High-content Screening Technology Research Articles

Repurposing of a library for high-content screening of inhibitors against Echinococcus granulosus

BackgroundCystic echinococcosis (CE) is a zoonotic disease caused by the larval stage of the dog tapeworm Echinococcus granulosus sensu lato (E. granulosus), with a worldwide distribution. The current treatment strategy for CE is insufficient. Limited drug screening models severely hamper the discovery of effective anti-echinococcosis drugs.MethodsIn the present study, using high-content screening technology, we developed a novel high-throughput screening (HTS) assay by counting the ratio of propidium iodide-stained dead protoscoleces (PSCs) to the total number of PSCs. In vitro and ex vivo cyst viability assays were utilized to determine the effect of drugs on cyst viability.ResultsUsing the newly established HTS assay, we screened approximately 12,000 clinical-stage or The Food and Drug Administration (FDA)-approved small molecules from the Repurposing, Focused Rescue, and Accelerated Medchem (ReFRAME) library, as well as the LOPAC1280 and SelleckChem libraries, as a strategic approach to facilitate the drug discovery process. Initial screening yielded 173 compounds with anti-echinococcal properties, 52 of which demonstrated dose–response efficacy against E. granulosus PSCs in vitro. Notably, two agents, omaveloxolone and niclosamide, showed complete inhibition upon further validation in cyst and microcyst viability assays in vitro after incubation for 3 days, and in an ex vivo cyst viability assay using cysts isolated from the livers of mice infected with E. granulosus, as determined by morphological assessment.ConclusionsThrough the development of a novel HTS assay and by repurposing libraries, we identified omaveloxolone and niclosamide as potent inhibitors against E. granulosus. These compounds show promise as potential anti-echinococcal drugs, and our strategic approach has the potential to promote drug discovery for parasitic infections.Graphical

LP-35 Use of high-content screening technology for toxicological testing

LP-35 Use of high-content screening technology for toxicological testing

Guidelines for Application of High-Content Screening in Traditional Chinese Medicine: Concept, Equipment, and Troubleshooting

High-content screening (HCS) technology combines automated high-speed imaging hardware and single-cell quantitative analysis. It can greatly accelerate data acquisition in cellular fluorescence imaging and is a powerful research technique in traditional Chinese medicine (TCM). An increasing number of laboratories and platforms, including TCM laboratories, have begun utilizing HCS systems. However, this technology is still in its infancy in TCM research and there is a lack of sufficient experience with the associated concepts, instrument configurations, and analysis methods. To improve the understanding of HCS among researchers in the field of TCM, this paper summarizes the concept of HCS, software and hardware configuration, the overall research process, as well as common problems and related solutions of HCS in TCM research based on our team’s previous research experience, providing several research examples and an outlook on future perspectives, aiming to provide a technical guide for HCS in TCM research.

Photobiomodulation at Defined Wavelengths Regulates Mitochondrial Membrane Potential and Redox Balance in Skin Fibroblasts.

Starting from the discovery of phototherapy in the beginning of the last century, photobiomodulation (PBM) has been defined in late 1960s and, since then, widely described in different in vitro models. Robust evidence indicates that the effect of light exposure on the oxidative state of the cells and on mitochondrial dynamics, suggesting a great therapeutic potential. The translational scale-up of PBM, however, has often given contrasting and confusing results, mainly due to light exposure protocols which fail to adequately control or define factors such as emitting device features, emitted light characteristics, exposure time, cell target, and readouts. In this in vitro study, we describe the effects of a strictly controlled light-emitting diode (LED)-based PBM protocol on human fibroblasts, one of the main cells involved in skin care, regeneration, and repair. We used six emitter probes at different wavelengths (440, 525, 645, 660, 780, and 900 nm) with the same irradiance value of 0.1 mW/cm2, evenly distributed over the entire surface of the cell culture well. The PBM was analyzed by three main readouts: (i) mitochondrial potential (MitoTracker Orange staining), (ii) reactive oxygen species (ROS) production (CellROX staining); and (iii) cell death (nuclear morphology). The assay was also implemented by cell-based high-content screening technology, further increasing the reliability of the data. Different exposure protocols were also tested (one, two, or three subsequent 20 s pulsed exposures at 24 hr intervals), and the 645 nm wavelength and single exposure chosen as the most efficient protocol based on the mitochondrial potential readout, further confirmed by mitochondrial fusion quantification. This protocol was then tested for its potential to prevent H2O2-induced oxidative stress, including modulation of the light wave frequency. Finally, we demonstrated that the controlled PBM induced by the LED light exposure generates a preconditioning stimulation of the mitochondrial potential, which protects the cell from oxidative stress damage.

Linking chemicals, genes and morphological perturbations to diseases

The progress in image-based high-content screening technology has facilitated high-throughput phenotypic profiling notably the quantification of cell morphology perturbation by chemicals. However, understanding the mechanism of action of a chemical and linking it to cell morphology and phenotypes remains a challenge in drug discovery.In this study, we intended to integrate molecules that induced transcriptomic perturbations and cellular morphological changes into a biological network in order to assess chemical-phenotypic relationships in humans. Such a network was enriched with existing disease information to suggest molecular and cellular profiles leading to phenotypes.Two datasets were used for this study. Firstly, we used the “Cell Painting morphological profiling assay” dataset, composed of 30,000 compounds tested on human osteosarcoma cells (named U2OS). Secondly, we used the “L1000 mRNA profiling assay” dataset, a collection of transcriptional expression data from cultured human cells treated with approximately 20,000 bioactive small molecules from the Library of Integrated Network-based Cellular Signatures (LINCS). Furthermore, pathways, gene ontology terms and disease enrichments were performed on the transcriptomics data.Overall, our study makes it possible to develop a biological network combining chemical-gene-pathway-morphological perturbation and disease relationships. It contains an ensemble of 9989 chemicals, 732 significant morphological features and 12,328 genes. Through diverse examples, we demonstrated that some drugs shared similar genes, pathways and morphological profiles that, taken together, could help in deciphering chemical-phenotype observations.

Highly Expressing SCARA5 Promotes Proliferation and Migration of Esophageal Squamous Cell Carcinoma.

Background Thrombospondin type 1 domain-containing 7A (THSD7A) was reported to play a procancer role in esophageal squamous cell carcinoma (ESCC). The aim of the study was to screen the downstream functional genes of THSD7A and explore their functions in ESCC, based on the reported research into THSD7A function and on gene microarrays. Methods We adopted quantitative reverse-transcription polymerase chain reaction (qRT-PCR) and Celigo high-content screening (HCS) technology to screen the downstream genes of THSD7A. The expression level of target genes was examined by PCR, western blot, and immunohistochemistry (IHC). The effects of these target genes on ESCC malignant biological behavior were performed in vivo and in vitro. The Kaplan-Meier (K-M) survival analysis and Cox regression were used to analyze the prognostic significance of target genes in ESCC patients. Experiments in the literature on liver cancer (LC) were repeated to verify the functions of these genes in different tumors. We further explored the cancer-promoting mechanism of target genes in ESCC by sequencing of the genes' exons. Results Scavenger receptor class A member 5 (SCARA5) was proved to be the downstream driving gene of THSD7A. SCARA5 promoted cell proliferation and migration but inhibited apoptosis in ESCC. IHC results confirmed that SCARA5 expression in ESCC exceeded that in normal tissues. The K-M survival analysis indicated that SCARA5 expression quantity was not related to prognosis, but tumor volume and T classification were both the independent prognostic factors. Repetition of experiments in LC in the literature confirmed that SCARA5 had exactly opposite functions in EC and LC. Conclusion SCARA5 was related to the development and occurrence of ESCC. Our findings suggested that it was a potentially diagnostic individualized therapeutic target for ESCC in the future and that its application could possibly be combined with that of upstream THSD7A gene.

Advances in high-content screening applications in toxicology research

The toxicity data of chemicals and drugs increases rapidly, while the animal experimental-based tests method could not meet the current demand of health risk assessment. The high-throughput screening techniques based on in vitro alternative models, integrating with computational methods and information technology to establish toxicity tests strategy promises to address this problem. High-content screening (HCS) technology uses automated microscopy and quantitative image platforms to perform multi-parameter and high-throughput phenotypic analysis via a visualization and quantification manner, and to quickly and effectively assess toxicity and prioritization of chemicals, which promotes the development of in vitro toxicity tests and computational toxicology. HCS technology has been included as an important tool for Toxicity Testing in the 21st Century (Tox21) and chemical risk prioritization. Its applications have been widely utilized in the research field of toxicity tests and chemical toxicity mechanisms. In this review, we describe the development of HCS technology, technical points, toxicological applications, and the future directions and challenges of HCS, so as to provide references for the toxicity testing technology and risk assessment methodology.

The Ribosomal Protein L28 Gene Induces Sorafenib Resistance in Hepatocellular Carcinoma.

BackgroundSorafenib is the first molecular-targeted drug for the treatment of advanced hepatocellular carcinoma (HCC). However, its treatment efficiency decreases after a short period of time because of the development of drug resistance. This study investigates the role of key genes in regulating sorafenib-resistance and elucidates the mechanism of drug resistance in hepatocellular carcinoma.MethodsThe HCC HepG2 cells were used to generate a sorafenib-resistant cell model by culturing the cells in gradually increasing concentration of sorafenib. RNA microarray was applied to profile gene expression and screen key genes associated with sorafenib resistance. Specific targets were knockdown in sorafenib-resistant HepG2 cells for functional studies. The HCC model was established in ACI rats using Morris hepatoma3924A cells to validate selected genes associated with sorafenib resistance in vivo.ResultsThe HepG2 sorafenib-resistant cell model was successfully established. The IC50 of sorafenib was 9.988μM in HepG2 sorafenib-resistant cells. A total of 35 up-regulated genes were detected by expression profile chip. High-content screening technology was used and a potential drug-resistance related gene RPL28 was filtered out. After knocking down RPL28 in HepG2 sorafenib-resistant cells, the results of cell proliferation and apoptosis illustrated that RPL28 is the key gene involving in drug resistance. Furthermore, it was found that both RNA and protein expression of RPL28 increased in HepG2 sorafenib-resistant specimens of Morris Hepatoma rats. In addition, the expression of proliferative protein Ki-67 increased in sorafenib-resistant cells.ConclusionOur study suggested that RPL28 is a key gene inducing sorafenib resistance in HCC and could be a potential target for the treatment of drug-resistant HCC.

High-Content Screening Differentiation and Maturation Analysis of Fetal and Adult Neural Stem Cell-Derived Oligodendrocyte Precursor Cell Cultures.

The main hurdle in developing drug screening techniques for assessing the efficacy of therapeutic strategies in complex diseases is striking a balance between in vitro simplification and recreating the complex in vivo environment, along with the main aim, shared by all screening strategies, of obtaining robust and reliable data, highly predictive for in vivo translation. In the field of demyelinating diseases, the majority of drug screening strategies are based on immortalized cell lines or pure cultures of isolated primary oligodendrocyte precursor cells (OPCs) from newborn animals, leading to strong biases due to the lack of age-related differences and of any real pathological condition or complexity. Here we show the setup of an in vitro system aimed at modeling the physiological differentiation/maturation of neural stem cell (NSC)-derived OPCs, easily manipulated to mimic pathological conditions typical of demyelinating diseases. Moreover, the method includes isolation from fetal and adult brains, giving a system which dynamically differentiates from OPCs to mature oligodendrocytes (OLs) in a spontaneous co-culture which also includes astrocytes. This model physiologically resembles the thyroid hormone-mediated myelination and myelin repair process, allowing the addition of pathological interferents which model disease mechanisms. We show how to mimic the two main components of demyelinating diseases (i.e., hypoxia/ischemia and inflammation), recreating their effect on developmental myelination and adult myelin repair and taking all the cell components of the system into account throughout, while focusing on differentiating OPCs. This spontaneous mixed model, coupled with cell-based high-content screening technologies, allows the development of a robust and reliable drug screening system for therapeutic strategies aimed at combating the pathological processes involved in demyelination and at inducing remyelination.

High-Content Screening for the Detection of Drug-Induced Oxidative Stress in Liver Cells.

Drug-induced liver injury (DILI) remains a major cause of drug development failure, post-marketing warnings and restriction of use. An improved understanding of the mechanisms underlying DILI is required for better drug design and development. Enhanced reactive oxygen species (ROS) levels may cause a wide spectrum of oxidative damage, which has been described as a major mechanism implicated in DILI. Several cell-based assays have been developed as in vitro tools for early safety risk assessments. Among them, high-content screening technology has been used for the identification of modes of action, the determination of the level of injury and the discovery of predictive biomarkers for the safety assessment of compounds. In this paper, we review the value of in vitro high-content screening studies and evaluate how to assess oxidative stress induced by drugs in hepatic cells, demonstrating the detection of pre-lethal mechanisms of DILI as a powerful tool in human toxicology.

Patient-Derived Induced Pluripotent Stem Cell-Based Models in Parkinson's Disease for Drug Identification.

Parkinson’s disease (PD) is a common progressive neurodegenerative disorder characterized by loss of striatal-projecting dopaminergic neurons of the ventral forebrain, resulting in motor and cognitive deficits. Despite extensive efforts in understanding PD pathogenesis, no disease-modifying drugs exist. Recent advances in cell reprogramming technologies have facilitated the generation of patient-derived models for sporadic or familial PD and the identification of early, potentially triggering, pathological phenotypes while they provide amenable systems for drug discovery. Emerging developments highlight the enhanced potential of using more sophisticated cellular systems, including neuronal and glial co-cultures as well as three-dimensional systems that better simulate the human pathophysiology. In combination with high-throughput high-content screening technologies, these approaches open new perspectives for the identification of disease-modifying compounds. In this review, we discuss current advances and the challenges ahead in the use of patient-derived induced pluripotent stem cells for drug discovery in PD. We address new concepts implicating non-neuronal cells in disease pathogenesis and highlight the necessity for functional assays, such as calcium imaging and multi-electrode array recordings, to predict drug efficacy. Finally, we argue that artificial intelligence technologies will be pivotal for analysis of the large and complex data sets obtained, becoming game-changers in the process of drug discovery.

Using High-Content Screening to Generate Single-Cell Gene-Corrected Patient-Derived iPS Clones Reveals Excess Alpha-Synuclein with Familial Parkinson's Disease Point Mutation A30P.

The generation of isogenic induced pluripotent stem cell (iPSC) lines using CRISPR-Cas9 technology is a technically challenging, time-consuming process with variable efficiency. Here we use fluorescence-activated cell sorting (FACS) to sort biallelic CRISPR-Cas9 edited single-cell iPSC clones into high-throughput 96-well microtiter plates. We used high-content screening (HCS) technology and generated an in-house developed algorithm to select the correctly edited isogenic clones for continued expansion and validation. In our model we have gene-corrected the iPSCs of a Parkinson’s disease (PD) patient carrying the autosomal dominantly inherited heterozygous c.88G>C mutation in the SNCA gene, which leads to the pathogenic p.A30P form of the alpha-synuclein protein. Undertaking a PCR restriction-digest mediated clonal selection strategy prior to sequencing, we were able to post-sort validate each isogenic clone using a quadruple screening strategy prior to generating footprint-free isogenic iPSC lines, retaining a normal molecular karyotype, pluripotency and three germ-layer differentiation potential. Directed differentiation into midbrain dopaminergic neurons revealed that SNCA expression is reduced in the gene-corrected clones, which was validated by a reduction at the alpha-synuclein protein level. The generation of single-cell isogenic clones facilitates new insights in the role of alpha-synuclein in PD and furthermore is applicable across patient-derived disease models.

Review of high-content screening applications in toxicology.

High-content screening (HCS) technology combining automated microscopy and quantitative image analysis can address biological questions in academia and the pharmaceutical industry. Various HCS experimental applications have been utilized in the research field of in vitro toxicology. In this review, we describe several HCS application approaches used for studying the mechanism of compound toxicity, highlight some challenges faced in the toxicological community, and discuss the future directions of HCS in regards to new models, new reagents, data management, and informatics. Many specialized areas of toxicology including developmentaltoxicity, genotoxicity, developmental neurotoxicity/neurotoxicity, hepatotoxicity, cardiotoxicity, and nephrotoxicity will be examined. In addition, several newly developed cellular assay models including induced pluripotent stem cells (iPSCs), three-dimensional (3D) cell models, and tissues-on-a-chip will be discussed. New genome-editing technologies (e.g., CRISPR/Cas9), data analyzing tools for imaging, and coupling with high-content assays will be reviewed. Finally, the applications of machine learning to image processing will be explored. These new HCS approaches offer a huge step forward in dissecting biological processes, developing drugs, and making toxicology studies easier.

Abstract 2173: Developing a high-throughput, high content screening technology to identify new anticancer drug treatments

Abstract The high attrition rate of new chemotherapeutic drugs represents the single most important challenge in anticancer drug development. With only one drug out of 10, 000 test compounds gaining FDA-approval, attrition is often attributed to failure of in vitro drug efficacy to translate to clinical success. The overall drug discovery and development pipeline can be dramatically improved with new, innovative assays of high pharmacological relevance that can shorten the discovery/preclinical phase and ensure rapid translation to clinic. We report the development of a three-dimensional phenotypic drug screening platform based on spheroidsMARY-X. The tight, compact spheroidsMARY-X form spontaneously due to innate molecular determinants with proliferative cells on the outer periphery and a dormant tumor cell core. These features mimic the intratumoral biological complexities of tumor tissue and thus provide a relevant in vitro model for drug discovery with a high degree of predictive value. Significantly, this model allows analysis of drug targeting of dormant tumor cells, the difficult to treat resistant or highly refractory cell population. Drug response can be reliably assessed using brightfield microscopy and can be quantified by measuring spheroid circularity (i.e., dissolution index). Evaluation of soluble factor concentration is performed using molecular rotors through restricted intramolecular rotation upon protein binding with a subsequent quantum yield. Overall, we present a 3-dimensional anticancer assay that provides a rapid, efficient, relevant platform for the evaluation of drug response and allows identification of small molecule hits in both proliferative and more importantly, dormant tumor cells, the major cause of relapse and recurrent metastatic disease. Citation Format: Andrew T. Milcarek, Kevin Daus, Mary L. Alpaugh. Developing a high-throughput, high content screening technology to identify new anticancer drug treatments [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2173.

Identification and Correction of Additive and Multiplicative Spatial Biases in Experimental High-Throughput Screening

Data generated by high-throughput screening (HTS) technologies are prone to spatial bias. Traditionally, bias correction methods used in HTS assume either a simple additive or, more recently, a simple multiplicative spatial bias model. These models do not, however, always provide an accurate correction of measurements in wells located at the intersection of rows and columns affected by spatial bias. The measurements in these wells depend on the nature of interaction between the involved biases. Here, we propose two novel additive and two novel multiplicative spatial bias models accounting for different types of bias interactions. We describe a statistical procedure that allows for detecting and removing different types of additive and multiplicative spatial biases from multiwell plates. We show how this procedure can be applied by analyzing data generated by the four HTS technologies (homogeneous, microorganism, cell-based, and gene expression HTS), the three high-content screening (HCS) technologies (area, intensity, and cell-count HCS), and the only small-molecule microarray technology available in the ChemBank small-molecule screening database. The proposed methods are included in the AssayCorrector program, implemented in R, and available on CRAN.

Detecting and removing multiplicative spatial bias in high-throughput screening technologies.

Considerable attention has been paid recently to improve data quality in high-throughput screening (HTS) and high-content screening (HCS) technologies widely used in drug development and chemical toxicity research. However, several environmentally- and procedurally-induced spatial biases in experimental HTS and HCS screens decrease measurement accuracy, leading to increased numbers of false positives and false negatives in hit selection. Although effective bias correction methods and software have been developed over the past decades, almost all of these tools have been designed to reduce the effect of additive bias only. Here, we address the case of multiplicative spatial bias. We introduce three new statistical methods meant to reduce multiplicative spatial bias in screening technologies. We assess the performance of the methods with synthetic and real data affected by multiplicative spatial bias, including comparisons with current bias correction methods. We also describe a wider data correction protocol that integrates methods for removing both assay and plate-specific spatial biases, which can be either additive or multiplicative. The methods for removing multiplicative spatial bias and the data correction protocol are effective in detecting and cleaning experimental data generated by screening technologies. As our protocol is of a general nature, it can be used by researchers analyzing current or next-generation high-throughput screens. The AssayCorrector program, implemented in R, is available on CRAN. makarenkov.vladimir@uqam.ca. Supplementary data are available at Bioinformatics online.

A High-Content Screening Technology for Quantitatively Studying Podocyte Dynamics.

Podocytes form the visceral layer of a kidney glomerulus and express a characteristic octopus-like cellular architecture specialized for the ultrafiltration of blood. The cytoskeletal dynamics and structural elasticity of podocytes rely on the self-organization of highly interconnected actin bundles, and the maintenance of these features is important for the intact glomerular filtration. Development of more differentiated podocytes in culture has dramatically increased our understanding of the molecular mechanisms regulating podocyte actin dynamics. Podocytes are damaged in a variety of kidney diseases, and therapies targeting podocytes are being investigated with increasing efforts. Association between podocyte damage and disease severity-or between podocyte recovery and the performance of therapeutic molecules-have been the venues of research for years. In this perspective, more standardized high--content screening has emerged as a powerful tool for visualization and analysis of podocyte morphology. This high-throughput fluorescence microscopy technique is based on an automated image analysis with simultaneous detection of various phenotypes (multiplexing) across multiple phenotypic parameters (multiparametric). Here, we review the principles of high-content screening technology and summarize efforts to carry out small compound screen using podocytes.

Validation of a Multiparametric, High-Content-Screening Assay for Predictive/Investigative Cytotoxicity: Evidence from Technology Transfer Studies and Literature Review.

A multiparametric, live-cell, high-content-screening (HCS) cytotoxicity assay was first demonstrated in 2006 ( Arch. Toxicol. 2006 , 80 , 580 - 604 ) to be highly concordant with human hepatotoxicity, including idiosyncratic hepatotoxicities and other target organ toxicities in contrast to historical assays. The success of the assay was attributed to its simultaneous measurement of multiple appropriate "cytobiomarkers": use of human cells with xenometabolic competence for toxicities mediated by metabolites, 72 h exposure to enable expression of slower-acting toxicants, exposure to a wide-range of concentrations from 30- to 100-fold the efficacious concentration, and normalizing the in vitro cytotoxic concentration to an estimate of the in vivo concentration of exposure. An overwhelming volume of evidence has accumulated over the last 10 years to support this approach as necessary in predictive toxicology. Equivalent assays have now been successfully applied in ∼50 studies across a wide variety of toxicants, toxicities, cell types, and disciplines. Review herein of the wider literature on cytotoxicity since the first assay was reported 100 years ago supports the selection of key cytobiomarkers along a final common pathway of cell injury, including cell proliferation, mitochondrial activity, apoptosis, lysosomal mass, oxidative stress, and cell membrane permeability. HCS studies without inclusion of such key cytobiomarkers or without testing to sufficiently high concentration have not been as successful. Furthermore, a subset of the original toxicants has been reanalyzed herein using the original HCS assay and has confirmed their high sensitivities and specificities across locations, HCS technologies, staff, laboratories, and time. A protocol is demonstrated for operational validation of the assay within laboratories to demonstrate proficiency and quality management.

Discovery of a Novel, Isothiazolonaphthoquinone-Based Small Molecule Activator of FOXO Nuclear-Cytoplasmic Shuttling.

FOXO factors are tumour suppressor proteins commonly inactivated in human tumours by posttranslational modifications. Furthermore, genetic variation within the FOXO3a gene is consistently associated with human longevity. Therefore, the pharmacological activation of FOXO proteins is considered as an attractive therapeutic approach to treat cancer and age-related diseases. In order to identify agents capable of activating FOXOs, we tested a collection of small chemical compounds using image-based high content screening technology. Here, we report the discovery of LOM612 (compound 1a), a newly synthesized isothiazolonaphthoquinone as a potent FOXO relocator. Compound 1a induces nuclear translocation of a FOXO3a reporter protein as well as endogenous FOXO3a and FOXO1 in U2OS cells in a dose-dependent manner. This activity does not affect the subcellular localization of other cellular proteins including NFkB or inhibit CRM1-mediated nuclear export. Furthermore, compound 1a shows a potent antiproliferative effect in human cancer cell lines.

Classifying and segmenting microscopy images with deep multiple instance learning.

Motivation: High-content screening (HCS) technologies have enabled large scale imaging experiments for studying cell biology and for drug screening. These systems produce hundreds of thousands of microscopy images per day and their utility depends on automated image analysis. Recently, deep learning approaches that learn feature representations directly from pixel intensity values have dominated object recognition challenges. These tasks typically have a single centered object per image and existing models are not directly applicable to microscopy datasets. Here we develop an approach that combines deep convolutional neural networks (CNNs) with multiple instance learning (MIL) in order to classify and segment microscopy images using only whole image level annotations.Results: We introduce a new neural network architecture that uses MIL to simultaneously classify and segment microscopy images with populations of cells. We base our approach on the similarity between the aggregation function used in MIL and pooling layers used in CNNs. To facilitate aggregating across large numbers of instances in CNN feature maps we present the Noisy-AND pooling function, a new MIL operator that is robust to outliers. Combining CNNs with MIL enables training CNNs using whole microscopy images with image level labels. We show that training end-to-end MIL CNNs outperforms several previous methods on both mammalian and yeast datasets without requiring any segmentation steps.Availability and implementation: Torch7 implementation available upon request.Contact: oren.kraus@mail.utoronto.ca