Profiling sulfate content of polysaccharides in seaweed species using a ligand-assisted ¹H-NMR assay

Abstract

This study aimed to use an existing ligand (imidazole)-assisted ¹H-nuclear magnetic resonance (¹H-NMR) spectroscopy protocol to determine sulfate content of seaweed sulfated saccharides. Twenty seaweed samples – spanning nine species and multiple geographic locations – were analyzed. Extracts were obtained by immersing freeze-dried seaweeds in boiling water, followed by alginate precipitation. Remaining supernatants contained natural crude polysaccharide (CP) fraction, including sulfated polysaccharides. To eliminate free CP carboxy groups that can interfere with the complex formation of imidazole with CP sulfate groups, the carboxy groups were chemically reduced to hydroxymethyl groups prior to ¹H-NMR spectroscopy. Resultant sulfate content estimates were comparable to those obtained via a conventional barium-rhodizonate assay. Findings indicate that the imidazole-assisted ¹H-NMR spectroscopy is an appropriate method for estimation of seaweed CP sulfate content.

Introduction

Seaweeds have been regarded as food items by coastal communities of the Asia-Pacific region (including Japan, South Korea, and China) for decades (Ale et al., 2011b). Recently, the bioactivities of seaweed cell wall sulfated polysaccharides have sparked extensive research. These compounds exhibit anti-tumor (Koyanagi et al., 2003), anti-coagulant (Mansour et al., 2019), and immunomodulatory effects (Miyazaki et al., 2019), among others. Generally speaking, seaweed sulfated polysaccharide bioactivities are related to their structural properties, including molecular size and sugar composition (Melo et al., 2002). Additionally, bioactivity is dependent on the presence of sulfate groups, which alter polysaccharide conformation, charge, and solubility (Patankar et al., 1993; Sun et al., 2018; Wang and Zhang, 2009). Notably, polysaccharides with a greater number of sulfate groups exhibit higher levels of bioactivity (Koyanagi et al., 2003; Sun et al., 2018; You et al., 2010). Therefore, sulfate group quantitation is important in evaluating sulfated polysaccharide functional potential.

A common method for determining sulfate content of sulfated polysaccharides is initial acid hydrolysis (to liberate inorganic sulfate), followed by the precipitation using barium ions (Jun et al., 2018; Koyanagi et al., 2003; Sun et al., 2018; You et al., 2010). Because of the relationship between sulfate groups and bioactivity, isolation methods should ideally prevent sulfate group loss or polysaccharide structural alteration (Hahn et al., 2012). However, the barium method employs extensive extraction and purification procedures, which may lead to polysaccharide desulfation and degradation (Wang and Chen, 2016). Consequently, reproducibility of this method is low. A more suitable and reliable method – lacking extensive extraction (e.g., acid-based) and purification procedures – for polysaccharide sulfate content evaluation is required (Ale et al., 2011a; Hahn et al., 2012). We recently established an imidazole-assisted ¹H-NMR spectroscopy protocol for estimation of the sulfate content of sulfated saccharides and successfully evaluated the sulfate content of commercially available fucoidan (≥ 95% pure) extracted from Fucus vesiculosus (Bak et al., 2020). As a logical next step, the present study aimed to apply this novel ligand (imidazole)-assisted ¹H-NMR assay to the estimation of sulfate content of seaweed natural crude polysaccharide (CP). Nine seaweed species spanning various geographic locations were sampled, and their boiling-water extracts were subjected to the ligand-assisted ¹H-NMR assay without any tedious purifications.

Materials and Methods

Reagents and specimens    Sodium borohydride (NaBH₄), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), L-ascorbic acid, and imidazole were obtained from Nacalai Tesque Inc. (Kyoto, Japan). Additionally, 3-(trimethylsilyl) propionic-2,2,3,3-d₄ acid (TSP-d₄) was purchased from Wako Pure Chemical Co. (Osaka, Japan). Deuterium oxide (D₂O, 99.8 atom% D) and sodium rhodizonate were obtained from Kanto Chemical Co. (Tokyo, Japan).

Nine species of pickled seaweed (one green, one red, and seven brown) were purchased from a local Japanese market. The green seaweed was Aosa-Nori (Monostroma nitidum). The red seaweed was Nori (Porphyra yezoensis). The brown seaweeds were as follows: Akamoku (Sargassum horneri), Gagome-Kombu (Kjellmaniella crassifolia), Ishi-Mozuku (Sphaerotrichia firma), Kombu (Saccharina japonica), Mekabu (Sporophyll of Undaria pinnatifida), Okinawa-Mozuku (Cladosiphon okamuranus), and Wakame (Undaria pinnatifida). All were harvested from different sites (Table 1).

Table 1. Comparison of polysaccharide sulfate content among commercially available seaweeds as estimated by an imidazole-assisted ¹H-NMR assay versus a conventional barium-rhodizonate assay

Seaweed		Harvest site	Sulfate content (%)*
Common name	Scientific name		NMR	Barium-rhodizonate**
Aosa-Nori	Monostroma nitidum	Aichi	7.61	7.94 ± 0.64 (7.33–8.59)
Nori	Porphyra yezoensis	Shizuoka	7.41	7.50 ± 0.66 (7.09–8.26)
Akamoku	Sargassum horneri	Aichi	7.17	7.37 ± 0.75 (6.79–8.21)
		Miyagi	7.31	6.82 ± 0.77 (6.21–7.67)
		Nagasaki	4.61	5.01 ± 0.40 (4.57–5.36)
		Mie	3.77	4.37 ± 0.59 (3.73–4.86)
Gagome-Kombu	Kjellmaniella crassifolia	Sapporo	10.4	11.1 ± 1.18 (10.1–12.4)
Ishi-Mozuku	Sphaerotrichia firma	Aomori	9.68	9.67 ± 0.30 (9.49–10.0)
		Niigata	16.3	16.0 ± 1.80 (14.7–18.1)
Kombu	Saccharina japonica	Iwate	4.35	4.43 ± 0.78 (3.71–5.24)
Mekabu	Sporophyll of Undaria pinnatifida	Miyagi	3.73	4.23 ± 0.90 (3.36–5.15)
		Shimane	7.56	7.35 ± 0.87 (6.42–8.15)
Okinawa-Mozuku	Cladosiphon okamuranus	Miyako Island	4.65	4.05 ± 0.71 (3.28–4.68)
		Tsuken Island	4.97	4.51 ± 0.82 (3.86–5.42)
		Ishigaki Island	4.81	5.35 ± 0.58 (4.68–5.68)
		Kume Island	4.04	4.32 ± 0.40 (3.94–4.72)
Wakame	Undaria pinnatifida	Miyagi	2.79	3.36 ± 0.95 (2.38–4.26)
		Nagasaki	5.53	5.77 ± 0.69 (5.02–6.35)
		Tokushima	3.85	3.70 ± 0.85 (2.74–4.33)
		Iwate	2.45	3.10 ± 0.83 (2.29–3.93)

*  Sulfate content is expressed as a percentage of polysaccharide dry mass.

**  Values represent means of three repeated measurements ± SD; total value range is provided in parentheses.

Preparation of CP fraction    Seaweed specimens were washed with tap water to desalt, followed by lyophilization. Dried seaweed was crashed, and 20–25 g of the sample was suspended in 300 mL of boiling-water for 1 h to obtain CP fraction. The suspension was then centrifuged (10 810 g, 20 min, 10 °C) and the supernatant was collected. This procedure was repeated three times. The pooled supernatants were concentrated via rotary evaporation. To precipitate alginate, 50 wt% calcium chloride solution was added. The mixture was centrifuged (13 980 g, 30 min, 10 °C) and the alginate-free supernatant was dialyzed (membrane MW cut off: 12,000; Nacalai Tesque) against 10 L of deionized water over a period of three days. The dialysates were concentrated, and lyophilized. Dried CP powder was weighed.

Chemical reduction of CP carboxy groups    Free CP carboxy groups may interfere with the formation of the complex between imidazole and CP sulfate groups (Bak et al., 2020). Therefore, prior to the ligand-assisted ¹H-NMR assay, CP carboxy groups were reduced with EDC and NaBH₄, according to the method described by Nagaoka et al. (1999), with certain modifications. Briefly, a 20 mg/mL CP solution was made up in deionized water. Thereafter, 100 mg of EDC was gradually added to 5 mL of this solution, and the pH was adjusted to 4.75–5.0 using 0.1 M HCl while stirring continuously for 2 h. Then, 500 mg of NaBH₄ was slowly added while stirring, and the pH was adjusted to 7.0 using 6 M HCl over a period of 1 h. A few drops of ethanol were added to prevent foaming. The solution was dialyzed (membrane MW cut off: 12,000; Nacalai Tesque) against deionized water over a period of two days. The dialysate was then passed through a TOYO-PAK IC-SP M cartridge (TOSOH Co., Tokyo, Japan) to remove remaining cations (e.g., Na⁺ and Ca²⁺), and the filtrate was lyophilized. Dried reduced CP was dissolved in D₂O for subsequent ¹H- and ¹³C-NMR measurements. Reduction of CP carboxy to hydroxymethyl groups was confirmed by the presence of ¹³C-NMR spectra at 59.0 ppm (hydroxymethyl group) and by the absence of 178.1 ppm (carboxy group) (Cheng and Neiss, 2012; Nagaoka et al., 1999).

¹H- and ¹³C-NMR spectroscopies    One-dimensional ¹H- and ¹³C-NMR spectroscopies were performed at 25 °C using an ECS-400 spectrometer (JEOL, Tokyo, Japan) to observe changes in chemical shift (Δδ) of the imidazole proton (H4,5) and to confirm the loss of carboxy signal in the reduced CP, respectively. For ¹H-NMR, a single pulse sequence was used to acquire NMR spectra, employing the following parameters: acquisition time of 2.18 s, 16,384 data acquisition points, 8–512 scans, a relaxation delay of 5 s, and spinning at 15 Hz. For ¹³C-NMR, a single pulse sequence was used to acquire NMR spectra, employing the following parameters: an acquisition time of 2.18 s, 32,768 data acquisition points, 20 000–30 000 scans, a relaxation delay of 2 s, and spinning at 15 Hz. As an external standard, TSP-d₄ was used. To prevent interference of TSP-d₄ carboxy groups with the interaction between imidazole and CP sulfate groups, a 3.5 mm-stem coaxial insert NMR sample tube (Nihon Seimitsu Scientific Co., Tokyo, Japan) containing TSP-d₄ dissolved in D₂O was inserted into a 5 mm-NMR sample tube (Nihon Seimitsu Scientific Co.) containing 400 µL of sample solution, across all samples.

Estimation of sulfate content    To estimate CP sulfate content via imidazole-assisted ¹H-NMR spectroscopy, a range of concentrations of imidazole (0.1–4.0 mmol imidazole per g of CP sample) dissolved in D₂O was mixed with reduced CP sample dissolved in D₂O (0.08 mg/mL). Imidazole δ values were monitored and plotted against imidazole concentration. Based on the inflection point of the plot, a regression equation was used to estimate the sulfate content of each sample (Bak et al., 2020). To explore the sensitivity of the proposed ligand-assisted ¹H-NMR assay, LOD for imidazole ¹H-NMR signal at the scanning number of 512 were determined. Accuracy of sulfate content estimation by the ¹H-NMR assay was confirmed by comparing values to those obtained from native CP using a conventional barium-rhodizonate method (Silvestri et al., 1982). Briefly, acid hydrolysis of 10 mg of CP in 3 M HCl for 2 h at 100 °C preceded the assay, using sodium sulfate as a standard.

Data analysis    While triplicate ¹H-NMR measurements were performed in this study, because δ value is constant under fixed NMR conditions, there was no difference in δ value among replication tests. The conventional assay was performed in triplicate for each sample. Sulfate content values are presented as the mean ± standard deviation (SD).

Results and Discussion

Natural seaweed cell wall polysaccharides consist largely of alginate having carboxy groups and sulfated fucans and/or galactan, which account for 40–50% of the dry mass of whole seaweeds (Jun et al., 2018). Alginate removal from seaweed sample was thus crucial for the estimation of sulfate content using an imidazole-assisted ¹H-NMR assay, since carboxy groups also form the complex with imidazole as a ligand (Bak et al., 2020). In this study, alginate was removed from the seaweed samples by calcium chloride precipitation. To confirm the complete removal of any interfering carboxy groups remaining in CP samples used in this study, further chemical reduction treatment of CP samples with EDC and NaBH₄ was performed. As shown in Fig. 1A, ¹³C-NMR measurement revealed the disappearance of carboxy groups in CP samples by the chemical reduction treatment; the observed 178.1 ppm from carboxy groups disappeared after the treatment, while the reduced form (hydroxymethyl groups) was newly observed at 59.0 ppm (Cheng and Neiss, 2012; Nagaoka et al., 1999).

Fig. 1.

¹H-NMR-aided estimation of sulfate content in seaweed polysaccharides. Imidazole-assisted ¹H-NMR assay was performed for the estimation of sulfate content in crude polysaccharides (CP) extracted from Monostroma nitidum harvested from a site of Aichi. Chemical reduction of carboxy groups in CP with EDC and NaBH₄ to hydroxymethyl groups was confirmed by ¹³C-NMR measurements (A). Resulting reduced CP was taken forward as an analyte at a fixed concentration of 0.08 mg/mL. Varying imidazole concentration between 0.25 and 3.5 mmol imidazole per g of analyte altered the chemical shift (δ) of imidazole protons (H4,5) (B). Imidazole concentration was plotted against the δ value (C).

To explore the influence of chemical reduction treatment on sulfate content estimation, the conventional barium-rhodizonate method was applied to determine the sulfate content in native (non-reduced) and reduced CPs. Although data were not shown, sulfate content values of native (7.94 ± 0.64% of total dry mass, n = 3) versus reduced CP from Aichi M. nitidum (7.07 ± 0.71% of total dry mass, n = 3) did not differ significantly by the Student's t-test (p = 0.19), indicating that the reduced CP may be applicable for the estimation of sulfate content by the present imidazole-assisted ¹H-NMR assay.

Imidazole was chosen as a ligand, given its ionic interaction and 1:1 complexing with a sulfate group of sulfated saccharides (Bak et al., 2020). To avoid imidazole proton signal broadening (Forshed et al., 2005) by polysaccharide synergistic viscosity – due to polymer entanglement - at higher concentrations (> 0.1 mg/mL) (Secouard et al., 2007), reduced CP concentration was fixed at 0.08 mg/mL for downstream analyses. Different imidazole concentrations were added to this concentration of analyte to assess impact on ¹H-NMR performance (Fig. 1B). In the proposed ligand-assisted ¹H-NMR assay, the estimable CP sulfate content is strongly dependent on LOD for imidazole. The LOD for imidazole target proton (H4,5) was found to be 0.011 mmol imidazole per g of reduced CP sample at the scanning number of 512, indicating that the proposed assay showed sufficient sensitivity to estimate the CP sulfate content. In this study, an appropriate scanning number (8–512), allowing to obtain ¹H-NMR signals of the imidazole proton (H4,5) with higher signal-to-noise ratio (> 3:1), was employed (Lacey et al., 1999).

In the presence of reduced CP derived from Aichi M. nitidum, the observed δ value of the imidazole proton (H4,5) is 7.485 ppm (up to concentrations of 0.75 mmol imidazole per g of reduced CP sample). At increasing concentrations above 0.75 mmol/g, a concentration-dependent upfield shift of imidazole proton resonance occurs (Fig. 1B). Notably, the δ value of the imidazole proton shifts linearly (R₂ = 0.985) in the range of 0.88–3.5 mmol/g (Fig. 1C). The inflection point predicted by the regression equation indicates the 1:1 complex formation of imidazole with sulfate group (Bak et al., 2020), showing the sulfate content of 0.79 ± 0.00 mmol/g (i.e. 7.61 ± 0.00% of total dry mass) for Aichi M. nitidum (Fig. 1C). Sulfate content values estimated via the ¹H-NMR assay were compared to those determined by the conventional bariumrhodizonate assay. Sulfate content of Aichi M. nitidum-derived CP as estimated NMR-based method (7.61 ± 0.00%) corresponded well with that by the barium-rhodizonate method (7.94 ± 0.64% of total dry mass, n=3) (Table 1). Since the parameter δ is constant under fixed NMR conditions, sulfate content value by the present ¹H-NMR assay may be highly accurate when compared to that by the conventional assay. Regardless of variation in species and geographic location, for all species, estimated CP sulfate content values via the ¹H-NMR-based assay were in the range of those obtained via triplicate application of the barium-rhodizonate assay.

Taken together, results of the present study demonstrate successful application of the novel imidazole-assisted ¹H-NMR assay to estimate the sulfate content of seaweed polysaccharides, without the need for extensive purification which may alter CP architecture or chemistry.

Acknowledgements    This study was supported in part by a Grant-in-Aid for Scientific Research (C) awarded to Y.M. (No. JP17K07819) from the Japan Society for the Promotion of Science (JSPS) Research, as well as a JSPS Research Fellowship for Young Scientists awarded to J.B. (No. JP19J20015).

Conflict of interest    The authors have declared that no competing interests exist.

References

•  Ale,  M.T.,  Mikkelsen,  J.D., and  Meyer,  A.S. (2011a). Designed optimization of a single-step extraction of fucose-containing sulfated polysaccharides from Sargassum sp. J. Appl. Phycol., 24, 715-723.
•  Ale,  M.T.,  Mikkelsen,  J.D., and  Meyer,  A.S. (2011b). Important determinants for fucoidan bioactivity: a critical review of structure-function relations and extraction methods for fucose-containing sulfated polysaccharides from brown seaweeds. Mar. Drugs, 9, 2106-2130.
•  Bak,  J.,  Miyazaki,  Y.,  Nakano,  H., and  Matsui,  T. (2020). Ligand-aided ¹H nuclear magnetic resonance spectroscopy for non-destructive estimation of sulfate content in sulfated saccharides. Anal. Sci., 36, 1269-1274.
•  Cheng,  H.N. and  Neiss,  T.G. (2012). Solution NMR spectroscopy of food polysaccharides. Polym. Rev., 52, 81-114.
•  Forshed,  J.,  Erlandsson,  B., and  Jacobsson,  S.P. (2005). Quantification of aldehyde impurities in poloxamer by ¹H NMR spectrometry. Anal. Chim. Acta, 552, 160-165.
•  Hahn,  T.,  Lang,  S.,  Ulber,  R., and  Muffler,  K. (2012). Novel procedures for the extraction of fucoidan from brown algae. Process Biochem., 47, 1691-1698.
•  Jun,  J.Y.,  Jung,  M.J.,  Jeong,  I.H.,  Yamazaki,  K.,  Kawai,  Y., and  Kim,  B.M. (2018). Antimicrobial and antibiofilm activities of sulfated polysaccharides from marine algae against dental plaque bacteria. Mar. Drugs, 16, 301.
•  Koyanagi,  S.,  Tanigawa,  N.,  Nakagawa,  H.,  Soeda,  S., and  Shimeno,  H. (2003). Oversulfation of fucoidan enhances its anti-angiogenic and antitumor activities. Biochem. Pharmacol., 65, 173-179.
•  Lacey,  M.E.,  Subramanian,  R.,  Olson,  D.L.,  Webb,  A.G., and  Sweedler,  J.V. (1999). High-resolution NMR spectroscopy of sample volumes from 1 nL to 10 mL. Chem. Rev., 99, 3133-3152.
•  Mansour,  M.B.,  Balti,  R.,  Yacoubi,  L.,  Ollivier,  V.,  Chaubet,  F., and  Maaroufi,  R.M. (2019). Primary structure and anticoagulant activity of fucoidan from the sea cucumber Holothuria polii. Int. J. Biol. Macromol., 121, 1145-1153.
•  Melo,  M.R.S.,  Feitosa,  J.P.A.,  Freitas,  A.L.P., and  De Paula,  R.C.M. (2002). Isolation and characterization of soluble sulfated polysaccharide from the red seaweed Gracilaria cornea. Carbohydr. Polym., 49, 491-498.
•  Miyazaki,  Y.,  Iwaihara,  Y.,  Bak,  J.,  Nakano,  H.,  Takeuchi,  S.,  Takeuchi,  H.,  Matsui,  T., and  Tachikawa,  D. (2019). The cooperative induction of macrophage activation by fucoidan derived from Cladosiphon okamuranus and β-glucan derived from Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun., 516, 245-250.
•  Nagaoka,  M.,  Shibata,  H.,  Kimura-Takagi,  I.,  Hashimoto,  S.,  Kimura,  K.,  Makino,  T.,  Aiyama,  R.,  Ueyama,  S., and  Yokokura,  T. (1999). Structural study of fucoidan from Cladosiphon okamuranus Tokida. Glycoconj. J., 16, 19-26.
•  Patankar,  M.S.,  Oehninger,  S.,  Barnett,  T.,  Williams,  R.L., and  Clark,  G.F. (1993). A revised structure for fucoidan may explain some of its biological activities. J. Biol. Chem., 268, 21770-21776.
•  Rani,  V.,  Shakila,  R.J.,  Jawahar,  P., and  Srinivasan,  A. (2017). Influence of species, geographic location, seasonal variation and extraction method on the fucoidan yield of the brown seaweeds of Gulf of Mannar, India. Indian J. Pharm. Sci., 79, 65-71.
•  Secouard,  S.,  Grisel,  M., and  Malhiac,  C. (2007). Flavour release study as a way to explain xanthan–galactomannan interactions. Food Hydrocoll., 21, 1237-1244.
•  Silvestri,  L.J.,  Hurst,  R.E.,  Simpson,  L., and  Settine,  J.M. (1982). Analysis of sulfate in complex carbohydrates. Anal. Biochem., 123, 303-309.
•  Sun,  Y.,  Gong,  G.,  Guo,  Y.,  Wang,  Z.,  Song,  S.,  Zhu,  B.,  Zhao,  L., and  Jiang,  J. (2018). Purification, structural features and immunostimulatory activity of novel polysaccharides from Caulerpa lentillifera. Int. J. Biol. Macromol., 108, 314-323.
•  Wang,  C.Y. and  Chen,  Y.C. (2016). Extraction and characterization of fucoidan from six brown macroalgae. J. Mar. Sci. Technol., 24, 319-328.
•  Wang,  X. and  Zhang,  L. (2009). Physicochemical properties and antitumor activities for sulfated derivatives of lentinan. Carbohydr. Res., 344, 2209-2216.
•  You,  S.,  Yang,  C.,  Lee,  H., and  Lee,  B. (2010). Molecular characteristics of partially hydrolyzed fucoidans from sporophyll of Undaria pinnatifida and their in vitro anticancer activity. Food Chem., 119, 554-559.