Back

Microfluidic Co-Culture for Modeling Human Joint Inflammation in Osteoarthritis Research

Mirazi, H.; Wood, S. T.

2025-01-29 bioengineering
10.1101/2025.01.27.635089 bioRxiv
Show abstract

Here we present a microfluidic model that allows for co-culture of human osteoblasts, chondrocytes, fibroblasts, and macrophages of both quiescent (M0) and pro-inflammatory (M1) phenotypes, maintaining initial viability of each cell type at 24 h of co-culture. We established healthy (M0-based) and diseased (M1-based) joint models within this system. An established disease model based on supplementation of IFN-{gamma} and LPS in cell culture media was used to induce an M1 phenotype in macrophages to recapitulate inflammatory conditions found in OA. Cell viability was assessed using NucBlue Live and NucGreen Dead fluorescent stains, with mean viability of 83.9% {+/-} 14% and 83.3% {+/-} 12% for healthy and diseased models, respectively, compared with 93.3% {+/-} 4% for cell in standard monoculture conditions. Cytotoxicity was assessed via a lactate dehydrogenase (LDH) assay and showed no measurable increase in LDH release into the culture medium under co-culture conditions, indicating that neither model promotes a loss of cell membrane integrity due to cytotoxic effects. Cellular metabolic activity was assessed using a PrestoBlue assay and indicated increased cellular metabolic activity in co-culture, with levels 5.9 {+/-} 3.2 times mean monolayer cell metabolic activity levels in the healthy joint model and 5.3 {+/-} 3.4 times mean monolayer levels in the diseased model. Overall, these findings indicate that the multi-tissue nature of in vivo human joint conditions can be recapitulated by our microfluidic co-culture system at 24 h and thus this model serves as a promising tool for studying the pathophysiology of rheumatic diseases and testing potential therapeutics.

Matching journals

The top 13 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.