Orbital Mixer: Using Atomic Orbital Features for Basis-Dependent Prediction of Molecular Wavefunctions.

Leveraging ab initio data at scale has enabled the development of machine learning models capable of extremely accurate and fast molecular property prediction. A central paradigm of many previous studies focuses on generating predictions for only a fixed set of properties. Recent lines of research instead aim to explicitly learn the electronic structure via molecular wavefunctions, from which other quantum chemical properties can be directly derived. While previous methods generate predictions as a function of only the atomic configuration, in this work we present an alternate approach that directly purposes basis-dependent information to predict molecular electronic structure. Our model, Orbital Mixer, is composed entirely of multi-layer perceptrons (MLPs) using MLP-Mixer layers within a simple, intuitive, and scalable architecture that achieves competitive Hamiltonian and molecular orbital energy and coefficient prediction accuracies compared to the state-of-the-art.

Dynamic loading of human engineered heart tissue enhances contractile function and drives a desmosome-linked disease phenotype.

The role that mechanical forces play in shaping the structure and function of the heart is critical to understanding heart formation and the etiology of disease but is challenging to study in patients. Engineered heart tissues (EHTs) incorporating human induced pluripotent stem cell (hiPSC)-derived cardiomyocytes have the potential to provide insight into these adaptive and maladaptive changes. However, most EHT systems cannot model both preload (stretch during chamber filling) and afterload (pressure the heart must work against to eject blood). Here, we have developed a new dynamic EHT (dyn-EHT) model that enables us to tune preload and have unconstrained contractile shortening of >10%. To do this, three-dimensional (3D) EHTs were integrated with an elastic polydimethylsiloxane strip providing mechanical preload and afterload in addition to enabling contractile force measurements based on strip bending. Our results demonstrated that dynamic loading improves the function of wild-type EHTs on the basis of the magnitude of the applied force, leading to improved alignment, conduction velocity, and contractility. For disease modeling, we used hiPSC-derived cardiomyocytes from a patient with arrhythmogenic cardiomyopathy due to mutations in the desmoplakin gene. We demonstrated that manifestation of this desmosome-linked disease state required dyn-EHT conditioning and that it could not be induced using 2D or standard 3D EHT approaches. Thus, a dynamic loading strategy is necessary to provoke the disease phenotype of diastolic lengthening, reduction of desmosome counts, and reduced contractility, which are related to primary end points of clinical disease, such as chamber thinning and reduced cardiac output.

Engineering aligned human cardiac muscle using developmentally inspired fibronectin micropatterns.

Cardiac two-dimensional tissues were engineered using biomimetic micropatterns based on the fibronectin-rich extracellular matrix (ECM) of the embryonic heart. The goal of this developmentally-inspired, in vitro approach was to identify cell-cell and cell-ECM interactions in the microenvironment of the early 4-chambered vertebrate heart that drive cardiomyocyte organization and alignment. To test this, biomimetic micropatterns based on confocal imaging of fibronectin in embryonic chick myocardium were created and compared to control micropatterns designed with 2 or 20 µm wide fibronectin lines. Results show that embryonic chick cardiomyocytes have a unique density-dependent alignment on the biomimetic micropattern that is mediated in part by N-cadherin, suggesting that both cell-cell and cell-ECM interactions play an important role in the formation of aligned myocardium. Human induced pluripotent stem cell-derived cardiomyocytes also showed density-dependent alignment on the biomimetic micropattern but were overall less well organized. Interestingly, the addition of human adult cardiac fibroblasts and conditioning with T3 hormone were both shown to increase human cardiomyocyte alignment. In total, these results show that cardiomyocyte maturation state, cardiomyocyte-cardiomyocyte and cardiomyocyte-fibroblast interactions, and cardiomyocyte-ECM interactions can all play a role when engineering anisotropic cardiac tissues in vitro and provides insight as to how these factors may influence cardiogenesis in vivo.

Photopatterned biomolecule immobilization to guide three-dimensional cell fate in natural protein-based hydrogels.

Hydrogel biomaterials derived from natural biopolymers (e.g., fibrin, collagen, decellularized extracellular matrix) are regularly utilized in three-dimensional (3D) cell culture and tissue engineering. In contrast to those based on synthetic polymers, natural materials permit enhanced cytocompatibility, matrix remodeling, and biological integration. Despite these advantages, natural protein-based gels have lagged behind synthetic alternatives in their tunability; methods to selectively modulate the biochemical properties of these networks in a user-defined and heterogeneous fashion that can drive encapsulated cell function have not yet been established. Here, we report a generalizable strategy utilizing a photomediated oxime ligation to covalently decorate naturally derived hydrogels with bioactive proteins including growth factors. This bioorthogonal photofunctionalization is readily amenable to mask-based and laser-scanning lithographic patterning, enabling full four-dimensional (4D) control over protein immobilization within virtually any natural protein-based biomaterial. Such versatility affords exciting opportunities to probe and direct advanced cell fates inaccessible using purely synthetic approaches in response to anisotropic environmental signaling.

Thermofluidic heat exchangers for actuation of transcription in artificial tissues.

Spatial patterns of gene expression in living organisms orchestrate cell decisions in development, homeostasis, and disease. However, most methods for reconstructing gene patterning in 3D cell culture and artificial tissues are restricted by patterning depth and scale. We introduce a depth- and scale-flexible method to direct volumetric gene expression patterning in 3D artificial tissues, which we call "heat exchangers for actuation of transcription" (HEAT). This approach leverages fluid-based heat transfer from printed networks in the tissues to activate heat-inducible transgenes expressed by embedded cells. We show that gene expression patterning can be tuned both spatially and dynamically by varying channel network architecture, fluid temperature, fluid flow direction, and stimulation timing in a user-defined manner and maintained in vivo. We apply this approach to activate the 3D positional expression of Wnt ligands and Wnt/ß-catenin pathway regulators, which are major regulators of development, homeostasis, regeneration, and cancer throughout the animal kingdom.

Engineering Heart Morphogenesis.

Recent advances in stem cell biology and tissue engineering have laid the groundwork for building complex tissues in a dish. We propose that these technologies are ready for a new challenge: recapitulating cardiac morphogenesis in vitro. In development, the heart transforms from a simple linear tube to a four-chambered organ through a complex process called looping. Here, we re-examine heart tube looping through the lens of an engineer and argue that the linear heart tube is an advantageous starting point for tissue engineering. We summarize the structures, signaling pathways, and stresses in the looping heart, and evaluate approaches that could be used to build a linear heart tube and guide it through the process of looping.

Differentiation of Cardiomyocytes from Human Pluripotent Stem Cells Using Monolayer Culture.

Human pluripotent stem cells (PSCs) are a promising cell source for cardiac tissue engineering and cell-based therapies for heart repair because they can be expanded in vitro and differentiated into most cardiovascular cell types, including cardiomyocytes. During embryonic heart development, this differentiation occurs under the influence of internal and external stimuli that guide cells to go down the cardiac lineage. In order to differentiate PSCs in vitro, these or similar stimuli need to be provided in a controlled manner. However, because it is not possible to completely recapitulate the embryonic environment, the factors essential for cardiac differentiation of PSCs in vitro need to be experimentally determined and validated. Since PSCs were first developed, significant progress has been made in optimizing techniques for their differentiation toward cardiomyocytes. In this review, we will summarize recent advances in these techniques, with particular focus on monolayer-based methods that have improved the efficiency and scalability of cardiomyocyte differentiation.