153x Filetype PDF File size 0.17 MB Source: www.ehu.eus
PeptideSequencingby Secondaryarticle EdmanDegradation Article Contents . Introduction JohnBryanSmith,CelltechChiroscienceplc,Slough,UK . TheBasisoftheMethod . TheRoleofPeptideSequencingwithEdmanChemistry Edmanchemistryisalong-establishedapproachtodeterminationoftheprimarystructure, or sequence, of polypeptides. Introduction sion(seethenextsection),andcouldundertakesolid-phase Bythemiddle of the twentieth century it was known that sequencing as well. Later developments were automated proteinswerecomposedofaminoacids,butnothowthese (‘on-line’) phenylthiohydantoin-amino acid identification were joined together. Were they arranged in blocks of by high-performance liquid chromatography, and later similar residues? or randomly mixed together? or in still, morethanonereactioncartridgesothatonesequence repeated patterns? At about that time, methods were run could automatically be followed by another. Today, devised to provide an answer to these questions:namely, manual sequencing is rarely done. Automation has that each individual protein has its constituent amino acid broughtsavingsinresourcesandtimeand,withminimiza- residues arranged linearly in a unique order, or sequence. tion of interference by oxygen and water and improved ThestrategyofSangerandcolleaguesforthesequencingof reagents and reaction conditions, a marked increase has insulin was to characterize series of small overlapping beenachievedintheefficiencyofthewholeprocess.Inthis peptides produced by cleavage of the parent molecule. waytheamountofsamplerequiredtoobtainasignificant Determination of the overall amino acid content and the sequenceisnowordersofmagnitudelessthanitwas,atlow identity of the amino- (N-)terminal residue for each or subpicomolar levels. peptide allowed deduction of the sequence of the whole molecule(Sanger,1959).Analternativeapproachwasthat describedbyPehrEdman(1950).Thisalloweddetermina- TheBasisoftheMethod tion of extended sequences of peptides or whole proteins, and has been used widely up to the present day. The method employs a series of chemical reactions to remove Peptide sequencing by Edman chemistry may be divided and identify the amino acid residue that is at the N- into steps as follows, and as illustrated in Figure 1. terminus of the polypeptide chain, i.e. the residue with a free a-amino group. At the same time, the next residue in Coupling the sequence is made available and subjected to the same round of chemical reactions. Reiteration of this process Phenyl isothiocyanate (PITC) reacts with an a-amino reveals the sequence of the polypeptide. group(orinthecaseofprolylresiduewithaniminogroup) Originally described as a manual method, it was later at the N-terminal end of the polypeptide chain, to form a partly automated (Edman and Begg, 1967) in an instru- phenylthiocarbamyl derivative of the terminal residue. menttermeda‘sequenator’,thedesignofwhichwasusedin Basic conditions are required for this reaction. Edman the commercially available ‘spinning cup’ sequenator originally used pyridine to generate a pH of 8.6. marketed by Beckman in the late 1960s. In 1971 Laursen Alternatives used since include ‘Quadrol’, trimethylamine describedadifferentdesignofautomatedpeptide‘sequen- and N-methylpiperidine. Clearly, a free a-amino group is cer’ that used the same chemistry but with the sample required for this reaction to occur. If the amino group has covalently attached to a solid support that minimized the become modified and will no longer react with PITC, the loss that can occur during steps involving extraction of polypeptide is said to be ‘blocked’. Many proteins are reaction products from solution (Laursen, 1971). This blocked by acetyl, formyl or pyroglutamyl groups, for ‘solid-phase’ sequencing approach was useful for sequen- instance.Variousremedialtechniqueshavebeendescribed cing of short peptides that were especially easily lost in to deblock such molecules (Hirano et al., 1977). Blockage extraction steps. A significant advance came with the ‘gas- may also occur artificially during the handling of the phase’ peptide sequenator (Hewick et al., 1981), so-called sampleinthelaboratory.Tracecontaminantsincommer- because some reagents were delivered as vapour. First cial preparations of non-ionic detergents are known to commercialized as the Applied Biosystems model 470A, cause blockage, for instance. Again, a peptide may not the design included automation of the process of conver- haveanN-terminusatall:cyclosporin is just one example ofanaturalpeptidethatiscyclic.Incaseswherethereisno ENCYCLOPEDIAOFLIFESCIENCES/&2001MacmillanPublishersLtd,NaturePublishingGroup/www.els.net 1 PeptideSequencingbyEdmanDegradation 1 2 Currently, trifluoroacetic acid (TFA) is used for this R O R H NCHC NH CH peptide cleavage reaction. Conditions are as anhydrous as is 2 practically possible in order to minimize acid hydrolysis at + pointswithinthepolypeptidechain.Wherethishappens,a Ph N C S new N-terminus is generated and becomes subject to sequencing.Thisthencontributestobackground‘noise’in thesubsequentanalysis.Minimizationofthisacidcleavage leads to longer and clearer sequence runs. 1 2 R O R Ph NH C NH CH C NH CH peptide Conversion S + The ATZ residue is separated from the peptide by H (trifluoroacetic acid) extraction in organic solvent (ethyl acetate or chlorobu- 2 tane), andisthenconvertedtoamorestableformtoallow R better analysis. Conversion to the more stable phenylthio- 1 peptide R CH C O + H N CH hydantoin(PTH)formisdoneinaqueousacid(25%TFA, 2 HN S v/v in water). Some modified amino acid residues (such as anilinothiazolinone C shortened peptide glycosylatedasparaginyl)maybepoorlysolubleinorganic (ATZ) NH solvent andsogiveablankattheircorrespondingpointin amino acid thesequence.Insolid-phasesequencing,wherethepeptide Ph is covalentlylinkedtoasolidsupport,alternativemeansof + extraction of the ATZ residue may be tried, in the H O + H (trifluoroacetic acid ) 2 knowledge that the remaining peptide will not also be extracted and lost (giving large drops in yield). 1 R CH C O HN N Ph Analysis of PTH residues phenylthiohydantoin C (PTH) S The PTH residue generated by each cycle of Edman amino acid chemistry is typically identified by chromatography, Figure 1 EdmanchemistryforN-terminalsequencingofpolypeptides. originally thin-layer chromatography and latterly re- versed-phase high-performance liquid chromatography. free N-terminus, the polypeptide must be cleaved by The PTH amino acid residue derived from each cycle in chemical or enzymatic means to yield fragments that do turn is identified and quantified by comparison with havefree N-termini. On a practical note, it is necessary to standards, and the sequence is described by the order of avoidcontaminationofthesamplewithamine-containing residues from the N- to the C-terminus. If radioactive nonpeptidic species, such as trizma base, since these, too, amino acid residues are present they may be detected by mayreactwithPITCandgenerateproductsthatinterfere their activity at this stage. with subsequent analysis. Reaction of PITC also occurs withtracesofwaterandothermoleculesthataredifficultto Theefficiencyofsequencing exclude completely from the reaction, so byproducts such asdiphenylthioureaarecommonlyfound.Minimizationof The different amino acid residues, being structurally side reactions such as these helps to improve the efficiency different, react at each stage with different degrees of ofthechemistryandsubsequentanalysis,andtheabilityto efficiency. The overall efficiency (‘repetitive yield’) is less dothis is one of the advantages of automated sequencing than100%(usuallyoftheorderof95%),sooverthecourse methodsoveroldermanualmethods. ofanumberofcyclestheyieldofsequencedeclines,andthe degreeofstagger,or‘lag’,graduallyincreases.Atthesame Cleavage time,theamountofbackgroundnoiseincreases.Whenthe sequence signal reaches the level of the background, the In the presence of strong acid, cleavage occurs at the first sequence becomes uninterpretable. The number of cycles peptide bond, giving the peptide (minus the first residue) at which this occurs may be small or large (50 or more), and the liberated first residue as the anilinothiazolinone depending to a great extent on the size and amino acid (ATZ)form.Onceotherreactantsandproductshavebeen content of the polypeptide itself, since this dictates the washed away, the shortened polypeptide can be taken degreeofrandomacidhydrolysisandothersidereactions. throughanotherroundofcouplingandcleavagetorelease Thetimetakentosequenceasampleisconsiderablyless the second residue, and so on in a cyclical fashion. than it used to be – the time for one cycle of Edman 2 ENCYCLOPEDIAOFLIFESCIENCES/&2001MacmillanPublishersLtd,NaturePublishingGroup/www.els.net PeptideSequencingbyEdmanDegradation chemistry has been reduced to 20 minutes. Current particular states of development or disease. For this sequencers may have more than one reaction cartridge, purpose, a short sequence just a few residues long is all sothemachinescanbeusedforafull24hourseachdayand that is necessary for screening against databases of known generate much information in a short time. gene sequences, when used in conjunction with other The amount of sample required currently to provide a properties such as molecular weight, pI or masses of sequence of a few to 20 or so residues is of the order of 1 peptides. Peptide sequencing also has an important role in picomole or less. Automated sequencers can accept quality assurance of ‘biopharmaceuticals’, recombinant samples that are covalently bound to solid supports, such therapeutic proteins that are the result of the molecular as synthesized peptides remaining on the supports on biology revolution. which they were made, or samples that are noncovalently Edmanchemistryisnowastandardmethodforpeptide boundtoglassfibreorpolyvinylidene difluoride (PVDF). sequencing.Inrecentyearsithasbeencomplementedbya Thelatter maybeusedtotrapproteinstransferredfroma varietyofmassspectrometricmethodsthathavebeen,and polyacrylamide gel, a rapid method for resolution of continuetobe,refinedsuchthattheycandeterminemasses complex mixtures, and this approach to sample prepara- of proteins and of peptides derived from them, and can tion is very common. determinepeptidesequencefromthepatternsoffragmen- The simplest case is one where there is just one tation from peptide into individual amino acid residues. polypeptide in the sample, but it is common to have While the methods may complement each other today, it differently processed forms of the same protein or other could be that in the next decade or so the mass spectro- proteins present as contaminants. These mixtures can be metrymethodwillreplacetheEdmanchemistryapproach. resolved into individual sequences if they are in signifi- Nevertheless,itremainsacredittoEdman’schemistrythat cantlydifferentquantities,ifoneofthesequencesisknown it hasprovedsorobustandgivensuchvaluableserviceover andmaybesubtractedfromtheother(s),orif,inthecaseof decades of rapid technological change in the field of a variously processed protein, the frameshift or delayed biochemistry. repeat in the sequence can be spotted. Again, mixtures of sequencescanbeidentifiedbyscreeningagainsttherapidly References growing database sequences. EdmanP(1950)Methodfordeterminationoftheaminoacidsequencein peptides. Acta Chimica Scandinavica 4:283–293. TheRoleofPeptideSequencingwith EdmanPandBeggG(1967)Aproteinsequenator.EuropeanJournalof Biochemistry 1:80–91. EdmanChemistry HewickRM,HunkapillarMW,HoodLEandDreyerWJ(1981)Agas– liquid solid phase peptide and protein sequenator. Journal of The purpose of peptide sequencing has changed over the Biological Chemistry 256:7990–7997. decades. Originally the aim was to determine the sequence Hirano H, Komatsu S and Tsunasawa S (1977) On membrane ofaprotein,whollyorinpart,inordertobetterunderstand deblocking of proteins. In:Smith BJ (ed.) Methods in Molecular Biology,vol.64;ProteinSequencingProtocols,pp.285–292.Totowa, its structureandfunction.Theadventofmolecularbiology NJ:HumanaPress. providedanalternativeandquickerwaytodothis,viathe Laursen R (1971) Solid-phase Edman degradation. An automatic sequencingoftherespectivegene.Partialproteinsequence peptide sequencer. European Journal of Biochemistry 20:89–102. wasrequiredfordesignoftheoligonucleotideprobesused Sanger F (1959) Chemistry of insulin. Science 129:1340–1344. in the process of gene cloning and for confirmation that isolated clones were indeed the relevant ones. As sequen- FurtherReading cing of whole genomes progresses, we approach the time whenallgene(andthereforeall protein) sequences will be ReimDFandSpeicherDW(1995)N-terminalanalysisofproteinsand known.Currently,however,knowledgeofthesequenceof peptides. In:Coligan JE, Dunn BM, Ploegh HL, Speicher DW and a gene does not give us complete information on matters WingfieldPT(eds)CurrentProtocolsinProteinScience,vol.2,section that are significant in the function of a protein, such as 11.10. Chichester:Wiley. inter- and intramolecular disulfide bonding patterns or Smith BJ (ed.) (1997) Protein Sequencing Protocols. Methods in modification and processing events. Determination of Molecular Biology, vol. 64. Totowa, NJ:Humana Press. SmithBJandChapmanJR(1998)Proteinsequencing.In:RapleyRand protein sequence has a role in providing this, and in Walker JM (eds) Molecular Biomethods Handbook, pp. 503–525. identification of proteins that are found to correlate with Totowa,NJ:HumanaPress. ENCYCLOPEDIAOFLIFESCIENCES/&2001MacmillanPublishersLtd,NaturePublishingGroup/www.els.net 3
no reviews yet
Please Login to review.