b University of Chinese Academy of Sciences, Beijing 100049, China
Cellular pseudopodia, dynamic actin-rich projections of the plasma membrane, play essential roles in core physiological processes such as cell adhesion, migration, signal transduction, and immune response. These structures constantly undergo nanoscale morphological changes in real-time, forming distinct substructures to fulfill various physiological functions [1–3]. Filopodia, composed of elongated actin filaments, primarily function as cellular “antennas”, aiding in adhesion, migration, and development by detecting and gathering biomechanical information [4–6]. In contrast, lamellipodia are broad, flat extensions at the leading edge of motile cells, contributing to cell adhesion and migration [7]. Additionally, cells can form complex networks of intercellular channels through pseudopodia, including (tunneling nanotubes)-nanoscale (TNTs), actin-rich, transient tubes facilitating intercellular communication and substance exchange, typically measuring 50–200 nm [8]. The minute size and intricate architecture of cellular pseudopodia challenge traditional optical microscopes, which often lack the resolution to reveal their ultrastructural details. Although techniques such as structured illumination microscopy (SIM) and other super-resolution methods have been used to visualize pseudopodial structures, limitations in spatial resolution and imaging speed hinder their ability to capture the rapid dynamics of pseudopodia in real-time [9–16]. Therefore, innovative super-resolution strategies that merge high spatial resolution with adequate temporal fidelity are crucial for accurately visualizing pseudopodia dynamics at the nanoscale.
Single-molecule localization microscopy (SMLM) has emerged as a powerful solution for achieving high-resolution imaging by localizing individual fluorophores, enabling spatial resolution improvements to 2–20 nm [17–20]. A critical factor in achieving such resolution is the switching of fluorescent molecules between bright (ON) and dark (OFF) states [21–24]. Conventional strategies to facilitate this switching involve the use of high concentrations of exogenous nucleophiles or redox modulators, as well as high-power or ultraviolet (UV) laser irradiation [25]. However, these methods are associated with high cytotoxicity and potential interference with cell physiology, which limits their effectiveness for dynamic super-resolution imaging of pseudopodia in living cells. Self-blinking dyes offer an alternative by facilitating spontaneous ON/OFF switching through reversible intramolecular spirocyclization based on pH-dependent [26,27]. These dyes simplify single-molecule imaging by removing the need for complex photoactivation, thereby enhancing imaging simplicity and duration. The first self-blinking dye, HMSiR, synthesized by Urano et al., has a pKcycl of 5.8 and has been used for repetitive time-lapse super-resolution imaging of microtubules in live cells for up to 1 h [28]. Subsequently, various environment-sensitive membrane probes based on HMSiR have been developed to image multiple structures with high spatiotemporal resolution [29,30]. Nonetheless, current self-blinking dyes often face limitations, including nonspecific blinking and insufficient target specificity, which can introduce background noise and false localization signals.
To address these challenges, we developed a blinkogenic probe exhibiting self-blinking activation upon molecular recognition and excellent photostability (Fig. 1a and Fig. S1 in Supporting information). This probe has been successfully applied for long-term super-resolution imaging of various cellular structures, including the nucleus and mitochondria, demonstrating its capability for high-precision SMLM imaging. In this study, we extend the use of this blinkogenic probe, HM-DS655-Halo, to super-resolution imaging of pseudopodia. The target-activated blinking of the probe enabled precise tracking of pseudopodia dynamics in live cells. By leveraging specific self-blinking activation with a Lifeact-HaloTag fusion protein, we monitored real-time growth and fusion of filopodia, as well as cell migration and shrinkage. Additionally, we identified two distinct fusion modes between filopodia and lamellipodia, and decoded the formation and interactions of TNTs with filopodia at the nanoscale (Fig. 1b). This work highlights the application of our probe in capturing real-time changes in pseudopodia, introducing a new method for dynamic super-resolution imaging of cellular structures undergoing rapid and complex morphological transformations.
|
Download:
|
| Fig. 1. (a) Schematic illustration of blinkogenic probes. (b) SMLM imaging of cellular pseudopodia dynamics with blinkogenic probes. | |
To verify the low background and ensure accurate localization, we first conducted SMLM imaging of various intracellular structures using blinkogenic probes. Through three dimensions (3D)-SMLM imaging of the nucleus (Fig. 2a), we were able to identify individual H2B proteins with a localization precision of 25.4 nm (Figs. 2a–c and Fig. S2 in Supporting information). The super-resolution image clearly delineated the outline of the nucleolus, highlighting the specificity and low blinking background of our probes. Additionally, we achieved SMLM imaging of the nucleolus by localizing nucleophosmin 1 (NMP1)-HaloTag, revealing multiple fibrillar center (FC)-formed empty structures with diameters ranging from 300 nm to 600 nm (Fig. S3 in Supporting information). The highly dynamic outer mitochondrial membrane was also resolved using HM-DS655-Halo labeled on TOMM20 (Figs. 2d and e). An average of 1022 photons were detected, yielding a localization precision of 17.8 nm (Fig. 2f, Figs. S4 and S5 in Supporting information). Mitochondrial fusion was observed as indicated by the white arrow in Fig. S6 (Supporting information), where two mitochondria merged into a single structure. The actin structure, reconstructed from 10,000 frames, also revealed clear microfilaments with a high localization precision of 19.1 nm and 830 photons per event (Figs. 2g–i and Fig. S7 in Supporting information). SMLM imaging revealed a filopodia width of 72.5 nm, compared to the 444 nm width observed with widefield imaging (Fig. 2h), further demonstrating the high blinkogenicity and localization precision of our probes.
|
Download:
|
| Fig. 2. Super-resolution imaging of cellular structures in live HeLa cells stained with HM-DS655-Halo (50.0 nmol/L). (a, b) 3D-SMLM imaging of nucleus marked with H2B-HaloTag fusion protein. (c) Histograms of localization accuracy per single molecule per frame in (a). (d, e) SMLM imaging of mitochondria marked with TOMM20-HaloTag fusion protein. (f) Histograms of localization accuracy per single molecule per frame in (d). (g) SMLM imaging and wide-field imaging of actin marked with Lifeact-HaloTag fusion protein. WF, wide field. (h) Intensity distributions across microfilament (green line) in (g). (i) Histograms of localization accuracy per single molecule per frame in (g). Scale bar: 2.0 µm. | |
Encouraged by the blinkogenicity of our probe, we focused on deciphering the dynamic behavior of cellular pseudopodia, which are highly active during cell migration, infection, and intercellular communication. The growth of filopodia was firstly captured. The filopodium labeled by the cyan arrow continued to extend over 200 s at a nearly uniform speed of 0.005 µm/s (Figs. 3a and b). Additionally, we observed strong interactions between filopodia, particularly in the form of head-to-head contacts. The filopodium marked with a pink dotted line exhibited notable bending during interaction to maintain prolonged contact. To quantify this bending, we measured the length of this pseudopod (R1) and the straight-line distance (R2) from the contact site to the pseudopod's origin, using the R1/R2 ratio to assess pseudopod curvature. Before 100 s, the filopodium labeled by the pink dotted line remained in a relaxed state with consistent R1 and curvature values (Fig. 3c and Fig. S8 in Supporting information). After 120 s, however, the curvature increased from 1.08 to 1.41 at a rate of 0.0033 s–1, suggesting intense interactions that could facilitate signal transduction or material exchange between filopodia. We also tracked two sequential fusion events among three neighboring filopodia in real-time. The filopodium marked by the pink arrow began moving toward a middle filopodium, initiating the first fusion event at 80 s (Fig. 3d). This newly generated filopodium continued moving toward the third filopodium, with a second fusion occurring at 160 s. Notably, prior to the second fusion, the two filopodia engaged in a sustained head-to-head interaction for nearly 80 s.
|
Download:
|
| Fig. 3. Cellular pseudopodia dynamics observed with SMLM imaging. (a) Time-colored SMLM image of filopodia and the contact mode between two filopodia. Time dependence of filopodia growth labeled by cyan arrow (b) and the curvature of filopodia along pink dotted line (c). (d) Time-colored SMLM image of filopodia and the sequential fusion mode. Time-colored SMLM image of cell migration (e) and shrinkage (g). (f) Time dependence of cell migration area. (h) Time dependence of area in pink dotted box. Scale bar: 2.0 µm. | |
In contrast to filopodia, lamellipodia are primarily responsible forward. They also mediate cell-extracellular matrix adhesion byadhesions to stabilize cell positioning. Using blinkogenic probes, we also monitored real-time dynamic behaviors of lamellipodia, such as cell migration and shrinkage. Time-colored SMLM imaging, where red indicates the initial time and blue the end time, revealed the migration process (indicated by an arrow in Fig. 3e) with the cell edge, marked by pink dotted lines, advancing along the green arrow. We measured the area of the region of interest over 200 s, observing that cell migration began at 40 s and ended at 150 s (Fig. 3f), with a peak migration rate of 0.016 µm2/s. Additionally, the SMLM image in Fig. 3g depicted the cell shrinkage process in three directions, marked by a reddish hue at the cell edge. We measured the rate of area reduction within a dashed box, observing that unlike migration, the cell initially shrank at a nearly uniform rate of 0.015 µm2/s before stabilizing around 100 s (Fig. 3h). Briefly, our blinkogenic probes demonstrated remarkable performance in dynamic SMLM imaging of cell protrusions, effectively capturing and monitoring the formation, fusion, and interactions of filopodia, as well as lamellipodia migration and shrinkage. The interaction between filopodia and lamellipodia plays a vital role in cellular migration and morphological dynamics. By leveraging their distinct structures and functions, these cellular extensions work together to enhance motility and shape changes. Through long-term SMLM imaging, we identified two distinct patterns of filopodia fusing with lamellipodia (Figs. 4a and b). In pattern 1, a filopodium establishes a contact point with a lamellipodium, followed by a gradual fusion process. While in pattern 2, a filopodium may first bend overall and approach the lamellipodium before integrating and fusing with it. Fig. 4c illustrated this process, where the filopodium indicated by the orange arrow forms a fusion site (labelled by the cyan arrow) between 40 s and 60 s. The two structures gradually merged, completing the fusion by 100 s, exemplifying the characteristics of pattern 1. Fortunately, we observed an alternative fusion mode within the same field of view. The filopodium marked by the pink arrow began to bend at 60 s and moved closer to the cell edge. Fusion occured between 140 s and 160 s, resulting in a protrusion at the cell edge, outlined by the pink dotted line.
|
Download:
|
| Fig. 4. (a) Time-colored SMLM image of fusion between filopodia and lamellipodia. (b) Two distinct patterns of filopodia fusing with lamellipodia. (c) Time series SMLM images of fusion between filopodia and lamellipodia. Scale bar: 2.0 µm. | |
TNTs are actin-rich structures that typically emerge from double filopodial bridges (DFBs) and play a crucial role in intercellular communication and material transport. Our SMLM imaging revealed multiple filopodia extending between adjacent cells, displaying highly dynamic behaviors (Fig. 5a). In Fig. 5b, we illustrated the process of TNT formation through the fusion of filopodia and their subsequent movement along lamellipodia. Two filopodia, indicated by green dotted lines at 0 s, gradually approached each other and merged into a single filopodium at 60 s. This newly formed filopodium then interacted with the lamellipodium of cell-1, establishing a connection that leads to the formation of TNT. Following this, the newly formed TNT gradually moved to the right along the lamellipodium, with the angle between it and the adjacent TNTs, marked by green arrows, progressively increasing. Moreover, we observed a dense network of TNTs interconnecting multiple cells and captured the interactive processes between filopodia and TNTs (Fig. 5c). Notably, the filopodium labeled with the pink dotted arrow gradually elongated from 0.66 µm to 1.76 µm and briefly interacted with the TNTs at 120 s, highlighting the role of filopodia in both TNT formation and dynamic interactions within this complex cellular network (Fig. 5d).
|
Download:
|
| Fig. 5. (a) SMLM imaging of neighboring intercellular communication. (b) Time-colored and time series SMLM images of filopodia between two cells of ROI in (a). (c) SMLM imaging of dynamics between multiple cellular pseudopods. (d) Time-colored and time series SMLM images of filopodia between two cells of ROI in (c). Scale bar: 2.0 µm. | |
In conclusion, our blinkogenic probe, showing self-blinking activation upon molecular recognition, has enabled long-term SMLM imaging of diverse cellular structures. The excellent blinkogenicity allowed for accurate tracking of pseudopodial dynamics in live cells, including filopodia, lamellipodia as well as TNTs. The growth of filopodia was recorded, showing a speed of 0.005 µm2/s. Besides, highly curved filopodia was found which demonstrated the intense “head to head” contact between two filopodia. SMLM imaging of lamellipodia provided insights into cell migration and shrinkage, capturing real-time movement rates. We also had identified two distinct fusion modes between filopodia and lamellipodia: the first involved the formation of a single contact site followed by fusion, while the second entailed overall bending of filopodia to approach lamellipodia before fusing. Additionally, we successfully utilized this method to track the formation of TNTs and their interactions with filopodia. This work demonstrates the potential of our probe to capture real-time changes in pseudopodia, establishing a novel approach for dynamic super-resolution imaging of cellular structures that undergo rapid and complex morphological transformations.
Declaration of competing interestThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
CRediT authorship contribution statementAoxuan Song: Formal analysis, Data curation. Qinglong Qiao: Writing – original draft, Supervision, Funding acquisition, Formal analysis, Conceptualization. Ning Xu: Data curation. Yiyan Ruan: Data curation. Wenhao Jia: Data curation. Xiang Wang: Data curation. Zhaochao Xu: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.
AcknowledgmentsThis work is supported by the National Natural Science Foundation of China (Nos. 22225806, 22078314, 22278394, 22378385) and Dalian Institute of Chemical Physics (Nos. DICPI202227, DICPI202436).
Supplementary materialsSupplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110643.
| [1] |
P.K. Mattila, P. Lappalainen, Nat. Rev. Mol. Cell Biol. 9 (2008) 446-454. DOI:10.1038/nrm2406 |
| [2] |
E. Génot, B. Gligorijevic, Eur. J. Cell Biol. 93 (2014) 367-379. |
| [3] |
J. Su, Y. Song, Z. Zhu, et al., Signal Transduct. Target. Ther. 9 (2024) 196. |
| [4] |
N. Leijnse, Y.F. Barooji, M.R. Arastoo, et al., Nat. Commun. 13 (2022) 1636. |
| [5] |
T. Nishimura, T. Oyama, H.T. Hu, et al., Dev. Cell. 56 (2021) 842-859. |
| [6] |
N. Leijnse, L.B. Oddershede, P.M. Bendix, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) 136-141. DOI:10.1073/pnas.1411761112 |
| [7] |
M. Krause, A. Gautreau, Nat. Rev. Mol. Cell Biol. 15 (2014) 577-590. DOI:10.1038/nrm3861 |
| [8] |
M. Chang, O.C. Lee, G. Bu, et al., Sci. Adv. 8 (2022) eabj3995. |
| [9] |
J. Li, Q. Qiao, Y. Ruan, et al., Chin. Chem. Lett. 34 (2023) 108266. |
| [10] |
M. Nozumi, Y. Sato, M. Nishiyama-Usuda, et al., J. Neurochem. 168 (2024) 2974-2988. DOI:10.1111/jnc.16162 |
| [11] |
F. Hu, D. Zhu, H. Dong, et al., iScience 25 (2022) 100514. |
| [12] |
Q. Qiao, W. Liu, W. Chi, et al., Aggregate 4 (2023) e258. |
| [13] |
W. Zhou, Y. Tao, Q. Qiao, et al., ACS Sens. 9 (2024) 3170-3177. DOI:10.1021/acssensors.4c00486 |
| [14] |
K. An, Q. Qiao, W. Zhou, et al., Anal. Chem. 96 (2024) 5985-5991. DOI:10.1021/acs.analchem.4c00342 |
| [15] |
W. Zhou, Q. Qiao, Y. Tao, et al., Sens. Actuat. B: Chem. 376 (2023) 132980. |
| [16] |
Z. Li, Q. Qiao, N. Xu, et al., Chin. Chem. Lett. 35 (2024) 108824. |
| [17] |
J. Chen, W. Liu, X. Fang, et al., Chin. Chem. Lett. 33 (2022) 5042-5046. |
| [18] |
B. Turkowyd, D. Virant, U. Endesfelder, Anal. Bioanal. Chem. 408 (2016) 6885-6911. DOI:10.1007/s00216-016-9781-8 |
| [19] |
T. Chen, B. Dong, K. Chen, et al., Chem. Rev. 117 (2017) 7510-7537. DOI:10.1021/acs.chemrev.6b00673 |
| [20] |
Q. Qiao, W. Liu, Y. Zhang, et al., Angew. Chem. Int. Ed. 61 (2022) e202208678. |
| [21] |
S. Culley, D. Albrecht, C. Jacobs, et al., Nat. Method. 15 (2018) 263-266. DOI:10.1038/nmeth.4605 |
| [22] |
W. Chi, D. Tan, Q. Qiao, et al., Angew. Chem. Int. Ed. 62 (2023) e202306061. |
| [23] |
Q. Qiao, W. Liu, J. Chen, et al., Angew. Chem. Int. Ed. 61 (2022) e202202961. |
| [24] |
W. Liu, J. Chen, Q. Qiao, et al., Chin. Chem. Lett. 33 (2022) 4943-4947. |
| [25] |
M. Lelek, M.T. Gyparaki, G. Beliu, et al., Nat. Rev. Methods Primers. 1 (2021) 39. |
| [26] |
W. Chi, Q. Qiao, C. Wang, et al., Angew. Chem. Int. Ed. 59 (2020) 20215-20223. DOI:10.1002/anie.202010169 |
| [27] |
N. Lardon, L. Wang, A. Tschanz, et al., J. Am. Chem. Soc. 143 (2021) 14592-14600. DOI:10.1021/jacs.1c05004 |
| [28] |
S.n. Uno, M. Kamiya, T. Yoshihara, et al., Nat. Chem. 6 (2014) 681-689. DOI:10.1038/nchem.2002 |
| [29] |
H. Takakura, Y. Zhang, R.S. Erdmann, et al., Nat. Biotechnol. 35 (2017) 773-780. DOI:10.1038/nbt.3876 |
| [30] |
L.A. Chu, C.H. Lu, S.M. Yang, et al., Nat. Commun. 10 (2019) 4762. |
2025, Vol. 36 

