Differential ¹⁴N/¹⁵N-Labeling of Peptides Using N-Terminal Charge Derivatization with a High-Proton Affinity for Straightforward de novo Peptide Sequencing

Abstract

While de novo peptide sequencing is essential in many situations, it remains a difficult task. This is because peptide fragmentation results in complicated and often incomplete product ion spectra. In a previous study, we demonstrated that N-terminal charge derivatization with 4-amidinobenzoic acid (Aba) resulted in improved peptide fragmentation under low-energy CID conditions. However, even with this derivatization, some ambiguity exists, due to difficulties in discriminating between N- and C-terminal fragments. In this study, to specifically identify b-ions from complex product ion spectra, the differential ¹⁴N/¹⁵N-labeling of peptides was performed using Aba derivatization. ¹⁵N-Labeled Aba was synthesized in the form of a succinimide ester. Peptides were derivatized individually with ¹⁴N-Aba or ¹⁵N-Aba and analyzed by ESI-MS/MS using a linear ion trap-Orbitrap hybrid FTMS system. The N-terminal fragments (i.e., b-ions) were then identified based on m/z differences arising from isotope labeling. By comparing the spectra between ¹⁴N- and ¹⁵N-Aba derivatized peptides, b-ions could be successfully identified based on the m/z shifts, which provided reliable sequencing results for all of the peptides examined in this study. The method developed in this study allows the easy and reliable de novo sequencing of peptides, which is useful in peptidomics and proteomics studies.

INTRODUCTION

Recent advances in various tandem mass spectrometric techniques have enabled the determination of the amino acid sequences of peptides.¹⁾ In addition, the development of comprehensive databases of protein primary sequences predicted from genomic information of various organisms has boosted the efficiency with which proteins can be identified by mass spectrometry.²⁾ In this database-dependent method, the assignment of each fragment ion in a product ion spectrum is not necessary, and identification is basically achieved by comparing the observed product ion spectra of enzymatically-digested peptides with those obtained by theoretical calculations based on the primary sequences in databases. As primary sequence databases continue to grow, database-dependent methods are expected to be more powerful and convenient tools for analyses of peptides/proteins. However, this workflow is possible only if the sequence of the protein under study is present in the database. Thus, amino acid sequencing without the assistance of a database (de novo sequencing) is still essential for identifying peptides and proteins in many situations.³⁾

De novo sequencing is generally carried out by interpreting differences in mass between series of consecutive ions of N- or C-terminal fragments (basically b- or y-ions, respectively) generated by collision-induced dissociation (CID). Since tryptic peptides always contain Arg or Lys residues at the C-terminus, y-ions are preferentially generated because of their basicity, allowing spectra to be easily interpreted. However, when proteins are digested with other enzymes such as chymotrypsin, basic residues are located at various positions within the digested peptides. Unlike tryptic digests, the fragmentation of these peptides provides complicated and often incomplete product ion spectra, making de novo sequencing a troublesome task. This is also true for peptidomic analysis, in which peptides are generally analyzed without trypsin digestion.⁴⁾

To overcome these problems in de novo sequencing, chemical modifications at the N-terminus of the peptide have been attempted in order to provide more consistent and informative fragmentation.⁵⁾ The introduction of a positively charged group, such as quaternary and tertiary ammonium,⁶⁾ or quaternary phosphonium groups,⁷⁾ at the N-terminus is known to enhance the relative intensity of N-terminal fragments in product ion spectra, facilitating the formation of a consecutive series of a- or b-ions. Recently, our group introduced another type of derivatization to improve fragmentation under low-energy CID conditions, in which a substructure with a high proton affinity, such as a guanidino or amidino group, is included. These structures are thought to be suitable for this purpose because the fragmentation of peptides under low-energy CID conditions is triggered by mobile protons from basic sites of peptides.⁸⁾ When singly-charged ions were selected as a precursor, mobile protons are provided mainly from the derivatization groups. After amide bond dissociation, mobile protons move back to the N-terminus to generate b-ions, if its proton affinity is higher than the C-terminal fragments. Of the structures tested, 4-amidinobenzoic acid (Aba) was found to have the appropriate proton affinity to serve as both a donor and an acceptor of mobile protons, which would provide “readable” fragmentation by increasing the number and intensity of b-ions.⁹⁾ However, some y-ions were still observed for cases of peptides containing highly basic residues even after derivatization, in which doubly-charged ions were selected as a precursor. This could lead to the incorrect assignment of b-ions in some cases.

Differential isotope labeling techniques have been investigated, in attempts to specifically identify a single series of fragments from complex spectra.^10–12) By introducing isotopes into the N- or C-terminus, either fragment can be specifically identified based on differences in mass caused by isotope labeling. For example, specific C-terminal ¹⁸O-labeling was accomplished when proteins were enzymatically digested in the presence of H₂¹⁸O.^13,14) Although this method allows the facile assignment of C-terminal fragments without the need for peptide derivatization, the fragmentation pattern itself is not improved. Furthermore, since enzymatic digestion is essential for this method, its application to peptidomics studies would be difficult, in which intact peptides are generally dealt with. Another differential isotope-labeling method reported previously is N-terminal acetylation using deuterated acetic anhydride.¹⁰⁾ Unlike ¹⁸O-labeling, this method is applicable to any sample. However, blocking of the N-terminal amino group results in a decrease in the yield of protonated molecules due to its reduced proton affinity, which could affect the sensitivity of detection of peptides. Although fragmentation efficiency is slightly improved by N-terminal acetylation, the type of product ion formed in both N-acetylated and intact peptides is reported to be similar.¹⁵⁾ Isotope-labeling using N,N-dimethylation of amino groups was also reported to facilitate de novo sequencing. This method not only provides the characteristic isotope signature in the N-terminal fragments but also increases the basicity of the N-terminus.^12,16) However, because the basicity of the N,N-dimethylamino group is lower than that of an amidino group, its effect on enhancing b-ion intensities could be limited. In addition, this reaction also takes place at Lys side chains, and additional treatments such as masking of the amino group by guanidination¹⁶⁾ or digestion with a Lys-N protease to produce peptides with Lys residues at the N-terminus¹²⁾ is necessary to avoid the complexity arising from multiple labeling. Thus, when N-terminal modification is employed for differential isotope labeling, its effect on fragmentation must be taken into account, which is associated with the availability of mobile protons.

In this study, we present a novel differential isotope labeling technique using N-terminal derivatization with Aba, in which a stable isotope of nitrogen (¹⁵N) is incorporated into the amidino group of Aba. The usefulness of this method for specific identification of b-ions in product ion spectra was demonstrated using high-resolution mass spectrometers, such as an Orbitrap instrument, to establish a reliable and straightforward de novo sequencing method.

EXPERIMENTAL

Chemicals

4-Amidinobenzoyloxysuccinimide (Aba-NHS) was synthesized as previously described.⁹⁾ Angiotensin I was purchased from Calbiochem (Darmstadt, Germany). Other peptides used in this study were synthesized in our laboratory using standard Fmoc solid-phase chemistry. Lysozyme was obtained from Sigma-Aldrich (St. Louis, MO). Acetonitrile and formic acid used for LC-MS/MS analysis were purchased from Kanto Chemical (Tokyo, Japan) and Wako Pure Chemical Industries, Ltd. (Osaka, Japan), respectively. Water used for LC-MS/MS analysis was produced by a Milli-Q gradient system (Millipore, Billerica, MA). Trypsin (mass spectrometry grade) was purchased from Wako Pure Chemical Industries, Ltd. ¹⁵N-Ammonium chloride (99% ¹⁵N) was obtained from Cambridge Isotope Laboratories (Andover, MA).

Synthesis of ¹⁵N-4-amidinobenzoyloxysuccinimide (¹⁵N-Aba-NHS)

¹⁵N-4-Amidinotoluene

p-Tolunitrile (7.0 g, 60 mmol) dissolved in 30 mL of methanol was added to the solution of sodium methoxide (0.37 g, 6.6 mmol) in 10 mL of methanol, and reacted for 39 h at room temperature under an atmosphere of argon. To this reaction mixture, ¹⁵N-ammonium chloride (1.0 g, 20 mmol) was added and the reaction was allowed to proceed for 5 h at room temperature. After evaporation of the solvent, the residue was suspended in diethylether, and the supernatant was discarded. This operation was repeated a total of five times to yield ¹⁵N-4-amidinotoluene (2.16 g, 81%).

¹H-NMR (400 MHz, DMSO-d₆) δ 2.40 (3H, s), δ 7.41 (2H, d, J=7.5 Hz), δ 7.74 (2H, d, J=7.9 Hz), δ 8.40 (3H, br s). ESI-MS: [M+H]⁺ (m/z) Calculated: 136.1, Found: 136.0

¹⁵N-4-Amidinobenzoic acid

¹⁵N-4-Amidinotoluene (1.0 g, 7.5 mmol) was dissolved in an aqueous solution of sodium hydrate (0.1 M). Potassium permanganate (2.3 g, 20 mmol) was added to this solution and reacted for 22 h at 40°C. After the reduction of the remaining sodium permanganate with sodium bisulfate, the solution was acidified with 3 M hydrochloric acid. The solution was filtrated, and the filtrate was evaporated. The residue was applied to silica gel flush column chromatography eluted with chloroform/methanol/trifluoroacetic acid (90 : 10 : 1) to obtain ¹⁵N-4-amidinobenzoic acid (0.27 g, 22%).

¹H-NMR (400 MHz, DMSO-d₆) δ 7.91 (2H, d, J=8.0 Hz), δ 8.12 (2H, d, J=8.0 Hz), δ 9.36–9.60 (3H, m), δ 13.47 (H, s). ESI-MS: [M+H]⁺ (m/z) Calculated: 166.1, Found: 166.0

¹⁵N-4-Amidinobenzoyloxysuccinimide

¹⁵N-4-Amidinobenzoic acid (165 mg, 1 mmol) was dissolved in 10 mL of N,N-dimethylformamide (DMF) and 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) (192 mg, 1 mmol). To the solution, N-hydroxysuccinimide (NHS) (115 mg, 1 mmol) were added and reacted for 40 h at room temperature. The solution was evaporated, and the residue was applied to silica gel flush column chromatography eluted with ethyl acetate/chloroform/trifluoroacetic acid (90 : 10 : 1) to obtain ¹⁵N-4-amidinobenzoyloxysuccinimide (73 mg, 28%, incorporation of ¹⁵N into one of nitrogen atoms in the amidino group: 97.1%).

¹H-NMR (400 MHz, DMSO-d₆) δ 2.92 (4H, s), δ 8.04 (2H, d, J=8.6 Hz), δ 8.31 (2H, d, J=8.2 Hz), δ 9.40–9.72 (3H, m). ESI-MS: [M+H]⁺ (m/z) Calculated: 263.1, Found: 262.9

Peptide derivatization

For the derivatization of peptides, non-labeled or ¹⁵N-labeled 4-amidinobenzoyloxysuccinimide was dissolved in phosphate buffer (final 50 mM, pH 5.9 for peptides 1 and 3, pH 8.2 for peptides 2, 4, and 5) at a final concentration of 10 mM, and mixed with peptides (10 μM) for approximately 20 min at room temperature. The derivatized peptides were desalted using a MonoSpin C18 cartridge (GL Sciences, Tokyo, Japan) and analyzed by LC-MS/MS as shown below.

Protein digestion and derivatization

Lysozyme was reduced with dithiothreitol (10 mM) and its cysteine residues were carbamidomethylated by reaction with iodoacetamide (20 mM). The sample was digested with trypsin at 37°C overnight. The resulting peptide fragments were desalted using a MonoSpin C18 cartridge. The peptide fragments were dissolved in a 50 mM phosphate buffer (pH 8.2) and mixed with non-labeled or ¹⁵N-labeled 4-amidinobenzoyloxysuccinimide for approximately 20 min at room temperature. Derivatized peptides were desalted using a MonoSpin C18 cartridge and analyzed by LC-MS/MS as shown below.

Mass spectrometric analysis

LC/MS/MS analysis was performed on an LTQ Orbitrap Velos (Thermo Fisher Scientific, Waltham, MA) equipped with an ESI probe. HPLC separations were performed with a TSK-ODS V100 column (2.1×50 mm for model peptides and 2.1×150 mm for protein digests, TOSOH, Tokyo, Japan) on an Agilent 1200 HPLC system (Agilent Technologies, Santa Clara, CA). The column was eluted with a mixture of 0.1% formic acid in water (solvent A) and acetonitrile (solvent B). For the analysis of derivatized peptides, except for peptide 4, a 3-min linear gradient from 5% to 50% solvent B at a flow rate of 0.3 mL/min was used. For analysis of derivatized peptide 4, a 25-min linear gradient from 5% to 50% solvent B at a flow rate of 0.2 mL/min was used after 5 min of isocratic elution with 5% solvent B. For the analysis of protein digests, a 40-min linear gradient from 2% to 50% solvent B at a flow rate of 0.2 mL/min was used after 2 min of isocratic elution with 2% solvent B. Mass spectra were obtained in a product ion scan mode under low-energy CID conditions using a normalized collision energy of 35%. Doubly charged ions were automatically selected as a precursor ion with an isolation width of 2 Da. The resolution was set to 7,500 or 30,000 at m/z 400. Other mass spectrometric conditions were as follows: spray voltage=3 kV; sheath gas flow=50 and auxiliary gas flow=15; heated capillary temperature=230°C; predictive automatic gain control (AGC) enabled; S-lenz RF level=31 or 67%.

RESULTS AND DISCUSSION

In a previous paper, we demonstrated that charge-derivatization with 4-amidinobenzoic acid (Aba) at the peptide N-terminus significantly improved peptide fragmentation by enhancing the intensity of b-ions.⁹⁾ However, in some cases, the identification of b-ions was still not very easy due to the overlap of b-ions with other series of ions, such as y-ions, especially when multiply charged ions were selected as a precursor. In the case of angiotensin I, derivatization with Aba permitted the complete series of b-ions to be detected, but several y-ions were observed with a sufficient intensity to complicate the spectrum because two His residues are present in the C-terminal half (Fig. 2a). In the present study, to distinguish N-terminal fragments (b-ions) from C-terminal fragments (y-ions), we utilized isotope-labeled Aba as a derivatization reagent. A stable isotope of nitrogen (¹⁵N) was incorporated into an amidino moiety of Aba (¹⁵N-Aba, Fig. 1), which provides a mass shift of 0.997 Da for N-terminal fragments when compared with fragments derivatized with non-labeled Aba (¹⁴N-Aba). Since this mass shift overlaps with isotope peaks arising from an intrinsic ¹³C isotope, it is difficult to distinguish between them when peptides derivatized with a mixture of ¹⁴N- and ¹⁵N-Aba. Therefore, it is necessary to perform MS/MS analysis individually for each derivatized peptide to detect the mass shifts, although it is possible to distinguish incorporated isotopes from natural isotopes using a software tool.¹⁷⁾ In addition, the utilization of high-resolution mass spectrometers, such as an Orbitrap mass spectrometer, permits the reliable identification of fragments when ions with similar m/z values are present in product ion spectra. The effectiveness of parallel ¹⁴N/¹⁵N-Aba derivatization for the unambiguous identification of b-ions was evaluated using several model peptides in the following sections.

Fig. 2. Product ion spectra of Aba-derivatized angiotensin I. Comparison of spectra between ¹⁴N- and ¹⁵N-Aba derivatization (a), and enlarged views of the spectra (b and c).

Fig. 1. Structure of ¹⁵N-4-amidinobenzoyloxysuccinimide.

Effect of ¹⁴N/¹⁵N-Aba derivatization on de novo sequencing of angiotensin I

The effect of parallel ¹⁴N/¹⁵N-Aba derivatization was first evaluated using angiotensin I. This peptide contains three basic amino acid residues that hamper efficient fragmentation (Supplemental Fig. 1a). Basic residues such as Arg, His and Lys often suppress the dissociation of amide bonds that are located adjacent to those residues, because the protons required for bond dissociation are repelled by the positive charge of their side chains.¹⁸⁾ In addition, the amide bond C-terminal to Pro is rarely dissociated because of the nature of its structure.¹⁹⁾ These features resulted in a low intensity of several b-ions in the product ion spectrum, even after ¹⁴N-Aba derivatization (Fig. 2a). In addition, the intensity of several y-ions was enhanced by the presence of the basic residues. To correctly identify b-ions, the peptides were individually derivatized with ¹⁵N-Aba, and the resulting spectrum was compared with that of peptides derivatized with ¹⁴N-Aba. Doubly charged ions (m/z 721.87 and 722.37 for ¹⁴N- and ¹⁵N-Aba, respectively) were chosen as a precursor for the MS/MS analysis, and similar product ion spectra were obtained for both derivatized peptides (Fig. 2a). A complete series of b-ions were observed in both spectra by virtue of Aba-derivatization, as shown in the previous report⁹⁾ (Table 1). Closer inspection of the spectra revealed that several fragment ions showed different m/z values between ¹⁴N- and ¹⁵N-Aba derivatized peptides. These m/z shifts (0.997 and 0.499 Da for singly and doubly charged ions, respectively) were detected only for b-ions (Figs. 2b, 2c and Table 1). For example, the y₈ ion was observed without any m/z shift (m/z 1025.556 and 1025.555 for both derivatized peptides), but the m/z value of the b₇ ion was different between ¹⁴N- and ¹⁵N-Aba derivatized peptides (m/z 1027.509 and 1028.507 for ¹⁴N- and ¹⁵N-Aba, respectively, Fig. 2b). The difference of 0.998 Da matched the theoretical value, indicating that these fragments contain the Aba-derivatized N-terminus. Likewise, the b₃ ion showed an m/z shift whereas the y₄ ion remained unchanged (Fig. 2c). Moreover, high-resolution mass spectra contributed to the reliable assignment of fragments, especially for discriminating between ions with similar m/z values. For example, Aba derivatization resulted in the formation of the b₇ ion as a doubly-charged ion at m/z 514.259, which was overlapped with the isotopic ion of singly-charged y₄ at m/z 514.286 (Fig. 2c). A weak signal at m/z 514.259 was also observed in the ¹⁵N-Aba derivatized peptide, but this is likely due to the remaining ¹⁴N-Aba (∼3%) in ¹⁵N-Aba. Discrimination between these two ions is not possible when a low-resolution mass spectrometer is used, where the m/z shift of the b₇ ion cannot be detected. Combined with the advantages of high-resolution mass spectrometry, differential ¹⁴N/¹⁵N labeling of Aba promises to greatly facilitate the unambiguous interpretation of peptide fragmentation, leading to a more reliable de novo sequencing method.

Table 1. List of fragment ions showing mass shifts in the product ion spectra between ¹⁴N- and ¹⁵N-Aba derivatized angiotensin I.

m/z	14N-Aba	262.083	418.184	465.733	514.259	517.252	587.793	656.323	680.315	793.399	930.457	1027.509	1174.579
	15N-Aba	263.079	419.181	466.231	514.758	518.248	588.292	656.821	681.312	794.396	931.455	1028.507	1175.577
Difference		0.996	0.997	0.498	0.499	0.996	0.499	0.498	0.997	0.997	0.998	0.998	0.998
Assignment		b1	b2	b62+	b72+	b3	b82+	b92+	b4	b5	b6	b7	b8

Evaluation of the effect of ¹⁴N/¹⁵N-Aba derivatization using various model peptides

To evaluate the usefulness of ¹⁴N/¹⁵N-Aba derivatization, we performed sequencing of various types of peptides that contain basic residues, which often provide complicated product ion spectra. Peptide 2 (TDVNGDGR), a model for tryptic digests, contains an Arg residue at the C-terminus. As tryptic digests possess an Arg or Lys residue at the C-terminus, y-ions are generally favorably produced without derivatization. However, the assignment of y-ions was not straightforward in the case of peptide 2, because several b-ions were also observed at significant intensities (Supplemental Fig. 1b). After derivatization with ¹⁴N-Aba, a complete series of b-ions as well as y-ions were detected in the product ion spectrum, but the intensities of several y-ions were still high enough to hamper the straightforward assignment of fragment ions (Fig. 3a). By comparing the spectrum with that obtained using ¹⁵N-Aba derivatization, all of the b-ions were clearly identified based on their mass shifts, allowing straightforward de novo sequencing (Fig. 3b).

Fig. 3. Product ion spectrum of the ¹⁴N-Aba-derivatized peptide 2 (a), and a list of fragment ions produced, showing mass shifts in the spectra between ¹⁴N- and ¹⁵N-Aba derivatization (b). Doubly-charged ions (m/z 490.21 and 490.71 for ¹⁴N- and ¹⁵N-Aba, respectively) were selected as a precursor.

Peptide 3 (INSAKDDAAGLAIA) has a relatively long sequence, and contains a Lys residue in the middle of the sequence. Without derivatization, the intensity of most of the b- and y-ions of this peptide were relatively low, making de novo sequencing difficult (Supplemental Fig. 1c). After derivatization with ¹⁴N-Aba, all of the b-ions were generated, but some ambiguity continued to exist as a result of the formation of several y-ions particularly in the high m/z region because of the presence of a Lys residue in the N-terminal part (Fig. 4a). However, a comparison of the spectrum with that obtained using ¹⁵N-Aba derivatization allowed b-ions to be unambiguously identified, as shown in Fig. 4b.

Fig. 4. Product ion spectrum of the ¹⁴N-Aba-derivatized peptide 3 (a), and a list of fragment ions produced, showing mass shifts in the spectra between ¹⁴N- and ¹⁵N-Aba derivatization (b). Doubly-charged ions (m/z 738.38 and 738.88 for ¹⁴N- and ¹⁵N-Aba, respectively) were selected as a precursor.

Peptide 4 (SFLLRN-NH₂) contains an Arg residue in the middle of the sequence, and its C-terminus is amidated. These substructures are often found in naturally occurring bioactive peptides. The underivatized peptide did not provide all of the b- and y-ion series (Supplemental Fig. 1d). After ¹⁴N-Aba derivatization, a complete series of b-ions was generated, but the intensity of the y-ions remained high because of the presence of an Arg residue at the C-terminal part. Again, the parallel use of ¹⁵N-Aba facilitated the unambiguous identification of b-ions for this peptide (Fig. 5).

Fig. 5. Product ion spectrum of ¹⁴N-Aba-derivatized peptide 4 (a), and a list of fragment ions produced, showing mass shifts in the spectra between ¹⁴N- and ¹⁵N-Aba derivatization (b). Doubly-charged ions (m/z 447.75 and 448.25 for ¹⁴N- and ¹⁵N-Aba, respectively) were selected as a precursor.

Peptide 5 (AAGLQIANRLTS) also contains an Arg residue in the middle of the sequence, and gave a complicated fragmentation pattern without Aba derivatization (Supplemental Fig. 1e). After ¹⁴N-Aba derivatization, all of the b-ions except the b₈ ion were observed in the product ion spectrum (Fig. 6). In this case, the presence of an Arg residue at the fourth position from the C-terminus enhanced the intensity of y₄ to y₁₁ ions. Since amide bond cleavage between the eighth and ninth residues was hampered by the neighboring Arg residue, the b₈ ion was not generated in significant intensities.¹⁸⁾ In such cases, however, the corresponding y-ions can compensate for the missing b-ions. By comparing the product ion spectra, the y₄ ion, which is complementary to the b₈ ion, was identified as a fragment ion without m/z shifts between the ¹⁴N- and ¹⁵N-Aba derivatized peptide 5, and the full sequence was successfully determined. The adverse effects of Arg residues on fragmentation can also be diminished by modifying the guanidino group of the Arg side chain to decrease its proton affinity.²⁰⁾

Fig. 6. Product ion spectrum of ¹⁴N-Aba-derivatized peptide 5 (a), and a list of fragment ions produced, showing mass shifts in the spectra between ¹⁴N- and ¹⁵N-Aba derivatization (b). Doubly-charged ions (m/z 680.87 and 681.37 for ¹⁴N- and ¹⁵N-Aba, respectively) were selected as a precursor.

Application to de novo sequencing of proteins

To explore the potential of the method described in this study, we applied it to the de novo sequencing of a protein digest, using lysozyme as a model protein. The protein was digested with trypsin, and the resulting peptide mixture was derivatized individually with light or heavy Aba. Each derivatized peptide mixture was analyzed by LC/MS/MS using the data-dependent scan mode. As a result, seven peptides were detected as ions with mass shifts, and their product ion spectra were obtained (Table 2). Among them, the full sequences of four peptides were easily determined by using only information for the b-ions. For the other three peptides, some of the b-ions were not clearly observed, but their full sequences were successfully determined with the aid of their complimentary y-ions, as in the case of peptide 5. Sequence coverage was not high in this analysis because some digested peptide fragments are too large or too small to permit their analysis MS/MS. Further optimization of sample preparations, such as the utilization of different enzymes and desalting methods, should improve the coverage. Consequently, the method described in this study promises to facilitate the straightforward de novo sequencing of peptides or proteins from organisms with unsequenced genomes.

Table 2. Result of de novo sequencing of tryptic digests of lysozyme.

Sequence1	Precursor ion (m/z)2	Position
	376.71	1–5
	377.21
	340.82	15–21
	341.15
	736.34	22–33
	736.84
	787.85	34–45
	788.35
	950.44	46–61
	950.94
	570.22	62–68
	570.72
	596.30	117–125
	596.80

¹Only the ions used for sequence determination were indicated. ²Upper: ¹⁴N-Aba derivatized peptide, bottom: ¹⁵N-Aba derivatized peptide. Doubly charged ions were selected as a precursor, except for fragment (15–21), for which a triply charged ion was used.

CONCLUSION

In this study, we investigated de novo peptide sequencing strategies using differential ¹⁴N/¹⁵N-labeling coupled with N-terminal charge derivatization with Aba. In a previous study, we showed that Aba derivatization confers a well-balanced proton affinity to the peptide N-terminus, which serves as a donor of mobile protons and also an acceptor of protons for the preferential generation of b-ions.⁹⁾ In addition to this effect, differential ¹⁴N/¹⁵N-labeling using Aba enabled the unambiguous identification of b-ions for all peptides used in this study, which provided more reliable sequencing results. Furthermore, the use of a high-resolution mass spectrometer contributed to the accurate assignment of each fragment ion, particularly when fragment ions with similar m/z values were present in the spectrum. The method developed in this study allows the easy and reliable de novo sequencing of peptides, which is useful in peptidomics and proteomics studies.

Since the difference in molecular mass between ¹⁴N- and ¹⁵N-labeled Aba is only 1 Da, MS/MS analysis was performed individually for each derivatized peptide. However, the introduction of two or more isotopes (¹⁵N or ¹³C) into Aba would enable a single analysis of peptides derivatized with a mixture of non-labeled and isotope-labeled Aba. In addition, the detection of mass shifts was performed manually in this study, which required some time in terms of assigning b-ions. To further increase the efficiency of de novo sequencing using this method, the development of a software tool for automated detection of mass shifts due to ¹⁵N-labeling would be required to assign key fragments in a more convenient manner.

Please cite this article as: Y. Nihashi, et al., Differential ¹⁴N/¹⁵N-Labeling of Peptides Using N-Terminal Charge Derivatization with a High-Proton Affinity for Straightforward de novo Peptide Sequencing, Mass Spectrometry, 2: A0024 (2013); DOI: 10.5702/massspectrometry.A0024