Protein interaction - (Jul/19/2005 )
Hi guys,
I have a really basic question in fact. I am working on ABC transporter, my main point of view of this subject is molecular biology and bacteriology but I would like to know a bit about proteins interaction. My ABC Transporter is composed by different copies of the different part of the system. I would like to know which part interact with which part.
Do you know how can do that simply? By bioinformatic or any experiment?
Thanks a lot
Fred
If you've got the setup for it, yeast-2-hybrid is a decent first-approximation. It requires a lot of cloning and the generation of a library, but it's a pretty standard method.
Alternatively, if you've got a good idea of which components might interact with which others, you could attempt purification with a fusion protein and do a pulldown. Attach, say, Glutathione-S-transferase (GST) onto the N terminus of one of your components, bind that to a glut-agarose column, then run the remaining components (or whole cell lysate) over the column. Follow that with extensive washing followed by elution and see what coelutes.
These are both fairly simple, but also time consuming, mostly due to multiple stages of cloning. I'm also not familiar with the specific system, so if there are a lot of membrane components, you'd want to try to just clone out the soluble portions if at all possible.
Alternatively, if you've got a good idea of which components might interact with which others, you could attempt purification with a fusion protein and do a pulldown. Attach, say, Glutathione-S-transferase (GST) onto the N terminus of one of your components, bind that to a glut-agarose column, then run the remaining components (or whole cell lysate) over the column. Follow that with extensive washing followed by elution and see what coelutes.
These are both fairly simple, but also time consuming, mostly due to multiple stages of cloning. I'm also not familiar with the specific system, so if there are a lot of membrane components, you'd want to try to just clone out the soluble portions if at all possible.
Thank you for your help, that are good idea.
I think also to yeast two hybrid but it consumes so much time and there is a lot control for this kind of experiment.
The second method look like really interesting because we have already proteins for some part of ABC transporter of interest. But I have an other question if I take just the soluble portions of the membrane component, is it possible to have the right folding of that protein. Can I be sure by that?
Thank you aludlam
Fred
As with any protein engineering, it's not guaranteed. You'd have to check expression levels independently, but in theory the membrane insertion part doesn't influence the folding of the non-membrane portion of the protein. It's more a question of how soluble the non-membrane portion is on its own. Of course, if you express it as a fusion construct, the fusion protein could help solubilize it as well.
The down side to this experiment is that it's more work per individual segment. If you did a yeast-2-hybrid, once you get your cloning done you can look at a lot of things simultaneously, though it's prone to false positives. The fusion construct requires individual cloning and protein purification, but it also enables you to optimize binding conditions, and you're not as likely to generate false positives.