TECPR1 is activated by damage-induced sphingomyelin exposure to mediate noncanonical autophagy.

Cells use noncanonical autophagy, also called conjugation of ATG8 to single membranes (CASM), to label damaged intracellular compartments with ubiquitin-like ATG8 family proteins in order to signal danger caused by pathogens or toxic compounds. CASM relies on E3 complexes to sense membrane damage, but so far, only the mechanism to activate ATG16L1-containing E3 complexes, associated with proton gradient loss, has been described. Here, we show that TECPR1-containing E3 complexes are key mediators of CASM in cells treated with a variety of pharmacological drugs, including clinically relevant nanoparticles, transfection reagents, antihistamines, lysosomotropic compounds, and detergents. Interestingly, TECPR1 retains E3 activity when ATG16L1 CASM activity is obstructed by the Salmonella Typhimurium pathogenicity factor SopF. Mechanistically, TECPR1 is recruited by damage-induced sphingomyelin (SM) exposure using two DysF domains, resulting in its activation and ATG8 lipidation. In vitro assays using purified human TECPR1-ATG5-ATG12 complex show direct activation of its E3 activity by SM, whereas SM has no effect on ATG16L1-ATG5-ATG12. We conclude that TECPR1 is a key activator of CASM downstream of SM exposure.

Multiomics analysis of naturally efficacious lipid nanoparticle coronas reveals high-density lipoprotein is necessary for their function.

In terms of lipid nanoparticle (LNP) engineering, the relationship between particle composition, delivery efficacy, and the composition of the biocoronas that form around LNPs, is poorly understood. To explore this we analyze naturally efficacious biocorona compositions using an unbiased screening workflow. First, LNPs are complexed with plasma samples, from individual lean or obese male rats, and then functionally evaluated in vitro. Then, a fast, automated, and miniaturized method retrieves the LNPs with intact biocoronas, and multiomics analysis of the LNP-corona complexes reveals the particle corona content arising from each individual plasma sample. We find that the most efficacious LNP-corona complexes were enriched with high-density lipoprotein (HDL) and, compared to the commonly used corona-biomarker Apolipoprotein E, corona HDL content was a superior predictor of in-vivo activity. Using technically challenging and clinically relevant lipid nanoparticles, these methods reveal a previously unreported role for HDL as a source of ApoE and, form a framework for improving LNP therapeutic efficacy by controlling corona composition.

Understanding Intracellular Biology to Improve mRNA Delivery by Lipid Nanoparticles.

Poor understanding of intracellular delivery and targeting hinders development of nucleic acid-based therapeutics transported by nanoparticles. Utilizing a siRNA-targeting and small molecule profiling approach with advanced imaging and machine learning biological insights is generated into the mechanism of lipid nanoparticle (MC3-LNP) delivery of mRNA. This workflow is termed Advanced Cellular and Endocytic profiling for Intracellular Delivery (ACE-ID). A cell-based imaging assay and perturbation of 178 targets relevant to intracellular trafficking is used to identify corresponding effects on functional mRNA delivery. Targets improving delivery are analyzed by extracting data-rich phenotypic fingerprints from images using advanced image analysis algorithms. Machine learning is used to determine key features correlating with enhanced delivery, identifying fluid-phase endocytosis as a productive cellular entry route. With this new knowledge, MC3-LNP is re-engineered to target macropinocytosis, and this significantly improves mRNA delivery in vitro and in vivo. The ACE-ID approach can be broadly applicable for optimizing nanomedicine-based intracellular delivery systems and has the potential to accelerate the development of delivery systems for nucleic acid-based therapeutics.

Precise and systematic end group chemistry modifications on PAMAM and poly(l-lysine) dendrimers to improve cytosolic delivery of mRNA.

Here, we aimed to chemically modify PAMAM dendrimers using lysine as a site-selective anchor for successfully delivering mRNA while maintaining a low toxicity profile. PAMAM dendrimers were multi-functionalised by amidation reactions in a regioselective, quantitative and stepwise manner with carefully selected property-modifying surface groups. Alternatively, novel lysine-based dendrimers were prepared in the same manner with the aim to unlock their potential in gene delivery. The modified dendrimers were then formulated with Cy5-EGFP mRNA by bulk mixing via liquid handling robotics across different nitrogen to phosphate ratios. The resulting dendriplexes were characterised by size, charge, mRNA encapsulation, and mRNA binding affinity. Finally, their in-vitro delivery activity was systematically investigated across key cellular trafficking stages to relate chemical design to cellular effect. We demonstrate our findings in different cell lines and benchmarked relative to a commercially available transfection agent, jetPEI®. We demonstrate that specific surface modifications are required to generate small, reliable and well-encapsulated positively charged dendriplex complexes. Furthermore, we show that introduction of fusogenic groups is essential for driving endosomal escape and achieving cellular delivery and translation of mRNA in these cell lines.

Interaction Kinetics of Individual mRNA-Containing Lipid Nanoparticles with an Endosomal Membrane Mimic: Dependence on pH, Protein Corona Formation, and Lipoprotein Depletion.

Lipid nanoparticles (LNPs) have emerged as potent carriers for mRNA delivery, but several challenges remain before this approach can offer broad clinical translation of mRNA therapeutics. To improve their efficacy, a better understanding is required regarding how LNPs are trapped and processed at the anionic endosomal membrane prior to mRNA release. We used surface-sensitive fluorescence microscopy with single LNP resolution to investigate the pH dependency of the binding kinetics of ionizable lipid-containing LNPs to a supported endosomal model membrane. A sharp increase of LNP binding was observed when the pH was lowered from 6 to 5, accompanied by stepwise large-scale LNP disintegration. For LNPs preincubated in serum, protein corona formation shifted the onset of LNP binding and subsequent disintegration to lower pH, an effect that was less pronounced for lipoprotein-depleted serum. The LNP binding to the endosomal membrane mimic was observed to eventually become severely limited by suppression of the driving force for the formation of multivalent bonds during LNP attachment or, more specifically, by charge neutralization of anionic lipids in the model membrane due to their association with cationic lipids from earlier attached LNPs upon their disintegration. Cell uptake experiments demonstrated marginal differences in LNP uptake in untreated and lipoprotein-depleted serum, whereas lipoprotein-depleted serum increased mRNA-controlled protein (eGFP) production substantially. This complies with model membrane data and suggests that protein corona formation on the surface of the LNPs influences the nature of the interaction between LNPs and endosomal membranes.

SimSearch: A Human-in-The-Loop Learning Framework for Fast Detection of Regions of Interest in Microscopy Images.

OBJECTIVE: Large-scale microscopy-based experiments often result in images with rich but sparse information content. An experienced microscopist can visually identify regions of interest (ROIs), but this becomes a cumbersome task with large datasets. Here we present SimSearch, a framework for quick and easy user-guided training of a deep neural model aimed at fast detection of ROIs in large-scale microscopy experiments. METHODS: The user manually selects a small number of patches representing different classes of ROIs. This is followed by feature extraction using a pre-trained deep-learning model, and interactive patch selection pruning, resulting in a smaller set of clean (user approved) and larger set of noisy (unapproved) training patches of ROIs and background. The pre-trained deep-learning model is thereafter first trained on the large set of noisy patches, followed by refined training using the clean patches. RESULTS: The framework is evaluated on fluorescence microscopy images from a large-scale drug screening experiment, brightfield images of immunohistochemistry-stained patient tissue samples, and malaria-infected human blood smears, as well as transmission electron microscopy images of cell sections. Compared to state-of-the-art and manual/visual assessment, the results show similar performance with maximal flexibility and minimal a priori information and user interaction. CONCLUSIONS: SimSearch quickly adapts to different data sets, which demonstrates the potential to speed up many microscopy-based experiments based on a small amount of user interaction. SIGNIFICANCE: SimSearch can help biologists quickly extract informative regions and perform analyses on large datasets helping increase the throughput in a microscopy experiment.

Physiology, pathology and the biomolecular corona: the confounding factors in nanomedicine design.

The biomolecular corona that forms on nanomedicines in different physiological and pathological environments confers a new biological identity. How the recipient biological system's state can potentially affect nanomedicine corona formation, and how this can be modulated, remains obscure. With this perspective, this review summarizes the current knowledge about the content of biological fluids in various compartments and how they can be affected by pathological states, thus impacting biomolecular corona formation. The content of representative biological fluids is explored, and the urgency of integrating corona formation, as an essential component of nanomedicine designs for effective cargo delivery, is highlighted.

Inhibition of SGLT2 Preserves Function and Promotes Proliferation of Human Islets Cells In Vivo in Diabetic Mice.

Dapagliflozin is a sodium-glucose co-transporter 2 (SGLT2) inhibitor used for the treatment of diabetes. This study examines the effects of dapagliflozin on human islets, focusing on alpha and beta cell composition in relation to function in vivo, following treatment of xeno-transplanted diabetic mice. Mouse beta cells were ablated by alloxan, and dapagliflozin was provided in the drinking water while controls received tap water. Body weight, food and water intake, plasma glucose, and human C-peptide levels were monitored, and intravenous arginine/glucose tolerance tests (IVarg GTT) were performed to evaluate islet function. The grafted human islets were isolated at termination and stained for insulin, glucagon, Ki67, caspase 3, and PDX-1 immunoreactivity in dual and triple combinations. In addition, human islets were treated in vitro with dapagliflozin at different glucose concentrations, followed by insulin and glucagon secretion measurements. SGLT2 inhibition increased the animal survival rate and reduced plasma glucose, accompanied by sustained human C-peptide levels and improved islet response to glucose/arginine. SGLT2 inhibition increased both alpha and beta cell proliferation (Ki67+glucagon+ and Ki67+insulin+) while apoptosis was reduced (caspase3+glucagon+ and caspase3+insulin+). Alpha cells were fewer following inhibition of SGLT2 with increased glucagon/PDX-1 double-positive cells, a marker of alpha to beta cell transdifferentiation. In vitro treatment of human islets with dapagliflozin had no apparent impact on islet function. In summary, SGLT2 inhibition supported human islet function in vivo in the hyperglycemic milieu and potentially promoted alpha to beta cell transdifferentiation, most likely through an indirect mechanism.

Endosomal escape of delivered mRNA from endosomal recycling tubules visualized at the nanoscale.

Delivery of exogenous mRNA using lipid nanoparticles (LNPs) is a promising strategy for therapeutics. However, a bottleneck remains in the poor understanding of the parameters that correlate with endosomal escape versus cytotoxicity. To address this problem, we compared the endosomal distribution of six LNP-mRNA formulations of diverse chemical composition and efficacy, similar to those used in mRNA-based vaccines, in primary human adipocytes, fibroblasts, and HeLa cells. Surprisingly, we found that total uptake is not a sufficient predictor of delivery, and different LNPs vary considerably in endosomal distributions. Prolonged uptake impaired endosomal acidification, a sign of cytotoxicity, and caused mRNA to accumulate in compartments defective in cargo transport and unproductive for delivery. In contrast, early endocytic/recycling compartments have the highest probability for mRNA escape. By using super-resolution microscopy, we could resolve a single LNP-mRNA within subendosomal compartments and capture events of mRNA escape from endosomal recycling tubules. Our results change the view of the mechanisms of endosomal escape and define quantitative parameters to guide the development of mRNA formulations toward higher efficacy and lower cytotoxicity.

Quantitative intracellular retention of delivered RNAs through optimized cell fixation and immunostaining.

Detection of nucleic acids within subcellular compartments is key to understanding their function. Determining the intracellular distribution of nucleic acids requires quantitative retention and estimation of their association with different organelles by immunofluorescence microscopy. This is particularly important for the delivery of nucleic acid therapeutics, which depends on endocytic uptake and endosomal escape. However, the current protocols fail to preserve the majority of exogenously delivered nucleic acids in the cytoplasm. To solve this problem, by monitoring Cy5-labeled mRNA delivered to primary human adipocytes via lipid nanoparticles (LNP), we optimized cell fixation, permeabilization, and immunostaining of a number of organelle markers, achieving quantitative retention of mRNA and allowing visualization of levels that escape detection using conventional procedures. The optimized protocol proved effective on exogenously delivered siRNA, miRNA, as well as endogenous miRNA. Our protocol is compatible with RNA probes of single molecule fluorescence in situ hybridization (smFISH) and molecular beacon, thus demonstrating that it is broadly applicable to study a variety of nucleic acids in cultured cells.

Deep-learning models for lipid nanoparticle-based drug delivery.

Background: Early prediction of time-lapse microscopy experiments enables intelligent data management and decision-making. Aim: Using time-lapse data of HepG2 cells exposed to lipid nanoparticles loaded with mRNA for expression of GFP, the authors hypothesized that it is possible to predict in advance whether a cell will express GFP. Methods: The first modeling approach used a convolutional neural network extracting per-cell features at early time points. These features were then combined and explored using either a long short-term memory network (approach 2) or time series feature extraction and gradient boosting machines (approach 3). Results: Accounting for the temporal dynamics significantly improved performance. Conclusion: The results highlight the benefit of accounting for temporal dynamics when studying drug delivery using high-content imaging.

A Novel Graphene Quantum Dot-Based mRNA Delivery Platform.

During the last decades, there has been growing interest in using therapeutic messager RNA (mRNA) together with drug delivery systems. Naked, unformulated mRNA is, however, unable to cross the cell membrane and is susceptible to degradation. Here we use graphene quantum dots (GQDs) functionalized with polyethyleneimine (PEI) as a novel mRNA delivery system. Our results show that these modified GQDs can be used to deliver intact and functional mRNA to Huh-7 hepatocarcinoma cells at low doses and, that the GQDs are not toxic, although cellular toxicity is a problem for these first-generation modified particles. Functionalized GQDs represent a potentially interesting delivery system that is easy to manufacture, stable and effective.

Rapid development of cloud-native intelligent data pipelines for scientific data streams using the HASTE Toolkit.

BACKGROUND: Large streamed datasets, characteristic of life science applications, are often resource-intensive to process, transport and store. We propose a pipeline model, a design pattern for scientific pipelines, where an incoming stream of scientific data is organized into a tiered or ordered "data hierarchy". We introduce the HASTE Toolkit, a proof-of-concept cloud-native software toolkit based on this pipeline model, to partition and prioritize data streams to optimize use of limited computing resources. FINDINGS: In our pipeline model, an "interestingness function" assigns an interestingness score to data objects in the stream, inducing a data hierarchy. From this score, a "policy" guides decisions on how to prioritize computational resource use for a given object. The HASTE Toolkit is a collection of tools to adopt this approach. We evaluate with 2 microscopy imaging case studies. The first is a high content screening experiment, where images are analyzed in an on-premise container cloud to prioritize storage and subsequent computation. The second considers edge processing of images for upload into the public cloud for real-time control of a transmission electron microscope. CONCLUSIONS: Through our evaluation, we created smart data pipelines capable of effective use of storage, compute, and network resources, enabling more efficient data-intensive experiments. We note a beneficial separation between scientific concerns of data priority, and the implementation of this behaviour for different resources in different deployment contexts. The toolkit allows intelligent prioritization to be `bolted on' to new and existing systems - and is intended for use with a range of technologies in different deployment scenarios.

A high-throughput Galectin-9 imaging assay for quantifying nanoparticle uptake, endosomal escape and functional RNA delivery.

RNA-based therapies have great potential to treat many undruggable human diseases. However, their efficacy, in particular for mRNA, remains hampered by poor cellular delivery and limited endosomal escape. Development and optimisation of delivery vectors, such as lipid nanoparticles (LNPs), are impeded by limited screening methods to probe the intracellular processing of LNPs in sufficient detail. We have developed a high-throughput imaging-based endosomal escape assay utilising a Galectin-9 reporter and fluorescently labelled mRNA to probe correlations between nanoparticle-mediated uptake, endosomal escape frequency, and mRNA translation. Furthermore, this assay has been integrated within a screening platform for optimisation of lipid nanoparticle formulations. We show that Galectin-9 recruitment is a robust, quantitative reporter of endosomal escape events induced by different mRNA delivery nanoparticles and small molecules. We identify nanoparticles with superior escape properties and demonstrate cell line variances in endosomal escape response, highlighting the need for fine-tuning of delivery formulations for specific applications.

High-Throughput Automation of Endosomolytic Polymers for mRNA Delivery.

In recent years, there has been an increasing interest in designing delivery systems to enhance the efficacy of RNA-based therapeutics. Here, we have synthesized copolymers comprised of dimethylaminoethyl methacrylate (DMAEMA) or diethylaminoethyl methacrylate (DEAEMA) copolymerized with alkyl methacrylate monomers ranging from 2 to 12 carbons, and developed a high throughput workflow for rapid investigation of their applicability for mRNA delivery. The structure activity relationship revealed that the mRNA encapsulation efficiency is improved by increasing the cationic density and use of shorter alkyl side chains (2-6 carbons). Minimal cytotoxicity was observed when using DEAEMA-co-BMA (EB) polyplexes up to 18 h after dosing, independent of a poly(ethylene glycol) (PEG) first block. The lowest molecular weight polymer (EB10,250) performed best, exhibiting greater transfection than polyethyenimine (PEI) based upon the number of cells transfected and mean intensity. Conventional investigations into the performance of polymeric materials for mRNA delivery is quite tedious, consequently limiting the number of materials and formulation conditions that can be studied. The high throughput approach presented here can accelerate the screening of polymeric systems and paves the way for expanding this generalizable approach to assess various materials for mRNA delivery.

Effect of complexing lipids on cellular uptake and expression of messenger RNA in human skin explants.

Messenger RNA (mRNA) represents a promising next-generation approach for both treatment and vaccination. Lipid based particles are one of the most investigated delivery systems for mRNA formulations. Here we explore how the complexing lipid affects uptake and translation of lipoplex-delivered RNA in resident cells in human skin explants and, we explore a more modular delivery system that utilizes mRNA added to pre-formed nanoparticles prior to dosing. We prepared formulations of lipoplexes with ionizable, cationic or zwitterionic lipids, externally complexed these with mRNA, and observed which cells internalized and/or expressed the mRNA over 72 h after intradermal injections into primary, human, skin explants. Using a flow cytometry panel to assess cellular phenotypes, mRNA uptake and mRNA expression, we found that, unlike other cell types, adipocytes expressed mRNA efficiently at 4 and 24 h after mRNA-lipoplex injection and contributed the greatest proportion of total RNA-encoded protein expression, despite being the lowest frequency cell type. Other cell types (epithelial cells, fibroblasts, T cells, B cells, dendritic cells, monocytes, NK cells, Langerhans cells, and leukocytes) had increasing mRNA expression over the course of 72 h, irrespective of lipoplex formulation. We observed that overall charge of the particle, but not the complexing lipid classification, was predictive for the pattern of mRNA uptake and expression among resident cell types in this model. This study provides insight into maximizing protein expression, using modular mRNA lipoplexes that are more compatible with product development, in a clinically relevant, human skin explant model.

Extracting quantitative biological information from bright-field cell images using deep learning.

Quantitative analysis of cell structures is essential for biomedical and pharmaceutical research. The standard imaging approach relies on fluorescence microscopy, where cell structures of interest are labeled by chemical staining techniques. However, these techniques are often invasive and sometimes even toxic to the cells, in addition to being time consuming, labor intensive, and expensive. Here, we introduce an alternative deep-learning-powered approach based on the analysis of bright-field images by a conditional generative adversarial neural network (cGAN). We show that this is a robust and fast-converging approach to generate virtually stained images from the bright-field images and, in subsequent downstream analyses, to quantify the properties of cell structures. Specifically, we train a cGAN to virtually stain lipid droplets, cytoplasm, and nuclei using bright-field images of human stem-cell-derived fat cells (adipocytes), which are of particular interest for nanomedicine and vaccine development. Subsequently, we use these virtually stained images to extract quantitative measures about these cell structures. Generating virtually stained fluorescence images is less invasive, less expensive, and more reproducible than standard chemical staining; furthermore, it frees up the fluorescence microscopy channels for other analytical probes, thus increasing the amount of information that can be extracted from each cell. To make this deep-learning-powered approach readily available for other users, we provide a Python software package, which can be easily personalized and optimized for specific virtual-staining and cell-profiling applications.

Aligned nanofiber scaffolds improve functionality of cardiomyocytes differentiated from human induced pluripotent stem cell-derived cardiac progenitor cells.

Cardiac progenitor cells (CPCs), capable of differentiating into multiple cardiac cell types including cardiomyocytes (CMs), endothelial cells, and smooth muscle cells, are promising candidates for cardiac repair/regeneration. In vitro model systems where cells are grown in a more in vivo-like environment, such as 3D cultures, have been shown to be more predictive than 2D culture for studying cell biology and disease pathophysiology. In this report, we focused on using Wnt inhibitors to study the differentiation of human iPSC-CPCs under 2D or 3D culture conditions by measuring marker protein and gene expression as well as intracellular Ca2+ oscillation. Our results show that the 3D culture with aligned nanofiber scaffolds, mimicing the architecture of the extracellular matrix of the heart, improve the differentiation of iPSC-CPCs to functional cardiomyocytes induced by Wnt inhibition, as shown with increased number of cardiac Troponin T (cTnT)-positive cells and synchronized intracellular Ca2+ oscillation. In addition, we studied if 3D nanofiber culture can be used as an in vitro model for compound screening by testing a number of other differentiation factors including a ALK5 inhibitor and inhibitors of BMP signaling. This work highlights the importance of using a more relevant in vitro model and measuring not only the expression of marker proteins but also the functional readout in a screen in order to identify the best compounds and to investigate the resulting biology.

Identification and analyses of inhibitors targeting apolipoprotein(a) kringle domains KIV-7, KIV-10, and KV provide insight into kringle domain function.

Increased plasma concentrations of lipoprotein(a) (Lp(a)) are associated with an increased risk for cardiovascular disease. Lp(a) is composed of apolipoprotein(a) (apo(a)) covalently bound to apolipoprotein B of low-density lipoprotein (LDL). Many of apo(a)'s potential pathological properties, such as inhibition of plasmin generation, have been attributed to its main structural domains, the kringles, and have been proposed to be mediated by their lysine-binding sites. However, available small-molecule inhibitors, such as lysine analogs, bind unselectively to kringle domains and are therefore unsuitable for functional characterization of specific kringle domains. Here, we discovered small molecules that specifically bind to the apo(a) kringle domains KIV-7, KIV-10, and KV. Chemical synthesis yielded compound AZ-05, which bound to KIV-10 with a Kd of 0.8 µm and exhibited more than 100-fold selectivity for KIV-10, compared with the other kringle domains tested, including plasminogen kringle 1. To better understand and further improve ligand selectivity, we determined the crystal structures of KIV-7, KIV-10, and KV in complex with small-molecule ligands at 1.6-2.1 Å resolutions. Furthermore, we used these small molecules as chemical probes to characterize the roles of the different apo(a) kringle domains in in vitro assays. These assays revealed the assembly of Lp(a) from apo(a) and LDL, as well as potential pathophysiological mechanisms of Lp(a), including (i) binding to fibrin, (ii) stimulation of smooth-muscle cell proliferation, and (iii) stimulation of LDL uptake into differentiated monocytes. Our results indicate that a small-molecule inhibitor targeting the lysine-binding site of KIV-10 can combat the pathophysiological effects of Lp(a).