Department of Physiology, Development and Neuroscience

PhD Projects on Offer

(Dr Richard Adams, PDN & Dr Alexandre Kabla, Department of Engineering)

Advanced imaging methods now allow us to follow and analyse the movements of many hundreds of cells during the morphogenetic development of tissues (Blanchard et al, Nature Methods). Emerging from these studies are stereotypical patterns of cell behaviours that correlate with the overall shape changes of the tissue. Comparing the development of mutant or experimentally-manipulated animals with normal development is showing that in many case we see evidence for redundancy or compensative cell behaviours within tissues, involving in particular a number of mechanical factors. An important next step in these studies is to investigate those mechanical properties such that realistic models for morphogenesis can be developed. In this project we plan to develop ways to directly measure the mechanical properties of tissues by applying local extrinsic forces to embryos, while measuring cell and tissue movements. A pilot study has demonstrated the feasibility embedding cell-size magnetic spheres into growing embryos to perform such studies. This interdisciplinary project will involve the development and application of instrumentation and analytical tools to perform a full study on developing tissues in culture and zebrafish embryos. This project will suit a student with a strong background in the physical sciences or engineering. For further information, please contact Dr R Adams (rja46@cam.ac.uk) or Dr A Kabla (ajk61@cam.ac.uk).

This project deals with fundamental questions of animal development.

This project is interdisciplinary, involving the development of understanding of the physical properties of biological tissues and cell behaviours, the development of instrumentation and experimental design and the fabrication of instruments. This project would suit a student from the physical sciences with an interest in biology. The work will be conducted and supervised jointly between PDN and Engineering.

Computer modeling of embryo morphogenesis

(Dr Alexandre Kabla & Dr Richard Adams)

A position exists for a PhD studentship to work on the kinematics and biomechanics of embryo morphogenesis. The early development of an animal results from of a highly complex sequence of interactions within and between cells to transform a fertilised egg into a differentiated and structurally-elaborate functional embryo. A major challenge is to discover how the behaviour of individual cells is controlled and how they collectively change the shape of tissues. This studentship is an integral part of an EPSRC-funded project to develop a universal framework to measure and interpret animal morphogenesis. The project is an interdisciplinary collaboration between the Departments of Engineering and Physiology, Development & Neuroscience, University of Cambridge, and the School of Engineering and Applied Sciences, Harvard University. The studentship will involve formulating and exploring numerical simulations of tissues to explore non-invasive ways to obtain mechanical information about the morphogenetic process.

References:

Blanchard et al. Tissue tectonics: morphogenetic strain rates, cell shape change and intercalation. Nature Methods (2009) vol. 6 (6) pp. 458-64 

Dynamics of the Notch response

(Prof Sarah Bray, PDN; with Dr Steve Russell, Systems Biology Institute; Prof Simon Tavare, DAMTP) 

Characterising the regulatory circuits underpinning genomic responses to signalling and developing the computational tools necessary for the predictive modelling of these circuits requires experimentally tractable systems with well-developed genomics resources. We have been exploiting an ex vivo model in Drosophila cells for generating quantitative data on the cellular response Notch signaling by performing a fine-scaled time course of the genome-wide transcriptional changes following a pulse of Notch activation. This reveals different “patterns” of responding genes. To gain a mechanistic insight into these different responses we analyze the occupancy of the DNA-binding complex associated with Notch pathway activation over a similar time period (using ChIP). Combining these data will allow us to distinguish (1) genes with a high probability of being primary targets (2) different binding kinetics at different loci and how these correlate with specific patterns of transcriptional activity. From these combined data we aim to identify the underlying regulatory circuits and connectivity in the network (e.g., the presence of feedback loops), using a Bayesian analysis of stochastic models and their associated stochastic differential equations. Regulatory circuits and modules identified through these models will be tested by perturbing down the function of nodal components with RNAi, or by co-expressing specific factors, allowing us to further refine the models in light of these interventions.

Decoding the Notch signal

(Professor Sarah Bray)

 Intercellular signalling via the Notch pathway is essential at many steps in the development of an animal as well as during post-natal life. These included roles in neural specification and in maintaining stem cell lineages. While the correct deployment of Notch is crucial, its incorrect activity is associated with many types of diseases including dimentias and cancers. Activation of Notch elicits a proteolytic cleavage, releasing the Notch intracellular domain (Nicd), which enters the nucleus and collaborates directly with the DNA-binding proteins  to regulate transcription.  Thus, effects on gene expression are a primary consequence of activating the pathway.  A major focus of the work in our lab is towards understanding the regulation and functions of the Notch pathway. Working with the Drosophila model, because of its simplicity, we have most recently been taking a genome-wide approach to identify direct targets of Notch pathway activation.  Our analysis uncovers many novel targets, with a high degree of conservation with human genes. Current projects are focussed on investigating the regulation and function of these targets or on further exploiting genomic approaches to discover more about the outputs from the Notch pathway in different contexts.

References:

Bray, S.J. (2006) Notch Signalling: a simple pathway becomes complex. Nature Reviews in Molecular Cell Biology 7: 678-689.

---

Early development of the human placenta; potential role of the endometrial gland secretions

(Professor Graham Burton) 

The role of the endometrial glands during human early pregnancy has largely been ignored once implantation is complete.  However, we have recently demonstrated that the glands are an important source of nutrients for the conceptus during the first trimester.  The glandular secretions also contain a rich mix of growth factors and cytokines, and so play a role in regulating early placental development, modulating trophoblast proliferation and differentiation.  In other species, such as the sheep and rabbit, there is good evidence that the secretions are essential for promoting growth of the conceptus.  It is also known that in these species signals from the conceptus increase the production of glandular secretions to meet its demands.  Our interests now are to identify how endometrial gland secretions are regulated in human early pregnancy, their composition, and their effects on trophoblast behaviour.  The project may tackle one or more of these topics depending on the interests of the student.  It is likely that the project will involve a mix of cell culture, cell signalling analysis and molecular biology

References:

Burton , G.J., Jauniaux, E. and Charnock-Jones, D.S. (2007) Human early placental development: potential roles of the endometrial glands. Placenta, 28 Suppl. A, S64-69.

Genomic imprinting and the regulation of adult neurogenesis.

(Professor Anne Ferguson-Smith)

Using a genetic mouse model, we have recently shown that the imprinted atypical Notch ligand delta-like homologue 1, DLK1, is required for normal adult neurogenesis. DLK1 is expressed in two major isoforms – a memebrane-bound form and a secreted form. Dlk1 mutant mice have compromised neurogenesis. Our results have shown that the secreted form is an important factor in the neurogenic niche. In contrast, the membrane-bound isoform acts in the stem cell compartment where, unusually, the gene is expressed from both parental chromosomes. This absence of imprinting is important for the normal neurogenic potential of the stem cells.

Developmental programming of heart disease by prenatal hypoxia and oxidative stress.

(Dr Dino Giussani)

Heart disease is the greatest killer in the UK today, imposing a substantial burden on the nation’s health and wealth.  The concept that smoking and obesity increase the risk of heart disease is familiar to all of us. However, it does little to explain why some develop the disease and others do not.  Hence, in addition to the genetic basis of cardiovascular disease, another concept has now become established - one of developmental programming.  This states that a component of both the cardiovascular health we enjoy and the risk of heart disease in adult life can be determined before birth by the quality of our prenatal development. In turn, the quality of the intrauterine environment is largely determined by the available nutrient and oxygen supply to the growing young.  In this programme of work, we have put forward the hypothesis that oxidative stress in the fetal cardiovascular system underlies the molecular basis via which prenatal hypoxia contributes to the developmental programming of cardiovascular disease.  If true, treatment with antioxidants of pregnancies complicated with reduced oxygen delivery to the fetus, such as during pre-eclampsia or placental insufficiency should diminish the hypoxia-induced origins of cardiovascular disease.  The data may hasten translation to relatively simple clinical interventions to not only treat the mother, but also her progeny in complicated pregnancy.  This will reduce the burden not only of IUGR, but also of early origins of cardiovascular disease, thereby having a major clinical, economic and social impact on health.  For further information, please visit: http://www.pdn.cam.ac.uk/staff/giussani/

References:

GIUSSANI, D.A. & GARDNER, D.S. (2004).  The impact of acute and longterm hypoxia on fetal cardiovascular development.  Frontiers in Nutritional Sciences, No.2. Fetal Nutrition and Adult Disease: Programming of chronic disease through fetal exposure to undernutrition. Ed. LANGLEY-EVANS, S.C.pp. 55-85.

GIUSSANI, D.A. (2005).  Hypoxia, fetal growth and developmental origins of health and disease.  In: Early Life Origins of Health and Disease, Springer Science + Business Media; Landes Bioscience/Eurekah.com. Ed. WINTOUR, E.M. & OWENS, J. 18: 221-224. ISBN: 1-58706-294-1.

FOWDEN, A.L., GIUSSANI, D.A. & FORHEAD, A.J. (2006).  Intrauterine programming: Causes and consequences. Physiology 21:29-37.

GIUSSANI, D.A. (2006).  Prenatal hypoxia:  Relevance to developmental origins of health and disease.  In Developmental Origins of Health and Disease, Cambridge University Press. Ed. Gluckman, P.D. & Hanson, M.A. pp 178-190.

GIUSSANI, D.A., SALINAS, C.E., VILLENA, M. & BLANCO, C.E. (2007). The role of oxygen in prenatal growth: studies in the chick embryo. Journal of Physiology 585(Pt 3), 911-7.

(Professor Bill Harris) 

Using a variety of imaging and genetic techniques projects in this lab exploring how distinct cell types arise in the retina of zebrafish embryos.   We are interested in the mode of division, the motors that drive particular aspects of morphogenesis and laminar specific targetting as well as the upstream controls controls on these pathways, including how specific sublineages arise from multipotent retinal progenitors.

Molecular Embryogenesis of the Visual System

(Professor Bill Harris) 

Our lab is interested in questions such as nervous system comes from in the embryo, how it grows to the right size and shape, how stem cells turn into more committed neuronal progenitors and how these cells know when to leave the cycle and differentiate into particular cell types, and what the mechanisms are that allow these cells to become properly polarised, branched, and integrated into the neural circuitry.   The visual systems of Xenopus and zebrafish are ideal for such questions because of their embryological, molecular and genetic accessibility to experimentation, combined with the possibility of in vivo time-lapse imaging.

References:

Norden C, Young S, Link BA, Harris WA. 2009 Actomyosin is the main driver of interkinetic nuclear migration in the retina. Cell. 138(6):1195-208.

Agathocleous M, Iordanova I, Willardsen MI, Xue XY, Vetter ML, Harris WA, Moore KB. 2009

A directional Wnt/beta-catenin-Sox2-proneural pathway regulates the transition from proliferation to differentiation in the Xenopus retina. Development. 136(19):3289-99.

Agathocleous M, Harris WA 2009 From progenitors to differentiated cells in the vertebrate retina.

Annu Rev Cell Dev Biol.25:45-69. 

Cardiac remodeling and associated changes in cardiac conduction and susceptibility to arrhythmogenesis with age.

(Professor Chris Huang)

This project will use murine models to determine the extent of atrial remodeling at the level of ion channel expression and function, and of cardiac conduction with increasing age, and the implications this has for the orderly initiation and conduction of excitation from the sino-atiral to the atrioventricular node, or its breakdown into arrhythmogenesis.  The studies will involve (1) a characterization of the expression and distribution of ion channel and related gene products in the cardiac conduction system using quantitative PCR and immunohistochemical analysis their correlation with (2) electrophysiological phenotypes determined by electrical mapping abnd conduction study, in both whole Langendorf perfused hearts and in in intact sino-atrial node (SAN)/atrioventricular (A-V) junction preparations and with (3) patch clamping studies particularly for Na+ channel properties in isolated ventricular and atrial cells. The findings will be drawn together into coherent hypothesis for alterations in electrical conduction and excitation with age through (4) modeling studies of the spread of action potential through two-dimensional models of cardiac tissue. The findings will be of possible translational importance in view of the large number of individuals over the age of 60 who show electrical abnormalities, including atrial fibrillation, in the atria.

Recent relevant publications:

1)    Lei, M., Goddard, C., Liu, J., Léoni, A-L., Royer, A., Fung, S. S. M., Xiao, G., Ma, A., Zhang, H., Charpentier, F., Vandenberg, J. I., Colledge, W. H., Grace, A. A. & Huang, C. L-H. (2005). Sinus node dysfunction following targeted disruption of the murine cardiac sodium channel gene, SCN5A.  Journal of Physiology. 567(Pt 2):387-400.

2)    van Veen^, T. A. B., Stein, M., Royer, A., Quang, K. L., Charpentier, F., Colledge, W. H., Huang , C. L-H., Grace, A. A., Escande, D., de Bakker, J. M. T., van Rijen , H. V. M. (2005). Impaired impulse propagation in Scn5a knockout mice: Synergistic contribution of excitability, connexin expression and tissue architecture in relation to aging.  Circulation. 112(13):1927-1935.

3)    Lei, M., Zhang, H., Grace, A. A. Huang, C. L-H. (2007). SCN5a and sinoatrial node pacemaker function.  Cardiovascular Research. 74, 356-365.

 Energetic and redox homeostasis in hypoxic heart and muscle

(Dr Andrew Murray) 

The aim of our research is to better understand the cellular mechanisms that underlie abnormalities in energy metabolism that occur in heart and skeletal muscle in metabolic disease states such as heart failure and diabetes. More specifically, this project concerns the roles that tissue hypoxia, and consequent oxidative stress, might play in the transcriptional control of energetics. 

Heart failure is characterized by systemically high plasma free fatty acids, systemic and local hypoxia and oxidative stress. The exact role of each of these stresses in the progression of heart failure, and the development of mitochondrial dysfunction in heart and skeletal muscle is not yet clearly established. In models of heart failure it is difficult to tease apart the effects of hypoxia from those of an altered metabolic milieu and it is impossible, therefore, to establish cause and effect, in the context of failure, or to identify what constitutes an adaptation or maladaption. 

Mitochondria are the end-consumers of oxygen in the body, and likely modulate the metabolic adaptation to cellular hypoxia by decreasing oxygen dependency at the electron transport chain. This adaptation could involve metabolic switches towards the use of more oxygen-efficient substrates (e.g. glucose instead of fatty acids), improved coupling of the processes of oxidation and phosphorylation at the inner membrane or redistribution of mitochondrial populations within the cell to minimize oxygen concentration gradients. Such changes may be brought about via alterations in the transcription of genes controlled by the hypoxia-inducible factor (HIF) transcription factors. 

This project will use chamber-induced hypoxia in whole animals and muscle samples from humans acclimatising to high altitude as part of the second Xtreme Everest expeditions (www.xtreme-everest.co.uk) to investigate mechanisms of adaptation. Techniques will include isolated, working heart perfusions, mitochondrial respiration, western blotting and RT-PCR. Students with experience of these techniques are particularly encouraged to apply.

References:

AJ Murray, RE Anderson, GC Watson, GK Radda, K Clarke. Uncoupling proteins in human heart. Lancet 364: 1786-1788, 2004. 

AJ Murray, M Panagia, D Hauton, GF Gibbons, K Clarke. Plasma free fatty acids and peroxisome proliferator-activated receptor α in the control of myocardial uncoupling proteins. Diabetes  54: 3496-3502, 2005.

AJ Murray, CA Lygate, MA Cole, S Neubauer, K Clarke. Insulin resistance, abnormal energy metabolism and increased ischemic damage in the chronically infarcted rat heart. Cardiovascular Res 71: 149-157, 2006. 

AJ Murray, LM Edwards, K Clarke. Invited Review: Mitochondria in Heart Failure. Curr Opin Clin Nutr Metab Care 10: 704-711, 2007.

AJ Murray. Metabolic adaptation of skeletal muscle to high altitude hypoxia: how new technologies could resolve the controversies. Genome Medicine In Press 2009

Prefronto-amygdala circuitry and its role in anxiety.

(Professor Angela Roberts)

Fear is viewed as a biologically adaptive physiological and behavioural response to the actual or anticipated occurrence of an explicit threatening stimulus. In contrast, anxiety is induced by uncertain expectations of threat, is triggered by less explicit cues and is characterised by a more diffuse state of distress. However, it should be recognised that under uncertainty to the expectancy of threat, a certain level of anxious behaviour is adaptive, so the question then becomes, when is anxiety pathological? We would argue that it becomes pathological when, for example, it interferes with learning. Recently we have demonstrated just such interference, with overly anxious animals displaying inappropriate anxiety responses to safety cues. The orbitofrontal cortex is implicated in pathological anxiety since hypo- and hyper-activity has been reported in the medial and lateral OFC, respectively, of humans suffering from a variety of anxiety disorders. Other prefrontal regions have also been implicated. One recent theory suggests that clinically anxious individuals show both trait and state anxiety and that this may be characterised by both impoverished recruitment of prefrontal attentional control mechanisms and exaggerated amygdala responsiveness to threat related stimuli, with neither one on their own producing trait anxiety. Thus, this project will (i) study the contribution of distinct regions of OFC to anxiety in animals;using a paradigm that will allow the identification of abnormal anxiety responses to safety signals and (ii) test the hypothesis that high anxiety may be a product of both reduced top-down (prefrontal) and enhanced bottom-up (e.g. amygdala) regulation.

References:

Reekie Y. L., Braesicke K., Man M., Roberts A.C. (2008) Uncoupling of behavioral and autonomic responses following lesions of the primate orbitofrontal cortex. Proceedings of the National Academy of Sciences. 105:9787-92.

 

Analysis of cell and tissue behaviours during morphogenesis in Drosophila embryos.

(Dr Bénédicte Sanson)

The reshaping of sheets of cells is a key part of embryonic development, in both vertebrates and invertebrates. For example in vertebrates, tissue reshaping happens when the main body axis elongates and when the neural tube folds at the beginning of embryonic development. Failure to complete these tissue shape changes correctly has severe consequences leading to birth defects or complete failure to develop. In collaboration with Richard Adams and colleagues, we have applied quantitative methods to understand the mechanisms behind tissue reshaping, using Drosophila embryos as a model (References 1, 2).  From this work, two main lines of research emerge. The first one is to investigate how mesoderm invagination imposes an external stress on the adjacent ectoderm, which contributes to its elongation. This will be tackled by analysing cell and tissue behaviours in mutants defective in mesoderm invagination. A second avenue is to investigate how antero-posterior (AP) patterning influences the mechanical properties of the deforming ectoderm. This will be approached by looking for correlations between the cell cytoskeleton dynamics and the cell and tissue behaviours that we have already characterized in wild-type and several AP patterning mutants.

References:

(1) Butler LC, Blanchard GB, Kabla AJ, Lawrence NJ, Welchman DP, Mahadevan L, Adams RJ, Sanson B. (2009) Cell shape changes indicate a role for extrinsic tensile forces in Drosophila germ-band extension. Nature Cell Biology, 11: 859-64.

(2) Blanchard GB, Kabla AJ, Schultz NL, Butler LC, Sanson B, Gorfinkiel N, Mahadevan L, Adams RJ. (2009) Tissue tectonics: morphogenetic strain rates, cell shape change and intercalation. Nature Methods, 6:458-64.

Systems Modelling of organ morphogenesis in mammals

(Dr Paul Schofield) 

The aim of this project is to develop multiscale models of one of the following processes involved in organ morphogenesis:

(i) the mesenchyme to epithelial transition, or

(ii) the process of edge fusion seen, for example, in folding tubulogenesis, palatal shelf fusion and neural tube closure.

Models will be derived from our existing knowledge of the molecular participants in the processes under scrutiny and compared with the spatial and temporal expression patterns of these participants in a wide range of tissues by gathering data from organ-specific gene expression studies. Boolean models for different tissue instances of the same process will be tested under different conditions to identify the effect of organ contexts on apparently similar morphogenetic events. The project will make extensive use of public data resources, together with the gene expression and phenotype data in Mouse Genome Informatics. The systems models will be used to predict the effect of loss or gain of function of participant molecules on the developmental outcomes in different tissues. Predictions thus made will be compared with the developmental phenotypes of mouse mutants in those genes and where available the subsequent changes in gene expression patterns in affected tissues.

(Dr Octavian Voiculescu)

The project aims at developing a quantitative and predictive model of the amniote (chick) embryo at gastrula and early neurula stages. We have identified two cell behaviours driving the global cell movements occuring prior and during gastrulation: epithelial cell intercalation in a defined region of the epiblast (Voiculescu et al. Nature 449: 1049 (2007)) and cooperative cell ingression (Voiculescu, Lau & Stern, in preparation). We have implemented these behaviours into an object-oriented model of the chick gastrula, which can account for the global cell movements in the embryo. In the first part of the project, we will build a quantitative and predictive models of gastrulation using cellular and finite-element paradigms. These models will be confronted with experiments to quantify the cellular processes involved. Second, we will incorporate experimental data on the individual cell behaviours generating the morphogenetic forces at early neurula stages in the most appropriate model.

The role of Insulator Elements in chromatin architecture and the regulation of gene expression

(Dr Rob White)

Insulators are key elements involved in the subdivision of the genome into functional domains and have been implicated in both genome architecture and regulation.  They form boundaries to chromatin domains and confine the activities of regulatory elements, such as enhancers, so that they interact with appropriate target genes. In this project we will study the role of regulation of insulator function in the control of gene expression and the orchestration of dynamic chromatin topology.  In Drosophila, several proteins required for insulator function have been identified and insulator elements have been mapped genome-wide. We will investigate the following questions: 

• How dynamic is the association of insulator components with the genome?  To what extent do insulators represent a constant architectural framework for genome regulation and to what extent are dynamic changes in insulator complexes associated with changes in gene activity and specific regulation?  Are some components stable, providing an insulator “platform”, and others dynamic, regulating insulator function at specific sites?
• What is the role of insulator components in mediating chromatin folding and organising the chromatin topology associated with gene regulation at specific loci?
•  

We will use genomic technologies (e.g. ChIP-array and ChIP-seq) to map insulator components and will use chromatin conformation capture to analysis chromatin topology.  This project will provide fundamental insight into the dynamic functional organisation of the genome.

References:

Holohan, E. E., Kwong, C., Adryan, B., Bartkuhn, M., Herold, M., Renkawitz, R., Russell, S. and White, R. (2007) CTCF Genomic Binding Sites in Drosophila and the Organisation of the Bithorax Complex. PLoS Genet 3:e112.

Adryan, B.,Woerfel, G., Birch-Machin, I., Gao, S., Quick, M.,, Meadows, L., Russell, S. and White, R. (2007) Genomic mapping of Suppressor of Hairy-wing binding sites in Drosophila. Genome Biol. 8:R167.

The first cell fate decisions in the mouse embryo: formation of pluripotent embryonic and differentiated extra-embryonic tissues

(Dr Magdalena Zernicka-Goetz)

The first cell fate decision in the mouse embryo generates inside cells that retain pluripotency while outside cells differentiate into the first extra-embryonic tissue, trophectoderm. The second fate decision generates epiblast that will contribute all cells to the future body of an animal and the second extra-embryonic tissues, primitive endoderm (PE). Until now it has remained unknown how this decision is reached. We recently found that consecutive division waves forming outside and inside cells impact upon both first and second fate choices. The first wave of such divisions preferentially generates epiblast, while subsequent waves yield PE. Thus, the age of the outside mother-cell affects the fate of her inside daughter. Moreover, the identity of a particular generation of inside cells is largely preserved as those cells which are not correctly positioned, move to join cells of the same origin and fate. The aim of this project is to reveal the mechanism that allows cells to sort to their correct positions. Time-lapse imaging will enable cell position, shape, division orientation and cell movement to be followed when the PE and epiblast are set apart. Clones of cells with different properties, such as differing E-cadherin levels, will be generated to examine whether differential cell adhesion might form the basis for cell sorting. Since most cells that fail to correct their position undergo cell death, we will also ask whether the inhibition of apoptosis, by culturing embryos in a presence of specific inhibitor, would change allocation of cells to these distinct fates. It should shed light onto mechanisms for eliminating cells that are in a ‘wrong position’ and cannot sort. Recent work suggests that metabolically weaker cells are engulfed by their neighbours. Using high resolution confocal microscopy, labelled cells undergoing apoptosis will be followed to detect potential engulfment by neighbouring cells in the mouse embryo.

References:

1.   Zernicka-Goetz M, Morris S, Bruce A. (2009): Making a firm decision: multifaceted regulation of cell fate in the early mouse embryo. Nat Rev Genet. 10(7):467-77

2.   Meilhac SM, Adams RJ, Morris SA, Danckaert A, Le Garrec JF, Zernicka-Goetz M. (2009) Active cell movements coupled to positional induction are involved in lineage segregation in the mouse blastocyst. Dev Biol. 332, 210-221.

3.   Jedrusik A, Parfitt DE, Guo G, Skamagki M, Grabarek JB, Johnson MH, Robson P, Zernicka-Goetz M. (2008). Role of Cdx2 and cell polarity in cell allocation and specification of trophectoderm and inner cell mass in the mouse embryo. Genes Dev. 22(19):2692-706.