2025 年 48 巻 6 号 p. 801-804
Oversulfated chondroitin sulfate (OSCS) is a chondroitin sulfate ester in which all hydroxyl groups have been converted to sulfuric acid esters. Here, we describe a rapid, novel synthesis method for OSCS. 1H-NMR analysis revealed that all signals derived from the protons of N-acetyl-galactosamine and glucuronic acid in the synthesized OSCS were shifted downfield due to sulfonation compared with the chondroitin sulfate (CS) starting material. Comparison of the edited heteronuclear single quantum correlation spectrum of the prepared OSCS with that of the Japanese Pharmacopoeia OSCS reference standard showed excellent agreement between the main correlation peaks. The proposed novel synthesis method for OSCS is faster and simpler than the conventional method.
Oversulfated chondroitin sulfate (OSCS) is a chondroitin sulfate derivative in which all hydroxyl groups have been converted to sulfate esters. The presence of OSCS as a contaminant in heparin sodium preparations leads to severe adverse clinical events, including rapid drops in blood pressure and acute inflammatory reactions, raising significant concerns with regard to pharmaceutical safety.1–3)
Various methods for OSCS synthesis have been reported, including the use of a tributylamine and sulfur-pyridinium trioxide complex, as described by Toida and colleagues.4–6) However, these conventional approaches often require labor-intensive procedures involving substitution of sodium chondroitin sulfate with a low molecular weight organic base such as tributylamine, carrying out a reaction to introduce sulfonic acid groups, followed by cation-exchange resin purification, all of which prolong the overall reaction time, making these approaches impractical for rapid synthesis.6) Thus, there is a need for a simpler and more efficient method to synthesize OSCS, especially for laboratory-scale applications.
In this study, we propose a novel, rapid synthesis method for OSCS. The proposed method can easily introduce sulfonic acid groups using cetylpyridinium chloride (CPC) and eliminates the need for cation-exchange resin purification while maintaining a high level of structural integrity. The synthesized OSCS was characterized using NMR spectroscopy and compared with the Japanese Pharmacopoeia (JP) OSCS reference standard to confirm its structural fidelity.
Chondroitin sulfate sodium salt derived from shark cartilage was provided by the Pharmaceutical and Medical Device Regulatory Science Society of Japan (Tokyo, Japan). A 1% aqueous solution of CPC monohydrate (hexadecylpyridinium chloride monohydrate, FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) was dissolved in ultrapure water. This stock solution was stored at room temperature and diluted 10 times with ultrapure water to give a 0.1% (w/w) working solution. The sulfonation reagent was prepared by dissolving sulfur trioxide pyridine complex (Sigma-Aldrich, Co., Ltd., St. Louis, MO, U.S.A.) in N,N-dimethylformamide (DMF, Wako Pure Chemical Corporation) to a concentration of 137.4 mg/0.8 mL. This solution turns yellow over time and thus small amounts (sufficient for about three experiments) were prepared as required and stored at room temperature.
Novel OSCS Synthesis MethodChondroitin sulfate sodium salt (10 mg) was weighed into several 15 mL centrifuge tubes and 10 mL of the above-described 0.1% CPC aqueous solution was added to form a chondroitin sulfate-CPC complex precipitate. After standing at 4°C for over 1 h, the samples were centrifuged (3740, KUBOTA, Osaka, Japan; 4°C, 2300 × g, 15 min), the supernatant was removed, and the precipitate was dried in a glass desiccator in the presence of diphosphorus pentoxide (Wako Pure Chemical Corporation) as a desiccant under reduced pressure using a dry vacuum pump (Rocker 300, Sibata Scientific Technology Ltd., Tokyo, Japan) overnight.
After the chondroitin sulfate-CPC complex in the desiccator became transparent, it was dissolved in 0.8 mL of DMF by vortex mixing. The sulfating reagent was added in 0.8 mL portions and the sample was mixed further. The sample was then placed in a dry block bath (HB150-S2, DLAB) set at 50°C, and heated for 1 h, with vortexing at 10-min intervals. NaCl (1.6 mL, 5 M) was added to each sample, followed by vortex mixing. The pH of the solution was then adjusted to 7 using 1 M NaOH. Sodium acetate-saturated ethanol (10 mL) was added to the samples, mixed by inversion and vortexing, and allowed to stand overnight at 4°C to form a precipitate. The product formed by ethanol precipitation was collected by centrifugation (4°C, 2300 × g, 15 min), the supernatant was removed, and the precipitate was air-dried. The dry precipitate was dissolved in 4 mL of ultrapure water. The dissolved precipitate from four sample tubes was dialyzed using a cellulose dialysis membrane (SnakeSkin Dialysis Tubing 88244, Thermo Fisher Scientific Inc., Waltham, MA, U.S.A.) with a molecular weight cutoff (MWCO) of 3.5 kD against 5 L of ultrapure water for more than 24 h (generally from 24 to 72 h) with magnetic stirring. The dialysis water was changed three times, after dialysis for 30 min or longer.
The dialysate was aliquoted into 50 mL centrifuge tubes (approximately 25 mL per tube) and placed tilted in a −20°C freezer overnight, followed by freeze-drying (FRD-82M, AGC Techno Glass Co., Ltd., Tokyo, Japan) for more than 2 d. The yield of the obtained dried product was measured.
NMR SpectroscopyThe dried product (10 mg) was weighed and completely dissolved in 750 µL of deuterium oxide (D2O) 99.96% (FUJIFILM Wako Pure Chemical Corporation), placed and sealed in an NMR sample tube, and measured using an NMR spectrometer (JNM-ECA600II, JEOL, Ltd., Tokyo, Japan) equipped with a 5-mm ROYAL probe at 45°C. NMR measurements were conducted at 600 MHz for 1H and 150 MHz for 13C (14.1 T). One-dimensional (1D) 1H-NMR spectra were acquired with a 45° pulse angle, a relaxation delay of 5 s, and spectral widths of 15 ppm using 32 k data points. The DHO (4.63 ppm) peak for D2O at 45°C was used as a reference. CRISIS adiabatic heteronuclear single quantum coherence (HSQC) sequence (multiplicity-edited HSQC, edited HSQC) data were obtained based on 2D 1H-13C HSQC spectra. All NMR spectra were processed using Delta NMR Processing and Control Software (JEOL).
HPLC Conditions for System Suitability Test as Outlined in Monograph “Heparin Sodium” in the JP, 18th Edition4,5)HPLC analysis was performed on a Prominence HPLC system (Shimadzu Corporation, Kyoto, Japan) using the conditions described below. The column used is the TSKgel DEAE-5PW HPLC column (Tosoh Bioscience LLC, Tokyo, Japan, 10 µm, 2.0 mm × 7.5 cm). For mobile phase A, 0.4 g of sodium dihydrogen phosphate dihydrate was dissolved in 1000 mL of water, and 10-times diluted phosphoric acid was added to adjust the pH to 3.0. For mobile phase B, 0.4 g of sodium dihydrogen phosphate dihydrate and 106.4 g of lithium perchlorate were dissolved in 1000 mL of water, and 10-times diluted phosphoric acid was added to adjust the pH to 3.0. The gradient was an initial 0–3 min with a mixture of 90% mobile phase A and 10% mobile phase B, followed by 3–15 min with a varying mixture of 90→0% mobile phase A and 10→100% mobile phase B. The injection volume was 20 µL, and the flow rate was 0.2 mL/min. The column temperature was 35°C, and detection was performed at a UV wavelength of 202 nm.
Previously reported OSCS synthesis methods4,5) require complicated operations such as cation-exchange resin column purification. Here, we precipitated the CS starting material using CPC and then dried the precipitate. Subsequent sulfonation with phosphorus pentoxide was achieved without cation-exchange resin column purification. We tested CPC concentrations of 0.1, 0.2, 0.3, 0.4, and 0.5%. No OSCS was synthesized using 0.4 and 0.5% CPC, and NMR analysis of the 0.3% CPC product showed many impurities. The 0.1% CPC precipitate gave the best results (data not shown), suggesting that CS sulfonation becomes more difficult as the CPC concentration increases. Drying the product after the sulfonation step is likely important for maximizing the yield of OSCS. We ultimately decided to precipitate the product using ethanol and to desalt it, followed by freeze-drying.
The sample was further desalted using dialysis membranes with 14 kD MWCO or 3.5 kD MWCO dialysis membranes. The recovery (final weight) using the 3.5 kD MWCO dialysis membrane was much better than that obtained using a 14 kD MWCO dialysis membrane, likely because the synthesized OSCS could not diffuse through the 3.5 kD MWCO dialysis membrane. The proposed synthesis method uses CPC precipitation, without the need for cation-exchange chromatography, providing a quicker, easier route to OSCS compared with previous synthesis methods.
Supplementary Figure S1 illustrates the differences between the conventional OSCS synthesis method6) and the proposed method. Unlike the traditional approach that requires strong cation-exchange chromatography and lengthy purification steps, the proposed method streamlines the process by utilizing CPC precipitation, thereby significantly reducing the complexity. The total synthesis time is reduced by approximately 48 h from 8 to 6 d, making it a more practical approach for laboratory use.
When 16 centrifuge tubes weighing 10 mg were prepared at the start of the experiment, the amount of OSCS obtained from 160 mg of raw material was approximately 80–100 mg (yield: 50–60%).
NMR AnalysisThe OSCS synthesized using the proposed method was evaluated using 1H-NMR and HSQC. The 1D 1H-NMR spectra of the parent CS and OSCS using are shown in Fig. 1 and Supplementary Fig. S2. In addition, 1D 13C-NMR spectra of the parent CS and OSCS are shown in Supplementary Fig. S3. Substantial structural heterogeneity associated with sulfonation at the 4- and/or 6-positions of the GalNAc residue is evident. In contrast, data for the OSCS synthesized using the present method show that all proton-derived signals are shifted to a lower magnetic field compared with the original CS used as a starting material, due to sulfonation of the hydroxyl groups. Comparison of the JP OSCS reference standard and the OSCS synthesized in the present study shows essentially identical chemical shifts of the signals derived from the protons of GalNAc. We analyzed the synthesized OSCS and original CS using edited HSQC (Fig. 2): the downfield shifts of signals associated with ring protons attached to the O-sulfonated carbons, such as GlcA H-2, H-3, and GalNAc H-4, H-6, confirm the sequence, and that the chemical shifts for A4 and A6 for CS converge to A6 in OSCS.
(A) A1: GalNAc-1, A2: GalNAc-2, A3: GalNAc-3, A4: GalNAcOS-4, A5: GalNAc-5, A6: GalNAcOS-6, U1: GlcA-1, U2: GlcA-2, U3: GlcA-3, U4: GlcA-4, U5: GlcA-5. (B, C) A1′: GalNAc-1, A2′: GalNAc-2, A3′: GalNAc-3, A4′: GalNAcS-4, A5′: GalNAc-5, A6′: GalNAcS-6, U1′: GlcA-1, U2′: GlcAOS-2, U3′: GlcAOS-3, U4′: GlcA-4, U5′: GlcA-5.
The OSCS synthesized using the proposed method was subjected to HPLC analysis using the methods described in the system suitability test in the monograph “Heparin Sodium” in the Japanese Pharmacopoeia, 18th edition.4,5) The results indicated successful detection of OSCS (Supplementary Fig. S4). The test for the degree of separation between heparin sodium and OSCS gave a value of more than 1.5 (Supplementary Fig. S5). The results indicated that dermatan sulfate, heparin, and OSCS were eluted in that order (Supplementary Fig. S6). As shown in Supplementary Fig. S7, a comparison of the JP OSCS reference standard and the novel synthesized OSCS shows essentially identical retention times and chromatographic profiles.
We have developed a novel, rapid OSCS synthesis method that eliminates the need for cation-exchange chromatography and minimizes processing complexity. By utilizing CPC precipitation, we achieved efficient sulfonation without requiring extensive purification steps. The conventional OSCS synthesis process typically requires several days due to the extensive amount of purification needed. In contrast, the proposed method can significantly reduce the overall synthesis time by approximately 48 h from 8 to 6 d, making it a practical alternative for laboratory-scale production.
Our findings demonstrate that the synthesized OSCS exhibits excellent structural similarity to the JP OSCS reference standard, as confirmed by NMR spectroscopy. This approach provides a valuable tool for researchers in glycosaminoglycan chemistry and pharmaceutical sciences. The NMR analysis revealed that all signals derived from the protons of N-acetyl-galactosamine and glucuronic acid in OSCS were shifted to a lower magnetic field due to sulfonation, compared with the signals for CS, the raw material. Furthermore, a comparison of the HSQC spectrum of the OSCS prepared in this study with that of the JP OSCS reference standard showed excellent agreement between the main correlation peaks (Figs. 1, 2). This novel OSCS synthesis method is therefore expected to be useful for preparing JP OSCS reference standards.
This study was supported by a Grant from the Pharmaceutical and Medical Device Regulatory Science Society of Japan.
The authors declare no conflict of interest.
This article contains supplementary materials.
