Our Research

Research Overview

The human body contains over 10 trillion cells spanning hundreds of morphologically distinct cell types. These cells must work together for our bodies to function correctly. However, it remains a mystery how such an enormous diversity of cells coordinate their behaviors.

Tissue structure - or the composition and physical arrangement of cells, extracellular matrix, and diffusible molecules - helps to coordinate cellular behaviors by organizing the flow of chemical, mechanical, and electrical information between cells. Thus, building tissue structure correctly and maintaining tissue structure over time are prerequisites for engineering functional organs and stopping the progression of diseases like cancer.

We are interested in three general questions about how tissue structure forms and functions:

(i) How does tissue structure form through the process of self-organization?

(ii) How does tissue structure help cells to arrive at collective decisions and to organize collective behaviors?

(iii) How does tissue structure breakdown during the progression of diseases like cancer?

To answer these questions we take a synthetic approach, studying and engineering human tissues from the bottom-up. This approach allows us to measure and perturb the molecular and physical properties of individual cells, reconstitute them into living tissue, then observe their interactions to reveal the underlying "principles" guiding their self-organizaton and decision making. We focus primarily on the cells and tissues of the human breast, and our work incorporates experimental principles from the chemical, biological, and engineering sciences.

Chemically Programmed Assembly of 3D Tissues

Engineering complex human tissues from simpler cellular building blocks remains difficult. One important challenge is defining the initial arrangment of cells and ECM molecules in three dimensions so that the cells' own program of self-organization can take over. We have developed several tools to facilitate this process.


The first is a chemical approach for patterning tissues that incorporate multiple cell types, arranged with high spatial precision, and assembled from the bottom-up and across three-dimensions. Cells are functionalized with chemical “Velcro” – short DNA oligonucleotides that impart specific adhesive properties between neighboring cells. Hybridization of complementary DNA sequences enables the assembly of multicellular structures with defined cell-cell contacts. We call this approach DNA Programmed Assembly. Our current efforts aim to extend the synthetic capabilities of DNA Programmed Assembly, so as to help more robustly guide tissue self-organization, with the goal of building fully integrated and functional human tissue. For example, we are working to program the assembly of the human mammary gland from purified cellular components in vitro. We are using the assembled structures as models to learn how cell-cell signaling networks regulate the behaviors of cells in vivo, and to reveal how these networks breakdown or are co-opted during the early stages of breast cancer.


We are also developing tools to precisely define the absolute number and composition of cells when place into culture. In collaboration with Adam Abate's lab at UCSF, we developed a technique called Printed Droplet Microfluidics. At its core, this technique provides a quantitatively precise means of combining any combination of cells together into a small volume of liquid. Once a combination is prepared it is transfered to physiologically relevant conditions for long-term study.

Self-organization

Cells are living materials – their properties are not static but rather change dynamically in response to their surroundings. This property of cells as materials gives them the capacity to self-organize into a wide variety of three dimensional structures. Indeed, the capacity of cells to self-organize is critical to their normal developmental and their ability to self-repair. A better understanding of how tissues self-organize will improve our ability to synthesize tissues and organs in the lab, and suggest new strategies to slow the breakdown of tissue structure that contributes to the initiation and progression of diseases such as cancer.

We are working to understand the physical principles used by cells to self-organize robustly into specific tissue structures. For example, we are characterizing the active material properties of the interfaces between cell types in the human mammary epithelium to reveal the forces that guide their self-organization. We are particularly interested in how these properties render self-organization robust to the heterogeneous and dynamic cell behaviors that are characteristic of mammary gland development. We are also investigating how genes involved in breast cancer alter these properties, increasing the probability of developing invasive disease.


The mammary gland self-organizes correctly in extracellular matrix (left), but into an inverted structure in non-adhesive microenvironments such as agarose (right).


We are also investigating how tissues regulate the pattern and extent of folding during their self-organization. This project has two components. In the first, we are working to engineer the autonomous folding of tissues using using cells that contract in a manner that causes the tissue to fold. In the second, we are collaborating with Ophir Klein's lab to understand the molecular and physical mechanism through which a pattern of folds form during the morphogenesis of intestinal villi. Our goal is to place this process under engineering control and to improve the regeneration of the gut.


Tissue origami: fibroblasts (green) generate forces that transform a collagen-rich gel (black) into an origami-like four-fold junction.

Collective behaviors and decision making

Individual cells can behave differently, even when in the same tissue. These behavioral differences can combine in non-intuitive ways to drive new and emergent behaviors at the level of cell communities. For example, many cellular behaviors related to growth, motility, differentiation and adhesion are controlled by signaling pathways downstream of the protein Ras. Ras integrates signals impinging on cells from their extracellular environment, and is also one of the most frequently mutated proteins in cancer. We demonstrated that sustained patterns of heterogeneity (also known as cell-to-cell variability) in signaling downstream of Ras are sufficient to drive several emergent behaviors that are not observed when the pathways are activated homogeneously among all the cells of a tissue. These emergent behaviors include dissemination of single hypermotile cells, basal cell extrusions, as well as collective and motile protrusions (see video below). These phenomona are frequently associated with aggressive tumors, a context where heterogeneity can become exaggerated.



We are also working to understand how different types of cells communicate within tissues to arrive at collective decisions. In order to investigate collective cell decision making, we are integrating human tissue models (organoids) with new multiplexed single cell analysis tools developed in the lab (MULTI-seq). These tools allow us to investigate how groups of interacting cells within a tissue change their collective decisions as we alter tissue composition, snip lines of cell-cell communication, or perturb the gene-regulatory networks that integrate these lines of communication. We are focusing our efforts on immune cell activation (in collaboration with Matt Thomson at Caltech) and hormone signaling in the human breast (in collaboration with Thea Tlsty at UCSF).

Nanoscale probes

Many cell-surface recptors for extracellular signals can organize into larger, nanoscale structures. How do these nanoscale structures affect the propagation of signals from the outside to the inside of the cell? This has been a difficult question to answer because few tools are available to observe or alter the organization and dynamics of these signaling complexes at the nanoscale. We are developing new tools to image (e.g. monovalent quantum dots) and perturb (e.g. DNA-scaffolded multivalent ligands) the dynamics and organization of these receptors with a focus on the biology of the human Notch and Epidermal Growth Factor Receptors. Our goal is to provide new insight into the spatial organization of these signaling complexes, and how their organization contributes to cell-cell interactions affecting cell-fate decisions and cancer.