ARTICLE IN PRESS
FOOD
HYDROCOLLOIDS Food Hydrocolloids 21 (2007) 329–337 www.elsevier.com/locate/foodhyd
Characterization and properties of Acacia senegal (L.) Willd. var. senegal with enhanced properties (Acacia (sen) SUPER GUMTM): Part 2—Mechanism of the maturation process Hiromitsu Aokia,b, Saphwan Al-Assaf a, Tsuyoshi Katayamab, Glyn O. Phillipsa,c, a
Phillips Hydrocolloids Research Centre, The North East Wales Institute, Plas Coch, Mold Road, Wrexham LL11 2AW, UK b San-Ei Gen F.F.I., Inc., 1-1-11 Sanwa-cho, Toyonaka, Osaka 561-8588, Japan c Phillips Hydrocolloid Research Ltd., 45 Old Bond Street, London W1S 4AQ, UK Received 19 December 2005; accepted 13 April 2006
Abstract The molecular changes which accompany the maturation process to produce Acacia (sen) SUPER GUMTM are further described and demonstrate that Acacia senegal can be produced with precisely structured molecular dimensions. The controlling feature is the agglomeration of the proteinaceous components within the molecularly disperse system that is naturally occurring A. senegal gum to increase the amount of arabinogalactan protein (AGP) component. The new structural unit is investigated by enzymatic treatment of a number of Acacia (sen) SUPER GUMTM samples, which demonstrates that the AGP so formed is hydrolyzed by the enzyme protease in exactly the same way as for the AGP in control A. senegal which has not been matured. The rheological features of the matured A. senegal gum reflect the increase in molecular dimensions and indicate that the AGP formed is highly cross-linked. A model for the aggregation of the proteinaceous components is proposed. r 2006 Elsevier Ltd. All rights reserved. Keywords: Gum arabic; Acacia senegal; SUPER GUMTM; Maturation; Molecular weight; Enzyme digestion; GPC-MALLS
1. Introduction Previously in Part 1 of this series (Al-Assaf, Phillips, Aoki, & Sasaki, 2006), we reported the maturation process to produce a new series of standardized Acacia gums (designated generally Acacia (sen) SUPER GUMTM when derived from Acacia senegal and as also Acacia (sey) SUPER GUMTM when derived by the same process from Acacia seyal) which accelerates and enhances this same natural aggregation process, under strictly controlled conditions, which were worked out first at laboratory level, then pilot scale and finally at plant level (Hayashi, 2002). Such an aggregation process occurs when the tree grows older up to about 15 years (Idris, Williams, Corresponding author. Phillips Hydrocolloids Research Centre, The North East Wales Institute, Plas Coch, Mold Road, Wrexham LL11 2AW, UK. Tel.: +44 29 20 843298; fax: +44 29 20 843145. E-mail address:
[email protected] (G.O. Phillips).
0268-005X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodhyd.2006.04.002
& Phillips, 1998). The naturally occurring association of the smaller molecular weight arabinogalactan (AG) and glycoprotein (GP) units into larger units is taken a further stage to give larger molecular weight arabinogalactan protein (AGP) aggregates. By monitoring the molecular architecture of the gum at all stages, specific new products have been characterized. In all aspects this specially matured gum is chemically and molecularly identical to the base gum, but because of the difference in distribution of smaller units into larger aggregates, the physical and functional performance is greatly enhanced. In Part 1 of this series (Al-Assaf et al., 2006) the functional emphasis was placed on emulsification, since this property has proved the most high value application of commercial gum arabic and the most difficult to reproduce consistently. We also demonstrated that there is no restriction upon the particular molecular characteristics and protein distribution, which can be produced by the maturation process. Here, we have selected representative
ARTICLE IN PRESS 330
H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
samples with complete solubility for large-scale production with weight average molecular weights of 1.0 to ca. 2.5 106 g/mole. This paper further describes the molecular maturation changes which can be effected to enhance the performance of natural gum arabic, without introducing any new chemical groups and ensure that the material is of constant properties and performance. 2. Material and methods 2.1. Materials
A pulsed amperometoric detector (ED40 Electrochemical Detector, Dionex) was used. 2.5. Amino acids analysis The sample was hydrolyzed with 6 N hydrochloric acid containing 2-mercapt ethanol at 110 1C for 24 h. The hydrolyzed sample was analyzed by HPLC using an AApak Na II-S2 column (Jasco, Japan) with Amino buffer-II as mobile phase. Post-column o-phthalaldehyde (OPA)-fluorescent detection (excitation 345 nm, emission 455 nm) was used.
Gum arabic (Acacia senegal var. senegal) samples Lot No. FR-2876 (hand picked selected, grade in lump form) were provided by San Ei Gen F.F.I., Inc. (Osaka, Japan) and were originally obtained from the Gum Arabic Company in Sudan. The sample was mechanically kibbled and three matured gum arabic samples were prepared, according to the method reported in Part 1 of this series (Al-Assaf et al., 2006). The matured samples were labeled FR-2877, FR-2878, and FR-2979 and were spray dried following maturation. Additionally two commercial Acacia (sen) SUPER GUMTM samples (EM1 and EM2), produced in ton quantities were also used. Sodium chloride (99.94% for analysis), Tris(hydroxymethyl) aminomethane (99.94%), and hydrochloric acid (35.5–37.5%) were purchased from Fisher Scientific UK. Protease (Type XIV: Bacterial from Streptomyces griseus; Product No. P5147) was obtained from Sigma. Sulfuric acid, hydrochloric acid and 2-mercapt ethanol were purchased from Wako Pure Chemical Industry (Osaka, Japan). Distilled water (Bibby Merit W4000) was used for all experiments. Pullulan standard (P-50, molecular weight 4.73 104 g/mole) was obtained from Shodex, Japan (Tokyo, Japan).
2.6. Intrinsic viscosity
2.2. Loss on drying
The procedures used in this investigation are extensions of the basic technique described previously (Al-Assaf, Katayama, Phillips, Sasaki, & Williams, 2003). However, 1.0–4.0 mg/ml of test solution was used, after taking into account the loss on drying. The solution was agitated on a tube roller mixer (SRT-2, Jencons Scientific, Inc., UK) for at least 5 h to ensure that the sample fully dissolves and hydrated. The test solution was filtered using 0.45-mm nylon filter (Whatman, 13 mm). Astra for Windows software (version 4.90.07, Wyatt Technology Corporation) was used in the instrument control and data acquisition by an Agilent 1100 series G1314A UV detector (214 nm, Agilent Technologies), a DAWN EOS multi-angle laser light scattering detector (l0 ¼ 690 nm, Wyatt Technology Corporation) and an Optilabs rEX refractometer (Wyatt Technology Corporation). DAWN EOS was calibrated using toluene (SPECTRANAL 99.9%, Riedel-de Hae¨n). The Pullulan standard (molecular weight 47,300 g/mole) was used to normalize the detectors and for the determination of the delay volume. The delay volumes of the equipment were
The loss of drying was measured to determine water content of spray-dried samples. Around 1.5 g of test material was accurately weighed into a weighted vial and dried (105 1C, 4 h, SANYO convection oven MOV-212F). Then, the sample was moved to a desiccator to cool at room temperature. The loss of weight was used to calculate the solid content of the samples. 2.3. Specific rotation The sample was dissolved in distilled water and measured by a Jasco P-1020 Polarimeter. 2.4. Sugar analysis The sample was hydrolyzed with 2-N sulfuric acid at 100 1C for 2 h and neutralized with sodium hydroxide. HPLC analysis was performed using a Dionex Carbo Pac PA-1 with a 0.01 M sodium hydroxide mobile phase.
In all, 30 mg/ml gum arabic solution (as solid content) in 1.0 M and 0.2 M NaCl aqueous solution was prepared. The precise concentration of the sample was calculated based on a dry solid weight basis. The solution was filtered through 0.8 mm cellulose acetate membrane (Naglene, 25 mm). A calibrated Ubbelohde viscometer (Cannon Ubbelohde Semi-Micro Calib 75) was used for measuring in efflux time (measured in seconds) to calculate the relative viscosity based on the efflux time of the solvents at 25.0 1C. The efflux time was measured three times by allowing the sample solution to flow freely through bulb. Firstly, the efflux time of solvents was measured and that of the test solution (30 mg/ml, 2 ml) was determined. Then, the test solution in the viscometer was successively diluted by adding 0.4, 0.6, 1, 2 and the 4 ml of solvents and the efflux time of the diluted solution was measured. 2.7. Gel permeation chromatography—multi-angle laser light scattering (GPC-MALLS) analysis
ARTICLE IN PRESS H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
0.0296 ml between UV detector and DAWN EOS, and 0.1353 ml between DAWN EOS and refractometer. A Superose 6 10/300GL (Amersham Biosciences) was used with 0.2 M NaCl aqueous solution filtered though 0.22 mm Millipore filter as eluent at flow rate of 0.50 ml/min (Knauer, HPLC Pump K-501). An injection volume of 100 ml (Rheodyne 7725i) was used. A value of 0.141 ml/g was used for the differential refractive index increment (dn/dc). The injected mass was 0.40 mg for FR-2876, 0.20 mg for FR-2877, 0.15 mg for FR-2878 and 0.10 mg for FR-2879 to avoid light scattering saturation. Berry fitting method was used for analysis. p The Berryffi ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi fitting method relies on constructing a plot of Kc=RðyÞ against sin2(y/2) and fitting a polynomial in sin2(y/2) to the data. Berry fitting method is useful for large molecules, therefore, when the result of Zimm plots was over 50 nm; the result of Berry plots was adopted. There are a total of 18 detectors in the DAWN EOS instrument, but for aqueous solutions the signal from lower angle detectors, i.e. detectors #3–6 and #16–18, is very noisy and for some samples these readings may be deleted during data processing. The subset of signals from light scattering detectors # 6 (431)–# 16 (1421) was used in the analysis. 2.8. Protease treatment A total of 40 mg/ml test material solution (as solid content) in 10 mM Tris-HCl/200 mM NaCl (pH 7.5, Trisbuffered Saline, TBS) was prepared. However, 1.6 mg/ml protease solution in TBS was also prepared. In all, 1.0 ml of test material solution was mixed with 8.0 ml of TBS and 1.0 ml of protease solution (final concentration: test material 4 mg/ml, protease 0.16 mg/ml), and then the mixed solution was incubated at 37 1C for 24 h and analyzed by GPC-MALLS (Osman, Menzies, Williams, Phillips, & Baldwin, 1993).
This gum is a typical ‘‘hashab’’ sample from the Kordofan region of Sudan as previously described (Al-Assaf, Phillips, & Williams, 2005; Idris et al., 1998). The gum was matured to different levels, and the resulting products (A–E) were designated FR-2877 (A), FR-2878 (B) and FR-2979 (C) at a pilot scale. The two D and E products (Acacia (sen) SUPER GUMTM EM1 and Acacia (sen) SUPER GUMTM EM2) are the commercial designations protected by patent and by trademark, have been produced in ton quantities and subjected to intensive commercial evaluation in a wide range of products. 3.2. Chemical analysis The matured gums produced by accelerated maturation process were chemically identical to the gum as collected from the tree. They contained exactly the same specific optical rotation, sugar moieties and amino acids in the same proportions as control gum, which has not been subjected to the accelerated maturation process (Tables 1 and 2). 3.3. GPC-MALLS analysis Fig. 1 shows elution profile of control gum arabic FR2876 using the GPC-MALLS technique. The elution profile for A. senegal monitored using light scattering, refractive index and UV detection has been previously described (Al-Assaf et al., 2003, 2005). The light scattering response reflects the mass and concentration and shows two distinctive peaks. The first peak has a high response since it corresponds to the high molecular weight material (AGP) content. The second peak is broader with lower response and it accounts for the rest of the gum (90%).
3. Results and discussion 3.1. Production of matured gums at a pilot scale and a plant level
Table 1 shows a series of matured gums using as starting gum arabic with MW average of 6.2 105 g/mol (FR-2876).
331
The refractive index (RI) response also shows two peaks but the response is opposite to that in light scattering. This is because it is a concentration detector and since the AGP is only 10% of the total gum its peaks is smaller than that of the AG and GP which consists 90% of the total mass. The UV response shows three peaks. The first peak is for the AGP which has the protein core and the carbohydrate attached to it. The second peak appears as a
Table 1 Specific rotation and sugar composition (mol%) of the control and matured gum arabic Sample name (code no.)
Specific rotation (deg dm1 cm3 g1)
Gal
Ara
Rha
Uronic acid
Control gum arabic (FR-2876) Matured gum arabic A (FR-2877) Matured gum arabic B (FR-2788) Matured gum arabic C (FR-2789) Matured gum arabic D (Acacia (sen) SUPER GUMTM EM1) Matured gum arabic E (Acacia (sen) SUPER GUMTM EM2)
30 30 30 30 30 31
32.3 34.2 32.1 31.9 32.0 32.8
30.2 31.3 30.1 29.9 30.9 31.4
13.3 13.9 13.4 13.5 13.4 13.1
24.2 20.6 24.4 24.8 23.7 24.3
Gal, galactose; Ara, arabinose; Rha, rhamnose.
ARTICLE IN PRESS H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
332
Table 2 Amino acid composition (mol%) of the control and matured gum arabic Sample
%N
FR-2876 0.314 FR-2879 0.308 Acacia (sen) SUPER GUMTM EM2 0.325
Ala Arg Asp Cys Glu Gly His Hyp
Ile
Leu Lys Met Phe Pro
Ser
Thr Tyr Val
2.6 2.8 2.8
1.3 1.4 1.4
8.0 8.6 8.3
14.1 14.3 13.8
7.5 7.9 8.1
0.9 0.9 1.0
5.1 5.6 5.4
— — —
3.5 3.7 3.6
5.6 5.7 5.7
5.8 5.2 5.4
26.9 25.2 26.2
2.9 2.5 2.6
0.1 0.1 0.1
3.6 3.6 3.5
7.5 7.6 7.4
0.8 0.8 0.7
3.8 4.1 4.0
Fig. 1. GPC chromatogram of gum arabic as starting material (control gum arabic FR-2876) showing the light scattering (LS, 901), refractive index (RI) and UV at 214 nm. GPC analysis condition: column, superose 6; temperature, room temperature; mobile phase, 0.2 M NaCl; flow rate, 0.5 ml/min; sample, 4 mg/ml dissolved in the mobile phase; injected volume, 100 ml.
shoulder immediately after the AGP and corresponds to the AG. Finally the third peak elutes before the total volume and it corresponds to the GP. The GP peak is not detected on the light scattering (mass detector) since it has low molecular weight. Also it cannot be seen on the refractive index (concentration detector). The samples had weight average molecular weight (MW) increasing from 0.6 to 2.5 106 g/mol; with the AGP MW value increasing from 2.5 to 11.6 106 g/mol (FR-2879). The amount of AGP in these samples increases from 10.6% to 18.6% whereas the MW of the AG component changes very little—from 0.40 to 0.45 106 g/mol (Table 3). As indicated above the second peak represents the total of the two components: AG and the smaller molecular weight GP. The overlay of the elution chromatograms measured by the light scattering detector (Fig. 2a), illustrates the regular increase in molecular weight of the matured products. Fig 2b shows overlay of the elution chromatograms measured by the refractive index detector. These figures illustrate the changes in amounts of the various components as the maturation proceeds. It is evident that there is a decrease in the amount of the AG component, which link to give an increased amount of AGP (Fig. 2c). The corresponding changes measured by the ultraviolet
detector support this conclusion and assist in interpreting the overall changes. Maturation is thus a process which associates the GP of lower molecular weight (ca. 50,000 g/mol) and the AG (MW, ca. 350,000 g/mol) to give more and higher molecular weight AG protein. The process thus mimics the biological process, which produces more AG protein as the tree grows to 15 years, and the maturation, which continues upon storage of the gum after harvesting (Idris et al., 1998). Thus, all the original structural features of the harvested gum are retained by the maturation process. The change is completely associated with the increased formation of the AGP component by a physical aggregation process. Assuming that AGP has a molecular weight of 2.5 106 g/mol, and AG and GP is one component with molecular weight of 4 105 g/mol in control gum arabic, the number of molecules required for AGP in maturated gum (FR-2879) can be evaluated. On this basis the extent of re-organization necessary to form the amount AGP in the matured sample FR-2879, would require 2–3 units of AGP in the original gum to associate with 10–15 AG and GP moieties in the control gum. The actual final molecule has, of course, a wider polydispersity compared with the original gum and accommodates the remaining unaffected components (Fig. 3).
ARTICLE IN PRESS H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
333
Table 3 Molecular weight of control and matured gum arabic by GPC-MALLS analysis Sample
Processing
Molecular weight (MW, g/mol)
% Mass
Rg (nm)
Control gum arabic FR-2876
Total First peak (AGP) Second peak (AG+GP)
6.22 105 2.54 106 3.96 105
10.6 89.4
28 41 —
Matured gum arabic A FR-2877
Total First peak (AGP) Second peak (AG+GP)
1.23 106 6.58 106 4.13 105
13.2 86.8
59 67 —
Matured gum arabic B FR-2788
Total First peak (AGP) Second peak (AG+GP)
1.66 106 8.56 106 4.16 105
15.3 84.7
64 71 —
Matured gum arabic C FR-2789
Total First peak (AGP) Second peak (AG+GP)
2.54 106 1.16 107 4.50 105
18.6 81.4
85 90 —
Matured gum arabic D Acacia (sen) SUPER GUMTM EM1
Total First peak (AGP) Second peak (AG+GP)
1.08 106 5.98 106 3.90 105
12.3 87.7
64 74 —
Matured gum arabic E Acacia (sen) SUPER GUMTM EM2
Total First peak (AGP) Second peak (AG+GP)
1.77 106 7.84 106 4.16 105
18.2 81.8
68 75 —
3.4. Changes in physical characteristics The intrinsic viscosities of the control and matured gum are shown in Table 4 which reflect the increase in overall size of the matured samples. At low salt concentration (0.2 M NaCl) the value, as a result of expansion, is higher compared to higher salt concentration (1 M NaCl). The latter is usually used for the determination of intrinsic viscosity of gum arabic. However, in both salt conditions the value increased with increasing the molecular weight of the respective sample. The relationship between the intrinsic viscosity [Z] and the relative molecular mass (MW) is given by [Z] ¼ KMaW, the MarkHouwink equation where K and a are constants. The literature values (references as listed on the figure legend) of Mark-Houwink constants of A. senegal given by Anderson and Rahman (1967) were K ¼ 0.013 and a ¼ 0.54 and other workers values are shown in Fig. 4. The matured gums, in 1 M NaCl, have a slope close to zero and corresponding to Mark-Houwink constants of K ¼ 3.36, a ¼ 0.120. This deviation from the conventional behavior indicates that the matured gum has a highly branched structure, so that the protein when aggregating forms a more compact structure than would occur if the protein simply extended the polypeptide chain of the basic AGP. Information about molecular shape can also be obtained by plotting log MW versus log Rg determined by a GPC-MALLS analysis (Fig. 5). The slope of the line for the AGP plots are 0.3–0.4. Values for linear polymer being in the range of 0.5–0.6 (Zimm & Stockmayer, 1949), which further indicates the existence of branch linkages in AGP.
3.5. The linkage structure of the protein in Acacia (sen) SUPER GUM The untreated gum (FR-2876) and the matured gums (FR-2977-2879) were subjected to enzyme hydrolysis using protease, which has been shown to completely hydrolyze the protein in the AGP component (Flindt, Al-Assaf, Phillips, & Williams, 2005). Following this treatment the gum no longer will serve as an emulsifier in beverage systems (Randall, Phillips, & Williams, 1988). Fig. 6 shows GPC elution profiles of control gum arabic after enzyme hydrolysis. As a result of the enzyme hydrolysis, the high molecular weight AGP peak was removed. The quantitative changes are summarized in Table 5. Several consequential changes in molecular weight parameters should be noted. The weight average molecular weight (MW) of the matured gum is reduced to that of the enzyme hydrolyzed control gum at ca. 4 105 g/mol and the MW of the AGP after enzyme hydrolysis and control gum equate at ca. 2 106 g/mol. As the AGP is hydrolyzed there is a corresponding increase in proportion of lower molecular weight components. These observations support the previous observations that the AGP formed by the maturation process behaves identically with that present in the control gum and retains the polypeptide linkage as in the control. The changes are illustrated schematically in Fig. 7. In conclusion, we consider that the maturing of the gum leads to transfer of the protein associated with the lower molecular weight components to give larger concentrations of AGP. This is no different from that which is present naturally and can be hydrolyzed by the protease enzyme in exactly the same way as the AGP in the control gum.
ARTICLE IN PRESS 334
H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
Fig. 2. GPC overlay chromatogram of control and matured gum arabic showing the light scattering at 901 (a), refractive index (b) and UV at 214 nm (c). The data was normalized as injected mass of 0.4 mg to display on the same Y-axis scale.
ARTICLE IN PRESS H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
335
Control Acacia senegal
AGP
AG + GP
Total Mw
(FR-2876)
2.5 × 106 g/mol
4.0 × 10 5 g/mol
6.0 × 10 6 g/mol
(10wt%)
2 ~ 3
Matured gum (FR-2879)
(90wt%)
10 ~ 15
AGP
AG + GP
Total Mw
1.2 × 107 g/mol
4.0 × 10 5 g/mol
2.5 × 10 6 g/mol
(18wt%)
(82wt%)
Fig. 3. Schematic representative of maturation process.
Table 4 Intrinsic viscosity (cm3/g) of control and matured gum in NaCl aqueous solution Sample
1.0 M
0.2 M
Control gum arabic (FR-2876) Matured gum arabic A (FR-2877) Matured gum arabic B (FR-2788) Matured gum arabic C (FR-2789) Matured gum arabic D (Acacia (sen) SUPER GUMTM EM1) Matured gum arabic E (Acacia (sen) SUPER GUMTM EM2)
16.8 17.8 18.7 19.5 17.5 19.4
18.3 20.0 21.4 22.2 18.4 21.6
Fig. 4. Mark-Houwink plot of gum arabic (K) FR series, (’) Acacia (sen) SUPER GUMTM series, and Fenyo (1985), , Idris et al. (1998).
, Anderson and Rahman (1967);
, Vandevelde
ARTICLE IN PRESS 336
H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
Fig. 5. Log–log plot of RMS radius of gyration versus molecular weight. FR 2876 (a), FR 2877 (b), FR 2878 (c), FR 2879 (d).
Fig. 6. GPC chromatogram of gum arabic after protease treatment (control gum arabic FR-2876) showing the light scattering (LS, 901), refractive index (RI) and UV at 214 nm.
ARTICLE IN PRESS H. Aoki et al. / Food Hydrocolloids 21 (2007) 329–337
337
Table 5 Molecular weight (g/mol) and % mass of first (AGP) and second (AG+GP) peaks of gum arabic samples before and after protease treatment determined by GPC-MALLS Sample
Enzyme treatment
Total MW
AGP (first peak)
5
6
% Mass
AG+GP (second peak) 5
% Mass
Control gum arabic FR-2876
Before After
6.22 10 4.08 105
2.54 10 2.30 106
10.6 2.3
3.96 10 3.63 105
89.4 97.7
Matured sample A FR-2877
Before After
1.23 106 4.22 105
6.58 106 2.13 106
13.2 3.5
4.13 105 3.60 105
86.8 96.5
Matured sample B FR-2878
Before After
1.66 106 4.39 105
8.56 106 2.27 106
15.3 4.0
4.16 105 3.63 105
84.7 96.0
Matured sample C FR-2879
Before After
2.54 106 4.56 105
1.16 107 2.27 106
18.6 5.0
4.50 105 3.60 105
81.4 95.0
Matured sample D Acacia (sen) SUPER GUMTM EM1
Before After
1.08 106 3.92 105
5.98 106 2.12 106
12.3 3.5
3.90 105 3.41 105
87.7 96.5
Matured sample E Acacia (sen) SUPER GUMTM EM2
Before After
1.77 106 4.39 105
7.84 106 2.22 106
18.2 4.0
4.16 105 3.58 105
81.8 96.0
Protease
Maturation process
Protease
Fig. 7. Maturation process and protease treatment of gum arabic.
References Al-Assaf, S., Katayama, T., Phillips, G. O., Sasaki, Y., & Williams, P. A. (2003). Quality control of gum arabic. Foods & Food Ingredients Journal of Japan, 208(10), 771–780. Al-Assaf, S., Phillips, G.O., Aoki, H., & Sasaki, Y. (2006). Characterization and properties of Acacia senegal (L.) Willd. var. senegal with enhanced properties (Acacia (sen) SUPER GUMTM): Part 1. Controlled maturation of Acacia senegal var. senegal to increase
viscoelasticity, produce a hydrogel form and convert a poor into a good emulsifier. Food Hydrocolloids, in press, doi:10.1016/ j.foodhyd.2006.04.011. Al-Assaf, S., Phillips, G. O., & Williams, P. A. (2005). Studies on Acacia exudate gums. Parts I. The molecular weight of Acacia senegal gum exudates. Food Hydrocolloids, 19(4), 647–660. Anderson, D. M. W., & Rahman, S. (1967). Studies on uronic acid materials. Part XX. The viscosity-molecular weight relationship for Acacia gums. Carbohydrate Research, 4, 298–304. Flindt, C., Al-Assaf, S., Phillips, G. O., & Williams, P. A. (2005). Studies on acacia exudate gums: Part V. Structural features of Acacia seyal. Food Hydrocolloids, 19(4), 687–701. Hayashi, H. (2002). Enhancement method of gum arabic under the atmosphere of 30–100% of relative humidity at over 40 1C. Patent Japan 2002 130212; PCT-JP02/08144; WO03/093324A1; US 2005/ 0158440. Idris, O. H. M., Williams, P. A., & Phillips, G. O. (1998). Characterisation of the gums from Acacia senegal trees of different age and location using multi-detection gel permeation chromatograph. Food Hydrocolloids, 12(4), 379–389. Osman, M. E., Menzies, A. R., Williams, P. A., Phillips, G. O., & Baldwin, T. C. (1993). The molecular characterization of the polysaccharide gum from Acacia senegal. Carbohydrate Research, 246, 303–318. Randall, R. C., Phillips, G. O., & Williams, P. A. (1988). The role of the proteinaceous component on the emulsifying properties of gum arabic. Food Hydrocolloids, 2(2), 131–140. Vandevelde, M. C., & Fenyo, J. C. (1985). Macromolecular distribution of Acacia senegal gum (gum arabic) by size-exclusion chromatography. Carbohydrate Polymers, 5(4), 251–273. Zimm, B. H., & Stockmayer, W. H. (1949). The dimensions of chain molecules containing branches and rings. Journal of Chemical Physics, 17(10), 1301–1314.
