April 2, 2015
Speaker: Dr. Srivatsan Raman, Wyss Institute, Harvard Medical School
Engineering biosynthetic pathways for chemical production requires extensive optimization of the host cellular metabolic machinery. Using natural allosteric transcription factors (TF) as metabolite sensors, we report a general strategy that combines targeted genome-wide mutagenesis to generate pathway variants, together with accelerated evolution to enrich for rare high producers. Through up to four rounds of evolution, we increased production of naringenin by 36-fold to a final titer of 61 mg/L, the highest reported titer directly from glucose. The ability to design allosteric TFs to respond to new inducers opens the door to engineering biosynthetic pathways for thousands of valuable molecules that currently lack a natural sensor. However, designing allosteric TFs is a major challenge because it requires engineering a new protein-small molecule interaction without disrupting the allosteric function. Using the classical lac repressor TF (LacI), we developed a general scheme to engineer allosteric TFs responding to new inducers. We used computational protein design to generate candidates, synthesize DNA encoding 20,000 designed variants with chip-based oligonucleotides, and developed a two-step enrichment for allosterically-functional variants inducible by the target ligand. We redesigned LacI to respond to four new inducers: sucralose, lactitol, fucose and gentiobiose, with induction response comparable to the original LacI with IPTG. Besides metabolic engineering applications, this method allows building orthogonal switches for synthetic gene networks, sensors for single cell real-time metabolomics and regulators for targeted control of developmental pathways.
Srivatsan Raman is currently a Technology Development Fellow at the Wyss Institute and Dr. George Church’s research group at the Harvard Medical School. He is working on directed evolution methods for engineering new proteins and genomes. He obtained a Ph.D in Dr. David Baker’s research group at the University of Washington, Seattle on computational modeling of protein structures. He has Masters and Bachelors degrees in Chemical Engineering from Missouri University of Science and Technology and Baroda University, India, respectively.
Hosted by Ken Shepard.