Mass spectrometry signal enhancement by reductive amination

International Journal of Mass Spectrometry 387 (2015) 16–23

Contents lists available at ScienceDirect

International Journal of Mass Spectrometry journal homepage: www.elsevier.com/locate/ijms

Mass spectrometry signal enhancement by reductive amination Meng-Chieh Liu a , Yi-Reng Lin b , Mei-Fang Huang a , De-Cheng Tsai d,∗∗ , Shih-Shin Liang a,c,e,∗ a

Department of Biotechnology, College of Life Science, Kaohsiung Medical University, Kaohsiung, Taiwan Department of Biotechnology, Fooyin University, Kaohsiung, Taiwan Institute of Biomedical Science, National Sun Yat-Sen University, Kaohsiung, Taiwan d Division of Urology, Ten Chan General Hospital, Taoyuan, Taiwan e Center for Research, Resources and Development, Kaohsiung Medical University, Kaohsiung, Taiwan b c

a r t i c l e

i n f o

Article history: Received 10 March 2015 Received in revised form 23 June 2015 Accepted 24 June 2015 Available online 2 July 2015 Keywords: Reductive amination Signal enhancement Mass spectrometry Multiple reaction monitoring (MRM)

a b s t r a c t Organosulfur compounds (OSCs) subjected to reductive amination in the presence of formaldehyde exhibited increased mass spectrometry signal intensities. In this study, four OSCs including S-allyl cysteine, S-allylcysteinine sulfoxide, S-methylcysteine and S-ethylcysteine were generated using isotopic formaldehyde, and mass spectrometry signal intensities of modified and unmodified OSCs were compared. This comparison involved tandem mass spectrometry infusion and detection techniques, such as selected ion monitoring (SIM) and multiple reaction monitoring (MRM). The signal intensities of modified OSCs increased from 2.6 to 39.2 fold by infusion, from 50.0 to 479.6 fold by SIM, and from 146.4 to 2494.8 fold by MRM. Compounds bearing primary amine groups reacted with formaldehyde in high yield and underwent reductive amination in the presence of sodium cyanoborohydride to form a dimethyl group on these amine groups. The modified OSCs showed enhanced intensities because the electron donating dimethyl groups increase their basicity. This signal enhancement is expected to improve the limit of detection in absolute quantification and structural characterization. Therefore, reductive amination involving primary amine groups may find application in the enhancement of mass spectrometry signal intensities. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Comparative proteomics typically relies on stable isotope labeling and tandem mass spectrometry (MS) coupled with the shotgun approach or two-dimensional gel electrophoresis to achieve global protein identification and profiling [1–3]. Amine containing metabolite profile including 20 amino acids and 15 amines has been generated by stable isotope labeling through the reductive amination of primary amine groups in the presence of formaldehyde [4]. Glycosylation variants have been relatively quantified using isotopic formaldehyde [5]. Reductive amination, described as stable isotope dimethyl labeling, has also been utilized to enhance the signal intensity of

∗ Corresponding author at: Department of Biotechnology, College of Life Science, Kaohsiung Medical University, No. 100, Shih-Chuan 1st Road, Kaohsiung 80708, Taiwan. Tel.: +886 7 3121101 2153; fax: +886 7 3125339. ∗∗ Corresponding author at: Division of Urology, Ten Chan General Hospital, No. 155, Yanping Road, Taoyuan 32043, Taiwan. Tel.: +886 937 069458. E-mail addresses: [email protected] (D.-C. Tsai), [email protected] (S.-S. Liang). http://dx.doi.org/10.1016/j.ijms.2015.06.010 1387-3806/© 2015 Elsevier B.V. All rights reserved.

saccharides by matrix-assisted laser desorption ionization MS [6]. Reductively aminated oligosaccharides have been detected by high-performance liquid chromatography (HPLC)/electrospray ionization (ESI) MS [7,8] and capillary electrophoresis (CE) [9], and have shown MS signal enhancement. However, organic synthetic conditions are sometimes unsuitable for protein and peptide analysis. In the early period, however, mobile phase composition adjustment methods were popular for proton transfer efficiency evaluation. Proton scavengers, such as ammonium, methylamine, trimethylamine, diethylamine and triethylamine [10–12], were compared because proteins exhibiting a high charge state during mass spectrometry fragmentation by collision-induced dissociation were more sensitive than those presenting a lower charge state [13]. However, the detection of post-translational modifications, such as protein phosphorylation and glycosylation, and the identification of signal charge states require protein or peptide modification associated with ion charge enhancement. Such a charge enhancement has been observed by electron transfer dissociation (ETD) tandem MS [14,15]. Furthermore, the orifice diameter of the spray tip was altered

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

for high resolution nanospray MS [16]. Due to its high surface tension and low relative volatility, m-nitrobenzyl alcohol is an excellent solvent, and is often added to the mobile phase to enhance the ion-charging state [17–19]; however, its mechanism is unclear. In this way, it has proven useful in post-translational modification detection by ETD tandem MS [14,15]. However, higher m-nitrobenzyl alcohol concentrations result in shorter retention times. Mobile phase addition has produced a less obvious charge shift at the peptide level than direct protein or peptide modification by covalent bonding. This direct modification can generate trialkylammonium ions or blocked C-terminal position in proteins [20,21], but as it requires organic solvents, this can be difficult to handle [6,20,21]. Other methods have also been used to provide protein ion-charge distribution estimates using protein surface areas [22] and in different buffer solutions, such as triethylammonium acetate and triethylammonium bicarbonate to reduce the charge state [23,24]. In this study, a reductive amination approach was developed to enhance the quantitative signal for four representative organosulfur compounds (OSCs). Reductive amination has previously been utilized in comparative proteomics where experimental and control samples were allowed to react with fromaldehyde-D2 and H2 [3]. Here, four organosulfur compounds, including S-allyl cysteine (SAC), S-allylcysteinine sulfoxide (alliin), S-methylcysteine (SMC), and S-ethylcysteine (SEC), were subjected to dimethyl labeling to generate four fromaldehyde-D2 and four formaldehyde-H2 modified OSCs. Signal intensities of unmodified, D2 -fromaldehyde modified, and H2 -fromaldehyde modified OSCs were determined by infusion, selected ion monitoring (SIM), and multiple reaction monitoring (MRM). Modified OSCs exhibited 146.4- to 2494.8-fold signal intensity enhancement compared with unmodified samples. This new method may be utilized for small biomolecules that contain primary amine groups and may be adapted for absolute quantification. 2. Materials and methods 2.1. Chemicals and materials SAC was purchased from Tokyo Chemical Industry Co. (Tokyo, Japan). SMC, SEC, alliin, formaldehyde-H2 (36.5%–38% in H2 O), and formic acid (FA, 98%–100%) were purchased by Sigma (St. Louis, MO, USA). Sodium acetate (NaOAc), sodium cyanoborohydride (NaBH3 CN), and trifluoroacetic acid (TFA) were obtained from Sigma–Aldrich (St. Louis, MO, USA). Acetonitrile (MeCN, LC/MS grade) was purchased from Merck (Seelze, Germany). Hydrochloric acid, sodium hydroxide, and ethanol were acquired from J.T. Baker (Phillipsburg, NJ, USA). Formaldehyde-D2 (20% solution in D2 O) was obtained from Isotec Corp. (Miamisburg, OH, USA). Deionized water was produced using a Millipore system at a resistance of 18.2 M (Bedford, MA, USA). 2.2. Instrumentation Separation and detection were performed by ultra-highpressure liquid chromatography coupled with a tandem mass spectrometer (UHPLC–MS/MS) using a Thermo Finnigan Acella 1250 UHPLC system (Thermo Fisher Scientific Inc., Waltham, MA, USA). The MS system comprised a triple quadrupole MS (Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with an ESI ion source in positive ion mode. MS measurements were conducted at an applied voltage of 3000 V, and vaporization and capillary temperatures of 300 ◦ C and 350 ◦ C, respectively. Sheath gas and auxiliary gas pressure were set at 35 and 10, respectively, while the collision pressure was maintained at 1.5, and the collision energy

17

was adjusted between 18 and 25 V. MRM transitions were set to the m/z values of precursor and product ions (Table 1). UHPLC system control and MS data acquisition were achieved using the Xcalibur software (version 2.2, Thermo Finnigan Inc., San Jose, CA). Samples were sequentially injected into a 10 ␮L loop using an Acella 1250 autosampler and separated on a Shiseido CAPCELL PAK C18 MG II column (i.d. 1.5 mm × 150 mm, 3 ␮m, Tokyo, Japan). Mobile phases comprised 0.1% FA in water (A) and 0.1% FA in 100% MeCN (B), and the UHPLC flow rate was set to 200 ␮L/min. The UHPLC linear gradient was set as follows: 5% (B) for 2 min from injection, 5%–40% (B) for 7 min, 40%–95% (B) for 5 min, and held at 95% (B) for 2 min. 2.3. MS infusion analysis of organosulfur compounds Organosulfur compounds were dissolved in ethanol and 500 ␮g/mL solutions were prepared in de-ionized water. Individual OSC solutions (20 ␮L) were transferred into three tubes and their pH was adjusted to 5.6 using sodium acetate buffer (180 ␮L). First and second samples were allowed to react for 5 min with 10 ␮L 4% formaldehyde-H2 and 10 ␮L 4% formaldehyde-D2 solutions, respectively. Modified samples were reduced for 1 h using 0.6 M NaBH3 CN and their pH was adjusted to 2–3 using 10% TFA before mixing with their unmodified counterparts. Individual OSCs mixtures were infused into the mass spectrometer for m/z characterization. 2.4. UHPLC–MS/MS analysis In the triple quadrupole MS instrument, OSC fragmentations (precursor ions) were detected via MRM and SIM scanning modes to establish the MRM transitions. In the SIM mode, m/z values were set to 136, 164 and 168 for unmodified, formaldehyde-H2 , and formaldehyde-D2 modified SMC, respectively. Meanwhile, m/z values for other OSCs, such as SEC, SAC, and alliin, were set according to the precursor values listed in Table 1. Establishment of MRM transitions acquired the method in MS, and this method can refer to the fragmented spectra of SIM. Finally, MRM transitions were set according to MS/MS data as follows: 136 > 73 and 136 > 119 for SMC, 164 > 73 and 164 > 119 for H2 -labeled SMC, 168 > 73 and 168 > 119 for D2 -labeled SMC. Other MRM transitions are shown in Table 1. The statistics analysis was conducted using the Xcalibur Thermo LCquan software (version 2.7, Thermo Finnigan Inc., San Jose, CA). 3. Results and discussion 3.1. Reductive amination of organosulfur compounds Four OSCs were subjected to reductive amination in the presence of isotopic formaldehyde to evaluate the effectiveness of this labeling approach (Fig. 1). Signal enhancement was achieved by reductive amination of OSCs involving formaldehyde. The formation of two methyl groups exhibiting electron-donating properties increased the amine basicity. Specifically, individual isometric OSCs were in three Eppendrof tubes for reductive amination. After coupling with isotopic formaldehyde in pH 5.6 sodium acetate and subsequent NaBH3 CN reduction, unmodified, H2 -labeled, and D2 -labeled sample solutions were acidified and diluted using TFA before being combined to determine signal differences. The mixtures were analyzed by UHPLC–MS/MS (Fig. 2) using infusion, SIM, and MRM scan modes. 3.2. Detection of signal enhancement by infusion Infusion spectra clearly showed signal enhancement in all formaldehyde-H2 and formaldehyde-D2 modified OSCs (Fig. 3). SMethyl cysteine (SMC) showed m/z peaks at 136, 164, and 168, which were attributed to unmodified, H2 -labeled and D2 -labeled

18

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

Table 1 Modified and unmodified OSCs qualities in MRM transitions with m/z of precursor ion and m/z of product ions. Name

Molecular formula

Molecular weight (Da)

SMC H2 -SMC D2 -SMC SEC H2 -SEC D2 -SEC SAC H2 -SAC D2 -SAC Alliin H2 -Alliin D2 -Alliin

C4 H9 NO2 S C6 H13 NO2 S C6 H9 D4 NO2 S C5 H11 NO2 S C7 H15 NO2 S C7 H11 D4 NO2 S C6 H11 NO2 S C8 H15 NO2 S C8 H11 D4 NO2 S C6 H11 NO3 S C8 H15 NO3 S C8 H11 D4 NO3 S

135.80 163.24 167.26 149.21 177.26 181.29 161.22 189.28 193.30 177.22 205.08 209.30

MRM transitions Precursor ion

Product ions

136 164 168 150 178.2 182.2 162 190 194 178 206 210

73, 119 73, 119 73, 119 87, 133 87, 133 87, 133 73, 145 73, 145 73, 145 42, 88 70, 116 74, 120

Fig. 1. Modification of organosulfur compounds by reductive amination of fromaldehyde-D2 and formaldehyde-H2 .

Fig. 2. Schematic of comparative experiments between unmodified, D2 -labeled, and H2 -labeled OSCs detected using triple quadrupole mass spectrometer.

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

Fig. 3. Infusion tandem MS spectra of organosulfur compounds obtained by (A) SMC; (B) SEC; (C) SAC; (D) alliin.

19

20

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

Fig. 4. Fragmented MS spectra of SMC displaying product ions and predicted fragmented structures for m/z values of 73 and 119. (A) Unmodified SMC; (B) H2 -labeled SMC; (C) D2 -labeled SMC.

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

21

Fig. 5. Fragmented MS spectra of alliin showing product ions and predicted fragmented structures (A) unmodified alliin with fragmentation of m/z 42 and 88; (B) H2 -labeled alliin with fragmentation of m/z 70 and 116; (C) D2 -labeled alliin with fragmentation of m/z 74 and 120.

22

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

Fig. 6. Multiple reaction monitoring mass spectra of SAC for different MRM transitions (A) unmodified SAC with MRM transitions of 162 > 73 and 162 > 145; (B) H2 formaldehyde modified SAC with MRM transitions of 190 > 73 and 194 > 145; (C) D2 -formaldehyde modified SAC with MRM transitions of 194 > 73 and 194 > 145. Table 2 The signal enhancement statistical data of modified and unmodified OSCs in MS spectra including infusion, SIM, and MRM methods. OSCs

Unmodified

H-labeled

D-labeled

H/Un

Infusion/Signal intensity SMC SEC SAC Alliin

48,200 5000 19,900 14,900

OSCs

Unmodified

394,000 99,000 918,000 24,100

480,000 112,000 643,000 53,100

8.2 19.8 46.1 1.6

H-labeled

D-labeled

H/Un

SIM/Peak area SMC SEC SAC Alliin

1914 11,822 ND 689

801,331 3,374,264 5,823,108 31,982

OSCs

Unmodified

H-labeled

SMC SEC SAC Alliin

2396 22,355 16,975 6103

Fold average

10.0 22.4 32.3 3.6

9.1 21.1 39.2 2.6

D/Un

Fold average

540.4 299.7 ND 53.7

479.6 292.6 ND 50.0

Fold 1,034,400 3,543,437 7,339,641 36,971

418.7 285.4 ND 46.4

D-labeled

H/Un

6,435,877 7,458,732 17,599,399 778,501

2303.6 374.4 1047.9 165.2

MRM/Peak area 5,519,420 8,368,876 17,788,866 1,008,370

D/Un Fold

D/Un

Fold average

2686.1 333.6 1036.8 127.6

2494.8 354.0 1042.4 146.4

Fold

SMC, respectively (see red rectangles, Fig. 3A). Furthermore, the three SMC related compounds displayed different peak intensities. Meanwhile, signals increased 8.2- and 10.0-fold relative to unmodified SMC for the H2 - and D2 -labeled compounds, respectively. Infusion spectra and statistical data are shown in Fig. 3B–D for SAC, SEC, and alliin, respectively, and in Table 2.

3.3. Detection of signal enhancement by SIM The triple quadrupole tandem MS method required m/z fragmentation values of individual OSCs. Therefore, the SIM scan mode was first set to detect specific m/z values for unmodified, H2 -labeled, and D2 -labeled SMC. Peak areas belonging to these

M.-C. Liu et al. / International Journal of Mass Spectrometry 387 (2015) 16–23

three compounds were subtracted from all spectra and the statistical data were displayed in Table 2 (data not shown). The signals increased 50.0 to 479.6 times on average. 3.4. Establishment of the MRM method MRM transitions were determined by SIM. Specific m/z values, such as 136, 164, and 168 (precursor ions), were selected and fragmented in the collision chamber. MS spectra of unmodified, and H2 -labeled, D2 -labeled SMC are shown in Fig. 4A–C, respectively. The unmodified SMC, characterized by a m/z value if 136, showed product ion peaks at m/z values of 73 and 119, which may result from well-defined structures (Fig. 4A). Both labeled SMCs corresponding to m/z values of 164 and 168 showed the same product ion peaks at m/z values of 73 and 119 (Fig. 4B and C). The other OSCs were analyzed under the same conditions and MRM transitions were set as follows: 136 > 73 and 136 > 119 for SMC, 164 > 73 and 164 > 119 for H2 -labeled SMC, and 168 > 73 and 168 > 119 for D2 -labeled SMC. Their transitions are shown in Table 1. Alliin showed different MRM transitions. Modified and unmodified alliin exhibited differences of 28 and 32 Da because their product ions contained an amine group. Their fragmented spectra are shown in Fig. 5A–C. 3.5. Detection of signal enhancement by MRM Statistical data extracted from MRM peak areas clearly showed the signal enhancement phenomenon. Unmodified SEC exhibited a peak area of 16,975 (Fig. 6A) while its H- and D-labeled counterparts displayed peak areas of 1,778,866 (Fig. 6B) and 17,599,399 (Fig. 6C), respectively. The corresponding signal increases amounted to 1036.8 and 1042.4 times the original. The average enhancement ranged from 146.4 to 2494.8 times for OSCs (Table 2). 4. Conclusion In this study, a new signal enhancement method was developed by reductive amination using isotopic formaldehydes. To demonstrate the efficacy of this technique, the enhancement between unmodified and modified OSCs was calculated for four OSCs, namely SMC, SEC, SAC, and alliin. MS signal enhancements were clearly observed through infusion, SIM, and MRM scan modes. This suggests that this method may find application in analytical absolute quantification and structural characterization. Conflict of interests We certify that there is no conflict of interests in the study. Acknowledgements The authors are grateful to the Ten Chan General Hospital, Chung-Li and KMU Joint Research Project (ST102004) for funding in the research.

23

References [1] P.J. Boersema, T.T. Aye, T.A. van Veen, A.J. Heck, S. Mohammed, Triplex protein quantification based on stable isotope labeling by peptide dimethylation applied to cell and tissue lysates, Proteomics 8 (2008) 4624–4632. [2] P.J. Boersema, R. Raijmakers, S. Lemeer, S. Mohammed, A.J. Heck, Multiplex peptide stable isotope dimethyl labeling for quantitative proteomics, Nat. Protoc. 4 (2009) 484–494. [3] J.L. Hsu, S.Y. Huang, N.H. Chow, S.H. Chen, Stable-isotope dimethyl labeling for quantitative proteomics, Anal. Chem. 75 (2003) 6843–6852. [4] K. Guo, C. Ji, L. Li, Stable-isotope dimethylation labeling combined with LC-ESI MS for quantification of amine-containing metabolites in biological samples, Anal. Chem. 79 (2007) 8631–8638. [5] E. Gimenez, V. Sanz-Nebot, A. Rizzi, Relative quantitation of glycosylation variants by stable isotope labeling of enzymatically released N-glycans using [12C]/[13C] aniline and ZIC-HILIC-ESI-TOF-MS, Anal. Bioanal. Chem. 405 (2013) 7307–7319. [6] Y.L. Chang, S.K. Liao, Y.C. Chen, W.T. Hung, H.M. Yu, W.B. Yang, J.M. Fang, C.H. Chen, Y.C. Lee, Tagging saccharides for signal enhancement in mass spectrometric analysis, J Mass Spectrom 46 (2011) 247–255. [7] A. Broberg, High-performance liquid chromatography/electrospray ionization ion-trap mass spectrometry for analysis of oligosaccharides derivatized by reductive amination and N,N-dimethylation, Carbohydr Res 342 (2007) 1462–1469. [8] M. Tokizane, K. Sato, Y. Sakami, Y. Imori, C. Matsuo, T. Ohta, Y. Ito, One-pot reductive N-alkylation with carbonyl compounds to give tertiary amines via borane reduction of imines, Synthesis-Stuttgart (2010) 36–42. [9] A.J. Mort, E.M. Chen, Separation of 8-aminonaphthalene-1,3,6-trisulfonate (ANTS)-labeled oligomers containing galacturonic acid by capillary electrophoresis: application to determining the substrate specificity of endopolygalacturonases, Electrophoresis 17 (1996) 379–383. [10] R.R.O. Loo, R.D. Smith, Proton-transfer reactions of multiply-charged peptide and protein cations and anions, J. Mass Spectrom. 30 (1995) 339–347. [11] D. Touboul, M.C. Jecklin, R. Zenobi, Investigation of deprotonation reactions on globular and denatured proteins at atmospheric pressure by ESSI-MS, J. Am. Soc. Mass Spectrom. 19 (2008) 455–466. [12] D. Bagal, H. Zhang, P.D. Schnier, Gas-phase proton-transfer chemistry coupled with TOF mass spectrometry and ion mobility-MS for the facile analysis of poly(ethylene glycols) and PEGylated polypeptide conjugates, Anal. Chem. 80 (2008) 2408–2418. [13] R.D. Smith, J.A. Loa, C.J. Barinaga, C.G. Edmonds, H.R. Udseth, Collisional activation and collision-activated dissociation of large multiply charged polypeptides and proteins produced by electrospray ionization, J. Am. Soc. Mass Spectrom. 1 (1990) 53–65. [14] F. Kjeldsen, A.M. Giessing, C.R. Ingrell, O.N. Jensen, Peptide sequencing and characterization of post-translational modifications by enhanced ion-charging and liquid chromatography electron-transfer dissociation tandem mass spectrometry, Anal. Chem. 79 (2007) 9243–9252. [15] S.H. Lomeli, S. Yin, R.R. Ogorzalek Loo, J.A. Loo, Increasing charge while preserving noncovalent protein complexes for ESI-MS, J. Am. Soc. Mass Spectrom. 20 (2009) 593–596. [16] Y. Li, R.B. Cole, Shifts in peptide and protein charge state distributions with varying spray tip orifice diameter in nanoelectrospray Fourier transform ion cyclotron resonance mass spectrometry, Anal. Chem. 75 (2003) 5739–5746. [17] A.T. Iavarone, J.C. Jurchen, E.R. Williams, Effects of solvent on the maximum charge state and charge state distribution of protein ions produced by electrospray ionization, J. Am. Soc. Mass Spectrom. 11 (2000) 976–985. [18] A.T. Iavarone, J.C. Jurchen, E.R. Williams, Supercharged protein and peptide ions formed by electrospray ionization, Anal. Chem. 73 (2001) 1455–1460. [19] A.T. Iavarone, E.R. Williams, Collisionally activated dissociation of supercharged proteins formed by electrospray ionization, Anal. Chem. 75 (2003) 4525–4533. [20] C.J. Krusemark, B.L. Frey, P.J. Belshaw, L.M. Smith, Modifying the charge state distribution of proteins in electrospray ionization mass spectrometry by chemical derivatization, J. Am. Soc. Mass Spectrom. 20 (2009) 1617–1625. [21] B.L. Frey, C.J. Krusemark, A.R. Ledvina, J.J. Coon, P.J. Belshaw, L.M. Smith, Ion-ion reactions with fixed-charge modified proteins to produce ions in a single, very high charge state, Int J Mass Spectrom 276 (2008) 136–143. [22] I.A. Kaltashov, A. Mohimen, Estimates of protein surface areas in solution by electrospray ionization mass spectrometry, Anal. Chem. 77 (2005) 5370–5379. [23] C.J. Hogan Jr., J.A. Carroll, H.W. Rohrs, P. Biswas, M.L. Gross, Charge carrier field emission determines the number of charges on native state proteins in electrospray ionization, J. Am. Chem. Soc. 130 (2008) 6926–6927. [24] D. Lemaire, G. Marie, L. Serani, O. Laprevote, Stabilization of gas-phase noncovalent macromolecular complexes in electrospray mass spectrometry using aqueous triethylammonium bicarbonate buffer, Anal. Chem. 73 (2001) 1699–1706.