Even those who maintain that super-resolution microscopy is a powerful tool of biological discovery have admitted that it may have a bit of an image problem. For example, in a recent review, several of super-resolution microscopy’s positives were paired with potential negatives (Prakash et al. Philos. Trans. A. 2022; 380(2220): 20210110).
Although super-resolution microscopy is cited in a growing number of publications, many of these publications emphasize physics/optics rather than biology. Although super-resolution microscopy has produced spectacular images of cellular structures, many of them depict structures that are already well-characterized, creating the mistaken impression that biological applications may be limited. Although super-resolution microscopy has inspired multiple technological approaches, these have given rise to a potentially bewildering array of options.
Yet, as this review pointed out, super-resolution microscopy is emerging from the shadows of technological and operational complexity, attracting the interest of a wider array of potential users in the life sciences. The review’s authors, who included a biochemist and a molecular pathologist, mentioned several encouraging developments:
- A growing emphasis on improving the reliability and applicability of super-resolution microscopy techniques, as opposed to achieving the utmost degrees of resolution. For example, “soft” super-resolution technologies—or high-resolution technologies—such as Airyscan technology or structured illumination microscopy (SIM) can be used to push the resolution of standard confocal microscopy systems into the super-resolution range.
- The development of better fluorescent dyes, probes, and labeling tools. Such tools include transiently binding DNA-based probes, which are employed in points accumulation for imaging in nanoscale topography (DNA-PAINT). The DNA-based probes target specific proteins, enabling localization-based super-resolution microscopy. With techniques such as Stochastic Optical Reconstruction Microscopy (STORM) and Photoactivated Localization Microscopy (PALM), photoswitchable fluorophores can be activated and deactivated individually, allowing more controlled and efficient activation, and enhancing localization precision.
- The proliferation of artificial intelligence applications. Several commercial platforms are using artificial intelligence to derive insights from gently illuminated samples, reducing the phototoxicity and bleaching that can accompany harsh illumination, thereby facilitating the study of living cells. Also, artificial intelligence can simplify three-dimensional workflows and correlative approaches that combine findings from super-resolution microcopy and those from electron microscopy.
- The establishment of centralized core facilities. These facilities can spare research laboratories the difficulties of developing their own super-resolution microscopy capabilities.
All of these developments have contributed to new discoveries. Some of the more familiar discoveries concern chromatin organization, nuclear pore structure/function, cytoskeletal dynamics, and cell division mechanisms. Other important discoveries concern membrane receptor clustering, RNA localization and dynamics, virus-host interactions, and the formation of molecular assemblies such as DNA repair complexes, ribosomes, protein aggregates, and synaptic scaffolds.
Super-resolution techniques
All super-resolution techniques bypass the laws of diffraction that limit light microscopy to a resolution of a few hundred nanometers. However, different techniques get around these laws in different ways, such that certain techniques are better suited to certain applications. Because there are so many techniques, knowing which technique is best for a given application is challenging for many researchers.
To help researchers consider their options, scientists affiliated with the Edinburgh Super Resolution Imaging Consortium (ESRIC) have produced a guide that shows how the most commonly available super-resolution techniques may address biological questions of different types—specifically, structure localization, live-cell dynamics, and molecular interactions.
The researchers presented their guide in a recent review article (Valli et al. J. Biol. Chem. 2021; 297(1): 100791). Conveniently, the article presents a flow chart in which paths to different techniques are determined by answers to basic questions. The first-tier question is: What kind of biological information do you need? The answer determines which of the second-tier questions comes next: What size is your object of interest? What do you need to image? What type of interaction will be imaged? At subsequent tiers, more detailed questions appear.
Even more conveniently, the ESRIC has turned the flowchart into a web-based form. However, the review article offers a more thorough consideration of selection criteria. It states, “There is a lot to take into account: the specimen itself (size, live/fixed, thickness, opacity, etc.), the aspect of the specimen under study (i.e., cell structures, surfaces, single molecules), and the temporal sensitivity required (i.e., for molecular dynamics, interactions, etc.). Nevertheless, as with any method, each of those discussed in this review has inherent pros and cons. Moreover, there will always be the tradeoff between spatial and temporal resolution to consider for any imaging experiment.”
Yes, the flowchart is a simplified guide. But it helps clarify the distinctions between the different techniques, which can’t be reduced to anything so simple as degree of resolution. Indeed, resolution ranges overlap a fair amount between the different techniques, even between the “ensemble” techniques and the “single molecule” techniques.
The former include Structured Illumination Microscopy (SIM), which relies on patterned illumination and computational reconstruction, and Stimulated Emission Depletion (STED) microscopy, which relies on the selective activation and deactivation of fluorophores. The latter include STORM, which relies on the random activation of photoswitchable fluorophores, PALM, which relies on the sequential activation of photoactivatable fluorophores, and DNA-PAINT (described earlier).
Complicating matters further are hybrid techniques such as Reversible Saturable Optical Fluorescence Transitions (RESOLFT), which is a STED variant that relies on specific light wavelengths to switch fluorescent proteins into on and off states, and Minimal photon FLUX (MINFLUX), which combines single-molecule localization and fluorescence excitation control. MINFLUX can achieve extremely high resolution (down to 1 nm) even though it uses less light. (A donut-shaped laser beam roams the sample, seeking intensity valleys rather than intensity peaks to localize individual fluorophores.)
Platforms and applications
That’s enough general information. Let’s look at a couple of specific applications. One example comes from ONI, which began as an Oxford University spinout company in 2016, and which now maintains offices in San Diego, CA, and Oxford, UK. (The company is perhaps best known for its Nanoimager single-molecule localization microscopy platform.) Another example comes from Eikon Therapeutics, which was founded in 2019 and is headquartered in Hayward, CA. The company uses its proprietary single-molecule tracking (SMT) platform to drive its own development programs.
“We are particularly proud of the work from researchers at the Octopus Imaging Cluster at Central Laser Facility, UKRI-STFC Rutherford Appleton Laboratory, UK,” says Shaked Ashkenazi, content marketing manager, ONI. “They used our Nanoimager to develop FLImP [Fluorophore Localization Imaging with Photobleaching], which is an automated single-molecule imaging technique. One application of this technology is the study of EGFR, a protein that is frequently mutated in many tumors. Earlier this year, they published a paper that used FLImP to explain how tumors acquire resistance to EGFR-targeting drugs (Iyer et al. Nat. Commun. 2024; 15(1): 2130). This discovery can lead to better screenings and hopefully new therapies to tackle the resistance.”
Ashkenazi adds that ONI is working to bring super-resolution microscopy to all life sciences research laboratories, not just specialists: “We identified that extracting scientifically meaningful insights from pretty images is not trivial to many, and we invest many efforts in developing user-friendly software tools to help the researchers, namely our CODI platform and its application-specific workflows for straightforward analyses. Our Nanoimager is not bigger than your average countertop kitchen appliance. Its small footprint and user-friendly software provide a huge leap forward in the journey to make super-resolution a garden variety research technique.”
Eikon’s SMT technology is built on Nobel Prize–winning super-resolution fluorescence microscopy technology that was originally developed by Eric Betzig, PhD, one of the company’s founders. (He is also affiliated with the University of California, Berkeley; Lawrence Berkeley National Laboratory; and the Howard Hughes Medical Institute’s Janelia Research Campus.) Today, SMT encompasses a benchtop device, integrated software, and reagent kits.
With SMT, Eikon can determine how protein motion is mediated by transient interactions between a protein molecule and other components of the cellular environment. Essentially, SMT reveals a protein’s activities and functions—crucial information for protein-focused drug discovery. “Because SMT enables the study of almost any protein within the cellular environment, even unstructured proteins, we pursue difficult therapeutic targets that are hard to evaluate with traditional methods,” says Dan Anderson, chief Eikon’s scientific officer. “We are also using our platform to understand basic cell biology by studying the motion of key proteins in genomics screens to map protein-protein interactions at systems scale.”
SMT is also suitable for the study of protein population dynamics. “Within a cell, there may be several subpopulations of a protein of interest, each performing unique functions, and each moving in a different way,” Anderson notes. “By observing vast numbers of individual proteins moving throughout multiple cells, we build profiles of a target protein’s subpopulation dynamics, in the basal state or in the presence of perturbations. As a result, we gain nuanced insight into each protein subpopulation’s change in function. This is accomplished by capturing hundreds of thousands of protein trajectories across dozens of cells in less than a second, and we typically analyze hundreds of thousands of experimental conditions each week.”
At Eikon, advances are being made in optics, data analysis software, robotics automation, and biological methods. A study led by Anderson employed Oblique Line Scan (OLS) illumination, which uses a thin sheet of light generated by a laser to enhance optical sectioning and spatiotemporal resolution. It allows for clear imaging of fine structures and high-throughput investigations (Driouchi et al. bioRxiv. December 23, 2023). Also, with its robotics capabilities, the company can run thousands of experiments across multiple microscopes every day. Finally, the company’s custom reagents can facilitate assays that identify drug candidates for cancer, neurodegenerative and inflammatory disease.
Visualization and quantification
By combining advanced imaging with rigorous data analysis, super-resolution microscopy is generating quantitative information about absolute numbers of molecules, molecular distributions, spatiotemporal dynamics, and molecular interactions. Super-resolution microscopy data is also contributing to multimodal investigations and predictive computational models of cellular behavior and molecular mechanisms.
“[Biological discovery does not necessarily mean revealing a new structure,” Prakash et al. noted. “High(er) throughput/content, along with elaborate data analyses, are becoming increasingly important for cutting-edge research involving super-resolution microscopy, and whatever the new findings, these need then be confirmed with orthogonal methods. Ideally, super-resolved images and data spark researchers to think differently about their particular biological problem and to question long-held assumptions.”
