METHOD FOR DETERMINING CARBOHYDRATES STRUCTURE

Abstract

The present invention relates to a method for determining in an expedient manner and with minimal sample consumption the structure of an unknown carbohydrate by using ion mobility-mass spectrometry (IM-MS) in negative ionization mode and fragmentation and a database containing structures of carbohydrates and/or of the fragments of the negative ions of carbohydrates, and for each of the structures of the target carbohydrates the collision cross section value and the mass-to-charge ratio value of the negative ion thereof, and for each of the structures of the fragments of the negative ions of the target carbohydrates the collision cross section value and the mass-to-charge ratio value of the fragment of the negative ion of the target carbohydrate.

Claims

1. A method for determining the structure of a target carbohydrate by ion mobility-mass spectrometry in negative ionization mode comprising the steps: A) providing a sample containing the target carbohydrate; and B) providing a database comprising the structures of the target carbohydrates and/or of the fragments of the negative ions of the target carbohydrates and for each of the structures of the target carbohydrates the collision cross section value and the mass-to-charge ratio value of the negative ion thereof, and for each of the structures of the fragments of the negative ions of the target carbohydrates the collision cross section value and the mass-to-charge ratio value of the fragment of the negative ion of the target carbohydrate; and C) measuring the drift time value and the mass-to-charge ratio value of a negative ion of the target carbohydrate; and D) converting the drift time value measured at step C) to the corresponding collision cross section value; and E) comparing the collision cross section value determined at step D) and the mass-to-charge ratio value measured at step C) with the cross section values and mass-to-charge ratio values stored in the database; and F) determining the structure of the target carbohydrate.

2. The method according to claim 1, further comprising steps E1), E2), E3) and E4) performed after step E) and before step F): E1) subjecting a negative ion of the target carbohydrate to fragmentation to generate fragments of the negative ion of the target carbohydrate; E2) measuring the drift time values and the mass-to-charge ratio values of the fragments of the negative ion of the target carbohydrate, which were generated at step E1); E3) converting the drift time values measured at step E2) to the corresponding collision cross section values; and E4) comparing the collision cross section values determined at step E3) and the mass-to-charge ratio values measured at step E2) of the fragments of the negative ion of the target carbohydrate with cross section values and mass-to-charge ratio values stored in the database.

3. The method according to claim 1, wherein determining the structure of a target carbohydrate comprises determining the stereochemistry of each of the anomeric carbons of the target carbohydrate.

4. The method according to claim 1, wherein determining the structure of a target carbohydrate comprises determining the compositional structure of the target carbohydrate.

5. The method according to claim 2, further comprising step G) which is performed after step F): G) storing the structure of the target carbohydrate determined at step F) and the collision cross section value determined at step D) and mass-to-charge ratio value determined at step C) of the negative ion thereof in the database.

6. The method according to claim 1, wherein the negative ion of the target carbohydrate is selected from the group comprising the deprotonated ion of the target carbohydrate and the anion complex of the target carbohydrate with chloride (Cl.sup.), acetate (CH.sub.3CO.sub.2.sup.), monochloroacetate (CH.sub.2ClCO.sub.2.sup.), dichloroacetate (CHCl.sub.2CO.sub.2.sup.), trichloroacetate (CCl.sub.3CO.sub.2.sup.), trifluoroacetate (CF.sub.3CO.sub.2.sup.), formate (HCO.sub.2.sup.), monobromoacetate (CH.sub.2BrCO.sub.2.sup.), dibromoacetate (CHBr.sub.2CO.sub.2.sup.), tribromoacetate (CBr.sub.3CO.sub.2.sup.), bromochloroacetate (CBrClHCO.sub.2.sup.), chlorodibromoacetate (CClBr.sub.2CO.sub.2.sup.), bromodichloroacetate (CBrCl.sub.2CO.sub.2.sup.), nitrate (NO.sub.3.sup.) or phosphate (H.sub.2PO.sub.4.sup.).

7. The method according to claim 1, wherein the negative ion of the target carbohydrate is the deprotonated ion of the target carbohydrate.

8. The method according to claim 1, wherein the target carbohydrate is a synthetic carbohydrate or a carbohydrate isolated from natural sources.

9. The method according to claim 8, wherein the target carbohydrate is a synthetic carbohydrate and the synthetic carbohydrate further comprises a linker covalently bound to the anomeric carbon of the reducing end monosaccharide of the synthetic carbohydrate.

10. The method according to claim 1, wherein the target carbohydrate comprises between 1 and 50 monosaccharides.

11. The method according to claim 1, wherein the concentration of target carbohydrate in the sample provided at step A) is at least of 0.2 g/mL.

12. The method according to claim 1, wherein the fragmentation performed at step E1) results from collision induced dissociation (CID), electron-transfer dissociation (ETD) or electron-capture dissociation (ECD).

13. The method according to claim 1, wherein the sample containing the target carbohydrate provided at step A) contains at least a further target carbohydrate, which is isobaric with the target carbohydrate.

14. The method according to claim 13, further comprising step H), which is performed after step F) H) determining the relative concentration ratio of each of the target isobaric carbohydrates in the sample.

15. The method according to claim 4, further comprising step G) which is performed after step F): G) storing the structure of the target carbohydrate determined at step F) and the collision cross section value determined at step D) and mass-to-charge ratio value determined at step C) of the negative ion thereof in the database.

Description

DESCRIPTION OF THE FIGURES

[0091] FIG. 1: Structural features of complex carbohydrates. The composition (I) is defined by the monosaccharide content. Monosaccharide building blocks are often isomers, as shown for glucose and galactose, which differ only in their C-4 stereochemistry. Due to the many functional groups, the formation of a new glycosidic bond can occur at several positions, resulting in different connectivities

[0092] (II). Each glycosidic linkage is a new stereocenter that can have either or configuration (III).

[0093] FIG. 2: Correlation between signal intensity and IM peak width in mixtures of 2 and 4. ATDs of [MH].sup. ions from a mixture of <1% 4 and >99% 2. Measurements at high signal intensity can be used to qualitatively detect 4. At low intensity, however, 4 is discriminated leading to a signal, which is indistinguishable from the background.

[0094] FIG. 3: Arrival time distributions of carbohydrates 1-6 as different species in negative ion mode: The difference in percent between the most compact and the most extended isomer of each negative ion is given in percent. The largest CCS differences can be achieved using deprotonated ions and allows for the identification of the regioisomers (e.g. 3/6) and anomers (e.g. 2/3). A clear identification of the regioisomers with a terminal 1.fwdarw.3 or 1.fwdarw.4 glycosidic bond can be obtained for chloride adducts.

[0095] FIG. 4: Comparison of drift times and CCSs of structurally similar precursor ions and fragments. a) Mass spectra of 5 and 6, as well as MS/MS spectrum of 7 (-Gal-(1.fwdarw.3)--GlcNAc-(1.fwdarw.3)--Gal-(1.fwdarw.4)--Gal-(1.fwdarw.4)--Glc-L; L=C.sub.5H.sub.10NH.sub.2) in negative ion mode. The pentasaccharide 7 exhibits an identical core structure as the trisaccharide 6. CID of deprotonated 7 consequently leads to a fragment of identical mass as the deprotonated precursor ion of 6. b) Arrival time distributions of [MH].sup.=588 ions. The CID fragment arising from deprotonated 7 exhibits an identical drift time and CCS as the intact deprotonated trisaccharide 6. This indicates that glycans and glycan fragments with identical structure also exhibit identical CCSs.

[0096] FIG. 5: IM-MS differentiation and identification of the larger hexasaccharides 8 and 9. Deprotonated ions 8 and 9 show almost identical drift times and therefore, cannot be distinguished. However, smaller CID fragments containing five, four, and three monosaccharide building blocks (m/z 1017, 832, and 652, respectively) exhibit highly diagnostic drift times. At m/z 832 a double peak is observed for the branched oligosaccharide 8 (inset, black trace), because two isomeric fragments are formed. Both fragments can be detected simultaneously using IM-MS with cleavage at the 3-antenna being clearly preferred. The disaccharide fragments at m/z 467 and 364 are identical for 8 and 9 and consequently exhibit identical drift times.

[0097] FIG. 6: IM-MS analysis of a mixture of two target carbohydrates. Arrival time distributions of the trisaccharides 1-6 compared to the mixture of 5 and 30 as a) [MH].sup.=588 and b) [M+Cl].sup.=624 ions clearly reveals a content of about 5% of carbohydrate 30. Especially the drift time of the chloride adduct of 30 is very diagnostic, since it differs considerably from all other trisaccharides investigated here. Carbohydrate 30 could be detected by NMR, but due to the low concentration in the sample, an assignment of the structure cannot be performed.

[0098] FIG. 7: Arrival time distributions (ATDs) of isomeric milk sugars with and without aminopentanol linker (L). To investigate the influence of a linker to the separation of carbohydrates, the milk sugars lacto-N-tetraose (LNT, -Gal-(1.fwdarw.3)--GlcNAc-(1.fwdarw.3)--Gal-(1.fwdarw.4)-Glc) and lacto-N-neo-tetraose (LNnT, -Gal-(1.fwdarw.4)--GlcNAc-(1.fwdarw.3)--Gal-(1.fwdarw.4)-Glc) were synthesized with and without an aminopentanol linker (L). Mixtures of both isomer pairs were analyzed in the negative ion mode using IM-MS. The ATDs of the chloride adducts show for both mixtures only one drift peak, which does not enable the separation of the milk sugars. However, the drift times of the deprotonated ions clearly differ and can be used to identify each isomer. The separation of the isomers was possible with and without the presence of a linker and thus indicates that the differences in drift time predominantly result from differences in carbohydrate structure.

[0099] FIG. 8: Relative quantification of configurational trisaccharide isomers. Mixtures of the configurational isomers 2 and 3 were measured using IM-MS. a) The amount of 2 was kept constant while 3 was diluted to yield relative concentrations of 50%, 30%, 10%, 1%, 0.1%, and 0%. Minor components with relative concentrations as low as 0.1% can still be qualitatively detected. b) 3D plot showing the separation of anomers 2 and 3. The intensity is plotted using a logarithmic scale and impurities of 0.1% can be clearly identified without magnification. c) Plot of the relative concentration of 3 against the corresponding relative IM-MS intensity to illustrate the dynamic range of the method. A value of 0.5 represents equal amounts of 2 and 3, while 0 and 1 indicate the presence of only 2 or 3, respectively. The grey error bars correspond to the double standard deviation observed for three independent replicates.

[0100] FIG. 9: Arrival time distributions of the deprotonated trisaccharides 1-6, 33, 35-37 and the isobaric fragment ion 34*. For the investigated [MH].sup.=588 trisaccharide ions large differences in drift time are observed. When isomer pairs that differed only in the configuration of the last glycosidic bond (2/3, 5/6, 33/34*) were compared, the a anomer showed a shorter drift time compared to the anomer in all cases. Furthermore, carbohydrates that exhibit the same stereo- and regiochemistry of the glycosidic bonds, but a different composition of the building blocks (5/34*, 6/33) can be differentiated. Taken together this indicates that isomers, which differ in regio- or stereochemistry at one site typically exhibit drift time differences large enough to enable their differentiation.

[0101] The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples which follow represent techniques discovered by the inventor to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

[0102] Further modifications and alternative embodiments of various aspects of the invention will be apparent to those skilled in the art in view of this description. Accordingly, this description is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It is to be understood that the forms of the invention shown and described herein are to be taken as examples of embodiments. Elements and materials may be substituted for those illustrated and described herein, parts and processes may be reversed, and certain features of the invention may be utilized independently, all as would be apparent to one skilled in the art after having the benefit of this description of the invention. Changes may be made in the elements described herein without departing from the spirit and scope of the invention as described in the following claims.

EXAMPLES

[0103] A) Analysis of Carbohydrates Using Ion Mobility-Mass Spectrometry:

[0104] Ion mobility-mass spectrometry (IM-MS) experiments were performed on a travelling wave quadrupole/IMS/oa-ToF MS instrument, Synapt G2-S HDMS (Waters, Manchester, U.K.) (Int. J. Mass Spectrom. 2007, 1-12), which was mass calibrated prior to measurements using a solution of cesium iodide (100 mg/mL). IM-MS data analysis was performed using MassLynx 4.1, DriftScope 2.4 (Waters, Manchester, UK), and OriginPro 8.5 (OriginLab Corporation, Northampton) software. For IM-MS analysis compounds 1-9, milk sugar lacto-N-tetraose (LNT, -Gal-(1.fwdarw.3)--GlcNAc-(1.fwdarw.3)--Gal-(14)-Glc), milk sugar lacto-N-neo-tetraose (LNnT, -Gal-(1.fwdarw.4)--GlcNAc-(1.fwdarw.3)--Gal-(1.fwdarw.4)-Glc), 33-37, the crude mixture 5/30 were each dissolved in water/methanol (1:1, v/v) to a concentration of 1-10 mol/L.

TABLE-US-00002 TABLE 2 Structures of the saccharides analyzed by ion mobility-mass spectrometry in negative ionization mode. [00001] 1 [00002] 2 [00003] 3 [00004] 4 [00005] 5 [00006] 6 [00007] 7 [00008] 8 [00009] 9 [00010] 30 [00011] 33 [00012] 35 [00013] 36 [00014] 37

[0105] A nano-electrospray source (nESI) was used to ionize 3-5 L of sample from platinum-palladium-coated borosilicate capillaries prepared in-house. Typical settings were: source temperature, 20 C.; needle voltage, 0.8 kV; sample cone voltage, 25 V; cone gas, 0 L/h. The ion mobility parameters were optimized to achieve maximum resolution without excessive heating of the ions upon injection into the IM cell. Values were: trap gas flow, 2 mL/min; helium cell gas flow, 180 mL/min; IM gas flow, 90 mL/min; trap DC bias, 35 V; IM wave velocity, 800 m/s; IM wave height, 40 V. For MS/MS experiments the trap collision energy was increased to 30-60V.

[0106] IM-MS Spectra of each individual carbohydrate and three trisaccharide mixtures (6/3, 3/2 and 5/6) were recorded in negative ion mode. Arrival time distributions (ATD) were extracted from raw data using MassLynx and drift times were determined manually via Gaussian fitting using Origin 8.5. For the measurement of the individual carbohydrates, the m/z signal intensity was kept at approximately 10.sup.3 counts per second to avoid saturation and subsequent broadening of the corresponding drift peak. In order to avoid discrimination of a minor component, an average signal intensity of 10.sup.4 counts per second was used for the semi-quantitative assessment of mixtures (see FIG. 2). Under these conditions, minor components with relative concentrations below 1% can still be detected qualitatively, but a semi-quantitative assessment is no longer possible. For unknown mixtures, it is therefore required to acquire data at both high and low intensity settings. In the former case, minor components with relative concentrations below 1% can be qualitatively detected, while the latter case typically yields a better IM resolution and enables a semi-quantitative assessment (see FIG. 2). In addition, an acquisition at different intensity settings can help to evaluate mixtures in which the isomers cannot be fully resolved. For broad and inconclusive ATDs, a comparison with neighboring peaks of similar mass and charge can furthermore be used to distinguish between overlapping and saturated peaks. (Anal. Chem. 2013, 85, 5138-5145).

[0107] CCS estimations were performed using an established protocol and dextran as calibrant (Dextran MW=1000 and Dextran MW=5000, Sigma Aldrich) (Anal. Chem. 2013, 85, 5138-5145; Anal. Chem. 2014, 86, 10789-10795). The calibration solution consisted of 0.1 mg/mL dextran1000, 0.5 mg/mL dextran5000, and 1 mM NaH.sub.2PO.sub.4 in water:methanol (1:1, v/v). The calibrant and each sample were measured on a travelling wave Synapt instrument at five wave velocities in negative ion mode. Drift times where extracted from raw data by fitting a Gaussian distribution to the arrival time distribution of each ion and corrected for their m/z dependent flight time. CCS reference values of dextran were corrected for charge and mass and a logarithmic plot of corrected CCSs against corrected drift times was used as a calibration curve to estimate CCSs. One calibration curve was generated for every wave velocity and each ion polarity. The resulting five estimated CCSs for each sample ion were averaged. These measurements where repeated three times and the averaged values for different ions are presented in Table 3. The reported error corresponds to the standard deviation obtained for three independent replicates.

[0108] FIG. 3 illustrates the arrival time distributions of target carbohydrates 1-6 as different species in negative ion mode. FIGS. 4 and 5 show how the assignment of the structure to large carbohydrates can be performed using the method described herein on the basis of the cross collisions values and the mass-to-charge ratios of the fragments of the negative ions thereof. FIG. 6 shows how the structure of each of the target carbohydrates contained in an analyte containing several target carbohydrates (herein the analyte is a mixture of 5 and 30) can be determined by using the inventive method. FIG. 7 illustrates that the current method enables the determination of the structure of carbohydrates functionalized or not with a linker at the anomeric position.

TABLE-US-00003 TABLE 3 Estimated nitrogen CCSs (.sup.TWCCS.sub.N2) for trisaccharides 1-6 and 30. Each .sup.TWCCS.sub.N2 is an average of three independent measurements with the corresponding standard deviation STD. Target .sup.TWCCS.sub.N2 Negative .sup.TWCCS.sub.N2 Carbohydrate Negative Ion (.sup.2) STD (.sup.2) m/z Ion (.sup.2) STD (.sup.2) m/z 1 [M H].sup. 294.4 1.1 588 [M + Cl].sup. 244.4 1.1 624 2 [M H].sup. 249.8 1.5 588 [M + Cl].sup. 242.2 1.5 624 3 [M H].sup. 233.2 1.3 588 [M + Cl].sup. 244.5 1.2 624 4 [M H].sup. 237.4 0.9 588 [M + Cl].sup. 229.7 0.8 624 5 [M H].sup. 235.6 1.0 588 [M + Cl].sup. 227.3 0.8 624 6 [M H].sup. 219.9 1.6 588 [M + Cl].sup. 224.6 1.4 624 30 [M H].sup. 248.4 0.3 588 [M + Cl].sup. 256.7 0.2 624

[0109] B) Semiquantitative Analysis of Carbohydrates Mixtures.

[0110] For the semiquantitative analysis of anomeric trisaccharide mixtures a quantification experiment was performed using isomers 2 and 3. Stock solutions of 2 and 3 with identical concentration were prepared in water/methanol (1:1, v/v). Each stock solution was diluted individually to yield relative concentrations (rel. conc.) of 80, 56, 43, 25, 11, 5, 1, 0.1, and 0.01%. The serial dilutions were used to obtain isomer mixtures with concentration ratios x(3)=c[3]/(c[2]+c[3]) between 0 and 1 (see Table 4). A value of 0.5 represents equal amounts of 2 and 3, while 0 and 1 indicate the presence of only 2 or 3, respectively.

TABLE-US-00004 TABLE 4 Relative concentrations of 2 and 3 in the investigated mixtures and their corresponding relative concentration ratio x(3) = [3]/[3 + 2]. Measured relative intensities Int.sub.rel(3) = A[3]/ (A[2] + A[3]) were calculated from the drift peak areas of the deprotonated species [M H].sup. = 588.4 and the standard deviation (STD) was obtained from three independent replicates. rel. conc. 3 rel. conc. 2 theoretical x(3) measured Int.sub.rel(3) STD 1 100 0.01 0.04 0.011 5 100 0.05 0.07 0.004 11 100 0.10 0.10 0.007 25 100 0.20 0.18 0.005 43 100 0.30 0.27 0.005 56 100 0.36 0.35 0.005 80 100 0.44 0.42 0.010 100 100 0.50 0.49 0.007 100 80 0.56 0.55 0.016 100 56 0.64 0.60 0.005 100 43 0.70 0.69 0.008 100 25 0.80 0.78 0.005 100 11 0.90 0.89 0.010 100 5 0.95 0.93 0.012 100 1 0.99 0.97 0.007

[0111] To achieve constant experimental conditions, the semi-quantitative analysis was performed on a Synapt instrument equipped with an online nano-ESI source that was coupled to an ACQUITY UPLC System (Waters, Manchester, U.K.). Settings were: eluents, 0.1% formic acid in methanol/0.1% formic acid in water at a constant rate of 50%, flow rate 8 L/min, sample injection: 10 L. Data were acquired in negative ion mode with following settings: source temperature, 80 C.; needle voltage, 2.7 kV; sample cone voltage, 25 V; desolvation temperature 150 C., 430 cone gas, 0 L/h, nanoflow gas 1.3 bar, purge gas flow 500.0 mL/h. Ion mobility parameter were: trap gas flow, 0.4 mL/min; helium cell gas flow, 180 mL/min; IM gas flow, 90 mL/min; trap DC bias, 45 V; IM wave velocity, 800 m/s; IM wave height, 40 V.

[0112] Extraction of the ATD of the 588.4 m/z ion showed two separate arrival times, each of which corresponded to one of the two isomers. The area under the ATD is related to the concentration of the sample. Therefore, the theoretical concentration ratio x(3) was compared to the ratio of the drift time peak areas Int.sub.rel(3)=A[3]/(A[2]+A[3]) (see FIG. 8). A linear correlation was observed, demonstrating the semi-quantification of one isomer in the presence of another, down to contents of 1% of the minor component. Relative concentrations between 1 and 0.1% were still qualitatively detectable, but a determination of the relative content was not possible anymore due to detector saturation caused by the major component.