BIOLOGICAL MASS SPECTROMETRY, VOL. 20, 345-350 (1991)
Sequencing of Peptides Containing Alanine, Asparagine, Histidine, Isoleucine and Tryptophan by Partial Methanolysis and Fast Atom Bombardment Mass-Spectrometry Salvatore Foti Dipanimento di Scienze Chimiche dell’hiversita, V. Ie A. Dona 8, 95125 Catania, Italy
Rosaria Saletti Istituto per la Chimica e la Technologia dei Materiali Polimerici del C.N.R., V. le A. Doria 8,95125 Catania, Italy
Five peptides containing the amino acids alanine, asparagine, histidine, isoleucine and tryptophan were investigated by partial methanolysis and fast atom bombardment mass spectrometry in order to examine the behaviour of tbese amino acid residues under the conditions employed in the methanolytic step. The results obtained confirm that partial methanolysis prior to mass analysis increases considerably the content of sequence information in the mass spectra and that no secondary reactions occur in the residues of the amino acids now investigated, with tbe excep tion of the esterification of the glutamic acid carboxyl group and partial conversion of the asparagine amide group in the corresponding methyl ester.
INTRODUCTION The introduction of fast atom bombardment (FAB) as an ionization method for large, polar and/or involatile molecules has been successful in producing long-lasting mass spectra of underivatized peptides.’ The MH+ ions originated from peptides in the FAB process possess very little excess energy and are evenelectron ions, so that they are stable and undergo very little fragmentation. As a consequence, the FAB mass spectra of peptides are characterized by intense MH+ ion peaks, from which the determination of the molecular mass of the peptide can be readily accomplished. The high intensity of the MH’ ion peaks is also of advantage when mixtures are to be analysed, since in that case individual components of the mixture can be easily recognized, provided that they are all extracted from the matrix. Conversely, fragment ions carrying sequence information are usually very weak in the FAB mass spectra. Moreover, the fragmentation of the MH+ ions is at present unpredictable since the sites of fragmentation along the peptide backbone are variable from one peptide to another, and even within one peptide some components of one series of peaks can be clearly distinguishable and others absent. This prevents the formation of extended series of peaks in which the difference from one peak to the subsequent represents the mass of the consecutive amino acid residue. A further factor that adversely affects the reliable interpretation of FAB mass spectra of peptides is the interference of the peaks originated by the matrix. In practice, due to the cumulative weight of these contributions, extraction of the sequence from the FAB lOS2-9306/91/06034S6$05.00
0 1991 by John Wiley & Sons, Ltd.
mass spectrum of a peptide is difficult, especially when one is dealing with completely unknown peptides, and normally only results in a partial sequence’s2 To overcome the difficulties encountered in sequencing peptides by FAB mass spectrometry, several strategies based either on tandem mass spectrometry (MS/MS)3-8 or on chemical or enzymaticg-” degradation of the peptide have been reported. Partial methanolysis of relatively low-molecularweight peptides and FAB mass spectrometry of the resulting mixture has proved to be effective in increasing the content of structure information of the FAB mass spectra.l4 The data collected, however, have also shown that, under the conditions employed for methanolysis, secondary reactions can occur on some amino acid residues.l4 Since the peptides previously investigated contained only a few among the common protein amino acids, we have now studied five further low-molecular-weight peptides (Table 1) containing some amino acids not previously included, i.e. alanine, asparagine, histidine, isoleucine and tryptophan, in order to examine the behaviour of these residues under conditions of methanolysis and to extend further the data already obtained.
EXPERIMENTAL The peptides used were purchased from Sigma (St Louis, Missouri, USA). All the other chemicals were of the highest purity commercially available and were used without further purification unless otherwise stated. Received 3 May 1990 Revised 29 January I991
S. FOTI AND R. SALETTI
346
residue are indicated as CEl, CE2, ... CEn, if they contain the C-terminal amino acid, and as NEl, NE2, ... NEn, if they contain the N-terminal amino acid. These fragments are referred to as primary methanolytic fragments. Further methanolysis of these primary fragments would result in the formation of fragments which do not contain either the C-terminal or the N-terminal amino acid of the original sequence. These fragments would correspond formally to cleavage of two peptide bonds in the original sequence and are indicated here as secondary methanolytic fragments.
Table 1. Model peptides subjected to partial methanolysis
3 4
5
Amino acid sequlnce
Mol. W
Peptide
1 2
Angiotensinogen frag. 1-4 Human fibrin &chain frag. 1-4 Rat prothrombin precursor seq. 5-9 DNA-binding peptide
406 481
VAL-AM-AM-PHE VAL-ILE-H IS-ASN
465
GLY-HIS-ARG-PRO
649
PHE-LEU-GLU-G
460
LYS-TRP-LYS
LU-I
LE
Partial methanolysisof peptides 1-5
RESULTS
A 5 N solution of HCl in dry methanol was prepared by bubbling gaseous HCl through redistilled anhydrous methanol. The amount of HCl dissolved was determined by weighing the solution. The concentration was then adjusted by adding an appropriate volume of methanol. About 0.5 mg of each peptide was placed in a quartz vial and 150 pl of a freshly prepared 5 N solution of HCl in dry methanol was added. The vial was firmly stoppered and the mixture was allowed to react at 37 "C for 6 h, after which the reaction was stopped by cooling the solution in liquid nitrogen. The solvent was then removed by vacuum evaporation. The residue was taken up with methanol and about one-tenth of the solution was transferred to the FAB probe.
FAB mass spectra were recorded on a Kratos MS-50 instrument fitted with a standard FAB source. The samples (about 5-10 nmol for peptides 1-5) were deposited by evaporation from methanol solutions onto a copper probe tip and about 5 pl of glycerol were added. A beam of 7-9 KeV of xenon atoms, generated by an Ion Tech neutral gun, was impacted onto the sample. The source operating pressure was typically 10- torr. Spectra were recorded on ultraviolet (UV)-sensitive paper and calibrated manually using glycerol clusters as reference peaks.
The sequence ions appearing in the FAB mass spectra of the peptides investigated (Table 1) are reported in Table 2. The fragment ions are designated following Biemann's modification* of the Roepstorff and Fohlman notation." Only fragment ions that can be differentiated from the matrix are included in Table 2. In addition to the fragment ions collected in Table 2, the FAB spectra of peptides 1-5 generally contained peaks at MH+ - 15 and, with the exception of peptide 5, at MH' - 44.Sodium-cationized molecular ions are also observed in peptides 1,2 and 4. Immonium ions of some, but not all, of the constituent amino acids are generally found in the low-mass region. Examination of the fragment ions reported in Table 2 shows that, among the C-terminal fragments, y ions are frequently the more distinguishable, whereas x and z ions are practically absent. Among N-terminal fragments, a and b ions are more often encountered; c ions are found with sizeable intensity in peptides 2 and 3. The random formation and the absence of complete series of fragment ions make the interpretation of the sequence problematic. Furthermore the structural significance of the data in Table 2 is lowered by the interference of the matrix peaks or isobaric fragments. The FAB spectra of the methanolytic mixtures of peptides 1-5 are shown in Figs 2-6. Peaks marked with an asterisk are due to a matrix contribution. Details of the spectra are given below.
Nomenclature for the methanolytic fragments
Peptide 1
The nomenclature for designating the methanolytic fragments is given in Fig. 1. Amino acid residues are numbered starting from the N-terminus. Methyl ester fragments produced by cleavage at the carboxylic end of the first, second, ... nth amino acid
The MH' ion peak of the esterified peptide is present with high relative intensity at m/z 421 in the FAB spectrum of the mixture of partial methanolysis of 1 (Fig. 2). The three MH' signals for the CE-type methanolytic
FAB mass spectra
Rl
I
CE 1
C E2
I
I
11
R2
O 1 H~-CH-~NH-CH-]--
I
CEn I Rn+l
Rn
O II
I
I
I
NH-CH-JNH
-CH
-c
J0 \ OCH,
NE 1
NE2
NEn
Figure 1. Nomenclature for rnethanolytic fragments.
SEQUENCING OF PEPTIDES BY PARTIAL METHANOLYSIS AND FABMS
347
Table 2. Relative abundaoce of sequence ions in the FAB spectra of peptides 1-5‘ 1
MH+ a1
bl C1 X1
Y1 2 1
82 b2 c2 x2
Y2 22
a3
b3 c3 x3 Y3 23
2
407 (100) 72 (336)b 100 (-) 117 (-) 192 (-) 166 (40) 149 (24)b 143 (29) 171 (56) 188 (-) 263 (-) 237 (63) 220 (-) 214 (-) 242 (28) 259 (11) 334 (-) 308 (11) 291 (-)
3
482 (100) 72 (115)b 100 (-) 117 (-) 159 (-) 133 (-) 116 (-) 185 (420)b 213 (-) 230 (8) 296 (-) 270 (26) 253 (-) 322 (-) 350 (7) 367 (12) 409 (-) 383 (-) 366 (-)
a4
be c4 x4 Y4 24 a
6
4
466 (100) 30 (-) 58 (-) 75 (840)b 142 (-) 116 (-) 99 (-1 167 (56)b 195 (-) 212 (-1 298 (-) 272 (-) 255 (-) 323 (-) 351 (4)b 368 (-) 435 (-) 409 (2) 392 (-)
650 (100) 120 (66)b 148 (-) 165 (-) 158 (-) 132 (14) 115 (-) 233 (19) 261 (22)b 278 (-) 287 (-) 261 (22)b 244 (-) 362 (-) 390 (ll)b 407 (5) 416 (-) 390 (ll)b 373 (-) 491 (-) 519 (3) 536 (3) 529 (-) 503 (3) 486 (-)
461 (100) 101 (-)
129 (-) 146 (-) 173 (-) 147 (-) 130 (-) 287 (-) 315 (-) 332 (-) 359 (-) 333 (16) 316 (-)
Relative intensities (in parentheses) are calculated assuming the intensity of MH’ as
100;values are reported only for fragment ions distinguishablefrom the background. Possible contribution from another isobaric fragment.
40iI 203
loo
150
1.,,,i., II1
260
“CL
, I
250
322
132
I
251
203
I
I
3L
3dO
4bO
* /r
180
214
Figure 2. FAB mass spectrum of mixture of partial methanolysis of peptide 1.
fragments are clearly distinguishable, the most intense of them (m/z 180 and 251) being originated by cleavage at the carboxylic group of alanine. NE-type methanolytic fragments produce less intense signals at m/z 274 (NaY and 203 (NE2). A third signal (NEl, m/z 132) cannot be itrerentiated from the background.
A close inspection of the spectrum reveals three distinct signals at m/z 242, 171 and 143, which could be attributed to the b , , b, and a, FAB-induced fragments of the molecular ion, respectively. A metastable peak corresponding to the transition 171 to 143 is also observed in the spectrum.
S. FOTI AND R. SALETTI
348 CE3
CE2
CEl
MH+
511
V A EL
Sequencing of Peptides Containing Alanine, Asparagine, Histidine, Isoleucine and Tryptophan by Partial Methanolysis and Fast Atom Bombardment Mass-Spectrometry Salvatore Foti Dipanimento di Scienze Chimiche dell’hiversita, V. Ie A. Dona 8, 95125 Catania, Italy
Rosaria Saletti Istituto per la Chimica e la Technologia dei Materiali Polimerici del C.N.R., V. le A. Doria 8,95125 Catania, Italy
Five peptides containing the amino acids alanine, asparagine, histidine, isoleucine and tryptophan were investigated by partial methanolysis and fast atom bombardment mass spectrometry in order to examine the behaviour of tbese amino acid residues under the conditions employed in the methanolytic step. The results obtained confirm that partial methanolysis prior to mass analysis increases considerably the content of sequence information in the mass spectra and that no secondary reactions occur in the residues of the amino acids now investigated, with tbe excep tion of the esterification of the glutamic acid carboxyl group and partial conversion of the asparagine amide group in the corresponding methyl ester.
INTRODUCTION The introduction of fast atom bombardment (FAB) as an ionization method for large, polar and/or involatile molecules has been successful in producing long-lasting mass spectra of underivatized peptides.’ The MH+ ions originated from peptides in the FAB process possess very little excess energy and are evenelectron ions, so that they are stable and undergo very little fragmentation. As a consequence, the FAB mass spectra of peptides are characterized by intense MH+ ion peaks, from which the determination of the molecular mass of the peptide can be readily accomplished. The high intensity of the MH’ ion peaks is also of advantage when mixtures are to be analysed, since in that case individual components of the mixture can be easily recognized, provided that they are all extracted from the matrix. Conversely, fragment ions carrying sequence information are usually very weak in the FAB mass spectra. Moreover, the fragmentation of the MH+ ions is at present unpredictable since the sites of fragmentation along the peptide backbone are variable from one peptide to another, and even within one peptide some components of one series of peaks can be clearly distinguishable and others absent. This prevents the formation of extended series of peaks in which the difference from one peak to the subsequent represents the mass of the consecutive amino acid residue. A further factor that adversely affects the reliable interpretation of FAB mass spectra of peptides is the interference of the peaks originated by the matrix. In practice, due to the cumulative weight of these contributions, extraction of the sequence from the FAB lOS2-9306/91/06034S6$05.00
0 1991 by John Wiley & Sons, Ltd.
mass spectrum of a peptide is difficult, especially when one is dealing with completely unknown peptides, and normally only results in a partial sequence’s2 To overcome the difficulties encountered in sequencing peptides by FAB mass spectrometry, several strategies based either on tandem mass spectrometry (MS/MS)3-8 or on chemical or enzymaticg-” degradation of the peptide have been reported. Partial methanolysis of relatively low-molecularweight peptides and FAB mass spectrometry of the resulting mixture has proved to be effective in increasing the content of structure information of the FAB mass spectra.l4 The data collected, however, have also shown that, under the conditions employed for methanolysis, secondary reactions can occur on some amino acid residues.l4 Since the peptides previously investigated contained only a few among the common protein amino acids, we have now studied five further low-molecular-weight peptides (Table 1) containing some amino acids not previously included, i.e. alanine, asparagine, histidine, isoleucine and tryptophan, in order to examine the behaviour of these residues under conditions of methanolysis and to extend further the data already obtained.
EXPERIMENTAL The peptides used were purchased from Sigma (St Louis, Missouri, USA). All the other chemicals were of the highest purity commercially available and were used without further purification unless otherwise stated. Received 3 May 1990 Revised 29 January I991
S. FOTI AND R. SALETTI
346
residue are indicated as CEl, CE2, ... CEn, if they contain the C-terminal amino acid, and as NEl, NE2, ... NEn, if they contain the N-terminal amino acid. These fragments are referred to as primary methanolytic fragments. Further methanolysis of these primary fragments would result in the formation of fragments which do not contain either the C-terminal or the N-terminal amino acid of the original sequence. These fragments would correspond formally to cleavage of two peptide bonds in the original sequence and are indicated here as secondary methanolytic fragments.
Table 1. Model peptides subjected to partial methanolysis
3 4
5
Amino acid sequlnce
Mol. W
Peptide
1 2
Angiotensinogen frag. 1-4 Human fibrin &chain frag. 1-4 Rat prothrombin precursor seq. 5-9 DNA-binding peptide
406 481
VAL-AM-AM-PHE VAL-ILE-H IS-ASN
465
GLY-HIS-ARG-PRO
649
PHE-LEU-GLU-G
460
LYS-TRP-LYS
LU-I
LE
Partial methanolysisof peptides 1-5
RESULTS
A 5 N solution of HCl in dry methanol was prepared by bubbling gaseous HCl through redistilled anhydrous methanol. The amount of HCl dissolved was determined by weighing the solution. The concentration was then adjusted by adding an appropriate volume of methanol. About 0.5 mg of each peptide was placed in a quartz vial and 150 pl of a freshly prepared 5 N solution of HCl in dry methanol was added. The vial was firmly stoppered and the mixture was allowed to react at 37 "C for 6 h, after which the reaction was stopped by cooling the solution in liquid nitrogen. The solvent was then removed by vacuum evaporation. The residue was taken up with methanol and about one-tenth of the solution was transferred to the FAB probe.
FAB mass spectra were recorded on a Kratos MS-50 instrument fitted with a standard FAB source. The samples (about 5-10 nmol for peptides 1-5) were deposited by evaporation from methanol solutions onto a copper probe tip and about 5 pl of glycerol were added. A beam of 7-9 KeV of xenon atoms, generated by an Ion Tech neutral gun, was impacted onto the sample. The source operating pressure was typically 10- torr. Spectra were recorded on ultraviolet (UV)-sensitive paper and calibrated manually using glycerol clusters as reference peaks.
The sequence ions appearing in the FAB mass spectra of the peptides investigated (Table 1) are reported in Table 2. The fragment ions are designated following Biemann's modification* of the Roepstorff and Fohlman notation." Only fragment ions that can be differentiated from the matrix are included in Table 2. In addition to the fragment ions collected in Table 2, the FAB spectra of peptides 1-5 generally contained peaks at MH+ - 15 and, with the exception of peptide 5, at MH' - 44.Sodium-cationized molecular ions are also observed in peptides 1,2 and 4. Immonium ions of some, but not all, of the constituent amino acids are generally found in the low-mass region. Examination of the fragment ions reported in Table 2 shows that, among the C-terminal fragments, y ions are frequently the more distinguishable, whereas x and z ions are practically absent. Among N-terminal fragments, a and b ions are more often encountered; c ions are found with sizeable intensity in peptides 2 and 3. The random formation and the absence of complete series of fragment ions make the interpretation of the sequence problematic. Furthermore the structural significance of the data in Table 2 is lowered by the interference of the matrix peaks or isobaric fragments. The FAB spectra of the methanolytic mixtures of peptides 1-5 are shown in Figs 2-6. Peaks marked with an asterisk are due to a matrix contribution. Details of the spectra are given below.
Nomenclature for the methanolytic fragments
Peptide 1
The nomenclature for designating the methanolytic fragments is given in Fig. 1. Amino acid residues are numbered starting from the N-terminus. Methyl ester fragments produced by cleavage at the carboxylic end of the first, second, ... nth amino acid
The MH' ion peak of the esterified peptide is present with high relative intensity at m/z 421 in the FAB spectrum of the mixture of partial methanolysis of 1 (Fig. 2). The three MH' signals for the CE-type methanolytic
FAB mass spectra
Rl
I
CE 1
C E2
I
I
11
R2
O 1 H~-CH-~NH-CH-]--
I
CEn I Rn+l
Rn
O II
I
I
I
NH-CH-JNH
-CH
-c
J0 \ OCH,
NE 1
NE2
NEn
Figure 1. Nomenclature for rnethanolytic fragments.
SEQUENCING OF PEPTIDES BY PARTIAL METHANOLYSIS AND FABMS
347
Table 2. Relative abundaoce of sequence ions in the FAB spectra of peptides 1-5‘ 1
MH+ a1
bl C1 X1
Y1 2 1
82 b2 c2 x2
Y2 22
a3
b3 c3 x3 Y3 23
2
407 (100) 72 (336)b 100 (-) 117 (-) 192 (-) 166 (40) 149 (24)b 143 (29) 171 (56) 188 (-) 263 (-) 237 (63) 220 (-) 214 (-) 242 (28) 259 (11) 334 (-) 308 (11) 291 (-)
3
482 (100) 72 (115)b 100 (-) 117 (-) 159 (-) 133 (-) 116 (-) 185 (420)b 213 (-) 230 (8) 296 (-) 270 (26) 253 (-) 322 (-) 350 (7) 367 (12) 409 (-) 383 (-) 366 (-)
a4
be c4 x4 Y4 24 a
6
4
466 (100) 30 (-) 58 (-) 75 (840)b 142 (-) 116 (-) 99 (-1 167 (56)b 195 (-) 212 (-1 298 (-) 272 (-) 255 (-) 323 (-) 351 (4)b 368 (-) 435 (-) 409 (2) 392 (-)
650 (100) 120 (66)b 148 (-) 165 (-) 158 (-) 132 (14) 115 (-) 233 (19) 261 (22)b 278 (-) 287 (-) 261 (22)b 244 (-) 362 (-) 390 (ll)b 407 (5) 416 (-) 390 (ll)b 373 (-) 491 (-) 519 (3) 536 (3) 529 (-) 503 (3) 486 (-)
461 (100) 101 (-)
129 (-) 146 (-) 173 (-) 147 (-) 130 (-) 287 (-) 315 (-) 332 (-) 359 (-) 333 (16) 316 (-)
Relative intensities (in parentheses) are calculated assuming the intensity of MH’ as
100;values are reported only for fragment ions distinguishablefrom the background. Possible contribution from another isobaric fragment.
40iI 203
loo
150
1.,,,i., II1
260
“CL
, I
250
322
132
I
251
203
I
I
3L
3dO
4bO
* /r
180
214
Figure 2. FAB mass spectrum of mixture of partial methanolysis of peptide 1.
fragments are clearly distinguishable, the most intense of them (m/z 180 and 251) being originated by cleavage at the carboxylic group of alanine. NE-type methanolytic fragments produce less intense signals at m/z 274 (NaY and 203 (NE2). A third signal (NEl, m/z 132) cannot be itrerentiated from the background.
A close inspection of the spectrum reveals three distinct signals at m/z 242, 171 and 143, which could be attributed to the b , , b, and a, FAB-induced fragments of the molecular ion, respectively. A metastable peak corresponding to the transition 171 to 143 is also observed in the spectrum.
S. FOTI AND R. SALETTI
348 CE3
CE2
CEl
MH+
511
V A EL
