Toufic, That is my point. Most non-surfactant chemists (i.e., us) seem to feel that detergents have a fairly homogeneous micelle size (DDM micelle ~65-70 kDa), which is not true. You are talking about "average" micellar size, not the distribution. Micelles readily and rapidly disaggregate into free micelles, into partial micelles, they deform, or they cluster into aggregates, all in a manner that is dependent on the solute conditions. For some detergents (but not all), their effective CMC can change with local concentrations (as can occur at high salt or protein concentrations or with high lipid content). The real point is defining EXCESS free detergent, not the total concentration or the amount bound to the protein. There is no assay that determines protein-free detergent concentration, only the total detergent concentration.
A 100 KDa protein at 10 mg/mL is at 0.1 mM. Given your examples, the effective concentration FOR ONLY the protein-bound detergent is 40-100 times that. For DDM, the protein-bound detergent concentration is already ~23 -110 times its CMC, and for DM, that can be ~2.3 - 11 times its CMC. Now, we have to consider that this protein-bound detergent is in rapid equilibrium with an unknown amount of free detergent in both micellar and monomer forms. At the moment, we cannot rapidly determine all the parameters needed to find an analytical method to the problem of controlling precisely the detergent concentration, so we are stuck with an empirical one (try different methods till one works, i.e. you reproducibly get crystals) and off-the-cuff calculations (like above). But the real take-home message is that crystals WERE NEVER GROWN in drops containing 0.1 or 0.05% detergent, but 0.1 or 0.05% detergent PLUS whatever is bound to the protein. For some cases, that will be 2-5X the nominal detergent concentration. Finding an optimal detergent environment for membrane protein crystal growth depends on finding the sweet spot in this complex detergent equilibrium. This means distinguishing between the nominal detergent concentration (what you add to your buffers) and what you actually need to have in your protein solution to allow crystal growth. I think that much of my gray hair came from dealing with these %*($# detergents. Cheers, Michael **************************************************************** R. Michael Garavito, Ph.D. Professor of Biochemistry & Molecular Biology 603 Wilson Rd., Rm. 513 Michigan State University East Lansing, MI 48824-1319 Office: (517) 355-9724 Lab: (517) 353-9125 FAX: (517) 353-9334 Email: [email protected] **************************************************************** On Jul 15, 2013, at 10:52 AM, El Arnaout, Toufic wrote: > Hi, > That is true, Michael.. can't remember the reference, but in some study they > found that even if you use a higher MWCO (for example 100 kDa) with DDM > (micelle 65-70 kDa), there is still 70-80 % retained detergent in the protein > sample. > Raji, another thing I would like to say is that the right MW cut-off may > depend on the protein, the detergent micelle, but also the bound detergent > molecules per protein. It could be 40-50 molecules of DM let's say for > protein A, but 80-90 molecules for protein B. You can quickly evaluate the > concentration step for your protein using SDS-PAGE or UV abs, and for the > detergent there are many methods like colorimetric assays, TLC.. > Best wishes > t > > > toufic el arnaout > School of Medicine - 660 S Euclid Ave > Washington University in St. Louis > St Louis, MO 63110, USA > > > From: CCP4 bulletin board [[email protected]] on behalf of R. M. Garavito > [[email protected]] > Sent: Sunday, July 14, 2013 11:17 AM > To: [email protected] > Subject: Re: [ccp4bb] Concentrating purified membrane protein > > Raji, > > One point the most people forget about is that whenever you concentrate any > detergent-solubilized membrane protein is that you will ALWAYS concentrate > the detergent. So regardless of the MWCO, if the protein-detergent complex > concentrates, the overall detergent concentration also increases. What you > want to shoot for is balancing protein loss with obtaining a sample having > minimal EXCESS free detergent. While a concentration step with a Ni-column > followed by dialysis will work, so will a wise choice of concentrator. One > trick to moderate high excess detergent is just to avoid the need for a many > fold concentration where you will really concentrate the free detergent. > Nonetheless, we have used concentrators effectively, and a Ni-column followed > by dialysis as well, but if you still have too much free detergent, you can > always use a spin desalting column with G-25 sephedex to bring remove > detergent down to a nominal concentration. In all cases, you will lose some > protein. > > Good luck, > > Michael > > **************************************************************** > R. Michael Garavito, Ph.D. > Professor of Biochemistry & Molecular Biology > 603 Wilson Rd., Rm. 513 > Michigan State University > East Lansing, MI 48824-1319 > Office: (517) 355-9724 Lab: (517) 353-9125 > FAX: (517) 353-9334 Email: [email protected] > **************************************************************** > > > > > On Jul 14, 2013, at 10:29 AM, Raji Edayathumangalam wrote: > >> Thanks everyone for your responses. I definitely plan to save the >> flowthrough so we'll see what happens. My protein has a His tag and I did >> consider doing an affinity step for concentration except I do not want to >> have imidazole for some functional assays that I need to carry out with the >> protein. Just occurred to me that I could simply dialyze out the imidazole >> after the affinity step. >> >> Thanks again! >> Raji >> >> >> On Sat, Jul 13, 2013 at 8:47 PM, Raji Edayathumangalam <[email protected]> >> wrote: >> Hi Folks, >> >> Sorry for the non-ccp4 post. >> >> I have purified an 18kDa membrane protein and want to concentrate the >> protein from gel filtration fractions, which are in buffer containing 0.05% >> DDM (well above the CMC for DDM). My colleague was able to concentrate a >> 30kDa membrane protein using a 100kDa MWCO concentrator but I am not sure if >> I can do the same without losing protein in the flowthrough. On the other >> hand, if use too low a MWCO for the concentrator, then I'm concerned that I >> may end up concentrating the DDM and end up with too much detergent in the >> final sample. >> >> Any tips about how to concentrate my low MW protein without concentrating >> the DDM? >> >> Many thanks. >> Raji >> >> -- >> Raji Edayathumangalam >> Instructor in Neurology, Harvard Medical School >> Research Associate, Brigham and Women's Hospital >> Visiting Research Scholar, Brandeis University >> >> >> >> >> -- >> Raji Edayathumangalam >> Instructor in Neurology, Harvard Medical School >> Research Associate, Brigham and Women's Hospital >> Visiting Research Scholar, Brandeis University >> > >
