Background

Over the past decades, high-content analysis (HCA) has transformed multiple areas, including basic cell biology, stem cell research, toxicology, developmental biology, and drug discovery, extending even to complex three-dimensional (3D) models [2,3]. This transformation has been driven by advances in automated fluorescence microscopy, the development of versatile fluorescent probes, and rapid progress in computational methods such as image segmentation, feature extraction, and machine learning [4–6]. Together, these tools have created a powerful framework for cellular profiling, enabling the detection of subtle and multifactorial biological responses to perturbations, including small molecules, oligonucleotides, and nanoparticles, that remain inaccessible to conventional techniques [7,8].

A central strategy in HCA relies on fluorescence microscopy, using genetically encoded fluorescent proteins, immunoreagents, fluorescent dyes, or small-molecule probes. These assays typically follow one of two conceptual approaches. The first is a target-based strategy, in which a specific and anticipated biological response is monitored—for example, cellular assays focused on G protein–coupled receptors (GPCRs), ion channels, transporters, DNA damage, or reactive oxygen species (ROS) production [9]. The second approach is unbiased or agnostic to specific biological targets, instead relying on comprehensive staining of cellular architecture. The most prominent technique in this category is cell painting, which labels eight major subcellular structures and has played a key role in advancing the field over the past decade [10]. Most HCA workflows rely on fixed-cell samples, which offer several practical advantages such as ease of handling, compatibility with batch processing, and signal stability. However, fixation can introduce artifacts and eliminates the possibility of capturing live-cell dynamics [11,12]. Additionally, many existing protocols are complex and costly, limiting their accessibility and widespread adoption. Live-cell imaging, in contrast, enables the observation of dynamic biological processes in real time, including the tracking of cellular or subcellular structures, acquisition of kinetic data, analysis of reversible events, and circumvention of fixation-related limitations.

In this protocol, we present live cell painting (LCP), a versatile, cost-effective, and information-rich assay that is simple to implement. This method was established and validated in different cell lines, including Huh-7, MCF-7, PNT1A, and PC3, and can be applied to other cell lines with appropriate concentration adjustments [1]. The method leverages acridine orange (AO), a metachromatic fluorescent dye capable of labeling multiple cellular components in live cells, including nucleic acids and acidic intracellular vesicles (e.g., lysosomes, endosomes, and autophagosolysosomes, among others) (Figures 1 and 2). This dual-channel dye enables multiparametric imaging of various cellular structures, such as nuclei and nucleoli, cytoplasmic organization, cell shape, cell count, and vesicle distribution. The rich morphological and spatial information obtained from AO staining can be applied across a wide range of high-content applications, including phenotypic screening and profiling, target-based screening with morphological readouts, mechanism of action (MoA) elucidation, phenotype recovery assays, toxicity and safety profiling, functional genomics screens (e.g., RNAi or CRISPR), and assays tailored to specific cellular processes. Furthermore, this approach offers a powerful means to generate biologically grounded hypotheses from image-based data. However, like any method, LCP has certain limitations. As with most HCA approaches, it requires access to a fluorescence microscope and imaging-compatible multi-well plates, which may not be readily available in all laboratory settings. Additionally, AO is prone to photobleaching, and its concentration must be carefully optimized for each cell line to avoid cytotoxicity or nonspecific staining. While LCP offers rich morphological insights, complementary assays may still be necessary for a full understanding of cellular function and physiology. Despite these considerations, LCP remains a powerful, cost-effective, and accessible tool for live-cell imaging and high-content morphological profiling.

Figure 1. Influence of pH and acridine orange (AO) concentration on its fluorescence behavior in distinct cellular compartments. AO binds nucleic acids by intercalating as monomers into double-stranded DNA and associating electrostatically with single-stranded RNA, both of which result in green fluorescence emission. In acidic compartments, such as lysosomes, or under conditions of high AO concentration, the dye tends to self-aggregate into stacked complexes, shifting its emission from green to red. This dual fluorescence property enables AO to function as a sensitive probe for intracellular localization of nucleic acids, organelle pH, and dye concentration in live-cell imaging.

Figure 2. Fluorescence image of MCF-7 cells stained with 10 μM acridine orange (AO), acquired using a 20× objective. The green channel (GFP) highlights nuclei, nucleoli, and cytoplasm, while the red channel (PI) selectively labels acidic vesicular compartments (e.g., lysosomes, endosomes, and autophagolysosomes, among others). Brightness and contrast were adjusted to enhance visualization. The merged image displays the combined spatial distribution of both signals. Scale bar = 100 μm.