Deep learning with digital holographic microscopy discriminates apoptosis and necroptosis

Transformer-based spatial-temporal detection of apoptotic cell death in live-cell imaging

Intravital microscopy has revolutionized live-cell imaging by allowing the study of spatial-temporal cell dynamics in living animals. However, the complexity of the data generated by this technology has limited the development of effective computational tools to identify and quantify cell processes. Amongst them, apoptosis is a crucial form of regulated cell death involved in tissue homeostasis and host defense. Live-cell imaging enabled the study of apoptosis at the cellular level, enhancing our understanding of its spatial-temporal regulation. However, at present, no computational method can deliver robust detection of apoptosis in microscopy timelapses. To overcome this limitation, we developed ADeS, a deep learning-based apoptosis detection system that employs the principle of activity recognition. We trained ADeS on extensive datasets containing more than 10,000 apoptotic instances collected both in vitro and in vivo, achieving a classification accuracy above 98% and outperforming state-of-the-art solutions. ADeS is the first method capable of detecting the location and duration of multiple apoptotic events in full microscopy timelapses, surpassing human performance in the same task. We demonstrated the effectiveness and robustness of ADeS across various imaging modalities, cell types, and staining techniques. Finally, we employed ADeS to quantify cell survival in vitro and tissue damage in mice, demonstrating its potential application in toxicity assays, treatment evaluation, and inflammatory dynamics. Our findings suggest that ADeS is a valuable tool for the accurate detection and quantification of apoptosis in live-cell imaging and, in particular, intravital microscopy data, providing insights into the complex spatial-temporal regulation of this process.

Toward a predictive understanding of epithelial cell death

Epithelial cell death is highly prevalent during development and tissue homeostasis. While we have a rather good understanding of the molecular regulators of programmed cell death, especially for apoptosis, we still fail to predict when, where, how many and which specific cells will die in a tissue. This likely relies on the much more complex picture of apoptosis regulation in a tissular and epithelial context, which entails cell autonomous but also non-cell autonomous factors, diverse feedback and multiple layers of regulation of the commitment to apoptosis. In this review, we illustrate this complexity of epithelial apoptosis regulation by describing these different layers of control, all demonstrating that local cell death probability is a complex emerging feature. We first focus on non-cell autonomous factors that can locally modulate the rate of cell death, including cell competition, mechanical input and geometry as well as systemic effects. We then describe the multiple feedback mechanisms generated by cell death itself. We also outline the multiple layers of regulation of epithelial cell death, including the coordination of extrusion and regulation occurring downstream of effector caspases. Eventually, we propose a roadmap to reach a more predictive understanding of cell death regulation in an epithelial context.

Current trends in luminescence-based assessment of apoptosis

Apoptosis, the most extensively studied type of cell death, is known to play a crucial role in numerous processes such as elimination of unwanted cells or cellular debris, growth, control of the immune system, and prevention of malignancies. Defective regulation of apoptosis can trigger various diseases and disorders including cancer, neurological conditions, autoimmune diseases and developmental disorders. Knowing the nuances of the cell death type induced by a compound can help decipher which therapy is more effective for specific diseases. The detection of apoptotic cells using classic methods has brought significant contribution over the years, but innovative methods are quickly emerging and allow more in-depth understanding of the mechanisms, aside from a simple quantification. Due to increased sensitivity, time efficiency, pathway specificity and negligible cytotoxicity, these innovative approaches have great potential for both in vitro and in vivo studies. This review aims to shed light on the importance of developing and using novel nanoscale methods as an alternative to the classic apoptosis detection techniques.

Quantitative phase imaging for characterization of single cell growth dynamics

Quantitative phase imaging (QPI) has emerged as an indispensable tool in the field of biomedicine, offering the ability to obtain quantitative maps of phase changes due to optical path length delays without the need for contrast agents. These maps provide valuable information about cellular morphology and dynamics, unperturbed by the introduction of exogenous substances. In this review, a summary of recent studies that have focused on elucidating the growth dynamics of individual cells using QPI is presented. Specifically, investigations into cellular changes occurring during mitosis, the differentiation of cellular organelles, the assessment of distinct cell death processes (i.e., apoptosis, necrosis, and oncosis) and the precise measurement of live cell temperature are explored. Furthermore, the captivating applications of QPI in theragnostics, where its potential for transformative impact is prominently showcased, are highlighted. Finally, the challenges that need to be overcome for its wider adoption and successful integration into biomedical research are outlined.

CellDeathPred: a deep learning framework for ferroptosis and apoptosis prediction based on cell painting

Cell death, such as apoptosis and ferroptosis, play essential roles in the process of development, homeostasis, and pathogenesis of acute and chronic diseases. The increasing number of studies investigating cell death types in various diseases, particularly cancer and degenerative diseases, has raised hopes for their modulation in disease therapies. However, identifying the presence of a particular cell death type is not an obvious task, as it requires computationally intensive work and costly experimental assays. To address this challenge, we present CellDeathPred, a novel deep-learning framework that uses high-content imaging based on cell painting to distinguish cells undergoing ferroptosis or apoptosis from healthy cells. In particular, we incorporate a deep neural network that effectively embeds microscopic images into a representative and discriminative latent space, classifies the learned embedding into cell death modalities, and optimizes the whole learning using the supervised contrastive loss function. We assessed the efficacy of the proposed framework using cell painting microscopy data sets from human HT-1080 cells, where multiple inducers of ferroptosis and apoptosis were used to trigger cell death. Our model confidently separates ferroptotic and apoptotic cells from healthy controls, with an average accuracy of 95% on non-confocal data sets, supporting the capacity of the CellDeathPred framework for cell death discovery.

Reflectional quantitative differential phase microscopy using polarized wavefront phase modulation

Quantitative phase microscopy (QPM), as a label-free and nondestructive technique, has been playing an indispensable tool in biomedical imaging and industrial inspection. Herein, we introduce a reflectional quantitative differential phase microscopy (termed RQDPM) based on polarized wavefront phase modulation and partially coherent full-aperture illumination, which has high spatial resolution and spatio-temporal phase sensitivity and is applicable to opaque surfaces and turbid biological specimens. RQDPM does not require additional polarized devices and can be easily switched from reflectional mode to transmission mode. In addition, RQDPM inherits the characteristic of high axial resolution of differential interference contrast microscope, thereby providing topography for opaque surfaces. We experimentally demonstrate the reflectional phase imaging ability of RQDPM with several samples: semiconductor wafer, thick biological tissues, red blood cells, and Hela cells. Furthermore, we dynamically monitor the flow state of microspheres in a self-built microfluidic channel by using RQDPM converted into the transmission mode.

Application of Digital Holographic Microscopy to Analyze Changes in T-Cell Morphology in Response to Bacterial Challenge

Quantitative phase imaging (QPI) is a non-invasive, label-free technique used to detect aberrant cell morphologies caused by disease, thus providing a useful diagnostic approach. Here, we evaluated the potential of QPI to differentiate specific morphological changes in human primary T-cells exposed to various bacterial species and strains. Cells were challenged with sterile bacterial determinants, i.e., membrane vesicles or culture supernatants, derived from different Gram-positive and Gram-negative bacteria. Timelapse QPI by digital holographic microscopy (DHM) was applied to capture changes in T-cell morphology over time. After numerical reconstruction and image segmentation, we calculated single cell area, circularity and mean phase contrast. Upon bacterial challenge, T-cells underwent rapid morphological changes such as cell shrinkage, alterations of mean phase contrast and loss of cell integrity. Time course and intensity of this response varied between both different species and strains. The strongest effect was observed for treatment with S. aureus-derived culture supernatants that led to complete lysis of the cells. Furthermore, cell shrinkage and loss of circular shape was stronger in Gram-negative than in Gram-positive bacteria. Additionally, T-cell response to bacterial virulence factors was concentration-dependent, as decreases in cellular area and circularity were enhanced with increasing concentrations of bacterial determinants. Our findings clearly indicate that T-cell response to bacterial stress depends on the causative pathogen, and specific morphological alterations can be detected using DHM.

Machine learning-based detection of label-free cancer stem-like cell fate

The detection of cancer stem-like cells (CSCs) is mainly based on molecular markers or functional tests giving a posteriori results. Therefore label-free and real-time detection of single CSCs remains a difficult challenge. The recent development of microfluidics has made it possible to perform high-throughput single cell imaging under controlled conditions and geometries. Such a throughput requires adapted image analysis pipelines while providing the necessary amount of data for the development of machine-learning algorithms. In this paper, we provide a data-driven study to assess the complexity of brightfield time-lapses to monitor the fate of isolated cancer stem-like cells in non-adherent conditions. We combined for the first time individual cell fate and cell state temporality analysis in a unique algorithm. We show that with our experimental system and on two different primary cell lines our optimized deep learning based algorithm outperforms classical computer vision and shallow learning-based algorithms in terms of accuracy while being faster than cutting-edge convolutional neural network (CNNs). With this study, we show that tailoring our deep learning-based algorithm to the image analysis problem yields better results than pre-trained models. As a result, such a rapid and accurate CNN is compatible with the rise of high-throughput data generation and opens the door to on-the-fly CSC fate analysis.

Morphological signatures of actin organization in single cells accurately classify genetic perturbations using CNNs with transfer learning

The actin cytoskeleton plays essential roles in countless cell processes, from cell division to migration to signaling. In cancer cells, cytoskeletal dynamics, cytoskeletal filament organization, and overall cell morphology are known to be altered substantially. We hypothesize that actin fiber organization and cell shape may carry specific signatures of genetic or signaling perturbations. We used convolutional neural networks (CNNs) on a small fluorescence microscopy image dataset of retinal pigment epithelial (RPE) cells and triple-negative breast cancer (TNBC) cells for identifying morphological signatures in cancer cells. Using a transfer learning approach, CNNs could be trained to accurately distinguish between normal and oncogenically transformed RPE cells with an accuracy of about 95% or better at the single cell level. Furthermore, CNNs could distinguish transformed cell lines differing by an oncogenic mutation from each other and could also detect knockdown of cofilin in TNBC cells, indicating that each single oncogenic mutation or cytoskeletal perturbation produces a unique signature in actin morphology. Application of the Local Interpretable Model-Agnostic Explanations (LIME) method for visually interpreting the CNN results revealed features of the global actin structure relevant for some cells and classification tasks. Interestingly, many of these features were supported by previous biological observation. Actin fiber organization is thus a sensitive marker for cell identity, and identification of its perturbations could be very useful for assaying cell phenotypes, including disease states.

Real-time 3D imaging of ocean algae with crosstalk suppressed single-shot digital holographic microscopy

Digital holographic microscopy (DHM) has the potential to reconstruct the 3D shape of volumetric samples from a single-shot hologram in a label-free and noninvasive manner. However, the holographic reconstruction is significantly compromised by the out-of-focus image resulting from the crosstalk between refocused planes, leading to the low fidelity of the results. In this paper, we propose a crosstalk suppression algorithm-assisted 3D imaging method combined with a home built DHM system to achieve accurate 3D imaging of ocean algae using only a single hologram. As a key step in the algorithm, a hybrid edge detection strategy using gradient-based and deep learning-based methods is proposed to offer accurate boundary information for the downstream processing. With this information, the crosstalk of each refocused plane can be estimated with adjacent refocused planes. Empowered by this method, we demonstrated successful 3D imaging of six kinds of ocean algae that agree well with the ground truth; we further demonstrated that this method could achieve real-time 3D imaging of the quick swimming ocean algae in the water environment. To our knowledge, this is the first time single-shot DHM is reported in 3D imaging of ocean algae, paving the way for on-site monitoring of the ocean algae.

Panax notoginseng Saponins Protect Brain Microvascular Endothelial Cells against Oxygen-Glucose Deprivation/Resupply-Induced Necroptosis via Suppression of RIP1-RIP3-MLKL Signaling Pathway

Recently, necroptosis has emerged as one of the important mechanisms of ischemia stroke. Necroptosis can be rapidly activated in endothelial cells to cause vascular damage and neuroinflammation. Panax notoginseng saponins (PNS), an ingredient extracted from the root of Panax notoginseng (Burk.) F.H. Chen, was commonly used for ischemic stroke, while its molecular mechanism and targets have not been fully clarified. Our study aimed to clarify the anti-necroptosis effect of PNS by regulating RIP1-RIP3-MLKL signaling pathway in brain microvascular endothelial cells (BMECs) subjected to transient oxygen-glucose deprivation (OGD/resupply [R]). In vitro, the necroptosis model of rat BMECs was established by testing the effect of OGD/R in the presence of the pan-caspase inhibitor z-VAD-FMK. After administration of PNS and Nec-1, cell viability, cell death modality, the expression of RIP1-RIP3-MLKL pathway and mitochondrial membrane potential (Δψm) level were investigated in BMECs upon OGD/R injury. The results showed that PNS significantly enhanced cell viability of BMECs determined by CCK-8 analysis, and protected BMECs from necroptosis by Flow cytometry and TEM. In addition, PNS inhibited the phosphorylation of RIP1, RIP3, MLKL and the downstream expression of PGAM5 and Drp1, while similar results were observed in Nec-1 intervention. We further investigated whether PNS prevented the Δψm depolarization. Our current findings showed that PNS effectively reduced the occurrence of necroptosis in BMECs exposed to OGD/R by inhibition of the RIP1-RIP3-MLK signaling pathway and mitigation of mitochondrial damage. This study provided a novel insight of PNS application in clinics.

Immunogenic Cell Death and Role of Nanomaterials Serving as Therapeutic Vaccine for Personalized Cancer Immunotherapy

Immunogenic cell death (ICD) is a rapidly growing research area representing one of the emerging therapeutic strategies of cancer immunotherapy. ICD is an umbrella term covering several cell death modalities including apoptosis, necroptosis, ferroptosis and pyroptosis, and is the product of a balanced combination of adjuvanticity (damage-associated molecular patterns and chemokines/cytokines) and antigenicity (tumor associated antigens). Only a limited number of anti-cancer therapies are available to induce ICD in experimental cancer therapies and even much less is available for clinical use. To overcome this limitation, nanomaterials can be used to increase the immunogenicity of cancer cells killed by anti-cancer therapy, which in themselves are not necessarily immunogenic. In this review, we outline the current state of knowledge of ICD modalities and discuss achievements in using nanomaterials to increase the immunogenicity of dying cancer cells. The emerging trends in modulating the immunogenicity of dying cancer cells in experimental and translational cancer therapies and the challenges facing them are described. In conclusion, nanomaterials are expected to drive further progress in their use to increase efficacy of anti-cancer therapy based on ICD induction and in the future, it is necessary to validate these strategies in clinical settings, which will be a challenging research area.

Label-free, non-invasive, and repeatable cell viability bioassay using dynamic full-field optical coherence microscopy and supervised machine learning

We present a novel method that can assay cellular viability in real-time using supervised machine learning and intracellular dynamic activity data that is acquired in a label-free, non-invasive, and non-destructive manner. Cell viability can be an indicator for cytology, treatment, and diagnosis of diseases. We applied four supervised machine learning models on the observed data and compared the results with a trypan blue assay. The cell death assay performance by the four supervised models had a balanced accuracy of 93.92 ± 0.86%. Unlike staining techniques, where criteria for determining viability of cells is unclear, cell viability assessment using machine learning could be clearly quantified.

Soft Hydrogel Actuator for Fast Machine-Learning-Assisted Bacteria Detection

We demonstrate that our bio-electrochemical platform facilitates the reduction of detection time from the 3-day period of the existing tests to 15 min. Machine learning and robotized bioanalytical platforms require the principles such as hydrogel-based actuators for fast and easy analysis of bioactive analytes. Bacteria are fragile and environmentally sensitive microorganisms that require a special environment to support their lifecycles during analytical tests. Here, we develop a bio-electrochemical platform based on the soft hydrogel/eutectic gallium-indium alloy interface for the detection of Streptococcus thermophilus and Bacillus coagulans bacteria in various mediums. The soft hydrogel-based device is capable to support bacteria' viability during detection time. Current-voltage data are used for multilayer perceptron algorithm training. The multilayer perceptron model is capable of detecting bacterial concentrations in the 10⁴ to 10⁸ cfu/mL range of the culture medium or in the dairy products with high accuracy (94%). Such a fast and easy biodetection is extremely important for food and agriculture industries and biomedical and environmental science.

Investigating Morphological Changes of T-lymphocytes after Exposure with Bacterial Determinants for Early Detection of Septic Conditions

Sepsis is a leading cause of morbidity and mortality, annually affecting millions of people worldwide. Immediate treatment initiation is crucial to improve the outcome but despite great progress, early identification of septic patients remains a challenge. Recently, white blood cell morphology was proposed as a new biomarker for sepsis diagnosis. In this proof-of-concept study, we aimed to investigate the effect of different bacteria and their determinants on T-lymphocytes by digital holographic microscopy (DHM). We hypothesize that species- and strain-specific morphological changes occur, which may offer a new approach for early sepsis diagnosis and identification of the causative agent. Jurkat cells as a model system were exposed to different S. aureus or E. coli strains either using sterile determinants or living bacteria. Time-lapse DHM was applied to analyze cellular morphological changes. There were not only living bacteria but also membrane vesicles and sterile culture supernatant-induced changes of cell area, circularity, and mean phase contrast. Interestingly, different cellular responses occurred depending on both the species and strain of the causative bacteria. Our findings suggest that investigation of T-lymphocyte morphology might provide a promising tool for the early identification of bacterial infections and possibly discrimination between different causative agents. Distinguishing gram-positive from gram-negative infection would already offer a great benefit for the proper administration of antibiotics.

Immunogenic ferroptosis and where to find it?

Ferroptosis is a recently discovered form of regulated cell death that is morphologically, genetically, and biochemically distinct from apoptosis and necroptosis, and its potential use in anticancer therapy is emerging. The strong immunogenicity of (early) ferroptotic cancer cells broadens the current concept of immunogenic cell death and opens up new possibilities for cancer treatment. In particular, induction of immunogenic ferroptosis could be beneficial for patients with cancers resistant to apoptosis and necroptosis. However, ferroptotic cancer cells may be a rich source of oxidized lipids, which contribute to decreased phagocytosis and antigen cross-presentation by dendritic cells and thus may favor tumor evasion. This could explain the non-immunogenicity of late ferroptotic cells. Besides the presence of lactate in the tumor microenvironment, acidification and hypoxia are essential factors promoting ferroptosis resistance and affecting its immunogenicity. Here, we critically discuss the crucial mediators controlling the immunogenicity of ferroptosis that modulate the induction of antitumor immunity. We emphasize that it will be necessary to also identify the tolerogenic (ie, immunosuppressive) nature of ferroptosis, which can lead to tumor evasion.