Tech Showcase: Improving Membrane Protein Expression

“DNA makes RNA, and RNA makes Protein” is the central dogma underpinning all molecular biology. Proteins are the cogs in the machine that is the cell, performing many of the functions that make something “alive.” All life is defined in form and function by the information stored in its DNA. Some of this information forms genes, which the cell translates into proteins via messenger RNA.
Proteins are minute molecules comprised from a chain of amino acids. There are 20 canonical amino acids, and the order of these unit parts is defined by the sequence of the four nucleic acid bases that comprise the genetic code. Some amino acids are sticky, some make strong foundations, some like water, some hate water, some are reactive and some are unreactive. By virtue of these properties, each protein will fold upon itself in a complex origami, forming a unique 3D structure controlled by the order of its amino acids.
These 3D amino acid structures can be likened to tools that are used by the cell to perform a swath of functions. One family of proteins is the enzymes – they typically exist inside the cell. Enzymes act as catalysts, speeding up chemical reactions that otherwise would lack the energy to occur. Another family is the membrane proteins. These proteins sit within the outer walls of the cell, integrating into the fatty cell membrane. Some membrane proteins are the cell’s eyes and ears, providing perception by interacting with their surroundings; others are protectors, defending the cell from external threats.

An illustration of a chain of amino acids folded into a complex 3D structure, forming an antibody, a type of immune protein.

In the pursuit of understanding life, it is important to study how proteins function and interact – although each protein is nanoscopic, it is the hum of activity from billions of tiny protein parts that come together and make us…us.
To study proteins in the lab, it is typical that non-natural volumes of protein are required to amplify any measurable signal from their activity. Researchers, therefore, have devised methods to force cells to create vast quantities of any protein of interest.
High throughput gene synthesis technologies provided by Twist Bioscience have simplified this process, as one way of overproducing protein involves the introduction of the protein encoding gene into a laboratory cell line like Escherichia coli. With high throughput DNA synthesis, a combination of low pricing and high DNA quality allows researchers to study more proteins per experiment.
Once new genetic material encoding a protein of interest has been inserted into a cell, and the protein has been expressed, there are several ways to study it. Membrane proteins use the membrane to fold into their specific 3D shape, and therefore their function and properties are typically studied directly in the cell context.
Integral membrane proteins are a subset of membrane proteins that are deeply rooted into the fatty molecules that make up the outer cell surface. Many integral membrane proteins are extremely challenging to study. Cells can be produced that make lots of the protein of interest, however the cell will often have difficulty integrating useful, measurable volumes of the protein into its membrane.
This bottleneck in protein integration happens because integral membrane proteins have to be threaded through narrow membrane channels before their water-hating amino acids stitch the protein into the fatty layer. The speed at which this process happens depends largely on the protein’s amino acid sequence.

A diagram showing different types of membrane proteins (blue) integrated into a cell’s double layered membrane. Source: Wikicommons. Author: LadyofHats.

Researchers Professor William M. (Bil) Clemons Jr. and Professor Thomas Miller from the California Institute of Technology have developed a new technology that helps the scientific world better understand this integration process, published in the Journal of Biological Chemistry. "In this publication we experimentally demonstrated that we can computationally optimize integration of proteins into the membrane resulting in predictable improvements in expression,” Clemons explained.
To better understand the membrane integration bottleneck, the team built a model able to predict how efficiently a protein can travel through a narrow bacterial membrane channel called the Sec translocon. This model was then applied to better understand the dynamics of the difficult-to-study integral membrane protein, TatC, and predicted that modifications to this protein’s sequence could have an affect its membrane integration. The researchers then validated this model by producing a library of mutant TatC proteins representing 140 different modifications to the protein sequence.  
The effect on membrane integration efficiency by each modification was then tested by introducing the library into bacteria. When compared with their model, the specific types of protein modifications that would improve overall membrane protein integration were accurately predicted. In the world of membrane protein research, this is significant, as this model provides researchers with a tool that can pinpoint ways to push more of their tricky to study protein into the membrane, massively improving any measurable signals.
Until recently, the volume of DNA required for such research was inaccessible. "Success of this work depended on creating a statistically significant library of sequences, which was made possible by Twist Bioscience," added Clemons.
Now, with the help of high throughput DNA synthesis, this work has broken open the study of integral membrane proteins. Researchers will be able to rationally design proteins for improved expression, providing new ways to better our understanding of the nanoscopic systems that govern life. Ultimately such a tool could offer improvements a number of fields, including the study of disease and the development of new targeted medicines.