I have been in the distinctive business of making proteins, looking at proteins, or finding proteins that have special properties. I mostly try to make them myself, and over the years I’ve had lots of success—and many failures. It’s part of research. We all know, from DNA we get transcribed RNA, which is translated into a strand of amino acids, which by forces of physics fold into a functional protein—sometimes with some help from chaperones. However, it does not always work. Even in a living organism with high adaptability, success isn’t guaranteed. Aggregation is a waste of energy for the organism, and now it needs time and energy to recycle the materials.
In industry, about 40 percent of all pharmaceutical proteins are made with recombinant protein production in bacterial strains, while the remainder is mostly by use of mammalian cells. In the lab, we use E. coli the most, simply because many tools are available—from cloning vectors to E. coli strains from DNA plasmid production, to cloning kits, simple transformation procedures, and plasmids designed for high yields of recombinant protein production. Many different media and fermentation procedures are designed by scientists and manufactures to increase yields of correctly folded protein. It all seems so simple, but it’s not.
Even a condensed, incomplete list of tools illustrates that we need to make many choices to find the right parameters. For now, I will skip most tools and focus only on protein yield. In the lab, we need just a little bit of protein, but as a bioprocess engineer, I always need to consider the choice of the expression system on the available means of the laboratory or production facility, modification needs of the protein produced, and compatibility of the gene control system with the bioprocess for production later on.
In terms of protein yield, either as yield per cell or total yield, it is important to remember that the product formation must be determined experimentally, and may either be growth associated or non-growth associated. The specific product rate formation, qp (kg kg h-1), is given by qp = qs Yp/s, where qs (kg kg-1 h-1) is the specific substrate rate and Yp/s (kg kg-1) is the yield coefficient. The specific product rate formation can also be expressed in terms of growth association or non-growth association by means of the Luedeking-Piret model: qp = αµ + β. Here, µ is the specific growth rate, β = 0 gives a complete growth association, and α = 0 gives a complete non-growth association. Protein product formation is mainly growth associated, at least in wild type cells. Low growth rate, for example in E. coli, may hinder product formation due to the maintenance cost of the cell, while fast product formations may hinder correct folding of proteins. These problems are related to protein production in living cells, but less so in cell-free expression systems.
Important parameters to consider are again plenty. Are we using a simple shake flask or do we have access to a fermenter? In either case, we still have to choose if we grow the cells in a batch phase or use a fed-batch culture, i.e. if we grow the cells following sigmoidal growth curve or a linear growth curve with a fixed number for µ. Batch phase in shake flasks is very common because it is cheap, however it has its own unique problems. For example, using LB medium would be a poor choice for long expressions. Cell densities are often low (OD600 < 6) due to acidification of the medium, while the pH in Terrific Broth (TB) can go up above a pH of 8.5, indicating high ammonia production due to utilization of amino acids as carbon source. The growth rate of E. coli is dramatically reduced under a pH of 5, and thus will not produce recombinant protein—it will start recycling it for its own maintenance. The utilization of amino acids as a carbon source will also lower the amount of recombinant protein.
The design of the shake flask (i.e. baffled or round), the size of the opening, and the choice of the cover are equally important, since E. coli can only grow and make recombinant protein if the oxygen transfer rate is as high as possible. Switching from an aluminium cover to very porous paper can increase the final biomass manyfold.
In conclusion, aside from the choice of DNA vector, E. coli expression strain, growth temperature, inducing agent and amount, and co-expression of chaperones, parameters to control the growth rate are equally important. Growth media, a good oxygen transfer, and a good buffering system have proven in my work to be important elements to obtain high yields of correctly folded protein.
(c) Marco Casteleijn. 2017
“Preparing recombinant protein: Tips to improve growth rate and yield” was published originally on the 23rd of May, 2017, in the ‘The Q’ . Republished with permission.