We study the structure, function, and physiology of membrane proteins, with a focus on ion channels and the roles they play in electrical signaling, mechanosensation, and cell biology. We combine cryo-electron microscopy, electrophysiology, biochemistry, and pharmacology to understand how these proteins work at a molecular level and how they can be harnessed as tools or targeted as therapeutics. A summary of recent and ongoing work in the lab is below.
K2P K⁺ channels
Two-pore domain (K2P) potassium channels are the molecular basis for the background leak conductances that set resting membrane potential and regulate cellular excitability throughout the nervous system. Rather than acting as simple passive leaks, K2P channels are highly regulated by a diverse array of mechanical and chemical stimuli, making them dynamic sensors and integrators of the cellular environment. We study the mechanosensitive K2P channels TRAAK, TREK-1, and TREK-2, which are gated directly by physical tension in the lipid membrane, as well as the pH-regulated channels TASK-1, TASK-2, and TWIK-1. Using structural and functional approaches, we have determined how these channels sense and respond to diverse stimuli including membrane tension, protons, and pharmacological agents including anesthetics and cannabinoids. We are also interested in how disease-causing mutations in K2P channels alter channel function and contribute to human pathology, including mutations in TRAAK associated with FHEIG syndrome and mutations in TASK-1 linked to pulmonary arterial hypertension and developmental delay with sleep apnea. Key open questions include how lipids directly and indirectly gate K2Ps, how different gates interact allosterically, how channel activity is modulated in different physiological contexts, and whether these channels represent viable drug targets.
We discovered that TRAAK is highly enriched at nodes of Ranvier, the regularly spaced gaps in the myelin sheath where action potentials are regenerated during propagation along axons. This raises the exciting and largely unexplored possibility that mechanical forces are an intrinsic feature of axon physiology — potentially influencing how faithfully and efficiently signals travel through the nervous system. We are working to elucidate the role of TRAAK in nodes and what happens when it goes wrong in disease including by developing new molecular tools to measure membrane tension changes in living cells. More broadly, we are investigating the ultrastructure of myelinated axons and nodes in development and degeneration to better understand their form and function.
LRRC8 (SWELL) volume-regulated anion channels
LRRC8 proteins form volume-regulated anion channels (VRACs) that are ubiquitously expressed in vertebrate cells and essential for cell volume regulation, apoptosis, and a growing number of physiological and disease-relevant processes including insulin secretion and metabolic homeostasis. VRACs are heteromeric assemblies of the obligatory subunit LRRC8A with one or more of four partner subunits (LRRC8B–E), and subunit composition determines channel properties and physiological function. We have determined cryo-EM structures of homomeric LRRC8A channels and of LRRC8A:C and LRRC8A:D heteromers, and revealed how lipids can directly gate these channels closed. Despite their fundamental importance, many basic questions about VRACs remain unanswered: How do cells sense volume change and activate these channels? How does subunit identity control selectivity and gating? What assemblies are present in which cells? And how can we modulate VRACs with small molecules or other factors? Answering these questions is a major focus of ongoing work in the lab.
Channelrhodopsins
Channelrhodopsins are light-gated ion channels from algae and other microorganisms that have transformed neuroscience as optogenetic tools for controlling the activity of defined cell populations with light. We solved the cryo-EM structure of ChRmine, a potent red-shifted channelrhodopsin, revealing the structural basis for its exceptional properties and providing a blueprint for rational engineering of improved variants. In collaboration with the Adesnik lab at UC Berkeley, we have developed high-performance ChroME channelrhodopsins for spatially and temporally precise perturbation of large neuronal networks. In collaboration with the Flannery lab, we are evaluating newly developed channelrhodopsins for vision restoration in preclinical models of blindness. We are actively pursuing structural, mechanistic, and translational directions to address remaining open questions: How do these channels open and close? What determines their ion selectivity, kinetics, and spectral tuning? Can we engineer variants tailored for specific research or therapeutic applications?
Viral membrane proteins
Enveloped viruses encode membrane proteins that are essential for viral assembly, budding, and host cell entry, yet the structural and mechanistic bases for many of these functions remain poorly understood. We determined the structure of the SARS-CoV-2 M protein — the most abundant protein in the coronavirus envelope and the central organizer of virion assembly — and revealed that direct lipid interactions drive M protein conformational changes critical for virus production. We also solved the first structures of SARS-CoV-2 Orf3a. This work raises fundamental questions about how viruses co-opt and remodel cellular membranes, and opens the door to new antiviral strategies targeting membrane proteins essential to enveloped viruses.
