Inter-laboratory automation of the in vitro micronucleus assay using imaging flow cytometry and deep learning

Application of image-recognition techniques to automated micronucleus detection in the in vitro micronucleus assay

BACKGROUND

An in vitro micronucleus assay is a standard genotoxicity test. Although the technique and interpretation of the results are simple, manual counting of the total and micronucleus-containing cells in a microscopic field is tedious. To address this issue, several systems have been developed for quick and efficient micronucleus counting, including flow cytometry and automated detection based on specialized software and detection systems that analyze images.

RESULTS

Here, we present a simple and effective method for automated micronucleus counting using image recognition technology. Our process involves separating the RGB channels in a color micrograph of cells stained with acridine orange. The cell nuclei and micronuclei were detected by scaling the G image, whereas the cytoplasm was recognized from a composite image of the R and G images. Finally, we identified cells with overlapping cytoplasm and micronuclei as micronucleated cells, and the application displayed the number of micronucleated cells and the total number of cells. Our method yielded results that were comparable to manually measured values.

CONCLUSIONS

Our micronucleus detection (MN/cell detection software) system can accurately detect the total number of cells and micronucleus-forming cells in microscopic images with the same level of precision as achieved through manual counting. The accuracy of micronucleus numbers depends on the cell staining conditions; however, the software has options by which users can easily manually optimize parameters such as threshold, denoise, and binary to achieve the best results. The optimization process is easy to handle and requires less effort, making it an efficient way to obtain accurate results.

The buccal micronucleus cytome assay: New horizons for its implementation in human studies

In this report we provide a summary of the presentations and discussion of the latest knowledge regarding the buccal micronucleus (MN) cytome assay. This information was presented at the HUMN workshop held in Malaga, Spain, in connection with the 2023 European, Environmental Mutagenesis and Genomics conference. The presentations covered the most salient topics relevant to the buccal MN cytome assay including (i) the biology of the buccal mucosa, (ii) its application in human studies relating to DNA damage caused by environmental exposure to genotoxins, (iii) the association of buccal MN with cancer and a wide range of reproductive, metabolic, immunological, neurodegenerative and other age-related diseases, (iv) the impact of nutrition and lifestyle on buccal MN cytome assay biomarkers; (v) its potential for application to studies of DNA damage in children and obesity, and (vi) the growing prospects of enhancing the clinical utility by automated scoring of the buccal MN cytome assay biomarkers by image recognition software developed using artificial intelligence. The most important knowledge gap is the need of prospective studies to test whether the buccal MN cytome assay biomarkers predict health and disease.

Interpretation of in vitro concentration-response data for risk assessment and regulatory decision-making: Report from the 2022 IWGT quantitative analysis expert working group meeting

Quantitative risk assessments of chemicals are routinely performed using in vivo data from rodents; however, there is growing recognition that non-animal approaches can be human-relevant alternatives. There is an urgent need to build confidence in non-animal alternatives given the international support to reduce the use of animals in toxicity testing where possible. In order for scientists and risk assessors to prepare for this paradigm shift in toxicity assessment, standardization and consensus on in vitro testing strategies and data interpretation will need to be established. To address this issue, an Expert Working Group (EWG) of the 8th International Workshop on Genotoxicity Testing (IWGT) evaluated the utility of quantitative in vitro genotoxicity concentration-response data for risk assessment. The EWG first evaluated available in vitro methodologies and then examined the variability and maximal response of in vitro tests to estimate biologically relevant values for the critical effect sizes considered adverse or unacceptable. Next, the EWG reviewed the approaches and computational models employed to provide human-relevant dose context to in vitro data. Lastly, the EWG evaluated risk assessment applications for which in vitro data are ready for use and applications where further work is required. The EWG concluded that in vitro genotoxicity concentration-response data can be interpreted in a risk assessment context. However, prior to routine use in regulatory settings, further research will be required to address the remaining uncertainties and limitations.

Development of a micronucleus test using the EpiAirway™ organotypic human airway model

BACKGROUND

The use of organotypic human tissue models in genotoxicity has increased as an alternative to animal testing. Genotoxicity is generally examined using a battery of in vitro assays such as Ames and micronucleus (MN) tests that cover gene mutations and structural and numerical chromosome aberrations. At the 7th International Workshop on Genotoxicity Testing, working group members agreed that the skin models have reached an advanced stage of maturity, while further efforts in liver and airway models are needed [Pfuhler et al., Mutat. Res. 850-851 (2020) 503135]. Organotypic human airway model is composed of fully differentiated and functional respiratory epithelium. However, because cell proliferation in organotypic airway models is thought to be less active, assessing their MN-inducing potential is an issue, even in the cytokinesis-blocking approach using cytochalasin B (CB) [Wang et al., Environ. Mol. Mutagen. 62 (2021) 306-318]. Here, we developed a MN test using EpiAirway™ in which epidermal growth factor (EGF) was included as a stimulant of cell division.

RESULTS

By incubating EpiAirway™ tissue with medium containing various concentrations of CB, we found that the percentage of binucleated cells (%BNCs) almost plateaued at 3 μg/mL CB for 72 h incubation. Additionally, we confirmed that EGF stimulation with CB incubation produced an additional increase in %BNCs with a peak at 5 ng/mL EGF. Transepithelial electrical resistance measurement and tissue histology revealed that CB incubation caused the reduced barrier integrity and cyst formation in EpiAirway™. Adenylate kinase assay confirmed that the cytotoxicity increased with each day of culture in the CB incubation period with EGF stimulation. These results indicated that chemical treatment should be conducted prior to CB incubation. Under these experimental conditions, it was confirmed that the frequency of micronucleated cells was dose-dependently increased by apical applications of two clastogens, mitomycin C and methyl methanesulfonate, and an aneugen, colchicine, at the subcytotoxic concentrations assessed in %BNCs.

CONCLUSIONS

Well-studied genotoxicants demonstrated capability in an organotypic human airway model as a MN test system. For further utilization, investigations of aerosol exposure, repeating exposure protocol, and metabolic activation are required.

Machine learning approach for quantitative biodosimetry of partial-body or total-body radiation exposures by combining radiation-responsive biomarkers

During a large-scale radiological event such as an improvised nuclear device detonation, many survivors will be shielded from radiation by environmental objects, and experience only partial-body irradiation (PBI), which has different consequences, compared with total-body irradiation (TBI). In this study, we tested the hypothesis that applying machine learning to a combination of radiation-responsive biomarkers (ACTN1, DDB2, FDXR) and B and T cell counts will quantify and distinguish between PBI and TBI exposures. Adult C57BL/6 mice of both sexes were exposed to 0, 2.0-2.5 or 5.0 Gy of half-body PBI or TBI. The random forest (RF) algorithm trained on ½ of the data reconstructed the radiation dose on the remaining testing portion of the data with mean absolute error of 0.749 Gy and reconstructed the product of dose and exposure status (defined as 1.0 × Dose for TBI and 0.5 × Dose for PBI) with MAE of 0.472 Gy. Among irradiated samples, PBI could be distinguished from TBI: ROC curve AUC = 0.944 (95% CI: 0.844-1.0). Mouse sex did not significantly affect dose reconstruction. These results support the hypothesis that combinations of protein biomarkers and blood cell counts can complement existing methods for biodosimetry of PBI and TBI exposures.

Towards AI-driven longevity research: An overview

While in the past technology has mostly been utilized to store information about the structural configuration of proteins and molecules for research and medical purposes, Artificial Intelligence is nowadays able to learn from the existing data how to predict and model properties and interactions, revealing important knowledge about complex biological processes, such as aging. Modern technologies, moreover, can rely on a broader set of information, including those derived from the next-generation sequencing (e.g., proteomics, lipidomics, and other omics), to understand the interactions between human body and the external environment. This is especially relevant as external factors have been shown to have a key role in aging. As the field of computational systems biology keeps improving and new biomarkers of aging are being developed, artificial intelligence promises to become a major ally of aging research.

Imaging flow cytometry: a primer

Imaging flow cytometry combines the high throughput nature of flow cytometry with the advantages of single cell image acquisition associated with microscopy. The measurement of large numbers of features from the resulting images provides rich datasets which have resulted in a wide range of novel biomedical applications. In this primer we discuss the typical imaging flow instrumentation, the form of data acquired and the typical analysis tools that can be applied to this data. Using examples from the literature we discuss the progression of the analysis methods that have been applied to imaging flow cytometry data. These methods start from the use of simple single image features and multiple channel gating strategies, followed by the design and use of custom features for phenotype classification, through to powerful machine and deep learning methods. For each of these methods, we outline the processes involved in analyzing typical datasets and provide details of example applications. Finally we discuss the current limitations of imaging flow cytometry and the innovations which are addressing these challenges.

Combining imaging flow cytometry and machine learning for high-throughput schistocyte quantification: A SVM classifier development and external validation cohort

Background

Schistocyte counts are a cornerstone of the diagnosis of thrombotic microangiopathy syndrome (TMA). Their manual quantification is complex and alternative automated methods suffer from pitfalls that limit their use. We report a method combining imaging flow cytometry (IFC) and artificial intelligence for the direct label-free and operator-independent quantification of schistocytes in whole blood.

Methods

We used 135,045 IFC images from blood acquisition among 14 patients to extract 188 features with IDEAS® software and 128 features from a convolutional neural network (CNN) with Keras framework in order to train a support vector machine (SVM) blood elements’ classifier used for schistocytes quantification.

Finding

Keras features showed better accuracy (94.03%, CI: 93.75-94.31%) than ideas features (91.54%, CI: 91.21-91.87%) in recognising whole-blood elements, and together they showed the best accuracy (95.64%, CI: 95.39-95.88%). We obtained an excellent correlation (0.93, CI: 0.90-0.96) between three haematologists and our method on a cohort of 102 patient samples. All patients with schistocytosis (>1% schistocytes) were detected with excellent specificity (91.3%, CI: 82.0-96.7%) and sensitivity (100%, CI: 89.4-100.0%). We confirmed these results with a similar specificity (91.1%, CI: 78.8-97.5%) and sensitivity (100%, CI: 88.1-100.0%) on a validation cohort (n=74) analysed in an independent healthcare centre. Simultaneous analysis of 16 samples in both study centres showed a very good correlation between the 2 imaging flow cytometers (Y=1.001x).

Interpretation

We demonstrate that IFC can represent a reliable tool for operator-independent schistocyte quantification with no pre-analytical processing which is of most importance in emergency situations such as TMA.

Funding

None.

CRISPRthripsis: The Risk of CRISPR/Cas9-induced Chromothripsis in Gene Therapy

The Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/Cas9 nuclease system has allowed the generation of disease models and the development of therapeutic approaches for many genetic and non-genetic disorders. However, the generation of large genomic rearrangements has raised safety concerns for the clinical application of CRISPR/Cas9 nuclease approaches. Among these events, the formation of micronuclei and chromosome bridges due to chromosomal truncations can lead to massive genomic rearrangements localized to one or few chromosomes. This phenomenon, known as chromothripsis, was originally described in cancer cells, where it is believed to be caused by defective chromosome segregation during mitosis or DNA double-strand breaks. Here, we will discuss the factors influencing CRISPR/Cas9-induced chromothripsis, hereafter termed CRISPRthripsis, and its outcomes, the tools to characterize these events and strategies to minimize them.

Detection and Characterization of Circulating Tumor Cells Using Imaging Flow Cytometry—A Perspective Study

Simple Summary

Liquid biopsy is non-invasive approach used to prognose and monitor tumor progression based on the detection and examination of metastasis-related events found in the patients’ blood (such as circulating tumor cells (CTCs), extracellular vesicles, and circulating nucleic acids). Different ultrasensitive techniques are applied to study those events and the biology of tumor dissemination, which in the future might complement standard diagnostics. Here, we suggest that CTCs analysis could be improved by the usage of imaging flow cytometry, combining advantages of both standard flow cytometry (high-scale analysis) and microscopy (high resolution) to investigate detailed features of those cells. From this perspective, we discuss the potential of this technology in the CTC field and present representative images of CTCs from breast and prostate cancer patients analyzed with this method.

Abstract

Tumor dissemination is one of the most-investigated steps of tumor progression, which in recent decades led to the rapid development of liquid biopsy aiming to analyze circulating tumor cells (CTCs), extracellular vesicles (EVs), and circulating nucleic acids in order to precisely diagnose and monitor cancer patients. Flow cytometry was considered as a method to detect CTCs; however, due to the lack of verification of the investigated cells’ identity, this method failed to reach clinical utility. Meanwhile, imaging flow cytometry combining the sensitivity and high throughput of flow cytometry and image-based detailed analysis through a high-resolution microscope might open a new avenue in CTC technologies and provide an open-platform system alternative to CellSearch^®, which is still the only gold standard in this field. Hereby, we shortly review the studies on the usage of flow cytometry in CTC identification and present our own representative images of CTCs envisioned by imaging flow cytometry providing rationale that this novel technology might be a good tool for studying tumor dissemination, and, if combined with a high CTC yield enrichment method, could upgrade CTC-based diagnostics.

Rapid and automatic detection of micronuclei in binucleated lymphocytes image

Cytokinesis block micronucleus (CBMN) assay is a widely used radiation biological dose estimation method. However, the subjectivity and the time-consuming nature of manual detection limits CBMN for rapid standard assay. The CBMN analysis is combined with a convolutional neural network to create a software for rapid standard automated detection of micronuclei in Giemsa stained binucleated lymphocytes images in this study. Cell acquisition, adhesive cell mass segmentation, cell type identification, and micronucleus counting are the four steps of the software's analysis workflow. Even when the cytoplasm is hazy, several micronuclei are joined to each other, or micronuclei are attached to the nucleus, this algorithm can swiftly and efficiently detect binucleated cells and micronuclei in a verification of 2000 images. In a test of 20 slides, the software reached a detection rate of 99.4% of manual detection in terms of binucleated cells, with a false positive rate of 14.7%. In terms of micronuclei detection, the software reached a detection rate of 115.1% of manual detection, with a 26.2% false positive rate. Each image analysis takes roughly 0.3 s, which is an order of magnitude faster than manual detection.

Recent progress in cytometric technologies and their applications in ecotoxicology and environmental risk assessment

Environmental toxicology focuses on identifying and predicting impact of potentially toxic anthropogenic chemicals on biosphere at various levels of biological organization. Presently there is a significant drive to gain deeper understanding of cellular and sub-cellular mechanisms of ecotoxicity. Most notable is increased focus on elucidation of cellular-response networks, interactomes, and greater implementation of cell-based biotests using high-throughput procedures, while at the same time decreasing the reliance on standard animal models used in ecotoxicity testing. This is aimed at discovery and interpretation of molecular pathways of ecotoxicity at large scale. In this regard, the applications of cytometry are perhaps one of the most fundamental prospective analytical tools for the next generation and high-throughput ecotoxicology research. The diversity of this modern technology spans flow, laser-scanning, imaging, and more recently, Raman as well as mass cytometry. The cornerstone advantages of cytometry include the possibility of multi-parameter measurements, gating and rapid analysis. Cytometry overcomes, thus, limitations of traditional bulk techniques such as spectrophotometry or gel-based techniques that average the results from pooled cell populations or small model organisms. Novel technologies such as cell imaging in flow, laser scanning cytometry, as well as mass cytometry provide innovative and tremendously powerful capabilities to analyze cells, tissues as well as to perform in situ analysis of small model organisms. In this review, we outline cytometry as a tremendously diverse field that is still vastly underutilized and often largely unknown in environmental sciences. The main motivation of this work is to highlight the potential and wide-reaching applications of cytometry in ecotoxicology, guide environmental scientists in the technological aspects as well as popularize its broader adoption in environmental risk assessment.