Single Stage Tandem Mass Spectrometry Assignment of the C-5 Uronic Acid Stereochemistry in Heparan Sulfate Tetrasaccharides using Electron Detachment Dissociation
The analysis of heparan sulfate (HS) glycosaminoglycans presents many challenges, due to the high degree of structural heterogeneity arising from their non-template biosynthesis. Complete structural elucidation of glycosaminoglycans necessitates the unambiguous assignments of sulfo modifications and the C-5 uronic acid stereochemistry. Efforts to develop tandem mass spectrometric-based methods for the structural analysis of glycosaminoglycans have focused on the assignment of sulfo positions. The present work focuses on the assignment of the C-5 stereochemistry of the uronic acid that lies closest to the reducing end. Prior work with electron-based tandem mass spectrometry methods, specifically electron detachment dissociation (EDD), have shown great promise in providing stereo-specific product ions, such as the B3 ´ –CO2, which has been found to distinguish glucuronic acid (GlcA) from iduronic acid (IdoA) in some HS tetrasaccharides. The previously observed diagnostic ions are generally not observed with 2-O-sulfo uronic acids or for more highly sulfated heparan sulfate tetrasaccharides. A recent study using electron detachment dissociation and principal component analysis revealed a series of ions that correlate with GlcA versus IdoA for a set of 2-O-sulfo HS tetrasaccharide standards. The present work comprehensively investigates the efficacy of these ions for assigning the C-5 stereochemistry of the reducing end uronic acid in 33 HS tetrasaccharides. A diagnostic ratio can be computed from the sum of the ions that correlate to GlcA to those that correlate to IdoA.
KeywordsGlycosaminoglycans Heparan sulfate Uronic acids Stereochemistry FTICR Electron detachment dissociation
A large portion of the extracellular matrix and basement membranes are comprised of proteoglycans, composed of proteins covalently attached to the class of carbohydrates called glycosaminoglycans (GAGs) [1, 2, 3]. GAGs are linear negatively charged biopolymers, the basic building blocks of which consist of a repeating disaccharide sequence of an amino sugar and a uronic acid or galactose [4, 5]. They are categorized as either keratan sulfate (KS), chondroitin sulfate (CS), dermatan sulfate (DS), hyaluronan (HA), or heparan sulfate (HS) depending on their disaccharide repeating unit [1, 2, 3, 4, 5, 6]. Among these classes of GAGs, HS is the most structurally complex [6, 7]. They are initially synthesized in the Golgi apparatus as alternating disaccharide units of D-glucuronic acid and N-acetylated glucosamine [6, 8]. C5-epimerization of glucuronic acid (GlcA) to iduronic acid (IdoA) occurs during the biosynthesis of HS followed by a series of sulfo modifications [7, 8]. Sulfo modifications may occur at the 2-O position of the uronic acid, and the N-, 3-O, and 6-O positions of the glucosamine unit . These structural modifications often do not go to completion, producing HS chains with varying sequences of sulfation, acetylation, and IdoA/GlcA content [6, 9]. Despite these varying structural modifications, specific structural motifs on HS chains have been reported to bind target proteins with high specificity. These categories of proteins, called heparan sulfate binding proteins (HSBPs), include chemokines, cytokines, blood coagulation factors, such as serine proteases, cell adhesion proteins, growth factors, and morphogenetic factors [7, 10, 11]. Documented physiologic processes influenced by HS interactions with proteins include growth and development, cancer, inflammation, viral infectivity, and blood coagulation [12, 13, 14, 15]. The numerous biological functions of this bio-molecule continue to inspire research into structure–function relationships of HS oligomers. However, this research has been hampered by their enormous micro-heterogeneity and limited availability requiring very sensitive and robust analytical methods for their analysis.
Advanced analytical methods like nuclear magnetic resonance (NMR) spectroscopy have been used to determine sulfo modifications and the C-5 stereochemistry of the uronic acid in GAGs [16, 17]. However, the quantity and purity of GAGs extracted from natural sources are often not suitable for NMR analysis . These drawbacks make mass spectrometry an excellent alternative for GAG analysis. Negative electrospray ionization mass spectrometry offers an excellent platform for GAG analysis, offering high sensitivity, throughput, and accuracy [19, 20]. However, tandem mass spectrometry of GAGs, especially heparan sulfate, is often challenging due to variations in oligomer length, hexuronic acid stereochemistry, and sulfation heterogeneity[6, 21]. ESI-MS is able to determine the length, degree of sulfation of GAGs, and other features affecting the elemental composition. To determine sites of sulfation, N-acetylation, and hexuronic acid stereochemistry, more advanced methods are required. Recent advances in tandem mass spectrometric applications to GAGs using collision-induced dissociation (CID) [22, 23], infrared multiphoton dissociation (IRMPD) , electron induced dissociation (EID) , electron detachment dissociation(EDD) [26, 27, 28], and negative electron transfer dissociation (NETD)  have addressed some of the challenges encountered during their structural analysis. Inherent sulfo decomposition of the labile sulfate half ester group present in GAGs hinders structural characterization. This phenomenon occurs mostly during the ionization and ion activation stages of the experiment. Chemical derivatization , deprotonation of the acidic groups [23, 31, 32], and metal cation exchange [22, 23] have been reported to effectively reduce sulfo decomposition depending on the degree of sulfation. Recent CID MS/MS reports on Arixtra and highly sulfated heparan sulfate GAGs showed that one can obtain very rich and structurally informative product ion coverage by adding dilute NaOH to the spray solution [33, 34]. Electron-based activation methods, especially EDD, have shown great potential in providing highly informative cross-ring products as well as the corresponding glycosidic cleavages, which are essential for localization of sulfo positions [25, 26, 27]. Although mass spectrometry methodologies continue to gain ground in assigning sites of sulfo modifications, a remaining challenge has been the inability to discriminate diastereomers that differ by the chirality of the uronic acid C-5 center. Zaia and coworkers were the first to address this challenge. Using collision induced dissociation, they were able to differentiate chondroitin sulfate (CS) from dermatan sulfate (DS) in mammalian extracellular matrix, and quantitatively assign the amount of the diastereomers present in mixtures . The ability to assign the C-5 stereochemistry in uronic acid residues of 4-O-sulfo chondroitin sulfate epimers with varying oligomer lengths (dp4-dp10) based on diagnostic cross-ring ions 2,4An and 0,2Xn in CID mass spectra has also been reported by Kailemia et al. . Compared with heparan sulfates, chondroitin sulfates have a well-defined sulfation pattern; hence, the former requires a more sensitive activation method for stereochemistry assignments. EDD results for HS tetrasaccharides reported by Wolff et al. showed the possibility of obtaining a stereospecific ion B3′-CO2 for assigning the C-5 stereochemistry for moderately sulfated tetramers (0–0.25 sulfates per disaccharide) . Gas-phase separation of epimeric mixtures of HS tetrasaccharides using field asymmetric ion mobility spectrometry (FAIMS) followed by EDD fragmentation confirmed the presence of the B3′-CO2 ion for the GlcA-containing epimer . Recent EDD reports, however, showed that the presence of a sulfo group at the 2-O position of the uronic acid hinders production of the B3´-CO2 ion in GlcA-containing epimers . More recent work on the assignment of the C-5 stereochemistry for 2-O-sulfated HS tetramers (0.5–2.5 sulfates per disaccharide) revealed the possibility of assigning the C-5 stereochemistry of HS tetrasaccharides using a ratio combination of selected ions obtain from EDD-PCA experiments . Such analyses are useful when epimeric compounds are available. The scarcity of naturally occurring epimeric HS samples has motivated the development of a technique that assigns the stereochemistry of these tetramers without reference to their isomers. Here we present for the first time a more general approach in assigning the C-5 hexuronic stereochemistry for 33 HS tetrasaccharide standards from a single stage EDD tandem mass spectrum.
Thirty-three heparan sulfate tetrasaccharides standards were synthesized using a modular approach . All the compounds examined had their compositions confirmed using FTICR MS accurate mass measurement and had their structures confirmed by 1H NMR, HSQC, and COSY. Supplementary Figures 1–4 show the chemical structures of all 33 compounds.
Mass Spectrometry Analysis
EDD experiments were performed on a 9.4T Bruker Apex Ultra QeFTMS (Billerica, MA, USA) with a hollow cathode (HeatWave, Watsonville, CA, USA), which serves as the source of electrons for EDD; 0.1 mg/mL of each standard were injected at a rate of 120 μL/h in 50:50 methanol:H2O and ionized by a metal electrospray capillary (#G2427A; Agilent Technologies, Santa Clara, CA, USA). Where necessary, 0.1 mM NaOH was added to the spray solvent to enhance the intensity of preferred sodium adducted precursor ions for analysis. All the HS tetrasaccharides were analyzed in the negative ion mode. Each EDD experiment was repeated three times with almost similar results for each HS standard examined.
For the EDD experiment, multiply charged precursor ions were isolated in the external quadruple and accumulated for 1–4 s before injection into the FT-ICR cell. Precursor ion selections were refined using in-cell isolation with a coherent excitation frequency (CHEF) event. These ions were then irradiated with 19 eV electrons for 1 s. The extraction lens was set to –18.5 ± 0.5 V with the cathode heater at 1.5 A. Twenty-four acquisitions were signal averaged per spectrum; 512 K points were acquired for each spectrum, padded with one zero fill, and apodized using a sine bell window. Internal calibration was achieved using confidently assigned glycosidic product ions as internal calibrants, providing mass accuracy of <1 ppm. We report peaks with S/N >10 due to the large number of low-intensity product ions formed by EDD. All cross-ring and glycosidic product ions generated from the EDD experiment were assigned using accurate mass measurement and GlycoWorkbench . These ions are reported using the Domon and Costello nomenclature .
Results and Discussion
Determination of Sulfo Positions
Assignment of the Reducing End Hexuronic Acid Stereochemistry
Equation 1 computes the ratio of the sum of intensities of ions that are statistically validated as diagnostic for the presence of GlcA to the sum of intensities of ions diagnostic of IdoA. The factor of one-third applied to each sum was determined empirically to produce a positive value for the diagnostic ratio (DR) when GlcA is present at the second residue from the reducing end, and negative values when IdoA is present in the same position. The DR values computed for all the standards include fragments with sulfo losses from the selected ions. The discussion below examines the results for a comprehensive set of HS standards with 1 to 4 sulfo modifications.
Diagnostic Ratio Results for Mono-Sulfated HS Standards
Diagnostic Ratio Results for Di-Sulfated HS Standards
Diagnostic Ratio Results for Tri-Sulfated HS Standards
Diagnostic Ratio Results for the Tetra-Sulfated HS Standards
In this study, we have demonstrated the capability to assign the C-5 hexuronic acid stereochemistry from a single stage tandem mass spectrum using EDD. The diagnostic ratio provides the means to assign the stereochemistry of the uronic acid near the reducing end for HS tetrasaccharides. For all 33 tetramers that were examined, the diagnostic ratio was positive for GlcA near the reducing end, whereas those having IdoA residues near the reducing end had negative DR values. The smallest absolute value of DR for an IdoA-containing tetrasaccharide standard was –0.06 ± 0.01, and for a GlcA was 0.05 ± 0.01. These data show that the diagnostic ratio clearly distinguishes the uronic acid stereochemistry, not by comparison to a standard, but with a number derived directly from the data.
The applicability of this approach to typical analytical problems faced by glycosaminoglycan researchers remains to be demonstrated. Generally speaking, researchers are confronted with mixtures of GAG oligomers, and these would need to be resolved before the type of analysis presented in this paper could be performed, as the diagnostic ratio only has significance for single component samples. Secondly, the diagnostic ratio presented here has been demonstrated on tetramers that are alkylated at the reducing end. This modification breaks the symmetry of the structure and allows one to easily distinguish reducing end from non-reducing end fragments. This derivatization can be performed on real-world samples, but will require an extra step in the work-up procedure. Finally, this approach has been demonstrated only for assigning the stereochemistry of the uronic acid residue closest to the reducing end. Future work will focus on extending this approach to assign the C-5 hexuronic acid stereochemistry of additional residues in longer chain HS GAGs using EDD.
The authors are grateful for generous support from the National Institutes of Health, 2P41GM103390.
- 5.Esko, J.D., Kimata, K., Lindahl, U.: Proteoglycans and sulfated glycosaminoglycans. In: Essentials of Glycobiology. 2nd Edition.;A. Varki; Cold Spring Harbor Laboratory Press, pp. 229–248 Cold Spring Harbor, NY (2009)Google Scholar
- 14.Kleeff, J., Ishiwata, T., Kumbasar, A., Friess, H., Büchler, M.W., Lander, A.D., Korc, M.: The cell-surface heparan sulfate proteoglycan glypican-1 regulates growth factor action in pancreatic carcinoma cells and is overexpressed in human pancreatic cancer. J. Clin. Investig. 102, 1662 (1998)CrossRefGoogle Scholar
- 20.Zaia, J.: Mass spectrometry of oligosaccharides. Mass Spectrom. 23, 161–227 (2004)Google Scholar
- 21.Leach, F.E., Arungundram, S., Al-Mafraji, K., Venot, A., Boons, G.-J., Amster, I.J.: Electron detachment dissociation of synthetic heparan sulfate glycosaminoglycan tetrasaccharides varying in degree of sulfation and hexuronic acid stereochemistry. Int. J. Mass Spectrom. 330, 152–159 (2012)CrossRefGoogle Scholar
- 36.Kailemia, M.J., Patel, A.B., Johnson, D.T., Li, L., Linhardt, R.J., Amster, I.J.: Differentiating chondroitin sulfate glycosaminoglycans using CID; uronic acid cross-ring diagnostic fragments in a single stage of MS/MS. Eur. J. Mass Spectrom. (Chichester, England) 21, 275 (2015)CrossRefGoogle Scholar
- 37.Kailemia, M.J., Park, M., Kaplan, D.A., Venot, A., Boons, G.-J., Li, L., Linhardt, R.J., Amster, I.J.: High-field asymmetric-waveform ion mobility spectrometry and electron detachment dissociation of isobaric mixtures of glycosaminoglycans. J. Am. Soc. Mass Spectrom. 25, 258–268 (2014)CrossRefGoogle Scholar
- 38.Agyekum, I., Patel, A.B., Zong, C., Boons, G.-J., Amster, I.J.: Assignment of hexuronic acid stereochemistry in synthetic heparan sulfate tetrasaccharides with 2-O-sulfo uronic acids using electron detachment dissociation. Int. J. Mass Spectrom. 390, 163–169 (2015)Google Scholar