Abstract
Direct monitoring the multifaceted intracellular microenvironment is fundamental for deciphering localized cellular behavior and function. However, performing nanoscale spatiotemporal sensing of the dynamic physiochemical environment within a live cell has remained challenging, mainly limited by the lack of robust, biocompatible method which satisfies the requirements for spatiotemporal resolution, sensitivity, multi-physical sensing capability and long-term viability. Here we present a super-resolution-based quantum biosensing approach, namely Super-resolution Multi-physical field Imaging with Deconvolution, implemented in our custom-built setup which integrates nitrogen vacancy center-based quantum sensing with structured illumination microscopy. This enables quantitative, super-resolved, and long-term tracking of intracellular physical fields, featuring spatial resolution of 136 nm, temporal resolution of 150 ms, and field of view of 180×180 μm2. We further perform super-resolution multimodal imaging in a live cell, revealing the spatially heterogeneous reactive oxygen species dynamics under external stimulation. Our work enables the dynamic assessment of localized physiochemical dynamics under super-resolution quantum biosensing.
Subject terms: Imaging and sensing, Imaging techniques, Imaging, Mitochondria
This work presents a super-resolution quantum biosensing approach by integrating nitrogen vacancy (NV) center-based quantum nanodiamond sensors with structured illumination microscopy, which achieves super-resolution multi-physical imaging within live cells.
Introduction
Coordinated intracellular activities, precisely regulated in time and space, are essential for governing cell functions. Dysregulations of these processes within specialized cells reflect in their collective behaviors, resulting in physiological disorders and diseases1. A goal in cell biology is to acquire the multifaceted information of these several dynamic activities during cell signaling, metabolism and transcription, which occur within complex space of cells while maintaining internal homeostasis2–4. This requires a live cell technique that can monitor the intracellular microenvironment (such as temperature, reactive oxygen species (ROS), etc) in real time and at high spatiotemporal resolution5.
Conventional fluorescence-based techniques, employing molecular probes or genetically encoded biosensors, have been developed to measure intracellular pH6, temperature7, and ROS8. With the advent of super-resolution imaging, fluorescence methods have advanced spatial resolution into sub-organelle scale9–11. However, these approaches remain limited by photobleaching, phototoxicity12, and lack capacities of real-time quantifying the multi-physical field changes of cellular environment. Meanwhile, electrochemical approaches such as microelectrode arrays enable dynamic monitoring of electrophysiological activity or neurotransmitter release but lack adequate spatial resolution13. Therefore, there is a pressing need for developing tools that allow dynamic observation of intracellular activities with high spatiotemporal resolution.
Nitrogen-vacancy (NV) center offers a quantum sensing approach for localized physical field detection14,15. This bio-compatible quantum sensor features good photostability and possesses highly sensitive response to multiple physical quantities, which can operate under physiological conditions16–20. NV center in fluorescent nanodiamonds (FNDs) has recently enabled applications such as thermometry in neurons firing21, thermometry in metabolism22, adhesive forces measurement during cell migration23, and free radical detection in immune response24. Although functionalized FND probes were previously used to probe localized microenvironment and ensure close enough distance with the targeted organelle24, such approaches are still insufficient for capturing the dynamic, real-time interplay between adjacent intracellular regions. To address this, the spatiotemporal resolution of intracellular NV quantum sensing must be improved. Prior attempts have been made to integrate NV sensing with super-resolution microscopy in inert physical systems25–27. However, super-resolution quantum biosensing in live cells remains elusive, which needs to address the complex intracellular environment, the trade-off between spatial and temporal resolution, and the decoupling of multi-physical quantities. This process is crucial because coordinated intracellular activities including localized metabolism and redox regulation, typically involves free radical generation and thermal conversion within spatially confined microdomains28–30.
In this work, we demonstrate a live cell quantum sensing approach, termed Super-resolution Multi-physical field Imaging with Deconvolution (SMID), which integrates NV-based sensing with structured illumination microscopy (SIM)31,32. This integration enables super-resolved imaging of cellular microenvironments, with spatial resolution of 136 nm, temporal resolution of 150 ms, and allows hour-scale observation. Leveraging SMID, we achieved simultaneous in situ mapping of subcellular structures, thermal heterogeneities, and localized ROS levels. This approach reveals a spatiotemporal pattern in ROS dynamics upon drug stimulation, which fundamentally deviates from the conventional equilibrium-state paradigm of live-cell physicochemical homeostasis.
Results
The workflow of SMID
SMID is designed to realize the super-resolution quantum sensing in live cells (Supplementary Fig. 1). To quantify the biological activity states at distinct intracellular sites or organelles, it is important to precisely and parallelly sense the multi-physical field changes of localized physiochemical events in live cells. The concept of SMID was introduced in Fig. 1: the atomic-size NV centers in FNDs act as the quantum probes for imaging nanoscale multi-physical fields. Each probe senses its surrounding microenvironment including temperature, ions and ROS, in physiological processes such as metabolism, drug stimulation and signaling. Moreover, the density of these probes is controlled to ensure their distribution across different cellular regions (Supplementary Fig. 2). This enables the nanoscale multi-site mapping of local physical fields in a live cell. Then, SMID multiplies this nanoscale quantum sensing process by conducting super-resolution parallel readout of densely distributed FND sensors. Thus, this approach aims to transform the conventional spatiotemporally averaged measurements into super-resolved microenvironment maps, which allows to reveal previously inaccessible spatiotemporal heterogeneities underlying live cell dynamics.
Fig. 1. Concept of SMID.

Cellular activities such as cellular signaling, metabolism and drug response occur within localized microdomains, where the physicochemical gradients such as temperature, ROS and ion levels govern their function. To move from the ensemble averaged measurements, super-resolution live cell quantum sensing is developed to precisely and parallelly resolve the spatiotemporal dynamics of each microenvironments with NV quantum sensors.
However, the integration of NV sensing with super-resolution imaging faces multiple experimental challenges. First, given the maintenance of cellular activity during observation, the introduced phototoxicity must be reduced. Compared to other super-resolution approaches, SIM employs a relatively low excitation power, offering a compromise among imaging speed and phototoxicity, which is suitable for live cell imaging. Therefore, we integrated NV quantum sensing with SIM through a custom-built live cell setup, which generates structured illumination (λ = 532 nm) from a digital micromirror device (DMD). DMD’s high pattern rate and high resolution (1920 × 1080 pixels) enable fast patterned illumination for active control in super-resolution imaging and quantum sensing sequence that captures fast evolution of microenvironment (Suppl. Note 1 and Suppl. Fig. 3). For multi-physical imaging, we designed sensing sequences for decoupling different physical quantities, which involves the precise control of structured light illumination, microwave delivery and detection pulse exposure. This sequence facilitates the precise super-resolution manipulation and read-out of NV’s spin state via optically detected magnetic resonance (ODMR) and spin-lattice relaxation (T1) measurements (Suppl. Fig. 4). To implement this technique, we built a multi-modal microscope setup that simultaneously controls optical fields, spin manipulation and live cell operation (Suppl. Fig. 1 and Suppl. Fig. 3), featuring an excellent mechanical stability using a customized 3D-locking system.
Besides, we integrated a deconvolution algorithm as a preprocessing step, to specifically tune super-resolution image reconstruction towards multi-physical quantum sensing under inhomogeneous live cell conditions33. In this way, illumination intensity could be further reduced whilst achieving comparable image quality. The preliminary reconstructed image stacks acquired were parallel-processed to extract the parameters related to different physical fields, forming the super-resolution multi-physical images (Fig. 2a, and detailed in Methods and Suppl. Fig. 5).
Fig. 2. Workflow of SMID.

a Schematic of super-resolution multi-physical imaging workflow. Spatial distribution of mitochondria and quantum probes acquired by (b) wide-field imaging, (c) super-resolved imaging and (d) super-resolution multi-physical field imaging with three modalities. A specific area (box line) in three imaging is zoomed in to observe the details.
Together, we demonstrated three imaging modalities, (1) SMID-optical imaging (SMID-OPT), (2) SMID-temperature imaging (SMID-TEMP) and (3) SMID-relaxation imaging (SMID-T1). Compared with the conventional wide-field image (Fig. 2b) which hardly distinguishes FND assembles, and SIM images which only offer FND’s location information (Fig. 2c), SMID capitalizes its capacity in simultaneously resolving the spatial locations and physical environment of the quantum probes beyond diffraction limit (Fig. 2d), effectively promoting super-resolution imaging into quantum biosensing.
In vitro validation of SMID performance
To benchmark the imaging performance of SMID, we first conducted in-vitro validation experiments (Suppl. Note 2 and Suppl. Fig. 6). As shown in Fig. 3a, SMID clearly resolves two spatially distinct FNDs that are blurred in the corresponding conventional wide-field (WF) imaging. Analysis across the intensity profile of FND reveals their spacing apart distance, cross-validated with the scanning electron microscopy (SEM) result (Fig. 3b and Suppl. Fig. 7). To quantify the spatial resolution, we further conducted Fourier resolution correlation (FRC) analysis34,35 across the entire imaging area. The FRC result indicates that SMID achieves a spatial resolution of 136 nm, comparable to traditional Fair-SIM36 and HiFi-SIM37. This ensures its competing optical performance in the following sensing experiment (Fig. 3c).
Fig. 3. In vitro validation of SMID.

a Image of FNDs in SEM imaging, wide-field fluorescence imaging, wide-field RL‑deconvolved imaging and SMID imaging, respectively. b Intensity profiles of FNDs in SEM image and SMID image indicated by dotted line in (a). c FRC analysis of the wide field imaging and different super-resolved imaging approaches using FND samples. The dashed line corresponds to the 1/7 threshold. d The relationship between SMID-T1 value with ·OH concentration. All data are presented as the mean ± standard deviation from three independent experiments (n = 3). In the boxplot, the central line represents the median, while the central dot represents the mean. The lower and upper edges of the box correspond to the 25th and 75th percentiles respectively. The whiskers extend to the most extreme data points. e Four-point thermometry calibration curve of SMID-TEMP. All data are presented as the mean ± standard deviation from four independent experiments (n = 4). f, g WF-T1/SMID-T1 and WF-TEMP/SMID-TEMP imaging of same FND samples. A specific area (yellow box line) is zoomed in to observe the details. The scale bar of the zoomed areas is 300 nm. h SMID-TEMP result of gradient temperature on FND samples in vitro. The laser spot is focused on the Au waveguide (white dashed circle at lower-right corner).
T1 relaxation time refers to NV’s longitudinal spin-lattice relaxation time, which could be perturbed by the surrounding paramagnetic ions or radicals38,39. For capturing the swift changes in intracellular physiochemical dynamics (typically ~ms-min scale40), a self-designed imaging sequence was used to demonstrate SMID-T1 imaging that features millisecond temporal resolution (Suppl. Note 3). In this sequence, ‘single-τ’ method was utilized to accelerate imaging speed, repeated to gain adequate SNR for T1 imaging (Methods and Suppl. Note 4). Accordingly, the exposure time of each pattern was set at 50 ms. Given that a single SMID frame was reconstructed from nine patterns, with a ‘rolling processing mode’ where the latest six images contribute to the computation of subsequent new images41, our sequence allows SMID-T1 to be conducted under 150 ms. With this, we employed hydroxyl radical with gradient concentrations to validate the quantitative capabilities of SMID-T1 (Fig. 3d). The result demonstrates that T1 value is sensitive to the used hydroxyl radical concentration range, which is comparable to the physiological ROS levels42,43 (Suppl. Fig. 8 and Suppl. Fig. 9). Meanwhile, their relationship could be quantitively captured using our previously derived model44. Similarly, temperature changes have an impact on NV’s zero-field splitting value through affecting diamond’s lattice distortion45 (Suppl. Note 5). After cross-validating the performance with full-spectrum ODMR fitting, SMID-TEMP incorporated a four-point ODMR method for data acquisition to improve the temporal resolution of the data collection46,47 (Suppl. Fig. 10 and Suppl. Fig. 11). We further employed a standard temperature gradient calibration, confirming the correlation between temperature and D-value measured in SMID-TEMP (Fig. 3e).
Multimodal imaging result (SMID-TEMP and SMID-T1) of the FNDs is shown in Fig. 3f-g. Compared with the conventional wide-field images, our SMID accurately reveals the precise localization of the FND probes with sub-diffraction resolution, along with encoding their situated microscopic temperature (measured by ODMR) and magnetic noise environment (measured by T1), which is related to free radical concentrations. Notably, the observed discrepancy between WF‑ and SMID‑derived T1 values arises from signal blending in WF due to insufficient spatial resolution, whereas SMID-T1 resolves individual FNDs and eliminates cross‑talk.
Moreover, to demonstrate SMID’s spatially resolved measurement capability under induced gradient fields, we focused laser on waveguide to generate a temperature gradient on nearby FND samples and performed SMID‑TEMP imaging. As shown in Fig. 3h, the SMID‑TEMP image clearly reveals a temperature gradient with high spatial resolution: FNDs located closer to the heat source exhibit higher temperatures, which validate the capability of SMID‑TEMP to spatially resolve environmental changes.
SMID for live cell multimodal imaging
Next, to carry out the multimodal imaging in live cells, we developed a hybrid SMID-TEMP/SMID-T1 imaging sequence (Fig. 4b). The results show that our method allows independent, super-resolved mapping of optical signals, T1 and temperature within the same live cell (Fig. 4a), revealing the simultaneous spatial distribution of internalized FND quantum probes, subcellular temperature heterogeneities and intracellular ROS levels. Notably, FRC analysis demonstrates that both SMID-OPT and SMID-T1 achieve about twofold resolution enhancement over the optical diffraction limit of our system, enabling precisely correlating the nanoscale structures with their microenvironment (Fig. 4c).
Fig. 4. SMID for live cell multi-modal imaging.

a Multi-modal imaging results in a living Hela cell, comparing wide-field and SMID imaging. b Schematics of multi-modal imaging sequence. N and K refer to the loop numbers of different part of sequence. f refers to the sweeping frequencies. Sig. and Ref. correspond to signal and reference frames in sequence. c FRC analysis of wide field images and super-resolved images in live cell. The dashed line corresponds to the 1/7 threshold. d Mitochondria and FND quantum probes distributed in live cells under SMID-OPT and SMID-T1 imaging with a large field of view concatenated from 25 sub FOV images.
Due to SMID’s high temporal resolution, within a short acquisition time of 12 s, we captured a large field of view (FOV) at 180 × 180 μm2, composed of 5 × 5 sub FOVs. This rapid large-area imaging capability is crucial for the statistical analysis of the cellular populations while maintaining high spatial resolution (Fig. 4d). In this multimodal image, the spatial distribution between mitochondria and the FNDs is clearly visible (Suppl. Fig. 12), which can be used in subsequent studies to investigate the relationship between the intracellular microenvironments and the specific regions.
Moreover, constrained by the susceptibility of cellular activity to phototoxicity or chemical toxicity, live cell observations often face substantial challenges of prolonged observation, thereby limiting the analysis of slow-progressing processes48,49. To evaluate the long-term stability and dynamic sensing capability of SMID, we conducted hour-long time-lapse imaging and carried out the phototoxicity assessment (Suppl. Note 6, Suppl. Fig. 13 and Suppl. Fig. 14). Specifically, a self-developed three-dimension locking system was integrated into our setup to suppress the system drift, ensuring long-term acquisition stability (Suppl. Fig. 15). The result shows that during the observation, mitochondrial morphology remains well-preserved, indicating reduced phototoxic effects under our structural illumination conditions (Suppl. Fig. 14). Besides, the fluorescence intensity of FNDs remains remarkably stable, demonstrating its photostability. In contrast, the intensity of mitochondrial dye PK Mito Red exhibited progressive photobleaching, a limitation for organic dyes. Crucially, in SMID-T1 imaging, the T1 relaxation time reveals temporal fluctuations that reflect dynamic variations of physiochemical states in the local cellular microenvironment, while the concurrent NV fluorescence intensity remains stable (Suppl. Fig. 13). This distinct behavior confirms that the T1 fluctuations originated from environmental changes rather than probe photo-instability, highlighting the robust sensing capability and reliability of SMID-T1 for monitoring dynamic intracellular processes over extended durations (Suppl. Table 1).
Localized intracellular spatiotemporal dynamics
To quantitatively probe the dynamic activities in intracellular microenvironment, we conducted SMID-T1 imaging for FNDs located in different cellular regions, including mitochondria, cytoplasm, and nucleus (Fig. 5a). Typically, T1 imaging reveals the intracellular relative ROS level, which exhibits a spatial dependence. The results indicate that mitochondrial regions display higher ROS levels (shorter T1 value), which may correlate with the metabolic activity occurring within the mitochondria. Compared to WF-T1, SMID-T1 reveals more refined spatial variations, owing to its super-resolution capability (Fig. 5b).
Fig. 5. Live cell spatiotemporal dynamics.

a Image of live cell and quantum probes with mitochondria (purple), cytoplasm and nucleus. Each probe is numbered. b Statistic of T1 value distributions in different regions from (a). All data are presented as the mean ± standard deviation from three independent experiments (n = 3). T1 value changes of FND probes with adding SOD and CCCP under (c) wide-field imaging and (d) SMID imaging. The scale bar of the zoomed areas in (c, d) is 300 nm and 100 nm, respectively. e Different types of dynamic T1 changes with adding SOD and CCCP. f Cluster analysis of different types in (e). PC refers to principal component. PC1 primarily corresponds to the magnitude of T1 changes upon SOD and CCCP stimulation (IP and DP), which is related to the intensity of the cellular oxidative stress; PC2 mainly distinguishes subcellular compartments reflecting differences in the source of ROS production. g Distribution of different FNDs in live cell. h T1 intensity oscillations profiles of different FNDs in (g).
We next perturbed the cellular state to investigate the dynamic responses. Superoxide dismutase (SOD) was used to scavenge free radicals and mitigate oxidative stress50, while carbonyl cyanide 3-chlorophenylhydrazone (CCCP) was introduced to induce apoptosis and disrupt mitochondrial activity51. In the conventional WF-T1 measurements, the spatially-averaged T1 signal shows a steady state upward trend upon the SOD addition and a subsequent steady state downward trend after the CCCP introduction. In contrast, SMID-T1 unveils distinct types of T1 dynamics that are obscured in the corresponding WF-T1 result (Fig. 5c-d). This observation highlights SMID’s advantage in super-resolved capturing of local microenvironmental dynamics. Further analysis of all FNDs’ SMID-T1 results within the same cell identified five characteristic response types 1-5 (Fig. 5e). The primary distinctions between these 5 types were manifested in T1 increasing proportion (IP) response to SOD and decreasing proportion (DP) response to CCCP. To systematically classify these responses, we performed a clustering analysis based on IP, DP, FND particle size, and the cellular region of interest (Fig. 5f). This analysis reveals that each type of oxidative stress response exhibits a distinct regional preference, suggesting that specific microenvironments may drive the emergence of particular response phenotypes.
Particularly, type 5 exhibits unexpected oscillatory dynamics in ROS levels and can be influenced by drug stimulation, which is different from the conventional view of steady-state free radical generation or consumption (Fig. 5g–h). The observed dynamics may reflect the dynamic feedback within the cellular redox system, such as periodic activation of antioxidant defenses or rhythmic fluctuations in mitochondrial activity, which can periodically counteract and reduce ROS accumulation52,53.
Discussion
In this work, we present a SMID approach, demonstrating the super-resolution quantum sensing in live cells. This technique features high spatiotemporal resolution, biocompatibility and capacity for resolving multi-physical fields. Leveraging custom-developed imaging sequences, microscope setup and decoupling algorithms, SMID is specially adapted to NV-based imaging and is promising to explore future integration with emerging quantum sensing sequences.
Notably, the movement in biosensing from ensemble averages to molecular-level physicochemical dynamics entails: (1) rendering the indistinct cellular structures discernible with super-resolution imaging, (2) perceiving the previously unseen cellular activity using quantum sensing. The development of SMID has allowed the connection between these two domains. This may bring new insights into understanding cellular behaviors.
Looking forward, we anticipate that integrating SMID with rapidly developing quantum probes, such as fluorescent proteins54, will further accelerate quantum biosensing. By enabling the visualization of dynamic physicochemical gradients within living cells, super-resolution quantum biosensing would offer insights into the functional architecture of the cellular interior, bridging the gap between structural imaging and functional analysis at the nanoscale.
Methods
Living cell sample preparation
The full preparation process could be described into three steps: cell culture, FND intracellular internalization and mitochondrial staining.
Cell culture
The living cell samples used in this study were HeLa cells, cultured in Dulbecco’s Modified Eagle Medium (DMEM; Gibco) supplemented with 10% (v/v) fetal bovine serum (FBS; Gibco) and 1 % (v/v) penicillin–streptomycin. Cells were maintained at 37 °C in a humidified 5% CO₂ atmosphere. For routine passaging, cells were detached with 0.25 % trypsin-EDTA every 2-3 days upon reaching ~90% confluency. For imaging experiments, cells were trypsinized, centrifuged, and plated onto 24 mm glass coverslips in 35 mm dishes. The cells were seeded at an appropriate density to achieve approximately 70-80% confluence at the time of the experiment (Supplementary Fig. 16).
FND intracellular internalization
Following a 12 h attachment period, cells cultured on glass coverslips were exposed to 2 mL of fresh complete DMEM containing FNDs at a final concentration of 0.05 mg/mL. The FND working solution was prepared immediately prior to treatment by diluting the stock solution (1 mg/mL stock solution, 100 nm FNDs in ultrapure water, stored at 4 °C) 1:19 (v/v) with pre-warmed complete medium. Cells were subsequently incubated under standard culture conditions (37 °C, 5 % CO₂) for 20–24 h to promote FND internalization. Notably, the final FND concentration and the incubation time were determined through gradient concentration and incubation time pre-experiment, allowing optimal FND intracellular internalization while maintain cell viability (Suppl. Fig. 2). Furthermore, the T1 value of intracellular FND probes would be influenced both by the inherent properties of NV center and different intracellular environments, varying from ~101 μs to ~102 μs.
Mitochondrial staining
The PK Mito Red (PKMR) working solution was prepared by diluting the stock solution (200 µM) 1:1000 in pre-warmed complete DMEM to achieve a final concentration of 200 nM. Following aspiration of culture medium, cells on coverslips were washed twice with warm PBS (37 °C). Cells were then incubated with 1 mL of the 200 nM PKMR solution under standard culture conditions (37 °C, 5% CO₂) for 15 min. After removal of the staining solution and two additional PBS washes, cells were maintained in fresh complete medium and immediately subjected to live-cell imaging.
Cellular regulation
In regulation experiment, we treated the cells with carbonyl cyanide 3-chlorophenylhydrazone (CCCP, a drug that induces apoptosis) and superoxide dismutase (SOD, an enzyme that suppresses apoptosis by eliminating superoxide free radicals). When apoptosis inhibition is required, we added 1 mL of SOD solution (final concentration of 1000 U/mL); when apoptosis promotion is needed, add 1 mL of CCCP solution (final concentration of 10 µM).
Experimental workflow for SMID
To technically realize the SMID methodology, we developed a workflow which consists of several parts (Fig. 2). First, the DMD generates striped illumination patterns with three distinct orientations. For each orientation, the phase of the stripes is shifted in three steps with a fixed stripe period. To achieve high modulation contrast, the polarization direction of the illumination is aligned parallel to stripe orientation. In total, nine unique illumination patterns (3 angles × 3 phases) are used to reconstruct a single super-resolution image. Second, we would generate SMID-T1 and SMID-TEMP sequence through self-developed program, (sequence parameters detailed in Supplementary Notes 4). Last, through performing data processing of the image stacks and resolving the corresponding physical properties (Supplementary Notes 7 and Supplementary Software), we acquired super-resolution multi-physical imaging maps.
Supplementary information
Description of Additional Supplementary Files
Source data
Acknowledgements
We thank the Analysis Center of Agrobiology and Environmental Sciences (to X. Zheng and N. Rong), Micro and Nano Fabrication Center (to Jiabao Sun and Ziqing Zhang) at Zhejiang University for technical assistances. We thank Yile Sun (Zhejiang University), Chengrong Jiang (Zhejiang Normal University) and Chuankang Li (Zhejiang Normal University) for technical assistances.
Author contributions
Y.L., Y.Y. and K.X. contributed equally to this work. J.F., Y.L., Y.Y., L.Y. and K.X. designed the setup. Y.L. built the setup with help from L.Y and Y.Y. K.X. designed the algorithm and conducted the data processing. Y.L., Y.Y., K.X., Z.M., Y.X., Z.Z., X.W., C.Z., and W.Z. conducted and analyzed the experiments. J.F. conceptualized and supervised the work. Y.L., Y.Y. and J.F. wrote the manuscript with feedback from other authors.
Peer review
Peer review information
Nature Communications thanks Ettore Bernardi and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Funding
This work was financially supported by the National Natural Science Foundation of China (Grants 21974123 and 62205298), the Fundamental Research Funds for the Zhejiang Provincial Universities (Grant No.: 226202500087), and the Zhejiang Provincial Natural Science Foundation of China (Grant number: LQ23F050008). J.F. acknowledges the support from the New Cornerstone Science Foundation through the XPLORER PRIZE.
Data availability
The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Raw data in other formats are available from the corresponding author upon reasonable request. Source data are provided within this paper. Source data are provided with this paper.
Code availability
The codes used to perform SMID image reconstruction are provided within this paper. All data processing was achieved using MATLAB R2017b with Wavelet Toolbox (Win 10).
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Yuxian Lu, Yibo Yang, Ke Xu.
Supplementary information
The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-78321-2.
References
- 1.Sarikhani, E. et al. Nanosensors for real-time intracellular analytics. Nat. Nanotechnol.20, 1374–1387 (2025). [DOI] [PubMed] [Google Scholar]
- 2.Tan, Z. et al. Time-resolved fluorescent proteins expand fluorescent microscopy in temporal and spectral domains. Cell188, 1–19 (2025). [DOI] [PubMed] [Google Scholar]
- 3.Fu, P. C. et al. Super-resolution imaging of non-fluorescent molecules by photothermal relaxation localization microscopy. Nat. Photonics17, 330–337 (2023). [Google Scholar]
- 4.Shan, X. C. et al. Sub-femtonewton force sensing in solution by super-resolved photonic force microscopy. Nat. Photonics18, 998–998 (2024). [Google Scholar]
- 5.Pak, V. V. et al. Ultrasensitive genetically encoded indicator for hydrogen peroxide identifies roles for the oxidant in cell migration and mitochondrial function. Cell Metab.31, 642–653 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hou, J. T. et al. Fluorescent bioimaging of pH: from design to applications. Chem. Soc. Rev.46, 2076–2090 (2017). [DOI] [PubMed] [Google Scholar]
- 7.Okabe, K. et al. Intracellular temperature mapping with a fluorescent polymeric thermometer and fluorescence lifetime imaging microscopy. Nat. Commun.3, 705 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Niu, H. Y. et al. Photoinduced electron transfer (PeT) based fluorescent probes for cellular imaging and disease therapy. Chem. Soc. Rev.52, 2322–2357 (2023). [DOI] [PubMed] [Google Scholar]
- 9.Balzarotti, F. et al. Nanometer resolution imaging and tracking of fluorescent molecules with minimal photon fluxes. Science355, 606–612 (2017). [DOI] [PubMed] [Google Scholar]
- 10.Sigal, Y. M., Zhou, R. B. & Zhuang, X. W. Visualizing and discovering cellular structures with super-resolution microscopy. Science361, 880–887 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lu, Y. X., Zhao, P. L. & Feng, J. D. Spatiotemporal isolation based super-resolution microscopy. Chin. J. Chem.41, 1608–1623 (2023). [Google Scholar]
- 12.Hsieh, H. C. et al. Imaging 3D cell cultures with optical microscopy. Nat. Methods22, 1167–1190 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chung, W. G. et al. Liquid-metal-based three-dimensional microelectrode arrays integrated with implantable ultrathin retinal prosthesis for vision restoration. Nat. Nanotechnol.19, 688–697 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Balasubramanian, G. et al. Nanoscale imaging magnetometry with diamond spins under ambient conditions. Nature455, 648–651 (2008). [DOI] [PubMed] [Google Scholar]
- 15.Schirhagl, R., Chang, K., Loretz, M. & Degen, C. L. Nitrogen-vacancy centers in diamond: nanoscale sensors for physics and biology. Annu. Rev. Phys. Chem.65, 83–105 (2014). [DOI] [PubMed] [Google Scholar]
- 16.McGuinness, L. P. et al. Quantum measurement and orientation tracking of fluorescent nanodiamonds inside living cells. Nat. Nanotechnol.6, 358–363 (2011). [DOI] [PubMed] [Google Scholar]
- 17.Glenn, D. R. et al. Single-cell magnetic imaging using a quantum diamond microscope. Nat. Methods12, 736–738 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Le Sage, D. et al. Optical magnetic imaging of living cells. Nature496, 486–489 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kucsko, G. et al. Nanometre-scale thermometry in a living cell. Nature500, 54–58 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Xu, Y. et al. Live cell quantum multiphysiology enabled by a manipulable single nanodiamond. Natl. Sci. Rev.12, nwaf130 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Petrini, G. et al. Nanodiamond-quantum sensors reveal temperature variation associated to hippocampal neurons firing. Adv. Sci.10, 2202014 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lee, Y., Kim, K., Kim, D. & Lee, J. S. Organelle-specific quantum thermometry using fluorescent nanodiamonds: insights into cellular metabolic thermodynamics. J. Am. Chem. Soc.147, 13180–13189 (2025). [DOI] [PubMed] [Google Scholar]
- 23.Xu, F. et al. Quantum-enhanced diamond molecular tension microscopy for quantifying cellular forces. Sci. Adv.10, eadi5300 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Nie, L. et al. Quantum monitoring of cellular metabolic activities in single mitochondria. Sci. Adv.7, eabf0573 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rittweger, E., Han, K. Y., Irvine, S. E., Eggeling, C. & Hell, S. W. STED microscopy reveals crystal colour centres with nanometric resolution. Nat. Photonics3, 144–147 (2009). [Google Scholar]
- 26.Pfender, M., Aslam, N., Waldherr, G., Neumann, P. & Wrachtrup, J. Single-spin stochastic optical reconstruction microscopy. Proc. Natl. Acad. Sci.111, 14669–14674 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Xu, F. et al. Super-resolution enabled widefield quantum diamond microscopy. ACS Photonics11, 121–127 (2023). [Google Scholar]
- 28.Selim, N. A. & Wojtovich, A. P. Mitochondrial membrane potential and compartmentalized signaling: Calcium, ROS, and beyond. Redox Biol.86, 103859 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sies, H. & Jones, D. P. Reactive oxygen species (ROS) as pleiotropic physiological signalling agents. Nat. Rev. Mol. Cell Bio.21, 363–383 (2020). [DOI] [PubMed] [Google Scholar]
- 30.Singh, G., Vengayil, V., Khanna, A., Adhikary, S. & Laxman, S. Active control of mitochondrial network morphology by metabolism- driven redox state. Proc. Natl. Acad. Sci.122, e2421953122 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chen, X. et al. Super-resolution structured illumination microscopy reconstruction algorithms: a review. Light.: Sci. Appl.12, 172 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Gustafsson, M. G. L. Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy. J. Microsc.198, 82–87 (2000). [DOI] [PubMed] [Google Scholar]
- 33.Wallace, W., Schaefer, L. H. & Swedlow, J. R. A working person’s guide to deconvolution in light microscopy. Biotechniques31, 1076–1097 (2001). [DOI] [PubMed] [Google Scholar]
- 34.Banterle, N., Bui, K. H., Lemke, E. A. & Beck, M. Fourier ring correlation as a resolution criterion for super-resolution microscopy. J. Struct. Biol.183, 363–367 (2013). [DOI] [PubMed] [Google Scholar]
- 35.Nieuwenhuizen, R. P. J. et al. Measuring image resolution in optical nanoscopy. Nat. Methods10, 557–562 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Müller, M., Mönkemöller, V., Hennig, S., Hübner, W. & Huser, T. Open-source image reconstruction of super-resolution structured illumination microscopy data in ImageJ. Nat. Commun.7, 10980 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wen, G. et al. High-fidelity structured illumination microscopy by point-spread-function engineering. Light.: Sci. Appl.10, 70 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Simpson, D. A. et al. Electron paramagnetic resonance microscopy using spins in diamond under ambient conditions. Nat. Commun.8, 458 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Steinert, S. et al. Magnetic spin imaging under ambient conditions with sub-cellular resolution. Nat. Commun.4, 1607 (2013). [DOI] [PubMed] [Google Scholar]
- 40.Guo, Y. T. et al. Visualizing intracellular organelle and cytoskeletal interactions at nanoscale resolution on millisecond timescales. Cell175, 1430–1442 (2018). [DOI] [PubMed] [Google Scholar]
- 41.Zhao, W. S. et al. Sparse deconvolution improves the resolution of live-cell super-resolution fluorescence microscopy. Nat. Biotechnol.40, 606–617 (2022). [DOI] [PubMed] [Google Scholar]
- 42.Kondengadan, S. M. & Wang, B. H. Quantitative factors introduced in the feasibility analysis of Reactive Oxygen Species (ROS)-sensitive triggers. Angew. Chem. Int. Ed.63, e202403880 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Lu, Q., Wu, Y. K. & Weil, T. Nanodiamond quantum sensors for probing free radical biology. Adv. Funct. Mater.36, e00003 (2026). [Google Scholar]
- 44.Yang, Y. B. et al. Quantum-sensing-enabled in situ imaging of free radicals in chemical reactions. Nat. Catal.9, 319–327 (2026). [Google Scholar]
- 45.Acosta, V. M. et al. Temperature dependence of the nitrogen-vacancy magnetic resonance in diamond. Phys. Rev. Lett.104, 070801 (2010). [DOI] [PubMed] [Google Scholar]
- 46.Aslam, N. et al. Quantum sensors for biomedical applications. Nat. Rev. Phys.5, 157–169 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Fujiwara, M. et al. Real-time estimation of the optically detected magnetic resonance shift in diamond quantum thermometry toward biological applications. Phys. Rev. Res.2, 043415 (2020). [Google Scholar]
- 48.Liu, T. L. et al. Observing the cell in its native state: Imaging subcellular dynamics in multicellular organisms. Science360, eaaq1392 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Lu, Z. et al. Long-term intravital subcellular imaging with confocal scanning light-field microscopy. Nat. Biotechnol.43, 569–580 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Wang, Y., Branicky, R., Noë, A. & Hekimi, S. Superoxide dismutases: dual roles in controlling ROS damage and regulating ROS signaling. J. Cell Biol.217, 1915–1928 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Wang, Y. Q., Nartiss, Y., Steipe, B., McQuibban, G. A. & Kim, P. K. ROS-induced mitochondrial depolarization initiates PARK2/PARKIN-dependent mitochondrial degradation by autophagy. Autophagy8, 1462–1476 (2012). [DOI] [PubMed] [Google Scholar]
- 52.Li, B. W. et al. Redox regulation: mechanisms, biology and therapeutic targets in diseases. Signal Transduct. Tar.10, 72 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Murphy, M. P. Mitochondrial dysfunction indirectly elevates ROS production by the endoplasmic reticulum. Cell Metab.18, 145–146 (2013). [DOI] [PubMed] [Google Scholar]
- 54.Feder, J. S. et al. A fluorescent-protein spin qubit. Nature645, 73–79 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Description of Additional Supplementary Files
Data Availability Statement
The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Raw data in other formats are available from the corresponding author upon reasonable request. Source data are provided within this paper. Source data are provided with this paper.
The codes used to perform SMID image reconstruction are provided within this paper. All data processing was achieved using MATLAB R2017b with Wavelet Toolbox (Win 10).
