ACR Meeting Abstracts

ACR Meeting Abstracts

Abstract Number: 0051

Identification of Biomarkers and Deregulated Pathways in Psoriatic Arthritis Through Proteomic Analysis of Synovial Fluid

Jacqueline Lai1, Sara Rahmati2, Raam Sivakumar2, Katerina Oikonomopoulou3, Fatima Abji4 and Vinod Chandran5, 1McMaster University, Hamilton, ON, Canada, 2University Health Network, Toronto, ON, Canada, 3Krembil Research Institute, Toronto, ON, Canada, 4University of Toronto, University Health Network, Toronto, ON, Canada, 5Schroeder Arthritis Institute, Krembil Research Institute, Toronto Western Hospital, Toronto, ON, Canada

Meeting: ACR Convergence 2021

Keywords: , , , ,

Session Information

Date: Saturday, November 6, 2021

Title: Spondyloarthritis Including PsA – Basic Science Poster (0046–0068)

Session Type: Poster Session A

Session Time: 8:30AM-10:30AM

Background/Purpose: Psoriatic arthritis (PsA) is an inflammatory arthritis that currently lacks diagnostic biomarkers. We hypothesized that there are differences in the protein intensities in the synovial fluid (SF) of patients with osteoarthritis (OA), rheumatoid arthritis (RA) and PsA. We aimed to identify markers for PsA by proteomic analysis of knee joint SF. Our objectives were to identify proteins and pathways differentially expressed between:

1. OA and the inflammatory arthritides (PsA and RA).
2. PsA and RA.

Methods: Mass spectrometry was used to identify the SF proteome from 10 OA, 10 PsA, and 10 RA patients. All patients satisfied the classification criteria for their disease. First, we used student’s t-test for differential intensity analysis (p-value < 0.05; fold change > 1.5) to identify potential protein biomarkers of each disease. Next, we performed protein-protein interaction (PPI) network and pathway enrichment analysis to obtain biological insights on systems biology of these diseases.

Results: Aim 1: 4 proteins were present and 45 were absent in OA SF compared to PsA and RA. Among those present in all conditions, we detected 56 and 66 proteins with differential intensities in OA compared to PsA and RA, respectively. Hierarchical clustering and principal component analysis confirmed the power of these proteins in distinguishing OA from the inflammatory arthritides (Fig 1). In both comparisons, enriched pathways were involved in inflammation and immune response. Additionally, PPI network analysis of both lists revealed connected subnetworks of proteins with high fold change enriched in immune-system and extracellular matrix (Fig 2a,b). In both comparisons, VIM showed the highest fold change and was central in connecting other high fold change proteins.

Aim 2: 22 proteins with differential intensities between PsA and RA were identified. In addition to immune system pathways, these proteins were enriched in cell metabolism pathways, such as glycolysis which has been previously suggested to distinguish PsA and RA. PPI analysis identified three densely connected network modules (Fig 3a). All upregulated proteins in PsA formed one dense module enriched in complement and clotting pathways (Fig 3b). We identified one highly densely connected subnetwork enriched in transcription-related pathways (Fig 3c) which has four interactions with one upregulated protein (VCAM1). All other proteins were downregulated and have many interactions with the upregulated module. These proteins were highly enriched in glycolysis and glycogenesis, followed by immune pathways (Fig 3d). Previous research has shown interplay between glycolysis and complement pathways. Unlike in Aim 1, VIM shows small but statistically significant differences.

Conclusion: We have identified differential proteins and pathways between PsA, OA, and RA. VIM was a central protein in connecting proteins with differential intensities between OA and the inflammatory arthritides. Additionally, we identified three large network modules enriched in complement, transcription, and glycolysis pathways that may distinguish PsA from RA. Our results suggest disrupted interplay between complement and glycolysis pathways in PsA vs RA.

Figure 1- Hierarchical clustering (a) and principal component analysis (b) distinguish OA from the inflammatory arthritides based on mass spectrometry data of proteins with significant differential intensities in at least one of the three comparisons.

Figure 2. Comparison between OA and the inflammatory arthritides a) OA vs RA b) OA vs PsA. To build the network, we first annotated the physical PPI network (obtained from http://iid.ophid.utoronto.ca) with differential intensities. We expanded the list of differential proteins by relaxing the thresholds for fold change and p-value. Instead, they must interact with at least another differential protein. Next, we selected proteins whose PPI count (degree) and/or fold change was within the top 10th percentile of protein degrees and/or fold changes in each network. Finally, we arranged all other proteins in a big circular layout around these selected proteins. Selected proteins with a high degree form a densely connected network module in both comparisons. In addition, they maintain physical connectivity among proteins with high fold change. PPIs within the proteins inside the large circular layout are shown in red. Networks are visualized in http://navigator.ophid.utoronto.ca/navigatorwp.

Figure 3. Network and pathway analysis of differential proteins between SF of PsA vs RA. a) To build the network, we first annotated the physical PPI network with differential intensities. We expanded the list of differential proteins by relaxing the thresholds for fold change and p-value. Instead, they must interact with at least another differential protein. Analysis of this network revealed three densely connected modules involved in different pathway groups. Module 1: covers all of the upregulated proteins in PsA compared to RA (arranged in the smallest circular subnetwork); module 2: consists of downregulated proteins with high intra-module connectivity (formed mainly by histone proteins); and module 3: includes all downregulated proteins not in module 2 (arranged on the large network circle). Most of these proteins have interactions with at least one upregulated protein (module 1). b) Key-terms in titles of pathways enriched in proteins of module 1 shows strong enrichment in complement pathways. c) Module 2 is enriched in pathways related to transcription. Polymerase, RNA, and Promoter are three major keywords in the titles of these pathways. This module was also connected to only one upregulated protein in the whole network – VCAM1. d) Proteins in module 3 are enriched in glycolysis and gluconeogenesis. Most proteins in module 3 have PPIs with proteins in module 1 that suggest differential interplay between complement and glycolysis pathways in PsA vs RA. Pathway and key-term analysis are done by http://ophid.utoronto.ca/pathDIP/ and word-clouds are made by http://www.edwordle.net.

Disclosures: J. Lai, None; S. Rahmati, None; R. Sivakumar, None; K. Oikonomopoulou, None; F. Abji, None; V. Chandran, Abbvie, 1, 2, 5, Amgen, 1, 2, 5, Eli Lilly, 1, 2, 5, BMS, 2, 5, Janssen, 1, 2, 5, Novartis, 1, 2, 5, Pfizer, 1, 2, 5, AstraZeneca, 12, Spousal employment, Celgene, 2, 5, UCB Pharma, 2, 5.

To cite this abstract in AMA style:

Lai J, Rahmati S, Sivakumar R, Oikonomopoulou K, Abji F, Chandran V. Identification of Biomarkers and Deregulated Pathways in Psoriatic Arthritis Through Proteomic Analysis of Synovial Fluid [abstract]. Arthritis Rheumatol. 2021; 73 (suppl 9). https://acrabstracts.org/abstract/identification-of-biomarkers-and-deregulated-pathways-in-psoriatic-arthritis-through-proteomic-analysis-of-synovial-fluid/. Accessed .

« Back to ACR Convergence 2021

ACR Meeting Abstracts - https://acrabstracts.org/abstract/identification-of-biomarkers-and-deregulated-pathways-in-psoriatic-arthritis-through-proteomic-analysis-of-synovial-fluid/