Chemistry as biology by design

Abstract

On 3 October 2018, Prof. Frances H. Arnold of the California Institute of Technology (Pasadena, USA) was announced as a recipient of the Nobel Prize in Chemistry for her pioneering work on directed evolution. Her corecipients, George P. Smith of the University of Missouri in Columbia (USA) and Sir Gregory P. Winter of the MRC Laboratory of Molecular Biology in Cambridge (United Kingdom), developed phage display and used this approach to evolve antibodies with enhanced binding properties, respectively. The recognition of Arnold with Smith and Winter may lead one to believe that her work has focused on medical applications. However, the Nobel committee lists as the prize motivation: ‘for the directed evolution of enzymes’.1 Enzymes, nature's catalysts, are essential for molecular transformations, and the ability to radically alter – or even create – their catalytic function is key to the true emergence of a bioeconomy, built upon principles of renewability and sustainability. It is no secret that Biology is quite a remarkable Chemist. The rich history of natural products as therapeutic agents provides ample evidence of the exquisite ability of enzymes to break and make chemical bonds, ultimately forming complex structures such as those found in antibiotics and antifungals that have proven essential for human health and development. While many of these molecules are accessible through synthetic organic chemistry, most are not practically so. That is, the large number of reaction steps and low overall yields of chemical synthesis make fermentation the preferred method of production. Beyond natural products, other biochemicals have been efficiently produced commercially through fermentation, including various amino acids (Wendisch et al., 2016), citric and lactic acids (Chen and Nielsen, 2016) and, of course, the old standby ethanol. In some cases, like ethanol, little manipulation of the microbial host is required to achieve acceptable production metrics, with process design as the only engineering input to the system. In others, metabolic engineering comes into play, through which organisms (typically microbes) are modified to increase product titre (i.e. concentration), rate (i.e. productivity, in units of concentration per time) and yield (i.e. fraction of available substrate converted to the target product). All of these molecules, however, have one thing in common: they are all naturally formed through established biosynthetic pathways. Does this mean that only those compounds first identified as natural biological metabolites are amenable to the fermentation-based approach to chemical synthesis? Of course not – that would be too boring! Enzymologists have known for years that biocatalysts do not conform to the one-lock/one-key model often presented in first-year biochemistry courses. This knowledge has been exploited to produce a number of molecules through single-step reactions, perhaps most notably in the synthesis of pharmaceuticals (Bornscheuer et al., 2012). The plasticity of enzymes has also been exploited in complete metabolic pathways, enabling the design of novel biosynthetic routes (Atsumi et al., 2008; Martin et al., 2013). In many cases, however, the catalytic activity of the enzyme when working on a non-native substrate is too low to be practically useful. Enter directed enzyme evolution (Zhang et al., 2010). Computational protein design has even made it possible to create enzymes that perform reactions not found in nature (Rothlisberger et al., 2008). Perhaps, not surprisingly, these enzymes frequently perform quite poorly. What to do to make them stronger, faster, better? Directed evolution (Khersonsky et al., 2010)! Proceeding directly to evolution has also enabled the construction of an enzyme that performs a reaction not found in nature, one that forms bonds between carbon and silicon, two earth-abundant elements (Kan et al., 2016). Taken together, what do these technological advancements already achieved a