Biomagnetic Separation: Thinking Bigger, Part II

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

By Lluis M. Martinez, PhD. CSO, SEPMAG

1 of 2 < 1 | 2 View on one page

Biomagnetic separation has many advantages as a biopharmaceutical downstream purification technology. When using magnetic beads coated with the right biomolecules, the protein of interest can be captured within seconds and retained by a magnetic force while the supernatant is washed out. Once the magnetic field is removed, the beads can be re-suspended in a clean buffer. The protein can then be eluted and the magnetic force applied again. The beads will be retained and the extracted supernatant will only contain the purified protein in the clean buffer. By removing the magnetic force, the separated magnetic carriers can be re-suspended in the appropriate buffer and reused several times.

However, when it comes to the practical application of biomagnetic protein purification, there are limitations due to its lack of efficiency at large volumes. Most of the successful situations in protein purification - when using magnetic beads - are limited to small volumes, mainly using immunocapture or His-tag. Even for larger volume downstream processes, the early steps are developed at small scale, focusing on the immunocapture process. At this stage, a lot of effort is dedicated to develop the protocol (buffers composition, incubation time, elution conditions), but the separation process itself receives little attention. For the small tubes used at this scale, magnetic separation seems to work fine (high recovery rate, no aggregation problems, short separation time) with a classical separator or even a simple permanent magnet. Thus, the separation process is usually defined by the time needed to have the solid phase moved to the retaining position and the buffer transparent.

The problem with this approach is that, when the process is scaled up, the separation time is no longer valid. The capture and elution are much less efficient at high volume and, even if longer separation times or larger separators are used, the situation does not improve. Even worse, many new problems appear; for example irreversible aggregation, or the degrading of the batch-to-batch consistency.

The usual reaction is to blame the magnetic beads, which initiates a series of long (and unpleasant) discussions with the providers. Nevertheless, new magnetic beads, buffer changes or applying longer separation time are useless. The process never reaches the efficiency necessary for a real-world application. These frustrating experiences have spread the idea that magnetic separation is not a suitable technique at large volumes.

The root of the problem is not the biomagnetic separation technology itself, but the bad validation and specification of the process. In addition to the buffer composition, incubation time and conditions, magnetic beads characteristics and concentration, it is necessary to correctly define the magnetic separation conditions. When parameterizing a biomagnetic separation process, we need to do something more than just define the separation time.

The first step to correctly validate the process is to focus on the relevant parameters. The basic point is to know the magnetic force that drives the beads in the suspension to the retention area. A simple permanent magnet (and almost all the classical magnetic separators) generates a magnetic field that changes with the distance. Hence, it generates a force proportional to the spatial variation of the field (technically speaking, the force depends on the gradient of the product of the magnetic moment of the bead and the applied magnetic field). This magnetic force will push the beads in one direction, which is against the drag force generated by the buffer viscosity. The magnetic bead speed is defined by the balance between both. When the same magnetic force is applied, the beads speed are lower at the higher viscosity media, i.e., the same beads would be 3-4 times slower when suspended in whole blood than when suspended in water buffers.

For standard separators, the magnetic force is highly dependent on the distance to the magnets, so its value is not the same over all the working volume. This fact may not always be noticeable at a small volume, but when the working volume scales up, farther beads will experience very low magnetic force (and move slowly) and nearest beads will experience extremely high magnetic force. As the magnetic force decreases almost exponentially with the distance, larger vessels imply that the beads need to travel longer distances moving at a much lower initial speed. To keep the recovery rates high enough to make the downstream process economically feasible, the separation time for capturing the farthest beads will also need to be exponentially longer. On top of that, the beads located initially near the retention area can quickly reach the retention area (high force implies high speed), and they will remain under high force during exponentially longer times. The consequence would be that this fraction of beads would aggregate irreversibly.

This is the most usual cause of the in-batch inconsistency issue when scaling up the magnetic separation downstream process. Notice then, that for biomagnetic protein purification, it is necessary to use separation for the capture of the beads (with the molecule of interest attached) just after the incubation. Following that, wash them several times to remove debris and impurities. Finally, after changing the buffer conditions for eluting the protein, it is necessary to add another biomagnetic separation step to capture the beads and extract the buffer containing the purified protein. Recovery rates not close to 100%, large separation times, and lack of consistency, make this technology inefficient and uneconomical for biopharmaceutical downstream processes.

1 of 2 < 1 | 2 View on one page
Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments