Single Molecule Spectroscopy and Superresolution Imaging XVII

Spectroscopic single-molecule localization microscopy (sSMLM) combines super-resolution microscopy and spectroscopy. Its single molecule sensitivity and high spectral precision have made it uniquely valuable for several applications, including multicolor imaging, chemical characterization, polarity sensing, and multiplexed single particle tracking. However, widespread adoption is hampered by a lack of standardization in optical implementation, calibration techniques, and image processing. We demonstrate our lab’s efforts to develop tools that simplify adoption and optimize photon efficiency, including protocols for calibration techniques, a user-friendly imageJ plugin for image processing, and a fabricated monolithic beam splitter and prism designed to fit into a microscope body with minimal optical alignment.

The ability to measure fluorophore molecular orientation and mobility can provide valuable information on the local physical and chemical environment. Analysis of polarized images can determine the orientation and mobility of fluorophores with well-defined uncertainties and bias, as well as enabling the use of two to four times greater emitter density per image frame than PSF engineering methods do. This study presents extensive Monte Carlo simulation and experimental data to determine quantitatively the degree of coupling among orientations and orientational mobility, and the dependence of orientational uncertainty on photon count and background noise. These results may be used for the rational design of experimental protocols and conditions to yield the levels of precision and accuracy necessary to effectively explore a wide range of physical phenomena.

The retinal pigment epithelium (RPE) is a monolayer of pigmented cells critical for sight and any intracellular damage in this monolayer will compromise the integrity of the entire eye, leading to blindness and other health problems. Super-resolution microscopy provides an opportunity to image damaged and healthy RPE tissue down to the molecular scale. Previous studies are limited in scope, relying on cultured RPE cells due to the difficulties in imaging tissue slices. Here, we report the first super-resolution imaging of flat-mounted whole albino mouse retinal pigment epithelium using single-molecule localization microscopy (SMLM). After optimizing the labeling protocols, we imaged Phalloidin, β-Tubulin, ZO-1 Tight Junctions, Peroxisome marker (PMP70), Histone-H4, and TOM20 mitochondrial proteins both separately and in simultaneous two-color spectroscopic SMLM (sSMLM) imaging of Phalloidin and ZO-1 as well as TOM20 and β-Tubulin. This work lays the foundation for future investigations of multiple intracellular interactions within damaged RPE monolayers at the molecular level.

We present a novel technique for volumetric super-resolution imaging. Our technique, which is based on the principles of single molecule localization microscopy, utilizes a mirror cavity with a series of pinholes on one of the mirrors allowing for simultaneous optical sectioning of different imaging planes. In addition, we employ a unique machine learned algorithm for 3D localization of events that occur between different imaging planes. Our technique enables high-resolution imaging of thicker volumes than what is currently available using other single molecule localization techniques.

Stochastic optical reconstruction microscopy (STORM) achieves super-resolution imaging by blinking individual dye molecules in thiol-containing media. While STORM is well-established for imaging thin biological specimens, its application to thick tissues has been limited by light scattering. Mounting media with an oil refractive index have shown promise in improving imaging depth and resolution in optical microscopy, but such buffers have not been explored for STORM. Here, we present a 3-pyridinemethanol-based STORM buffer with a refractive index matching standard immersion oil. Our buffer demonstrates comparable performance to conventional STORM buffers and exceptional stability for 5 weeks, enabling imaging of numerous cells on a single slide and larger field-of-view imaging. With perfect index matching, our buffer simplifies imaging and holds potential for lightsheet STORM applications in thick tissues.

We present our latest results on a structured illumination microscope (SIM) implementation using individual microelectromechanical systems (MEMS) micromirrors with three-axis full angular, radial and phase control of the illumination pattern in the sample. Results of a simultaneous multi-colour 2D SIM and 3D SIM implementation are shown with digital system adjustment to select the optimal imaging conditions and adapt to variable microscope objectives used in the system. Calibration and cell images of 2D and 3D samples are shown to verify the resolution enhancement and axial sectioning potential.

Fluorophore labels that transiently and repetitively bind to a target (“exchangeable” or “renewable” labels) lead to a continuous renewal of the fluorescence signal. This dynamic labeling approach minimizes photobleaching and was beneficially exploited in various super-resolution microscopy methods. Here, we report two new developments using exchangeable fluorophores: first, we report fast and long-time live-cell super-resolution microscopy using a weak-affinity protein label and a neural network. Second, we report a novel design for exchangeable DNA labels that show higher brightness and lower background. Together, these two developments further increase the application range for exchangeable fluorophore labels in super-resolution fluorescence microscopy.

Fluorescence laser-scanning microscopy (FLSM) is a widely utilized tool in life-science research. In recent years, this technique has undergone a profound transformation, thanks to the introduction of novel single-photon avalanche diode (SPAD) array detectors. This study reveals the exciting possibilities of combining the SPAD array detector with single-molecule techniques. We propose a real-time single-molecule tracking architecture, where the SPAD array effortlessly localizes the molecule of interest, and the beam scanning architecture effectively maintains the molecule at the center of the microscope's detection volume. This approach enables comprehensive three-dimensional tracking throughout the entire cell, offering valuable insights into molecular nano-environments, interactions, and structural changes through fluorescence lifetime information. Furthermore, utilizing the same FLSM system, we present a novel sequential structure illumination single-molecule localization microscope (similar to MINFLUX). This advanced technique achieves localization precision in the few-nanometer range while simultaneously providing the molecule's fluorescence lifetime.

Self-Interference Digital Holography (SIDH) enables imaging of incoherently emitting objects over large axial ranges with sub-diffraction resolution in all three dimensions. SIDH can be combined with Single-Molecule Localization Microscopy (SMLM) for 3D imaging over a large axial range with nanometer precision. Because SIDH captures the phase of the light field, aberrations are recorded in the hologram. Here, we report on the development of computational aberration correction for SIDH, capable of correcting optical aberrations over a large axial range. Our sensorless wavefront correction method can be widely applied for SMLM imaging to achieve super-resolution imaging and three-dimensional particle tracking.

Oligomers of disease-causing amyloid proteins (such as Alzheimer’s Amyloid beta or ’Aβ’) are generally amphiphilic and their interactions with lipid membranes are possibly the origin of their toxicity. However, how oligomers of different stoichiometries or different mutants differ in their interaction with the membrane, and how these differences correlate with their toxicity, has largely remained beyond the reach of existing experimental techniques. Here we use Q-SLIP, a single-molecule tool that can resolve the surface exposure of different parts of individual oligomers, and different radical-labeled lipids that act as quenchers, to address these questions.

Defect engineering of 2D materials offers enormous opportunities to tune material properties. This presentation will show two types of substitutional defects in 2D materials, self-limited along the out-of-plane and in-plane directions, respectively. The first type is atomic substitution: a nitrogen atom substituting a chalcogen atom in 2D transition metal dichalcogenides (TMDs), which yields new distinct photoluminescence features well separated from the free excitons of 2D TMDs. The second type is layer substitution: an entire layer of chalcogen atoms in 2D TMD substituted by another type of chalcogen atoms, namely, Janus TMDs. Due to the intrinsic vertical dipole, Janus TMDs form unconventional interaction with adjacent materials including other 2D material layers. These unconventional interactions were probed by optical signature changes such as ultra-low frequency Raman modes and photoluminescence yield change. Engineering such substitutional defects in 2D materials promises potential for optoelectronic devices and quantum information platforms.

The development of super-resolution imaging techniques have extended the resolving power to around 10 - 50 nm. However, most current super-resolution imaging techniques need exogenous fluorescent dyes as imaging contrast, whose essential weakness of labeling includes imprecise spatial localization, and perturbation of the sample. In 2017, Dong et al., demonstrated that the intrinsic fluorescence of DNA under visible light excitation has similar photo-switching properties to the organic dyes used in single-molecule localization microscopy. In this paper, we measured the fluorescence spectra of poly-G (guanine) of different lengths (5, 8, 12, 16 base-pair), 20 base-pair single-stranded DNA molecules (poly-A, G, C, T), as well as double-stranded DNA (AT chain, GC chain), under multiple wavelengths. The 20 base-pair AT, GC spectra can be classified with an accuracy of more than 90%, which demonstrates the molecular specificity of the double stranded DNA polymers via its intrinsic fluorescence. Our work paves the way for developing spectroscopic intrinsic-contrast localization optical nanoscopy for chromatin study.

G-quadruplexes (GQs) are essential guanine-rich secondary structures found in DNA and RNA, playing crucial roles in genomic maintenance and stability. Recent studies have unveiled GQs in the intergenic regions of the E. coli genome, suggesting their biological significance and potential as anti-microbial targets. Here, we investigated the interaction between single-stranded DNA (ssDNA) containing multiple GQ-forming sequences and E. coli SSB with varying lengths of "sticky ends." Using multiple techniques, we explored E. coli SSB binding to ssDNA and the structural changes of these oligonucleotides upon protein binding. Additionally, we utilized smFRET to probe the conformational changes of GQ-ssDNA structures upon SSB binding. Our results provide detailed insights into how SSB gains access to various GQ-ssDNA and how this homo-tetrameric protein wraps GQ-ssDNA in multiple distinct binding modes. These findings shed light on the molecular mechanism underlying biological processes involving DNA, offering potential applications in understanding and targeting these interactions for future research and antimicrobial strategies.

Integration Host Factor (IHF) is a nucleoid-associated protein involved in various DNA processes, including DNA packaging, viral DNA integration, and recombination. IHF binds to duplex DNA with a 13 bp consensus sequence, inducing a significant bend upon binding. While Wild-type IHF (WtIHF) facilitates gene regulation and foreign DNA integration into the host genome, the engineered version, Single-chain IHF (ScIHF), has genetic engineering and biotechnological applications. Our study investigated IHF's interaction with Holliday junctions (HJ) crucial for homologous recombination. We found high-affinity binding of WtIHF and ScIHF to HJs in the presence of consensus sequences, indicating a structure-based recognition mechanism. Using microscale thermophoresis, circular dichroism and Single molecule Förster resonance energy transfer (SmFRET), we determined the impact of both proteins on the dynamicity of HJ conformations. Our results reveal that IHF stabilizes the open conformation of the junction, suggesting a role in facilitating branch migration and strand exchange during DNA repair and recombination processes.

In localization microscopy, the positions of individual nanoscale point emitters (e.g. fluorescent molecules) are determined at high precision from their point-spread functions (PSFs). This enables highly precise single/multiple-particle-tracking, as well as super-resolution microscopy, namely single molecule localization microscopy (SMLM). In this talk I will describe advances to localization microscopy that we have recently achieved using deep learning, both in analysis (image processing) and in optimal imaging-system design. Specific topics to be discussed include: volumetric (3D) SMLM and single particle tracking by deep-learning-based PSF engineering, high-throughput in-flow colocalization in live cells, dynamic SMLM (blinking-to-video), and optical genome mapping. A novel method for additive-manufacturing of phase masks for wavefront shaping will also be discussed.

In recent years, Image Scanning Microscopy (ISM) has emerged as a powerful technique for achieving super-resolution bio-imaging across various applications. Particularly noteworthy is the implementation of a single-photon detector array, enabling the utilization of Lifetime Image Scanning Microscopy, which has proven to be highly effective. In our study, we present a novel approach that combines ISM with direct Stochastic Optical Reconstruction Microscopy (dSTORM), resulting in a doubling of the localization precision in Single Molecule Localization Microscopy (SMLM). Additionally, we capitalize on the available lifetime information provided by ISM, allowing for multilabel fluorescence measurements without the detrimental effects of chromatic aberration, even at resolutions significantly surpassing the diffraction limit. Moreover, we introduce a freely available add-on to previously employed open-source tools for single particle tracking and localization, enhancing the accessibility and utility of our methodology. This add-on serves as a valuable resource for the research community, facilitating the adoption and further advancement of the combined ISM and dSTORM technique.

Simultaneous measurement of the 3D orientation and the 3D position of a single fluorescent molecule can be achieved by Point Spread Function (PSF) engineering. However, this 5D problem is complex to optimize and time consuming when solved with classical approaches. To overcome this problem, we developed a deep learning approach that allows us to obtain an optimized phase mask as well as an Analysis Neural Network that estimates the 5 parameters of immobilized single molecules with a reduced computation time. Our method shows an axial precision of about 30 nm and an orientation precision of about 10 degrees, and it can be applied to complex problems such as molecular orientation in membranes.

This study explores uncertainties in fluorescence labeling, a complication in Single-Molecule Localization Microscopy (SMLM) image interpretation. We examine variability caused by antibody and fluorophore attachment, orientation, and photobleaching, focusing on protein tagging and indirect immunofluorescence, techniques known for their specificity but prone to introducing variable label densities. We use a Monte Carlo (MC) model to simulate SMLM images, providing a 'ground truth' for comparison. This model also investigates the balance between labeling size and density, considering the possibility of single fluorophore attachment in protein tagging and multiple fluorophores in indirect immunofluorescence. We propose methods to quantify the effects of labeling strategies on image quality and accuracy, considering parameters such as labeling linker length and fluorophore photoswitching. Our work enhances the accuracy of SMLM image interpretation and guides the selection of labeling strategies, advancing super-resolution microscopy.

In order to study complex biomachinery processes at a single molecule level, it is paramount to develop new platforms capable of tethering the analytes of interest while keeping anti-fouling properties at high concentrations. This work explores the potential of fluorous brushes to overcome the limitations of the current methods. Following our previous work on mass photometry, where we reported virtually zero non-specific binding of proteins on fluorous substrates, we have used perfluorinated tails to tether DNA oligonucleotides onto fluorous brushes for TIRF measurements. This way, we have been able to characterise, for the first time, the dynamic nature of the fluorous-fluorous interactions, and have developed custom-made software to track the diffusion of these tethered molecules.

Electro-optic fluorescence lifetime imaging (EO-FLIM) uses resonantly-driven optical modulators to gate images at 40 MHz onto standard scientific cameras, allowing lifetime estimation at every image pixel in parallel. We demonstrate that EO-FLIM may be applied to wide-field imaging of a genetically encoded voltage indicator expressed in Drosophila neurons in vivo, achieving kilohertz frame rates and < 5 ps lifetime resolution. Action potentials and sub-threshold voltage features are studied in lifetime and are mapped throughout neuronal structures. Standard techniques rely on measuring ΔF/F to quantify neuron activity. FLIM instead allows measurement of an absolute change in fluorescence lifetime without normalizing to a moving average baseline. Since EO-FLIM measures a ratio of image intensities, it also provides significant improvements in signal-to-noise ratio and rejects technical noise and motion artifacts from voltage recordings. Next steps in EO-FLIM technology development will be shown including a light-sheet FLIM microscope and applications to functional imaging in mammalian cells.

Fluorescence Lifetime Imaging (FLIM) has become more attractive in recent years as it offers increased specificity in many assays as well as the possibility of multiplexing the read out of many markers with a small number of detectors. Here we present how FLIM modalities are implemented in Luminosa, the new single-photon counting confocal microscope by PicoQuant. Thanks to a dynamic binning format and GPU-based algorithms FLIM images of 1024x1024 can be analysed in a few seconds. The FLIM analysis workflow suggests the best fitting model based on statistical arguments and requires minimal user interaction making these modalities become accessible to new users who can then confidently start working with FLIM and incorporate it into their research toolbox combining the strengths of phasor plots with decay fitting.

Time Correlated Single Photon Counting (TCSPC) has been historically subject to the count-rate vs distortion tradeoff. Several attempts to work around this limitation have been reported in the literature, either based on multichannel systems or on post-processing correction algorithms. In this work, we’ll show how distortion can be avoided by exploiting additional information on the system status acquired during the whole experiment. We’ll provide evidence that a new research line can finally combine all the advantages of TCSPC with very high speed. Starting from on-field results, we’ll present the novel technique providing design guidelines for next-generation ultrafast TCSPC acquisition systems.

Time-correlated single-photon counting (TCSPC) allows to achieve picosecond-precision measurements for low-light signals. However, TCSPC suffers from pile-up distortion, constraining the acquisition rate to 1-5% of the laser rate. To overcome the issue, our research focuses on high-rate TCSPC methodologies: in 2017 we reported a hardware acquisition approach, that has been translated into a real system, guaranteeing low distortion at 32 Mcps. This talk provides an overview on the research project, and in particular on the two validation campaigns carried out in fluorescence and lidar measurements, and on our first on-field experiment, i.e. the application of the technique to a single-pixel camera.

Using a novel imaging device—NCam—fluorophore lifetime measurements can be captured simultaneously with wide-field microscopy methods. Because NCam records single-photon events with spatial and temporal information, the localization precision can be improved compared to camera-based imaging or scanning confocal instruments. We demonstrate this new imaging capability by examining the fluorescence behavior of quantum dots.

DNA nanotechnology enables the construction of large self-assembled structures by DNA origami. Here, we present several applications of how DNA nanotech can create functionality on the nanoscale that might alternatively be realized by the construction of very complicated microscopes. We show more than 1000fold fluorescence enhancement by DNA origami nanoantennas that is used for attomolar single-molecule detection of nucleic acids on simple portable optical systems We also show how molecular forces can be measured by a DNA origami force clamp without physical connection to the macrosocpic world. With DNA nanostructures, we finally demonstrate the benefits of using the fluorescence lifetime in superresolution measurements when combining graphene-energy-transfer with pMINFLUX superresolution microscopy and DNA PAINT.

Genome organization has been shown to regulate transcriptional activity and be involved in cell state transitions in both prokaryotes and eukaryotes. However, beyond the nucleosome level, the connection between external cues, genome reorganization, and cellular reprogramming, has been inconsistent and lacks structural details of chromatin remodeling. Our lab develops and utilizes super-resolution chromatin tracing which visualizes and describes the shape of tens to hundreds of chromosomal loci within single cells. In this presentation I will describe how we are now using this technology to explain the high degree of structural variability of the chromatin and how that may contribute to its function in a broad range of biological systems, from simple bacteria to human.

Enzymes are cellular protein machines using a variety of conformational changes to power fast biochemical catalysis. Our goal is to exploit the single-spin properties of the luminescent NV (nitrogen-vacancy) center in nanodiamonds to reveal the dynamics of an active enzyme. Specifically attached to the membrane enzyme FoF1-ATP synthase, the NV sensor will report the adenosine triphosphate (ATP)-driven full rotation of Fo motor subunits in ten consecutive 36° steps. Conformational dynamics are monitored using either a double electron-electron resonance scheme or NV- magnetometry with optical readout or using NV- relaxometry with a superparamagnetic nanoparticle as the second marker attached to the same enzyme. First, we show how all photophysical parameters like individual size, charge, brightness, spectral range of fluorescence and fluorescence lifetime can be determined for the NV- center in a single nanodiamond held in aqueous solution by a confocal anti-Brownian electrokinetic trap (ABEL trap). Stable photon count rates of individual nanodiamonds and the absence of blinking allow for observation times in solution exceeding hundreds of seconds.

Oblique plane microscopy-based single molecule localization microscopy (obSTORM) shows promise for super-resolution imaging in thick biological samples. However, the Gaussian point spread function (PSF) model used in previous studies limits imaging resolution and axial localization range in obSTORM due to poor fitting with actual PSF shapes. To overcome these limitations, we employed cubic splines to construct a more precise PSF model. This refined model enhances three-dimensional localization precision, improving obSTORM imaging of mouse retina tissues. It increases imaging resolution by approximately 1.2 times, enables seamless stitching of single molecules across optical sections, and doubles the sectional interval in volumetric obSTORM imaging by extending the usable section thickness. The cubic spline PSF model offers a promising approach for achieving faster and more accurate volumetric obSTORM imaging of biological specimens.

Single Molecule Localization Microscopy (SMLM) and its ability to resolve < 100 nm structures has generated an ever-growing demand in biomedical research. This technique is highly relevant when trying to gain better understanding of cellular machinery in infectious models. The Imaging and Modelling Unit led by C. Zimmer developed an open optical and computational method based on Zernike Optimised Localisation Approach (ZOLA) enabling 3D localization of single molecules using point spread function (PSF) modifiers in the detection path. This technique offers different performances and trade-off depending on the required application. This unique flexibility is relevant when dealing with various types of samples and models as those presented to an Imaging core facility. We will present how the Unit of technology and Services (UtechS) Photonic Bio Imaging (PBI), the imaging platform of the Institut Pasteur in Paris has conducted the technological transfer of ZOLA 3D from a research laboratory to a Bio Safety Level 2 (BSL2) ISO 9001 core facility. This will make flexible 3D super-resolution imaging accessible to a wide range of biological projects , including the study of pathogens.

Low exercise capacity is highly correlated with skeletal muscle dysfunction and metabolic disorders, such as obesity, diabetes, and cardiovascular disease. Analysis of frozen liver tissue by NAD(P)H and Lipofuscin FLIM and multiphoton autofluorescence microscopy showed that citosolic NADH mainly exists in bound form beside higher amounts of lipofuscin in an aged rat model of inborn low versus high capacity runners (LCR/HCR). NAD(P)H and Lipofuscin FLIM imaging might offer a sensitive, predictive approach to study early effects of metabolic syndrome and ageing.

Quantitative single molecule and time-resolved fluorescence techniques offer new insights into many samples from various research areas such as • dynamic structural biology • cellular mechanisms driven by phase separation • virology Using examples from these areas, I will show how single-molecule FRET (smFRET), fluorescence (cross-)correlation spectroscopy (F(C)CS), and time-resolved anisotropy can be combined to get a comprehensive picture of the sample under investigation. In order to get reliable, quantitative results, many factors like laser power, or spectral bleedthrough need to be considered. I will discuss these, together with ways to increase the accuracy and reproducibility of experiments, with a focus on smFRET studies. Ultimately, findings from in vitro studies need to be linked to studies in cells and tissues. I will showcase how this can be achieved by taking FCS into cells, and by going from smFRET to FRET imaging, and from time-resolved anisotropy to anisotropy imaging.

We present a novel, high-resolution design of FPGA-based Time-Correlated Single Photon Counting (TCSPC) and time-tagging electronics featuring multiple synchronized input channels. The triggering mechanism for this instrument can be initiated by either an edge trigger or a vertex-finding Constant Fraction Discriminator (CFD), ensuring compatibility with a wide range of single photon detectors. To optimize timing, such as for Superconducting Nanowire Single Photon Detectors (SNSPD), the inputs can be configured as edge triggers. Conversely, for superior performance with Hybrid Photodetectors (HPD) or Micro Channel Plates (MCP), they can be configured as vertex-finding Constant Fraction Discriminators (CFD). The presented time tagging electronics achieves an exceptional digital time resolution of 1 ps and a single channel timing uncertainty of 2 ps rms, coupled with an exceptionally short dead-time. This unique combination opens the door to pioneering high-resolution TCSPC applications and significantly enhances the precision of the well-established high-speed fluorescence lifetime imaging method, RapidFLIM.

SPADλ is a linear single-photon detector array with 320×1 single-photon avalanche diode (SPAD) pixels, featuring thermo-electric cooling for reduced noise. These SPADs offer a low dark count rate and wide detection spectrum. Equipped with microlenses, they achieve a peak photon detection efficiency of 45% at 520 nm. This system can count photons at 4 Gcps and provides time-tagging and time-gating for time-resolved detection. With 80 TDC channels, it achieves time-tagging precision averaging better than 130 ps full width half maximum (FWHM). Ideal for flow cytometry, fluorescence lifetime imaging (FLIM), and Raman spectroscopy applications.

Super-resolution microscopy has revolutionized visualizing sub-diffraction limit structures, but it mainly focuses on labeled reconstructions, leaving valuable unlabeled information untapped. Correlative microscopy combines different methods to capture distinct features and their relationships. However, co-registering features at different resolutions poses challenges. We developed an algorithm enabling correlation and co-registration between microscopy techniques. Validated using Partial Wave Spectroscopy (PWS) and Stochastic Optical Reconstruction Microscopy (STORM), we tested it on real mammalian cell (HeLa) images to explore the relationship between PWS-captured chromatin packing domains and STORM-visualized histone modifications. Combining super-resolution microscopy and PWS, our methodology unravels complex biological phenomena across resolution scales, enhancing our understanding of cellular structures and epigenetic regulation. With further development, correlative microscopy holds promise for comprehensive investigations into the intricate world of subcellular architecture.

Spatial distribution and relative localization of proteins in cells can be an essential clue for the health of an organism. Thus, capturing high-resolution images of living cells is a vital tool for diagnostics in medicine and biology. Fluorescence Lifetime Imaging Microscopy (FLIM) allows for distinguishing different fluorescent markers by their specific decay-time after excitation. This allows to discriminate up to three different markers within the same spectral detection band. Image Scanning Microscopy (ISM) provides twice the resolution of a confocal microscope by replacing the single-point detector by a multi-pixel detector, e.g. a emCCD, or sCMOS camera. Due to the robustness and relative simplicity of the method, an extension to realize fluorescence lifetime imaging in combination with ISM seems straight-forward. Combining ISM with FLIM (FL-ISM) enables super-resolution microscopy with fluorescence lifetime based multiplexing for live-cell and tissue imaging. Our results show simultaneous fluorescence lifetime multiplexing for up to six different structures in two spectral regions. At the same time, we are keeping acquisition times comparable to other FLIM techniques.

Accurate characterization of the microscopic point spread function (PSF) is crucial for high-performance localization microscopy (LM). Traditionally, LM assumes a spatially invariant PSF to simplify the modeling of the imaging system. However, for large field of view (FOV) imaging, it becomes important to account for the spatially variant nature of the PSF. Despite efforts in characterizing field dependence, such as interpolating Zernike-polynomial-based pupil functions, there is still a demand for an efficient and accurate method. In this work, we introduce a spatially variant 3D PSF generator (PPG3D), based on Principal Component Analysis, accompanied by a localizer for LM. Our simulations and experimental results demonstrate PPG3D’s effectiveness with various PSFs, including astigmatism, double-helix, tetrapod, etc. This enables, through single molecule localization microscopy, super-resolution imaging of mitochondria and microtubules with high fidelity over a large FOV. A comparison of PPG3D with three other PSF generators for 3D LM reveals a three-fold improvement in accuracy and operates approximately a hundred times faster.