Biopharmaceutical Manufacturing and the Power of Synthetic Biology
Synthetic biology has advanced the scope and scale with which biologically derived therapeutics can be developed.
Elnur/Stock.Adobe.com
Traditional small-molecule drug discovery-the mainstay of pharmaceutical R&D for more than a century and a half-has become increasingly challenging, as the “low hanging fruit” of universally effective therapeutics have been slowly exhausted. The only real potential areas for major growth in the small-molecule drug sector are now the discovery of new drug targets that come from furthering our understanding of disease and the development of precision medicines targeting specific subpopulations of patients.
Fortunately for the industry, the rise of biotherapeutics is making up for this gap in traditional discovery pipelines, providing greater specificity, new avenues of therapy, and the possibility of combating previously untreatable diseases. Somewhat less fortunately, the commercial development and large-scale production of biological agents are far more complex than classical chemical syntheses. This means that many novel biotherapies never make the transition from the lab bench to the clinic, simply because the requisite biological molecules can’t be produced in the purity or quantity required.
Pick a host...
Biotherapeutic production generally relies on inserting genetic elements encoding the peptide, protein, or antibody of interest into a host organism capable of producing it in large quantities. There are a wide range of potential cellular hosts that can be employed for bioproduction-each with its own unique advantages and limitations. At present, Escherichia coli (E. coli) or Chinese hamster ovary (CHO) cells are the “go to” systems for the majority of labs looking to produce biological molecules that are ultimately intended for clinical applications. However, more often than not, this is because they are the host organisms researchers are most familiar with, rather than the optimal choice of expression system. This generally isn’t an issue in an upstream R&D setting, where only small quantities of the biotherapeutic are required, but can lead to yield and solubility issues when attempting to scale up production for preclinical and clinical studies, or for manufacturing. This is a real stumbling block for many potential biotherapeutics, as the resources required to scale up production in an inefficient or unsuitable host can make it economically unviable.
... any host
Foreseeing and overcoming the various issues that can arise during the biomanufacture of novel drug products requires a broad understanding of the various host organisms available, their strengths, and, crucially, their weaknesses. While there are no set rules to determine which host may be the most effective for the production of a given biomolecule, a broad expertise and holistic understanding of each organism’s metabolism can help to avoid problems downstream in product development. Experience is crucial here, as is having access to an extensive toolbox of synthetic biology (synbio) and metabolic engineering technologies (Table I) designed to optimize gene expression and production of the target molecule.
Table I. Summary of synthetic biology approaches to enhance biologics production. PCR is polymerase chain reaction. KOs are knock-outs. DSP is downstream processing.
The major challenge for drug discovery groups-or even biotech companies-attempting to develop and exploit these technologies in-house is that most simply don’t have the breadth of knowledge necessary for success. They are, no doubt, experts in their specific technology or disease area, but the scale-up production of biotherapeutics requires a very different skill set, covering aspects of synthetic and molecular biology, fermentation, and chemistry as well as good manufacturing practice (GMP) manufacturing.
The true power of synbio for biomanufacturing comes from the ability to combine commercially validated synbio tools (Figure 1) with other techniques and technologies to enhance the overall production system. For example, transcriptomics and proteomics-and even more traditional bioengineering techniques, such as ultraviolet (UV) mutagenesis-can be used to identify further enhancements that improve the hosts’ overall production capabilities.
Figure 1. Technology readiness level cascade for the commercial validation of synbio tools. DSP is downstream processing. cGMP is current good manufacturing practices. QMS is quality management system.
The expression capacity of the system can then be further increased through strategies such as ribosome engineering-to uplift overall gene expression-as well as the over-expression of regulatory genes, and the suppression or knockout of genes that may compete for resources, or could metabolize or destabilize the final product.
The need for a holistic approach
The main challenge in biomanufacturing is that scaling up production of therapeutics requires not just an understanding of the individual elements involved in the process-cell culture maintenance and fermentation, genetic manipulation, codon optimization, directed evolution, metabolic engineering, transcriptomics, and proteomics, etc.-but also how all of these components interact as part of a holistic biomanufacturing system. Worse still, optimizing these bioproduction systems often requires the ability to screen hundreds or thousands of individual strains or mutants to identify the highest yielding candidates. With no hard and fast rules on which vector, insertion locus, individual strain, or even host organism may be best suited to the production of a specific biological product, it is easy to see why so many biotherapeutics never make it to the clinic.
There are no shortcuts to overcoming this issue, but experience can mean the difference between success and failure. There are obviously numerous ways of tackling any given challenge that may arise during biomanufacturing scale-up, so working with an experienced partner is often the fastest way to get to an efficient and effective solution, without the cost and delays of following numerous dead ends. Outlined below are a number of issues commonly encountered during biotherapeutic development and scale up, with examples of various approaches to address them using different host organisms and synthetic biology approaches.
Protein solubility issues
One of the most common challenges encountered when attempting to scale up bioproduction is the solubility of the target protein or peptide. The simplest, and often fastest, way to overcome this is to express your product in an alternative host organism. However, this requires an understanding of which hosts might be appropriate, the tools to design, build, and test a suitable expression vector, and the capacity to perform strain selection and growth optimization. While this may sound a daunting task for most labs, specialist synthetic biotechnology companies routinely perform this type of project, and have the knowledge and experience to rapidly transfer production of a target biomolecule into a new host. For example, switching production from E. coli to Pichia pastoris (P. pastoris) may eliminate solubility issues encountered in early scale-up studies, allowing the high yielding production and secretion of a high purity product for GMP manufacturing.
Expression screening limitations
Protein engineering for improved therapeutic function is often a numbers game; the more gene variants that can be screened, the higher the chance of success. E. coli is often a good choice of host for this application-at least during the early development phase-due to the plethora of bioengineering tools available for this host. However, effective variant screening requires rapid, cost-effective assays that enable the qualitative or semi-quantitative assessment of target protein expression across large numbers of individual transformants. Unlike the functional or kinetic assays used in downstream characterization studies, these variant library screening assays must offer a quick and easy “yes/no” assessment of every transformant, and generally employ simple colorimetric or fluorimetric detection.
The challenge here is that the common approach of using a generic reporter gene-linked expression assay-combining the target of interest with the expression of a reporter protein of some sort-is often not appropriate when optimizing biotherapeutic production, and so a predictive assay tailored to the individual target protein must be developed. This is often the rate-limiting step in the identification and selection of improved gene variants, which can significantly delay development, or even halt progress entirely if not deployed successfully. Working with a partner that has extensive experience in designing and performing this type of screening assay can dramatically accelerate timelines, generating lead candidates much sooner. Figure 2 shows an example of a screening assay designed to improve transcription/translation of a novel amine oxidase in E. coli. The assay, which was developed and validated in just a few weeks, uses a colorimetric substrate (added to the culture media) to assess the expression of an enzyme involved in the in-vitro biocatalysis of an advanced pharmaceutical intermediate. Combined with design of experiment (DoE) approaches and an automated workflow, this type of assay allows both the selection of high-yielding transformants and the determination of optimal expression conditions in a very short timeframe.
The development of efficient screening assays can also help to improve the product itself, by enabling larger studies to be performed than would otherwise be possible. High assay throughput provides the opportunity for further product optimization, potentially in combination with machine learning or artificial intelligence strategies, to identify potential enhancements or improvements in the design of the target biomolecule.
Figure 3. Synthetic biology and high throughput screening can be combined with machine learning to aid in the development of new biotherapeutics. FACS is fluorescence activated cell sorting. IP is intellectual property. DSP is downstream processing.
For example, advanced screening has been successfully combined with machine learning to further enhance the potency of epidermicins, a new class of antimicrobial peptides capable of rapidly killing potentially harmful bacteria, including drug-resistant species (Amprologix) (Figure 3) (1).
Post-translational expression optimization
Looking beyond cell culture maintenance and the use of alternative transcription/translation elements of protein expression, there are a number of other factors to consider when attempting to scale up bioproduction. Even for a well-characterized host such as Bacillus subtilis (B. subtilis)-which is widely used in the production of enzymes-there are various opportunities to improve the overall biomanufacturing process in culture media composition, protein secretion, and endogenous protease gene knockout. Being able to extract the target protein from the bioreactor is obviously essential for efficient production, avoiding the need to destroy the culture to harvest the product. A number of secretion systems already exist for B. subtilis, so selecting a strain that excretes the heterologous target protein will help to simplify downstream purification.
It is also important to consider the stability of the target protein within the expression system. For example, Bacillus sp. have a number of protease enzymes that could cause the degradation of the target protein before it has even left the cell. Choosing knockout mutants that lack the major proteases can therefore increase overall yields by improving target protein stability. Another strategy to increase protein stability is to create a more favorable environment for your product. For example, co-expression of chaperonins has been successfully used in the production of recombinant factor VIII in CHO cells by allowing high levels of expression without leading to misfolding or aggregation.
Failure to consider these kinds of post-translational factors could significantly reduce the overall yield, and therefore performance, of a biomanufacturing system. Even beyond this, there are practical considerations when choosing the best expression system or host for a target biotherapeutic. For example, many Bacillus species are spore forming, which is an undesirable trait in a closely regulated GMP environment. The development of asporogenic strains has mitigated this issue, simplifying culture management and, consequently, reducing biomanufacturing costs.
Using the whole toolbox
Advances in synthetic biology have made it possible to coordinate the expression of multiple genes to allow the creation of novel biochemical pathways. This allows the host organism to produce large quantities of a heterologous product that would normally be outside of its metabolic “repertoire”. It is only by understanding each of these elements, while still considering the system as a whole, that biomanufacturing can reach its true potential, and working with a knowledgeable and experienced synbio partner is the best way to ensure successful scale up of novel biotherapeutics, on time and on budget.
Reference
1. Ingenza, “Amprologix Secures Funding to Develop New Antibiotic with Ingenza,” Press Release, Feb. 5, 2019.
About the author
David McElroy*, PhD, david.mcelroy@ingenza.com, is chief business officer at Ingenza Ltd, UK.
Article Details
BioPharm International
Vol. 33, No. 5
May 2020
Pages: 22–27
Citation
When referring to this article, please cite it as D. McElroy, “Biopharmaceutical Manufacturing and the Power of Synthetic Biology,” BioPharm International 33 (5) 2020.
