Chemically bonded maltose via click chemistry as stationary phase for HILIC

Abstract

A HILIC stationary phase was synthesized through bonding maltose on silica particles via click chemistry. A set of representative polar compounds including sugars, amino acids and small peptides were employed to evaluate chromatographic properties of Click Maltose. The results illustrated the surface adsorption mechanism from the direct hydrogen bonding interaction between sugars and Click Maltose played an important role on retention behavior of sugars under HILIC mode. The chromatographic behavior of anions on Click Maltose was consistent with HILIC/WAX mechanism, while the retention of cations was depended on electrostatic repulsion-hydrophilic interaction chromatography (ERLIC) mechanism. The Click Maltose stationary phase was successfully used in the separation of standard samples of polar compounds and complex samples of oligosaccharides from the natural product. The Click Maltose also exhibited good reproducibility and stability, which were helpful to further application. Our study confirmed the advantages and application potential of bonded oligosaccharides for HILIC separation.

---

Introduction

Carbohydrates are the most extensive molecules in the biological world, which have been considered as energy sources and structural materials.¹ The biological functions of carbohydrates have been gaining more attention, and the realizations of such functions are based on the interaction between carbohydrates and other bio-molecules.² In order to study those biological functions, a critical strategy is immobilization of carbohydrates to solid surfaces.³ The reactions of Diels-Alder and thiol addition were employed to graft sugars to self-assembled monolayers (SAMs),⁴ but one of the drawbacks of these reactions was low efficiency, especially on solid surface. It was necessary to develop an efficient and chemoselective method for immobilization of sugars.

A “click chemistry” strategy, termed by Sharpless,⁵ has several characteristics of high yields, operating in a variety of conditions and producing only inoffensive byproducts. Among several types of reactions, the reaction of Cu(I)-catalyzed 1,3-dipolar cycloadditon of terminal azide to terminal alkyne is considered as a typical click method, which can be operated in aqueous environment at room temperature.⁶ The advantages of this 1,3-dipolar cycloaddition used for sugar immobilization are high reaction efficiency and high stereoselectivity. Therefore, immobilization of carbohydrates on the surface can be achieved through this well-developed “click chemistry” strategy. Wong et al. exploited this method to immobilize oligosaccharides to the microtiter well surface for profiling the special interaction between antibodies and receptors.^3,7 This reaction was also employed to prepare carbohydrate SAMs to understand the carbohydrate-protein interactions, reported by Wang et al.^2b

Carbohydrates have been widely used as separation media in chromatography, both as support materials and bonded stationary phases, for the separation of complex samples. Polysaccharides such as dextran that are functionalized with exchange groups have been grafted on agarose gels to prepare ion exchanger for protein chromatography.⁸ Polysaccharides (such as cellulose and amylose) have been immobilized on silica surface used as stationary phases for chiral separations.⁹ In the process for immobilization of polysaccharides, a number of reaction sites of polysaccharides have reacted with solid surfaces' active groups. Obviously, those polysaccharides grafted on solid surfaces are highly cross-linked, which is not favorable to keep their spatial conformation. Due to the special configuration, cyclodextrins (CDs) as cyclic oligosaccharides are commonly used as chiral stationary phases (CSPs).¹⁰ Since Armstrong first demonstrated the chemically bonded β-CD stationary phases,¹¹ with advances in immobilization technologies, CD-based stationary phases have been widely applied in HPLC.^1,12 The applications of non-cyclic oligosaccharides (including disaccharides) used as chromatographic stationary phases have been reported in a few literatures. Three examples have illustrated the use of silica materials containing covalently bonded non-cyclic oligosaccharides as affinity chromatographic supports for the isolation and purification of lectins.¹³ Compared with the rigid structure of CDs, the conformation of non-cyclic oligosaccharides is relative flexible. Non-cyclic oligosaccharides are also hydrophilic due to a number of hydroxyl groups contained in their structures. Because of their hydrophilic and flexible characteristics, non-cyclic oligosaccharides immobilized on silica surfaces are potential stationary phases for hydrophilic interaction liquid chromatography (HILIC).

HILIC was defined by Alpert in 1990¹⁴ and has been gaining intense interest in the last few years. It has been employed for many applications of separating peptides, pharmaceuticals, carbohydrates, etc.¹⁵ A variety of silica-based stationary phases have been developed for HILIC separations such as neat silica gel or bonded with amino-, diol-, cyano-, amide- and zwitterionic sulfobetaine ligands.^15b,16 However, few studies have been concerned about the immobilization of non-cyclic oligosaccharides to silica surface to prepare stationary phases for HILIC separation of polar compounds. Our previous work has reported on the synthesis of silica materials bonded maltose (Click Maltose) used as HILIC stationary phases through “click chemistry strategy”.¹⁷ The reaction of Cu(I)-catalyzed 1,3-dipolar cycloadditon of azide to alkyne has been employed to introduce maltose to silica surface and this reaction has been proved to be stereospecific selective with high yields. However, the chromatographic properties and the potential applications of Click Maltose have not been researched in detail.

The objective of this study is to illustrate the synthesis method based on “click chemistry” strategy and further investigate the chromatographic properties of Click Maltose. A set of typical polar compounds such as carbohydrates, amino acids and small peptides will be selected for the evaluations of chromatographic performance of Click Maltose. The applications of Click Maltose as stationary phase for the separation of polar compounds and complex samples in natural product will be researched. Finally, the reproducibility and stability of Click Maltose will be illustrated.

Results and discussion

Preparation and characterization of maltose stationary phase

According to “click chemistry” strategy, the silica support terminated azide group reacted with 1-propargyl-O-maltose to synthesize Click Maltose. FT-IR and solid state ¹³C CP/MAS NMR were employed to characterize Click Maltose. The peak at 2100 cm⁻¹ disappeared on the FT-IR spectra (not shown) after immobilization of maltose on azide-silica surface, which indicated that the azide group on silica was almost reacted.¹⁸ Solid state ¹³C CP/MAS NMR was used to obtain more information of the surface chemistry of Click Maltose (Fig. 1). Three groups of signals at 9.1, 27.2, and 49.9 ppm were three carbon atoms belonging to the three carbon atoms near the silica surface. The signals at 123.1 and 143.9 ppm were two carbon atoms in the structure of triazole ring respectively. The signal at 63.1 ppm was carbon atom linked to triazole ring and oxygen atom. The signal at 103.2 ppm was C1 atom on glucose unit of maltose. The peak at 73.7 ppm was responding to carbon atoms on maltose. The results demonstrated that maltose was successfully immobilized on silica surface and the designed structure of the stationary phase has been achieved.

Fig. 1 Solid state ¹³C CP/MAS NMR spectra of the Click-Maltose stationary phase.

The results of elemental analysis and the surface concentration of azide-silica and click maltose were shown in Table 1. The surface concentration of the azide group on the azide-silica was 3.35 µmol m⁻² based on the nitrogen content of azide-silica, according to the method reported by Kibbey and Meyerhoff.¹⁹ The surface concentration of maltose was 1.75 µmol m⁻², calculated from the increasing carbon content of silica surface.

Table 1 Elemental analysis of the azide-silica and Click Maltose

Stationary phases	C%	N%	Surface coverage [µmol m^−2]
Azide-silica	4.654	3.450	3.35
Click Maltose	11.87	2.542	1.75

---

Retention of sugars on Click Maltose stationary phase

Sugars containing a huge number of hydroxyl groups were commonly selected as typical polar compounds for evaluation of performance of stationary phase under HILIC mode. Seven sugars including mono-, di- and trisaccharides were used for studying their retention behavior on Click Maltose stationary phase. The percentage of water in mobile phase was changed from 20% to 40%, while keeping ammonium acetate (pH 4.7) constant at 10 mM. The natural logarithmic of retention factors (k) was plotted versus the water content in mobile phase, as shown in Fig. 2(a). The retention behaviors of seven sugars all exhibited typical HILIC characteristics and a decrease trend of retention factors was shown as the water content increased.²⁰

Fig. 2 (a) Plot of sugar retention factors versus water content in mobile phase and (b) The natural logarithmic plots of lnkversus water content in mobile phase for seven sugars.

According to the study of Alpert, the retention mechanism of HILIC was controlled by partitioning between mobile phase and a water-rich layer immobilized on polar stationary phase.¹⁴ The relationship established²¹ for partitioning mechanism was

lnk = a + cC_B (1)

where C_B was the content of stronger member of a binary mobile phase.

However, Fig. 2(a) showed that the relationship between lnk and C_B was not linear. The retention behaviors of sugar on Click Maltose were not consistent to a pure partitioning mechanism but there were some other interactions.

According to the previous investigations of retention behavior of HILIC,²² regression analysis was carried out based on eqn (2).

lnk = a + blnC_B (2)

The natural logarithmic plot of lnkversus water content in mobile phase was showed in Fig. 2(b). The linear relationship between lnk and lnC_B was observed and the regression coefficients of sugars were 0.9995∼0.9999 (Table 2). It was further demonstrated that the retention behaviors of carbohydrates on Click Maltose were not controlled by a pure partitioning mechanism but that there were some other interactions. The interactions may attribute to the direct hydrogen bonding between sugars and bonded maltose on the silica surface. The similar conclusion was drawn in the previous report on the mechanism of oligosaccharides separation.¹ The retention functions were beneficial to the method development in the separation of sugars. The results are also helpful to understand the retention mechanism of HILIC. Hydrogen bonding may play important roles in the separation process of HILIC, especially in the separation of hydroxyl-containing compounds and use of hydroxyl bonded stationary phases.

Table 2 Data for the study of the retention behavior of sugars on Click Maltose stationary phase

Sugars	Slope	Intercept	R ^2
Sorbose	−2.184	−2.350	0.9996
Glucose	−2.407	−2.499	0.9999
Turanose	−3.012	−2.873	0.9999
Cellobiose	−3.138	−2.911	0.9998
Trehalose	−3.273	−2.973	0.9999
Melezitose	−3.664	−3.241	0.9996
Raffinose	−3.885	−3.350	0.9995

---

Retention behaviors of ionic solutes on Click Maltose stationary phase

The representative polar compounds of amino acids and peptides were employed to investigate the retention behavior of ionic solutes on Click Maltose. The mobile phase contained 65% (v/v) ACN with various ammonium formate concentrations ranging from 5 to 35 mM. As buffer salt concentrations increased, the retention times of neutral solutes (FGGF, LGG and F) were almost unchanged, acidic compounds (EE and E) decreased and basic solutes (GGH, KG, R, K and H) increased, as shown in Fig. 3(a). EE and E were negatively charged and their retention times decreased as buffer salt concentrations increased. It indicated that they were affected by electrostatic attraction interactions and those interactions were suppressed as buffer salt concentrations increased.²³ On the other hand, basic compounds (GGH, KG, R, K and H) were positively charged and their retention times increased, which implied that they were affected by electrostatic repulsion interactions that were weakened as buffer salt concentrations increased.²⁴

Fig. 3 The effect of buffer concentration (a) and pH (b) on retention behaviors of small peptides and amino acids on Click Maltose.

According to other studies, there was a water-rich layer on polar solid phase surface and the retention behavior of solute was mainly controlled by partitioning mechanism.¹⁴ The thickness of the water-rich layer increased as the salt concentration increased because a greater number of salt ions were driven to the polar solid phase surface leading to stronger retention of solutes.²⁰ However, our results described different retention behavior of acidic solutes, which was similar to that of in ion-exchange chromatography. The differences in retention behavior of acidic compounds indicated that besides hydrophilic interaction, the ionic solutes were affected by electrostatic interactions (containing attraction and repulsion) which were from the positively charged stationary phase surface. Because maltose bonded on silica surface was a neutral molecule, the positive charges on the stationary phase surface were provided by the triazoles ring introduced in the process of “click chemistry” synthesis. The pK_a of the triazole is 9.3²⁵ and it could be protonated within the present pH range, leading to the Click Maltose stationary phase being positively charged. The results proved that besides hydrophilic interaction provided by the bonded maltose, the triazoles ring also contributed to retention behaviors of ionic solutes on Click Maltose just like the beneficial effect of the triazoles linkage in drug discovery due to its large dipole moment.²⁶

The effect of pH of buffer solution on retention was investigated to further demonstrate the effect of the triazoles ring on chromatographic behavior of ionic solutes. The pH values of ammonium formate changed from 4.4 to 3.0 while keeping ACN content in mobile phase constant at 65%. As pH values decreased, the retention times of neutral solutes (FGGF, LGG and F) were almost unchanged, while a decrease trend of retention times of other peptides and amino acids (including both acidic and basic) was observed, as shown in Fig. 3(b).

Glutamic acid (E) that was negatively charged was not eluted within 60 min at pH 4.4, which indicated that the electrostatic attraction interaction between protonated triazoles and anion was strong and the triazoles played an important role on retention behavior of anion solutes. When pH values decreased, changed from 3.7 to 3.0, close to isoelectric points of EE and E, the ionization of EE and E was suppressed, thus electrostatic attraction and hydrophilic interactions were weakened, leading to their rapid decrease of retention (Fig. 3(b)).

As for basic compounds (GGH, KG, R, K and H), their retention behaviors were mainly controlled by two interactions. One was electrostatic repulsion with protonated triazoles and the other was hydrophilic interaction. When pH values decreased, basic solutes were protonated and became more hydrophilic. It resulted in stronger hydrophilic interaction and positive contribution to their retention. On the other hand, electrostatic repulsion also increased due to the more positive charges bearing on the basic solutes, resulting in retention decreased. A gradual decreased trend of basic compounds retention was illustrated in Fig. 3(b). It indicated that the retention behavior of basic compounds was affected by electrostatic repulsion interaction seriously and protonated triazoles also played an important role on the retention of cations.

The results showed that the retention behavior of anions on Click Maltose was controlled by HILIC/WAX (weak anion exchange) mechanism,²³ while the chromatographic behavior of cations depended on electrostatic repulsion-hydrophilic interaction chromatography (ERLIC) mechanism.²⁴ The Click Maltose with surface positively charged exhibited the potential of simultaneously separating neutral, acidic and basic compounds under HILIC mode.

Applications

The separation of carbohydrates in conventional RP liquid chromatography has encountered challenges due to their poor retention in RP mode,^15c which was attributed to the existence of huge number of hydroxyl groups in the structure. In this study, mixtures of mono-, di- and trisaccharides were separated on Click Maltose, as shown in Fig. 4. Seven sugars were successfully separated in 15 min.

Fig. 4 Separation of a mixture of mono-, di- and trisaccharides on Click Maltose stationary phase. Mobile phase A, 100 mM ammonium acetate, pH 4.7, mobile phase B, acetonitrile, mobile phase C, water. 0–15 min, A/B/C, 10/75/15 (v/v/v) →10/70/20 (v/v/v), column temperature was 30 °C. Flow rate: 1.0 mL min⁻¹, ELS detector: gas pressure 30 psi, tube temperature 85 °C, gain 100. Solutes 1 sorbose, 2 glucose, 3 turanose, 4 cellobiose, 5 trehalose, 6 melezitose, 7 raffinose.

According to the above results, Click Maltose with positively charged surface had the potential of simultaneously separating neutral, acidic and basic compounds. A baseline separation of eight nucleic acid bases and nucleosides was achieved on Click Maltose (Fig. 5). Further study used a mixture of neutral, acidic and basic small peptides, which were successfully separated on Click Maltose (Fig. 6). Mobile phases contained 15 mM ammonium formate (pH 3.0) and the percentage of ACN changed from 67% to 63% in 10 min.

Fig. 5 Separation of eight nucleic acid bases and nucleosides on Click Maltose stationary phase. Mobile phase A, 100 mM ammonium formate, pH 3.0, mobile phase B, acetonitrile, mobile phase C, water. 0–12 min, A/B/C, 10/85/5(v/v/v), 12–15 min, 10/85/5 (v/v/v) →10/70/20 (v/v/v), 15–20 min, A/B/C, 10/70/20 (v/v/v) →10/60/30 (v/v/v), column temperature was 30 °C. Flow rate: 1.0 mL min⁻¹, UV detection: 260 nm. Solutes 1 5-methyluridine, 2 uridine, 3 adenosine, 4 inosine, 5 xanthosine, 6 cytidine, 7 guanosine, 8 1-methyladenosine.

Fig. 6 Separation of five small peptides on Click Maltose stationary phase. Mobile phase A, 100 mM ammonium formate, pH 3.0, mobile phase B, acetonitrile, mobile phase C, water. 0–10 min, A/B/C, 15/67/18 (v/v/v) →15/63/22 (v/v/v), column temperature was 30 °C. Flow rate: 1.0 mL min⁻¹, ELS detector: gas pressure 30 psi, tube temperature 85 °C, gain 100. Solutes 1 FGGF, 2 LGG, 3 Angiotensin III, 4 EE, 5 GGH.

The final example was performed to further illustrate the use of Click Maltose in oligosaccharides separation. Fig. 7 showed the separation of oligosaccharides in Rehmannia glutinosa Libosch. Compounds 1–3 were sucrose, raffinose and stachyose respectively and compounds 4–6 were unknown pentose.

Fig. 7 Separation of oligosaccharides in Rehmannia glutinosa Libosch on Click Maltose stationary phase. Sample was pretreated with SPE. Fractions 1–3 (a–c) were eluted with 5%, 10% and 15% ACN respectively. Mobile phase A, 100 mM ammonium acetate, pH 4.7, mobile phase B, acetonitrile, mobile phase C, water. 0–30 min, A/B/C, 10/75/15 (v/v/v) →10/60/30 (v/v/v), column temperature was 30 °C. Flow rate: 1.0 mL min⁻¹, ELS detector: gas pressure 30 psi, tube temperature 85 °C, gain 100. Compounds 1–3 were sucrose, raffinose and stachyose respectively. Compounds 4–6 were unknown pentose.

Reproducibility and stability

The reproducibility and the chemical stability are important parameters of stationary phases. Five standard solutes were selected to investigate the reproducibility and the chemical stability of Click Maltose. Fig. 8 showed the chromatograms of continuous 10 injections. The relative standard deviations (RSDs) of the five solutes were all less than 0.2%. It demonstrated a good reproducibility of Click Maltose.

Fig. 8 Chromatograms of continuous 10 times injections. Solutes and RSDs of retention times were indicated on peaks. Mobile phase A, 100 mM ammonium formate, pH 3.0, mobile phase B, acetonitrile, mobile phase C, water. 0–14 min, A/B/C, 10/85/5 (v/v/v), column temperature was 30 °C, flow rate: 1.0 mL min⁻¹, UV detection: 260 nm.

In addition, the Click Maltose stationary phase was continuously eluted by a solution of 20 mM ammonium formate (pH 3.0) for 3 days to test the chemical stability. There were no obvious changes of the retention times of standard compounds (Fig. 9). Intra-day and inter-day reproducibility of retention times for standard compounds on Click Maltose were also measured and the RSDs were less than 0.5% (Table 3). It showed good stability of Click Maltose.

Fig. 9 Retention stability of standard compounds on Click Maltose. The chromatographic column was eluted by a solution of 20 mM ammonium formate (pH 3.0) for 3 days.

Table 3 Intra-day and inter-day reproducibility of retention times for standard compounds on Click Maltose stationary phase

Samples	Intra RSD (%)^a			Inter RSD (%)^b
	24 h	48 h	72 h
a Intra-day reproducibility was measured by injecting the standard solution three times in a single day. b Inter-day reproducibility was measured by analyzing the retention times of standard compounds over three different days.
Uracil	0.18	0.00	0.18	0.15
Uridine	0.12	0.12	0.21	0.34
Cytosine	0.20	0.13	0.28	0.30
Cytidine	0.32	0.25	0.44	0.45
Orotic acid	0.26	0.21	0.39	0.29

---

Experimental

Chemicals and materials

Spherical silica (5 µm particle size, 10 nm pore size, 300 m²/g surface areas) was obtained from Fuji Silysia Chemical (Aichi, Japan). Maltose was purchased from Juyuan Biotechnology (Shanghai, China). 3-Chloropropyl triethoxysilane was purchased from ABCR (Karlsruhe, Germany) and sodium azide was obtained from Tianjing Chemical Reagents (Tianjing, China). Propargylamine was purchased from Acros (Fair Lawn, NJ., USA). Water was from a Milli-Q water purification system (Billerica, MA, USA), and acetonitrile of HPLC grade was purchased from Fisher (Fair Lawn, NJ., USA). Solutes for chromatographic evaluation were analytical quality, dissolved in methanol/water (1:1, v/v) solution.

Synthesis of maltose intermediates

The synthetic route of maltose intermediates was showed in Fig. 10.²⁷ Acetyl maltose was first prepared and then alkynyl group was introduced at C1 position by the substitution reaction of propargylalcohol and acetylated maltose with Lewis acid FB₃ catalysed. Finally, other acetyl groups were deacetylated by the reaction with methylamine to obtain 1-propargyl-O-maltose.

Fig. 10 The sythetic routes of intermediates of maltose: (i) Ac₂O, NaAc, 140 °C, 2 h; (ii) propargylalcohol, BF₃/Et₂O, CH₂Cl₂, 0–5 °C, 1 h, r.t., 4 h; (iii) MeNH₂/MeOH, RT, 12 h.

(i) Acetyl maltose. To an acetic anhydride solution was added maltose (20.0 g) and NaAc (2.0 g). The mixture was refluxed for 2 hours. The solution was cooled to room temperature, poured into ice water to obtain solid and then recrystallized in methanol to get acetyl maltose product (19.6 g, 49.4% yield). ¹H-NMR (400 MHz, [²H6]dimethyl sulfoxide (DMSO-d₆))δ: 2.01–2.13 (m, 24 H), 4.02–4.08 (m, 3 H), 4.20–4.25 (m, 3 H), 4.41 (d, 1 H, J = 10 Hz), 4.88 (dd, 1 H, J = 10.4 Hz, 8 Hz), 4.94 (dd, 1 H, J = 10.4 Hz, 4 Hz), 5.05 (t, 1 H, J = 10 Hz), 5.29 (t, 1 H, J = 10 Hz), 5.37 (d, 1 H, J = 4 Hz), 5.53 (t, 1 H, J = 9.2 Hz), 5.99 (d, 1 H, J = 8.4 Hz). ¹³C-NMR (100 MHz, [²H6]dimethyl sulfoxide (DMSO-d₆))δ: 20.71, 61.77, 63.10, 68.14, 68.44, 69.32, 69.84, 71.12, 72.40, 73.80, 74.23, 90.95, 95.95, 169.14, 169.61, 169.74, 169.96, 170.01, 170.31, 170.45, 170.57. Q/TOF-MS m/z 701.2451 for [M + Na]⁺, calcd (C₂₈H₃₈O₁₉Na) 701.1905.

(ii) 1-Propargyl-acetyl maltose. Acetyl maltose (15.0 g) was dissolved in 250 mL dichloromethane, and then 1.6 mL propargylalcohol and 4.2 mL boron trifluoride/ether were added. The mixture was reacted at 0–5 °C for 1 hour, and stirred at room temperature for another 4 hours. K₂CO₃ (7.5 g) was added and the solution continuously stirred for 30 minutes. The unreacted solid was filtered out and the filtrate was concentrated to obtain solid product. The solid was recrystallized in dichloromethane/n-hexane to get 1-propargyl-acetyl maltose (12.1 g, 81% yield). ¹H-NMR (400 MHz, CDCl₃)δ: 2.01–2.15 (m, 21 H), 2.48 (t, 1 H, J = 2.4 Hz), 3.72 (dt, 1 H, J = 9.6 Hz, 3.6 Hz), 6.94–3.96 (m, 1 H), 4.01–4.05 (m, 2 H), 4.22–4.28 (m, 2 H), 4.36 (d, 2 H, J = 2.4 Hz), 4.51 (dd, 1 H, J = 12 Hz, 2.4 Hz), 4.79–4.88 (m, 3 H), 5.06 (t, 1 H, J = 9.2 Hz), 5.29 (t, 1 H, J = 9.2 Hz), 5.36 (t, 1 H, J = 10 Hz), 5.42 (d, 1 H, J = 4 Hz). ¹³C-NMR (100 MHz, [²H6]dimethyl sulfoxide (DMSO-d₆))δ: 20.71, 20.78, 20.84, 20.87, 20.98, 21.01, 56.18, 61.83, 63.23, 68.18, 68.42, 69.30, 69.87, 71.71, 71.87, 74.06, 74.60, 78.30, 79.60, 95.81, 97.83, 169.61, 169.65, 169.96, 170.04, 170.32, 170.46, 170.59. Q/TOF-MS m/z 697.2393 for [M + Na]⁺, calcd (C₂₉H₃₈O₁₈Na) 697.1956.

(iii) 1-Propargyl-O-maltose. 1-propargyl-acetyl maltose (10.0 g) was dissolved in 20 mL methanol solution, and 60 mL methylamine alcohol was slowly dropwised. The mixture was stirred at room temperature for 24 hours and then concentrated to get oily solid. The solid was purified by silica gel column to obtain 1-propargyl-O-maltose (5.7 g, 75% yield). ¹H-NMR (400 MHz, [²H6]dimethyl sulfoxide (DMSO-d₆))δ: 2.54 (m, 1 H), 3.03–3.05 (m, 2 H), 3.20–3.24 (m, 2 H), 3.42–3.72 (m, 8 H), 4.26–4.30 (m, 2 H), 4.39 (dd, 1 H, J = 16 Hz, 2.4 Hz), 4.51–4.55 (m, 2 H), 4.90 (t, 2 H, J = 6.4 Hz), 5.02 (d, 1 H, J = 3.6 Hz), 5.24 (d, 1 H, J = 5.2 Hz), 5.43 (d, 1 H, J = 6.4 Hz), 5.52 (d, 1 H, J = 3.2 Hz). ¹³C-NMR (100 MHz, [²H6]dimethyl sulfoxide (DMSO-d₆))δ: 55.47, 61.11, 61.20, 70.29, 72.84, 73.23, 73.69, 73.92, 75.72, 76.76, 77.82, 79.90, 80.29. Q/TOF-MS m/z 403.1416 for [M + Na]⁺, calcd (C₁₅H₂₄O₁₁Na) 403.1216.

Immobilization of 1-propargyl-O-maltose and column packing

The immobilization of 1-propargyl-O-maltose on silica surface was showed in Fig. 11. 1-propargyl-O-maltose (3.1 g) was dissolved in 160 mL water/methanol (1:1, v/v). Azide-silica (5.0 g), CuSO₄ (0.075 g) and NaAsc (0.24 g) were added into solution. The mixture was stirred at 25 °C for 48 hours and then the solid was filtered, washed with methanol, water, 10% EDTA solution, water and acetone successively. The solid was dried at 60 °C for 12 hours to obtain Click Maltose stationary phase. The Click maltose silica was slurry-packed into a stainless steel column (150 mm × 4.6 mm I.D.) with methanol as slurry solvent under a pressure of 60 MPa. The separation efficiency is 60 000 plates/m at 0.5 mL min⁻¹, determined by cytidine with mobile phase containing 10 mM ammonium formate (pH 3.0) in 85% ACN.

Fig. 11 Immobilization of 1-propargyl-O-maltose on the silica surface.

Instrumentation and method

Experiments were performed on a Waters HPLC system, which consisted of a Waters 2695 HPLC pump, a Waters 2996 diode array detection (DAD), and a Waters 2420 evaporative light scattering detection (ELSD) system (Waters, Milford, MA, USA). The Empower workstation software was used for data recording. A Nicolet 20 DXB FT-IR (Madison, WI, USA) was used to get IR spectra. Data of solid state ¹³C cross polarization/magic-angle spinning (CP/MAS) NMR were obtained on a Bruker DSX 300 NMR Spectrometer (300 MHz, 7.0 T) (Karlsruhe, Germany), and the chemical shifts of ¹³C were referenced to tetramethylsilane (TMS). Elemental analysis characterization was performed on a Vario EL III elemental analysis system (Elementar, Germany).

The flow rate was 1.0 mL min⁻¹, and the column temperature was 30 °C. The dead time was 1.68 min, determined by injecting 2 µL methanol with mobile phase of ACN/water (1:1, v/v). The stock solution of ammonium acetate (100 mM) and ammonium formate (100 mM) were prepared by dissolving appropriate buffer salts in water. A ternary mobile phase system was used in experiments. Mobile phase A was ammonium buffer. Mobile phase B was ACN. Mobile phase C was water. In the experiment of studying the retention behavior of sugars, the percentage of water in mobile phase was changed from 20% to 40%, while keeping ammonium acetate (pH 4.7) constant at 10 mM. Ammonium formate (pH 3.0) was used in the study of retention behavior of ionic solutes. The mobile phase first contained 65% (v/v) ACN with various ammonium formate concentrations changed from 5 to 35 mM and then the pH values of the mobile phase changed from 4.4 to 3.0 while keeping ACN content constant at 65%.

Sampling of oligosaccharides in natural products

The natural product of Rehmannia glutinosa Libosch was employed to evaluate chromatographic performance of Click Maltose stationary phase. The sample was pretreated by solid phase extraction (SPE) and the material of SPE was porous graphitic carbon (PGC), purchased from Lanzhou Institute of Chemical Physics. The cartridge was washed by methanol, followed by water. The sample was loaded onto PGC cartridge, washed with water and eluted with 5%, 10%, and 15% (v/v) ACN respectively. Three collected fractions were dried by a rotary evaporator at 65 °C and dissolved in ACN/water (50:50, v/v) solution for further analysis.

Conclusions

In this work, a Click Maltose was synthesized based on a “click chemistry” method. The results of FT-IR, solid state ¹³C NMR and elemental analysis proved the successful immobilization of maltose on the silica surface and the designed chemistry of the stationary phase was obtained. The representative polar compounds of sugars, amino acids and small peptides were employed to evaluate the chromatographic properties of Click Maltose stationary phase. The retention behaviors of sugars on Click Maltose were consistent with typical HILIC characteristics. The retention equations demonstrated that surface adsorption mechanism resulted from the direct hydrogen bonding interaction between the sugars and Click Maltose played an important role on the retention of sugars. The retention of anions on Click Maltose depended on HILIC/WAX mechanism, while the chromatographic behavior of cations was controlled by ERLIC mechanism. Based on the above chromatographic properties of Click Maltose, it was successfully used in the separations of standard samples of polar compounds and complex samples of oligosaccharides in the natural product. In addition, the Click Maltose exhibited good reproducibility and stability which were beneficial to application. The unique chromatographic properties of Click Maltose and application examples, together with the synthetic methods, presented in this paper demonstrated the advantages and application potential of bonded oligosaccharides in HILIC. The further investigations of Click Maltose in the separation or enrichment of polar compounds, such as glycopeptides, nucleotides, glycosaminoglycans, etc., have been undergoing in our laboratory and will be reported in due course.

Acknowledgements

Financial support from the Natural Science Foundation of China (Grant: 20775079, 20805046) and Foundation for Distinguished Young Scholars (No. 20825518) from the National Natural Sciences Foundation of China.

Notes and references

1. A. Berthod, S. S. C. Chang, J. P. S. Kullman and D. W. Armstrong, Talanta, 1998, 47, 1001.
2. (a) E. A. Smith, W. D. Thomas, L. L. Kiessling and R. M. Corn, J. Am. Chem. Soc., 2003, 125, 6140; (b) Y. Zhang, S. Z. Luo, Y. J. Tang, L. Yu, K. Y. Hou, J. P. Cheng, X. Q. Zeng and P. G. Wang, Anal. Chem., 2006, 78, 2001; (c) C. C. Wang, Y. L. Huang, C. T. Ren, C. W. Lin, J. T. Hung, J. C. Yu, A. L. Yu, C. Y. Wu and C. H. Wong, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 11661.
3. F. Fazio, M. C. Bryan, O. Blixt, J. C. Paulson and C. H. Wong, J. Am. Chem. Soc., 2002, 124, 14397.
4. (a) B. T. Houseman and M. Mrksich, Chem. Biol., 2002, 9, 443; (b) B. T. Houseman, E. S. Gawalt and M. Mrksich, Langmuir, 2003, 19, 1522.
5. H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed., 2001, 40, 2004.
6. (a) V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596; (b) P. Appukkuttan, W. Dehaen, V. V. Fokin and E. Van der Eycken, Org. Lett., 2004, 6, 4223.
7. (a) M. C. Bryan, F. Fazio, H. K. Lee, C. Y. Huang, A. Chang, M. D. Best, D. A. Calarese, C. Blixt, J. C. Paulson, D. Burton, I. A. Wilson and C. H. Wong, J. Am. Chem. Soc., 2004, 126, 8640; (b) C. Y. Huang, D. A. Thayer, A. Y. Chang, M. D. Best, J. Hoffmann, S. Head and C. H. Wong, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 15.
8. (a) Y. Y. Tao and G. Carta, J. Chromatogr., A, 2008, 1211, 70; (b) M. C. Stone, Y. Y. Tao and G. Carta, J. Chromatogr., A, 2009, 1216, 4465; (c) E. J. Suda, K. E. Thomas, T. M. Pabst, P. Mensah, N. Ramasubramanyan, M. E. Gustafson and A. K. Hunter, J. Chromatogr., A, 2009, 1216, 5256.
9. (a) B. X. Yao, F. P. Zhan, G. Y. Yu, Z. F. Chen, W. J. Fan, X. P. Zeng, Q. L. Zeng and W. Weng, J. Chromatogr., A, 2009, 1216, 5429; (b) L. L. Zhou, C. Welch, C. Lee, X. Y. Gong, V. Antonucci and Z. H. Ge, J. Pharm. Biomed. Anal., 2009, 49, 964.
10. (a) S. Li and W. C. Purdy, Chem. Rev., 1992, 92, 1457; (b) J. Szejtli, Chem. Rev., 1998, 98, 1743.
11. (a) D. W. Armstrong, Chromatographic separation material - comprises cyclodextrin bonded to silica gel through functional organo-silane, US patent US4539399-A ( 1985); (b) D. W. Armstrong, T. J. Ward, R. D. Armstrong and T. E. Beesley, Science, 1986, 232, 1132.
12. (a) L. F. Zhang, Y. C. Wong, L. Chen, C. B. Ching and S. C. Ng, Tetrahedron Lett., 1999, 40, 1815; (b) X. H. Lai and S. C. Ng, Tetrahedron Lett., 2003, 44, 2657; (c) X. H. Lai and S. C. Ng, Tetrahedron Lett., 2004, 45, 4469; (d) Y. Liu, S. Urgaonkar, J. G. Verkade and D. W. Armstrong, J. Chromatogr., A, 2005, 1079, 146.
13. (a) W. C. Lee, C. C. Hsiao and R. C. Ruaan, J. Chem. Technol. Biotechnol., 1995, 64, 66; (b) A. Heeboll-Nielsen, M. Dalkiaer, J. J. Hubbuch and O. R. T. Thomas, Biotechnol. Bioeng., 2004, 87, 311; (c) M. Ortega-Munoz, J. Lopez-Jaramillo, F. Hernandez-Mateo and F. Santoyo-Gonzalez, Adv. Synth. Catal., 2006, 348, 2410.
14. A. J. Alpert, J. Chromatogr., A, 1990, 499, 177.
15. (a) M. J. Omaetxebarria, P. Hagglund, F. Elortza, N. M. Hooper, J. M. Arizmendi and O. N. Jensen, Anal. Chem., 2006, 78, 3335; (b) X. D. Wang, W. Y. Li and H. T. Rasmussen, J. Chromatogr., A, 2005, 1083, 58; (c) C. Antonio, T. Larson, A. Gilday, I. Graham, E. Bergstrom and J. Thomas-Oates, Rapid Commun. Mass Spectrom., 2008, 22, 1399; (d) H. Schlichtherle-Cerny, M. Affolter and C. Cerny, Anal. Chem., 2003, 75, 2349.
16. (a) D. V. McCalley, J. Chromatogr., A, 2007, 1171, 46; (b) B. A. Olsen, Elsevier Science Bv, Seattle, Washington, 2000, pp. 113; (c) T. Yoshida and T. Okada, J. Chromatogr., A, 1999, 840, 1; (d) G. Karlsson, S. Winge and H. Sandberg, J. Chromatogr., A, 2005, 1092, 246; (e) M. S. S. Curren and J. W. King, J. Chromatogr., A, 2002, 954, 41; (f) W. Jiang and K. Irgum, Anal. Chem., 2002, 74, 4682.
17. Z. M. Guo, A. W. Lei, Y. P. Zhang, Q. Xu, X. Y. Xue, F. F. Zhang and X. M. Liang, Chem. Commun., 2007, 2491.
18. T. Lummerstorfer and H. Hoffmann, J. Phys. Chem. B, 2004, 108, 3963.
19. C. E. Kibbey and M. E. Meyerhoff, Anal. Chem., 1993, 65, 2189.
20. Y. Guo and S. Gaiki, J. Chromatogr., A, 2005, 1074, 71.
21. P. Hemstrom and K. Irgum, J. Sep. Sci., 2006, 29, 1784.
22. (a) L. R. Snyder and H. Poppe, J. Chromatogr., A, 1980, 184, 363; (b) P. Nikitas, A. Pappa-Louisi and P. Agrafiotou, J. Chromatogr., A, 2002, 946, 33.
23. J. Lin, X. C. Lin and Z. G. Xie, J. Chromatogr., A, 2009, 1216, 801.
24. A. J. Alpert, Anal. Chem., 2008, 80, 62.
25. J. A. Dean, in Lange's Handbook of Chemistry, 15th edition, pp. 8.69.
26. K. B. Sharpless, J. Biotechnol., 2008, 136, S3.
27. (a) H. B. Mereyala and S. R. Gurrala, Carbohydr. Res., 1998, 307, 351; (b) A. K. Pathak, V. Pathak, J. M. Riordan, S. S. Gurcha, G. S. Besra and R. C. Reynolds, Carbohydr. Res., 2004, 339, 683.

---

Footnote

† The first two authors contributed equally to this work.

---