SC-Track: a robust cell-tracking algorithm for generating accurate single-cell lineages from diverse cell segmentations.

Computational analysis of fluorescent timelapse microscopy images at the single-cell level is a powerful approach to study cellular changes that dictate important cell fate decisions. Core to this approach is the need to generate reliable cell segmentations and classifications necessary for accurate quantitative analysis. Deep learning-based convolutional neural networks (CNNs) have emerged as a promising solution to these challenges. However, current CNNs are prone to produce noisy cell segmentations and classifications, which is a significant barrier to constructing accurate single-cell lineages. To address this, we developed a novel algorithm called Single Cell Track (SC-Track), which employs a hierarchical probabilistic cache cascade model based on biological observations of cell division and movement dynamics. Our results show that SC-Track performs better than a panel of publicly available cell trackers on a diverse set of cell segmentation types. This cell-tracking performance was achieved without any parameter adjustments, making SC-Track an excellent generalized algorithm that can maintain robust cell-tracking performance in varying cell segmentation qualities, cell morphological appearances and imaging conditions. Furthermore, SC-Track is equipped with a cell class correction function to improve the accuracy of cell classifications in multiclass cell segmentation time series. These features together make SC-Track a robust cell-tracking algorithm that works well with noisy cell instance segmentation and classification predictions from CNNs to generate accurate single-cell lineages and classifications.

Elevated basal AMP-activated protein kinase activity sensitizes colorectal cancer cells to growth inhibition by metformin.

Expression and activity of the AMP-activated protein kinase (AMPK) α1 catalytic subunit of the heterotrimeric kinase significantly correlates with poor outcome for colorectal cancer patients. Hence there is considerable interest in uncovering signalling vulnerabilities arising from this oncogenic elevation of AMPKα1 signalling. We have therefore attenuated mammalian target of rapamycin (mTOR) control of AMPKα1 to generate a mutant colorectal cancer in which AMPKα1 signalling is elevated because AMPKα1 serine 347 cannot be phosphorylated by mTORC1. The elevated AMPKα1 signalling in this HCT116 α1.S347A cell line confers hypersensitivity to growth inhibition by metformin. Complementary chemical approaches confirmed this relationship in both HCT116 and the genetically distinct HT29 colorectal cells, as AMPK activators imposed vulnerability to growth inhibition by metformin in both lines. Growth inhibition by metformin was abolished when AMPKα1 kinase was deleted. We conclude that elevated AMPKα1 activity modifies the signalling architecture in such a way that metformin treatment compromises cell proliferation. Not only does this mutant HCT116 AMPKα1-S347A line offer an invaluable resource for future studies, but our findings suggest that a robust biomarker for chronic AMPKα1 activation for patient stratification could herald a place for the well-tolerated drug metformin in colorectal cancer therapy.

pcnaDeep: a fast and robust single-cell tracking method using deep-learning mediated cell cycle profiling.

SUMMARY: Computational methods that track single cells and quantify fluorescent biosensors in time-lapse microscopy images have revolutionized our approach in studying the molecular control of cellular decisions. One barrier that limits the adoption of single-cell analysis in biomedical research is the lack of efficient methods to robustly track single cells over cell division events. Here, we developed an application that automatically tracks and assigns mother-daughter relationships of single cells. By incorporating cell cycle information from a well-established fluorescent cell cycle reporter, we associate mitosis relationships enabling high fidelity long-term single-cell tracking. This was achieved by integrating a deep-learning-based fluorescent proliferative cell nuclear antigen signal instance segmentation module with a cell tracking and cell cycle resolving pipeline. The application offers a user-friendly interface and extensible APIs for customized cell cycle analysis and manual correction for various imaging configurations. AVAILABILITY AND IMPLEMENTATION: pcnaDeep is an open-source Python application under the Apache 2.0 licence. The source code, documentation and tutorials are available at https://github.com/chan-labsite/PCNAdeep. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

An updated view on the centrosome as a cell cycle regulator.

The centrosome is a multifunctional organelle that is known primarily for its microtubule organising function. Centrosomal defects caused by changes in centrosomal structure or number have been associated with human diseases ranging from congenital defects to cancer. We are only beginning to appreciate how the non-microtubule organising roles of the centrosome are related to these clinical conditions. In this review, we will discuss the historical evidence that led to the proposal that the centrosome participates in cell cycle regulation. We then summarize the body of work that describes the involvement of the mammalian centrosome in triggering cell cycle progression and checkpoint signalling. Then we will highlight work from the fission yeast model organism, revealing the molecular details that explain how the spindle pole body (SPB, the yeast functional equivalent of the centrosome), participates in these cell cycle transitions. Importantly, we will discuss some of the emerging questions from recent discoveries related to the role of the centrosome as a cell cycle regulator.

Cell cycle heterogeneity directs spontaneous 2C state entry and exit in mouse embryonic stem cells.

Mouse embryonic stem cells (ESCs) show cell-to-cell heterogeneity. A small number of two-cell-like cells (2CLCs) marked by endogenous retrovirus activation emerge spontaneously. The 2CLCs are unstable and they are prone to transiting back to the pluripotent state without extrinsic stimulus. To understand how this bidirectional transition takes place, we performed single-cell RNA sequencing on isolated 2CLCs that underwent 2C-like state exit and re-entry, and revealed a step-by-step transitional process between 2C-like and pluripotent states. Mechanistically, we found that cell cycle played an important role in mediating these transitions by regulating assembly of the nucleolus and peri-nucleolar heterochromatin to influence 2C gene Dux expression. Collectively, our findings provide a roadmap of the 2C-like state entry and exit in ESCs and also a causal role of the cell cycle in promoting these transitions.

Dialogue between centrosomal entrance and exit scaffold pathways regulates mitotic commitment.

The fission yeast scaffold molecule Sid4 anchors the septum initiation network to the spindle pole body (SPB, centrosome equivalent) to control mitotic exit events. A second SPB-associated scaffold, Cut12, promotes SPB-associated Cdk1-cyclin B to drive mitotic commitment. Signals emanating from each scaffold have been assumed to operate independently to promote two distinct outcomes. We now find that signals from Sid4 contribute to the Cut12 mitotic commitment switch. Specifically, phosphorylation of Sid4 by NIMAFin1 reduces Sid4 affinity for its SPB anchor, Ppc89, while also enhancing Sid4's affinity for casein kinase 1δ (CK1δ). The resulting phosphorylation of Sid4 by the newly docked CK1δ recruits Chk2Cds1 to Sid4. Chk2Cds1 then expels the Cdk1-cyclin B antagonistic phosphatase Flp1/Clp1 from the SPB. Flp1/Clp1 departure can then support mitotic commitment when Cdk1-cyclin B activation at the SPB is compromised by reduction of Cut12 function. Such integration of signals emanating from neighboring scaffolds shows how centrosomes/SPBs can integrate inputs from multiple pathways to control cell fate.

Removal of centrosomal PP1 by NIMA kinase unlocks the MPF feedback loop to promote mitotic commitment in S. pombe.

BACKGROUND: Activation of the Cdk1/cyclin B complex, also known as mitosis-promoting factor (MPF), drives commitment to mitosis. Interphase MPF is inhibited through phosphorylation of Cdk1 by Wee1-related kinases. Because Cdc25 phosphatases remove this phosphate, Cdc25 activity is an essential part of the switch that drives cells into mitosis. The generation of a critical "trigger" of active MPF promotes a positive feedback loop that employs Polo kinase to boost Cdc25 activity and inhibit Wee1, thereby ensuring that mitotic commitment is a bistable switch. Mutations in the spindle pole body (SPB) component Cut12 suppress otherwise lethal deficiencies in Cdc25. RESULTS: Cut12 harbors a bipartite protein phosphatase 1 (PP1) docking domain. Mutation of either element alone suppressed the temperature-dependent lethality of cdc25.22, whereas simultaneous ablation of both allowed cells to divide in the complete absence of Cdc25. Late G2 phase phosphorylation between the two elements by MPF and the NIMA kinase Fin1 blocked PP1(Dis2) recruitment, thereby promoting recruitment of Polo to Cut12 and the SPB and elevating global Polo kinase activity throughout the cell. CONCLUSIONS: PP1 recruitment to Cut12 sets a threshold for Polo's feedback-loop activity that locks the cell in interphase until Cdc25 pushes MPF activity through this barrier to initiate mitosis. We propose that events on the SPB (and, by inference, the centrosome) integrate inputs from diverse signaling networks to generate a coherent decision to divide that is appropriate for the particular environmental context of each cell. PP1 recruitment sets one or more critical thresholds for single or multiple local events within this switch.

Centrosomal MPF triggers the mitotic and morphogenetic switches of fission yeast.

Activation of mitosis-promoting factor (MPF) drives mitotic commitment. In human cells active MPF appears first on centrosomes. We show that local activation of MPF on the equivalent organelle of fission yeast, the spindle pole body (SPB), promotes Polo kinase activity at the SPBs long before global MPF activation drives mitotic commitment. Artificially promoting MPF or Polo activity at various locations revealed that this local control of Plo1 activity on G2 phase SPBs dictates the timing of mitotic commitment. Cytokinesis of the rod-shaped fission yeast cell generates a naive, new, cell end. Growth is restricted to the experienced old end until a point in G2 phase called new end take off (NETO) when bipolar growth is triggered. NETO coincided with MPF activation of Plo1 on G2 phase SPBs (ref. 4). Both MPF and Polo activities were required for NETO and both induced NETO when ectopically activated at interphase SPBs. NETO promotion by MPF required polo. Thus, local MPF activation on G2 SPBs directs polo kinase to control at least two distinct and temporally separated, cell-cycle transitions at remote locations.

Functional characterisation and drug target validation of a mitotic kinesin-13 in Trypanosoma brucei.

Mitotic kinesins are essential for faithful chromosome segregation and cell proliferation. Therefore, in humans, kinesin motor proteins have been identified as anti-cancer drug targets and small molecule inhibitors are now tested in clinical studies. Phylogenetic analyses have assigned five of the approximately fifty kinesin motor proteins coded by Trypanosoma brucei genome to the Kinesin-13 family. Kinesins of this family have unusual biochemical properties because they do not transport cargo along microtubules but are able to depolymerise microtubules at their ends, therefore contributing to the regulation of microtubule length. In other eukaryotic genomes sequenced to date, only between one and three Kinesin-13s are present. We have used immunolocalisation, RNAi-mediated protein depletion, biochemical in vitro assays and a mouse model of infection to study the single mitotic Kinesin-13 in T. brucei. Subcellular localisation of all five T. brucei Kinesin-13s revealed distinct distributions, indicating that the expansion of this kinesin family in kinetoplastids is accompanied by functional diversification. Only a single kinesin (TbKif13-1) has a nuclear localisation. Using active, recombinant TbKif13-1 in in vitro assays we experimentally confirm the depolymerising properties of this kinesin. We analyse the biological function of TbKif13-1 by RNAi-mediated protein depletion and show its central role in regulating spindle assembly during mitosis. Absence of the protein leads to abnormally long and bent mitotic spindles, causing chromosome mis-segregation and cell death. RNAi-depletion in a mouse model of infection completely prevents infection with the parasite. Given its essential role in mitosis, proliferation and survival of the parasite and the availability of a simple in vitro activity assay, TbKif13-1 has been identified as an excellent potential drug target.

The role of the Kinesin-13 family protein TbKif13-2 in flagellar length control of Trypanosoma brucei.

TbKif13-2, a member of the microtubule-depolymerising Kinesin-13 family was localised at the tip of the flagellum in Trypanosoma brucei. Its predicted activity suggested a role in the regulation of axonemal length. However, using gene deletion and overexpression of TbKif13-2 we show that, in procyclic T. brucei, this kinesin has only a very limited effect on flagellar length. Gene deletion resulted in no significant elongation of the flagellum and overexpression only slightly decreased flagellar length and the rate of growth of a new flagellum during cell division. This is in contrast to studies in Leishmania major, where overexpression of the TbKif13-2 homologue resulted in a significant length reduction of the flagellum. Knockout of TbKif13-2 has, however, an effect on the initial growth of the emerging new flagellum. In conclusion, we show that TbKif13-2 has only a marginal impact on flagellar length in T. brucei.