BroadE: Sample prep for proteomics

so up to now Steve has told us how the mass spectrometer works how things go into the mass spectrometer the fundamentals of the mass spectrometer in some of the words and things that we use resolution accuracy sensitivity is something that he wants me to get into I don't know if I will karl has told you how we look at what's being fragmented in the spectrometer what we see how we analyze the peptides that we see but I think a lot of you here are thinking well I have a biological problem I have a sample and I want to get some information out of it help me get my sample into the mass spectrometer and that's kind of the idea of what I'm going to show you today and then tell you in very broad strokes what what the main types of discovery experiments are Hasmik will get later on today into the targeted ones and instead we'll show some examples tomorrow I've kept my talk short for a couple of reasons one is that I'm the only thing that stands between you and the coffee break and two that I want to allow for you to stop me ask me questions and just make this very interactive and because I'll probably mess up the here we go okay so in general the experiment and just to reiterate is we start with our sample in general we like to have sample versus control and Jake will get a little bit into this later today we then put this into the high-resolution HPLC which will separate peptides as they go into the mass spectrometer this is the inlet that Steve was talking about we generally do high resolution mass spectrometry to give us the spectra DMS one this is the parent ion which then gets fragmented from which we get sequence identification as Carl just explained the data processing and analysis and from there we get peptide and protein identification and tomorrow Kasper will get into how we're working towards getting those proteins that we are identifying into networks in pathways that will make sense to the to us as experimentalists so I will focus on this part here we start from a very calm sample mixture it will most generally be either a cell lysate part of an organism it could be a part of an organ it could be a sub cellular organelle we want to generate proteins a mixture of proteins for from them then digest peas into peptides that can then be fractionated at the peptide level cleaned up and go into the spectrometer why is this so important Steve got at the sensitivity issue in general the proteins that you're interested in are going to have varying amounts within a very complex mixture that also has proteins in various amounts so this is what we normally call the dynamic range problem we have huge ranges of abundances in a very complex mixture and we want to be able to sample them in the spectrometer as they're coming out from the HPLC so sample preparation is very different from experiment to experiment and sample to sample but the underlying thing is you want to be able to from your biological system make a mixture of proteins that will then be easily a sequence and identified by the spectrometer one of the things that we do sometimes is fraction eight at the protein level and this is again in order to minimize the dynamic range of your sample the less complex your sample is the easier it will be for the mass spectrometer to sequence the peptides as they come along so now I will walk us through each of these steps and what are the methods that we normally use within the group and maybe touch briefly about some that we don't use but the vast majority are the things that we routinely do within the group so cell tissue sub cellular fractionation solubalization what you want to do is lyse yourself in order to get a nice mixture of the proteins that you care about and this can be done in two ways using denaturing buffers and those are buffers that what we'll do is just break the quaternary structure of your protein to generate a nice single polypeptide chain that can then be digested these are normally based on care chops and most likely urea or when it in hydrofluoric ts in the buffer can also be used and what what this will do is just completely unfold and denature protein that's great because then you have a nice single chain that can be chopped up by the end of proteases and it's very useful when you're doing complete protein experiments or post translational modification but when you lose the secondary structure of your protein you end up losing any information that might be about how the protein is talking to another protein is making protein protein interactions protein small molecules or protein DNA so the other sets of conditions that we use are non denied ring buffers these are most generally Reaper based buffer which is just risk chloride buffer sometimes you might have mops you might have heaps it's just a Belfort that has much milder conditions with a little bit of detergent to help in the solubilization detergents are not our friends and I will get into this and keep bringing it back every single time we generally do not like to have detergents but sometimes we can't get around them so we keep them at low concentrations we also use the oxy code so this is very useful when we're doing affinity proteomics types experiments many times you might have an organelle or or part of a tissue that is very hard to get into solution it were to get the proteins into solutions so we preface these by using some physical disruption methods these can be either sonication grinding with a mortar and pestle or cryo polarization which is what we've gotten into more and more within the group which basically is freezing your sample and then hitting it really hard with a hammer to make little little tiny pieces so that these buffers that we're using can get in there and extract our proteins that we want to generate one of the things you have to keep in mind is the minute you lyse your cells and you open and you release all the proteins you're also releasing protease as phosphatases all sorts of enzymes that might modify your protein in ways that you might not like if you're looking for post translational modifications you might want to preserve those you don't want to release phosphatases that way then it changed the nature of your sample so it's very important to think about what additives in each ayat' protease inhibitors you could produce omics possible classification hitter's producing inhibitors protease inhibitors in general so this is something that again is very dependent on what type of experiment you're doing but the general ideas remember you're releasing a lot of enzymes when you break up your proteins again you've seen this picture before Steve mentioned that called mention it but urea which is what we normally use to soluble eyes underneath our proteins in solution with time and heat will form an ammonium cyanate iron which will commemorate your proteins so if your solubilizing your proteins don't heat in urea don't heat them and if you do just make sure you are aware of it and you know that most likely your proteins will blow turmeric or emulate it so then as Carl was saying you can include that in your search engine and have more of a likelihood of finding them if you don't know that you did if you don't know that you did it or you've done it you might not be able to identify a wide number of proteins or peptides fractionation at the protein level again the whole idea is to keep your sample as minimally complex as possible in order to allow the spectrometer to look at what's coming through it so when we fraction ate at the protein level before we digest them the first thing we do is reduce an alkyl aid Steve and Karl both mentioned it I'll get into it in a little bit but we run an SDS page where we separate the proteins by molecular weight what this does is twofold one it allows us to see what the dynamic range of our sample is and two separate things that are very dark so we would segregate this band from something like here in order to not swamp the instrument with this the peptides coming from this protein when it's trying to find the low abundance proteins so it's very useful when you have the ability to run a gel to be able to look and separate your proteins by molecular weight and determine what the dynamic range of the sample is the other advantage of running a gel is that it helps to clean up your sample so if you do have detergents that come from your sample preparation running them through the gel will help clean them up we normally use BIST crystals and we with Coomassie staining other ways of separating at the protein level could be size exclusion chromatography which you don't routinely use that also separates by molecular weight or isoelectric focusing at the protein level which just separates the proteins according to their p i-- what we routinely do if we want to fraction a that the protein level is run an SDS gel again before we even run them on a gel or if we're doing a new solution digest the first thing we want to do is break apart these disulfide bonds which are generating secondary structures and we don't want them to refold so the first thing we do is alkylate them and the regions we most commonly use in the lab are d TT and iodoacetamide that way now we really have a nice long chain where the proteases can come in and cut in the peptides and so here we go the protease is the most commonly used is trypsin this it's very specific so we know where it cuts it also leaves an average-size peptide of about 9 residues which is very helpful for identifying them and being able to map that peptide who easily uniquely to a protein and it also gives it a small enough size that it can be sequenced and the other thing about tryptic peptides is that their charges around 2 plus so those are easily ionizable and and easy to determine what in the search engine why would you use other ones as Steve mentioned lycée staff has been all of these the reason for using them is if you have one specific protein that you would like to see full coverage you might think about combining these two and then generating overlapping peptides that have different chromatographic or ionization states so now you can start really mapping a determined protein that you're interested in in to a very high degree of confidence and look at all the possible peptides another reason is that you might be looking for in a particular protein a specific post translational modification that you have an idea of where it could be and trypsin might not cut or give you the peptides of the length or the desired coverage to find that posters modification so in general we use trypsin but depending on what you're doing you might want to start getting crafty and thinking about which ones you want to combine or what you want to use or using a combination of these then comes fractionation at the peptide level again you when you soluble eyes you lyse and you digest a complex mixture of protease you're going to end up with an even more complex mixture of peptides and you might have peptides coming from a very abundant protein that are completely masking peptides from a very non abundant protein so the idea is to start fractionating divide and conquer and the ways we do these are by either ion exchange in which we use a strong cation or anion exchange column and we allude with either a pH or a salt gradient the other way is isoelectric focusing based on the P I so now the pet each peptide has a p i if you have carrier and flights that are immobilized on a pH strip and then you apply current your peptides will travel and once they find the P I there on P I their charge will be neutralized and they will stay there so now you've been able to separate your peptides according to their isoelectric point and the other very common way of doing it is by hydrophobicity reversed phase in which increasingly polar solvent is separating your peptides from the hydrophobic surface and so peptides that are hydrophilic or water-loving will elute first peptides that are more hydrophobic will be retain more and if you play around with a pH you can start getting different cell activities and be able to decide how you you separate the peptides all of these are prior to going into the mass spectrometer so after you've separated your peptides in this way they will go into this high-performance HPLC at high pressures and again go through most likely a reverse phase separation before going into a mass spectrometer at this point is where you would enrich if you are looking for a post translational modification so Phillip and namorada will get into this into much more detail too but once you've got your peptides you can go in and enrich for the post translational modification you want either by an antibody against the modification or affinity of that post translational modification has a case of possible peptides desalting and cleaning up this is I have a whole slide with almost nothing in here but pictures but it's because it's extremely important the key to an L CMS experiment is clean samples and minimally complex samples we want to get rid of the salts salts are charged they will end up suppressing the ionization of your sample and will also end up cutting very technical word but you will end up not being able to introduce sample into the spectrometer because salts end up accumulating at the entrance of the instrument so what we normally use our pipette tips that we load with reverse face c18 material but there's a slew of commercially available products as well depending on the scale of your experiment you could use cartridges or plates for multiplexing and well we're in the subject of cleaning up and what's important what are the things that are enough and led to an experiment what do we want to avoid salts as I mentioned they can cause attic formation which cause showed so we don't normally look for them and so they'll end up complicating your spectra how do we deal with it we remove them we use desalting columns and if we can reuse ammonium bicarbonate which sublimates and will not accumulate in your sample the biggest worst enemy are detergents and polymers they form M over Z polymers which will use the mass spectrometer will spend all its time sequencing those polymers you will never get to see what was in your sample the other thing is they tend to accumulate on the column on the spectrometer we've had instruments down for about a week just because we introduced it detergent in it and we're down for a week just because we didn't remove the detergents and they can also suppress ionization so if we can we avoid them if we cannot avoid them we go into a gel to clean them up we might not even want to fraction it at the protein level but we'll run them into the gel excise that plug and that's a way of cleaning it up there's also a ion exchange chromatography that can be done to remove some of the ionic detergents but this is probably our biggest enemy there are other things that we try to avoid like non volatile organic solvents and those we either get rid of by cleaning them if they're they normally come from affinity enrichment experiments in which your proteins are already bound to a solid face so they can be washed away other things are strong as it's like TFA and TFA is something that people don't normally think about because it's such a good iron pairing reagent it gives you beautiful chromatography you would think it's a great thing to have but because it's an ion pair and reagent it will suppress the signals from the analyte so what we use is formic acid it gives us nice peak shapes we're able to get chromatography using it but it won't suppress our signal and then the infamous our famous characters and that's just basically skin dust and we can't really avoid them they're very very abundant they're in the air but there are ways that we can do to avoid them use clean materials when you're putting your gels in there when you're handling your sample use gloves tie your hair try not to cough and spit in your sample which sounds like a very prosaic thing but it makes a huge difference because if not all you're going to be seeing are skin proteins so now the most common are most important classes of experiments in discovery proteomics I am NOT going to get into targeted methods those will be talked about later but there they bingley mainly fall into global proteomics which is what you want to know what proteins are there and it - what abundance relative abundance and have they changed because you've returned them in some way or another mobile clothes translational modifications have those proteins been modified post-translationally what are these modifications and to what extent aren't they're present when you treat your sample some way or not an affinity proteomics which is when you want to see what a certain protein is talking to is interacting with is interacting with another protein in a complex with a small molecule with a nucleic acid and in very very simplified terms this is what I mean when I say a global protium you start with a protein mixture it could be two protein mixtures sampling and control and you end up with a comprehensive list of proteins and how their expression has changed either by time with time with perturbation or not and when we talk about global proteomics if you're starting with about 200 micrograms of a cell lysate you might able be able to identify 8,000 proteins if you start from 200 micrograms of Tisha about 10,000 proteins deplete at plasma and this is very important when you're talking about biofluids most of them plasma serum they will have huge amounts of immunoglobulins and albumin so again the dynamic range problem if you did not deplete of those very abundant proteins that's all you would be sequencing in your master experiment so if you start from about 200 micrograms of plasma that has been depleted of these components you can end up with about 3,000 proteins being identified when you're talking about a global post translational modification again you start with a protein mixture and you end up with a comprehensive list of proteins with post translational modifications these the ability to quantify and determine the PTMs is given by the depending on what PTM you're talking about what input amount you need and how many sites you can find in a deep coverage experiment and this means comprehensive fractionation at the peptide level if you think of that phosphorylation you can start with 1 2 minutes of say a cell lysate you might end up with 20,000 plus of peptides a set elation you need more material and you might be able to get about 3000 saturated peptides ubiquitination starting with 5 to 10 minutes you might end up with 20,000 ubiquitinated sites and the third type is affinity proteomics where you're looking at protein protein interactions sometimes disrupted by small molecules protein nucleic acid that is a protein talking to DNA RNA or protein small molecule which has been known up to now is target ID but it's a subset of these because conceptually these three are exactly the same all you're doing is going fishing with the bait whether you're paid as a protein nucleic acid or smallmouth you're mobilizing your bait and going to see in the cell lysate of interest what it is binding to so in here your your input is really a protein mixture that has been simplified by enriching it with the bait at the protein level and what you end up with is a rank ordered list of protein binders to that particular bait and with that oh ok so we're going back to what the steps are soluble eyes fraction eight at the protein level if necessary digest into peptides fraction eight at the peptide level clean it reduce the complexity make sure you're not suppressing ionization and for each of these three types of experiments you might have different things I touched upon this when you're using a global proteome you want to salt you can soluble ice and care traps you can use physical disruptions you might need detergents some of the things you might want to do if you want to decrease the complexity of your sample at the cell lysis is do some sub cellular fractionation if you're doing a global PTM experiment again you will soluble ice with any of these and remember that if you're looking for a specific PTM you have to make sure that none of the enzymes you're releasing is affecting that and when you're doing affinity enrichment you need to make sure that you're using mild solubilization techniques you can't really do nature your protein because then you lose any information you might have fractionation at the protein level that is most commonly done when you're doing affinity enrichment because you want to see how enriching your protein comp mixture by that bait is changing the dynamic range of your of your sample it's not equally distributed amongst the molecular weights of your proteins you can do some fractionation at the protein level if any in any of these other methods but it's not as important most generally Global protium and global PTM the digestion is done in solution whereas when you're doing affinity enrichment because you've fractionated at the protein level you will do it in jail and then fraction again this part here is depending on how deep you want to go in your proteome you will do more or less fractionation to allow the spectrometer to have more time to see what's the peptides I repeating myself but I think it's important to realize that that's where the most impact you can do on sensitivity is when you're doing for PTMs you will also fraction eight at the peptide level but it will also take that point of your experimental design to enrich for the PTM that you're looking for and again at the peptide level once you've you've fractionated at the protein level you might not need to do such extensive fractionation after the enrichment with the bait and the protein molecular weight separation and lastly this is something you have to keep in mind which is what is the goal of your experiment you could end up with a catalog of proteins just a list of proteins that were identified in your sample and then you're off and running and you can look and decide how to prioritize them or you could want to have a prioritized list of candidates that will generate or answer a hypothesis and if this is what you want you will want to come back after the break and listen to Jake talk about quantitative proteomics because that is the difference between a catalog of proteins that you need to make sense of some way or another or using quantitative proteomics which will give you an extra false discovery rate which is the likelihood of that protein being important in your data set and rank ordered or just all of them the same and with that I will take questions and they have given us a very long coffee break you

Channel: Broad Institute

Views: 10,116

Keywords: Broad Institute, Monica Schenone, BroadE, Proteomics, Sample prep

Id: Zaqt9Jo-U-M

Length: 24min 11sec (1451 seconds)

Published: Mon Sep 30 2013