Cell segmentation in images without structural fluorescent labels.

High-content screening (HCS) provides an excellent tool to understand the mechanism of action of drugs on disease-relevant model systems. Careful selection of fluorescent labels (FLs) is crucial for successful HCS assay development. HCS assays typically comprise (a) FLs containing biological information of interest, and (b) additional structural FLs enabling instance segmentation for downstream analysis. However, the limited number of available fluorescence microscopy imaging channels restricts the degree to which these FLs can be experimentally multiplexed. In this article, we present a segmentation workflow that overcomes the dependency on structural FLs for image segmentation, typically freeing two fluorescence microscopy channels for biologically relevant FLs. It consists in extracting structural information encoded within readouts that are primarily biological, by fine-tuning pre-trained state-of-the-art generalist cell segmentation models for different combinations of individual FLs, and aggregating the respective segmentation results together. Using annotated datasets that we provide, we confirm our methodology offers improvements in performance and robustness across several segmentation aggregation strategies and image acquisition methods, over different cell lines and various FLs. It thus enables the biological information content of HCS assays to be maximized without compromising the robustness and accuracy of computational single-cell profiling.

High-content imaging-based pooled CRISPR screens in mammalian cells.

CRISPR (clustered regularly interspaced short palindromic repeats)-based gene inactivation provides a powerful means for linking genes to particular cellular phenotypes. CRISPR-based screening typically uses large genomic pools of single guide RNAs (sgRNAs). However, this approach is limited to phenotypes that can be enriched by chemical selection or FACS sorting. Here, we developed a microscopy-based approach, which we name optical enrichment, to select cells displaying a particular CRISPR-induced phenotype by automated imaging-based computation, mark them by photoactivation of an expressed photoactivatable fluorescent protein, and then isolate the fluorescent cells using fluorescence-activated cell sorting (FACS). A plugin was developed for the open source software µManager to automate the phenotypic identification and photoactivation of cells, allowing ∼1.5 million individual cells to be screened in 8 h. We used this approach to screen 6,092 sgRNAs targeting 544 genes for their effects on nuclear size regulation and identified 14 bona fide hits. These results present a scalable approach to facilitate imaging-based pooled CRISPR screens.

Doublecortin Is Excluded from Growing Microtubule Ends and Recognizes the GDP-Microtubule Lattice.

Many microtubule (MT) functions are mediated by a diverse class of proteins (+TIPs) at growing MT plus ends that control intracellular MT interactions and dynamics and depend on end-binding proteins (EBs) [1]. Cryoelectron microscopy has recently identified the EB binding site as the interface of four tubulin dimers that undergoes a conformational change in response to ß-tubulin GTP hydrolysis [2, 3]. Doublecortin (DCX), a MT-associated protein (MAP) required for neuronal migration during cortical development [4, 5], binds to the same site as EBs [6], and recent in vitro studies proposed DCX localization to growing MT ends independent of EBs [7]. Because this conflicts with observations in neurons [8, 9] and the molecular function of DCX is not well understood, we revisited intracellular DCX dynamics at low expression levels. Here, we report that DCX is not a +TIP in cells but, on the contrary, is excluded from the EB1 domain. In addition, we find that DCX-MT interactions are highly sensitive to MT geometry. In cells, DCX binding was greatly reduced at MT segments with high local curvature. Remarkably, this geometry-dependent binding to MTs was completely reversed in the presence of taxanes, which reconciles incompatible observations in cells [9] and in vitro [10]. We propose a model explaining DCX specificity for different MT geometries based on structural changes induced by GTP hydrolysis that decreases the spacing between adjacent tubulin dimers [11]. Our data are consistent with a unique mode of MT interaction in which DCX specifically recognizes this compacted GDP-like MT lattice.

Reed-Sternberg cells form by abscission failure in the presence of functional Aurora B kinase.

Large multinucleated Reed-Sternberg cells (RS) and large mononucleated Hodgkin cells (H) are traditionally considered to be the neoplastic population in classical Hodgkin lymphoma, (cHL) and postulated to promote the disease. However, the contribution of these larger cells to the progression of cHL remains debatable. We used established cHL cell lines and cHL cellular fractions composed of small mononucleated cells only or enriched in large RS/H cells to investigate RS/H cell origin and to characterize the cells which they derive from. We confirm that the small mononucleated cells give rise to RS/H cells, and we show that the latter proliferate significantly more slowly than the small cells. By using live-cell imaging, we demonstrate that binucleated RS cells are generated by failure of abscission when a few small cells attempt to divide. Finally, our results reveal that the small mononucleated cells are chromosomally unstable, but this is unlikely to be related to a malfunctioning chromosomal passenger protein complex. We propose that the small mononucleated cells, rather than the RS/H cells, are the main drivers of cHL.

Induction of focal adhesions and motility in Drosophila S2 cells.

Focal adhesions are dynamic structures that interact with the extracellular matrix on the cell exterior and actin filaments on the cell interior, enabling cells to adhere and crawl along surfaces. We describe a system for inducing the formation of focal adhesions in normally non-ECM-adherent, nonmotile Drosophila S2 cells. These focal adhesions contain the expected molecular markers such as talin, vinculin, and p130Cas, and they require talin for their formation. The S2 cells with induced focal adhesions also display a nonpolarized form of motility on vitronectin-coated substrates. Consistent with findings in mammalian cells, the degree of motility can be tuned by changing the stiffness of the substrate and was increased after the depletion of PAK3, a p21-activated kinase. A subset of nonmotile, nonpolarized cells also exhibited focal adhesions that rapidly assembled and disassembled around the cell perimeter. Such cooperative and dynamic fluctuations of focal adhesions were decreased by RNA interference (RNAi) depletion of myosin II and focal adhesion kinase, suggesting that this behavior requires force and focal adhesion maturation. These results demonstrate that S2 cells, a cell line that is well studied for cytoskeletal dynamics and readily amenable to protein manipulation by RNAi, can be used to study the assembly and dynamics of focal adhesions and mechanosensitive cell motility.

Condensin regulates the stiffness of vertebrate centromeres.

When chromosomes are aligned and bioriented at metaphase, the elastic stretch of centromeric chromatin opposes pulling forces exerted on sister kinetochores by the mitotic spindle. Here we show that condensin ATPase activity is an important regulator of centromere stiffness and function. Condensin depletion decreases the stiffness of centromeric chromatin by 50% when pulling forces are applied to kinetochores. However, condensin is dispensable for the normal level of compaction (rest length) of centromeres, which probably depends on other factors that control higher-order chromatin folding. Kinetochores also do not require condensin for their structure or motility. Loss of stiffness caused by condensin-depletion produces abnormal uncoordinated sister kinetochore movements, leads to an increase in Mad2(+) kinetochores near the metaphase plate and delays anaphase onset.

A promoter-hijack strategy for conditional shutdown of multiply spliced essential cell cycle genes.

We describe a method for the isolation of conditional knockouts of essential multiply spliced genes in which the entire body of the gene downstream of the ATG start codon is left untouched but can be switched off rapidly and completely by adding tetracycline to the culture medium. The approach centers on a "promoter-hijack" strategy in which the gene's promoter is replaced with a minimal promoter responsive to the tetracycline-repressible transactivator (tTA). Elsewhere in the genome, a cloned fragment of the gene's promoter is used to drive expression of a tTA. Thus, the gene is essentially regulated by its own promoter but through the intermediary tTA. Using this strategy, we generated a conditional knockout of chromokinesin KIF4A, an important mitotic effector protein whose mRNA is multiply spliced and whose cDNA is highly toxic when overexpressed in cells. We used chicken DT40 cells, but the same strategy should be applicable to ES cells and, eventually, to mice.

Condensin and Repo-Man-PP1 co-operate in the regulation of chromosome architecture during mitosis.

The reversible condensation of chromosomes during cell division remains a classic problem in cell biology. Condensation requires the condensin complex in certain experimental systems, but not in many others. Anaphase chromosome segregation almost always fails in condensin-depleted cells, leading to the formation of prominent chromatin bridges and cytokinesis failure. Here, live-cell analysis of chicken DT40 cells bearing a conditional knockout of condensin subunit SMC2 revealed that condensin-depleted chromosomes abruptly lose their compact architecture during anaphase and form massive chromatin bridges. The compact chromosome structure can be preserved and anaphase chromosome segregation rescued by preventing the targeting subunit Repo-Man from recruiting protein phosphatase 1 (PP1) to chromatin at anaphase onset. This study identifies an activity critical for mitotic chromosome structure that is inactivated by Repo-Man-PP1 during anaphase. This activity, provisionally termed 'regulator of chromosome architecture' (RCA), cooperates with condensin to preserve the characteristic chromosome architecture during mitosis.

Molecular characterization of PDGFR-alpha/PDGF-A and c-KIT/SCF in gliosarcomas.

Gliosarcomas are rare and poorly characterized malignant brain tumors that exhibit a biphasic tissue pattern with areas of gliomatous and sarcomatous differentiation. These tumors are histological variants of glioblastoma, displaying a similar genetic profile and dismal prognosis. Up-regulation of PDGFR subfamily of tyrosine kinase members, PDGFR-alpha and c-Kit, and their intracellular effectors RAS/RAF/MAPK has a crucial role in the cancer development. In addition, signal transduction mediated by activating mutations of c-Kit and PDGFR can be effectively blocked by specific tyrosine kinase inhibitors, such as Imatinib mesylate. The aim of this study was to characterize the molecular alterations of PDGFR signaling in gliosarcomas. Six cases were analyzed by immunohistochemistry for the expression of PDGFR-alpha, c-Kit and their ligands PDGF-A and SCF, respectively. The cases were further evaluated for the presence of activating mutations of PDGFR-alpha (exons 12 and 18) and c-kit (exons 9, 11, 13, and 17), as well as B-RAF (exons 11 and 15). Expression of PDGF-A was found in all cases and co-expression of PDGFR-alpha was observed in three cases. Four cases showed expression of SCF, and c-Kit was observed only in one case that also expressed SCF. Generally, immunoreaction predominates in the glial component. The mutational analysis of PDGFR-alpha showed the presence of an IVS17-50insT intronic insertion in two cases, one of them also with a 2472C > T silent mutation; this silent mutation was also found in another case. Glioma cell line analysis of IVS17-50insT insertion showed no influence on PDGFR-alpha gene splicing. No mutations were detected in c-kit and B-RAF oncogenes. Our results indicate that activating mutations of PDGFR-alpha, c-kit and B-RAF are absent in gliosarcomas. Nevertheless, the presence of a PDGFR-a/PDGFA and c-Kit/SCF autocrine/paracrine stimulation loop in a proportion of cases, supports the potential role of specific tyrosine kinase inhibitors in the treatment of gliosarcomas.