Automated Cell Lineage Reconstruction using Label-Free 4D Microscopy.

Here we describe embGAN, a deep learning pipeline that addresses the challenge of automated cell detection and tracking in label-free 3D time lapse imaging. embGAN requires no manual data annotation for training, learns robust detections that exhibits a high degree of scale invariance and generalizes well to images acquired in multiple labs on multiple instruments.

Novel metrics reveal new structure and unappreciated heterogeneity in Caenorhabditis elegans development.

High throughput experimental approaches are increasingly allowing for the quantitative description of cellular and organismal phenotypes. Distilling these large volumes of complex data into meaningful measures that can drive biological insight remains a central challenge. In the quantitative study of development, for instance, one can resolve phenotypic measures for single cells onto their lineage history, enabling joint consideration of heritable signals and cell fate decisions. Most attempts to analyze this type of data, however, discard much of the information content contained within lineage trees. In this work we introduce a generalized metric, which we term the branch edit distance, that allows us to compare any two embryos based on phenotypic measurements in individual cells. This approach aligns those phenotypic measurements to the underlying lineage tree, providing a flexible and intuitive framework for quantitative comparisons between, for instance, Wild-Type (WT) and mutant developmental programs. We apply this novel metric to data on cell-cycle timing from over 1300 WT and RNAi-treated Caenorhabditis elegans embryos. Our new metric revealed surprising heterogeneity within this data set, including subtle batch effects in WT embryos and dramatic variability in RNAi-induced developmental phenotypes, all of which had been missed in previous analyses. Further investigation of these results suggests a novel, quantitative link between pathways that govern cell fate decisions and pathways that pattern cell cycle timing in the early embryo. Our work demonstrates that the branch edit distance we propose, and similar metrics like it, have the potential to revolutionize our quantitative understanding of organismal phenotype.

Novel metrics reveal new structure and unappreciated heterogeneity in C. elegans development.

High throughput experimental approaches are increasingly allowing for the quantitative description of cellular and organismal phenotypes. Distilling these large volumes of complex data into meaningful measures that can drive biological insight remains a central challenge. In the quantitative study of development, for instance, one can resolve phenotypic measures for single cells onto their lineage history, enabling joint consideration of heritable signals and cell fate decisions. Most attempts to analyze this type of data, however, discard much of the information content contained within lineage trees. In this work we introduce a generalized metric, which we term the branch distance, that allows us to compare any two embryos based on phenotypic measurements in individual cells. This approach aligns those phenotypic measurements to the underlying lineage tree, providing a flexible and intuitive framework for quantitative comparisons between, for instance, Wild-Type (WT) and mutant developmental programs. We apply this novel metric to data on cell-cycle timing from over 1300 WT and RNAi-treated Caenorhabditis elegans embryos. Our new metric revealed surprising heterogeneity within this data set, including subtle batch effects in WT embryos and dramatic variability in RNAi-induced developmental phenotypes, all of which had been missed in previous analyses. Further investigation of these results suggests a novel, quantitative link between pathways that govern cell fate decisions and pathways that pattern cell cycle timing in the early embryo. Our work demonstrates that the branch distance we propose, and similar metrics like it, have the potential to revolutionize our quantitative understanding of organismal phenotype.

Microparticle traction force microscopy reveals subcellular force exertion patterns in immune cell-target interactions.

Force exertion is an integral part of cellular behavior. Traction force microscopy (TFM) has been instrumental for studying such forces, providing spatial force measurements at subcellular resolution. However, the applications of classical TFM are restricted by the typical planar geometry. Here, we develop a particle-based force sensing strategy for studying cellular interactions. We establish a straightforward batch approach for synthesizing uniform, deformable and tuneable hydrogel particles, which can also be easily derivatized. The 3D shape of such particles can be resolved with superresolution (<50 nm) accuracy using conventional confocal microscopy. We introduce a reference-free computational method allowing inference of traction forces with high sensitivity directly from the particle shape. We illustrate the potential of this approach by revealing subcellular force patterns throughout phagocytic engulfment and force dynamics in the cytotoxic T-cell immunological synapse. This strategy can readily be adapted for studying cellular forces in a wide range of applications.

PCP and SAX-3/Robo Pathways Cooperate to Regulate Convergent Extension-Based Nerve Cord Assembly in C. elegans.

Formation and resolution of multicellular rosettes can drive convergent extension (CE) type cell rearrangements during tissue morphogenesis. Rosette dynamics are regulated by both planar cell polarity (PCP)-dependent and -independent pathways. Here we show that CE is involved in ventral nerve cord (VNC) assembly in Caenorhabditis elegans. We show that a VANG-1/Van Gogh and PRKL-1/Prickle containing PCP pathway and a Slit-independent SAX-3/Robo pathway cooperate to regulate, via rosette intermediaries, the intercalation of post-mitotic neuronal cell bodies during VNC formation. We show that VANG-1 and SAX-3 are localized to contracting edges and rosette foci and act to specify edge contraction during rosette formation and to mediate timely rosette resolution. Simultaneous loss of both pathways severely curtails CE resulting in a shortened, anteriorly displaced distribution of VNC neurons at hatching. Our results establish rosette-based CE as an evolutionarily conserved mechanism of nerve cord morphogenesis and reveal a role for SAX-3/Robo in this process.

The Regulatory Landscape of Lineage Differentiation in a Metazoan Embryo.

Elucidating the mechanism of cell lineage differentiation is critical for our understanding of development and fate manipulation. Here we combined systematic perturbation and direct lineaging to map the regulatory landscape of lineage differentiation in early C. elegans embryogenesis. High-dimensional phenotypic analysis of 204 essential genes in 1,368 embryos revealed that cell lineage differentiation follows a canalized landscape with barriers shaped by lineage distance and genetic robustness. We assigned function to 201 genes in regulating lineage differentiation, including 175 switches of binary fate choices. We generated a multiscale model that connects gene networks and cells to the experimentally mapped landscape. Simulations showed that the landscape topology determines the propensity of differentiation and regulatory complexity. Furthermore, the model allowed us to identify the chromatin assembly complex CAF-1 as a context-specific repressor of Notch signaling. Our study presents a systematic survey of the regulatory landscape of lineage differentiation of a metazoan embryo.

Small sample sorting of primary adherent cells by automated micropallet imaging and release.

Primary patient samples are the gold standard for molecular investigations of tumor biology yet are difficult to acquire, heterogeneous in nature and variable in size. Patient-derived xenografts (PDXs) comprised of primary tumor tissue cultured in host organisms such as nude mice permit the propagation of human tumor samples in an in vivo environment and closely mimic the phenotype and gene expression profile of the primary tumor. Although PDX models reduce the cost and complexity of acquiring sample tissue and permit repeated sampling of the primary tumor, these samples are typically contaminated by immune, blood, and vascular tissues from the host organism while also being limited in size. For very small tissue samples (on the order of 10(3) cells) purification by fluorescence-activated cell sorting (FACS) is not feasible while magnetic activated cell sorting (MACS) of small samples results in very low purity, low yield, and poor viability. We developed a platform for imaging cytometry integrated with micropallet array technology to perform automated cell sorting on very small samples obtained from PDX models of pancreatic and colorectal cancer using antibody staining of EpCAM (CD326) as a selection criteria. These data demonstrate the ability to automate and efficiently separate samples with very low number of cells.

Dynamics and evolution of ß-catenin-dependent Wnt signaling revealed through massively parallel clonogenic screening.

Wnt/ß-catenin signaling is of significant interest due to the roles it plays in regulating development, tissue regeneration and disease. Transcriptional reporters have been widely employed to study Wnt/ß-catenin signal transduction in live cells and whole organisms and have been applied to understanding embryonic development, exploring oncogenesis and developing therapeutics. Polyclonal heterogeneity in reporter cell lines has historically been seen as a challenge to be overcome in the development of novel cell lines and reporter-based assays, and monoclonal reporter cell lines are commonly employed to reduce this variability. A375 cell lines infected with a reporter for Wnt/ß-catenin signaling were screened over short (<6) and long (>25) generational timescales. To characterize phenotypic divergence over these time-scales, a microfabricated cell array-based screen was developed enabling characterization of 1119 clonal colonies in parallel. This screen revealed phenotypic divergence after <6 generations at a similar scale to that observed in monoclonal cell lines cultured for >25 generations. Not only were reporter dynamics observed to diverge widely, but monoclonal cell lines were observed with seemingly opposite signaling phenotypes. Additionally, these observations revealed a generational-dependent trend in Wnt signaling in A375 cells that provides insight into the pathway's mechanisms of positive feedback and self-inhibition.

Capture and 3D culture of colonic crypts and colonoids in a microarray platform.

Crypts are the basic structural and functional units of colonic epithelium and can be isolated from the colon and cultured in vitro into multi-cell spheroids termed "colonoids". Both crypts and colonoids are ideal building blocks for construction of an in vitro tissue model of the colon. Here we proposed and tested a microengineered platform for capture and in vitro 3D culture of colonic crypts and colonoids. An integrated platform was fabricated from polydimethylsiloxane which contained two fluidic layers separated by an array of cylindrical microwells (150 µm diameter, 150 µm depth) with perforated bottoms (30 µm opening, 10 µm depth) termed "microstrainers". As fluid moved through the array, crypts or colonoids were retained in the microstrainers with a >90% array-filling efficiency. Matrigel as an extracellular matrix was then applied to the microstrainers to generate isolated Matrigel pockets encapsulating the crypts or colonoids. After supplying the essential growth factors, epidermal growth factor, Wnt-3A, R-spondin 2 and noggin, 63 ± 13% of the crypts and 77 ± 8% of the colonoids cultured in the microstrainers over a 48-72 h period formed viable 3D colonoids. Thus colonoid growth on the array was similar to that under standard culture conditions (78 ± 5%). Additionally the colonoids displayed the same morphology and similar numbers of stem and progenitor cells as those under standard culture conditions. Immunofluorescence staining confirmed that the differentiated cell-types of the colon, goblet cells, enteroendocrine cells and absorptive enterocytes, formed on the array. To demonstrating the utility of the array in tracking the colonoid fate, quantitative fluorescence analysis was performed on the arrayed colonoids exposed to reagents such as Wnt-3A and the γ-secretase inhibitor LY-411575. The successful formation of viable, multi-cell type colonic tissue on the microengineered platform represents a first step in the building of a "colon-on-a-chip" with the goal of producing the physiologic structure and organ-level function of the colon for controlled experiments.

Microfluidic chemical cytometry of peptide degradation in single drug-treated acute myeloid leukemia cells.

Microfluidic systems show great promise for single-cell analysis; however, as these technologies mature, their utility must be validated by studies of biologically relevant processes. An important biomedical application of these systems is characterization of tumor cell heterogeneity. In this work, we used a robust microfluidic platform to explore the heterogeneity of enzyme activity in single cells treated with a chemotherapeutic drug. Using chemical cytometry, we measured peptide degradation in the U937 acute myeloid leukemia (AML) cell line in the presence and absence of the aminopeptidase inhibitor Tosedostat (CHR-2797). The analysis of 99 untreated cells revealed rapid and consistent degradation of the peptide reporter within 20 min of loading. Results from drug-treated cells showed inhibited, but ongoing degradation of the reporter. Because the device operates at an average sustained throughput of 37 ± 7 cells/h, we were able to sample cells over the course of this time-dependent degradation. In data from 498 individual drug-treated cells, we found a linear dependence of degradation rate on amount of substrate loaded superimposed upon substantial heterogeneity in peptide processing in response to inhibitor treatment. Importantly, these data demonstrated the potential of microfluidic systems to sample biologically relevant analytes and time-dependent processes in large numbers of single cells.

Probing protein adsorption onto mercaptoundecanoic acid stabilized gold nanoparticles and surfaces by quartz crystal microbalance and zeta-potential measurements.

The adsorption characteristics of three proteins [bovine serum albumin (BSA), myoglobin (Mb), and cytochrome c (CytC)] onto self-assembled monolayers of mercaptoundecanoic acid (MUA) on both gold nanoparticles (AuNP) and gold surfaces (Au) are described. The combination of quartz crystal microbalance measurements with dissipation (QCM-D) and pH titrations of the zeta-potential provide information on layer structure, surface coverage, and potential. All three proteins formed adsorption layers consisting of an irreversibly adsorbed fraction and a reversibly adsorbed fraction. BSA showed the highest affinity for the MUA/Au, forming an irreversibly adsorbed rigid monolayer with a side-down orientation and packing close to that expected in the jamming limit. In addition, BSA showed a large change in the adsorbed mass due to reversibly bound protein. The data indicate that the irreversibly adsorbed fraction of CytC is a monolayer structure, whereas the irreversibly adsorbed Mb is present in form of a bilayer. The observation of stable BSA complexes on MUA/AuNPs at the isoelectric point by zeta-potential measurements demonstrates that BSA can sterically stabilize MUA/AuNP. On the other hand, MUA/AuNP coated with either Mb or CytC formed a reversible flocculated state at the isoelectric point. The colloidal stability differences may be correlated with weaker binding in the reversibly bound overlayer in the case of Mb and CytC as compared to BSA.