A Decade of Chromatography: A Powerful Technology Reasserts Itself

Chromatographic separations are vital both to the analysis of biological macromolecules and to their manufacturing. When properly applied, chromatography provides exquisite specificity in separating different molecules from solution based on their size, electrical charge, or other physicochemical properties. Large liquid chromatographic (LC) columns remove host-cell nucleic acids, endotoxins, viruses, and process intermediates from harvest material. Combine high-pressure liquid chromatography (HPLC) with mass spectrometric (MS) or ultraviolet–visible (UV–vis) spectroscopic detection, and you can qualify and quantify macromolecules in such complex biological mixtures. Apply Fourier-transform infrared (FTIR) detection to gas chromatographic analysis, and you convert data into wave form, making it easy to compare spectra for confirming the identity of raw materials or determining leachable/extractable components in single-use equipment.

This column appeared on the cover of BPI in February 2004. ()

Speaking of disposables, this is one technology that has been problematic for conversion from reusable to single-use formats. Glass or stainless steel LC columns are regenerated a number of times to offset their cost (1,2,3). And it’s not just the columns themselves, but also the chromatographic resins packed inside them that can be very expensive depending on their construction and chemistry. Disposable, prepacked columns are available, however, from a growing number of vendors. And automation is making inroads with the ÄKTAexplorer systems from GE Healthcare. Protein A and Beyond

When it comes to expensive media, the most notorious (and celebrated) are protein A affinity resins. Because of the strong affinity that certain antibodies have for binding protein A (due to their role in immune response), it is so useful that despite its cost, high-yield protein A affinity capture is the first in line for a purification platform (4, 5). That means it handles the largest, least-refined process stream (after clarification removes debris) — and delivers 99% purity in a single step (6).

Even so, the expense combined with engineering and practical difficulties (6, 7) are making some process developers in this era of high-titer production wonder whether some alternative might be a better choice than protein A. Among those are mercaptoethyl-pyridine ligand–based affinity resins (7). Thus far, however, protein A remains king (8, 9).

We call them resins, but chromatographic media can be made from a number of materials. These began with cellulose and over the years have expanded to include carbohydrate polyethers, ceramics, silicates, and various polymers (10, 11). Different vendors sell different types. Membranes and Monoliths

Traditional column chromatography doesn’t lend itself easily to single-use mode. Two variations on a theme have emerged for single use: membrane adsorbers (12,13,14) and monoliths (11, 15, 16). The former are not really chromatography per se; they use binding chemistry, but not in a flow-through, bind-and-elute mode. More accurately, they function as a specialized form of tangential-flow or depth filtration.

Monoliths, however, “are chromatography media cast as a single integrated unit” rather than as porous particles packed in a column (12). They offer many attractive features: an ability to maintain high resolution and capacity regardless of flow rate and molecular size; and low-shear fractionation for shear-sensitive products such as DNA plasmids, live viruses, and labile proteins (15). BIA Separations is known for this technology. Process Development

Like every other aspect of biotherapeutics production and processing, development of chromatographic capture and polishing steps is changing due to the FDA’s quality by design (QbD) initiative (17, 18). New statistical and other modeling approaches are speeding optimization, while advanced analytical methods help process engineers find and fix problems better than ever before (19,20,21,22,23,24,25,26,27,28,29).

For example, column packing procedures can affect results in both analytical and process chromatography (19,20,21,22). Different media/supports require different methods (19). Larger companies and those performing high-throughput analytics can make use of automation (20). Too-high sample loads or flow rates can push solutes through too fast, blurring the unstable interface between fluids of different viscosities (21). And packing is just one aspect of chromatographic operations that’s subject to scale-down optimization (22,23,24,25,26,27,28,29).

People Are Talking About Protein A and Monoclonal Antibodies

Late in March, 2012, Nick Hutchinson of Parker Hannifin in the United Kingdom initiated an interesting discussion on the BioProcessing Professionals Linked-In group by asking, “Is Protein A Chromatography here to stay? Are people still actively looking to develop Protein A–free processes for MAbs?” Many answers followed, covering the gamut from yes to no. Here are a few comments (all opinions of the individuals rather than their organizations, of course).

H. Fai Poon of SAFC in St. Louis, MO called protein A “the most expensive part of manufacturing. Elimination will be good.” But Shalvi Arora, a student at Sardar Patel University in India, replied, “Although it’s expensive, it deletes later steps [and ultimately] costs the same. Hence it should not be deleted.”

Tim Breece of XOMA elaborated: “Protein A costs about $0.50/g of MAb processed when a column is used to its full lifetime. Trying something like simulated moving bed (SMB) [technology] will reduce the cost further. This expense is irrelevant compared with the selling price of MAbs. When you add in the extra development costs and reduced yield for alternatives, Protein A is significantly cheaper.”

Mukesh Mayani of Therapure Biopharma Inc. in Canada says his company is investigating recombinant Protein A. “It’s expensive, costing ∼20% of the total consumable cost for MAb production. But I think it will remain the choice at least for some time.” He believes “the reality favors Tim’s view at the moment. Many issues to be considered include regulatory aspects and availability of in-depth information of resin for adaptation. Novel media need to demonstrate technical challenges such as selectivity of target MAb among aggregates and other variants under optimized conditions, sanitizability, holding binding capacity upon recycling, ligand-leaching tendency, compressibility, MAb monomer recovery and aggregation (if any), level of nonspecific binding in capture step,” and so on. “A standard/reliable assay to determine ligand leachate is usually required. I have come across many alternative developments that sound very interesting, though!”

Sylvio Bengio of Pall Life Sciences in France: “Protein A is like rock n’ roll: here to stay (at least a few years)!” He went on to call it “definitely the most selective method in one step. However, if you can afford spending some time in optimization, mixed-mode chromatography (e.g., MEP HyperCel) can give good purity/yield results too, but the capacity might be less (∼30 mg/mL).” After protein A capture steps, he said, “these mixed-mode sorbents give excellent contaminant and DNA removal results (HCPs, aggregates)…. But as we all know, the industry is very conservative, and standard chromatography platforms using Protein A (typically followed by two chromatography steps) remain the standard. SMB has been around for many years, but its broad adoption seem to force a number of process ‘habit’ changes that few people seem to be ready to make in standard situations for antibodies. We should no
t forget that IgGs are not the most difficult proteins to purify!”

Michiel Ultee of Laureate Pharma was candid. “People have been whining about the high cost of protein A for decades. Let’s face it, nothing does as much in one step to purify your IgG antibody in a standard, platform-suitable process. In that one step, you begin with a mixture of host-cell proteins, DNA, lipids, and media components and recover what is essentially pure IgG by SDS-PAGE and HPLC-SEC. You could with much effort develop alternatives with enough development, but why bother? Sure, occasional issues arise such as aggregation with acidic elution, but there are methods for dealing with this.”

Andreas Kage of AptaRes AG, says his company offers MAb purification with a protein-A–like small-molecule ligand that doesn’t require acidic elution. “The molecule has a high temperature stability (>100 °C) and renaturating capability after alkaline treatment. This approach offers a new versatile purification technology for MAbs that are sensitive to acidic treatment. We also offer the custom tailored development of other affinity resins to purify other proteins.”

Barry Rosenblatt of SME Biotech Consulting said, “The paradigm for replacement of any resin should require a ‘quantum’ increase in productivity (yield, throughput, purity). This is difficult when looking at replacing protein A. Yes, it is expensive, but the value equation is still very attractive, especially with the incremental improvements achieved over the past few years. The trick is to decrease the price sufficiently to allow it to fit into the ‘disposable’ column category. Perhaps introducing new recombinant versions and challenges from new resin manufacturers will put pressure on the market to drive the cost down.”

Kage wondered whether “such a ‘quantum’ could be also the possibility of a nondenaturing elution from the resin,” preventing denatured single molecules that aren’t apparent using SDS-PAGE or in-process size-exchange chromatography (SEC). “Assuming that the costs are in the range of protein A resin, or a ‘quantum’ defined by not only molecular but also functional purity using a resin with a cost effective anti-idiotypic resin,” he went on, “this might also address the problem of possible clonal alterations.”

And Rosenblatt replied, “The ‘quantum’ would need to be defined by the company. If a small amount of denatured single molecules poses a threat to the product, then it would be a major determinant. That would need to be balanced by the potential loss of viral clearance, normally afforded by denaturing buffers used in protein A elution. Leaching of anti-idiotypic ligands can be more problematic in a final product than a small amount of denatured material. Pros and cons must be fully assessed before leaping into a new platform.” Anything But …?

A few years ago, we began to hear a strange rallying cry that would have been unthinkable when BioProcess International began: “ABC” for “anything but chromatography” (30). As downstream process groups faced the challenges of high-titer expression and highly concentrated product streams, they began to look “outside the box” for answers. But not only is chromatography familiar to regulators and well known for its power, it also has many years of engineering and optimization behind it.Other techniques face an up-hill road trying to replace it.

Meanwhile, chromatographers haven’t stopped improving their technologies. Expanded-bed adsorption is already used in a number of market-approved processes (31). And perhaps the most exciting recent advancement has been in combined chemistries for multidimensional separations (32,33,34,35,36,37).

Sometimes thought of as an “alternative to protein A,” multimodal resins put two different chromatographic chemistries to work in separating biotherapeutics from their contaminants: typically hydrophobic interactions and ion exchange (32). They provide “unique selectivities that are not achievable by single-mode sorbents used sequentially, so they enable some proteins to be purified when single-mode sorbent combinations fail” (33). Some can even reduce endotoxins to clinically acceptable levels (33).

Multimodal (mixed-mode) chromatography is no new idea. Hydroxyapatite (HA) was the first, combining cation exchange and metal affinity in the 1950s. “HA’s selectivity was recognized as unique from its introduction, but a lack of practical knowledge concerning its binding mechanisms long delayed the development of scouting pathways that fully revealed its abilities. That discouraged process developers who might have benefited from its capabilities. As those pathways were defined, it became possible to control each binding mechanism, and HA has emerged as the most broadly capable process option for removing fragments and high levels of aggregates from antibody preparations” (36).

The development of HA mirrors that of chromatography overall: It is a powerful method that’s been around the block, improving and expanding its applications as time goes by. Don’t expect its importance in bioprocessing to fade any time soon.

About the Author

Author Details
Cheryl Scott is senior technical editor of BioProcess International; [email protected].

REFERENCES

1.) Sofer, G. 2003. Establishing Resin Lifetime: Key Issues and Regulatory Positions. BioProcess Int. 1:64-69.

2.) Ng, PK, and V. McLaughlin. 2007. Regeneration Studies of Anion-Exchange Chromatography Resins. BioProcess Int. 5:52-56.

3.) Sofer, G, and J. Yourkin. 2007. Cleaning and Cleaning Validation in Process Chromatography: Current Industry Practices and Future Prospects. BioProcess Int. 5:72-82.

4.) Gottschalk, U. 2003. Biotech Manufacturing Is Coming of Age. BioProcess Int. 1:54-62.

5.) Grönberg, A. 2007. A Strategy for Developing a Monoclonal Antibody Purification Platform. BioProcess Int. 5:48-55.

6.) Shukla, AA. 2005. Strategies To Address Aggregation During Protein A Chromatography. BioProcess Int. 3:36-44.

7.) Sellick, I. 2005. Chromatography Advisor #3: Economic Benefits of Protein A Alternatives. BioProcess Int. 3:68-70.

8.) Langer, ES. 2008. Quantifying Trends Toward Alternatives to Protein A. BioProcess Int. 6:72.

9.) Thillaivinayagalingam, P. 2011. Revisiting Protein A Chromatography. BioProcess Int. 10:36-39.

10.) Santambien, P. 2003. Effective Protein Capture in Fluidized-Bed Mode with Zirconia-Based Beads. BioProcess Int. 1:46-59.

11.) Gagnon, P. 2006. A Ceramic Hydroxyapatite-Based Purification Platform. BioProcess Int. 4:50-60.

12.) Gottschalk, U, S Fischer-Frueholz, and O. Reif. 2004. Membrane Adsorbers: A Cutting Edge Process Technology at the Threshold. BioProcess Int. 2:56-65.

13.) Sellick, I. 2005. Chromatography Advisor #4: Capturing Very Large Biomolecules with Membrane Chromatography. BioProcess Int. 3:58-59.

14.) Lim, JAC. 2007. Economic Benefits of Single-Use Membrane Chromatography in Polishing: A Cost of Goods Model. BioProcess Int. 5:48-56.

15.) Gagnon, P. 2010. Monoliths Open the Door to Key Growth Sectors. BioProcess Int. 8:20-23.

16.) Gagnon, P. 2008. The Emerging Generation of Chromatography Tools for Virus Purification. BioProcess Int. 6:S24-S30.

17.) Gavin, D, and P. Gagnon. 2006. Building Process Control into Chromatographic Purification of Viruses, Part 1: Qualification of Critical Manufacturing Components. BioProcess Int. 4:22-30.

18.) Gavin, D, and P. Gagnon. 2006. Building Process Control into Chromatographic Purification of Viruses, Part 2: Purification As a Tool for Enhancing Process Control. BioProcess Int. 4:28-34.

19.) Sellick, I. 2004. Chromatography Advisor #2: The Advancing Science of Column Packing. BioProcess Int. 2:60-62.

20.) Bloomingburg, G, and P. Gandhi. 2005. Engineering Design Considerations for Column Packing in Large-Scale Biotechnology Facilities. BioProcess Int. 3:44-51.

21.) Shaliker, RA. 2007. How Viscous Fingering Can Spoil Your Separation: And You May Not Even Suspect It. BioProcess Int. 5:32-37.

22.) Crawford, M, J Stevens, and L. Roenneburg. 2007. Optimizing Sample Load Capacity and Separation Through a Series of Short Prep Columns. BioProcess Int. 5:40-46.

23.) Sellick, I. 2006. Chromatography Advisor #5: Process Proteomics Explained. BioProcess Int. 4:66-67.

24.) Guo, W. 2008. Statistical Approach to IgG Binding on a Strong Cation Exchanger. BioProcess Int. 6:82-86.

25.) Engstrand, C. 2010. Rapid and Scalable Microplate Development of a Two-Step Purification Process. BioProcess Int. 8:58-66.

26.) Hitchcock, AG. 2010. Scale-Up of a Plasmid DNA Purification Process. BioProcess Int. 8:46-54.

27.) Westerbeg, K. 2011. Model-Assisted Process Development for Preparative Chromatography Applications. BioProcess Int. 9:48-56.

28.) Ljunglöf, A, K. Eriksson, and T. Frigård. 2011. Rapid Process Development for Purification of a MAb. BioProcess Int. 9:62-68.

29.) Forss, A. 2011. Optimization, Robustness, and Scale-Up of MAb Purification. BioProcess Int. 9:64-69.

30.) Rosin, L. 2008. Anything But Chromatography? BioProcess Int. 6:74.

31.) May, T, and K. Pohlmeyer. 2011. Improving Process Economy with Expanded-Bed Adsorption Technology. BioProcess Int. 9:32-36.

32.) Sellick, I. 2006. Chromatography Advisor #6: Mixed-Mode Sorbents. BioProcess Int. 4:66-68.

33.) Lees, A. 2009. Purifying a Recalcitrant Therapeutic Recombinant Protein with a Mixed-Mode Chromatography Sorbent. BioProcess Int. 7:42-48.

34.) Jin, Z. 2005. A Method for Automated Multistep (Multidimensional) Purification Processes for Protein Recovery. BioProcess Int. 3:68-70.

35.) Eriksson, K. 2009. MAb Contaminant Removal with a Multimodal Anion Exchanger. BioProcess Int. 7:52-56.

36.) Gagnon, P. 2010. Minibodies and Multimodal Chromatography Methods. BioProcess Int. 8:26-35.

37.) Snyder, MA. 2011. Working with a Powerful and Robust Mixed-Mode Resin for Protein Purification. BioProcess Int. 9:50-53.