Top-down and back up: Modeling chromosome organization

The organization and segregation of chromosomes are vital cellular processes. However, understanding chromosome organization in living organisms is challenging because these polymers are both highly compacted and far from thermodynamic equilibrium. In this thesis, we use a combination of top-down data-driven methods, bottom-up mechanistic modeling, and polymer theory to gain insight into the organization and segregation of bacterial and human mitotic chromosomes.

Chapter 1 first introduces the reader to important concepts of chromosome biology, and then reviews previous physical models for bacterial chromosome organization. We discuss three classes of models: ones that impose large-scale constraints on a polymer model; ones that model how bottom-up mechanisms affect chromosome organization; and finally, ones that are inferred directly from experimental data.

Chapter 2 discusses two models for bacterial chromosome organization which predict similar pairwise contact frequencies, yet qualitatively different three-point contact statistics. We explain how these differences arise by deriving approximations for three-point contact frequencies in three distinct physical limits, and by showing that the two models are best modeled by different approximations.

Chapters 3 and 4 discuss how loop extrusion by structural maintenance of chromosome (SMC) complexes can both help organize and segregate bacterial chromosomes. SMC complexes are motor proteins that latch onto the chromosome at a point and actively reel in a loop of DNA. Loop extrusion is emerging as a critical mechanism of chromosome organization in both eukaryotes and prokaryotes.

In Chapter 3, we show that specific off-loading of loop-extruders can explain the orientation of Escherichia coli chromosomes within the cell. We find that this mechanism, which does not rely on anchoring the chromosome to the cell wall, gives rise to accurate positioning of genes within the cellular confinement.

In Chapter 4, we revisit an earlier hypothesis that bacterial chromosome segregation could be a purely entropic process. We show that at intermediate replication stages, purely entropic forces drive bacterial chromosomes towards unsegregated states. However, we also show that specific loading of loop-extruders at the origin of replication, as seen in many bacteria, can redirect entropic forces for segregation.

In Chapter 5, we present a data-driven model for the organization of a replicating bacterial chromosome. Our model is inferred from time-course Hi-C data and can be used to sample 3D chromosome configurations across the cell cycle. After validating our model against independent experiments, we use it to make data-driven inferences about chromosome organization throughout the cell cycle. We also compare our predictions to simple mechanistic models.

In Chapter 6, we use polymer theory to interpret experimental data on extracted human mitotic chromosomes. By probing the force-extension behavior of mitotic chromosomes, we find that they exhibit abnormally gradual stiffening. We explain this behavior by constructing a Hierarchical Worm-Like Chain model, in which components with a large range in elastic properties stiffen sequentially. Our model suggests that human mitotic chromosomes exhibit great structural heterogeneity. Further experiments suggest that the viscosity and elasticity of mitotic chromosomes arise from two distinct mechanisms: an elastic central scaffold surrounded by a looser effective solution of chromatin.