Seamus Holden

DAOSTORM: an algorithm for high- density super-resolution microscopy

We first investigated the qualitative performance of each algorithm for images of Alexa Fluor 647–immunolabeled microtubules in fixed COS-7 cells. We recorded data at high imaging density using total internal reflection fluorescence microscopy and direct (d)STORM photoswitching conditions5 (100 ms integration time, ~4,000 photons fluorophore–1 frame–1). We plotted localizations on raw images, illustrating the characteristic performance of each algorithm (Fig. 1a). SA1 only localized isolated molecules, which were fitted with small localization error. SA2 localized a larger fraction of the molecules but yielded large localization errors for overlapping molecules. DAOSTORM outperformed both sparse algorithms, identifying almost all molecules with small localization error. We quantified the performance of each algorithm by analyzing simulations of randomly distributed surface-immobilized fluorophores6. We compared observed localizations to simulated positions, calculating the recall5 and localization error at different imaging densities. Recall is the percentage of simulated fluorophores detected. Localization error is the root-mean-square distance between a localization and the simulated position. We also measured the precision5 and redundancy (Supplementary Methods), which did not vary substantially. DAOSTORM substantially outperformed the sparse algorithms in simulations at high signal-to-noise ratio typical of STORM data (bright organic fluorophores, 5,000 photons molecule–1 frame–1; Fig. 1b-c). SA1 showed poor recall at high density, with imaging density at half-maximum recall, ρHM, of 1.2 molecule μm –2. However, SA1 yielded small localization errors even at high imaging density because most overlapping molecules were rejected. SA2 had better recall performance (ρHM = 3.4 molecules μm –2) but gave large localization errors even at low imaging density (>0.1 molecules μm–2). In contrast, DAOSTORM gave small localization errors similar to the other ‘precise’ algorithm, SA1, together with a sixfold improvement in recall performance (ρHM = 7.5 molecules μm –2). For simulations at low signal-to-noise ratio typical of photoactivated localization microscopy data (fluorescent proteins, 200 photons molecule–1 frame–1; DAOSTORM: an algorithm for highdensity super-resolution microscopy

Treadmilling by FtsZ filaments drives peptidoglycan synthesis and bacterial cell division

Coordinating cell wall synthesis and cell division Most bacteria are protected by peptidoglycan cell walls, which must be remodeled to split the cell. Cell division requires the tubulin homolog FtsZ, a highly conserved cytoskeletal polymer that specifies the future site of division. Bisson-Filho et al. and Yang et al. found that the dynamic treadmilling of FtsZ filaments controls both the location and activity of the associated cell wall synthetic enzymes. This creates discrete sites of cell wall synthesis that circle around the division plane to divide the cell. Science, this issue p. 739, p. 744 Bacterial cytokinesis is controlled by the circumferential treadmilling of FtsZ and FtsA filaments that drives the insertion of new cell wall. The mechanism by which bacteria divide is not well understood. Cell division is mediated by filaments of FtsZ and FtsA (FtsAZ) that recruit septal peptidoglycan-synthesizing enzymes to the division site. To understand how these components coordinate to divide cells, we visualized their movements relative to the dynamics of cell wall synthesis during cytokinesis. We found that the division septum was built at discrete sites that moved around the division plane. FtsAZ filaments treadmilled circumferentially around the division ring and drove the motions of the peptidoglycan-synthesizing enzymes. The FtsZ treadmilling rate controlled both the rate of peptidoglycan synthesis and cell division. Thus, FtsZ treadmilling guides the progressive insertion of new cell wall by building increasingly smaller concentric rings of peptidoglycan to divide the cell.

High throughput 3D super-resolution microscopy reveals Caulobacter crescentus in vivo Z-ring organization

Significance The bacterial cytoskeletal protein FtsZ, which forms a constricting “Z-ring” during cell division, is the major cytoskeletal protein involved in cell division in almost all prokaryotes, and is a key next-generation antibiotic target. However, the small size of the Z-ring, approximately 500 nm in diameter, makes it difficult to observe in vivo. We provide a quantitative nanoscale picture of Z-ring organization in live cells; these results improve our understanding of the structural and force-generating roles of FtsZ in bacterial cell division. To achieve this, we created an automated modality of superresolution fluorescence microscopy, allowing high-throughput live cell microscopy at nanoscale resolution; this technique should be broadly useful in prokaryotic and eukaryotic systems. We created a high-throughput modality of photoactivated localization microscopy (PALM) that enables automated 3D PALM imaging of hundreds of synchronized bacteria during all stages of the cell cycle. We used high-throughput PALM to investigate the nanoscale organization of the bacterial cell division protein FtsZ in live Caulobacter crescentus. We observed that FtsZ predominantly localizes as a patchy midcell band, and only rarely as a continuous ring, supporting a model of “Z-ring” organization whereby FtsZ protofilaments are randomly distributed within the band and interact only weakly. We found evidence for a previously unidentified period of rapid ring contraction in the final stages of the cell cycle. We also found that DNA damage resulted in production of high-density continuous Z-rings, which may obstruct cytokinesis. Our results provide a detailed quantitative picture of in vivo Z-ring organization.

Multiscale Spatial Organization of RNA Polymerase in Escherichia coli

Nucleic acid synthesis is spatially organized in many organisms. In bacteria, however, the spatial distribution of transcription remains obscure, owing largely to the diffraction limit of conventional light microscopy (200-300 nm). Here, we use photoactivated localization microscopy to localize individual molecules of RNA polymerase (RNAP) in Escherichia coli with a spatial resolution of ∼40 nm. In cells growing rapidly in nutrient-rich media, we find that RNAP is organized in 2-8 bands. The band number scaled directly with cell size (and so with the chromosome number), and bands often contained clusters of >70 tightly packed RNAPs (possibly engaged on one long ribosomal RNA operon of 6000 bp) and clusters of such clusters (perhaps reflecting a structure like the eukaryotic nucleolus where many different ribosomal RNA operons are transcribed). In nutrient-poor media, RNAPs were located in only 1-2 bands; within these bands, a disproportionate number of RNAPs were found in clusters containing ∼20-50 RNAPs. Apart from their importance for bacterial transcription, our studies pave the way for molecular-level analysis of several cellular processes at the nanometer scale.

Defining the Limits of Single-Molecule FRET Resolution in TIRF Microscopy

Single-molecule FRET (smFRET) has long been used as a molecular ruler for the study of biology on the nanoscale (∼2-10 nm); smFRET in total-internal reflection fluorescence (TIRF) Förster resonance energy transfer (TIRF-FRET) microscopy allows multiple biomolecules to be simultaneously studied with high temporal and spatial resolution. To operate at the limits of resolution of the technique, it is essential to investigate and rigorously quantify the major sources of noise and error; we used theoretical predictions, simulations, advanced image analysis, and detailed characterization of DNA standards to quantify the limits of TIRF-FRET resolution. We present a theoretical description of the major sources of noise, which was in excellent agreement with results for short-timescale smFRET measurements (<200 ms) on individual molecules (as opposed to measurements on an ensemble of single molecules). For longer timescales (>200 ms) on individual molecules, and for FRET distributions obtained from an ensemble of single molecules, we observed significant broadening beyond theoretical predictions; we investigated the causes of this broadening. For measurements on individual molecules, analysis of the experimental noise allows us to predict a maximum resolution of a FRET change of 0.08 with 20-ms temporal resolution, sufficient to directly resolve distance differences equivalent to one DNA basepair separation (0.34 nm). For measurements on ensembles of single molecules, we demonstrate resolution of distance differences of one basepair with 1000-ms temporal resolution, and differences of two basepairs with 80-ms temporal resolution. Our work paves the way for ultra-high-resolution TIRF-FRET studies on many biomolecules, including DNA processing machinery (DNA and RNA polymerases, helicases, etc.), the mechanisms of which are often characterized by distance changes on the scale of one DNA basepair.

Identifying Molecular Dynamics in Single-Molecule FRET Experiments with Burst Variance Analysis

Histograms of single-molecule Förster resonance energy transfer (FRET) efficiency are often used to study the structures of biomolecules and relate these structures to function. Methods like probability distribution analysis analyze FRET histograms to detect heterogeneities in molecular structure, but they cannot determine whether this heterogeneity arises from dynamic processes or from the coexistence of several static structures. To this end, we introduce burst variance analysis (BVA), a method that detects dynamics by comparing the standard deviation of FRET from individual molecules over time to that expected from theory. Both simulations and experiments on DNA hairpins show that BVA can distinguish between static and dynamic sources of heterogeneity in single-molecule FRET histograms and can test models of dynamics against the observed standard deviation information. Using BVA, we analyzed the fingers-closing transition in the Klenow fragment of Escherichia coli DNA polymerase I and identified substantial dynamics in polymerase complexes formed prior to nucleotide incorporation; these dynamics may be important for the fidelity of DNA synthesis. We expect BVA to be broadly applicable to single-molecule FRET studies of molecular structure and to complement approaches such as probability distribution analysis and fluorescence correlation spectroscopy in studying molecular dynamics.

FALCON: fast and unbiased reconstruction of high-density super-resolution microscopy data

Super resolution microscopy such as STORM and (F)PALM is now a well known method for biological studies at the nanometer scale. However, conventional imaging schemes based on sparse activation of photo-switchable fluorescent probes have inherently slow temporal resolution which is a serious limitation when investigating live-cell dynamics. Here, we present an algorithm for high-density super-resolution microscopy which combines a sparsity-promoting formulation with a Taylor series approximation of the PSF. Our algorithm is designed to provide unbiased localization on continuous space and high recall rates for high-density imaging, and to have orders-of-magnitude shorter run times compared to previous high-density algorithms. We validated our algorithm on both simulated and experimental data, and demonstrated live-cell imaging with temporal resolution of 2.5 seconds by recovering fast ER dynamics.

Single‐Molecule DNA Biosensors for Protein and Ligand Detection

Transcription factors (TFs) are sequence-specific DNA-binding proteins that control much of gene expression. TFs are natural biosensors and switches, translating chemical and physical signals (temperature shifts, light exposure, chemical concentrations, redox status) into transcriptional changes by modulating the binding of RNA polymerase to promoter DNA. Since changes in TF levels underlie fundamental biological processes such as DNA repair and cell-cycle progression, alterations in the levels of active TFs both lead to and indicate disease; for example, mutations in transcription factor p53 contribute to the rapid growth of cancer cells and, owing to their prevalence (p53 is mutated in roughly 50% of all human tumors), they have served as cancer biomarkers. Thus, methods for the sensitive detection and quantitation of TFs provide both fundamental information about gene regulation and a platform for diagnostics. TF detection often involves gel-based assays and Western blotting; although helpful in characterizing TF–DNA interactions, these assays are tedious, expensive, and qualitative, and consume large quantities of sample. Enzyme-linked immunosorbent assays (ELISAs) are more sensitive and offer higher throughput, but they require many preparation and signal-amplification steps for the detection of lowabundance TFs. Amplification is also required in the proximity-based ligation assay, making it incompatible with TF detection in living cells and diagnostic settings that demand results within minutes. An additional TF detection assay is based on fluorescence resonance energy transfer (FRET) between two doublestranded DNA (dsDNA) fragments containing fluorescently labeled single-stranded complementary overhangs (“molecular beacons”). In the presence of TF, the DNAs associate, resulting in donor fluorophore quenching as a result of FRET. This assay still requires significant amounts of sample and cannot detect low-abundance TFs; and because of the short dynamic range of FRET (1–10 nm), it also requires close proximity among the fluorophore, the quencher, and the protein–DNA interface, increasing the likelihood of steric interference with protein–DNA binding and complicating sensor design. Moreover, placing the fluorophore and the quencher on either side of the protein-binding site (usually 15–30 base pairs (bp) in length) on DNA results in very low FRET signals for most TFs. Here, we use alternating-laser excitation (ALEX) spectroscopy to detect TFs and small molecules by means of the TF-dependent coincidence of fluorescently labeled DNA. Like the molecular-beacon assay, our method is based on TFdriven DNA association, is rapid, and requires no amplification. However, our assay can detect pm levels of TFs in small amounts of sample, and it is FRET-independent, bypassing the need to optimize fluorophore position or know the structural details of TF–DNA binding; this flexibility in labeling ensures unperturbed TF–DNA binding. Using ALEX, we demonstrate TF and small-molecule detection, assay multiplexing, and suitability for analysis of complex biological samples. In our assay (Figure 1a,b), the full DNA-binding site for a TF is split in two (as in Ref. [5]): the left half-site (H1) and the right half-site (H2). Each site contains half of the TF-binding determinants and short, complementary 3’-overhangs. H1 is labeled with a “green” fluorophore (“G”) to give half-site H1, whereas H2 is labeled with a spectrally distinct “red” fluorophore (“R”) to give H2. In the absence of TF and at DNA concentrations of roughly 10–100 pm, H1 and H2 diffuse independently and associate only transiently. In contrast, in the presence of a TF that binds to the fully assembled DNA site, H1 and H2 diffuse as a complex (H1TF-H2; Figure 1a, bottom). We detect TF-dependent DNA coincidence using ALEX spectroscopy, wherein single molecules are excited by two lasers in an alternating fashion, with each laser capable of directly exciting either a G or a R fluorophore. ALEX allows molecular sorting on two-dimensional histograms of apparent FRET efficiency E* (a fluorescence ratio that reports on interfluorophore proximity) and probe stoichiometry S (a fluorescence ratio that reports on molecular stoichiometry). A search for all R-labeled molecules (i.e., G–R molecules [*] Dr. K. Lymperopoulos, R. Crawford, J. P. Torella, Dr. M. Heilemann, Dr. L. C. Hwang, S. J. Holden, Dr. A. N. Kapanidis Biological Physics Research Group, Department of Physics University of Oxford, Clarendon Laboratory Parks Road, Oxford, OX1 3PU (United Kingdom) E-mail: a.kapanidis1@physics.ox.ac.uk Dr. K. Lymperopoulos Current address: BioQuant Institute, Cellnetworks Cluster Ruprecht-Karls Universit t Heidelberg 69120 Heidelberg (Germany)

3D high-density localization microscopy using hybrid astigmatic/ biplane imaging and sparse image reconstruction

Localization microscopy achieves nanoscale spatial resolution by iterative localization of sparsely activated molecules, which generally leads to a long acquisition time. By implementing advanced algorithms to treat overlapping point spread functions (PSFs), imaging of densely activated molecules can improve the limited temporal resolution, as has been well demonstrated in two-dimensional imaging. However, three-dimensional (3D) localization of high-density data remains challenging since PSFs are far more similar along the axial dimension than the lateral dimensions. Here, we present a new, high-density 3D imaging system and algorithm. The hybrid system is implemented by combining astigmatic and biplane imaging. The proposed 3D reconstruction algorithm is extended from our state-of-the art 2D high-density localization algorithm. Using mutual coherence analysis of model PSFs, we validated that the hybrid system is more suitable than astigmatic or biplane imaging alone for 3D localization of high-density data. The efficacy of the proposed method was confirmed via simulation and real data of microtubules. Furthermore, we also successfully demonstrated fluorescent-protein-based live cell 3D localization microscopy with a temporal resolution of just 3 seconds, capturing fast dynamics of the endoplasmic recticulum.

PALMsiever: a tool to turn raw data into results for single-molecule localization microscopy

During the past decade, localization microscopy (LM) has transformed into an accessible, commercially available technique for life sciences. However, data processing can be challenging to the non-specialist and care is still needed to produce meaningful results. PALMsiever has been developed to provide a user-friendly means of visualizing, filtering and analyzing LM data. It includes drift correction, clustering, intelligent line profiles, many rendering algorithms and 3D data visualization. It incorporates the main analysis and data processing modalities used by experts in the field, as well as several new features we developed, and makes them broadly accessible. It can easily be extended via plugins and is provided as free of charge open-source software. Contact: thomas.pengo@gmail.com