Contributed by K. E. Hernandez, Ph.D. candidate working for Professor Frances Arnold, Division of Chemistry and Chemical Engineering, California Institute of Technology
Catalysts are important tools in green chemistry because they enable reduced-waste manufacturing methods by accelerating reactions where every reactant is incorporated into the product. Many catalysts that facilitate enantioselective bond formation are made from rare and non-renewable materials, such as palladium or rhodium; replacing these catalysts with renewable alternatives would allow for the more environmentally-friendly production of key chemicals. We believe that enzymes can meet this acute need for new, low-cost, sustainable catalysts, and in doing so, further advance and expand green chemistry.
In Frances Arnold’s lab, we start from natural proteins to develop new enzymes. Though a major drawback to enzymes is their limited reaction scope, expanding the range of reactions that can be catalyzed enzymatically will facilitate the widespread adoption of biocatalysis. Our team of chemists, biologists and engineers has expanded the reactions catalyzed by naturally-occurring proteins to include reactions unknown in the biological world using a process called ‘directed evolution.’
The powerful algorithm of evolution allows organisms to continuously update their catalytic repertoires with useful, new capabilities — think antibiotic resistance, or the ability to degrade many manmade compounds. Humans have been capitalizing on evolution to engineer desirable traits into biological systems for thousands of years: Everything from corn to cats has had its genomes altered through artificial selection and breeding to produce favorable phenotypes. Our innovation in Frances Arnold’s lab is that we do it with molecules. In the lab, we evolve enzymes by putting pressure on them to perform non-native functions that may not be useful to a bacterium, but are useful to us. Directed evolution mimics evolution through artificial selection accelerated in the laboratory setting by focusing on individual genes expressed in fast-growing microorganisms. We introduce mutations to parent proteins sourced from nature and screen the daughter proteins for increased activity for a desired reaction. We then use the ones with increased activity as parents for the next round of mutation and selection, and continue until we reach the desired activity and selectivity.
Although we can greatly enhance activity with directed evolution, a new enzyme activity has to come from somewhere. Thus, our starting proteins have to have at least small levels of the new activity in order to act as a starting point for the evolution of a novel enzyme. This is how nature creates new enzymes – we just follow the same recipe.
We have found that cytochrome P450s, whose native functions include monooxygenation, are a wonderful source of non-natural activities. We have engineered these proteins to carry out carbene and nitrene transfer reactions known to chemistry (e.g., olefin cyclopropanation, aziridination), but not found in biology. In recent projects, we have used a wider range of heme proteins as starting points to further expand the reactions catalyzed by biocatalysts.
Our lab has made a cyclopropanating enzyme that produces the chiral precursor to the antidepressant medication levomilnacipran (1), and we have pushed these industrially relevant biocatalysts to synthesize pharmaceutical precursors at larger scales. One recent project focused on making a chiral cyclopropane intermediate used in the synthesis of ticagrelor, a medication used to prevent the reoccurrence of heart attacks. We identified a truncated globin from Bacillus subtilis that catalyzes this reaction (Figure 1) at low levels and also showed some selectivity for producing the desired diastereomer of the ticagrelor cyclopropane precursor from ethyl diazoacetate and 3,4-difluorostyrene (2).
This enzyme variant underwent evolution through the mutagenesis of many residues within its heme-binding pocket. Screening libraries of mutant enzymes identified beneficial mutations Y25L, T45A and Q49A, which improved the activity and selectivity of the catalyst so that it yields the desired ticagrelor cyclopropane almost exclusively. The catalyst does not even have to be purified because the reaction proceeds well in whole bacterial cells that express the evolved enzyme. After testing a range of reaction conditions, we found that the slow addition of the whole-cell catalyst and ethyl diazoacetate solutions to 3,4-difluorostyrene gave virtually a single isomer (>99 percent dr, 98 percent ee) of the ticagrelor precursor in 79 percent yield in preparative scale reactions. This work demonstrates how directed evolution can rapidly optimize a newly discovered biocatalytic activity, olefin cyclopropanation, to synthesize useful products in high selectivity and yield.
Another recent project has focused on engineering a biocatalyst with a novel carbon-silicon bond forming activity that has also never been found in nature. C-Si bonds are seen in medicinal chemistry, imaging agents, elastomers, and high tech consumer products, such as televisions screens. Until now, the only methods used to create these bonds enantioselectively have relied on multistep chemical syntheses to prepare chiral reagents or chiral transition metal complexes. An iron-based catalyst had never been reported for this carbene insertion reaction. Then, postdoctoral researcher Jennifer Kan and her team discovered that cytochrome c from Rhodothermus marinus could catalyze the reaction between ethyl 2-diazopropanoate and phenyldimethylsilane to form the chiral organosilicon product with high enantioselectivity (Figure 2A, 3). They performed saturation mutagenesis on three residues within the active site that they thought were likely to influence enzyme activity, and through this, discovered the triple mutant V75T M100D M103E, which allowed the catalyst to form the C-Si bond with very high turnover numbers and enantioselectivity (>1500 turnovers, >99 percent ee).
Testing the engineered enzyme against a panel of silane and diazo reagents, Kan’s team discovered that the mutations in the triple mutant were broadly activating. The engineered enzyme was shown to catalyze the formation of 20 organosilicon products, most of which were obtained as a single enantiomer (Figure 2B). The triple mutant was also shown to have high selectivity for carbon-silicon bond formation over the formation of carbon-nitrogen bonds in the same substrates.
It is fascinating to see that at least some of nature’s vast catalog of proteins can be evolved — with only a few mutations — to efficiently create chemical bonds not known in biology and that the new biocatalysts can access areas of the chemical space that biology has not explored.
The future for biocatalysis is bright – companies embrace the technology to replace wasteful stoichiometric processes and catalytic processes that rely on costly and unsustainable rare metals (4,5). As the community continues to discover and develop new enzymes, the applications of biocatalysts for sustainable chemical production will grow. This year, our lab has explored a wide range of natural protein diversity to develop useful enzymes with wide-reaching applications. Through this, we have demonstrated that nature has the capacity to quickly produce and optimize catalysts for novel reactions — all we have to do is ask the right question and then evolve.
“The Nexus Blog” is a sister publication of “The Nexus” newsletter. To sign up for the newsletter, please email firstname.lastname@example.org, or if you have an ACS ID, login to your email preferences and select “The Nexus” to subscribe.
To read other posts, go to Green Chemistry: The Nexus Blog home.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.