Extracellular information is communicated to intracellular mediators by activation of membrane-embedded proteins, called receptors, on the cell surface. Receptors represent the gateway through which cells sense and respond to their environment. Most physiologically important processes are, at some level, initiated by the engagement of an extracellular molecule with the extracellular regions of cell surface receptors in a highly specific fashion, in order to activate intracellular signal transduction cascades and subsequent genetic programs. Cell surface receptors are also the major therapeutic entry points for drug development by the pharmaceutical industry.
My laboratory is investigating structural and functional aspects of cell surface receptor recognition and activation, in receptor-ligand systems with relevance to human health and disease. Collectively, our receptor programs blend traditional structural approaches with ligand engineering and discovery. Our goal is to paint a detailed mechanistic picture—from the outside to the inside of a cell—of how ligand binding is structurally coupled to receptor activation. How is the static aspect of recognition linked with the dynamic aspects of activation? We combine this structural and mechanistic information with protein engineering to modulate the strength and specificity of downstream intracellular signaling and thereby control cell fate in novel ways. We also use proteomics to discover ligands for orphan receptors.
Thematically, my laboratory focuses on "shared" receptors—receptors that are activated by a variety of structurally diverse ligands and give rise to both unique and redundant signaling outcomes. Such systems are also referred to as pleiotropic, and one of our principal interests is to elucidate the basis of functional specificity by these highly pleiotropic and cross-reactive receptor systems. We are particularly interested in shared receptors at the interface between immunity, neurobiology, and development.
The mechanisms of cell surface receptor activation, which are highly diverse, usually involve some combination of ligand-induced oligomerization and conformational change. Yet, we currently have an appreciation for only a small fraction of the potential receptor-ligand interaction paradigms across the human genome. We have endeavored to characterize novel receptor activation mechanisms by determining crystal structures of extracellular receptor-ligand complexes. The structural studies have been complemented by companion biochemical, in vivo, and engineering experiments to probe the relevance of the structural information to the proteins in their native environments.
We have described new paradigms for recognition and activation of a range of shared pleiotropic cell surface receptors that generally fall into the categories of adaptive immunity, neural signaling, and development. Examples include cytokine receptors, developmental morphogens, vasoactive hormone receptors, neurotrophin receptors, semaphorin receptors, and T cell receptors (TCRs). These receptors play critical roles in immune regulation (e.g., IL-2, IL-4), stem cell biology (Wnt, CD47), cancer (gp130, Frizzled), autoimmunity (TCR/MHC), and neural growth and axon guidance (neurotrophins, semaphorin/plexin). We continue to investigate the structural diversity of receptor activation mechanisms by targeting receptor systems for which structural information does not exist. Our long-term goal, however, is to probe more deeply the systems for which we have gained structural "access." We are attempting to use mechanism-based ligand-engineering strategies as well as combinatorial ligand libraries to modulate receptor activation. Because receptors are transmembrane proteins, we are examining receptor molecules fully loaded with both extracellular ligand and intracellular adapter molecules as intact membrane proteins.
Within the immune system, we investigate both antigen and cytokine-driven phases of immunity. With respect to the latter, we have delved deeply into pleiotropic cytokines that regulate immunity. For example, the shared cytokine receptors gp130 and common γ chain are signaling receptors for almost 15 different hematopoietic cytokine ligands with critical functions in the immune system (e.g., IL-2, IL-4, and IL-6). What is the structural basis for this cross-reactivity, and how can the engagement of the different cytokines lead to qualitatively different signaling outcomes? Our studies have shown that gp130 cross-reacts with different ligands, such as interleukin-6 and leukemia inhibitory factor, through a uniquely accommodating surface chemistry in its structurally rigid binding site, rather than through conformational change. It also appears that ligand recognition by gp130 is coupled to a bending of the receptor as it enters the membrane, in order to position the intracellular segments optimally for signaling. In contrast, common γ chain appears to have evolved some semblance of a cytokine recognition code to engage IL-2, IL-4, and IL-15 in similar fashions. Our current cytokine studies are focused in two directions: (1) "tuning" the immune system by engineering cytokines with diverse signaling properties, and (2) understanding better how cytokine-receptor interactions result in intracellular Janus kinase (JAK) and STAT activation. For this latter goal, we are visualizing, through electron microscopy, the architecture of the entire transmembrane gp130 receptor in both quiescent and activated forms.
Regarding the antigen-specific receptors that control adaptive immunity, my lab has a long-standing interest in structural aspects of T cell recognition of peptide-MHC (major histocompatibility complex). Currently we are focused on the apparent ability of TCRs to react with any MHC molecule. This property raises the question of whether there is a coevolved, germline-derived recognition code between TCR and MHC molecules that has so far been elusive. This fundamental question is linked to the larger issue of whether the geometric manner by which TCRs recognize pMHC has an impact on TCR signaling. In other words, does structure matter, or is the TCR/pMHC ligation even structurally indiscriminate? We have developed a novel methodology to use yeast-displayed peptide-MHC libraries to discover new TCR ligands. This TCR-ligand "medicinal chemistry" gives us an opportunity not only to explore the role of antigen recognition in TCR signaling but also to use these libraries to discover the natural ligands for orphan TCRs. For example, we are attempting to identify peptide ligands for TCRs derived from tumor-infiltrating lymphocytes (TILs) that are resident in tumors.
In development, receptor-ligand pleiotropy is critical to context-dependent signaling. For example, Wnt signaling through Frizzled is a critical modulator of cell fate, and deregulation of this pathway is closely linked to many cancers. Mammalian genomes encode 20 Wnts and 10 Frizzled receptors, but we do not understand the basis of Wnt specificity given this pleiotropy. Our recent structural studies gave us a first snapshot of how Wnt engages Frizzled and shed insight into Wnt/Fz cross-reactivity. Given the centrality of Wnt in human biology, we are attempting to engineer Frizzled-specific Wnts that we might use to unravel the functional consequences of Wnt specificity and that might serve as potential therapeutics. Further studies in this system are also aimed at understanding the elusive nature of Frizzled signaling, which has alternatively been linked to G proteins as well as β-catenin stabilization. Structural studies of higher order Wnt/Fz complexes containing signaling adapters will clarify these outstanding questions.
An important example of receptor sharing in the nervous system involves semaphorins, which can act as ligands for both neuropilin and plexin receptors in order to mediate repulsive signaling during axonal growth. However, semaphorins are also active in both the immune and nervous systems, where interactions with plexins mediate a common activity of inducing cell homing and pathfinding through the shared receptor plexinC1. In fact, viruses express mimics of semaphorins in order to interfere with immune semaphorins and enhance the virus survival. We found that mammalian and viral Sema proteins engage plexinC1 in nearly identical fashions, inducing plexinC1 into an activated dimer for signaling. The basis of the cross-reactivity for plexinC1 appears to be that the virus hijacks the mammalian Sema and improves the plexin-binding site so as to "out-compete" the human Sema for receptor engagement.
Finally, a growing interest in our lab is receptor "deorphanization." The vast majority of cell surface receptors do not have known ligands. It is important to develop methodologies to identify endogenous partners of orphan receptors and ligands. In a pilot study, by taking a pairwise interaction approach, we have discovered many new receptor-ligand pairs in Drosophila melanogaster. We hope to escalate this work to a genome-wide receptor deorphanization program in the near future.
This work is also supported by grants from the National Institutes of Health.
As of January 16, 2013