 Hello, we're now looking at the seventh installment and the final installment of our series on mass spectrometry basics. We've talked about amino acids and proteins in the first lecture, we talked about the basic concepts of mass spectrometry and ion sources in the second lecture, third lecture was about the mass analyzers, fourth lecture was about the detectors, fifth lecture was about the Fourier transform ion cyclotron resonance instrument and the orbit trap that summed up the whole mass spec bit and then the sixth and previous lecture we talked about tandem mass spectrometry and how we fragment ions into smaller bits and then analyze these and now we're going to have a closer look at this fragmentation and that will conclude our basics of mass spectrometry. So without further ado let's have a look. Let's start with something very basic. We have a peptide and this peptide carries a positive charge. If we use electrospace sources, multiple positive charges. Now these positive charges actually are protons, so hydrogen atoms without their electron is a single proton and they're somewhere on this peptide. Now the hypothesis, the assumption is that the protons can in principle travel, so they don't have to be located very precisely in place, they have the ability to move along the entire chemical structure of the peptide. Then we make a very important hypothesis and that is that the location where the proton is most often found will be the weakest point in the whole peptide structure and so if we fragment or when we fragment with CID or HCD remember these are the fragmentation methods that are ergodic that where the energy spreads out across the whole backbone and they can find the weakest point that these weakest points will be found and will trigger a fragmentation event. So knowing which kind of fragments we can expect to see and there are usually B and Y ions so we've established that these are the weakest bonds but then which ones of these B and Y ions will fragment that we try to reduce to this very simple question where do we think the proton will be and if we know where the proton is likely to be we know where the fragmentation is likely to occur. Now how do we figure out where this proton will be? Well a proton being somewhere has something to do with the basicity of that location. If this location is very basic it means it's very ready to accept a proton which again means that the proton there is very energetically stable at that location. Now how do we figure out basicity if you think back to our very first lecture when we had this table of amino acid properties we had this the PKB so this was the minus logarithm of the equilibrium base constant of the amino group of an amino acid and we had the PKX which was the basicity or acidity of the side chain. Now I've plotted these numbers from the original first lecture in a graph this time. What we're looking at is we're looking at the acidity equilibrium constant of the carboxyl group which we're not going to bother with but it's just here for completeness and then we're looking at the amino group of each individual residue of each individual amino acid. So here are the amino acids and this is the PKB of those amino acids and then here these blue bars there are the PKXs. As you can see only a few amino acids actually have a PKX. Remember that something like valine or isoleucine or leucine do not have a PKX because they do not act as bases or acids they won't accept or lose a proton. Now what you should remember is that the side chains they're always available because of course the side chains of the amino acids are there but this amino group is only present at the amino terminal residue in the peptide. So the head of the peptide that amino acid still has its own free amine all the other amines have now been joined together in the peptide bond. What is interesting to see is of course that arginine as we know has the highest PKB so it is a location that very strongly attracts protons. The second most attractive position if you like for a proton is lysine side chain but lysine is very close in attractiveness for a proton as the proline n-terminus. So any peptide that starts on a proline will have a really nice location for a proton on the n-terminus and if that peptide should end on a lysine there will also be an equally attractive position here. Now imagine that there are two positions where a proton can be held reasonably easily and roughly equally stable. That actually means that there is a very good probability that the protons will be highly localized. They will either spend a lot of time here or they will spend a lot of time here. If there's an arginine at the end and remember arginine and lysine why are they so special? We use trypsin to cleave proteins into peptides and trypsin cleaves c-terminally or after lysine or arginine so the vast majority of our peptides will end in lysine or arginine. If there's an arginine here the proton will be very strongly localized there. Remember arginine is about two units of pkx higher than lysine which is in fact a hundred fold more basic. So a hundred fold better location for a proton. So with all of this in mind let's have a look whether we can now apply our knowledge about the specificity of all these different locations and with the hypothesis of this where the proton is that's where things will break and see if this hypothesis holds. This hypothesis in fact is called the mobile proton model. So we are assuming that we have a model of a peptide and in this model the key factor is the fact that the proton is mobile and can in principle travel around and so the ability of the proton to travel around will determine how well how completely something fragments. Now this theory was proven by a series of very elegant papers by Wiesoki et al and earned Wiesoki the ABRF award amongst many other awards and rightfully so because this is very interesting study that is very simple to understand and is very elegant in its design. Let's have a close look at how this works. What we have here on this plot is the collision energy we have put in roughly stating the collision energy is this voltage or is related to the voltage we put across the collision cell. Now the higher the voltage the faster the ions will move. The faster they will move the more energy they will have the more force they have when they smack into the collision gas. So we actually add much more energy and the question we're going to ask ourselves is how much energy do we need to invest in the collisions in order to get all possible fragments and when we get all possible fragments this means that we have a complete coverage of the sequence we have the y1 ion we have the y2 ion the y3 the y4 the y5 we have all these ions somewhere in the spectrum so we've completely blown everything sufficiently apart to see the full ladder of all possible y ions. It's a great way to see how easy it is to fragment. Now remember a more mobile proton should make for easier fragmentation so the more mobile a proton is if the model is correct of course the the less energy we need to get to a high amount of coverage. Now let's see we have two peptides here they are very specifically built for this purpose one has the arginine at the end terminus so it doesn't look like a triptych peptide but if that doesn't matter it's a matter to prove a hypothesis and it has a phenylalanine which is not very interesting from a base perspective because it's it doesn't even have a pkx at the c-terminus and here we have one with a proline at the end terminus and the arginine at the c-terminus. Now in fact we've just moved the arginine into the back and that is also shown by the way that is written we've just swapped the location of the arginine. Now if we have a singly charged version of this peptide that means there's only one proton that can possibly be mobile. The singly charged proton is going to be at the arginine remember arginine at this pkx of 12 and a bit so it's going to hold on to that proton and that proton is going to be very very happy on that arginine you want the denium group so the proton is going to stick around here or here. The result is the proton is not very mobile the proton is happy on the arginine and it's not going anywhere so we don't have a mobile proton it's going to be very difficult to break all the other bonds because the proton never gets to be at these places and so we'll need to invest a lot of energy to get these things broken and that's exactly what we see we have to start at the collision energy of 60 electron volts before we see fragments and then we need to go all the way up to almost 120 for both of them so both the triangles and the circles to get a hundred percent coverage so that's a lot of energy no surprises there so far the hypothesis holds if the proton is not mobile it's very tricky to get the thing to fragment and if you have only one location where the proton is is happy to be and if you have only one proton it doesn't even matter where that location is now things change dramatically when we go to the doubly charged version of this peptide the doubly charged version of this peptide actually has of course still one proton at the arginine side chain but the other proton now no longer feels happy on that arginine that is very simple because two positive charges even on a very basic side will start to repel each other so one proton is going to be at the arginine and the other one will have to go find a new place to be now have a look at this first peptide where is that second proton going to go there isn't a single side chain here that has a reasonable pkx in fact none of them has so the proton doesn't really find another place where it's happy and where it can just sit comfortably at a low energy state it's going to essentially wander around looking for such a place never finding any because this one is already taken and it's the only one there so we here have in a doubly charged peptide we have a very very mobile proton one that is locked down one that is mobile now look what happens this is the triangle plot we start seeing fragmentation already at about 30 electron volts and we get complete coverage really quickly this is a very very sharp curve and we get that at 60 electron volts which is where for the singly charged version we saw the first signs of fragmentation so it's very obvious the mobile proton that we have created here by adding two protons to the sequence yield a lot of coverage very quickly at low energy it still shows the mobile proton hypothesis has merit by the way these curves that you see here are the ones that we've talked about for detectors there are sigmoid curves they don't just show up in detectors now to to finish the whole story let's have a look at this other peptide the circle and doubly charged state obviously one of the protons is going to be on the arginine and it's not going anywhere i think we've agreed on that we've seen that on the singly charged version now the second proton is going to try and find the location to be to feel happy so it it's not going to be happy on any of these residues because they don't have any pkx so they're not basic in any way but look at this we have an n-terminal proline and remember that the n-terminal proline the amino group which is cyclical there that amino group is pretty basic so our second proton is going to find a happy home on the n-terminus as a result of this low energy state that can occupy here on the proline it's not going to travel around so we've got two lockdown protons one here and one there and that and that situation leads to poorer fragmentation because we don't have a wandering proton like we had in the first case in the trial case and look what happens to the collision energies needed to get good fragmentation for this second peptide with the arginine at the c-terminal and the proline at the n-terminus the circle it is very high and in fact you only reach full coverage at the same collision energies we had for the singly charged state so it shows very conclusively that the mobile proton hypothesis is very very valuable it the paper actually goes beyond that and then takes the next step which is to take the same peptide sequences and add an additional arginine now the two arginines are sufficiently widely spaced that each of them can harbor a proton without the protons deflecting each other in fact in peptide backbones usually the side chains point opposite directions which means that the protons are actually pretty far apart and they're perfectly happy where they are and as we would expect a doubly charged version of these peptides fragments exactly as badly as the singly charged version of the one with the single arginine because again one proton is going to live on the first arginine the second proton is going to live on the second arginine and the same here and none of these two sequences will have a mobile proton so this is quite convincing evidence of the mobile proton hypothesis now i have to tell you something else about this which is that arginine and lysine are very different kinds of amino acids because arginine is a hundred times more basic what you will see in an arginine ending peptide is that it's very heavily biased towards y ions if you have a lysine at the end to see terminus it will also have y ions but it will probably have a few more b ions and it will typically fragment more easily because the proton on a lysine is more mobile than the proton on an arginine the energy well the energy valley in which the proton sits in arginine is much deeper and much more comfortable than the one in lysine it's a hundred times deeper a hundred times more comfortable so it's easier to dislodge that proton from a lysine and therefore these peptides tend to fly apart a little bit easier and a little bit more comprehensively so lysine peptides are shifted a little bit back compared to arginine peptides and of course anything with a proline at the end terminus is harder to fragment because that creates a second valley for a proton to sit and not be mobile so you will see these patterns if you care to look for them another thing that you can notice is strange patterns when you have an internal lysine a mist cleavage or an internal arginine that too will localize the charge and you will see a very prominent break a very strong signal for the fragment right adjacent to the internal arginine lysine and you will see fewer fragments outside of that range these kind of things are very typical another typical thing that you tend to see with fragmentation and again look at a lot of spectra which is so easy because there are so many in the public domain a great software to look at them you can see that a proline somewhere internally is usually a site that gives a very very prominent fragment ion the reason for this is that the proline bond is actually very weak so a proline to something else bond we talked about how this is a helix breaker and how this is a special kind of bond with assist trans location so it can act a bit like a hinge that bond is super weak which means that it's very easy to break the peptide there which means that in the majority of cases the peptide will break at the proline bond so if there is an internal proline expect a really big peak there for the fragment spectrum because that's the place where the backbone will break most likely finally we talked about hcd very briefly at the end of the last lecture and hcd this higher energy collisional induced association is a very special kind of fragmentation when it comes to the kind of fragments we expect for the simple reason that b ions which are generated whenever you split a peptide into you get a bi and a y ion these bi ions are actually very unstable if you go back to the previous lecture you have a look at the structure again of these ions you'll see that they end on a co but usually we don't see these kind of bonds usually our carbon has a double bond to the oxygen and then has another bond to another group and then of course it has the bond that links it to the alpha carbon as well so the carbon atom needs four bonds and the four electrons to share and the oxygen only needs two so what people draw is usually a carbon atom connected to an oxygen atom by three bonds so in exchange of three electrons which puts a positive charge of the oxygen and is extremely unlikely to happen because oxygen is quite electronegative and really likes to to have electrons closely located to it as a result this is a very unstable kind of compound a y ion on the other hand you will see that it is stuck with an nh group but that nh group if you put an additional proton there becomes nh2 which is like a mini peptide and that is a far more stable configuration so what we see is when we break a peptide according to by breakage so according to the peptide bond we will get a very stable y ion relatively speaking and a very unstable bi ion so bi ions if you give them a little bit of energy they will refragment and they will they will very quickly break themselves down in hcd where we have this higher energy most of the bi ions are gone they don't show up in the spectrum anymore with a very notable exception which is the b2 ion so what is special about the b2 ion it starts with an amine an nh2 of the original n-terminus of the peptide alpha carbon so the first central carbon of the first residue the carboxyl carbon of the first residue the amine of the second residue the central carbon of the second residue and then finally the carboxyl carbon of the second residue so let's count we've got one two three four five six carbon atoms carbon nitrogen atoms in a row that actually cyclizes and becomes a six ring structure and that six ring structure is very very stable so what you see there is a cyclization of the b2 ion which keeps it intact it becomes strong enough to withstand further fragmentation all the other bi ions the longer bi ions it's very hard to cyclize because the the ring structures will be very big and also because it would be very difficult for the ends to meet up and join together anyway and these will deteriorate through subsequent fragmentation steps into the b2 which will then cyclize and lock in so all the bi ions get compressed if you like into this b2 ion sometimes you can see a b3 ion as well the b3 ion is already a lot bigger it's nine residues big ring but it can happen and you see some of it so that is something to look out for in hcd spectra very few bi ions but a pretty prominent b2 ion nonetheless because the b2 ion is special it is much more stable than all the other ions it's really interesting literature on that but we just don't have time to go into it but if you're interested if you just look on gas phase chemistry of the b2 ion you will find a lot of interesting studies on that so remember the mobile proton hypothesis allows us to explain a lot of the observed patterns and fragmentation spectra based on the sequence which is very valuable when you're trying to interpret a sequence of a sequence to spectrum assignments or a peptide to spectrum matches more commonly referred to it actually allows you to say based on the sequence do i expect this is there an internal protein do i expect a big peak and is there a big peak you can also look at a lysine at the c-terminus we should see more b ions and a better spread of y ions if we see an arginine at the enter at the c-terminus we can say ah we would expect a lot of y ions especially the ones close to the arginine and so forth so this can help you interpret things if you understand this and if you have a basic notion of this and of the pk axis and the pkb's you should be able with a few example data sets to quickly learn what kind of spectra you're supposed to see for certain analytes this is very interesting you get control of your data and you get to learn how the mass spectrometry actually functions on your sample now i'll just finish up with this last consideration remember that in maldi we had singly charged peptides so that was really great because we didn't have to worry about the charge state assignment in electrospray we could have two plus three plus four plus charge states and that would be really difficult to sort out later then again we learned that with high resolution instruments which is essentially anything about apart from an ion trap these days we can look at the isotopes and the distances between the isotopes will reveal the charge state so it wasn't quite a bit of quite as big a problem anymore now the one plus thing from maldi comes back to create a little problem for us if our peptide has a single charge and we fragment the peptide somewhere we now have a b ion and a y ion the problem is we have only one charge so where does the charge go it has to either go to the b ion making a b ion or it has to go to the y ion and whichever part b or y gets the charge becomes an ion the other thing becomes an uncharged molecule and disappears from the mass spectrometry so the efficiency of fragment recovery in maldi is necessarily half because you will definitely lose one of the two fragment ions you will not be able to see both this is really problematic because it creates less dense less informative spectra i say problematic it's let's say it's annoying but rather than problematic and it's less efficient as a result mind you what would you expect a b ion or a y ion to capture the charge that you can figure out from the sequence and from the pkx if we have a lysine on this end and a proline on that end i would say the chances are roughly 50 50 and it might depend on which of the fragments is the bigger fragment and the bigger the fragment the more chance for the proton to be happy essentially if however you have an arginine here and an apolar amino acid or a negative amino acid with a really low pkb for the n-terminus there then the proton is going to make an energetic choice it's going to end up here on the y ion this is one of the reasons as well why you see a lot of y ions with arginine containing peptides because it is more likely that the charges stay on the y ion on the other hand if we have c id sorry electoral spray it has nothing to do with the fragmentation per se if we have electoral spray ionization when we have two plus three plus four plus charges obviously we can split the charge over the y and the b ion and in fact this is not such a silly idea because if we have a fragment here which is say four or five amino acids long and we squeeze two charges there remember one can be on the lysine or arginine but the other one has to find another nice place to live and that may be hard on such a short peptide to find another place but on the b ion we still have the original amino terminus which is a very nice place in general for the proton to live so it is more likely to yield two singly charged ion ions rather than a single two plus ion on either side nevertheless this can be possible and it really depends on the sequence and you can probably figure it out if you look at the pkx and pkb's you can have a good good idea of whether or not this pattern that you see on the fragment ions makes sense this could be very hydrophobic and negatively charged at the end terminus and it will not be very interesting for a charge this one could have a proline at the end terminus and an arginine here and so two charges would definitely end up on this side so again knowing about the basicity of the side chain and of the end terminus and knowing a little bit about the mobile proton model and thinking a little bit about the composition of your ions in the gas phase can help you a lot with interpreting your spectrum so with that we come to the end of this lecture and we actually come to the end of the entire series on mass-petrometry basics I do hope you've enjoyed it and it was educational for you and I would like to thank you for staying with us throughout the entire course bye