Dear dpsk, I asked a very similar question here about a month ago, so I'll post a summary of the replies.
In our case, the protein was known to exist in a biologically relevant monomer-dimer equilibrium, with the amount of monomer becoming vanishingly small at high concentrations (e.g. in our crystallisation assays). We obtained several crystal forms, one of which contained the monomer and not the dimer. Several people commented that the phenomenon isn't unusual - after all what crystallises is not necessarily the predominant form of the molecule but rather those species that have properties favouring a crystalline phase in the given conditions. Upon nucleation, mass action will replenish the pool of whatever crystallises at the expense of the predominant species, so crystals can continue to grow even though at any one time the concentration of the crystallising species in solution is very low. In spite of this, well-documented cases and/or systematic studies in the literature seem to be fairly rare. Below is a list of replies with relevant cases. HTH, best wishes, Sebastiaan Werten. Luca Jovine: Sure - crystallization selects what packs, which may constitute a very minor fraction of what you set up drops with. 1DUH is an example... F.Xavier Gomis-Rüth: we indeed found something very similar with selecase (see http://www.ncbi.nlm.nih.gov/pubmed/25159620). This is a highly specific metallopeptidase, which is also a metamorphic protein that transits between several stable states, among which only the monomer is the active species. We managed to trap 5 different crystal structures (monomers, three distinct dimers and tetramers)...and all this in a small 110-residue protein! Rick Lewis: your crystallisation conditions could quite easily affect the MD equilibrium. we had something similar a few years ago, when i was working with a protein that was monomeric in solution 'til it was phosphorylated when it then dimerised. in the crystals, however, we had monomeric phospho-protein and (domain-swapped) dimeric non-phosphoprylated. took some auc runs in conditions that mirrored the xtl conditions to sort it out. have a look here: http://www.ncbi.nlm.nih.gov/pubmed/11851334 i don't know of a more systematic study of multiple examples; i guess people worry about their own problems first! leo brady had something similar but different a few years ago with the structure of a mis-folded cd2 variant, that was only a minor fraction of the prep http://www.ncbi.nlm.nih.gov/pubmed/7638192 Roger Rowlett: some of the allosteric carbonic anhydrases stubbornly crystallize only in the T-state, despite crystallization conditions that are known to preferentially stabilize the R-state, and for which the predominant R-state population can be confirmed by other methods. Emilia C. Arturo (specifically commenting on proteins that not only crystallise in different states but also change their conformation/function depending on oligomerisation): Our lab has termed a type of protein sequence that does this sort of thing a "morpheein", and there is currently only one bonafide morpheein in the family, though we are in the progress of validating this term for at least one other enzyme. A description of morpheeins can be found here: http://en.wikipedia.org/wiki/Morpheein At Wednesday, 27-05-2015 on 04:56 dpsk wrote: Hi, Recentely,we determined a monomer protein structure which looks like an dimer in solution via size-exclusion chromatography and SAXS, have you guys met such a situation before? or some references, Thank you for your reading Best wishes,
