Tissue engineering approaches have the potential to increase the physiologic relevance of human iPS-derived cells, such as cardiomyocytes (iPS-CM). However, forming Engineered Heart Muscle (EHM) typically requires >1 million cells per tissue. Existing miniaturization strategies involve complex approaches not amenable to mass production, limiting the ability to use EHM for iPS-based disease modeling and drug screening. Micro-scale cardiospheres are easily produced, but do not facilitate assembly of elongated muscle or direct force measurements. Here we describe an approach that combines features of EHM and cardiospheres: Micro-Heart Muscle (μHM) arrays, in which elongated muscle fibers are formed in an easily fabricated template, with as few as 2,000 iPS-CM per individual tissue. Within μHM, iPS-CM exhibit uniaxial contractility and alignment, robust sarcomere assembly, and reduced variability and hypersensitivity in drug responsiveness, compared to monolayers with the same cellular composition. μHM mounted onto standard force measurement apparatus exhibited a robust Frank-Starling response to external stretch, and a dose-dependent inotropic response to the β-adrenergic agonist isoproterenol. Based on the ease of fabrication, the potential for mass production and the small number of cells required to form μHM, this system provides a potentially powerful tool to study cardiomyocyte maturation, disease and cardiotoxicology in vitro.

No abstract

Human induced pluripotent stem cell derived cardiomyocytes (hiPSC-CM) are a promising in vitro tool for drug development and disease modeling, but their immature electrophysiology limits their diagnostic utility. Tissue engineering approaches involving aligned and 3D culture enhance hiPSC-CM maturation but are insufficient to induce electrophysiological maturation. We hypothesized that recapitulating postnatal switching of the heart's primary adenosine triphosphate source from glycolysis to fatty acid oxidation could enhance maturation of hiPSC-CM. We combined hiPSC-CM with microfabrication to create 3D cardiac microphysiological systems (MPS) that enhanced immediate microtissue alignment and tissue specific extracellular matrix production. Using Robust Experimental design, we identified a maturation media that allowed the cardiac MPS to correctly assess false positive and negative drug response. Finally, we employed mathematical modeling and gene expression data to explain the observed changes in electrophysiology and pharmacology of MPS exposed to maturation media. In contrast, the same media had no effects on 2D hiPSC-CM monolayers. These results suggest that systematic combination of biophysical stimuli and metabolic cues can enhance the electrophysiological maturation of hiPSCderived cardiomyocytes. Results and Discussion Robust Design Experiments Indicate Optimal Carbon Sourcing for Mature Beating PhysiologyWe have developed a microfabricated cardiac MPS, employing hiPSC-CM, for drug testing (Mathur et al. 2015). In the present study, we formed cardiac MPS that that mimicked the mass composition of the human heart by combining 80% hiPSC-CM and 20% hiPSC-SC (Supplemental Methods, Fig. S1-2). We employed Robust Experimental Design to screen for the effects of glucose, oleic acid, palmitic acid, and albumin (bovine serum albumin, BSA) levels on hiPSC-CM maturity ( Table 1). MPS were incubated with different fatty-acid media for ten days, at which time their beating physiology and calcium flux were assessed. Optimal media would reduce automaticity (e.g. reduce spontaneous beating rate), while also reducing the interval between peak contraction and peak relaxation (a surrogate for APD) in field-paced tissues (Fig. 1A-C), while maintaining a high level of beating prevalence during pacing (defined as the percent of the tissue with substantial contractile motion; Fig. 1D). In general, beating prevalence was consistent with calcium flux amplitude (Fig. 1D,F), and beating interval correlated with rate-corrected Full-Width Half Maximum calcium flux time, FWHMc; Fig. 1B,E).

This work presents a study measuring and characterizing the brain ultrashort-T2* component in patients suffering from MS. Our results show that the measured T2* relaxation time of the ultrashort-T2* component is significantly lower in almost all white matter ROIs measured. Fitted signal curves for the fractional component suggest the fractional component fits well in certain patients but not as well in others suggesting a more robust fitting model may be necessary for using this UTE relaxometry technique in patients with MS.

Despite the proliferation of deep learning techniques for accelerated MRI acquisition and enhanced image reconstruction, the construction of large and diverse MRI datasets continues to pose a barrier to effective clinical translation of these technologies. One major challenge is in collecting the MRI raw data (required for image reconstruction) from clinical scanning, as only magnitude images are typically saved and used for clinical assessment and diagnosis. The image phase and multi-channel RF coil information are not retained when magnitude-only images are saved in clinical imaging archives. Additionally, preprocessing used for data in clinical imaging can lead to biased results. While several groups have begun concerted efforts to collect large amounts of MRI raw data, current databases are limited in the diversity of anatomy, pathology, annotations, and acquisition types they contain. To address this, we present a method for synthesizing realistic MR data from magnitude-only data, allowing for the use of diverse data from clinical imaging archives in advanced MRI reconstruction development. Our method uses a conditional GAN-based framework to generate synthetic phase images from input magnitude images. We then applied ESPIRiT to derive RF coil sensitivity maps from fully sampled real data to generate multi-coil data. The synthetic data generation method was evaluated by comparing image reconstruction results from training Variational Networks either with real data or synthetic data. We demonstrate that the Variational Network trained on synthetic MRI data from our method, consisting of GAN-derived synthetic phase and multi-coil information, outperformed Variational Networks trained on data with synthetic phase generated using current state-of-the-art methods. Additionally, we demonstrate that the Variational Networks trained with synthetic k-space data from our method perform comparably to image reconstruction networks trained on undersampled real k-space data.