Inferring cell type innovations by phylogenetic methods-concepts, methods, and limitations

Integrating phylogenies into single-cell RNA sequencing analysis allows comparisons across species, genes, and cells

Comparisons of single-cell RNA sequencing (scRNA-seq) data across species can reveal links between cellular gene expression and the evolution of cell functions, features, and phenotypes. These comparisons evoke evolutionary histories, as depicted by phylogenetic trees, that define relationships between species, genes, and cells. This Essay considers each of these in turn, laying out challenges and solutions derived from a phylogenetic comparative approach and relating these solutions to previously proposed methods for the pairwise alignment of cellular dimensional maps. This Essay contends that species trees, gene trees, cell phylogenies, and cell lineages can all be reconciled as descriptions of the same concept-the tree of cellular life. By integrating phylogenetic approaches into scRNA-seq analyses, challenges for building informed comparisons across species can be overcome, and hypotheses about gene and cell evolution can be robustly tested.

Inferring chromatin accessibility during murine hematopoiesis through phylogenetic analysis

OBJECTIVE

Diversification of cell types and changes in epigenetic states during cell differentiation processes are important for understanding development. Recently, phylogenetic analysis using DNA methylation and histone modification information has been shown useful for inferring these processes. The purpose of this study was to examine whether chromatin accessibility data can help infer these processes in murine hematopoiesis.

RESULTS

Chromatin accessibility data could partially infer the hematopoietic differentiation hierarchy. Furthermore, based on the ancestral state estimation of internal nodes, the open/closed chromatin states of differentiating progenitor cells could be predicted with a specificity of 0.86-0.99 and sensitivity of 0.29-0.72. These results suggest that the phylogenetic analysis of chromatin accessibility could offer important information on cell differentiation, particularly for organisms from which progenitor cells are difficult to obtain.

Evolutionary Origins of the Oligodendrocyte Cell Type and Adaptive Myelination

Oligodendrocytes are multifunctional central nervous system (CNS) glia that are essential for neural function in gnathostomes. The evolutionary origins and specializations of the oligodendrocyte cell type are among the many remaining mysteries in glial biology and neuroscience. The role of oligodendrocytes as CNS myelinating glia is well established, but recent studies demonstrate that oligodendrocytes also participate in several myelin-independent aspects of CNS development, function, and maintenance. Furthermore, many recent studies have collectively advanced our understanding of myelin plasticity, and it is now clear that experience-dependent adaptations to myelination are an additional form of neural plasticity. These observations beg the questions of when and for which functions the ancestral oligodendrocyte cell type emerged, when primitive oligodendrocytes evolved new functionalities, and the genetic changes responsible for these evolutionary innovations. Here, I review recent findings and propose working models addressing the origins and evolution of the oligodendrocyte cell type and adaptive myelination. The core gene regulatory network (GRN) specifying the oligodendrocyte cell type is also reviewed as a means to probe the existence of oligodendrocytes in basal vertebrates and chordate invertebrates.

Inferring changes in histone modification during cell differentiation by ancestral state estimation based on phylogenetic trees of cell types: Human hematopoiesis as a model case

Revealing the landscape of epigenetic changes in cells during differentiation is important for understanding the development of organisms. In this study, to infer such epigenetic changes during human hematopoiesis, ancestral state estimation based on a phylogenetic tree was applied to map the epigenomic changes in six kinds of histone modifications onto the hierarchical cell differentiation process of hematopoiesis using epigenomes of eight types of differentiated hematopoietic cells. The histone modification changes inferred during hematopoiesis showed that changes that occurred on the branches separating different cell types reflected the characteristics of hematopoiesis in terms of genomic position and gene function. These results suggested that ancestral state estimation based on phylogenetic analysis of histone modifications in differentiated hematopoietic cells could reconstruct an appropriate landscape of histone modification changes during hematopoiesis. Since integration of the inferred changes of different histone modifications could reveal genes with specific histone marks such as active histone marks and bivalent histone marks on each internal branch of cell-type trees, this approach could provide valuable information for understanding the cell differentiation steps of each cell lineage.

Quartet-based inference of cell differentiation trees from ChIP-Seq histone modification data

Understanding cell differentiation-the process of generation of distinct cell-types-plays a pivotal role in developmental and evolutionary biology. Transcriptomic information and epigenetic marks are useful to elucidate hierarchical developmental relationships among cell-types. Standard phylogenetic approaches such as maximum parsimony, maximum likelihood and neighbor joining have previously been applied to ChIP-Seq histone modification data to infer cell-type trees, showing how diverse types of cells are related. In this study, we demonstrate the applicability and suitability of quartet-based phylogenetic tree estimation techniques for constructing cell-type trees. We propose two quartet-based pipelines for constructing cell phylogeny. Our methods were assessed for their validity in inferring hierarchical differentiation processes of various cell-types in H3K4me3, H3K27me3, H3K36me3, and H3K27ac histone mark data. We also propose a robust metric for evaluating cell-type trees.

Cross-Species Analysis of Single-Cell Transcriptomic Data

The ability to profile hundreds of thousands to millions of single cells using scRNA-sequencing has revolutionized the fields of cell and developmental biology, providing incredible insights into the diversity of forms and functions of cell types across many species. These technologies hold the promise of developing detailed cell type phylogenies which can describe the evolutionary and developmental relationships between cell types across species. This will require sampling of many species and taxa using single-cell transcriptomics, and methods to classify cell type homologies and diversifications. Many tools currently exist for analyzing single cell data and identifying cell types. However, cross-species comparisons are complicated by many biological and technical factors. These factors include batch effects common to deep-sequencing approaches, well known evolutionary relationships between orthologous and paralogous genes, and less well-understood evolutionary forces shaping transcriptome variation between species. In this review, I discuss recent developments in computational methods for the comparison of single-cell-omic data across species. These approaches have the potential to provide invaluable insight into how evolutionary forces act at the level of the cell and will further our understanding of the evolutionary origins of animal and cellular diversity.

Evolution of neuronal types and families

•

Solving nervous system evolution requires cross-species comparison of neuronal types.

•

Neuronal types are commonly defined by their specific structure and function.

•

We provide an operational definition of cell types that allows evolutionary comparison.

•

The identity of neuronal types is best reflected by specifying transcription factors.

•

Families of related neuronal types are conserved across large evolutionary distances.

Major questions in the evolution of neurons and nervous systems remain unsolved, such as the origin of the first neuron, the possible convergent evolution of neuronal phenotypes, and the transition from a relatively simple decentralized nerve net to the complex, centralized nervous systems found in modern bilaterian animals. In recent years, comparative single-cell transcriptomics has opened up new research avenues addressing these issues. Here, we review recent conceptual progress toward an evolutionary definition of cell types, and how it facilitates the identification and large-scale comparison of neuronal types and neuron type families from single-cell data — with the family of GABAergic neurons in distinct parts of the vertebrate forebrain as prime example. We also highlight strategies to infer cell type-specific innovation, so-called apomeres, from single-cell data.

The Transcriptomic Evolution of Mammalian Pregnancy: Gene Expression Innovations in Endometrial Stromal Fibroblasts

The endometrial stromal fibroblast (ESF) is a cell type present in the uterine lining of therian mammals. In the stem lineage of eutherian mammals, ESF acquired the ability to differentiate into decidual cells in order to allow embryo implantation. We call the latter cell type "neo-ESF" in contrast to "paleo-ESF" which is homologous to eutherian ESF but is not able to decidualize. In this study, we compare the transcriptomes of ESF from six therian species: Opossum (Monodelphis domestica; paleo-ESF), mink, rat, rabbit, human (all neo-ESF), and cow (secondarily nondecidualizing neo-ESF). We find evidence for strong stabilizing selection on transcriptome composition suggesting that the expression of approximately 5,600 genes is maintained by natural selection. The evolution of neo-ESF from paleo-ESF involved the following gene expression changes: Loss of expression of genes related to inflammation and immune response, lower expression of genes opposing tissue invasion, increased markers for proliferation as well as the recruitment of FOXM1, a key gene transiently expressed during decidualization. Signaling pathways also evolve rapidly and continue to evolve within eutherian lineages. In the bovine lineage, where invasiveness and decidualization were secondarily lost, we see a re-expression of genes found in opossum, most prominently WISP2, and a loss of gene expression related to angiogenesis. The data from this and previous studies support a scenario, where the proinflammatory paleo-ESF was reprogrammed to express anti-inflammatory genes in response to the inflammatory stimulus coming from the implanting conceptus and thus paving the way for extended, trans-cyclic gestation.