Background

The discovery of super-resolution imaging methods has enabled optical microscopy to overcome the conventional light diffraction limit, defined by Ernst Karl Abbe in 1873. These methods have completely revolutionized the field of optical fluorescence microscopy, tremendously enhancing its resolution power [1]. Single-molecule localization microscopy (SMLM) techniques rely on the precise determination of fluorophore positions, with subpixel accuracy, based on the emission from a small molecule subset, which can be distinguished from nearby fluorophores in the dark (non-emitting) state, thus allowing the reconstruction of a high-resolution image [2,3]. The strategy applied to achieve the fluorophore on/off state switching is specific to each SMLM technique, which mainly includes photo-activated localization microscopy (PALM) and stochastic optical reconstruction microscopy (STORM). Briefly, PALM can also be applied to live-cell imaging and relies on the use of photoactivatable/switchable proteins, which can be activated with near-UV/violet light (typically ~405 nm) or photo-converted from a shorter to a longer wavelength form. The detection of emission by single molecules is achieved using low-intensity activation or switching laser cycles, which turn on or convert only a subset of fluorophores at a time, until they are photobleached [4]. In contrast, STORM exploits organic fluorophores such as Alexa Fluor–conjugated antibodies (e.g., Alexa Fluor 647), which undergo reversible photo switching in suitable imaging buffers; therefore, it is suitable for fixed sample imaging [5].

STORM is a single-molecule high-resolution microscopy technique that allows the visualization of unresolved details of many cellular structures, including clathrin-coated pits, mitochondria, synaptic structures, and neuronal spectrin, which forms, with actin and associated proteins, a periodic structure in neurons [6–9]. In addition, the technique relies on fluorophores blinking, defined as the stochastic switching of a photosensitive molecule between fluorescent (on) and non-fluorescent (off) states, enabling the precise localization of individual emitters, before they enter a long-lived dark state or become photobleached. The localization coordinates of many individuals emitting fluorophores, each associated with respective spatial coordinates, allow the reconstruction of a high-resolution image [10]. The implementation of advanced imaging methods for broader laboratory use highlights the need for rigorous validation strategies and quantitative performance metrics supporting protocol optimization and cross-platform reproducibility in quantitative imaging workflows [11].

The current protocol describes a complete workflow, including sample preparation, image acquisition, data processing, and quality assessment to obtain high-quality single and dual-color STORM images of different cellular structures such as tubulin, lysosomes, mitochondria, and the actin cytoskeleton. Here, we employed commercially available primary antibodies and secondary antibodies conjugated to Alexa Fluor 647 switchable fluorophores, which show excellent blinking properties and represent the most widely used dyes for STORM imaging. In addition, for dual-color experiments, the actin cytoskeleton was labeled using a phalloidin-Alexa Fluor 488, which allows high-quality reconstruction of the cytoskeleton, although its blinking properties are lower than those of the Alexa Fluor 647 counterpart [12]. For imaging, an oxygen-scavenging imaging buffer, which maintains a reducing chemical environment ideal for the dye photo conversion, is used. The choice to perform dual-color STORM using an Alexa Fluor 647/Alexa Fluor 488 combination was dictated by the excitation–emission filter of our system, as well as the compatibility of the two dyes with the imaging buffer, which needs to be adjusted and optimized according to the chemical properties of the fluorophores.

Despite the growing interest in super-resolution microscopy, the implementation of STORM often requires specialized instrumentation and technical expertise that may not be readily available in many laboratories. This protocol addresses this gap by providing a workflow that enables STORM imaging using widely available microscopy platforms with minimal technical modifications. By lowering the technical barriers to SMLM implementation, this approach aims to facilitate broader adoption of super-resolution imaging across research laboratories. Such accessibility could benefit diverse applications, including studies of cellular architecture, mechanotransduction, and tissue organization, where nanoscale imaging can provide insights into biological processes such as osteocyte responses to mechanical stress [13]. STORM imaging requires customized microscopes or commercially available systems, which are specifically designed for SMLM experiments, including the ONI imager (www.oni.bio.com) or the Abbelight platform (www.abbelight.com). This represents a limitation to the spreading of this powerful microscopy technique. In order to overcome this limitation, we designed this protocol where image acquisition is performed on a conventional epifluorescence/total internal reflection (TIRF) microscope, commonly available in many laboratories or imaging facilities, adapted for STORM imaging. Crucial adaptations include the use of a 160×/1.43 NA oil-immersion objective and a high-power mode that concentrates the laser beam onto a small region of the sample, ensuring sufficient intensity to drive fluorophores into the dark state. This allows us to overcome the limitation of the low laser power compared to the system designed for STORM, which shows high-power laser beams. Indeed, power ranges (at samples) for 638 and 488 nm laser lines are on the order of 250 mW for ONI or 40–500 mW for Abbelight microscopes, compared to 25–30 mW for our system used here. In addition, implementing a 1.6× magnification lens and a 4×4 binning camera mode allowed us to achieve a 100-nm effective pixel size optimal for reliable molecule detection. Moreover, the use of TIRF microscopy, which allows regulation of laser penetration depth within the sample [14], helps with background removal, optimizing the molecule localization during image analysis. Image reconstruction and post-processing drift correction, as well as the evaluation of the quality of molecule localization, were performed using the ThunderSTORM open-source Fiji plugin tool [15]. The quality of STORM reconstructed images was assessed using the NanoJ-SQUIRREL open-source Fiji plugin [16], comparing the corresponding TIRF image.

These developments aim to improve the accessibility of STORM imaging and reduce technical barriers to its implementation, facilitating broader adoption of super-resolution microscopy in standard laboratory environments. Such efforts align with recent methodological advances emphasizing translational robustness and standardized experimental workflows to ensure reproducibility across research settings [17]. We believe that this protocol will be particularly valuable to the microscopy community, especially for beginner STORM users, as it can be implemented on widely available microscopy platforms already available in many laboratories or imaging facilities. With minor technical adjustments, these platforms can be adapted for SMLM imaging, avoiding substantial financial investments.