June 1, 2015 (Vol. 35, No. 11)

How Silica Can Be Used to Improve Biochromatography and Bioprocessing

The influence of particle and pore size has been recognized since the publication of the first theories of chromatography. Ideally, and in very general terms, a process should maximize the contact between the chromatographic media and the biological molecule being purified while maintaining the mobility of the latter during the elution step.

Several studies have provided a detailed account of how pore size and pore size distribution affect the dynamic binding capacity of a recently developed strong anion exchange Q-Silica with a particle size of 70 µm.

It was found that the dynamic binding capacity to albumin steeply increases when the median pore size increases from 250 to 800 Å. Conversely, at median pore sizes higher than 1,000 Å, there is a moderate decrease in the dynamic binding capacity. This nonlinear relationship between the dynamic binding capacity and the median pore is likely the result of multiple factors.

The initial increase in dynamic binding capacity is likely the result of an increase in the accessibility of albumin to the pores. The subsequent decrease in dynamic binding capacity is driven by a reduction in the total BET surface area arising from the increase in median pore size (total surface area is inversely proportional to median pore diameter). In addition to this, the surface bonding stoichiometry was found to be dependent on the median pore size.

One of the obvious changes in performance arising from a variation in the surface bonding stoichiometry is a change in the dynamic binding capacity. Another more subtle change is the possibility of an increased ligand stoichiometry imposing a steric hindrance in the accessibility to a pore. Steric hindrance can play a significant role in chromatography especially with larger and larger proteins.

When compared to other types of commercial silica gel, Davisil® silica gels (W. R. Grace) have very broad pore size distribution. As shown in Figure 1, D1250 is a silica with a median pore size of 1,250 Å and has pores with sizes ranging from 300 to 4,000 Å. Contrary to some common belief, this broad pore size distribution has been proven to be beneficial with about 40% higher binding of albumin than other narrow pore size distribution commercial grades with similar ligand bonding processes. It can be rationalized that broad pore size distribution allows better accessibility of proteins into pores of various sizes. There are very few process scale particles that have pore sizes greater than 1,500 Å.

Although the effects of particle and median pore sizes of silica have to be tested during the development of each new silica-based chromatographic media, some of the conclusions obtained during the development of a particular bio-chromatography product may be ported across different types of chromatography.

For example, the recently released silica-based ProVance® Protein A media is also based in particle sizes of 70 µm with pore sizes slightly bigger (1,250 Å) than the anion exchange Q-Silica. The high and robust dynamic binding capacity of ProVance Protein A media to antibodies at different flow rates (Figure 2) strongly suggests that there is good accessibility to the media’s pores and particles.

Figure 1. Pore size distribution of Davisil XWP silica 1250 measured by mercury intrusion.

Column Packing of Silica-Based Media

The efficiency of column packing always affects the outcome of a biochromatography.  During column packing, a homogeneous consolidated bed is first formed, followed by a compression step at high flow rate. The higher the flow rate used at the compression step, the more stable the column will be. The flow rate used is only limited by the deformability of the media particles, which puts noncompressible silica at a significant advantage over flexible agarose beads.
As a result of a high flow rate compression step during packing, a typical silica-based media packed column will have higher theoretical plates per meter. The quality product specifications for a silica-based ProVance Protein A column is 4,600 plates per meter, compared to the 1,500 plates per meter offered by an equivalent column packed with agarose-based protein A media. The higher plates per meter in the ProVance Protein A columns provide narrower elution peaks than columns packed with agarose media.
Whether biochromatography is performed as a capture, purification, or polishing step, the proper selection of media and good column packing practices are a good starting point. Implicit in the performance and column integrity of the selected column, there are considerations like the robustness of the method being implemented.
One such measure of robustness is how the method operates at different flow rates. An additional merit of silica-based media is how well it can run at high flow rates. As seen in Figure 3, at high flow rates the pressure remains significantly lower in the silica-based ProVance Protein A than in the agarose-based media. This is because silica-based media is noncompressible. The high flow rates at which silica-based media can be used to generate a significant gain in productivity by accelerating loading particularly for very low titer feedstreams or by reducing washing and elution times.
Many screening studies focus heavily on characterizing the binding capacity of chromatography resins. Affinity-based chromatography resins also serve a very important role in concentrating the product. Using silica-based resin that delivers sharper elution profiles can further increase the concentration effect. A quick elution with low tailing in the chromatogram peak is advantageous not only for the decrease in processing time but also to ensure that there is not a process impairment due to the handling of extra volumes of a diluted purified material.

Figure 2. Dynamic binding capacity vs. column residence times of a silica-based ProVance Protein A column tested with an antibody at different concentrations.

Bioprocessing with Silica-Based Media

One of the major bioprocess trends in recent years is the use of disposable technologies. Although first limited to liquid handling, single-use technologies are already starting to spread into chromatography processes. The low cost of silica could be a tipping point in driving the disruption as a solid support matrix in single-use chromatography.  Using a ProVance protein A column can achieve a 60% reduction in media costs, compared to an identical operation using an agarose protein A media. Compounded in these savings are the elimination of labor and time to pack, clean, and validate columns.

Another emerging trend is continuous chromatography. In a basic continuous chromatography setup, columns are connected in series and the feedstock is loaded continuously. Once the first column is saturated with product, it is then disconnected from the feedstock flow, washed, eluted, and regenerated. While this happens, the feedstock is still being loaded into the columns downstream. The disconnection, washing, elution, and regeneration cycle is repeated with each saturated column while the other columns are still connected to the feedstock. Once a particular column has completed its cycle, it is then reconnected to the feedstock. Continuous chromatography increases productivity, saves costly buffers, and makes full use of all the chromatography media’s dynamic binding capacity.

An optimized continuous chromatography setup may require a high flow rate for the washing, elution, and regeneration steps. As noted above, silica-based media can operate at higher flow rates. The other desirable quality of a chromatography column to be used in continuous chromatography is a steady performance after multiple cycles of use, which also include cleaning. Reliable cleaning-in-place methods are now available for silica-based chromatographic media.

Figure 3. Pressure vs. flow characteristics of various Protein A media based in silica (ProVance) or agarose (Leading Agarose and ProA 4FF).

Julio Cabrera, Ph.D. ([email protected]), is biopurification technology manager at W.R. Grace.

Previous articleA Sweet Tooth Is HIV’s Achilles Heel
Next articleSimplifying Protein Analysis