Dipnath Baidyaroy, Director, Strategic Alliances, Business Operations
Stefanie Ng Minor, Manager, Product Management, Business Operations
Revolutionary advances in digital technology, genomics and sequencing have fueled tremendous advances in protein engineering. Enhanced understanding of the complex relationships between protein structure and function has unveiled virtually limitless opportunities for engineering proteins that meet precise performance specifications. Using highly advanced techniques that combine the disciplines of biochemistry, chemistry, recombinant DNA technology, structural biology, biochemical engineering and information technology, protein engineering enables scientists to create enzymes that are far superior to those available from nature, sometimes even absent in nature.
A key resultant of the protein revolution is the rapid development and production of high-performing enzymes that serve custom applications. Protein engineers are designing and producing optimized proteins that dramatically reduce the cost and improve the quality of pharmaceuticals and food ingredients, while also creating novel biotherapeutics and enabling the sequencing of minute quantities of DNA for in vitro diagnostics. As protein engineering techniques continue to evolve, suppliers of these services are increasingly able to deliver optimized proteins that precisely match their clients’ desired performance specifications — and can do so faster and less expensively than ever before.
The rationale for engineered enzymes
Scientists developing the manufacturing route to any chemical — be it an active pharmaceutical ingredient (API), a food ingredient, a fragrance or a flavor molecule, or even a commodity chemical — often need to assemble a series of chemical reactions that convert an inexpensive raw material to the desired commercial compound. These chemical reactions are catalyzed by chemical or biological catalysts.
A wide variety of catalysts occur in nature and have long been used in various ways to enable or speed up reactions. Until recently, most catalysts were transition metal based, with limited catalytic scope and selectivity. Moreover, metal based catalysts are often toxic or rare and are usually not very optimizable beyond their initial performance. Additionally, reaction conditions for chemocatalysts are usually much harsher than for biocatalysts, as the reactions must occur at elevated temperatures and at higher pressure. Chemocatalytic processes also generate more waste and thus increase costs.
By contrast, biological catalysts (ie. enzymes) are malleable and can be manipulated significantly through the process of protein engineering. Engineered enzymes have the significant advantages of simplicity, scalability and stereoselectivity, enabling the production of chirally pure compounds — an especially important consideration in API manufacture, as most drugs are chiral compounds. Additionally, biocatalytic processes use standard lab or plant equipment, take place within ambient temperature and normal pressure ranges, and do not require complex controls. These advantages provide a major benefit for companies that outsource manufacturing to contract manufacturers.
The most important benefit of biocatalysts may lie in the fact that engineers can devise the most cost-effective and sustainable route of manufacture, and then engineer the protein catalyst to fit this “ideal” process; this feature is often unavailable for chemocatalysts, for which a suboptimal process has to fit the catalyst.
The CodeEvolver® protein engineering technology
CodeEvolver®, is a protein engineering technology platform developed by Codexis that enables the modification of proteins to enhance specific performance characteristics. This technology allows Codexis scientists to first identify the best natural enzyme that can perform the desired chemical transformation, and then use the proprietary CodeEvolver® protein engineering platform to improve the enzyme. Using bioinformatics tools, Codexis scientists manipulate the sequence of the initial enzyme to create a library of variant enzymes that contain mutations at specific positions. Screening of this library enables identification of variant enzymes with improved performance for the desired chemical reaction. Analysis of the library identifies the specific advantageous mutations that lead to these improved traits. With this information in hand, scientists can then combine the beneficial mutations to produce the best possible variant enzyme.
The CodeEvolver® protein engineering process consists of in silico screening, introduction of function-driven mutations, high-throughput screening and sequencing, machine learning to identify a productive combination of mutations, and finally, the human expertise of the scientific team. This process is repeated iteratively for compounded improvements until all the desired traits are incorporated into the final enzyme. The technology reflects an accelerated evolutionary process: something that happens in nature, but is harnessed in the lab for faster and specific improvements at a molecular level.
Another advantage of the CodeEvolver® protein engineering platform is the speed at which an optimized enzyme can be delivered into a customer’s commercial process. Once an improved enzyme has been engineered, the technology enables rapid scale-up of the enzyme manufacture, in some cases allowing customers to manufacture product at metric ton levels using the biocatalytic route within a three- to six-month period. Customers can also access Codexis’ CodeEvolver® technology by purchasing an off-the-shelf screening kit to assess the feasibility of a biocatalytic route in their own labs.
One of the most powerful elements of the CodeEvolver® protein engineering technology is that it is an additive construct, in that every cycle of variations and protein optimization generates more data. This benefit allows Codexis to undertake each successive project with more knowledge, facilitating the rapid identification of candidate proteins from nearly infinite possibilities, as well as the engineering and optimization of those proteins to meet specific customer needs. Today, Codexis can accommodate 15 unique protein engineering projects in parallel, with each project taking an average of three to six months to achieve its intended performance targets. By contrast, at the time Codexis was incorporated in 2002, a single project might take two years or more to reach fruition.
Evidence of customer benefits
Enzyme optimization can yield benefits for a variety of companies in the life sciences, food industries and other applications. For example, implementation of the CodeEvolver® protein engineering technology resulted in engineering of a biocatalyst that had zero starting activity to a commercially-relevant biocatalyst in less than 12 months, enabling a major pharmaceutical company that faced capacity constraints in its existing supply chain due to increasing demand for a blockbuster Type 2 diabetes medication to avoid building a new factory.
In this case, Codexis developed a new, high-performing enzyme catalyst to replace an expensive, toxic chemocatalyst that was used to produce the drug’s API. The biocatalytic solution eliminated several steps from the manufacturing process, increased the measured productivity of the process by 53% and decreased energy usage by 19%. Perhaps most importantly, use of the new biocatalyst helped the pharmaceutical company avoid the cost of building a second factory to meet the rising demand for the drug, significantly reducing capital expenditure requirements.
Another notable example involves a leading food industry supplier seeking a healthier ingredient that would meet demanding caloric and sensory targets. The new ingredient also needed to fulfill requirements for speed of commercialization, safety and lower production costs. Using the CodeEvolver® protein engineering platform, Codexis worked with the supplier to engineer a biocatalyst that would meet the customer’s goals and enable a simple, cost-efficient production process. The project resulted in attainment of commercial targets in seven months, and the engineered biocatalyst produced a 70-fold improvement in catalyst stability under challenging process conditions. The solution cut costs by 90% and enabled commercial production of the healthy ingredient less than two years after the first project discussion.
In addition to engineering proteins as biocatalysts, the CodeEvolver® technology is also being applied to create novel, targeted biotherapeutic solutions. Phenylketonuria (PKU) is a rare, inherited genetic disorder that strikes approximately one in 15,000 newborns in the United States. The disorder is characterized by a deficiency in the enzyme that converts the essential amino acid phenylalanine into tyrosine, resulting in the accumulation of high levels of phenylalanine in the brain, where it causes serious neurological problems including intellectual disability, seizures and cognitive and behavioral problems. Whereas a screening test for PKU has been available for more than 50 years, treatment has essentially been limited to dietary control.
Codexis has engineered a biotherapeutic enzyme candidate for treatment of PKU that compensates for the absence of the patient’s missing natural enzyme that metabolizes phenylalanine in the body. The engineered enzyme’s stability in the gastrointestinal tract enables convenient oral dosage, and pharmacokinetics analysis has revealed a greater than 50-fold improvement in stability in vitro, compared to the natural enzyme. Four preclinical models have demonstrated efficacy, and Codexis aims to initiate human trials in 2018.
Unsurprisingly, much of the innovation in protein engineering and enzyme optimization is driven by customers seeking a competitive edge in a highly dynamic commercial environment. In their quest to be first-in-class or first-to-market, every customer that approaches Codexis wants to increase manufacturing efficiencies to drive costs down and improve product quality, to create a better targeted biotherapeutic, or to improve sensitivity and precision on their next-generation sequencing (NGS) workflows. Codexis’ success in multiple market application of their proprietary CodeEvolver® protein engineering technology allows customers to seek a protein engineering solution to deliver on the customers’ precise specifications. Moreover, greater customer awareness of the benefits of advanced protein engineering increases the demand for engineered enzyme solutions, and the accumulated experience expands Codexis’ knowledge base exponentially with each project.
As the benefits of advanced protein engineering and protein optimization become more widely known, it is reasonable to expect continued innovation in this field well into the future. In addition to leveraging optimized enzyme solutions in the manufacture of APIs and food and beverage ingredients, cost- and resource-conscious customers will also seek to apply protein engineering to renovate existing or develop new small-molecule drugs, conjugated antibodies, animal feed enzymes, molecular diagnostic enzymes and biotherapeutics.
In particular, customers in the life sciences fields will benefit from the maturation of methods to produce more sensitive, fluid-based molecular diagnostic tests from optimized NGS enzymes, which will enable less invasive detection of cancer and other diseases. Such development may give fresh momentum to the personalized medicine movement as greater numbers of patients and their physicians seek clearer, more precise diagnoses based on the molecular signature of their diseases. Plus, protein engineers’ initial forays into optimizing proteins as biotherapeutics, like Codexis’ biologic lead candidate for PKU, may yield advances in the treatment of various orphan and various diseases.
Finally, advances in protein engineering and evolving customer needs will launch exploration in applications beyond biocatalysis, biotherapeutics and medical diagnostics. As the protein engineering toolbox expands, so, too, will engineers’ ability to improve the functionality and stability of proteins of interest. Indeed, the range of possibilities is as vast as the protein landscape itself.