No abstract

Multienzyme complexes, or metabolons, are assemblies or clusters of sequential enzymes that naturally exist in metabolic pathways. These nanomachineries catalyze the conversion of metabolites more effectively than the freely floating enzymes by minimizing the diffusion of intermediates in vivo. Bioengineers have devised synthetic versions of multienzyme complexes in cells to synergize heterologous biosynthesis, to improve intracellular metabolic flux, and to achieve higher titer of valuable chemical products. Here, we utilized orthogonal protein reactions (SpyCatcher/SpyTag and SnoopCatcher/SnoopTag pairs) to covalently assemble three key enzymes in the mevalonate biosynthesis pathway and showed 5-fold increase of lycopene and 2-fold increase of astaxanthin production in Escherichia coli. The multienzyme complexes are ellipsoidal nanostructures with hollow interior space and uniform thickness and shapes. Intracellular covalent enzyme assembly has yielded catalytic nanomachineries that drastically enlarged the flux of carotenoid biosynthesis in vivo. These studies also deepened our understanding on the complexity of hierarchical enzyme assembly in vivo.

Intracellular reactions on nonenzymatic proteins that activate cellular signals are rarely found. We report one example here that a designed peptide derivative undergoes a nucleophilic reaction specifically with a cytosolic PDZ protein inside cells. This reaction led to the activation of ephrin-B reverse signaling, which subsequently inhibited SDF-1 induced neuronal chemotaxis of human neuroblastoma cells and mouse cerebellar granule neurons. Our work provides direct evidence that PDZ-RGS3 bridges ephrin-B reverse signaling and SDF-1 induced G protein signaling for the first time.

Targeted covalent inhibitors of protein-protein interactions differ from reversible inhibitors in that the former bind and covalently bond the target protein at a specific site of the target. The site specificity is the result of the proximity of two reactive groups at the bound state, for example, one mild electrophile in the inhibitor and a natural cysteine in the target close to the ligand binding site. Only a few pharmaceutically relevant proteins have this structural feature. Grb2, a key adaptor protein in maintaining the ERK activity via binding Sos1 to activated RTKs, is one: the N-terminal SH3 domain of Grb2 (Grb2) carries a unique solvent-accessible cysteine Cys close to its Sos1-binding site. Here we report the design of a peptide-based antagonist (a reactive peptide) that specifically binds to Grb2 and subsequently undergoes a nucleophilic reaction with Cys to form a covalent bond thioether, to block Grb2-Sos1 interaction. Through rounds of optimization, we eventually obtained a dimeric reaction reactive peptide that can form a covalent adduct with endogenous Grb2 protein inside the cytosol of SK-BR-3 human breast cancer cells with pronounced inhibitory effect on cell mobility and viability. This work showcases a rational design of Grb2-targeted site-specific covalent inhibitor and its pronounced anticancer effect by targeting Grb2-Sos1 interaction.