Structural elucidation of rhamnogalacturonans from flaxseed hulls

Carbohydrate Research 362 (2012) 47–55

Contents lists available at SciVerse ScienceDirect

Carbohydrate Research journal homepage: www.elsevier.com/locate/carres

Structural elucidation of rhamnogalacturonans from flaxseed hulls Ke-Ying Qian a,b, Steve W. Cui b,⇑, John Nikiforuk c, H. Douglas Goff a a b c

Department of Food Science, University of Guelph, Guelph, ON, Canada N1G 2W1 Guelph Food Research Centre, Agriculture and Agri-Food Canada, 93 Stone Rd. W., Guelph, ON, Canada N1G 5C9 Eastern Cereal and Oilseed Research Centre, Agriculture and Agri-Food Canada, 960 Carling Avenue, Ottawa, Canada K1A 0C6

a r t i c l e

i n f o

Article history: Received 25 June 2012 Received in revised form 10 August 2012 Accepted 11 August 2012 Available online 19 August 2012 Keywords: Flaxseed gum Acidic fraction Rhamnogalacturonan Methylation analysis 1D and 2D NMR

a b s t r a c t The structure of acidic fraction gum (AFG) from flaxseed hulls was elucidated by methylation analysis and 1D/2D NMR spectroscopy. This acidic fraction was separated from water-soluble flaxseed gum using anion-exchange chromatography. AFG consisted of a rhamnogalacturonan-I (RG-I) backbone that features diglycosyl repeating units, ?2)-a-L-Rhap-(1?4)-a-D-GalpA-(1?. Rhamnosyl residues (38.2%) were the most abundant neutral sugar component. It was present mainly as unbranched (16.5%) and branched (19.5%) ?2)-a-L-Rhap-(1? at O-3. Most of its branches were terminated by monosaccharides, a/b-DGalp-(1? (19.6%), a-L-Fucp-(1? (4.5%) or b-D-Xylp-(1? (3.1%). However, when this branching site was occasionally appended with ?4)-a-D-GalpA-(1? or ?2)-a-L-Rhap-(1?, side chains may consist of rhamnogalacturonan-I (RG-I), homorhamnan (HR) or a mixture of both. AFG was highly branched as indicated by its high degree of branching (0.55). A possible structure of AFG was proposed:

(HR, RG-I, and HG refer to homorhamnan, rhamnogalacturonan-I, and homogalacturonan, respectively. The locations of HR, RG-I, and HG are interchangeable; (m+n)/(n+i)1.5. The substitution rate of R1 is 54%. R1 is mostly monosaccharide (a/b-D-Galp-(1?, a-L-Fucp-(1? or b-D-Xylp-(1?). R1 may also occasionally be a longer side chain with more than two residues beginning with ?4)-a-GalpA-(1? or ?2)-aL-Rhap-(1?, wherein the side-chain structure may be similar to part of the main chain.) Ó 2012 Published by Elsevier Ltd.

1. Introduction In our previous study, soluble flaxseed gum extracted from flaxseed hulls was separated into a neutral and an acidic fraction using anion-exchange chromatography.1 The acidic fraction gum (AFG) was favored for its low viscous properties as a potential dietary fiber fortifier to be included into food systems in a large amount without over-texturization. Methylation analysis in previous studies found that linear ?4)-a-D-GalpA-(1?, and unbranched or branched ?2)-a-L-Rhap-(1? at O-3 were the main linkages in the AFG from flaxseed.2,3 Rhamnogalacturonan-I (RG-I) backbone features repeating units of alternatively distributed ?2)-a-LRhap-(1? and ?4)-a-D-GalpA-(1?. RG-I, RG II, and homogalactu⇑ Corresponding author. Tel.: +1 519 780 8028; fax: +1 519 829 2600. E-mail addresses: [email protected] (K.-Y. Qian), [email protected] (S.W. Cui), [email protected] (J. Nikiforuk), [email protected] (H.D. Goff). 0008-6215/$ - see front matter Ó 2012 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.carres.2012.08.005

ronan (HG) comprise the three main structural elements of pectin.4,5 In general, pectins encompass hetero-polysaccharides containing at least 65% of galacturonic acid-based units according to the FAO and EU stipulation.6 Either arabinose or galactose is the second-most abundant neutral sugar component in pectins with no more than 4% of rhamnose and fucose typically.7 Although AFG from flaxseed was referred to as pectic polysaccharides, RG-I,3 this fraction differed from pectins in its higher rhamnose content and lower galacturonic acid content. Interest in polysaccharides as potential sources for anti-tumor agents has increased in recent decades due to their fewer side effects. Modified pectin was reported to show (in vitro) anti-tumor bioactivities, probably due to the existence of galactans from side chains in the hairy region.8,9 Revealing the fine structure of polysaccharides assists in further tracing their potential functionalities and/or bioactivities. The detailed structure of AFG from flaxseed has not been reported. This study focused on the structure of

48

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

Table 1 Partially methylated aditol acetate (PMAA) sugar residue derivatives of the acidic fraction gum Deduced Residue 3

R R RT

?2,3)- L-Rhap-(1? ?2)- L-Rhap-(1?

R34/4

?2,3,4)/?2,4)- L-Rhap-(1? Rhamnose ?4)-D-GalpA-(1? ?2,4)- D-GalpA-(1? ?3,4)- D-GalpA-(1? Galacturonic Acid

GA GA2 GA3 GT G’ FT

L-Rhap-(1?

RT⁄

Mol %⁄⁄

PMAA

1.307 0.957 0.603

19.5 16.5 1.2

4-Me Rhap 3,4-Me2 Rhap 2,3,4-Me3 Rhap

1.379

0.9 38.1 22.9 1.1 0.8 24.8 19.6

Rhap-(OAC)5 /3-Me Rhap

1.543 1.900 1.764

D-Galp-(1?

1.117

?6)- D-Galp-(1? Galactose

1.703

L-Fucp-(1?

X X34 X2/3 A AT

2,3,4,6-Me4 Galp

0.754

1.2 20.8 4.5

2,3,4-Me3 Fucp

D-Xylp-(1?

0.784

4.5 3.1

2,3,4-Me3 Xylp

?4)- D-Xylp-(1? ?2,3,4)- D-Xylp-(1? ?2,4)/3,4)- D-Xylp-(1? Xylose ?3)- L-Araf-(1?

1.250 2.054 1.646

Fucose XT

2,3,6-Me3 GalpA 3,6-Me2 GalpA 2,6-Me2 GalpA

1.054 0.642

L-Araf-(1?

Arabinose ⁄

2.4 1.3 0.9 7.7 1.8 0.8

2,3,4-Me3 Galp

2,3,-Me2 Xylp Xylp-(OAC)5 3/2-Me Xylp 2,5-Me2 Araf 2,3,5-Me3 Araf

2.6 ⁄⁄

(RT : Retention time is relative to 2,3,4,6-Me4 Glc (14.591 min); Mol % : molar ratio of each sugar residue is based on the percentage of its peak area; Letter labeled by a superscript T refers to a terminal residue; The superscript numbers follow the letters indicate the branching sites in these residues; Data for sugar residues less than 0.6% are not shown; Linkages below 2% were not included in the proposed repeating unit for AFG in Fig. 7)

AFG from flaxseed hulls using methylation-GC–MS analysis and 1D/2D NMR spectroscopy, including homonuclear 1H/1H correlation spectroscopy (COSY, TOCSY) and nuclear overhauser effect spectroscopy (NOESY), heteronuclear 1H/13C multiple-quantum coherence spectroscopy (HMQC) and heteronuclear 1H/13C multiple bond correlation spectroscopy (HMBC). 2. Results and discussion 2.1. Methylation and GC–MS of partially methylated alditol acetate (PMAA) of the acidic fraction gum (AFG) Methylation-GC–MS analysis of the AFG showed that this fraction composed of four hexosyl residues (rhamnose (38.1%), galacturonic acid (24.8%), galactose (20.8%), and fucose (4.5%)) and two pentosyl residues (xylose (7.7%) and arabinose (2.6%)) as listed in Table 1. Four major linkage patterns constituted 78.5% of the total sugar residues and they are: ?2,3)-L-Rhap-(1? (19.5%), ?2)-LRhap-(1? (16.5%), ?4)-D-GalpA-(1? (22.9%), and D-Galp-(1? (19.6%). The molar ratios of total non-reducing terminal residues and total branching points were 29.2 and 25.8%, respectively. The degree of branching (DB) of AFG was equal to 0.55 as calculated according to the equation below10,11:

DB ¼ ðNT þ NB Þ=ðNT þ NB þ NL Þ ¼ ð29:2 þ 24:5Þ=ð29:2 þ 24:5 þ 44:8Þ ¼ 0:55

ð1Þ

where NT, NB, and NL are the molar percentage of the terminal, branched, and linear residues, respectively. The DB value of a linear chain equals 0, whereas that of a fully branched polymer is 1. The DB value (0.55) of AFG indicated it was highly branched. 2.2. 1D and 2D NMR analysis of AFG More than six peaks are shown in the anomeric region (4.3– 5.2 ppm) of the 1H NMR spectrum (Fig. 1a). Two peaks at 1.08 and 1.11 ppm arose from the proton resonances of methyl groups in fucose and rhamnose, respectively; their corresponding reso-

nances in the 13C spectrum (Fig. 1b) are at 16.2 and 17.6 ppm, respectively. The spin system for each sugar residue of AFG was assigned according to the COSY spectrum (Fig. 2), with assistant/confirmative information from the TOCSY (Fig. 3) and HMQC spectra (Fig. 4). The sequence of the linkages of sugar residues was inferred from the HMBC (Fig. 5) and NOESY spectra (Fig. 6). The assignment of each sugar residue and their relative sequences are discussed in the following sections before a possible structure of the repeating unit of AFG is proposed. 2.2.1. Rhamnose residues Rhamnosyl residues (38.2%) in AFG include four linkages as listed in decreasing order of predominance in Table 1: ?2,3)-a-LRhap-(1? (R3, 19.5%), ?2)-a-L-Rhap-(1? (R, 16.5%), and minor amount of T-a-Rhap-(1? (RT, 1.2%) and ?2,3,4)/?2,4)-a-L-Rhap(1? (R34/4, 0.9%). The a-configuration of all Rhap units was inferred from their common chemical shifts of anomeric carbon (99.6 ppm) and proton (5.11 ppm) as shown in the anomeric region of HMQC spectrum (Fig. 4a). The spin units for ?2)-a-L-Rhap-(1? (R) and ?2,3)-a-L-Rhap(1? (R3) are fully assigned in COSY, TOCSY, and HMQC spectra (Figs. 2–4 & Table 2). The H-2 (4.05 ppm), H-3 (3.83/3.85 ppm), and H-4 (3.50 ppm) chemical shifts of the R3 shifted downfield by 0.1–0.2 ppm, due to glycosylation at O-3 position.12,13 The substitution at O-3 positions of rhamnosyl residues instead of the typical O-4 positions, as found in pectin and soluble soybean polysaccharides, was also reported based on methylation analysis of flaxseed mucilage in earlier studies.3,14 Both cross-peaks at 3.83/72.2 and 3.85/77.9 ppm in HMQC (Fig. 4) were tentatively assigned to H-3/C-3 correlation of R3. This splitting may be due to the substitution by different sugar units, according to HMBC (Fig. 5) and NOESY (Fig. 6) spectra. The chemical shifts of glycosylated carbon (C-3) moved downfield by 1.7 and 7.5 ppm (Table 2), respectively. The former value was far smaller than the value of 7–10 ppm reported for rhamnosyl residues branched at O-3 position15–18; other than this, the provisional assignment was in good agreement with literature data.17,19–21

49

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

a R3/R /αGT 5.11

FT 5.02

GA 4.89

T βG T

/X

X4

4.44

4.31

R3/R

FT

1.11 (H-6)

1.08 (H-6)

GA 4.62 (H-5)

(ppm)

b

R3/R GA T αG

99.5

98.7 98.9

102.0

GA

R3/R

173.6 174.0 174.5 (C-6)

17.6 (C-6)

FT 16.2 (C-6)

(ppm) Figure 1. 1H (a) and 13C (b) NMR spectra of the acidic fraction gum (ppm). (R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; bGT:b-D-Galp-(1?; T T T aG : a-D-Galp-(1?; F : a-L-Fucp-(1?; X : b-D-Xylp-(1?; X: ?4)-b-D-Xylp-(1?.)

2.2.2. Galacturonic acid residue The presence of three peaks at 173.6–174.5 ppm from carboxyl group in 13C 1D NMR spectrum (Fig. 1b) is the evidence for the existence of galacturonic acid (GA, 24.8%). Linear ?4)-D-GalpA(1? (GA, 22.9%) accounted for 92% of this residue, with the rest singly substituted at O-2 or O-3 position (Table 1). A full assignment for ?4)-a-D-GalpA-(1? (Table 2) was in good agreement with previous reports.15–17,19 2.2.3. Galactose residues The second most abundant neutral sugar residue is galactose (20.8%, Table 1), 94% of which is T-Galp-(1? (GT, 19.6%). The cross-peak at 5.12/102.1 ppm in HMQC (Fig. 4a) arose from H-1/ C-1 correlation of a-D-Galp-(1? (aGT), and its full assignment was in accordance with literature data.20,22,23 Two cross-peaks at 4.40/97.4 and 4.44/97.2 ppm in HMQC (Fig. 4a) may arise from H-1/C-1 connectivity of b-D-Galp-(1? (bGT),20,24,25 though both could be only equivocally assigned according to COSY and TOCSY spectra (Figs. 2 and 3) due to the low intensity in the rest of spectra. The a-configuration was predominant, however, the ratio of both configurations could not be estimated due to the overplayed anomeric proton chemical shifts of a-D-Galp-(1? and Rhap (5.12 ppm), or of b-D-Galp-(1? and b-D-Xylp-(1?(4.44 ppm) as shown in Figure 1.

2.2.4. Fucose residue Terminal a-L-Fucp-(1? (FT, 4.5%) was the only linkage for fucosyl residue. Correlations of H-1/H-2 and H-5/H-6 (Figs. 2 and 3) could be assigned according to literature data.26,27 2.2.5. Minor residues Xylose (7.7%) was present mainly as b-D-Xylp-(1? (XT, 3.1%) and ?4)-b-D-Xylp-(1? (X, 2.4%). The rest were ?4)-b-D-Xylp(1? branched at O-2, O-3, or both. Two group of cross-peaks at 4.44-4.46/3.20 and 4.31-4.33/3.20 ppm in COSY (Fig. 2) may arise from XT and X, respectively.28–30 However, full assignment was impossible due to the low intensity of signals. The small portion (2.6%) of arabinofuranose existed as ?3)-a-LAraf-(1? (A, 1.8%) or a-L-Araf-(1? (AT, 0.8%). The chemical shift for H-1/H-2 correlation of a-L-Araf-(1? was reported to be close to that of rhamnose,28–30 thus it could not be assigned. The cross-peak at 5.12/4.17 ppm in COSY (Fig. 2) may arise from H-1/ H-2 correlation of ?3)-a-L-Araf-(1? as the H-2 chemical shift moved to downfield by 0.1–0.2 ppm due to the glycosylation at O-3.31 2.2.6. Linkage sequence of AFG Heteronuclear multiple bond correlation (HMBC) spectrum shows correlation between protons and carbons 2–3 bonds away.

50

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

XT12 X12 t βG 12

FT12 T

αG

12 GA12

R12 R312 A12

Figure 2. Key fragments of COSY spectrum of the acidic fraction gum. (Intra-ring COSY correlations are labeled by residue abbreviations and two unseparareted numbers of correlating protons; R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; a/bGT: a/b-D-Galp-(1?; FT: a-L-Fucp-(1?; XT: b-D-Xylp-(1?; X: ?4)-b-DXylp-(1?; A: ?3)-a-Araf-(1?.)

Nuclear overhauser effect spectroscopy (NOESY) correlates nuclei through space (distance smaller than 5 Å). Both methods are helpful to reveal the glycosylic linkages between sugar residues, although intra-residue connectivities are also included. The interresidue and intra-residue correlations observed are listed in Table 3. The most intense inter-residue connectivities assigned in Figure 5 were C-1R⁄/H-4GA (99.9/4.32 ppm, R⁄: R or R3) and H1R⁄/C-4GA (5.12/78.3 ppm), indicating the abundance of diglycosyl repeating unit ?2)-a-L-Rhap-(1?4)-a-D-GalpA-(1?. Links between ?4)-a-D-GalpA-(1? itself, i.e. homogalacturonan (HG) regions, were evident by H-1/H-4 correlation of ?4)-a-D-GalpA(1? (4.89/4.32 ppm, Fig. 6), though its evidence in HMBC (C-1/H4GA, 99.0/4.32 ppm, Fig. 5) overplayed with C-1R⁄/H-4GA, (99.9/ 4.32 ppm). However, the amount of HG should be limited due to the low GalpA content and its predominant occurrence in repeating unit ?2)-a-L-Rhap-(1?4)-a-D-GalpA-(1?. Strong correlations of

H-1/C-2R⁄ (5.12/77.5–78.0 ppm) and C-1/H-2R (99.7/3.99 ppm) observed in Figure 5 may derive from both intra- and inter-residue connectivities between rhamnosyl residues. However, the existence of a homorhamnan (HR) region could still be confirmed due to the abundance of rhamnosyl residues. Given the above evidence, a backbone consisting of RG-I possibly intervened by small amount of HG and HR could be drawn. The branching site of R3 (19.5%) at O-3 was mostly substituted by monosaccharides, for example, a/b-D-Galp-(1? (a/bGT, 19.6%), a-L-Fucp-(1? (FT, 4.5%), and b-D-Xylp-(1? (XT, 3.1%) (Table 1). The C-3 chemical shifts of R3 varied with substitution by different sugar units (Table 3, Figs. 5 and 6). However, when it was substituted with ?2)-a-L-Rhap-(1? or ?4)-a-D-GalpA-(1?, a longer side chain consisting of more than two residues may induced. The fine structure of pectin is still in dispute due to its complexity and heterogeneity between plants and tissues6. The controversy between conventional and recently proposed alternative structural

51

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

FT56

R 16 R3 16

R3/R

26

FT

36 46

46 56

T βG

12 13

T αG

R15 16 12/13 R13 /14 R315 R313 15 R12

36 56

GA

R14

R314

26

13 12

13 14

12

15

15 13

R312

14

Figure 3. Key fragments of TOCSY spectrum of the acidic fraction gum. (Intra-ring TOCSY correlations are labeled by residue abbreviations and two unseparated numbers of correlating protons; R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; a/bGT: a/b-D-Galp-(1?; FT: a-L-Fucp-(1?.)

models of pectin is whether the HG domain is inserted into its backbone or linked as a side chain (Fig. 8). Similar question also existed in revealing the structure of AFG. A combination of both models was tentatively adopted. Given the above evidence, a possible repeating unit of AFG (Fig. 7) is proposed.

existed when substitution with ?4)-D-GalAp-(1? or ?2)-a-LRhap-(1? was 3-linked to 1,2-linked rhamnosyl residues. 4. Experimental 4.1. Materials

3. Conclusions The acidic fraction gum (AFG) from flaxseed hulls was a pectic polysaccharide as investigated by methylation analysis and 2D NMR spectroscopy. AFG contained a backbone consisting of rhamnogalacturonan-I which might be interrupted by small amount of homorhamnan or homogalacturonan. However, AFG differed from pectin in a much higher rhamnose (38.2%) and much lower galacturonic acid content (25.4%), thus it had limited amount of HG region. This fraction also featured its high amount of monogalactosyl branches 3-linked to half of 1,2-linked rhamnosyl residues. However longer side chains with more than two residues also

Soluble flaxseed gum was isolated by aqueous extraction form flaxseed hulls (variety Bethune) supplied by Natunola Health Biosciences (Winchester, Ontario, Canada). The acidic fraction gum (AFG) was separated from soluble flaxseed gum using anionexchange chromatography and purified by protease hydrolysis to remove proteinaceous contaminant.1 AFG was a heterogeneous polysaccharide as indicated by its molecular weight distribution from size-exclusion chromatography. It consisted mainly of three portions of peak molecular weight: 1510, 341, and 6.6 kDa, respectively.1All chemicals were of reagent grade unless otherwise specified.

52

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

a

b

T βG 1

R31 R1

T αG 1

FT6

GA1

FT1

R3/R6

c

T αG 6 T T αG 3 αG 2

GA2

GA5 GA3

3 R33 R 5 R5 3 R4 GT4 αGT5 R2 α

GA4

R32

R4

R33

Figure 4. Key fragments of HMQC spectrum of the acidic fraction gum. (Intra-ring H/C correlations are labeled by residue abbreviation and the number of correlating proton/ carbon; R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; a/bGT: a/b-D-Galp-(1?; FT: a-L-Fucp-(1?.)

4.2. Methylation analysis The acidic fraction gum (AFG) was reduced before methylation. AFG was reduced by sodium borodeuteride (NaBD4) instead

of sodium borohydride (NaBH4) in order to label each sugar residue reduced from uronic acid by two deuterium nuclei (–CD2OH). The labeled sugar residue differed from a corresponding native one in an increased mass-to-charge ratio by 2 for each corre-

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

H-1/C-2/3αGT H-1αGT/C-3R* (H-1/C-3R*)

53

H-1FT/C-3R3 H-1/C-2GA H-1/C-3GA H-1βGT(XT)/C-3R3

H-1GA/C-3R3

H-1αGT/C-3R3 H-1R*/C-4GA (H-1/C-2R*) C-1/H-4GA C-1GA/H-2R *

C-1R /H-4GA C-1/H-2R

Figure 5. Key fragment of HMBC spectrum of the acidic fraction gum. (Inter-ring H/C correlations are labeled by correlating proton or carbon and its number followed by the residue abbreviation; R⁄: R & R3; R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; a/bGT: a/b-D-Galp-(1?; FT: a-L-Fucp-(1?; XT: b-D-Xylp-(1?.)

T αG 1 /R*1 (5.12)

FT1 (5.03)

T βG 1 /XT1 (4.46)

X1 /GA4 (4.32)

GA1 (4.90)

T 3 αG 1/R 3 & R1/R33

T 3 βG 1/R 3 T 3

X 1/R 3 FT1/R33 GA1/R33

R33 (3.83/3.85) R2 (3.97) R32 (4.05)

T

F 1/R2

GA1/R2 GA1/R32

GA4 (4.32)

R*1/GA4 GA1/GA4

Figure 6. Key fragment of NOESY spectrum of the acidic fraction gum. (Inter-ring H/C correlations are labeled by the residue abbreviation/its proton number; R⁄: R & R3; R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; a/bGT: a/b-D-Galp-(1?; FT: a-L-Fucp-(1?; XT: b-D-Xylp-(1?; X: ?4)-b-D-Xylp-(1?.)

sponding ionic fragment in the mass spectra. The procedure of reduction and methylation followed that of previous work.32

The resultant partially methylated alditol acetates (PMAAs) were injected into a GC–MS system (ThermoQuest Finnigan, San Diego,

54

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

Table 2 1 H/13C NMR chemical shifts for the acidic fraction gum in D2O at 70 °C (ppm) Deduced Residues

H-1/C-1

H-2/C-2

H-3/C-3

H-4/C-4

H-5/C-5

H-6/C-6

?2,3)-a-L-Rhap-(1?

5.11/99.6

4.05/78.0

3.50/74.1

3.74/70.5

1.16/17.5

R GA t aG t G b

?2)-a-L-Rhap-(1? ?4)-a-D-GalpA-(1? a-D-Galp-(1? b-D-Galp-(1?

FT XT X A

a-Fucp-(1?

5.12/99.6 4.89/99.0 5.12/102.1 4.40/97.4 4.44/97.2 5.02/102.3 4.44/97.2 4.31/97.4 5.12/-

3.97/77.5 3.76/69.0 3.67/69.5 3.32/— 3.38/— 3.64/— 3.20/— 3.20/— 4.17/—

3.83/72.2 3.85/77.9 3.73/70.5 3.97/71.1 3.69/69.6 3.48/— 3.48/— 3.60/— —/— —/— —/—

3.30/73.1 4.32/78.3 3.87/70.5 3.80/— 3.80/— —/— —/— —/— —/—

3.62/70.5 4.63/72.1 3.77/70.5 3.59/— 3.59/— 3.90/— —/— —/— —/—

1.13/17.5 /173.6-174.5 3.59/62.5 —/— —/— 1.08/16.2 —/— —/— —/—

R

3

b-D-Xylp-(1? ?4)-b-D-Xylp-(1? ?3)-Araf-(?

Table 3 Heteronuclear (HMBC) and homonuclear (TOCSY & NOESY) connectivities in the acidic fraction gum (ppm) Inter-residue connectivities

Intra-residue connectivities

HMBC

NOESY

HMBC

Atom

d

Atom

d

Atom

d

Atom

d

Atom

H-1 R⁄

5.12

4.90

99.7 99.9 99.0 99.0

78.3 77.5-78.0 3.99 4.32 4.32 3.97

H-1 GA

C-1 R⁄

C-4 GA C-2 R⁄ H-2 R H-4 GA GA H-4 H-2 R

H-1 FT H-1 R⁄ H-1 GA

5.03 5.12 4.90

H-2 H-2 H-2 H-4 H-4

3.97 4.05 3.97 4.32

H-1 H-1 H-1 H-4

C-1 GA

R R3 R GA GA

d

Atom

d

R⁄ R R3 GA

5.12 5.12 5.12 4.32

H-1 aGT

5.12

C-2 C-3 C-3 C-2 C-3 C-2/3

78.0 70.9 72.2 69.4 71.3 69.8

TOCSY (Appedants linked to C-3 R3) H-1 H-1 H-1 H-1 H-1 H-1

aG bG T

T

T

X R GA FT

5.12 4.44 4.44 5.12 4.90 5.03

(Appedants linked to C-3 R3) C-3 C-3 C-3 C-3 C-3 C-3

3

R R3 R3 R3 R3 R3

72.1/77.9 72.2 72.2 72.1 71.7 70.8

H-1 H-1 H-1 H-1 H-1 H-1

T

aG bG T

T

X R GA FT

5.12 4.46 4.46 5.12 4.90 5.03

Atom H-3 H-3 H-3 H-3 H-3 H-3

3

R R3 R3 R3 R3 R3

3.83 3.85 3.85 3.85 3.85 3.83

H-1 H-1 H-1 H-6 H-1 H-6

aG bG ⁄

T

T

R R⁄ GA FT

d

Atom

5.12 4.44 5.12 1.11 4.89 1.08

H-2,3,4,5,6 H-3,4 H-2,3,4,5 H-1,2,3,4,5 H-2,3,4 H-5,6

(R⁄: R and R3; R3: ?2,3)-a-L-Rhap-(1?; R: ?2)-a-L-Rhap -(1?; GA: ?4)-a-D-GalpA -(1?; bGT:b-D-Galp-(1?; aGT: a-D-Galp-(1?; FT: a-L-Fucp-(1?; XT: b-D-Xylp-(1?.)

Figure 7. Proposed repeating unit of the acidic fraction gum. (HR, RG-I and HG refer to homorhamnan, rhamnogalacturonan-I and homogalacturonan, respectively. The locations of HR, RG-I and HG are interchangeable; (m+n)/(n+i)1.5. The substitution rate of R1 is 54 %. R1 is mostly monosaccharide (a/b-D-Galp-(1?, a-L-Fucp-(1? or b-DXylp-(1?). R1 may also occasionally be a longer side chain with more than two residues beginning with ?4)-a-GalpA-(1? or ?2)-a-L-Rhap-(1?, wherein the side-chain structure may be similar to part of the main chain.)

CA) with an SP-2330 (Supelco, Bellefonte, Pa) column (30 m  0.25 mm, 0.2 mm film thickness, 160–210 °C at 2 °C/min, and then 210–240 °C at 5 °C/min) equipped with an ion trap MS detector. 4.3. NMR analysis Sample was dissolved in deuterium oxide (D2O, 80 °C, 1 h) and lyophilized for three times to replace the exchangeable protons with deuterons before being finally redissolved in D2O (3%) for NMR analysis. High-resolution 1H and 13C NMR spectra were

recorded in D2O at 500.13 and 125.78 MHz, respectively, on a Bruker ARX500 NMR spectrometer operating at 25 °C. A 5 mm inverse geometry 1H/13C/15N probe was used. Chemical shifts are reported relative to external standards trimethylsilyl propionate (TSP in D2O, 4.76 ppm, for 1H) and 1,4-dioxane (in D2O, 66.5 ppm, for 13 C). Homonuclear 1H/1H correlation spectroscopy (COSY, TOCSY) and nuclear overhauser effect spectroscopy (NOESY), heteronuclear 1H/13C multiple-quantum coherence spectroscopy (HMQC) experiments, and heteronuclear multiple bond correlation spectroscopy (HMBC) were run using the standard Bruker pulse sequence at 70 °C.

K.-Y. Qian et al. / Carbohydrate Research 362 (2012) 47–55

55

a

HG RG-I RG-II

b

Figure 8. Schematic representations of the conventional (a) and recently proposed alternative (b) structures of pectin. (RG-I: rhamnogalacturonan-I; RG-II: rhamnogalacturonan-II; HG: homogalacturonan; schematics are adapted from Willats & Knox et al. 2006.)

Acknowledgements The authors would like to thank Ms. Cathy Wang for her technical assistant. Special thanks are also due to Dr. Ying Wu, Professor Shao-Ping Nie and Qing-Bin Guo for their suggestions and support. References 1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15.

Qian, K. Y.; Cui, S. W.; Wu, Y.; Goff, H. D. Food Hydrocolloids 2012, 28, 275–283. Cui, W.; Mazza, G.; Biliaderis, C. G. J. Agric. Food Chem. 1994, 42, 1891–1895. Naran, R.; Chen, G. B.; Carpita, N. C. Plant Physiol. 2008, 148, 132–141. Beda, M. Y. Carbohydr. Polym. 2011, 86, 373–385. Ridley, B. L.; O’Neill, M. A.; Mohnen, D. Phytochemistry 2001, 57, 929–967. Willats, W. G. T.; Knox, J. P.; Mikkelsen, J. D. Trends Food Sci. Technol. 2006, 17, 97–104. Voragen, A. G. J.; Pilnik, W.; Thibault, J.-F.; Axelos, M. A. V.; Renard, M. G. C. Pectins. In Food Polysaccharides & Their Applications; Stephen, A. M., Ed., 5th ed.; CRC Press: New York, 1995; pp 287–339. Glinsky, V. V.; Raz, A. Carbohydr. Res. 2009, 344, 1788–1791. Morris, V.; Gromer, A.; Kirby, A.; Bongaerts, R.; Patrick, G. A. Food Hydrocolloids 2011, 25, 230–237. Hawker, C. J.; Lee, R.; Frechet, J. M. J. J. Am. Chem. Soc. 1991, 113, 4583– 4588. Tao, Y.; Zhang, L.; Yan, F.; Wu, X. Biomacromolecules 2007, 8, 2321–2328. Pawan, K. A. Phytochemistry 1992, 31, 3307–3330. Colquhoun, I. J.; Ruiter, G. A.; Schols, H. A.; Voragen, A. G. J. Carbohydr. Res. 1990, 206, 131–144. Muralikrishna, G.; Salimath, P. V.; Tharanathan, R. N. Carbohydr. Res. 1987, 161, 265–271. MacLean, L. L.; Vinogradov, E.; Pagotto, F.; Farber, J. M.; Perry, M. B. Carbohydr. Res. 2010, 345, 1932–1937.

16. Fedonenko, Y. P.; Konnova, O. N.; Zdorovenko, E. L.; Konnova, S. A.; Zatonsky, G. V.; Shashkov, A. S.; Ignatov, V. V.; Knirel, Y. A. Carbohydr. Res. 2008, 343, 810–816. 17. Ojha, A. K.; Maiti, D.; Chandra, K.; Mondal, S.; Roy, D. D. S. K.; Ghosh, K.; Islam, S. S. Carbohydr. Res. 2008, 343, 1222–1231. 18. Choma, A.; Komaniecka, I.; Sowinski, P. Carbohydr. Res. 2009, 344, 936–939. 19. Mikshina, P. V.; Gurjanov, O. P.; Mukhitova, F. K.; Petrova, A. A.; Shashkov, A. S.; Gorshkova, T. A. Carbohydr. Polym. 2012, 87, 853–861. 20. Sengkhamparn, N.; Bakx, E. J.; Verhoef, R.; Schols, H. A.; Sajjaanantakul, T.; Voragen, A. G. J. Carbohydr. Res. 2009, 344, 1842–1851. 21. Cui, W.; Eskin, M. N. A.; Biliaderis, C. G.; Marat, K. Carbohydr. Res. 1996, 292, 173–183. 22. Galbraith, L.; Sharples, J. L.; Wilkinson, S. G. Carbohydr. Res. 1999, 319, 204– 208. 23. Ahrazem, O.; Prieto, A.; Leal, J. A.; Jiménez-Barbero, J. O.; Bernabé, M. Carbohydr. Res. 2002, 337, 2347–2351. 24. Vidal, S. P.; Doco, T.; Williams, P.; Pellerin, P.; York, W. S.; O’Neill, M. A.; Glushka, J.; Darvill, A. G.; Albersheim, P. Carbohydr. Res. 2000, 326, 277–294. 25. Deng, C.; O’Neill, M. A.; York, W. S. Carbohydr. Res. 2006, 341, 474–484. 26. Shashkov, A. S.; Vinogradov, E. V.; Knirel, Y. A.; Nifant’ev, N. E.; Kochetkov, N. K.; Dabrowski, J.; Kholodkova, E. V.; Stanislavsky, E. S. Carbohydr. Res. 1993, 241, 177–188. 27. Hanniffy, O. M.; Shashkov, A. S.; Moran, A. P.; Prendergast, M. M.; Senchenkova, S. Y. N.; Knirel, Y. A.; Savage, A. V. Carbohydr. Res. 1999, 319, 124–132. 28. Höije, A.; Sandström, C.; Roubroeks, J. P.; Andersson, R.; Gohil, S.; Gatenholm, P. Carbohydr. Res. 2006, 341, 2959–2966. 29. Pastell, H.; Tuomainen, P. I.; Virkki, L.; Tenkanen, M. Carbohydr. Res. 2008, 343, 3049–3057. 30. Sun, Y.; Cui, S. W.; Gu, X.; Zhang, J. Carbohydr. Polym. 2011, 85, 615–621. 31. Westphal, Y.; Kühnel, S.; Waard, P.; Hinz, S. W. A.; Schols, H. A.; Voragen, A. G. J.; Gruppen, H. Carbohydr. Res. 2010, 345, 1180–1189. 32. Kang, J.; Cui, S. W.; Chen, J.; Phillips, G. O.; Wu, Y.; Wang, Q. Food Hydrocolloids 2011, 25, 1984–1990.