Dear all,
we all use prepacked chromatography columns for affinity binding in protein purification (most prominent metal affinity like IMAC). I always wondered whether it would be better to load the column with fast flow rates (3-5 ml/min) or slowly (around 0,5 ml/min).
Some say fast loading is advantageous because the protein comes into contact with a larger portion of the column material in shorter time, leading to more homogeneous binding, comparable to a batch process.
Others say slow binding prevents clogging and aggregating at the top of the column, where protein gets concentrated and tends to precipitate during loading...
I am trying to solve that case once and for all :)
Any contributions are more than welcome.
Best,
Peter
It always depends on the size of the column. The general rule is to keep the "residence time" at 10 min or longer. The "residence time" is the time of the exposure of each protein molecule to the resin beginning from entering the column and leaving the column. It usually provides enough time for any affinity interaction between the protein and the ligand on the column.
Hello. In my experience I prefer to load slowly because when I load with fast flow proteins can precipitate maybe for the change of concentration of buffers, metals, etc. You can use batch for your experiments (I prefer to use this mode)
but in this case load sample slowly in order to avoid any aggregation or precipitation, before starting the shake of your tube.
There may not be a universal residence time (column volumes/unit time) that applies to all contexts. In manufacturing processes, for example, the throughput (mass protein/mass resin/unit time) is extremely important because of cycle times, work shifts, and the delay time for up or down stream unit operations. Ni is not used that commonly in commercial manufacturing ( due to toxic Ni disposal guidelines) but Protein A certainly is. The commercial conditions used for Protein A or Ni may be a starting point to answer your question even though there are different: pre-packed columns, bead size distributions, ligand densities, column dimensions, % cross-linking of beads, bead porosity distributions, bead morphologies, protein concentrations, load viscosities, temperature, conductivities, exposures and lengths of His tags, naurally occurring His sites, etc. In an academic context there would likely be much more diversity in how investigators load their feedstocks onto the affinity columns because the columns are not being run under GLP or cGMP conditions which would include run, regeneration, and storage cylces that impact the resin capacity and packing for the next run. If your are referring to a theoretical context where there's one loading on an unused column and the Ni resin has no significant difference with any other manufacturer's resin, let's say Qiagen, then the load solution properties may need to firstly be described as well as the column format. Would you be able to load in identical ways the same His6 tagged protein expressed in two different systems where one is a crude feedstock from a bacterial homogenate having 30% glycerol, at 5C, contains ~100 mM arginine, 3 M urea, 10 mM b-ME, and 75 mM Tris whereas the other feedstock may be just the protein in PBS and conditioned mammalian cell growth media? The small scale columns used for Process Development will be run with a linear flow rate to optimize scale up requirements while using conditions that also avoid liabilities for product quality; processes are built around the protein's needs rather than the other way around.
Hi Peter,
As a rule of thumb, when carrying out affinity steps or binding steps, it is better to use a slower flow rate compared to the step of washes or even elution step. This makes sense since you are trying to maximize interactions between eg., a phosphopeptide and the resin or antigen and antibody etc.
That is what I would recommend.
Best,
Hediye.
There are situations in which a protein may have certain degradation liabilities and faster processing is desirable, even batch-wise processing. But if you want a column format and need to process quickly then you could follow the guidelines for linear flow rates or residence times based on DBC (dynamic binding capacity) at an acceptable percentage breakthrough. This empirical approach will give you the fastest load time specific to your sample's physical characteristics. You may even get some positive displacement chromatography effects.
- Badges
- Science method