DELE1 maintains muscle proteostasis to promote growth and survival in mitochondrial myopathy.

Mitochondrial dysfunction causes devastating disorders, including mitochondrial myopathy, but how muscle senses and adapts to mitochondrial dysfunction is not well understood. Here, we used diverse mouse models of mitochondrial myopathy to show that the signal for mitochondrial dysfunction originates within mitochondria. The mitochondrial proteins OMA1 and DELE1 sensed disruption of the inner mitochondrial membrane and, in response, activated the mitochondrial integrated stress response (mt-ISR) to increase the building blocks for protein synthesis. In the absence of the mt-ISR, protein synthesis in muscle was dysregulated causing protein misfolding, and mice with early-onset mitochondrial myopathy failed to grow and survive. The mt-ISR was similar following disruptions in mtDNA maintenance (Tfam knockout) and mitochondrial protein misfolding (CHCHD10 G58R and S59L knockin) but heterogenous among mitochondria-rich tissues, with broad gene expression changes observed in heart and skeletal muscle and limited changes observed in liver and brown adipose tissue. Taken together, our findings identify that the DELE1 mt-ISR mediates a similar response to diverse forms of mitochondrial stress and is critical for maintaining growth and survival in early-onset mitochondrial myopathy.

Invariant point message passing for protein side chain packing.

Protein side chain packing (PSCP) is a fundamental problem in the field of protein engineering, as high-confidence and low-energy conformations of amino acid side chains are crucial for understanding (and designing) protein folding, protein-protein interactions, and protein-ligand interactions. Traditional PSCP methods (such as the Rosetta Packer) often rely on a library of discrete side chain conformations, or rotamers, and a forcefield to guide the structure to low-energy conformations. Recently, deep learning (DL) based methods (such as DLPacker, AttnPacker, and DiffPack) have demonstrated state-of-the-art predictions and speed in the PSCP task. Building off the success of geometric graph neural networks for protein modeling, we present the Protein Invariant Point Packer (PIPPack) which effectively processes local structural and sequence information to produce realistic, idealized side chain coordinates using χ $$ \chi $$ -angle distribution predictions and geometry-aware invariant point message passing (IPMP). On a test set of ∼1400 high-quality protein chains, PIPPack is highly competitive with other state-of-the-art PSCP methods in rotamer recovery and per-residue RMSD but is significantly faster.

DELE1 promotes translation-associated homeostasis, growth, and survival in mitochondrial myopathy.

Mitochondrial dysfunction causes devastating disorders, including mitochondrial myopathy. Here, we identified that diverse mitochondrial myopathy models elicit a protective mitochondrial integrated stress response (mt-ISR), mediated by OMA1-DELE1 signaling. The response was similar following disruptions in mtDNA maintenance, from knockout of Tfam, and mitochondrial protein unfolding, from disease-causing mutations in CHCHD10 (G58R and S59L). The preponderance of the response was directed at upregulating pathways for aminoacyl-tRNA biosynthesis, the intermediates for protein synthesis, and was similar in heart and skeletal muscle but more limited in brown adipose challenged with cold stress. Strikingly, models with early DELE1 mt-ISR activation failed to grow and survive to adulthood in the absence of Dele1, accounting for some but not all of OMA1's protection. Notably, the DELE1 mt-ISR did not slow net protein synthesis in stressed striated muscle, but instead prevented loss of translation-associated proteostasis in muscle fibers. Together our findings identify that the DELE1 mt-ISR mediates a stereotyped response to diverse forms of mitochondrial stress and is particularly critical for maintaining growth and survival in early-onset mitochondrial myopathy.

Transfer learning to leverage larger datasets for improved prediction of protein stability changes.

Amino acid mutations that lower a protein's thermodynamic stability are implicated in numerous diseases, and engineered proteins with enhanced stability can be important in research and medicine. Computational methods for predicting how mutations perturb protein stability are, therefore, of great interest. Despite recent advancements in protein design using deep learning, in silico prediction of stability changes has remained challenging, in part due to a lack of large, high-quality training datasets for model development. Here, we describe ThermoMPNN, a deep neural network trained to predict stability changes for protein point mutations given an initial structure. In doing so, we demonstrate the utility of a recently released megascale stability dataset for training a robust stability model. We also employ transfer learning to leverage a second, larger dataset by using learned features extracted from ProteinMPNN, a deep neural network trained to predict a protein's amino acid sequence given its three-dimensional structure. We show that our method achieves state-of-the-art performance on established benchmark datasets using a lightweight model architecture that allows for rapid, scalable predictions. Finally, we make ThermoMPNN readily available as a tool for stability prediction and design.

In silico evolution of autoinhibitory domains for a PD-L1 antagonist using deep learning models.

There has been considerable progress in the development of computational methods for designing protein-protein interactions, but engineering high-affinity binders without extensive screening and maturation remains challenging. Here, we test a protein design pipeline that uses iterative rounds of deep learning (DL)-based structure prediction (AlphaFold2) and sequence optimization (ProteinMPNN) to design autoinhibitory domains (AiDs) for a PD-L1 antagonist. With the goal of creating an anticancer agent that is inactive until reaching the tumor environment, we sought to create autoinhibited (or masked) forms of the PD-L1 antagonist that can be unmasked by tumor-enriched proteases. Twenty-three de novo designed AiDs, varying in length and topology, were fused to the antagonist with a protease-sensitive linker, and binding to PD-L1 was measured with and without protease treatment. Nine of the fusion proteins demonstrated conditional binding to PD-L1, and the top-performing AiDs were selected for further characterization as single-domain proteins. Without any experimental affinity maturation, four of the AiDs bind to the PD-L1 antagonist with equilibrium dissociation constants (KDs) below 150 nM, with the lowest KD equal to 0.9 nM. Our study demonstrates that DL-based protein modeling can be used to rapidly generate high-affinity protein binders.

Transfer learning to leverage larger datasets for improved prediction of protein stability changes.

Amino acid mutations that lower a protein's thermodynamic stability are implicated in numerous diseases, and engineered proteins with enhanced stability are important in research and medicine. Computational methods for predicting how mutations perturb protein stability are therefore of great interest. Despite recent advancements in protein design using deep learning, in silico prediction of stability changes has remained challenging, in part due to a lack of large, high-quality training datasets for model development. Here we introduce ThermoMPNN, a deep neural network trained to predict stability changes for protein point mutations given an initial structure. In doing so, we demonstrate the utility of a newly released mega-scale stability dataset for training a robust stability model. We also employ transfer learning to leverage a second, larger dataset by using learned features extracted from a deep neural network trained to predict a protein's amino acid sequence given its three-dimensional structure. We show that our method achieves competitive performance on established benchmark datasets using a lightweight model architecture that allows for rapid, scalable predictions. Finally, we make ThermoMPNN readily available as a tool for stability prediction and design.

In silico evolution of protein binders with deep learning models for structure prediction and sequence design.

There has been considerable progress in the development of computational methods for designing protein-protein interactions, but engineering high-affinity binders without extensive screening and maturation remains challenging. Here, we test a protein design pipeline that uses iterative rounds of deep learning (DL)-based structure prediction (AlphaFold2) and sequence optimization (ProteinMPNN) to design autoinhibitory domains (AiDs) for a PD-L1 antagonist. Inspired by recent advances in therapeutic design, we sought to create autoinhibited (or masked) forms of the antagonist that can be conditionally activated by proteases. Twenty-three de novo designed AiDs, varying in length and topology, were fused to the antagonist with a protease sensitive linker, and binding to PD-L1 was tested with and without protease treatment. Nine of the fusion proteins demonstrated conditional binding to PD-L1 and the top performing AiDs were selected for further characterization as single domain proteins. Without any experimental affinity maturation, four of the AiDs bind to the PD-L1 antagonist with equilibrium dissociation constants (KDs) below 150 nM, with the lowest KD equal to 0.9 nM. Our study demonstrates that DL-based protein modeling can be used to rapidly generate high affinity protein binders.

Invariant point message passing for protein side chain packing.

Protein side chain packing (PSCP) is a fundamental problem in the field of protein engineering, as high-confidence and low-energy conformations of amino acid side chains are crucial for understanding (and designing) protein folding, protein-protein interactions, and protein-ligand interactions. Traditional PSCP methods (such as the Rosetta Packer) often rely on a library of discrete side chain conformations, or rotamers, and a forcefield to guide the structure to low-energy conformations. Recently, deep learning (DL) based methods (such as DLPacker, AttnPacker, and DiffPack) have demonstrated state-of-the-art predictions and speed in the PSCP task. Building off the success of geometric graph neural networks for protein modeling, we present the Protein Invariant Point Packer (PIPPack) which effectively processes local structural and sequence information to produce realistic, idealized side chain coordinates using χ-angle distribution predictions and geometry-aware invariant point message passing (IPMP). On a test set of ~1,400 high-quality protein chains, PIPPack is highly competitive with other state-of-the-art PSCP methods in rotamer recovery and per-residue RMSD but is significantly faster.

OMA1 mediates local and global stress responses against protein misfolding in CHCHD10 mitochondrial myopathy.

Mitochondrial stress triggers a response in the cell's mitochondria and nucleus, but how these stress responses are coordinated in vivo is poorly understood. Here, we characterize a family with myopathy caused by a dominant p.G58R mutation in the mitochondrial protein CHCHD10. To understand the disease etiology, we developed a knockin (KI) mouse model and found that mutant CHCHD10 aggregated in affected tissues, applying a toxic protein stress to the inner mitochondrial membrane. Unexpectedly, the survival of CHCHD10-KI mice depended on a protective stress response mediated by the mitochondrial metalloendopeptidase OMA1. The OMA1 stress response acted both locally within mitochondria, causing mitochondrial fragmentation, and signaled outside the mitochondria, activating the integrated stress response through cleavage of DAP3-binding cell death enhancer 1 (DELE1). We additionally identified an isoform switch in the terminal complex of the electron transport chain as a component of this response. Our results demonstrate that OMA1 was critical for neonatal survival conditionally in the setting of inner mitochondrial membrane stress, coordinating local and global stress responses to reshape the mitochondrial network and proteome.

Global sensitivity analysis informed model reduction and selection applied to a Valsalva maneuver model.

In this study, we develop a methodology for model reduction and selection informed by global sensitivity analysis (GSA) methods. We apply these techniques to a control model that takes systolic blood pressure and thoracic tissue pressure data as inputs and predicts heart rate in response to the Valsalva maneuver (VM). The study compares four GSA methods based on Sobol' indices (SIs) quantifying the parameter influence on the difference between the model output and the heart rate data. The GSA methods include standard scalar SIs determining the average parameter influence over the time interval studied and three time-varying methods analyzing how parameter influence changes over time. The time-varying methods include a new technique, termed limited-memory SIs, predicting parameter influence using a moving window approach. Using the limited-memory SIs, we perform model reduction and selection to analyze the necessity of modeling both the aortic and carotid baroreceptor regions in response to the VM. We compare the original model to systematically reduced models including (i) the aortic and carotid regions, (ii) the aortic region only, and (iii) the carotid region only. Model selection is done quantitatively using the Akaike and Bayesian Information Criteria and qualitatively by comparing the neurological predictions. Results show that it is necessary to incorporate both the aortic and carotid regions to model the VM.

Persistent instability in a nonhomogeneous delay differential equation system of the Valsalva maneuver.

Delay differential equations are widely used in mathematical modeling to describe physical and biological systems, often inducing oscillatory behavior. In physiological systems, this instability may signify (i) an attempt to return to homeostasis or (ii) system dysfunction. In this study, we analyze a nonlinear, nonautonomous, nonhomogeneous open-loop neurological control model describing the autonomic nervous system response to the Valsalva maneuver (VM). We reduce this model from 5 to 2 states (predicting sympathetic tone and heart rate) and categorize the stability properties of the reduced model using a two-parameter bifurcation analysis of the sympathetic delay (Ds) and time-scale (τs). Stability regions in the Dsτs-plane for this nonhomogeneous system and its homogeneous analog are classified numerically and analytically, identifying transcritical and Hopf bifurcations. Results show that the Hopf bifurcation remains for both the homogeneous and nonhomogeneous systems, while the nonhomogeneous system stabilizes the transition at the transcritical bifurcation. This analysis was compared with results from blood pressure and heart rate data from three subjects performing the VM: a control subject exhibiting sink behavior, a control subject exhibiting stable focus behavior, and a patient with postural orthostatic tachycardia syndrome (POTS) also exhibiting stable focus behavior. Results suggest that instability caused from overactive sympathetic signaling may result in autonomic dysfunction.