Protein optimization

Enzymes evolved by nature do not always live up to the performance criteria of industrial applications.

It is not always possible to use enzymes isolated from nature to solve a given industrial problem. Many industrial processes use conditions that put extreme demands on the enzyme used. Examples of such conditions are very high temperature, extreme high or low pH, or high concentrations of detergents. In these cases our enzyme optimization platform can be used to improve the desired property of the enzyme so that it will work under the condition imposed by the industrial process. Characteristics like temperature stability, solvent resistance, or even substrate specificity are among the parameters that we routinely alter in our enzyme products. Our protein optimization platform utilizes several advanced technologies, including protein modeling, protein engineering, directed molecular evolution, and-high throughput screening.
As a typical enzyme often consists of more than 300 amino acids, and as 20 possible amino acids can occupy each of these positions, the number of combinations that can be produced is almost endless. We therefore need to be smart in the way we generate our diversity. The simplest scenario is when the structure of an enzyme is known and our computational scientists are able to come up with suggestions for specific changes in the enzyme that will alter it toward the desired goal.
These changes are then introduced into the enzyme one by one by our protein engineering experts, and the altered proteins are purified and tested. A more complicated situation arises when the structure of the enzyme is known but the computational analysis is unable to suggest specific changes, merely suggesting regions of the enzymes where introduction of amino acid changes may hypothetically confer a desired change. In this case it is often not sufficient to make single enzyme variants, but instead we create libraries of enzyme variants in which the diversity we introduce is targeted toward the regions suggested by the analysis. Such enzyme libraries can contain from a few hundred enzyme variants up to millions or tens of millions of different enzymes. In these cases it is no longer possible to purify and analyze each individual member of the library; we must rely on intelligent prescreens and high-throughput screening with robotic equipment.


Protein modeling 

Combining knowledge of protein structure with the wealth of sequence information.
Detailed information on the molecular structure of enzymes is an important tool for changing or improving the natural activity of enzymes. On the basis of the three-dimensional structure of an enzyme it is possible to extract precise information about the enzyme’s properties and function.
As the three-dimensional structure of an enzyme is so closely linked to its properties, it is a key point in enzyme optimization to be able to reveal this structure. At Novozymes our researchers do this using X-ray crystallography and NMR techniques. This information has been used to develop new enzymes with amazing properties that can be of industrial benefit. Another computational approach is homology modeling, where the structure of a protein can be calculated if we know the structure of a close homologue.

Protein engineering   

Protein engineering has formed the basis of many of Novozymes’ most prominent products.
An enzyme consists of several hundreds of amino acids linked together in a delicate three-dimensional structure. This structure determines the properties of the enzyme such as reactivity, stability, and specificity.
Through extensive research we know the exact location of each amino acid in the three-dimensional structure of most of our enzymes. Based on years of experience with these enzymes and aided by specially programmed supercomputers we are able to suggest how to change one or a few of the amino acids in the enzyme so that the enzyme will function better in a desired application. In this way we can engineer an enzyme to meet the needs of our customers and optimize its function.
As soon as we have discovered possible ways to improve the three-dimensional structure of a specific enzyme, our scientists can construct these slightly altered enzymes by modifying the gene encoding of the enzyme. Engineered organisms then produce the modified enzyme, which is tested to evaluate whether the predictions have been correct.
Using protein engineering we have made enzymes that can withstand boiling, enzymes that work at extreme pH (both acidic and alkaline), enzymes that can withstand high concentrations of bleach, enzymes that have improved activity, and enzymes that work on several new substrates.

Directed molecular evolution

We use the same strategy as nature to develop new enzyme products. When conditions change in the environment, new organisms better adapted for survival under the new conditions emerge through the evolutionary process.
The underlying force behind the process of evolution is that slight modification of the DNA in the cells can cause the organisms to evolve new traits, including small changes in the enzymes produced by the organisms. Such modifications, known as mutations and recombinations respectively, are completely natural and occur spontaneously within all living organisms.
If our scientists are unable to solve a specific problem using our collection of enzymes isolated from nature, or to develop an enzyme using protein engineering, they can use a technique known as directed molecular evolution, which mimics the evolutionary process taking place in nature to create new improved enzymes. Directed molecular evolution is a highly efficient method for making improved enzyme solutions. By forcing enzymes to evolve in the laboratory many new enzyme products have been discovered. The first enzyme developed using such methods, a detergent lipase, was developed by Novozymes many years ago.