25 Years of Protein Expression

The first steps in the production of any protein are transfecting a cell line, achieving expression of functional protein, and selecting a high-expessing clone for production. Charles H. Squires, vice-president of discovery and external partnerships at Pfenex Inc., spoke with BioPharm International about advances in protein expression technology, and discussed the challenges yet to be overcome.

BioPharm: Can you describe how the process of protein expression has improved or changed over the past 10 years or so? What are one or two key technological advances that have made these changes possible?

Squires: New methods to more rapidly and reliably express proteins have proliferated over the past decade and a half. Innovations in this area have included the improvement of expression hosts to introduce more phenotypic capability to produce active, undegraded protein. For instance, hosts have been modified to overexpress folding modulators or mutated to eliminate proteolytic activities. Also, there has been a large increase in the number of these tools available commercially, putting many effective approaches within the reach of researchers with limited resources. The past several years have also seen the emergence of proprietary platform technologies, each of which offers some advantages in expressing difficult proteins or which delivers an advantageous downstream process. Examples include platforms such as a strain of Escherishia coli that secretes product to the extracellular space (Wacker Chemie AG), a toolbox of regulated promoters in the yeast Pichia pastoris (VTU Technologies), and a completely unique Pseudomonas fluorescens-based expression platform (Pfenex). The Pfenex platform has proven highly effective in expressing a variety of types of proteins and employs a high-throughput parallel processing technology in which a thousand or more unique expression strains are constructed and analyzed in about a month.

Similar types of improvements have been made in the screening of mammalian expression hosts for the extent of target protein expression, including fluorescence-activated cell sorting for the selection of productive clones. These technologies are particularly effective in the identification of cell lines expressing high levels of engineered monoclonal antibodies.

In addition to these types of improvements to host and expression strategy, advances in gene design that go beyond the traditional adherence to codon usage tables have been implemented, such as the technology developed by DNA 2.0. Expression data-taught, host-specific algorithms that modulate DNA sequences while keeping protein amino acid sequence constant may add yet another dimension to the list of effective protein production tools. The community of researchers involved in all aspects of protein production is certainly much better off than it was 20 years ago in its ability to provide solutions to protein expression, but it will take much more work understanding all aspects of the nature of protein expression to arrive at universally applicable principles. Those solutions will necessarily involve an ever widening toolbox of approaches that will need to be accessed through more advanced parallel processing tools. The principle difference now is that a multidimensional approach to gene and protein expression is available to the great majority of researchers, whereas 20 years ago the process was one dimensional and digital, giving only a yes or no result.

BioPharm: What parts of the protein expression process are still rate-limiting, and how do you see these challenges being addressed? What improvements do you expect to see in the future?

Squires: We anticipate that given the continued need for protein supply across the spectrum of applications, many of the existing weak links in protein expression will be strengthened. One of the most pressing unmet needs is production of small amounts of protein for research purposes. Often, dozens of proteins ranging from antigens for preclinical testing to related variants of a target protein for some form of differential testing are required. While only a small amount of protein, a few mg, is generally required for activity evaluation or crystallization for these types of studies, currently all of the cloning steps needed to get even that small amount of protein are identical to those required to assemble a strain or cell line for large-scale biopharmaceutical production. This is highly inefficient and a consistent bottleneck in the process. Clearly, there is a need to develop off-the-shelf solutions to fill this gap. New, precise, and more adaptable tools allowing more rapid cloning and characterization will need to be developed. These tools will necessarily include the development and adoption of more effective high throughput processing methods. Improvements in genetic, analytical, and processing methods practiced in high throughput, parallel modes will enable the researcher to access more tools and reach a wider range of solutions more rapidly.

Clearly, for these approaches to have an impact, methods to allow the researcher to identify productive clones at the earliest possible stage must be developed. Better reporter systems that may enable screening, even at the colony or single cell level, would be very valuable. Processing to obtain small amounts of protein would then benefit from the development of more efficient purification tags. These may also be envisioned to include reliable expression reporter sequences, sequences that promote proper protein folding, and ideally be efficiently and universally removable. Pfenex is exploring this segment, adapting its expression platform strategy specifically for custom reagents. Development in this area would be of great benefit not only to the core of researchers producing proteins for research purposes but the benefits would be amply felt in the area of target protein expression as well.

Another route to solve the problem of small-scale supply would be to adopt a cell-free approach such as that employed by Sutro Biopharma. Such an approach is enabled today by the relatively rapid, commercial availability of synthetic, gene-length DNA sequences. Developing an off-the-shelf cell-free system would jump the cloning and cell growth steps in the current processes and potentially deliver advantages of protein quality as well. An in vitro system would also be able to take advantage of the reporter and purification technologies mentioned above for in vivo systems. As more and more dimensions to protein production are added, the researchers' ability to deliver will continue to improve. The challenge will continue to evolve, however, as more and more types of hybrid or completely synthetic proteins are developed that may or may not have some evolutionarily defined path to proper folding and thus activity.