Hi Darren
I believe that the most frequently used detergent for protein crystallization (not including membrane proteins) is octyl-glucoside.
The most important parameter is the CMC of the detergent and the size of the micelles of free detergent if you have micelles around. These considerations do not apply if you are well under the CMC (only free molecules of detergent are present in solution). I believe tween-20 CMC is roughly 0.006% so you are at the border.
The problem is you have micelles of detergent is that depending on their size and the MW cutoff of the device you use for protein concentration you may end up concentrating both components protein and detergent and then this is when trouble starts. You may have an unknown amount of concentrated detergent above the CMC, this can be very problematic for crystallization because you may have a lot of phase separation in your drops; phase separation are usually (but not always) not desirable.
For membrane protein crystallization we are always facing this problem because we usually work in this regime of critical detergent concentrations. However for soluble proteins that need a little bit of detergent to remain stable this is not as usual.
OG (octyl glucoside) is a rather mild detergent with a CMC of about 20 mM (check ANATRACE catalog).
To simplify, maybe a little bit too much, the most important is at first to find the minimum amount of detergent you need to keep you sample stable and possibly stay as low under the CMC. And then may be go up in detergent concentration if you don't get the results expected. You can check that your complex stays functional using Biacore (in your specific case)
Other detergents very popular for non membrane proteins are DDM (dodecyl maltoside) or CHAPS (zwitterionic cholesterol derivative).
So in order of decreasing preference I would suggest OG, CHAPS (which is a totally different type of detergent) and DDM (this one has a very low CMC ~0.15mM I believe, so it is difficult to get rid of it, but it is fairly genlte).
An alternative to detergent are non-detergent sulfo-betaines they can sometime have the same protective effect without the trouble of detergents.
I hope this helps,
Cheers
Pascal F. Egea, PhD
University of California San Francisco
Department of Biochemistry and Biophysics
