A fast, fully automated cell segmentation algorithm for high-throughput and high-content screening

YOLOv5-FPN: A Robust Framework for Multi-Sized Cell Counting in Fluorescence Images

Cell counting in fluorescence microscopy is an essential task in biomedical research for analyzing cellular dynamics and studying disease progression. Traditional methods for cell counting involve manual counting or threshold-based segmentation, which are time-consuming and prone to human error. Recently, deep learning-based object detection methods have shown promising results in automating cell counting tasks. However, the existing methods mainly focus on segmentation-based techniques that require a large amount of labeled data and extensive computational resources. In this paper, we propose a novel approach to detect and count multiple-size cells in a fluorescence image slide using You Only Look Once version 5 (YOLOv5) with a feature pyramid network (FPN). Our proposed method can efficiently detect multiple cells with different sizes in a single image, eliminating the need for pixel-level segmentation. We show that our method outperforms state-of-the-art segmentation-based approaches in terms of accuracy and computational efficiency. The experimental results on publicly available datasets demonstrate that our proposed approach achieves an average precision of 0.8 and a processing time of 43.9 ms per image. Our approach addresses the research gap in the literature by providing a more efficient and accurate method for cell counting in fluorescence microscopy that requires less computational resources and labeled data.

Machine learning approach for discrimination of genotypes based on bright-field cellular images

Morphological profiling is a combination of established optical microscopes and cutting-edge machine vision technologies, which stacks up successful applications in high-throughput phenotyping. One major question is how much information can be extracted from an image to identify genetic differences between cells. While fluorescent microscopy images of specific organelles have been broadly used for single-cell profiling, the potential ability of bright-field (BF) microscopy images of label-free cells remains to be tested. Here, we examine whether single-gene perturbation can be discriminated based on BF images of label-free cells using a machine learning approach. We acquired hundreds of BF images of single-gene mutant cells, quantified single-cell profiles consisting of texture features of cellular regions, and constructed a machine learning model to discriminate mutant cells from wild-type cells. Interestingly, the mutants were successfully discriminated from the wild type (area under the receiver operating characteristic curve = 0.773). The features that contributed to the discrimination were identified, and they included those related to the morphology of structures that appeared within cellular regions. Furthermore, functionally close gene pairs showed similar feature profiles of the mutant cells. Our study reveals that single-gene mutant cells can be discriminated from wild-type cells based on BF images, suggesting the potential as a useful tool for mutant cell profiling.

Kinematic dexterity analysis of human-robot interaction of an upper limb rehabilitation robot

BACKGROUND

The definition of rehabilitation training trajectory is of great significance during rehabilitation training, and the dexterity of human-robot interaction motion provides a basis for selecting the trajectory of interaction motion.

OBJECTIVE

Aimed at the kinematic dexterity of human-robot interaction, a velocity manipulability ellipsoid intersection volume (VMEIV) index is proposed for analysis, and the dexterity distribution cloud map is obtained with the human-robot cooperation space.

METHOD

Firstly, the motion constraint equation of human-robot interaction is established, and the Jacobian matrix is obtained based on the speed of connecting rod. Then, the Monte Carlo method and the cell body segmentation method are used to obtain the collaborative space of human-robot interaction, and the VMEIV of human-robot interaction is solved in the cooperation space. Finally, taking the upper limb rehabilitation robot as the research object, the dexterity analysis of human-robot interaction is carried out by using the index of the approximate volume of the VMEIV.

RESULTS

The results of the simulation and experiment have a certain consistency, which indicates that the VMEIV index is effective as an index of human-robot interaction kinematic dexterity.

CONCLUSIONS

The VMEIV index can measure the kinematic dexterity of human-robot interaction, and provide a reference for the training trajectory selection of rehabilitation robot.

A Bacterial Toxin with Analgesic Properties: Hyperpolarization of DRG Neurons by Mycolactone

Mycolactone, a polyketide molecule produced by Mycobacterium ulcerans, is the etiological agent of Buruli ulcer. This lipid toxin is endowed with pleiotropic effects, presents cytotoxic effects at high doses, and notably plays a pivotal role in host response upon colonization by the bacillus. Most remarkably, mycolactone displays intriguing analgesic capabilities: the toxin suppresses or alleviates the pain of the skin lesions it inflicts. We demonstrated that the analgesic capability of mycolactone was not attributable to nerve damage, but instead resulted from the triggering of a cellular pathway targeting AT₂ receptors (angiotensin II type 2 receptors; AT₂R), and leading to potassium-dependent hyperpolarization. This demonstration paves the way to new nature-inspired analgesic protocols. In this direction, we assess here the hyperpolarizing properties of mycolactone on nociceptive neurons. We developed a dedicated medium-throughput assay based on membrane potential changes, and visualized by confocal microscopy of bis-oxonol-loaded Dorsal Root Ganglion (DRG) neurons. We demonstrate that mycolactone at non-cytotoxic doses triggers the hyperpolarization of DRG neurons through AT₂R, with this action being not affected by known ligands of AT₂R. This result points towards novel AT₂R-dependent signaling pathways in DRG neurons underlying the analgesic effect of mycolactone, with the perspective for the development of new types of nature-inspired analgesics.

Acid phosphatase 2 (ACP2) is required for membrane fusion during influenza virus entry

Influenza viruses exploit host factors to successfully replicate in infected cells. Using small interfering RNA (siRNA) technology, we identified six human genes required for influenza A virus (IAV) replication. Here we focused on the role of acid phosphatase 2 (ACP2), as its knockdown showed the greatest inhibition of IAV replication. In IAV-infected cells, depletion of ACP2 resulted in a significant reduction in the expression of viral proteins and mRNA, and led to the attenuation of virus multi-cycle growth. ACP2 knockdown also decreased replication of seasonal influenza A and B viruses and avian IAVs of the H7 subtype. Interestingly, ACP2 depletion had no effect on the replication of Ebola or hepatitis C virus. Because ACP2 is known to be a lysosomal acid phosphatase, we assessed the role of ACP2 in influenza virus entry. While neither binding of the viral particle to the cell surface nor endosomal acidification was affected in ACP2-depleted cells, fusion of the endosomal and viral membranes was impaired. As a result, downstream steps in viral entry were blocked, including nucleocapsid uncoating and nuclear import of viral ribonucleoproteins. Our results established ACP2 as a necessary host factor for regulating the fusion step of influenza virus entry.

High-Content Screening of Raw Actinomycete Extracts for the Identification of Antituberculosis Activities

The feasibility and relevance of screening a library of raw actinomycete extracts (ECUM library) for the identification of antituberculosis activities was assessed on 11,088 extracts using a multiple-screening approach. Each extract was first tested at two concentrations against noninfected macrophages as a control, then against Mycobacterium tuberculosis growing in broth medium as well as infecting murine macrophages. The screening results indicated a library of good quality with an apparent low proportion of cytotoxic extracts. A correlation was found between both bacterial assays, but the intracellular assay showed limitations due to low rates of cell survival. Several extracts of interest were highlighted by this multiple screening. A focus on the strain producing the two most effective revealed similarities with known producers of active molecules, suggesting the possibility of selecting relevant extracts using this strategy.

Nanoluciferase signal brightness using furimazine substrates opens bioluminescence resonance energy transfer to widefield microscopy

Fluorescence and bioluminescence resonance energy transfer (FRET, BRET) techniques are powerful tools for studying protein-protein interactions in cellular assays. In contrast to fluorescent proteins, chemiluminescent proteins do not require excitation light, known to trigger autofluorescence, phototoxicity, and photobleaching. Regrettably, low signal intensity of luciferase systems restricts their usage as they require specialized microscopes equipped with ultra low-light imaging cameras. In this study, we report that bioluminescence quantification in living cells using a standard widefield automated microscope dedicated to screening and high content analysis is possible with the newer luciferase systems, Nanoluciferase (Nluc). With such equipment, we showed that robust intramolecular BRET can be measured using a combination of Nluc and yellow fluorescent protein (YFP). Using the human Superoxide Dismutase 1 (SOD1) dimer model, we next validated that intermolecular BRET could be quantified at a single cell level. The enhanced signal brightness of Nluc enabling BRET imaging to widefield microscopy shows strong potential to open up single cell protein-protein interactions studies to a wider audience. © 2016 International Society for Advancement of Cytometry.

Iterative h-minima-based marker-controlled watershed for cell nucleus segmentation

Automated microscopy imaging systems facilitate high-throughput screening in molecular cellular biology research. The first step of these systems is cell nucleus segmentation, which has a great impact on the success of the overall system. The marker-controlled watershed is a technique commonly used by the previous studies for nucleus segmentation. These studies define their markers finding regional minima on the intensity/gradient and/or distance transform maps. They typically use the h-minima transform beforehand to suppress noise on these maps. The selection of the h value is critical; unnecessarily small values do not sufficiently suppress the noise, resulting in false and oversegmented markers, and unnecessarily large ones suppress too many pixels, causing missing and undersegmented markers. Because cell nuclei show different characteristics within an image, the same h value may not work to define correct markers for all the nuclei. To address this issue, in this work, we propose a new watershed algorithm that iteratively identifies its markers, considering a set of different h values. In each iteration, the proposed algorithm defines a set of candidates using a particular h value and selects the markers from those candidates provided that they fulfill the size requirement. Working with widefield fluorescence microscopy images, our experiments reveal that the use of multiple h values in our iterative algorithm leads to better segmentation results, compared to its counterparts. © 2016 International Society for Advancement of Cytometry.

Segmentation of the Clustered Cells with Optimized Boundary Detection in Negative Phase Contrast Images

Cell image segmentation plays a central role in numerous biology studies and clinical applications. As a result, the development of cell image segmentation algorithms with high robustness and accuracy is attracting more and more attention. In this study, an automated cell image segmentation algorithm is developed to get improved cell image segmentation with respect to cell boundary detection and segmentation of the clustered cells for all cells in the field of view in negative phase contrast images. A new method which combines the thresholding method and edge based active contour method was proposed to optimize cell boundary detection. In order to segment clustered cells, the geographic peaks of cell light intensity were utilized to detect numbers and locations of the clustered cells. In this paper, the working principles of the algorithms are described. The influence of parameters in cell boundary detection and the selection of the threshold value on the final segmentation results are investigated. At last, the proposed algorithm is applied to the negative phase contrast images from different experiments. The performance of the proposed method is evaluated. Results show that the proposed method can achieve optimized cell boundary detection and highly accurate segmentation for clustered cells.

[High content screening in chemical biology: overview and main challenges]

The last two decades have seen the development of high content screening (HCS) methodology and its adaptation for the evaluation of small molecules as drug candidates or their use as chemical tools for research purpose. HCS was initially set-up for the understanding of the mechanism of action of compounds by testing them on cell based-assays for pharmacological and toxicological studies. Since the last decade, the use of HCS has been extended to academic research laboratories and this technology has become the starting point for numerous projects aiming at the identification of molecular targets and cellular pathways for a given disease on which novel type of drugs could act. This screening approach relies on image capture of fluorescently labeled cells therefore generating a large amount of data that must be handled by appropriate automated image analysis methods and storage instrumentation. These latter in addition to the integration and data sharing are current challenges that HCS must still tackle.

Automated Cell Detection and Morphometry on Growth Plate Images of Mouse Bone

Microscopy imaging of mouse growth plates is extensively used in biology to understand the effect of specific molecules on various stages of normal bone development and on bone disease. Until now, such image analysis has been conducted by manual detection. In fact, when existing automated detection techniques were applied, morphological variations across the growth plate and heterogeneity of image background color, including the faint presence of cells (chondrocytes) located deeper in tissue away from the image's plane of focus, and lack of cell-specific features, interfered with identification of cell. We propose the first method of automated detection and morphometry applicable to images of cells in the growth plate of long bone. Through ad hoc sequential application of the Retinex method, anisotropic diffusion and thresholding, our new cell detection algorithm (CDA) addresses these challenges on bright-field microscopy images of mouse growth plates. Five parameters, chosen by the user in respect of image characteristics, regulate our CDA. Our results demonstrate effectiveness of the proposed numerical method relative to manual methods. Our CDA confirms previously established results regarding chondrocytes' number, area, orientation, height and shape of normal growth plates. Our CDA also confirms differences previously found between the genetic mutated mouse Smad1/5^CKO and its control mouse on fluorescence images. The CDA aims to aid biomedical research by increasing efficiency and consistency of data collection regarding arrangement and characteristics of chondrocytes. Our results suggest that automated extraction of data from microscopy imaging of growth plates can assist in unlocking information on normal and pathological development, key to the underlying biological mechanisms of bone growth.

Mycobacterial toxin induces analgesia in buruli ulcer by targeting the angiotensin pathways

Mycobacterium ulcerans, the etiological agent of Buruli ulcer, causes extensive skin lesions, which despite their severity are not accompanied by pain. It was previously thought that this remarkable analgesia is ensured by direct nerve cell destruction. We demonstrate here that M. ulcerans-induced hypoesthesia is instead achieved through a specific neurological pathway triggered by the secreted mycobacterial polyketide mycolactone. We decipher this pathway at the molecular level, showing that mycolactone elicits signaling through type 2 angiotensin II receptors (AT2Rs), leading to potassium-dependent hyperpolarization of neurons. We further validate the physiological relevance of this mechanism with in vivo studies of pain sensitivity in mice infected with M. ulcerans, following the disruption of the identified pathway. Our findings shed new light on molecular mechanisms evolved by natural systems for the induction of very effective analgesia, opening up the prospect of new families of analgesics derived from such systems.

Digital reconstruction of the cell body in dense neural circuits using a spherical-coordinated variational model

Mapping the neuronal circuits is essential to understand brain function. Recent technological advancements have made it possible to acquire the brain atlas at single cell resolution. Digital reconstruction of the neural circuits down to this level across the whole brain would significantly facilitate brain studies. However, automatic reconstruction of the dense neural connections from microscopic image still remains a challenge. Here we developed a spherical-coordinate based variational model to reconstruct the shape of the cell body i.e. soma, as one of the procedures for this purpose. When intuitively processing the volumetric images in the spherical coordinate system, the reconstruction of somas with variational model is no longer sensitive to the interference of the complicated neuronal morphology, and could automatically and robustly achieve accurate soma shape regardless of the dense spatial distribution, and diversity in cell size, and morphology. We believe this method would speed drawing the neural circuits and boost brain studies.

Dimerization, oligomerization, and aggregation of human amyotrophic lateral sclerosis copper/zinc superoxide dismutase 1 protein mutant forms in live cells

More than 100 copper/zinc superoxide dismutase 1 (SOD1) genetic mutations have been characterized. These mutations lead to the death of motor neurons in ALS. In its native form, the SOD1 protein is expressed as a homodimer in the cytosol. In vitro studies have shown that SOD1 mutations impair the dimerization kinetics of the protein, and in vivo studies have shown that SOD1 forms aggregates in patients with familial forms of ALS. In this study, we analyzed WT SOD1 and 9 mutant (mt) forms of the protein by non-invasive fluorescence techniques. Using microscopic techniques such as fluorescence resonance energy transfer, fluorescence complementation, image-based quantification, and fluorescence correlation spectroscopy, we studied SOD1 dimerization, oligomerization, and aggregation. Our results indicate that SOD1 mutations lead to an impairment in SOD1 dimerization and, subsequently, affect protein aggregation. We also show that SOD1 WT and mt proteins can dimerize. However, aggregates are predominantly composed of SOD1 mt proteins.

An image-based algorithm for precise and accurate high throughput assessment of drug activity against the human parasite Trypanosoma cruzi

We present a customized high content (image-based) and high throughput screening algorithm for the quantification of Trypanosoma cruzi infection in host cells. Based solely on DNA staining and single-channel images, the algorithm precisely segments and identifies the nuclei and cytoplasm of mammalian host cells as well as the intracellular parasites infecting the cells. The algorithm outputs statistical parameters including the total number of cells, number of infected cells and the total number of parasites per image, the average number of parasites per infected cell, and the infection ratio (defined as the number of infected cells divided by the total number of cells). Accurate and precise estimation of these parameters allow for both quantification of compound activity against parasites, as well as the compound cytotoxicity, thus eliminating the need for an additional toxicity-assay, hereby reducing screening costs significantly. We validate the performance of the algorithm using two known drugs against T.cruzi: Benznidazole and Nifurtimox. Also, we have checked the performance of the cell detection with manual inspection of the images. Finally, from the titration of the two compounds, we confirm that the algorithm provides the expected half maximal effective concentration (EC50) of the anti-T. cruzi activity.

An automatic method for robust and fast cell detection in bright field images from high-throughput microscopy

Background

In recent years, high-throughput microscopy has emerged as a powerful tool to analyze cellular dynamics in an unprecedentedly high resolved manner. The amount of data that is generated, for example in long-term time-lapse microscopy experiments, requires automated methods for processing and analysis. Available software frameworks are well suited for high-throughput processing of fluorescence images, but they often do not perform well on bright field image data that varies considerably between laboratories, setups, and even single experiments.

Results

In this contribution, we present a fully automated image processing pipeline that is able to robustly segment and analyze cells with ellipsoid morphology from bright field microscopy in a high-throughput, yet time efficient manner. The pipeline comprises two steps: (i) Image acquisition is adjusted to obtain optimal bright field image quality for automatic processing. (ii) A concatenation of fast performing image processing algorithms robustly identifies single cells in each image. We applied the method to a time-lapse movie consisting of ∼315,000 images of differentiating hematopoietic stem cells over 6 days. We evaluated the accuracy of our method by comparing the number of identified cells with manual counts. Our method is able to segment images with varying cell density and different cell types without parameter adjustment and clearly outperforms a standard approach. By computing population doubling times, we were able to identify three growth phases in the stem cell population throughout the whole movie, and validated our result with cell cycle times from single cell tracking.

Conclusions

Our method allows fully automated processing and analysis of high-throughput bright field microscopy data. The robustness of cell detection and fast computation time will support the analysis of high-content screening experiments, on-line analysis of time-lapse experiments as well as development of methods to automatically track single-cell genealogies.

Attributed relational graphs for cell nucleus segmentation in fluorescence microscopy images

More rapid and accurate high-throughput screening in molecular cellular biology research has become possible with the development of automated microscopy imaging, for which cell nucleus segmentation commonly constitutes the core step. Although several promising methods exist for segmenting the nuclei of monolayer isolated and less-confluent cells, it still remains an open problem to segment the nuclei of more-confluent cells, which tend to grow in overlayers. To address this problem, we propose a new model-based nucleus segmentation algorithm. This algorithm models how a human locates a nucleus by identifying the nucleus boundaries and piecing them together. In this algorithm, we define four types of primitives to represent nucleus boundaries at different orientations and construct an attributed relational graph on the primitives to represent their spatial relations. Then, we reduce the nucleus identification problem to finding predefined structural patterns in the constructed graph and also use the primitives in region growing to delineate the nucleus borders. Working with fluorescence microscopy images, our experiments demonstrate that the proposed algorithm identifies nuclei better than previous nucleus segmentation algorithms.

A high-content, live-cell, and real-time approach to the quantitation of ligand-induced β-Arrestin2 and Class A/Class B GPCR mobilization

We report the development of a method to analyze receptor and β-arrestin2 mobilization between Class A and B GPCRs via time-resolved fluorescent microscopy coupled with semiautomated high-content multiparametric analysis. Using transiently expressed, tagged β2-adrenergic receptor (β₂-AR) or parathyroid hormone receptor type 1 (PTH₁R), we quantified trafficking of the receptors along with the mobilization and colocalization of coexpressed tagged β-arrestin2. This classification system allows for exclusion of cells with nonoptimal characteristics and calculation of multiple morphological and spatial parameters including receptor endosome formation, β-arrestin mobilization, colocalization, areas, and shape. Stimulated Class A and B receptors demonstrate dramatically different patterns with regard to β-arrestin interactions. The method provides high kinetic resolution measurement of receptor translocation, which allows for the identification of the fleeting β-arrestin interaction found with β₂-AR agonist stimulation, in contrast to stronger mobilization and receptor colocalization with agonist stimulation of the PTH₁R. Though especially appropriate for receptor kinetic studies, this method is generalizable to any dual fluorescence probe system in which quantification of object formation and movement is desired. These methodologies allow for quantitative, unbiased measurement of microscopy data and are further enhanced by providing real-time kinetics.

Smart markers for watershed-based cell segmentation

Automated cell imaging systems facilitate fast and reliable analysis of biological events at the cellular level. In these systems, the first step is usually cell segmentation that greatly affects the success of the subsequent system steps. On the other hand, similar to other image segmentation problems, cell segmentation is an ill-posed problem that typically necessitates the use of domain-specific knowledge to obtain successful segmentations even by human subjects. The approaches that can incorporate this knowledge into their segmentation algorithms have potential to greatly improve segmentation results. In this work, we propose a new approach for the effective segmentation of live cells from phase contrast microscopy. This approach introduces a new set of "smart markers" for a marker-controlled watershed algorithm, for which the identification of its markers is critical. The proposed approach relies on using domain-specific knowledge, in the form of visual characteristics of the cells, to define the markers. We evaluate our approach on a total of 1,954 cells. The experimental results demonstrate that this approach, which uses the proposed definition of smart markers, is quite effective in identifying better markers compared to its counterparts. This will, in turn, be effective in improving the segmentation performance of a marker-controlled watershed algorithm.

The balance of apoptotic and necrotic cell death in Mycobacterium tuberculosis infected macrophages is not dependent on bacterial virulence

Background

An important mechanism of Mycobacterium tuberculosis pathogenesis is the ability to control cell death pathways in infected macrophages: apoptotic cell death is bactericidal, whereas necrotic cell death may facilitate bacterial dissemination and transmission.

Methods

We examine M.tuberculosis control of spontaneous and chemically induced macrophage cell death using automated confocal fluorescence microscopy, image analysis, flow cytometry, plate-reader based vitality assays, and M.tuberculosis strains including H37Rv, and isogenic virulent and avirulent strains of the Beijing lineage isolate GC1237.

Results

We show that bacterial virulence influences the dynamics of caspase activation and the total level of cytotoxicity. We show that the powerful ability of M.tuberculosis to inhibit exogenously stimulated apoptosis is abrogated by loss of virulence. However, loss of virulence did not influence the balance of macrophage apoptosis and necrosis – both virulent and avirulent isogenic strains of GC1237 induced predominantly necrotic cell death compared to H37Rv which induced a higher relative level of apoptosis.

Conclusions

This reveals that macrophage necrosis and apoptosis are independently regulated during M. tuberculosis infection of macrophages. Virulence affects the level of host cell death and ability to inhibit apoptosis but other strain-specific characteristics influence the ultimate mode of host cell death and alter the balance of apoptosis and necrosis.

Intensity quantile estimation and mapping--a novel algorithm for the correction of image non-uniformity bias in HCS data

MOTIVATION

Image non-uniformity (NU) refers to systematic, slowly varying spatial gradients in images that result in a bias that can affect all downstream image processing, quantification and statistical analysis steps. Image NU is poorly modeled in the field of high-content screening (HCS), however, such that current conventional correction algorithms may be either inappropriate for HCS or fail to take advantage of the information available in HCS image data.

RESULTS

A novel image NU bias correction algorithm, termed intensity quantile estimation and mapping (IQEM), is described. The algorithm estimates the full non-linear form of the image NU bias by mapping pixel intensities to a reference intensity quantile function. IQEM accounts for the variation in NU bias over broad cell intensity ranges and data acquisition times, both of which are characteristic of HCS image datasets. Validation of the method, using simulated and HCS microtubule polymerization screen images, is presented. Two requirements of IQEM are that the dataset consists of large numbers of images acquired under identical conditions and that cells are distributed with no within-image spatial preference.

AVAILABILITY AND IMPLEMENTATION

MATLAB function files are available at http://nadon-mugqic.mcgill.ca/.

High-content imaging of Mycobacterium tuberculosis-infected macrophages: an in vitro model for tuberculosis drug discovery

Macrophages are reservoirs for replicating mycobacterium during tuberculosis (TB) infections. In this study, small molecules to be developed as anti-tubercular treatments were investigated for their ability to kill intracellular bacteria in in vitro macrophage models. High-content imaging technologies offer a high-throughput method to quantify a drug's ability to inhibit Mycobacterium tuberculosis intracellular invasion and multiplication in host cells. Dedicated image analysis enables the automated quantification of infected macrophages, and compounds that inhibit mycobacteria proliferation can be tested using this method. Furthermore, the implementation of the assay in 384-well microtiter plates combined with automated image acquisition and analysis allows large-scale screening of compound libraries in M. tuberculosis-infected macrophages. Here we describe a high-throughput and high-content workflow and detail its utility for the development of new TB drugs.

Comparison of segmentation algorithms for fluorescence microscopy images of cells

The analysis of fluorescence microscopy of cells often requires the determination of cell edges. This is typically done using segmentation techniques that separate the cell objects in an image from the surrounding background. This study compares segmentation results from nine different segmentation techniques applied to two different cell lines and five different sets of imaging conditions. Significant variability in the results of segmentation was observed that was due solely to differences in imaging conditions or applications of different algorithms. We quantified and compared the results with a novel bivariate similarity index metric that evaluates the degree of underestimating or overestimating a cell object. The results show that commonly used threshold-based segmentation techniques are less accurate than k-means clustering with multiple clusters. Segmentation accuracy varies with imaging conditions that determine the sharpness of cell edges and with geometric features of a cell. Based on this observation, we propose a method that quantifies cell edge character to provide an estimate of how accurately an algorithm will perform. The results of this study will assist the development of criteria for evaluating interlaboratory comparability.

High content phenotypic cell-based visual screen identifies Mycobacterium tuberculosis acyltrehalose-containing glycolipids involved in phagosome remodeling

The ability of the tubercle bacillus to arrest phagosome maturation is considered one major mechanism that allows its survival within host macrophages. To identify mycobacterial genes involved in this process, we developed a high throughput phenotypic cell-based assay enabling individual sub-cellular analysis of over 11,000 Mycobacterium tuberculosis mutants. This very stringent assay makes use of fluorescent staining for intracellular acidic compartments, and automated confocal microscopy to quantitatively determine the intracellular localization of M. tuberculosis. We characterised the ten mutants that traffic most frequently into acidified compartments early after phagocytosis, suggesting that they had lost their ability to arrest phagosomal maturation. Molecular analysis of these mutants revealed mainly disruptions in genes involved in cell envelope biogenesis (fadD28), the ESX-1 secretion system (espL/Rv3880), molybdopterin biosynthesis (moaC1 and moaD1), as well as in genes from a novel locus, Rv1503c-Rv1506c. Most interestingly, the mutants in Rv1503c and Rv1506c were perturbed in the biosynthesis of acyltrehalose-containing glycolipids. Our results suggest that such glycolipids indeed play a critical role in the early intracellular fate of the tubercle bacillus. The unbiased approach developed here can be easily adapted for functional genomics study of intracellular pathogens, together with focused discovery of new anti-microbials.

Evolving generalized Voronoi diagrams for accurate cellular image segmentation

Analyzing cellular morphologies on a cell-by-cell basis is vital for drug discovery, cell biology, and many other biological studies. Interactions between cells in their culture environments cause cells to touch each other in acquired microscopy images. Because of this phenomenon, cell segmentation is a challenging task, especially when the cells are of similar brightness and of highly variable shapes. The concept of topological dependence and the maximum common boundary (MCB) algorithm are presented in our previous work (Yu et al., Cytometry Part A 2009;75A:289-297). However, the MCB algorithm suffers a few shortcomings, such as low computational efficiency and difficulties in generalizing to higher dimensions. To overcome these limitations, we present the evolving generalized Voronoi diagram (EGVD) algorithm. Utilizing image intensity and geometric information, EGVD preserves topological dependence easily in both 2D and 3D images, such that touching cells can be segmented satisfactorily. A systematic comparison with other methods demonstrates that EGVD is accurate and much more efficient.

Analysis of cellular phosphatidylinositol (3,4,5)-trisphosphate levels and distribution using confocal fluorescent microscopy

We have developed an immunocytochemistry method for the semiquantitative detection of phosphatidylinositol (3,4,5)-trisphosphate (PI(3,4,5)P3) at the cell plasma membrane. This protocol combines the use of a glutathione S-transferase-tagged pleckstrin homology (PH) domain of the general phosphoinositides-1 receptor (GST-GRP1PH) with fluorescence confocal microscopy and image segmentation using cell mask software analysis. This methodology allows the analysis of PI(3,4,5)P3 subcellular distribution in resting and epidermal growth factor (EGF)-stimulated HEK293T cells and in LIM1215 (wild-type phosphoinositide 3-kinase (PI3K)) and LIM2550 (H1047R mutation in PI3K catalytic domain) colonic carcinoma cells. Formation of PI(3,4,5)P3 was observed 5min following EGF stimulation and resulted in an increase of the membrane/cytoplasm fluorescence ratio from 1.03 to 1.53 for HEK293T cells and from 2.2 to 3.3 for LIM1215 cells. Resting LIM2550 cells stained with GST-GRP1PH had an elevated membrane/cytoplasm fluorescence ratio of 9.8, suggesting constitutive PI3K activation. The increase in the membrane/cytoplasm fluorescent ratio was inhibited in a concentration-dependent manner by the PI3K inhibitor LY294002. This cellular confocal imaging assay can be used to directly assess the effects of PI3K mutations in cancer cell lines and to determine the potential specificity and effectiveness of PI3K inhibitors in cancer cells.

High content screening identifies decaprenyl-phosphoribose 2' epimerase as a target for intracellular antimycobacterial inhibitors

A critical feature of Mycobacterium tuberculosis, the causative agent of human tuberculosis (TB), is its ability to survive and multiply within macrophages, making these host cells an ideal niche for persisting microbes. Killing the intracellular tubercle bacilli is a key requirement for efficient tuberculosis treatment, yet identifying potent inhibitors has been hampered by labor-intensive techniques and lack of validated targets. Here, we present the development of a phenotypic cell-based assay that uses automated confocal fluorescence microscopy for high throughput screening of chemicals that interfere with the replication of M. tuberculosis within macrophages. Screening a library of 57,000 small molecules led to the identification of 135 active compounds with potent intracellular anti-mycobacterial efficacy and no host cell toxicity. Among these, the dinitrobenzamide derivatives (DNB) showed high activity against M. tuberculosis, including extensively drug resistant (XDR) strains. More importantly, we demonstrate that incubation of M. tuberculosis with DNB inhibited the formation of both lipoarabinomannan and arabinogalactan, attributable to the inhibition of decaprenyl-phospho-arabinose synthesis catalyzed by the decaprenyl-phosphoribose 2′ epimerase DprE1/DprE2. Inhibition of this new target will likely contribute to new therapeutic solutions against emerging XDR-TB. Beyond validating the high throughput/content screening approach, our results open new avenues for finding the next generation of antimicrobials.

Tuberculosis is still a major threat to global health. The disease in humans is caused by a bacterium, Mycobacterium tuberculosis, and treatment of an infected individual requires more than six months of chemotherapy. Because such a long course of treatment is required, compliance is low, which can result in the development of multidrug resistant strains (MDR-TB) and even extremely resistant strains (XDR-TB). Identifying new drug targets and potential lead therapeutic compounds are needed to combat MDR-XDR-TB. We developed a new type of assay based on the visualization of mycobacterium replication within host cells and applied it for the search of compounds that are able to chase the pathogen from its hideout. As a result, we found 20 new series of drug candidates that are effective against the bacilli in its hiding place, potentially addressing a crucial aspect in the resilience of the disease. We also showed that one series of compounds acts by inhibiting a key enzyme required for the synthesis of an essential component from the mycobacterial cell wall that is not targeted by any of the commercially available antituberculosis drugs. Altogether, our results pave the way for development of the next generation of antibacterial agents.

Automatic analysis of dividing cells in live cell movies to detect mitotic delays and correlate phenotypes in time

Live-cell imaging allows detailed dynamic cellular phenotyping for cell biology and, in combination with small molecule or drug libraries, for high-content screening. Fully automated analysis of live cell movies has been hampered by the lack of computational approaches that allow tracking and recognition of individual cell fates over time in a precise manner. Here, we present a fully automated approach to analyze time-lapse movies of dividing cells. Our method dynamically categorizes cells into seven phases of the cell cycle and five aberrant morphological phenotypes over time. It reliably tracks cells and their progeny and can thus measure the length of mitotic phases and detect cause and effect if mitosis goes awry. We applied our computational scheme to annotate mitotic phenotypes induced by RNAi gene knockdown of CKAP5 (also known as ch-TOG) or by treatment with the drug nocodazole. Our approach can be readily applied to comparable assays aiming at uncovering the dynamic cause of cell division phenotypes.

Benzothiazinones kill Mycobacterium tuberculosis by blocking arabinan synthesis

New drugs are required to counter the tuberculosis (TB) pandemic. Here, we describe the synthesis and characterization of 1,3-benzothiazin-4-ones (BTZs), a new class of antimycobacterial agents that kill Mycobacterium tuberculosis in vitro, ex vivo, and in mouse models of TB. Using genetics and biochemistry, we identified the enzyme decaprenylphosphoryl-beta-d-ribose 2'-epimerase as a major BTZ target. Inhibition of this enzymatic activity abolishes the formation of decaprenylphosphoryl arabinose, a key precursor that is required for the synthesis of the cell-wall arabinans, thus provoking cell lysis and bacterial death. The most advanced compound, BTZ043, is a candidate for inclusion in combination therapies for both drug-sensitive and extensively drug-resistant TB.