BeatProfiler: Multimodal In Vitro Analysis of Cardiac Function Enables Machine Learning Classification of Diseases and Drugs.

Goal: Contractile response and calcium handling are central to understanding cardiac function and physiology, yet existing methods of analysis to quantify these metrics are often time-consuming, prone to mistakes, or require specialized equipment/license. We developed BeatProfiler, a suite of cardiac analysis tools designed to quantify contractile function, calcium handling, and force generation for multiple in vitro cardiac models and apply downstream machine learning methods for deep phenotyping and classification. Methods: We first validate BeatProfiler's accuracy, robustness, and speed by benchmarking against existing tools with a fixed dataset. We further confirm its ability to robustly characterize disease and dose-dependent drug response. We then demonstrate that the data acquired by our automatic acquisition pipeline can be further harnessed for machine learning (ML) analysis to phenotype a disease model of restrictive cardiomyopathy and profile cardioactive drug functional response. To accurately classify between these biological signals, we apply feature-based ML and deep learning models (temporal convolutional-bidirectional long short-term memory model or TCN-BiLSTM). Results: Benchmarking against existing tools revealed that BeatProfiler detected and analyzed contraction and calcium signals better than existing tools through improved sensitivity in low signal data, reduction in false positives, and analysis speed increase by 7 to 50-fold. Of signals accurately detected by published methods (PMs), BeatProfiler's extracted features showed high correlations to PMs, confirming that it is reliable and consistent with PMs. The features extracted by BeatProfiler classified restrictive cardiomyopathy cardiomyocytes from isogenic healthy controls with 98% accuracy and identified relax90 as a top distinguishing feature in congruence with previous findings. We also show that our TCN-BiLSTM model was able to classify drug-free control and 4 cardiac drugs with different mechanisms of action at 96% accuracy. We further apply Grad-CAM on our convolution-based models to identify signature regions of perturbations by these drugs in calcium signals. Conclusions: We anticipate that the capabilities of BeatProfiler will help advance in vitro studies in cardiac biology through rapid phenotyping, revealing mechanisms underlying cardiac health and disease, and enabling objective classification of cardiac disease and responses to drugs.

Engineered cardiac tissue model of restrictive cardiomyopathy for drug discovery.

Restrictive cardiomyopathy (RCM) is defined as increased myocardial stiffness and impaired diastolic relaxation leading to elevated ventricular filling pressures. Human variants in filamin C (FLNC) are linked to a variety of cardiomyopathies, and in this study, we investigate an in-frame deletion (c.7416_7418delGAA, p.Glu2472_Asn2473delinAsp) in a patient with RCM. Induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) with this variant display impaired relaxation and reduced calcium kinetics in 2D culture when compared with a CRISPR-Cas9-corrected isogenic control line. Similarly, mutant engineered cardiac tissues (ECTs) demonstrate increased passive tension and impaired relaxation velocity compared with isogenic controls. High-throughput small-molecule screening identifies phosphodiesterase 3 (PDE3) inhibition by trequinsin as a potential therapy to improve cardiomyocyte relaxation in this genotype. Together, these data demonstrate an engineered cardiac tissue model of RCM and establish the translational potential of this precision medicine approach to identify therapeutics targeting myocardial relaxation.

RNA and Protein Delivery by Cell-Secreted and Bioengineered Extracellular Vesicles.

Extracellular vesicles (EVs) are carriers of biological signals through export and delivery of RNAs and proteins. Of increasing interest is the use of EVs as a platform for delivery of biomolecules. Preclinical studies have effectively used EVs to treat a number of diseases. Uniquely, endogenous machinery within cells can be manipulated in order to produce desirable loading of cargo within secreted EVs. In order to inform the development of such approaches, an understanding of the cellular mechanisms by which cargo is sorted to EVs is required. Here, the current knowledge of cargo sorting within EVs is reviewed. Here is given an overview of recent bioengineering approaches that leverage these advances. Methods of externally manipulating EV cargo are also discussed. Finally, a perspective on the current challenges of EVs as a drug delivery platform is offered. It is proposed that standardized bioengineering methods for therapeutic EV preparation will be required to create a well-defined clinical product.

STK25 inhibits PKA signaling by phosphorylating PRKAR1A.

In the heart, protein kinase A (PKA) is critical for activating calcium handling and sarcomeric proteins in response to beta-adrenergic stimulation leading to increased myocardial contractility and performance. The catalytic activity of PKA is tightly regulated by regulatory subunits that inhibit the catalytic subunit until released by cAMP binding. Phosphorylation of type II regulatory subunits promotes PKA activation; however, the role of phosphorylation in type I regulatory subunits remain uncertain. Here, we utilize human induced pluripotent stem cell cardiomyocytes (iPSC-CMs) to identify STK25 as a kinase of the type Iα regulatory subunit PRKAR1A. Phosphorylation of PRKAR1A leads to inhibition of PKA kinase activity and increased binding to the catalytic subunit in the presence of cAMP. Stk25 knockout in mice diminishes Prkar1a phosphorylation, increases Pka activity, and augments contractile response to beta-adrenergic stimulation. Together, these data support STK25 as a negative regulator of PKA signaling through phosphorylation of PRKAR1A.

A microfluidic platform for the high-throughput study of pathological cardiac hypertrophy.

Current in vitro models fall short in deciphering the mechanisms of cardiac hypertrophy induced by volume overload. We developed a pneumatic microfluidic platform for high-throughput studies of cardiac hypertrophy that enables repetitive (hundreds of thousands of times) and robust (over several weeks) manipulation of cardiac µtissues. The platform is reusable for stable and reproducible mechanical stimulation of cardiac µtissues (each containing only 5000 cells). Heterotypic and homotypic µtissues produced in the device were pneumatically loaded in a range of regimes, with real-time on-chip analysis of tissue phenotypes. Concentrated loading of the three-dimensional cardiac tissue faithfully recapitulated the pathology of volume overload seen in native heart tissue. Sustained volume overload of µtissues was sufficient to induce pathological cardiac remodeling associated with upregulation of the fetal gene program, in a dose-dependent manner.