Tell me more about x-ray crystallography, Veeky Forums. I hear integral membrane proteins are quite a challenge.
Tell me more about x-ray crystallography, Veeky Forums. I hear integral membrane proteins are quite a challenge
Mason Cooper
Other urls found in this thread:
associomics.dpb.carnegiescience.edu
sciencedirect.com.sci-hub.cc
sci-hub.cc
nobelprize.org
youtube.com
jcrystal.com
jcrystal.com
biochem.utah.edu
twitter.com
Luis Martinez
Nice resolution.
Yes, they are very difficult to express, purify, and crystallize.
Blake Rogers
For some reason it's much harder to get E. coli to express membrane proteins than any other type, at least in any meaningful amount, without killing the cell.
Purification and crystallization are also tough because, since it's a membrane protein, correct conformation often depends on the presence of a lipid bilayer for the transmembrane domains. If you don't have that layer, you might not fold the protein correctly, which means any structure you crystallize might be just plain wrong.
Usually people get around this by expressing just the free domains of a protein, but this again has its issues. You don't know that the protein folded correctly without the transmembrane domain. Also, if your protein has a multi-pass structure and the tertiary structure depends on the loops between passes being present, again, you might not fold correctly.
We've run into this problem a lot in our lab because the membrane protein we're interested has a good deal of sequence similarity to bacterial histidine kinases, so we think it might actually be functionally active in E. coli in a way that kills the cell.
Chase Edwards
> we think it might actually be functionally active in E. coli in a way that kills the cell
Thats like one of the common lame excuses for laziness
>Hurr my protein is toxic, i cant express it, so i cant raise antibody
Not in your case of course
Isaiah Perez
We're not a biochem lab anyway so we don't actually care that much about the crystal structure. We just use epitope tags if we need to run westerns or fluorescent protein fusions for localization experiments.
Elijah Kelly
I thought most of the problem with expressing non-native membrane proteins in, say, E. coli, was due to codon bias. I have a question for this thread: say I do want to express a eukaryotic TMP, they require a motley of posttranslational modifications like glycosylation and palmitoylation for proper transfer and insertion. What model organisms are usually used for isolating a decent quantity of these proteins? I know there's been promising work with Xenopus oocytes that's really conducive to good batches of protein, but I don't see it happening in our lab.
Joshua Bailey
Insect cells? CHO?
Jason Hernandez
Codon optimization won't change the amino acid sequence, so you can optimize for whatever system you're expressing in without worry.
Your situation sounds like an absolute nightmare desu. I wouldn't even want to try that experiment unless I knew there were closely related PTM'ing enzymes present in the system that I had a good expectation would be active.
Maybe try expressing in yeast?
There was a paper a while back that tried to prepare a plant membrane interactome database (associomics.dpb.carnegiescience.edu
Christian Nguyen
It's difficult. The bottleneck really comes with getting a respectful, pure batch. There are a lot of other setbacks too. The literature on this protein is simply convoluted. Apparently, tagging this TMP with GFP gives it greater propensity to aggregate, there isn't much work on this protein with other receptors either, nor cross-comparison of other fluorescent receptor-binding proteins with wild-type, and wild-type alone has a rapid turnover rate and a very sad half-life.
There is this paper (sciencedirect.com.sci-hub.cc
So I'm just here trying to piece shit together. I probably won't need to crystallize it if it's that tortuous, 18kDa is good for NMR, no? We'll see where this takes me.
Alexander Miller
>tagging this TMP with GFP gives it greater propensity to aggregate
GFP can do that to pretty much any protein it's fused to, not just TMPs. But yeah that's a problem. Good luck m8. Sounds like NMR might be your best bet, yeah.