DNA mismatch repair (MMR)

DNA MMR processes: I Mismatch Recognition: MSHs search and recognize the mismatch. ➔ II Mismatch Signal Transmission: Mismatch finding is transmitted to a downstream site by MSH and MLH/PMS, for which ATP is required. ➔ III Strand Excision: Strand excision is initiated at a 5′-end (5′-directed strand excision) or a 3′-end (3′-directed strand excision) to release the mismatch. ➔ IV DNA Resynthesis: After the mismatch is eliminated, DNA polymerase fills the gap and Ligase joins two ends of ssDNAs. Components participating in each MMR process are listed on the table^[1]. The concept of mismatch repair (MMR) was formulated independently in 1964 to explain the removal of brominated nucleotides from DNA as well as gene conversion during genetic recombination. In the intervening 40 years, the field has developed incrementally, punctuated by a number of transformative genetic and biochemical studies. Two core MMR genes, MutS and MutL, have been conserved throughout life on earth. Defects in human MutS homologues (MSH) and MutL homologues (MLH/PMS) cause the common cancer predisposition Lynch syndrome or hereditary nonpolyposis colorectal cancer (LS/HNPCC). Work on the mechanism of MMR has been significantly aided by completely defined biochemical systems in vitro as well as several crystal snapshots that depict critical intermediates. It has been mired by unseemly biochemical conditions and misinterpretation. The contemporary use of real-time single molecule imaging has the potential to finally and fully resolve the mechanics of MMR^[1].

References

Tunneling nanotube

As an intercellular communication, tunneling nanotubes (TNTs) or membrane nanotubes are submicrometer thin routes that directly connect distant cells. TNTs mainly originate from plasma membrane protrusions containing filamentous actins (F-actins), but microtubule- or cytokeratin filament–associated TNTs have also been found. Organelles, vesicles, viruses, morphogens, RNAs, and receptors can be transported through this unique structure in various types of cells, including neuronal, immune, cancerous, epithelial, and stem cells, as well as in Drosophila tissue. Moreover, a recent study on primary cancer cells showed the pathological importance of TNT formation in tumor metastasis. TNTs may play an important role in long-distance communication between cells for human disease and immune systems. There are two types of Tunneling nanotube, being closed-end that connected by gap junctions and membrane open-end. In our lab, suggest the model about formation mechanism of closed-end TNT, in 2022.^[1] With HILO (Highly inclined and laminated optical sheet) microscopy, we can detect double filopodial bridges (DFBs) structure in single cell level. Also with super-resolution imaging technics, STORM and SRRF, the structure of DFBs is twisted in double helical structure. We suggest this torsional and shearing energies are important for DFB-to-TNT transition.

Translation initiation

Our lab is broadly interested in understanding how the structural dynamics of mRNA influence translation regulation. A growing body of evidence suggests that mRNA does not maintain a fixed “closed-loop” configuration during translation. Instead, the interactions between the 5' and 3' ends appear to be transient and dynamically regulated, reflecting a more flexible model of translation initiation and ribosome recycling^[2]. This perspective emphasizes the importance of studying mRNA structure and behavior in real time, as such dynamics can significantly impact translation efficiency and gene expression. By combining imaging techniques and molecular analysis, we aim to explore how mRNA–protein interactions and conformational changes contribute to post-transcriptional regulation in diverse biological contexts.

References

TIRFM (Total Internal Reflection Fluorescence Microscopy)

Total Internal Reflection Fluorescence (TIRF) microscopy is an advanced imaging technique that enables highly sensitive observation of fluorescent molecules near a surface. It is based on the optical phenomenon of total internal reflection, which occurs when light travels from a medium with a higher refractive index to one with a lower refractive index at an angle greater than the critical angle. Under these conditions, the light does not pass through the interface but is entirely reflected. Importantly, this reflection generates an evanescent wave—a near-field electromagnetic field that penetrates only ~100–200 nanometers into the lower refractive index medium. This shallow excitation field allows TIRF microscopy to selectively excite fluorophores located very close to the glass-water interface, dramatically reducing background fluorescence from molecules outside the evanescent field. As a result, the signal-to-noise ratio is significantly enhanced, making TIRF microscopy particularly powerful for observing single molecules, membrane-associated processes, and dynamic events at or near the cell surface.

FRET (Förster Resonance Energy Transfer)

Förster Resonance Energy Transfer (FRET) is a mechanism describing energy transfer between two light-sensitive molecules. A donor chromophore, initially in its electronic excited state, may transfer energy to an acceptor chromophore through nonradiative dipole–dipole coupling. The efficiency of this energy transfer is inversely proportional to the sixth power of the distance between donor and acceptor, making FRET extremely sensitive to small changes in distance. At the single-molecule level, FRET enables the observation of molecular heterogeneity and dynamic processes that are often hidden in ensemble measurements. By tracking FRET efficiency changes in real time, researchers can monitor conformational fluctuations, binding kinetics, and reaction pathways of individual biomolecules, providing unparalleled insights into their function and mechanisms^{[1] [2]}.

Super Resuolution Imaging

PALM (Photoactivated Localization Microscopy) and STORM (Stochastic Optical Reconstruction Microscopy) are super-resolution imaging techniques that overcome the diffraction limit of conventional light microscopy. Both methods rely on the precise localization of individual fluorophores that are stochastically activated and imaged over time. In PALM, photoactivatable fluorescent proteins are used, allowing controlled activation of a sparse subset of molecules, which are then imaged and localized with nanometer precision before being photobleached. STORM, on the other hand, typically uses synthetic fluorophores that can switch between dark and fluorescent states, enabling repeated imaging cycles. By compiling the positions of thousands of individual molecules, both PALM and STORM reconstruct high-resolution images of cellular structures with spatial resolutions down to ~20 nm, making them powerful tools for studying molecular organization at the nanoscale^[3].

Flow-Stretching Bead Assay

Single-molecule flow-stretching bead assay uses a well-defined drag force onto a micron-size super paramagnetic bead by regulated laminar flow. DNA molecules were immobilized on a surface of coverslip by one terminus and on the magnetic bead by the other terminus. DNA was stretched along the flow direction. When a protein binds to and diffuses along the DNA, the position and dynamics of the protein can be monitored in real time. This assay can monitor protein-DNA interactions such as diffusion, sliding, binding, pausing, and dissociation of single proteins^[4].

DNA Skybridge

Surface-condition‑independent single‑molecule imaging platforms have recently achieved remarkable advances, exemplified by the DNA skybridge system^[5]. This technique constructs a three-dimensional array of ~4 µm-high quartz barriers, anchoring each end of DNA molecules at the top of adjacent posts. Such a 3D configuration keeps the substrate surface out of focus during imaging and inherently generates a thin Gaussian light sheet aligned parallel to the stretched DNA. As a result, researchers can directly observe individual proteins or complexes diffusing along the DNA, with minimal background interference from the surface. The combination of extended DNA immobilization, suppression of surface noise, and sheet-like excitation enables high-throughput, high-resolution studies of DNA–protein interactions, mapping binding events, enzymatic activity, and transport dynamics at the single-molecule level. In contrast, the DNA Hanger assay focuses on enhancing sensitivity and specificity in single-molecule immunoassays ^[6]. By deploying a surface-minimized 3D scaffold—similar in spirit to the skybridge concept—it reduces nonspecific adsorption without requiring extensive blocking steps. This simplified format accelerates workflows and achieves sub-picomolar detection limits, thus offering a robust platform for low-abundance biomarker detection. The core innovation lies in minimizing the contact area between antibody-labeled entities and the surface, effectively suppressing background binding and increasing signal-to-noise ratio. This makes it particularly useful for rapid, sensitive assays in complex biological samples.

Deep Learning for Single-Molecule Fluorescence Analysis

Our lab leverages deep learning to push the boundaries of single-molecule fluorescence imaging. We have developed smDeepFLUOR^[7], a deep learning framework designed to classify molecular states by analyzing subtle spatiotemporal emission patterns of single fluorophores. Using 3D convolutional neural networks, smDeepFLUOR can accurately distinguish between different types of molecular interactions—even those with nearly identical intensity profiles—based on data acquired from total internal reflection fluorescence (TIRF) microscopy. One of the key innovations in our approach is the ability to analyze unlabeled proteins with high accuracy. This capability enables fluorescence-based molecular classification without the need for site-specific labeling, preserving native biological functionality and making the method broadly applicable to minimally perturbed systems. Looking ahead, we aim to extend this framework through transfer learning and self-supervised training to adapt to new molecular targets with limited labeled data. Future work will also explore incorporating z-stack imaging and spectrally multiplexed data to enhance chemical specificity and structural resolution. Through this work, we demonstrate how deep learning can transform single-molecule imaging into a powerful, label-efficient platform for exploring dynamic and complex molecular phenomena.

References