Dynamic BMP10 Release Reflects Atrial Fibrillation Burden in Human Atrial Engineered Heart Tissue

Atrial fibrillation (AF) burden plays a crucial role in determining clinical risk, yet it has been difficult to estimate using long-term rhythm monitoring. Researchers have investigated a potential biomarker for AF burden in a human atrial model: atrial engineered heart tissue (aEHT). This tissue is derived from human induced pluripotent stem cells and produces a protein called bone morphogenetic protein 10 (BMP10), which is associated with AF and its complications.

To assess the relationship between BMP10 release and AF burden, the researchers mimicked AF by optogenetically pacing mature aEHTs at high rates. There were two pacing regimes: intermittent pacing at a rate of 4 Hz for 4 hours every 2 days, representing ~10% AF burden, and continuous pacing at 4 Hz for 24 hours per day, representing 100% AF burden. After 18 days of high-rate pacing, the aEHTs underwent a 7-day recovery period without pacing.

BMP10 release was measured using enzyme-linked immunosorbent assay (ELISA), while contractile function was evaluated through video analysis. The researchers also performed RNA sequencing to study transcriptional remodeling in response to the AF burden and used quantitative PCR (qPCR) to analyze the effects of recovery.

The results revealed that high-rate optogenetic pacing, which mimics AF, led to a dynamic and burden-dependent BMP10 release. BMP10 concentrations in the aEHT medium were significantly increased under intermittent optogenetic pacing, mirroring a 10% AF burden, and reached their highest levels under continuous optogenetic pacing, representing a 100% AF burden. As the aEHTs recovered from the high-rate pacing, BMP10 release gradually declined back to control levels.

However, the researchers observed that contractile dysfunction was most severe after continuous pacing and only partially recovered within the 7-day recovery period. RNA sequencing indicated that pacing-induced changes in BMP10 release were associated with distinct burden-dependent transcriptional states. These transcripts were related to BMP/TGFβ signaling, atrial identity, calcium handling, contractile phenotype, and electrophysiological remodeling.

After recovery, BMP10 mRNA expression remained elevated, suggesting that the tissue had undergone molecular changes in response to the prior AF burden.

In conclusion, this study demonstrates that AF burden dynamically regulates BMP10 release in human aEHTs. BMP10 release serves as a secreted protein-based readout of current or recent atrial high-rate stress. Furthermore, persistent transcriptional changes indicate molecular memory of prior AF burden. These findings suggest that BMP10 release could serve as a sensitive biomarker for AF burden, providing valuable insights into the condition and potential therapeutic targets.