The simplicity of metazoan cell lineages

Reconstructing cell lineage trees with genomic barcoding: approaches and applications

In multicellular organisms, developmental history of cell divisions and functional annotation of terminal cells can be organized into a cell lineage tree (CLT). The reconstruction of the CLT has long been a major goal in developmental biology and other related fields. Recent technological advancements, especially those in editable genomic barcodes and single-cell high-throughput sequencing, have sparked a new wave of experimental methods for reconstructing CLTs. Here we review the existing experimental approaches to the reconstruction of CLT, which are broadly categorized as either image-based or DNA barcode-based methods. In addition, we present a summary of the related literature based on the biological insight provided by the obtained CLTs. Moreover, we discuss the challenges that will arise as more and better CLT data become available in the near future. Genomic barcoding-based CLT reconstructions and analyses, due to their wide applicability and high scalability, offer the potential for novel biological discoveries, especially those related to general and systemic properties of the developmental process.

Lineage motifs: developmental modules for control of cell type proportions

In multicellular organisms, cell types must be produced and maintained in appropriate proportions. One way this is achieved is through committed progenitor cells that produce specific sets of descendant cell types. However, cell fate commitment is probabilistic in most contexts, making it difficult to infer progenitor states and understand how they establish overall cell type proportions. Here, we introduce Lineage Motif Analysis (LMA), a method that recursively identifies statistically overrepresented patterns of cell fates on lineage trees as potential signatures of committed progenitor states. Applying LMA to published datasets reveals spatial and temporal organization of cell fate commitment in zebrafish and rat retina and early mouse embryo development. Comparative analysis of vertebrate species suggests that lineage motifs facilitate adaptive evolutionary variation of retinal cell type proportions. LMA thus provides insight into complex developmental processes by decomposing them into simpler underlying modules.

Symmetry and simplicity spontaneously emerge from the algorithmic nature of evolution

SignificanceWhy does evolution favor symmetric structures when they only represent a minute subset of all possible forms? Just as monkeys randomly typing into a computer language will preferentially produce outputs that can be generated by shorter algorithms, so the coding theorem from algorithmic information theory predicts that random mutations, when decoded by the process of development, preferentially produce phenotypes with shorter algorithmic descriptions. Since symmetric structures need less information to encode, they are much more likely to appear as potential variation. Combined with an arrival-of-the-frequent mechanism, this algorithmic bias predicts a much higher prevalence of low-complexity (high-symmetry) phenotypes than follows from natural selection alone and also explains patterns observed in protein complexes, RNA secondary structures, and a gene regulatory network.

Data-Theoretical Synthesis of the Early Developmental Process

Biological development is often described as a dynamic, emergent process. This is evident across a variety of phenomena, from the temporal organization of cell types in the embryo to compounding trends that affect large-scale differentiation. To better understand this, we propose combining quantitative investigations of biological development with theory-building techniques. This provides an alternative to the gene-centric view of development: namely, the view that developmental genes and their expression determine the complexity of the developmental phenotype. Using the model system Caenorhabditis elegans, we examine time-dependent properties of the embryonic phenotype and utilize the unique life-history properties to demonstrate how these emergent properties can be linked together by data analysis and theory-building. We also focus on embryogenetic differentiation processes, and how terminally-differentiated cells contribute to structure and function of the adult phenotype. Examining embryogenetic dynamics from 200 to 400 min post-fertilization provides basic quantitative information on developmental tempo and process. To summarize, theory construction techniques are summarized and proposed as a way to rigorously interpret our data. Our proposed approach to a formal data representation that can provide critical links across life-history, anatomy and function.

What machine learning can do for developmental biology

Developmental biology has grown into a data intensive science with the development of high-throughput imaging and multi-omics approaches. Machine learning is a versatile set of techniques that can help make sense of these large datasets with minimal human intervention, through tasks such as image segmentation, super-resolution microscopy and cell clustering. In this Spotlight, I introduce the key concepts, advantages and limitations of machine learning, and discuss how these methods are being applied to problems in developmental biology. Specifically, I focus on how machine learning is improving microscopy and single-cell 'omics' techniques and data analysis. Finally, I provide an outlook for the futures of these fields and suggest ways to foster new interdisciplinary developments.

Alignment of Cell Lineage Trees Elucidates Genetic Programs for the Development and Evolution of Cell Types

A full understanding of the developmental process requires fine-scale characterization of cell divisions and cell types, which are naturally organized as the developmental cell lineage tree (CLT). Technological breakthroughs facilitated determination of more CLTs, but complete comprehension of the data remains difficult without quantitative comparison among CLTs. We hereby quantified phenotypic similarity between CLTs using a novel computational method that exhaustively searches for optimal correspondence between individual cells meanwhile retaining their topological relationships. The revealed CLT similarities allowed us to infer functional similarity at the transcriptome level, identify cell fate transformations, predict functional relationships between mutants, and find evolutionary correspondence between cell types of different species. By allowing quantitative comparison between CLTs, our work is expected to greatly enhance the interpretability of relevant data and help answer the myriad of questions surrounding the developmental process.

•

Align cell lineage trees (CLTs) to search/quantify their phenotypic similarities

•

Aligning worm CLTs captured known genetic/developmental programs

•

Similarities between knockdown CLTs revealed functional relationships between genes

•

CLT alignments between species gave insight on the evolution of cell types

Lineage space and the propensity of bacterial cells to undergo growth transitions

The molecular makeup of the offspring of a dividing cell gradually becomes phenotypically decorrelated from the parent cell by noise and regulatory mechanisms that amplify phenotypic heterogeneity. Such regulatory mechanisms form networks that contain thresholds between phenotypes. Populations of cells can be poised near the threshold so that a subset of the population probabilistically undergoes the phenotypic transition. We sought to characterize the diversity of bacterial populations around a growth-modulating threshold via analysis of the effect of non-genetic inheritance, similar to conditions that create antibiotic-tolerant persister cells and other examples of bet hedging. Using simulations and experimental lineage data in Escherichia coli, we present evidence that regulation of growth amplifies the dependence of growth arrest on cellular lineage, causing clusters of related cells undergo growth arrest in certain conditions. Our simulations predict that lineage correlations and the sensitivity of growth to changes in toxin levels coincide in a critical regime. Below the critical regime, the sizes of related growth arrested clusters are distributed exponentially, while in the critical regime clusters sizes are more likely to become large. Furthermore, phenotypic diversity can be nearly as high as possible near the critical regime, but for most parameter values it falls far below the theoretical limit. We conclude that lineage information is indispensable for understanding regulation of cellular growth.

Quantifying Mosaic Development: Towards an Evo-Devo Postmodern Synthesis of the Evolution of Development via Differentiation Trees of Embryos

Embryonic development proceeds through a series of differentiation events. The mosaic version of this process (binary cell divisions) can be analyzed by comparing early development of Ciona intestinalis and Caenorhabditis elegans. To do this, we reorganize lineage trees into differentiation trees using the graph theory ordering of relative cell volume. Lineage and differentiation trees provide us with means to classify each cell using binary codes. Extracting data characterizing lineage tree position, cell volume, and nucleus position for each cell during early embryogenesis, we conduct several statistical analyses, both within and between taxa. We compare both cell volume distributions and cell volume across developmental time within and between single species and assess differences between lineage tree and differentiation tree orderings. This enhances our understanding of the differentiation events in a model of pure mosaic embryogenesis and its relationship to evolutionary conservation. We also contribute several new techniques for assessing both differences between lineage trees and differentiation trees, and differences between differentiation trees of different species. The results suggest that at the level of differentiation trees, there are broad similarities between distantly related mosaic embryos that might be essential to understanding evolutionary change and phylogeny reconstruction. Differentiation trees may therefore provide a basis for an Evo-Devo Postmodern Synthesis.

A morphospace for synthetic organs and organoids: the possible and the actual

Efforts in evolutionary developmental biology have shed light on how organs are developed and why evolution has selected some structures instead of others. These advances in the understanding of organogenesis along with the most recent techniques of organotypic cultures, tissue bioprinting and synthetic biology provide the tools to hack the physical and genetic constraints in organ development, thus opening new avenues for research in the form of completely designed or merely altered settings. Here we propose a unifying framework that connects the concept of morphospace (i.e. the space of possible structures) with synthetic biology and tissue engineering. We aim for a synthesis that incorporates our understanding of both evolutionary and architectural constraints and can be used as a guide for exploring alternative design principles to build artificial organs and organoids. We present a three-dimensional morphospace incorporating three key features associated to organ and organoid complexity. The axes of this space include the degree of complexity introduced by developmental mechanisms required to build the structure, its potential to store and react to information and the underlying physical state. We suggest that a large fraction of this space is empty, and that the void might offer clues for alternative ways of designing and even inventing new organs.

Determinative developmental cell lineages are robust to cell deaths

All forms of life are confronted with environmental and genetic perturbations, making phenotypic robustness an important characteristic of life. Although development has long been viewed as a key component of phenotypic robustness, the underlying mechanism is unclear. Here we report that the determinative developmental cell lineages of two protostomes and one deuterostome are structured such that the resulting cellular compositions of the organisms are only modestly affected by cell deaths. Several features of the cell lineages, including their shallowness, topology, early ontogenic appearances of rare cells, and non-clonality of most cell types, underlie the robustness. Simple simulations of cell lineage evolution demonstrate the possibility that the observed robustness arose as an adaptation in the face of random cell deaths in development. These results reveal general organizing principles of determinative developmental cell lineages and a conceptually new mechanism of phenotypic robustness, both of which have important implications for development and evolution.

Experimental evolution of multicellularity using microbial pseudo-organisms

In a major evolutionary transition to a new level of organization, internal conflicts must be controlled before the transition can truly be successful. One such transition is that from single cells to multicellularity. Conflicts among cells in multicellular organisms can be greatly reduced if they consist of genetically identical clones. However, mutations to cheaters that experience one round of within-individual selection could still be a problem, particularly for certain life cycles. We propose an experimental evolution method to investigate this issue, using micro-organisms to construct multicellular pseudo-organisms, which can be evolved under different artificial life cycles. These experiments can be used to test the importance of various life cycle features in maintaining cooperation. They include structured reproduction, in which small propagule size reduces within-individual genetic variation. They also include structured growth, which increases local relatedness within individual bodies. Our method provides a novel way to test how different life cycles favour cooperation, even for life cycles that do not exist.

Evolution of embryonic development in nematodes

BACKGROUND

Nematodes can be subdivided into basal Enoplea (clades 1 and 2) and more derived Chromadorea (clades 3 to 12). Embryogenesis of Caenorhabditis elegans (clade 9) has been analyzed in most detail. Their establishment of polarity and asymmetric cleavage requires the differential localization of PAR proteins. Earlier studies on selected other nematodes revealed that embryonic development of nematodes is more diverse than the essentially invariant development of C. elegans and the classic study object Ascaris had suggested. To obtain a more detailed picture of variations and evolutionary trends we compared embryonic cell lineages and pattern formation in embryos of all 12 nematode clades.

METHODS

The study was conducted using 4-D microscopy and 3-D modeling of developing embryos.

RESULTS

We found dramatic differences compared to C. elegans in Enoplea but also considerable variations among Chromadorea. We discovered 'Polarity Organizing Centers' (POCs) that orient cleavage spindles along the anterior-posterior axis in distinct cells over consecutive cell generations. The resulting lineally arranged blastomeres represent a starting point for the establishment of bilateral symmetry within individual lineages. We can discern six different early cleavage types and suggest that these variations are due to modifications in the activity of the POCs in conjunction with changes in the distribution of PAR proteins. In addition, our studies indicate that lineage complexity advanced considerably during evolution, that is we observe trends towards an increase of somatic founder cells, from monoclonal to polyclonal lineages and from a variable (position-dependent) to an invariable (lineage-dependent) way of cell fate specification. In contrast to the early phase of embryogenesis, the second half ('morphogenesis') appears similar in all studied nematodes. Comparison of early cleavage between the basal nematode Tobrilus stefanskii and the tardigrade Hypsibius dujardini revealed surprising similarities indicating that the presence of POCs is not restricted to nematode embryos.

CONCLUSIONS

The pattern of cleavage, spatial arrangement and differentiation of cells diverged dramatically during the history of the phylum Nematoda without corresponding changes in the phenotype. While in all studied representatives the same distinctive developmental steps need to be taken, cell behavior leading to these is not conserved.

Spatio-temporal reference model of Caenorhabditis elegans embryogenesis with cell contact maps

The nematode Caenorhabditis elegans has been used as a model for developmental biology for decades. Still, the few publicly available spatio-temporal (4D) data sets have conflicting information regarding variability of cell positions and are not well-suited for a standard 4D embryonic model, due to compression. We have recorded six uncompressed embryos, and determined their lineage and 4D coordinates, including nuclear radii, until the end of gastrulation. We find a remarkable degree of stability in the cell positions, as well as little rotational movement, which allowed us to combine the data into a single reference model of C. elegans embryogenesis. Using Voronoi decomposition we generated the list of all predicted cell contacts during early embryogenesis and calculated these contacts up to the approximately 150 cell stage, and find that about 1500 contacts last 2.5 min or longer. The cell contact map allows for comparison of multiple 4D data sets, e.g., mutants or related species, at the cellular level. A comparison of our uncompressed 4D model with a compressed embryo shows that up to 40% of the cell contacts can be different. To visualize the 4D model interactively we developed a software utility. Our model provides an anatomical resource and can serve as foundation to display 4D expression data, a basis for developmental systems biology.

Embryogenesis of Romanomermis culicivorax: an alternative way to construct a nematode

The current picture of embryonic development in nematodes is essentially shaped by Caenorhabditis elegans and its close relatives. As their pattern of embryogenesis is rather similar, it is often considered to be representative for the taxon Nematoda as a whole. Here we give for the first time a comprehensive description of embryonic development in an ancestrally diverged nematode. Romanomermis culicivorax differs strikingly from C. elegans with respect to cell division pattern, spatial arrangement of blastomeres and tissue formation. Our study reveals a number of unexpected phenomena. These include (i) unique polar interphase microtubule caps forming in early blastomeres destined to undergo asymmetric cleavages, suggesting the presence of a so far undescribed MTOC; (ii) embryonic cell lineages of reduced complexity with predominantly monoclonal sublineages, generating just a single tissue type; (iii) construction of major parts of the body from duplicating building blocks consisting of rings of cells, a pattern showing some resemblance to segmentation; (iv) prominent differences in cell fate assignment which can be best explained with a global shift affecting all somatic founder cells. In summary, our data indicate that during nematode evolution massive alterations in the developmental program took place of how to generate a juvenile.

Easily adaptable complexity measure for finite time series

We present a complexity measure for any finite time series. This measure has invariance under any monotonic transformation of the time series, has a degree of robustness against noise, and has the adaptability of satisfying almost all the widely accepted but conflicting criteria for complexity measurements. Surprisingly, the measure is developed from Kolmogorov complexity, which is traditionally believed to represent only randomness and to satisfy one criterion to the exclusion of the others. For familiar iterative systems, our treatment may imply a heuristic approach to transforming symbolic dynamics into permutation dynamics and vice versa.

Embryology of a planktonic tunicate reveals traces of sessility

A key problem in understanding deuterostome evolution has been the origin of the chordate body plan. A biphasic life cycle with a sessile adult and a free-swimming larva is traditionally considered ancestral in chordates with subsequent neotenic loss of the sessile adult stage. Molecular phylogenies challenged this view, suggesting that the primitive life cycle in chordates was entirely free-living as in modern day larvaceans. Here, we report the precise cell lineage and fate map in the normal embryo of the larvacean Oikopleura dioica, using 4D microscopy technique and transmission electron microscopy. We document the extraordinary rapidity of cleavage and morphogenetic events until hatching and demonstrate that--compared with ascidians--fate restriction occurs considerably earlier in O. dioica and that clonal organization of the cell lineage is more tightly coupled to tissue fate. We show that epidermal cells in the trunk migrate through 90 degrees, reminiscent of events in ascidian metamorphosis and that the axis of bilateral symmetry in the tail rotates in relation to the trunk. We argue that part of the tail muscle cells are ectomesodermal, because they are more closely associated with prospective epidermis than with other tissues in the cell lineage. Cladistic comparison with other deuterostomes suggests that these traits are derived within tunicates strengthening the hypothesis that the last common ancestor of tunicates had a sessile adult and thus support traditional morphology-derived scenarios. Our results allow hypothesizing that molecular developmental mechanisms known from ascidian models are restricted to fewer, yet identifiable, cells in O. dioica.

Gene promoter scan methodology for identifying and classifying coregulated promoters

A critical challenge of the postgenomic era is to understand how genes are differentially regulated. Genetic and genomic approaches have been used successfully to assign genes to distinct regulatory networks in both prokaryotes and eukaryotes. However, little is known about what determines the differential expression of genes within a particular network, even when it involves a single transcription factor. The fact that coregulated genes may be differentially expressed suggests that subtle differences in the shared cis-acting regulatory elements are likely to be significant. This chapter describes a method, termed gene promoter scan (GPS), that discriminates among coregulated promoters by simultaneously considering a variety of cis-acting regulatory features. Application of this method to the PhoP/PhoQ two-component regulatory system of Escherichia coli and Salmonella enterica uncovered novel members of the PhoP regulon, as well as regulatory interactions that had not been discovered using previous approaches. The predictions made by GPS were validated experimentally to establish that the PhoP protein uses multiple mechanisms to control gene transcription and is a central element in a highly connected network.

A generative bias towards average complexity in artificial cell lineages

The evolution of life on earth has been characterized by generalized long-term increases in phenotypic complexity. Although natural selection is a plausible cause for these trends, one alternative hypothesis--generative bias--has been proposed repeatedly based on theoretical considerations. Here, we introduce a computational model of a developmental system and use it to test the hypothesis that long-term increasing trends in phenotypic complexity are caused by a generative bias towards greater complexity. We use our model to generate random organisms with different levels of phenotypic complexity and analyse the distributions of mutational effects on complexity. We show that highly complex organisms are easy to generate but there are trade-offs between different measures of complexity. We also find that only the simplest possible phenotypes show a generative bias towards higher complexity, whereas phenotypes with high complexity display a generative bias towards lower complexity. These results suggest that generative biases alone are not sufficient to explain long-term evolutionary increases in phenotypic complexity. Rather, our finding of a generative bias towards average complexity argues for a critical role of selective biases in driving increases in phenotypic complexity and in maintaining high complexity once it has evolved.

LinMap: visualizing complexity gradients in evolutionary landscapes

This article describes an interactive visualization tool, LinMap, for exploring the structure of complexity gradients in evolutionary landscapes. LinMap is a computationally efficient and intuitive tool for visualizing and exploring multidimensional parameter spaces. An artificial cell lineage model is presented that allows complexity to be quantified according to several different developmental and phenotypic metrics. LinMap is applied to the evolutionary landscapes generated by this model to demonstrate that different definitions of complexity produce different gradients across the same landscape; that landscapes are characterized by a phase transition between proliferating and quiescent cell lineages where both complexity and diversity are maximized; and that landscapes defined by adaptive fitness and complexity can display different topographical features.

The choice of model organisms in evo-devo

Study of the model organisms of developmental biology was crucial in establishing evo-devo as a new discipline. However, it has been claimed that this limited sample of organisms paints a biased picture of the role of development in evolution. Consequently, judicious choice of new model organisms is necessary to provide a more balanced picture. The challenge is to determine the best criteria for choosing new model organisms, given limited resources.

A phylogenetic approach to mapping cell fate

Recent, surprising, and controversial discoveries have challenged conventional concepts regarding the origins and plasticity of stem cells, and their contributions to tissue regeneration, and highlight just how little is known about mammalian development in comparison to simpler model organisms. In the case of the transparent worm, Caenorhabditis elegans, Sulston and colleagues used a microscope to record the birth and death of every cell during its life, and the compilation of this "fate map" represents a milestone achievement of developmental biology. Determining a fate map for mammals or other higher organisms is more complicated because they are opaque, take a long time to mature, and have a tremendous number of cells. Consequently, fate mapping experiments have relied on tagging a progenitor cell with a dye or genetic marker in order to later identify its descendants. This approach, however, extracts little information because it demonstrates that a population of cells, all having inherited the same label, shares a common ancestor, but it does not reveal how cells in that population are related to one another. To avoid that problem, as well as technical limitations of current methods for mapping cell fate, we, and others, have developed a new strategy for retrospectively deriving cell fate maps by using phylogenetics to infer the order in which somatic mutations have arisen in the genomes of individual cells during development in multicellular organisms. DNA replication inevitably introduces mutations, particularly at repetitive sequences, every time a cell divides. It is thus possible to deduce the history of cell divisions by cataloging somatic mutations and phylogenetically reconstructing cell lineage. This approach has the potential to produce a complete mammalian cell fate map that, in principle, could describe the developmental lineage of any cell and help resolve outstanding questions of stem cell biology, tissue repair and maintenance, and aging.

A logical circuit for the regulation of fission yeast growth modes

Growth of fission yeast at the ends of its cylindrical cells switches from a monopolar to a bipolar mode, before it ceases during mitosis and cell division. Here we assume that these growth modes correspond to three stable states of an underlying regulatory circuit, which is a relatively simple and to a large degree autonomous subsystem of an otherwise complex cellular control system. We develop a switch-like logical circuit based on three elements defined as binary variables. Effects of circuit variables on each other are expressed in terms of logical operations. We analyse this circuit for its behavior ("phenotypes") after removing single or multiple operations ("mutants"). Known fission yeast polarity mutants such as those defective in the switch to bipolar growth can be classified based on these predicted 'phenotypes'. Differences in growth patterns between daughter cells in different bipolar growth mutants are also predicted by the circuit model. The model presented here should provide a useful framework to guide future experiments into mechanisms of cellular polarity. This paper illustrates the usefulness of simple logical circuits to describe and dissect features of complex regulatory processes such as the fission yeast growth patterns in both wild type and mutant cells.

Analysis of differentially-regulated genes within a regulatory network by GPS genome navigation

MOTIVATION

A critical challenge of the post-genomic era is to understand how genes are differentially regulated even when they belong to a given network. Because the fundamental mechanism controlling gene expression operates at the level of transcription initiation, computational techniques have been developed that identify cis regulatory features and map such features into expression patterns to classify genes into distinct networks. However, these methods are not focused on distinguishing between differentially regulated genes within a given network. Here we describe an unsupervised machine learning method, termed GPS for gene promoter scan, that discriminates among co-regulated promoters by simultaneously considering both cis-acting regulatory features and gene expression. GPS is particularly useful for knowledge discovery in environments with reduced datasets and high levels of uncertainty.

RESULTS

Application of this method to the enteric bacteria Escherichia coli and Salmonella enterica uncovered novel members, as well as regulatory interactions in the regulon controlled by the PhoP protein that were not discovered using previous approaches. The predictions made by GPS were experimentally validated to establish that the PhoP protein uses multiple mechanisms to control gene transcription, and is a central element in a highly connected network.

AVAILABILITY

The scripts and programs used in this work are accessible from the gps-tools.wustl.edu website. Data and predictions are available by request.