An EThcD-based approach for the reliable discrimination of isomeric leucine and isoleucine residues in peptide de novo sequencing procedure has been proposed. A multistage fragmentation of peptide ions was performed with Orbitrap Elite mass spectrometer in electrospray ionization mode. At the first stage, z-ions were produced by ETD or ETcaD fragmentation of doubly or triply charged peptide precursor ions. These primary ions were further fragmented by HCD with broad-band ion isolation, and the resulting w-ions showed different mass for leucine and isoleucine residues. The procedure did not require manual isolation of specific z-ions prior to HCD stage. Forty-three tryptic peptides (3 to 27 residues) obtained by trypsinolysis of human serum albumin (HSA) and gp188 protein were analyzed. To demonstrate a proper solution for radical site migration problem, three non-tryptic peptides were also analyzed. A total of 93 leucine and isoleucine residues were considered and 83 of them were correctly identified. The developed approach can be a reasonable substitution for additional Edman degradation procedure, which is still used in peptide sequencing for leucine and isoleucine discrimination.
In 1990s, mass spectrometry (MS) established itself as a preferential analytical tool for peptide sequencing, slowly replacing Edman degradation, the most reliable approach in earlier times [1, 2]. MS is more sensitive, faster, cheaper, and allows working with longer and/or modified peptides as well as their mixtures. The only issue that still often requires application of Edman degradation is discrimination of isomeric leucine and isoleucine residues. They have the same exact mass and cannot be distinguished even by high resolution mass spectrometers in MS2 mode. Quite often they are marked in new peptide sequences with L/I or X signs. There are no reliable data on the notably different biological activity of the similar peptides with only leucine/isoleucine substitutions. Our earlier 2D approach for prediction of possible activity types of frog peptides  places peptides with leucine or isoleucine into the same position on the map. Nevertheless, accurate sequencing is rather important for description of proteome (peptidome) of any organism. Thus, additional experiments are always required to resolve this problem.
Radical cations are usually less stable than even-electron cations and easily fragment further due to processes induced by the radical site . Such fragmentation of odd-electron a•- and z•- peptide ions leads to formation of d- and w-secondary ions, respectively [5, 6]; z•-ion with leucine on its N-terminus is transformed to w-ion by isopropyl radical loss, whereas z•-ion with isoleucine loses ethyl, and, to smaller extent, methyl radicals (according to the maximal alkyl loss rule ). Thus, w-ions formed by leucine and isoleucine side chains fragmentation have different masses, allowing reliable discrimination of these residues.
During the last three decades, a number of approaches were proposed for mass spectrometry-based discrimination of leucine and isoleucine residues [5–17]. Most of them are based on the observation of corresponding d- or w-ions in the spectra, although negative ion mass spectrometry  or an approach with copper complexes  may also be quite efficient. Alternatively, MSn experiments are used to obtain the required information out of immonium leucine/isoleucine ions [9, 10]. The benefits and drawbacks of these methods were summarized in our recent publication . In the same study, we have first proposed an MS3 method for leucine/isoleucine discrimination based on the following sequence: (a) isolation of a multiprotonated peptide molecule in electrospray ionization (ESI) spectrum and its fragmentation by electron transfer dissociation (ETD), (b) isolation of characteristic z•-ions with leucine or isoleucine at the N-terminus and their fragmentation by higher-energy collision dissociation (HCD), (c) searching for the corresponding w-ions in the resulting spectra. The method was tested on natural peptides isolated from skin secretion of Rana ridibunda frog. All of the 22 analyzed leucine/isoleucine residues were successfully determined. Noteworthy, the method solved a problem of radical site migration [19–21], which arises when leucine and isoleucine are placed in neighboring positions.
The main drawback of this method is that both fragmentation steps require manual isolation of fragmenting ions. The precursor ions of protonated peptides can be easily found in full ESI mass spectra, since they are usually the most intense ones. Nevertheless, the isolation of the targeted z-ions for further HCD fragmentation requires time-consuming manual analysis of ETD spectra.
Xiao et al.  used a combination of multistage mass spectrometric analysis (HCD-MSn) and ETD-HCD MS3 analysis with an Orbitrap Fusion mass spectrometer to efficiently and reliably distinguish leucine and isoleucine in proteins and peptides. Each ion must be isolated separately and fragmented in MSn mode (n = 2–5) either using ETD-HCD MS3 approach  or, if it fails, producing the corresponding immonium ion and triggering its fragmentation. A similar targeted HCD MS3 method was developed in  in combination with nanoflow LC. The proposed (HCD-MSn) approach is based on decision-making guideline, which is suitable when peptide sequence is already known (except for leucine/isoleucine ambiguities), but not for de novo peptide sequencing. Besides, MS4 and MS5 experiments may face sensitivity problem when dealing with real samples.
In the present study, we have improved ETD-HCD MS3 method , so that the isolation of z-ions is not required anymore. The procedure involves EThcD approach introduced by Frese et al.  instead of ETD-HCD used in previous studies [11, 18]. Since simultaneous HCD fragmentation of all ETD products is not possible with Orbitrap Elite, we had to split the entire mass range into several pieces, so-called “broadband windows.” The applicability of the proposed approach was checked using the doubly or triply charged precursor ions of the tryptic peptides from the digests of HSA and gp188. Leucine and isoleucine residues in 43 detected tryptic peptides were analyzed in this way. To estimate the possibility of the radical site migration, we also analyzed Brevinin 2Ec, a natural frog skin secretion peptide , and three synthetic peptides with adjacent positions of leucine and isoleucine residues.
The developed method allows reliable distinguishing between leucine and isoleucine residues without Edman degradation or other chemistry-based procedures and may be incorporated into the automatic MS sequencing engines.
Acetonitrile (HPLC gradient grade) was from Sigma-Aldrich (St Louis, Missouri, USA), formic acid (puriss. p.a., for HPLC) was from Fluka (Buchs, Switzerland), trypsin (modified, proteomics grade), iodoacetamide (IAA), and 2,2,2-trifluoroethanol (TFE) (Reagent Plus) were from Sigma-Aldrich (St.Louis, Missouri, USA). Ammonium bicarbonate (Ultra) and tris(carboxyethyl)phosphine (TCEP) were purchased from Fluka (Stenheim, Germany). Hydrochloric acid (purum) was from Chimmed (Moscow, Russia). Ultrapure water was prepared by Milli-Q water purification system (Millipore, Billerica, Massachusetts, USA).
Human serum albumin (HSA) was kindly provided by Professor Valery Shevchenko (N.N. Blokhin Cancer Research Center). The protein gp188 which corresponds to the recombinant endolysin  was kindly provided by Dr. Lidia Kurochkina (Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences). Both proteins were digested by trypsin. The theoretical set of trypsinolysis products was calculated using the website http://web.expasy.org/peptide_mass/.
Aliquots of 1 mg protein were taken from the protein solutions (10 mg/mL of HSA and 8.2 mg/mL of gp188), then 25 μL TFE + 25 μL 100 mM NH4HCO3 + 10 μ: of fresh 50 mM TCEP were added. This reaction mixture was incubated for 1 h at 60°C, and then cooled to 25°C. Five μL of fresh 84 mM aqueous IAA was added and mixed for 30 min at 25°C in the dark. Then 100 μL 100 mM NH4HCO3 + 300 μL H2O + trypsin solution in 1 mM of HCl (trypsin concentration 1 μg/μL, trypsin:protein ratio 1:100 by weight) were added and incubated for 18 h at 37°C. All studied tryptic peptides are listed in Tables 1 and 2.
Brevinin 2Ec (GILLDKLKNFAKTAGKGVLQSLLNTASCKLSGQC, monoisotopic mass 3516.9161) was isolated from Rana ridibunda frog skin secretion and purified by HPLC as described earlier . A synthetic peptide DILFLSDPDAR-NH 2 (monoisotopic mass 1259.6510) was kindly provided by Dr. Ludmila Alekseeva (Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences). A synthetic peptide Ac-IVGKGIVGVK-NH 2 (monoisotopic mass 1009.6649) was kindly provided by Dr. Irina Tarasova (V.L. Talrose Institute for Energy Problems of Chemical Physics, Russian Academy of Sciences).
Mass Spectrometric Analysis
Non-tryptic peptides were dissolved at 0.01 mg/mL in acetonitrile:water mixture (1:1) containing 1% formic acid. Tryptic hydrolyzates were diluted 1:100 with the same mixture. No chromatographic or any other separation of hydrolyzates was performed. Experiments were carried out with Orbitrap Elite mass spectrometer (Thermo Fisher Scientific GmbH, Bremen, Germany) with electrospray ionization source. Sheath gas flow rate was from 10 to 25 arbitrary units, auxiliary and sweep gas flow rate was set to zero, spray voltage 3.5 kV, capillary temperature 275°C, S-Lens rf Level 60%. Samples were introduced into the ion source using a syringe pump at 3 μL/min.
Mass spectra were acquired in a positive ionization mode. All masses were analyzed in Orbitrap mass analyzer with 120,000 resolving power. For ion trap, acquisition of 1e5 ions was allowed with maximum injection time of 500 ms in MSn mode (for ETD). For Orbitrap mass analyzer (FT), acquisition of 1e6 (full MS) or 1e5 (MSn, HCD) ions was allowed with maximum injection time of 1000 ms. For ETD reagent, AGC target of 6e5 was used with maximum injection time of 100 ms. All spectra were recorded with one microscan and registered continuously during 1 min, then averaged.
Leucine and Isoleucine Discrimination Procedure
Full mass spectra of tryptic hydrolyzates or non-tryptic peptide samples were recorded in the range 150–2000 Da. For every peptide analyzed, a peak of its triply charged ion was searched in the spectrum. If its intensity was adequate (104 or more), the ion was isolated (isolation width 4 Da) and fragmented by electron transfer dissociation with activation time 100 ms. If targeted z-ions were not found in ETD spectra of triply charged peptide ions, a supplemental collisional activation (ETcaD) fragmentation was performed with supplemental activation energy (SAE) of 25 .
If intensity of a peak of triply protonated peptide was below 104, a doubly charged ion of this peptide was isolated and fragmented by ETcaD with SAE of 20 and ETD activation time of 100 ms. If z-ions were absent, SAE was increased up to 25, which is the maximal possible value for Orbitrap Elite instrument. In the case of a negative result, ETD activation time was increased up to 500 ms.
Quadruply charged ions were isolated and fragmented by ETD (in the same manner as triply charged ions) only for Brevinin 2Ec and tryptic peptide LVRPEVDVMCTAFHDNEETFLK.
At the second stage a “broadband” HCD fragmentation was performed (i.e., fragmentation of all ETD/ETcaD product ions with m/z values within defined brackets). The “m/z for isolation” parameter was set to certain values (Table 3). Isolation width was 600, as this is the maximum for Orbitrap Elite mass spectrometer. For windows with isolation centers of 300 and 500, the maximal possible widths were 300 and 500, respectively. All used ranges for ion isolation (i.e., “broadband windows”) are listed in Table 3. Only the windows containing z-ions corresponding to leucine or isoleucine residues were selected for each peptide. In every used window, six EThcD experiments were performed with different values of normalized collision energy (NCE) for HCD: 0, 5, 10, 15, 20, and 25. The resulting w-ions were searched in EThcD spectra manually.
If the targeted w-ions were not detected in EThcD spectra recorded in a broadband window 2 (from 250 to 750 Da, Table 3), experiments were repeated in windows 2a, 2b, or 2c, depending on m/z of the fragmenting z-ion.
It is worth mentioning that all these divisions of the mass scale become unnecessary if there is a possibility to do HCD for the entire set of ETD products. In this case, all L/I residues may be established in one shot. This option is available in certain instruments of Orbitrap family, e.g., Orbitrap Fusion.
Results and Discussion
General Procedure for Leucine and Isoleucine Discrimination
Forty-three tryptic peptides in the digests of HSA and gp188 and three non-tryptic peptides were studied. Their length varied from 3 to 34 amino acid residues (the longest tryptic peptide contained 27 amino acids); their monoisotopic masses varied from 330.2237 to 3516.9161. Eleven leucine/isoleucine-containing peptides from the two proteins were not included in this study because their peak intensities in ESI spectrum were too low. A total of 93 leucine/isoleucine residues were analyzed, 83 of which were correctly identified (Table 4). Thus, the method proved its efficiency in about 90% of cases.
Full ESI mass spectrum of HSA tryptic digest contained an intense peak of m/z 754.0141 corresponding to the triply protonated 19-residue tryptic peptide EFNAETFTFHADICTLSEK with isoleucine at position 13 and leucine at position 16. This ion was isolated and subjected to ETD fragmentation. Resulting ETD spectrum revealed z 4 - and z 7-ions with targeted N-terminal residues (Figure 1a). To obtain w-ions characteristic for leucine or isoleucine, further fragmentation by HCD was triggered, as it was done in , but without preliminary isolation of z-ions. Thus, z 4 ion of m/z 460.2537 falls into window 2 (from 250 to 750 Da, see Table 3). EThcD spectrum demonstrating this z-ion as well as the corresponding w-ion is shown in Figure 1b. In the case of leucine, an isopropyl radical should be ejected from z-ion , and the mass of resulting w-ion should be diminished by 43.0548. An intense peak of m/z 417.1988 corresponding to this w-ion (within the limits of experimental error) is clearly seen. If isoleucine was in this position, w-ion would appear due to ethyl radical loss, with a mass diminished by 29.0391. No peak of m/z 431.2146 is observed in Figure 1b. Therefore, one can confirm that this peptide has leucine but not isoleucine in position 16. The residue in position 13 was assigned as isoleucine with z7 ion of m/z 834.4179, which falls into the window 3 (700 – 1300 Da). EThcD spectrum in Figure 1c demonstrates this z-ion as well as w-ion of m/z 805.3785 corresponding to ethyl radical loss. The loss of isopropyl radical from this z-ion is not observed.
The proposed approach is applicable for peptides with multiple leucine/isoleucine residues. Figure 2 demonstrates an example of peptide ITLGEISSTLYTTYYK with four leucine/isoleucine residues. Two other examples of long peptides with several leucine/isoleucine residues are presented in supplementary data (Supplementary Figures S1 and S2).
The use of “broadband” ion isolation for HCD fragmentation is the main improvement of the ETD-HCD method that we tested earlier on natural peptides . The manual search for z-ions in every ETD spectrum is time-consuming, and it would be rather difficult to combine it with automatic sequencing procedure or HPLC/MS analysis. Now six “broadband windows” are used, which cover entire m/z range from 150 to 2800. The windows are partially overlapped at the borders to prevent a risk of occasional missing of any z-ion (see Table 3). An approach for automatic leucine/isoleucine discrimination may be developed in the future. Besides that, the procedure may be simplified using Orbitrap Fusion mass spectrometers allowing for EThcD fragmentation over the whole mass range in a single experiment.
ETD Fragmentation with Supplemental Collisional Activation (ETcaD)
Only a small portion of tryptic peptides can attach three or more protons while being ionized by electrospray. Most of them form doubly charged ions, attaching two protons, one of which appears to be fixed at the C-terminal lysine or arginine residue, whereas another one, “mobile proton” , is initially positioned at the N-terminal amino group, but can migrate along the backbone. When ETD fragmentation of such ion is triggered, the cleavage of N–Cα bonds near N-terminus is more probable than of those located in C-terminal part. Thus, the peaks of heavier z-ions are usually more intense than those of smaller ones. Figure 2a illustrates ETD spectrum of doubly charged ion of ITLGEISSTLYTTYYK peptide from gp188 tryptic digest. No z-ions are observed below z12; so, only five N-terminal N–Cα bonds are cleaved. Therefore, this spectrum cannot be used for identification of leucine/isoleucine residues at positions 6 and 10.
This problem was solved by ETD fragmentation with supplemental collisional activation (ETcaD), a method described in . Figure 2b illustrates ETcaD spectrum of this peptide. In this case every z-ion peak is quite pronounced, except for the smallest ones (z 1 to z 3). Besides, all targeted w-ions are well represented in EThcD spectra (Figure 2c–e). The optimal SAE was experimentally defined to be 20. Its maximal allowed value in Orbitrap Elite instrument is 25, but SAE 20 was already enough to find z- (and w-) ions of all leucines and isoleucines in the studied peptides. The only exception was z2-ion of GTSSSTETLLR peptide. Its peak had very low intensity at SAE 20 but was much more pronounced at SAE 25. On the contrary, both leucine residues of ADEVLFLQGSK peptide showed higher S/N of w-ions peaks at SAE 20.
ETD spectra of triply charged peptide ions contained all targeted z-ions, so supplemental activation was not required. The only exception involved long 27-residue tryptic peptide SHCIAEVENDEMPADLPSLAADFVESK. The probability of mobile proton occurrence at a certain residue in long peptides is lower; therefore supplemental activation has to be applied.
ETD Fragmentation of Short Peptides
Leucine/isoleucine discrimination in short tryptic peptides of 7 residues or less can be easily achieved using ETD without supplementary collisional activation. ETD spectra of their doubly charged precursor ions contain all z-ions starting from z 2 (e.g., for the peptide YLYEIAR, Figure 3a). Moreover, HCD stage may be skipped, since ETD provides enough energy not only for formation of z-ions but for their further fragmentation in a way that allows for leucine/isoleucine discrimination. Figure 3b demonstrates that both z- and w-ions are present in ETD spectrum of pentapeptide LDELR. The extent of such z-ion fragmentation decreases with the increase of peptide length. It is still recorded in 7-residue peptides but becomes negligible starting from 8-residue ones. Conversely, a tripeptide LAK demonstrated complete transformation of z 3-ion to w 3 under ETD conditions, so the z 3-ion peak of m/z 315.2158 is not observed (Figure 3c). Since the excessive energy accumulated during ETD fragmentation is distributed along the entire peptide backbone, it is sufficient for the cleavage of weak bonds in the case of short peptides. The radical site initiated z-ion fragmentation requires low excessive energy, so it may occur without additional activation with HCD. This issue should always be taken into account when dealing with short peptides. It is quite similar to the “hot-ECD” approach proposed by Kjeldsen et al. , which was routinely used in our earlier studies on the natural frog peptides .
It is worth mentioning that mass spectra of short peptides may be additionally complicated due to cyclization during MS experiments. To prevent cyclization, N-terminal acetylation or any other simple N-terminal derivatization  should be done prior to analysis. Thus, acetylated peptide Ac-IVGKGIVGVK-NH 2 was studied in the present work.
Effect of NCE in HCD Fragmentation
NCE in HCD fragmentation necessary for obtaining w-ions is highly dependent both on z-ion mass and peptide length. Short peptides z-ions may release w-ions even without HCD (see above), or with HCD with minimal NCE (5 or lower). At the same time, singly charged w-ions of m/z over 1500 are hardly visible in EThcD spectra with NCE lower than 20. At NCE 30, different fragmentation pathways are extensively activated, and spectra become too complicated. Therefore, we did not perform EThcD at NCE 30 or more in our studies. To detect some large w-ions, we attempted to increase NCE up to 40, but never succeeded.
EThcD spectrum in Figure 1 reveals both z- and w-ions peaks. However, starting from a certain NCE, z-ions are no longer visible in EThcD spectra, since various pathways of their fragmentation are activated. To confirm the presence of w-ion for leucine or isoleucine, one certainly needs to know the exact mass of z-ion, and if z-ion peak is absent in EThcD spectrum, it can be detected either in ETD or EThcD spectra with lower NCE.
A further increase of NCE may cause fragmentation of w-ion itself with unavoidable decrease of its intensity. Fortunately, all w-ions analyzed in our studies were detected at NCE below 25, although the short ones were sometimes hidden among the numerous peaks of fragment ions of other types. Figure 4 illustrates this phenomenon. The z2-ion of HPDYSVLLLR peptide starts fragmenting forming w 2-ion without any activation, i.e., at NCE 0 (Figure 4, left panel). At NCE 10 (Figure 4, middle panel) z 2-ion peak is absent due to extensive fragmentation, but w 2-ion peak only slightly increases. Either alternative pathways of z-ion fragmentation take place (bypassing w-ion formation), or w-ion secondary fragmentation starts at that NCE. Finally, at NCE 25 (Figure 4, right panel), the peak of w-ion is lost among many other peaks as multiple fragmentation pathways of many ions are activated.
Limitations of the Method
The presence of z-ions with leucine/isoleucine residues at N-terminus in ETD/ETcaD spectra is critical. If a peptide contains many leucines and/or isoleucines, it is important that all the targeted z-ions would be visible in ETD/ETcaD spectra. If a z-ion is absent it is impossible to obtain the corresponding w-ion, and the method does not work.
Sometimes we could not observe the peaks of the targeted z-ions in ETD/ETcaD spectra registered under standard conditions (SAE 20, ETD activation time 100 ms). Usually they appeared after variation of one of these parameters (or both of them). Nevertheless, there were seven leucine/isoleucine z-ions that did not form. Noteworthy, their m/z values fall into one of two ranges: 300–450 (three z-ions) or >2000 (four z-ions).
Considering the method limitations, it is worth mentioning that 70% of unsuccessful leucine/isoleucine discriminations in the current study happened when dealing with z-ions of m/z beyond 2000. The kinetic shift increases when a fragmenting molecule becomes more complex, since an excessive internal energy should be distributed among many bonds [30, 31]. The peaks of heavy z-ions were of very low intensity, just above the noise level or even below it. Even if such z-ions were visible in ETD spectra, it was difficult to detect w-ions in resulting EThcD spectra. HCD fragmentation of large z-ions requires high NCE values, triggering alternative pathways of their fragmentation and resulting in considerable suppression of w-ions formation.
The problem of large z-ions was already mentioned by Xiao et al. . The authors have reported that it was difficult to obtain any good result working with z-ions of m/z over 1000. Nevertheless our data demonstrate the applicability of the proposed leucine/isoleucine discrimination performed with z-ions of m/z up to nearly 2000. Successful leucine/isoleucine discrimination was performed using z 16 ion of ITLGEISSTLYTTYYK peptide, (m/z 1836.9455, Figure 2) and z 16 ion of YTQLEDFYLTIFHPASVGK peptide (m/z 1820.9400, Supplementary Figure S3a). EThcD spectrum of z 18 ion of peptide NTGGIGLPTTSDASGYNVITALQR (m/z 1890.9760) at NCE 25 demonstrated w-ion at 1847.9211, which was only slightly above the noise (S/N = 2.5), but that was enough to confirm leucine in position 7 (Supplementary Figure S3b).
If m/z value of a targeted z-ion is over 2000, either z-ion or the corresponding w-ion is usually absent in the spectra, making the method not applicable. Only determination of leucine in position 2 of NLAFLGLYSLTVDGIWGNGTLSGLDK peptide was successful when dealing with z-ions of that type. In this case, we had to use Orbitrap analyzer at maximal resolving power of 480,000. Ion z25 had monoisotopic mass 2594.3639, and its more intense isotope peaks were very close to c25 ion peaks; the difference was 0.0275 Da (Supplementary Figure S3c). These peaks were poorly distinguishable at resolving power 120,000.
Since all analyzed peptides with molecular mass above 2000 formed triply charged precursors that were used for ETD, we looked for doubly charged z-ions in ETD spectra, as their m/z values should have been under 2000 Da. However, only z 21 2+ ion of peptide LVRPEVDVMCTAFHDNEETFLK was found at m/z 1317.6272 and allowed for identification of the target residue (Supplementary Figure S4).
There were several occasions when z-ions of “normal size range” (m/z from 300 to 2000) could not be easily fragmented by HCD with w-ion formation. Thus, w 7-ion of m/z 729.3629 was detected in EThcD spectra (NCE 20 or 25) of ELGLYTGQIDGVWGK peptide, although its intensity did not reach even that of two noise peaks (Figure 5). The peak of precursor z7-ion, however, was rather intense in the ETcaD spectrum. The use of higher resolution mode (120,000) allowed detecting indisputably this w-ion and even its M + 1 isotope peak by their exact m/z values (Figure 5, inset). So, residue 9 of this peptide has been confirmed as isoleucine, despite of low S/N ratio of the corresponding w-ion peak.
Radical Site Migration Problem
A radical site migration has been described earlier as a problem for leucine/isoleucine discrimination. It was reported in many studies when leucine and isoleucine residues were adjacent to each other [15, 16, 19–21]. A radical site of odd-electron z-ion may migrate to the next residue (so-called “hydrogen abstraction” ), and cause ejection of ethyl or isopropyl radical from its side chain. This would lead to incorrect identification, since EThcD spectra would contain peaks corresponding to both ethyl and isopropyl losses from the targeted z-ion. As we have shown in our previous study , such “false w-ions” generally appear at larger NCE values than “true w-ions.” Therefore, the radical site migration problem can be resolved by fragmentation with the lowest possible NCE value.
Now we have addressed this problem once again. None of tryptic peptides used in our studies contained “IL” combination, and “LI” was found in VFDEFKPLVEEPQNLIK and ALVLIAFAQYLQQCPFEDHVK peptides. We did not detect any signs of the radical site migration in the case of VFDEFKPLVEEPQNLIK peptide (Supplementary Figure S1). Even the shortest z3 ion of m/z 357.2637 revealed only correct w 3 -ion of m/z 314.2087. An alternative ion corresponding to ethyl loss (m/z 328.2248) was not observed even after HCD fragmentation at maximal NCE value. Another peptide, ALVLIAFAQYLQQCPFEDHVK, could not be used for this purpose, as z 18 -ion was not observed in ETD spectrum.
To study the problem, we analyzed two non-tryptic peptides with “IL” combination. Peptide DILFLSDPDAR-NH 2 demonstrated a radical site migration upon identification of isoleucine at position 2. EThcD spectrum at NCE 20 (Figure 6a) revealed peaks of z 10-ion of m/z 1129.6138, w 10-ion of m/z 1100.5752 (ethyl loss from z 10), and a peak of m/z 1086.5596 corresponding to isopropyl loss from z10 ion. This latter one, however, was not detected at NCE 15 (Figure 6b). So, the radical migration problem can still be solved by a proper NCE selection as was stated in our previous work .
Additionally, we tested natural peptide Brevinin 2Ec from Rana ridibunda frog skin secretion, one of the peptides used in our previous work . As it is seen from its sequence, GILLDKLKNFAKTAGKGVLQSLLNTASCKLSGQC, the isoleucine at position 2 is adjacent to two leucines, and both of them may contribute to isopropyl loss from z 33-ion due to radical site migration. ETD spectrum of the quadruply charged peptide precursor ion (m/z 880.2444) contained triply-charged z 33-ion of m/z 1148.9687. Its further fragmentation by HCD required fine-tuning of NCE value. At NCE 10, w 33-ion (m/z 1139.2894) peak was hardly visible, whereas at NCE 20 the “false w-ion” of m/z 1134.6171 appeared due to hydrogen migration. The optimal NCE value was 15, when w 33-ion peak was intense enough, but the radical migration was too weak to be observed (Supplementary Figure S5). Again, one can always choose a proper NCE value when w-ion is detected but radical site migration has not started yet.
There were several isoleucine identifications when additional peak with exact mass value corresponding to C3H7 • radical loss (–43.0548) from z-ion was observed, but there was no leucine at the adjacent position. First of all, this phenomenon was observed upon fragmentation of a short z2-ion IK•+ (peptides QIK and VFDEFKPLVEEPQNLIK, Table 1). As expected, ETD spectra of both peptides revealed z2-ion of m/z 244.1786. EThcD spectra contained correct w-ion of m/z 215.1396 along with a peak of ion of m/z 201.1240 corresponding to the loss of C3H7 • radical from z 2 -ion (Supplementary Figure S9). The mechanism of this process requires an independent study. On the other hand, three tryptic peptides with –LK at their C-termini (Table 1) never demonstrated C2H5 • loss from z 2-ion (LK•+), whereas w 2-ion of m/z 201.1240 was always detected in their EThcD spectra. Therefore, in this particular or similar case, leucine/isoleucine discrimination may be performed on the basis of the presence or absence of a peak of m/z 215.1396 corresponding to C2H5 • loss from z 2-ion.
Radical site migration may occur when leucines and isoleucines are separated with several residues, being rather far away from each other. When identifying isoleucine in position 15 of peptide NLAFLGLYSLTVDGIWGNGTLSGLDK, we have observed in EThcD spectra a peak of m/z 1201.5890 corresponding to C3H7 • loss from z 12-ion (i.e., 1244.6438 minus 43.0548). This peak appeared already at NCE 15, and its intensity was almost as high as that of w 12-ion of m/z 1215.6043. Fine-tuning in the range of NCE 10–15 allowed selecting proper energy (NCE13.7) when only the correct w 12-ion (m/z 1215.6043) was present (Supplementary Figure S6).
Finally, ion “z minus 43.0548” was observed when identifying isoleucine in position 6 of synthetic peptide Ac-IVGKGIVGVK-NH2. Supplementary Figure S7 shows that the ion peak of m/z 455.2991 corresponding to C3H7 • loss from z 5-ion (m/z 498.3539) appears in EThcD spectrum already at NCE 15; w 5 -ion of m/z 469.3146 is present at larger intensity at that NCE. Since this peptide does not contain leucine residues, we hypothesized that the peak at 455.2991 appeared due to isopropyl radical loss by the side chain of adjacent valine residue. To check this, we isolated z 5-ion after ETD stage and performed its HCD fragmentation with all NCE values up to 25. No peak at 455.2991 was observed at any NCE. Thus, we concluded that this peak in EThcD spectrum was not a fragment of z 5-ion, but of another ion present in ETD spectrum, and its exact mass value corresponding to isopropyl loss from z 5-ion was just a coincidence. Anyway, if both peaks “z minus 29.0391” and “z minus 43.0548” are present in EThcD spectrum, one should isolate corresponding z-ion manually and check by HCD its real product ions.
Therefore, when radical site migration is suspected (both alternative ions peaks are present in the spectrum) the following procedure may be applied: isolation of the corresponding z-ion (with standard isolation width, e.g., 4 Da) to perform ETD-HCD MS3 experiment with different NCEs. If only one w-ion is present in the obtained spectrum, the false ion comes from another primary ion and reliable identification of leucine/isoleucine is achieved. Otherwise fine stepwise NCE tuning from NCE when w-ions are absent to NCE when only the first w-ion appears is required. Supplementary Figure S6 illustrates that issue.
The isomeric leucine and isoleucine residues can be easily discriminated in the vast majority of tryptic peptides using Orbitrap Elite mass spectrometer. The shorter the peptide, the easier it is for the discrimination. The procedure includes broadband HCD fragmentation of primary z-ions obtained by ETD/ETcaD, which is routinely done in peptide sequencing using mass spectrometry. The isolation of each z-ion is not required; therefore, the targeted HCD fragmentation could be done automatically after each ETD/ETcaD step. Radical site migration, which sometimes represents a problem, may be reliably precluded by adjusting NCE. The proposed method should be quite helpful for proteomic studies.
Edman, P.: A method for the determination of the amino acid sequence in peptides. Arch. Biochem. 22, 475–476 (1949)
Edman, P., Begg, G.: A protein sequenator. Eur. J. Biochem. 1, 80–91 (1967)
Artemenko, K.A., Zubarev, A.R., Samgina, T.Y., Lebedev, A.T., Savitski, M.M., Zubarev, R.A.: Two-dimensional mass mapping as a general method of data representation in comprehensive analysis of complex molecular mixtures. Anal. Chem. 81, 3738–3745 (2009)
McLafferty, F.W., Turecek, F.: Interpretation of mass spectra, 4th edn. University Science Books, Mill Valley, CA (1993)
Johnson, R.S., Martin, S.A., Biemann, K., Stults, J.T., Watson, J.T.: Novel fragmentation process of peptides by collision-induced decomposition in a tandem mass spectrometer: differentiation of leucine and isoleucine. Anal. Chem. 59, 2621–2625 (1987)
Martin, S.A., Biemann, K.: Mass spectrometric determination of the amino acid sequence of peptides and proteins. Mass Spectrom. Rev. 6, 1–78 (1987)
Ramsey, S.L., Steinbornen, S.T., Waugh, R.J., Dua, S., Bowie, J.H.: A simple method for differentiating Leu and Ile in peptides. The negative‐ion mass spectra of M-h.− ions of phenylthiohydantoin Leu and Ile. Rapid Commun. Mass Spectrom. 9, 1241–1243 (1995)
Tao, W.A., Wu, L., Cooks, R.G.: Differentiation and quantitation of isomeric dipeptides by low-energy dissociation of copper(II)-bound complexes. J. Am. Soc. Mass Spectrom. 12, 490–496 (2001)
Nakamura, T., Nagaki, H., Ohki, Y., Kinoshita, T.: Differentiation of leucine and isoleucine residues in peptides by consecutive reaction mass spectrometry. Anal. Chem. 62, 311–313 (1990)
Armirotti, A., Millo, E., Damonte, G.: How to discriminate between leucine and isoleucine by low energy ESI-TRAP MSn. J. Am. Soc. Mass Spectrom. 18, 57–63 (2007)
Xiao, Y., Vecchi, M.M., Wen, D.: Distinguishing between leucine and isoleucine by integrated LC-MS analysis using an Orbitrap Fusion mass spectrometer. Anal. Chem. 88, 10757–10766 (2016)
Seymour, J.L., Turecek, F.: Distinction and quantitation of leucine-isoleucine isomers and lysine-glutamine isobars by electrospray ionization tandem mass spectrometry (MS(n), n = 2, 3) of copper(II)-diimine complexes. J. Mass Spectrom. 35, 566–571 (2000)
Kjeldsen, F., Haselmann, K.F., Sørensen, E.S., Zubarev, R.A.: Distinguishing of Ile/Leu amino acid residues in the PP3 protein by (hot) electron capture dissociation in Fourier transform ion cyclotron resonance mass spectrometry. Anal. Chem. 75, 1267–1274 (2003)
Fung, Y.M., Chan, T.W.: Experimental and theoretical investigations of the loss of amino acid side chains in electron capture dissociation of model peptides. J. Am. Soc. Mass Spectrom. 16, 1523–1535 (2005)
Han, H., Xia, Y., McLuckey, S.A.: Ion trap collisional activation of c- and z-ions formed via gas-phase ion/ion electron-transfer dissociation. J. Proteome Res. 6, 3062–3069 (2007)
Gupta, K., Kumar, M., Chandrashekara, K., Krishnan, K.S., Balaram, P.: Combined electron transfer dissociation-collision-induced dissociation fragmentation in the mass spectrometric distinction of leucine, isoleucine, and hydroxyproline residues in peptide natural products. J. Proteome Res. 11, 515–522 (2012)
Cook, S.L., Collin, O.L., Jackson, G.P.: Metastable atom‐activated dissociation mass spectrometry: leucine/isoleucine differentiation and ring cleavage of proline residues. J. Mass Spectrom. 44, 1211–1223 (2009)
Lebedev, A.T., Damoc, E., Makarov, A.A., Samgina, T.Y.: Discrimination of leucine and isoleucine in peptides sequencing with Orbitrap Fusion mass spectrometer. Anal. Chem. 86, 7017–7022 (2014)
Leymarie, N., Costello, C.E., O'Connor, P.B.: Electron capture dissociation initiates a free radical reaction cascade. J. Am. Chem. Soc. 125, 8949–8958 (2003)
O’Connor, P.B., Lin, C., Cournoyer, J.J., Pittman, J.L., Belyayev, M., Budnik, B.A.: Long-lived electron capture dissociation product ions experience radical migration via hydrogen abstraction. J. Am. Soc. Mass Spectrom. 17, 576–585 (2006)
Li, X., Lin, C., Han, L., Costello, C.E., O'Connor, P.B.: Charge remote fragmentation in electron capture and electron transfer dissociation. J. Am. Soc. Mass Spectrom. 21, 646–656 (2010)
Bagal, D., Kast, E., Cao, P.: Rapid distinction of leucine and isoleucine in monoclonal antibodies using nanoflow LCMSn. Anal. Chem. (2016). doi:10.1021/acs.analchem.6b03261
Frese, C.K., Altelaar, A.F., van den Toorn, H., Nolting, D., Griep-Raming, J., Heck, A.J., Mohammed, S.: Toward full peptide sequence coverage by dual fragmentation combining electron-transfer and higher-energy collision dissociation tandem mass spectrometry. Anal. Chem. 84, 9668–9673 (2012)
Samgina, T.Y., Artemenko, K.A., Gorshkov, V.A., Nielsen, M.L., Savitski, M.M., Zubarev, R.A., Lebedev, A.T.: ESI MS/MS sequencing of novel skin peptides from Ranid frogs containing disulfide bridges. Eur. J. Mass Spectrom. 13, 155–163 (2007)
Kurochkina, L.P., Semenyuk, P.I., Orlov, V.N., Robben, J., Sykilinda, N.N., Mesyanzhinov, V.V.: Expression and functional characterization of the first bacteriophage-encoded chaperonin. J. Virol. 86, 10103–10111 (2012)
Swaney, D.L., McAlister, G.C., Wirtala, M., Schwartz, J.C., Syka, J.E., Coon, J.J.: Supplemental activation method for high-efficiency electron-transfer dissociation of doubly protonated peptide precursors. Anal. Chem. 79, 477–485 (2007)
Wysocki, V.H., Tsaprailis, G., Smith, L.L., Breci, L.A.: Mobile and localized protons: a framework for understanding peptide dissociation. J. Mass Spectrom. 35, 1399–1406 (2000)
Samgina, T.Y., Artemenko, K.A., Gorshkov, V.A., Ogourtsov, S.V., Zubarev, R.A., Lebedev, A.T.: Mass spectrometric study of peptides secreted by the skin glands of the brown frog Rana arvalis from the Moscow region. Rapid Commun. Mass Spectrom 23, 1241–1248 (2009)
Samgina, T.Y., Kovalev, S.V., Gorshkov, V.A., Artemenko, K.A., Poljakov, N.B., Lebedev, A.T.: N-terminal tagging strategy for de novo sequencing of short peptides by ESI-MS/MS and MALDI-MS/MS. J. Am. Soc. Mass Spectrom. 21, 104–111 (2010)
Price, W.D., Schnier, P.D., Jockusch, R.A., Strittmatter, E.F., Williams, E.R.: Unimolecular reaction kinetics in the high-pressure limit without collisions. J. Am. Chem. Soc. 118, 10640–10644 (1996)
Laskin, J., Futrell, J.H.: Activation of large ions in FT-ICR mass spectrometry. Mass Spectrom. Rev. 24, 135–167 (2005)
The authors are thankful to Thermo Fisher Scientific Inc., Textronica AG group (Moscow, Russia), and personally to Professor Alexander Makarov for providing Orbitrap Elite mass spectrometer for this work. They also express their gratitude to Professor Valery Shevchenko (Cancer Research Center, Moscow), Dr. Irina Tarasova (Insitute for Energy Problems of Chemical Physics, Russian Academy of Sciences), Dr. Ludmila Alekseeva and Dr. Lidia Kurochkina (Institute of Bioorganic Chemistry, Russ. Acad. Sci.) for proteins and synthetic peptides used in this studies.
Sergey Zhokhov and Sergey Kovalyov contributed equally to this work.
Electronic supplementary material
Below is the link to the electronic supplementary material.
(DOC 1.67 MB)
About this article
Cite this article
Zhokhov, S.S., Kovalyov, S.V., Samgina, T.Y. et al. An EThcD-Based Method for Discrimination of Leucine and Isoleucine Residues in Tryptic Peptides. J. Am. Soc. Mass Spectrom. 28, 1600–1611 (2017). https://doi.org/10.1007/s13361-017-1674-3
- Leucine/isoleucine differentiation
- Tryptic peptides
- Peptide sequencing