Technologies to decode the multicellular networks within the human body
This is a story about how it is possible for collections of cells to physically assemble into coordinated multicellular systems. In other words, how millions of individual cells are able to physically interact with each other in an organized way, as well as how pathogens such as viruses can exploit these interaction points in order to infect the body. Its subject is principally centered around the proteins that cover the surfaces of human cells. These proteins have to bind with specific combinations of surface proteins on nearby cells, thereby establishing a complex ‘code’ of direct interactions possible between different cell populations and tissues. Some of these interactions trigger the exchange of signals that enable collections of multiple cells to coordinate complex behaviors such as immune responses, while others act as adhesive receptors that enable physical structure to emerge out of groups of cells. The influential roles of these surface proteins and their accessibility to systemic medications have also made them among the most effective targets for therapeutics, with surface proteins constituting a majority of all approved drug targets.
So far however, prior research has only pieced together a fragmented picture of the direct receptor links between cells and the functional roles they have. Surface receptors pose unique experimental challenges to study and historically have lacked systematic methods to measure, leading most studies to only consider receptors at small-scale without a global view to the larger system. In this thesis, I take a different approach. I will present my work to establish a series of technological tools and strategies that overcome these challenges, in order to make it possible to systematically build up from characterizing the function of individual receptor molecules all the way to reconstructing multicellular interaction networks across entire systems of human cells. These methodologies can be categorized into three sequential steps. First, testing the binding of pairs of surface proteins across large arrays to decode the ‘interactome’ between two cells. Second, using cell-based assays to annotate the broad functional consequences a surface interaction has. And third, to computationally integrate these diverse data sources in order to understand how interacting communities of cells are organized.
As my initial case study, I consider the question of how the distributed individual cells of the human immune system interact to produce a cohesive whole. By individually producing recombinant forms of most surface proteins detectable on white blood cells, I could assemble the first systematic and quantitative interaction network of these proteins, and in the process discover several novel interactions and reveal the identities of previously-unidentified binding partners for key immunomodulatory receptors. I could then adapt those recombinant proteins to experimentally manipulate live human immune cells in a multiplex microscopy technique, which revealed previously unknown interactions as having prominent roles in immune activation and leukocyte adhesion. I will show how these data can be integrated with high-resolution expression data in order to infer patterns of cell-to-cell connectivity throughout the human body, as well as to formulate a mathematical model that could predict the behavior of interacting cells from molecular first-principles.
In the second half of my thesis, I will explain how this series of methods I established can be adapted and extended to new contexts. I will describe a large-scale effort I led applying these methods to characterize the cell-to-cell interactions occurring within the human brain, which revealed unexpected new pathways by which glia can directly communicate with cortical neurons. I will then extend my approaches to reveal which interactions may play a causal role in driving human disease. To do so, I will first show computational methods I devised for leveraging human clinical genetics in order to pinpoint cell-to-cell processes underlying the pathology. In the final section, I will extend this to infectious diseases driven by host-pathogen interactions. For this, I will explain how the tools I established allowed me to rapidly respond to the COVID-19 pandemic by systematically profiling the surface proteins that act as host factors during infection by the novel coronavirus SARS-CoV-2. That work has led to the discovery of two pathogen-host interactions that have subsequently been independently linked to COVID-19 severity, as well as helped clarify the precise host receptors that SARS-CoV-2 utilizes when invading human cells.
From the combination of these technologies and approaches, I hope to provide a systematic and mechanistically-grounded foundation for deconstructing the emergence of biological function from the interacting communities of cells that make up the human body.