This is a short/sweet study showing where FTICR technology can get to in terms of native protein complexes.
Native mass spec is hard to do -- top down proteomics is hard to do -- the two together might be the mass spectrometrist Kobayashi Maru (reference -- like you need one...nerd...) ;)
Characterizing a 1.8MDa complex is nothing to scoff at. Nowhere near the record (which was around 10x higher last I looked). The fact that this team can not only get the mass but also top-down characterization is seriously impressive.
Unfortunately for me, my new facility retired the FTICR due in large part to crazy Helium prices before I got there -- but....one of these things just showed up...
It is the first one to be installed in the whole state of Maryland (and it's not mine -- it belongs to a really good friend who has tons of really cool intact/native stuff lined up for it. I am, however, hoping that a time will come that maybe she would really need access to, I dunno, ETHcD and we can work out a deal for some EMR run time!) FTICR or not, this is how native protein and complex characterization is gonna go down here!
Yeah...ETD/ECD on the native complex like the Nature paper above uses would be nice, but SID and (stepped) HCD gives you a lot of power. And the Helium/year for an FTICR would almost pay for a second EMR...so it seems like a fair trade off for now!
EDIT (1/28/18): Great! Almost no one has really yet read this awful post. (Blogger actually tells me how many people have seen this and the joke:useful ratio is way off.
This recent review in JASMS is an awesome short perspective on where we are right now in terms of Native protein mass spectrometry -- and where we need to get to.
This figure from it sums up why Native MS is so powerful.
Take the extreme complex on the end. If you're used to shotgun proteomics you are probably very aware of how much harder it is to manually pick through a z +4 peptide compared to a z+2. At 10,000 m/z the 490kDa native protein is still around z+50. Imagine how hard it is to make sense of it if it was around 1,000 m/z. It would have to pick up 500(5e2!) charges. And with 490kDa of protein -- it probably would!
Also consider this: in shotgun proteomics on the more sensitive instruments these days you are almost always seeing your peptide showing up as +2 and +3 (and if you look hard and it is a large peptide..maybe also +4) it's all pH, pKa gobblydygook that determines this. If you have 500 charges on something as your dominant peak, 499,501,498,502,etc.,etc., are going to be there (in decreasing likelihood -- translating to intensity) diluting your charge envelope until the maximum number of ions your C-trap let you load is still hardly enough to get you above the noise.
I think this is the point that is so nicely made in the image above.
One more since the espresso just kicked in. Check out this (by perspective of the studies mentioned above) much smaller intact problem -- intact antibodies (around 150kDa) nicely shown in this recent study.
I think this makes the clarification of the spectra more clear. The denatured mAB with 45 charges is clearly more complex than the same protein in native MS. Even doubling the number of charges makes it more difficult to determine structural modifications as large as glycans (~202 Da!). Still do-able, but harder than when your dominant peak is only 25 charges. This third paper spends a lot of time on antibody drug conjugates (ADCs) where they further complicate the intact antibody by adding things to it to turn the mAB into a disease-seeking missile for drug delivery. In case you're wondering, this doesn't make the mass of the mAB easier to resolve.