Biology subjects

Steventon, B.

Publications and source records attributed to Steventon, B..

24 records · Page 2Linked to original sources

bioRxiv2020

Cell-to-cell heterogeneity in Sox2 and Brachyury expression guides progenitor destiny by controllingtheir movements.

Although cell-to-cell heterogeneity in gene and protein expression within cell populations has been widely documented, we know little about its potential biological functions. We addressed this issue by studying progenitors that populate the posterior region of the vertebrate embryos, a cell population known for its capacity to self-renew or to contribute to the formation of the neural tube and paraxial mesoderm tissues. Posterior progenitors are characterized by the co-expression of Sox2 and Brachyury (Bra), two transcription factors related to neural and mesodermal lineages, respectively. In this study, we show that the respective levels of Sox2 and Bra proteins display a high degree of variability among posterior progenitors of the quail embryo. By developing forced expression and downregulation approaches, we further provide evidence that the value of the Sox2-to-Bra ratio in a given progenitor directs its choice of staying in place or exit the progenitor zone to generate neural or mesodermal cells. Time-lapse imaging together with mathematical modeling then reveal that variations of the Sox2-to-Bra ratio confer these cells heterogeneous motile behaviors. While high Bra levels display high motile properties that push cells to join the mesoderm, high levels of Sox2 tend to inhibit cell movement making cells get integrated into the neural tube. Our work thus provides evidence that the spatial heterogeneity of posterior progenitors, with regards to their relative contents of Sox2 and Bra and thus to their motile properties, is fundamental to maintain a pool of resident progenitors while others segregate to contribute to tissue formation. More broadly, our work reveals that heterogeneity among a population of progenitor cells is critical to ensure robust multi-tissue morphogenesis.

developmental biology↗

bioRxiv2020

A deep learning approach for staging embryonic tissue isolates with small data

Machine learning approaches are becoming increasingly widespread and are now present in most areas of research. Their recent surge can be explained in part due to our ability to generate and store enormous amounts of data with which to train these models. The requirement for large training sets is also responsible for limiting further potential applications of machine learning, particularly in fields where data tend to be scarce such as developmental biology. However, recent research seems to indicate that machine learning and Big Data can sometimes be decoupled to train models with modest amounts of data. In this work we set out to train a CNN-based classifier to stage zebrafish tail buds at four different stages of development using small information-rich data sets. Our results show that two and three dimensional convolutional neural networks can be trained to stage developing zebrafish tail buds based on both morphological and gene expression confocal microscopy images, achieving in each case up to 100% test accuracy scores. Importantly, we show that high accuracy can be achieved with data set sizes of under 100 images, much smaller than the typical training set size for a convolutional neural net. Furthermore, our classifier shows that it is possible to stage isolated embryonic structures without the need to refer to classic developmental landmarks in the whole embryo, which will be particularly useful to stage 3D culture in vitro systems such as organoids. We hope that this work will provide a proof of principle that will help dispel the myth that large data set sizes are always required to train CNNs, and encourage researchers in fields where data are scarce to also apply ML approaches. Author summaryThe application of machine learning approaches currently hinges on the availability of large data sets to train the models with. However, recent research has shown that large data sets might not always be required. In this work we set out to see whether we could use small confocal microscopy image data sets to train a convolutional neural network (CNN) to stage zebrafish tail buds at four different stages in their development. We found that high test accuracies can be achieved with data set sizes of under 100 images, much smaller than the typical training set size for a CNN. This work also shows that we can robustly stage the embryonic development of isolated structures, without the need to refer back to landmarks in the tail bud. This constitutes an important methodological advance for staging organoids and other 3D culture in vitro systems. This work proves that prohibitively large data sets are not always required to train CNNs, and we hope will encourage others to apply the power of machine learning to their areas of study even if data are scarce.

developmental biology↗

bioRxiv2018

Neuromesodermal Progenitors are a Conserved Source of Spinal Cord with Divergent Growth Dynamics

During gastrulation, embryonic cells become specified into distinct germ layers. In mouse, this continues throughout somitogenesis from a population of bipotent stem cells called neuromesodermal progenitors (NMps). However, the degree self-renewal is associated with NMps in the fast-developing zebrafish embryo is unclear. With a genetic clone tracing method, we labelled early embryonic progenitors and find a strong clonal similarity between spinal cord and mesoderm tissues. We then followed individual cell lineages by light-sheet imaging and reveal a common neuromesodermal lineage contribution to a subset of spinal cord tissue across the anterior-posterior body axis. An initial population subdivides at mid gastrula stages and is directly allocated to neural and mesodermal compartments during gastrulation. A second population in the tailbud undergoes delayed allocation to contribute to the neural and mesodermal compartment only at late somitogenesis. We suggest that NMps undergo vastly different rates of differentiation and growth in a species-specific manner.

developmental biology↗

bioRxiv2018

The chick caudo-lateral epiblast acts as a permissive niche for generating neuromesodermal progenitor behaviours

Neuromesodermal progenitors (NMps) are a population of bipotent progenitors that maintain competence to generate both spinal cord and paraxial mesoderm throughout the elongation of the posterior body axis. Recent studies have generated populations of NMp-like cells in culture and have been shown to differentiate to both neural and mesodermal cell fates when transplanted into either mouse or chick embryos. Here, we aim to compare the potential of mouse embryonic stem (ES) cell-derived progenitor populations to generate NMp behavior against both undifferentiated and differentiated populations. We define NMp behaviour as the ability of cells to i) contribute to a significant proportion of the anterior-posterior body axis, ii) enter into both posterior neural and somitic compartments and, iii) retain a proportion of the progenitor population within the posterior growth zone. We compare previously identified ES cell-derived NMp-like populations to undifferentiated mouse ES cells and find that they all display similar potentials to generate NMp behaviour in vivo. To assess whether this competence is lost upon further differentiation, we generated anterior and posterior embryonic cell types through the generation of 3D gastruloids and show that NMp competence is lost within the anterior (Brachyury negative) portion of the gastruloid. Taken together, this demonstrates that the chick caudo-lateral epiblast is a highly permissive environment for testing NMp competence and is therefore not suitable as a positive test of neuromesodermal progenitor identity (ie. specification). However, it does act as an appropriate system to test for the loss of NMp potential, and therefore offers insight as a functional test for the regulation of NMp competence in vivo.

developmental biology↗

bioRxiv2018

An Epiblast Stem Cell derived multipotent progenitor population for axial extension

The Caudal Lateral Epiblast of mammalian embryos harbours bipotent progenitors that contribute to the spinal cord and the paraxial mesoderm in concert with the elongation of the body axis. These progenitors, called Neural Mesodermal Progenitors (NMPs) are identified as cells coexpressing Sox2 and T/Brachyury, a criterion used to derive NMP-like cells from Embryonic Stem Cells in vitro. However, these progenitors do not self renew, as embryonic NMPs do. Here we find that protocols that yield NMP-like cells in vitro first produce a multipotent population that, additional to NMPs, generates progenitors for the lateral plate and intermediate mesoderm. We show that Epiblast Stem Cells (EpiSCs) are an effective source for these multipotent progenitors that are further differentiated by a balance between BMP and Nodal signalling. Importantly, we show that NMP-like cells derived from EpiSCs self renew in vitro and exhibit a gene expression signature similar to that of their embryo counterparts.

developmental biology↗

bioRxiv2017

Gastruloids develop the three body axes in the absence of extraembryonic tissues and spatially localised signalling

Establishment of the three body axes is a critical step during animal development. In mammals, genetic studies have shown that a combination of precisely deployed signals from extraembryonic tissues position the anteroposterior axis (AP) within the embryo and lead to the emergence of the dorsoventral (DV) and left-right (LR) axes. We have used Gastruloids, embryonic organoids, as a model system to understand this process and find that they are able to develop AP, DV and LR axes as well as to undergo axial elongation in a manner that mirror embryos. The Gastruloids can be grown for 160 hours and form derivatives from ectoderm, mesoderm and endoderm. We focus on the AP axis and show that in the Gastruloids this axis is registered in the expression of T/Bra at one pole that corresponds to the tip of the elongation. We find that localisation of T/Bra expression depends on the combined activities of Wnt/{beta}-Catenin and Nodal/Smad2,3 signalling, and that BMP signalling is dispensable for this process. Furthermore, AP axis specification occurs in the absence of both extraembryonic tissues and of localised sources of signalling. Our experiments show that Nodal, together with Wnt/{beta}-Catenin signalling, is essential for the expression of T/Bra but that Wnt signalling has a separable activity in the elongation of the axis. The results lead us to suggest that, in the embryo, the role of the extraembryonic tissues might not be to induce the axes but to bias an intrinsic ability of the embryo to break its initial symmetry and organise its axes.\n\nOne sentence summaryCulture of aggregates of defined number of Embryonic Stem cells leads to self-organised embryo-like structures which, in the absence of localised signalling from extra embryonic tissues and under the autonomous influence of Wnt and Nodal signalling, develop the three main axes of the body.

developmental biology↗