Quantification of Tryptic Peptides in Quadrupole Ion Trap Using High-Mass Signals Derived from Isotope-Coded N-Acetyl Dipeptide Tags
Isotope-labeled N-acetyl dipeptides (Ac-Xxx-Ala) are coupled to the primary amines of tryptic peptides and then analyzed by tandem mass spectrometry. Amide bond cleavage between Xxx and Ala provides both low- and high-mass isotope-coded signals for quantification of peptides. Especially, facile cleavage at the modified lysine side chain yields very strong high-mass quantitation signals in a noise-free region. Tagging tryptic peptides with isobaric N-acetyl dipeptides is a viable strategy for accurate quantification of proteins, which can be used with most quadrupole ion trap mass spectrometers carrying the 1/3 mass cut-off problem.
Key wordsIsobaric tag Quantitative proteomics High-mass quantitation signal Quadrupole ion trap mass spectrometry N-acetyl dipeptide
Isobaric tags have been widely used in the mass spectrometry (MS)-based quantification of proteins and peptides [1, 2, 3]. Peptides of interest are differentially labeled with isotope-coded tags and the resulting isotopomeric precursor ions are analyzed by tandem mass spectrometry (MS/MS) to identify the peptide sequence and to simultaneously quantify the amounts of differentially-labeled peptides. All of the isobaric tags available to date are designed to report quantitation signals in the m/z 100–250 region [4, 5, 6, 7, 8, 9, 10, 11, 12, 13]. However, this low-mass region is inaccessible by conventional quadrupole ion trap (QIT) because of the ~1/3 mass cut-off problem in ion trapping during resonant excitation of the precursor ion . Moreover, the m/z 100–250 region is ion-rich due to other small fragment peaks overlapping in this region . Although a couple of novel ion trapping methods have been recently developed to overcome this low-mass cut-off problem [15, 16, 17, 18], they are not directly applicable to most QIT mass spectrometers. Alternatively, it would be convenient to come up with a chemical method that provides quantitation signals in a high-mass, noise-free region. Herein, we present results from N-acetyl dipeptides coupled to the primary amines of tryptic peptides, which demonstrates the significance of high-mass quantitation signals for quantification of proteins.
H/D-isotopes are incorporated into the CH3/CD3 group either in acetyl or in alanine to report 2-plex quantitation signals separated by 3 Da. The variable amino acid Xxx can be chosen either from a natural amino acid  or from an artificial amino acid  to diversify MBITs (Xxx-tags). The variation of side chain RT shifts quantitation signals and modulates chemical properties of the tagged peptides.
To investigate quantitation signals from MBIT-linked peptides, we prepared two model peptides having the same sequence except for the amino acid at the C-terminus, LISFYAGR (1) and LISFYAGK (2). Their sequences were arbitrarily chosen from natural amino acids, excluding histidine, proline, aspartic acid, and glutamic acid to avoid specific fragmentation pathways . Of the various MBITs, the Gln-tag (Ac-Q-A) was conjugated to the model peptides. Amine-reactive coupling produced the arginine-terminated peptide 1 with one Gln-tag at the N-terminus (11) and the lysine-terminated peptide 2 with two Gln-tags (22), one at the N-terminus and another at lysine (Scheme 1). Superscript denotes the number of tags attached to the peptide. Resulting peptides were analyzed by using matrix-assisted laser desorption ionization (MALDI)-time-of-flight (TOF) and electrospray ionization (ESI)-QIT mass spectrometers. Tandem mass analyses of tagged peptides yield both low-mass (L/HbS) and high-mass (L/HyS) quantitation signals through the Gln–Ala peptide bond cleavage (see reference  for the nomenclature of peptide fragmentation). The elution profile of MBIT-linked peptides in liquid chromatography (LC) was examined with MBIT-linked peptides 11 and 22. Meanwhile, the performance of MBITs on quantification of proteins in a QIT mass spectrometer was demonstrated with a protein mixture containing bovine serum albumin (BSA), horse myoglobin, and human ubiquitin.
Preparation of the acid form of mass-balanced H/D-isotope tag (L/HMBIT) is described elsewhere . N-hydroxysuccinimide (NHS), hydroxylamine hydrochloride, N,N'-dimethylformamide (DMF, HPLC grade), trifluoroacetic acid (TFA), formic acid, and 4-hydroxy-α-cyano-cinnamic acid (HCCA) were purchased from Sigma-Aldrich (St. Louis, MO, USA). N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC) hydrochloride and modified trypsin were obtained from Pierce (Rockford, IL) and Promega (Madison, WI, USA), respectively. Model peptides, LISFYAGR (1), LISFYAGK (2), and AARLISFYAGK (3), were purchased from Peptron, Inc. (Daejeon, Korea). BSA was obtained from Merck (Darmstadt, Germany). Both myoglobin and ubiquitin were purchased from Sigma-Aldrich.
2.2 Preparation of Succinimidyl Ester of MBIT (MBIT-OSu)
L/HMBIT was dissolved in DMF to a final concentration of 80 mM. A solution containing EDC (140 mM) and NHS (160 mM) was prepared in DMF. The L/HMBIT-OH solution (6 μL) was mixed with the EDC/NHS solution (2 μL) in a 0.6 mL Eppendorf-tube to prepare an O-succinimidyl ester (OSu) form of MBIT. The L/HMBIT reagent was added in excess in order to consume EDC, thereby preventing the formation of active esters of target peptides in the subsequent peptide conjugation. A freshly-prepared L/HMBIT-OSu solution (35 mM) was promptly used for peptide conjugation without further purification.
2.3 Preparation of MBIT-Linked Peptides 11, 22, and 32
Each model peptide was dissolved in sodium bicarbonate buffer (100 mM, pH 8.1) to a final concentration of 250 μM. Each peptide solution (8 μL) was mixed with the LMBIT-OSu or HMBIT-OSu solution (8 μL) in an Eppendorf-tube. After stirring the mixture at room temperature for 2 h, hydroxylamine (100 mM in 100 mM sodium bicarbonate buffer, 8 μL) was added to the mixture in order to minimize side reactions on hydroxyl groups and to consume excess LMBIT-OSu or HMBIT-OSu reagent . The reaction was carried out for more than 6 h and terminated with 10% TFA (4 μL). The final volume of each sample solution was 28 μL. As-prepared MBIT-linked peptide solution was purified with ZipTip-μC18 column (Millipore, San Diego, CA, USA), and then further diluted to a proper concentration before MS and MS/MS.
2.4 Preparation of MBIT-Linked Peptide 21
L/HMBIT-linked peptide 32 was digested with trypsin to obtain peptide 21 conjugated with only one MBIT at lysine. L/HMBIT-linked peptide 32 was diluted in aqueous sodium bicarbonate buffer (80 mM) to the final concentration of 50 μM, and an aliquot (95 μL) was mixed with a trypsin solution (0.1 μg μL–1, 5 μL) and incubated at 37 °C for 12 h. Trypsin reaction was terminated by adding 10% TFA (10 μL). The tryptic digest of peptides was dried under vacuum, reconstituted in distilled water (10 μL), purified with ZipTip-μC18 column, and then diluted to a proper concentration before MS and MS/MS.
2.5 Quantitation Linearity Measurement
LMBIT- and HMBIT-linked peptides were diluted with acetonitrile/water/formic acid (50/50/0.5, vol/vol) to a final concentration of 10 μM and then mixed together in various volume ratios: LMBIT/HMBIT = 1, 4, 9, 16, 25, 36, 49, and 64. The volume of each mixture was 300 μL and the lowest concentration of HMBIT-linked peptide was approximately 154 nM. This premixed sample was analyzed by ESI-MS and MS/MS. Each MS/MS spectrum was acquired for 15 s.
2.6 MALDI and ESI Mass Analyses of Model Peptides
For MALDI, MBIT-linked peptides (1 μM) were mixed with HCCA (5 μg μL–1) in acetonitrile/water/TFA (50/50/0.1, vol/vol), and then the mixture (0.5 μL) was loaded on a MALDI plate. MALDI-MS and MS/MS were performed using a 4700 Proteomics Analyzer (TOF/TOF; AB SCIEX, Foster City, CA, USA). Air (1.5 × 10–7 torr) was used as collision gas for TOF/TOF. For ESI, MBIT-linked peptides (5 μM) were dissolved in acetonitrile/water/formic acid (50/50/0.5, vol/vol). ESI-MS and MS/MS were performed using a high-capacity ion trap (HCT, Bruker Daltonics, Germany). The sample solution was loaded on a syringe pump, and then sprayed through an electrospray emitter at the flow rate of ~1 μL min–1. Electrospray voltage was 3.5 kV under N2 nebulizer gas (5 psi). Helium (~1.5 × 10–5 torr) was used as collision gas for QIT-MS/MS.
2.7 LC Elution Profiles of MBIT-Linked Peptides
LC elution profiles were obtained through multiple-reaction monitoring (MRM) mode using a 2000 Q-TRAP triple-quadrupole mass spectrometer (AB SCIEX, Forster City, CA, USA) connected to reverse-phase nano-LC system (LC Packings, Sunnyvale, CA, USA). LMBIT- and HMBIT-linked peptides 11 and 22 were mixed together in a 1:1 ratio and diluted to the final concentration of 20 μM with a 0.1% TFA solution. The mixture was analyzed by nano-LC-ESI-MRM. The precursor ion was selected in Q1 and fragmented in Q2. Of the fragment ions, quantitation signal ions (either L/HbS or H/LyS) were selected in Q3 at unit resolution and their abundances were recorded every 0.8 s. LC running conditions are described in the Supplementary Material.
2.8 Quantification of Proteins
Three proteins (BSA, horse myoglobin, and human ubiquitin) were mixed in two different ratios (sample A and B). Sample A contains 4.0 mg mL–1 of BSA, 2.0 mg mL–1 of myoglobin, and 0.2 mg mL–1 of ubiquitin in aqueous sodium bicarbonate buffer (80 mM), whereas sample B contains 2.0 mg mL–1 of BSA, 0.5 mg mL–1 of myoglobin, and 0.4 mg mL–1 of ubiquitin. Each protein mixture (80 μL) was digested with trypsin (0.1 μg μL–1 of trypsin, 20 μL) for 18 h at 38 °C. Tryptic peptides of sample A and B (20 μL each) were mixed with LMBIT-OSu and HMBIT-OSu (20 μL each), respectively. Each MBIT–peptide mixture was stirred at room temperature for 2 h, and then treated with hydroxylamine (100 mM in 100 mM sodium bicarbonate buffer, 20 μL) for 6 h. The conjugation reaction was terminated with 10% TFA (10 μL). The MBIT-linked sample A and B (10 μL each) were mixed together, dried under vacuum, and reconstituted in 0.5% formic acid (10 μL). An aliquot of the mixture (5 μL) was analyzed by nano-LC-ESI-MS and MS/MS using an LTQ XL linear ion trap mass spectrometer (Thermo Scientific, Waltham, MA, USA) connected to a nano-LC system (Eksigent, Dublin, CA, USA). LC running conditions are described in the Supplementary Material. Peptide sequencing and protein identification was carried out by Mascot MS/MS ion search with a custom-built protein database containing 26,269 bovine, horse, and human proteins. Of the MS/MS peak lists, the precursor ions that yielded a tagging-signature ion ytagn+ indicating the loss of neutral Ac-Q-A (−244.1/n) were fed into the Mascot search. The m/z tolerance was 0.5 Da, and the MBIT tagging was considered as a variable modification at the N-terminus and lysine.
3 Results and Discussion
3.1 MALDI-TOF Mass Analyses of MBIT-Linked Peptides 11 and 22
3.2 ESI-QIT Mass Analyses of 11 and 22
Collisional activation of [11 + H]+ in QIT yields singly protonated L/HyS ions at m/z 997.6/1000.6 with a number of sequence ions, but no peaks in the low-mass cut-off range (Figure 2b). The relative abundance of L/HyS is only 1.8% of the total fragment ions. [11 + H]+ mainly undergoes a loss of ammonia from the protonated arginine  or a loss of water from the serine or tyrosine side chain [22, 23]. The tandem mass analysis of [11 + 2H]2+ also provides singly protonated L/HyS ions, but no peaks below the low-mass cut-off (Figure 2c).
In contrast to [11 + H]+, a series of b- and y-type sequence ions are abundantly produced from [11 + 2H]2+, suggesting that although one proton is fixed at arginine, another proton is mobile to facilitate the peptide backbone fragmentation [24, 25]. Nevertheless, the relative abundance of singly protonated L/HyS is only 2.2% of the total fragment ions. The measured [HyS]/[LyS] ratios from [11 + H]+ and [11 + 2H]2+ are 0.87 and 0.55, respectively, both of which deviate significantly from the premixed LMBIT/HMBIT ratio of 1.0. The singly protonated LyS ion from both [11 + H]+ and [11 + 2H]2+ overlaps with the isotope pattern of the b7 ion that results from a loss of the C-terminal arginine (174.1 Da) (inset of Figure 2b and c). By subtracting the b7 isotope, we can obtain the correct [HyS]/[LyS] ratio of 1.05 and 1.02 from [11 + H]+ and [11 + 2H]2+, respectively. In the case of Gln-tag, the bñ1 ions of arginine-terminated peptides can always interfere with singly protonated LyS ions. Although one can avoid this accidental overlap by using other MBITs such as His- and Phe-tags, potential overlap between quantitation signals and other ions needs to be examined carefully when various MBITs are employed to quantify complex peptide mixtures.
By contrast, the QIT-MS/MS spectra of [22 + H]+ and [22 + 2H]2+ report strong L/HyS ions at m/z 1213.7/1216.7 with the 1:1 intensity ratio (Figure 2d and e). L/HbS ions at m/z 171.1/174.1 are not detected due to the low-mass cut-off. In the case of [22 + H]+, both water- and tag-loss ion peaks are abundant, whereas sequence ions are not. For [22 + 2H]2+, sequence ions are quite abundant. The relative abundance of a pair of L/HyS ions from [22 + H]+ is 48% and that from [22 + 2H]2+ is 18%. Relative abundances of sequence ions, water-loss and tag-loss ion peaks from [22 + H]+ are 11%, 21%, and 20%, respectively, whereas those from [22 + 2H]2+ are 65%, 5%, and 12%, respectively. Most importantly, high-mass L/HyS ions appear in a noise-free region without any overlap with sequence ions as all of the b1–b7 and y1–y7 sequence ions fall between btag and ytag ions. These singly protonated L/HyS ions are derived from a loss of Ac-Q from [22 + H]+ and a loss of [Ac-Q + H]+ from [22 + 2H]2+. Meanwhile, doubly protonated L/HyS2+ ions at m/z 607.3/608.8 result from a loss of neutral Ac-Q from [22 + 2H]2. These L/HyS2+ ions can also be used as quantitation signals.
Evidently, reporting both bS and complementary yS signal ions is a unique feature of N-acetyl dipeptide tags [10, 11]. To the contrary, other isobaric tags based on piperazine [4, 5, 8, 9], piperidine [6, 7], or tertiary amine [12, 13] derivatives report strong low-mass quantitation signals, but no complementary high-mass signals.
3.3 ESI-QIT MS3 Analyses of 22 and MS2 Analyses of 21
To substantiate the major formation of L/HyS(K) ions through the Gln–Ala cleavage at the lysine side chain, we prepared peptide 21 having only one Gln tag at lysine. For this, another model peptide AARLISFYAGK (3) was conjugated with two Gln-tags, one at the N-terminus and another at the lysine side chain, and then digested with trypsin to cleave off Ac-QAAAR. The QIT-MS/MS spectra of [21 + H]+ and [21 + 2H]2+ from the 1:1 mixture of LMBIT- and HMBIT-linked peptide 21 show strong H/LyS(K) ion peaks in the high-mass region (Figure 3c and d), which are almost identical to the fragmentation patterns of [22 + H]+ and [22 + 2H]2+ (Figure 2d and e), respectively. In the case of [21 + 2H]2+, the complementary low-mass H/LbS ions are also detected above the low-mass cut-off. Apparently, N-acetyl dipeptide appended to the lysine side chain is a primary source of strong L/HyS signals from both 22 and 21.
3.4 Quantitation Linearity Using High-Mass H/LyS Signals from Lysine-Tagged Peptides
3.5 LC Elution Profiles of LMBIT- and HMBIT-Linked Peptides
3.6 Quantification of a Protein Mixture
We present a strategy for accurate quantification of peptides and proteins in quadrupole ion trap using high-mass isotope-coded signals derived from N-acetyl dipeptide tags. Differentially labeled MBIT-linked tryptic peptides are co-eluted in LC with little H/D isotope effects. Isobaric MBIT-linked peptides result in low-mass b-type signal ions in the MALDI-TOF/TOF spectra and high-mass y-type signal ions in the ESI-QIT-MS/MS spectra, which is complementary to each other. Of the MBIT-linked tryptic peptides, singly protonated lysine-terminated peptides yield strong high-mass quantitation signal ions whose relative abundances are nearly 50% of the total fragment ions. N-acetyl dipeptide tags allow accurate identification and quantification of tryptic peptides, regardless of the type of mass spectrometers.
The authors are grateful for the support from the Functional Proteomics Center (grant FPR08A1-040). They also thank to the Ulsan National Institute of Science and Technology Central Research Facility (UCRF), the Peptide Library Support Facility (PLSF) at the POSTECH Biotechnology Center, and D. H. Hwang for technical assistance.
- 4.Ross, P.L., Huang, Y.N., Marchese, J.N., Williamson, B., Parker, K., Hattan, S., Khainovski, N., Pillai, S., Daniels, S., Purkayastha, S., Juhasz, P., Martin, S., Barlet-Jones, M., He, F., Jacobson, A., Pappin, D.: Multiplexed Protein Quantitation in Saccharomyces cerevisiae Using Amine-Reactive Isobaric Tagging Reagents. Mol. Cell. Proteomics 3, 1154–1169 (2004)CrossRefGoogle Scholar
- 5.Choe, L., D’Ascenzo, M., Relkin, N.R., Pappin, D., Ross, P., Williamson, B., Guertin, S., Pribil, P., Lee, K.H.: 8-Plex Quantitation of Changes in Cerebrospinal Fluid Protein Expression in Subjects Undergoing Intravenous Immunoglobulin Treatment for Alzheimer's Disease. Proteomics 7, 3651–3660 (2007)CrossRefGoogle Scholar
- 8.Li, S., Zeng, D.: CILAT–A New Reagent for Quantitative Proteomics. Chem. Commun. 2181–2183 (2007)Google Scholar
- 12.Zeng, D., Li, S.: Revival of Deuterium-Labeled Reagents for Protein Quantitation. Chem. Commun. 3369–3371 (2009)Google Scholar
- 18.Köcher, T., Pichler, P., Schutzbier, M., Stingl, C., Kaul, A., Teucher, N., Hasenfuss, G., Penninger, J.M., Mechtler, K.: High Precision Quantitative Proteomics Using iTRAQ on an LTQ Orbitrap: A New Mass Spectrometric Method Combining the Benefits of All. J. Proteome Res. 8, 4743–4752 (2009)CrossRefGoogle Scholar