Hi,

Simple case first: I am assuming your proteins are >90% pure and you have lots of them recombinantly expressed. 36 and 50kDa are difficult to resolve (S75 or S200) but that is not your purpose. Follow the elution of the 36kDa protein (on S75) with and without the complex. If it binds the 15kDa protein it will be shifted from its profile and elute slightly early.

It is much easier to do this if you have all three individually. If this is the case use the 15kDa protein and follow its elution profile with the other partners.

Precautions:

1. Size exclusion (gel filtration) is about shape too.

2. The binding affinities of the partners should be kept in mind.

3. Your buffer and instrument conditions should be identical for comparison. Calibrate the column with very well known markers thoroughly.

4. Use more than one wavelength if possible. Keep flow rates same between experiments. Repeat your experiment.

You should think of other ways of solving this. Do not be technique limited.

Good luck

You may not be able to get baseline resolution between the elution profiles of 50 kDa and 36 kDa proteins by gel filtration, but you may be able to separate the peaks a little. But I agree with Sujay that you may find another technique to be more suitable to the purpose. Fluorescence resonance energy transfer and chemical crosslinking occur to me.

Hi Devi,

Unless you are trying to do this preparatively I think gel filtration is probably the wrong technique to use.

If I understand your question correctly this appears to be a question about protein-protein interactions.

It may be interesting for you to try immobilising  your 36 kDa protein on an SPR chip and passing over your complex (and ideally also the free proteins).

You may also try using AUC of the mixture of your three proteins - you should be able to get pretty accurate molecular weights and relative abundances of your complexes.

SEC-MALS would effectively be like trying a super accurate gel filtration as you can separate on the basis of gel-filtration like you're proposing, and the light scattering will give you a much more accurate molecular weight of the components of each peak.

Hope this helps & best of luck.

With MW this close I would recommend to use very low flow rates to increase resolution on superdex. We have found that resolution keeps improving with flow rates going as low as 2 cm/h. I agree with the advise above to run the proteins individually if available. And to measure at different wavelength to be able to distinguish between the proteins in the mixture by looking at ratios. Suggested wavelength to use are 254, 280 and 305 nm. 

Hello Sujay,

A long time ago, when I was still a PhD student (in the 80s), I carried out a gel exclusion chromatography on a column 2-meter high and 5-cm wide to separate two proteisn fo close MW. It worked quite fien and the baseline resolution was excellent. But prepare a 2-m high chromatography column (no bublle in the gel etc.) was quiet challengning too and I ahd to repeat teh loading three times before getting a well-packed column.

There are several other ways to do so today, but you need specialized expensive devices.

Best regards,

Prof Philippe Urban

Connecting several shorter columns in series works almost as well as one long. Just try to minimize the tubing between columns. In smaller scale using Tricorn columns you can connect them directly to each other with a male-male connector. For XK columns let the second column run upflow to avoid a long tubing between the columns. The band broadening contributed by the extra tubing is probably compensated by better packing of the two shorter columns compared to one 2-meter anyway.

I will suggest if you want to check protein protein interaction using chromatography, to run anion exchange chromatography. If your complex co-elutes with the 36kDa protein, it will be another evidence (of course could mean they have the same charge, if you know the sequences you coudl calculate the isoelectric point of the complex and of teh 36 kDa protein). Thsi chromatography based on charge separation might help to discriminate, Coelution is one of the criterium for protein protein interaction.

Problem using high length column is that you might dilute your sample and as dilution favors dissocitaion you might be in trouble.

Would you consider using electrophoresis instead of chromatography? It might better suit your needs.