I'm in complete awe of what an absurd amount of work this new preprint represents, but even more than that -- wow -- what a shining example of how we need to move to the surfacome to figure how how cells are responding stuff.....
So much work here...where to start....
Okay -- so they used Dr. Josip Blonder's favorite cell line in the world (and I'm pumped to see his work recognized in a study this great) -- there is some reason that MCF10a is important and I know he told me 100 times. I think that it might not have any of these cancer genes activated or something. And that makes sense in the context of this study.
This great multi-institution team messed with 6 different oncogenes in this cell line
AND
they messed the cells up with drugs that inhibit some of the downstream processes of these oncogenes. See the matrix expanding?
SILAC is employed here in some experiments, and maybe I'll clarify which ones in my head as I'm going through it.
Proteomes were directly harvested the way we'd normally do -- just from 20x10e6 cells, in troplicate....which, don't quote me, I'm pretty sure qualifies as more or less a fuckload. The digests were then enriched for glycopeptides using a lectin kit.
In case the lectin kit didn't pull down enough stuff they also did SAX chromatography to enrich for glycopeptides as well.
Okay -- so now they've got a more or less normal glycoproteomics setup for the cells, albeit on a pretty large scale.
There are two ways to go after glycopeptides -- enrich pre-digest or post-digest, as I just described. they also do the other one -- cause -- you know, why not? This is the method Josip uses, I'm pretty sure, where you oxidize the sugars and then use a biotin thingy to pull out all the proteins --- AHA!! this is the part where the SILAC is employed and this is what they consider the surfaceome. The glycoproteome is more of for filling in the background. The SILAC is for gold standard quan of the surfaceome changes.
The SILAC surfaceome is studied with a Q Exactive Plus. The glycoproteome and...well...proteome...you've got to do that while you're here, right? Goes on the super fast Lumos system using the restricted mass range method some people from this group demonstrated just a few weeks ago. The Lumos also has activated ETD, cause...well....of course it does....
Results?
This is such a cool paper to read through. Take this statement "Proliferative oncogenes cause large changes in the surfaceome that are diverse in detail but have common functional themes."
That's the title of figure 2 and it's just marvelous. There are patterns to carcinogenesis (that's a word, right?) Yes, there are a couple genes that if they're messed up, you're screwed, but there are so many checks and back up systems that cancer is a disease of lots of things going awry most of the time.
I might have rambled about this recently, but it's always going down through the MAP Kinase things. Or at least it always seems to. How do you interpret that? I don't think you do. I think you go to the surfaceome and see how hundreds of proteins are changing in dozens of different GLYCOSYLATION THINGS and that's the fine tuned responses. All that stuff downstairs in the cytoplasm? That's so far from the action that it seems a little silly to be doing that ERK western blot.
If you can get very far in this one without being inspired that (glyco?)proteomics has finally arrived and it's time to get out of your chair and go use it to fix medicine, you're crazier than I am.
...and whoever made those figures deserves an honorary art degree or something....and should put a tutorial online about how to keep a certain preprint server from rendering your images in 41 dpi....
No comments:
Post a Comment