Derivatization Using Dimethylamine for Tandem Mass Spectrometric

Derivatization Using Dimethylamine for Tandem. Mass Spectrometric Structure Analysis of. Enzymatically and Acidically Depolymerized. Methyl Cellulose...
0 downloads 0 Views 288KB Size
Anal. Chem. 2005, 77, 2948-2959

Derivatization Using Dimethylamine for Tandem Mass Spectrometric Structure Analysis of Enzymatically and Acidically Depolymerized Methyl Cellulose Dane Momcilovic,† Herje Schagerlo 1 f,‡ Daniel Ro 1 me,§ Magnus Jo 1 rnte´n-Karlsson,| Karl-Erik Karlsson,⊥ ⊥ ‡ † Bengt Wittgren, Folke Tjerneld, Karl-Gustav Wahlund, and Gunnar Brinkmalm*,⊥

Departments of Technical Analytical Chemistry, Biochemistry, and Bioorganic Chemistry, Lund University, P.O. Box 124, S-221 00 Lund, Sweden, AstraZeneca R&D Lund, S-221 87 Lund, Sweden, and AstraZeneca R&D Mo¨lndal, S-431 83 Mo¨lndal, Sweden

Structure analysis of partially depolymerized methyl cellulose was performed by nanoelectrospray ionization tandem mass spectrometry (nano-ESI-MS/MS) and by matrix-assisted laser desorption/ionization tandem mass spectrometry (MALDI-MS/MS). Dimethylamine (DMA) was used for the first time as a reducing end derivatization reagent for oligosaccharides. This is an attractive reagent since it could be easily removed from the reaction mixture. Most important it also introduces a basic functional group that increased the sensitivity in both MALDI and nano-ESI. Depolymerization was made in two ways: one by the cellulose selective endoglucanase 5A from Bacillus agaradhaerens (Ba Cel5A) and the other by trifluoroacetic acid. The DMA derivatives formed both protonated and sodiated molecules in nano-ESI and MALDI. Tandem MS of protonated molecules yielded predominantly Y fragments from which the distribution of the substituents in the oligomers could be measured. Fragments obtained in tandem MS of sodiated molecules provided information regarding the positions of the substituents within the anhydroglucose units (AGUs). It was found that Ba Cel5A could cleave glucosidic bonds also if the AGU on the reducing side of the bond was fully methylated. The combination of DMA derivatization and tandem MS was demonstrated as a tool for the characterization of endoglucanase selectivity. Mass spectrometry (MS) has evolved to become a frequently applied technique for analysis of oligosaccharides during the past decades.1-3 Utilizing techniques such as matrix-assisted laser * To whom corresponance should be addressed. Tel.: +46-31-7761013. Fax: +46-31-7763834. E-mail: [email protected]. † Department of Technical Analytical Chemistry, Lund University. ‡ Department of Biochemistry, Lund University. § Department of Bioorganic Chemistry, Lund University. | AstraZeneca R&D Lund. ⊥ AstraZeneca R&D Mo ¨lndal. (1) Harvey, D. J. Mass Spectrom. Rev. 1999, 18, 349-451. (2) Zaia, J. Mass Spectrom. Rev. 2004, 23, 161-227. (3) Reinhold: V. N.; Reinhold: B. B.; Costello, C. E. Anal. Chem. 1995, 67, 1772-1784.

2948 Analytical Chemistry, Vol. 77, No. 9, May 1, 2005

desorption/ionization (MALDI), electrospray ionization (ESI) and fast-atom bombardment (FAB), large oligosaccharide molecules can be ionized with relatively little fragmentation.4-6 This increases the possibility of performing molar mass determinations as well as structure analysis by MS. Tandem MS has been demonstrated to be very useful for determining the sequence, branching, linkage type, and also the positions of substituents in oligosaccharides.1,7,8 In many cases such information is sufficient to determine the oligosaccharide structure. Reducing end derivatization has frequently been utilized to improve mass spectrometry of oligosaccharides.9-11 It can increase the sensitivity in both MALDI and ESI and also facilitates structure analysis since the mass of the reducing end sugar unit is changed. Mass spectrometry has been demonstrated as an useful approach for chemical characterization of partially depolymerized cellulose and other polysaccharide derivatives.12-15 Arisz et al. studied the substituent distribution in partially depolymerized methyl cellulose (MC) by gas chromatography and FAB-MS.16 Recently, Tu¨ting et al. performed a study where regioselectively methylated maltooligosaccharides were subjected to ESI tandem MS.17 It was shown that the positions of the substituents within (4) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299-2301. (5) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64-71. (6) Barber, M.; Bordoli, R. S.; Elliott, G. J.; Sedgwick, R. D.; Tyler, A. N. Anal. Chem. 1982, 54, 645A-657A. (7) Zaia, J.; Li, X.-Q.; Chan, S.-Y.; Costello, C. E. J. Am. Soc. Mass Spectrom. 2003, 14, 1270-1281. (8) Limberg, G.; Ko¨rner, R.; Buchholt, H. C.; Christensen, T. M. I. E.; Roepstorff, P.; Mikkelsen, J. D. Carbohydr. Res. 2000, 327, 321-332. (9) Ku ¨ ster, B.; Naven, T. J. P.; Harvey, D. J. Rapid Commun. Mass Spectrom. 1996, 10, 1645-1651. (10) Yoshino, K.-I.; Takao, T.; Murata, H.; Shimonishi, Y. Anal. Chem. 1995, 67, 4028-4031. (11) Okamoto, M.; Takahashi, K.-I.; Doi, T. Rapid Commun. Mass Spectrom. 1995, 9, 641. (12) Mischnick, P.; Heinrich, J.; Gohdes, M. Das Papier 1999, 53, 739-743. (13) Momcilovic, D.; Wittgren, B.; Wahlund, K.-G.; Karlsson, J.; Brinkmalm, G. Rapid Commun. Mass Spectrom. 2003, 17, 1107-1115. (14) Momcilovic, D.; Wittgren, B.; Wahlund, K.-G.; Karlsson, J.; Brinkmalm, G. Rapid Commun. Mass Spectrom. 2003, 17, 1116-1124. (15) Richardson, S.; Nilsson, G. S.; Bergquist, K.-E.; Gorton, L.; Mischnick, P. Carbohydr. Res. 2000, 328, 365-373. (16) Arisz, P. W.; Kauw, H. J. J.; Boon, J. J. Carbohydr. Res. 1995, 271, 1-14. 10.1021/ac048194e CCC: $30.25

© 2005 American Chemical Society Published on Web 04/06/2005

the anhydroglucose units (AGUs) influenced the relative yield of the various fragment types. One approach to characterize the substituent distribution in cellulose derivatives is to perform enzymatic depolymerization by endoglucanases.18 This has been performed by Saake et al. for MC by size fractionation of the enzymatic hydrolysate and thereafter acidic depolymerization of the fractions.19 Subsequent analysis by high-performance anion-exchange chromatography with pulsed amperometric detection provided information regarding the degree of substitution (DS) in the fractions. Limberg et al. studied block sequences of galacturonic acid in pectins.8 This was performed by enzymatic depolymerization with exo- and endopolygalacturonase. The positions of the methyl esterfied galacturonic acid units in the oligomers thus obtained were determined by ESI tandem MS. Nojiri and Kondo studied an endoglucanase from Trichoderma viride in enzymatic depolymerization of regioselectively substituted methyl celluloses.20 In this study, it was determined that the enzyme could cleave glucosidic bonds between AGUs that were methylated at the C-6 position. Glucosidic bonds in methyl cellulose with substituents at both the C-2 and C-3 positions were not cleaved. To our knowledge, characterization of the substituent distribution of partially depolymerized cellulose derivatives has up to date not been performed by combination of endoglucanase depolymerization and tandem MS. In nature, endoglucanases are part of the metabolism of many microorganisms. In the process when the glucosidic bonds are cleaved, subsites in the enzyme active site attach to the cellulose polymers through interactions with several AGUs. Figure 1a shows the -n to +n scheme proposed for the sugar-binding subsites in glycosyl hydrolases proposed by Davies et al.21 The AGU interacting with the -1 subsite becomes the new reducing end glucose unit (GU), and the one in the +1 subsite becomes the new nonreducing end AGU after the cleavage. For cellulose derivatives, the introduced substituents may prevent hydrogen bond formation or constitute sterical obstacles for the sugar-binding subsites and thereby hinder the cleavage of certain glucosidic bonds. Yet, the type of substituents as well as their positions within the AGUs may determine the possibility of cleaving the glucosidic bonds. Due to this selectivity, the composition of the oligomers formed is determined by the applied enzyme and the substituents in the original intact cellulose derivative. Therefore, to be able to apply enzymes for characterization of the substituent distribution in cellulose derivatives, it is very important to first characterize the enzyme selectivity. The enzyme that we have applied is a purified cellulose selective endoglucanase belonging to family 5 from Bacillus agaradhaerens (Ba Cel5A).23 The structure of the enzyme with oligosaccharides in the active site has been described elsewhere.24,25 These studies have also indicated that the enzyme has (17) Tu ¨ ting, W.; Adden, R.; Mischnick, P. Int. J. Mass Spectrom. 2004, 232, 107115. (18) Karlsson, J.; Momcilovic, D.; Wittgren, B.; Schu ¨ lein, M.; Tjerneld, F.; Brinkmalm, G. Biopolymers 2002, 63, 32-40. (19) Saake, B.; Lebioda, S.; Puls, J. Holzforschung 2004, 58, 97-104. (20) Nojiri, M.; Kondo, T. Macromolecules 1996, 29, 2392-2395. (21) Davies, G. J.; Wilson, K. S.; Henrissat, B. Biochem. J. 1997, 321, 557-559. (22) Domon, B.; Costello, C. E. Glycoconjugate J. 1988, 5, 397-409. (23) Henrissat, B. Biochem. J. 1991, 280, 309-316. (24) Davies, G. J.; Dauter, M.; Marek Brzozowski, A. M.; Bjørnvad, M. E.; Andersen, K. V.; Schu ¨ lein, M. Biochemistry 1998, 37, 1926-1932. (25) Varrot, A.; Schu ¨ lein, M.; Davies, G. J. J. Mol. Biol. 2000, 297, 819-828.

Figure 1. Schematic illustration of the nomenclature for sugarbinding subsites in glycosyl hydrolases proposed by Davies et al.21 (a). Structure of DMA-derivatized cellotriose (b). For methyl cellulose, the “R” is either a methyl group or a hydrogen. Domon and Costello’s nomenclature for assigning fragments obtained in tandem MS of oligosaccharides was applied.22 The numbers given are used to assign each carbon atom within the AGU.

five or six sugar-binding subsites (Figure 1a). Cohen et al. have studied Ba Cel5A depolymerization of carboxymethyl cellulose by liquid chromatography/ESI-MS and size exclusion chromatography.26 In this work, we introduce dimethylamine (DMA) as a reagent for reducing end derivatization of methylated oligosaccharides obtained by acidic or enzymatic depolymerization. Tandem MS has been performed to demonstrate the usefulness of DMA derivatization and to characterize the selectivity of Ba Cel5A. Reducing end derivatization provides identification of the end of the oligosaccharide being studied, i.e., reducing or nonreducing end. This is of significant analytical importance, since information regarding the location of the substituents in the oligomer can be obtained and, thus, the enzyme selectivity characterized. Moreover, DMA derivatization increases the sensitivity in MS and also the relative intensities of ions formed through protonation. Since these ions are fragmented predominantly through cleavage of the glucosidic bonds, the corresponding tandem mass spectra become easy to interpret. EXPERIMENTAL SECTION Chemicals. Methyl cellulose SM-1500 (viscosity type 1500 cP, lot number 103674), SM-15 (viscosity type 15 cP, lot number 109614), and SM-4 (viscosity type 4 cP, lot number 907540) with the stated degrees of substitution 1.80, 1.78, and 1.76, respectively, (26) Cohen, A.; Schagerlo¨f, H.; Nilsson, C.; Melander, C.; Tjerneld, F.; Gorton, L. J. Chromatogr., A 2004, 1029, 87-95.

Analytical Chemistry, Vol. 77, No. 9, May 1, 2005

2949

were from Shin-Etsu Chemical Co. (Tokyo, Japan). The MALDI matrix 2,5-dihydroxybenzoic acid (DHB) and the 2.0 mol L-1 DMA solution in methanol (MeOH) were from Aldrich (Steinheim, Germany). Cellooligosaccharide standards, NaBH3CN, and NaBH4 were from Sigma (St. Louis, MO). Trifluoroacetic acid (TFA) was purchased from Riedel-de Hae¨n (Seelze, Germany), and the H2O was from an ELGA Maxima (18.2 MΩ‚cm, Vivendi Water Systems, High Wycombe, U.K.). All other chemicals were of analytical grade. All chemicals were used without further purification. Acidic Depolymerization. Partial acidic depolymerization with TFA was performed according to Momcilovic et al.14 with slight modifications. The reaction time was 120 min. and the solution obtained from the depolymerization was freeze-dried, redissolved in H2O, and then freeze-dried again. This procedure was repeated until the depolymerized MC was obtained as a solid white powder. Conversion of Reducing Ends to Alditols. Prior to enzymatic depolymerization, the reducing end GUs of the MC were converted into their alditol derivative by reduction with NaBH4. To an aqueous solution of MC (1 g L-1) NaBH4 was added to a molar ratio NaBH4/MC of 10 000:1. The reaction was performed in a 1.5-mL screw-capped glass vial for 24 h at room temperature with continuous shaking in an HLC Heating ThermoMixer 130 R (Haep Labor Consult, Bovenden, Germany). Thereafter 100 µL of acetic acid (HAc) was added. The solvent was evaporated under a stream of N2 and heating to 50 °C. To the residue 500 µL of MeOH/HAc (9:1, v/v) was added and the glass vial was shaken for 5 min. This solution was then evaporated under a stream of N2 at room temperature. The procedure was repeated 3 times with MeOH/HAc (9:1, v/v) and thereafter three times with MeOH. The remaining residue was redissolved in H2O (5 mL) and dialyzed in a Spectra/Por membrane MWCO: 6-8000 (Spectrum Medical Industries Inc., Laguna Hills, CA) against H2O (3 L), which was exchanged daily for 7 days. Thereafter the solution was freeze-dried and redissolved in H2O to be used for enzymatic depolymerization. Enzymatic Depolymerization. The purified cellulose selective family 5 endoglucanase from B. agaradhaerens (Ba Cel5A) with a molar mass of 44 703 g mol-1 was a kind gift from the late Dr. Martin Schu¨lein (Novozymes, Bagsværd, Denmark). To 2 mL of the MC solution, 1 g L-1 in H2O, Ba Cel5A was added to a final concentration of 1 µmol L-1. Depolymerization was carried out for 72 h at room temperature, whereafter the solutions were filtered using a centrifuge and Nanosep 10-kDa Omega filters (Pall, Ann Arbor, MI). The permeate was evaporated using a Christ rotation vacuum concentrator model 2-18 (Martin Christ GmbH, Osterode am Harz, Germany). DMA Derivatization. To 500 µL of DMA (2.0 mol L-1 in MeOH) 63 µL of HAc was added. Thereafter, 40 µL of this mixture was mixed with 40 µL of partially depolymerized MC (2.5 g L-1 in H2O) and 120 µL of MeOH in a 1.5-mL screw-capped glass vial. This solution was heated for 5 h at 80 °C in an HLC Heating ThermoMixer 130 R with constant shaking, whereafter 10 µL of NaBH3CN (10 g L-1, 159 mmol L-1) in MeOH was added and the resulting mixture heated to 80 °C for another 4 h. Thereafter, 500 µL of H2O was added and the resulting solution was subjected to continuous shaking in ambient temperature for 12 h. The solvent was evaporated under a stream of N2 and heating to 50 2950

Analytical Chemistry, Vol. 77, No. 9, May 1, 2005

°C and the residue redissolved in 140 µL of 300 mmol L-1 NH4Ac (pH 5)/acetonitrile (ACN) 70/30 (v/v). This solution was then used in the chromatographic purification. DMA derivatization of the cellooligosaccharides was performed by the same procedure. Purification of DMA Derivatives. Purification of the DMA derivatives was performed by size exclusion chromatography. The HPLC pump was a Hewlett-Packard Series 1050 (Agilent Technologies, Palo Alto, CA) equipped with an online degasser. A TSKgel G 3000 PW column (30 cm × 7.5 mm i.d., TosoHaas Bioseparation Specialists, Stuttgart, Germany) with 300 mmol L-1 NH4Ac (pH 5)/ACN (70:30, v/v) as mobile phase at a flow rate of 0.5 mL min-1 was used for the separation. The resolution window of this column is approximately 100-50 000 g mol-1 as calibrated for poly(ethylene oxide)s. It can be expected that for linear oligoand polysaccharides the range may be shifted to slightly higher molar masses. Samples were injected using a Rheodyne model 7010 valve injector (Rheodyne, Berkeley, CA) fitted with a 100-µL sample loop. The detector was a Shimadzu RID10A refractive index detector (Shimadzu, Kyoto, Japan). After determination of the elution times for the DMA derivatives and the unwanted components, the detector was disconnected from the column and volume fractions were then collected. The fractions were evaporated under a stream of N2 and heating to 50 °C. Nano-ESI Mass Spectrometry. Nano-ESI-MS experiments were performed on a Thermo Finnigan TSQuantum (Thermo Electron Corp., San Jose, CA) triple quadrupole mass spectrometer fitted with a modified Thermo Finnigan nano-ESI interface. The modification is a thin metal wire that is spot welded onto the back of the needle mount and inserted through the opening. The glass capillary emitter is then slid onto the wire, which delivers the potential to the liquid. With this arrangement there is no need for conducting emitters, and the problem with evaporation of the metal layer on the emitter is remedied. The DMA derivatized MC oligosaccharides were dissolved at 1 g L-1 in 0.1% HAc (in H2O)/ ACN (1:1, v/v) or in 1 mmol L-1 sodium acetate (in H2O)/ACN (1/1, v/v). The emitters (type ES381, Proxeon Biosystems, Odense, Denmark) were filled with ∼20 µL of analyte solution. The spray voltage was 1000 V. For ion selection in tandem MS, only the monoisotopic peak was selected. This was performed by setting the width of the mass window in the first quadrupole to 0.7 m/z units for singly charged ions and 0.2 m/z units for doubly charged ions. Collision-induced dissociation (CID) was performed with Ar as collision gas. The collision energy was set so that the relative intensity of the parent ion became 10-30% of that of the most abundant fragment ion. MS3 experiments were performed on a Finnigan MAT 900S double-focusing sector mass spectrometer equipped with a quadrupole ion trap (Thermo Electron Corp., Bremen, Germany). The precursor ion was selected using the sector, and subsequent fragmentation, isolation, further fragmentation, and scanning was performed in the trap. The nomenclature proposed by Domon and Costello was used to assign the fragments obtained in tandem MS.22 This nomenclature as well as the assignment of the carbon atoms within the AGU is shown in Figure 1b. Unless stated otherwise, data presented originate from the MC SM-1500.

Figure 2. Mass spectra of DMA-derivatized acidically depolymerized methyl cellulose obtained by nano-ESI-MS (a) and MALDI-MS (b). The analyte concentration was 1 g L-1 in 0.1% HAc (in H2O)/ACN (1:1, v/v) for nano-ESI-MS and 1 g L-1 in H2O for MALDI-MS. The large numbers indicate the DP of the ions within each cluster, and the superscript range indicates the number of methyl groups. In (a), the peaks originating from doubly charged ions with even DPs are located within the peak clusters from the singly charged ions. The doubly charged series are marked with asterisks.

MALDI mass spectrometry. MALDI-MS experiments were performed on a Micromass Q-Tof Ultima Global (Waters, Manchester, U.K.) quadrupole time-of-flight hybrid equipped with a N2 laser emitting at 337 nm. The acceleration voltage in the TOF mass spectrometer was 7.2 kV. The microchannel plate detector voltage was set to 2100 V. The laser was operated at 10 Hz and without any attenuation of the energy. The time-of-flight analyzer was typically operated at a resolution of 10 000 fwhm, and spectra were acquired over a mass range of m/z 200-2000. For ion selection in tandem MS, the quadrupole mass window was ∼4 m/z units wide. Hence, the monoisotopic peak as well as the next two isotopic peaks were selected. Argon was used as collision gas for the CID experiments, and the collision energy was adjusted so that the relative signal intensities of the strongest fragment ions were ∼50% of that of the parent ion. The DMA derivatives were dissolved at 1 g L-1 in H2O and the DHB at 10 g L-1 in H2O. MALDI sample preparation through drying of the sample droplet at reduced pressure was performed according to a procedure described earlier13 with some modifications. After application of the matrix/analyte solution to the MALDI target, it was placed in a homemade gastight plexiglass chamber, which was then connected to a vacuum pump. The solvent evaporated within 30 s. Mass spectra were processed with five-point average smoothing. Unless stated otherwise, data presented originate from the MC SM-1500. RESULTS AND DISCUSSION Nano-ESI Mass Spectrometry. Figure 2a shows a full scan (m/z 200-1500) nano-ESI mass spectrum of DMA-derivatized acidically depolymerized methyl cellulose. Peaks originating from several ion types were detected. The most abundant ions were those formed through protonation. Molecules in the DP range

1-7 were detected for these. Peaks originating from sodiated molecules were also detected, but the relative intensities of those were significantly lower than for the protonated molecules. However, the relative intensities of the peaks originating from the sodiated molecules could be increased by adding 1 mmol L-1 sodium acetate (NaAc) instead of HAc to the analyte solution. The same DPs were detected for the sodiated as for the protonated molecules. Singly charged ions could only be detected for DP