OPTIMAL: An OPTimized Imaging Mass cytometry AnaLysis framework for benchmarking segmentation and data exploration

Analysis of imaging mass cytometry (IMC) data and other low-resolution multiplexed tissue imaging technologies is often confounded by poor single-cell segmentation and suboptimal approaches for data visualization and exploration. This can lead to inaccurate identification of cell phenotypes, states, or spatial relationships compared to reference data from single-cell suspension technologies. To this end we have developed the "OPTimized Imaging Mass cytometry AnaLysis (OPTIMAL)" framework to benchmark any approaches for cell segmentation, parameter transformation, batch effect correction, data visualization/clustering, and spatial neighborhood analysis. Using a panel of 27 metal-tagged antibodies recognizing well-characterized phenotypic and functional markers to stain the same Formalin-Fixed Paraffin Embedded (FFPE) human tonsil sample tissue microarray over 12 temporally distinct batches we tested several cell segmentation models, a range of different arcsinh cofactor parameter transformation values, 5 different dimensionality reduction algorithms, and 2 clustering methods. Finally, we assessed the optimal approach for performing neighborhood analysis. We found that single-cell segmentation was improved by the use of an Ilastik-derived probability map but that issues with poor segmentation were only really evident after clustering and cell type/state identification and not always evident when using "classical" bivariate data display techniques. The optimal arcsinh cofactor for parameter transformation was 1 as it maximized the statistical separation between negative and positive signal distributions and a simple Z-score normalization step after arcsinh transformation eliminated batch effects. Of the five different dimensionality reduction approaches tested, PacMap gave the best data structure with FLOWSOM clustering out-performing phenograph in terms of cell type identification. We also found that neighborhood analysis was influenced by the method used for finding neighboring cells with a "disc" pixel expansion outperforming a "bounding box" approach combined with the need for filtering objects based on size and image-edge location. Importantly, OPTIMAL can be used to assess and integrate with any existing approach to IMC data analysis and, as it creates .FCS files from the segmentation output and allows for single-cell exploration to be conducted using a wide variety of accessible software and algorithms familiar to conventional flow cytometrists.

Deep Learning-based Image Cytometry Using a Bit-pattern Kernel-filtering Algorithm to Avoid Multi-counted Cell Determination

BACKGROUND/AIM

In pathology, the digitization of tissue slide images and the development of image analysis by deep learning have dramatically increased the amount of information obtainable from tissue slides. This advancement is anticipated to not only aid in pathological diagnosis, but also to enhance patient management. Deep learning-based image cytometry (DL-IC) is a technique that plays a pivotal role in this process, enabling cell identification and counting with precision. Accurate cell determination is essential when using this technique. Herein, we aimed to evaluate the performance of our DL-IC in cell identification.

MATERIALS AND METHODS

Cu-Cyto, a DL-IC with a bit-pattern kernel-filtering algorithm designed to help avoid multi-counted cell determination, was developed and evaluated for performance using tumor tissue slide images with immunohistochemical staining (IHC).

RESULTS

The performances of three versions of Cu-Cyto were evaluated according to their learning stages. In the early stage of learning, the F1 score for immunostained CD8⁺ T cells (0.343) was higher than the scores for non-immunostained cells [adenocarcinoma cells (0.040) and lymphocytes (0.002)]. As training and validation progressed, the F1 scores for all cells improved. In the latest stage of learning, the F1 scores for adenocarcinoma cells, lymphocytes, and CD8⁺ T cells were 0.589, 0.889, and 0.911, respectively.

CONCLUSION

Cu-Cyto demonstrated good performance in cell determination. IHC can boost learning efficiencies in the early stages of learning. Its performance is expected to improve even further with continuous learning, and the DL-IC can contribute to the implementation of precision oncology.

Assessment of Ploidy Status in Oral Potentially Malignant Disorders - A Systematic Review

Malignant and potentially malignant epithelial lesions are often associated with various abnormalities such as epithelial dysplasia, abnormal DNA content, loss of heterozygosity, and chromosomal number aberrations. Screening and early detection of such abnormalities facilitates proper care and also helps to prevent further progression of potentially malignant lesions to malignancy. In such way, the presence of DNA aneuploidy in oral potentially malignant disorders (OPMDs) may serve as an indicator for the malignant transforming potential. Various assessment methods have been proposed to find the DNA ploidy status of cells. This current systematic review is mainly designed to assess the importance of ploidy status in OPMD while measuring the feasibility of using this biomarker for evaluating the hazard of malignant transformation. As an upshot of this systematic review, we can conclude that use of DNA ploidy status can serve as an independent bio-marker for predicting the malignant transformation of lesions. Furthermore, as a future scope the use of DNA ploidy analysis in normal mucosa of smokers will help to assess the malignancy risk and this technique might also help to predict the genetic predisposition of patients with malignancy.

In situ cell division and mortality rates of SAR11, SAR86, Bacteroidetes, and Aurantivirga during phytoplankton blooms reveal differences in population controls

Net growth of microbial populations, that is, changes in abundances over time, can be studied using 16S rRNA fluorescence in situ hybridization (FISH). However, this approach does not differentiate between mortality and cell division rates. We used FISH-based image cytometry in combination with dilution culture experiments to study net growth, cell division, and mortality rates of four bacterial taxa over two distinct phytoplankton blooms: the oligotrophs SAR11 and SAR86, and the copiotrophic phylum Bacteroidetes, and its genus Aurantivirga. Cell volumes, ribosome content, and frequency of dividing cells (FDC) co-varied over time. Among the three, FDC was the most suitable predictor to calculate cell division rates for the selected taxa. The FDC-derived cell division rates for SAR86 of up to 0.8/day and Aurantivirga of up to 1.9/day differed, as expected for oligotrophs and copiotrophs. Surprisingly, SAR11 also reached high cell division rates of up to 1.9/day, even before the onset of phytoplankton blooms. For all four taxonomic groups, the abundance-derived net growth (-0.6 to 0.5/day) was about an order of magnitude lower than the cell division rates. Consequently, mortality rates were comparably high to cell division rates, indicating that about 90% of bacterial production is recycled without apparent time lag within 1 day. Our study shows that determining taxon-specific cell division rates complements omics-based tools and provides unprecedented clues on individual bacterial growth strategies including bottom-up and top-down controls. IMPORTANCE The growth of a microbial population is often calculated from their numerical abundance over time. However, this does not take cell division and mortality rates into account, which are important for deriving ecological processes like bottom-up and top-down control. In this study, we determined growth by numerical abundance and calibrated microscopy-based methods to determine the frequency of dividing cells and subsequently calculate taxon-specific cell division rates in situ. The cell division and mortality rates of two oligotrophic (SAR11 and SAR86) and two copiotrophic (Bacteroidetes and Aurantivirga) taxa during two spring phytoplankton blooms showed a tight coupling for all four taxa throughout the blooms without any temporal offset. Unexpectedly, SAR11 showed high cell division rates days before the bloom while cell abundances remained constant, which is indicative of strong top-down control. Microscopy remains the method of choice to understand ecological processes like top-down and bottom-up control on a cellular level.

Single-cell segmentation in bacterial biofilms with an optimized deep learning method enables tracking of cell lineages and measurements of growth rates

Bacteria often grow into matrix-encased three-dimensional (3D) biofilm communities, which can be imaged at cellular resolution using confocal microscopy. From these 3D images, measurements of single-cell properties with high spatiotemporal resolution are required to investigate cellular heterogeneity and dynamical processes inside biofilms. However, the required measurements rely on the automated segmentation of bacterial cells in 3D images, which is a technical challenge. To improve the accuracy of single-cell segmentation in 3D biofilms, we first evaluated recent classical and deep learning segmentation algorithms. We then extended StarDist, a state-of-the-art deep learning algorithm, by optimizing the post-processing for bacteria, which resulted in the most accurate segmentation results for biofilms among all investigated algorithms. To generate the large 3D training dataset required for deep learning, we developed an iterative process of automated segmentation followed by semi-manual correction, resulting in >18,000 annotated Vibrio cholerae cells in 3D images. We demonstrate that this large training dataset and the neural network with optimized post-processing yield accurate segmentation results for biofilms of different species and on biofilm images from different microscopes. Finally, we used the accurate single-cell segmentation results to track cell lineages in biofilms and to perform spatiotemporal measurements of single-cell growth rates during biofilm development.

A novel image-based method for simultaneous counting of Lactobacillus and Saccharomyces in mixed culture fermentation

Mixed microorganism cultures are prevalent in the food industry. From sour beer to sourdough bread, mixed culture popularity has been growing steadily. A variety of microbiological mixtures have been used in these unique fermenting processes to create distinctive flavor profiles and potential health benefits. Although mixed cultures may seem ubiquitous, they are typically not well characterized, which may be due to the lack of user-friendly and simple measurement tools. Image-based cytometry systems have been employed to count bacteria or yeast cells directly and rapidly in plastic chambers, as opposed to traditional cell counting methods that can take days to grow and count. In this work, we aim to develop a novel image cytometry method to distinguish and enumerate mixed cultures of yeast and bacteria in beer products. We demonstrate that the proposed image cytometry method can be used to rapidly distinguish and simultaneously count the bacteria and yeast cells mixed cultures. The Cellometer X2 from Nexcelom was used to automate the counting of Lactobacillus plantarum and Saccharomyces cerevisiae in a mixed culture using fluorescent dyes and size exclusion image analysis algorithm. The proposed image cytometry method was tested with a series of experiments. (1) Yeast and bacteria monoculture titration, (2) mixed culture with various ratios, and (3) monitoring a Berliner Weisse mixed culture fermentation. All experiments were validated by comparing directly to manual counting of yeast and bacteria colony formation on agar plates. The two counting methods were highly comparable with ANOVA analysis showing p-value > 0.05 for all trials. The image cytometry method was the most accurate for yeast concentrations between 105-107 cells/mL and 106-108 cells/mL for lactic acid bacteria. Overall, the novel image cytometry method was able to distinguish and count mixed cultures consistently and accurately, which may provide better characterization and assessment of mixed culture brewing applications that may produce more consistent and higher quality products.

The mechanistic and structural role of von Willebrand factor in endotoxemia-enhanced deep vein thrombosis in mice

BACKGROUND

Although the concept of immunothrombosis has established a link between inflammation and thrombosis, the role of inflammation in the pathogenesis of deep vein thrombosis remains to be fully elucidated. Further, although various constituents of venous thrombi have been identified, their localizations and cellular and molecular interactions are yet to be combined in a single, multiplexed analysis.

OBJECTIVES

The objective of this study was to investigate the role of the von Willebrand factor (VWF) in inflammation-associated venous thrombosis. We also performed a proof-of-concept study of imaging mass cytometry to quantitatively and simultaneously analyze the localizations and interactions of 10 venous thrombus constituents.

METHODS

We combined the murine inferior vena cava stenosis model of deep vein thrombosis with the lipopolysaccharide model of endotoxemia. We also performed a proof-of-concept study of imaging mass cytometry to assess the feasibility of this approach in analyzing the structural composition of thrombi.

RESULTS

We found that lipopolysaccharide-treated mice had significantly higher incidences of venous thrombosis, an effect that was mitigated when VWF was inhibited using inhibitory αVWF antibodies. Our detailed structural analysis also showed that most thrombus components are localized in the white thrombus regardless of endotoxemia. Moreover, although endotoxemia modulated the relative representation and interactions of VWF with other thrombus constituents, the scaffolding network, comprised VWF, fibrin, and neutrophil extracellular traps, remained largely unaffected.

CONCLUSIONS

We observe a key role for VWF in the pathogenesis of inflammation-associated venous thrombosis while providing a more comprehensive insight into the molecular interactions that constitute the architecture of venous thrombi.

Correlative Multi-Modal Microscopy: A Novel Pipeline for Optimizing Fluorescence Microscopy Resolutions in Biological Applications

The modern fluorescence microscope is the convergence point of technologies with different performances in terms of statistical sampling, number of simultaneously analyzed signals, and spatial resolution. However, the best results are usually obtained by maximizing only one of these parameters and finding a compromise for the others, a limitation that can become particularly significant when applied to cell biology and that can reduce the spreading of novel optical microscopy tools among research laboratories. Super resolution microscopy and, in particular, molecular localization-based approaches provide a spatial resolution and a molecular localization precision able to explore the scale of macromolecular complexes in situ. However, its use is limited to restricted regions, and consequently few cells, and frequently no more than one or two parameters. Correlative microscopy, obtained by the fusion of different optical technologies, can consequently surpass this barrier by merging results from different spatial scales. We discuss here the use of an acquisition and analysis correlative microscopy pipeline to obtain high statistical sampling, high content, and maximum spatial resolution by combining widefield, confocal, and molecular localization microscopy.

From Double-Strand Break Recognition to Cell-Cycle Checkpoint Activation: High Content and Resolution Image Cytometry Unmasks 53BP1 Multiple Roles in DNA Damage Response and p53 Action

53BP1 protein has been isolated in-vitro as a putative p53 interactor. From the discovery of its engagement in the DNA-Damage Response (DDR), its role in sustaining the activity of the p53-regulated transcriptional program has been frequently under-evaluated, even in the case of a specific response to numerous DNA Double-Strand Breaks (DSBs), i.e., exposure to ionizing radiation. The localization of 53BP1 protein constitutes a key to decipher the network of activities exerted in response to stress. We present here an automated-microscopy for image cytometry protocol to analyze the evolution of the DDR, and to demonstrate how 53BP1 moved from damaged sites, where the well-known interaction with the DSB marker γH2A.X takes place, to nucleoplasm, interacting with p53, and enhancing the transcriptional regulation of the guardian of the genome protein. Molecular interactions have been quantitatively described and spatiotemporally localized at the highest spatial resolution by a simultaneous analysis of the impairment of the cell-cycle progression. Thanks to the high statistical sampling of the presented protocol, we provide a detailed quantitative description of the molecular events following the DSBs formation. Single-Molecule Localization Microscopy (SMLM) Analysis finally confirmed the p53–53BP1 interaction on the tens of nanometers scale during the distinct phases of the response.

Genome Size of Life Forms of Araceae-A New Piece in the C-Value Puzzle

The genome size of an organism is an important trait that has predictive values applicable to various scientific fields, including ecology. The main source of plant C-values is the Plant DNA C-values database of the Royal Botanic Gardens Kew, which currently contains 12,273 estimates. However, it covers only 2.9% of known angiosperm species and has gaps in the life form and geographic distribution of plants. Only 4.5% of C-value estimates come from researchers in Central and South America. This study provides 41 new C-values for the aroid family (Araceae), collected in the Piedras Blancas National Park area in southern Costa Rica, including terrestrial, epiphytic and aquatic life forms. Data from our study are combined with C-value entries in the RBGK database for Araceae. The analysis reveals a wider range of C-values for terrestrial aroids, consistent with other terrestrial plants, a trend toward slightly lower C-values for epiphytic forms, which is more consistent for obligate epiphytes, and comparatively low C-values for aquatic aroids.

Measurement of agonist-induced Ca2+ signals in human airway smooth muscle cells using excitation scan-based hyperspectral imaging and image analysis approaches

Ca²⁺ and cAMP are ubiquitous second messengers known to differentially regulate a variety of cellular functions over a wide range of timescales. Studies from a variety of groups support the hypothesis that these signals can be localized to discrete locations within cells, and that this subcellular localization is a critical component of signaling specificity. However, to date, it has been difficult to track second messenger signals at multiple locations. To overcome this limitation, we utilized excitation scan-based hyperspectral imaging approaches to track second messenger signals as well as labeled cellular structures and/or proteins in the same cell. We have previously reported that hyperspectral imaging techniques improve the signal-to-noise ratios of both fluorescence measurements, and are thus well suited for the measurement of localized Ca²⁺ signals. We investigated the spatial spread and intensities of agonist-induced Ca²⁺ signals in primary human airway smooth muscle cells (HASMCs) using the Ca²⁺ indicator Cal520. We measured responses triggered by three agonists, carbachol, histamine, and chloroquine. We utilized custom software coded in MATLAB and Python to assess agonist induced changes in Ca²⁺ levels. Software algorithms removed the background and applied correction coefficients to spectral data prior to linear unmixing, spatial and temporal filtering, adaptive thresholding, and automated region of interest (ROI) detection. All three agonists triggered transient Ca²⁺ responses that were spatially and temporally complex. We are currently analyzing differences in both ROI area and intensity distributions triggered by these agonists. This work was supported by NIH awards P01HL066299, K25HL136869, and R01HL137030 and NSF award MRI1725937.

Novel Hyperspectral imaging approaches allow 3D measurement of cAMP signals in localized subcellular domains of human airway smooth muscle cells

Studies of the cAMP signaling pathway have led to the hypothesis that localized cAMP signals regulate distinct cellular responses. Much of this work focused on measurement of localized cAMP signals using cAMP sensors based upon Fӧrster resonance energy transfer (FRET). FRET-based probes are comprised of a cAMP binding domain sandwiched between donor and acceptor fluorophores. Binding of cAMP triggers a conformational change which alters FRET efficiency. In order to study localized cAMP signals, investigators have targeted FRET probes to distinct subcellular domains. This approach allows detection of cAMP signals at distinct subcellular locations. However, these approaches do not measure localized cAMP signals per se, rather they measure cAMP signals at specific locations and typically averaged throughout the cell. To address these concerns, our group implemented hyperspectral imaging approaches for measuring highly multiplexed signals in cells and tissues. We have combined these approaches with custom analysis software implemented in MATLAB and Python. Images were filtered both spatially and temporally, prior to adaptive thresholding (OTSU) to detect cAMP signals. These approaches were used to interrogate the distributions of isoproterenol and prostaglandin-triggered cAMP signals in human airway smooth muscle cells (HASMCs). Results demonstrate that cAMP signals are spatially and temporally complex. We observed that isoproterenol- and prostaglandin-induced cAMP signals are triggered at the plasma membrane and in the cytosolic space. We are currently implementing analysis approaches to better quantify and visualize the complex distributions of cAMP signals. This work was supported by NIH P01HL066299, R01HL058506, and S10RR027535.

The MTT Assay: Utility, Limitations, Pitfalls, and Interpretation in Bulk and Single-Cell Analysis

The MTT assay for cellular metabolic activity is almost ubiquitous to studies of cell toxicity; however, it is commonly applied and interpreted erroneously. We investigated the applicability and limitations of the MTT assay in representing treatment toxicity, cell viability, and metabolic activity. We evaluated the effect of potential confounding variables on the MTT assay measurements on a prostate cancer cell line (PC-3) including cell seeding number, MTT concentration, MTT incubation time, serum starvation, cell culture media composition, released intracellular contents (cell lysate and secretome), and extrusion of formazan to the extracellular space. We also assessed the confounding effect of polyethylene glycol (PEG)-coated gold nanoparticles (Au-NPs) as a tested treatment in PC-3 cells on the assay measurements. We additionally evaluated the applicability of microscopic image cytometry as a tool for measuring intracellular MTT reduction at the single-cell level. Our findings show that the assay measurements are a result of a complicated process dependant on many of the above-mentioned factors, and therefore, optimization of the assay and rational interpretation of the data is necessary to prevent misleading conclusions on variables such as cell viability, treatment toxicity, and/or cell metabolism. We conclude, with recommendations on how to apply the assay and a perspective on where the utility of the assay is a powerful tool, but likewise where it has limitations.

Development of an Image-Based HCS-Compatible Method for Endothelial Barrier Function Assessment

The recent renascence of phenotypic drug discovery (PDD) is catalyzed by its ability to identify first-in-class drugs and deliver results when the exact molecular mechanism is partially obscure. Acute respiratory distress syndrome (ARDS) is a severe, life-threatening condition with a high mortality rate that has increased in frequency due to the COVID-19 pandemic. Despite decades of laboratory and clinical study, no efficient pharmacological therapy for ARDS has been found. An increase in endothelial permeability is the primary event in ARDS onset, causing the development of pulmonary edema that leads to respiratory failure. Currently, the detailed molecular mechanisms regulating endothelial permeability are poorly understood. Therefore, the use of the PDD approach in the search for efficient ARDS treatment can be more productive than classic target-based drug discovery (TDD), but its use requires a new cell-based assay compatible with high-throughput (HTS) and high-content (HCS) screening. Here we report the development of a new plate-based image cytometry method to measure endothelial barrier function. The incorporation of image cytometry in combination with digital image analysis substantially decreases assay variability and increases the signal window. This new method simultaneously allows for rapid measurement of cell monolayer permeability and cytological analysis. The time-course of permeability increase in human pulmonary artery endothelial cells (HPAECs) in response to the thrombin and tumor necrosis factor α treatment correlates with previously published data obtained by transendothelial resistance (TER) measurements. Furthermore, the proposed image cytometry method can be easily adapted for HTS/HCS applications.

Development and validation of a risk model for noninvasive detection of cancer in oral potentially malignant disorders using DNA image cytometry

OBJECTIVE

To elucidate whether DNA aneuploidy was an independent discriminator for carcinoma within oral potentially malignant disorders (OPMDs), and further establish and validate a risk model based on DNA aneuploidy for the detection of oral cancer.

METHODS

A total of 810 consecutive patients with OPMD were prospectively enrolled from March 2013 to December 2018, and divided into a training set (n = 608) and a test set (n = 202). Brushing and biopsy samples from each patient were processed by DNA-DNA image cytometry and histopathological examination, respectively.

RESULTS

DNA aneuploidy of an outside DNA index ≥ 3.5 in OPMD was an independent marker strongly associated with malignant risk [adjusted odds ratio: 13.04; 95% confidence interval (CI): 5.46-31.14]. In the training and test sets, the area under the curve (AUC) was 0.87 (95% CI: 0.82-0.91) and 0.77 (95% CI: 0.57-0.97), respectively, for detecting carcinoma in OPMD patients. The independent risk factors of lateral/ventral tongue and non-homogenous type combined with a risk model built with a multivariate logistic regression revealed a more favorable diagnostic efficacy associated with the training set (AUC: 0.93; 95% CI: 0.91-0.96) and test set (AUC: 0.94; 95% CI: 0.90-0.98). The sensitivity and specificity of carcinoma detection within OPMD was improved to 100% and 88.1%, respectively.

CONCLUSIONS

This large-scale diagnostic study established a risk model based on DNA aneuploidy that consisted of a noninvasive strategy with lateral/ventral tongue and non-homogenous features. The results showed favorable diagnostic efficacy for detecting carcinoma within OPMD, irrespective of the clinical and pathological diagnoses of OPMD. Multicenter validation and longitudinal studies are warranted to evaluate community practices and clinical applications.

Comparison of Ki-67 Labeling Index Patterns of Diffuse Large B-Cell Lymphomas and Burkitt Lymphomas Using Image Analysis: A Multicenter Study

Diffuse large B-cell lymphoma (DLBCL) is the most common high-grade B-cell lymphoma found in Korea; it manifests with a variety of cellular morphologies and a high proliferation index. It is difficult to differentiate between DLBCL and Burkitt lymphoma (BL) based on immunohistochemistry, histology, and Epstein-Barr virus infection status owing to the overlap in findings. In this study, we performed comparative morphometric analysis to understand the proportional difference in Ki-67 staining between DLBCL and BL. We analyzed Ki-67-stained slides of 103 DLBCLs and 29 BLs that were pathologically confirmed using a three-tier classification system (negative, 1+, 2+, and 3+) to compare Ki-67 expression between BL and activated B-cell and germinal center B-cell subtypes of DLBCL and DLBCL with high proliferation indices (>90% of 2+ and 3+ cells). Patients with DLBCL were older than those with BL (62.1 versus 51.0 years). The number and proportion of negative cells (passenger and true negative cells) were significantly lower in BLs than those in DLBCLs (337.4, 5.9% versus 690.3, 12.4%). The number and proportion of 3+ cells were significantly higher in BLs than those in DLBCLs (5213.6, 96.3% versus 3132.4, 62.0%). BLs and DLBCLs with a high proliferation index showed similar results as those between BLs and overall DLBCLs. We were able to differentiate BLs and DLBCLs with 98.1% sensitivity and 100.0% specificity using an optimal cut-off of 97.9% of 2+/3+ Ki-67-positive cells. Thus, the Ki-67 labeling index may be a good differential biomarker for DLBCLs and BLs.

High-Throughput Image Cytometry Detection Method for CAR-T Transduction, Cell Proliferation, and Cytotoxicity Assays

Chimeric antigen receptor (CAR)-T cell therapy has drawn much attention due to its recent clinical success in B-cell malignancies. In general, the CAR-T cell discovery process consists of CAR identification, T-cell activation, transduction, and expansion, as well as assessment of CAR-T cytotoxicity. The current evaluation methods for the CAR-T discovery process can be time-consuming, low-throughput and requires the preparation of multiple sacrificial samples in order to produce kinetic data. In this study, we employed the use of a plate-based image cytometer to monitor anti-CAIX (carbonic anhydrase IX) G36 CAR-T generation and assess its cytotoxic potency of direct and selective killing against CAIX⁺ SKRC-59 human renal cell carcinoma cells. The transduction efficiency and cytotoxicity results were analyzed using image cytometry and compared directly to flow cytometry and Chromium 51 (⁵¹ Cr) release assays, showing that image cytometry was comparable against these conventional methods. Image cytometry method streamlines the assays required during the CAR-T cell discovery process by analyzing a plate of T cells from CAR-T generation to in vitro functional assays with minimum disruption. The proposed method can reduce assay time and uses less cell samples by imaging and analyze the same plate over time without the need to sacrifice any cells. The ability to monitor kinetic data can allow additional insights into the behavior and interaction between CAR-T and target tumor cells. © 2020 International Society for Advancement of Cytometry.

SAPHIR: a Shiny application to analyze tissue section images

Study of cell populations in tissues using immunofluorescence is a powerful method for both basic and medical research. Image acquisitions performed by confocal microscopy notably allow excellent lateral resolution and more than 10 parameter measurements when using spectral or multiplex imaging. Analysis of such complex images can be very challenging and easily lead to bias and misinterpretation. Here, we have developed the Shiny Analytical Plot of Histological Image Results (SAPHIR), an R shiny application for histo-cytometry using scatterplot representation of data extracted by segmentation. It offers many features, such as filtering of spurious data points, selection of cell subsets on scatterplot, visualization of scatterplot selections back into the image, statistics of selected data and data annotation. Our application allows to characterize labeled cells, from their phenotype to their number and location in the tissue, as well as their interaction with other cells. SAPHIR is available from: https://github.com/elodiegermani/SAPHIR.

SAPHIR: a Shiny application to analyze tissue section images

Study of cell populations in tissues using immunofluorescence is a powerful method for both basic and medical research. Image acquisitions performed by confocal microscopy notably allow excellent lateral resolution and more than 10 parameter measurements when using spectral or multiplex imaging. Analysis of such complex images can be very challenging and easily lead to bias and misinterpretation. Here, we have developed the Shiny Analytical Plot of Histological Image Results (SAPHIR), an R shiny application for histo-cytometry using scatterplot representation of data extracted by segmentation. It offers many features, such as filtering of spurious data points, selection of cell subsets on scatterplot, visualization of scatterplot selections back into the image, statistics of selected data and data annotation. Our application allows to characterize labeled cells, from their phenotype to their number and location in the tissue, as well as their interaction with other cells. SAPHIR is available from: https://github.com/elodiegermani/SAPHIR.

Using Cytometry for Investigation of Purinergic Signaling in Tumor-Associated Macrophages

Tumor-associated macrophages are widely recognized for their importance in guiding pro-tumoral or antitumoral responses. Mediating inflammation or immunosuppression, these cells support many key events in cancer progression: cell growth, chemotaxis, invasiveness, angiogenesis and cell death. The communication between cells in the tumor microenvironment strongly relies on the secretion and recognition of several molecules, including damage-associated molecular patterns (DAMPs), such as adenosine triphosphate (ATP). Extracellular ATP (eATP) and its degradation products act as signaling molecules and have extensively described roles in immune response and inflammation, as well as in cancer biology. These multiple functions highlight the purinergic system as a promising target to investigate the interplay between macrophages and cancer cells. Here, we reviewed purinergic signaling pathways connecting cancer cells and macrophages, a yet poorly investigated field. Finally, we present a new tool for the characterization of macrophage phenotype within the tumor. Image cytometry emerges as a cutting-edge tool, capable of providing a broad set of information on cell morphology, expression of specific markers, and its cellular or subcellular localization, preserving cell-cell interactions within the tumor section and providing high statistical strength in small-sized experiments. Thus, image cytometry allows deeper investigation of tumor heterogeneity and interactions between these cells. © 2020 International Society for Advancement of Cytometry.

Slug Is A Surrogate Marker of Epithelial to Mesenchymal Transition (EMT) in Head and Neck Cancer

Background: Epithelial to mesenchymal transition (EMT) promotes therapy resistance in head and neck cancer (HNC) cells. In this study, EMT was quantified in HNC tumor samples by the cellular co-localization of cytokeratin/vimentin, E-cadherin/β-catenin and by Slug expression. Methods: Tissue samples from HNC patients were stained with antibody pairs against cytokeratin/vimentin and E-cadherin/β-catenin. Epithelial–mesenchymal co-localization was quantified using immunofluorescence multichannel image cytometry. Double positivity was confirmed using confocal microscopy. Slug was semi-quantified by 2 specialists and quantified by bright field image cytometry. Results: Tumor samples of 102 patients were investigated. A loss of E-cadherin positive cells (56.9 ± 2.6% vs. 97.9 ± 1.0%; p < 0.0001) and E-cadherin/β-catenin double positive cells (15.4 ± 5.7% vs. 85.4 ± 1.2%; p < 0.0001) was observed in tumor samples. The percentage of Slug positive cells was increased in tumor samples (12.1 ± 3.6% vs. 3.2 ± 2.6%; p = 0.001). Ordinal Slug scores judged by two specialists closely correlated with percentage of Slug-positive cells (Spearman’s rho = 0.81; p < 0.001). Slug score correlated negatively with the percentage of E-cadherin positive cells (r = 0.4; p = 0.006), the percentage of E-cadherin/β-catenin positive cells (r = 0.5; p = 0.001) and positively with cytokeratin/vimentin positive cells (r = 0.4, p = 0.003). Conclusion: EMT can be assessed in HNC tumor probes by cytokeratin/vimentin co-expression and loss of E-cadherin/β-catenin co-expression. Slug score provides a convenient surrogate marker for EMT.

A High-Throughput Image Cytometry Method for the Formation, Morphometric, and Viability Analysis of Drug-Treated Mammospheres

The nonadherent mammosphere assay has been commonly used to investigate cancer stem cell activities in breast cancers that have the ability to form tumorspheres and maintain tumor growth. The sphere formation step is critical, in that it enables the construction of the mammosphere models for downstream assays. The mammosphere assay has also been used to assess the effects of drug treatment on the tumorspheres formed from primary cancer cells or cell lines. Traditionally, the mammosphere formation has been evaluated by standard microscopy systems that required external software for additional analyses. However, this method can be time-consuming and low-throughput, thus impractical for high-throughput characterization of mammosphere models and screening for potential therapeutic cancer drugs. To overcome these challenges, we developed a plate-based high-throughput method to rapidly analyze mammospheres in whole wells using the Celigo Image Cytometer. The method is employed to characterize mammosphere formation and morphology for adherent and nonadherent propagation of four breast cancer cell lines (MCF7, MDA-MB-436, MDA-MB-231, and SKBR3). Next, the dose-dependent effects of four small molecule drugs (doxorubicin, paclitaxel, 8-quinolinol, and salinomycin) are characterized based on sphere formation and viability stained with calcein AM and propidium iodide. We observed growth and morphometric differences between adherent and nonadherent propagation of the four cell lines. Furthermore, drug treatments induced various effects on mammosphere formation, morphology, and viability. The proposed image cytometry method provides a useful tool suitable for high-throughput characterization and analysis of mammospheres, which can improve assay efficiency when investigating the formation capabilities and drug-induced cytotoxicity effects.

Automated imaging cytometry reveals dysplastic indices of colonic serrated adenomas

Aim:

Left-sided colonic serrated adenomas (L-SAs) were evaluated for aneuploidy using automated imaging cytometry to quantify DNA content and compared with normal colonic tissues (NCT), tubular adenomas (TA), left-sided hyperplastic polyps (L-HP) and adenocarcinomas.

Materials & methods:

We used standard paraffin-embedded Feulgen-stained tissue sections.

Results:

The mean DNA index (DI) of NCT was 0.95, L-HP was 1.08, TA was 1.22, L-SA was 1.11 and adenocarcinomas was 1.46. DI of L-SA was statistically higher than that of NCT, but not statistically different from L-HP.

Conclusion:

This study demonstrates that DIs correlate with the described neoplastic progression of L-SA, TA and L-SA compared with NCT and suggests that L-SA may be involved in a chromosome instability pathway of neoplastic progression.

Colon cancer remains a deadly disease, with a significant burden of illness to patients and healthcare systems. While most precursor lesions will not necessarily produce cancers, they vary in histology and potential for neoplastic progression. Aneuploidy or abnormal chromosomal content of a cell is considered a marker for chromosomal instability and neoplastic progression. However, conventional methods of assessment can be laborious, costly and may even underestimate its malignant potential if the lesion is focal, small and surrounded by normal stromal cells in the sampled tissue. We used a nuclear stain to detect and quantify aneuploidy on conventionally prepared colonic precancerous histological slides and in particular assessed serrated and hyperplastic polyps of the left colon. When compared with normal tissues, we determined that there was aneuploidy in these lesions, which supports the underappreciated assumption that these lesions manifest chromosomal instability.

Diagnostic accuracy of brush biopsy-based cytology for the early detection of oral cancer and precursors in Fanconi anemia

BACKGROUND

Individuals with Fanconi anemia (FA) have a 500-fold to 700-fold elevated risk, much earlier onset, and limited therapeutic options for oral squamous cell carcinoma (SCC) compared with the general population. The early detection of SCC, or preferably its precursors, is mandatory to retain curative therapeutic options. Due to frequent synchronic and metachronic oral lesions, tissue biopsies, as usually recommended by guidelines, often are not feasible. In the current study, an alternative strategy for early detection using oral brush biopsy-based cytology was validated regarding its diagnostic accuracy.

METHODS

Over a 12-year period, the oral cavities of a large cohort of 713 individuals with FA were inspected systematically and brush biopsy-based cytology of 1233 visible oral lesions was performed. In cases of inconclusive cytology, analysis of DNA ploidy was performed whenever possible. The results were correlated to a long-term clinicopathological follow-up reference standard.

RESULTS

A total of 737 lesions were suitable for statistical analysis, including 86 lesions with at least high-grade oral epithelial dysplasia in 30 patients. For cytology, the sensitivity and specificity were 97.7% and 84.5%, respectively. Additional analysis of DNA ploidy increased the sensitivity and specificity to 100% and 92.2%, respectively.

CONCLUSIONS

Careful inspection of the oral cavity of individuals with FA followed by brush biopsy-based cytology appears to identify visible oral, potentially malignant and malignant lesions that warrant treatment. Approximately 63% of SCC and precursor lesions are detected at a noninvasive or early stage. Negative cytology or a lack of DNA aneuploidy can exclude high-grade oral epithelial dysplasia or SCC with high accuracy and thus reduce the need for invasive diagnostic biopsies.

New Antifungal Susceptibility Test Based on Chitin Detection by Image Cytometry

The antifungal susceptibility tests used in clinical laboratories have several limitations. We developed a new test, SensiFONG, based on the detection of chitin levels after exposure to antifungal drugs. The optimal culture conditions were 30°C for 6 h for yeast strains and 26°C for 16 h for molds. The strains were exposed to a range of echinocandin or azole concentrations. Chitin was stained with calcofluor white. The percentage of fungal cells with high chitin levels was determined with an automatic epifluorescence microscope. The SensiFONG results were compared to those with the EUCAST method. Image acquisition and analysis were performed with ScanR software. Fifty-nine strains (28 Candida albicans, 17 Candida glabrata, and 14 Aspergillus fumigatus) were analyzed. Thresholds for the classification of strains as resistant or susceptible were determined for each fungal species. The strains displaying an increase in chitin content of ≥32% for C. albicans, ≥6% for C. glabrata, and ≥17% for A. fumigatus were considered susceptible. The application of these thresholds to all 59 strains resulted in a sensitivity of 0.87, 0.93, and 1.00 and a specificity of 0.93, 0.84, and 0.82 for C. albicans, C. glabrata, and A. fumigatus, respectively. The correlation between the results obtained in the SensiFONG and EUCAST assays was excellent. We developed a new test, SensiFONG, based on a new concept. While current assays assess growth inhibition, our test detects changes in chitin levels after exposure to antifungal drugs. Here, we present preliminary results and we propose a proof of concept of this methodology.

Deep Learning in Image Cytometry: A Review

Artificial intelligence, deep convolutional neural networks, and deep learning are all niche terms that are increasingly appearing in scientific presentations as well as in the general media. In this review, we focus on deep learning and how it is applied to microscopy image data of cells and tissue samples. Starting with an analogy to neuroscience, we aim to give the reader an overview of the key concepts of neural networks, and an understanding of how deep learning differs from more classical approaches for extracting information from image data. We aim to increase the understanding of these methods, while highlighting considerations regarding input data requirements, computational resources, challenges, and limitations. We do not provide a full manual for applying these methods to your own data, but rather review previously published articles on deep learning in image cytometry, and guide the readers toward further reading on specific networks and methods, including new methods not yet applied to cytometry data. © 2018 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.

Image cytometry as an alternative to flow cytometry for the transplant histocompatibility crossmatch assay

The lymphocyte crossmatch is currently the only cell-based compatibility assay performed by histocompatibility laboratories for transplant purposes. While in many transplant programs the complement-dependent cytotoxicity crossmatch (CDCXM) remains in use, when available, the flow cytometry crossmatch (FCXM) is the method of choice because of its superior sensitivity and specificity. Unfortunately, the maintenance and cost of a flow cytometer is a considerable limitation for small histocompatibility laboratories. Therefore, in this study, we evaluated the use of the Cellometer Vision CBA image cytometer (Nexcelom Bioscience LLC, Lawrence, Massachusetts) as an alternative instrument to perform the crossmatch assay. The 3-color FCXM protocol was modified into two separate 2-color panel image cytometry crossmatches (IXMs), one for T cells and one for B cells. After initial serum and cell incubation, a cocktail consisting of PE/Cy5-conjugated anti-human CD3 or CD19 and PE-conjugated anti-human IgG F(ab')₂ was added to the T cell and B cell panels, respectively. The final cell preparation was added to a separate counting chamber. Images were captured using the Cellometer Vision CBA, an image cytometer designed for cell counting, size analysis and fluorescence intensity measurement. Thirty-nine IXMs were performed and compared with the FCXM. We obtained a concordance sensitivity of 94.1% and 100% and specificity of 100% and 88.9% for T cells and B cells, respectively. The linearity of the system was verified using dilutions of a sample containing known donor-specific anti-HLA antibodies (DSA) against the target cells. This feasibility study demonstrates that the FCXM test could be easily adapted to the Cellometer Vision CBA image cytometer without compromising specificity and sensitivity. The low instrumentation cost, minimal maintenance, and simple operation allow for efficient implementation or transition from the FCXM to the IXM method.

Performance evaluation of an automated image-based fluorescence CD4+ cell analyzer

BACKGROUND

Although AIDS-related mortality has declined since the introduction of antiretroviral therapy (ART), HIV/ AIDS patients are predominantly present in developing countries that lack high-cost diagnostic devices and human expertise.

OBJECTIVE

New methods for counting CD4+ cells cost-effectively are needed to replace conventional flow cytometry-based diagnosis.

METHODS

We developed a CD4+ cell analyzer, ADAMII, which is a benchtop fluorescence image-based CD3+/4+ cell counting analyzer. It bears a three-channel light source and performs CD3+/4+ counting assays. The automatic 3D stage captures a maximum of 136 images that are subsequently processed and analyzed using a software integrated into the system.

RESULTS

Results obtained using ADAMII were compared with data obtained by conventional methods using a FACSCalibur flow cytometer and the point-of-care PIMA CD4 analyzer. Both comparisons between ADAMII vs. FACS and ADAMII vs. PIMA data yielded a strong correlation with an R2 value of 0.98, which ensures the feasibility of CD4 test by ADAMII.

CONCLUSIONS

The proposed method using ADAMII can be easily employed in resource-limited areas to replace conventional flow cytometers, which are expensive and require highly trained staff.

Image cytometry DNA ploidy analysis: Correlation between two semi-automated methods

BACKGROUND AND OBJECTIVES

Quantitation of cell DNA content, DNA ploidy, has been established as a research and prognostic technique for decades. A variety of instruments have been used although only a few commercially available systems have established quality assurance and published outcome data. The aim of this study was to compare two automated systems.

METHODS

Nuclear monolayers were obtained from 112 oral biopsies by enzyme digestion and Feulgen staining. These were scanned on both the Fairfield and the Ploidy Work Station (PWS) systems. The overall ploidy diagnosis, number of epithelial nuclei, coefficient of variation (CV) and 5c exceeding rate (5CER) were compared by quantile-quantile plots, t test, Wilcoxon and Spearman's tests.

RESULTS

The PWS system identified more nuclei (p < 0.0001) at a lower CV (p < 0.0001). Using the PWS system, fewer samples were classified as indeterminate. No difference between 5CER was found between systems (p > 0.54). There was complete concordance between the two systems in terms of DNA ploidy diagnosis.

CONCLUSIONS

The PWS system is comparable to the Fairfield system for determination of DNA ploidy and has advantages that may lead to improved performance.

An Experimental Insight into Extracellular Phosphatases - Differential Induction of Cell-Specific Activity in Green Algae Cultured under Various Phosphorus Conditions

Extracellular phosphatase activity (PA) has been used as an overall indicator of P depletion in lake phytoplankton. However, detailed insights into the mechanisms of PA regulation are still limited, especially in the case of acid phosphatases. The novel substrate ELF97 phosphate allows for tagging PA on single cells in an epifluorescence microscope. This fluorescence-labeled enzyme activity (FLEA) assay enables for autecological studies in natural phytoplankton and algal cultures. We combined the FLEA assay with image analysis to measure cell-specific acid PA in two closely related species of the genus Coccomyxa (Trebouxiophyceae, Chlorophyta) isolated from two acidic lakes with distinct P availability. The strains were cultured in a mineral medium supplied with organic (beta-glycerol phosphate) or inorganic (orthophosphate) P at three concentrations. Both strains responded to experimental conditions in a similar way, suggesting that acid extracellular phosphatases were regulated irrespectively of the origin and history of the strains. We found an increase in cell-specific PA at low P concentration and the cultures grown with organic P produced significantly higher (ca. 10-fold) PA than those cultured with the same concentrations of inorganic P. The cell-specific PA measured in the cultures grown with the lowest organic P concentration roughly corresponded to those of the original Coccomyxa population from an acidic lake with impaired P availability. The ability of Coccomyxa strains to produce extracellular phosphatases, together with tolerance for both low pH and metals can be one of the factors enabling the dominance of the genus in extreme conditions of acidic lakes. The analysis of frequency distribution of the single-cell PA documented that simple visual counting of ‘active’ (labeled) and ‘non-active’ (non-labeled) cells can lead to biased conclusions regarding algal P status because the actual PA of the ‘active’ cells can vary from negligible to very high values. The FLEA assay using image cytometry offers a strong tool in plankton ecology for exploring P metabolism.

Quantitative Multiplex Immunohistochemistry Reveals Myeloid-Inflamed Tumor-Immune Complexity Associated with Poor Prognosis

Here, we describe a multiplexed immunohistochemical platform with computational image processing workflows, including image cytometry, enabling simultaneous evaluation of 12 biomarkers in one formalin-fixed paraffin-embedded tissue section. To validate this platform, we used tissue microarrays containing 38 archival head and neck squamous cell carcinomas and revealed differential immune profiles based on lymphoid and myeloid cell densities, correlating with human papilloma virus status and prognosis. Based on these results, we investigated 24 pancreatic ductal adenocarcinomas from patients who received neoadjuvant GVAX vaccination and revealed that response to therapy correlated with degree of mono-myelocytic cell density and percentages of CD8⁺ T cells expressing T cell exhaustion markers. These data highlight the utility of in situ immune monitoring for patient stratification and provide digital image processing pipelines to the community for examining immune complexity in precious tissue sections, where phenotype and tissue architecture are preserved to improve biomarker discovery and assessment.

Statistical performance of image cytometry for DNA, lipids, cytokeratin, & CD45 in a model system for circulation tumor cell detection

Detection of circulating tumor cells (CTCs) in a blood sample is limited by the sensitivity and specificity of the biomarker panel used to identify CTCs over other blood cells. In this work, we present Bayesian theory that shows how test sensitivity and specificity set the rarity of cell that a test can detect. We perform our calculation of sensitivity and specificity on our image cytometry biomarker panel by testing on pure disease positive (D⁺ ) populations (MCF7 cells) and pure disease negative populations (D^- ) (leukocytes). In this system, we performed multi-channel confocal fluorescence microscopy to image biomarkers of DNA, lipids, CD45, and Cytokeratin. Using custom software, we segmented our confocal images into regions of interest consisting of individual cells and computed the image metrics of total signal, second spatial moment, spatial frequency second moment, and the product of the spatial-spatial frequency moments. We present our analysis of these 16 features. The best performing of the 16 features produced an average separation of three standard deviations between D⁺ and D^- and an average detectable rarity of ∼1 in 200. We performed multivariable regression and feature selection to combine multiple features for increased performance and showed an average separation of seven standard deviations between the D⁺ and D^- populations making our average detectable rarity of ∼1 in 480. Histograms and receiver operating characteristics (ROC) curves for these features and regressions are presented. We conclude that simple regression analysis holds promise to further improve the separation of rare cells in cytometry applications. © 2017 International Society for Advancement of Cytometry.

Spatial relationship between chromosomal domains in diploid and autotetraploid Arabidopsis thaliana nuclei

Polyploids constitute more than 80% of angiosperm plant species. Their DNA content is often further increased by endoreplication, which occurs as a part of cell differentiation. Here, we explore the relationship between 3D chromatin architecture, number of genome copies and their origin in the model plant, Arabidopsis thaliana. Spatial proximity between pericentromeric, interstitial and subtelomeric domains of chromosomes 1 and 4 was quantified over a range of distances. The results indicate that average nuclear volume as well as chromatin density increase with the genome copy number. Similar dependence is observed when association of homologous chromosomes (in 2C/ endopolyploid nuclei) and sister chromatid separation (in endopolyploid nuclei) is studied. Moreover, clusters of chromosomal domains are detectable at the spatial scale above microscopy resolution. Subtelomeric, interstitial and pericentromeric chromosomal domains are affected to different extent by these processes, which are modulated by endopolyploidy. This factor influences fusion of heterochromatin as well. Nonetheless, local chromatin architecture of Arabidopsis thaliana depends mainly on endopolyploidy level, and to lesser extend on polyploidy.

Robust microbial cell segmentation by optical-phase thresholding with minimal processing requirements

High-throughput imaging with single-cell resolution has enabled remarkable discoveries in cell physiology and Systems Biology investigations. A common, and often the most challenging step in all such imaging implementations, is the ability to segment multiple images to regions that correspond to individual cells. Here, a robust segmentation strategy for microbial cells using Quantitative Phase Imaging is reported. The proposed method enables a greater than 99% yeast cell segmentation success rate, without any computationally-intensive, post-acquisition processing. We also detail how the method can be expanded to bacterial cell segmentation with 98% success rates with substantially reduced processing requirements in comparison to existing methods. We attribute this improved performance to the remarkably uniform background, elimination of cell-to-cell and intracellular optical artifacts, and enhanced signal-to-background ratio-all innate properties of imaging in the optical-phase domain. © 2017 International Society for Advancement of Cytometry.

Large population cell characterization using quantitative phase cytometer

A major challenge in cellular analysis is the phenotypic characterization of large cell populations within a short period of time. Among various parameters for cell characterization, the cell dry mass is often used to describe cell size but is difficult to be measured directly with traditional techniques. Here, we propose an interferometric approach based on line-focused beam illumination for high-content precision dry mass measurements of adherent cells in a non-invasive fashion-we call it quantitative phase cytometry (QPC). Besides dry mass, abundant cellular morphological features such as projected area, sphericity, and phase skewness can be readily extracted from the QPC interferometric data. To validate the utility of our technique, we demonstrate characterizing a large population of ∼10⁴ HeLa cells. Our reported QPC system is envisioned as a promising quantitative tool for label-free characterization of a large cell count at single cell resolution. © 2017 International Society for Advancement of Cytometry.

Label-free high temporal resolution assessment of cell proliferation using digital holographic microscopy

Cell proliferation assays are widely applied in biological sciences to understand the effect of drugs over time. However, current methods often assess cell population growth indirectly, that is, the cells are not actually counted. Instead other parameters, for example, the amount of protein, are determined. These methods often also demand phototoxic labels, have low temporal resolution, or employ end-point assays, and frequently are labor intensive. We have developed a robust and label-free kinetic cell proliferation assay with high temporal resolution for adherent cells using digital holographic microscopy (DHM), one of many quantitative phase microscopy techniques. As no labels or stains are required, and only very low intensity illumination is necessary, the technique allows for noninvasive continuous cell counting. Only two image processing settings were adjusted between cell lines, making the assay practical, user friendly, and free of user bias. The developed direct assay was validated by analyzing cell cultures treated with various concentrations of the anti-cancer drug etoposide, a well-established topoisomerase inhibitor that causes DNA damage and leads to programmed cell death. After treatment, the unstained adherent cells were nondestructively imaged every 30 min for 36 h inside a cell incubator. In the recorded time-lapse image sequences, individual cells were automatically identified to provide detailed growth curves and growth rate data of cell number, confluence, and average cell volume. Our results demonstrate how these parameters facilitate a deeper understanding of cell processes than what is achievable with current single-parameter and end-point methods. © 2017 International Society for Advancement of Cytometry.

DNA ploidy and cell cycle protein expression in oral squamous cell carcinomas with and without lymph node metastases

BACKGROUND

Oral squamous cell carcinoma (OSCC) is the most frequently occurring malignant tumour in the oral cavity. OSCC arises because of multiple genetic alterations. Cell cycle aberrations and aneuploidy are reportedly among the main characteristics of cancer cells and are associated with aggressive growth and poor prognosis.

METHODS

The study sample included 47 non-metastasised and 39 metastasised primary OSCC, with matched positive cervical lymph nodes and 17 normal oral mucosa samples. Tissue microarrays (TMAs) were prepared with a minimum of three cores from each case. TMA sections were cut and immunostained with MCM2, Ki-67, geminin and cyclin D1 antibodies. DNA image analysis was performed on the whole tissue section before TMAs were created.

RESULTS

The results revealed that there were no differences in cell cycle protein expression in different areas of the tumours or between the metastatic and non-metastatic carcinomas. None of the cell cycle proteins showed significant differences between the lymph node metastasis and the primary OSCC, except for Ki-67. Geminin/Ki-67 ratio showed significant difference between metastatic and non-metastatic tumours. Aneuploidy was detected in all (100%) cases of OSCC. Similarly, all lymph node samples (39 cases) were aneuploid.

CONCLUSION

The results suggest that although there was dysregulation of cell cycle regulatory proteins, only Ki-67 and the MCM2/Ki-67 and geminin/Ki-67 ratios may have prognostic significance in oral cancer. DNA ploidy alone was not specific and may not be a good tool to evaluate prognosis or metastatic progression in oral cavity carcinomas.

A Novel Multiparametric Drug-Scoring Method for High-Throughput Screening of 3D Multicellular Tumor Spheroids Using the Celigo Image Cytometer

Three-dimensional (3D) tumor models have been increasingly used to investigate and characterize cancer drug compounds. The ability to perform high-throughput screening of 3D multicellular tumor spheroids (MCTS) can highly improve the efficiency and cost-effectiveness of discovering potential cancer drug candidates. Previously, the Celigo Image Cytometer has demonstrated a novel method for high-throughput screening of 3D multicellular tumor spheroids. In this work, we employed the Celigo Image Cytometer to examine the effects of 14 cancer drug compounds on 3D MCTS of the glioblastoma cell line U87MG in 384-well plates. Using parameters such as MCTS diameter and invasion area, growth and invasion were monitored for 9 and 3 d, respectively. Furthermore, fluorescent staining with calcein AM, propidium iodide, Hoechst 33342, and caspase 3/7 was performed at day 9 posttreatment to measure viability and apoptosis. Using the kinetic and endpoint data generated, we created a novel multiparametric drug-scoring system for 3D MCTS that can be used to identify and classify potential drug candidates earlier in the drug discovery process. Furthermore, the combination of quantitative and qualitative image data can be used to delineate differences between drugs that induce cytotoxic and cytostatic effects. The 3D MCTS-based multiparametric scoring method described here can provide an alternative screening method to better qualify tested drug compounds.

Connected component masking accurately identifies the ratio of phagocytosed and cell-bound particles in individual cells by imaging flow cytometry

Innate immune cell-mediated recognition, capture, and engulfment of large particulate targets such as bacteria is known as phagocytosis. This highly dynamic cellular process involves a series of steps including receptor-mediated target binding, phagocytic cup formation, pseudopod extension, and phagosome closure, which depend on distinct actin polymerization events. Using flow cytometry, precise determination of target locations relative to cell membranes (i.e., surface-bound vs. fully engulfed/internalized) during the phagocytic process is difficult to quantify. Here, we describe the application of new analysis features within the IDEAS® software to distinguish internalized and surface-bound particles on individual cells with a high degree of accuracy and reproducibility. Through the use of connected component masks, the accurate discrimination of surface-bound beads versus those internalized is clearly demonstrated. In addition, we were able to further analyze the ratio of beads that had been surface-bound or internalized within individual cells. This novel method of analyzing the phagocytic process provides more accurate determination of target-cell interactions that will assist in examination of the signalling events that occur during the various stages of phagocytosis. © 2017 International Society for Advancement of Cytometry.

Virtual cell imaging: A review on simulation methods employed in image cytometry

The simulations of cells and microscope images thereof have been used to facilitate the development, selection, and validation of image analysis algorithms employed in cytometry as well as for modeling and understanding cell structure and dynamics beyond what is visible in the eyepiece. The simulation approaches vary from simple parametric models of specific cell components-especially shapes of cells and cell nuclei-to learning-based synthesis and multi-stage simulation models for complex scenes that simultaneously visualize multiple object types and incorporate various properties of the imaged objects and laws of image formation. This review covers advances in artificial digital cell generation at scales ranging from particles up to tissue synthesis and microscope image simulation methods, provides examples of the use of simulated images for various purposes ranging from subcellular object detection to cell tracking, and discusses how such simulators have been validated. Finally, the future possibilities and limitations of simulation-based validation are considered. © 2016 International Society for Advancement of Cytometry.

Trypan blue as an affordable marker for automated live-dead cell analysis in image cytometry

The aim of the present study was to combine image cytometry and trypan blue (TB) exclusion staining for a reproducible high-throughput detection of dead cells, enabling TB as an inexpensive marker, to be affordable for many studies and creating the possibility to combine fluorochromes without or with less spectral overlap. Capillary blood was drawn from a healthy volunteer, red blood cells were lysed and leukocyte cell death was induced. Samples were stained with CD45-FITC, CD14-PE, TB and DAPI, and then analyzed using image cytometry (iCys). TB quenching control tests were performed using DAPI and CD45-FITC. Images were generated in .TIF and .JPEG format using iCys image cytometer. The images were analyzed using CellProfiler (CP) modules to optimize the analysis based on the aims of each phase of this study. CellProfiler Analyst (CPA) was used to classify cells throughout machine learning and to calculate sensibility of the classification. A sensitivity of 0.94 for dead cells and 0.99 for live cells was calculated using CPA. We did not see any quenching effects of the FITC staining. DAPI signal was reduced in the presence of TB. The results of the present study revealed that TB serves as a dead cell marker in an image cytometric analysis, being able to be combined with other fluorescence markers without loss of fluorescence intensity signal or overlapping emission spectrum. SCANNING 38:857-863, 2016. © 2016 Wiley Periodicals, Inc.

High-Throughput 3D Tumor Spheroid Screening Method for Cancer Drug Discovery Using Celigo Image Cytometry

Oncologists have investigated the effect of protein or chemical-based compounds on cancer cells to identify potential drug candidates. Traditionally, the growth inhibitory and cytotoxic effects of the drugs are first measured in 2D in vitro models, and then further tested in 3D xenograft in vivo models. Although the drug candidates can demonstrate promising inhibitory or cytotoxicity results in a 2D environment, similar effects may not be observed under a 3D environment. In this work, we developed an image-based high-throughput screening method for 3D tumor spheroids using the Celigo image cytometer. First, optimal seeding density for tumor spheroid formation was determined by investigating the cell seeding density of U87MG, a human glioblastoma cell line. Next, the dose-response effects of 17-AAG with respect to spheroid size and viability were measured to determine the IC₅₀ value. Finally, the developed high-throughput method was used to measure the dose response of four drugs (17-AAG, paclitaxel, TMZ, and doxorubicin) with respect to the spheroid size and viability. Each experiment was performed simultaneously in the 2D model for comparison. This detection method allowed for a more efficient process to identify highly qualified drug candidates, which may reduce the overall time required to bring a drug to clinical trial.

Quantitative Assessment of Retinopathy Using Multi-parameter Image Analysis

A multi-parameter quantification method was implemented to quantify retinal vascular injuries in microscopic images of clinically relevant eye diseases. This method was applied to wholemount retinal trypsin digest images of diabetic Akita/+, and bcl-2 knocked out mice models. Five unique features of retinal vasculature were extracted to monitor early structural changes and retinopathy, as well as quantifying the disease progression. Our approach was validated through simulations of retinal images. Results showed fewer number of cells (P = 5.1205e-05), greater population ratios of endothelial cells to pericytes (PCs) (P = 5.1772e-04; an indicator of PC loss), higher fractal dimension (P = 8.2202e-05), smaller vessel coverage (P = 1.4214e-05), and greater number of acellular capillaries (P = 7.0414e-04) for diabetic retina as compared to normal retina. Quantification using the present method would be helpful in evaluating physiological and pathological retinopathy in a high-throughput and reproducible manner.

The DNA index as a prognostic tool in hilar cholangiocarcinoma

BACKGROUND AND OBJECTIVES

Due to the devastating prognosis of patients suffering from hilar cholangiocarcinoma (HCCA) valid prognostic factors are urgently needed to guide treatment decisions in a personalized concept. The aim of this study was to analyze the predictive value of the DNA index in a large single-center cohort of patients undergoing resection of HCCA.

METHODS

A total of 154 patients who underwent resection of HCCA were included in this prospective study. The DNA index was assessed by image cytometry of fresh tumor samples and correlated, as well as standard histopathological parameters, with patient survival.

RESULTS

The median DNA index was 1.61 ± 0.32. Univariate survival analysis identified eight parameters including DNA index, but not DNA ploidy as prognostic markers. In the Cox proportional hazard model DNA index (P = 0.021), tumor size (P = 0.029) and lymph nodes status (P = 0.039) could be shown to be independent predictors of patient survival.

CONCLUSION

The DNA index represents an independent prognostic marker in HCCA which is superior to most standard histopathological factors. Since the DNA index can be assessed not only post- but also preoperatively, it might be a potential tool in the preoperative decision-making process.

Adaptation of image cytometry methodology for DNA ploidy analysis of cervical epithelium samples: a pilot study

OBJECTIVE

To determine DNA ploidy in the cervical specimens of patients revealing a suspicion of cancer by image analysis performed by using a combination of commercial analysis software, conventional microscopy, and certified filters.

MATERIALS AND METHODS

This study followed a prospective design. Cervical samples were obtained from 20 patients undergoing routine screening in the Gynecologic-Oncology Unit of the University Hospital of the Federal University of Minas Gerais, Brazil. Three slides were prepared for each case and the DNA content was determined by image cytometry, post Feulgen staining. DNA ploidy, as well as events exceeding 5C and 9C, was assessed according to the guidelines and algorithms prescribed for diagnostic interpretation by the European Society for Analytical Cellular Pathology.

RESULTS

By employing the adapted tool, identification of the lesions with euploid and aneuploid profiles was possible. Abnormal DNA content was found in 65% of the cases (13/20), with 45% (9/20) presenting nuclei with >5C content and 20% (4/20) with >9C content. In the analyses conducted in this study, the coefficient of variation with respect to DNA quantity was lower than the 5% threshold recommended by the European Society for Analytical Cellular Pathology.

CONCLUSION

Image cytometry of the cervical specimens revealed DNA aneuploidy, most probably resulting from chromosomal alterations and appearing as precancerous lesions in 65% of the cases. The adaptations implemented in this study, enabled the DNA-image cytometry to become more accessible, enhancing its extended use as an adjuvant strategy for the early screening of the cervical epithelium samples during routine analyses.

High-throughput detection of DNA double-strand breaks using image cytometry

Assessment of γH2AX expression for studying DNA double-strand break formation is often performed by manual counting of foci using immunofluorescence microscopy, an approach that is laborious and subject to significant foci selection bias. Here we present a novel high-throughput method for detecting DNA double-strand breaks using automated image cytometry assessment of cell average γH2AX immunofluorescence. Our technique provides an expedient, high-throughput, objective, and cost-effective method for γH2AX analysis.

BRCA1 mutation site may be linked with nuclear DNA ploidy in BRCA1-mutated ovarian carcinomas

AIMS

BRCA1 has a role in maintaining normal nuclear DNA content during cell division and its inactivation may result in DNA aneuploidy and cancer progression. BRCA1-linked breast cancers are more aneuploid and have a worse prognosis, but this has not been elucidated in ovarian cancers. This study explores the potential difference in ploidy status between BRCA1-mutated and sporadic ovarian carcinomas. It also explores the potential association between BRCA1 mutation site and DNA ploidy status.

METHODS

This study compared DNA ploidy status of tumor blocks from 23 BRCA1-mutated ovarian carcinomas with that of 23 sporadic ovarian carcinomas matched for histologic subtype, patient age, stage and grade. DNA content of the nuclei was measured by Feulgen-Schiff staining followed by image cytometry and compared.

RESULTS

BRCA1-linked tumors with a stop codon closer to the N-terminal (between 1 and 500 aa; 6/6, 100%) had a significantly higher frequency of nondiploidy compared with those with stop codon above 500 aa (7/12, 58%) (P = 0.033). A diploid peak was detected in 28% of BRCA1-mutated ovarian cancers and in 33% of sporadic ovarian cancers.

CONCLUSIONS

The present study concluded that ovarian tumors with mutations closer to the N-terminal of BRCA1 may have a higher risk of DNA aneuploidy. There is no significant difference between BRCA1-mutated and sporadic ovarian carcinomas with respect to the DNA content.

Confocal microscopy for high-resolution and high-content analysis of the cell cycle

Optical fluorescence microscopy offers a wide range of technological solutions to address many questions in biomedical research. Spatial resolution has been greatly improved by the use of confocal microscopes, providing a 3-D analysis of the intracellular space. Automation has contributed to make confocal analysis available for high-content image cytometry studies. However, the storage, browsing, and analysis of the amount of data generated can challenge the feasibility of such studies. Presented in this chapter is a multistep acquisition and analysis protocol that can bypass such difficulties by an analysis-driven data collection. Cell-cycle analysis of low-resolution data can be employed to select cell populations of interest that can then be imaged at extremely high resolution and subjected to high-content analysis.

High-resolution cytometry for high-content cell cycle analysis

One of the major limitations of flow cytometry (FCM) is the absence of an intracellular view. Automated microscopy and image analysis, together with technological developments, led to new approaches in cytometry that bypass the above limitation, introducing high resolution, high content, and large statistical sampling. However, few attempts have been made, until now, to translate the wide repertoire of FCM assays into high-content image screening. This unit describes the implementation of an acquisition and analysis protocol for evaluation of the cell cycle by automated microscopy. The approach grants the possibility to perform simultaneous analysis of a high number of different parameters. A large part of this unit is devoted to the description of hardware features that can optimize the recorded information together with the acquisition and analysis procedures employed to produce good-quality data.

Analysis of centromere signal patterns in breast cancer cells with chromosomal instability using image cytometry combined with centromere fluorescence in situ hybridization

Fluorescence in situ hybridization (FISH) with centromeric probes is a method used to detect chromosomal instability (CIN), a hallmark of most cancers. However, no studies thus far have investigated the relationship between centromeric FISH signals and the cell cycle in cancer cells. In this study, the chromosome content in each cell cycle phase was evaluated with respect to the number of centromeric FISH signals in two breast cancer cell lines and eight surgically resected breast cancer specimens using image cytometry. Variations in chromosome number were detected at each phase of the cell cycle but were not associated with proliferative capacity in the cell lines. Furthermore, the chromosome doubling frequency differed in each cell line and clinical specimen. These results reveal two aspects of centromeric FISH signal variation in breast cancers that exhibit CIN, and suggest that chromosome doubling is a remarkable occurrence that may increase the heterogeneity of tumors.