A systematic analysis of regression models for protein engineering.

To optimize proteins for particular traits holds great promise for industrial and pharmaceutical purposes. Machine Learning is increasingly applied in this field to predict properties of proteins, thereby guiding the experimental optimization process. A natural question is: How much progress are we making with such predictions, and how important is the choice of regressor and representation? In this paper, we demonstrate that different assessment criteria for regressor performance can lead to dramatically different conclusions, depending on the choice of metric, and how one defines generalization. We highlight the fundamental issues of sample bias in typical regression scenarios and how this can lead to misleading conclusions about regressor performance. Finally, we make the case for the importance of calibrated uncertainty in this domain.

Neonatal intestinal mucus barrier changes in response to maturity, inflammation, and sodium decanoate supplementation.

The integrity of the intestinal mucus barrier is crucial for human health, as it serves as the body's first line of defense against pathogens. However, postnatal development of the mucus barrier and interactions between maturity and its ability to adapt to external challenges in neonatal infants remain unclear. In this study, we unveil a distinct developmental trajectory of the mucus barrier in preterm piglets, leading to enhanced mucus microstructure and reduced mucus diffusivity compared to term piglets. Notably, we found that necrotizing enterocolitis (NEC) is associated with increased mucus diffusivity of our large pathogen model compound, establishing a direct link between the NEC condition and the mucus barrier. Furthermore, we observed that addition of sodium decanoate had varying effects on mucus diffusivity depending on maturity and health state of the piglets. These findings demonstrate that regulatory mechanisms governing the neonatal mucosal barrier are highly complex and are influenced by age, maturity, and health conditions. Therefore, our results highlight the need for specific therapeutic strategies tailored to each neonatal period to ensure optimal gut health.

Deconvoluting the Effect of Cell-Penetrating Peptides for Enhanced and Controlled Insertion of Large-Scale DNA Nanopores.

DNA nanopores have emerged as powerful tools for molecular sensing, but the efficient insertion of large DNA nanopores into lipid membranes remains challenging. In this study, we investigate the potential of cell-penetrating peptides (CPPs), specifically SynB1 and GALA, to enhance the insertion efficiency of large DNA nanopores. We constructed SynB1- or GALA-functionalized DNA nanopores with an 11 nm inner diameter and visualized and quantified their membrane insertion using a TIRF microscopy-based single-liposome assay. The results demonstrated that incorporating an increasing number of SynB1 or GALA peptides into the DNA nanopore significantly enhanced the membrane perforation. Kinetic analysis revealed that the DNA nanopore scaffold played a role in prearranging the CPPs, which facilitated membrane interaction and pore formation. Notably, the use of pH-responsive GALA peptides allowed highly efficient and pH-controlled insertion of large DNA pores. Furthermore, single-channel recording elucidated that the insertion process of single GALA-modified nanopores into planar lipid bilayers was dynamic, likely forming transient large toroidal pores. Overall, our study highlights the potential of CPPs as insertion enhancers for DNA nanopores, which opens avenues for improved molecule sensing and the controlled release of cargo molecules.

Deep learning assisted single particle tracking for automated correlation between diffusion and function.

Sub-cellular diffusion in living systems reflects cellular processes and interactions. Recent advances in optical microscopy allow the tracking of this nanoscale diffusion of individual objects with an unprecedented level of precision. However, the agnostic and automated extraction of functional information from the diffusion of molecules and organelles within the sub-cellular environment, is labor-intensive and poses a significant challenge. Here we introduce DeepSPT, a deep learning framework to interpret the diffusional 2D or 3D temporal behavior of objects in a rapid and efficient manner, agnostically. Demonstrating its versatility, we have applied DeepSPT to automated mapping of the early events of viral infections, identifying distinct types of endosomal organelles, and clathrin-coated pits and vesicles with up to 95% accuracy and within seconds instead of weeks. The fact that DeepSPT effectively extracts biological information from diffusion alone illustrates that besides structure, motion encodes function at the molecular and subcellular level.

SEMORE: SEgmentation and MORphological fingErprinting by machine learning automates super-resolution data analysis.

The morphology of protein assemblies impacts their behaviour and contributes to beneficial and aberrant cellular responses. While single-molecule localization microscopy provides the required spatial resolution to investigate these assemblies, the lack of universal robust analytical tools to extract and quantify underlying structures limits this powerful technique. Here we present SEMORE, a semi-automatic machine learning framework for universal, system- and input-dependent, analysis of super-resolution data. SEMORE implements a multi-layered density-based clustering module to dissect biological assemblies and a morphology fingerprinting module for quantification by multiple geometric and kinetics-based descriptors. We demonstrate SEMORE on simulations and diverse raw super-resolution data: time-resolved insulin aggregates, and published data of dSTORM imaging of nuclear pore complexes, fibroblast growth receptor 1, sptPALM of Syntaxin 1a and dynamic live-cell PALM of ryanodine receptors. SEMORE extracts and quantifies all protein assemblies, their temporal morphology evolution and provides quantitative insights, e.g. classification of heterogeneous insulin aggregation pathways and NPC geometry in minutes. SEMORE is a general analysis platform for super-resolution data, and being a time-aware framework can also support the rise of 4D super-resolution data.

Probing Activation and Conformational Dynamics of the Vesicle-Reconstituted ß2 Adrenergic Receptor at the Single-Molecule Level.

G-protein-coupled receptors (GPCRs) are structurally flexible membrane proteins that mediate a host of physiological responses to extracellular ligands like hormones and neurotransmitters. Fine features of their dynamic structural behavior are hypothesized to encode the functional plasticity seen in GPCR activity, where ligands with different efficacies can direct the same receptor toward different signaling phenotypes. Although the number of GPCR crystal structures is increasing, the receptors are characterized by complex and poorly understood conformational landscapes. Therefore, we employed a fluorescence microscopy assay to monitor conformational dynamics of single ß2 adrenergic receptors (ß2ARs). To increase the biological relevance of our findings, we decided not to reconstitute the receptor in detergent micelles but rather lipid membranes as proteoliposomes. The conformational dynamics were monitored by changes in the intensity of an environmentally sensitive boron-dipyrromethene (BODIPY 493/503) fluorophore conjugated to an endogenous cysteine (located at the cytoplasmic end of the sixth transmembrane helix of the receptor). Using total internal reflection fluorescence microscopy (TIRFM) and a single small unilamellar liposome assay that we previously developed, we followed the real-time dynamic properties of hundreds of single ß2ARs reconstituted in a native-like environment─lipid membranes. Our results showed that ß2AR-BODIPY fluctuates between several states of different intensity on a time scale of seconds, compared to BODIPY-lipid conjugates that show almost entirely stable fluorescence emission in the absence and presence of the full agonist BI-167107. Agonist stimulation changes the ß2AR dynamics, increasing the population of states with higher intensities and prolonging their durations, consistent with bulk experiments. The transition density plot demonstrates that ß2AR-BODIPY, in the absence of the full agonist, interconverts between states of low and moderate intensity, while the full agonist renders transitions between moderate and high-intensity states more probable. This redistribution is consistent with a mechanism of conformational selection and is a promising first step toward characterizing the conformational dynamics of GPCRs embedded in a lipid bilayer.

Everything, everywhere, almost at once.

A new platform that can follow the movement of individual proteins inside millions of cells in a single day will help contribute to existing knowledge of cell biology and identify new therapeutics.

Constitutive Endolysosomal Perforation in Neurons allows Induction of α-Synuclein Aggregation by Internalized Pre-Formed Fibrils.

The endocytic pathway is both an essential route of molecular uptake in cells and a potential entry point for pathology-inducing cargo. The cell-to-cell spread of cytotoxic aggregates, such as those of α-synuclein (α-syn) in Parkinson's Disease (PD), exemplifies this duality. Here we used a human iPSC-derived induced neuronal model (iNs) prone to death mediated by aggregation in late endosomes and lysosomes of endogenous α-syn, seeded by internalized pre-formed fibrils of α-syn (PFFs). This PFF-mediated death was not observed with parental iPSCs or other non-neuronal cells. Using live-cell optical microscopy to visualize the read out of biosensors reporting endo-lysosome wounding, we discovered that up to about 10% of late endosomes and lysosomes in iNs exhibited spontaneous constitutive perforations, regardless of the presence of internalized PFFs. This wounding, absent in parental iPSCs and non-neuronal cells, corresponded to partial damage by nanopores in the limiting membranes of a subset of endolysosomes directly observed by volumetric focused ion beam scanning electron microscopy (FIB-SEM) in iNs and in CA1 pyramidal neurons from mouse brain, and not found in iPSCs or in other non-neuronal cells in culture or in mouse liver and skin. We suggest that the compromised limiting membranes in iNs and neurons in general are the primary conduit for cytosolic α-syn to access PFFs entrapped within endo-lysosomal lumens, initiating PFF-mediated α-syn aggregation. Significantly, eradicating the intrinsic endolysosomal perforations in iNs by inhibiting the endosomal Phosphatidylinositol-3-Phosphate/Phosphatidylinositol 5-Kinase (PIKfyve kinase) using Apilimod or Vacuolin-1 markedly reduced PFF-induced α-syn aggregation, despite PFFs continuing to enter the endolysosomal compartment. Crucially, this intervention also diminished iN death associated with PFF incubation. Our results reveal the surprising presence of intrinsically perforated endo-lysosomes in neurons, underscoring their crucial early involvement in the genesis of toxic α-syn aggregates induced by internalized PFFs. This discovery offers a basis for employing PIKfyve kinase inhibition as a potential therapeutic strategy to counteract synucleinopathies.

Deep learning assisted single particle tracking for automated correlation between diffusion and function.

Sub-cellular diffusion in living systems reflects cellular processes and interactions. Recent advances in optical microscopy allow the tracking of this nanoscale diffusion of individual objects with an unprecedented level of precision. However, the agnostic and automated extraction of functional information from the diffusion of molecules and organelles within the sub-cellular environment, is labor-intensive and poses a significant challenge. Here we introduce DeepSPT, a deep learning framework to interpret the diffusional 2D or 3D temporal behavior of objects in a rapid and efficient manner, agnostically. Demonstrating its versatility, we have applied DeepSPT to automated mapping of the early events of viral infections, identifying distinct types of endosomal organelles, and clathrin-coated pits and vesicles with up to 95% accuracy and within seconds instead of weeks. The fact that DeepSPT effectively extracts biological information from diffusion alone indicates that besides structure, motion encodes function at the molecular and subcellular level.

Realizing integration in structural biology: The 2022 ISBUC Annual Meeting.

This meeting report presents the 2022 Annual Meeting of the cluster for Integrative Structural Biology at the University of Copenhagen (ISBUC) and discusses the cluster approach to interdisciplinary research management. This approach successfully facilitates cross-faculty and inter-departmental collaboration. Innovative integrative research collaborations ignited by ISBUC, as well as research presented at the meeting, are showcased.

Single-Particle Tracking of Thermomyces lanuginosus Lipase Reveals How Mutations in the Lid Region Remodel Its Diffusion.

The function of most lipases is controlled by the lid, which undergoes conformational changes at a water-lipid interface to expose the active site, thus activating catalysis. Understanding how lid mutations affect lipases' function is important for designing improved variants. Lipases' function has been found to correlate with their diffusion on the substrate surface. Here, we used single-particle tracking (SPT), a powerful tool for deciphering enzymes' diffusional behavior, to study Thermomyces lanuginosus lipase (TLL) variants with different lid structures in a laundry-like application condition. Thousands of parallelized recorded trajectories and hidden Markov modeling (HMM) analysis allowed us to extract three interconverting diffusional states and quantify their abundance, microscopic transition rates, and the energy barriers for sampling them. Combining those findings with ensemble measurements, we determined that the overall activity variation in the application condition is dependent on surface binding and lipase mobility when bound. Specifically, the L4 variant with a TLL-like lid and wild-type (WT) TLL displayed similar ensemble activity, but WT bound stronger to the surface than L4, while L4 had a higher diffusion coefficient and thus activity when bound to the surface. These mechanistic elements can only be de-convoluted by our combined assays. Our findings offer fresh perspectives on the development of the next iteration of enzyme-based detergent.

Dimensional Reduction for Single-Molecule Imaging of DNA and Nucleosome Condensation by Polyamines, HP1α and Ki-67.

Macromolecules organize themselves into discrete membrane-less compartments. Mounting evidence has suggested that nucleosomes as well as DNA itself can undergo clustering or condensation to regulate genomic activity. Current in vitro condensation studies provide insight into the physical properties of condensates, such as surface tension and diffusion. However, methods that provide the resolution needed for complex kinetic studies of multicomponent condensation are desired. Here, we use a supported lipid bilayer platform in tandem with total internal reflection microscopy to observe the two-dimensional movement of DNA and nucleosomes at the single-molecule resolution. This dimensional reduction from three-dimensional studies allows us to observe the initial condensation events and dissolution of these early condensates in the presence of physiological condensing agents. Using polyamines, we observed that the initial condensation happens on a time scale of minutes while dissolution occurs within seconds upon charge inversion. Polyamine valency, DNA length, and GC content affect the threshold polyamine concentration for condensation. Protein-based nucleosome condensing agents, HP1α and Ki-67, have much lower threshold concentrations for condensation than charge-based condensing agents, with Ki-67 being the most effective, requiring as low as 100 pM for nucleosome condensation. In addition, we did not observe condensate dissolution even at the highest concentrations of HP1α and Ki-67 tested. We also introduce a two-color imaging scheme where nucleosomes of high density labeled in one color are used to demarcate condensate boundaries and identical nucleosomes of another color at low density can be tracked relative to the boundaries after Ki-67-mediated condensation. Our platform should enable the ultimate resolution of single molecules in condensation dynamics studies of chromatin components under defined physicochemical conditions.

Enhanced hexamerization of insulin via assembly pathway rerouting revealed by single particle studies.

Insulin formulations with diverse oligomerization states are the hallmark of interventions for the treatment of diabetes. Here using single-molecule recordings we firstly reveal that insulin oligomerization can operate via monomeric additions and secondly quantify the existence, abundance and kinetic characterization of diverse insulin assembly and disassembly pathways involving addition of monomeric, dimeric or tetrameric insulin species. We propose and experimentally validate a model where the insulin self-assembly pathway is rerouted, favoring monomeric or oligomeric assembly, by solution concentration, additives and formulations. Combining our practically complete kinetic characterization with rate simulations, we calculate the abundance of each oligomeric species from nM to mM offering mechanistic insights and the relative abundance of all oligomeric forms at concentrations relevant both for secreted and administrated insulin. These reveal a high abundance of all oligomers and a significant fraction of hexamer resulting in practically halved bioavailable monomer concentration. In addition to providing fundamental new insights, the results and toolbox presented here can be universally applied, contributing to the development of optimal insulin formulations and the deciphering of oligomerization mechanisms for additional proteins.

Heterogeneous and Surface-Catalyzed Amyloid Aggregation Monitored by Spatially Resolved Fluorescence and Single Molecule Microscopy.

Amyloid aggregation is associated with many diseases and may also occur in therapeutic protein formulations. Addition of co-solutes is a key strategy to modulate the stability of proteins in pharmaceutical formulations and select inhibitors for drug design in the context of diseases. However, the heterogeneous nature of this multicomponent system in terms of structures and mechanisms poses a number of challenges for the analysis of the chemical reaction. Using insulin as protein system and polysorbate 80 as co-solute, we combine a spatially resolved fluorescence approach with single molecule microscopy and machine learning methods to kinetically disentangle the different contributions from multiple species within a single aggregation experiment. We link the presence of interfaces to the degree of heterogeneity of the aggregation kinetics and retrieve the rate constants and underlying mechanisms for single aggregation events. Importantly, we report that the mechanism of inhibition of the self-assembly process depends on the details of the growth pathways of otherwise macroscopically identical species. This information can only be accessed by the analysis of single aggregate events, suggesting our method as a general tool for a comprehensive physicochemical characterization of self-assembly reactions.

In vitro and in vivo immunogenicity assessment of protein aggregate characteristics.

The immunogenicity risk of therapeutic protein aggregates has been extensively investigated over the past decades. While it is established that not all aggregates are equally immunogenic, the specific aggregate characteristics, which are most likely to induce an immune response, remain ambiguous. The aim of this study was to perform comprehensive in vitro and in vivo immunogenicity assessment of human insulin aggregates varying in size, structure and chemical modifications, while keeping other morphological characteristics constant. We found that flexible aggregates with highly altered secondary structure were most immunogenic in all setups, while compact aggregates with native-like structure were found to be immunogenic primarily in vivo. Moreover, sub-visible (1-100 µm) aggregates were found to be more immunogenic than sub-micron (0.1-1 µm) aggregates, while chemical modifications (deamidation, ethylation and covalent dimers) were not found to have any measurable impact on immunogenicity. The findings highlight the importance of utilizing aggregates varying in few characteristics for assessment of immunogenicity risk of specific morphological features and may provide a workflow for reliable particle analysis in biotherapeutics.

Morphological integrity of insulin amyloid-like aggregates depends on preparation methods and post-production treatments.

Protein aggregates are often varying extensively in their morphological characteristics, which may lead to various biological outcomes, such as increased immunogenicity risk. However, isolation of aggregates with a specific morphology within an ensemble is often challenging. To gain vital knowledge on the effects of aggregate characteristics, samples containing a single morphology must be produced by direct control of the aggregation process. Moreover, the formed aggregates need to be in an aqueous solution suitable for biological assays, while keeping their morphology intact. Here we evaluated the dependence of morphology and integrity of amyloid-like fibrils and spherulites on preparation conditions and post-treatment methods. Samples containing either amyloid-like fibrils or spherulites produced from human insulin in acetic acid solutions are dependent on the presence of salt (NaCl). Moreover, mechanical shaking (600 rpm) inhibits spherulite formation, while only affecting the length of the formed fibrils compared to quiescent conditions. Besides shaking, the initial protein concentration in the formulation was found to control fibril length. Surprisingly, exchanging the solution used for aggregate formation to a physiologically relevant buffer, had a striking effect on the morphological integrity of the fibril and spherulite samples. Especially the secondary structure of one of our spherulite samples presented dramatic changes of the aggregated ß-sheet content after exchanging the solution, emphasizing the importance of the aggregate stability. These results and considerations have profound implications on the data interpretation and should be implemented in the workflow for both fundamental characterization of aggregates as well as assays for evaluation of their corresponding biological effects.

A blind benchmark of analysis tools to infer kinetic rate constants from single-molecule FRET trajectories.

Single-molecule FRET (smFRET) is a versatile technique to study the dynamics and function of biomolecules since it makes nanoscale movements detectable as fluorescence signals. The powerful ability to infer quantitative kinetic information from smFRET data is, however, complicated by experimental limitations. Diverse analysis tools have been developed to overcome these hurdles but a systematic comparison is lacking. Here, we report the results of a blind benchmark study assessing eleven analysis tools used to infer kinetic rate constants from smFRET trajectories. We test them against simulated and experimental data containing the most prominent difficulties encountered in analyzing smFRET experiments: different noise levels, varied model complexity, non-equilibrium dynamics, and kinetic heterogeneity. Our results highlight the current strengths and limitations in inferring kinetic information from smFRET trajectories. In addition, we formulate concrete recommendations and identify key targets for future developments, aimed to advance our understanding of biomolecular dynamics through quantitative experiment-derived models.

Direct observation of heterogeneous formation of amyloid spherulites in real-time by super-resolution microscopy.

Protein misfolding in the form of fibrils or spherulites is involved in a spectrum of pathological abnormalities. Our current understanding of protein aggregation mechanisms has primarily relied on the use of spectrometric methods to determine the average growth rates and diffraction-limited microscopes with low temporal resolution to observe the large-scale morphologies of intermediates. We developed a REal-time kinetics via binding and Photobleaching LOcalization Microscopy (REPLOM) super-resolution method to directly observe and quantify the existence and abundance of diverse aggregate morphologies of human insulin, below the diffraction limit and extract their heterogeneous growth kinetics. Our results revealed that even the growth of microscopically identical aggregates, e.g., amyloid spherulites, may follow distinct pathways. Specifically, spherulites do not exclusively grow isotropically but, surprisingly, may also grow anisotropically, following similar pathways as reported for minerals and polymers. Combining our technique with machine learning approaches, we associated growth rates to specific morphological transitions and provided energy barriers and the energy landscape at the level of single aggregate morphology. Our unifying framework for the detection and analysis of spherulite growth can be extended to other self-assembled systems characterized by a high degree of heterogeneity, disentangling the broad spectrum of diverse morphologies at the single-molecule level.

Single Vesicle Fluorescence-Bleaching Assay for Multi-Parameter Analysis of Proteoliposomes by Total Internal Reflection Fluorescence Microscopy.

Reconstitution of membrane proteins into model membranes is an essential approach for their functional analysis under chemically defined conditions. Established model-membrane systems used in ensemble average measurements are limited by sample heterogeneity and insufficient knowledge of lipid and protein content at the single vesicle level, which limits quantitative analysis of vesicle properties and prevents their correlation with protein activity. Here, we describe a versatile total internal reflection fluorescence microscopy-based bleaching protocol that permits parallel analysis of multiple parameters (physical size, tightness, unilamellarity, membrane protein content, and orientation) of individual proteoliposomes prepared with fluorescently tagged membrane proteins and lipid markers. The approach makes use of commercially available fluorophores including the commonly used nitrobenzoxadiazole dye and may be applied to deduce functional molecular characteristics of many types of reconstituted fluorescently tagged membrane proteins.