Biology & Life Sciences

mRNA-Loaded Lipid Nanoparticles for Cardiac Repair: Screening Platforms and the Delivery Challenge

A heart-on-chip microphysiological system now enables rapid screening of lipid nanoparticle-mRNA formulations for cardiac delivery, predicting in vivo transfection efficacy. This platform addresses a critical bottleneck in developing mRNA therapeutics for heart disease.

By ORAA Research
This blog summarizes research trends based on published paper abstracts. Specific numbers or findings may contain inaccuracies. For scholarly rigor, always consult the original papers cited in each post.

The success of mRNA-lipid nanoparticle (LNP) vaccines against COVID-19 demonstrated that mRNA therapeutics can be manufactured and deployed at scale. Extending this platform to cardiovascular diseaseโ€”the leading cause of death globallyโ€”faces a fundamental biological obstacle: LNPs injected systemically accumulate overwhelmingly in the liver, not the heart. A recent study by Neiman et al. (2025) in Nature Biomedical Engineering introduces a microphysiological screening platform that could accelerate the search for cardiac-tropic LNP formulations.

The Cardiac Delivery Problem

After myocardial infarction, the adult human heart loses cardiomyocytesโ€”the contractile cells of the heart muscleโ€”and replaces them with fibrotic scar tissue. Unlike liver or skin, cardiac muscle has negligible regenerative capacity. mRNA therapeutics could theoretically address this by delivering transcription factor cocktails that reprogram cardiac fibroblasts into functional cardiomyocytes, or by encoding growth factors that stimulate endogenous repair.

The challenge is getting the mRNA there. Standard LNPs are rapidly cleared to the liver by apolipoprotein E (ApoE)-mediated uptake. The heart receives perhaps 1-2% of systemically administered LNPs, a fraction far too low for therapeutic effect. Direct intramyocardial injection is possible but invasive, technically demanding, and difficult to distribute uniformly across damaged tissue.

Soroudi et al. (2024) provide a comprehensive review of the LNP-mRNA landscape for cardiovascular diseases in the Journal of Controlled Release. They catalog strategies for redirecting LNP tropism away from the liver, including SORT (selective organ targeting through lipid engineering), ASSET (anchored secondary scFv enabling targeting), and surface functionalization with cardiac-homing peptides. Each approach modifies the LNP surface to change its biodistribution, but systematically evaluating these modifications has been slow because the primary readoutโ€”cardiac delivery efficiencyโ€”requires animal experiments.

The Microphysiological Screening Platform

Neiman et al. (2025) address this bottleneck with a heart-on-chip platform designed specifically for screening LNP-mRNA formulations. The system uses human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) organized into functional cardiac microtissues within microfluidic devices that recapitulate key aspects of the cardiac microenvironment: contractile function, fluid flow, and tissue architecture.

The critical contribution is the demonstration that transfection efficiency in this microphysiological system predicts in vivo cardiac transfection in animal models. This is not a trivial finding. Many in vitro systemsโ€”standard 2D cell cultures, for exampleโ€”poorly predict in vivo LNP behavior because they lack the biological barriers (endothelial lining, extracellular matrix, flow dynamics) that determine real-world delivery.

The screening workflow enables parallel evaluation of multiple LNP formulationsโ€”varying lipid composition, ionizable lipid identity, PEG density, and targeting ligandsโ€”with readouts of both transfection efficiency and cardiomyocyte viability. Formulations that transfect the heart-on-chip model efficiently can then be prioritized for animal testing, substantially reducing the number of in vivo experiments required.

Why This Matters for mRNA Cardiac Therapeutics

Several therapeutic applications are converging on cardiac mRNA delivery:

Direct cardiac reprogramming. Delivering mRNA encoding transcription factors (Gata4, Mef2c, Tbx5, and others) could convert scar-forming fibroblasts into cardiomyocyte-like cells in situ, potentially restoring contractile function after infarction. Animal studies have demonstrated proof of concept, but translation requires efficient and specific delivery to the infarct border zone.

Transient CAR-T generation. Epstein and colleagues demonstrated that LNP-delivered mRNA can transiently reprogram T cells to express chimeric antigen receptors targeting cardiac fibrosis markers. Unlike DNA-based CAR-T therapy, mRNA-based expression is inherently transient, reducing the risk of sustained autoimmune attack on cardiac tissue.

Genome editing. CRISPR-Cas9 components delivered via mRNA-LNPs could correct genetic cardiomyopathies. The transient nature of mRNA expression is actually advantageous here, as persistent Cas9 expression increases off-target editing risk.

For all these applications, the rate-limiting step is the same: cardiac-specific delivery with sufficient efficiency.

Critical Assessment

Predictive validity has limits. While the correlation between in vitro and in vivo performance is encouraging, the heart-on-chip system does not capture all barriers to cardiac delivery: the pulmonary first-pass effect, immune cell clearance, protein corona formation in blood, and coronary vascular architecture. Formulations optimized on the chip may still fail in vivo for reasons the model cannot detect.

Scale of screening remains modest. The microfluidic format enables higher throughput than animal experiments but lower throughput than high-throughput plate-based screens. Whether the physiological fidelity justifies the throughput tradeoff depends on the specific optimization campaign.

hiPSC-CM maturity is a concern. Stem cell-derived cardiomyocytes are notoriously immature compared to adult human cardiomyocytes, differing in gene expression, sarcomere organization, and electrophysiology. LNP uptake mechanisms may differ between fetal-like hiPSC-CMs and adult cardiomyocytes in vivo.

Clinical translation path is long. Even with optimized formulations, regulatory requirements for cardiac mRNA therapeuticsโ€”demonstrating safety in a vital organ with no regenerative backupโ€”are stringent. The gap between identifying a promising LNP formulation and obtaining clinical approval is measured in years.

Open Questions

  • Can the microphysiological platform be extended to model the infarct border zone specifically, where fibroblasts and surviving cardiomyocytes coexist?
  • How well do LNP formulations optimized for hiPSC-CMs perform in adult primary cardiomyocytes from patient tissue?
  • What is the minimum cardiac transfection efficiency required for therapeutic benefit in direct reprogramming applications?
  • Can this screening approach be combined with computational LNP design tools to create a closed-loop optimization pipeline?

Closing Reflection

The heart-on-chip screening platform from Neiman et al. represents a practical engineering solution to a biological bottleneck. By providing a physiologically relevant filter between LNP design and animal testing, it could accelerate the iterative optimization cycle that cardiac mRNA therapeutics require. The ultimate questionโ€”whether any LNP formulation can deliver mRNA to the human heart with sufficient efficiency for clinical benefitโ€”remains open. But the tools for answering it are becoming substantially more capable.


References (2)

Neiman, G., et al. (2025). A microphysiological system for screening lipid nanoparticle-mRNA complexes predicts in vivo heart transfection efficacy. Nature Biomedical Engineering.
Soroudi, S., et al. (2024). Lipid nanoparticle (LNP) mediated mRNA delivery in cardiovascular diseases: Advances in genome editing and CAR T cell therapy. Journal of Controlled Release.

Explore this topic deeper

Search 290M+ papers, detect research gaps, and find what hasn't been studied yet.

Click to remove unwanted keywords

Search 7 keywords โ†’