Session F44: Delbruck Prize Session

Prize Talk: Max Delbruck Prize in Biological PhysicsGeometry and Genetics

The application of quantitative methods to biological problems faces the choice of how much detail to include and the generality of the conclusions. The middle ground entails some use phenomenology, a well-regarded approach in physics. A sampling of examples will be presented from my work in developmental biology, to give a flavor of what is possible. The phenomenon of canalization is a license to develop models that are quantitative and dynamic yet do not begin from an enumeration of the relevant genes. Modern mathematics (ie post 1960), under the rubric of 'dynamical systems', has many similarities to experimental embryology and allows the enumeration of categories of dynamical behaviors. Examples from stem cell differentiation will illustrate how systems with a few variables can be fit to cell state transitions and the self-organizing capacity of cell aggregates, 'organoids'. Geometric arguments alone suffice to enumerate a short list of 'typical' parameter spaces, i.e., phase diagrams for how states transform into each other. Phenomenology of the sort envisioned is essential to bridge the scales from the cell, to tissue, to embryo, by breaking the system into blocks that can be separately parameterized.   

Mechanics of embryonic self-organization

Embryonic development is a largely self-organized process, that can restore its course following perturbations, through cell-cell communication. While this can be mediated by diffusing molecules, it is increasingly appreciated that the mechanical forces that shape an embryo can also function as long-range signals. This dual role of mechanics is illustrated in the early development of birds, which shares many aspects with that of other vertebrates, yet is easily observed and manipulated. I will present a study of avian gastrulation, in which we have shown that the forces driving morphogenesis self-organize and impinge on gene expression in the establishment of the embryonic axis. It is tempting to speculate that mechanical feedback, implicated here in the proportioning of the early embryo, is pervasive in development.   

Modulating signals to pattern complex tissues

The blastocyst of the human embryo forms a single-layer epithelial sheet of cells called the epiblast. This epiblast gives rise to the entire human body over the course of development. The patterning of the epiblast is achieved through a sequence of symmetry-breaking events driven by signals. The identity of the signals is often known. However, their spatiotemporal profiles, certainly in the case of human embryos and, more broadly, for most vertebrate embryos, are not. One might think of these morphogen signal profiles as establishing boundary conditions for a complex underlying dynamical system, driving it to pattern itself. 

Embryonic stem cells are obtained from the blastocyst of the embryo and can be cultured indefinitely. These cells have the capability to generate all the cell types of the body. Can we pattern these cells by learning and establishing the correct spatiotemporal boundary conditions to generate complex human tissues? We will address this question using a combination of statistical inference approaches and experiments on human embryonic stem cells.    

Imaging protein translation from micron to atomic scales

Translation is an evolutionarily conserved process in which the ribosome, mRNA and tRNA coordinately synthesize new proteins. Here, I focus on two emerging technologies for studying the complex dynamics of translation at scales ranging from nanometer to atomic resolution. Single particle tracking of mRNAs and nascent peptide using Halo dyes revealed where translation of secretory and membrane proteins occurs on ER. The findings uncovered a novel coordination between ER and lysosomes in the patterning and regulation of translation of secretory/membrane proteins involving local release of amino acids and other factors from lysosomes adjacent to ER. To study translation at the atomic scale, we employed High Resolution Template Matching (HRTM) to examine ribosomes at different stages of peptide elongation in cryo-EM images of intact human cells. Combining reconstructions across 41 native conformations, we obtained a high-resolution movie of ribosome dynamics revealing ligand movements during polypeptide chain elongation and spring-like intramolecular motion. Together, these new tools open-up a plethora of questions related to translation and its mechanism that can now be studied in intact cells at the nanometric/atomic level.   

Loop extrusion, chromatin crosslinking, and the geometry, topology and mechanics of chromosomes and nuclei

The chromosomes of cells are based on tremendously long DNA molecules that must be replicated and then physically separated to allow successful cell division. Our group uses biophysical and mathematical approaches to study chromosome structure and dynamics. A key emerging feature of chromosome organization is the role of active chromatin loop formation, or "loop extrusion" as a mechanism underlying chromosome compaction, individualization, and segregation, as well as maintenance of regulatory chromatin loops. I will discuss experiments on SMC complexes (condensin, cohesin and SMC5/6 in eukaryotes), currently thought to be the loop-extruding elements, and how those experiments inform us on the mechanisms of these novel chromatin-organizing motors. For whole chromosomes, compaction via formation of tightly packed loops is combined with chromatin-chromatin interactions to achieve individualization of chromosomes and separation of adjacent chromatids during cell division. I will also discuss our group's studies of the role of chromosomal epigenetic marks in control of the structure and integrity of the cell nucleus, with a focus on the role of the balance between heterochromatin and euchromatin in controlling nuclear mechanics. I will discuss experiments on human cell nuclei that have established that the epigenetic "reader" HP1 plays a key role in controlling nuclear mechanics, presumably via "crosslinking" H3K9me2,3 marks on nucleosomes and thereby stabilizing mechanically robust heterochromatin domains. Our current experiments are probing the mechanics of the centromere, a highly robust complex of proteins and DNA, which appears to be uniquely resistant to disruption of its nucleic acid content.