Geometric Cues Research Articles

Deep endpoints focusing network under geometric constraints for end-to-end biometric measurement in fetal ultrasound images

Biometric measurements in fetal ultrasound images are one of the most highly demanding medical image analysis tasks that can directly contribute to diagnosing fetal diseases. However, the natural high-speckle noise and shadows in ultrasound data present big challenges for automatic biometric measurement. Almost all the existing dominant automatic methods are two-stage models, where the key anatomical structures are segmented first and then measured, thus bringing segmentation and fitting errors. What is worse, the results of the second-stage fitting are completely dependent on the good performance of first-stage segmentation, i.e., the segmentation error will lead to a larger fitting error. To this end, we propose a novel end-to-end biometric measurement network, abbreviated as E2EBM-Net, that directly fits the measurement parameters. E2EBM-Net includes a cross-level feature fusion module to extract multi-scale texture information, a hard-soft attention module to improve position sensitivity, and center-focused detectors jointly to achieve accurate localizing and regressing of the measurement endpoints, as well as a loss function with geometric cues to enhance the correlations. To our knowledge, this is the first AI-based application to address the biometric measurement of irregular anatomical structures in fetal ultrasound images with an end-to-end approach. Experiment results showed that E2EBM-Net outperformed the existing methods and achieved the state-of-the-art performance.

PrimitivePose: Generic Model and Representation for 3D Bounding Box Prediction of Unseen Objects

A considerable amount of research is concerned with the challenging task of estimating three-dimensional (3D) pose and size for multi-object indoor scene configurations. Many existing models rely on a priori known object models, such as 3D CAD models and are therefore limited to a predefined set of object categories. This closed-set constraint limits the range of applications for robots interacting in dynamic environments where previously unseen objects may appear. This paper addresses this problem with a highly generic 3D bounding box detection method that relies entirely on geometric cues obtained from depth data percepts. While the generation of synthetic data, e.g. synthetic depth maps, is commonly used for this task, the well-known synth-to-real gap often emerges, which prohibits transition of models trained solely on synthetic data to the real world. To ameliorate this problem, we use stereo depth computation on synthetic data to obtain pseudo-realistic disparity maps. We then propose an intermediate representation, namely disparity-scaled surface normal (SN) images, which encodes geometry and at the same time preserves depth/scale information unlike the commonly used standard SNs. In a series of experiments, we demonstrate the usefulness of our approach, detecting everyday objects on a captured data set of tabletop scenes, and compare it to the popular PoseCNN model. We quantitatively show that standard SNs are less adequate for challenging 3D detection tasks by comparing predictions from the model trained on disparity alone, SNs and disparity-scaled SNs. Additionally, in an ablation study we investigate the minimal number of training samples required for such a learning task. Lastly, we make the tool used for 3D object annotation publicly available at: https://preview.tinyurl.com/3ycn8v5k . A video showcasing our results can be found at: https://preview.tinyurl.com/dzdzabek .

Deep learning-driven diagnosis: A multi-task approach for segmenting stroke and Bell's palsy

Strong efforts have been undertaken to enhance the diagnosis and identification of diseases that cause facial paralysis, such as Bell's palsy and stroke, because of their detrimental social effects. Stroke is one of the most serious and potentially fatal conditions among the major cardiovascular disorders. We are introducing a deep-learning-based method for early diagnosis of facial paralysis diseases such as stroke and Bell's palsy. Recognizing the costs associated with traditional diagnostic techniques like magnetic resonance tomography (MRI) and computed tomography (CT) scan images, our model employs a multi-task network, integrating face parsing, facial asymmetry parsing, and category enhancement. Spatial inconsistencies are addressed via a depth-map estimation module that leverages an instance-specific kernel approach. To clarify the boundaries of facial components, we use category edge detection with a foreground attention module, generating generic geometric structures and detailed semantic cues. Our model is trained on two datasets, comprising individuals with regular smiles and those with one-sided facial weakness. This cost-effective, easily accessible solution can streamline the diagnostic process, minimizing data gaps, and reducing needless rescreening and intervention costs.

Fibre-infused gel scaffolds guide cardiomyocyte alignment in 3D-printed ventricles.

Hydrogels are attractive materials for tissue engineering, but efforts to date have shown limited ability to produce the microstructural features necessary to promote cellular self-organization into hierarchical three-dimensional (3D) organ models. Here we develop a hydrogel ink containing prefabricated gelatin fibres to print 3D organ-level scaffolds that recapitulate the intra- and intercellular organization of the heart. The addition of prefabricated gelatin fibres to hydrogels enables the tailoring of the ink rheology, allowing for a controlled sol-gel transition to achieve precise printing of free-standing 3D structures without additional supporting materials. Shear-induced alignment of fibres during ink extrusion provides microscale geometric cues that promote the self-organization of cultured human cardiomyocytes into anisotropic muscular tissues in vitro. The resulting 3D-printed ventricle in vitro model exhibited biomimetic anisotropic electrophysiological and contractile properties.

Length limitation of astral microtubules orients cell divisions in murine intestinal crypts

Planar spindle orientation is critical for epithelial tissue organization and is generally instructed by the long cell-shape axis or cortical polarity domains. We introduced mouse intestinal organoids in order to study spindle orientation in a monolayered mammalian epithelium. Although spindles were planar, mitotic cells remained elongated along the apico-basal (A-B) axis, and polarity complexes were segregated to basal poles, so that spindles oriented in an unconventional manner, orthogonal to both polarity and geometric cues. Using high-resolution 3D imaging, simulations, and cell-shape and cytoskeleton manipulations, we show that planar divisions resulted from a length limitation in astral microtubules (MTs) which precludes them from interacting with basal polarity, and orient spindles from the local geometry of apical domains. Accordingly, lengthening MTs affected spindle planarity, cell positioning, and crypt arrangement. We conclude that MT length regulation may serve as a key mechanism for spindles to sense local cell shapes and tissue forces to preserve mammalian epithelial architecture.

Alignment behavior of nerve, vascular, muscle, and intestine cells in two- and three-dimensional strategies.

By harnessing structural hierarchical insights, plausibly simulate better ones imagination to figure out the best choice of methods for reaching out the unprecedented developments of the tissue engineering products as a next level. Constructing a functional tissue that incorporates two-dimensional (2D) or higher dimensions requires overcoming technological or biological limitations in order to orchestrate the structural compilation of one-dimensional and 2D sheets (microstructures) simultaneously (in situ). This approach enables the creation of a layered structure that can be referred to as an ensemble of layers or, after several days of maturation, a direct or indirect joining of layers. Here, we have avoided providing a detailed methodological description of three-dimensional and 2D strategies, except for a few interesting examples that highlight the higher alignment of cells and emphasize rarely remembered facts associated with vascular, peripheral nerve, muscle, and intestine tissues. The effective directionality of cells in conjunction with geometric cues (in the range of micrometers) is well known to affect a variety of cell behaviors. The curvature of a cell's environment is one of the factors that influence the formation of patterns within tissues. The text will cover cell types containing some level of stemness, which will be followed by their consequences for tissue formation. Other important considerations pertain to cytoskeleton traction forces, cell organelle positioning, and cell migration. An overview of cell alignment along with several pivotal molecular and cellular level concepts, such as mechanotransduction, chirality, and curvature of structure effects on cell alignments will be presented. The mechanotransduction term will be used here in the context of the sensing capability that cells show as a result of force-induced changes either at the conformational or the organizational levels, a capability that allows us to modify cell fate by triggering downstream signaling pathways. A discussion of the cells' cytoskeleton and of the stress fibers involvement in altering the cell's circumferential constitution behavior (alignment) based on exposed scaffold radius will be provided. Curvatures with size similarities in the range of cell sizes cause the cell's behavior to act as if it was in an in vivo tissue environment. The revision of the literature, patents, and clinical trials performed for the present study shows that there is a clear need for translational research through the implementation of clinical trial platforms that address the tissue engineering possibilities raised in the current revision. This article is categorized under: Infectious Diseases > Biomedical Engineering Neurological Diseases > Biomedical Engineering Cardiovascular Diseases > Biomedical Engineering.

Polarimetric Multi-View Inverse Rendering.

A polarization camera has great potential for 3D reconstruction since the angle of polarization (AoP) and the degree of polarization (DoP) of reflected light are related to an object's surface normal. In this paper, we propose a novel 3D reconstruction method called Polarimetric Multi-View Inverse Rendering (Polarimetric MVIR) that effectively exploits geometric, photometric, and polarimetric cues extracted from input multi-view color-polarization images. We first estimate camera poses and an initial 3D model by geometric reconstruction with a standard structure-from-motion and multi-view stereo pipeline. We then refine the initial model by optimizing photometric rendering errors and polarimetric errors using multi-view RGB, AoP, and DoP images, where we propose a novel polarimetric cost function that enables an effective constraint on the estimated surface normal of each vertex, while considering four possible ambiguous azimuth angles revealed from the AoP measurement. The weight for the polarimetric cost is effectively determined based on the DoP measurement, which is regarded as the reliability of polarimetric information. Experimental results using both synthetic and real data demonstrate that our Polarimetric MVIR can reconstruct a detailed 3D shape without assuming a specific surface material and lighting condition.

ROIFormer: Semantic-Aware Region of Interest Transformer for Efficient Self-Supervised Monocular Depth Estimation

The exploration of mutual-benefit cross-domains has shown great potential toward accurate self-supervised depth estimation. In this work, we revisit feature fusion between depth and semantic information and propose an efficient local adaptive attention method for geometric aware representation enhancement. Instead of building global connections or deforming attention across the feature space without restraint, we bound the spatial interaction within a learnable region of interest. In particular, we leverage geometric cues from semantic information to learn local adaptive bounding boxes to guide unsupervised feature aggregation. The local areas preclude most irrelevant reference points from attention space, yielding more selective feature learning and faster convergence. We naturally extend the paradigm into a multi-head and hierarchic way to enable the information distillation in different semantic levels and improve the feature discriminative ability for fine-grained depth estimation. Extensive experiments on the KITTI dataset show that our proposed method establishes a new state-of-the-art in self-supervised monocular depth estimation task, demonstrating the effectiveness of our approach over former Transformer variants.

Biased Random Walk Model of Neuronal Dynamics on Substrates with Periodic Geometrical Patterns

Neuronal networks are complex systems of interconnected neurons responsible for transmitting and processing information throughout the nervous system. The building blocks of neuronal networks consist of individual neurons, specialized cells that receive, process, and transmit electrical and chemical signals throughout the body. The formation of neuronal networks in the developing nervous system is a process of fundamental importance for understanding brain activity, including perception, memory, and cognition. To form networks, neuronal cells extend long processes called axons, which navigate toward other target neurons guided by both intrinsic and extrinsic factors, including genetic programming, chemical signaling, intercellular interactions, and mechanical and geometrical cues. Despite important recent advances, the basic mechanisms underlying collective neuron behavior and the formation of functional neuronal networks are not entirely understood. In this paper, we present a combined experimental and theoretical analysis of neuronal growth on surfaces with micropatterned periodic geometrical features. We demonstrate that the extension of axons on these surfaces is described by a biased random walk model, in which the surface geometry imparts a constant drift term to the axon, and the stochastic cues produce a random walk around the average growth direction. We show that the model predicts key parameters that describe axonal dynamics: diffusion (cell motility) coefficient, average growth velocity, and axonal mean squared length, and we compare these parameters with the results of experimental measurements. Our findings indicate that neuronal growth is governed by a contact-guidance mechanism, in which the axons respond to external geometrical cues by aligning their motion along the surface micropatterns. These results have a significant impact on developing novel neural network models, as well as biomimetic substrates, to stimulate nerve regeneration and repair after injury.

Early committed polarization of intracellular tension in response to cell shape determines the osteogenic differentiation of mesenchymal stromal cells.

Within the heterogeneous tissue architecture, a comprehensive understanding of how cell shapes regulate cytoskeletal mechanics by adjusting focal adhesions (FAs) signals to correlate with the lineage commitment of mesenchymal stromal cells (MSCs) remains obscure. Here, via engineered extracellular matrices, we observed that the development of mature FAs, coupled with a symmetrical pattern of radial fiber bundles, appeared at the right-angle vertices in cells with square shape. While circular cells aligned the transverse fibers parallel to the cell edge, and moved them centripetally in a counter-clockwise direction, symmetrical bundles of radial fibers at the vertices of square cells disrupted the counter-clockwise swirling and bridged the transverse fibers to move centripetally. In square cells, the contractile force, generated by the myosin IIA-enriched transverse fibers, were concentrated and transmitted outwards along the symmetrical bundles of radial fibers, to the extracellular matrix through FAs, and thereby driving FA organization and maturation. The symmetrical radial fiber bundles concentrated the transverse fibers contractility inward to the linkage between the actin cytoskeleton and the nuclear envelope. The tauter cytoskeletal network adjusted the nuclear-actomyosin force balance to cause nuclear deformability and to increase nuclear translocation of the transcription co-activator YAP, which in turn modulated the switch in MSC commitment. Thus, FAs dynamically respond to geometric cues and remodel actin cytoskeletal network to re-distribute intracelluar tension towards the cell nucleus, and thereby controlling YAP mechanotransduction signaling in regulating MSC fate decision. STATEMENT OF SIGNIFICANCE: We decipher how cellular mechanics is self-organized depending on extracellular geometric features to correlate with mesenchymal stromal cell lineage commitment. In response to geometry constrains on cell morphology, symmetrical radial fiber bundles are assembled and clustered depending on the maturation state of focal adhesions and bridge with the transverse fibers, and thereby establishing the dynamic cytoskeletal network. Contractile force, generated by the myosin-IIA-enriched transverse fibers, is transmitted and dynamically drives the retrograde movement of the actin cytoskeletal network, which appropriately adjusts the nuclear-actomyosin force balance and deforms the cell nucleus for YAP mechano-transduction signaling in regulating mesenchymal stromal cell fate decision.

Bone-on-a-Chip: Biomimetic Models Based on Microfluidic Technologies for Biomedical Applications.

With the increasing importance of preclinical evaluation of newly developed drugs or treatments, in vitro organ or disease models are necessary. Although various organ-specific on-chip (organ-on-a-chip, or OOC) systems have been developed as emerging in vitro models, bone-on-a-chip (BOC) systems that recapitulate the bone microenvironment have been less developed or reviewed compared with other OOCs. The bone is one of the most dynamic organs and undergoes continuous remodeling throughout its lifetime. The aging population is growing worldwide, and healthcare costs are rising rapidly. Since in vitro BOC models that recapitulate native bone niches and pathological features can be important for studying the underlying mechanism of orthopedic diseases and predicting drug responses in preclinical trials instead of in animals, the development of biomimetic BOCs with high efficiency and fidelity will be accelerated further. Here, we review recently engineered BOCs developed using various microfluidic technologies and investigate their use to model the bone microenvironment. We have also explored various biomimetic strategies based on biological, geometrical, and biomechanical cues for biomedical applications of BOCs. Finally, we addressed the limitations and challenging issues of current BOCs that should be overcome to obtain more acceptable BOCs in the biomedical and pharmaceutical industries.

Gaussian similarity-based adaptive dynamic label assignment for tiny object detection

Benefiting from the advanced deep learning techniques, significant achievements have been made in generic object detection. Tiny object detection (TOD) is a challenging task in computer vision due to the low resolution, insufficient geometric cues, and high noise levels. A recent trend for detectors is introducing more granular label assignment strategies to provide promising supervision information for classification and regression. However, most previous Intersection-Over-Union (IoU) based methods suffer from two main drawbacks, including (1) low tolerance of IoU for bounding box deviations in tiny objects and (2) deficient guidance for optimization caused by inter-sample and intra-sample imbalance. We propose two novel components to address these problems: the Gaussian probabilistic distribution-based fuzzy similarity metric (GPM) and the adaptive dynamic anchor mining strategy (ADAS). GPM aims to address the issue of inaccurate similarity measurement between small bounding boxes and pre-defined anchors, providing a more accurate basis for label assignment. ADAS adopts a dynamically adjusted strategy for label assignment to address the distribution bias between positive and negative samples, ensuring that the label assignment is consistent with the distribution of objects in the image. Extensive experiments are conducted on AI-TODv2 and other tiny object detection datasets to evaluate the proposed ADAS-GPM method’s performance. The results demonstrate that incorporating ADAS-GPM into an anchor-based object detector yields significant outperformance over state-of-the-art methods on the challenging AI-TODv2 benchmark. The proposed ADAS-GPM method exhibits good results, clearly demonstrating its validity and potential.

When landmarks are not enough

Including geometric spatial cues in an environment can help reverse the difficulties with spatial navigation experienced by children and older adults.

Abstract 4303: High-throughput 3D tumoroid models for immunotherapy and drug discovery

Abstract The tumor microenvironment (TME) is a complex 3D cellular system comprised of a diverse amalgam of cells. To characterize the TME, we developed a 3D cell culture platform which facilitates imaging-based analysis and cell viability for extended durations. We used this platform to collect in-situ spatiotemporal measurements of cytokine concentrations in the tumor vicinity, track cellular activity, and model local tumor invasion of glioblastoma (GBM) using fast-scanning confocal microscopy. The mechanism by which we achieve long-term 3D culture is perfusion, modeled as flow through a porous medium. Perfusion facilitates regulated transport of nutrients and waste within the soft microgel medium: Liquid-Like Solids (LLS). The interstitial space is tuned to mimic a capillary bed, and surface bioconjugation of the microgels promotes cellular adhesion and migration. To measure cytokine concentrations, GBM tumoroids were grown in LLS and printed into a dispersion of ELISA beads in LLS. The mechanical stability of LLS ensures the tumor and ELISA beads remain stationary without impeding cellular activity. Cytokine on and off-rates were referenced alongside measured bead fluorescence intensities and positions to fit spatiotemporal reaction-diffusion models. Fast-scanning confocal microscopy facilitated in-situ observation of the evolutionary dynamics of tumor progression. Co-culture of patient-derived tumor explants and autologous PBMCs printed into type I collagen-bioconjugated LLS enabled studies of cancer-immune interactions. In-situ cytokine measurements revealed local IL-8 concentrations reached a maximum value of 2 ng ml−1 after 10 hours. A cellular production rate was estimated at 2 molecules cell−1 s−1. Invasive behavior into the proximal space was determined to be super-diffusive; off-lattice agent-based simulations indicated this behavior is a result of the confinement of invasive fronts to the microgel interstitium. The invading glioblastoma cells used anchorage-dependent migration and were guided by geometric cues to traverse the porous bioconjugated LLS network. Cancer-immune interaction studies revealed average CD8+ speeds greater than 2.8 µm min−1 and both chemotaxis and chemokinetic behavior. CD8+ T-cell killing rates were estimated at approximately 3 cancer cells hr−1 initially, monotonically falling over 12 hours to roughly 1 cell hr−1. The development of the in-situ 3D ELISA assay and imaging-based analysis techniques have enabled the tracking of tumor-immune cell interactions, observation of dynamic tumor progression, and local cytokine concentration profiling at appreciable spatiotemporal resolutions. The combination of a physiologically relevant 3D culture platform with the capacity for in-situ qualitative and quantitative observation may lead to new, powerful, preclinical models that allow for interrogation of the TME and decrease the rate of false discovery. Citation Format: Duy Nguyen, Alexander McGhee, Diego Pedro, Alfonso Pepe, Matthew Schaller, Ryan Smolchek, Jack Famiglietti, Stephanie Warrington, W. Gregory Sawyer. High-throughput 3D tumoroid models for immunotherapy and drug discovery. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4303.

Fabrication of Scratched Nanogrooves for Highly Oriented Cell Alignment and Application as a Wound Healing Dressing.

Using improper wound care materials may cause impaired wound healing, which can involve scar formation and infection. Herein, we propose a facile method to fabricate a cell-alignment scaffold, which can effectively enhance cell growth and migration, leading to the reproduction of cellular arrangements and restoration of tissues. The principle is scratching a diamond lapping film that gives uniaxial nanotopography on substrates. Cells are seeded to follow the geometric cue via contact guidance, resulting in highly oriented cell alignment. Remarkable biocompatibility is also demonstrated by the high cell viability on various substrates. In vivo studies in a wound healing model in mice show that the scratched film supports directed cell guidance on the nanostructure, with significantly reduced wound areas and inhibition of excessive collagen deposition. Rapid recovery of the epidermis and dermis is also shown by histological analyses, suggesting the potential application of the scratching technique as an advanced wound dressing material for effective tissue regeneration.

Real-Time Globally Consistent 3D Reconstruction with Semantic Priors.

Maintaining global consistency continues to be critical for online 3D indoor scene reconstruction. However, it is still challenging to generate satisfactory 3D reconstruction in terms of global consistency for previous approaches using purely geometric analysis, even with bundle adjustment or loop closure techniques. In this paper, we propose a novel real-time 3D reconstruction approach which effectively integrates both semantic and geometric cues. The key challenge is how to map this indicative information, i.e. semantic priors, into a metric space as measurable information, thus enabling more accurate semantic fusion leveraging both the geometric and semantic cues. To this end, we introduce a semantic space with a continuous metric function measuring the distance between discrete semantic observations. Within the semantic space, we present an accurate frame-to-model semantic tracker for camera pose estimation, and semantic pose graph equipped with semantic links between submaps for globally consistent 3D scene reconstruction. With extensive evaluation on public synthetic and real-world 3D indoor scene RGB-D datasets, we show that our approach outperforms the previous approaches for 3D scene reconstruction both quantitatively and qualitatively, especially in terms of global consistency.

Landmark-based spatial navigation across the human lifespan.

Human spatial cognition has been mainly characterized in terms of egocentric (body-centered) and allocentric (world-centered) wayfinding bhavior. It was hypothesized that allocentric spatial coding, as a special high-level cognitive ability, develops later and deteriorates earlier than the egocentric one throughout lifetime. We challenged this hypothesis by testing the use of landmarks versus geometric cues in a cohort of 96 deeply phenotyped participants, who physically navigated an equiangular Y maze, surrounded by landmarks or an anisotropic one. The results show that an apparent allocentric deficit in children and aged navigators is caused specifically by difficulties in using landmarks for navigation while introducing a geometric polarization of space made these participants as efficient allocentric navigators as young adults. This finding suggests that allocentric behavior relies on two dissociable sensory processing systems that are differentially affected by human aging. Whereas landmark processing follows an inverted-U dependence on age, spatial geometry processing is conserved, highlighting its potential in improving navigation performance across the lifespan.

Abstract P5-11-01: Epithelial-Mesenchymal Plasticity is Regulated by Inflammatory Signaling Networks Coupled to Cell Morphology

Abstract Purpose: During development and homeostatic processes such as wound repair, certain cells undergo a remarkable process where they radically transform in cell shape and state, from epithelial to mesenchymal cells. This ability is referred to as ‘Epithelial-Mesenchymal Plasticity’ (EMP), triggered by both mechanical (i.e. loss of cell-cell contact) and soluble cues (i.e. TGFβ), and is absolutely essential in both embryonic and adult organisms. Dysregulation of EMP also occurs in cancer; where tumor cells undergo EMP to become metastatic, stem-like, and drug resistant. Critically, increased EMP correlates with increased cancer severity. However, it is largely a black box as to how EMP is regulated and how epithelial cells sense physical, geometrical, and soluble cues in their environment to assume a mesenchymal fate. This work inferred that cell shape is a determinant of not only fates – but of long-term outcomes. That is, we provide mechanistic explanations between cell context, environment, and cancer severity. In this work, we investigated how mechanical and soluble cues are coupled to the dynamics of signaling pathways that regulate transcriptional and post-transcriptional events that underpin EMP. Especially with regards to mechanical cues, we attempt to unlock the ‘black box’ as to how changes in adhesion, ECM (Extracellular Matrix) stiffness, and environment geometry are coupled to the transcriptional events that drive EMP. Results: We show that changes in cell and nuclear shape result from the actions of the cytoskeleton and important drivers of EMP in upregulating ‘interlocking’ networks that promote EMP-driving inflammation and suppressing insulin signaling. Using a combination of cell biology, proteomics, and new statistical methods, we provide a systems biology model demonstrating: Cell shape → MT bound Kinesin-1 activity and nuclear shape → inflammation (IKK, JNK), insulin signaling (IRS), and YAP/TAZ → EMP Our work connects observable changes in phenotype to causal changes in signaling network architecture and cell fates. We used an integrative -omic approach to analyze tumors from breast cancer patients. We identified a novel tumor suppressor – JAM3 – whose loss is associated with altered nuclear shape in vivo, inhibiting JAM3 in cells, or stimulation with canonical EMP inducer TGFβ to promote EMP, and thus changes the organization of microtubules and alters nuclear shape. During EMP we observe there is upregulation of a pro-inflammatory, insulin resistant, signaling network that is predictive of mesenchymal states across cancer. EMP following JAM3 depletion and/or TGFβ stimulation is rescued by inhibition of Kinesin-1 motors. This rescue is explained by changes in inflammatory and insulin signaling. We show that while Kinesin-1 activity is responsible for upregulation in canonical signalling and network ‘hubs’, changes in nuclear shape upregulate ‘effectors’ of these hubs. Thus, microtubules and nuclei differentially regulate different parts of ‘interlocking’ networks. Conclusions: This work has integrated image-omics, comprehensive global proteomics, and quantitative cell biology to provide a mechanistic ¬and systems-level understanding of how epithelial cells differentiate into mesenchymal forms during disease development and progression. This work is of major significance for three reasons. First, it shows how cell shape can mechanistically regulate cell fates on an unprecedented systems-level. Second, we identify an EMP network that is conserved across cancers and may indeed be conserved across both normal and diseased mesenchymal cells. Indeed, we speculate that different types of diseased cells may all share the same network. Finally, we introduce the concept of interlocking networks – where hubs and effectors are regulated by different cellular components. Our work has been extensively validated, using chemical and genetic approaches and in vivo model of in human breast cancer. Citation Format: Zheng Yin. Epithelial-Mesenchymal Plasticity is Regulated by Inflammatory Signaling Networks Coupled to Cell Morphology [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-11-01.

Bioconjugation of COL1 protein on liquid-like solid surfaces to study tumor invasion dynamics.

Tumor invasion is likely driven by the product of intrinsic and extrinsic stresses, reduced intercellular adhesion, and reciprocal interactions between the cancer cells and the extracellular matrix (ECM). The ECM is a dynamic material system that is continuously evolving with the tumor microenvironment. Although it is widely reported that cancer cells degrade the ECM to create paths for migration using membrane-bound and soluble enzymes, other nonenzymatic mechanisms of invasion are less studied and not clearly understood. To explore tumor invasion that is independent of enzymatic degradation, we have created an open three-dimensional (3D) microchannel network using a novel bioconjugated liquid-like solid (LLS) medium to mimic both the tortuosity and the permeability of a loose capillary-like network. The LLS is made from an ensemble of soft granular microgels, which provides an accessible platform to investigate the 3D invasion of glioblastoma (GBM) tumor spheroids using in situ scanning confocal microscopy. The surface conjugation of the LLS microgels with type 1 collagen (COL1-LLS) enables cell adhesion and migration. In this model, invasive fronts of the GBM microtumor protruded into the proximal interstitial space and may have locally reorganized the surrounding COL1-LLS. Characterization of the invasive paths revealed a super-diffusive behavior of these fronts. Numerical simulations suggest that the interstitial space guided tumor invasion by restricting available paths, and this physical restriction is responsible for the super-diffusive behavior. This study also presents evidence that cancer cells utilize anchorage-dependent migration to explore their surroundings, and geometrical cues guide 3D tumor invasion along the accessible paths independent of proteolytic ability.

Gradient Nanoconfinement Facilitates Binding of Transcriptional Factor NF-κB to Histone- and Protamine-DNA Complexes.

Mechanically induced chromosome reorganization plays important roles in transcriptional regulation. However, the interplay between chromosome reorganization and transcription activities is complicated, such that it is difficult to decipher the regulatory effects of intranuclear geometrical cues. Here, we simplify the system by introducing DNA, packaging proteins (i.e., histone and protamine), and transcription factor NF-κB into a well-defined fluidic chip with changing spatical confinement ranging from 100 to 500 nm. It is uncovered that strong nanoconfinement suppresses higher-order folding of histone- and protamine-DNA complexes, the fracture of which exposes buried DNA segments and causes increased quantities of NF-κB binding to the DNA chain. Overall, these results reveal a pathway of how intranuclear geometrical cues alter the open/closed state of a DNA-protein complex and therefore affect transcription activities: i.e., NF-κB binding.