Screening and Identification of Acidic ... - ACS Publications

Figure 4 Analysis of the extract from bovine colostrum. (A) Mass spectrum of the MeOH layer from the crude extract of bovine colostrum. The most abund...
6 downloads 0 Views 123KB Size
Anal. Chem. 2004, 76, 203-210

Screening and Identification of Acidic Carbohydrates in Bovine Colostrum by Using Ion/Molecule Reactions and Fourier Transform Ion Cyclotron Resonance Mass Spectrometry: Specificity toward Phosphorylated Complexes Christopher J. Petzold,† Michael D. Leavell,† and Julie A. Leary*

Department of Chemistry, University of California, Berkeley, California 94720

A screening method was developed for the identification of acidic saccharides from biological mixtures utilizing gas-phase derivatization and mass spectrometry. Phosphorylated compounds were differentiated from other acidic species by exploiting the selective reactivity of chlorotrimethylsilane with the phosphate ions (phosphorylated compounds shift by 72 Da, allowing rapid compound detection). A 13-component mock mixture was used to demonstrate the viability of the method, and a detection limit of 600 nM (30 fmol) was determined. This method was applied to the identification of acidic compounds from bovine colostrum. To further verify the selectivity of the ion/molecule reaction, exact mass measurements were used to determine the elemental composition of 14 compounds. Eight novel acidic carbohydrate species were observed in bovine colostrum, six of which have never been reported previously in milks. Tandem mass spectrometric experiments allowed compound characterization for two of these components. Identification and characterization of chemical species are some of the most demanding yet crucial tasks in bioanalytical chemistry. Development of methods to rapidly and accurately identify components in biological mixtures is of great interest due to the large quantity of samples to be analyzed and the desire to implement high-throughput techniques.1-7 The analysis of specific components is further complicated by the fact that analytes are * Corresponding author. E-mail: [email protected]. Phone: (510) 643-6499. Fax: (510) 642-9295. † These authors contributed equally to this work. (1) Steen, H.; Pandey, A. Trends Biotechnol. 2002, 20, 361-364. (2) Kyranos, J. N.; Cai, H.; Wei, D.; Goetzinger, W. K. Curr. Opin. Biotechnol. 2001, 12, 105-111. (3) Barbarin, N.; Mawhinney, D. B.; Black, R.; Henion, J. J. Chromatogr., B 2003, 783, 73-83. (4) Lindon, J. C.; Nicholson, J. K.; Wilson, I. D. J. Chromatogr., B 2000, 748, 233-258. (5) Berger, S. J.; Lee, S. W.; Anderson, G. A.; Pasa-Tolic, L.; Tolic, N.; Shen, Y. F.; Zhao, R.; Smith, R. D. Anal. Chem. 2002, 74, 4994-5000. (6) Smith, R. D.; Anderson, G. A.; Lipton, M. S.; Pasa-Tolic, L.; Shen, Y. F.; Conrads, T. P.; Veenstra, T. D.; Udseth, H. R. Proteomics 2002, 2, 513523. (7) Dass, C. Principles and practice of biological mass spectrometry; John Wiley: New York, 2001. 10.1021/ac034682v CCC: $27.50 Published on Web 11/25/2003

© 2004 American Chemical Society

often of low abundance and occur in complex biological mixtures. Frequently, separation of the analyte from a complex mixture requires more time and effort than actually identifying the unknown.8 Consequently, analytical techniques that are capable of screening a complex mixture for particular analytes or types of analytes are highly desirable. Mass spectrometry (MS) is a technique that is well-suited for complex mixture analysis, as is evidenced by a recent report where 11 000 compounds were resolved in a single mass spectrum.9 Furthermore, MS is a rapid technique that does not require a large amount of sample handling prior to analysis, a constraint especially important for biological applications.7,10,11 Identification of unknown species via mass spectrometry typically involves stable isotope incorporation,12-14 derivatization prior to MS analysis,15,16 or exact mass measurements.17,18 By far the most common form of mass spectrometric screening involves monitoring the loss of specific fragments or appearance of specific ions following tandem mass spectrometry.19,20 Implementation of any of the techniques described above yields a wealth of information, yet unambiguous identification and characterization of a metabolite is often still elusive. (8) Venn, R. F. Principles and practice of bioanalysis; Taylor & Francis: London; New York, 2000. (9) Hughey, C. A.; Rodgers, R. P.; Marshall, A. G. Anal. Chem. 2002, 74, 41454149. (10) Valaskovic, G. A.; Kelleher, N. L.; Little, D. P.; Aaserud, D. J.; McLafferty, F. W. Anal. Chem. 1995, 67, 3802-3805. (11) Hofstadler, S. A.; Severs, J. C.; Smith, R. D.; Swanek, F. D.; Ewing, A. G. Rapid Commun. Mass Spectrom. 1996, 10, 919-922. (12) Leis, H. J.; Fauler, G.; Raspotnig, G.; Windischhofer, W. J. Mass Spectrom. 2000, 35, 1100-1104. (13) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aebersold, R. Nat. Biotechnol. 1999, 17, 994-999. (14) Mougous, J. D.; Leavell, M. D.; Senaratne, R. H.; Leigh, C. D.; Williams, S. J.; Riley, L. W.; Leary, J. A.; Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 17037-17042. (15) Oda, Y.; Nagasu, T.; Chait, B. T. Nat. Biotechnol. 2001, 19, 379-382. (16) Zhou, H. L.; Watts, J. D.; Aebersold, R. Nat. Biotechnol. 2001, 19, 375378. (17) Bossio, R. E.; Marshall, A. G. Anal. Chem. 2002, 74, 1674-1679. (18) Guan, S. H.; Marshall, A. G.; Scheppele, S. E. Anal. Chem. 1996, 68, 4671. (19) Busch, K. L.; Glish, G. L.; McLuckey, S. A. Mass spectrometry/mass spectrometry: techniques and applications of tandem mass spectrometry; VCH Publishers: New York, 1988. (20) Yost, R. A.; Fetterolf, D. D. Mass Spectrom. Rev. 1983, 2, 1-45.

Analytical Chemistry, Vol. 76, No. 1, January 1, 2004 203

Phosphorylated species such as oligosaccharides, nucleic acids, and posttranslationally modified proteins are ubiquitous in biological systems and have many biological functions including the following: roles in signal transduction,21 as molecular recognition markers,22 or as metabolic intermediate.23 As a result, mass spectrometric methods for the analysis of phosphorylated species are of great interest, as is evidenced by two recent reviews on the subject.24,25 Various tandem mass spectrometric methods, including collision-induced dissociation,26-29 infrared multiphoton dissociation,30,31 postsource decay,32,33 and electron capture dissociation,34,35 are often used to screen mixtures or characterize phosphorylated analytes. Typically, phosphorylated components are identified by monitoring specific fragmentations (i.e., neutral loss of HPO3 (80 Da) or H3PO4 (98 Da) for positively charged ions, appearance of H2PO4- (m/z 97) for negatively charged ions) resulting from various dissociation processes. While useful when known species are being analyzed, monitoring the loss of an 80or 98-Da neutral fragment or the appearance of m/z 97 is not necessarily sufficient to establish the presence of a phosphate group due to isobaric interferences from unknown compounds (e.g., sulfated compounds).17 In situations where fragmentation of the unknowns fails to adequately screen for phosphorylated analytes due to isobaric interferences, exact mass measurements can be implemented. Bossio and Marshall have utilized the high-resolution capabilities of FTICR mass spectrometry to differentiate between sulfated and phosphorylated ions (mass difference of 0.0095 Da).17 However, these experiments are not possible using low-resolution instruments. Furthermore, the number of reasonable elemental compositions possible for an ion increases drastically with mass, resulting in lower confidence in a particular elemental composition.18 Complementary to the gas-phase methods, there are a variety of condensed-phase derivatization methods that have been implemented to denote the position of the phosphate moiety prior to MS analysis.15,16,36 For example, Chait and co-workers have introduced a strategy whereby phosphate moieties are replaced by a biotin affinity handle, enriched with avidin, and analyzed by mass spectrometry.15 In another study, Aebersold and co-workers (21) Hunter, T. Cell 2000, 100, 113-127. (22) Kornfeld, S.; Mellman, I. Annu. Rev. Cell Biol. 1989, 5, 483-525. (23) Voet, D.; Voet, J. G. Biochemistry; Wiley: New York, 1990. (24) McLachlin, D. T.; Chait, B. T. Curr. Opin. Chem. Biol. 2001, 5, 591-602. (25) Mann, M.; Ong, S. E.; Gronborg, M.; Steen, H.; Jensen, O. N.; Pandey, A. Trends Biotechnol. 2002, 20, 261-268. (26) Steen, H.; Kuster, B.; Fernandez, M.; Pandey, A.; Mann, M. Anal. Chem. 2001, 73, 1440-1448. (27) Salomon, A. R. F.; Scott, B.; Brill, L. M.; Brinker, A.; Phung, Q. T.; Ericson, C.; Sauer, K.; Brock, A.; Horn, D. M.; Schultz, P. G.; Peters, E. C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100. (28) Schlosser, A.; Pipkorn, R.; Bossemeyer, D.; Lehmann, W. D. Anal. Chem. 2001, 73, 170-176. (29) Carr, S. A.; Huddleston, M. J.; Annan, R. S. Anal. Biochem. 1996, 239, 180192. (30) Flora, J. W.; Muddiman, D. C. Anal. Chem. 2001, 73, 3305-3311. (31) Flora, J. W.; Muddiman, D. C. J. Am. Chem. Soc. 2002, 124, 6546-6547. (32) Talbo, G. H.; Suckau, D.; Malkoski, M.; Reynolds, E. C. Peptides 2001, 22, 1093-1098. (33) Annan, R. S.; Huddleston, M. J.; Verma, R.; Deshaies, R. J.; Carr, S. A. Anal. Chem. 2001, 73, 393-404. (34) Shi, S. D. H.; Hemling, M. E.; Carr, S. A.; Horn, D. M.; Lindh, I.; McLafferty, F. W. Anal. Chem. 2001, 73, 19-22. (35) Stensballe, A.; Jensen, O. N.; Olsen, J. V.; Haselmann, K. F.; Zubarev, R. A. Rapid Commun. Mass Spectrom. 2000, 14, 1793-1800. (36) Li, W.; Boykins, R. A.; Backlund, P. S.; Wang, G. Y.; Chen, H. C. Anal. Chem. 2002, 74, 5701-5710.

204

Analytical Chemistry, Vol. 76, No. 1, January 1, 2004

utilized a synthetic strategy to protect and enrich phosphorylated peptides, followed by mass spectrometric analysis.16 While these techniques are clearly powerful, the additional sample handling/ quantities needed, the time required, and the multiple synthetic steps employed in these methods might limit general use especially for biological applications. Furthermore, these derivatization schemes are optimized for phosphorylated peptides and their potential utility for other phosphorylated species may be affected by differing reactivity or matrix effects. Therefore, a gas-phase technique that circumvents the limitations encountered with the condensed-phase methods is of considerable analytical interest.37-44 We describe herein the development and utilization of ion/ molecule reactions coupled to FTICR mass spectrometry as a means to identify phosphorylated compounds in complex biological mixtures. The developed methodology is generally applicable to complex mixtures and can be implemented on the majority of mass spectrometers. Trimethylsilyl chloride (TMSCl) was used as a neutral reagent to react selectively with phosphorylated oligosaccharides both in a mock mixture and in a crude extract from bovine colostrum. Bovine colostrum is known to be rich in various saccharide species, many of which are acidic.45-47 It is thought that the saccharide species present in colostrum (human, bovine, etc.), provide protection from pathogenic viruses and bacteria.45,47-49 Consequently, investigation of the acidic saccharide species from this sample could add insight into the biological function of colostrum. The method is rapid and sensitive and aids in elemental composition assignment of unknown compounds. Furthermore, gas-phase derivatization of phosphorylated unknowns in combination with tandem mass spectrometry yields more structural information than tandem mass spectrometry of the underivatized phosphorylated species. Using this methodology and exact mass measurements, several novel phosphorylated oligosaccharides not previously reported in milks were identified. EXPERIMENTAL SECTION Chemicals and Materials. All chemicals were obtained from Sigma (St. Louis, MO) except for ∆uronic acidR(1f4)GlcNAc6S, N-acetylneuraminic acid dimer, disialyl-lacto-N-tetraose (Calbiochem, La Jolla, CA), chlorotrimethylsilane (Aldrich, Milwaukee, WI), ammonium acetate, chloroform, and methanol (Fisher, Fairborn, NJ). All solvents were of HPLC grade. The anionexchange resin (AG-1 × 2, hydroxide form) was obtained from Bio-Rad Laboratories (Hercules, CA). Dr. Edward DePeters (UC Davis) donated the bovine colostrum sample. The lipophospho(37) Green, M. K.; Lebrilla, C. B. Mass Spectrom. Rev. 1997, 16, 53-71. (38) Brodbelt, J. S. Mass Spectrom. Rev. 1997, 16, 91-110. (39) Grigorean, G.; Lebrilla, C. B. Anal. Chem. 2001, 73, 1684-1691. (40) Gronert, S.; O’Hair, R. A. J. J. Am. Soc. Mass Spectrom. 2002, 13, 10881098. (41) Leavell, M. D.; Kruppa, G. H.; Leary, J. A. Anal. Chem. 2002, 74, 26082611. (42) Reid, G. E.; McLuckey, S. A. J. Mass Spectrom. 2002, 37, 663-675. (43) Petucci, C.; Guler, L.; Kenttamaa, H. I. J. Am. Soc. Mass Spectrom. 2002, 13, 362-370. (44) Leeck, D. T.; Ranatunga, T. D.; Smith, R. L.; Partanen, T.; Vainiotalo, P.; Kenttamaa, H. I. Int. J. Mass Spectrom. Ion Processes 1995, 141, 229-240. (45) Gopal, P. K.; Gill, H. S. Br. J. Nutr. 2000, 84, S69-S74. (46) Jensen, R. G. Handbook of milk composition; Academic Press: San Diego, 1995. (47) Urashima, T.; Saito, T.; Nakamura, T.; Messer, M. Glycoconjugate J. 2001, 18, 357-371. (48) Lingwood, C. A. Curr. Opin. Chem. Biol. 1998, 2, 695-700. (49) Sharon, N.; Ofek, I. Glycoconjugate J. 2000, 17, 659-664.

Figure 1. Selectivity of the ion/molecule reagent TMSCl for phosphate functional groups: (A) Reaction of TMSCl with glucose 6-phosphate (m/z 259.02; Glc6P); (B), (C) mass spectrum of glucose 6-sulfate (m/z 259.01; Glc6S) and N-acetylneuraminic acid (m/z 308.10; NeuAc), respectively, following TMSCl introduction.

glycan analogue, LPG-1 ([Gal(1-4)Man-PO4]2Gal(1-4)Man-O(CH2)8-CHCH2), was provided by Professor Michael A. J. Ferguson and Dr. Andrei Nikolaev (University of Dundee, Scotland, U.K.).50 Mass Spectrometry. All experiments were performed on a Bruker-Daltonics (Billerica, MA) Apex II FTICR mass spectrometer equipped with a 7.0-T magnet and an electrospray ionization source (Analytica, Branford, CT). All commercial samples were sprayed from 1:1 MeOH/H2O at a concentration of between 600 nM and 30 µM. Spectra are composed of between 256K and 1M data points and are an average of between 4 and 32 scans. The spectra were acquired on the FTICR data station operating Xmass version 5.0.10 (Bruker Daltonics). Elemental compositions of unknown compounds were determined through exact mass measurements made through internal calibration. For ion/molecule reactions, chlorotrimethylsilane was pulsed into the ICR analyzer cell for 0.1-5 s. The reagent pulsing time was adjusted to maximize reaction product formation. Tandem mass spectrometry experiments for structural analysis of phosphorylated components were performed as described previously.51 Bovine Colostrum Extraction. The colostrum sample was collected from a Holstein cow 12 h after the cow calved and was stored frozen until needed for analysis. Crude carbohydratecontaining fractions were prepared by shaking 40 mL of colostrum with 165 mL of chloroform/methanol (2:1) for 1 h at 4 °C. After centrifugation (3500g, 1 h) at 4 °C, the MeOH layer was collected. To enrich the acidic components, a 500-µL aliquot of the MeOH layer was incubated with the anion-exchange resin (25 mg) for 1 h. The resin was then washed twice with 1000 µL of H2O to (50) Nikolaev, A. V.; Rutherford, T. J.; Ferguson, M. A. J.; Brimacombe, J. S. J. Chem. Soc., Perkin Trans. 1 1995, 1977-1987. (51) Leavell, M. D.; Leary, J. A. J. Am. Soc. Mass Spectrom. 2003, 14, 323-331.

remove the nonacidic components. The acidic components were then released by incubating the resin with 500 µL of 100 mM ammonium acetate (pH 7.0). The resin was removed by centrifugation, and the remaining solution (containing the acidic components released from the resin) was diluted with an equal volume of MeOH and analyzed on the mass spectrometer. RESULTS AND DISCUSSION Three acidic monosaccharides were employed to ascertain the selectivity of the ion/molecule reagent, TMSCl, for the common acidic modifications to carbohydrates. Functionalized monosaccharides, rather than oligosaccharides, were first employed to determine the intrinsic reactivity of the ion/molecule reagent, therefore minimizing complicating factors such as steric hindrance. These monosaccharides, glucose 6-phosphate (Glc6P; Figure 1A), glucose 6-sulfate (Glc6S; Figure 1B), and N-acetyl neuraminic acid (NeuAc; Figure 1C) are functionalized with phosphate, sulfate, or carboxylic acid moieties, respectively. Upon reaction of singly charged (deprotonated) [Glc6P - H]- (m/z 259.02) with TMSCl, one reaction product results (Figure 1A). Conversely, when TMSCl was pulsed into the ICR cell containing deprotonated sulfated or carboxylated carbohydrates, the complementary reaction was not observed. For example, even when a 3-s reaction time was used (a 30-fold longer introduction time) with either singly charged [Glc6S - H]- (m/z 259.01; Figure 1B) or [NeuAc - H]- (m/z 308.10; Figure 1C), the commensurate reaction products, m/z 331.05 and 380.14, or other reaction products were not observed. Thus, it appears that TMSCl is selectively reactive with phosphate moieties and is unreactive with singly deprotonated sulfate or carboxylic acid moieties. The reactivity observed for singly charged phosphorylated components is consistent with that reported previously for singly Analytical Chemistry, Vol. 76, No. 1, January 1, 2004

205

Table 1. Compounds Included in the Mock Mixture and Their Observed Reactivity functional groupsc compounda

no.

unreacted m/zb

shift (+TMSCl)

GlcNH2-1-P Glc6S Glc1P GlcNAc1P NeuAc Glc1,6diP Mal-1-P ∆UA-GlcNAc-6S NeuAc2 3′-SL LPG-1 DSLNT 6′-SLN

1 2 3 4 5 6 7 8 9 10 11 12 13

258.041259.011259.021300.051308.101339.001421.081458.061599.201632.201643.182643.712673.231-

Y N Y Y N N Y N N N Y Y N

R

P

370.081-

1

S

C

1 331.061372.091-

1 1

493.121-

2 1 1

697.192697.722-

1 2 1

2 2 1

OH

NH2

X X X X X X X X X X X X X

X

NAc

X X

X X X X X

a Compound abbreviations: GlcNH -1-P, glucosamine 1-phosphate; Glc6S, glucose 6-sulfate; Glc1P, glucose 1-phosphate; GlcNAc1P, N2 acetylglucosamine 1-phosphate; NeuAc, N-acetylneuraminic acid; Glc1,6diP, glucose 1,6-diphosphate; Mal-1-P, maltose 1-phosphate; ∆UA-GlcNAc6S, ∆uronic acidR(1f4)GlcNAc6S; NeuAc2, N-acetylneuraminic acid dimer; 3′-SL, 3′-sialyllactose; LPG-1, [Gal(1-4)Man-PO4]2Gal(1-4)Man-O(CH2)8-CHCH2;50 DSLNT, disialyl-lacto-N-tetraose; 3′-SLN, 3′-N-acetylsialyllactosamine. b Ions are singly or doubly charged as indicated by the superscript next to their observed m/z. c Functional group abbreviations: P, phosphate; S, sulfate; C, carboxylic acid; OH, alcohol; NH2, amine; NAc, amide. The number of acidic functional groups is specified, while only the presence (X) of nonacidic functional groups is indicated.

charged species.51,52 The product ion, m/z 331.06, results from the addition of TMSCl followed by elimination of HCl. The reaction is sufficiently fast, as is evidenced by the short reagent introduction time (100 ms). Earlier studies have demonstrated the reactivity of TMSCl with phosphorylated oligonucleotides52 and phosphorylated mono- and oligosaccharides,41,51,53 where analogous reactivity trends were observed. The analytical utility of the selective reactivity was subsequently investigated in a more complex system. A 13-component mixture was made of several phosphate-, sulfate-, and carboxylic acidcontaining saccharides (Table 1). This mock mixture also contained a diverse set of functional groups including alcohol, amine, and amide groups. Thus, the reactivity of TMSCl with various neutral and acidic functional groups as well as the interplay between these functionalities could be determined. Furthermore, several compounds containing two acidic groups were also introduced, including diphosphates and dicarboxylic acid functionalities, as well as mixed acidic species. A mass spectrum of the mock mixture prior to reaction with TMSCl is shown in Figure 2A, and the numbered ions in the spectrum correspond to the compounds as numbered in Table 1. Consistent with the previous experiments, it is apparent that only the phosphorylated compounds react with the TMSCl (Figure 2B). This is evidenced by a decrease in the relative abundance of monophosphorylated compounds 1, 3, 4, 7, and a diphosphorylated compound 11, with respect to the other singly charged unphosphorylated compounds. The 72-Da shift, which corresponds to the addition/elimination reaction discussed above, permits rapid identification of the phosphorylated components in the mixture. Additionally it was determined that the limits of detection for the phosphorylated monosaccharides was 30 fmol, corresponding to a solution concentration of 600 nM.

It is important to note that 2 (Glc6S) and 3 (Glc1P) are isobars differing by only 0.0095 Da, requiring a resolution of ∼26 000 for separation (inset, Figure 2A, B). While this resolution is clearly achievable by FTICR mass spectrometry, as has been shown by Marshall and co-workers in their study of phosphorylated and sulfated peptide isobars,17,54 it is not achievable on lower resolution instruments. Thus, reaction with TMSCl can be utilized to distinguish between phosphorylated compounds and other isobaric species on lower resolution instruments. Consequently, we implemented the ion/molecule reaction methodology on an Applied Biosystems/MDS Sciex Q-Trap instrument by introducing the ion/molecule reagent into the collision cell (q2 region, ∼1 mTorr) through the CAD gas connection. Reaction between Gal6P and TMSCl was observed by mass selecting m/z 259 ([Gal6P - H]-) in the first quadrupole and scanning for reaction product ions in the linear ion trap (Figure 3). The observed reactivity is consistent with that observed in the experiments accomplished with the FTICR mass spectrometer. Future work will focus on application of the ion/molecule methodology for screening complex mixtures by using the Q-Trap instrument. One interesting result of the application of the ion/molecule reaction methodology to the mock mixture was the discovery that TMSCl is not reactive with singly charged compounds that contain both phosphate and another acidic functionality (phosphate, sulfate, or carboxylate). This is illustrated by 6, Glc1,6diP, where no reaction was observed. Previous work by Damrauer and coworkers suggests that the nucleophilicity of the ion becomes the limiting factor for species with proton affinities lower than that of Cl-.55 While the proton affinity of (MeO)2PO2- is essentially the same as Cl-, it is likely that multiply acidic species have even lower proton affinities. Indeed, O’Hair and McLuckey did not

(52) O’Hair, R. A. J.; McLuckey, S. A. Int. J. Mass Spectrom. Ion Processes 1997, 162, 183-202. (53) Leavell, M. D.; Kruppa, G. H.; Leary, J. A. Int. J. Mass Spectrom. 2003, 222, 135-153.

(54) He, F.; Hendrickson, C. L.; Marshall, A. G. Anal. Chem. 2001, 73, 647650. (55) Damrauer, R.; DePuy, C. H.; Bierbaum, V. M. Organometallics 1982, 1, 1553-1554.

206 Analytical Chemistry, Vol. 76, No. 1, January 1, 2004

Figure 2. Selectivity of the ion/molecule reaction for a complex mixture. Compounds are identified by number; compare to Table 1 for compound identity. (A) Mass spectrum (MS1) of the mock mixture prior to addition of TMSCl. (B) After the ion/molecule reaction phosphorylated components shift by 72 Da as indicated by a “+”. The insets show a magnified view of the region around m/z 259.015, where two monosaccharide isobars glucose 1-sulfate (m/z 259.01; Glc1S, 2) and glucose 1-phosphate (m/z 259.02; Glc1P, 3) are resolved, and the phosphorylated compound 3 reacts selectively with TMSCl.

Figure 3. Mass-selected galactose 6-phosphate (m/z 259.2; Gal6P) reacted with TMSCl in the collision cell of an Applied Biosystems/MDS Sciex Q-Trap instrument yielding TMS-modified Gal6P (m/z 331). The observed reactivity is consistent with that of the FTICR experiments.

observe reaction between TMSCl and singly charged di- and trinucleotides with multiple phosphate groups. However, a reaction does occur with a doubly charged/doubly phosphorylated or a doubly charged/doubly carboxylated ion as is illustrated by 11 (LPG-1) and 12 (DSLNT), respectively (Table 1). Since the reactivity differs significantly for these species, compounds

containing multiple acidic functionalities and multiple charges are currently being addressed in another investigation. With the ion/molecule reaction methodology in place, bovine colostrum was used to test the applicability of the method to a more complex biological mixture. The MeOH layer from a bovine colostrum extract was sprayed into the mass spectrometer, Analytical Chemistry, Vol. 76, No. 1, January 1, 2004

207

Figure 4. Analysis of the extract from bovine colostrum. (A) Mass spectrum of the MeOH layer from the crude extract of bovine colostrum. The most abundant ion, m/z 377.09, is identified as [lactose + Cl]-. (B) Enrichment of the acidic components by anion-exchange affinity purification as is illustrated by the decrease in the abundance of lactose (denoted by a “/”) and the increase of other ions in the spectrum. Compounds are indicated by number; compare to Table 2 for compound identity. (C) Upon reacting the enriched mixture with TMSCl, phosphorylated components shift by the requisite 72 Da as is indicated by a “+”. The insets (in B and C) show selective reaction of the phosphorylated disaccharide (m/z 421.0656; 17) as compared to the sulfated disaccharide (m/z 421.0756, 18) as is shown in the inset (both spectra in B and C are magnified by 5-fold on the y-axis so low-abundance compounds can be observed).

generating the spectrum in Figure 4A. Only three ionic species, m/z 377.09, 632.20, and 719.20, were observed; these correspond to [lactose + Cl]-, [X′-SL (where X′ ) 3 or 6, a mixture of isomers exists)]-, and [2lactose + Cl]-, respectively. Lactose is known to be the major carbohydrate species in bovine colostrum, existing at a concentration 1000-fold greater (∼130 mM) than the next abundant species, the sialyllactoses,45,46 which in turn have been shown to be ∼100-fold greater in concentration than other acidic oligosaccharides.56 Since other acidic oligosaccharides species were not observed, further enrichment of these species from the crude extract was required. To enhance the abundance of the acidic components, the extract was subjected to anion-exchange affinity enrichment. Upon incubating the MeOH layer (from the above extraction) with the anion-exchange resin, washing away the nonacidic components (e.g., lactose), releasing the acidic species from the resin, and spraying the mixture into the FTICR, the spectrum in Figure 4B was obtained. It is immediately apparent that the abundance of the nonacidic components has decreased significantly, as is evidenced by the low abundance of m/z 377.09 ([lactose + Cl]-, denoted by an asterisk in the spectrum). Many other species were observed after enrichment that were not observed previously in the crude MeOH extract. Some of these compounds, 19, 2225, and 27, have been identified previously and are summarized in Table 2.45,47 It is important to note that these compounds contain different types of acidic groups, including phosphate and carboxylic acid functionalities. Furthermore, as is discussed below, a newly (56) Parkkinen, J.; Finne, J. In Complex Carbohydrates; Ginsburg, V., Ed.; Academic Press: Orlando, 1987; Vol. 138, pp 289-300.

208 Analytical Chemistry, Vol. 76, No. 1, January 1, 2004

identified sulfated disaccharide (in bovine colostrum) was also observed. Observation of several previously identified acidic compounds indicates that the anion-exchange procedure is general to all acidic functionalities and can therefore be applied to the mixture. In addition to the previously characterized saccharides, many additional carbohydrates were also detected in the bovine colostrum. To distinguish the phosphorylated components from the other acidic compounds, the ion/molecule reaction methodology was employed. Upon reaction with TMSCl, several of the compounds were observed to shift by the requisite 72 Da. Through exact mass measurements, these compounds were shown to have the following compositions: HexP (14), HexNAcP (15), Hex2P (19), and Hex1HexNAc1P (20). Other compounds that were not observed to shift were identified by exact mass measurements with the following compositions: HexPS (16), HexP2 (17), Hex2S (18), Hex2P2 (21), and NeuAc1Hex3P (26). All of the compounds showed only a small error upon exact measurements (