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

Chromatography Resins Must Balance Capacity and Selectivity against Productivity and Quality

For at least 25 years, large-scale biochromatography resins have been sold, to a significant degree, on the basis of binding capacity. But capacity is just one piece of the productivity puzzle.

Chris Pohl, vice president, chromatography chemistry, Thermo Fisher Scientific, refers to resin capacity as “a moving target.”

He says the term “capacity” is itself inherently vague because manufacturers tend to measure it in terms of breakthrough capacity, which does not translate to actual practice. “Breakthrough capacity only reflects what you would see if the only material on the column was the molecule of interest,” Pohl says.

Dynamic capacity, which relates to the maximum quantity of loaded protein that will allow recovery of pure product, is often significantly less. “Dynamic capacity depends on how you do the experiment, whereas breakthrough capacity is more reproducible,” Pohl explains. “That is why industry uses the breakthrough number.”

Both breakthrough and dynamic capacities have slowly risen over the years, not by simply increasing the number of binding sites, or by enhancing the specificity of binding sites to their targets—but by balancing these properties against other factors such as carryover, kinetics, recovery, and protein stability. Blindly increasing capacity without regard to these factors usually affects some other resin property adversely.

“The trend is to increase these other desirable parameters in preference to focusing solely on capacity,” Pohl tells GEN. For example, studies show that introducing a spacer between the resin’s backbone polymer and the affinity site improves recovery and reduces protein denaturization at the cost of absolute capacity because spacers take up physical room on resins.

“There’s always have a tradeoff between resin designs that provide the best chromatographic properties and highest absolute capacity,” Pohl continues. Doubling current dynamic binding capacity may be possible given current resin manufacturing technology, but only if designers ignore other performance factors. Tradeoffs occur even when maximizing such apparently straightforward characteristics as cleanability, reusability, and ease of regeneration. These properties, Pohl observes, are also at odds with each other.

“Materials that can handle harsh cleaning with concentrated hydroxide don’t stand up to those cleaning regimens indefinitely,” Pohl cautions, “nor do they always provide optimal protein recoveries.” He gives as an example stationary phases based on divinylbenzene, which endure many cleaning cycles but tend to have poor protein recovery. One variable that resin manufacturers have tried, with some success, is manipulating the resin particles’ pore size to maximize favorable binding interactions while minimizing protein becoming trapped within beads, which lowers recovery.

Twenty-five years ago, resin manufacturers and users believed that resins needed to be scalable from analytic chromatography through large-scale manufacturing. Such scalability, it was thought, would provide time savings through seamless transition from the analytical lab to production.

Pohl observes that since then, analytical and preparative media have sharply diverged. Analytic chromatographers care only about resolution and sensitivity, not capacity, whereas preparative chromatography is all about putting large quantities of material onto a column and getting as much of it off as possible. Resin particle sizes have parted ways as well. One would be hard-pressed to find HPLC columns that use particles larger than 10 microns, or production-worthy media particles much smaller than 25 microns.

A scientist at Thermo Fisher Scientific loads her samples into the autosampler of an ultra-high pressure liquid chromatography system for LC-MS analysis.

Overall Productivity

According to John Daicic, Ph.D., department director for bioprocess media at GE Healthcare, the goal of chromatography media design is overall productivity, which consists of different factors depending on the resin and the unit operation. “You can gain significant productivity with capture media based on capacity and flow rate,” he says, “but you must tightly control selectivity and resolution to provide customers with productivity and robustness.”

Where capacity is the dominant productivity factor for capture resins, aggregate removal and polishing rely more on selectivity. Improving productivity for protein A resins also arises from in-process resin stability, cleanability, flow characteristics, elution properties, and lifetime cycle capabilities. “Alkali stability is also at the top of many customers’ requirements,” Dr. Daicic notes.

Flow properties dictate how quickly a purification can be run and, thus, determine the target protein’s column residence time. The effect of longer residence time is typically higher selectivity and capacity. Consequently, requirements for processes with shorter residence times may present challenges. Similarly, elution pH may present quality issues with certain labile proteins.
“Capacity is still up there in terms of importance,” Dr. Daicic explains, “but there are more subtle parameters to work with as well.”

Affinity Resins: A Different Story

When EMD Millipore launched tentacle-based chromatography resins 25 years ago, these products were rated at an IgG binding capacity of 100 g/L. Competitors soon followed with high-capacity media of their own. Since then, EMD Millipore has improved on its ion exchange and mixed-mode resins. GE Healthcare, Tosoh, and others have introduced refinements, too.

“Companies claimed that they had achieved capacity improvements, but in fact the capacities were comparable to previous-generation resins,” says Matthias Jöhnck, Ph.D., head of chromatography R&D at EMD Millipore. “The real difference is we can now achieve that high binding capacity at significantly higher flow rates. This was achieved by utilizing larger particle sizes combined with improved surface modifications.” As with analytical chromatography, smaller particles pack more tightly and oppose buffer flow.

Innovation in ion exchange resins, Dr., Jöhnck continues, is about engineering base beads, particle sizes, and surface chemistry. “Binding capacity,” he notes, “is important but not a dominant factor.”

Anion exchange binding capacity improvements were more palpable, compared with cation exchangers, from one generation of resins to the other. Particles also became more rigid, resulting in a doubling of column bed heights from 20 cm to 40 cm.

Like Thermo’s Chris Pohl, Dr. Jöhnck does not put much stock in theoretical capacity values. Calculated binding capacity for one of EMD Millipore’s ion exchange resins is 280 g/L, and he has achieved 250 g/L in the lab. Regardless, Dr. Jöhnck contends that such capacity figures are of “limited value” to EMD Millipore’s customers: “If you bind such quantities of protein, you will have elution issues and reduced selectivity. That is why we’re not aiming at improving binding capacity. We’re looking for higher selectivity.”

Affinity resins are a different story. These resins make a dominant contribution to downstream processing costs. The battle among producers centers on lifecycle stability and binding capacity at appropriate flow rates. This is reflected by capacities for protein A resins rising over the last 20 years from 20 g/L to 55 g/L for monoclonal antibodies.

Biochromatography resin manufacturers will continue to focus on capacity and selectivity, but not at the expense of quality and overall productivity. “There may be physico-chemical limits to resin capacities,” says GE’s John Daicic, “but we haven’t reached them yet.”

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