GENETIC MARKERS OF EARLY EQUINE OSTEOARTHRITIS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1014770
• Project Start Date: Dec 15, 2017
• Project End Date: Sep 30, 2019
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF ILLINOIS
2001 S. Lincoln Ave.
URBANA,IL 61801

Performing Department
Veterinary Research & Extension

Non Technical Summary
Osteoarthritis (OA) is the most common cause of chronic lameness in horses, placing a significant economic burden on the equine industry. Despite widespread awareness of this disease, development of effective tools for early diagnosis and therapeutic intervention remains elusive. We hypothesize that changes in gene expression in the synovium (tissue lining the joints) provide the earliest reflection of disease-related changes in the joint and can thus provide insight into the onset and progression of early OA. Synovium is known to be a key player in OA due to its role in inflammation and further, is easily collected via minimally-invasive approaches, making it an ideal target for a future clinically-applicable tissue-based diagnostic test. In this project, we will use quantitative genomic techniques (RNAseq) to identify genes that may play an important role in the development and early progression of OA utilizing an equine experimental model that induces mild disease consistent with early naturally-occurring OA. These high-throughput research tools allow for comprehensive identification of gene expression changes in paired arthritic and control synovial tissue samples. The proposed work will generate crucial preliminary data and lay the foundation for future studies exploring the specific mechanisms by which altered genes expression results in functional changes within the joint. This is an important first step towards our long-term goal of identifying novel diagnostic markers and therapeutic targets that could lead to a reduction in both individual animal morbidity and the economic impact of OA in agricultural species.

Animal Health Component

50%

Research Effort Categories

Basic

50%

Applied

50%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
303	3810	1080	100%

Knowledge Area
303 - Genetic Improvement of Animals;

Subject Of Investigation
3810 - Horses, ponies, and mules;

Field Of Science
1080 - Genetics;

Keywords

osteoarthritis

horse

transcriptomics

differential expression

animal model

Goals / Objectives
We hypothesize that early OA-related changes in the joint are reflected by upregulation of synovial expression of genes involved in cell proliferation and stimulus response pathways. Altered gene expression can be reliably quantified using comprehensive high-throughput sequencing (RNAseq), and pathway analysis provides a framework for evaluating interactions of differentially expressed genes and associated key cellular processes involved in disease pathogenesis. In Objective 1 we will determine the relative importance of key cellular processes in early OA pathogenesis. We will use both gene-based (Sub-Objective 1A) and network-based approaches (Sub-Objective 1B) to analyze differential expression in paired control and arthritic samples.

Project Methods
Synovial samples werepreviously collected(under an approved IACUC protocol from another institution) from the metacarpophalangeal joints of 11 adult horses before (control) and 16 weeks after (arthritic) experimental induction of OA, as well as from sham-operated joints in five of these horses. Samples were placed in RNAlater (Qiagen, Valencia, CA) and stored at -80 degrees C. RNA will be isolated from the frozen samples and submitted for RNAseq using paired-end sequencing on an Illumina HiSeq 2500 sequencer with a total output of approximately 40-50 million reads per sample.Quality control and data analysis will be carried out using an established best practices bioinformatics pipeline. Briefly, quality control, followed by alignment of sequence data and identification of splice junctions, will be performed using Trimmomatic and TopHat2 software, respectively. Assembled transcriptome data will be deposited with GenBank for manual curation and public release. Quantified gene expression in control, arthritic, and sham-operated samples will be normalized by Cufflinks using fragments per kilobase per million reads mapped (FPKM) to address expected inter-individual variation. Expressed genes will be functionally annotated using the DAVID bioinformatics database (http://david.abcc.ncifcrf.gov/) to identify those clustering within the cell proliferation (Gene Ontology [GO]:0008283) and stimulus response (GO:0051716) pathways. Differential expression between paired samples will be analyzed using DEGseq. Samples will be compared both within individuals (control and arthritic samples from the same joint) and between time points (all control versus all arthritic samples). Sham-operated joints will be compared to both normal and arthritic samples. Statistical significance for differential expression at a FDR of 10% after Benjamini-Hochberg multiple testing adjustment will be set at p < 0.05. Genes from the cell proliferation and stimulus response pathways that are significantly differentially expressed in arthritic samples compared to control samples will be analyzed using STRING and IPA to visualize interactions and evaluate putative functional effects.In Sub-aim 1b, FPKM values for all genes in the RNAseq data will be used to construct co-expression networks using the open source network analysis software Camoco (Co-Analysis of Molecular Components) (https://github.com/schae234/camoco). Genes with > 20% missingness or < 5 FPKM will be removed as part of the program's quality control pipeline. State-specific co-expression networks will be constructed across all control, arthritic, and sham samples. Network interactions will be quantified by calculating the Pearson correlation coefficient between each possible pair of genes. A global significance threshold of Z ≥ 3 will be set on co-expression interactions. Basic network statistics, including gene degree (the number of significantly co-expressed interactions of a gene) and network density will be calculated. Once assembled, co-expression networks will be tested for gene enrichment within annotated GO pathways. Statistical significance for enrichment after Bonferroni correction for multiple testing will be set at p < 0.05. Pathways that are enriched in the arthritic samples, but not in the control or sham samples will be considered of putative functional importance in disease development and/or progression.

Progress 12/15/17 to 09/30/19

Outputs
Target Audience:Members of the target audience included the international community of osteoarthritis researchers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project provided training for an undergraduate Animal Sciences student. How have the results been disseminated to communities of interest?Data were presented as abstracts at three meetings in 2019. One meeting was regional, one national, and one international. All were aimed at orthopedic researchers. Kemper, A.M., Trumble, T.N., Boyce, M.K., Brown, M.P., McCoy, A.M. Altered synovial gene expression reflects early changes in post-traumatic osteoarthritis in a novel animal model. Orthopedic Research Society, Austin, TX, February 2019. [poster]; published in Osteoarthritis Cart. 2019; 27(Suppl. 1):S291-S292. Kemper, A.M., Trumble, T.N., Boyce, M.K., Brown, M.P., McCoy, A.M. Altered synovial gene expression reflects early changes in post-traumatic osteoarthritis in a novel animal model. OARSI World Congress on Osteoarthritis, Toronto, ON, May 2019. [poster]. McCoy, A.M., Kemper, A.M., Trumble, T.N. Differential gene expression analysis reveals pathways important in early post-traumatic osteoarthritis in an equine model. Orthopedic Research Society Great Lakes/Midwest Regional Symposium, Chicago, IL, August 2019. [poster]. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Post-traumatic osteoarthritis (PTOA) is common and its progression cannot be reversed by currently available drugs. The ability to identify disease onset and predict progression are priorities in improving the management of PTOA. Our objective was to quantify gene expression changes in the synovium of OA-affected joints in a novel large animal model specifically designed to recapitulate early PTOA. We hypothesized that early PTOA-related changes in the joint are reflected by altered synovial gene expression, particularly in genes that fall within pathways related to inflammation and cellular response to stimuli.RNA was successfully extracted from 28 samples (6 preOA, 11 OA, 11 sham). Sequencing yielded 15.7-29.4 million paired-end reads per sample. 'Sham' and 'preOA' samples were not different and were grouped. 321 genes were upregulated and 351 genes were downregulated in OA synovium compared to unaffected. Gene ontology (GO) terms related to extracellular matrix (ECM) organization and growth factor binding were overrepresented among DE genes. There were 20 significantly enriched pathways; these included pathways involved in ECM turnover, O-glycosylation of TSR domain-containing proteins, and growth factor signaling.Most enriched pathways and overrepresented GO terms in our data reflect a state of high metabolic activity and tissue turnover in OA-affected tissue, suggesting efforts at healing and restoring homeostasis. TSR domain-containing proteins play a role in many processes including inflammation, development, and wound healing. Limitations of this study include a small sample size and capture of a single time point post-injury. Additionally, changes in gene expression do not always result in changes in protein expression; work to address this point is ongoing. DE genes falling within key pathways may represent potential diagnostic markers or therapeutic targets for PTOA.

Publications

Progress 12/15/17 to 09/30/18

Outputs
Target Audience: Nothing Reported Changes/Problems:Five samples were too small/too degraded for us to isolate RNA of sufficient quantity/quality to sequence, thus we will not be able to analyze as many samples as originally planned. What opportunities for training and professional development has the project provided?This project provided an opportunity for training an undergraduate student in molecular genetics techniques, including RNA isolation. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?Differential gene expression analysis and functional annotation of differentially expressed genes is ongoing. Network analysis is planned during the next reporting period.

Impacts
What was accomplished under these goals? RNA was successfully isolated from 28 samples from 11 individuals; these represented 6 control (pre-OA), 11 sham, and 11 affected (OA) samples. RNA was sequenced using an Illumina HiSeq4000, yielding ~16-29 million paired-end reads per sample. Reads underwent quality control prior to mapping to EquCab3.0 with STAR. Normalized gene counts were used for differential expression analysis. Gene ontology (GO) annotation for all differentially expressed genes was assigned using PANTHER. A comparison of differentially expressed (DE) genes in this data set with those reported in previous data sets that used end-stage tissue showed minimal overlap in the gene set. Furthermore, DE genes in the current data set were enriched for pathways related to immunity/inflammation and tissue homeostasis, while previously reported DE genes were enriched for pathways related to proteolysis and fatty acid metabolism. This suggests that the current data reflects a different stage in the pathogenesis of OA than does previously reported data. Analysis using the gene-based approach (Objective 1A) is ongoing, while network-based analyses are planned (Objective 1B).

Publications