August 1, 2008 (Vol. 28, No. 14)

Pete Gagnon CSO BIA Separations

Continuing Improvements Extend Potential for Future Applications

According to Georges Guiochon, Ph.D., professor and distinguished scientist in the department of chemistry at the University of Tennessee, “monolithic columns will some day become the main workhorse of chromatographic separation.”

His remark was made in reference to analytical applications, but the same trend is evolving in preparative circles and it was strongly in evidence at the “Third International Symposium on Monoliths,” which was recently held in Portoroz, Slovenia.

Numerous therapeutic proteins, plasmids, and viral particles purified with monolith-based procedures are already in or entering clinical trials. The industrial benefits these procedures exemplify seem certain to attract a growing number of applications.

The conference chairman, Alois Jungbauer, Ph.D., professor at the University of Natural Resources and Applied Life Sciences in Vienna, defined a monolith as a continuous stationary-phase cast as a homogeneous column in a single piece. Monoliths are further characterized by a highly interconnected network of channels, most with sizes ranging from 1 to 5 µm. The adsorptive surface is directly accessible to solutes as they pass through the column.

Traditional chromatography media are prepared as porous particles and packed in columns. Most of the adsorptive surface area resides within pores with sizes ranging from 5 to 100 nm. These contrasts lead to fundamental differences in operating characteristics and separation performance.

Mass transport of solute molecules into and out of the pores in particle-based chromatography media occurs by diffusion. Diffusion is slow, increasingly so for larger solutes. Dynamic binding capacity and resolution diminish with increasing flow rate and solute size. Mass transport to and from the channel surfaces in monoliths is convective, driven by the physical flow of the mobile phase through the column. Dynamic capacity and resolution remain relatively constant regardless of flow rate or solute size.

Practical Value

These features have immediate practical value for industrial applications. One of the handicaps in downstream processing of antibodies has been the long residence time required to achieve high binding capacity with protein A affinity on porous particle media. The need for slow flow rates has often been attributed to kinetic limitations imposed by a low-affinity constant for IgG, but experimental data indicates that inefficient mass transport is the primary obstacle; binding reaches equilibrium 10 times faster on monolithic protein A. Dynamic binding capacity of DNA favors monoliths by 30- to 50-fold. Dynamic capacities in excess of 40 mg recombinant IgM/mL have been demonstrated on monolithic ion exchangers at flow rates of 15 column volumes per minute. This corresponds to a linear flow rate of 4,500 cm/hr on a 5 cm high bed of porous particles with the same volume.

All of these data speak to the potential for convective chromatography media, including membranes, to break the much-discussed bottleneck in downstream processing, the importance of which was highlighted in a presentation by Uwe Gottschalk, Ph.D.,vp of purification technologies from Sartorius Stedim Biotech. A presentation by Validated Biosystems indicated that although the dynamic capacity of current protein A monoliths is only about 10–12 g/L, continuous processing on a single 8 L radial flow monolith could purify 20 kg of IgG from 20,0000 L of cell culture supernatant in 27 hours, three times faster than a 19 L column of conventional protein A media with a capacity of 35 g/L.

Maarten Pennings, group manager at Tarpon Biosystems, presented another model projecting the ability of a 10 liter simulated moving bed (BioSMB™ ) array of 800 mL protein A monoliths to process 75 kg of IgG in 72 hours, the same time as a conventional column with a volume of 90 liters. The efficiency of BioSMB would enable the monolith system to achieve this result with buffer consumption equivalent to the conventional system despite its lower capacity. A BioSMB array of 8 L monoliths could process 75 kg in less than eight hours.

Both presentations concluded that the large number of process cycles in these models make it economical to dispose of the used monoliths after processing a single lot of IgG, thereby avoiding the need to develop and validate sanitization procedures.

Yow-Pin Lim, M.D., Ph.D., president and CSO of ProThera Biologics, discussed a monolith-based purification of Inter-alpha trypsin inhibitor, a 250 Kd plasma protein reported to be a highly effective treatment for sepsis and anthrax. He remarked on the ability of monoliths to accelerate process development, enabling more experiments in one day than can be completed with conventional media in a week.

“The speed of monoliths encouraged us to evaluate a much broader range of conditions than is typical during early-phase process development. The result was an ultrahigh efficiency anion exchange process that achieves greater than 90% purity and 75% recovery from plasma in a single step,” he said.

Another important difference between monoliths and particles concerns the void volume in particle columns. About 40% of a well-packed column is void space. Buffer follows the path of least resistance, flowing preferentially between the particles, and disfavoring solute contact with the particle surface, thereby depressing capacity.

An additional consequence of void space is the formation of eddies; vortexes created by differential friction between particle surfaces and the deeper void space. Eddies cause interstitial mixing that erodes resolution as solute zones flow down the column, and they create shear forces that may damage labile molecules. Eddy dispersion remains constant at higher flow rates, but shear increases in direct proportion.

Monoliths have no void volume; flow is laminar and they do not develop eddies.

The benefits of low-shear separations were highlighted in several presentations. Fatima Plieva, Ph.D., R&D scientist at Protista Biotechnology demonstrated passage of whole blood through a monolith with 10–200 µm channels without hemolysis. Similar monoliths have been used to capture von Willebrand Factor directly from whole blood, to capture inclusion bodies from high-debris lysates, and to capture live yeast and bacteria.

Kornelia Schriebl, Ph.D., research scientist at the Bioprocessing Technology Institute in Singapore, presented on stem cell purification and also remarked on the importance of low-shear chromatography supports for maintaining viability. In his keynote presentation, Charles Lutsch, Ph.D., director of downstream process development at Sanofi Pasteur, emphasized the importance of mild handling and rapid processing times for the manufacture of live and attenuated viruses and noted the ability of convective chromatography media to support both.

Conservation of Viability

Elizabeth Maurer, Ph.D., manager of process development at AVIR Green Hills Biotech, demonstrated conservation of viability for a replication-deficient influenza vaccine through the course of a monolith-based purification process. Recovery from the monolith was 19% higher than a membrane adsorber and 27% higher than particle-based media. Dynamic binding capacity was 10- to 100-fold higher than particle-based media.

Jochen Urthaler, Ph.D., head of in-process control at Boehringer Ingelheim, presented a Phase III scale-up example of a DNA plasmid.

Mark Etzel, Ph.D., professor at the University of Wisconsin, reviewed experimental data directed toward elucidating models of virus retention on convective supports. He also shared results showing the ability of experimental salt-tolerant anion exchange monoliths to improve viral clearance.

One such monolith was able to achieve phage clearance of 6.1 LRV in a buffer lacking sodium chloride, and still achieved 5.5 and 3.5 LRV at 50 and 150 mM sodium chloride. A QA monolith demonstrated 5.6 LRV in the absence of sodium chloride but no clearance at 50 or 150 mM. These results have important ramifications for removal of nonenveloped retrovirus from therapeutic antibodies and other recombinant proteins, and also offer new tools for purification of viral vectors and vaccines.

Intuitive Model

The current generation of preparative monoliths has been configured to accommodate large biomolecules such as plasmids, viral particles, and large proteins. This is apparent in the aforementioned data describing low capacity for IgG but fourfold higher capacity for IgM.

Dr. Etzel offered an intuitive model for higher capacities of larger solutes in systems unencumbered by diffusion. He contrasted the mass of marbles that can be packed on a tabletop versus the mass of bowling balls packed on an identical surface.

Although current monolith capacities adequately support industrial purification of large biomolecules, it was acknowledged that their growth into downstream processing of smaller solutes such as IgG will require higher capacity.

Vida Frankovic, research and development scientist at BIA Separations

tions.com), presented preliminary data demonstrating a threefold increase of BSA capacity from 24 to 80 mg/mL. This remains modest compared to recently improved particle-based exchangers, but in the context of monoliths’ extraordinary throughput capabilities, it nevertheless signals their expansion into an increasing diversity of industrial applications.

A pre-conference workshop highlighted the ability of monoliths to enable application of selectivity modifiers that compromise mass transport in particle-based columns. Dynamic capacity of an IgG monoclonal was doubled on a cation exchange monolith in the presence of polyethylene glycol (PEG) but reduced by half on a porous particle support.

Selectivity Modifiers

The increase on the monolith was attributed to the thermodynamic effects of preferential exclusion of PEG by the protein and solid-phase surfaces. The same mechanism operates with porous particles, but the elevated viscosity of PEG reduces diffusivity. This depresses diffusive mass transport and cancels out the hoped-for improvement. The effects of PEG were also shown to increase with protein size.

An example showed the ability of PEG to enable baseline resolution between native and aggregated IgM on a cation-exchange monolith. Shuichi Yamamoto, Ph.D., of Yamaguchi University in Japan, commented that size-dependent enhancement with PEG is likewise observed with single- and double-stranded DNA.

Discussion suggested that this technique may enhance discrimination among viral fragments, partially assembled capsids, intact particles, and aggregates.

Frantisek Svec, Ph.D., facility director of the Molecular Foundry at the Lawrence Berkeley National Laboratory, and Tatiana Tennikova, Ph.D., of the Russian Academy of Sciences in St. Petersburg, both pioneers in the field of monolith synthesis and applications, noted that monoliths bring together the most dynamic research elements in the field of chromatography: polymer synthesis, channel architecture, surface texture, chemical modification, mass transport, and applications.

Accordingly, monoliths are attracting many gifted new scientists to the field. It is only natural that they should also attract the most challenging separation issues in industry and extend industrial capabilities into areas that have been previously unimaginable.

It will be most interesting to observe their continuing evolution toward fulfillment of Prof. Guiochon’s prediction.

Pete Gagnon is a purification process development consultant and CSO at Validated Biosystems. Web: validated.com. Email [email protected].

He is also GEN editorial advisor for downstream processing. A copy of the symposium book of abstracts can be obtained at www.monolith-events.com/archive/2006/abstract_book.asp.

Previous articleBristol-Myers Squibb Bids $4.5B for ImClone
Next articleCary Institute of Ecosystem Studies Is Recipient of $750K EPA Award