Cytiva has a handbook about size exclusion chromatography, including Sephadex and other resins. You can find it here:

https://cdn.cytivalifesciences.com/api/public/content/digi-11639-pdf

For Sephadex G-25, the separation range is 1,000 - 5,000 for globular proteins. The separation also depends on the shape of the molecules.

Sephadex is used for low-pressure chromatography only. For FPLCs, other resins are needed. Sephadex can also be used for centrifuge (spin) columns for desalting.

Dear Glen Hinckley

i'm not sure the origin of the numbers related to each sepadex resins however when you are selecting a SEC resin you have to take in consideration:

- Exclusion limit;

- Separation range of MW;

- Resolution;

- Conterpressure;

Sephadex for example are generally used for desalting, where you need a resing with low resolution, low conterpressure and the different resins are G-10, G25, G-50 differ mainly for the exclusion limit.

If you would like to run a SEC with better resolution on the basis of the range of MW you can select

Sepacryl-Superose-Supedex.

Superdex (High resolution --> high conterpressure --> High cost)

Supedex 75 good for separation of protein in the rante 70KDa-5KDa while Superdex200 for separation of protein in the range 600KDa-30KDa

Attached an extract from the handbook suggested by Adam B Shapiro those could be helpfull

good luck

Manuele

I understand that I can look at the manufacturer's specs but -and I hate to be the old curmudgeon here- those numbers used to mean something. G-25 meant a 25 kDa MWCO. That has apparently changed?

I recently tested several of our Lab columns (most are a bit older, but still work okay).

It's best to look up the manufacturers instructions about the exclusion limits etc.

However, it's noteworthy that separation strength is highly dependant on sample volume loaded.

Additionally, if your proteins are not globular, that also changes how the separation is going to be affected.

Furthermore, (I think only relatively few people know/consider this), if your proteins interact, they will run quite differently. Some proteins (like Lectins, Lipases, etc.) may also interact with the column material itself, making separation almost impossible. This can be circumvented by increasing buffer strength (e.g. by using 2x TBS [100 mM TRIS, 300 mM NaCl]).

All those caveats aside, I've found the following separation strength for some common SEC Columns:

TSK-2000: 5 - 150 kDa according to manufacturer

TSK-3000: 10 - 500 kDa (according to manufacturer)

TSK-4000: 20 - 7000 kDa (according to manufacturer, lower overall resolution)

all TSK Columns are unsuitable for preparative work (>200 µL Sample or very high concentrations), are not stable over pH 7.5 and sharply decrease in separation strength below 15 kDa (even TSK2000). Due to their low overall volume they are great for analysis of small samples, though.

Superdex Peptide (new Name Superdex 30 Increase): 0.1 - 7 kDa; you can actually separate proteins of 6 - 20 kDa as well. This will need more care (lower sample volume, maybe you need to do multiple iterations), but the column can handle a variety of buffers and you can even use highly concentrated samples (up to 100 mg/mL was the highest I've ever used), but you sensor will probably get saturated.

Superdex 75: 3 - 70 kDa; the same as Superdex Peptide. Good separation of proteins >70 kDa and those < 70 kDa (e.g. IgG and BSA); almost perfect between 60 and 6 kDa, still great below 6 kDa.

However, I've had problems with Lipases, as they seem to stick to the column material even with increased salt concentrations.

Superose 12: 1 - 300 kDa; Similar to Superdex. Very good separation > 65 kDa (BSA); below that it becomes severely lacking in performance. Although Salts still get separated very nicely.

As far as I remember, the higher numbers roughly equal the exclusion limit, which is not true for the lower numbers, e.g. 50 and below. A 2009 paper (https://www.bioprocessintl.com/chromatography/50-years-of-sephadex-media) suggests that the designation G-25 indicates that 1 g crosslinked dextran will absorb 1.5 g water, although there seems to be no reference to the source of this particular piece of information.