Seeing the Unseen: Super-Resolution Microscopy in Protein Aggregation Research

Abstract

Super-resolution microscopy surpasses the diffraction limit and enables the visualization of biomolecular structures with unprecedented detail. These techniques have been widely used in many scientific areas, including cell biology, genomics, microbiology, and material science. In the field of protein aggregation, a process intimately linked to numerous neurodegenerative diseases, the high spatial resolution of super-resolution microscopy enables the direct observation of the fine structure of different species, ranging from small oligomers to mature aggregates, providing insights into molecular aggregation mechanisms and the pathology of neurodegenerative diseases, such as Parkinson’s, Alzheimer’s, and Huntington’s disease. In this review, we outline the principles of three major super-resolution microscopy techniques, including stimulated emission depletion (STED), structured illumination microscopy (SIM), and single-molecule localization microscopy (SMLM), and compare their respective strengths and limitations in studying protein aggregation. We then highlight the recent applications of these techniques in studying protein aggregation, with a focus on aggregate morphology, dynamic formation processes, and interactions with cellular components.

Introduction

Protein aggregation is a widespread biological phenomenon, with profound implications across various domains of our daily lives and industry, e.g., milk production, , industrial and pharmaceutical manufacturing. − Under certain conditions such as stress, genetic mutations, or aging, proteins may misfold and assemble into insoluble aggregates, particularly amyloid aggregates. This aberrant aggregation process is intimately linked to a wide spectrum of neurodegenerative diseases, including Parkinson’s disease (PD), Alzheimer’s disease (AD) and Huntington’s disease. − Aggregated protein species can severely disrupt cellular homeostasis by interfering with essential biochemical pathways, compromising cell membrane integrity, and impairing organelle function, and provoking chronic inflammatory responses. Collectively, these pathological changes cause neuronal dysfunction and eventually neuronal cell death contributing to the progressive decline of neurological function.

Deciphering the molecular aggregation mechanisms is vital for understanding disease ethology, and for the development of effective therapeutic interventions for neurodegenerative disorders. Protein aggregation typically follows a nucleation-dependent polymerization pathway, which is generally divided into the “on-pathway” and the “off-pathway”, and can be described by three kinetic phases. The first phase is an initial, relatively slow lag phase, during which protein monomers form soluble oligomeric intermediates through the “on-pathway”. This is followed by a rapid growth or elongation phase, in which the oligomers aggregate into protofibrils. The process terminates in a final plateau phase, where the mature aggregates are predominantly stable. Although initial aggregation via the “off-pathway” results in a similar kinetic profile, the monomers usually grow directly into amorphous aggregates. − The process terminates in a final plateau phase, where the mature aggregates are predominantly stable. Protein aggregation is a highly heterogeneous process, with each step leading to numerous structural variations, resulting in diverse morphologies of protein aggregates or intermediates. Depending on their morphology and structural order, aggregates can be broadly classified into three categories: amorphous aggregates, which lack specific high-order structure; amyloid fibrils, which consist of highly ordered β-sheets; and amyloid oligomers that contain a combination of ordered β-sheets and native-like structures (see Figure ). Understanding the complex aggregation process and the resulting structural diversity at the molecular level is essential for linking specific aggregate species to cellular toxicity and disease progression, and for developing targeted therapeutic strategies.

1.

Simplified example of protein aggregation via the “on” and “off” pathways, demonstrating the rudimentary structural differences between the main categories of aggregates; oligomers, protofibrils, amyloid fibrils, and amorphous aggregates.

A wide range of biophysical and imaging techniques have been employed to characterize protein aggregation at different levels of resolution and throughput. Conventional readouts like absorbance or fluorescence spectroscopy, scattering methods (dynamic light scattering (DLS), small-angle X-ray (SAXS), neutron scattering), − transmission electron microscopy (TEM), atomic force microscopy (AFM), and confocal microscopy, have been used to study protein aggregation. Each technique offers distinct strengths and limitations.

Fluorescence spectroscopy based on fluorescent amyloid-binding probes, such as Thioflavin T (ThT), which selectively bind to β-sheet structures and exhibit enhanced fluorescence upon binding, remains a cornerstone for monitoring aggregation kinetics. − However, these measurements rely on ensemble averaging and thus provide information only on the average assembly kinetics of heterogeneous samples. As a result, they cannot resolve the behavior of individual aggregates and are challenged to resolve distinct subpopulations within the sample. Additionally, the probe concentration can influence the rate of aggregation, and there can be competitive binding between the probe and other additives, such as small inhibitors, , complicating quantitative information extraction. DLS conveniently measures the average hydrodynamic size distribution of the aggregates; however, it has a demanding sample preparation due to dust sensitivity and a low data output. SAXS, while is valuable for resolving the shape of aggregates, their diversity and conformation changes, it is a lengthy process and often with low throughput. TEM and AFM can image the structure of aggregates to subnanometer resolution, and high-speed AFM has also been used to monitor protein aggregation in real time – albeit for a few minutes. However, the sample and surface preparation are complicated, and may require further optimization. Additionally, these methods are not suitable for detecting protein aggregation in live cells.

Traditional fluorescence microscopy techniques, including widefield and laser scanning confocal microscopy, have also been utilized to study protein aggregation. − Yet, these methods are constrained by the diffraction limit of light (∼200 nm), which precludes the visualization of molecular-scale details of aggregate formation and growth. Consequently, while they provide valuable contextual information, they fall short in resolving the nanoscale heterogeneity and dynamics of protein aggregates.

Super-resolution microscopy (SRM), which breaks through the diffraction limit of conventional optical microscopy and allows visualization of biomolecules at the nanoscale to the single-molecule level, has become a vital tool in studying protein aggregation. − By achieving spatial resolutions well below 100 nm, and in some modalities even approaching molecular-scale precision, SRM provides an unprecedented window into the structural and dynamic landscape of protein aggregation, enabling the direct observation of heterogeneous populations of aggregates ranging from early stage oligomers to amyloid fibrils (Figure ), in situ and in real time. − The capacity to resolve individual aggregates and monitor their formation pathways has revealed previously inaccessible aspects of the aggregation process, such as the nucleation sites, the morphology of transient species, and the spatial organization of aggregates even within the cellular environment. − As a result, super-resolution microscopy has emerged as an indispensable tool for dissecting the multiscale complexity of protein aggregation, paving the way for novel mechanistic insights and potential therapeutic strategies.

In this review, we first briefly introduce several common SRM techniques, including stimulated emission depletion (STED), , structured illumination microscopy (SIM), , and single-molecule localization microscopy (SMLM), − through a comparative approach, highlighting their principles, strengths and limitations. We then examine the key advances these SRM techniques have contributed to protein aggregation research over the past five years, emphasizing their transformative impact on the field.

Super-Resolution Microscopy

SRM encompasses a suite of innovative imaging techniques that overcome the diffraction limit of conventional optical microscopy, enabling nanoscale visualization of biomolecules with unprecedented spatial precision. The widely used super-resolution techniques can be broadly classified into three categories: STED, SIM, and SMLM. In addition to these optical-based methods, expansion microscopy (ExM) offers an alternative strategy by physically expanding the specimen through isotropic swelling of a polymer matrix, enabling fine structures to be resolved with resolution 20–70 nm. − Together, these techniques allow high-resolution imaging of biological structures at the nanoscale.

Stimulated Emission Depletion (STED)

STED utilizes stimulated emission from an excitation laser and a depletion laser shaped into a characteristic doughnut profile, to effectively compress the point spread function (PSF) of the emitting point to a subdiffraction-limit volume (Figure , left), thus bypassing the limitation of optical diffraction. The lateral resolution of STED can reach 30–80 nm. Since STED uses optical methods to achieve super-resolution, it does not require complex postprocessing, offering in essence the potential of real-time imaging capabilities and making it well-suited for dynamic studies of protein aggregation. However, it typically needs high-intensity laser illumination (∼MW/cm²), which results in more photobleaching and phototoxicity of the samples, especially during live-cell imaging or prolonged acquisition sessions.

2.

Schematics of the fundamental optical framework (top row) and the imaging process (bottom row) for the major super-resolution techniques STED (left), SIM (middle), and SMLM (right).

Structured Illumination Microscopy (SIM)

SIM enhances spatial resolution by illuminating samples with high-frequency patterned light at multiple orientations and phases, generating interference patterns (Moiré patterns) that effectively encode subdiffraction-limit information into observable spatial frequencies (Figure , middle). Computational reconstruction of multiple phase-shifted images results in approximately a 2-fold enhancement in spatial resolution compared to conventional optical microscopy. SIM has undergone significant advancements over the past 20 years. For example, in 2005, Gustafsson et al. proposed Saturated Structured Illumination Microscopy (SSIM), which leverages the nonlinear response of fluorophores under high-intensity patterned illumination, and generates higher-order harmonic signals that allow extraction of finer structural details, significantly improving the lateral resolution beyond conventional SIM limits to approximately 50 nm. 3D SIM, developed in 2008, extends conventional SIM by employing patterned illumination projected along multiple orientations and depths within the sample, allowing volumetric images to be reconstructed with approximately twice the lateral and axial resolutions compared to traditional microscopy. The resolution of 3D SIM has been further enhanced to 100–120 nm by improving the imaging setup. , Notably, SIM offers a favorable balance between resolution, speed, and phototoxicity, enabling gentle, volumetric imaging of fragile biological systems.

Single-Molecule Localization Microscopy (SMLM)

SMLM achieves super-resolution by temporally isolating fluorescence emission from individual molecules, typically through stochastic switching between fluorescent (″on″) and dark (″off″) states (Figure , right). , By controlling the random photoswitching of fluorophores and activating only a small subset of them at any given time, their positions can be localized with high precisionoften down to 10–20 nm. , Iteratively imaging the same field of view over thousands of cycles captures localization data for all molecules, allowing reconstruction of a super-resolution image with a resolution of ∼20–40 nm. Different SMLM approaches use various fluorophores or labeling strategies: Photoactivated localization microscopy (PALM) uses photoactivatable fluorescent proteins as fluorophores, (direct) stochastic optical reconstruction microscopy ((d)­STORM) uses blinking dyes or pairs of organic dyes as fluorophores, while point accumulation for imaging in nanoscale topography (PAINT), such as DNA-PAINT and Exchange-PAINT, implements temporarily binding fluorophores and enables multiplexed super-resolution imaging of different targets, by using the same fluorescence channel after exchanging different complementary imager strands labeled with the same dye. − PAINT methods offer several advantages, including reduced photobleaching, tunable binding kinetics, and compatibility with both fixed and expanded samples. Unlike STED microscopy, SMLM generally does not rely on specialized imaging hardware. However, its reliance on extended imaging acquisition times and intensive computational reconstruction can limit temporal resolution and throughput.

Each of these SRM toolboxes offers distinct advantages and has its own limitations for interrogating different facets of the protein aggregation process (see Table ). STED microscopy excels in providing immediate, high-resolution structural insights, particularly beneficial for visualizing detailed structures of aggregates within biological samples; however, the imaging area is small, and it requires high-intensity illumination, which can cause significant photobleaching and phototoxicity. SIM offers moderately improved resolution in both lateral and axial dimensions with relatively gentle illumination, enabling effective and rapid imaging of delicate structures and live-cell studies with minimal photodamage. SMLM achieves the highest spatial resolution through precise molecular localization, making it ideal for detecting aggregates including oligomers and small amorphous aggregates, and tracking the dynamics of the aggregation processes at the single-molecule level, though it requires sophisticated data analysis and computational resources, and is limited by slower acquisition speeds.

1. Comparison of Super-Resolution Techniques and Their Suitability for Protein Aggregation Studies.

SRM techniques	STED	SIM	SMLM
resolution	∼30–80 nm	∼100–130 nm	∼20–40 nm
imaging speed	moderate (seconds to minutes per frame depending on the scanning rate of its confocal platform)	fast (subseconds to seconds per frame)	slow (minutes per image due to required molecule localization steps)
fluorophore options	photostable fluorophores	broad compatibility	photoswitchable dyes (dSTORM); photoactivable fluorescent proteins (PALM); antibody/DNA-linked fluorophore (PAINT).
phototoxicity	moderate to high (intense depletion laser)	low to moderate (gentle structured illumination)	moderate to high (prolonged imaging and strong illumination, usually needs imaging buffer)
strengths	high spatial resolution (∼30–80 nm); rapid imaging relative to SMLM; good for observation of dense structures.	fast, gentle imaging suitable for live cells; compatible with common fluorophores; ideal for dynamic processes.	highest spatial resolution (∼20–40 nm); precise localization of single molecules; detailed structural information.
limitations	photobleaching and phototoxicity concerns; requires highly photostable dyes; specialized laser setups and optical alignment.	limited resolution; cannot resolve finest structural details; artifacts from image reconstruction.	slow image acquisition; specialized fluorophores required; computationally intensive postprocessing.
suitability for specific aggregation contexts	effective for moderately sized, stable aggregates where the balance of resolution and speed is needed; ideal for fixed or minimally dynamic samples.	optimal for live-cell, dynamic aggregation events; good for observing aggregation kinetics.	ideal for high-resolution structural studies, e.g., amyloid fibrils and oligomers; suitable for detailed analysis of fixed, static samples; possible for observing aggregation kinetics (e.g., REPLOM).

Together, these super-resolution methodologies have advanced our understanding of protein aggregation, offering direct recordings of the structural transitions and spatial dynamics of aggregates, ranging from the molecular to the cellular scales. In the following section, we explore recent advances in the application of SRM to protein aggregation research, focusing on studies from the past five years that have not only expanded mechanistic understanding but also uncovered novel biological insights with implications for disease pathology and therapeutic strategies.

Application of Super-Resolution Microscopy in Protein Aggregation

SRM has made significant advances in the field of protein aggregation. These techniques enable the direct observation of molecular structures and dynamic events with high spatial and temporal resolution, that are usually inaccessible with conventional microscopy or ensemble methods. For example, due to their ability to resolve structures below the diffraction limit, SRM techniques can detect crucial small oligomers implicated in toxicity and disease pathology. By providing nanoscale-level visualization of aggregate morphology development, these techniques enable researchers to identify and characterize critical aggregation intermediates, such as elucidating structural transitions during the aggregation process, and closely examine the interactions between protein aggregates and cellular components. These are important to reveal insights into aggregation pathways, cellular interactions, and molecular mechanisms underlying disease progression. In this section, we review recent key findings obtained by applying STED, SIM and SMLM (see Table ), and emphasizing how these techniques have deepened our understanding of protein aggregation and its pathology in neurodegenerative diseases, and facilitating the development of therapeutic strategies.

2. Overview of the Recent Application of Super-Resolution Techniques to Protein Aggregation.

protein	SRM techniques	application/study topic
α-synuclein	STED	morphology; aggregation pathways; toxicity and interaction with the membrane; , distribution
α-synuclein	SMLM	morphology; − ,, aggregation pathways; , interaction with proteasomes
α-synuclein	SIM	interactions with acetylated α-tubulin
Aβ	STED	morphology/structure; , interaction with the membrane
Aβ	SMLM	morphology ,,,
Aβ	SMLM & SIM	Aβ-receptor interaction
tau	STED	GVBs-tau pathology
tau	SMLM	morphology/size; ,,,, aggregation pathways
TAR DNA-binding protein 43 (TDP-43)	SMLM	aggregation pathways
hen egg white lysozyme (HEWL)	SMLM	aggregation pathway
insulin	SMLM	aggregation pathway −
TDP-43	STED	interaction with SGs
human light chain (hLC) protein	STED	morphology
serum amyloid A	STED	interaction between ASC and SAA
p62/sequestome1	3D SIM	structure of the assembly

Direct Visualization of the Structure and Morphology of Protein Aggregates

Protein aggregation involves diverse molecular pathways, resulting in a variety of structurally distinct aggregated species, such as oligomers, amyloid fibrils, and amorphous aggregates etc., which vary in size, morphology, and biochemical properties (see Figure ). Revealing fine structural details of various aggregates is essential for understanding protein aggregation mechanisms and their role in disease pathology. SRM has emerged as an invaluable tool to visualize directly and study the structures of protein aggregates both in vitro and in cells. ,,

Historically, the pathological hallmark deposits such as neurofibrillary tangles and amyloid plaques were considered the primary elements in neurodegenerative diseases like AD. However, the advent of SRM has challenged this view by revealing that oligomers are generally more toxic to cells than mature aggregates. , Examples are shown in Figure , which are the representative super-resolution images of oligomers and fibrils from different proteins. These findings have prompted a paradigm shift in the field, directing increasing attention toward the structural and functional properties of intermediate aggregate species. By enabling the visualization of these small, dynamic, and previously elusive oligomeric forms, SRM has provided powerful new insights into the early stages of protein aggregation and their pathological relevance. Several illustrative studies below further demonstrate how SRM has been used to dissect the structural complexity and biological roles of protein aggregates.

3.

Representative super-resolution images of oligomers and fibrils from (A) p62, (B) α-synuclein, (C) tau, (D) hLC, (E) Aβ, (F) TDP-43, and (G) insulin. (A) SIM imaging of p62 in ALIS cells after LPS stimulation for 8, 16, and 24 h, as compared to control (untreated) cells. Adapted with permission from ref . Copyright 2024 Bhatnagar et al. The American Society for Cell Biology. (B) STED imaging of primary rat cortical neurons treated with type-B* prefibrillar oligomers (left) and short fibrils (right), with cell membranes in red and α-synuclein in green. The smaller aggregates are mostly internalized in the cell (left), while the larger aggregates are found outside the cell (right). Adapted with permission from ref . Copyright 2021 Springer Nature. (C) SMLM imaging tau in clone 4.1 cells expressing 4R-P301L-GFP tau, with the Voronoi segmentation showed for the zoomed-in regions, which displays nanoclusters (magenta circles), fibrillary structures (green circles), branched fibrils (yellow circles), and conglomerate NFT-like structures (white circles). Reprinted with permission from ref . Copyright 2021 PNAS. (D) STED imaging of hLC aggregates prepared with (left to right) TCEP, DTT, GSH reduction, and by heating. Adapted with permission from ref . Copyright 2022 Elsevier B. V. (E) Conventional (left) and SMLM (right) imaging of amyloid-β aggregates; the merged imaging of AT630 (green) and anti-Aβ antibody 6E10 (red) stained the App^NL‑G‑F mouse brain tissue. Reprinted with permission from ref . Copyright Morten et al., published by PNAS. (F) RRM1-2 (a domain of TDP-43) aggregates probed using Apt-1 and imaged by aptamer DNA-PAINT. Reprinted with permission from ref . Copyright 2022 Springer Nature. (G) dSTORM images of human insulin aggregates. The pseudocolor scale, ranging from 0 to 400, corresponds to the density of neighboring events within a 100 nm radius sphere from the localization. Adapted with permission from ref . Copyright 2022 Springer Nature.

Building on these findings, STED has played a key role for visualizing the fine structure of amyloid aggregates with nanoscale precision. Studies using STED to detect amyloid beta (Aβ) plaques in brain tissue have provided critical insights into the structure of fibrils and nonfibrillar species, such as oligomers, which may play an important role in Aβ plaque stability and toxicity. , Additionally, STED imaging of pathological human multiple myeloma light chain (hLC) aggregates revealed that the final branched morphologies were highly variable depending on the reducing agent (see Figure D). This contributes to our understanding of how protein aggregation under reducing conditions influences disease progression.

In parallel, SMLM has been utilized to investigate the relationship between aggregate size and toxicity of α-synuclein, showing that aggregates smaller than 450 nm can cross the plasma membrane and are more cytotoxic than larger fibrils. Combining SMLM with AFM has allowed researchers to study the sizes of α-synuclein aggregates, and found that α-synuclein aggregates in the serum and cerebrospinal fluid of PD patients range from 20 to 200 nm, with a significantly increased number of large aggregates than controls. Furthermore, a super-resolution imaging platform based on SMLM has been developed to detect and quantify α-synuclein aggregates in skin biopsies with nanoscale spatial resolution. All these SRM works suggest that α-synuclein aggregates can serve as a reliable biomarker for PD diagnosis.

As protein aggregates are usually rich in β-sheet structures, amyloid reactive dyes such as ThT that can temporarily bind to the β-sheet structures, are used together with SMLM to detect the morphology of aggregates at the nanoscale. , These methods avoided the artifacts caused by fixed fluorescent labels, however they cannot specifically recognize a certain type of protein aggregate. For example, they cannot distinguish α-synuclein aggregates from Aβ aggregates when these aggregates coexist in the system. Recently, Klenerman’s lab addressed this issue by using a single-molecule pull-down-based assay that integrates antibody-based immunoprecipitation with super-resolution fluorescence imaging DNA-PAINT, to sensitively detect α-synuclein and Aβ aggregates in human serum. They further applied the assay to identify tau aggregates in post-mortem brain tissues and biofluids from AD patients as well as from healthy controls. By employing a panel of antibodies, it was possible to distinguish between different aggregates and detect compositional profiles of the individual aggregates. Utilizing this method, they further investigated how aggregate size and inflammation affect the clearance of Aβ and tau aggregates in human-induced pluripotent stem cells.

Collectively, SRM techniques have revolutionized our mechanistic understanding of protein aggregation by revealing critical structural heterogeneity, size dependent toxicity, and selective biological behaviors of protein aggregates. − , These advancements not only facilitate the identification and characterization of pathological aggregates at unprecedented resolution, but also significantly contribute to our ability to pinpoint early biomarkers and understand disease progression at the molecular level. Moving forward, SRM will likely play a crucial role in the development of refined diagnostic tools and targeted therapeutic strategies, ultimately leading to improved interventions and patient outcomes in neurodegenerative diseases.

Detection of the Aggregation Processes and Growth Heterogeneity

Protein aggregation is a complex, multistage process involving heterogeneous growth pathways and kinetics. This high degree of heterogeneity results in aggregates with different morphologies and pathological roles in neurodegenerative diseases. Investigating the aggregation pathways and dynamics is crucial to understanding the relationship between the growth pathway and the resulting morphology and properties of aggregates. This will reveal the aggregation mechanism and the pathology of diseases such as AD and PD. ,

SRM has been pivotal in capturing the dynamic processes of protein aggregation, providing critical insights into the kinetics, growth heterogeneity, and mechanistic pathways underlying aggregation. ,,, A widely adopted strategy involves the use of preformed small aggregates or oligomers as seeds that act as nucleation centers to initiate or accelerate aggregation. Generally, the seeds are labeled with one type of fluorophore and the monomers with another. By utilizing two-color SRM to image aggregates at different incubation times, it is possible to analyze the structural evolution and aggregation kinetics at the single aggregate level (Figure A). ,,, This dual-color approach not only provides quantitative information about growth rates and aggregate morphology but also reveals directional growth patterns and the spatial dynamics of monomer incorporation.

4.

Examples of super-resolution techniques for observing the protein aggregation process. (A) Schematic of the short (top) and long (bottom) seeds used to initiate α-synuclein fibrillation (left) and two-color STED imaging of the resulting fibrils, showing that the seed size influences the aggregation process (right). Adapted with permission from ref . Copyright 2022 American Chemical Society. (B) dSTORM imaging of P301S tau fibrillation in HEK293 cells at discrete time intervals. Adapted with permission from ref . Copyright 2023 Dimou et al., published by Elsevier. (C) Schematic of REPLOM (top) and direct imaging of the formation of human insulin (HI) aggregates. Adapted with permission from ref . Copyright 2022 Springer Nature.

Complementary to the two-color strategy, single-color SRM techniques offer powerful means to investigate protein aggregation processesparticularly in cellular environments where multicolor labeling may be challenging or biologically disruptive. For instance, antibody-based DNA-PAINT is able to monitor the self-assembly of exogenous α-synuclein in SH-SY5Y cells, and dSTORM can observe tau aggregation in HEK cells and in neurons (see Figure B). These studies revealed the early stages of protein aggregation mechanisms in cells. In addition, SMLM has allowed researchers to find the critical size threshold for α-synuclein fibrillation is 70 monomers. Aggregates exceeding this size underwent a structural conversion into fibrils, while aggregates smaller than this size remained as protofibrils or globular oligomers. This finding suggests the existence of a critical tipping point, beyond which aggregates rapidly propagate and amplify within cells.

Moreover, recent works demonstrate that combining SRM with rationally designed aptamers or special antibodies (such as conformation-sensitive antibodies) has the potential to image and monitor protein aggregation, opening new avenues for early disease detection and drug development work. , For example, a novel aptamer Apt-1, utilized in conjunction with DNA-PAINT, successfully enabled tracking the aggregation of TAR DNA-binding protein 43 (TDP-43), and distinguishing the distinct structural transitions of TDP-43 aggregates with high resolution.

The studies outlined above have primarily captured protein oligomers or fibrils at discrete time points, providing valuable but inherently stepwise snapshots of the aggregation process. However, recent advances in fluorophore design and super-resolution imaging techniques have paved the way for continuous, real-time visualization of protein aggregation dynamics at nanometer resolution (∼50 nm). ,

Notably Fan et al. exploited the aggregation-induced emission (AIE) properties of amyloid reactive fluorophore PD-BZ–OH, and combined with SMLM to track the fibrillation of Hen egg white lysozyme (HEWL) in situ, providing valuable information about the dynamic aggregation events at nanoscale.

Our group recently developed a super-resolution method termed REal-time kinetics via binding and Photobleaching LOcalization Microscopy (REPLOM) based on SMLM. Typically, protein monomers are covalently labeled with organic dyes, such as Alexa Fluor 647, and mixed with unlabeled monomers. Initially, small protein condensates, such as cores, form and attach to the surface. The spatial position of each fluorophore is accurately localized before it undergoes photobleaching. By optimizing imaging conditions and omitting imaging buffer, rapid chromophore bleaching is ensured shortly after binding, allowing individual binding events to be temporally resolved. As aggregation progresses, additional monomers bind to the growing core, increasing the size of the aggregate. Each labeled binding event produces a diffraction-limited fluorescent spot, the position of which can be precisely extractedproviding both spatial and temporal resolution of the aggregation process (see Figure C). This method allows direct observation of the real-time formation of protein aggregates, and quantification of the existence and abundance of diverse morphologies as well as their heterogeneous growth kinetics. −

These developments mark a significant milestone, enabling researchers for the first time to move beyond static imaging and directly observe the formation, morphological transitions, and the heterogeneous growth kinetics of protein aggregates in real time.

Time series or real-time imaging by SRM enables visualization and measurement of the size differences and morphological development in various species, e.g. oligomers, and oligomer-amplified products, at the nanometer scale. − ,, These techniques elucidate heterogeneous aggregation pathways and their relevance to the properties and functions of protein aggregates, thereby revealing the complex mechanistic details of protein aggregation. These insights deepen our understanding in the pathological development processes associated with aggregation-related diseases, guiding the discovery of novel therapeutic targets and informing the development of strategies aimed at effectively mitigating or preventing neurodegenerative diseases.

Investigations of the Interactions between Aggregates and Cellular Components

The structure, distribution, and aggregation dynamics of protein aggregates are affected by the intracellular environment. Understanding the interactions between protein aggregates and cellular components is critical for unraveling the mechanisms underlying aggregate toxicity and disease progression. SRM provides a powerful means to investigate protein aggregation and its cellular context with nanoscale spatial resolution. This enables researchers to resolve individual aggregates and cellular components with remarkable precision. With these capabilities, SRM has uncovered critical insights into the localization and dynamics of aggregates within cells, as well as their interactions with various cellular components. −

Several of the studies discussed in the previous sections also explored the interactions between protein aggregates and cellular components, such as membranes , and proteasomes. Here, we highlight key advances enabled by SRM in investigating the complex biological interactions underlying pathological protein aggregation. These studies indicate that SRM has been instrumental in uncovering how pathological protein aggregation interacts with key cellular pathways implicated in neurodegenerative disorders.

The link between protein aggregation and cellular stress response is one of the central theme in neurodegenerative disorders. Stress granules (SGs) are protective cellular structures that play a key role in responding to acute stress. They have been implicated as potential nucleation sites for protein aggregation linked to neurodegenerative disorders. SRM combined with single-molecule tracking demonstrated that the stress-related mobility loss of TDP-43 occurs independently of SGs. This finding suggested that stress-induced biophysical states of TDP-43 may predispose it to pathological aggregation, establishing a direct link between cellular stress responses and protein aggregation pathways.

Cell-surface receptors represent another critical modulatory interface. In AD, toxic Aβ oligomers bind to receptor proteins or lipids on the surface of neurons, initiating neurotoxic signaling. Studies using SRM have directly observed the Aβ-receptor (including PrP^C, FcγRIIb and LilrB2) interactions, revealing that these receptors do not bind Aβ randomly but target specific structural features present at early and dynamic aggregation stages (see Figure A). Their interactions may drive early neurotoxicity by amplifying toxic signaling at the synapse, offering a mechanistic explanation for receptor-mediated vulnerability in AD.

5.

Examples of the application of structured illumination microscopy (SIM) to study protein aggregations. (A) Schematic and resulting images of monomeric Cy3-labeled amyloid-β elongating into fibrils from preformed Cy5-labeled seeds and cellular prion proteins (PrP) apparent binding to the growing fibril end, indicating that PrP inhibits elongation. Adapted with permission from ref . Copyright 2021 Springer Nature. (B) Distribution of α-synuclein and acetylated α-tubulin along a bundle of microtubule, with a zoom-in (right) of the section marked by a white outline (left). Adapted with permission from ref . Copyright 2023 Calogero et al., published by MDPI.

Lysosomal dysfunction is another critical element connecting protein aggregation to cellular dysfunction. STED has been used to study Granulovacuolar Degeneration Bodies (GVBs), membrane-bound lysosomal structures that accumulate in neurons of patients with neurodegenerative disorders, and their connection to tau pathology. The ultrastructure of tau-induced GVBs and their lysosomal characteristics resolved by SRM at nanoscale resolution directly linked tau pathology to lysosomal dysfunction.

Inflammatory signaling also contributes to protein aggregation. Study using STED has shown that inflammasome adaptor protein ASC (apoptosis-associated speck-like protein containing a caspase recruitment domain) colocalized with SAA aggregates (serum amyloid A – a protein that associates with the systemic disease of amyloid A amyloidosis). This suggested that ASC may act as a structural scaffold or nucleation center, which initiates or enhances SAA aggregation. This finding highlights the potential of inflammatory responses to initiate or amplify protein aggregation.

Additionally, cytoskeletal modifications may also influence protein aggregation. Investigation using SIM has revealed the colocalization between acetylated α-tubulin and α-synuclein along microtubules, demonstrating that increased acetylation significantly promoted α-synuclein oligomerization (see Figure B). It suggested that acetylated α-tubulin may act as a scaffold or aggregation-promoting surface, highlighting the importance of cytoskeletal in the pathology of neurodegenerative disease.

Collectively, these SRM studies have clarified how cellular environment modulates the localization, dynamics, and distribution of aggregates. These insights highlight the critical role of cellular components or signals, such as cellular stress responses, membrane receptors, organelles, inflammation, and cytoskeletal architecture, in shaping the aggregation landscape and toxicity of pathological protein species. SRM continues to provide unparalleled insights into these complex interactions, offering not only mechanistic understanding but also potential avenues for therapeutic intervention aimed at disrupting harmful aggregate–cellular interactions.

Quantitative Analysis in SRM

Quantitative analysis of protein aggregation captured by SRM usually needs to segment and classify the amyloid structures first. This is usually done by automated or semiautomated image analysis algorithms, such as Enhanced Classification of Localized Point clouds by Shape Extraction (ECLiPSE) and Segmentation and MORphological fingErprinting (SEMORE), which allows the identification of individual aggregates, including oligomers, fibrils and spherulites etc. Due to the high resolution of SRM, we are able to extract the precise size of individual aggregates through their areas or lengths, as well as their numbers or abundance. The number of protein monomers per aggregate can also be estimated by quantifying the localization numbers of individual aggregate detected by SMLM or by measuring the fluorescence intensity ratio between single aggregate and monomers. Time series snapshot or real-time imaging by SRM enable the tracking of dynamic changes of protein aggregates, thus quantifying the growth kinetics or nucleation rates. , Dual- or multicolor SRM enables precise spatial mapping and colocalization analysis of different protein aggregate species or organelles, often employing coefficients like Pearson’s , or Manders’ to quantify interactions or proximity at the nanometer scale. Statistical analysis is integral throughout to compare quantitative features such as sizes, morphology, and distributions of aggregate species between conditions (e.g., control vs disease, seeded vs unseeded), ,, revealing how nanoscopic aggregate features correlate with toxicity, disease progression, or response to cellular stress. Overall, the quantitative analysis in SRM indicate hidden structural motifs or clustering behaviors within aggregate populations, advancing a more nuanced understanding of aggregation-related processes in health and disease.

Conclusions and Future Perspectives

In this review, we have summarized the principles of three major SRM techniquesSTED, SIM and SMLMand reviewed their recent applications in the field of protein aggregation, over the last five years. Protein aggregation is inherently heterogeneous, resulting in multiple species of protein aggregates that vary in dimension, morphology, and pathological potential. The dimension and morphology of protein aggregates are determined by the aggregation pathway, which in turn influence their behaviors, and the onset and progression of neurodegenerative diseases. The ability of SRM to surpass the diffraction limit enables the identification of diverse aggregate species, especially the small oligomers with sizes below the diffraction limit, the morphology of which is impossible to detect with conventional optical microscopy. By monitoring the morphological evolution of aggregates at the nanoscale, SRM reveals critical insights into aggregation pathways. In addition, these super-resolution techniques can map the distribution of aggregates in cells, uncovering cellular targets, and elucidating the pathological mechanisms. These findings enhance our understanding of protein aggregation pathology and guide the development of targeted therapeutic strategies.

While SRM has significantly advanced our understanding of protein aggregation, several promising research and technological directions remain essential for further progress. Minimizing photobleaching and phototoxicity is a key step for long-time dynamic or live-cell imaging. Achieving this requires developing fluorophores with high brightness and high photostability, , exploiting amyloid-specific probes such as AmyBlink-1 and Amytracker, − as well as AIE fluorophores, ,, and optimizing labeling methods with high specificity to aggregates. , Moreover, these advanced fluorophores and labeling/detecting techniques will enable SRM to detect early stage aggregation biomarkers associated with neurodegenerative diseases, facilitating timely and accurate diagnosis and intervention.

Advances in imaging technologies, especially 3D imaging capability with high spatial and temporal resolution are crucial for the simultaneous observation of many aggregates and capturing their dynamic events and interactions within complex cellular environments. One solution is to combine SRM to lattice light-sheet microscopy (LLSM), which addresses the challenge of imaging live-cell dynamics with minimal phototoxicity by illuminating samples with a thin sheet of light structured into a lattice pattern, , providing high-speed volumetric imaging with improved axial resolution (∼400–500 nm). The hybrid SRM-LLSM techniques such as SIM-LLSM excel in capturing rapid biological processes, making it particularly suited for observing protein aggregation dynamics in live cells and complex tissue environments over extended periods.

Integrating SRM with complementary biophysical and biochemical methods would provide more comprehensive insights into the aggregation process and its implications for disease pathology. ,,,,, In addition, the advancement and broader accessibility of next-generation SRM technologies also has great potential in the field of protein aggregation. For example, MINFLUX (Minimal Photon FLUxes) is one of the latest advancements in super-resolution imaging, it combines elements of SMLM and STED to achieve record-breaking spatial precision as fine as 1–3 nm. , This unprecedented resolution allows direct observation of molecular-scale interactions and dynamic processes, making MINFLUX a powerful tool for studying the nucleation and growth of protein aggregates with unparalleled detail. Although the complexity and cost of instrumentation currently limit widespread adoption, MINFLUX holds immense promise for future investigations of protein aggregation at the single-molecule level.

Furthermore, the growing complexity and volume of data generated by SRM have driven the integration of machine learning (ML) algorithms into image analysis workflows. − Future developments should prioritize creating advanced ML algorithms capable of automated, high-throughput processing of diverse protein aggregate morphologies and dynamic behaviors.

The continued technological and methodological developments in SRM are expected to unlock deeper insights into the fundamental mechanisms of protein aggregation, paving the way for improved diagnostic and therapeutic approaches for neurodegenerative diseases and other protein aggregation-related disorders.

Glossary

Vocabulary Section

Resolution

The ability of an imaging system to distinguish two closely spaced points as separate objects.

Super-resolution microscopy

A family of optical imaging techniques that surpass the diffraction limit of conventional light microscopy, enabling visualization of structures at the nanometer scale.

Protein aggregation

The process by which proteins self-assemble into soluble oligomers or insoluble aggregates, such as amyloid fibrils. It is implicated in a spectrum of neurodegenerative diseases, such as Alzheimer’s, Parkinson’s, and Huntington’s disease.

Oligomers

Small assemblies containing a few protein monomers that form during the early stages of aggregation. Oligomers are generally more toxic to cells than mature aggregates.

Amyloid fibrils

Highly ordered, insoluble, β-sheet-rich protein fibers. The accumulation and deposition of amyloid fibrils are a hallmark of neurodegenerative diseases.

M.Z. provided the initial concept; M.J.M.T. and M.Z. contributed to writingoriginal draft preparation; M.J.M.T., J.C., N.S.H. and M.Z. contributed to writingreview and editing; M.J.M.T. prepared the figures; All authors read and approved the final version of the manuscript.

The authors declare no competing financial interest.