# Identification of potential biomarkers related to the Bitong mixture in osteoarthritis based on bioinformatics and network pharmacology, and exploration of the mechanism involved

## TL;DR

This study identifies MMP9, MMP2, and SPP1 as potential biomarkers for Bitong Mixture in treating osteoarthritis and explores their role in immune and matrix regulation.

## Contribution

The study introduces a novel integrative approach combining bioinformatics and experimental validation to identify OA biomarkers for Bitong Mixture.

## Key findings

- MMP9, MMP2, and SPP1 were identified as potential OA biomarkers related to Bitong Mixture.
- BM treatment reduced OA symptoms and expression of these biomarkers in rats.
- Luteolin from BM showed strong binding affinity to MMP9, suggesting a possible mechanism.

## Abstract

Bitong Mixture (BM) has shown efficacy in alleviating pain in knee osteoarthritis (OA) in clinical practice; however, the molecular mechanisms underlying its therapeutic effects remain to be fully elucidated. This study aimed to identified BM-related OA biomarkers and explore their functional implications.

An integrative strategy combining bioinformatics prediction and experimental validation was used. Biomarkers were screened from public OA transcriptomic data using differential expression analysis, network pharmacology, and machine learning. Their functions were explored via enrichment and immune infiltration analyses. Molecular docking predicted interactions between herbal compounds and targets. Single-cell analysis characterized biomarker expression in chondrocyte subsets. A rat OA model and reverse transcription quantitative polymerase chain reaction (RT-qPCR) were employed for in vivo validation.

Bioinformatic prediction identified three potential biomarkers: MMP9, MMP2, and SPP1. They demonstrated certain diagnostic performance for OA and were implicated in pathways related to extracellular matrix organization and immune regulation. Immune analysis revealed significant correlations, notably between MMP2 and activated dendritic cells (cor = 0.66) and between SPP1 and CD4+ central memory T cells (cor = -0.75). Molecular docking suggested strong binding affinity between luteolin (a BM component) and MMP9. Single-cell analysis indicated high expression of these potential biomarkers in hypertrophic chondrocytes, inflammatory chondrocytes, and fibrochondrocytes. In vivo validation confirmed that BM alleviated OA symptoms and histopathological damage in rats. RT-qPCR results showed that BM treatment alleviated the OA-induced upregulation of MMP9, MMP2, and SPP1 expression.

MMP9, MMP2, and SPP1 are potential therapeutic biomarkers for BM in OA. The efficacy of BM may be attributed to its regulation of extracellular matrix remodeling and immune responses, which provides a possible mechanistic explanation for its clinical use.