top lncRNAs differentially expressed in SSc fibroblasts (CYTOR, ENSG00000287853, LINC01503, MIR193-BHG, ZNF295-AS1). Subsequent RT-PCR confirmation analysis in cultured fibroblasts showed a significant 1.89-fold increase of LINC01503 basal gene expression in SSc fibroblasts as compared to HC fibroblasts (n=8, p< 0.05). TGF-β1 stimulation of fibroblasts led to a 2.96-fold increase of LINC01503 gene expression in SSc cells (n=6, p< 0.01). Conversely, basal MIR193-BHG expression was not significantly altered between SSc and HC fibroblasts, MIR193-BHG expression was significantly induced by TGF-β1 stimulation 1.42-fold in HC cells (p< 0.001) and by 1.39-fold in SSc cells (p< 0.01) respectively. Knockdown of LINC01503 in SSc fibroblasts followed by bulk RNA-seq analysis showed TAPT1 and DUSP6 among the top differentially expressed genes. TAPT1 protein instability has been linked to impaired collagen fibril formation. DUSP6 negatively regulates members of the Mitogen-Activated Protein Kinase (MAPK) family. Knockdown of MIR193-BHG in SSc fibroblasts lead to the downregulation of several extracellular matrix genes (SULF1, RB1, B4GALT1, CCDC80, CRISPLD2, FBLN5, ADAMTS5, RUNX1, FLRT2, VIT). GSEA showed “extracellular matrix organization”, “extracellular structure organization” and “extracellular matrix” among the top enriched gene sets.

Conclusion: Our targeted bioinformatic analysis revealed previously overlooked lncRNAs LINC01503 and MIR193-BHG, which may play a role in SSc pathogenesis. These findings contribute to a deeper understanding of disrupted regulatory processes in SSc skin fibrosis and highlight the importance of exploring lncRNAs to better understand SSc pathomechanisms.

Table 1: Top 5 lncRNA targets and their respective values for several metrics. Base mean = mean count in all fibroblasts. Mean HC = mean count in all HC derived fibroblasts. Mean SSc = mean count in all SSc derived fibroblasts. Log2FC = log2 fold change between HC and SSc gene expression, positive values indicate more abundance in SSc derived cells. Padj = adjusted p value. Biotype = Classification of the gene based on Ensembl annotation using BioMart. No. Cells = number of fibroblasts that express the gene.

Figure 1: Gene expression differences between HC and SSc sc-RNA seq fibroblast data. Volcano plot representing the gene expression difference (log2 fold change) of non-coding RNA between SSc and HC fibroblasts and their respective adjusted p-values (Benjamini and Hochberg). Genes that show absolute log2 fold changes > 1 and p.adj <= 0.05 are displayed in red. MIR193-BHG and LINC01503 are indicated. Positive log2 fold change means the gene is more abundant in SSc patient fibroblasts.

Figure 2: Effects of LINC01503 and MIR193-BHG knockdown on gene expression in SSc derived fibroblasts. (A) Heat map showing centered and row scaled gene expression of differently expressed genes (rows) per sample (columns) upon LINC01503 knockdown. TAPT1 and DUSP6 are indicated in red. (B) Bar plot showing significantly downregulated processes upon MIR103-BHG knockdown. Count = Number of differentially expressed genes affiliated with each process. Processes related to extracellular matrix are indicated in red.

Disclosures: S. Wagner: None; E. Pachera: None; G. Bonazza: None; C. Geiss: None; A. Hofman: None; L. Much: None; L. Li: None; P. Bearzi: None; A. Hoffmann-Vold: Arxx Therapeutics, 2, Boehringer Ingelheim, 2, 5, 6, 12, Support for travel, Genentech, 2, Janssen, 2, 5, 6, Medscape, 2, 6, 12, Support for travel, Merck/MSD, 2, Novartis, 6, Pliant Therapeutics, 2, Roche, 2, 6, 12, Support for travel, Werfen, 2; O. Distler: 4P-Pharma, 2, “mir-29 for the treatment of systemic sclerosis” (US8247389, EP2331143), 10, AbbVie, 2, Acceleron, 2, Alcimed, 2, Altavant Sciences, 2, Amgen, 2, AnaMar, 2, Arxx, 2, AstraZeneca, 2, Bayer, 2, 6, Blade Therapeutics, 2, Boehringer Ingelheim, 2, 5, 6, Citrus AG, 12, Co-founder, Corbus Pharmaceuticals, 2, CSL Behring, 2, EMD Serono, 2, ERS/EULAR Guidelines, 12, Co-Chair, EUSTAR, 12, President, FOREUM Foundation, 12, Chair of Executive Committee, Galapagos, 2, Glenmark, 2, Gossamer, 2, Hartmann Müller Foundation, 12, Member Board of Trustees, Horizon, 2, Janssen, 2, 6, Kymera, 2, 5, Lupin, 2, Medscape, 2, 6, Merck/MSD, 2, Miltenyi Biotec, 2, Mitsubishi Tanabe, 2, 5, Nkarta Inc., 2, Novartis, 2, Orion, 2, Prometheus Biosciences, 2, Redxpharma, 2, Roivant, 2, Swiss Academy of Medical Sciences, 12, Senat Member, Swiss Clinical Quality Management in Rheumatic Diseases, 12, Member Board of Trustees, Topadur, 2, UCB, 2.

To cite this abstract in AMA style:

Wagner S, Pachera E, Bonazza G, Geiss C, Hofman A, Much L, Li L, Bearzi P, Hoffmann-Vold A, Distler O. Single-Cell RNA SequencingBioinformatic Pipeline Optimized for Non-Coding GenesIdentified LINC01503 and MIR193-BHG as Potential Pathogenic Effectors in Systemic Sclerosis Fibroblasts [abstract]. Arthritis Rheumatol. 2024; 76 (suppl 9). https://acrabstracts.org/abstract/single-cell-rna-sequencingbioinformatic-pipeline-optimized-for-non-coding-genesidentified-linc01503-and-mir193-bhg-as-potential-pathogenic-effectors-in-systemic-sclerosis-fibroblasts/. Accessed .

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/single-cell-rna-sequencingbioinformatic-pipeline-optimized-for-non-coding-genesidentified-linc01503-and-mir193-bhg-as-potential-pathogenic-effectors-in-systemic-sclerosis-fibroblasts/