What has proteomics done for biology lately? You mean today? Well -- how 'bout figuring out the protein turnover dynamics for over 9,000 (!! over 9,000 !!!) different proteins in B-cells, Natural
Why is it important? Proteostasis is a critical component of understanding mammalian biology and perturbation of the natural processes is the center of many diseases. Also, all of the cells this study works with are important in their own right.
This dynamic SILAC approach used in the paper is a major improvement over anything I've seen before on this topic -- they can assess protein turnover in a huge dynamic range from a time perspective, assessing proteins that have half-lives as short as 10 hours or as looong as 1,000 hours!
This study has "systematic" in it's title. This translates here into "a ridiculous amount of work". Just when I think I've got this under wraps, and I can't get any more impressed, I realized that they also apply this turnover analysis to protein complexes. How do all the proteins that construct the nuclear pore complex cycle through degradation/replacement in every one of these cell lines? Oh. Like this.
Yes, I know I get excited about a lot of stuff that I read. My basal level state probably appears to exist somewhere between "in complete awe" and "totally blown away" to outside observers, but this study is biology text book altering level stuff and I'm having a lot of trouble putting it down and going to work this morning.
Important note -- This study also features major improvements on the already awesome isobarquant Python software package that can be directly downloaded from links within this great paper.
All the RAW and processed data is available. Since it is 571 RAW files (!!!) and I've already used 2 screenshots from this paper...
Oh no. I'm not done yet. (What time is it?)
In this study the authors use a Q Exactive (Plus, I think) and if you are going through the methods you'll notice that they use a higher MS/MS target value than I bet you're using. 1e6.
Now. Let's think about this one for a second. Why do we keep our target values lower? Because we're cramming a lot of positively charged things into a little tiny space (the C-trap and Orbitrap). I wrote Dr. Makarov once for advice on maximum ions for SIM scans and he said that I would start to see space charging if I went above 5e4 (I can't remember what instrument I was on, but I promise you I kept that email. I didn't frame it, or anything I'm not that weird.)
But that is a SIM scan, right? My only proof that my ion is my ion is that it has a perfect mass and ion distribution. Any shifts from that and I'm in trouble (some FDA pesticide assays on SIM scans require <1ppm mass accuracy for positive ID). This is MS/MS scan. We're using a much more wobbly tolerance, typically allowing 0.02 Da.
If I go to my new favorite bookmark, the RedElephant, it tells me that on a 200Da fragment ion, a 0.02Da shift is 100ppm! Even when these guys purposely tried to space charge an Orbitrap I think they couldn't force it to get 30ppm out on a SIM scan (I forget all the details and I really should go to work sometime...)
So...if we couldn't possibly space charge our ions out of whack why wouldn't we go for a higher target value for our MS/MS ions? Sure, maybe it is overkill, but if there is no downside?
Okay, so check this out. They didn't just go for 1e6 without evaluating it. They've thoroughly vetted this target level. And it isn't a good idea for reporter ions where you need perfectly focused fragment ion masses.
(I need to read more ACS)
But it appears just fine for everything else. You bet your sweet peppy I'm going to run some head-to-head comparisons as soon as I find some open time on one of our instruments.