Dipeptide-Based Carbohydrate Receptors and Polymers for

Research Article www.acsami.org

Dipeptide-Based Carbohydrate Receptors and Polymers for Glycopeptide Enrichment and Glycan Discrimination Guangyan Qing,†,§ Xiuling Li,‡,§ Peng Xiong,† Cheng Chen,‡ Mimi Zhan,† Xinmiao Liang,*,‡ and Taolei Sun*,†,⊥ †

State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, People’s Republic of China ‡ Key Laboratory of Separation Science for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, People’s Republic of China ⊥ School of Chemistry, Chemical Engineering and Life Science, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, People’s Republic of China S Supporting Information *

ABSTRACT: Glycoproteomics identifies and catalogs protein glycosylation and explores its impact on protein conformations and biofunctions. However, these studies are restricted by the bottleneck to enrich low-abundance glycopeptides from complex biosamples and the difficulties in analyzing glycan structures by mass spectrometry. Here, we report dipeptide as a simple but promising carbohydrate binding platform to tackle these problems. We build a hydropathy-index-based strategy for sequence optimization and screen out three optimal dipeptide sequences from 54 types of dipeptides. The optimized dipeptide-based homopolymers display excellent performance (e.g., selectivity up to ∼70% for real biosamples and strong anti-interference capacity capable of resisting 1000-fold bovine serum albumin interference) in glycopeptide enrichment. Meanwhile, our polymers exhibit high-efficiency chromatographic separation toward oligosaccharides with different compositions, polymerization degrees and even their linkage isomers. This brings another attractive feature that our materials can discriminate subtly variable glycan structures of glycopeptides, especially, isomeric glycosidic linkages. These features provide a solid foundation to analyze the complex glycan structures and glycosites simultaneously, which will benefit future development of glycoproteomics and glycobiology. KEYWORDS: polymer, interface, glycoproteomics, carbohydrate, enrichment

1. INTRODUCTION Glycoproteomics has been arousing more and more interest, because of its significant importance in life science and medicine.1,2 Increasing evidence have indicated that cancers and other major diseases are closely related to abnormal glycosylation of proteins;3−6 until now, most of the cancer biomarkers applied in clinical medicine are glycoproteins.7,8 Glycoproteomic studies rely strongly on the mass spectrometry (MS) analysis9−11 of glycopeptides digested from total proteins in biosamples. This, however, is largely hampered by the bottleneck to enrich ultralow abundance glycopeptides from the hydrolysates containing large amounts of nonglycopeptides (NGs) and other impurities.12−14 Glycopeptides possess one or more saccharide moieties (glycans) that are covalently attached to peptide chains. Although several types of carbohydrate receptors have been used to develop glycopeptide enrichment materials and methods,15,16 they usually show unsatisfied enrichment selectivity and/or other problems.17,18 For example, lectin-affinity chromatography (LAC)19 is generally recognized © 2016 American Chemical Society

as the most promising enrichment method, because of the high specific binding of lectins (saccharide-binding proteins) to saccharides. However, high specificity makes each lectin effective only for a narrow subset of glycopeptides and vast amounts of other information will be lost.20 Meanwhile, various artificial enrichment methods, such as boronate-affinity chromatography (based on reversible binding with cisdiols),21,22 hydrazide chemistry (aldehyde-based chemical reaction),23−25 TiO2 affinity (chelation binding with negatively charged peptides),9,26,27 hydrophilic interaction chromatography (difference in hydrophilicity),18,28 or the combination of these methods have been developed and widely used in glycoproteomics. However, each method has its unique advantages and disadvantages.15−17 Therefore, it is necessary Received: June 28, 2016 Accepted: August 8, 2016 Published: August 8, 2016 22084

DOI: 10.1021/acsami.6b07863 ACS Appl. Mater. Interfaces 2016, 8, 22084−22092

Research Article

ACS Applied Materials & Interfaces to develop new carbohydrate-binding platform to solve the challenges of glycopeptide enrichment. On the other hand, the glycans of glycopeptides are highly complex with variable compositions, sequences, linkages, and other structures.29 Nevertheless, slight differences in glycans may induce very different life and disease processes.30 For example, prostate cancer features an increased expression of core-fucosylated biantennary glycans and α-2,3-sialylation,31 while α-2,6-sialylation is overexpressed in breast and some other cancers.32 Although the glycan compositions can be analyzed by MS, MS cannot directly give the information on isomerization and branch structures of glycans.33 Therefore, another challenge in glycoproteomic studies is how to discriminate different linkage isomers and other glycan structures at the same time of enrichment. This, however, is very difficult, which requires elaborate molecular design34 for carbohydrate receptors. Here, we show that this may be solved in a simple way. Saccharides possess extended, complex structures with hydroxyls as the main functional groups arrayed in different stereochemistry.34 While multiple chiral centers lead to large scope of variation, differences among them are sometimes subtle. This makes the development of synthetic receptors35−37 for saccharide discrimination very difficult, which is even more challenging when highly variable glycan structures of glycopeptides are involved.38 It conventionally requires sophisticated molecular design and/or well-defined spatial structures. As a typical example, lectins39−42 bind to saccharides specifically via multiple hydrogen bonds (H-bonds) from particular amino acid residues with definite spatial arrangements (Figure 1A). Inspired by this, here, we show that oligopeptides can act as a promising saccharide binding platform for glycopeptide analysis. Amino acid residues of oligopeptides provide multiple complementary H-bonding sites43−45 for saccharide hydroxyls, leading to good affinities between the two components. The flexible stereoconformations of oligopeptides can adapt differentially to various saccharides, thus enabling saccharide discrimination via the variation in affinities; notably, the lack of definite conformations also avoids the problem of excessive specificity as lectins. Finally, easily designable sequence enables fine adjustment of saccharide/ oligopeptide interactions, thereby facilitating the optimization of receptors to achieve the best performance. We found that these features could be fully manifested on dipeptides,46,47 and further elongating the chain length would exponentially increase the numbers of sequences; thus, we focused primarily on dipeptides here.

Figure 1. Dipeptide-based carbohydrate receptors. (A) Binding model of an earthworm lectin (Protein Databank (PDB) No. 2DS0) to sialylactose, comprised of 8 sets of H-bonds (green dashed lines) contributed by particular amino acid residues arranged in well-defined spatial structures. (B) Results of the orthogonal investigation of association constants (Ka) between 15 typical dipeptides and 7 model monosaccharides and homologues, obtained by fluorescent titration experiment. (C) Optimized binding models between PD and Neu5Ac (left) or glucose (right). PD had stronger binding (6 sets of H-bonds) to Neu5Ac than to glucose (3 sets of weak H-bonds). (D) Hydropathy plot of the discretization index (D) for 54 dipeptides investigated in this study. The plot shows that dipeptides with hydropathy indices between −0.2 and −1.2 have the lowest D values (inside the red dashed circle). (E) Corresponding hydropathy plot of the Ka(max)/ Ka(min) ratio (R). The combination of panels (D) and (E) produced four optimized sequences with small D and large R values: DY, YD, PD, and PE. (F) Association constants with Neu5Ac for 54 dipeptides. Interestingly, YD, DY, and PD also showed affinities to Neu5Ac that were apparently higher than those of most of the other dipeptides. (G) Chemical structure of poly-PD-modified mesoporous silica gel.

2. RESULTS AND DISCUSSION 2.1. Dipeptide Sequence Screening. Considering the great variation in saccharide structures, we targeted seven important monosaccharides and homologues that constitute the main structural units of glycopeptides: N-acetyl-neuraminic acid (Neu5Ac, the most common member of sialic acid (SA) family, which is a class of neuraminic acid derivatives that are always expressed at the distal ends of glycan chains in vertebrate cell surfaces48,49), mannose, glucose, galactose, Nacetylated glucosamine (GlcNAc), N-acetylated galactosamine (GalNAc), and fucose (their chemical structures are shown in Scheme S1 in the Supporting Information (SI)). Subsequently, we studied their binding with 54 types of dipeptides. An orthogonal investigation (Scheme S2 in the SI) indicated that the association constants (Ka) between dipeptides and 22085


Research Article

ACS Applied Materials & Interfaces

Figure 2. Characterization data of poly-PD@SiO2. (A) Thermogravimetric analysis (TGA) curve of amino-modified silica gel (black) or poly-PDmodified silica gel (red). (B) Pore size distribution of the silica gel before (black) and after (red) poly-PD modification, obtained from the Brunauer−Emmett−Teller (BET) adsorption isotherm curves. These data indicated that the polymer film thickness was ∼10 nm inside the silica pores. (C) pH-dependent zeta potential change of poly-PD@SiO2 at 20 °C. (D) Carbon (upper part) and nitrogen (lower part) element spectra of poly-PD thin film grafted on the silicon substrate, obtained from X-ray photoelectron spectroscopy (XPS) measurement. (E, F) Atomic force microscopy image (panel (E)) of poly-PD grafted silicon substrate and the corresponding section profile (panel (F)) along the green line shown in panel (E). (This indicated that the polymer film thickness was ∼10 nm.) (G) Water droplet profile on the poly-PD grafted silicon substrate; the static contact angle was 45° ± 2°, which indicated the hydrophilicity of the polymer surface.

This brought significant difficulty in finding the optimal sequence from more than 400 types of combinational sequences of dipeptides. To analyze these data, we developed a statistical algorithm (Scheme S3 in the SI) using the largest to the smallest ratio (R = Ka(max)/Ka(min)) to describe the variation range for each set of seven Ka values for a given dipeptide, as well as a discretization index (D) to characterize their dispersity (a smaller D represents a better dispersity, see Table S2 in the SI). Apparently, smaller D and larger R values are favorable for the discrimination of different saccharides. The hydropathy of amino acid side chains is an important factor that affects the interactions between various amino acid residues and other behaviors including H-bond formation with guests.50 Hydropathy index (HI) represents a quantitative description of this characteristic and has been used extensively to predict the three-dimensional (3-D) structures of proteins and their allocations in cell membranes.51 In this study, we attempted to utilize HI as a quantitative indicator to build a correlation between dipeptide sequences and their saccharide discrimination abilities (D and R values), in which dipeptide HIs were defined as the summation of HI values for the two amino acid residues based on the Eisenberg consensus scale.51 Figure 1D

monosaccharides were generally good; the Ka values ranged from hundreds to tens of thousands of liters per mole (L mol−1), which was consistent with other supramolecular systems involving carbohydrate binding35 (the calculation method is described in the SI and the detailed Ka data are shown in Table S1 in the SI; dimethyl sulfoxide was used as a solvent in order to guarantee the solubility of both dipeptide and monosaccharide). Moreover, the selectivity for different saccharides was distinct, which was strongly dependent on the dipeptide sequences (Figure 1B). For example, Ala-Pro had the highest affinity to GalNAc but the lowest affinity to galactose with a selectivity of 3.5:1, whereas those for Asp-Asn were Neu5Ac and GalNAc (selectivity = 13:1), respectively. Optimized models employing density functional theory (DFT) (Figure 1C, Gaussian, B3LYP, at 6-31G level, using water as a solvent) and hydrogen nuclear magnetic resonance (1H NMR) spectroscopy (Figures S2−S10 in the SI) verified that multiple H-bonds played crucial roles in the complexation between dipeptides and monosaccharides. The orthogonal investigation presented a comprehensive profile for dipeptide/saccharide interactions, which, however, produced a large amount of disordered Ka data (Figure 1B). 22086


Research Article

ACS Applied Materials & Interfaces shows the hydropathy plot of D values for 54 typical dipeptides chosen according to three types of amino acid combinations, namely, hydrophilic/hydrophilic, hydrophilic/hydrophobic, and hydrophobic/hydrophobic; the order of the residues was also considered. We found that dipeptides with HI values between −0.2 and −1.2 had the lowest D values, and all possessed a highly hydrophilic amino acid residue with strong H-bonding ability, e.g., Asp (D), Glu (E), or Asn (N), and a residue with some hydrophobicity, e.g., Tyr (Y), His (H), or Pro (P). Among these combinations, four sequences (i.e., DY, YD, PD, and PE) simultaneously had large R values (Figure 1E), which indicated great potential of these sequences in saccharide separation. Interestingly, we noticed that approximately onethird of the studied dipeptides showed larger association constants with Neu5Ac than with the other six monosaccharide substrates, in which the values for YD, DY, and PD were higher than those for most of the other dipeptides (Figure 1F). 2.2. Saccharide Separation and Binding Mechanism Study. Generally, saccharide separation is challenging, especially for complex oligosaccharides with different degrees of polymerization and glycosidic linkages that contain abundant biological information, although there have been several successful materials system for saccharide separation.52,53 Compared with monolayers, functional polymers54−56 can provide larger surface area and sufficient interaction sites, which have been extensively used in chromatographic materials.57 We thus synthesized homopolymers from propylene-acrylated monomers (Scheme S4 and Figure S12 in the SI) of the optimized dipeptides, which were coated on porous silica surfaces, acting as the stationary phase. Figure 1G and Figure 2 display the chemical structure of poly-PD modified silica gel (poly-PD@SiO2) and its characterization data. The corresponding data of poly-PE@SiO2 or poly-YD@SiO2 are shown in Scheme S5 and Figures S13 and S14 in the SI. We first evaluated their chromatographic separation properties for saccharides, using a column of PE-based polymer (poly-PE) as an example. Poly-PE column exhibited much longer retention time to Neu5Ac than the other neutral monosaccharides. Further experimentation indicated that poly-PE column could well distinguish Neu5Ac and other SAs with different chemical compositions (see Figure 3A, as well as Scheme S6 in the SI). Notably, the high-efficiency separation of mixtures containing three pairs of linkage isomers of oligosaccharides (Figure 3B, as well as Scheme S7 in the SI) indicated that our polymer could even distinguish between α(2−3) and α(2−6) linkages for SAs and between α(1−3) and α(1−4) linkages for galactose. This result is noteworthy because these are important glycosidic linkages for SAs and galactose in glycoconjugates, which play substantially different yet important roles in life and disease processes58,59 and are difficult to recognize chemically. Excellent separation was also observed for highly complex oligosaccharides. As shown in Figures 3C and 3D, fructo-oligosaccharides and galactooligosaccharides (chemical structures are shown in Scheme S8 in the SI) were separated into more than 20 constituents (tetra- to pentacosa-saccharides and higher) and 7 constituents (di- to octa-saccharides), respectively, with different degrees of polymerization. Polymers based on other optimized sequences, e.g., poly-PD (lower parts of each panel in Figure 3) and polyYD (Figure S15 in the SI), also showed very good saccharide separation properties. We also used other dipeptide-based polymers (e.g., poly-DD, poly-EE, poly-PP, poly-LL, and polyFF) to perform the experiments, but found that their separation

Figure 3. Chromatographic separation of saccharides using stationary phases based on poly-PE (upper part for each panel) and poly-PD (lower parts). (A) Combined chromatograms of different SAs under the same chromatographic conditions: Neu5Ac (SA1), Neu5Ac methyl ester (SA2), 2,4,7,8,9-penta-O-acetyl-N-acetyl-neuraminic acid methyl ester (SA3), N-glycolyl-neuraminic acid (SA4), N-acetyl-2,3didehydro-2-deoxyneuraminic acid (SA5). (B) Separation of three pairs of linkage isomer mixtures for SA-based oligo-saccharides: Neu5Ac-α(2−3)Gal-β(1−4)Glc (SA6), Neu5Ac-α(2−6)Gal-β(1− 4)Glc (SA7), Neu5Ac-α(2−3)Gal-β-4-methoxyphenyl glycoside (SA8), Neu5Ac-α(2−6)Gal-β-4-methoxyphenyl glycoside (SA9), Gal-β(1−3)[Neu5Ac-α(2−6)]GalNac-β-p-nitrophenyl (SA10), and Gal-β(1−4)[Neu5Ac-α(2−6)]GalNac-β-p-nitrophenyl (SA11). The separations in panel (B) were all performed under the same conditions. (C, D) Fructo-oligosaccharides (panel (C)) and galactooligosaccharides (panel (D)) were separated into more than 20 and 7 constituents, respectively, with different degrees of polymerization (labeled by numbers). See Schemes S6−S8 in the SI for molecular structures, and Table S3 in the Supporting Information for chromatographic conditions.

performance was rather poor, indicating that the dipeptide sequence optimization is critically important. Furthermore, 1H NMR titration experiment was performed to investigate the binding behavior of PD with Neu5Ac-α(2−3)Gal-β(1−4)Glc (SA6) or Neu5Ac-α(2−6)Gal-β(1−4)Glc (SA7) in D2O at 20 °C. As shown in Figures 4A and 4B, almost all C−H protons in SA7 incurred remarkable changes when it interacted with PD. By comparison, the corresponding changes in chemical shifts were much different in the complexation of SA6 with PD (see Figures S6 and S7 in the SI), which might induce distinct chromatographic retention behaviors of the linkage isomers of SA6 and SA7 on the poly-PD column. In addition, quantum chemistry calculation (Figure 4C) revealed that PD could insert into the pocket region of SA7, this tightly binding mode was further stabilized by 9 sets of strong hydrogen bonds. Therefore, we postulated that the differential saccharide/ dipeptide binding played a more important role in the saccharide separation than that of the hydrophilic interaction. 2.3. Glycopeptide Enrichment Performance. Next, we selected poly-PD as an example to study the effect of our polymers in glycopeptide enrichment. Sialylated glycopeptides (SGs) play pivotal roles in cell−cell and cell−microenvironment interactions.60 Excessive expression of SGs has been extensively observed in the sera from several types of cancers.8,31,32 Fetuin (a sialylated glycoprotein) has been widely used as a model system to evaluate the efficiency of 22087


Research Article


with different interference levels of BSA. To the best of our knowledge, the anti-interference level for other glycopeptide enrichment materials is normally less than 100-fold (molar ratio) BSA.61 However, for poly-PD in our experiments, digests with 200-fold BSA interference level, 39 glycopeptide signals were identified, which contained 33 SG signals (Figure S16B in the SI). To our surprise, the identification number remained at the same level (33 glycopeptide signals containing 29 SG signals), even when the interference level was increased 1000fold (Figure 5A). Interestingly, we even identified 7 unknown glycopeptide signals (red stars in Figure 5A; their compositions were identified by tandem MS, as shown in Figure S17 in the SI) from the trace impurities in BSA. This has never been reported by other materials, because of insufficient enrichment capacities, which further demonstrated the power of our polymer in glycopeptide enrichment. Moreover, experiments using a standard glycopeptide (Scheme S9 in the SI) showed a recovery as high as 87.9% ± 1.9% (Table S7 in the SI), illustrating very good controllability and reversibility for glycopeptide adsorption/desorption processes on poly-PD. These results displayed an excellent performance for poly-PD in glycopeptide enrichment markedly superior to that of commercial materials (Figures S21 and S22 in the SI). Notably, poly-YD, poly-DY, and poly-PE also exhibited good performance in glycopeptide enrichment (Figures S17−S20 in the SI). However, control experiments using other dipeptide-based polymers (e.g. poly-DD, poly-EE, poly-PP, poly-LL, and polyFF) showed very poor enrichment performance, which further proved the rationality and importance of dipeptide sequence optimization. 2.4. Discrimination of Linkage Isomers of Glycopeptides. As aforementioned, poly-PE column showed highefficiency separation toward oligosaccharides with different polymerization degrees and glycosidic linkage isomers (Figure 3). We further studied its effect on glycopeptide discrimination, using glycopeptides enriched directly from fetuin digests by the same column as an example. We found that glycopeptides with identical peptide chain but different glycans exhibited very different retention times, which were strongly influenced by the number of SA units (Figures 5B−F). Peptide chains also have an obvious influence on the retention time of glycopeptides (Figures 5B and 5G). More interestingly, poly-PE could even discriminate glycosidic linkages of glycopeptides, and as shown in Figures 5B−G, each glycopeptide was further separated into two or more isomer peaks. This brought substantially more and accurate glycan information that conventional methods are difficult to provide.15,16 It is widely acknowledged that slight differences between linkage isomers of glycan may induce distinct biofunctions of glycopeptides, even cellular behaviors.31,32,61,62 Accurate discrimination and separation of these linkage isomers of glycopeptides by our material not only can help biologists collect sufficient amounts of samples with high purities for analyzing their elaborate glycan structures, but also remarkably decrease the complexity of glycopeptide mixture, which may facilitate the discovery of more valuable glycosylation sites.14,15 2.5. Glycopeptide Enrichment in Real Biosamples. Compared with the model protein samples mentioned above, glycopeptide enrichment in real biosamples is substantially more challenging, because of higher complexity and the wide dynamic range of protein concentrations. Therefore, we used HeLa cell line as an example to assess the performance of polyPD in real biosamples. The enrichment selectivity (ES) in real

Figure 4. 1H NMR titration experiment to investigate binding mode of Pro-Asp (PD) with Neu5Ac-α(2−6)Gal-β(1−4)Glc (SA7). (A, B) Partial 1H NMR spectra of PD (spectrum a), SA7 (spectrum b), and its mixtures with different molar ratios (spectrum (c), 1:0.5; spectrum (d), 1:1; spectrum (e), 1:2; spectrum (f), 1:3) of PD in D2O at 20 °C. (Concentration of SA7 = 1 × 10−3 mol L−1; the attribution of each C− H proton is displayed in Scheme S7 in the SI). The hydroxyl groups in saccharides might participate in the complexation with PD; however, these signals could not be observed, because these active hydroxyl protons have been exchanged by D2O. (C) Optimized interaction model of acrylated PD with SA7, obtained from quantum chemistry calculation (Gaussian, density functional theory (DFT), B3LYP, at the 3-21G level). Hydrogen bonds with different lengths are indicated by green dashed lines.

glycopeptide enrichment. We demonstrated the high-performance enrichment of SGs on poly-PD under the optimized conditions (Table S4 in the SI) using tryptic digests of fetuin 22088


Research Article


Figure 5. Enrichment and separation of glycopeptides in model samples. (A) Typical mass spectrum of tryptic digests of fetuin with 1000-fold BSA interference after enrichment by poly-PD. It produced 33 glycopeptide signals (29 were SGs). Corresponding glycan structures are shown here. Other detailed information is presented in Tables S5 and S6 in the SI. Red stars show the glycopeptide signals (all are SGs) from the impurities of BSA. Their glycan compositions (Figure S17 in the SI) were confirmed by high-accuracy tandem MS. (B−G) Typical extracted ion chromatograms showing the separation of glycopeptides using a poly-PE based column. Glycopeptides in panels (B), (C), and (F) had the same peptide chain (P08) but different glycans with two, two, and three terminal SAs, respectively. Those in panels (D) and (E) had the same peptide chain of P06 but different glycans with three and four terminal SAs, respectively. Glycopeptide in panel (G) had the same glycan as that in panel (B), but a different peptide chain (P10). The corresponding serial numbers of MS signals are labeled in the square brackets. Interestingly, each glycopeptide in panels (B)−(G) split into two or more peaks, which corresponded to different linkage isomers of glycans: solid blue square denotes GlcNAc, solid yellow square denotes GalNAc, solid green circle denotes mannose, solid yellow circle denotes galactose, and solid red diamond denotes sialic acid. The peptide sequences are denoted as follows: P06, RPTGEVYDIEIDTLETTCHVLDPTPLAN(99)CSVR; P08, KLCPDCPLLAPLN(156)DSR; and P10, LCPDCPLLAPLN(156)DSR.

biosamples is defined as the ratio of identified glycopeptides to total number of peptides detected by MS. For LAC and other nondestructive chromatographic methods, the ES is usually unsatisfactory.62,63 Interestingly, we used a simple dispersive solid-phase extraction protocol (Scheme S10 in the SI) and achieved an average ES of ∼70% for the whole-cell analysis of HeLa cell line samples (Figure 6). This was already comparable to or even higher than those obtained via hydrazide chemistry methods64 or by other combined enrichment procedures.15,63,65 For three replicates of the experiment, the sample amounts were ∼100 μg of total protein each. After enrichment, three replicates produced 528, 477, and 504 identified glycopeptides, respectively (as determined using the MaxQuant software), with a false discovery rate of 1% at both the peptide and site levels. Through comparison with the glycosylation site information collected in the UniProt knowledge base and relevant references,66,67 we found that 40% of the identified glycosylation sites were new sites. Furthermore, a dataset analysis (Figure S24 in the SI) showed substantial overlap between the identified glycopeptides among these replicates: ∼80% overlap between any two single experiments; each of these covered ∼71% of the combined identified glycosites, on

Figure 6. Glycopeptide enrichment from HeLa cell lysate with polyPD. Experiment was conducted in triplicate. Sample amount for each replicate: ∼100 μg of total protein, 4 mg poly-PD. Black and red columns show the numbers of glycopeptides and glycosylation sites identified via mass spectrometry, respectively; blue columns show the corresponding enrichment selectivity (ES).

average, for three replicates. Considering HeLa S3 cell is a mature and well-studied cell sample, the high ratio of new glycosylation sites, satisfactory ES value, and good coverage for 22089


ACS Applied Materials & Interfaces glycopeptides indicated the good potential of our material applied in large-scale and in-depth glycoproteomics analysis, which usually involves in more-complex tissue samples (e.g., mouse liver or human liver)66,67 and the combination of multiple enrichment methods.

ACKNOWLEDGMENTS

■

REFERENCES

(1) Hart, G. W.; Copeland, R. J. Glycomics Hits the Big Time. Cell 2010, 143, 672−676. (2) Doerr, A. Glycoproteomics. Nat. Methods 2011, 9, 36. (3) Alper, J. Glycobiology−Turning Sweet on Cancer. Science 2003, 301, 159−160. (4) Dube, D. H.; Bertozzi, C. R. Glycans in Cancer and Inflammation−Potential for Therapeutics and Diagnostics. Nat. Rev. Drug Discovery 2005, 4, 477−488. (5) Szymanski, C. M.; Wren, B. W. Protein Glycosylation in Bacterial Mucosal Pathogens. Nat. Rev. Microbiol. 2005, 3, 225−237. (6) Rudd, P. M.; Elliott, T.; Cresswell, P.; Wilson, I. A.; Dwek, R. A. Glycosylation and the Immune System. Science 2001, 291, 2370−2376. (7) Reis, C. A.; Osorio, H.; Silva, L.; Gomes, C.; David, L. Alterations in Glycosylation as Biomarkers for Cancer Detection. J. Clin. Pathol. 2010, 63, 322−329. (8) Pinho, S. S.; Reis, C. A. Glycosylation in Cancer: Mechanisms and Clinical Implications. Nat. Rev. Cancer 2015, 15, 540−555. (9) Palmisano, G.; Lendal, S. E.; Engholm-Keller, K.; Leth-Larsen, R.; Parker, B. L.; Larsen, M. R. Selective Enrichment of Sialic AcidContaining Glycopeptides Using Titanium Dioxide Chromatography with Analysis by HILIC and Mass Spectrometry. Nat. Protoc. 2010, 5, 1974−1982. (10) Desaire, H. Glycopeptide Analysis, Recent Developments and Applications. Mol. Cell. Proteomics 2013, 12, 893−901. (11) Kolli, V.; Schumacher, K. N.; Dodds, E. D. Engaging Challenges in Glycoproteomics: Recent Advances in MS-Based Glycopeptide Analysis. Bioanalysis 2015, 7, 113−131. (12) Mariño, K.; Bones, J.; Kattla, J. J.; Rudd, P. M. A Systematic Approach to Protein Glycosylation Analysis: A Path through the Maze. Nat. Chem. Biol. 2010, 6, 713−723. (13) Wuhrer, M.; Catalina, M. I.; Deelder, A. M.; Hokke, C. H. Glycoproteomics Based on Tandem Mass Spectrometry of Glycopeptides. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2007, 849, 115− 128. (14) Garbis, S.; Lubec, G.; Fountoulakis, M. Limitations of Current Proteomics Technologies. J. Chromatogr. A 2005, 1077, 1−18. (15) Huang, J. F.; Wang, F. J.; Ye, M. L.; Zou, H. F. Enrichment and Separation Techniques for Large-Scale Proteomics Analysis of the Protein Post-Translational Modifications. J. Chromatogr. A 2014, 1372, 1−17. (16) Ongay, S.; Boichenko, A.; Govorukhina, N.; Bischoff, R. Glycopeptide Enrichment and Separation for Protein Glycosylation Analysis. J. Sep. Sci. 2012, 35, 2341−2372. (17) Chen, C. C.; Su, W. C.; Huang, B. Y.; Chen, Y. J.; Tai, H. C.; Obena, R. P. Interaction Modes and Approaches to Glycopeptide and Glycoprotein Enrichment. Analyst 2014, 139, 688−704. (18) Yu, L.; Li, X. L.; Guo, Z. M.; Zhang, X. L.; Liang, X. M. Hydrophilic Interaction Chromatography Based Enrichment of Glycopeptides by Using Click Maltose: A Matrix with High Selectivity and Glycosylation Heterogeneity Coverage. Chem.Eur. J. 2009, 15, 12618−12626. (19) Pohleven, J.; Štrukelj, B.; Kos, J. Affinity Chromatography, Vol. 3; Magdeldin, S., Ed.; InTech: Rijeka, Czech Republic, 2012; 49 pp.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b07863. Experimental details for the synthesis of diverse dipeptide monomers, polymerization procedures, fluorescent titration experiment for calculating association constants, 1H NMR titration for investigating the binding modes, preparation of dipeptide-based polymeric chromatographic columns, trypsin digestion of proteins, glycopeptide enrichment, determination of the recovery rate for glycopeptide, separation of glycopeptides with different linkage isomers and glycan structures, enrichment of glycopeptides from HeLa cell lysate and human serum (PDF)

■

■

This work was supported by the Major State Basic Research Development Program of China (973 Program) (No. 2013CB933002), the National High Technology Research and Development Program of China (No. 2012AA020203), China National Funds for Distinguished Young Scientists (No. 51325302), the National Natural Science Foundation of China (Nos. 51533007, 51521001, 21135005, 21275114, 51473131, 21475129). G.Y.Q. acknowledges Hubei Provincial Department of Education for financial assistance through the “Chutian Scholar” Program and Hubei Provincial Natural Science Foundation of China No. (2014CFA039).

3. CONCLUSIONS In conclusion, the above results reveal two attracting features of our materials: (i) high-efficiency enrichment property toward glycopeptides and (ii) the power to discriminate their elaborate glycan structures simultaneously. This is fundamentally different from conventional studies of glycopeptide enrichment materials,62,68 which mainly focus on only the former property. However, the latter property helps to acquire the glycan information precisely, and the combination of these two features in one material makes it possible to build a direct association between glycosylation sites and the glycans attached to them, which cannot be obtained easily by the separate analysis processes but is especially important for in-depth study of glycosylation and its biofunctions. Therefore, our materials may represent a novel material system that can really bridge the gap between proteomics and glycomics, which is a goal that glycoproteomics always pursues.29 Furthermore, the principle for carbohydrate-binding receptor design and the hydropathybased searching strategy for dipeptide optimization proposed in this study indicate a clear direction to develop other oligopeptide-based materials with higher performances. Besides, sialylated glycopeptides (SGs) are highly expressed on tumor cell membranes.59 As a benefit from accurate discrimination capacities of our materials toward SGs, the selectivity of materials/biodevices for tumor cells may obtain remarkable improvement when our polymers are converged, which may facilitate many fantastic applications in tumor cell detection, capture, and separation.69,70

■

Research Article

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (X. Liang). *E-mail: [email protected] (T. Sun). Author Contributions §

These authors contributed equally to this work.

Notes

The authors declare no competing financial interest. 22090


Research Article

ACS Applied Materials & Interfaces (20) Gabius, H. J.; Siebert, H. C.; André, S.; Jiménez-Barbero, J.; Rüdiger, H. Chemical Biology of the Sugar Code. ChemBioChem 2004, 5, 740−764. (21) Li, H. Y.; Liu, Z. Recent Advances in Monolithic Column-Based Boronate-Affinity Chromatography. TrAC, Trends Anal. Chem. 2012, 37, 148−161. (22) Wang, J. X.; Wang, Y. N.; Gao, M. X.; Zhang, X. M.; Yang, P. Y. Multilayer Hydrophilic Poly(Phenol-Formaldehyde Resin)-Coated Magnetic Graphene for Boronic Acid Immobilization as a Novel Matrix for Glycoproteome Analysis. ACS Appl. Mater. Interfaces 2015, 7, 16011−16017. (23) Zhang, H.; Li, X. J.; Martin, D. B.; Aebersold, R. Identification and Quantification of N-Linked Glycoproteins Using Hydrazide Chemistry, Stable Isotope Labeling and Mass Spectrometry. Nat. Biotechnol. 2003, 21, 660−666. (24) Cao, Q. C.; Ma, C.; Bai, H. H.; Li, X. Y.; Yan, H.; Zhao, Y.; Ying, W. T.; Qian, X. H. Multivalent Hydrazide-Functionalized Magnetic Nanoparticles for Glycopeptide Enrichment and Identification. Analyst 2014, 139, 603−609. (25) Bai, H. H.; Fan, C.; Zhang, W. J.; Pan, Y. T.; Ma, L.; Ying, W. T.; Wang, J. H.; Deng, Y. L.; Qian, X. H.; Qin, W. J. A pH-Responsive Soluble Polymer-Based Homogeneous System for Fast and Highly Efficient N-Glycoprotein/Glycopeptide Enrichment and Identification by Mass Spectrometry. Chem. Sci. 2015, 6, 4234−4241. (26) Larsen, M. R.; Jensen, S. S.; Jakobsen, L. A.; Heegaard, N. H. H. Exploring the Sialiome Using Titanium Dioxide Chromatography and Mass Spectrometry. Mol. Cell. Proteomics 2007, 6, 1778−1787. (27) Palmisano, G.; Parker, B. L.; Engholm-Keller, K.; Lendal, S. E.; Kulej, K.; Schulz, M.; Schwämmle, V.; Graham, M. E.; Saxtorph, H.; Cordwell, S. J.; Larsen, M. R. A Novel Method for the Simultaneous Enrichment, Identification, and Quantification of Phosphopeptides and Sialylated Glycopeptides Applied to a Temporal Profile of Mouse Brain Development. Mol. Cell. Proteomics 2012, 11, 1191−1202. (28) Zheng, J. N.; Xiao, Y.; Wang, L.; Lin, Z.; Yang, H. H.; Zhang, L.; Chen, G. N. Click Synthesis of Glucose-Functionalized Hydrophilic Magnetic Mesoporous Nanoparticles for Highly Selective Enrichment of Glycopeptides and Glycans. J. Chromatogr. A 2014, 1358, 29−38. (29) Cummings, R. D.; Pierce, J. M. The Challenge and Promise of Glycomics. Chem. Biol. 2014, 21, 1−15. (30) Nonaka, M.; Fukuda, M. N.; Gao, C.; Li, Z.; Zhang, H. T.; Greene, M. I.; Peehl, D. M.; Feizi, T.; Fukuda, M. Determination of Carbohydrate Structure Recognized by Prostate-Specific F77 Monoclonal Antibody through Expression Analysis of Glycosyltransferase Genes. J. Biol. Chem. 2014, 289, 16478−16486. (31) Saldova, R.; Fan, Y.; Fitzpatrick, J. M.; Watson, R. W. G.; Rudd, P. M. Core Fucosylation and α2−3 Sialylation in Serum N-Glycome Is Significantly Increased in Prostate Cancer Comparing to Benign Prostate Hyperplasia. Glycobiology 2011, 21, 195−205. (32) Alley, W. R.; Novotny, M. V. Glycomic Analysis of Sialic Acid Linkages in Glycans Derived From Blood Serum Glycoproteins. J. Proteome Res. 2010, 9, 3062−3072. (33) Wollscheid, B.; Bausch-Fluck, D.; Henderson, C.; O’Brien, R.; Bibel, M.; Schiess, R.; Aebersold, R.; Watts, J. D. Mass-Spectrometric Identification and Relative Quantification of N-Linked Cell Surface Glycoproteins. Nat. Biotechnol. 2009, 27, 378−386. (34) Ferrand, Y.; Crump, M. P.; Davis, A. P. A Synthetic Lectin Analog for Biomimetic Disaccharide Recognition. Science 2007, 318, 619−622. (35) Mazik, M. Molecular Recognition of Carbohydrates by Acyclic Receptors Employing Noncovalent Interactions. Chem. Soc. Rev. 2009, 38, 935−956. (36) Sun, X. L.; James, T. D. Glucose Sensing in Supramolecular Chemistry. Chem. Rev. 2015, 115, 8001−8037. (37) Rios, P.; Carter, T. S.; Mooibroek, T. J.; Crump, M. P.; Lisbjerg, M.; Pittelkow, M.; Supekar, N. T.; Boons, G. J.; Davis, A. P. Synthetic Receptors for the High-Affinity Recognition of O-GlcNAc Derivatives. Angew. Chem., Int. Ed. 2016, 55, 3387−3392. (38) Garces, F.; Sok, D.; Kong, L.; McBride, R.; Kim, H. J.; SayeFrancisco, K. F.; Julien, J. P.; Hua, Y. Z.; Cupo, A.; Moore, J. P.;

Paulson, J. C.; Ward, A. B.; Burton, D. R.; Wilson, I. A. Structural Evolution of Glycan Recognition by A Family of Potent HIV Antibodies. Cell 2014, 159, 69−79. (39) Weis, W. I.; Drickamer, K. Structural Basis of Lectin− Carbohydrate Recognition. Annu. Rev. Biochem. 1996, 65, 441−473. (40) Fadda, E.; Woods, R. J. Molecular Simulations of Carbohydrates and Protein−Carbohydrate Interactions: Motivation, Issues and Prospects. Drug Discovery Today 2010, 15, 596−609. (41) Antonik, P. M.; Volkov, A. N.; Broder, U. N.; Lo Re, D.; van Nuland, N. A. J.; Crowley, P. B. Anomer-Specific Recognition and Dynamics in a Fucose-Binding Lectin. Biochemistry 2016, 55, 1195− 1203. (42) Fujimoto, Y. K.; Green, D. F. Carbohydrate Recognition by the Antiviral Lectin Cyanovirin-N. J. Am. Chem. Soc. 2012, 134, 19639− 19651. (43) Gale, P. A.; Caltagirone, C. Anion Sensing by Small Molecules and Molecular Ensembles. Chem. Soc. Rev. 2015, 44, 4212−4227. (44) Edwards, S. J.; Valkenier, H.; Busschaert, N.; Gale, P. A.; Davis, A. P. High-Affinity Anion Binding by Steroidal Squaramide Receptors. Angew. Chem., Int. Ed. 2015, 54, 4592−4596. (45) Nowick, J. S. Exploring β-Sheet Structure and Interactions with Chemical Model Systems. Acc. Chem. Res. 2008, 41, 1319−1330. (46) Qing, G. Y.; Sun, T. L. Chirality-Triggered Wettability Switching on a Smart Polymer Surface. Adv. Mater. 2011, 23, 1615−1620. (47) Cocinero, E. J.; Carcabal, P.; Vaden, T. D.; Davis, B. G.; Simons, J. P. Exploring Carbohydrate−Peptide Interactions in the Gas Phase: Structure and Selectivity in Complexes of Pyranosides with NAcetylphenylalanine Methylamide. J. Am. Chem. Soc. 2011, 133, 4548− 4557. (48) Varki, A. Uniquely Human Evolution of Sialic Acid Genetics and Biology. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 8939−8946. (49) Varki, A. Glycan-Based Interactions Involving Vertebrate SialicAcid-Recognizing Proteins. Nature 2007, 446, 1023−1029. (50) Garrett, R. H.; Grisham, C. M. Biochemistry; Cengage Learning Press: Boston, 2008; 251 pp. (51) Eisenberg, D. Three-Dimensional Structure of Membrane and Surface Proteins. Annu. Rev. Biochem. 1984, 53, 595−623. (52) Ikegami, T.; Tomomatsu, K.; Takubo, H.; Horie, K.; Tanaka, N. Separation Efficiencies in Hydrophilic Interaction Chromatography. J. Chromatogr. A 2008, 1184, 474−503. (53) Ruhaak, L. R.; Deelder, A. M.; Wuhrer, M. Oligosaccharide Analysis by Graphitized Carbon Liquid Chromatography−Mass Spectrometry. Anal. Bioanal. Chem. 2009, 394, 163−174. (54) Ma, M. M.; Guo, L.; Anderson, D. G.; Langer, R. Bio-Inspired Polymer Composite Actuator and Generator Driven by Water Gradients. Science 2013, 339, 186−189. (55) Hou, X.; Yang, F.; Li, L.; Song, Y. L.; Jiang, L.; Zhu, D. B. A Biomimetic Asymmetric Responsive Single Nanochannel. J. Am. Chem. Soc. 2010, 132, 11736−11742. (56) Mano, J. F. Stimuli-Responsive Polymeric Systems for Biomedical Applications. Adv. Eng. Mater. 2008, 10, 515−527. (57) Guiochon, G. Monolithic Columns in High-Performance Liquid Chromatography. J. Chromatogr. A 2007, 1168, 101−168. (58) Weinstein, J.; de Souza-e-Silva, U.; Paulson, J. C. Sialylation of Glycoprotein Oligosaccharides N-linked to Asparagine. Enzymatic Characterization of a Galβ1→3(4)GlcNAc α2→3 Sialyltransferase and A Galβ1→4GlcNAc α2→6 Sialyltransferase from Rat Liver. J. Biol. Chem. 1982, 257, 13845−13853. (59) Schauer, R. Achievements and Challenges of Sialic Acid Research. Glycoconjugate J. 2000, 17, 485−499. (60) Stowell, S. R.; Ju, T. Z.; Cummings, R. D. Protein Glycosylation in Cancer. Annu. Rev. Pathol.: Mech. Dis. 2015, 10, 473−510. (61) Jiang, B.; Liang, Y.; Wu, Q.; Jiang, H.; Yang, K. G.; Zhang, L. H.; Liang, Z.; Peng, X. J.; Zhang, Y. K. New GO-PEI-Au-L-Cys ZICHILIC Composites: Synthesis and Selective Enrichment of Glycopeptides. Nanoscale 2014, 6, 5616−5619. (62) Alley, W. R., Jr.; Mann, B. F.; Novotny, M. V. High-Sensitivity Analytical Approaches for the Structural Characterization of Glycoproteins. Chem. Rev. 2013, 113, 2668−2732. 22091


Research Article

ACS Applied Materials & Interfaces (63) Zielinska, D. F.; Gnad, F.; Wiśniewski, J. R.; Mann, M. Precision Mapping of an In Vivo N-Glycoproteome Reveals Rigid Topological and Sequence Constraints. Cell 2010, 141, 897−907. (64) Liu, L.; Yu, M.; Zhang, Y.; Wang, C.; Lu, H. Hydrazide Functionalized Core−Shell Magnetic Nanocomposites for Highly Specific Enrichment of N-Glycopeptides. ACS Appl. Mater. Interfaces 2014, 6, 7823−7832. (65) Sugahara, D.; Kaji, H.; Sugihara, K.; Asano, M.; Narimatsu, H. Large-Scale Identification of Target Proteins of A Glycosyltransferase Isozyme by Lectin-IGOT-LC/MS, An LC/MS-Based Glycoproteomic Approach. Sci. Rep. 2012, 2, 680. (66) Zhu, J.; Sun, Z.; Cheng, K.; Chen, R.; Ye, M. L.; Xu, B.; Sun, D. G.; Wang, L. M.; Liu, J.; Wang, F. J.; Zou, H. F. Comprehensive Mapping of Protein N-Glycosylation in Human Liver by Combining Hydrophilic Interaction Chromatography and Hydrazide Chemistry. J. Proteome Res. 2014, 13, 1713−1721. (67) Medzihradszky, K. F.; Kaasik, K.; Chalkley, R. J. Tissue-Specific Glycosylation at the Glycopeptide Level. Mol. Cell. Proteomics 2015, 14, 2103−2110. (68) Zhang, Y.; Zhang, C.; Jiang, H. C.; Yang, P. Y.; Lu, H. J. Fishing the PTM Proteome with Chemical Approaches Using Functional Solid Phases. Chem. Soc. Rev. 2015, 44, 8260−8287. (69) Li, Y. Y.; Lu, Q. H.; Liu, H. L.; Wang, J. F.; Zhang, P. C.; Liang, H. G.; Jiang, L.; Wang, S. T. Antibody-Modified Reduced Graphene Oxide Films with Extreme Sensitivity to Circulating Tumor Cells. Adv. Mater. 2015, 27, 6848−6854. (70) Li, Q. K.; Gabrielson, E.; Askin, F.; Chan, D. W.; Zhang, H. Glycoproteomics Using Fluid-Based Specimens in the Discovery of Lung Cancer Protein Biomarkers: Promise and Challenge. Proteomics: Clin. Appl. 2013, 7, 55−69.

22092


Dipeptide-Based Carbohydrate Receptors and Polymers for

Recommend Documents