UCLA

Methanol, a common derivative of natural gas, is increasingly attractive as a chemical feedstock due to its low cost and abundance. Among the technologies that converts methanol to value added chemicals, biological conversion has the advantage of mild reaction conditions low capital cost. Several methanol assimilation pathways can be found in nature. However, carbon yield for production of acetyl-CoA derived products using these pathways has intrinsic limits due to CO2 lost in pyruvate decarboxylation or ATP requirement. To address the carbon yield problem, as our first Aim, a synthetic pathway termed methanol condensation cycle (MCC) was designed and demonstrated. MCC is an ATP-independent pathway that is capable of stoichiometric conversion of methanol to acetyl-CoA, a precursor to many longer chain products. With a cell-free system, a catalytic cycle of MCC was confirmed using 13C labeled methanol. The in vitro production was further optimized by adjusting the key enzymes to avoid “kinetic trap”, where excess amount of certain enzymes would reduce the robustness of the cycle. As a result, the Aim was accomplished by demonstrating conversion of methanol to ethanol and 1-butanol with high carbon yield (80% and 50%, respectively).

During our work in realizing MCC, poor catalytic efficiency of NAD-dependent methanol dehydrogenase (Mdh) had been a major hurdle. The enzyme catalyzes a thermodynamically unfavorable methanol oxidation reaction and has higher activity towards ethanol and 1-butanol instead of methanol. Therefore, our second Aim was to bioprospect and engineer Mdhs for improved methanol oxidation activity. We first identified and characterized Mdh2 from Cupriavidus necator N-1 with significant activity towards methanol. We then employed directed evolution with high-throughput screening to further enhance methanol activity and specificity. As a result, the engineered Mdh2 variant CT4-1 showed 6-fold higher Kcat/Km for methanol and 10-fold lower Kcat/Km for 1-butanol.

With the experience in in vitro methanol conversion to chemical products and Mdh engineering, our third Aim was to engineer E. coli for methanol dependent growth and production. We constructed an E. coli strain that couples methanol assimilation to growth. By disrupting ribulose-phosphate 3-epimerase or ribose-5-phosphate isomerase in the pentose phosphate pathway, the strain was unable to utilize xylose or ribose as sole carbon source. Expression of methanol assimilation enzymes from the ribulose monophosphate pathway allowed the strain to co-utilize methanol with either one of the pentoses. This strain was termed as “synthetic methanol auxotrophy” since it cannot grow without methanol. This strain allowed us to optimize methanol assimilation by evolving for faster methanol-dependent growth. Our best strains were able to utilize methanol for growth to an OD600 of 4.0 in 30 hrs with methanol and xylose co-assimilation at a molar ratio of about 1:1. Genome sequencing and reversion of mutations indicates that mutations on genes encoding for adenylate cyclase (cyaA) and the fromaldehyde detoxification operon (frmABR) are necessary for the growth phenotype. The methanol auxotrophy strain was further engineered to produce ethanol or 1-butanol to final titers of 4.6 g/L and 2.0 g/L, respectively. We also demonstrated the utility of this strain as a selection platform to significantly decrease residual methanol amount in growth medium. The synthetic methanol auxotrophy strain represents a useful platform for engineering methanol conversion.