CA2406356C - Method and product for the sequence determination of peptides using a mass spectrometer - Google Patents

Abstract

Method is described for sequencing polyptptides by forming peptide ladders comprising a series of polypeptides in which adjacent members of the series vary by one amino acid residue and determining the identity and position of each amino acid in the polypeptide by mass spectroscopy.

Description

y ~ 1 METHOD AND PRODUCT FOR THE SEQUENCE DETERMINATION
OF PEPTIDES USING A MASS SPECTROMETER
RELATED APPLICATION
This application is a continuation in part of copending and cosmonly owned application s~trial number 07/891,177 filed May 29, 1992.
FIELD OF THE. INVENTION
This invention relates to rapid and efficient methods for sequencing formed or forming polypeptides utilizing a mass spectrometer.
Pailypeptides are a class of compounds composed of a(-amino acid residues chemically bonded together by amide linkages with elfaination of water between the c~rboxy group of one amino acid and the amino group of another.
amino acid. A polypeptide is thus a polymer of ~-amino acid residues which may contain a large number of such residues. Peptides are similar to polypeptides, except that they are comprised of a lesser number of a(-amino acids. There is no clear-cut distinction between polypeptides and peptides. For convenience, fn this disclosure and claims, the term "polypeptide" will be used to refer generally to peptides and polyp~ptfdes.

WO 93/4834 ~ PCT/US93/05070 Proteins are polypeptide chains folded into a defined three dimensional structure. They are c~pl~x high polymers containing carbon, hydrogen, nitrogen, and sulfur and are comprised of linear chains of amino acids connected by peptide links. They are siailar to polypeptides, but of a much higher molecular weight.
for a complete understanding of physiological reactions involving proteins it is often nscessary to understand their structure. There are a number of facets to the structure of proteins. These are the primary structure which is concerned with amino acid sequence in the protein chain and the secondary, tertiary and quaternary structures which generally relate to the three dimensional configuration of ~5 proteins. This invention is concerned with s~qusncing polypeptides to assist in determining the prirxr structure of proteins. It provides a facile and accurate procedure for sequencing polypeptides. ~t is also applicable to sequencing the amino acid residues at the termini of.proteins.
Many procedures have been used over the years to deteranine the amino acid sequence, i.e. the primary structure, of polyp~ptides and proteins. At the present time, the best method available for such determfnati~ons WO 93I2~34 ~ PCT/LJS93f~0 is the Edman degradation. In this procedure, one amino terminal amino acid residue at a tame is removed f~c~s a polypeptide to be analyzed. That amino acid is normally identified by reverse phase high performance liquid chromatography (I;PLC), but recently mass spectrosetric procedures have been described for this purpose (1).
The Edman degradation cycle is repeated for each successive terminal amino acid residue until the complete polypeptide hes been degraded. The procedure is tedious and time consuming. Each sequential removal of a terminal amino acid requires 20 to 30 minutes.
Hence, with a polypeptide of even moderate length, say for example 5o amino acid residues, a ssque~rce determination may require many hours. The ~oosdura bas been automated. The automated machines are available as sequenators, but it still requires an una~ocepta~le amount of time to carry out a sequence analysis.
Although the procedure is widely employed, ort~s rich required less time and which yielded information about a broader range of modified or unusual amino acid residues present in a polypeptide would be very useful to the art. A process which can be used to sequence individual members of mixtures of polypeptides would be particularly useful.

WO 93/~A$3~ ~ ~'CT/US93111~5070 Recent advances in the art of mass sp'ctro~copy have made it possible to obtain characterizing data Eros extremely small amounts of polypeptide samples. it is, for example, presently possible because of the sensitivity and precision of available instruments to obtain useful data utilizing from picomole to subpicomole amounts of products to be analyzed.
Further, the incipient ion-trap technologies pr~eise even better sensitivities, and have already been demonstrated to yield useful spectra in the l0-15 to l0 16 sample range.
In general, both electrospray and matrix-assisted laser desorption ionizaton methods mainly generate intact molecular ions. The resolution of the 75 electrospray quadrupole instruments is about 1 in 2,000 and that of the laser desorption time-of-flight instruments about 1 in 400. Both techniques give sta~ss accuracies of about 1 in 10-20,000 (i.e. +/- 0.~10 or better). There are proposed modificdtion~ of tfae-of-flight analyzer that may improve the resolution by up to factor of l0-fold, and markedly improve the sensitivity of that technique.

W'U 93/34 ~ ~ ~.ty~93/OSOTO
r These techniques yield mass measur:menus accurate to +/- 0.2 atomic mass units, or better. These capabilities mean that, by employing.the process of this invention, the polypeptide itself whether already formed or as it is being formed can be sequenced more readily, with greater sped, sensitivity,_and precision, than the amino acid derivative released by stepwise degradation techniques such as the Edman degradation. ~!s will be explained in more detail below, the process of this invention employs a novel technique of sequence determination in which a mixture containing a family of ~fragments~, each differing by.a single amino acid residue is produced and thereafter analysed by mass spectroscopy.
SU~RY OF THE INVENTION
This invention provides a mothoct for the sequential analysis of polypeptides which may be already formed or are being formed by producing under controlled conditions, from the formed polypeptide or from the segments of the polypeptide as it is being formed, a mixture containing a series of adjacent polypptides in which each member of the series differs from the next adjacent meatier by one amino acid residue. The mixture is then subjected to mass spectrometric analysis to generate a spectrum in which the peaks represent the WO 93/14834 ~ PCT/L~593/0!5070 ssparate ors of the series. The difference: in molecular sass between such adjacent ~aeabers fled with the position of the peaks in the spectrum for such adjacent members is indicative of the identity of the said amino acid residue and of its position in the gain of the formed or forming polypeptide.
The process of this invention which utilizes controlled cycling of reaction conditions to produce peptide ladders of predictable structure is to be contrasted with previous methods employing mass spectroscopy including exopeptidase digestion on uncontrolled chemical degradation. See references 2-5.
Because of the uncontrolled nature of these previous methods, only incomplete sequence information could be obtained.
BRIEF DESCRIPTION OF THE DRiIWINGS
Fig. 1 indicates a family or mixture of polypeptides (peptide ladder, as defined hereinafter.) derived from a single formed polypeptide containim~ n amino acid residues. The mixture is analyzed in accordance with this invention to determine the amino acid sequence of the original polypeptide. Each amino WO 9~ ~ PCT/U593/0~78 acid in the sequence is denoted by a number with tt~e numbering starting at the amino terminal of thl peptide.
X denotes a terminating group.
Fig. 2 is an idsalized mass spectrum of the peptide ladder of a polypeptide similar to the family shown in Fig. 1.
Fig. 3. shows the reactions involved in generating a peptide ladder from a formed polypeptide for analysis utilizing phenyl isothiocyanate (PITC) as the coupling reagent arid phenyl isocyanate (PIC) as the terminating reagent.
Fig. 4 is a more precise summary of the process shown in Fig. 3.
Fig. 5 is an idealised mass spectrum of peptide t5 ladders obtained from a mi~cture of two formed polypeptides one of which is identified as A, the other as B.
Fig. 6 is a positive ion, matrix assisted laser desorption mass spectrum of the formed polypeptide [Glut]fibrinopeptide H.

VVO 93J~4834 ~ PCT/L~593/05670 Fig. 7 is a positive ion matrix assisted laser desorption spectrum of [Glut]fibrinopeptide 8 after 7 cycles of sequential reactions in accordance with an embodiment this invention in which a formed polypeptide is degraded in a controled manner to produce a mixture containing a peptide ladder:
Fig. 8 is the spectrum o! the peptide ladder in the region 87-67 obtained from the mixture 99-67 in Example 2.
Fig. 9 is the spectrum of the mixture 6b-33 obtained in Example 2. v Fig. 10 is a spectrum of the low mass region obtained from the mixture 66-33 obtained in Exaa~le 2 showing the side reaction products formed duri~ tht synthesis of HIV-1 protease.
Fig. 11 is a spectrum of the reaction mixtut~e obtained in Example 3.
Figs. 127 and 128 show the reaction support systsm employed in an embodiment of the inventions which permits multiple simultaneous sequencing of polypeptfdes.

1~HH0 93/4834 ~ PGT/liS93/OS070 Figs. 13A and 138 are the mass spectra of the peptide ladders formed from both phosphorylated (1Z11) and unphosphorylated (128) 16 residue peptides containing a serine residue.
Fig. 14 shows the spectrum of a protein ladder generated by incomplete Edman degradation:
Fig. 15 shows the spectrum of the mixture obtained in Example 4.
As will be explained in more detail below, Ffgs. 8 through to are spectra obtained in the sequencing of a forming polypeptida employing the process of this invention.
The invention will be more easily understood if certain of the terms used in this specification and claims are defined.
The term "polypeptids" is used herein in a gens~ric sense to describe both high and low molecular wsigh!t products comprising linear covalent polymers of amino acid residues. As the description of this invention proceeds, it will be seen that mixtures are produced which may contain individual components containing 100 WO 93/244 . pCf/L9S93/05D70 or more amino acid residues or as few as one or two such residues. Conventionally, such low molecular weight products would be referred to a amino acids, dipaptid~s, tripeptides, etc. liowaver, for convenience hsrsin, all 5 such products will be referred to as polypoptfdas sinv~s the mixtures which are prepared for mass spactrosetric analysis contain such components together with products of sufficiently high molecular weight to be conventionally identified as polypieptides.
The term ~formed polypeptide~ refers to an existing polypeptide which is to be sequenced. It refers, for example to [Glut]fibrinopeptids B which is sequenced for purposes of illustration in Example 1. The process of the invention is, of course; most useful for ssquemcing the primary structure of unknown polypeptides isolated, for example, by reverse phase I;PLC of an enzyoatic digest from a protein.
The term forming polypeptide~ refers to such polypeptides as they are being formed for example by solid phase synthesis as illustrated in Example 2.
The ttr~ ~paptida laddar~ refers to a mixtures containing a aeries of polypsptidss produced by the processes described'herein either from a foraad or a WO 931 ~ ~ ' PC'f/U593/DS070 forcing polypeptide. As will ba seen from the various figures and understood from this description of the invention, a peptide ladder comprises a mixture of polypeptides in which the various components of the mixture differ from the next adjacent member of the series by the molecular mass of one amino acid residue.
~ coupling reagents is a reactant which forms a reaction product with a terminal amino acid residue of a polypeptide to be sequenced and is subsequently removed together with the residue.
A terminating reagents is a reactant which similarly forms a reaction product with a terminal amino acid of polypeptide and is stable to subsequent cycling procedures.
DETAILED DESCRIPTION OF THE INVENTION
There are several procedures for building peptide ladders, some applicable to the sequencing of formed polypeptides, others to sequencing of polypeptidas as they are being formed.
One such process will be understood from~a study of Fig. 3 which shows an embodiment of the invention which is applicable to formed polypsptides. The figure shows VI~O 93/l~ ~ ' hCT/L~S9310~56'10 ' . 12 the sequencing of an original formed poiypeptide which may contain any number of amino acid residues, even as many as 50 or more. The polypeptide is shown here by way of illustration as containing thr.e residues, each residue with a side chain represented by Rl, RZ or R3 in accordance with conventional practice.
The significant feature of this embodiment of the invention, as illustrated in the figure, is that the reaction conditions are cycled to produce a peptide to ladder in the final mixture. The final mixture is analyzed by mass spectroscopy to determine the exact mass of the components of the ladder, thereby to accumulate the information necessary to sequence the original polypeptide.
The skilled artisan will recognise that this procedure of sequencing a formed polypeptide makes use of degradation chemistry, but is based on a naw principle, i.e. the original polypeptide is aa~ployed to generate a family of fragments, each differing by a 2p single amino acid as shown in Fig. 1 wherein X
represents a terminating agent. Typically X will be a terminating agent that is resistant to all subsequent reactions or manipulations in tJie cyclic degradation VItO 93/~~8.14 ~ . PCT/US93/05070 process of this invention: As will be described below, in connection with another embodibent of this invention, X may also ba hydrogen.
In the process illustrated in Fig. 3, PITC is the coupling reagent and PIC is the t~rsinati~ reagent.
From such a family or peptide ladder of terminatsd molecular species prepared as outlined in the figure, the amino acid sequence can be simply read out in a single mass spectrometry operation, based on the mass differences between the intact molecular ions.
Fu=thermore, because of the sensitivity of modern mass spectrometers, the accuracy of the amino acid ssquatice thus determined is unaffected, over a wide range (5-fold or more), by the amount of each molecular species t5 present in the mixture.
Fig. 2 shows an idealized mass spectrum of a peptide ladder in which each peak is rep~esentatiy. of one member of a series of terminated polypeptides each member of which differs from the adjacent member by one amino acid residue.
Thus, for example, if the peaks of the highest mass in Fig. 2 represent a polypeptide, the first five members of which at the amino terminal end may be:

VV4 93/1834 ~ PCT/l?S93/O50?0 Glyl-Leu-Val-Phe-Alas-, the next peak of lower mass would represent Leu2-Val-Phe-Ala5-Subsequent peaks would represent products with one ia~ss amino acid residue. The difference in mass between adjacent members of the series would be indicative o!
the amino acid residue removed. The difference in aolecular mass between the first product on the right and the adjacent product would correspond to a glycine residue. subsequent peaks show the sequential removal of leucine, valise, phenylalanine and alanine residues thus establishing the sequence of these amino acid residues in the original polypeptide.
Fig. 3 illustrates a practical sequence of reactions by which the idealized procedure of Figs. 1 and 2 can be conducted utilizing PITC and PIC e1 the reagents for sequencing an original formed polypeptide by cycling reaction conditions to produce a peptide ladder for spectrometric analysis.
20' In the first step of the sequwncing procedure the original polypeptide is reacted with a mixture of PITC
and PIC under basic conditions: A large molar excess of each reagent is employed. A much,larger amount of PITc than of PIC is utilized so as to be certain t3~at at each W~ 931~d834 ~ , ~/US93>05~70 cycle of the pra~cedure most of the available polyp~ptids reacts with the coupling agent but that a small measurable fraction of the available peptide rea~ets with the terminating reagent. The fraction rsa~cted with tbs 5 terminating agent will be detorminad by the relative activities of the coupling agent and th~ terminating agent, and the molar ratio of the two roegsnts.
The first reaction products which fo~~m during the basic step of the cycle comprise a mixture of original polypeptide terminated with PIC (PC-polypeptide) and an original polypeptide terminated with PITC (P'fC-polypeptide). The PIC terminated polypeptide (FaC-polypeptide) is stable or essentially stable under ail subsequent reaction conditions with the rmsult thtt it 15 will be present in a measureable amount in the final mixture when that mixture is ready for analysis.
The next step in the procedure is to subj~act the PTC-polypeptide/PC-polypeptide mixture to acid conditions whereupon a reaction product separates from the PTC-polypeptide. This reaction product contains the terminal amino acid residue of the original peptide.
The separation of this product results in the~.foruetion W4 93/24 ~ . ~ PCT/US93/05070 of a new polypeptide which, because the terminal amino acid has bean cleaved contains one less wino acid than the original polypeptide.
The reaction mixture formed at the end of this cycle contains as the principal products:
1. unreacted coupling and terminating reagents, 2. a first reaction product which is the reaction product between the original polypeptide and the terminnting reagent. it is a PC terminated polypeptids (PC-polypeptide).
3. a new polypeptide from which the amino terminal amino acid residue hae been ~~1.
t5 The skilled artisan will readily understand that sequential repeats of the cycle just described will result in the formation of a mixture which contains as the principal msasureable components a series of PC-polypsptides each member of which contains one less aaino acid residue than the next higher aasber of the series. The member of the series with the highest molecular mass will be the first reaction product between the original polypsptide and the terminating ~p 93/?A~4 ~ . P'G'f/US93/05070 .

reagent. The molecular mass of ench subsequent rea~c~tion product in tt~e series will ba the molecular mass of the next higher adjacent member of the series minus the molecular mass of the terminal amino acid residue removed by reaction with the PITC. The molecular mass of the PIC, blocking group or any other blocking group selected is irrelevant to the spectrometric analysis since the identity of each amino acid residue removed from the next adjacent peptide is determined by differences in molecular mass. These differences identify the amino acid residue, and the position of that mesas difference in the spectrum data set deffines the position of the identified residue in the original polypeptide.
A constant 5~ termination of the available polypeptide at each cycle for ten cycles of the described chemistry would yield a peptide ladder in which the mole fraction of the original polypeptide after each cycle would be approximately MOhE

FRACTION

(X)-i-2-3-4-5-6-7-8-9-10-li-12-.........-n-(OH) .050 (X)-2-3-4-5-6-7-8-9-10-il-12-..........-n-(OH-) .048 (X)-3-4-5-6-7-8-9-10-11-12-..........-n-(OH) .045 (X)-4-5-6-7-8-9-l0-li-12-..........-n-(OH) .043 (X)-5-6-7-8-9-10-li-12-..........-n-(OH) .041 (X)-6-7-8-9-10-11-12-..........-n-(OH) .039 (X)-7-8-9-10-11-12-..........-n-(OH) .037 (X)-8-9-10-il-12-..........-n-(OH) .035 WO 93/34834 ~ PGT/13593/05070 (X)-9-i0-ii-12-..........-n-(OH) .033 (X)-10-ii-12-..........-n-(O~i) .031 (X)-ii-12-............-n-(Oii) .60 remain The differences in molecular mass between each successive member of the series in the peptide ladder can be readily determined with high precision by mass spectroscopy.
With relatively low molecular weight polypeptides, it is possible to repeat each cycle without removal of unreacted PITC or PIC. However, as illustrated in Example 1, it is generally preferred to remove unreacted coupling and terminating reagents at the ,completion of each cycle. Such removal may also include romoval of the cleavage reaction product between the coupling reagent and the terminal amino acid.
Fig. 4 is a more precise summary of the procedure illustrated in Fig. 3 and described in detail above. It specifically illustrates the process utilizing a ~one pots technique. In the figure ~AA~ stands for amino acid and ATZ represents 5-anilinothiazolinons. The other symbols have the same meaning as above.

W4 93/Z4834 ~ , ~ PCT/US93/03070 The figure illustrates the preparation of a peptide ladder from a fors~d polypeptide using controlled ladder-generating chemistry. The stepwise degradation is conducted with a saall asount of PIC and a aajor proportion of PITC. Successive cycles of peptide ladder generating chemistry are performed as described above without intermediate isolation or analysis of released amino acid derivatives. Finally the mixture containing the peptide ladder is read out in one step by laser desorption time-of-flight mass spectrometry (LDHS).
The coupling and terminating reagents are not liaited to the pair described above. Those skilled in the art can readily select other equfvaient reagents.
of course, the procedure can be adapted to either the wino terminal or the carboxy terminal o~ the polypeptide under analysis.
Another procedure for constructing a peptide ladder from a formed polypeptide is to conduct each cycle in a manner to insure incomplete termination. The process is similar to the above described procedure except that only a coupling reagent is employed and the peptide ladder comprises a series of polypeptides none of which is terminated with a terminating reagent but each of which differs from the adjacent member of the aeri,eis by WO 93/t483d ~ ~ PCT/U593/050 70 , ~ 20 one amino acid residue. In this procedure, X of Fig. 1 is hydrogen. The principle of this embodiment of the invention is that only the coupling reagent is amploy~d in the cycle, and the extant of reaction is limited for example by limiting reaction times so that all of the original forced polypeptide does not rsect. ~s a result, after the cycle has been moved to the acid step, the reaction mixture produced will contain:
1. Unreacted PITC, 2. The reaction product of PITC and the terminal amino acid residue with which it has reacted (PTC-polypeptide), 3. Unreacted original formed polypeptide, 4. A polypeptide with one less amino acid residue ~5 than the original polypeptide.
It will be apparent that by suitable adjusnt of reaction conditions; continued repetition of the cycle any selected number of times will produce a desired peptide ladder similar to the ladder produced in the procedure which eaploys both coupling and terminating reagents except that the polypeptide members of the ladder are not end blocked with a terminating reagent.
This process is similarly applicable to a mixture of polypeptiaes.

WO 9314 ~ PCT/US93/05074 J~nother procedure for generating a peptide ladder with only one reagent involves termination by side reaction. In one such process, PITC is employed as a coupling reagent; and, under controlled conditions o#' oxidation, a small amount of PITC terminated polyQeptide is converted to stable PIC terminnted peptide to form a peptide ladder after a selected number of cycles. The key to this aspect o! the invention is the controlled oxidation of a small amount of the PITC terminated polypeptide to form PIC terminated polypeptide which i=
stable, or essentially stable, under subsequent reactions conditions.
To describe the process with more specificity, the reaction steps are as follows:
1. React the polypeptide to be ssqwnced under basic conditions with an excess of PITiC
to convert substantially all of the polypeptide to PITC terminated polypeptide (PTC-polypeptide).
2. React the PTC-polypeptide with a controlled amount of oxygen to convert a small portion of the PTC-polypeptide, say. St, to PC-polypeptide while leaving the balance unchanged.

WO 93/~a4834 ~ ~ PCT/U593/05070 3. Cycle the mixture to the acid step to cleave the PITC bound terminal amino acid from the PTC-polypeptide and leave a polypeptide with one less amino acid residue than the original polypeptide.
4. Repeat tile cycle any selected number of times to generate a peptide ladder for mass spectrometric analysis.
A very significant practical advantage of the process of this invention is that it is possible to sequence a plurality of peptides in one reaction system.
This advantage arises principally from the high degree of accuracy that is possible because of the recent advances in mass spectroscopy.
This aspect of the invention will be understood by reference to Figs. 12A and 12H which show a suitable device for producing a plurality of peptide ladders. In the figure, 1 is a reaction support member shown in the form of a cylinder with a holding basin 2 and a through bore 3 permitting the passage of chemicals. A series of absorbent members or discs 4, for axample absorbent WO 93/?A$34 ~ PCT/US93/05070 ' 23 meabranes era supported by a thin filter member 5 which may be simply a glass fiber or other suitable filttr material.
In practice, the support member would be in a closed system adapted to permit the appropriate reactants for the preparation of a peptide ladder on each disc to contact each polypeptide to be eequa~ed.
After each step of the cycle, the reactants exit the support member through the bore 3. The reactants are i0 delivered to the reaction zone by any conventional pumping system of the type employed to collect reactant;
from a series of reservoirs., mix them and pass the mixture through a delivery nozzle.
Sequencing of forced polyp~ptides on sappimrs i~nobilized on a solid support, ns in the this embodiment of the invention is especially advantageous because it is applicable to very small amounts of total sample and because there are reduced handling losses and increased recoveries.
As applied to the system illustrated in the figures, any convenient number of polypeptides tro be sequenced are separately absorbed on separate discs 4 which may be, for example, an absorbent membrane such as WO 93/-f~183~1 ~ PCT/L~S93/05070 ' 2q tha cati~ni~c, hydrophilic, charge modified polyvinylidene flupride membrane available from Millipore Corp. as Imobilon CD.
The discs are spaced apart on the filter paler 5 which is supported over the through bore 3 on support aember 1 which is then placed in a closed system to conduct the controlled cyclic reactions appropriate to the production of a peptide ladder in accordance with this invention.
The amount of polypeptide absorbed on each segment may be as small as one picomole or even less.
Generally, it is from about.i to about 10 pi~comoles.
In a typical operation, 1 to 10 picoaoles of each polypepti~de are separately absorbed on the sela~ted membrane discs and placed separately on the filter deer which is then placed on the support member a: shown. The peptides are subjected to the PITC/PIC/base/acid cycle described above to generate a peptide ladder on each disc. Each separate peptide ladder containing mixture to be analyzed may be extracted from each separate membrane with an organic solvent containing a small VI~O 93/?,~3d ~ . ~.;./~93/05070 amount of surfactant. one useful extraction solvent is 2.5~ trifluoroacetic acid in a 1:i mixture of acetonitrile and 1-O-n-octyl-~ -glucopyranoside.
Fig. 14 shows the spectrum obtained using the 5 absorbent membrane technology coupled with incomplete termination described above. To generate the peptide ladder which was analyzed, 50 picomoles of [Glu-1]
fibrinopeptide 8 on Immobilon-CD membrane was applied to ABI-471A protein sequences (Applied Biosystem). The 10 sequences was programmed using 5.5 minute cycle time with a cartridge temperature of 56°C so as to insure incomplete reaction at each cycle. Six cycles were performed. Under these conditions, a reaction yield of about 56~ was estimated. The resulting peptide ladder 15 is comprised of free N-terminal amines.
This example illustrates the speed with which the sequencing can be performed. Similar spectra were obtained with a total loading of only 1 picomole of polypeptide on the membrane.
20 Although this multiple, simultaneous, sequence analysis of separate formed polypeptides utilizing the same chemical reagents for separate reactions with the said polypeptides has been specifically described by WO 93/24834 ~ ~ PCT/U593/OSfTO
' 26.
reference to the use of a mixture of specitic.aoupling and terminating reagents in the same reaction zone, it will be apparent that the process is equally applicable to the other processes described above.
The system is; of course, applicable to the use of only one disc for the sequencing of a polypeptide or polypeptide mixture.
Although the discs are shown separately on the support, they may also be stacked ox replaced with a column of suitably absorbent packing materials.
Further, there may be a number of support members in one device and the chemicals fed to the separate support members through a manifold system so that instead of only one reaction zone, there may be a plurality of reaction zones to still further increase the number of polypeptidas which can be simultaneously sequenced.
An especially important embodiment of this invention is that it provides a method of locating covalent modifications on a polypeptide chain particularly post translational modifications of biologically important products which on chemical or WO 93/1834 ~ ~ PCT/US93/05090 ' 27 enzymatic hydrolysis produce polypeptidss which are phosphorylated, aceylated, glycosylatad, cross-linked by disulfide bonds or otherwise modified. Such polypeptides are referred to in this specification and claims as modified polypaptides~.
The inability to directly identify, locate, and quantify modified amino acid residues such as phosphorylated residues in a modified polypeptide is a major shortcoming of standard sequencing methods, and has imposed major limitations on currently important areas of biological research, such as mechanisms of signal transduction. The process of this invention has general application to the direct identification of post-translation modifications present in a peptide chain being sequenced. A modified amino acid residue that is stable to the conditions used in generating the peptide ladder from a formed peptide reveals itself as an additional mass difference at the site of the covalent modification. As described above, from the 2p mass difference, both the position in the amino acid sequence and the mass of the modified amino acid can be determined. The data generated can provide unambiguous identification of the chemical nature of the post translational modification.

WO 93/Z4$34 . PCT/US93/05070 ' ~ 28 A typical example of #:his aspect of the intention is the analysis of both phosphorylated and unphosphorylated forms of the 15 residue peptide LRRASGLIYNNTLMAR amide prepared by the method of Schnolzer et al (9) containing a phosphorylatsd satins residue prepared by enzymatic reaction using 3~, 5~-cyclic AMP-dependent kinase. After ten cycles of PITC/PIC chemistry on each form of the peptide using the procedures described above and illustrated in Example 1, the two separate sequence-defining fragment mixtures (peptide ladders) were each read out by laser desorption mass spectrometry. The resulting protein ladder data sets are shown in Figs. 13A and 13B. Again, the sass differences define the identity and order of the amino acids. For the phosphopeptide (Fig. 13A); a mass difference of 166.7 daltons was observed for the fifth amino acid from the N-terminal, compared with the mass difference of 87.0 for the same residue in the unphosphorylated peptide (Fig: 138). This measured mass difference corresponds to a phosphyorylated satins residue, calculated mass 167.1 daltons. Thus, the protein ladder sequencing method has directly identified and located a Ser(Pi) at position five in the peptide.
There was no detectable loss of phosphate from tie WO 93/~ ~ PCT/~93/05070 phosphoserine residue, which has besn regard in the art as the most sensitive and unstable of the phosphorylated amino acids.
Altough only ten cycles of ladder generating chemistry were performed, sequence-defining fragments corresponding to eleven residues ware observed, apparently arising from a small amount of premature cleavage (l0). This side reaction which can have serious consequences for standard Edman methods, has no effect on the ladder sequencing approach.
A specific and very important advantage of this invention is that it is not limited to analysis of one polypeptide. Mixtures of polypeptides can be analyzed simultaneously in one reaction vessel. Each polypeptide will give a separate spectrum as shown in idealized form in Fig. 4. In this figure, the molecular masses of the original components of the mixture differ by any arbitrary mass difference. Each of the separate spectra can be analyzed as described above even though there aay be appreciable overlapping in molecular mass among the polypeptides to be sequenced. This will be clenr from the figure. As a result, it is possible to sequence wo 93r.. ~ pcrit~93ioso~o proteins by analyzing mixtures of palypeptides obtained by chemical or enzymatic hydrolysis of the protein. The process can be outlined as follows:
Protein sample in quantities of nanomoles or less 5 enzymatic or chemical hydrolysis fragments separate - s.~.
10 by HPLC or gel electrophormais collection of separated peptides parallel cyclic l adder 15 generating chemistry mixture of peptide ladders mass spectrometry 20 readout analysis of data In most cases, gel electrophoresis will be employed to separate proteins and HPLC to separate polypeptides.
Thus, for example, a protein mixture can be separated 25 into its protein components by electrophoresis and each separate component sequenced by digestion into polypeptides, separation and ladder sequencing in WO 9314834 ~ ~ PCT/US93/OS070 accordance with the process of this invention to yield data from which the sequence of the entire protein can be deduced. The process of the invention may also be employed to obtain extensive data relating to the S primary structure of intact proteins at their amino or carboxy terminals.
There follows a description of the application of this invention to a forming peptide.
Stepwise solid phase peptide synthesis involves the assembly of a protected peptide chain by repetition of a series of chemical steps (the synthetic cycle) which results in the addition of one amino acid residue to as amino acid or peptide chain bound to a support, usually a rsin such as methylbenzhydrylamine. The final polypeptide chain is built up one residue at a time, usually from the C-terminal, by repetition of the synthetic cycle. As is well known to peptide chemists, the solid phase synthetic method does not always proceed according to plan. For any of a number of reasons, some of the polypeptfde formed may terminate before the finnl product is produced. For example, a synthesis designed to produce a polypeptide containing twenty amino acid residues may produce as side products a variety of wo 93~aa83a . ~ PCT/Li593l05070 polypeptides containing lesser numbers of amino acid residues, e.g. tripeptides, octapeptides and dodecapeptides.
To utilize the advantages of this invention in solid phase synthesis, polypeptide-resin samples are collected after each cycle of amino acid addition.
Mixing approximately equal amounts of all samples obtained in the course of a synthesis yields a peptide ladder containing all possible lengths of resin bound polypeptide. Cleavage of the resin from such a mixture produces a mixture of free polypeptide chains of all possible lengths containing a common carboxy or amino terminal. Usually, stepwise solid phase synthesis proceeds starting from the carboxy terminal. in these cases, the resulting peptide ladder will contain polypeptides all having a common carboxy terminal.
Consideration of the steps involved in the production of a heptapeptide will explain the procedure.
It the heptapeptide to be produced is of the structure:
Alal-Val-Gly-Leu-Phe-Ala-Gly~, the first synthetic step is the attachment of Gly to tie resin, usually with a spacer molecule betw~ean the resin and the Gly. The next step is the attachment of Nd -blocked Ala to the Gly following well kno~m, coupling WO 93/24534 ~ ~ PiCf/U593/45070 and deblocking procedures so that the synths:is i~
controlled. The cycle is repeated to form the heptapeptide on the resin from which it may be isolated by standard methods.
In accordance with the procedure of this invention, a small sample of polypeptide attached to resin is removed after each cycle. After completion of the synthesis, the seven samples are added together to produce a peptide ladder which contains the following ~0 components.
Gly-Resin Ala-Gly-Resin Phe-Ala-Gly-Resin Leu-Phe-Ala-Gly-Resin Gly-Leu-Phe-Ala-Gly-Resin Val-Gly-Leu-Phe-Ala-Gly-Resin Ala-Val-Gly-Leu-Phe-Ala~Gly-Resin The mixture is then treated, for example with hydrogen fluoride to generate a resin-free peptide ladder which is analyzed mass spectrometrically to assure that the final heptapeptide is of the desired amino acid structure.

WO 93!24834 ~ ~ PC'T/L)S93/05070 ' 34 one possible type of side reaction in stepwise solid phase synthesis is low level blocking at a particular residue (step) in the synthesis.
It will be apparent that each has occurred and mixed separate sample collected subsequent to the step at which a side reaction such as low level blocking has occurred above during the assembly of the final polypeptide will contain a portion of such terminated side product with the result that the amount of such terminated peptide is amplified in the final mixture as prepared for mass spectrometric analysis. Thus, for example, if for some reason such as low level blocking there was a termination of some polypeptide at the decapeptide stage in a synthesis designed to produce a 20-residue polypeptide, the simple from each subsequent synthetic cycle would contain terminated decapeptide and the final analytical sample would contain a 10-fold amplification of this side product. The inforaation obtained by this method of analysis is very useful in designing optimum procedures for synthesizing polypeptides, especially those of high molecular weight.
One adaptation of this invention to solid phase synthesis is illustrated in Example 2.

WO 931~B3d ~ ~ PCT/LiS93/05070 optionally, the peptide train samples collected as described above may be assayed colorimetrically, for example by a ninhydrin procedure to determine reaction yields prior to mixing to form a peptide ladder. This 5 procedure provides a complimentary method of controlling and assessing the process.
In the foregoing process, a sample of polypeptide attached to the resin is collected at each step of the synthetic cycle far the preparation of the #inal 10 analytical mixture. An alternative procedure for preparing the final sample is deliberate termination of a small portion of the forming peptide at each step of the synthetic cycle followed by removal of all of the peptides from the resin to form the analytical mixture 15 directly.
This can be.accomplished by utilizing, instead of one reversibly blocked amino acid residue at each step in the cycle, a mixture of the selected amino acid residue one portion of which is stable under the 20 reaction conditions, another portion of which is susceptible to removal of the blocking group under controlled conditions.

WO 93/24834 l ~ PCT/tiS93/05070 If, for example, the amino acid residue to be added to the forming polypeptide is alanine, tha peptide i~ond could be formed utilizing a mixture of Boc-alanine and Fmoc-alanine in which the carboxyl group is in the appropriate form for reaction, for example in the dorm of an hydroxybenzotriazole ester. After the peptide bond has been formed, one of the blocking groups, the removable group, can ba removed under conditions such that the other blocking group remains intact.
Repetition of this cycle will result in the formation of the desired polypeptide on the resin together with a peptide ladder comprising a series of polypeptides each member of which is joined to the resin and is t~etminated by the selected blocking group.
The procedure will be more readily understood by reference to the prepazation of a specific polypeptide such as:
Glyl-Phe-Ala-Lau-IleS.
The chemistry involved in the preparation of such pentapeptide is standard solid phase polypsptide synthesis applied in such a manner as to produce a peptide ladder. As applied to this invention, by way of example, the C-terminal amino acid residue would be WO 93/t4~63d ~ PGT/1i593/05070 ' 37 joined to the resin, typically through a linker, as a mixture containing a major proportion of t-Boc-isoleucine and a minor proportion of Fmoc-i~soleucine, e.g. in a 19:1 ratio.
The t-Hoc blocking group is next removed with an acid such as trifluoroacetic acid. Since the ~'moc group is stable under acid conditions the Fmoc-isoleucine attached to the resin will retain its blocking group and will be stable to all subsequent reactions.
0 In the next step of this synthesis, a 19:1 mixture of Boc- leucine and P'moc-leucine will be joined to the Ile-Resin, and the Hoc blocking group selectively removed under acid conditions. As a result of this step in the synthetic cycle, the state of the resin may be ~5 indicated by:
Fmoc-ile-Resin Fmoc-Leu-Ile-Resin Leu-Ile-Rasin Repetition of these reactions will result in a 20 final resin mixture comprising a peptide ladder which may be represented by:

WO 93/~d834 , PCT/(~93/05fl7~
' 38 Fmoc-Ile-Resin Fmoc-Leu-Ile-Resin Fmoc-Ala-Leu-Ile-Resin Fmoc-Phe-Ala-Leu-Ile-Resin Fmoc-Gly-Phe-Ala-Lsu-Ile-Resin Gly-Phe-Ala-Leu-Iie-Resin This peptide mixture is removed from the resin by standard solid phase procedures which, optionally, will also remove the Fmoc group to produce an analytical sa~sple ready for analysis by mass spectroscopy as described above.
The peptide ladder can also be formed by the reverse procedure of employing Fmoc as the removable group and t-Boc as the terminating group.
The adaptation of this invention to solid phase synthesis techniques is illustrated in Example 3 and ~'ig: Z1 Any blocking group stable to the conditiona of chain assembly synthesis can be used in this application of the invention. For example, acetic acid could be added to each reversibly N-protected amino acid in a WO 93/1834 ~ ~ PCT/tlS93/f5070 ' 39 stepwise solid phase synthesis in an aaount suitable to cause a few percent permanent blocking of the growing peptide chain at each step of the synthesis. The mass of the blocking group is without effect on the ability to read out the sequence of the peptide synthesized since the readout relies on mass differences between adjacent members of the polypeptide series as described above.
Using the procedures described, each i~ividual resin bead carries the mixture of target full-length peptide arid the peptide ladder. Typically each bead carries from 1 to 1o or more picomoles of polypeptides.
Thus, cleavage of the products from a single bead permits the direct determination of the sequence of the ~5 polypeptide on that bead.
It is recognized that the foregoing procedures are described in an idealized form which does not include possible interference by other functional groups such as the hydroxyl group in tyrosine and-serine, the dextral carboxyl groups in dicarboxylic amino acids or the ~extra~ amino groups in dibasic amino acids. This method of description has been adopted to avoid unnecessarily lengthening the specification. The artisan will recognize the problems which will be WO 93/4834 ~ ~ PGT/ZIS93/f50'~0 . . 40 introduced by the other functional group: and will kTiow how to deal with them utilizing techniques well known to peptide chemists.
It will also be recognized that the procedures described have been applied to relatively small polypeptides. They are equally applicable to large polypeptides. For example, if the forming polypeptide is one which contains twenty or more amino acid residues, it may be expedient to sequence the pentapeptide, the decapeptide and the pentadecapeptide to bs certain that the synthesis is going aocordfng to plan.
A variety of other chemical reaction syst~ems~can be employed to generate peptide ladders for analysis in ~5 accordance with this invention.
It will be recognized that there are a number of significant advantages to the processes of this invention. For example, the demands on yield of the chemical degradation reactions are much less stringent and more readily achieved than by wet chemical stepwise degradation techniques such as the Edman degradation is which low molecular weight derivatives are recover8d and analyzed at each chemical step. Other advantages WO 93/~483d ~ PCT/1iS93/tf~070 ' 41 include accuracy, speed, convenience, sample reco~rery, and the ability to recognize modifications in the peptide such as phosphorylation. Relatively unsophisticated and inexpensive mass spectrometric equipment, a.g. time of flight; single quadrupole; etc.
can be used.
By employing the process of this invention, it is routinely possible to sequence polypeptides containing or more amino acid residues from one picomole, yr 10 even a smaller amount of a polypeptide in one hour or less including cyclic degradation, mass spectrometry, and interpretation.
The processes described may be readily automated i.e., carried out for example in microtiter plates, using an x, y, z chemical robot. Furthermore, the determination of amino acid sequence from mass spectrometric data obtained from the protein sequencing ladders is readily carried out by simple computer algorithms. The process of the invention therefore includes coaputer read-out of the spectra of the peptide ladders prcducad.

WO 93/24834 ~ PCT/US93/05070 The skilled artisan will recognize that t~rete are some limitations to the procea of the invention as described above.
For example, some pairs of amino acids such as laucine and isoleucine have the same molecular weights.
Therefore, they can not be distinguish~d by mass differences of terminated polypeptides in a series:
There are several procedures for avoiding this difficulty. One is to differentiate them by CDNA
sequencing. They are highly degenerate codons, so they can be accommodated by inosine substitution in DNA
probes/primers for isolation/identification of the corresponding gene. This limitation will have little impact on practical application of the invention.
Further, several amino acids differ by only 1 emu.
This places stringent requirements on accuracy of mass determination. However, this invention utilizes a determination of mass differences between adjacent peaks, not a determination of absolute masses. Since mass differences can be determined with great accuracy by mass spectroscopy, the limitation will also be of little practical significance.

Wfl 93/24834 ~ ~ PCT/C~S93/05070 Finally, samples which are blocked at the amino or carboxy terminal may not be susceptible to the generation of peptide ladders. This problem can be circumvented by chemical or enzymatic fragmentation of the blocked polypeptide chain to yield unblocked segments which can be separately analyzed.
The following non-limiting examples are given by way of illustration only and are not to be considered a~
limitations of the invention many apparent variations of which may be made without departing from the spf~cit or scope thereof.
Exa~tple 1 Sectuencing of [Glul~Fibrino~tide B
[Glut]Fibrinopeptide B was purchased from Sigma Chemical Co. (St. Louis, Mo.). The reported sequence was: Glul-Gly-Val-Asn-Asp-Asn-aGlu-Glu fly-Phel~-Ph~e-Ser-Ala-Argl4. Matrix assisted laser dssorption mass spectrometry gave MW 1570.6 dalton (Calculated: 1570.8 dalton) and showed high purity of the starting peptiale.
A mixture of PITC plus 5% v/v phenylisocyanate PhC was us~d in the coupling step. PIC reacts with the NH~-of a polypeptide chain to yield an N~ -phenylcarba~nyl-peptide which is stable to tha conditions of the ~dman WO 93/~~B34 ~ F'(,'f/I~S93/0"5070 ' 4~
degradation. A modification of a standard manual ~dman degradation procedure (6) was used. All reactions wars carried out in the sameØ5mL polypropylene microfugs tube under a blanket of dry nitrogen. Peptide (200pmoles to 10 nmole) was dissolved in 20u1 of pyridine/water (i:iv/v; pH10.1); 20uL of coupling reagent containing PITC:PIC:pyridine:hexafluoroisopropanol (20:1:76:4 v/v) was added to the reaction vial. The coupling reaction was allowed to proceed at 50°C for 3 minutes. The coupling reagents and non-peptide coproducts ware extracted by addition of 300uL of heptane:ethyl acetate (l0:iv/v), gentle vortexing, followed by centrifugation to separate the phases. The upper phase was aspirated and discarded. This washing procedure was repeated once, followed by washing twice with hsptane:ethyl acetate (2:iv/v). The remaining solution containing the peptide products was dried on a vacuum centrifuge. The cleavage step was carried out by addition of 20uL of anhydrous trifluoroacstic acid to the dry residue in the reaction vial and reaction at 5o°C for 2 minutes, followed by drying on a vacuum centrifuge. Coupling-wash-cleavage steps were regeated for a predetermined number of cycles. The low MW ATZ/PT~i derivatives released at each cycle were not separated/analysed.
Finally, the total product mixture was subjset~ed to an WO 93f24834 ~ PCTfIdS93fDS9'f0 additional treatment with PIC to convert any remaining unblocked peptides to their phenyicarbamyl derivatives.
In this final step, the sample was dissolved in 20uL of trimsthylamine/water (25~wt/wt) in pyridine (i:lv/v);
5 20uL of PIC/pyridine/IiFIP (1:76:4v/v) was added to the reaction vial. The coupling reaction was carried out at 50°C for 5 min. The reagents were extracted as described above. After the last cycle of ladder generating chemistry, the product mixture was dissolved in O.i~ aqueous trifluoroacetic acid: acetonitrile (2:1, v/v). A 1uL aliquot ( 250pmol total peptide, assuming no losses) was mixed with 9uL of ~-cyano-4-hydroxy-cinnammic acid (5g/L in 0.1~ trifluoroacetic acid:
acetonitrile, 2:i v/v), and l.OuL of this mixture of ~5 total peptide products (25pmo1) and aatrix was applied to the probe tip and dried in a stream of air at room temperature. Mass spectra ware acquired in positive ion mode using a laser desorption time-of-flight instrument constructed at The Rockefellsr University (7). Ths 20 spectra resulting from 200 pulses at a wavelength of 355nm, 15 mJ per pulse, were acquired over 80 seconds and added to give a mass spectrum of the protein sequencing ladder shown in Fig. 7. Masses were calculated using matrix peaks of known mass as.
25 calibrants.

WO 93/t4834 , ~ PCT/U593/05090 fb Peptide seuuence read-out. Positive ion (MAhDMS) spectra of [Glut]Fibrinopeptide B is shown in Fig. 6. A
protonated molecular ion [M+H] was observed at m/z 1572.5 (calculated value is 1571.8).
Its positive ion MAhDMS spectrum of the reaction mixture obtained after seven cycles is shown in Fig. 6.
Each of the peaks in the spectrum represents a related phenylcarbamoylpeptide derivative in the peptide ladder (except a few peaks which will discussed later). The ~0 amino acid sequence can be easily read-out from the mass difference of adjacent two peaks. for instance, the mass difference are 129.1, 56.9, and 99.2 between peak$
at m/z 1690.9 and 1561.8, peaks at m/z 1561.8 and 1'504.9 and peaks at m/z 1504.9 and 1405.7. Which correspond to glutamic acid (ca. 129.12), glycine (ca. 57.05) and vaiine (ca. 99.13) residues, respectively. One set of paired peaks gives mass difference 119.0 (1062.1-943.1) which corresponds to the phenylcarbamoyl group. ~n other words, these two peaks represent one piece of Zp peptide with or without phenylcarbamoyl group. Peak at m/z 1553.8 corresponds partially blocked peptide with pyroglutamic acid at the N-terminus. This results from cyclization of the N-terminal Glu under the reaction WO 93124834 ~ ~ PCTfidS93f05070 ~7 conditions used, such products are readily identified from the accurately measured mass and know chemical reaction tendencies.
Example 2 Stepwise solid phase synthesis of the 99 amino acid residue polypeptide chain corresponding to the monomer of the HIV-1 protease (SF2 isolate):
PQITLWQRPLVTIRIGGQLKEALLDTGADDTVLHEMNLPGKWKPKMI~GGI"GGFIX9~t QYDQIPVEI(Aba)GHKAIGTVLVGPTPVNIIGRNLLTQIG(Aba)TLNF99 [where Aba = °(-amino-n-butyric acid] was undertaken.
Highly optimised Boc-chemistry instrument-assisted stepwise assembly of the protected peptide chain ~raa carried out on a resin support, according to the metfiod described by S.B.H. Kent (8). Samples (3-8mg, about i5 lumole each) were taken after each cycle of amino acid addition. The protected peptide-resin samples were mixed in three batches of consecutive samples: (number corresponds to the amino acid after which sample was taken, i.e. residue number in the target sequence.) 99-67; 66-33; 32-1: The first such mixture contained the peptises:

WO 93/24834 ~ ~ PCTIUS93/~'5070 ~l 8 99-Resin 98-99-Resin 97-98-99-Resin 96-97-98-99-Resin . . . . (etc. ) .
70: . . 96-97-98-99-Resin 69-70. . . . 96-97-98-99-Resin 68-69-70. . . . 96-97-98-99-Resin 67-68-69-70. . . . 96-97-98-99-Resin Similarly for the other two mixtures. The mixed batches of peptide-resin were deprotected and cleaved with IiF (i hours, at OoC, plus 5% cresol/5%/thiacresol). The products were precipitated with diethyl ether, dissolved in acetic acid-water 950/50%, v/v) and then lyophilized.
Each peptide mixture was dissolved in 0.1% TFA, 1 uL of the peptide mixture (10 uM per p8pt~d.ie component) was added to 9uL of 4-hydroxy- -cyanoci~nnamic acid in a 1:2 (v/v) ratio of 30% acetonitrile/0.1% aqueous trifluoroacetic acid. 0.5uL of the resulting mixture was applied to the mass spectrometer probe and inserted into the instrument (7). The spectra shown in Figs. 8 and 9 are the result of adding the data of each of 100 wo 9~m ~ ~ Pcriu5g3ro5o~o laser shots performed at a rats of 2.5 laser shots/second. Figure 8 shows the mass spectrum obtained from the mixture resulting from cleaving mixed sampl,as from residues 99-67 of the synthesis: Fig. 9 shows the mass spectrum obtained from the mixture resulting from cleaving mixed samples from residues 66-33 of the synthesis: Table 1 shows the measured mass differences between consecutive peaks of a selection of these peaks and compares them with the mass differences calculated from known sequences of the target peptides. The agreements are sufficiently close to allow confirmation of the correctness of the synthesis:
Figure 11 shows mass spectra of the mixture obtained from mixed samples from residues (66-33) of the synthesis.
The sequence of the assembled polypeptide chain can be read out in a straightforward fashion from the mass differences between consecutive peaks in the mass spectra of the peptide mixture. This confirmed the sequence of amino acids in the peptide chain actually synthesized. The identity of the amino acids as determined by such mass differences is shown in Table 1 WO 9314834 ~ ~ PCT/US93/050T0 Tabte 1. the identify of amino acid by the mass differences in protein ladder seqvwrx~cltg using matrix-assisted laser desotption mass spectrometry.
Amino Mass DifferenceDeviation Amino Mass Differencedeviation Aad Measured, Da) Acid (Meastued, ( Da) 5 Leu~ 113.3 . ~ 114.8 ~ ,0.3 ~ 0.1 Aso ~ ~

Glu" _ 129.7 ~ GIn6' t 28.7 ~ 0 6 ~ 0.6 ~

Glu' 129.5 ~ ~ 113.2 ~ 0.0 ~ - llea ~.4 ~

Mete 130.8 -0:4 ~ 97.0 ~ -0.1 ~ Prom ~

Asn~ t 15.0 ~ Val~' 99.4 ~ 0,3 ~ 0.9 Leu' 1124 ~ ~ 128:G ~ -0.5 ~ -0.8 Glum ~

Prop 97.9 ( ~~ 113.3 ~ 0.1 ~ 0.8 Item ~

Gly' 56.1 ~ j~ 84.9 ~ -0.2 ~ -0.9 Aba~' ~

~ys" - 1 ~.1 ~0.0 j~ 57.0 . ~ 0.0 ~ Gty ~

TrQ'~ 186.4 ~0.2 ~~ 137.3 ~ ~ 02 ~ His' , 15 ~Ys" ~ 128.2 ~ ~~ 127.8 ( X0.4' 0.0 Lys' ~

Pro" ' 97.1 ' ~~Ala" ~ 71.4 ~ 0.3 0.0 ~

I_ys' ~ 128.0 ( ~ Ilen ~ 113.4 , 0.2 -0.2 Met" ~ 131.9 ~0.7 G1y" Sfi.B ( -0.2 Ile" ' 112.6 -0.6 Thr" ; 101.1 ( 0.0 2 ' Gly'~ ( 57.9 ~ ~~Val's~ 99.2 ' ' 0.1 0 0.9 G1y' ~ 56.3 ~ -fl.7 ~ ~eu~ ~ 113.1 ~ -0.1 Ile' f 112.4 ~ -0.8 ~ Val'~~ 99.1 ~ 0 0 GIv3' ~ 57.6 ~ 0.6 ~ Gly' ~ 57.1 ~ 0.1 Glv~ ~ 57.5 ~ ~~Pro' 97.2 ~ d.1 0.5 2 Phe~ ~ 147.3 ~ 0.1 ~ Thr ~ 101.1 0.0 ' Ile' ' 112.5 ~ ' Pro' ~ 97.1 ~ 0:0 -fl.7 Lys~ ~ 128.9 ~ ~ ~ f 99.2 ~ 0.1 0.8 Val~ y Val~ ; 99.0 ( f ~ 113.8 ~ 0.3 -0.1 Asn" j Arg~' ! 156.2 ~ ~ ~ 113.4 ~ 0.2 0.0 tte~' Gin~ ~ ~ 28.4 - ' ~~ ~ 113.1 ~ 0:0 0.3 ilea T yrs' ; 1626 ~ f) ~ 57.1 ~ 0 O
-0.6 Glv VVO 93/2 ~ ~ PCT/tlS93/05070 In addition, terminated by-products (where the peptide chain has become blocked and doss not grow anymore) are present in every peptide-resin sample taken after the step in which the block occurred. Thus, there is an amplification factor equal to the number of resin samples in the batch after the point of termination.
This can be seen in Fig. l0 (samples #66-33) which contains a peak at 3339Ø This corresponds to the peptide 71-99, 3242.9 (N~terminal His71) plus 96.1 0 dalton. The characteristics mass, together with knowledge of the chemistry used in the synthesis identifies the blocking group as CF3C0-(97.1-H = 96.1 dalton). The observed by product is the trifluoroacetyl-peptide, N'°~-Tfa-(71-99). The ratio of the amount of this component to the average amount of the other components is about 2:1. There wets 34 samples combined in this sample. Thus, the terminated byproduct N°~-Tfa-(71-99) had occurred at a level of about 5mol~. This side reaction, specific to the N-terminal His-peptide chain, has not previously been reported. This illustrates the important sensitivity advantage provided by this amplification effect in detecting terminated pegtidss. Such byproducts are not readily detected by any other means.

WO 93/?4834 ~ ~ PCT/tiS93/05070 Exa~ple 3 Boc/Fmoc Terminations Synthesis of the peptide LRRAFGLIGNNPLMAR-amide was performed manually on a 0.2 mmol scale using p-methylbenzhydrylamine resin and 0.8 mmoles amino acid (95 mol%
N-°< -Boc, 5 mol% N-"< -FSnoc) according to the ink neutralization methods of Schnolzer et al (9). The following side chain protecting groups were used: 8oc-Arg, tosyl; Fmoc-Arg, 2,3,6-trimethyl-9-methoxybenzenesulfonyl (Mtr). Fmoc-Arg(Mtr) was used for its greater stability in trifluoroacetic acid (TFA).
After completion of the chain assembly, Fmoc groups were removed using 50% piperidine/DMF, followed by 8oc grog removal in TFA. The peptide fragments were then cleaved from the resin by treatment with HF-10% p-cresol (OoC, 1 hour). The resulting crude peptide products were precipitated and washed with ether, dissolved in 50%
acetic acid, diluted with water and lyophilized. The mass spectra of the reaction mixture thus produced ie shown in Fig. 11.

W4 93/24834 ~ ~ PCTlUS93/g5070 ' 53 Exal4ple 4 Post-ninhYdrin Experiment The machine-assisted assembly of the peptide LRRASGLIYNNPI~tAR-amide was performed according to the in situ neutralization methods of Schnolzer and Kent (9) on a 0.25 mmol scale using I~HA resin and 2.2 mmol N-~ -Boc amino acids . The following side chain protecting groups were used: Arg, tosyl; Asn, xanthyl; Ser, benzyl(Bzl); Tyr, bromobenzyloxycarbonyl(BrZ). Resin samples were collected at each step in the synthesis and each sample was individually subjected to the quantitative ninhydrin reaction. These samples were then pooled and the Boc groups removed in neat TFA. Cleavage of the peptide fragments from the resin was performed by treatment with HF-10~ p-cresol (0C, 1 hour). The resulting crudre peptide products were precipitated and Washed with ether, dissolved in 50~ acetic acid, diluted with water and lyophillized. The mass spectrum of the mixture is shown in Fig. 15:
CITATIONS
The following publications are referred to in this specifications. The complete disclosure of each of them is hereby incorporated by references.

WO 93/24x34 ~ ~ PCT/ C~S9310'5070 1. Aebersold et al, Protein Science 1, X94 (1992) 2. R: Self, A. Parents, Biomed. Mass_Sp~ectrom. 10, 78 (1983) 3. L.A. Smith, R.M. Caprioli, Biomed. Mass ~pectroW.
10, 98 (1983) 4. B.T. Chait, T. Chaudhary, F.H. Field, ~Methods in Protain sequence Analysis 1986, R.A. Walsh, ed., 8umana Press 1987, pp. 483-493, and uncontrolled chemical degradation 5. A. Tsugita, K. Takamoto, M. Kamo; H. Iwadate, Eur.
J. Biochem. 206, 691 (1992) 6. G.E. Tarr (1977), in Methods En2Ymoloqy 47, 355.

7. R.C. Beavis and B.T: Chait (1989), Rapid Commun.
Mass Spectrom. 3, 233.

8. S.B.H. Rent, Annual Rev. Hiochem. 5?, 957-984 (1988) 9. Schnolzer et al, Int. J. Peptide Protein Res. 40, 1992, 180-193 10. W.A. Schroeder, Meth. Enzymol. 25, 298 (1972)

Claims (13)

WHAT IS CLAIMED IS:

1. A method for identifying a covalent modification of an amino acid residue in a polypeptide chain comprising:
(1) detecting a mass difference between a formed polypeptide and a modified polypeptide by mass spectrometry, wherein the modified polypeptide comprises a covalent modification of an amino acid residue in the formed polypeptide, and (2) identifying the covalent modification from the mass difference.

2. The method of claim 1 wherein the covalent modification is phosphorylation.

3. The method of claim 1 wherein the covalent modification is acetylation.

4. The method of claim 1 wherein the covalent modification is glycosylation.

5. The method of claim 1 wherein the covalent modification is a disulfide bond.

6. The method of claim 1 further comprising:
(i) producing reaction mixtures from the formed and the modified polypeptide, each reaction mixture containing a peptide ladder comprising a series of adjacent polypeptides in which each member of the series differs from the next adjacent member by one amino acid residue;
(ii) determining the differences in molecular mass between adjacent members of each series by mass spectroscopy, said differences coupled with the positions of said adjacent members in the respective series being indicative of the identity and position of the amino acid residue in the formed or modified polypeptide;
(iii) identifying an amino acid residue in the modified polypeptide that comprises a covalent modification wherein the amino acid residue is identified by having an additional mass difference compared with the corresponding amino acid residue in the formed polypeptide.

7. The method of claim 6 wherein producing the peptide ladders comprises:

(1) reacting the formed and modified polypeptide with a molar excess of a pair of reagents comprising a coupling reagent and a terminating reagent each of which forms a reaction product with a terminal amino acid residue of the formed or modified polypeptide to be analyzed under a first reaction condition; the reaction product generated between the terminating reagent and the terminal amino acid residue of the formed or modified polypeptide being stable under all subsequent reaction conditions; the reaction product generated between the coupling reagent and terminal amino acid residue of the polypeptide to be analyzed being removable as a cleavage product from the original formed or modified polypeptide under a second reaction condition;
(2) changing the reaction conditions so that the cleavage product separates, thereby to form a reaction mixture comprising:
i. unreacted coupling and terminating reagents, ii. a first reaction product which is the reaction product between the original formed or modified polypeptide and the terminating reagent, iii. a newly formed polypeptide from which the terminal amino acid residue has been removed;
(3) repeating steps (1) and (2) any selected number of cycles thereby to form a final mixture which comprises:
reaction product between the original formed or modified polypeptide and the terminating reagent, ii, a peptide ladder which is series of adjacent reaction products which is formed by reaction between the terminating reagent and the terminal amino acid residue of a fraction of the newly generated polypeptide of each cycle, and (4) determining the differences in molecular mass between adjacent members of the series of reaction products by mass spectroscopy, said differences being equal to the molecular mass of the amino acid residue cleaved from the original formed or modified polypeptide and from each subsequent formed or modified polypeptide of the series, said differences coupled with the positions of said adjacent members in the mass spectrum being indicative of the identity and position of that amino acid residue in the original formed or modified polypeptide.

8. The method of claim 7 wherein the coupling agent is PITC, the terminating agent is PIC, the first set of reaction conditions comprises basic conditions and the second set of conditions comprises acidic conditions.

9. The method of any of claims 1, 2, 3, 4, 5, 6, 7 or 8 wherein the formed polypeptide and the modified polypeptide are analyzed simultaneously in a mixture.

10. The method of claims 1, 2, 3, 4, 5, 6, 7, or 8, wherein said laser desorption mass spectrometry comprises:
a) applying the formed polypeptide or the modified polypeptide to a matrix, said matrix comprising a-cyano-4-hydroxy-cinnammic acid; and b) applying laser light to irradiate the matrix at a wavelength sufficient to desorb the sample from the matrix.

11. The method of claim 10 wherein said laser light is at a wavelength of about 355 nm.

12. The method of claim 10, wherein the formed polypeptide and the modified polypeptide are analyzed simultaneously in a mixture.

13. The method of claim 1 where in the mass spectroscopy is by laser desorption mass spectroscopy.