An objective comparison of cell-tracking algorithms.

We present a combined report on the results of three editions of the Cell Tracking Challenge, an ongoing initiative aimed at promoting the development and objective evaluation of cell segmentation and tracking algorithms. With 21 participating algorithms and a data repository consisting of 13 data sets from various microscopy modalities, the challenge displays today's state-of-the-art methodology in the field. We analyzed the challenge results using performance measures for segmentation and tracking that rank all participating methods. We also analyzed the performance of all of the algorithms in terms of biological measures and practical usability. Although some methods scored high in all technical aspects, none obtained fully correct solutions. We found that methods that either take prior information into account using learning strategies or analyze cells in a global spatiotemporal video context performed better than other methods under the segmentation and tracking scenarios included in the challenge.

The Effect of Uncoated SPIONs on hiPSC-Differentiated Endothelial Cells.

BACKGROUND: Endothelial progenitor cells (EPCs) were indicated in vascular repair, angiogenesis of ischemic organs, and inhibition of formation of initial hyperplasia. Differentiation of endothelial cells (ECs) from human induced pluripotent stem cells (hiPSC)-derived endothelial cells (hiPSC-ECs) provides an unlimited supply for clinical application. Furthermore, magnetic cell labelling offers an effective way of targeting and visualization of hiPSC-ECs and is the next step towards in vivo studies. METHODS: ECs were differentiated from hiPSCs and labelled with uncoated superparamagnetic iron-oxide nanoparticles (uSPIONs). uSPION uptake was compared between hiPSC-ECs and mature ECs isolated from patients by software analysis of microscopy pictures after Prussian blue cell staining. The acute and long-term cytotoxic effects of uSPIONs were evaluated by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay) and Annexin assay. RESULTS: We showed, for the first time, uptake of uncoated SPIONs (uSPIONs) by hiPSC-ECs. In comparison with mature ECs of identical genetic background hiPSC-ECs showed lower uSPION uptake. However, all the studied endothelial cells were effectively labelled and showed magnetic properties even with low labelling concentration of uSPIONs. uSPIONs prepared by microwave plasma synthesis did not show any cytotoxicity nor impair endothelial properties. CONCLUSION: We show that hiPSC-ECs labelling with low concentration of uSPIONs is feasible and does not show any toxic effects in vitro, which is an important step towards animal studies.

A benchmark for comparison of cell tracking algorithms.

MOTIVATION: Automatic tracking of cells in multidimensional time-lapse fluorescence microscopy is an important task in many biomedical applications. A novel framework for objective evaluation of cell tracking algorithms has been established under the auspices of the IEEE International Symposium on Biomedical Imaging 2013 Cell Tracking Challenge. In this article, we present the logistics, datasets, methods and results of the challenge and lay down the principles for future uses of this benchmark. RESULTS: The main contributions of the challenge include the creation of a comprehensive video dataset repository and the definition of objective measures for comparison and ranking of the algorithms. With this benchmark, six algorithms covering a variety of segmentation and tracking paradigms have been compared and ranked based on their performance on both synthetic and real datasets. Given the diversity of the datasets, we do not declare a single winner of the challenge. Instead, we present and discuss the results for each individual dataset separately. AVAILABILITY AND IMPLEMENTATION: The challenge Web site (http://www.codesolorzano.com/celltrackingchallenge) provides access to the training and competition datasets, along with the ground truth of the training videos. It also provides access to Windows and Linux executable files of the evaluation software and most of the algorithms that competed in the challenge.

Performance and sensitivity evaluation of 3D spot detection methods in confocal microscopy.

Reliable 3D detection of diffraction-limited spots in fluorescence microscopy images is an important task in subcellular observation. Generally, fluorescence microscopy images are heavily degraded by noise and non-specifically stained background, making reliable detection a challenging task. In this work, we have studied the performance and parameter sensitivity of eight recent methods for 3D spot detection. The study is based on both 3D synthetic image data and 3D real confocal microscopy images. The synthetic images were generated using a simulator modeling the complete imaging setup, including the optical path as well as the image acquisition process. We studied the detection performance and parameter sensitivity under different noise levels and under the influence of uneven background signal. To evaluate the parameter sensitivity, we propose a novel measure based on the gradient magnitude of the F1 score. We measured the success rate of the individual methods for different types of the image data and found that the type of image degradation is an important factor. Using the F1 score and the newly proposed sensitivity measure, we found that the parameter sensitivity is not necessarily proportional to the success rate of a method. This also provided an explanation why the best performing method for synthetic data was outperformed by other methods when applied to the real microscopy images. On the basis of the results obtained, we conclude with the recommendation of the HDome method for data with relatively low variations in quality, or the Sorokin method for image sets in which the quality varies more. We also provide alternative recommendations for high-quality images, and for situations in which detailed parameter tuning might be deemed expensive.

A simple Fourier filter for suppression of the missing wedge ray artefacts in single-axis electron tomographic reconstructions.

The limited specimen tilting range that is typically available in electron tomography gives rise to a region in the Fourier space of the reconstructed object where experimental data are unavailable - the missing wedge. Since this region is sharply delimited from the area of available data, the reconstructed signal is typically hampered by convolution with its impulse response, which gives rise to the well-known missing wedge artefacts in 3D reconstructions. Despite the recent progress in the field of reconstruction and regularization techniques, the missing wedge artefacts remain untreated in most current reconstruction workflows in structural biology. Therefore we have designed a simple Fourier angular filter that effectively suppresses the ray artefacts in the single-axis tilting projection acquisition scheme, making single-axis tomographic reconstructions easier to interpret in particular at low signal-to-noise ratio in acquired projections. The proposed filter can be easily incorporated into current electron tomographic reconstruction schemes.

Determining Omics spatiotemporal dimensions using exciting new nanoscopy techniques to assess complex cell responses to DNA damage: part A--radiomics.

Recent ground-breaking developments in Omics have generated new hope for overcoming the complexity and variability of biological systems while simultaneously shedding more light on fundamental radiobiological questions that have remained unanswered for decades. In the era of Omics, our knowledge of how genes and proteins interact in the frame of complex networks to preserve genome integrity has been rapidly expanding. Nevertheless, these functional networks must be observed with strong correspondence to the cell nucleus, which is the main target of ionizing radiation. Nuclear architecture and nuclear processes, including DNA damage responses, are precisely organized in space and time. Information regarding these intricate processes cannot be achieved using high-throughput Omics approaches alone, but requires sophisticated structural probing and imaging. Based on the results obtained from studying the relationship between higher-order chromatin structure, DNA double-strand break induction and repair, and the formation of chromosomal translocations, we show the development of Omics solutions especially for radiation research (radiomics) (discussed in this article) and how confocal microscopy as well as novel approaches of molecular localization nanoscopy fill the gaps to successfully place the Omics data in the context of space and time (discussed in our other article in this issue, "Determining Omics Spatiotemporal Dimensions Using Exciting New Nanoscopy Techniques to Assess Complex Cell Responses to DNA Damage: Part B--Structuromics"). Finally, we introduce a novel method of specific chromatin nanotargeting and speculate future perspectives, which may combine nanoprobing and structural nanoscopy to observe structure-function correlations in living cells in real time. Thus, the Omics networks obtained from function analyses may be enriched by real-time visualization of Structuromics.

Determining Omics spatiotemporal dimensions using exciting new nanoscopy techniques to assess complex cell responses to DNA damage: part B--structuromics.

Recent groundbreaking developments in Omics and bioinformatics have generated new hope for overcoming the complexity and variability of (radio)biological systems while simultaneously shedding more light on fundamental radiobiological questions that have remained unanswered for decades. In the era of Omics, our knowledge of how genes and dozens of proteins interact in the frame of complex signaling and repair pathways (or, rather, networks) to preserve the integrity of the genome has been rapidly expanding. Nevertheless, these functional networks must be observed with strong correspondence to the cell nucleus, which is the main target of ionizing radiation. Information regarding these intricate processes cannot be achieved using high-throughput Omics approaches alone; it requires sophisticated structural probing and imaging. In the first part of this review, the article "Giving Omics Spatiotemporal Dimensions Using Exciting New Nanoscopy Techniques to Assess Complex Cell Responses to DNA Damage: Part A--Radiomics," we showed the development of different Omics solutions and how they are contributing to a better understanding of cellular radiation response. In this Part B we show how high-resolution confocal microscopy as well as novel approaches of molecular localization nanoscopy fill the gaps to successfully place Omics data in the context of space and time. The dynamics of double-strand breaks during repair processes and chromosomal rearrangements at the microscale correlated to aberration induction are explained. For the first time we visualize pan-nuclear nucleosomal rearrangements and clustering at the nanoscale during repair processes. Finally, we introduce a novel method of specific chromatin nanotargeting based on a computer database search of uniquely binding oligonucleotide combinations (COMBO-FISH). With these challenging techniques on hand, we speculate future perspectives that may combine specific COMBO-FISH nanoprobing and structural nanoscopy to observe structure-function correlations in living cells in real-time. Thus, the Omics networks obtained from function analyses may be enriched by real-time visualization of Structuromics.

Trajectories and nuclear arrangement of PML bodies are influenced by A-type lamin deficiency.

BACKGROUND INFORMATION: Promyelocytic leukaemia (PML) bodies are specific nuclear structures with functional significance for acute promyelocytic leukaemia. In this study, we analysed the trajectories of PML bodies using single-particle tracking. RESULTS: We observed that the recovery of PML protein after photobleaching was ATP dependent in both wild-type (wt) and A-type lamin-deficient cells. The movement of PML bodies was faster and the nuclear area occupied by particular PML bodies was larger in A-type lamin-deficient fibroblasts compared with their wt counterparts. Moreover, dysfunction of the LMNA gene increased the frequency of mutual interactions between individual PML bodies and influenced the morphology of these domains at the ultrastructural level. As a consequence of A-type lamin deficiency, PML protein accumulated in nuclear blebs and frequently appeared at the nuclear periphery. CONCLUSIONS: We suggest that the physiological function of lamin A proteins is important for events that occur in the compartment of PML bodies. This observation was confirmed in other experimental models characterised by lamin changes, including apoptosis or the differentiation of mouse embryonic stem cells.

Hybrid detectors improved time-lapse confocal microscopy of PML and 53BP1 nuclear body colocalization in DNA lesions.

We used hybrid detectors (HyDs) to monitor the trajectories and interactions of promyelocytic leukemia (GFP-PML) nuclear bodies (NBs) and mCherry-53BP1-positive DNA lesions. 53BP1 protein accumulates in NBs that occur spontaneously in the genome or in γ-irradiation-induced foci. When we induced local DNA damage by ultraviolet irradiation, we also observed accumulation of 53BP1 proteins into discrete bodies, instead of the expected dispersed pattern. In comparison with photomultiplier tubes, which are used for standard analysis by confocal laser scanning microscopy, HyDs significantly eliminated photobleaching of GFP and mCherry fluorochromes during image acquisition. The low laser intensities used for HyD-based confocal analysis enabled us to observe NBs for the longer time periods, necessary for studies of the trajectories and interactions of PML and 53BP1 NBs. To further characterize protein interactions, we used resonance scanning and a novel bioinformatics approach to register and analyze the movements of individual PML and 53BP1 NBs. The combination of improved HyD-based confocal microscopy with a tailored bioinformatics approach enabled us to reveal damage-specific properties of PML and 53BP1 NBs.

Acetylation-dependent nuclear arrangement and recruitment of BMI1 protein to UV-damaged chromatin.

Polycomb group (PcG) proteins, organized into Polycomb bodies, are important regulatory components of epigenetic processes involved in the heritable transcriptional repression of target genes. Here, we asked whether acetylation can influence the nuclear arrangement and function of the BMI1 protein, a core component of the Polycomb group complex, PRC1. We used time-lapse confocal microscopy, micro-irradiation by UV laser (355 nm) and GFP technology to study the dynamics and function of the BMI1 protein. We observed that BMI1 was recruited to UV-damaged chromatin simultaneously with decreased lysine acetylation, followed by the recruitment of heterochromatin protein HP1ß to micro-irradiated regions. Pronounced recruitment of BMI1 was rapid, with half-time τ = 15 sec; thus, BMI1 is likely involved in the initiation step leading to the recognition of UV-damaged sites. Histone hyperacetylation, stimulated by HDAC inhibitor TSA, suppression of transcription by actinomycin D, and ATP-depletion prevented increased accumulation of BMI1 to γH2AX-positive irradiated chromatin. Moreover, BMI1 had slight ability to recognize spontaneously occurring DNA breaks caused by other pathophysiological processes. Taken together, our data indicate that the dynamics of recognition of UV-damaged chromatin, and the nuclear arrangement of BMI1 protein can be influenced by acetylation and occur as an early event prior to the recruitment of HPß to UV-irradiated chromatin.

Arrangement of nuclear structures is not transmitted through mitosis but is identical in sister cells.

Although it is well known that chromosomes are non-randomly organized during interphase, it is not completely clear whether higher-order chromatin structure is transmitted from mother to daughter cells. Therefore, we addressed the question of how chromatin is rearranged during interphase and whether heterochromatin pattern is transmitted after mitosis. We additionally tested the similarity of chromatin arrangement in sister interphase nuclei. We noticed a very active cell rotation during interphase, especially when histone hyperacetylation was induced or transcription was inhibited. This natural phenomenon can influence the analysis of nuclear arrangement. Using photoconversion of Dendra2-tagged core histone H4 we showed that the distribution of chromatin in daughter interphase nuclei differed from that in mother cells. Similarly, the nuclear distribution of heterochromatin protein 1ß (HP1ß) was not completely identical in mother and daughter cells. However, identity between mother and daughter cells was in many cases evidenced by nucleolar composition. Moreover, morphology of nucleoli, HP1ß protein, Cajal bodies, chromosome territories, and gene transcripts were identical in sister cell nuclei. We conclude that the arrangement of interphase chromatin is not transmitted through mitosis, but the nuclear pattern is identical in naturally synchronized sister cells. It is also necessary to take into account the possibility that cell rotation and the degree of chromatin condensation during functionally specific cell cycle phases might influence our view of nuclear architecture.

Human embryonic stem cells suffer from centrosomal amplification.

Propagation of human embryonic stem cells (hESCs) in culture tends to alter karyotype, potentially limiting the prospective use of these cells in patients. The chromosomal instability of some malignancies is considered to be driven, at least in part, by centrosomal overamplification, perturbing balanced chromosome segregation. Here, we report, for the first time, that very high percentage of cultured hESCs has supernumerary centrosomes during mitosis. Supernumerary centrosomes were strictly associated with an undifferentiated hESC state and progressively disappeared on prolonged propagation in culture. Improved attachment to culture substratum and inhibition of CDK2 and Aurora A (key regulators of centrosomal metabolism) diminished the frequency of multicentrosomal mitoses. Thus, both attenuated cell attachment and deregulation of machinery controlling centrosome number contribute to centrosomal overamplification in hESCs. Linking the excessive number of centrosomes in mitoses to the ploidy indicated that both overduplication within a single cell cycle and mitotic failure contributed to generation of numerical centrosomal abnormalities in hESCs. Collectively, our data indicate that supernumerary centrosomes are a significant risk factor for chromosome instability in cultured hESCs and should be evaluated when new culture conditions are being implemented.

Endonuclease G interacts with histone H2B and DNA topoisomerase II alpha during apoptosis.

Apoptosis is a natural form of cell death involved in many physiological changes in the cell. Defects in the process of apoptosis can lead to serious diseases. During some apoptotic pathways, proteins apoptosis-inducing factor (AIF) and endonuclease G (EndoG) are released from the mitochondria and they translocate into the cell nuclei, where they probably participate in chromatin degradation together with other nuclear proteins. Exact mechanism of EndoG activity in cell nucleus is still unknown. Some interacting partners like flap endonuclease 1, DNase I, and exonuclease III were already suggested, but also other interacting partners were proposed. We conducted a living-cell confocal fluorescence microscopy followed by an image analysis of fluorescence resonance energy transfer to analyze the possibility of protein interactions of EndoG with histone H2B and human DNA topoisomerase II alpha (TOPO2a). Our results show that EndoG interacts with both these proteins during apoptotic cell death. Therefore, we can conclude that EndoG and TOPO2a may actively participate in apoptotic chromatin degradation. The possible existence of a degradation complex consisting of EndoG and TOPO2a and possibly other proteins like AIF and cyclophilin A have yet to be investigated.

Knockdown of apoptosis-inducing factor disrupts function of respiratory complex I.

Recent findings suggest that apoptotic protein apoptosis-inducing factor (AIF) may also play an important non-apoptotic function inside mitochondria. AIF was proposed to be an important component of respiratory chain complex I that is the major producer of superoxide radical. The possible role of AIF is still controversial. Superoxide production could be used as a valuable measure of complex I function, because the majority of superoxide is produced there. Therefore, we employed superoxide-specific mitochondrial fluorescence dye for detection of superoxide production. We studied an impact of AIF knockdown on function of mitochondrial complex I by analyzing superoxide production in selected cell lines. Our results show that tumoral telomerase-positive (TP) AIF knockdown cell lines display significant increase in superoxide production in comparison to control cells, while a non-tumoral cell line and tumoral telomerase-negative cell lines with alternative lengthening of telomeres (ALT) show a decrease in superoxide production. According to these results, we can conclude that AIF knockdown disrupts function of complex I and therefore increases the superoxide production in mitochondria. The distinct effect of AIF depletion in various cell lines could result from recently discovered activity of telomerase in mitochondria of TP cancer cells, but this hypothesis needs further investigation.

SUV39h-independent association of HP1 beta with fibrillarin-positive nucleolar regions.

Heterochromatin protein 1 (HP1), which binds to sites of histone H3 lysine 9 (H3K9) methylation, is primarily responsible for gene silencing and the formation of heterochromatin. We observed that HP1 beta is located in both the chromocenters and fibrillarin-positive nucleoli interiors. However, HP1 alpha and HP1 gamma occupied fibrillarin-positive compartments to a lesser extent, corresponding to the distinct levels of HP1 subtypes at the promoter of rDNA genes. Deficiency of histone methyltransferases SUV39h and/or inhibition of histone deacetylases (HDACi) decreased HP1 beta and H3K9 trimethylation at chromocenters, but not in fibrillarin-positive regions that co-localized with RNA polymerase I. Similarly, SUV39h- and HDACi-dependent nucleolar rearrangement and inhibition of rDNA transcription did not affect the association between HP1 beta and fibrillarin. Moreover, the presence of HP1 beta in nucleoli is likely connected with transcription of ribosomal genes and with the role of fibrillarin in nucleolar processes.

Chromocentre integrity and epigenetic marks.

The epigenetic modification of histones dictates the formation of euchromatin and heterochromatin domains. We studied the effects of a deficiency of histone methyltransferase, SUV39h, and trichostatin A-dependent hyperacetylation on the structural stability of centromeric clusters, called chromocentres. We did not observe the expected disintegration of chromocentres, but both SUV39h deficiency and hyperacetylation in SUV39h+/+ cells induced the re-positioning of chromocentres closer to the nuclear periphery. Conversely, TSA treatment of SUV39h-/- cells re-established normal nuclear radial positioning of chromocentres. This structural re-arrangement was likely caused by several epigenetic events at centromeric heterochromatin. In particular, reciprocal exchanges between H3K9me1, H3K9me2, H3K9me3, DNA methylation, and HP1 protein levels influenced chromocentre nuclear composition. For example, H3K9me1 likely substituted for the function of H3K9me3 in chromocentre nuclear arrangement and compaction. Our results illustrate the important and interchangeable roles of epigenetic marks for chromocentre integrity. Therefore, we propose a model for epigenetic regulation of nuclear stability of centromeric heterochromatin in the mouse genome.

SUV39h- and A-type lamin-dependent telomere nuclear rearrangement.

Telomeres are specialized chromatin structures that are situated at the end of linear chromosomes and play an important role in cell senescence and immortalization. Here, we investigated whether changes in histone signature influence the nuclear arrangement and positioning of telomeres. Analysis of mouse embryonic fibroblasts revealed that telomeres were organized into specific clusters that partially associated with centromeric clusters. This nuclear arrangement was influenced by deficiency of the histone methyltransferase SUV39h, LMNA deficiency, and the histone deacetylase inhibitor Trichostatin A (TSA). Similarly, nuclear radial distributions of telomeric clusters were preferentially influenced by TSA, which caused relocation of telomeres closer to the nuclear center. Telomeres also co-localized with promyelocytic leukemia bodies (PML). This association was increased by SUV39h deficiency and decreased by LMNA deficiency. These differences could be explained by differing levels of the telomerase subunit, TERT, in SUV39h- and LMNA-deficient fibroblasts. Taken together, our data show that SUV39h and A-type lamins likely play a key role in telomere maintenance and telomere nuclear architecture.

Prediction of localization and interactions of apoptotic proteins.

During apoptosis several mitochondrial proteins are released. Some of them participate in caspase-independent nuclear DNA degradation, especially apoptosis-inducing factor (AIF) and endonuclease G (endoG). Another interesting protein, which was expected to act similarly as AIF due to the high sequence homology with AIF is AIF-homologous mitochondrion-associated inducer of death (AMID). We studied the structure, cellular localization, and interactions of several proteins in silico and also in cells using fluorescent microscopy. We found the AMID protein to be cytoplasmic, most probably incorporated into the cytoplasmic side of the lipid membranes. Bioinformatic predictions were conducted to analyze the interactions of the studied proteins with each other and with other possible partners. We conducted molecular modeling of proteins with unknown 3D structures. These models were then refined by MolProbity server and employed in molecular docking simulations of interactions. Our results show data acquired using a combination of modern in silico methods and image analysis to understand the localization, interactions and functions of proteins AMID, AIF, endonuclease G, and other apoptosis-related proteins.

Deep-Learning-Based Segmentation of Small Extracellular Vesicles in Transmission Electron Microscopy Images.

Small extracellular vesicles (sEVs) are cell-derived vesicles of nanoscale size (~30-200 nm) that function as conveyors of information between cells, reflecting the cell of their origin and its physiological condition in their content. Valuable information on the shape and even on the composition of individual sEVs can be recorded using transmission electron microscopy (TEM). Unfortunately, sample preparation for TEM image acquisition is a complex procedure, which often leads to noisy images and renders automatic quantification of sEVs an extremely difficult task. We present a completely deep-learning-based pipeline for the segmentation of sEVs in TEM images. Our method applies a residual convolutional neural network to obtain fine masks and use the Radon transform for splitting clustered sEVs. Using three manually annotated datasets that cover a natural variability typical for sEV studies, we show that the proposed method outperforms two different state-of-the-art approaches in terms of detection and segmentation performance. Furthermore, the diameter and roundness of the segmented vesicles are estimated with an error of less than 10%, which supports the high potential of our method in biological applications.