Cryo-EM images of phase-separated lipid bilayer vesicles analyzed with a machine-learning approach.

Lateral lipid heterogeneity (i.e., raft formation) in biomembranes plays a functional role in living cells. Three-component mixtures of low- and high-melting lipids plus cholesterol offer a simplified experimental model for raft domains in which a liquid-disordered (Ld) phase coexists with a liquid-ordered (Lo) phase. Using such models, we recently showed that cryogenic electron microscopy (cryo-EM) can detect phase separation in lipid vesicles based on differences in bilayer thickness. However, the considerable noise within cryo-EM data poses a significant challenge for accurately determining the membrane phase state at high spatial resolution. To this end, we have developed an image-processing pipeline that utilizes machine learning (ML) to predict the bilayer phase in projection images of lipid vesicles. Importantly, the ML method exploits differences in both the thickness and molecular density of Lo compared to Ld, which leads to improved phase identification. To assess accuracy, we used artificial images of phase-separated lipid vesicles generated from all-atom molecular dynamics simulations of Lo and Ld phases. Synthetic ground-truth data sets mimicking a series of compositions along a tieline of Ld + Lo coexistence were created and then analyzed with various ML models. For all tieline compositions, we find that the ML approach can correctly identify the bilayer phase with >90% accuracy, thus providing a means to isolate the intensity profiles of coexisting Ld and Lo phases, as well as accurately determine domain-size distributions, number of domains, and phase-area fractions. The method described here provides a framework for characterizing nanoscopic lateral heterogeneities in membranes and paves the way for a more detailed understanding of raft properties in biological contexts.

Role of a Pdlim5:PalmD complex in directing dendrite morphology.

Neuronal connectivity is regulated during normal brain development with the arrangement of spines and synapses being dependent on the morphology of dendrites. Further, in multiple neurodevelopmental and aging disorders, disruptions of dendrite formation or shaping is associated with atypical neuronal connectivity. We showed previously that Pdlim5 binds delta-catenin and promotes dendrite branching. We report here that Pdlim5 interacts with PalmD, a protein previously suggested by others to interact with the cytoskeleton (e.g., via adducin/spectrin) and to regulate membrane shaping. Functionally, the knockdown of PalmD or Pdlim5 in rat primary hippocampal neurons dramatically reduces branching and conversely, PalmD exogenous expression promotes dendrite branching as does Pdlim5. Further, we show that each proteins' effects are dependent on the presence of the other. In summary, using primary rat hippocampal neurons we reveal the contributions of a novel Pdlim5:PalmD protein complex, composed of functionally inter-dependent components responsible for shaping neuronal dendrites.

Role of a Pdlim5:PalmD complex in directing dendrite morphology.

Neuronal connectivity is regulated during normal brain development with the arrangement of spines and synapses being dependent on the morphology of dendrites. Further, in multiple neurodevelopmental and aging disorders, disruptions of dendrite formation or shaping is associated with atypical neuronal connectivity. We showed previously that Pdlim5 binds delta-catenin and promotes dendrite branching (Baumert et al., J Cell Biol 2020). We report here that Pdlim5 interacts with PalmD, a protein previously suggested by others to interact with the cytoskeleton (e.g., via adducin/ spectrin) and to regulate membrane shaping. Functionally, the knockdown of PalmD or Pdlim5 in rat primary hippocampal neurons dramatically reduces branching and conversely, PalmD exogenous expression promotes dendrite branching as does Pdlim5. Further, we show that effects of each protein are dependent on the presence of the other. In summary, using primary rat hippocampal neurons we reveal the contributions of a novel Pdlim5:PalmD protein complex, composed of functionally inter-dependent components responsible for shaping neuronal dendrites.

Enhanced presynaptic mitochondrial energy production is required for memory formation.

Some of the prominent features of long-term memory formation include protein synthesis, gene expression, enhanced neurotransmitter release, increased excitability, and formation of new synapses. As these processes are critically dependent on mitochondrial function, we hypothesized that increased mitochondrial respiration and dynamics would play a prominent role in memory formation. To address this possibility, we measured mitochondrial oxygen consumption (OCR) in hippocampal tissue punches from trained and untrained animals. Our results show that context fear training significantly increased basal, ATP synthesis-linked, and maximal OCR in the Shaffer collateral-CA1 synaptic region, but not in the CA1 cell body layer. These changes were recapitulated in synaptosomes isolated from the hippocampi of fear-trained animals. As dynamin-related protein 1 (Drp1) plays an important role in mitochondrial fission, we examined its role in the increased mitochondrial respiration observed after fear training. Drp1 inhibitors decreased the training-associated enhancement of OCR and impaired contextual fear memory, but did not alter the number of synaptosomes containing mitochondria. Taken together, our results show context fear training increases presynaptic mitochondria respiration, and that Drp-1 mediated enhanced energy production in CA1 pre-synaptic terminals is necessary for context fear memory that does not result from an increase in the number of synaptosomes containing mitochondria or an increase in mitochondrial mass within the synaptic layer.

Rapid Synthesis of Cryo-ET Data for Training Deep Learning Models.

Deep learning excels at cryo-tomographic image restoration and segmentation tasks but is hindered by a lack of training data. Here we introduce cryo-TomoSim (CTS), a MATLAB-based software package that builds coarse-grained models of macromolecular complexes embedded in vitreous ice and then simulates transmitted electron tilt series for tomographic reconstruction. We then demonstrate the effectiveness of these simulated datasets in training different deep learning models for use on real cryotomographic reconstructions. Computer-generated ground truth datasets provide the means for training models with voxel-level precision, allowing for unprecedented denoising and precise molecular segmentation of datasets. By modeling phenomena such as a three-dimensional contrast transfer function, probabilistic detection events, and radiation-induced damage, the simulated cryo-electron tomograms can cover a large range of imaging content and conditions to optimize training sets. When paired with small amounts of training data from real tomograms, networks become incredibly accurate at segmenting in situ macromolecular assemblies across a wide range of biological contexts.

p120-catenin subfamily members have distinct as well as shared effects on dendrite morphology during neuron development in vitro.

Dendritic arborization is essential for proper neuronal connectivity and function. Conversely, abnormal dendrite morphology is associated with several neurological pathologies like Alzheimer's disease and schizophrenia. Among major intrinsic mechanisms that determine the extent of the dendritic arbor is cytoskeletal remodeling. Here, we characterize and compare the impact of the four proteins involved in cytoskeletal remodeling-vertebrate members of the p120-catenin subfamily-on neuronal dendrite morphology. In relation to each of their own distributions, we find that p120-catenin and delta-catenin are expressed at relatively higher proportions in growth cones compared to ARVCF-catenin and p0071-catenin; ARVCF-catenin is expressed at relatively high proportions in the nucleus; and all catenins are expressed in dendritic processes and the soma. Through altering the expression of each p120-subfamily catenin in neurons, we find that exogenous expression of either p120-catenin or delta-catenin correlates with increased dendritic length and branching, whereas their respective depletion decreases dendritic length and branching. While increasing ARVCF-catenin expression also increases dendritic length and branching, decreasing expression has no grossly observable morphological effect. Finally, increasing p0071-catenin expression increases dendritic branching, but not length, while decreasing expression decreases dendritic length and branching. These distinct localization patterns and morphological effects during neuron development suggest that these catenins have both shared and distinct roles in the context of dendrite morphogenesis.

Experiment and Simulation Reveal Residue Details for How Target Binding Tunes Calmodulin's Calcium-Binding Properties.

We aim to elucidate the molecular mechanism of the reciprocal relation of calmodulin's (CaM) target binding and its affinity for calcium ions (Ca2+), which is central to decoding CaM-dependent Ca2+ signaling in a cell. We employed stopped-flow experiments and coarse-grained molecular simulations that learn the coordination chemistry of Ca2+ in CaM from first-principle calculations. The associative memories as part of the coarse-grained force fields built on known protein structures further influence CaM's selection of its polymorphic target peptides in the simulations. We modeled the peptides from the Ca2+/CaM-binding domain of Ca2+/CaM-dependent kinase II (CaMKII), CaMKIIp (293-310) and selected distinctive mutations at the N-terminus. Our stopped-flow experiments have shown that the CaM's affinity for Ca2+ in the bound complex of Ca2+/CaM/CaMKIIp decreased significantly when Ca2+/CaM bound to the mutant peptide (296-AAA-298) compared to that bound to the wild-type peptide (296-RRK-298). The coarse-grained molecular simulations revealed that the 296-AAA-298 mutant peptide destabilized the structures of Ca2+-binding loops at the C-domain of CaM (c-CaM) due to both loss of electrostatic interactions and differences in polymorphic structures. We have leveraged a powerful coarse-grained approach to advance a residue-level understanding of the reciprocal relation in CaM, that could not be possibly achieved by other computational approaches.

Visualizing lipid membrane structure with cryo-EM: past, present, and future.

The development of electron cryomicroscopy (cryo-EM) has evolved immensely in the last several decades and is now well-established in the analysis of protein structure both in isolation and in their cellular context. This review focuses on the history and application of cryo-EM to the analysis of membrane architecture. Parallels between the levels of organization of protein structure are useful in organizing the discussion of the unique parameters that influence membrane structure and function. Importantly, the timescales of lipid motion in bilayers with respect to the timescales of sample vitrification is discussed and reveals what types of membrane structure can be reliably extracted in cryo-EM images of vitrified samples. Appreciating these limitations, a review of the application of cryo-EM to examine the lateral organization of ordered and disordered domains in reconstituted and biologically derived membranes is provided. Finally, a brief outlook for further development and application of cryo-EM to the analysis of membrane architecture is provided.

Optimization of cryo-electron microscopy for quantitative analysis of lipid bilayers.

Cryogenic electron microscopy (cryo-EM) is among the most powerful tools available for interrogating nanoscale structure of biological materials. We recently showed that cryo-EM can be used to measure the bilayer thickness of lipid vesicles and biological membranes with subangstrom precision, resulting in the direct visualization of nanoscopic domains of different thickness in multicomponent lipid mixtures and giant plasma membrane vesicles. Despite the great potential of cryo-EM for revealing the lateral organization of biomembranes, a large parameter space of experimental conditions remains to be optimized. Here, we systematically investigate the influence of instrument parameters and image postprocessing steps on the ability to accurately measure bilayer thickness and discriminate regions of different thickness within unilamellar liposomes. This unique application of cryo-EM places particular demands on image acquisition optimization and analysis due to the facts that 1) each vesicle is a different size with different curvature, 2) the domains in each vesicle can be heterogenous in size, and 3) the random orientation of vesicles amplifies the variability of domain size in projected images. We also demonstrate a spatial autocorrelation analysis to extract additional information about lateral heterogeneity.

Lipidomic atlas of mammalian cell membranes reveals hierarchical variation induced by culture conditions, subcellular membranes, and cell lineages.

Lipid membranes are ubiquitous biological organizers, required for structural and functional compartmentalization of the cell and sub-cellular organelles. Membranes in living cells are compositionally complex, comprising hundreds of dynamically regulated, distinct lipid species. Cellular physiology requires tight regulation of these lipidomic profiles to achieve proper membrane functionality. While some general features of tissue- and organelle-specific lipid complements have been identified, less is known about detailed lipidomic variations caused by cell-intrinsic or extrinsic factors. Here, we use shotgun lipidomics to report detailed, comprehensive lipidomes of a variety of cultured and primary mammalian membrane preparations to identify trends and sources of variation. Unbiased principle component analysis (PCA) shows clear separation between cultured and primary cells, with primary erythrocytes, synaptic membranes, and other mammalian tissue lipidomes sharply diverging from all cultured cell lines and also from one other. Most broadly, cultured cell membrane preparations were distinguished by their paucity of polyunsaturated lipids. Cultured mammalian cell lines were comparatively similar to one another, although we detected clear, highly reproducible lipidomic signatures of individual cell lines and plasma membrane (PM) isolations thereof. These measurements begin to establish a comprehensive lipidomic atlas of mammalian cells and tissues, identifying some major sources of variation. These observations will allow investigation of the regulation and functional significance of mammalian lipidomes in various contexts.

Novel phospho-switch function of delta-catenin in dendrite development.

In neurons, dendrites form the major sites of information receipt and integration. It is thus vital that, during development, the dendritic arbor is adequately formed to enable proper neural circuit formation and function. While several known processes shape the arbor, little is known of those that govern dendrite branching versus extension. Here, we report a new mechanism instructing dendrites to branch versus extend. In it, glutamate signaling activates mGluR5 receptors to promote Ckd5-mediated phosphorylation of the C-terminal PDZ-binding motif of delta-catenin. The phosphorylation state of this motif determines delta-catenin's ability to bind either Pdlim5 or Magi1. Whereas the delta:Pdlim5 complex enhances dendrite branching at the expense of elongation, the delta:Magi1 complex instead promotes lengthening. Our data suggest that these complexes affect dendrite development by differentially regulating the small-GTPase RhoA and actin-associated protein Cortactin. We thus reveal a "phospho-switch" within delta-catenin, subject to a glutamate-mediated signaling pathway, that assists in balancing the branching versus extension of dendrites during neural development.

Exploring the F-actin/CPEB3 interaction and its possible role in the molecular mechanism of long-term memory.

Dendritic spines are tiny membranous protrusions on the dendrites of neurons. Dendritic spines change shape in response to input signals, thereby strengthening the connections between neurons. The growth and stabilization of dendritic spines is thought to be essential for maintaining long-term memory. Actin cytoskeleton remodeling in spines is a key element of their formation and growth. More speculatively, the aggregation of CPEB3, a functional prion that binds RNA, has been reported to be involved in the maintenance of long-term memory. Here we study the interaction between actin and CPEB3 and propose a molecular model for the complex structure of CPEB3 and an actin filament (F-actin). The results of our computational modeling, including both energetic and structural analyses, are compared with novel data from peptide array experiments. Our model of the CPEB3/F-actin interaction suggests that F-actin potentially triggers the aggregation-prone structural transition of a short CPEB3 sequence by zipping it into a beta-hairpin form. We also propose that the CPEB3/F-actin interaction might be regulated by the SUMOylation of CPEB3, based on bioinformatic searches for potential SUMOylation sites as well as SUMO interacting motifs in CPEB3. On the basis of these results and the existing literature, we put forward a possible molecular mechanism underlying long-term memory that involves CPEB3's binding to actin, its aggregation, and its regulation by SUMOylation.

The ubiquitin ligase UBE4B regulates amyloid precursor protein ubiquitination, endosomal trafficking, and amyloid ß42 generation and secretion.

The extracellular accumulation of amyloid ß (Aß) fragments of amyloid precursor protein (APP) in brain parenchyma is a pathological hallmark of Alzheimer's disease (AD). APP can be cleaved into Aß on late endosomes/multivesicular bodies (MVBs). E3 ubiquitin ligases have been linked to Aß production, but specific E3 ligases associated with APP ubiquitination that may affect targeting of APP to endosomes have not yet been described. Using cultured cortical neurons isolated from rat pups, we reconstituted APP movement into the internal vesicles (ILVs) of MVBs. Loss of endosomal sorting complexes required for transport (ESCRT) components inhibited APP movement into ILVs and increased endosomal Aß42 generation, implying a requirement for APP ubiquitination. We identified an ESCRT-binding and APP-interacting endosomal E3 ubiquitin ligase, ubiquitination factor E4B (UBE4B) that regulates APP ubiquitination. Depleting UBE4B in neurons inhibited APP ubiquitination and internalization into MVBs, resulting in increased endosomal Aß42 levels and increased neuronal secretion of Aß42. When we examined AD brains, we found levels of the UBE4B-interacting ESCRT component, hepatocyte growth factor-regulated tyrosine kinase substrate (Hrs), were significantly decreased in AD brains. These data suggest that ESCRT components critical for membrane protein sorting in the endocytic pathway are altered in AD. These results indicate that the molecular machinery underlying endosomal trafficking of APP, including the ubiquitin ligase UBE4B, regulates Aß levels and may play an essential role in AD progression.

Direct label-free imaging of nanodomains in biomimetic and biological membranes by cryogenic electron microscopy.

The nanoscale organization of biological membranes into structurally and compositionally distinct lateral domains is believed to be central to membrane function. The nature of this organization has remained elusive due to a lack of methods to directly probe nanoscopic membrane features. We show here that cryogenic electron microscopy (cryo-EM) can be used to directly image coexisting nanoscopic domains in synthetic and bioderived membranes without extrinsic probes. Analyzing a series of single-component liposomes composed of synthetic lipids of varying chain lengths, we demonstrate that cryo-EM can distinguish bilayer thickness differences as small as 0.5 Å, comparable to the resolution of small-angle scattering methods. Simulated images from computational models reveal that features in cryo-EM images result from a complex interplay between the atomic distribution normal to the plane of the bilayer and imaging parameters. Simulations of phase-separated bilayers were used to predict two sources of contrast between coexisting ordered and disordered phases within a single liposome, namely differences in membrane thickness and molecular density. We observe both sources of contrast in biomimetic membranes composed of saturated lipids, unsaturated lipids, and cholesterol. When extended to isolated mammalian plasma membranes, cryo-EM reveals similar nanoscale lateral heterogeneities. The methods reported here for direct, probe-free imaging of nanodomains in unperturbed membranes open new avenues for investigation of nanoscopic membrane organization.

The role of the Arp2/3 complex in shaping the dynamics and structures of branched actomyosin networks.

Actomyosin networks give cells the ability to move and divide. These networks contract and expand while being driven by active energy-consuming processes such as motor protein walking and actin polymerization. Actin dynamics is also regulated by actin-binding proteins, such as the actin-related protein 2/3 (Arp2/3) complex. This complex generates branched filaments, thereby changing the overall organization of the network. In this work, the spatiotemporal patterns of dynamical actin assembly accompanying the branching-induced reorganization caused by Arp2/3 were studied using a computational model (mechanochemical dynamics of active networks [MEDYAN]); this model simulates actomyosin network dynamics as a result of chemical reactions whose rates are modulated by rapid mechanical equilibration. We show that branched actomyosin networks relax significantly more slowly than do unbranched networks. Also, branched networks undergo rare convulsive movements, "avalanches," that release strain in the network. These avalanches are associated with the more heterogeneous distribution of mechanically linked filaments displayed by branched networks. These far-from-equilibrium events arising from the marginal stability of growing actomyosin networks provide a possible mechanism of the "cytoquakes" recently seen in experiments.

Molecular Dynamics Ensemble Refinement of Intrinsically Disordered Peptides According to Deconvoluted Spectra from Circular Dichroism.

We have developed a computational method of atomistically refining the structural ensemble of intrinsically disordered peptides (IDPs) facilitated by experimental measurements using circular dichroism spectroscopy (CD). A major challenge surrounding this approach stems from the deconvolution of experimental CD spectra into secondary structure features of the IDP ensemble. Currently available algorithms for CD deconvolution were designed to analyze the spectra of proteins with stable secondary structures. Herein, our work aims to minimize any bias from the peptide deconvolution analysis by implementing a non-negative linear least-squares fitting algorithm in conjunction with a CD reference data set that contains soluble and denatured proteins (SDP48). The non-negative linear least-squares method yields the best results for deconvolution of proteins with higher disordered content than currently available methods, according to a validation analysis of a set of protein spectra with Protein Data Bank entries. We subsequently used this analysis to deconvolute our experimental CD data to refine our computational model of the peptide secondary structure ensemble produced by all-atom molecular dynamics simulations with implicit solvent. We applied this approach to determine the ensemble structures of a set of short IDPs, that mimic the calmodulin binding domain of calcium/calmodulin-dependent protein kinase II and its 1-amino-acid and 3-amino-acid mutants. Our study offers a, to our knowledge, novel way to solve the ensemble secondary structures of IDPs in solution, which is important to advance the understanding of their roles in regulating signaling pathways through the formation of complexes with multiple partners.

Loss of PTEN-induced kinase 1 (Pink1) reduces hippocampal tyrosine hydroxylase and impairs learning and memory.

Phosphatase and tensin homolog (PTEN)-induced kinase 1 (Pink1) is involved in mitochondrial quality control, which is essential for maintaining energy production and minimizing oxidative damage from dysfunctional/depolarized mitochondria. Pink1 mutations are the second most common cause of autosomal recessive Parkinson's disease (PD). In addition to characteristic motor impairments, PD patients also commonly exhibit cognitive impairments. As the hippocampus plays a prominent role in cognition, we tested if loss of Pink1 in mice influences learning and memory. While wild-type mice were able to perform a contextual discrimination task, age-matched Pink1 knockout (Pink1-/-) mice showed an impaired ability to differentiate between two similar contexts. Similarly, Pink1-/- mice performed poorly in a delayed alternation task as compared to age-matched controls. Poor performance in these cognitive tasks was not the result of overt hippocampal pathology. However, a significant reduction in hippocampal tyrosine hydroxylase (TH) protein levels was detected in the Pink1-/- mice. This decrease in hippocampal TH levels was also associated with reduced DOPA decarboxylase and dopamine D2 receptor levels, but not post-synaptic dopamine D1 receptor levels. These presynaptic changes appeared to be selective for dopaminergic fibers as hippocampal dopamine beta hydroxylase, choline acetyltransferase, and tryptophan hydroxylase levels were unchanged in Pink1-/- mice. Administration of the dopamine D1 receptor agonist SKF38393 to Pink1-/- mice was found to improve performance in the context discrimination task. Taken together, our results show that Pink1 loss may alter dopamine signaling in the hippocampus, which could be a contributing mechanism for the observed learning and memory impairments.

On the Mechanism of Bilayer Separation by Extrusion, or Why Your LUVs Are Not Really Unilamellar.

Extrusion through porous filters is a widely used method for preparing biomimetic model membranes. Of primary importance in this approach is the efficient production of single bilayer (unilamellar) vesicles that eliminate the influence of interlamellar interactions and strictly define the bilayer surface area available to external reagents such as proteins. Submicroscopic vesicles produced using extrusion are widely assumed to be unilamellar, and large deviations from this assumption would impact interpretations from many model membrane experiments. Using three probe-free methods-small angle X-ray and neutron scattering and cryogenic electron microscopy-we report unambiguous evidence of extensive multilamellarity in extruded vesicles composed of neutral phosphatidylcholine lipids, including for the common case of neutral lipids dispersed in physiological buffer and extruded through 100-nm diameter pores. In such preparations, only ∼35% of lipids are externally accessible and this fraction is highly dependent on preparation conditions. Charged lipids promote unilamellarity as does decreasing solvent ionic strength, indicating the importance of electrostatic interactions in determining the lamellarity of extruded vesicles. Smaller extrusion pore sizes also robustly increase the fraction of unilamellar vesicles, suggesting a role for membrane bending. Taken together, these observations suggest a mechanistic model for extrusion, wherein the formation of unilamellar vesicles involves competition between bilayer bending and adhesion energies. The findings presented here have wide-ranging implications for the design and interpretation of model membrane studies, especially ensemble-averaged observations relying on the assumption of unilamellarity.

Assemblies of calcium/calmodulin-dependent kinase II with actin and their dynamic regulation by calmodulin in dendritic spines.

Calcium/calmodulin-dependent kinase II (CaMKII) plays a key role in the plasticity of dendritic spines. Calcium signals cause calcium-calmodulin to activate CaMKII, which leads to remodeling of the actin filament (F-actin) network in the spine. We elucidate the mechanism of the remodeling by combining computer simulations with protein array experiments and electron microscopic imaging, to arrive at a structural model for the dodecameric complex of CaMKII with F-actin. The binding interface involves multiple domains of CaMKII. This structure explains the architecture of the micrometer-scale CaMKII/F-actin bundles arising from the multivalence of CaMKII. We also show that the regulatory domain of CaMKII may bind either calmodulin or F-actin, but not both. This frustration, along with the multipartite nature of the binding interface, allows calmodulin transiently to strip CaMKII from actin assemblies so that they can reorganize. This observation therefore provides a simple mechanism by which the structural dynamics of CaMKII establishes the link between calcium signaling and the morphological plasticity of dendritic spines.

Visualization of the type III secretion mediated Salmonella-host cell interface using cryo-electron tomography.

Many important gram-negative bacterial pathogens use highly sophisticated type III protein secretion systems (T3SSs) to establish complex host-pathogen interactions. Bacterial-host cell contact triggers the activation of the T3SS and the subsequent insertion of a translocon pore into the target cell membrane, which serves as a conduit for the passage of effector proteins. Therefore the initial interaction between T3SS-bearing bacteria and host cells is the critical step in the deployment of the protein secretion machine, yet this process remains poorly understood. Here, we use high-throughput cryo-electron tomography (cryo-ET) to visualize the T3SS-mediated Salmonella-host cell interface. Our analysis reveals the intact translocon at an unprecedented level of resolution, its deployment in the host cell membrane, and the establishment of an intimate association between the bacteria and the target cells, which is essential for effector translocation. Our studies provide critical data supporting the long postulated direct injection model for effector translocation.