Membrane Chromatography

Some researchers contend membrane chromatography might be the next technology to supersede fixed-bed methods and conventional resin- or particulate-based methods. The reason is primarily based on the more open pore structure of a membrane compared to a particulate structure where there is higher mass transfer with lower pressure resistance. Unfortunately, as we discuss later the benefits have not been wholesale at the commercial level.

Membrane chromatography is also claimed to have higher productivity with a lower cost of goods attached. It can also be scaled up from a ml volume to a 1 litre device. In recent years, membrane chromatography systems are being considered as an alternative polishing step especially in the biopharmaceutical industry but not so in the food industry (Demmer & Nussbaumer, 1999; Knudsen et al., 2001: Phillips et al., 2005).

Membrane chromatography has not reached the same heady heights in terms of applications as the resin-based column, batch and continuous systems that are available. The bead-based systems are generally more desirable irrespective of the issues that bedevil them.

Membrane filtration has been explored as an alternative method to overcome the inherent flow distribution issues, the limitations from poor flow-rates and of cleaning especially when the resin is ‘in situ’ (Gosh, 2002). The problem with the membrane chromatography system is its lower adsorption capacity and the poor performances of the devices used. That implies low loading capacity, membrane fouling and suboptimal fluid distribution which gets worse with scaling-up (Gebauer et al., 1997). Interestingly, these are almost the same physical issues that are claimed for fixed bed chromatography systems.

Over the years though, better membranes suited to specific applications, improved surface chemistries and greater thought into the design of the modules and the devices has meant improved performance. It’s worth bearing in the mind that the same developments and improvements have also been occurring in packed bed (fixed-bed systems) so the balanced between fixed- and membrane systems has not changed that much over the intervening years.

For process-scale operations, many membrane layers can be used. It’s not uncommon to see 15-, 20- layer membrane modules for contaminant removal and for virus clearance.

The issue with any affinity membrane system though compared to its resin counterpart is the low binding capacity which trumps any benefits from a high flux.

Membrane systems are ideal in the polishing type of chromatography as opposed to the capture approach where the former is about the removal of trace impurities. Trace impurities include endotoxins, viruses, rogue nucleic acids etc.

In this situation the adsorption capacity is not an issue (Phillips et al., 2005). They should be more widely used for removal of endotoxins and DNA. It’s feasible to remove greater than 4 log
removal of mammalian viruses, 3 log removal of endotoxin and DNA, and greater than 1 log removal of host cell protein. These systems can come in single-use disposable systems eliminating the need for expensive regeneration.

A novel Protein-A affinity membrane system supplied by Sartorius has been tested for binding immunoglobulin G (Boi et al., 2008). An affinity membrane chromatography system allows for a greater exposure of the active area of the ligand and this greater accessibility to the antibody because the ligand is not required to be bound to the resin or the bead. Any target proteins pass through the membrane via convective transport which means they come into close contact with the ligands. There are no limitations from diffusion with this system when compared to a bead-based column say.

Membrane filtration generally has better or higher flow throughputs which leads to reduced processing times. A recent creation was a high-capacity affinity membrane of a regenerated cellulose microporous matrix to which was bound recombinant Protein-A.

Membrane Producers

Membranes used for conventional filtration can be modified to become adsorption membranes. Gore® is a brand of chromatography devices which exploit membrane systems. Sartorius have Sartobind® membranes made from stabilized regenerated cellulose.  Pall have the Mustang Q and S membranes which are flat sheet polyethersulphone membranes modified with quaternary amine and sulfonic acid functionalities. The unit comes in a 5L XT 5000 module containing 16 layers of pleated 0.8 micron PES membranes.

References

Boi, C., Dimartino, S., Sarti, G.C. (2008) Performance of a new Protein A affinity membrane for the primary recovery of antibodies. Biotechnol Prog. 24 pp. 640–647 (Article).

Demmer, W.; Nussbaumer, D. (1999) Large-scale membrane adsorbers. J. Chromatogr. A  852, pp. 73–81

Gebauer, K.; Thommes, J.; Kula, M. (1997) Plasma protein fractionation with advanced membrane adsorbents. Biotechnol. Bioeng. 54, pp. 181–189

Ghosh, R. (2002) Protein separation using membrane chromatography: opportunities and challenges. J. Chromatogr. A 952, pp. 13–27 (Article)

Knudsen, H. L.; Fahrner, R. L.; Xu, Y.; Norling, L. A.; Blank, G. S. (2001) Membrane ion-exchange chromatography for process scale antibody purification. J. Chromatogr. A 907, pp. 145–154.

Phillips, M.; Cormier, J.; Ferrence, J.; Dowd, C.; Kiss, R.; Lutz, H.; Carter, J. (2005) Performance of a membrane adsorber for trace impurity removal in biotechnology manufacturing. J. Chromatogr. A .1078, pp. 74–82.