Nur Aininie
Yusoh
a,
Martin R.
Gill
*b and
Xiaohe
Tian
*a
aDepartment of Radiology, Huaxi MR Research Center (HMRRC), Institution of Radiology and Medical Imaging, West China Hospital of Sichuan University, Sichuan University, Chengdu, Sichuan, China. E-mail: xiaohe.t@wchscu.cn
bDepartment of Chemistry, Faculty of Science and Engineering, Swansea University, Swansea, UK. E-mail: m.r.gill@swansea.ac.uk
First published on 21st February 2025
Super-resolution microscopy (SRM) has transformed biological imaging by overcoming the diffraction limit, offering nanoscale visualization of cellular structures and processes. However, the widespread use of organic fluorescent probes is often hindered by limitations such as photobleaching, short photostability, and inadequate performance in deep-tissue imaging. Metal complexes, with their superior photophysical properties, including exceptional photostability, tuneable luminescence, and extended excited-state lifetimes, address these challenges, enabling precise subcellular targeting and long-term imaging. Beyond imaging, their theranostic potential unlocks real-time diagnostics and treatments for diseases such as cancer and bacterial infections. This review explores recent advancements in applying metal complexes for SRM, focusing on their utility in visualizing intricate subcellular structures, capturing temporal dynamics in live cells and elucidating in vivo spatial organization. We emphasize how rational design strategies optimize biocompatibility, organelle specificity, and deep-tissue penetration, expanding their applicability in multiplexed imaging. Furthermore, we discuss the design of various metal nanoparticles (NPs) for SRM, along with emerging hybrid nanoscale probes that integrate metal complexes with gold (Au) scaffolds, offering promising avenues for overcoming current limitations. By highlighting both established successes and potential frontiers, this review provides a roadmap for leveraging metal complexes as versatile tools in advancing SRM applications.
Key learning points(1) Understand the fundamental concepts of super-resolution microscopy (SRM) and its key techniques.(2) Gain insights into the limitations of conventional fluorescent probes and how metal complexes address these challenges. (3) Explore the structural design and tuneable properties of metal complexes as imaging agents. (4) Learn about the recent advances in using metal complexes for nanoscale imaging in biology and medicine. (5) Identify challenges and future directions for integrating metal complexes into SRM for broader applications. |
A major breakthrough in overcoming these limitations emerged with the development and application of super-resolution microscopy (SRM) techniques, which surpass the classical diffraction limit of conventional light microscopy to achieve nanometre-scale resolution.3–5 SRM has revolutionized biological imaging by expanding the capabilities of light microscopy, allowing visualization of subcellular structures and molecular processes (Fig. 1), at a level of detail previously accessible only through electron microscopy (EM). Notably, SRM retains the advantages of optical microscopy, including sample preservation, imaging flexibility, and target specificity.3–5 These techniques now enables ultrastructural characterization of a broad range of cellular components, with the flexibility to use diverse fluorescent dyes targeting multiple biomolecules simultaneously – a key advantage over EM.4–6 This is particularly valuable for studying complex biological systems, as it integrates seamlessly with techniques such as co-localization analysis, fluorescence resonance energy transfer (FRET), and fluorescence recovery after photobleaching (FRAP), enabling dynamic investigations in living cells.7 Furthermore, SRM allows for the real-time visualization of interactions between labelled biomolecules, providing detailed insights into the dynamics of cellular responses to various stimuli. For example, it has been used to study the spatial organization of protein complexes, track intracellular transport processes, and observe the kinetics of membrane trafficking in live cells.3–6,8 These capabilities provide insights into molecular behaviour under native conditions and offer direct experimental feedback for modelling complex biological interactions. As a result, SRM has transformed the study of complex biological systems, from intracellular signalling to tissue architecture, making it indispensable in fields such as virology,9 immunology,10,11 neuroscience,12,13 and cancer research.14,15
![]() | ||
Fig. 1 Super-resolution imaging of subcellular structures of eukaryotic cells showing (i) nucleolus, (ii) chromatin substructures, (iii) lipid droplets, (iv) cytoskeletal components, (v) mitochondria, (vi) endoplasmic reticulum (ER), (vii) ribosome, and (viii) lysosome. Image (i) reprinted from Biosensors and Bioelectronics, 203, Liu, J. et al., Nucleolar RNA in action: Ultrastructure revealed during protein translation through a terpyridyl manganese(II) complex, 114058, Copyright 2022, with permission from Elsevier. Images (ii), (v) and (vii) reprinted (adapted) with permission from ref. 16–18, respectively. Copyright 2017, 2017 and 2024, respectively, American Chemical Society. Images (iii), (vi) and (viii) reproduced from ref. 19–21, respectively, with permission from the Royal Society of Chemistry. Image (iv) adapted from ref. 22 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Advanced Materials published by John Wiley & Sons Ltd. © The Author 2020. |
![]() | ||
Fig. 2 Comparison of the simplified light paths, underlying principles, lateral resolutions, analysis complexity, and probe selection between confocal imaging and various super-resolution techniques and their variants. The reported resolutions for each technique are approximate and can vary depending on experimental conditions and fluorophore properties. Redrawn with permission from ref. 35. Copyright 2022 American Chemical Society. SIM: structured illumination microscopy, STED: stimulated emission depletion, SMLM: single-molecule localization microscopy, 2D: two-dimensional, 3D: three-dimensional, SSIM: saturated structured-illumination microscopy, PALM: photoactivated localization microscopy, STORM: stochastic optical reconstruction microscopy, dSTORM: direct STORM, SOFI: super-resolution optical fluctuation imaging, GSD: ground state depletion, PAINT: points accumulation for imaging in nanoscale topography, DNA-PAINT: DNA-based-PAINT, FRET-PAINT: fluorescence resonance energy transfer-PAINT, res-PAINT: reservoir-PAINT, and MINFLUX: minimal photon fluxes. |
In conventional confocal laser scanning microscopy (CLSM), image information is acquired sequentially by scanning a focused excitation laser across the sample plane (Fig. 2). In contrast, SRM distinguishes fluorophores through the manipulation of fluorescent on and off states. Among SRM techniques, SMLM involves the most complex analysis, STED is of intermediate complexity, and SIM is the least complex.24 SIM enhances resolution by illuminating samples with patterned light, such as striped patterns, which generate Moiré fringes.36–38 These encode high-resolution information into low-frequency beat patterns. By capturing images with different pattern orientations and processing them using Fourier analysis, SIM reconstructs a higher-resolution image. Compared to conventional wide-field (WF) microscopy, approximately two-fold improvement in lateral resolution was achieved using SIM. This approach is effective for visualizing fine details in both fixed and live cells. STED nanoscopy, on the other hand, achieves super-resolution by combining fluorescence with stimulated emission. A fluorescent molecule is excited by a laser pulse, but a doughnut-shaped STED beam depletes fluorescence in the surrounding area, confining emission to a small central region.39,40 This narrows the point spread function (PSF) and enhances spatial resolution. While SIM and STED rely on optical modulation to improve resolution, SMLM achieves nanometre precision by imaging individual fluorophores sequentially and localizing their positions based on the centroids of their emission PSFs, though it requires longer acquisition times. For instance, in PALM, photoactivatable or photoswitchable fluorophores – initially non-fluorescent or weakly fluorescent – are selectively activated and cycled between fluorescent on and off states during illumination, enabling precise localization.26,27 Similarly, STORM uses fluorescent molecules that stochastically switch between bright and dark states, allowing for the reconstruction of high-resolution images by localizing a small subset of fluorophores at a time.27 Overall, due to variations in imaging principles, data acquisition requirements, and fluorophore properties, the detection speed differs significantly between these techniques. Typically, SIM and STED methods are preferred for live-cell imaging.
Recent advancements, such as stimulated emission double depletion (STEDD),41 surface-migration emission depletion (SMED),42 scanning switch-off microscopy (SSM),43 have further advanced super-resolution fluorescence imaging. Building upon the principles of STED, these techniques offer improvements in resolution, specificity, and reduced phototoxicity, albeit with added complexity and the need for specialized equipment, making them ideal for specific imaging applications where conventional STED falls short.
In addition to this, combining optical microscopy with deep learning methods, such as ANNA-PALM – a neural network-based approach – enables significantly faster and less invasive high-throughput live-cell super-resolution imaging.47 By minimizing phototoxicity and increasing imaging speed, ANNA-PALM allows for real-time visualization of dynamic processes, advancing our understanding of disease progression at the molecular level. Looking forward, integrating AI- or machine learning-driven tools for complex dataset analysis may enhances data extraction, supporting advanced and rapid super-resolution imaging.48,49 Moreover, multimodal imaging approaches that combine optical imaging techniques with modalities like magnetic resonance imaging (MRI), X-ray Raman spectroscopy, or volume electron microscopy, offer nanoscale visualization of samples and molecular-level insights, revealing unprecedented information about cellular and tissue states.50 As SRM methods continue to develop, it is essential to match study requirements such as resolution needs, sample type, and imaging conditions, with the most suitable imaging technique.
Technique | Fluorophore requirements | ||
---|---|---|---|
SIM | STED | SMLM | |
Quantum yield (Φ) | High | Moderate | Very high |
Lifetime | Moderate | 500 ps–5 ns | Variable, with dynamic blinking/photo-switching properties |
Photostability | Moderate | High (to withstand depletion laser intensity) | Very high (to ensure prolonged imaging over many cycles) |
Despite these advantages and recent advancements in imaging technology, established imaging agents face limitations that hinder their full potential in SRM applications. Many conventional fluorescent probes emit in the shorter wavelength range, where tissue scattering and absorption limit their effectiveness for deep-tissue imaging. These probes also tend to suffer from nonspecific background signals, which reduce imaging contrast and specificity. Moreover, poor photostability and susceptibility to photobleaching under intense light diminish their effectiveness for long-term imaging. Organic fluorophores generally exhibit emission decay times ranging from 500 ps to 5 ns, which restricts the depletion laser pulse width to a narrow window, typically around 200 to 300 ps. While quantum dots hold promise, they present challenges such as cytotoxicity (e.g., those containing heavy metals like cadmium), blinking behaviour that complicates continuous imaging, difficulties in achieving stable and biocompatible functionalization, and limited tissue penetration within the visible spectrum.61
Their luminescent properties are attributed to charge-transfer processes, such as metal-to-ligand charge-transfer (MLCT), ligand-to-metal charge-transfer (LMCT), ligand-to-ligand charge-transfer (LLCT), intraligand (IL), intraligand charge-transfer (ILCT), ligand-to-metal–metal charge transfer (LMMCT), and metal–metal-to-ligand charge-transfer (MMLCT) states (Fig. 3).66,71 Furthermore, the incorporation of a heavy metal centre with a high spin–orbit coupling (SOC) constant in these complexes facilitates efficient intersystem crossing (ISC) from singlet to triplet excited states upon photoexcitation (Fig. 3). The spin-forbidden transition from the lowest triplet excited state (T1) to the singlet ground state (S0) subsequently gives rise to phosphorescence. Based on their attractive photophysical properties, numerous metal complexes have been developed as cell imaging agents for a wide range of molecular targets.72–74
![]() | ||
Fig. 3 A simplified Jablonski diagram illustrating key processes in transition metal complexes, including transitions from the lowest triplet excited state (T1) to the singlet ground state (S0). |
Recently, metal complexes based on transition metals (e.g., manganese, ruthenium, rhenium, osmium, iridium, and platinum), lanthanides (e.g., europium), and main group or post-transition metals (e.g., zinc and tin), have been explored for applications in SRM (Fig. 4A). This is because the ability of metal complexes to undergo controlled photoactivation or photoconversion,75 makes them ideal for super resolution imaging. Moreover, these complexes typically exhibit strong phosphorescence in solution at room temperature, with emission lifetimes in the submicrosecond to microsecond range – significantly longer than the nanosecond-scale lifetimes of organic fluorophores. Their emission is also marked by large Stokes shifts, which minimize self-quenching and brightness reduction at high local concentrations – challenges commonly observed with organic fluorophores. This property, along with reduced spectral overlap, makes them particularly suitable for multiplex imaging. Additionally, many transition metal complexes demonstrate high photostability, allowing for continuous irradiation and enabling their use in long-term and real-time imaging applications within living systems. Their long-lived triplet excited states, particularly in phosphorescent Ru(II) and Ir(III) complexes, facilitate time-gated imaging, reducing background fluorescence and improving signal-to-noise ratios.72,73 For instance, Ru(II) polypyridyl complexes (RPCs) have long-lived excited states, typically lasting between 200 and 1000 ns.19 However, metal complexes generally exhibit lower quantum yields than organic probes, which can result in reduced brightness. Despite this limitation, advances in ligand design and a deeper understanding of their photophysical properties have enabled the development of metal complexes with competitive or even superior performance in specific imaging applications. Notably, the strong photostability, tuneable luminescence, extended excited-state lifetimes, and resistance to photobleaching of metal complexes allow for continuous-single molecule monitoring and real-time imaging of cellular structures and processes. These properties surpass the limitations of conventional fluorescent materials, establishing metal complexes as ideal photoluminescent probes for precise, long-term imaging in SRM. Fig. 4B compares images of cells stained with Ru(II), Ir(III), and Zn(II) complexes, captured using conventional CLSM and SRM. The figure demonstrates that metal complexes serve as useful fluorescent probes, and the acquisition of significantly higher-resolution images with SRM compared to CLSM.
![]() | ||
Fig. 4 (A) Periodic table highlighting the metals used to construct metal complexes discussed in this review for super-resolution imaging. (B) Comparative visualization of images of cells stained with metal complexes, captured using confocal laser scanning microscopy (CLSM) and super resolution microscopy. Cells were treated with (i) Ru(II) complexes (5 μM, 30 min), (ii) Ir(III) complexes (50 μM, 4 h) and (iii) Zn(II) complex (10 μM). Images (i) reproduced from ref. 76 with permission from the Royal Society of Chemistry. Images (ii) reprinted (adapted) with permission from ref. 16. Copyright 2017 American Chemical Society. Images (iii) reprinted (adapted) with permission from ref. 77. Copyright 2021 American Chemical Society. |
Although many reviews provide comprehensive perspectives on metal complexes as functional imaging agents,62–67 there is a lack of high-profile reviews specifically addressing their applications in super-resolution imaging. In this review, we discuss the applications of metal complexes in SRM for visualizing intricate subcellular structures, tracking temporal dynamics in live cells, exploring in vivo spatial organization, and detecting bacterial infections. We highlight the functional tunability of these complexes, achieved through rational design strategies that enhance biocompatibility, cellular uptake, specific organelle targeting and deep tissue penetration while preserving desirable photophysical properties. Additionally, we briefly discuss the design of various metal nanoparticles (NPs) for SRM, as well as emerging approaches for developing nanosized hybrid probes based on gold (Au) scaffolds and transition metal complexes. Although the application of these hybrid systems in SRM remains underexplored, it represents a promising strategy for translating the attractive luminescent properties of metal complexes into nanoscale platforms.
Interestingly, the cationic and lipophilic properties of metal complexes can facilitate efficient cellular uptake and preferential accumulation in hydrophobic cytoplasmic organelles, such as mitochondria, driven by the strong negative membrane potential of these organelles. Recent studies have explored Zn(II) complexes as promising probes for SRM in mitochondrial imaging, offering advantages such as earth abundance, structural rigidity, biocompatibility, low cytotoxicity, and tuneable brightness achieved through ligand design.74 In 2016, Tang et al. developed a Zn(II)-salen complex (1) where salen = 2,3-bis[(4-dialkylamino-2-hydroxybenzylidene)amino]but-2-enedinitrile, that operates via photooxidation of thioethers to sulfoxides (Fig. 5A).80 This complex demonstrates large two-photon absorption (2PA) cross sections (ca. 200 GM), a quantum yield of up to 0.8 (high brightness), and low cytotoxicity which is particularly suited for STORM technique. As a result, complex 1 achieved ∼12 nm localization precision for mitochondria in COS7 cells under STORM imaging, with fluorescence intensity retained at 10% after brief continuous scanning. Notably, its stable photophysical properties enabled efficient mitochondrial imaging without the need for redox chemicals or oxygen scavengers. This study marks a milestone as the first Zn(II) complex used in SRM, particularly using STORM. Terpyridine ligands, known for their strong chelating ability, provide robust coordination to metal ions, while their planar, aromatic structure promotes π–π stacking interactions that enhance cellular uptake and may facilitate mitochondrial localization.81,82 Moreover, in Zn(II) complexes, the d10 Zn(II) ion coordinates to the nitrogen atoms of the terpyridine core in a tridentate fashion, forming symmetrical structures that can exhibit nonlinear optical properties. In 2023, Su and co-workers developed two Zn(II) complexes (2 and 3) with modified terpyridyl ligands that vary in alkyl chain lengths.83 Live-cell studies confirmed the mitochondrial accumulation of complexes 2 and 3, and STED nanoscopy enabled detailed imaging of mitochondria, revealing the characteristic wrinkled structure of the inner membrane structure, known as cristae (Fig. 5A).83
![]() | ||
Fig. 5 Super-resolution imaging of subcellular structures, including mitochondria, lysosomes, and lipid droplets, was achieved using metal complexes in cultured cells. (A) (left) Structures of complexes 1–7. (right) (i) STED images of HeLa cervical cancer cells stained with complexes 2 or 3 (1 μM, 15 min), with magnified images showing the mitochondrial cristae. Reproduced from ref. 83 with permission from the Royal Society of Chemistry. (ii) SIM imaging of HeLa cells treated with complex 4 (50 μM, 4 h). Reprinted (adapted) with permission from ref. 16. Copyright 2017, American Chemical Society. (iii) and (iv) STED imaging of MCF7 breast cancer cells stained with complexes 6 or 7 (6 μM), showing mitochondrial structures. Reprinted (adapted) with permission from ref. 17 and 84, respectively. Copyright 2017 and 2020, respectively, American Chemical Society. (B) (left) Structure of complex 8. (right) STED imaging of HepG2 liver cancer cells stained with complex 8 (10 μM), showing lipid droplets (yellow) and mitochondrial structure (blue). Reproduced from ref. 20 with permission from the Royal Society of Chemistry. (C) (top) Structure of complex 9. (bottom) Long-term super-resolution radial fluctuation (SRRF) imaging of live HeLa cells showing lysosomal labelling with complex 9. Reproduced from ref. 21 with permission from the Royal Society of Chemistry. |
In addition to Zn(II) complexes, Ir(III) complexes, the third-row transition metal complexes, have proven highly suitable for SRM. In 2017, Shewring and co-workers developed an Ir(III)-based complex (4) featuring the [Ir(C^N)2(N^N)] motif, where the N^N unit is an anionic pyridyl-triazole chelate.16 Complex 4 exhibited stable Ir–N and Ir–C bonds, achieving photostable and low-cytotoxic mitochondrial labelling, marking a milestone as the first Ir(III) complexes used in SRM. SIM imaging demonstrated this complex provided high mitochondrial contrast at concentrations of 50 μM (Fig. 5A). Compared to conventional MitoTracker orange, complex 4 was found to exhibit a much greater photostability with a small intensity decay of 2.6%, whereas MitoTracker orange had decayed by 32.1%. Similarly, in 2018, Chen and co-workers developed Ir(III) complex ([Ir(mpq)2(mbbt)]+, 5), where mpq = 2-methyl-3-phenylquinoxaline and mbbt = 2-(4′-methyl-[2,2′-bipyridin]-4-yl)benzo[d]thiazole, demonstrating high specificity, photostability, and good cell permeability.85 Using SIM imaging, complex 5 achieved ∼80 nm resolution, allowing clear visualization of mitochondrial cristae, comparable to other studies.86–88
The Thomas group has developed several Ru(II)-based complexes for imaging cellular structures, including mitochondria. In 2017, they utilized a dinuclear, membrane-permeable Ru(II) complex (6), coordinated with the ditopic ligands tppz and tatpp (tppz = tetrapyrido[3,2-a:2′,3′-c:3′′,2′′-h:2′′,3′′-j]phenazine and tatpp = tetraazatetrapyrido[3,2-a:2′3′-c:3′′,2′′-l:2′′′,3′′′-n]pentacene), which demonstrated excellent photostability, a large Stokes shift, and intrinsic subcellular targeting, localizing to mitochondria in both live and fixed cells at low concentrations (Fig. 5A).17 By utilizing the highest magnifications available, complex 6 was shown to localize within the intermembrane space of the mitochondria. Ru(II)dppz complexes are well-known for their “light-switch” effect, characterized by weak luminescence in aqueous solution that becomes significantly enhanced upon binding to double-stranded DNA.89 This increase in emission is attributed to the shielding of the phenazine nitrogen atoms from interactions with water molecules, which suppresses nonradiative decay pathways. Subsequently, in 2020, the same group synthesized a dinuclear complex (7) composed of Ru(II)(dppz) and Re(I)(dppz) moieties (dppz = dipyrido[3,2-a:2′,3′-c]phenazine), linked by an N,N′-bis(4-pyridylmethyl)-1,6-hexanediamine tether.84 Complex 7 showed concentration-dependent mitochondrial localization and optimal photophysical properties, facilitating SIM and STED imaging of mitochondria (Fig. 5A). Interestingly, this study demonstrated that the cellular uptake and subcellular localization of these complexes are influenced by both the chemical composition and length of the linker.
Of interest, multi-colour SRM probes (e.g., Qdot 605 quantum dots),90 have further enhanced cellular imaging by enabling simultaneous visualization of multiple structures and processes in real time through imaging microscopy. These probes allow tracking the structural localization, dynamic interactions, spatial relationships, and functional changes within living cells, supporting high-content imaging essential for studying signal transduction, organelle interactions, and metabolic activities. Recently, Yuan et al. developed a rare-earth europium(III)-based complex, [Eu(TTA)3L3] (8), where TTA = 2-thenoyltrifluoroacetone and L3 = 1,10-phenanthroline-5,6-diamine, which functions by transitioning from a “turn-off” to a “turn-on” fluorescence state.20 Remarkably, complex 8 has proven effective in dual-targeting STED applications, particularly for studying interactions between lipid droplets and mitochondria (Fig. 5B), aligning well with the commercial probes Nile red (for lipid droplets) and MitoTracker deep red (for mitochondria). These interactions are particularly important for elucidating processes such as fatty acids oxidation, lipid storage, and cellular autophagy. While previous studies have elucidated the capabilities of rare earth metal-based complexes in disease diagnosis and treatment,91 this study represents a milestone as the first to employ a lanthanide-series rare-earth metal in SRM. This group is also the first to introduce a dual-targeting metal-based probe for SRM, paving the way for the future design and application of such innovative probes.
Lysosomal imaging at high resolution helps to understand processes like phagocytosis and autophagy, typically using conventional probes such as Lysotracker green and Lysotracker red, which selectively accumulate in acidic lysosomal compartments.92,93 However, concerns persist regarding their stability in response to changes in the lysosomal microenvironment, susceptibility to lysosomal membrane permeabilization (LMP), and photostability. In addition to these limitations, these probes can interfere with lysosomal function, cause leaching during live-cell imaging, and pose challenges for quantitative analysis, highlighting the need for the design of new chemical probes for lysosomal targeting. Compared to visible light, the near-infrared (NIR) region (700–1000 nm) is particularly advantageous for SRM as it reduces photodamage and phototoxicity, increases penetration depth, and minimizes cell autofluorescence.94,95 NIR light also facilitates long-term, real-time imaging by significantly reducing photobleaching and photodamage to biological samples. Recently, a morpholine-conjugated Zn(II)-salen complex (9) with emission in the NIR region has been developed for lysosomal imaging, showing high photostability and versatility in SRRF nanoscopy.21 Strikingly, this complex provides detailed mapping of lysosomal distribution and motility in live cells (Fig. 5C). The high-brightness, NIR-emissive and non-toxic properties of complex 9 minimize cell damage whilst providing a superior signal-to-noise ratio.
The exceptional photostability, high DNA selectivity and chirality of RPCs, which influences luminescent intensity and cellular phototoxicity, are advantageous for both high-resolution imaging and therapeutic applications. In particular, RPCs are valuable in cancer photodynamic therapy (PDT) due to their ability to release biologically active molecules upon photoinduced irradiation. A notable example is the RPC TLD1433, which is used for treating bladder cancer patients.96 In 2016, Byrne and co-workers developed several luminescent, organelle-targeted RPCs peptide conjugates, including [Ru(bpy)2-phen-Ar-ER]9+ (10), [Ru(bpy)2-phen-Ar-Arg8]10+ (11), and [Ru(dppz)(bpy)-bpy-Ar-NLS]6+ (12), where bpy = 2,2-bipyridine and phen = 1,10-phenanthroline (Fig. 6A).19 These complexes were designed to selectively target various organelles via signal peptides, bypassing the need for membrane permeabilization. Their large Stokes shift aligns with the depletion laser wavelengths, reducing anti-Stokes excitation, while their extended fluorescence lifetime widens the gating window, making them ideal for gated STED imaging. As such, RPC 10 reveals the tubular ER architecture with high resolution using STED nanoscopy (Fig. 6A), while both RPCs 10 and 11 facilitated high-quality STED imaging of actin filaments (Fig. 6C).19 Among these complexes, RPC 10, being the most lipophilic complex, exhibits preferential enrichment in the membrane-dense structures of the ER, demonstrating its promising application for studying the role of ER in maintaining cellular physiological functions. Compared with the widely used AlexaFluor 532 dyes, these RPCs demonstrated enhanced photostability with two-fold improvement in image resolution in both continuous-wave (CW) and gated STED. Notably, these RPCs were among the first metal-based luminophores used in STED nanoscopy, setting a precedent for using metal complexes in SRM.
![]() | ||
Fig. 6 Super-resolution imaging of subcellular structures, including the endoplasmic reticulum (ER), ribosomes, and cytoskeletal components, using metal complexes in cultured cells. (A) (left) Structure of complex 10. (right) STED images of HeLa cells stained with complex 10 (70 μM, 4 h), showing the ER. Reproduced from ref. 19 with permission from the Royal Society of Chemistry. (B) (left) Structure of complex 13. (right) Top: Two-colour STED imaging of HepG2 cells co-stained with complex 13 (red, 5 μM, ribosomes) and ER-tracker (green, ER), showing the localization of ribosomes relative to the ER. Bottom: STED imaging of HepG2 cells stained with complex 13 (5 μM), revealing ribosomal structures. Reprinted (adapted) with permission from ref. 18. Copyright 2024 American Chemical Society. (C) (top) Structures of complexes 11, 14 and 15. (bottom) (a) STED and time-gated STED (15 ns) imaging of fixed HepG2 cells stained with complex 14 (5 μM), showing microtubule structures. Adapted from ref. 22 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Advanced Materials published by John Wiley & Sons Ltd. © The Author 2020. (b) (i) STED imaging of live human foetal lung HFL1 cells stained with complex 15 (5 μM, 60 min), showing actin filaments. Reprinted from Sensors and Actuators: B. Chemical, 388, Liu, T. et al., An Iridium(III) complex revealing cytoskeleton nanostructures under super-resolution nanoscopy and liquid-phase electron microscopy, 133839, Copyright 2023, with permission from Elsevier. (ii) STED, continuous-wave (CW)-STED, and gated CW-STED (gCW-STED) imaging of HeLa cells stained with complexes 10 (70 μM) and 11 (70 μM, 4 h), showing actin filaments. Reproduced from ref. 19 with permission from the Royal Society of Chemistry. |
Detailed visualization of ribosome distribution on the rough ER provides insights into protein synthesis dynamics. Current ribosome probes for SRM include fluorescent dyes, such as cyanine 3 (Cy3) or Alexa Fluor, and genetically encoded tags like green fluorescent protein (GFP), along with ribosomal ribonucleic acid (RNA) (rRNA)-targeting probes.97 However, genetically encoded tags may interfere with ribosome function or expression restricting their use in SRM. Recently, Liu et al. designed an Ir(III) complex (13) with azide- and benzoic acid-modified terpyridine ligands for the selective labelling of ribosomes in live cells (Fig. 6B).18 The azide ligand, functionalized with a positively charged quaternary ammonium salt at its end, enhances aqueous solubility. Similarly, the introduction of a carboxylic acid group further improved the solubility of the complexes and likely enhanced biocompatibility by reducing cytotoxicity. With a robust Ir(III) metal core and high photostability, complex 13 enables super-resolution STED imaging of ribosomes on the rough ER with ∼40 nm resolution, sufficient to resolve structures at the scale of eukaryotic ribosomes (25–30 nm in diameter).
To visualize cytoskeletal components, the Tian group developed another Ir(III) complex (14) coordinated with vanillin (4-hydroxy-3-methoxybenzaldehyde) and phenanthroline ligands, where phenanthroline imparts yellow emission.22 Vanillin reacts with o-aminothiophenol to form thiazole derivatives, creating a conjugated system with push–pull electronic effects that enhance fluorescence. These ligands facilitate hydrogen bonding and hydrophobic interactions (e.g., π–π stacking with amino acids), which stabilize fluorescence by reducing torsional movement in the extended conjugated system, resulting in a “light-switch” effect. STED imaging with complex 14 under optimized conditions of maximum STED laser power and extended detection gating time allows for super-resolution of microtubules in live cells, achieving ∼30 nm of lateral resolution, attributed to its long-lived and photostable fluorescence properties (Fig. 6C). This resolution, measured as 29 ± 11 nm, aligns with the expected optical resolution in STED nanoscopy, considering the actual microtubule diameter of ∼20 nm.98 Subsequently, in 2023, Liu and her co-workers further advanced this area with an Ir(III) complex [Ir(dmobtz)2(H2O)2]+ where dmobtz = 4,6-dimethyl-2-(benzo[d]thiazol-2-yl)phenol (15), that selectively binds actin, exhibiting a “light-switch” fluorescence response upon actin binding.99 This feature enables high-resolution STED imaging of actin in live cells while preserving actin structure and cellular integrity (Fig. 6C). This is a remarkable feat, considering that common probes for cytoskeletal components, such as phalloidin conjugates for actin, tubulin-specific antibodies for microtubules, and vimentin or keratin antibodies for intermediate filaments, may interfere with cytoskeletal function and cause artifacts in live-cell imaging due to probe overexpression or leaching.100
![]() | ||
Fig. 7 Super-resolution imaging of genetic structures and materials using metal complexes in cultured cells. (A) (top) Structure of complex 12. (bottom) (i) gCW-STED image of HeLa cells stained with complex 12 (40 μM, 24 h) showing chromosomal DNA. Reproduced from ref. 19 with permission from the Royal Society of Chemistry. (ii) dSTED image of MCF7 cells treated with complex 16 (150 μM) captures a single cell entering mitosis. Reprinted (adapted) with permission from ref. 102. Copyright 2021 American Chemical Society. (iii) STED image of HeLa cells stained with complex 17 (100 μM, 30 min) shows chromosomal DNA during metaphase. Adapted from ref. 103 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Nucleic Acids Research published by Oxford University Press. © The Author 2023. (B) (top) Structure of complex 18. (bottom) STED imaging of HepG2 cells treated with complex 18 (10 μM) reveals histones in the nucleus. Reprinted (adapted) with permission from ref. 77. Copyright 2021 American Chemical Society. (C) (top) Structures of complexes 16, 17, 19, 20 and 21. (bottom) (i) SIM and STED images of MCF7 cells stained with complex 6 (500 μM) reveal nuclear DNA. Reprinted (adapted) with permission from ref. 17. Copyright 2017 American Chemical Society. (ii) STED image of HeLa cells stained with complex 12 (40 μM, 24 h) shows nuclear DNA. Reproduced from ref. 19 with permission from the Royal Society of Chemistry. (iii) dSTED image of MCF7 cells treated with complexes 16 (150 μM) and 19 shows nuclear DNA. Reprinted (adapted) with permission from ref. 102. Copyright 2021 American Chemical Society. (iv) STED image of HeLa cells stained with complex 17 (100 μM, 30 min) visualizes nuclear DNA during the cell cycle. Adapted from ref. 104 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Nucleic Acids Research published by Oxford University Press. © The Author 2023. (v) STED images of HepG2 cells stained with complexes 20 and 21 (5 μM, 30 min), showing nuclear DNA. Reproduced from ref. 76 with permission from the Royal Society of Chemistry. |
Ruthenium complexes are designed to target DNA, binding to specific sites and disrupting cellular processes, making them valuable for applications in cancer therapy and molecular imaging. Previously, the Thomas group showed that complex 6 could target mitochondria.17 Interestingly, further detailed studies revealed its effective nuclear DNA imaging capabilities in live cells at concentrations of ∼20 mM (Fig. 7C).17 Thus, this complex is suitable for dual-colour imaging of mitochondria and nuclear DNA in two-colour SIM (2C-SIM), STED, and 3D-STED in both fixed and live cells. Strikingly, the emission from the bound complex 6 is long-lived (180 ns), resulting in high photostability that surpasses those of conventional super-resolution probes. Subsequently, in 2021, they developed heteroleptic metal complexes containing tpphz and TAP ligands, where tpphz = tetrapyrido[3,2-a:2′,3′-c:3′′,2′′-h:2′′′,3′′′-j]phenazine and TAP =1,4,5,8-tetraazaphenanthrene.102 For instance, the dinuclear Os(II) complex [(Os(TAP)2)2(tpphz)]4+ (16), which emits in the NIR region (λmax = 780 nm) from an Os → TAP 3MLCT excited state. This complex has demonstrated photostability, cellular permeability, and low toxicity, making it suitable for NIR STED nanoscopy of nuclear DNA (Fig. 7C).102 This study also reported that the dinuclear Ru(II) complex [(Ru(TAP)2)2(tpphz)]4+ (19) binds with high affinity to duplex and quadruplex DNA, allowing high-resolution imaging of nuclei in fixed cells (Fig. 7C).102
Recently, the Gill group synthesized two RPCs [Ru(bpy)2(10,11-dmdppz)]2+ (10,11-dmdppz = 10,11-dimethyl-dipyridophenazine) and [Ru(5,5′-dmbpy)2(10,11-dmdppz)]2+ (5,5′-dmbpy = 5,5′-dimethyl-bpy) (complexes 20 and 21, respectively), which bind duplex DNA with high affinity (DNA binding constants, Kb up to 5.7 × 107 M−1).76 Utilizing STED nanoscopy, these RPCs facilitated super-resolution imaging of the nuclear DNA and the plasma membrane, attributable to their robust MLCT luminescence (Fig. 7C). Although DNA-targeting probes are valuable tools for imaging, however, binding to DNA can disrupt replication, transcription and repair, potentially causing cytotoxicity or undesirable cellular responses. It is known that mode of DNA binding significantly influences the extent of interference. For instance, groove binders are generally less disruptive to cellular processes, while intercalators can deform the DNA helix, hindering interactions with essential proteins.105 As such, to maintain cell viability and ensure reliable experimental results, probes that act via intercalation should be used at concentrations that minimize such disruptions.
Visualizing RNA enables real-time tracking of gene expression, RNA localization, investigation of RNA dynamics within cells, and examination of disease-associated abnormalities, particularly in cancer and viral infections.106 The nucleolus, a nuclear substructure responsible for rRNA processing during interphase, plays an important role in cytosolic protein synthesis. Despite its importance, the nucleolar ultrastructure remains underexplored, and there is limited insight into RNA-ribosome interactions within the nucleolus. Moreover, traditional fluorescence microscopy often fails to capture nanoscale RNA structures due to limited probe permeability and interference from DNA.
Addressing these limitations, Liu et al. developed a terpyridyl Mn(II) complex (22) that was able to selectively label nucleolar RNA in live cell nuclei (Fig. 8A), marking a milestone as the first Mn(II) complex used in SRM.107 Like complex 13, complex 22 contains azide ligands, each with a positively charged quaternary ammonium salt at its end. This design enhances cellular uptake, likely through electrostatic interactions with negatively charged membranes, and facilitates nuclear accumulation, possibly through passive diffusion. As a result, complex 22 was shown to localize in RNA-rich regions. Moreover, Mn(II), regarded as a biosafe metal, contributes superior photostability to the probes, enabling them to withstand prolonged laser exposure under STED nanoscopy. Using STED imaging, complex 22 enables super-resolution imaging of RNA-rich nucleolar components, including dense fibrillar components (DFC) and fibrillar centres (FC) (Fig. 8A).
![]() | ||
Fig. 8 Super-resolution imaging of subcellular structures and genetic components using metal complexes in cultured cells. (A) (left) Structures of complexes 22 and 23. (right) (i) STED image of HeLa cells treated with complex 22 (10 μM, 30 min), showing the nucleolus, and 3D-STED image revealing the ultra-detailed structure of a single nucleolus, highlighting distinct regions within the nucleolus. Reprinted from Biosensors and Bioelectronics, 203, Liu, J. et al., Nucleolar RNA in action: Ultrastructure revealed during protein translation through a terpyridyl manganese(II) complex, 114058, Copyright 2022, with permission from Elsevier. (ii) STED images of HepG2 cells stained with complex 23, showing the nucleus, with magnified images focusing on the nucleoplasmic region and nucleolus. Reproduced from ref. 108 with permission from the Royal Society of Chemistry. (B) (top) Structure of complex 24. (bottom) STED images of HepG2 cells stained with complex 24 (10 μM, 10 min), showing the nucleus and nuclear histidine. Reproduced from ref. 109 with permission from the Royal Society of Chemistry. (C) (top-left) Structure of complex 25. (i) SIM images of HeLa cells stained with complex 17 (100 μM, 30 min, mitochondrial DNA, red), Tomo20-dronpa (mitochondrial membrane, green), and MitoTracker Green (mitochondria, green). Adapted from ref. 104 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Nucleic Acids Research published by Oxford University Press. © The Author 2023. (ii) STED images of HepG2 cells stained with complex 25, showing mitochondrial structure, with a magnified image displaying ultra-detailed visualization of a single mitochondrion. Reproduced from ref. 103 with permission from the Royal Society of Chemistry. |
Subsequently, Feng and co-workers advanced the field of nucleolar RNA imaging by developing a thiophene-based terpyridine Zn(II) complex (23) with a donor–π–acceptor structure, which exhibits aggregation-induced emission (AIE) properties,108 where the complex becomes highly fluorescent upon aggregation due to the restriction of intramolecular motions. Complex 23 displayed a large Stokes shift, high photostability, and three-photon absorption (3PA) activity, with fluorescence enhanced by hydrophobic and π–π interactions. Similar to complex 22, the quaternary ammonium group in complex 23 facilitates electrostatic interactions with the negatively charged RNA backbone, thereby enhancing cell membrane permeability. The octahedral geometry of complex 22 further contributes to its ability to interact specifically with nucleolar RNA, enabling precise imaging using STED nanoscopy, through a combination of structural complementarity and photophysical properties (Fig. 8A). In another study, Zhang et al. developed an Ir(III) complex (24) with a benzoic acid-modified terpyridine ligand, characterized by high 2PA efficiency, which selectively targets nuclear histidine.109 The carboxylic acid group in complex 24 increases electron delocalization, solubility, biocompatibility, and nuclear membrane permeability. Also, the planarity of the phenanthroline group enables histidine targeting, while the protonatable pyridine group improves water solubility and forms stabilizing hydrogen bonds with histidine. Using STED nanoscopy, complex 24 facilitated high-resolution nuclear histidine imaging with an improved signal-to-noise ratio in comparison to confocal microscopy (Fig. 8B). While histidine is not genetic material, imaging nuclear histidine is key to studying the function and regulation of nuclear proteins involved in key cellular processes.
Mitochondrial DNA (mtDNA), a double-stranded circular molecule encoding 37 essential genes that resides in the mitochondrial matrix and along the cristae surfaces, plays a key role in cellular energy conversion and metabolic regulation.110 Alterations in mtDNA function are linked to key physiological changes, making mtDNA imaging useful as a disease biomarker, though live-cell mtDNA-targeting probes remain limited. In 2018, Shen and co-workers addressed this gap by developing another Zn(II) complex (25) with thiophene-based terpyridine ligands, which showed high biocompatibility and photostability.103 The thiophene unit served as a π-conjugated bridge, enhancing electron transport, luminescence, and mitochondrial targeting. STED nanoscopy with complex 25 enabled visualization of folded cristae, capturing unique cristae details and revealing mtDNA distribution at a spatial resolution of 40–70 nm (Fig. 8C). Recently, Huang et al. showed that the DNA “light-switch” RPC of [Ru(phen)2dppz]Cl2 (17), exhibits exceptional SIM and STED imaging of nuclear DNA in live cells when paired with chlorophenolic counter-anions to form lipophilic ion-pairs, which facilitate nuclear permeability (Fig. 7C).104 Interestingly, after initially localizing to the nucleus, complex 17 was observed to migrate to the central lumen of the mitochondria, where mtDNA is located, following medium refreshment. There, it selectively bound to mtDNA, thereby demonstrating dual targeting of both nuclear and mitochondrial DNA. As shown in Fig. 8C, high-resolution mtDNA imaging in live cells stained with complex 17 was observed using SIM nanoscopy (Fig. 8C). This study revealed detailed mtDNA distribution, showing that each mitochondrion contains 1–8 mtDNA molecules, with nucleoid clusters often located near the nucleus. Notably, this study is the first to employ a metal complex as an mtDNA probe to achieve such high-resolution mtDNA imaging with STED nanoscopy. Although chlorophenols are not explicitly restricted for use in therapeutic agents, their toxicity, environmental persistence, and regulatory scrutiny render them unfavourable, emphasizing the need for safer, more biocompatible alternatives;111 consequently, this finding should be interpreted with caution.
Due to the therapeutic potential of organotin(IV) complexes as antibiotics, in 2018, Hu et al. synthesized a novel organotin(IV) coumarin-based complex, 26, which showed 2PA cross-sections in the NIR region.114 The cyano group, with its strong electron-withdrawing ability, was incorporated into the ligand to facilitate coordination with tin(IV), thereby forming highly stable organotin complexes. The cyano group also modulates the push–pull electronic structure of the entire molecule, enabling fine-tuning of its photophysical properties. Super-resolution imaging of Gram-negative bacteria Escherichia coli (E. coli) using STED nanoscopy demonstrated that complex 26 was strongly adsorbed onto the cell walls and even penetrated into the cells (Fig. 9A). Moreover, complex 26 was shown to interact with bacterial membranes, inducing the production of reactive oxygen species (ROS), which ultimately leads to cell death.
![]() | ||
Fig. 9 Super-resolution imaging of bacterial cell components using metal complexes. (A) (left) Structures of complexes 26–28. (right) Top: SIM images of Escherichia coli (E. coli) EC958 stained with 1.6 μM complex 27 for 5 min (i), and 3.6 μM complex 28 (ii) for 60 min. Bottom: STED images of E. coli EC958 stained with 50 μg mL−1 complex 26 (iii) and 1.6 μM complex 27 (i) for 10 min. Images (i) reprinted (adapted) with permission from ref. 115. Copyright 2019 American Chemical Society. Image (ii) reproduced from ref. 116 with permission from the Royal Society of Chemistry. Image (iii) reprinted (adapted) with permission from ref. 114. Copyright 2018, American Chemical Society. (B) (top-left) Structures of complexes 29 and 30. (right) Top: SIM images of Staphylococcus aureus (S. aureus) SH1000 cells stained with 4 μM complex 27 (i) for 5 min and 3.6 μM complex 28 (ii) for 60 min. Bottom: STED images of S. aureus SH1000 cells stained with 4 μM complex 27 (i) for 60 min and 3.6 μM complex 28 (ii) for 120 min. Images (i) reproduced from ref. 117 with permission from the Royal Society of Chemistry. Images (ii) reproduced from ref. 116 with permission from the Royal Society of Chemistry. (iii) SIM images of methicillin-resistant S. aureus (MRSA) stained with complexes 29 and 30 for 60 min. Reproduced from ref. 118 with permission from the Royal Society of Chemistry. (C) (left) Structure of complex 31. (right) SIM images of Bacillus cereus (B. cereus) stained with complex 31 (10 μM, green) and MitoTracker red (3 μM, red) for 15 min. Reproduced from ref. 119 with permission from the Royal Society of Chemistry. |
Subsequently in 2019, Smitten et al. developed a luminescent dinuclear Ru(II) complex [(Ru(3,4,7,8-tetramethyl-1,10-phenanthroline)2)2(tpphz)][PF6]4 (27) which displayed potent antimicrobial activity against multidrug-resistant Gram-negative E. coli strain EC958 ST131.115 STED nanoscopy revealed its initial localization in cellular membranes, with subsequent movement toward cell poles (Fig. 9A). Complex 27 also showed antimicrobial activity in Gram-positive bacteria Staphylococcus aureus (S. aureus) SH1000, including methicillin-resistant S. aureus (MRSA) strains, with STED nanoscopy confirming its presence within or on the membrane rather than binding to peptidoglycan within the cell wall (Fig. 9B).117 At the improved resolutions provided by 3D-STED, complex 27 showed distinctive patterning across the S. aureus membrane, with the bacterial membrane and DNA identified as therapeutic targets. In 2020, they developed another luminescent, mononuclear Ru(II) complex (28) containing tpphz and dppz ligands which exhibited activity in both Gram-negative bacteria and Gram-positive bacteria (Fig. 9A and B).116 Super-resolution SIM nanoscopy showed complex 28 targets bacterial DNA, which was confirmed by morphological changes in treated E. coli, such as increased cell length and multinucleation (Fig. 9A). This is similar to effects seen with DNA replication-disrupting antibiotics such as ampicillin. Apart from Ru(II) complexes, this group also demonstrated that Os(II) complexes [Os(pytz)3](PF6)2 (where pytz = pyrazine-2-carbaldehyde thiosemicarbazone), extracted as the chloride salts of their meridional and facial isomers (termed 29 and 30, respectively), displayed antimicrobial activity against MRSA strains.118 Subsequent super-resolution imaging experiments demonstrate high colocalization of the two enantiomers with bacterial DNA (Fig. 9B), marking a milestone as the first Os(II) complexes used in SRM. Overall, these findings illustrate that the lipophilicity of these complexes influences their cellular uptake and localization, which can vary from the nucleus to the cell membrane.
In 2021, Ranieri and co-workers designed a metal-naphthalimide scaffold (31), which incorporates a platinum(II) metal centre and a luminescent 4-amino-1,8-naphthalimide group via a pyridine ancillary ligand.119 This complex exhibited stability and suitability for bacterial imaging, as confirmed by SIM and nanoscale secondary ion mass spectrometry (nanoSIMS). These techniques demonstrated the internal uptake of the complex in live Bacillus cereus (B. cereus), with nanoSIMS providing high-resolution chemical mapping of the metal complex within the bacteria. Of interest, SIM imaging revealed that complex 31 showed sub-cellular localization within well-defined inclusions resembling lipid bodies (Fig. 9C), where the lipidic nature of these organelles was further confirmed by staining with the lipophilic dye BODIPY 493/503. Overall, these studies demonstrate the dual therapeutic-imaging applications of metal complexes for treating and detecting bacterial infections. Of note, their development as imaging agents may offer a quicker route to clinical application compared to therapeutic development, owing to differing regulatory requirements and the minimal amounts of metal complex needed for imaging.
![]() | ||
Fig. 10 Super-resolution imaging of spatial organization in vivo using metal complexes. (A) SRRF image of lysosomes throughout the entire Caenorhabditis elegans (C. elegans) labelled with complex 9. Reproduced from ref. 21 with permission from the Royal Society of Chemistry. (B) Top: STED images of brain hippocampus sections stained with complex 14. Bottom: 3D rendering of STED images and magnified regions showing neuronal subunits. Adapted from ref. 22 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Advanced Materials published by John Wiley & Sons Ltd. © The Author 2020. (C) STED image of mouse heart tissue stained with complex 15, showing fibril structure with magnified images revealing vertical and parallel actin filaments. Reprinted from Sensors and Actuators: B. Chemical, 388, Liu, T. et al., An Iridium(III) complex revealing cytoskeleton nanostructures under super-resolution nanoscopy and liquid-phase electron microscopy, 133839, Copyright 2023, with permission from Elsevier. (D) (top) Structure of complex 32. (bottom) STED image of myelin sheaths along neuronal axons in the mouse brain labelled with complex 32, with the magnified image showing a single myelin wrapping an axon fibre. Reproduced from ref. 120 with permission from the Royal Society of Chemistry. |
In 2019, the Tian group designed another Mn(II) complex ([Mn((4-((4-([2,2′:6′,2′′-terpyridin]4′yl)phenyl)(phenyl)amino)-phenyl)methanol)2]2+, 32) with versatile imaging capabilities.120 The triphenylamine group, known for its electron-rich nature and excellent photoelectric properties, is complemented by a hydroxymethyl group that enhances solubility and increases electron density. Mn(II), with its d5 electron configuration, forms stable octahedral core by coordinating with the nitrogen atoms of 2,2′:6′,2′′-terpyridine (tpy) derivatives, which are essential for assembling supramolecular architectures with enhanced 2PA responses. Remarkably, complex 32 exhibited deep tissue penetration in thick mouse brain samples enabling STED imaging and enhancing the magnetic resonance signal. STED micrographs revealed discontinuous myelin ultrastructure along neuronal axons, providing detailed insight into neuronal organization and possible regions of demyelination (Fig. 10D). Of interest, the dual functionality of this complex as a fluorescence/MRI multimodal probe demonstrates its potential application in clinical diagnostics. The significance of such probes was further demonstrated by Yan et al., who developed an activatable small-molecule NIR fluorescence/MRI bimodal probe triggered by endogenous alkaline phosphatase (ALP).121 This probe enables real-time, high-sensitivity imaging and precise localization of ALP activity in live tumour cells and mice, with promising applications for image-guided surgical resection of tumour tissues.
![]() | ||
Fig. 11 Super-resolution imaging of temporal dynamics using metal complexes. (A) STED imaging of ultrastructural changes in living cells after incubation with complex 2 (left) or 3 (right), followed by laser irradiation to induce mitochondrial reactive oxygen species (ROS). Reproduced from ref. 83 with permission from the Royal Society of Chemistry. (B) Representative trajectories of individual lysosome movement labelled with complex 9, with the starting point marked by a white dot and the end position indicated by a white arrowhead for peripheral and perinuclear regions, respectively. Reproduced from ref. 21 with permission from the Royal Society of Chemistry. (C) Dynamics of lysosomes and lysosome-related organelles (LROs) from the single-cell stage to mature C. elegans, labelled with complex 9. Reproduced from ref. 21 with permission from the Royal Society of Chemistry. (D) Left: 3D-STED live-cell imaging of the cytosolic region, where ribosomes are labelled with complex 13. Right: Time-lapse 3D-STED imaging showing complex 13-stained cells under insulin treatment. Reprinted (adapted) with permission from ref. 18. Copyright 2024, American Chemical Society. (E) Cells treated with histone deacetylase activator (ITSA 1) to induce histone acetylation, incubated with complex 18, and imaged using STED to construct 3D micrographs. The selected region of interest (ROI) highlights histone condensation, aggregates, and cavity regions. ITSA 1-treated cells stained with complex 18 show micronucleus formation and nuclear misshaping. Reprinted (adapted) with permission from ref. 77. Copyright 2021, respectively, American Chemical Society. (F) STED images of HeLa cells incubated with complex 22 (10 μM, 30 min, red) and a ribosome marker (blue), revealing a single nucleolus with ribosomes under various physiological conditions: 6 h of insulin treatment, 6 h of starvation, and normal culture conditions. Reprinted from Biosensors and Bioelectronics, 203, Liu, J. et al., Nucleolar RNA in action: Ultrastructure revealed during protein translation through a terpyridyl manganese(II) complex, 114058, Copyright 2022, with permission from Elsevier. |
The Salam group further studied the in vivo dynamics of lysosomes using metal complexes in super-resolution imaging.21 Tracking lysosomal dynamics is essential as lysosomes regulate cellular homeostasis, with their motility and localization are associated with key processes such as phagocytosis and autophagy.93 Complex 9, for example, tracks lysosomal movement and speed in both cancerous and normal cells (Fig. 11B), while also enabling real-time visualization of lysosomal processes in C. elegans, including fission, fusion, and “kiss-and-run” events (transient interactions where lysosomes briefly touch and exchange material before separating) (Fig. 11C), using SRM imaging.21 In another study, complex 13 facilitated real-time imaging of rRNA dynamics in live cells (Fig. 11D).18 Using STED, this study revealed rRNA distribution and aggregation, showing higher ribosome density near the cell nucleus and membrane compared to the central cytosol. This distribution is likely linked to localized protein synthesis, as the rough ER is near the nucleus and plasma membrane.
Strikingly, complex 18 revealed nanoscale histone acetylation dynamics, specifically showing three morphological variations – condensation, aggregation, and cavity formation – linked to histone function during cell division (Fig. 11E).77 Further STED observations showed that these variations were accompanied by the apparent misshaping of micronuclei and nuclei. Given that abnormalities in histone function are often associated with tumorigenesis and other diseases,122 super-resolution imaging of histone biology during cell division holds promise for clinical diagnosis.
Interestingly, complex 22 was shown to elucidate the dynamics of nucleolar RNA and its relationship with ribosomes during protein synthesis (Fig. 11F).107 This study is the first to observe the distribution of ribosomes within individual nucleoli during the stimulation of protein translation and autophagic cell death. In 2020, Liu et al. developed a dinuclear platinum complex [(Pt(dien))2L](NO3)5, 33, where dien = diethylenetriamine, with a tripodal luminescent ligand L = 2-(4-(bis(4-(pyridin-4-yl)phenyl)ami-no)styryl)-1-methylquinolin-1-ium iodide (Fig. 12).123 Complex 33, characterized by its large Stokes shift and high resistance to photobleaching, serves as an ideal SRM probe with minimal background interference. Under SRM imaging and upon light stimulation, complex 33 exhibits a photoactivated escape mechanism, transitioning from autolysosomes to the nucleus unveiling a novel transport pathway that could enhance platinum drug targeting. This photo-selective control over nuclear access offers a promising strategy to improve the specificity and efficacy of platinum-based drugs in live-cell applications.
Furthermore, the synthesis of metal complexes often involves intricate procedures requiring precise coordination and functional group incorporation for targeting. However, preparation of metal complexes generally requires fewer synthetic steps than organic fluorescent dyes and also avoids the genetic encoding required for fluorescent proteins. As ligand selection forms a key aspect of molecular design, synthesis can be modular, where quantitative synthetic methods facilitate high-throughput workflows. For example, Kench et al. described a high-throughput approach for the synthesis and identification of iridium complexes.127 Additionally, Orsi and co-workers employed machine learning to predict the photophysical and biological properties of metal complexes, further accelerating the discovery and optimization of new metal complexes.128
Another significant limitation is the lack of biological specificity. Metal complexes often exhibit nonspecific binding to cellular components increasing background noise and reducing imaging specificity. Conjugation with biomolecules (e.g., antibodies, peptides, aptamers) and surface coatings like polyethylene glycol (PEG) or biomimetic shells can enhance specificity and reduce off-target effects.129 While this has not yet been widely demonstrated in SRM, it remains an area of significant interest. Previous studies have also highlighted the inefficiency of cellular uptake in some metal complexes.130 Recent advances discussed in this review demonstrate that ligand modifications can enhance cellular permeability while maintaining low cytotoxicity. Likewise, although the stability of metal complexes in biological environments can be compromised by degradation, interactions with cellular components, or chelation by endogenous molecules,65 rational ligand design can prevent such interactions and improve stability.
Moreover, another notable challenge is the complexity of correlating structural design with biological targets, as the targeting mechanisms of some complexes can exhibit variability and unpredictability.65 Addressing this issue requires enhanced molecular target validation, including advanced screening techniques and mechanistic studies, to improve the accuracy of subcellular localization and the monitoring of specific biological processes. Furthermore, imaging subcellular structures is challenged by organelle density, dynamics, and chemical environments. Dense organelles (e.g., nucleus and mitochondria) can quench fluorescence or restrict metal complex diffusion, while rapidly moving structures, such as vesicles, require complexes with fast binding kinetics and high photostability.131 Organelles with acidic environments, such as lysosomes, also necessitate optimization of metal complex properties for intracellular conditions.131 In addition, due to potential toxicity, ROS generation, and immune responses, the biocompatibility of metal complexes in in vivo imaging remains a major concern.
![]() | ||
Fig. 13 (A) Strategies to optimize the optical characteristics of metal nanoparticles (NPs) by altering their composition or core–shell architecture, physical properties (size and shape), surface chemistry and targeting ligands. (B) Structure of gold (Au) nanocrystals-streptavidin with the hydrogen (H)-bond network on the surface. Reprinted (adapted) with permission from ref. 136. Copyright 2022, American Chemical Society. (C) The design of AuNPs (GNP-PNA-A647). AuNP is covalently linked to a peptide nucleic acid (PNA) probe through a polyethylene glycol (PEG)-linker molecule. The peptide nucleic acids (PNA) probe has an Alexa-647 fluorescent molecule at the N-terminus. Reprinted (adapted) with permission from ref. 137. Copyright 2017, American Chemical Society. (D) Schematic composition of Cy5@AuNPs. Adapted from ref. 138 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Theranostics published by Ivyspring International Publisher. © The Author 2020. (E) Schematic illustration of the preparation of Halo-Tag AuNPs (GNP-Atto565-fR8-CA) Reprinted (adapted) with permission from ref. 139. Copyright 2024, American Chemical Society. |
Moreover, modifications to the properties of NPs can influence their biophysical and chemical properties, including cellular uptake, biodistribution, stability, and functional activity. Unlike small molecules, metal NPs exhibit size- and shape-dependent behaviours, enabling distinct interactions with biological systems, such as enhanced permeability, retention effects, and selective targeting. These properties make metal NPs particularly suitable for advanced imaging techniques, including SRM, fluorescence-guided surgery,140 and surface-enhanced Raman spectroscopy (SERS).141 Their optical versatility also supports multiple imaging modalities, such as fluorescence, dark-field, and EM, further expanding their applications in SRM. Interestingly, Syed et al. demonstrated that AuNPs and AgNPs can provide 3D imaging contrast in intact and transparent tissues using light-sheet fluorescence microscopy (FLSM).142 This study successfully demonstrated molecular imaging of blood vessels, the mapping of lesions and immune cells in a mouse model of multiple sclerosis and nanodrug carrier tracking within tumours, highlighting the applicability of these probes for SRM applications.
Furthermore, the high surface area-to-volume ratio of metal NPs allows functionalization with diverse ligands, drugs, or imaging agents, enabling precise targeting of biomolecules or organelles to improve imaging specificity and sensitivity. For example, although Au nanoclusters are valued for their biocompatibility and strong luminescent properties for SRM, they often require surface modifications to reduce nonspecific binding and cytotoxicity. This need is exemplified by two separate studies that visualized cellular tubulin and lysosomes in live cells.136,143 Yang et al. demonstrated that the abundant –NH2 and –COOH groups on the surface of Au nanoclusters act as functional sites for biomolecule conjugation (Fig. 13B), facilitating targeted imaging of cellular components such as tubulin, lysosomes and amyloid-β (Aβ) aggregates using STED imaging (Fig. 14A).136 Similarly, Yadav et al. reported that conjugating bovine serum albumin (BSA) to Au nanoclusters is highly effective for SRRF imaging of lysosomes, enabling the visualization of lysosomal with a measured diameter of ∼59 nm, which corresponds to the smallest lysosomes observed in HeLa cells (Fig. 14B).143
![]() | ||
Fig. 14 Metal NPs for super-resolution imaging of subcellular structures. (A) STED imaging of tubulin (top-left), lysosomes (top-right) and Aβ-aggregates (bottom) labelled with Au nanocrystals-streptavidin. Reprinted (adapted) with permission from ref. 136. Copyright 2022, American Chemical Society. (B) SRRF images of Au nanoclusters-labelled lysosomes in HeLa cells. Reprinted (adapted) with permission from ref. 143. Copyright 2020, American Chemical Society. (C) (a) Image of chromosomes (blue) with telomeres taken using CLSM (red) and dSTORM (white). (b) dSTORM image of a telomere, where each dot represents an event. (c) Magnified image of a telomere using CLSM (red) and dSTORM (white). Reprinted (adapted) with permission from ref. 137. Copyright 2017, American Chemical Society. (D) Top: SIM image of HeLa cells stained with Cy5@AuNPs showing lysosomes. Bottom: SIM images of HeLa cells stained with Cy5@AuNPs during kiss-and-run, fusion and fission processes in HeLa cells. Adapted from ref. 138 under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). Theranostics published by Ivyspring International Publisher. © The Author 2020. (E) Top: SIM images of U-2 OS cells stained with surface-functionalized Halo-Tag AuNPs (GNP-Atto565-fR8-CA) showing ER (left), ER-mitochondria dynamics (centre) and ER-lysosomes dynamics (right). Bottom: Time-lapse SIM images of ER dynamics in transfected U-2 OS cells stained with Halo-Tag AuNPs. Reprinted (adapted) with permission from ref. 139. Copyright 2024, American Chemical Society. |
Various studies have employed AuNPs for SRM, despite their large size, to visualize lysosomes, human telomeres and ER in live cells.137–139,144–146 Using dSTORM, Jeynes and co-workers measured the biophysical characteristics of individual telomeres (average length ∼ 10 kbps and diameters ∼ 60–300 nm) using AuNPs as probes (Fig. 13C and 14C).137 However, despite providing strong contrast, AuNPs can generate background noise and nonspecific signals, impairing the clarity of SRM images. To address this, the Qiu group developed surface-modified AuNPs by functionalizing their surfaces with biocompatible PEG, with cyanine 5 (Cy5) further conjugated to the terminal end of the PEG chain (Fig. 13D).138 SIM imaging of these surface engineered-AuNPs revealed excellent photostability and lysosome localization, enabling lysosome labelling for more than three days (Fig. 14D). Xu et al. recently developed surface-functionalized Halo-Tag AuNPs (GNP-Atto565-fR8-CA) for long-term imaging of ER dynamics in living cells using SIM nanoscopy (Fig. 13E and 14E).139 These AuNPs were functionalized with the cell-penetrating peptide fR8 facilitating efficient cytoplasmic delivery and addressing the challenge of vesicular entrapment. Overall, these studies demonstrate that AuNPs and AgNPs exhibit size-dependent, optical absorption and scattering properties, while surface functionalization significantly enhances their stability and photostability. In another study, Liang et al. developed lanthanide-doped NPs enabling background-suppressed STED imaging across all-NIR spectral bands, achieving a lateral resolution of less than 20 nm and zero photobleaching.147 Notably, the all-NIR regime of these NPs facilitated high-contrast deep-tissue imaging at a depth of ∼50 μm with a spatial resolution of ∼70 nm.
Of particular interest, AuNPs, with their unique optical properties and versatile surface chemistry, can integrate multiple imaging or targeting agents, making them ideal multifunctional nanoplatforms for imaging and therapeutic applications.148 The AuNP scaffold serves as a multimodal probe detectable via optical and electron microscopy, a distinct advantage for 100 nm AuNPs, which can be visualized as individual, nonaggregated particles within the microscope diffraction limit. Moreover, AuNP scaffolds allow the attachment of hundreds to thousands of metal complexes on a single nanoparticle, producing a strong and characteristic luminescent signal. Strikingly, functionalization of AuNPs with luminescent metal-based probes, such as Ru(II),149–152 Ir(III),153–155 Os(II),156 and Fe(III),157 has been shown to significantly improve spatial resolution in imaging biological systems. Despite their potential to address key challenges in SRM, including sensitivity, resolution, and performance in complex biological environments, the application of AuNP-metal complex hybrid systems in super-resolution imaging remains underexplored.
In this review, we summarize recent progress on metal complexes coordinated with various metals and tuneable ligands for super-resolution imaging of diverse subcellular structures, enabling the tracking of temporal dynamics in live cells. These complexes provide high-contrast, high-resolution imaging at greater tissue depths in vivo. Compared to conventional fluorescent probes, they offer distinct advantages, including high photostability, tuneable emission properties, and extended excited-state lifetimes, achieved through specific ligand design. These distinct properties of metal complexes position them as indispensable tools for high-resolution imaging and theranostic applications, with demonstrated efficacy in addressing cancer and bacterial infections. Moreover, the emerging approaches of nanosized hybrid probes based on Au scaffolds and transition metal complexes represent a promising strategy for translating the attractive luminescent properties of metal complexes into nanoscale platforms.
Current research efforts focus on expanding the range of molecular species that can be imaged simultaneously. This goal can be approached by developing multi-organelle-targeted fluorescent probes for super-resolution imaging, enabling the investigation of interactions between subcellular organelles.161 However, cells contain thousands of distinct genes and molecules that collectively determine cellular behaviour and function, whereas traditional multicolour imaging typically visualizes only a few molecular species at a time. Recent advances in high-throughput genomic-scale imaging are beginning to address this limitation. For instance, single-cell fluorescence in situ hybridization (FISH),162,163 and in situ sequencing,164,165 now allows simultaneous imaging of RNAs from over 1000 genes within individual cells. Similar multiplexing capabilities may soon extend to DNA and protein imaging. Integrating these methods with SRM could enable genomic-scale, super-resolution imaging, providing detailed insights into molecular networks, signalling pathways, and cellular functions at unprecedented scales. Nonetheless, the applicability of metal complexes in these contexts remains to be determined.
Concurrently, there is a growing interest in designing multifunctional metal complexes that serve both as imaging and therapeutic agents, allowing for real-time monitoring of treatment responses.63,65,67,68,72 Integrating these complexes with advanced imaging techniques, such as SRM, could facilitate precise visualization of disease progression and continuous assessment of therapeutic efficacy, paving the way for effective clinical diagnostics and therapeutic interventions.
In addition to super-resolution imaging, integrating techniques such as FRET with SRM holds considerable promise for visualizing real-time molecular interactions at the nanoscale by detecting proximity-dependent energy transfer between donor and acceptor fluorophores.7 Moreover, advances in single-molecule tracking and hybrid methodologies that combine SRM with live-cell imaging,166 or biosensors,167 further enable the study of dynamic molecular interactions within the complex and heterogeneous cellular environment.
The growing interest in super-resolution 3D tissue imaging is driven by its ability to capture cells and biomolecules in their native context, enabling the study of biological systems in their natural state while preserving dynamic functional and morphological processes. However, achieving super-resolution imaging in living, awake, and behaving animals remains challenging due to motion artifacts. Metal complexes exhibit biodistribution patterns governed by size, charge, lipophilicity, and coordination environment,66,130 while organic fluorescent dyes and proteins diffuse more uniformly but specificity is a major concern. Positively charged cationic complexes target negatively charged cellular components (e.g., mitochondrial membranes, nucleic acids), while lipophilic ones passively diffuse across membranes but risk off-target accumulation in perfused organs (liver, spleen, kidneys), affecting imaging precision. It is still the case that rigorous pharmacokinetic studies are required to assess the biodistribution, clearance, and long-term safety of metal complexes for broader in vivo applications. Of note, successful applications of SRM have been demonstrated in models such as zebrafish,168,169 and mice,170–174 highlighting the translatability of this technique in living organisms. These advances pave the way for exploring the applicability of metal-complex-based fluorescent probes in these advanced in vivo settings using SRM. A promising approach involves optimizing the structure of metal-based probes to refine their optical properties for the NIR region, enabling deeper tissue penetration and minimizing background autofluorescence for clearer and more precise in vivo imaging.
Another promising approach involves achieving optical transparency in live animals through the use of absorbent molecules,175 combined with 3D SRM imaging. This integration enables deeper and more detailed visualization of intact biological systems. By enhancing light penetration, reducing background noise, and minimizing scattering, optical transparency preserves image quality, thereby offering novel insights into large-scale biological questions, including neuronal wiring in mammalian brains and developmental processes in human embryos. These advancements highlight the need for developing new chemical probes specifically designed for 3D imaging of optically transparent tissues.
Most importantly, current efforts to reduce the toxicity of metal complexes for live-cell super-resolution imaging focus on limiting ROS production by using less reactive ligand frameworks, incorporating targeting ligands or protective coatings, and enhancing photostability via selective energy transfer mechanisms.130 For example, Li et al. demonstrated that incorporating biothiol-responsive disulfide-linked PEG chains into Ir(III) complexes improves their biocompatibility by enhancing solubility and reducing nonspecific interactions.176 Similarly, Deng et al. showed that Ir(III) PEG micelles enhance tumour targeting and prolong circulation time by reducing nonspecific protein interactions.177 These findings highlight the role of ligand modifications, such as PEGylation or biodegradable coordination frameworks, in minimizing toxicity. Some metal complexes are engineered to remain inert until activated by specific biological triggers, thus reducing toxicity during circulation while enabling precise imaging.66,178 For instance, Liu et al. designed an afterglow luminescent probe that requires only short-term excitation, reducing the need for prolonged external sensitization and thereby minimizing phototoxicity – an important consideration for deep-tissue imaging.179 Additionally, Shum et al. explored Ru(II) tetrazine complexes utilizing a two-step targeting strategy, enabling selective accumulation and controlled activation in cancer cells, thereby reducing systemic toxicity.180 In addition, metal leaching can be mitigated through strong chelation, electron-rich functional groups, and steric protection. Additional approaches include substituting heavy metals with biocompatible alternatives (e.g., Zn or lanthanides), tuning photophysical properties to avoid interference with cellular chromophores, and encapsulating complexes in biocompatible nanoparticles for shielding and controlled release. Moreover, using low-energy excitation (e.g., NIR excitation) and optimized imaging conditions (e.g., reduced laser power, shorter exposure times) further minimizes photodamage. As such, incorporating these considerations into probe design, along with extensive optimization, can establish metal complexes as highly effective tools for SRM.
2D | Two-dimensional |
2PA | Two-photon absorption |
3D | Three-dimensional |
3PA | Three-photon absorption |
Ag | Silver |
AIE | Aggregation-induced emission |
ALP | Alkaline phosphatase |
Au | Gold |
Aβ | Amyloid-β |
B. cereus | Bacillus cereus |
BSA | Bovine serum albumin |
C. elegans | Caenorhabditis elegans |
CLSM | Confocal laser scanning microscopy |
CW-STED | Continuous-wave STED |
Cy | Cyanine |
DFC | Dense fibrillar components |
DNA-PAINT | DNA-based PAINT |
dSTORM | Direct STORM |
E. coli | Escherichia coli |
EM | Electron microscopy |
ER | Endoplasmic reticulum |
FC | Fibrillar centres |
FLSM | Light-sheet fluorescence microscopy |
fps | Frames per second |
FRAP | Fluorescence recovery after photobleaching |
FRET | Fluorescence resonance energy transfer |
gCW-STED | Gated CW-STED |
GFP | Green fluorescent protein |
GSD | Ground state depletion |
IL | Intraligand |
ILCT | Intraligand charge-transfer |
ISC | Intersystem crossing |
LLCT | Ligand-to-ligand charge-transfer |
LMMCT | Ligand-to-metal–metal charge transfer |
LMCT | Ligand-to-metal charge-transfer |
LMP | Lysosomal membrane permeabilization |
LRO | Lysosome-related organelles |
LSPR | Localized surface plasmon resonance |
MINFLUX | Minimum photon flux |
MLC | Mitochondria-lysosome contact |
MLCT | Metal-to-ligand charge-transfer |
MMLCT | metal–metal-to-ligand charge-transfer |
MRI | Magnetic resonance imaging |
MRSA | Methicillin-resistant S. aureus |
mtDNA | Mitochondrial DNA |
NanoSIMS | Nanoscale secondary ion mass spectrometry |
NIR | Near-infrared |
NP | Nanoparticle |
PAINT | Point accumulation for imaging in nanoscale topography |
PALM | Photoactivated localization microscopy |
PDT | Photodynamic therapy |
PEG | Polyethylene glycol |
PIRET | Plasmon-induced resonance energy transfer |
PSF | Point spread function |
resPAINT | Reservoir-PAINT |
RNA | Ribonucleic acid |
ROS | Reactive oxygen species |
RPC | Ruthenium(II) polypyridyl complexes |
rRNA | Ribosomal RNA |
S. aureus | Staphylococcus aureus |
S0 | Singlet ground state |
SERS | Surface-enhanced Raman spectroscopy |
SIM | Structured illumination microscopy |
SiO2 | Silica |
SMED | Surface-migration emission depletion |
SMLM | Single-molecule localization microscopy |
SOC | Spin–orbit coupling |
SOFI | Super-resolution optical fluctuation imaging |
SplitSMLM | Multi-colour SMLM with a spectral image splitter |
SPT | Single-particle tracking |
SRM | Super-resolution microscopy |
SRRF | Super-resolution radial fluctuation |
SSM | Scanning switch-off microscopy |
STED | Stimulated emission depletion |
STEDD | Stimulated emission double depletion |
STORM | Stochastic optical reconstruction microscopy |
T1 | Triplet excited state |
TEM | Transmission electron microscopy |
This journal is © The Royal Society of Chemistry 2025 |