DBlink: dynamic localization microscopy in super spatiotemporal resolution via deep learning.

Single-molecule localization microscopy (SMLM) has revolutionized biological imaging, improving the spatial resolution of traditional microscopes by an order of magnitude. However, SMLM techniques require long acquisition times, typically a few minutes, to yield a single super-resolved image, because they depend on accumulation of many localizations over thousands of recorded frames. Hence, the capability of SMLM to observe dynamics at high temporal resolution has always been limited. In this work, we present DBlink, a deep-learning-based method for super spatiotemporal resolution reconstruction from SMLM data. The input to DBlink is a recorded video of SMLM data and the output is a super spatiotemporal resolution video reconstruction. We use a convolutional neural network combined with a bidirectional long short-term memory network architecture, designed for capturing long-term dependencies between different input frames. We demonstrate DBlink performance on simulated filaments and mitochondria-like structures, on experimental SMLM data under controlled motion conditions and on live-cell dynamic SMLM. DBlink's spatiotemporal interpolation constitutes an important advance in super-resolution imaging of dynamic processes in live cells.

DeepOM: single-molecule optical genome mapping via deep learning.

MOTIVATION: Efficient tapping into genomic information from a single microscopic image of an intact DNA molecule is an outstanding challenge and its solution will open new frontiers in molecular diagnostics. Here, a new computational method for optical genome mapping utilizing deep learning is presented, termed DeepOM. Utilization of a convolutional neural network, trained on simulated images of labeled DNA molecules, improves the success rate in the alignment of DNA images to genomic references. RESULTS: The method is evaluated on acquired images of human DNA molecules stretched in nano-channels. The accuracy of the method is benchmarked against state-of-the-art commercial software Bionano Solve. The results show a significant advantage in alignment success rate for molecules shorter than 50 kb. DeepOM improves the yield, sensitivity, and throughput of optical genome mapping experiments in applications of human genomics and microbiology. AVAILABILITY AND IMPLEMENTATION: The source code for the presented method is publicly available at https://github.com/yevgenin/DeepOM.

DeepSTORM3D: dense 3D localization microscopy and PSF design by deep learning.

An outstanding challenge in single-molecule localization microscopy is the accurate and precise localization of individual point emitters in three dimensions in densely labeled samples. One established approach for three-dimensional single-molecule localization is point-spread-function (PSF) engineering, in which the PSF is engineered to vary distinctively with emitter depth using additional optical elements. However, images of dense emitters, which are desirable for improving temporal resolution, pose a challenge for algorithmic localization of engineered PSFs, due to lateral overlap of the emitter PSFs. Here we train a neural network to localize multiple emitters with densely overlapping Tetrapod PSFs over a large axial range. We then use the network to design the optimal PSF for the multi-emitter case. We demonstrate our approach experimentally with super-resolution reconstructions of mitochondria and volumetric imaging of fluorescently labeled telomeres in cells. Our approach, DeepSTORM3D, enables the study of biological processes in whole cells at timescales that are rarely explored in localization microscopy.

Publisher Correction: DeepSTORM3D: dense 3D localization microscopy and PSF design by deep learning.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

VIPR: vectorial implementation of phase retrieval for fast and accurate microscopic pixel-wise pupil estimation.

In microscopy, proper modeling of the image formation has a substantial effect on the precision and accuracy in localization experiments and facilitates the correction of aberrations in adaptive optics experiments. The observed images are subject to polarization effects, refractive index variations, and system specific constraints. Previously reported techniques have addressed these challenges by using complicated calibration samples, computationally heavy numerical algorithms, and various mathematical simplifications. In this work, we present a phase retrieval approach based on an analytical derivation of the vectorial diffraction model. Our method produces an accurate estimate of the system's phase information, without any prior knowledge about the aberrations, in under a minute.

A unique octameric structure of Axe2, an intracellular acetyl-xylooligosaccharide esterase from Geobacillus stearothermophilus.

Geobacillus stearothermophilus T6 is a thermophilic, Gram-positive soil bacterium that possesses an extensive and highly regulated hemicellulolytic system, allowing the bacterium to efficiently degrade high-molecular-weight polysaccharides such as xylan, arabinan and galactan. As part of the xylan-degradation system, the bacterium uses a number of side-chain-cleaving enzymes, one of which is Axe2, a 219-amino-acid intracellular serine acetylxylan esterase that removes acetyl side groups from xylooligosaccharides. Bioinformatic analyses suggest that Axe2 belongs to the lipase GDSL family and represents a new family of carbohydrate esterases. In the current study, the detailed three-dimensional structure of Axe2 is reported, as determined by X-ray crystallography. The structure of the selenomethionine derivative Axe2-Se was initially determined by single-wavelength anomalous diffraction techniques at 1.70 Šresolution and was used for the structure determination of wild-type Axe2 (Axe2-WT) and the catalytic mutant Axe2-S15A at 1.85 and 1.90 Šresolution, respectively. These structures demonstrate that the three-dimensional structure of the Axe2 monomer generally corresponds to the SGNH hydrolase fold, consisting of five central parallel ß-sheets flanked by two layers of helices (eight α-helices and five 310-helices). The catalytic triad residues, Ser15, His194 and Asp191, are lined up along a substrate channel situated on the concave surface of the monomer. Interestingly, the Axe2 monomers are assembled as a `doughnut-shaped' homo-octamer, presenting a unique quaternary structure built of two staggered tetrameric rings. The eight active sites are organized in four closely situated pairs, which face the relatively wide internal cavity. The biological relevance of this octameric structure is supported by independent results obtained from gel-filtration, TEM and SAXS experiments. These data and their comparison to the structural data of related hydrolases are used for a more general discussion focusing on the structure-function relationships of enzymes of this category.

Large-FOV 3D localization microscopy by spatially variant point spread function generation.

Accurate characterization of the microscopic point spread function (PSF) is crucial for achieving high-performance localization microscopy (LM). Traditionally, LM assumes a spatially invariant PSF to simplify the modeling of the imaging system. However, for large fields of view (FOV) imaging, it becomes important to account for the spatially variant nature of the PSF. Here, we propose an accurate and fast principal components analysis-based field-dependent 3D PSF generator (PPG3D) and localizer for LM. Through simulations and experimental three-dimensional (3D) single-molecule localization microscopy (SMLM), we demonstrate the effectiveness of PPG3D, enabling super-resolution imaging of mitochondria and microtubules with high fidelity over a large FOV. A comparison of PPG3D with a shift-variant PSF generator for 3D LM reveals a threefold improvement in accuracy. Moreover, PPG3D is approximately 100 times faster than existing PSF generators, when used in image plane-based interpolation mode. Given its user-friendliness, we believe that PPG3D holds great potential for widespread application in SMLM and other imaging modalities.

This Microtubule Does Not Exist: Super-Resolution Microscopy Image Generation by a Diffusion Model.

Generative models, such as diffusion models, have made significant advancements in recent years, enabling the synthesis of high-quality realistic data across various domains. Here, the adaptation and training of a diffusion model on super-resolution microscopy images are explored. It is shown that the generated images resemble experimental images, and that the generation process does not exhibit a large degree of memorization from existing images in the training set. To demonstrate the usefulness of the generative model for data augmentation, the performance of a deep learning-based single-image super-resolution (SISR) method trained using generated high-resolution data is compared against training using experimental images alone, or images generated by mathematical modeling. Using a few experimental images, the reconstruction quality and the spatial resolution of the reconstructed images are improved, showcasing the potential of diffusion model image generation for overcoming the limitations accompanying the collection and annotation of microscopy images. Finally, the pipeline is made publicly available, runnable online, and user-friendly to enable researchers to generate their own synthetic microscopy data. This work demonstrates the potential contribution of generative diffusion models for microscopy tasks and paves the way for their future application in this field.

Depth-enhanced high-throughput microscopy by compact PSF engineering.

High-throughput microscopy is vital for screening applications, where three-dimensional (3D) cellular models play a key role. However, due to defocus susceptibility, current 3D high-throughput microscopes require axial scanning, which lowers throughput and increases photobleaching and photodamage. Point spread function (PSF) engineering is an optical method that enables various 3D imaging capabilities, yet it has not been implemented in high-throughput microscopy due to the cumbersome optical extension it typically requires. Here we demonstrate compact PSF engineering in the objective lens, which allows us to enhance the imaging depth of field and, combined with deep learning, recover 3D information using single snapshots. Beyond the applications shown here, this work showcases the usefulness of high-throughput microscopy in obtaining training data for deep learning-based algorithms, applicable to a variety of microscopy modalities.

Crystallization and preliminary crystallographic analysis of Axe2, an acetylxylan esterase from Geobacillus stearothermophilus.

Acetylxylan esterases are part of the hemi-cellulolytic system of many microorganisms which utilize plant biomass for growth. Xylans, which are polymeric sugars that constitute a significant part of the plant biomass, are usually substituted with acetyl side groups attached at position 2 or 3 of the xylose backbone units. Acetylxylan esterases hydrolyse the ester linkages of the xylan acetyl groups and thus improve the ability of main-chain hydrolysing enzymes to break down the sugar backbone units. As such, these enzymes play an important part in the hemi-cellulolytic utilization system of many microorganisms that use plant biomass for growth. Interest in the biochemical characterization and structural analysis of these enzymes stems from their numerous potential biotechnological applications. An acetylxylan esterase (Axe2) of this type from Geobacillus stearothermophilus T-6 has recently been cloned, overexpressed, purified, biochemically characterized and crystallized. One of the crystal forms obtained (RB1) belonged to the tetragonal space group I422, with unit-cell parameters a = b = 110.2, c = 213.1 Å. A full diffraction data set was collected to 1.85 Å resolution from flash-cooled crystals of the wild-type enzyme at 100 K using synchrotron radiation. A selenomethionine derivative of Axe2 has also been prepared and crystallized for single-wavelength anomalous diffraction experiments. The crystals of the selenomethionine-derivatized Axe2 appeared to be isomorphous to those of the wild-type enzyme and enabled the measurement of a full 1.85 Å resolution diffraction data set at the selenium absorption edge and a full 1.70 Å resolution data set at a remote wavelength. These data are currently being used for three-dimensional structure determination of the Axe2 protein.

Near index matching enables solid diffractive optical element fabrication via additive manufacturing.

Diffractive optical elements (DOEs) have a wide range of applications in optics and photonics, thanks to their capability to perform complex wavefront shaping in a compact form. However, widespread applicability of DOEs is still limited, because existing fabrication methods are cumbersome and expensive. Here, we present a simple and cost-effective fabrication approach for solid, high-performance DOEs. The method is based on conjugating two nearly refractive index-matched solidifiable transparent materials. The index matching allows for extreme scaling up of the elements in the axial dimension, which enables simple fabrication of a template using commercially available 3D printing at tens-of-micrometer resolution. We demonstrated the approach by fabricating and using DOEs serving as microlens arrays, vortex plates, including for highly sensitive applications such as vector beam generation and super-resolution microscopy using MINSTED, and phase-masks for three-dimensional single-molecule localization microscopy. Beyond the advantage of making DOEs widely accessible by drastically simplifying their production, the method also overcomes difficulties faced by existing methods in fabricating highly complex elements, such as high-order vortex plates, and spectrum-encoding phase masks for microscopy.

A new family of carbohydrate esterases is represented by a GDSL hydrolase/acetylxylan esterase from Geobacillus stearothermophilus.

Acetylxylan esterases hydrolyze the ester linkages of acetyl groups at positions 2 and/or 3 of the xylose moieties in xylan and play an important role in enhancing the accessibility of xylanases to the xylan backbone. The hemicellulolytic system of the thermophilic bacterium Geobacillus stearothermophilus T-6 comprises a putative acetylxylan esterase gene, axe2. The gene product belongs to the GDSL hydrolase family and does not share sequence homology with any of the carbohydrate esterases in the CAZy Database. The axe2 gene is induced by xylose, and the purified gene product completely deacetylates xylobiose peracetate (fully acetylated) and hydrolyzes the synthetic substrates 2-naphthyl acetate, 4-nitrophenyl acetate, 4-methylumbelliferyl acetate, and phenyl acetate. The pH profiles for k(cat) and k(cat)/K(m) suggest the existence of two ionizable groups affecting the binding of the substrate to the enzyme. Using NMR spectroscopy, the regioselectivity of Axe2 was directly determined with the aid of one-dimensional selective total correlation spectroscopy. Methyl 2,3,4-tri-O-acetyl-ß-d-xylopyranoside was rapidly deacetylated at position 2 or at positions 3 and 4 to give either diacetyl or monoacetyl intermediates, respectively; methyl 2,3,4,6-tetra-O-acetyl-ß-d-glucopyranoside was initially deacetylated at position 6. In both cases, the complete hydrolysis of the intermediates occurred at a much slower rate, suggesting that the preferred substrate is the peracetate sugar form. Site-directed mutagenesis of Ser-15, His-194, and Asp-191 resulted in complete inactivation of the enzyme, consistent with their role as the catalytic triad. Overall, our results show that Axe2 is a serine acetylxylan esterase representing a new carbohydrate esterase family.

Quantifying cell-cycle-dependent chromatin dynamics during interphase by live 3D tracking.

The study of cell cycle progression and regulation is important to our understanding of fundamental biophysics, aging, and disease mechanisms. Local chromatin movements are generally considered to be constrained and relatively consistent during all interphase stages, although recent advances in our understanding of genome organization challenge this claim. Here, we use high spatiotemporal resolution, 4D (x, y, z and time) localization microscopy by point-spread-function (PSF) engineering and deep learning-based image analysis, for live imaging of mouse embryonic fibroblast (MEF 3T3) and MEF 3T3 double Lamin A Knockout (LmnaKO) cell lines, to characterize telomere diffusion during the interphase. We detected varying constraint levels imposed on chromatin, which are prominently decreased during G0/G1. Our 4D measurements of telomere diffusion offer an effective method to investigate chromatin dynamics and reveal cell-cycle-dependent motion constraints, which may be caused by various cellular processes.

3D printable diffractive optical elements by liquid immersion.

Diffractive optical elements (DOEs) are used to shape the wavefront of incident light. This can be used to generate practically any pattern of interest, albeit with varying efficiency. A fundamental challenge associated with DOEs comes from the nanoscale-precision requirements for their fabrication. Here we demonstrate a method to controllably scale up the relevant feature dimensions of a device from tens-of-nanometers to tens-of-microns by immersing the DOEs in a near-index-matched solution. This makes it possible to utilize modern 3D-printing technologies for fabrication, thereby significantly simplifying the production of DOEs and decreasing costs by orders of magnitude, without hindering performance. We demonstrate the tunability of our design for varying experimental conditions, and the suitability of this approach to ultrasensitive applications by localizing the 3D positions of single molecules in cells using our microscale fabricated optical element to modify the point-spread-function (PSF) of a microscope.

Three-dimensional localization microscopy in live flowing cells.

Capturing the dynamics of live cell populations with nanoscale resolution poses a significant challenge, primarily owing to the speed-resolution trade-off of existing microscopy techniques. Flow cytometry would offer sufficient throughput, but lacks subsample detail. Here we show that imaging flow cytometry, in which the point detectors of flow cytometry are replaced with a camera to record 2D images, is compatible with 3D localization microscopy through point-spread-function engineering, which encodes the depth of the emitter into the emission pattern captured by the camera. The extraction of 3D positions from sub-cellular objects of interest is achieved by calibrating the depth-dependent response of the imaging system using fluorescent beads mixed with the sample buffer. This approach enables 4D imaging of up to tens of thousands of objects per minute and can be applied to characterize chromatin dynamics and the uptake and spatial distribution of nanoparticles in live cancer cells.

Ultrasensitive Refractometry via Supercritical Angle Fluorescence.

Refractometry, namely, the measurement of refractive index (RI), provides information about various sample properties such as concentrations and molecular structure. One physical phenomenon which enables precise determination of a sample's RI in a microscope is the supercritical-angle fluorescence. This effect is observed when the fluorescence from an emitter near a glass-medium interface is captured by an objective lens with a high numerical aperture. The materials' index mismatch creates a distinguishable transition in the intensity pattern at the back focal plane of the objective that changes proportionally to the RI of the media. Here, we present a refractometry approach in which the fluorophores are preattached to the bottom surface of a microfluidic channel, enabling highly sensitive determination of the RI using tiny amounts of liquid (picoliters). With this method, we attained a standard deviation of 3.1 × 10-5 and a repeatability of 2.7 × 10-5 RI units. We first determine the capabilities of the device for glycerol-water solutions and then demonstrate the relevance of our system for monitoring changes in biological systems. As a model system, we show that we can detect single bacteria ( Escherichia coli) and measure population growth.

A myovirus encoding both photosystem I and II proteins enhances cyclic electron flow in infected Prochlorococcus cells.

Cyanobacteria are important contributors to primary production in the open oceans. Over the past decade, various photosynthesis-related genes have been found in viruses that infect cyanobacteria (cyanophages). Although photosystem II (PSII) genes are common in both cultured cyanophages and environmental samples 1-4 , viral photosystem I (vPSI) genes have so far only been detected in environmental samples 5,6 . Here, we have used a targeted strategy to isolate a cyanophage from the tropical Pacific Ocean that carries a PSI gene cassette with seven distinct PSI genes (psaJF, C, A, B, K, E, D) as well as two PSII genes (psbA, D). This cyanophage, P-TIM68, belongs to the T4-like myoviruses, has a prolate capsid, a long contractile tail and infects Prochlorococcus sp. strain MIT9515. Phage photosynthesis genes from both photosystems are expressed during infection, and the resultant proteins are incorporated into membranes of the infected host. Moreover, photosynthetic capacity in the cell is maintained throughout the infection cycle with enhancement of cyclic electron flow around PSI. Analysis of metagenomic data from the Tara Oceans expedition 7 shows that phages carrying PSI gene cassettes are abundant in the tropical Pacific Ocean, composing up to 28% of T4-like cyanomyophages. They are also present in the tropical Indian and Atlantic Oceans. P-TIM68 populations, specifically, compose on average 22% of the PSI-gene-cassette carrying phages. Our results suggest that cyanophages carrying PSI and PSII genes are likely to maintain and even manipulate photosynthesis during infection of their Prochlorococcus hosts in the tropical oceans.

Preliminary crystallographic analysis of a double mutant of the acetyl xylo-oligosaccharide esterase Axe2 in its dimeric form.

Xylans are polymeric sugars constituting a significant part of the plant cell wall. They are usually substituted with acetyl side groups attached at positions 2 or 3 of the xylose backbone units. Acetylxylan esterases are part of the hemicellulolytic system of many microorganisms which utilize plant biomass for growth. These enzymes hydrolyze the ester linkages of the xylan acetyl groups and thus improve the accessibility of main-chain-hydrolyzing enzymes and their ability to break down the sugar backbone units. The acetylxylan esterases are therefore critically important for those microorganisms and as such could be used for a wide range of biotechnological applications. The structure of an acetylxylan esterase (Axe2) isolated from the thermophilic bacterium Geobacillus stearothermophilus T6 has been determined, and it has been demonstrated that the wild-type enzyme is present as a unique torus-shaped octamer in the crystal and in solution. In order to understand the functional origin of this unique oligomeric structure, a series of rational noncatalytic, site-specific mutations have been made on Axe2. Some of these mutations led to a different dimeric form of the protein, which showed a significant reduction in catalytic activity. One of these double mutants, Axe2-Y184F-W190P, has recently been overexpressed, purified and crystallized. The best crystals obtained belonged to the orthorhombic space group P212121, with unit-cell parameters a = 71.1, b = 106.0, c = 378.6 Å. A full diffraction data set to 2.3 Šresolution has been collected from a flash-cooled crystal of this type at 100 K using synchrotron radiation. This data set is currently being used for the three-dimensional structure analysis of the Axe2-Y184F-W190P mutant in its dimeric form.

The abp gene in Geobacillus stearothermophilus T-6 encodes a GH27 ß-L-arabinopyranosidase.

In this study we demonstrate that the abp gene in Geobacillus stearothermophilus T-6 encodes a family 27 glycoside hydrolase ß-L-arabinopyranosidase. The catalytic constants towards the chromogenic substrate pNP-ß-L-arabinopyranoside were 0.8±0.1 mM, 6.6±0.3 s(-1), and 8.2±0.3 s(-1) mM(-1) for K(m), k(cat) and k(cat)/K(m), respectively. (13)C NMR spectroscopy unequivocally showed that Abp is capable of removing ß-L-arabinopyranose residues from the natural arabino-polysaccharide, larch arabinogalactan. Most family 27 enzymes are active on galactose and contain a conserved Asp residue, whereas in Abp this residue is Ile67, which shifts the specificity of the enzyme towards arabinopyranoside.

Potential for phosphite and phosphonate utilization by Prochlorococcus.

Phosphonates (Pn) are diverse organic phosphorus (P) compounds containing C-P bonds and comprise up to 25% of the high-molecular weight dissolved organic P pool in the open ocean. Pn bioavailability was suggested to influence markedly bacterial primary production in low-P areas. Using metagenomic data from the Global Ocean Sampling expedition, we show that the main potential microbial contributor in Pn utilization in oceanic surface water is the globally important marine primary producer Prochlorococcus. Moreover, a number of Prochlorococcus strains contain two distinct putative Pn uptake operons coding for ABC-type Pn transporters. On the basis of microcalorimetric measurements, we find that each of the two different putative Pn-binding protein (PhnD) homologs transcribed from these operons possesses different Pn- as well as inorganic phosphite-binding specificities. Our results suggest that Prochlorococcus adapt to low-P environments by increasing the number of Pn transporters with different specificities towards phosphite and different Pns.