General introduction: lipid codes, organelles, and contact sites
Cells write spatial instructions on membranes using compositional “lipid codes.” These codes are shaped and read at inter-organelle contact sites, where lipids and ions are exchanged without fusion, creating landscapes that govern receptor clustering, vesicle biogenesis, signaling amplitude, organelle quality control, and cargo fate across the cell.
This framework is broader than any single pathway. Antigen presentation, cytokine trafficking, mitochondrial maintenance, endolysosomal maturation, and secretory routing are all constrained by lipid identity and contact dynamics. In this program we apply that general framework through two application arenas: the licensing of NLRP3 inflammasome assembly on permissive endomembranes, and the decision of nascent insulin granules to be secreted or degraded by lysosomes.
Our recent work and current directions
In beta cells, we established that newly formed insulin granules can be selectively degraded in lysosomes, providing nutrients to lysosomal signaling hubs and thereby tuning autophagy and secretion. This revealed a granular quality control axis, distinct from canonical autophagy, and connected lipid exchange at the TGN to the decision between secretion and degradation. In diabetic contexts, this pathway is heightened and contributes to progressive insulin content loss, establishing direct disease relevance.
In innate immunity, we showed that specific phosphorylation events act as activating switches for NLRP3. More recently, we demonstrated that, upon activation, NLRP3 is recruited to endosomes where lipid-enriched membranes provide a permissive platform for inflammasome assembly. This positioned membrane context, not just soluble checkpoints, as the critical variable governing inflammasome competence.
Together, these contributions link kinase signaling and metabolic cues to defined membrane identities, establishing that lipid codes and organelle contacts license complex cellular decisions.
Program for 2025 to 2028
The program contains two interlocked streams.
Stream 1: Grammar of the permissive endosome in NLRP3 activation More details…
We will define the lipid and protein composition and biophysical properties that make an endosomal membrane permissive for NLRP3. Endogenous models and reconstitution on simplified membranes will be used to identify the rules that govern inflammasome licensing and assembly.
Stream 2: Contact-coded fate of insulin granules More details…
We will establish how TGN lipid reprogramming drives selective routing of nascent insulin granules toward secretion or lysosomal degradation. By manipulating contact-site dynamics, we will test whether restoring these processes can preserve insulin content in diabetic islets.
In vivo toolkit and endpoints
Reporter systems and in vivo models will be used to visualize recruitment, assembly, and granule fate under physiological conditions. Pharmacologic and genetic modulation of contact-site regulators will be introduced in controlled settings, with disease models providing context. Functional outputs will include inflammation, insulin content, and secretion, all under ethical and statistically robust designs.
Integration and coherence
The two streams serve as case studies of a single question. How do lipid codes and inter-organelle contacts license or veto the assembly of macromolecular machines and the routing of selective membrane traffic. Because the logic is shared, the toolkit remains portable across diverse cellular contexts.
Outlook
By treating lipid codes and contact sites as first principles, we aim to produce a mechanistic framework that predicts when and where inflammasomes assemble and how endocrine cells decide between secretion and degradation. This unified approach is scientifically rigorous and directly relevant to inflammatory and metabolic disease.