Label-free detection of neuronal differentiation in cell populations using high-throughput live-cell imaging of PC12 cells

Biocompatible, Europium-Doped Fluorapatite Nanoparticles as a Wide-Range pH Sensor

The development of a simple, biocompatible, pH sensor with a wide range of detection, using a single fluorescent probe is highly important in the medical field for the early detection of diseases related to the pH change of tissues and body fluids. For this purpose, europium-doped fluorapatite (FAP: Eu) nanoparticles were synthesized using the coprecipitation method. Doping with the rare earth element europium (Eu) makes the non-luminescent phosphate mineral fluorapatite, luminescent. The luminous response of the sample upon dissolution in hydrochloric acid (HCl), in highly acidic to weakly basic media, makes it a potential pH sensor. A linear variation was observed with an increase in pH, in both the total intensity of emission and the R-value or the asymmetry ratio. The ratiometric pH sensing enabled by the variation in R-value makes the sensor independent of external factors. The structural, optical, and photoluminescent (PL) lifetime analysis suggests a particle size-dependent pH sensing mechanism with the changes in the coordinated water molecules around the Eu3+ ion in the nanoparticle. Given its exceptional biocompatibility and pH-dependent fluorescence intensity for a wide range of pH from 0.83 to 8.97, the probe can be used as a potential candidate for pH sensing of biological fluid.

Eggshell Derived Europium Doped Hydroxyapatite Nanoparticles for Cell Imaging Application

Hen's eggshell, a biological waste product, was turned into a cell imaging probe: europium doped hydroxyapatite (HAp: Eu) nanoparticle using hydrothermal method. Luminescence of the synthesized nanoparticle was studied for various doping concentrations of the lanthanide ion europium (Eu³⁺). Eu doped HAp showed a hexagonal crystal structure and rod-shaped morphology. Well-defined emission peaks of europium, corresponding to the substitution of Eu³⁺ at the Ca²⁺(I) site of HAp, were confirmed from the samples' photoluminescence (PL) spectra. Good biocompatibility up to 500 μg/mL of the samples indicates their potential applications in bioimaging. Synthesized nanoparticles were internalized and used for in vitro imaging of the PC12 cells without any surface modification. The materials' use as a potential in vivo imaging agent is proposed from the haemolysis study.

Immunofluorescence Image Feature Analysis and Phenotype Scoring Pipeline for Distinguishing Epithelial-Mesenchymal Transition

Epithelial–mesenchymal transition (EMT) is an essential biological process, also implicated in pathological settings such as cancer metastasis, in which epithelial cells transdifferentiate into mesenchymal cells. We devised an image analysis pipeline to distinguish between tissues comprised of epithelial and mesenchymal cells, based on extracted features from immunofluorescence images of differing biochemical markers. Mammary epithelial cells were cultured with 0 (control), 2, 4, or 10 ng/mL TGF-β1, a well-established EMT-inducer. Cells were fixed, stained, and imaged for E-cadherin, actin, fibronectin, and nuclei via immunofluorescence microscopy. Feature selection was performed on different combinations of individual cell markers using a Bag-of-Features extraction. Control and high-dose images comprised the training data set, and the intermediate dose images comprised the testing data set. A feature distance analysis was performed to quantify differences between the treatment groups. The pipeline was successful in distinguishing between control (epithelial) and the high-dose (mesenchymal) groups, as well as demonstrating progress along the EMT process in the intermediate dose groups. Validation using quantitative PCR (qPCR) demonstrated that biomarker expression measurements were well-correlated with the feature distance analysis. Overall, we identified image pipeline characteristics for feature extraction and quantification of immunofluorescence images to distinguish progression of EMT.

Emerging machine learning approaches to phenotyping cellular motility and morphodynamics

Cells respond heterogeneously to molecular and environmental perturbations. Phenotypic heterogeneity, wherein multiple phenotypes coexist in the same conditions, presents challenges when interpreting the observed heterogeneity. Advances in live cell microscopy allow researchers to acquire an unprecedented amount of live cell image data at high spatiotemporal resolutions. Phenotyping cellular dynamics, however, is a nontrivial task and requires machine learning (ML) approaches to discern phenotypic heterogeneity from live cell images. In recent years, ML has proven instrumental in biomedical research, allowing scientists to implement sophisticated computation in which computers learn and effectively perform specific analyses with minimal human instruction or intervention. In this review, we discuss how ML has been recently employed in the study of cell motility and morphodynamics to identify phenotypes from computer vision analysis. We focus on new approaches to extract and learn meaningful spatiotemporal features from complex live cell images for cellular and subcellular phenotyping.

Probing the characteristics and biofunctional effects of disease-affected cells and drug response via machine learning applications

Drug-induced transformations in disease characteristics at the cellular and molecular level offers the opportunity to predict and evaluate the efficacy of pharmaceutical ingredients whilst enabling the optimal design of new and improved drugs with enhanced pharmacokinetics and pharmacodynamics. Machine learning is a promising in-silico tool used to simulate cells with specific disease properties and to determine their response toward drug uptake. Differences in the properties of normal and infected cells, including biophysical, biochemical and physiological characteristics, plays a key role in developing fundamental cellular probing platforms for machine learning applications. Cellular features can be extracted periodically from both the drug treated, infected, and normal cells via image segmentations in order to probe dynamic differences in cell behavior. Cellular segmentation can be evaluated to reflect the levels of drug effect on a distinct cell or group of cells via probability scoring. This article provides an account for the use of machine learning methods to probe differences in the biophysical, biochemical and physiological characteristics of infected cells in response to pharmacokinetics uptake of drug ingredients for application in cancer, diabetes and neurodegenerative disease therapies.

Image-derived models of cell organization changes during differentiation and drug treatments

PC12 cells are a popular model system to study changes driving and accompanying neuronal differentiation. While attention has been paid to changes in transcriptional regulation and protein signaling, much less is known about the changes in organization that accompany PC12 differentiation. Fluorescence microscopy can provide extensive information about these changes, although it is difficult to continuously observe changes over many days of differentiation. We describe a generative model of differentiation-associated changes in cell and nuclear shape and their relationship to mitochondrial distribution constructed from images of different cells at discrete time points. We show that the model accurately represents complex cell and nuclear shapes and learn a regression model that relates cell and nuclear shape to mitochondrial distribution; the predictive accuracy of the model increases during differentiation. Most importantly, we propose a method, based on cell matching and interpolation, to produce realistic simulations of the dynamics of cell differentiation from only static images. We also found that the distribution of cell shapes is hollow: most shapes are very different from the average shape. Finally, we show how the method can be used to model nuclear shape changes of human-induced pluripotent stem cells resulting from drug treatments.

Correlation and Comparison of Cortical and Hippocampal Neural Progenitor Morphology and Differentiation through the Use of Micro- and Nano-Topographies

Neuronal morphology and differentiation have been extensively studied on topography. The differentiation potential of neural progenitors has been shown to be influenced by brain region, developmental stage, and time in culture. However, the neurogenecity and morphology of different neural progenitors in response to topography have not been quantitatively compared. In this study, the correlation between the morphology and differentiation of hippocampal and cortical neural progenitor cells was explored. The morphology of differentiated neural progenitors was quantified on an array of topographies. In spite of topographical contact guidance, cell morphology was observed to be under the influence of regional priming, even after differentiation. This influence of regional priming was further reflected in the correlations between the morphological properties and the differentiation efficiency of the cells. For example, neuronal differentiation efficiency of cortical neural progenitors showed a negative correlation with the number of neurites per neuron, but hippocampal neural progenitors showed a positive correlation. Correlations of morphological parameters and differentiation were further enhanced on gratings, which are known to promote neuronal differentiation. Thus, the neurogenecity and morphology of neural progenitors is highly responsive to certain topographies and is committed early on in development.

Machine learning and computer vision approaches for phenotypic profiling

Grys et al. review computer vision and machine-learning methods that have been applied to phenotypic profiling of image-based data. Descriptions are provided for segmentation, feature extraction, selection, and dimensionality reduction, as well as clustering, outlier detection, and classification of data.

With recent advances in high-throughput, automated microscopy, there has been an increased demand for effective computational strategies to analyze large-scale, image-based data. To this end, computer vision approaches have been applied to cell segmentation and feature extraction, whereas machine-learning approaches have been developed to aid in phenotypic classification and clustering of data acquired from biological images. Here, we provide an overview of the commonly used computer vision and machine-learning methods for generating and categorizing phenotypic profiles, highlighting the general biological utility of each approach.

Morphological Evaluation of Nonlabeled Cells to Detect Stimulation of Nerve Growth Factor Expression by Lyconadin B

The success of drug development is greatly influenced by the efficiency of drug screening methods. Recently, phenotype-based screens have raised expectations, based on their proven record of identifying first-in-class drugs at a higher rate. Although fluorescence images are the data most commonly used in phenotype-based cell-based assays, nonstained cellular images have the potential to provide new descriptive information about cellular responses. In this study, we applied morphology-based evaluation of nonlabeled microscopic images to a phenotype-based assay. As a study case, we attempted to increase the efficiency of a cell-based assay for chemical compounds that induce production of nerve growth factor (NGF), using lyconadin B as a model compound. Because the total synthesis of lyconadin B was accomplished very recently, there is no well-established cell-based assay scheme for further drug screening. The conventional cell-based assay for evaluating NGF induction requires two types of cells and a total of 5 days of cell culture. The complexity and length of this assay increase both the risk of screening errors and the cost of screening. Our findings show that analysis of cellular morphology enables evaluation of NGF induction by lyconadin B within only 9 h.

Boolean Modeling Reveals the Necessity of Transcriptional Regulation for Bistability in PC12 Cell Differentiation

The nerve growth factor NGF has been shown to cause cell fate decisions toward either differentiation or proliferation depending on the relative activity of downstream pERK, pAKT, or pJNK signaling. However, how these protein signals are translated into and fed back from transcriptional activity to complete cellular differentiation over a time span of hours to days is still an open question. Comparing the time-resolved transcriptome response of NGF- or EGF-stimulated PC12 cells over 24 h in combination with protein and phenotype data we inferred a dynamic Boolean model capturing the temporal sequence of protein signaling, transcriptional response and subsequent autocrine feedback. Network topology was optimized by fitting the model to time-resolved transcriptome data under MEK, PI3K, or JNK inhibition. The integrated model confirmed the parallel use of MAPK/ERK, PI3K/AKT, and JNK/JUN for PC12 cell differentiation. Redundancy of cell signaling is demonstrated from the inhibition of the different MAPK pathways. As suggested in silico and confirmed in vitro, differentiation was substantially suppressed under JNK inhibition, yet delayed only under MEK/ERK inhibition. Most importantly, we found that positive transcriptional feedback induces bistability in the cell fate switch. De novo gene expression was necessary to activate autocrine feedback that caused Urokinase-Type Plasminogen Activator (uPA) Receptor signaling to perpetuate the MAPK activity, finally resulting in the expression of late, differentiation related genes. Thus, the cellular decision toward differentiation depends on the establishment of a transcriptome-induced positive feedback between protein signaling and gene expression thereby constituting a robust control between proliferation and differentiation.

Rapid Assessment and Visualization of Normality in High-Content and Other Cell-Level Data and Its Impact on the Interpretation of Experimental Results

When investigators monitor effects on a population of cells following a perturbation, these events rarely occur in a classical normal (or Gaussian) distribution. A normal distribution is, however, explicitly assumed for events within a single well, in which mean values per well are used as an assay metric and, in general, measures of assay robustness, such as the Z' score and the V factor. Such analysis is not possible for many technologies; however, high-content screening (HCS) measures events of individual cells, which are averaged over the well. These individual cell-level measurements may be analyzed separately. This study quantifies the extent of nonnormality in experimental samples and their effects on determining the EC50 of a test compound and the assay robustness statistics. The results, based on five sets of publicly available data, indicate that the Z' or V-factor score can be improved by as much as 0.44 more than standard calculations, and the EC50 of a dose-response curve can be lowered by as much as fivefold when nonparametric methods are used, but not all data sets show a significant improvement. The effect on analysis depends in part on whether the greatest shift from normality occurs in the upper or lower range of the dose-response curve.

Machine learning in cell biology - teaching computers to recognize phenotypes

Recent advances in microscope automation provide new opportunities for high-throughput cell biology, such as image-based screening. High-complex image analysis tasks often make the implementation of static and predefined processing rules a cumbersome effort. Machine-learning methods, instead, seek to use intrinsic data structure, as well as the expert annotations of biologists to infer models that can be used to solve versatile data analysis tasks. Here, we explain how machine-learning methods work and what needs to be considered for their successful application in cell biology. We outline how microscopy images can be converted into a data representation suitable for machine learning, and then introduce various state-of-the-art machine-learning algorithms, highlighting recent applications in image-based screening. Our Commentary aims to provide the biologist with a guide to the application of machine learning to microscopy assays and we therefore include extensive discussion on how to optimize experimental workflow as well as the data analysis pipeline.

HCS-Neurons: identifying phenotypic changes in multi-neuron images upon drug treatments of high-content screening

Background

High-content screening (HCS) has become a powerful tool for drug discovery. However, the discovery of drugs targeting neurons is still hampered by the inability to accurately identify and quantify the phenotypic changes of multiple neurons in a single image (named multi-neuron image) of a high-content screen. Therefore, it is desirable to develop an automated image analysis method for analyzing multi-neuron images.

Results

We propose an automated analysis method with novel descriptors of neuromorphology features for analyzing HCS-based multi-neuron images, called HCS-neurons. To observe multiple phenotypic changes of neurons, we propose two kinds of descriptors which are neuron feature descriptor (NFD) of 13 neuromorphology features, e.g., neurite length, and generic feature descriptors (GFDs), e.g., Haralick texture. HCS-neurons can 1) automatically extract all quantitative phenotype features in both NFD and GFDs, 2) identify statistically significant phenotypic changes upon drug treatments using ANOVA and regression analysis, and 3) generate an accurate classifier to group neurons treated by different drug concentrations using support vector machine and an intelligent feature selection method. To evaluate HCS-neurons, we treated P19 neurons with nocodazole (a microtubule depolymerizing drug which has been shown to impair neurite development) at six concentrations ranging from 0 to 1000 ng/mL. The experimental results show that all the 13 features of NFD have statistically significant difference with respect to changes in various levels of nocodazole drug concentrations (NDC) and the phenotypic changes of neurites were consistent to the known effect of nocodazole in promoting neurite retraction. Three identified features, total neurite length, average neurite length, and average neurite area were able to achieve an independent test accuracy of 90.28% for the six-dosage classification problem. This NFD module and neuron image datasets are provided as a freely downloadable MatLab project at http://iclab.life.nctu.edu.tw/HCS-Neurons.

Conclusions

Few automatic methods focus on analyzing multi-neuron images collected from HCS used in drug discovery. We provided an automatic HCS-based method for generating accurate classifiers to classify neurons based on their phenotypic changes upon drug treatments. The proposed HCS-neurons method is helpful in identifying and classifying chemical or biological molecules that alter the morphology of a group of neurons in HCS.