Bacteria and viruses are nature’s shapeshifters, able to alter the make-up of their surface proteins in order to evade recognition by host immune systems. This strategy increases the virulence of these infectious agents and enables them to re-infect hosts, despite any previous encounters with the same pathogen. For example, the Neisseria genus of bacteria can vary the composition of their pili – protein polymers on their surface that play critical roles in bacterial adhesion [1]; whereas Streptococci bacteria often vary their expression of M-protein [2], an anti-phagocytic virulence factor that also blocks an immune process called opsonization. Such bacterial surface proteins are prime targets for vaccine development, and being able to capture and study them is a vital element of this work.
Like bacteria, viruses also exhibit great amounts of surface plasticity, as exemplified by Human Immunodeficiency Virus (HIV) – its structural mutability remains one of the major obstacles to its effective eradication [3]. This variability makes viral coat proteins important targets for the development of anti-viral vaccines; however, viruses have also grown to become indispensable to the developing field of gene therapy, proving effective tools for the transfer of genetic material into mammalian cells. Adenoviral vectors are currently the most widely-used vector for this purpose, being employed in a quarter of all gene therapy trials [4]. This popularity is not only attributable to the vector’s safety profile (the genome rarely integrates into host chromosomes), but its large capacity for foreign genetic material (up to 37 kilobases) and high infection efficiency. In addition, current production systems can generate high titers of adenovirus; however, like bacterial surface proteins, adenoviral vectors are tricky to purify on any significant scale despite their availability and abundance.
Here we will discuss some of the challenges and provide tips for purifying either bacterial proteins or viruses using a mixed-mode chromatography approach.
Purification hurdles
A major challenge to performing large-scale purification of adenoviral vector constructs is the large size of these targets; weighing in at a hefty 165 MDa and measuring 0.1 μm in diameter, these viral vectors are comprised of over 2,700 different subunits represented by a huge array of different charge variants [4]. Generally unstable at low pH too (unless significant amounts of NaCl are present [5], these characteristics present some tough obstacles for vector purification.
With individual bacterial surface antigens, size is not so much the problem but variance remains an issue. Bacteria can be sorted into sub-species groups called serotypes, depending on the composition of certain surface proteins. The challenge with many of the existing purification protocols targeting bacterial surface antigens is that they are serotypespecific. Multivalent vaccine manufacturing may start with fermentation of a mixture of bacterial serotypes, making any protocol that favors a certain population of these disadvantageous. Plus, many of these protocols have many steps and are time-consuming, impacting yield and compromising protein activity – a worry also applicable to adenoviral vectors.
Purification of a single, stable target protein is challenging enough, but when the target comprises a heterogeneous mix of uncharacterized bacterial surface proteins or a large, multi-protein protozoan, which purification strategy do you opt for?
Mixed-mode solutions
The best way to approach a new purification, especially if you’re planning to expand in scale further down the line, is to identify which modes of interaction work best for your target(s). This means testing a range of products available on the market. If feasible, investigating many different resins for your initial purification step should allow you to identify the optimum chromatography medium. For this you need to use a system that allows you to gauge how much of your target molecule is retained by the chromatography column versus how much ends up in your flow-through; furthermore, you need to assess how much of the captured protein or virus is released upon addition of your elution buffer. If your column is great at binding your target, but yields very little of it at the elution stage, you need to continue your search.
You should always include at least one mixed-mode chromatography resin in your initial screen. These resins often perform well with adenoviral vectors and mixed serotype bacterial surface antigens as they are based on media supports that have been functionalized with ligands capable of multiple modes of interaction, including ion exchange, metal affinity, and hydrophobic interactions. It may be that a mixed-mode resin capable of both hydrophobic interaction and cation exchange, like BioRad’s Nuvia™ cPrime™ resin, is the best fit for your viral particle purification workflow. This type of resin supports the binding of adenoviral vectors such as Ad5-E1 in low salt and low pH conditions. You also need to look for a resin that gives you a good clearance of feed stream contaminants such as any residual host cell proteins, especially if you wish to purify material for clinical/therapeutic use.
Optimal buffers
Investigating different sample loading, washing and elution conditions with your chosen resin before moving on is also important. Especially with sample-loading conditions, testing different dilution factors can enhance target binding to your initial chromatography resin. Although dilution increases sample load volume and hence loading time, you may achieve significant improvements in binding capacity for minimal additional dilutions. You can then examine binding properties under different concentrations of your feed buffer. The wash and elution buffers can also be examined using a similar step-wise approach, eventually transitioning to either a single step or a gradient approach. Examining each buffer in this step-wise manner may be time-consuming at first, but it will pay dividends in time saved and product yield further up the production scale.
Think ahead
Another tip is to think ahead to the next stage of the purification process as early as possible. Think about which polishing resins are most compatible with the resin in your initial capture step. Can you introduce your eluted sample straight onto the next chromatography column? Better yet, can it form a closed system with the first resin, i.e. does it retain your target in the presence of your elution buffer? No matter your target, reducing the amount of eluate manipulation required to perform downstream will enhance the scalability of your purification procedure.
Most of the testing described here is easiest to perform with a target that’s straightforward to track, so consider doing your early scale-up work with, for example, a viral vector which expresses a fluorescent protein instead of your ultimate payload. However, when fine tuning later stages, such as wash buffer composition for the final polishing step, it’s best to switch to traditional methods like SDS-PAGE. This is because changes in contaminant profiles may not be readily detected on chromatograms, but bands of any contaminating proteins will be visible on an SDS-PAGE gel. Even small changes in wash buffer concentrations can provide important additional impurity clearances.
Rounding up
In combination with the approaches described above, a single mixed-mode column can replace two or more single mode columns, streamlining your purification workflow and improving the yield of your target species. The mixed-mode option enables high-quality purification of complex targets, such as adenovirus, in just two steps, yielding active, concentrated products with purity, HCP levels and DNA contamination levels comparable to those of existing clinical grade products. Furthermore, mixed-mode resins enhance the scalability and efficiency of workflows.
To find out more on how you can apply mixed-mode resins to adenovirus vector or bacterial surface antigen purification workflows, view this webinar from Mark A. Snyder, Process R&D Applications Group Manager at BioRad.
References
[1] Swanson J, Belland RJ, Hill SA. Neisserial surface variation: how and why? Curr Opin Genet Dev.2,805 (1992). [PMID: 1360853]
[2] Lannergård J, Gustafsson MC, Waldemarsson J, Norrby-Teglund A, Stålhammar-Carlemalm M, Lindahl G. The Hypervariable region of Streptococcus pyogenes M protein escapes antibody attack by antigenic variation and weak immunogenicity. Cell Host Microbe. 10, 147 (2011). [PMID: 21843871]
[3] Rambaut A, Posada D, Crandall KA, Holmes EC. The causes and consequences of HIV evolution. Nature Rev. Gen. 5, 52 (2004). [PMID: 14708016]
[4] Nemerow GR, Steward PL, Reddy VS. Structure of human adenovirus. Curr Opin. Virol. 2, 115 (2012). [PMID: 22482707]
[5] Rexroad J, Evans RK, Middaugh CR. Effect of pH and ionic strength on the physical stability of adenovirus type 5. J. Pharm. Sci. 95, 237 (2006). [PMID: 16372304]
Related Products from: Bio-Rad