Biomagnetic Separation, Thinking Bigger Part I

July 23, 2015
Contrary to popular belief, biomagnetic separation is ready for large-scale processing

Separation is a key process used in life science. Cells, proteins and genetic material of interest for medical and pharmaceutical use are rarely obtained in a purified form. The molecules we are interested in are usually in a matrix that contains numerous other biological substances. We therefore need to extract/isolate these molecules from the supernatant before work can start. Even high-value pharma processes such as protein purification, the downstream process (i.e. purification) can lead to losses of 50-80 percent of the product. 

When working on a small scale, using magnetic beads or particles as solid support for separation has become increasingly popular, particularly when it involves immunocapture. Coated with the right antibody, magnetic carriers can capture the molecule of interest in seconds and retain this by magnetic force while washing out the supernatant. When the magnetic field is removed, the beads can then be re-suspended in a clean buffer.

With such obvious benefits, it makes sense to use this technology on a larger scale. It is commonly believed that biomagnetic separation is not suitable for large volumes, which may explain why people are so doggedly attached to centrifugation, filtration and packaged columns—even though the methodology is slow, complex, requires extremely expensive equipment and can be a cleaning nightmare.

One life science industry has already tackled these questions. Efficiency, rapidity and simplicity of biomagnetic separation are among the reasons behind the success of chemiluminescent immunoassays (CLIA). The rapid growth of this market has prompted In-Vitro Diagnostics (IVD) manufacturers to scale up magnetic-bead processes to cope with demand. The lessons learned by this industry can be very helpful in other life science companies and cell capture or protein purification may be of benefit to processes in the IVD industry for coping with volumes up to tens of liters.

Coated with the right antibody, magnetic carriers can capture the molecule of interest in seconds and retain this by magnetic force while washing out supernatants.

At the beginning of the current century, the IVD industry had similar challenges to those now facing other life science companies. Attempts made to scale up biomagnetic separation processes beyond a few milliliters proved unsuccessful. The paradox was that the main cause of failure lay in its success with small volumes. Easy implementation of biomagnetic separation at milliliter scale had allowed researchers to focus on selecting the right magnetic beads and how to coat them with the antibody or biomolecule. Given that almost any magnetic separator (or even a simple magnet) would capture the beads in few seconds, little emphasis is placed on the workings of the process. Descriptions of the biomagnetic separation conditions were limited to a description of the magnet and, at most, the separation time. When the volume was scaled up, the initial attempts used a ‘larger magnet, usually with no more specification than the material (NdFeB magnet) or the required magnetic field on the surface. However, with this approach, the scaled-up biomagnetic separation process did not work as expected. Separation time increased exponentially, magnetic beads losses were significant and the beads became irreversibly aggregated. The process is no longer fast, reliable or consistent technology when the working volume is increased.

Around 2000, when my team and I were approaching the problem, IVD-manufacturers were considering several options. One was to replicate the small volume process. To increase production by a factor of 10 it was necessary to build 10 production lines. This option involved large-scale investment and proportionally increased labor costs with virtually no economy of scale. Worse still, a new problem arose: The need to guarantee batch-to-batch consistency between the different lines. As one of our customers said, “It is far simpler to validate one ten-liter batch than to guarantee that 10, one-liter batches are the same.”

The problem was so significant that some IVD-manufacturers considered using tangential filtration to separate the coated beads from the incubation buffer. This investment may have made economic sense despite the complex process (flow, pressure and temperature, etc., all of which need to be controlled ...) and running costs were relatively high.

Our discussions with IVD production managers focused on finding the ideal solutions from both technical and economical standpoints. The result of these discussions was clear. These operations-centered managers wanted a large-volume permanent magnet-based system with 100 percent magnetic bead recovery, a separation time of few minutes and perfect in-batch and batch-to-batch consistency. Permanent magnet-based systems would mean no maintenance and no running costs.

New approaches are challenging the notion that magnetic separation is suitable only for small-scale production.

High recovery meant high magnetic force, even on the beads which are farther away from the retention area and a high-enough magnetic retention force to prevent aspiration of magnetic beads when the supernatant was pumped out. In-batch consistency would mean that all the magnetic beads in the vessel would be subjected to the same force, avoiding crushed beads and forming clumps due to high forces while other are gently attracted. Finally, batch-to-batch consistency would require a well-validated process and, if possible, a means of monitoring the process for traceability of every single batch.

This list of requirements seems to pose a big challenge, as many of them are, and at first sight, often contradictory. Then, there’s the temptation to look for trade-offs. What is the acceptable loss rate? Could the separation time be increased to one hour? Should we find more aggressive re-suspension techniques to disaggregate the clumps?

However, closer analysis of the requirements reveal that no trade-offs were necessary. All we needed to do was the homework that was neglected when working on a small scale to understand how magnetic beads move under the influence of magnetic fields. This would enable us to correctly parameterize the biomagnetic separation process, validate the right conditions and then define the characteristics of the separation systems, regardless the working volume.

Editors note: Because downstream biopharmaceutical processing efficiency is so critical, the topic merits further examination. To learn how and why biomagnetic separation is so effective, please look for Part II online and in print in the very near future.

About the Author

Lluis M. Martinez | PhD. CSO