G Model


BIOMAC 4433 1–6

International Journal of Biological Macromolecules xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac







Structure of a shear-thickening polysaccharide extracted from the New Zealand black tree fern, Cyathea medullaris May S.M. Wee a , Lara Matia-Merino a , Susan M. Carnachan b , Ian M. Sims b , Kelvin K.T. Goh a,∗ a




Institute of Food Nutrition and Human Health, Massey University, Private Bag 11 222, Palmerston North 4442, New Zealand The Ferrier Research Institute, Victoria University Wellington, 69 Gracefield Road, Lower Hutt 5040, New Zealand


a r t i c l e

8 23

i n f o

a b s t r a c t


Article history: Received 13 March 2014 Received in revised form 18 June 2014 Accepted 18 June 2014 Available online xxx

10 11 12 13 14 15

Keywords: Glucuronomannan Structural analysis 18 Rheology 19 NMR 20 Black tree fern 21 22 Q4 Mamaku 16 17

A shear-thickening water-soluble polysaccharide was purified from mucilage extracted from the fronds of ¯ the New Zealand black tree fern (Cyathea medullaris or ‘mamaku’ in Maori) and its structure characterised. Constituent sugar analysis by three complementary methods, combined with linkage analysis (of carboxyl reduced samples) and 1 H and 13 C nuclear magnetic resonance spectroscopy (NMR) revealed a glucuronomannan comprising a backbone of 4-linked methylesterified glucopyranosyl uronic acid and 2-linked mannopyranosyl residues, branched at O-3 of 45% and at both O-3 and O-4 of 53% of the mannopyranosyl residues with side chains likely comprising terminal xylopyranosyl, terminal galactopyranosyl, nonmethylesterified terminal glucopyranosyl uronic acid and 3-linked glucopyranosyl uronic acid residues. The weight-average molecular weight of the purified polysaccharide was ∼1.9 × 106 Da as determined by size-exclusion chromatography coupled with multi-angle laser light scattering (SEC–MALLS). The distinctive rheological properties of this polysaccharide are discussed in relation to its structure. © 2014 Published by Elsevier B.V.

1. Introduction


The black tree fern (Cyathea medullaris), also known as mamaku ¯ in Maori, is one of the most common native flora found throughout damp lowland forests in New Zealand. Growing up to 20 m, it is one of the world’s tallest tree ferns. The pith at the centre of the trunk, the bases of the frond stems, as well as the uncurled part of the new shoots are edible [1]. Apart from being a food source for ¯ the Maori, mamaku is also traditionally used for medicinal purposes, either by consumption or external application [2]. When bruised, the wounded exposed stem exudes a brownish-red gum with a slimy consistency which has been used to treat swellings, bruises and wounds [1,3]. The starchy white pith within the fronds is also rich in translucent mucilage which is released when the pith is pulped. The mucilaginous matter has been extracted from the stem pith of mamaku fronds using hot water and shown to exhibit unique rheological behaviour, such as shear-thickening, a high extensional viscosity and the rod-climbing effect [4]. Such rheological behaviour is rare in naturally-occurring polysaccharides. In addition, the origin of their unique rheological properties is poorly

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43


∗ Corresponding author. Tel.: +64 06 3569099x. . .; fax: +64 06 3505657. E-mail addresses: [email protected], [email protected] (K.K.T. Goh).

understood because of the structural complexity of the biopolymer fraction. In contrast, the relationship between rheological properties (shear-thickening in particular) of synthetic polymers and their chemical structure is generally well understood. Telechelic polymers (polymers with associative groups at the chain ends) such as hydrophobically modified ethoxylated urethane exhibit shearthickening properties due to the terminal hydrophobic groups, forming transient associations under shear [5]. The dipole-dipole interactions between ionic groups in ionomers (polymers with ionic groups at wide intervals along a non-polar backbone) in low polarity solvents also lead to shear-thickening behaviour [6]. Clearly, the position of these groups within the polymer chain drives the interactions that occur during shear and hence the resultant rheology. The mechanism behind shear-thickening in these synthetic polymers can be systematically deduced because their molecular structures are well-defined. Chemical analysis of mamaku mucilage showed that it is comprised of ∼10% w/w non-starch polysaccharide. The polysaccharide fraction is rich in uronic acid and consists of the neutral sugars galactose, xylose and arabinose [4]. Other constituents present in the mucilage are mainly low molecular weight simple sugars (∼44% w/w) and minerals (∼16% w/w). The non-starch polysaccharide component of the mamaku mucilage is believed to be the macromolecule responsible for the distinctive rheological properties, although any attempts to understand the molecular origin of

http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032 0141-8130/© 2014 Published by Elsevier B.V.

Please cite this article in press as: M.S.M. Wee, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032

44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68

G Model BIOMAC 4433 1–6

ARTICLE IN PRESS M.S.M. Wee et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx



its rheology need to be accompanied by fundamental knowledge on the primary structure of the polysaccharide. In this paper, as part of our on-going studies of the rheological properties of mamaku mucilage, we have purified mamaku polysaccharide and analysed its composition and structure. Examination of the rheological properties of the purified mamaku polysaccharide suggests that the non-starch polysaccharide fraction is responsible for the shear-thickening properties of mamaku mucilage reported previously. The rheological properties are discussed in relation to the structure of the purified mamaku polysaccharide.


2. Experimental


2.1. Extraction and purification of mamaku mucilage

69 70 71 72 73 74 75 76 77 78

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101



The mamaku fronds (∼40 kg) were harvested from the grounds of Massey University, Palmerston North in June 2011 and the mucilage extracted and freeze dried (1% w/w yield) as described by Goh et. al. [4]. The native mucilage was reconstituted with distilled water (3% w/w solution) and centrifuged (250,000 × g, 1 h, 20 ◦ C) to remove large aggregates and fine insoluble particulates. The soluble material was treated with thermostable ␣-amylase from Bacillus licheniformis (EC, Megazyme, Bray, Ireland), 35 units at 100 ◦ C for 6 min and amyloglucosidase from Aspergillus niger (EC, Megazyme), 330 units at 50 ◦ C for 30 min to hydrolyse starch and then with Sevag reagent (chloroform/1-butanol, 1/4 v/v) to remove protein [7]. The de-proteinised extract was treated overnight with 80% v/v ethanol. The resulting precipitate was collected by centrifugation and then rehydrated overnight in Milli-Q water (50 g). Residual ethanol was removed by rotary evaporation and the water-soluble material was dialysed (12–14 kDa molecular weight cut-off, SpectraPor) then freeze-dried to give purified mamaku polysaccharide (13% w/w yield). The yield is expressed as: yield(%) =

amount of purified freeze − dried material × 100% amount of native freeze − dried material

2.2. General analyses


Total uronic acid content was measured colorimetrically by the m-hydroxydiphenyl method [8] using glucuronic acid (0–12.5 ␮g) as the standard. Moisture, protein, ash and starch contents were analysed by an accredited chemical laboratory (Nutritional Laboratory, Institute of Food, Nutrition & Human Health, Massey University) using the convection oven 105 ◦ C (AOAC 930.15, 925.10), Leco total combustion (AOAC 968.06), furnace 550 ◦ C (AOAC 942.05) and ␣-amylase (AOAC 996.11) methods, respectively.


2.3. Constituent sugar analysis

104 105 106 107 108 109 110 111

114 115 116 117 118 119 120 121 122 123 124 125

The constituent monosaccharide composition of the purified mamaku polysaccharide was determined by three complementary methods following acid hydrolysis: Method 1: Samples (1 mg, in duplicate) were hydrolysed with aqueous TFA (2 M, 500 ␮L, containing 0.406 mg/mL myo-inositol, 120 ◦ C, 1 h), filtered (0.2 ␮m, hydrophilic PTFE) hydrolysates, dried and neutralised by addition of 2 M NH4 OH (200 ␮L). The neutralised hydrolysates were then reduced, acetylated and analysed by GC–MS as described by Carnachan et al. [9]. Weight calibration constants were determined from a seven sugar standard mix (derivatised at the same time as the samples) following the TAPPI standard method T249 cm-85 [10]. Myo-inositol was used as an

internal standard. Monosaccharide yields were based on the mean values of duplicate samples and were expressed as weight percent anhydro-sugar because this is the form of sugar present in a polysaccharide. Method 2: Samples (1 mg) were hydrolysed with methanolic acid (3 N, 500 ␮L, 80 ◦ C overnight) and aqueous TFA (2 M, 500 ␮L, containing 0.86 mg/mL d-allose as an internal standard, 120 ◦ C, 1 h). The hydrolysates were reduced and acetylated to form alditol acetates and analysed as above. Method 3: The monosaccharides resulting from hydrolysis with methanolic acid and aqueous TFA from method 2 were analysed by high-performance anion-exchange chromatography (HPAEC). Samples (20 ␮L) dissolved in distilled water (0.5 mg mL−1 ) were separated at 30 ◦ C on a CarboPac PA-1 (4 × 250 mm) column equilibrated in 25 mM NaOH and eluted with simultaneous gradients of NaOH (25–10 mM from 0 to 10 min, then 10–100 mM from 10 to 30 min and held to 55 min) and sodium acetate (0–500 mM NaOH from 30 to 55 min) at a flow rate of 1 mL min−1 and monitored by pulsed amperometric detection, using the Dionex standard carbohydrate waveform. 2.4. Glycosyl linkage analysis Prior to glycosyl linkage analysis, uronic acid and methylesterified uronic acid residues were reduced using a two-step carboxyl reduction method as described by Sims and Bacic [11]. Briefly, purified mamaku polysaccharide (10 mg) was dissolved in 500 mM imidazole–HCl (10 mL, pH 8.0), cooled to 4 ◦ C and reduced with NaBD4 . Excess NaBD4 was destroyed by addition of acetic acid and the samples were dialysed (2 kDa molecular weight cut-off) for 24 h against distilled water and freeze-dried. The free uronic acids were then activated by addition of 1-cyclohexyl-3-(2-morpholinoethyl)-carbodiimidemetho-p-toluenesulfonate (400 ␮L, 500 mg mL−1 ) and reduced overnight with either NaBD4 or NaBH4 . The carboxyl-reduced samples were dialysed against distilled water and freeze-dried. Carboxyl-reduced samples (1 mg, in duplicate) were methylated with the method of Ciucanu and Kerek [12] except that samples were dispersed in DMSO (200 ␮L). After extraction into chloroform, the methylated samples were hydrolysed with 2.5 M TFA, reduced and acetylated before analysis by GC–MS as described above. 2.5. NMR spectroscopy Purified mamaku polysaccharide was exchanged with deuterium by freeze-drying with D2 O (99.9 atom%) three times. Samples were dissolved in D2 O and 1 H and 13 C (both 1 H coupled and decoupled) spectra were recorded on a Bruker Avance DPX500 spectrometer at 90 ◦ C. The 1 H and 13 C chemical shifts were measured relative to an internal standard of Me2 SO (1 H, 2.70 ppm; 13 C, 39.5 ppm) [13]. Assignments were made from double quantum filtered (DQF) COSY, heteronuclear multiple quantum coherence (HMQC) COSY and HMQC TOCSY experiments and by comparing the spectra with published data. 2.6. Size-exclusion chromatography–multi-angle laser light scattering (SEC–MALLS) The molecular weight of the purified mamaku polysaccharide was determined using size-exclusion chromatography coupled with multi-angle laser light scattering (SEC–MALLS). Samples (5 mg mL−1 in 0.1 M NaNO3 ) were centrifuged (14,000 × g, 10 min) before injection (100 ␮L) and eluted with 0.1 M NaNO3 (0.7 mL min−1 , 60 ◦ C) from two columns (TSK-Gel G5000PWXL and G4000PWXL , 300 × 7.8 mm, Tosoh Corp., Tokyo, Japan) connected in series using a Waters 2690 Alliance separations module. The eluted

Please cite this article in press as: M.S.M. Wee, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032

126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145


147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164


166 167 168 169 170 171 172 173 174 175

176 177

178 179 180 181 182 183 184 185

G Model BIOMAC 4433 1–6

ARTICLE IN PRESS M.S.M. Wee et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx


material was detected using a Waters 490E variable wavelength detector (280 nm), a DAWN-EOS multi-angle laser light scattering detector with a laser at 690 nm (Wyatt Technology Corp., Santa Barbara, CA) and a Waters 2410 refractive index monitor. The data for molecular weight determination was analysed using ASTRA software (v4.73.04, Wyatt Technology Corp.) selected based on the elution volume between 9.4 and 12.2 mL of the chromatogram. An incremental refractive index (dn/dc) of 0.141 mL g−1 determined experimentally was used in the analysis.


2.7. Rheological measurements

186 187 188 189 190 191 192 193

196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215

The viscosity curves of the purified mamaku polysaccharide at different concentrations were measured using a Paar Physica rheometer MCR 301 (Anton-Paar, Graz, Austria) in controlled shear rate (CSR) mode at 20.0 ◦ C with the cone and plate geometry (CP 40-4 and P-PTD200/56, gap = 0.049 ␮m). The temperature of the rheological measurement was controlled using a Peltier system to an accuracy of ±0.1 ◦ C. The samples were allowed to rest for 10 min after loading into the geometry prior to starting the test. This was done to reduce any effects of shear history and at the same time for temperature equilibration. To obtain the plot of concentration dependence of zero shear rate viscosity, zero-shear viscosities of dilute mamaku solutions (0.01–0.1% w/w) were measured using an Ubbelohde capillary viscometer (viscometer no. 100, K59, Cannon Instrument Co., U.S.A.). The zero shear rate viscosities in the semi-dilute regime (0.6–1.4% w/w) were estimated by fitting (using Solver, Microsoft Excel, method of least squares) the simplified 1−n Cross equation ( = 0 /(1 + () ˙ )) to the complex viscosity as a function of angular frequency (0.1 to 10 rad s−1 ) obtained from dynamic oscillatory shear measurements at 1% strain amplitude viscoelastic region) as described in an earlier study [4].


3. Results and discussion


3.1. Purification and composition of mamaku polysaccharide

218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245

The total sugar content (38.4% w/w) of the native mucilage fraction (dry weight basis) was close to that reported previously (44.3%) [4]. The uronic acid content was 5.7% w/w determined by colorimetric analysis. Constituent neutral sugar analysis showed 30.3% w/w glucose, with small amounts of galactose (1.5% w/w), mannose (0.4% w/w), xylose (0.3% w/w) and arabinose (0.2% w/w). Proton NMR spectroscopy (data not shown) showed a series of sharp signals from 5.3 to 3.2 ppm, consistent with the presence of low molecular sugars, probably sucrose. It is noted that the high amount of glucose was partly due to the presence of starch (10.3% w/w). In addition, the amount of protein and ash were 2.9% w/w and 17.1% w/w, respectively. In order to obtain a purified polysaccharide fraction, the mucilage was dialysed (to remove the smaller molecular weight components) and then subjected to enzymatic digestion of starch with amylase/amyloglucosidase and treatment with Sevag reagent to remove protein. This was then followed by precipitation with ethanol and dialysis as described earlier. Compositional analysis showed that the purified mamaku polysaccharide contained 6.3% w/w protein, 13.7% w/w ash, and 3.3% w/w moisture. Starch was not detected in the purified mucilage. Based on these values, the nonstarch polysaccharide was estimated to be approximately 77% w/w. The constituent sugar composition of the purified mamaku polysaccharide determined by the three complementary methods are in fairly good agreement (Table 1). Analysis by HPAEC, following hydrolysis with methanolic HCl/TFA, gave mostly glucuronic acid, galactose, mannose and xylose, with small amounts of other sugars. The total sugar content determined by HPAEC was 88.9% w/w,


Table 1 Constituent sugar composition of purified mamaku polysaccharide determined three methods. Weight %a

Sugar Me-HCl/TFA

Rhamnose Fucose Arabinose Xylose Mannose Galactose Glucose Glucuronic acid Galacturonic acid Total neutral sugar Uronic acid Total sugar a b c





1.2 0.8 1.8 5.1 12.5 15.4 0.1 50.0 1.8 37.1 51.8 88.9

0.5 0.3 0.8 3.0 12.5 15.7 3.8 n.db n.db 36.6 53.0c 89.6

0.8 0.4 0.9 3.8 6.4 13.0 5.1 n.db n.db 30.4 53.0c 83.4

Values are the averages of duplicate analyses. n.d., Not determined. Determined colorimetrically.

with 37.1% w/w neutral sugars. Analysis by GC, following hydrolysis with either methanolic HCl/TFA, or TFA alone, gave neutral sugar contents of 36.6% or 30.4%, respectively. Both analyses showed mostly galactose, mannose, glucose and xylose, although hydrolysis with TFA alone gave lower amounts of mannose and galactose than with methanolic HCl/TFA. The lower yield of these sugars may be due to their linkage to uronic acids and thus remained more resistant to acid hydrolysis [14]. The higher glucose content observed when hydrolysates were analysed by GC was probably due to the formation of glucuronolactone, which was then reduced to glucose [15]. In terms of the uronic acid content, the colorimetric analysis gave 53% w/w which is in good agreement with that determined by HPAEC (51.8%). The total sugar content, including the uronic acid determined colorimetrically, was 89.6% (methanolic HCl and TFA) and 83.4% (TFA only). 3.2. Structural analyses of purified mamaku polysaccharide As the constituent sugar analyses of the purified mamaku polysaccharide showed a high uronic acid content and preliminary NMR experiments indicated that a proportion of these uronic acids were methylesterified, these residues were reduced to their respective 6,6 -dideuterio labelled neutral sugars prior to linkage analysis (Table 2). The analysis of the purified mamaku polysaccharide showed high proportions of 4-linked methylesterified glucopyranosyl uronic acid (4-GlcpA), 2,3- and 2,3,4-linked mannopyranosyl (2,3- and 2,3,4-Manp) residues, consistent with the presence of a glucuronomannan comprising of a backbone of 4-GlcpA and 2Manp, branched at O-3 of 45% and at both O-3 and O-4 of 53% of the Manp residues [16,17]. The other major linkages detected were terminal xylopyranosyl (T-Xylp), terminal galactopyranosyl (T-Galp) and non-esterified 3-GlcpA residues. Other linkages observed (3-, 6- and 3,6-Galp) were typical of type II arabinogalactans that are typically terminated variously by rhamnopyranosyl (Rhap), arabinofuranosyl (Araf), arabinopyranosyl (Arap) and Galp residues [13]. The 13 C NMR spectrum of the purified mamaku polysaccharide showed C-1 signals from 98.9 to 104.0 ppm, together with a weak C-1 signal at 109.8 ppm (Fig. 1A) and 1 H NMR showed H-1 signals from 4.47 to 5.37 ppm (Fig. 1B). The DEPT-135 experiment showed three methylene signals at 60.3, 62.1 and 65.9 ppm, with cross-peaks in the HSQC spectrum at 3.94/3.83, 3.77/3.64 and 4.00/3.33 ppm, respectively. The spectra were assigned on the basis of HSQC, COSY and DEPT-135 experiments and by comparison with published spectra of similar molecules [15,18–20].

Please cite this article in press as: M.S.M. Wee, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032

246 247 248 249 250 251 252 253 254 255 256 257 258 259 260


262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287

G Model


BIOMAC 4433 1–6

M.S.M. Wee et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx


Table 2 Glycosyl linkage composition of carboxyl-reduced mamaku polysaccharide. Sugar Rhap Fucp Arap Araf Xylp


Glcp Manp GlcpA


Deduced linkagea Terminal Terminal 2Terminal 2Terminal Terminal 24Terminal 363,642,32,3,4Terminal 34-


Relative amount (mol%)b 2.9 0.8 0.7 2.3 1.0 1.0 8.9 1.8 2.8 14.8 0.6 0.9 1.0 1.1 9.2 10.9 2.3 5.9 27.9 (100) 3.2

C1 β-GlcAp

C1 β-Galp


C6 C1 α-Manp


C3 α-Manp





a Terminal Rhap deduced from 1,5-di-O-acetyl-6-deoxy-2,3,4-tetra-Omethylrhamnitol, etc. b Values are the averages of duplicate determinations (value in parentheses is the degree of methylesterification). c Comprises linkages present ≤0.5 mol%.

Signals at C-1/H-1 98.9–99.5/5.37 were assigned to ␣-dManp residues and a C-3/H-3 signal at 83.4/3.75 was consistent with O-3 substitution of these residues as observed in the gly290 cosyl linkage analysis (Table 2). Similarly, signals at C-1/H-1 291 102.5–102.8/4.54–4.59 were assigned to ␤-d-GlcpA residues. The 292 presence of a carbonyl signal in the 13 C NMR spectrum at 171.0 ppm 293 and absence of a signal at 175 ppm (data not shown), together 294 with an intense O-methyl signal (13 C 54.2 ppm, 1 H 3.88 ppm) 295 indicated that the GlcpA residues were completely esterified and 296 confirmed the glycosyl linkage analysis that showed that nearly all 297 of the 4-GlcpA residues were methylesterified. Signals at C-1/H298 1 103.9/4.70 were assigned to ␤-d-Galp residues and the weak 299 C-1 signal (109.8 ppm) in the 13 C NMR spectrum was assigned 300 to an ␣- l-Araf residue [13]. Other anomeric signals were not 301 assigned, but probably represent other residues (e.g. Xylp and 3302 GlcpA) observed in the glycosyl linkage analysis (Table 2). The 303 methylene signal at 65.9 ppm with H-5 cross peaks at 4.00 and 304 3.33 ppm was assigned to T-Xylp, which was the major xylose link305 age observed [21]. Similarly, the methylene signals at 60.3 ppm 306 (H-1 3.94/3.83 ppm) and 62.1 (H-1 3.77/3.64) were assigned to ␣307 d-Manp and ␤-d-Galp residues, respectively. The presence of small 308 proportions of Rhap and fucopyranosyl (Fucp) residues observed 309 in the linkage analysis was supported by the presence of two C310 methyl signals (13 C 17.3 and 16.1 ppm, 1 H 1.30 and 1.26 ppm). A 311 possible structure of the mamaku polysaccharide is illustrated in 312 Fig. 4. 313 Glucuronomannans, comprising of a repeating backbone of -4)314 Q5 ␤-d-GlcpA-(1 → 2)-␣-d-Manp-(1 → - are commonly found as gum 315 and mucilage polysaccharides. For example, glucuronomannans 316 are components of a number of exudate gums, including gum Ghatti 317 [22] and Anogeissus leiocarpus gum [23], as well as gum exudates 318 from the Hakea species [24] and from Vochysia spp. trees [15,19,20]. 319 They have also been extracted from the stem pith of Actinidia deli320 ciosa (kiwifruit) [16,17,25], fronds of Asplenium australasicum fern 321 [26], the mucin of Drosera binata [27] and the fruit of Auricularia 322 auricala [28]. These polysaccharides usually have heterogeneous 323 side-chains comprising variously of Araf, Arap, Galp, Fucp and Xylp 324 residues attached to O-3 of the backbone residues. Thus, it is prob325 able that the purified mamaku polysaccharide comprises a similar 326





288 289

H1 β-GlcAp

H1 α-Manp

H1 β-Galp



Fig. 1. Selected regions of the mamaku polysaccharide.







C-NMR (A) and 1 H-NMR (B) spectra of purified

Fig. 2. Molecular weight analysis by size-exclusion coupled with multi-angle laser light scattering (SEC–MALLS) of purified mamaku polysaccharide.

Please cite this article in press as: M.S.M. Wee, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032

G Model BIOMAC 4433 1–6

ARTICLE IN PRESS M.S.M. Wee et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx


R 4 1Man2 à 1GlcA(Me)4 à 1Man2 à 1GlcA(Me)4 3 3 R


Fig. 4. Possible structure of mamaku polysaccharide; R = T-Xylp, T-Galp or T-GlcpA and other more complex oligosaccharides containing sugars with linkages shown in Table 2.

smaller than that determined previously (3.2 × 106 Da, polydisper-


1/2 sity index of 1.08 and (rg2 )z of 144 nm) for the native mucilage [32].


The higher molecular weight value and Z-average RMS radius could be due to the presence of aggregates in the native mucilage. These contaminants were markedly reduced in the purified samples.

3.4. Rheological properties of the purified mamaku polysaccharide

Fig. 3. Viscosity curves of purified mamaku at concentrations of 0.2–1.4% (A); dashed lines represent viscosity of a 5% native mamaku solution before purification. Concentration dependence of zero-shear viscosity (B); unfilled symbols ( ) represent viscosities measured using an Ubbelohde dilution capillary viscometer; filled symbols () are zero-shear viscosities in the dilute regime estimated by fitting the cross equation to the complex viscosity curves.

327 328 329 330 331 332 333 334 335 336 337 338

339 340

341 342 343 344 345 346 347

repeating backbone structure, but differs from most of the reported structures in that all of the Manp residues bear one or two side chains, and no side chain is attached to GlcpA residues. The role of glucuronomannans in gums and mucilages may be related to their functionality in the plant. Uronic acids improve the solubility of polysaccharides and increase their water holding capacity due to stronger hydrogen bonds formed between the polar carboxyl groups and water [29]. This enables the polysaccharide to hydrate rapidly and form a protective layer on the plant or store reserves of water during droughts [30]. The slimy consistency of mucilages may also act as lubrication within the plant during water transport [31]. 3.3. Size-exclusion chromatography–multi-angle laser light scattering (SEC–MALLS) The chromatogram (Fig. 2) shows a single large light scattering (LS; 90◦ ) peak overlapped with an RI signal peak detected at elution volume range of ∼9–12 mL. The low UV absorbance in this peak suggested that this material is mostly polysaccharide. The weight-average molecular weight (Mw ) of the purified mamaku polysaccharide was determined to be 1.944 × 106 Da with a low polydispersity index (Mw /Mn ) of 1.034 and a Z-average RMS radius 1/2


(rg2 )z

of 94 nm based on the Zimm model. This molecular weight is

The viscosity of the purified mamaku polysaccharide was tested at various concentrations from 0.2 to 1.4% (Fig. 3A). Shearthickening occurred at a concentration above 0.3% and was accompanied by a large increase in viscosity. The onset of shearthickening (generally at shear rates of ∼10 s−1 ) was shifted to lower shear rates as the concentration increased. The same phenomenon was observed for the native solution, where shear thickening was seen only at concentrations above 4% [4]. The viscosity of a 5% native solution (dashed lines on Fig. 3A) corresponds approximately to a 0.5% purified solution and is consistent with the native mucilage containing about ∼10% non-starch polysaccharides [4]. The data suggested strongly that the glucuronomannan polymer, purified in the present study, is responsible for the shear-thickening behaviour of the native mucilage. Fig. 3B shows the zero-shear viscosities of the purified material in the dilute and semi-dilute regions. A power law equation (y = a × xb ) was fitted to the data for each concentration regime. The exponent of concentration dependence (b) in the dilute and semi-dilute region is 1.6 and 3.1, respectively, similar to most other random coil polysaccharides (dilute: C1.4 ; semi-dilute:C3.3 ) [33]. The coil overlap concentration (the intersection between the two fit lines) was estimated as 0.35% which coincided with the onset concentration for shearthickening. Shear-thickening is an uncommon rheological property among natural biopolymers. Many polysaccharides, including glucuronomannans such as gum Ghatti, exhibit typical shear-thinning behaviour [34]. However, many of the glucuronomannans have not been characterised for their rheological properties. The mucilage from Drosera binata (a carnivorous plant) which contains a glucuronomannan with a relatively simple structure displays complex flow mechanisms (e.g. extensional viscoelasticity, capillary thinning) in order to trap its prey [27,35]. Rheological properties of polymers are largely dependent on the intra- or intermolecular interactions between chains (polymer–polymer) and solvent (polymer–water). Hydrophobic interactions may arise from the methylesterified groups ( COOCH3 ) on the 4-GlcpA residues [36]. Morris et al. [37] found that in pectin, methylesterification of galacturonic acid residues does not significantly affect molecular weight. However it does alter the hydrodynamic properties and chain conformation. Hence, chain stiffness (based on intrinsic viscosity) is decreased with increasing degree of esterification. The non-methylesterified terminal and 3-GlcpA residues could interact via electrostatic interactions (with ions) and hydrogen bonding. In addition, the hydroxyl groups of the sugars along the backbone chain could also

Please cite this article in press as: M.S.M. Wee, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032

351 352 353

354 355

356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400

G Model BIOMAC 4433 1–6

ARTICLE IN PRESS M.S.M. Wee et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx



participate in hydrogen bonding to the glucuronic acid residues and other sugar molecules. The resultant rheological behaviour of the purified mamaku polysaccharide would be a balance of these intra and inter molecular chain–chain and chain–solvent interactions. Shear-thickening at intermediate shear rates followed by shearthinning at high shear rates is a common phenomenon among associative polymers, where the associations between attractive groups lead to the formation of reversible physical bonds [38]. The origin of shear-thickening in the mamaku polysaccharide is likely to be similar to that of associative polymers [4,41]. Therefore, a possible hypothesis for the shear-thickening is that the non-methylesterified GlcpA groups on the side chains act as ‘sticker groups’ which associate via hydrogen bonding or electrostatic complex formation with ions during shear [39]. Such intermolecular interaction would give rise to an increase in viscosity at certain shear. Since the charged groups are most likely located on the side-chains and not in the backbone, the application of shear would uncoil the polymer chain and expose these groups for intermolecular interaction to occur. Hydrophobic aggregation between methylesterified groups on the backbone could also contribute to shear-thickening as in the case of high molecular weight poly(ethylene oxide) [40]. However, it is not likely to be the dominant interaction for shear-thickening in this case as an increase in temperature (5–50 ◦ C) did not promote shearthickening [41,42]. Based on the current findings, further work on the structure–function relationship of the purified mamaku polysaccharide is underway to elucidate the mechanism behind shear-thickening.


4. Conclusions

401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428


Structural characterisation of the purified mamaku polysaccharide molecule indicated that it is a glucuronomannan backbone (methylesterified 4-GlcpA with 2,3- and 2,3,4-linked Manp) with branched sugar side-chains of galactose, arabinose, nonmethylesterified glucuronic acid and other simple sugars at the O-3 and O-4 of the mannose residues. The specific positioning of the non-methylesterified glucuronic acid in the side-chains and methylesterified uronic acid residues along the backbone may be the cause for shear-thickening as a result of coordinated electrostatic, hydrogen and hydrophobic interactions between the polymer molecules.



431 432 433 434 435 436 437 438 439 440

This research was supported by the New Zealand Ministry of Business, Innovation and the Environment and Massey University Doctoral Scholarship for May Wee. Ian Sims and Susan Carnachan 445 Q7 acknowledge the Ministry of Science and Innovation, New Zealand 446 for financial support. We are grateful to Herbert Wong, Callaghan 447 Innovation, for NMR spectroscopy services. Thanks also to Felicity 448 Jackson and Leiza Turnbull (Nutrition Laboratory, Massey Univer449 sity) for carrying out the compositional analyses. 443

Q6 444



[1] A. Crowe, A Field Guide to the Native Edible Plants of New Zealand, Godwit, 451 Birkenhead, Auckland, 1952. 452 [2] S.G. Brooker, R.C. Cambie, R.C. Cooper, New Zealand Medicinal Plants, Reeds 453 Books, Auckland, 1987. 454 [3] T. Foster, Plant Heritage New Zealand: te whakapapa o nga rakua: Interpreting 455 the Species Features of Native Plants, Raupo, North Shore, 2008. 456 [4] K.K.T. Goh, L. Matia-Merino, C.E. Hall, P.J. Moughan, H. Singh, Biomacro457 molecules 8 (2007) 3414–3421. 458 [5] K.C. Tam, R.D. Jenkins, M.A. Winnik, D.R. Bassett, Macromolecules 31 (1998) 459 4149–4159. 460 [6] S. Nomula, S., Ma, S.L. Cooper, in, 2000, pp. 127-142. Q8 461 [7] M.G. Sevag, Biochem. Z. 273 (1934) 419–429. 462 [8] N. Blumenkrantz, G. Asboe Hansen, Anal. Biochem. 54 (1973) 484–489. 463 [9] S.M. Carnachan, T.J. Bootten, S. Mishra, J.A. Monro, I.M. Sims, Food Chem. 133 464 (2012) 132–139. 465 [10] TAPPI, TAPPI Standards Method T249 cm-85, Carbohydrate Composition of 466 Extractive-free Wood and Wood Pulp by Gas–Liquid Chromatography, TAPPI, Q9 467 1985. 468 [11] I.M. Sims, A. Bacic, Phytochemistry 38 (1995) 1397–1405. 469 [12] I. Ciucanu, F. Kerek, Carbohydr. Res. 131 (1984) 209–217. 470 [13] I.M. Sims, R.H. Furneaux, Carbohydr. Polym. 52 (2003) 423–431. 471 [14] L.D. Melton, B.G. Smith, Isolation of plant cell walls and fractionation of cell 472 wall polysaccharides, in: Current Protocols in Food Analytical Chemistry, John 473 Wiley & Sons, Inc., 2001. 474 [15] R. Wagner, F.F. Simas, G.L. Sassaki, M. Iacomini, M.A. da Silva, P.A.J. Gorin, Car475 bohydr. Polym. 72 (2008) 382–389. 476 [16] R.J. Redgwell, M.A. O’Neill, R.R. Selvendran, K.J. Parsley, Carbohydr. Res. 153 477 (1986) 97–106. 478 [17] R.J. Redgwell, M.A. O’Neill, R.R. Selvendran, K.J. Parsley, Carbohydr. Res. 153 479 (1986) 107–118. 480 [18] Y. Honda, H. Inaoka, A. Takei, Y. Sugimura, K. Otsuji, Phytochemistry 41 (1996) 481 1517–1521. 482 [19] R. Wagner, F.F. Simas, G.C.Z. Pereira, A. Angeli, J.O. Brito, S.M. Woranovicz483 Barreira, C.L. Delgobo, G.L. Sassaki, M. Iacomini, P.A.J. Gorin, Carbohydr. Polym. 484 485 69 (2007) 512–521. 486 [20] R. Wagner, S.M. Woranovicz-Barreira, M. Iacomini, C.L. Delgobo, N.M. Pimentel, 487 P.A.J. Gorin, Carbohydr. Polym. 57 (2004) 269–275. 488 [21] I.M. Sims, R. Newman, Carbohydr. Polym. 63 (2006) 379–384. [22] G.O. Aspinall, E.L. Hirst, A. Wickstrøm, J. Chem. Soc. (1955) 1160–1165 Q10 489 490 (Resumed). 491 [23] G.O. Aspinall, J.J. Carlyle, J.M. McNab, A. Rudowski, J. Chem. Soc. C: Org. (1969) 492 840–845. 493 [24] P.F.K. Eagles, A.M. Stephen, S.C. Churms, Phytochemistry 34 (1993) 709–713. 494 [25] R.J. Redgwell, Phytochemistry 22 (1983) 951–956. 495 [26] L.S. Lai, H.Y. Liang, Food Hydrocolloids 26 (2012) 344–349. 496 [27] D.C. Gowda, G. Reuter, R. Schauer, Phytochemistry 21 (1982) 2297–2300. 497 [28] Z. Li, D. Jiang, H. Bi, D. Liu, S. Jang, Y. Zhou, Nat. Prod. Commun. 7 (2012) 498 1507–1510. 499 [29] R.L. Whistler, Adv. Chem. Ser. (1973) 242–255. 500 [30] A.E. Clarke, R.L. Anderson, B.A. Stone, Phytochemistry 18 (1979) 521–540. 501 [31] U. Zimmermann, J.J. Zhu, F.C. Meinzer, G. Goldstein, H. Schneider, G. Zimmer502 mann, R. Benkert, F. Thurmer, P. Melcher, D. Webb, A. Haase, Bot. Acta 107 503 (1994) 218–229. 504 [32] K.K.T. Goh, L. Matia-Merino, D.N. Pinder, C. Saavedra, H. Singh, Food Hydrocol505 loids 25 (2011) 286–292. 506 [33] E.R. Morris, A.N. Cutler, S.B. Ross-Murphy, D.A. Rees, J. Price, Carbohydr. Polym. 507 1 (1981) 5–21. 508 [34] L. Kaur, J. Singh, H. Singh, J. Food Sci. 74 (2009) E328–E332. 509 [35] P. Erni, M. Varagnat, G.H. McKinley, in: A. Co, L.G. Leal, R.H. Colby, A.J. Giacomin 510 (Eds.), 15th International Congress on Rheology—The Society of Rheology 80th 511 Annual Meeting, Pts 1 and 2, 2008, pp. 579–581. 512 [36] D. Oakenfull, A. Scott, J. Food Sci. 49 (1984) 1093–1098. 513 [37] G.A. Morris, T.J. Foster, S.E. Harding, Food Hydrocolloids 14 (2000) 227–235. 514 [38] M. Rubinstein, A.V. Dobrynin, Trends Polym. Sci. 5 (1997) 181–186. 515 [39] A.L. Kjoniksen, M. Hiorth, B. Nystrom, Eur. Polym. J. 41 (2005) 761–770. 516 [40] D. Rivero, L.M. Gouveia, A.J. Mueller, A.E. Saez, Rheol. Acta 51 (2012) 13–20. 517 [41] L. Matia-Merino, K.K.T. Goh, H. Singh, Carbohydr. Polym. 87 (2012) 131–138. 518 [42] D. Oakenfull, D.E. Fenwick, Aust. J. Chem. 30 (1977) 741–752.

Please cite this article in press as: M.S.M. Wee, et al., Int. J. Biol. Macromol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.06.032

Structure of a shear-thickening polysaccharide extracted from the New Zealand black tree fern, Cyathea medullaris.

A shear-thickening water-soluble polysaccharide was purified from mucilage extracted from the fronds of the New Zealand black tree fern (Cyathea medul...
814KB Sizes 0 Downloads 5 Views