Biology Reference
In-Depth Information
6
CHAPTER
Computational Protein
Design for Synthetic Biology
Florian Richter
1
and David Baker
2
1
Institut für Biologie, Humboldt-Universität zu Berlin, Berlin, Germany
2
Department of Biochemistry, University of Washington, Seattle, Washington, USA
INTRODUCTION
The objective of (computational) protein design and engineering is to create proteins with
functions that are not available in natural proteomes. In many synthetic biology
applications, just like in natural organisms, proteins are the workhorses that carry out the
actual desired functions. But since natural proteins and the functionalities they exhibit were
evolved in facilitating the survival and maintenance of cells and organisms, the synthetic
biologist
101
s toolbox is limited to functions necessary for that purpose. And while the set of
naturally available proteins is already very large and can be used to design cells and
organisms with novel properties, synthetic biology would benefit tremendously from
transcending this barrier and being able to design proteins with functional properties that so
far have not naturally evolved. For example, if one wants to engineer a strain of
Escherichia
coli
that produces a certain small molecule of interest, one is dependent on the existence of
an enzymatic synthesis pathway for said molecule. However, if the molecule of interest is
artificial, such as a drug or biofuel candidate, it is unlikely that a natural synthesis pathway
exists. In this case, a novel enzyme needs to be designed that catalyzes the desired reaction.
'
In a way, a synthetic biologist without the capability to create custom-tailored proteins is
like an architect who is limited to using only naturally occurring materials like mud, wood,
and stones to erect buildings. And while buildings with these materials are good enough for
certain applications, the development of more advanced materials like brick, steel, and glass
immensely increased the size and type of possible buildings. Similarly, once proteins with
new functions can be reliably engineered, synthetic biology will take a huge leap forward.
Computational protein design (CPD) is by no means the only method available to engineer
proteins. Other approaches such as directed evolution, which is discussed elsewhere in this
topic, have been employed to obtain impressive results. Computational design does,
however, have some unique advantages that allow it to address problems not amenable to
directed evolution, as we will demonstrate in this chapter. Conversely, if sufficient high-
throughput assays for the function of interest exist, directed evolution is better suited to
improve proteins starting from a threshold level of initial activity. Thus, computational
design and directed evolution are perfectly complementary, and we anticipate that these two
methods will often be used hand-in-hand when designing new proteins for real-life
applications.
