Macromolecular Dimensions and Mechanical

Properties of Monolayer Films of

Sonorean Mesquite Gum

Yolanda L. Lopez-Franco,1 Miguel A. Valdez,*2 Javier Hernandez,1 Ana M. Calderon de la Barca,1 Marguerite Rinaudo,3

Francisco M. Goycoolea1*

1 Laboratory of Biopolymers. Centro de Investigacion en Alimentacion y Desarrollo, A.C. (C.I.A.D., A.C.)P.O. Box 1735 Hermosillo, Sonora, 83000, Mexico

2 Departamento de Investigacion en Polımeros y Materiales, Departamento de Fısica, Universidad de Sonora. Blvd.Transversal y Rosales, 83000, Hermosillo, Sonora, MexicoE-mail: valdez@fisica.uson.mx

3 Centre de Recherches sur les Macromolecules Vegetales, C.N.R.S. affiliated with University Joseph Fourier – B.P. 53,38041, Grenoble, Cedex 9, France

Received: April 22, 2004; Revised: June 25, 2004; Accepted: June 25, 2004; DOI: 10.1002/mabi.200400055

Keywords: gum arabic; interfaces; light scattering; mesquite gum; monolayers

Introduction

Mesquite gum is a natural polysaccharide exudated by the

bark of Prosopis spp. trees. This gum was collected by the

Seri Indians settled in the coast of the Gulf of California in

the state of Sonora, Mexico, who used it to prepare eye

drops.[1] In this region of Mexico, mesquite gum (locally

known as ‘‘chucata’’) is sourced mostly from P. velutina

wild trees and is still used in various domestic applications

(e.g., as a hat hardener) and is commonly chewed in the

rural areas.[2] In the state of San Luis Potosi, mesquite gum

is also collected predominantly from P. laevigata wild trees.

It has also been commercialised in limited amounts to be

used as emulsifier in soft drinks as a substitute of gum

arabic, an extensively used hydrocolloid. In previous

studies, it has been demonstrated that the functionality of

mesquite gum is comparable and even superior in certain

conditions to that of gum arabic.[3–7] As a matter of fact, the

Summary: Mesquite gum sourced from Prosopis velutinatrees and gum arabic (Acacia spp.) were characterized usinglight scattering and Langmuir isotherms. Both gum materialswere fractionated by hydrophobic interaction chromatogra-phy, yielding four fractions for both gums: FI, FIIa, FIIb andFIII in mesquite gum and FI, FII, FIIIa and FIIIb in gumarabic. In mesquite gum, the obtained fractions had differentprotein content (7.18–38.60 wt.-%) and macromoleculardimensions (Mw � 3.89� 105–8.06� 105 g �mol�1, RG�48.83–71.11 nm, RH� 9.61–24.06 nm) and architecturegiven by the structure factor (RG/RH ratio �2.96–5.27). Themechanical properties of Langmuir monolayers at the air-water interface were very different on each gum and theirfractions. For mesquite gum, the most active species at theinterface were those comprised in Fractions IIa and IIb andIII, while Fraction I the p/A isotherm lied below that of thewhole gum. In gum arabic only Fraction III developed greatersurface pressure at the same surface per milligram of materialthan whole gum. This is rationalized in terms of structuraldifferences in both materials. Mesquite gum tertiary structureseems to fit best with an elongated polydisperse macrocoil

in agreement with the ‘‘twisted hairy rope’’ proposal forarabinogalactan proteoglycans.

Compression isotherms of gum arabic and mesquite gum.

Macromol. Biosci. 2004, 4, 865–874 DOI: 10.1002/mabi.200400055 � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper 865

Ministry of Health of Mexico has granted authorization for

use of mesquite gum in soft drinks, tablets and feeds.

Chemically, mesquite gum is a highly branched complex

proteoglycan. The primary structure of the polysaccharide

component has been described in detail:[8–10] a central

backbone comprised of b(1! 3)-linked D-galactose resi-

dues, to which side oligosaccharide chains of varying size

are attached at O(6); these branches contain predominantly

D-galactose and L-arabinose (known to occur in both

furanose and pyranose forms) and minor proportions of

D-glucoronate, 4-O-methyl-D-glucuronate and L-rhamnose.

A very similar sugar residue composition is shared by

the carbohydrate components of gum arabic from

Acacia senegal and other exudate gums from the acacia

species.[11–16] Besides the carbohydrate components,

mesquite gum bears a proteinaceous component that

accounts for about �4% of its weight.[7] Hence mesquite

gum can be regarded as an arabinogalactan proteoglycan

(AGP) of the Type II,[17] the same type as gum arabic and

related gums. Up to now, the macromolecular structure of

mesquite gum has been rationalized as highly branched

driving it to a globular disordered conformation in solution.

This would account for its high solubility in water and

Newtonian rheology even at very high polymer concentra-

tion (ca. 50%).[18] However, in this study we revise this

proposal and favor a more elongated structure.

Although the overall composition of carbohydrate

domains of mesquite gum seems to be closely related

to that of gum arabic, both materials can distinctively be

distinguished by different sets of antibodies which spe-

cifically recognize non-reducing chain termini of the

peripheral chains,[19] as well as by crossed immunoelec-

trophoresis.[20] Besides, studies by electrophoresis with

Yariv reagent embedded in the second dimension, have

shown that AGPs from gum arabic differ from those of

mesquite gum[21] and other AGPs of wild tomato.[22]

Fractionation of gum arabic using hydrophobic affinity

chromatography demonstrated that three main fractions can

be isolated.[23] Analysis of the fractions showed that each

contained similar proportions of the various sugars and

differed principally in their molecular masses and protein

contents. The bulk of the gum (84.4 wt.-% of the total) was

shown to be comprised by the so-called arabinogalactan

(AG) fraction with weight-average molecular weight (Mw)

of 2.78� 105 from light scattering measurements and

low in protein (0.35 wt.-%). The second major fraction

(10.4 wt.-% of the total), or AGP, had a higher molecular

weight of 1.45� 106 and contained a greater proportion of

protein (11.18 wt.-%). The third minor fraction (1 wt.-% of

the total), or glycoprotein fraction (GP), had a molecular

weight of 2.5� 105 and the higher protein content (50%

wt.-%). The proteinaceus component of the first two

fractions had similar amino acid distributions, with

hydroxyproline and serine being the most abundant. Work

by Williams et al.,[24] using 1H and 13C NMR spectroscopy

and methylation analysis showed that there were no major

differences in the sugar constituents of the above three

fractions. In a recent study,[25] mesquite gum from

P. laevigata produced both in vitro from stem segments of

mesquite tree and in situ by wild trees was fractionated by

hydrophobic affinity chromatography. Five fractions were

identified for both materials. In the case of gum from wild

trees, Fraction I represented 85 wt.-% of the total gum, had a

molecular mass of 9.3� 105 as determined by gel

permeation chromatography (GPC) (calibrated with dex-

tran standards) and was low in protein (2.3 wt.-%).

Fractions IIa and IIb which represented �11 wt.-% of the

total gum, were found with molecular masses of 6.9� 105

and 5.9� 105, respectively and low protein content (1.4 and

4.3 wt.-%, respectively). Two minor fractions, IIIa and IIIb,

accounted for 3.7 and 1.8 wt.-% of the total gum, had

molecular masses of 1.1� 105 and 3.5� 104, respectively,

and accounted for the greater protein content (35.7 and

50.0 wt.-%, respectively).

For gum arabic, the tertiary structure of the fraction with

species of largest molecular mass (Mw � 1.0� 106), has

been described in terms of a ‘wattle blossom’ macro-

molecular assembly,[23,26,27] by virtue of which few (�5)

discrete polysaccharide domains of Mw � 2� 105 are held

together by a short peptide backbone chain. The key

experimental evidence to support this hypothesis showed

that hydrolysis of the fraction with largest molecular mass

with a protease produced a five-fold reduction in molecular

mass and these fractions were no longer prone to

proteolysis. Alternatively, a ‘‘twisted hairy rope’’ structure

has also been proposed, whereby a linear polypeptide

backbone bears polysaccharide domains in a repeating

pattern.[28] In this proposal, the gum arabic glycoproteins

are represented by a statistical model with a fundamental

7 kDa subunit, where polysaccharide side chains attach to a

polypeptide backbone (Hyp/Pro-rich) of >400 residues

(�150 nm long; 5 nm diameter; axial ratio �30:1) in a

highly regular and ordered fashion (every 10 to 12 residues

repetitive peptide units), forming a twisted hairy rope.

Evidence to this proposal has come mostly from transmis-

sion electron microscopy (TEM) and it is consistent with

the fact that large macromolecules such as AGPs, migrate

through a primary cell wall with a nominal molar mass

cutoff of 68 kDa. Neither of these type of large assemblies

appear to exist in mesquite gum. Instead, the experimental

evidence available to date, argues in favor that mesquite

gum comprises predominantly individual loose proteogly-

can domains. This hypothesis is consistent with differences

in rheology, hydrodynamic molecular dimensions, and area

per molecule adsorbed to a oil-water interface when both

materials have been subjected to similar test condi-

tions.[18,20,21] It is accepted that the complex heterogeneous

and amphiphilic structure of gum arabic is central to its

unique functionality and utilitarian value, namely as an

emulsifier and stabilizer of citrus flavor based soft-

866 Y. L. Lopez-Franco, M. A. Valdez, J. Hernandez, A. M. Calderon de la Barca, M. Rinaudo, F. M. Goycoolea

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

drinks.[29,30] In a recent study, the mechanical properties of

monolayers spread at an air-water interface for whole and

fractionated gum arabic from A. senegal and A. seyal

fractions were investigated. In both gums it was observed

that the maximal surface pressure of whole gum was higher

than the corresponding parameter for each of their fractions,

implying that there was a synergistic effect between the

various proteoglycan molecular species comprising the

whole gum.[31] Such studies have not so far been undertaken

in mesquite gum.

Like gum arabic, mesquite gum also forms and stabilizes

oil-in-water emulsions and has the ability to encapsulate

orange citrus oil during spray drying.[20,32,33] This study

aims to glean further insight to the structure and functional

properties of mesquite gum sourced from the predominant

mesquite species of the plains of the Sonorean Desert,

P. velutina. The gum was fractionated using standard

hydrophobic affinity chromatography in order to keep in

line with previous studies in gum arabic and mesquite gum.

Multi-angle laser light scattering and dynamic light

scattering were used to characterize the molecular dimen-

sions and the architecture of the obtained gum fractions.

The behavior of the individual fractions in monolayers

adsorbed at the air-water interface was also studied using a

Langmuir balance in order to gain understanding of the

relationship between the macromolecular structure of

the various fractions and the phenomena involved in the

adsorption process. The gained knowledge is compared

with the properties of gum arabic characterized under

similar experimental protocol.

Experimental Part

Mesquite gum sample was a batch of hand sorted tears (Grade‘‘A’’) selected from a gum stock gathered in the spring of 1998from various local suppliers in Hermosillo, Sonora, Mexico.The gum arabic (Acacia spp.) sample and the rest of thechemical substances were all analytical grade and obtainedfrom Sigma-Aldrich Chemicals (St Louis, MO). Milli-Q waterwas obtained by a filtration system (Millipore Inc. Mexico) andused throughout. Mesquite gum sample was pre-treated bydissolving it in water at �4.5 vol.-%, filtered through acellulose filter in a pilot plant system (Zeta Plus1, Cuno Inc.USA) and freeze-dried until utilized.

Hydrophobic Interaction Chromatography

A column of dimensions 2.5� 24 cm was packed with phenyl-Sepharose CL–4B gel (Pharmacia, Uppsala, Sweden) using4.2 mol � dm�3 NaCl as eluent solution. A 100 cm3 volume ofa 10 wt.-% solution of mesquite gum in 4.2 mol � dm�3

NaCl was applied to the column and eluted successively by4.2 mol � dm�3 NaCl, 2.0 mol � dm�3 NaCl and finally water.The flow rate was set to 40 cm3 � h�1. Absorbance wasmonitored at 280 nm by using a low-pressure chromatographicsystem (Bio-Rad, CA, USA). The obtained fractions weredialyzed extensively against water and freeze-dried.

Protein

The protein content in the various samples was determinedusing a combustion N analyzer LECO Mod. FP 528 (LecoMexico). Conversion factors of 6.53 and 6.60 were used inthe calculations of mesquite gum[34] and gum arabic,[35]

respectively.

Total Sugars

The content of total sugars was determined by the phenol-sulfuric method.[36]

Optical Rotation

Optical rotation measurements were recorded for 1 wt.-%aqueous gum solutions in a cell of path length 1 dmthermostated at 20� 0.1 8C. Angular rotation measurementswere recorded using a Perkin Elmer 341 polarimeter.

Light Scattering Measurements

The static and dynamic light scattering measurements wereperformed using an ALV-5000 digital correlator system(Langen-GmbH, Germany) fitted with a temperature controlset at 25� 0.1 8C. The scattered light, vertically polarized withan l0¼ 632 nm Argon laser (30 mW), was measured atdifferent angles in the range of 408–1508. The reduced elasticscattering I(q)/KC, with K¼ 4p2n0

2(dn/dc)2(I90/R90)/l04NA, was

measured in scattering angle steps of 108, where n0 is therefractive index of the standard (toluene), I90and R90are theintensity and the Rayleigh ratio of the standard at y¼ 908,respectively, C is the sample concentration in g � cm�3, I(q) isthe intensity scattered by the sample after subtraction ofthe scattering contribution of solvent, and NA is Avogadro’snumber. The increment of the refractive index (dn/dc) wastaken as 0.150 ml � g�1, a value previously found for varioussamples of gum arabic in several previous studies.[37]

This same dn/dc value was also used for mesquite gum andfor the fractions of both materials. The scattering angle q isgiven by q¼ (4pn/l0)sin(y/2) and n is the refractive index ofthe medium (n¼ 1.33). The data were analyzed using the wellestablished Zimm equation.[38]

Kc

Ry¼ 1

MwPðyÞþ 2A2C ðC ! 0; y ! 0Þ ð1Þ

where Ry is the Rayleigh ratio for the angle y, A2 is the secondvirial coefficient of the molecule, Mw is the weight-averagemolecular mass and P(y) is the form factor which depends onthe shape of molecules and is approached for small q by1 � 1

3R2

Gq2 and RG is the radius of gyration of the molecule.Several form factors were tested for different geometriesaccording to the information given in the software of theinstrument (ALV 5000/E/WIN Software).

The hydrodynamic radius, RH, was obtained for dilutedsamples from dynamic light scattering measurements at anincidence angle of 908 through the Stokes-Einstein relation:

D0 ¼ kBT

6pZRH

ð2Þ

Macromolecular Dimensions and Mechanical Properties of Monolayer Films of Sonorean Mesquite Gum 867

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

where kB is the Boltzmann constant, T is the absolute temper-ature and Z the viscosity of the solvent. Samples were dissolvedin 20-mmol � dm�3 NaCl solutions at 3, 5, 7 and 10 mg �ml�1.Solutions were filtered twice with 0.8mm filters (Millipore) andonce with 0.2 mm filter.

p/A Isotherms

Compression isotherms (p/A diagrams) were established using aNima Langmuir-Blodgett balance (Model 611) at 25� 0.1 8C.Instrument surface tension precision was 0.1 mN �m�1. Thecompression rate was kept constant at 50 cm2 �min�1 and thesubphase was 1 mol � dm�3 KCl solution.

Avolume of 100 ml aqueous solution (10 wt.-%) of a samplewas spread onto the water surface using a Hamilton microsyringe with its needle very near the surface. Purity of thesubphase was checked by compression of the surface withoutsample. The surface was regarded as clean until the pressurewas lower than 0.5 mN �m�1. The elapsed time after spreadingthe sample and before compressing was 40 min. The pH valuesof whole mesquite gum and whole gum arabic solutions were�4.5 and �4.3, respectively.

Results and Discussion

Hydrophobic Interaction ChromatographicFractionation

Figure 1 shows the elution chromatographs corresponding

to the separation of mesquite gum and gum arabic in order

of increasing hydrophobicity. The observed UV profiles

(l¼ 280 nm) are proportional to the proteic fraction

associated with the eluting species. Gum recovery, protein

contents and specific optical rotation for these fractions are

given in Table 1. Notice that both materials yield three

major fractions following step-wise elution with 4.2 mol �dm�3 NaCl, 2.0 mol � dm�3 NaCl and water. The peak

described by the least hydrophobic component in mesquite

gum (Fraction I) shows an almost identical elution pattern

to the equivalent fraction of gum arabic. This peak

represents the major component of mesquite gum material,

comprising 94.77 wt.-% of the recovered gum and has a

7.18 wt.-% protein content. This fraction can be regarded as

the AG fraction. The second peak in mesquite gum eluted

with 2.0 mol � dm�3 NaCl and appeared split in further two

sub-fractions. These were labeled as IIa and IIb and studied

separately and can be regarded as the constituents of an

AGP fraction. Fraction IIa accounted for 2.49 wt.-% of the

Table 1. Physicochemical analysis of fractions of mesquite gum (P. velutina) and gum (Acacia spp.), isolated by hydrophobic affinitychromatography.

Material Fraction Eluent Gum recovery Total sugars Protein [a]D58920

% % %

Mesquite Gum Whole 93.13 5.90a) þ55.10(P. velutina) I 4.2 mol � dm�3 NaCl 94.77 91.50 7.18a) þ57.15

IIa 2 mol � dm�3 NaCl 2.49 76.80 23.20b) þ46.64IIb 2 mol � dm�3 NaCl 1.17 n.d.c) n.d.c) n.d.c)

III H2O 1.56 61.40 38.60b) þ27.61Gum Arabic Whole 99.20d) 2.24d) �28.77(Acacia spp.) I 4.2 mol � dm�3 NaCl 89.08 97.00d) 0.35d) �26.29

II 2 mol � dm�3 NaCl 9.40 91.20d) 11.18d) �32.52IIIa H2O 1.42 45.20d) 47.30d) n.d.c)

IIIb H2O 0.10 n.d.c) 50.00d) n.d.c)

a) Protein conversion factor of 6.53 for mesquite gum[34] and 6.60 for gum arabic.[35]

b) Calculated by difference from the total sugar content.c) n.d. not determined because sample amount was not enough.d) Randall et al.[23]

Figure 1. Elution UV profile (l¼ 280 nm) of mesquite gum andgum arabic following hydrophobic interaction chromatography onPhenyl-Sepharose CL-4B. Fractions were eluted (step-wise) using4.2 mol � dm�3 NaCl, 2.0 mol � dm�3 NaCl and MilliQ water.

868 Y. L. Lopez-Franco, M. A. Valdez, J. Hernandez, A. M. Calderon de la Barca, M. Rinaudo, F. M. Goycoolea

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

recovered gum and had a 23.2 wt.-% protein content, while

the recovered amount of Fraction IIb was not enough to

allow its analytical characterization. A third fraction was

eluted as a single peak and represented 1.56 wt.-% of the

recovered weight of the gum and had the highest protein

content, 38.6 wt.-%, and can be compared with the GP

fraction proposed for gum arabic. In the case of gum arabic,

the first major peak represented 89.08 wt.-% of the material

(Fraction I or AG fraction). The second peak, Fraction II

(AGP fraction), eluted as single monomodal profile,

represented 9.40 wt.-%, while the third peak, Fraction III

(GP fraction) appeared split in two fractions and both

accounted for 1.52 wt.-%. The results found for fractionated

gum arabic compare very well with those previously

documented data using an identical experimental protocol

for hydrophobic chromatographic separation.[23,29,37]

While the results on mesquite gum almost coincide with

those of Orozco-Villafuerte et al.,[25] differing essentially

in that Fraction III is reported as eluting as two components.

This difference may stem on subtle structural differences in

the glycoprotein components associated with the distinctive

species botanical origin.

Further physicochemical information of mesquite gum

and gum arabic and their various comprising fractions was

provided by the data given in Table 1 of the specific optical

rotation, [a]D20. It is interesting to notice that as the hydro-

phobic nature of the constituent species increases in

mesquite gum, the material becomes less dextrorotatory.

It can be argued that this is a consequence of the varying

composition in the protein fraction or in the nature of the

sugar fraction. It is also shown that [a]D20 is nearly inde-

pendent on the protein yield.

Macromolecular Dimensions Characterization

Until recently there was almost no biophysical data about

the size, conformation and molecular weight of mesquite

gum. We have used multi-angle laser light scattering to

deepen the present understanding of the macromolecular

properties of mesquite gum and its fractions and compare

them with those of gum arabic. Figure 2 shows representa-

tive Zimm plots obtained for both whole materials obtained

by static light scattering used to derive RG and Mw data.

Gum arabic has been investigated for several decades

due to its widespread use worldwide. According to Randall

et al.,[23] the AGP fraction of gum arabic has an RH¼22.8� 0.5 nm corresponding to a molecular mass of

1.45� 106. The AG fraction had a RH¼ 9.2� 0.5 nm

corresponding to a molecular mass of 2.79� 105. Similar

results were found by Osman et al.[39] They identified the

individual AG, AGP and GP components and found a RH for

the AG fraction between 8 and 12 nm with a small propor-

tion of the AGP fraction with an RH¼ 30 nm.

Islam et al.[40] found that the molecular weight of the

whole gum arabic was about 560 000 and their results

were consistent with the ‘‘wattle blossom’’ proposal to

describe the tertiary structure of the AGP fraction by virtue

of which a number of polysaccharide domains are linked to

a common polypeptide chain, a model proposed originally

by Fincher et al.[17] for AGPs. Later, Connolly et al.[26]

calculated the molecular mass of the blocks of polysacchar-

ide obtained after treatment with pronase of the order of

2� 105. Williams and Langdon[41] undertook GPC studies

monitoring the column eluent simultaneously by UV,

refractive index, on-line photon correlation spectroscopy

and on-line multi-angle laser light scattering (MALLS)

detection, measured the RH and RG of gum arabic. They

found that the ratio RG/RH was a function of the shape of the

molecule. For spheres this ratio has a value of 0.788 while

for coils and rods it has a value >2.0. A ratio RG/RH� 1 for

all the eluting species was consistent with the proposal of

the highly branched compact structure. One of the few

works that has addressed the value of RH of whole mesquite

Figure 2. Zimm plot of gum arabic (a) and mesquite gum (b) in2 mmol � dm�3 NaCl at 25 8C.

Macromolecular Dimensions and Mechanical Properties of Monolayer Films of Sonorean Mesquite Gum 869

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

gum is by Goycoolea et al.[18] Using dynamic light scat-

tering to investigate the dependence of RH on the ionic

strength, the values of RH were found to vary from 8.2 to

10.6 nm. However, no data of RG were available to date on

mesquite gum and its fractions.

The results for Mw, RH and RG on mesquite gum and gum

arabic and their fractions are given in Table 2. As observed

for the case of gum arabic, the results for Mw for the

Fractions I and II are similar to the ones found by Randall

et al.[23] A difference of 20% is found in the case of our Mw

for the whole gum arabic and the value found by Islam

et al.[40] When we compared our own results with the ones

found by Williams and Langdon,[41] it was found that the

values of the ratio RG/RH� 1� 0.5, for whole gum and

Fraction III of gum arabic were in reasonably good

agreement. However, note that for the Fraction II of gum

arabic, RG/RH> 2, a result that is in conflict with the idea

that the AGP fraction has a predominantly branched

compact structure assumed by the ‘wattle blossom’

assembly. Indeed, a ‘‘twisted hairy rope’’ model has been

proposed to account for the tertiary structure of gum arabic

AGP fraction as a competing model to the ‘wattle blossom’

one.[28] Our values agree reasonably well with the

previously reported Mw values for mesquite gum fractions

separated using a GPC system calibrated with protein and

dextran standards.[25] However, the use of standards may

not give very accurate results in elongated macromolecules.

The value of RH found in this study for whole mesquite

gum is very close to the one previously found.[18] Yet, the

ratio RG/RH> 2 for whole and fractionated species, indicate

the possibility of an elongated structure for all the fractions

of mesquite gum. This result suggests the existence of

important differences in the structure between the corres-

ponding fractions of mesquite gum and gum arabic.

Figure 3a and 3b show the unweighted size (RH) dis-

tribution curves of both gum materials and their constituent

fractions. The various macromolecular species that com-

prise gum arabic show broad dispersion in size spanning

three orders of magnitude (2 to 1 000 nm). There is close

similarity in the monomodal distribution curves of whole

gum and Fraction I, as this fraction accounts for most of

the weight of the gum. Close inspection of the curves show

that Fraction I species have overall somewhat lower RH

values than do the species comprising the bulk material. In

turn, it can be appreciated that Fraction II has the narrowest

dispersion and it is also monomodal. Further inspection of

the data, reveals that molecular species of Fraction IIIa are

split under a bimodal size distribution. Fraction IIIb appears

to be of similar size than the larger species of Fraction IIIa.

In the case of mesquite gum, the registered size distribution

curves of the bulk material and its fractions (Figure 3b)

described very similar patterns to those of gum arabic.

Indeed, Fraction III also showed a clear bimodal size

distribution.

Dynamic light scattering studies with native mesquite

gum and gum arabic and their fractions indicate that all the

samples show some degree of polydispersity, so that our

Table 2. Molecular size for whole and fractionated mesquitegum in comparison with whole and fractionated gum Arabic.

Material Fraction Mw RG RH RG/RH

g �mol�1 nm nm

Mesquite gum Whole 3.86� 105 50.47 9.48 5.32(P. velutina) I 3.89� 105 49.82 9.61 5.27

IIa 4.84� 105 55.30 13.17 4.20IIb 3.82� 105 48.83 15.68 3.08III 8.06� 105 71.11 24.06 2.96

Gum Arabic Whole 4.39� 105 35.02 22.36 1.57(Acacia spp.) I 2.92� 105 52.26 14.96 3.49

II 1.19� 106 47.64 18.52 2.57IIIa 8.88� 105 40.93 43.58 0.94IIIb 1.34� 106 69.93 66.20 1.06

Figure 3. Unweighted probability distribution curves of thehydrodynamic radius of gum arabic and its fractions (a) andmesquite gum and its fractions (b).

870 Y. L. Lopez-Franco, M. A. Valdez, J. Hernandez, A. M. Calderon de la Barca, M. Rinaudo, F. M. Goycoolea

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

results of hydrodynamic radius must be interpreted

cautiously and as average values. Due to this polydispersity

and/or aggregation, perhaps as a consequence of the

polyelectrolyte character of the samples and concentration,

there are more than two relaxation modes in the correlation

curves and hence it makes very difficult to differentiate

them in pure slow and a fast relaxation modes, under a

treatment similar to the one conducted by Buhler and

Rinaudo on chitosan, a semiflexible biopolymer, using a

sample of low polydispersity (Mw=Mn ¼ 1.3).[42]

Static laser light scattering was used to characterize the

average molecular weight and the architecture (radius of

gyration) of the obtained gum fractions. By computing

the factor form for different molecular structures with

the available different configurations of the instrument, we

observed that whole mesquite gum and its fractions were

best fit by a model for a polydisperse coil structure with a

ratio Mw=Mn ¼ 2. In Figure 4 we show the Kratky function,

P(U)*U2 vs U for the three fractions of mesquite gum

(Figure 4d, 4e and 4f) and gum arabic (Figure 4a, 4b and 4c),

where U¼ q�Rg. Inspection of the curves of mesquite

gum Fractions I (Figure 4d), IIb (Figure 4e) and III

(Figure 4f) shows that in all cases the experimental data

deviate significantly from a globular (hard sphere) structure

and follow closely the polydisperse coil model. The result

for Fraction I is not very clear because of the large

fluctuation occurred in the experiment. While in the case of

gum arabic, Fractions I (Figure 4a), II (Figure 4b) and IIIa

(Figure 4c) all could be fitted by the three different models,

including the hard sphere one.

p/A Isotherms

Whole mesquite gum and gum arabic isotherms were

obtained by spreading the samples and compressing the film

balanceand then measuring thechange in thesurfacepressure

at the air-water interface. The corresponding compression

isotherms of the three fractions of both arabic and mesquite

gums are shown in Figure 5a and Figure 5b, respectively. It is

important to stress the fact that gum samples in the present

study were not dissolved in a mixture of 2-propanol/water as

in the method used by Fauconnier et al.[31]

Inspection of Figure 5a shows that the bulk material and

three fractions of gum arabic present essentially the same

behavior at large surface area per milligram of gum. Further,

as the surface areas are reduced during compression there is a

gradual increase in surface pressure and some kind of

condensed phase is invariably attained in all cases. Closer

inspection shows that the Fraction IIIa isotherm lies at

somewhat greater values than whole gum and the rest of the

Figure 4. Form factor for fractions I (a), II (b) and IIIa (c) from gum arabic and fractions I (d), IIb (e) and III (f) frommesquite gum obtained from static and dynamic light scattering.

Macromolecular Dimensions and Mechanical Properties of Monolayer Films of Sonorean Mesquite Gum 871

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

fractions even when both isotherms follow very similar

profiles. This is probably due to the fact that this fraction

contributes mostly to the hydrophobic properties of the gum

arabic, as it bears the highest protein content (Table 1).

Differences in amino acid composition profiles among the

various fractions, may also underlie their capacity to adsorb

at the air-water interface. It is expected that non-polar amino

acids will contribute significantly to the monolayer isotherm.

Another way of analyzing the p/A curves is by comparing

the surface area per milligram of gum at a fixed similar

surface pressure (5 mN �m�1). We observe that curves of

Fractions I and II of gum arabic achieve this pressure at

smaller areas than do whole gum and Fraction III species.

This could be directly related to the protein content, size

and/or shape (i.e., related to RG/RH) of the involved

molecules adsorbed at the interface. Synergy between the

AG, AGP and GP fractions of gum arabic has been sug-

gested to account for the greater surface pressure at the end

of compression (maximal surface pressure) attained by

whole gum arabic, which was higher than the corresponding

parameter for each fraction.[31] Our results seem to confirm

this hypothesis, as the contribution of Fraction III in weight

terms to the weight of whole gum is very small indeed so as

to claim an additive effect. Notice that the GP fraction

(Fraction IIIa and IIIb), has the greater RG values and lower

RG/RH ratio, indicative of a large yet spherical species. The

consequence of these features on the adsorption behavior

might be to allow closer packing at the air-water interface,

thus effectively attaining a lower surface tension (i.e.,

greater surface pressure) at smaller areas. Fraction II, or

AGP fraction, has a protein content of 11.18 wt.-% and a

RG/RH ratio of 2.57 (Table 2), indicative of a different

conformation than the spheroid one of Fraction IIIa and

IIIb, which is presumably closer to that of an elongated rod.

Inspection of the p/A isotherms of mesquite gum

(Figure 5b) reveals important differences in monolayer

properties among the various analyzed fractions. Notice

that isotherm of Fraction I lies consistently below the one of

the whole gum and starts to rise at lower surface area

(�0.003 m2 �mg�1). By contrast, isotherms of Fraction IIa,

IIb and III start to show elevation of the surface pressure at

the onset of the compression cycle. Besides, Fraction IIa

attains the highest surface pressure (�37 mN �m�1) at the

end of the compression cycle compared with the other

fractions, including those of gum arabic. In order to account

for the observed mechanical properties of the monolayers of

mesquite gum and its fractions differences in protein

content, macromolecular dimensions and conformation of

the constituent species need be considered. Fraction I, with

7.18 wt.-% protein content produces the lowest surface

pressure, roughly in line with Fraction I of gum arabic.

Although the protein content of this fraction appears to be

greater by an order of magnitude than Fraction I of gum

arabic, the presence of protein by itself is not sufficient

so as to modify the interfacial adsorption. Therefore, the

importance of the conformation and topology of the

molecule and the hydrophilic/hydrophobic sites adsorbed

at the interface is highlighted. It is interesting to notice that

Fraction I of mesquite gum had a large deviation from the

globular conformation as assessed from the RG/RH ratio

�5.27, pointing towards an elongated structure. This

coincides with the isotherms of the fractions of gum arabic,

where also Fraction I with the highest value of RG/RH give

the lowest surface pressure value at smaller areas. By

contrast, the species curves with lower RG/RH ratios start

rising at larger areas (Fractions IIb and III).

Comparing the behavior for the isotherms of both whole

gums, we observe that their isotherms reach the surface

pressure of 5 mN �m�1 at the approximate the same area

(�0.002 m2 �mg�1). In spite of the fact that two isotherms

of the fractions of mesquite gum reached the highest

pressure, gum arabic reached a higher surface pressure at

the smallest compressed area compared with the value

reached by the whole mesquite gum. This can be explained

with the fact that mesquite gum contains a somewhat larger

Figure 5. Compression isotherms of gum arabic and its fractions(a) and mesquite gum and its fractions (b).

872 Y. L. Lopez-Franco, M. A. Valdez, J. Hernandez, A. M. Calderon de la Barca, M. Rinaudo, F. M. Goycoolea

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

proportion of the Fraction I, which does not produce a

noticeable surface pressure as can be observed in Figure 5b.

The analysis of the monolayer mechanical properties was

complemented by deriving the parameters for the limiting

area, namely A0 and e. A0 denotes the limiting area which is

determined by extrapolation at the intersection of the abs-

cissa axis with the tangent of the isotherm atp¼ 2 mN �m�1

and e ¼ �A dpdA

is the film elasticity measured from the

isotherm curve at the final part (pressure around

20 mN �m�1) of the isotherm.

From the results shown for both gums in Table 3, we

observe that A0 follows the same behavior for both fractions

of the gums. The value for A0 grows in the order Fraction

I< Fraction II< Fraction III. Whole and Fraction I of both

gums share very close A0 values among themselves. The

result agrees with the one found by Fauconnier et al.[31] for

two different samples of gum arabic. However, in our case,

the value of A0 for the whole samples is larger than the one

for the corresponding A0. This result agrees well with the

fact that the RH of Fraction I is the smallest for both gums

fractions. A0 indicates monolayer expansion and in our case

we observe that Fraction III is the most expanded for both

gums. In the case of gum arabic, Fraction III together

with Fraction II (AGP fraction) give the most important

contribution to the interfacial properties of whole gum

arabic isotherm.[31]

We also notice that fractions of both mesquite and arabic

samples show the same behavior for monolayer expansion.

This is larger for Fraction III, then follows Fraction II and

this value is shorter for Fractions I. This is according

with the behavior of the hydrodynamic radius as also

noticed by Fauconnier et al.[31] Comparing A0 for each

fraction of both gums, we notice that A0 is almost the same

for the whole gums and is larger for Fractions II and III of

mesquite gum than the corresponding fractions of gum

arabic, indicating strong differences of the molecules at the

air-water interface.

In the case of the film elasticity e; this parameter

measures the film elasticity and the resistance to the change

of area.[43] We observe in Table 3 that most of the fractions

and the whole gum of gum arabic produce more resistant

films than the corresponding ones of mesquite gum. Only

the film of Fraction IIa of mesquite gum results more

resistant than any other measured film of both gums. As

mentioned by Fauconnier et al.[31] this behavior explains

the excellent emulsifying properties of gum arabic. We

compared the behavior of both whole gums at the air-water

interface by performing one complete cycle by compres-

sing and expanding to the original area. Figure 6 shows

the results of this experiment for both gums. Clearly, the

hysteresis area is very small probably because of the

flexibility of molecules at the interface.[44] On the other

hand, the curves do not return to the original pressure, this

would mean that compressing and expanding molecules at

the interface produces some changes on the molecular

structure that could improve their hydrophobicity. These

Table 3. Limiting area and film elasticity extracted fromisotherm curves of gum arabic (Acacia spp.) and its fractionsand mesquite gum (P. velutina) and its fractions.

Material Fraction A0 e

m2 �mg�1 mN �m�1

Mesquite Gum Whole 3.50� 10�3 19.20(P. velutina) I 2.37� 10�3 20.00

IIa 5.10� 10�3 34.50IIb 5.97� 10�3 12.01III 5.64� 10�3 5.26

Gum Arabic Whole 3.55� 10�3 33.40(Acacia spp.) I 2.88� 10�3 24.80

II 4.30� 10�3 25.28IIIa 4.95� 10�3 25.44IIIb n.d.a) n.d.a)

a) n.d. not determined because sample amount was insufficient.

Figure 6. Cycle of compression and expansion of the spreadmonolayer of gum arabic (a) and mesquite gum (b).

Macromolecular Dimensions and Mechanical Properties of Monolayer Films of Sonorean Mesquite Gum 873

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

phenomena indicate that both samples behave as real mono-

layers at the air-water interface and that most of molecules

at the interface stay bounded at the interface after com-

pressing and expanding the area of the film without loosing

any appreciable amount of molecules at the interface.

Conclusion

The characterization of mesquite gum showed differences

in its chemical surface properties in comparison with gum

arabic. According to the structure factor and light scattering

evidence, the macromolecular structure of mesquite gum

resembles a polydisperse macrocoil in agreement with the

‘‘twisted hairy rope’’ proposal advanced for gum arabic

AGP in previous studies. Such structural differences be-

tween the constituent macromolecular species comprising

both materials underlie also different adsorption and mec-

hanical properties when they are adsorbed as a monolayer at

the air-water interface. They also confirm that mesquite

gum is an effective natural emulsifier.

Acknowledgements: CONACYT is gratefully acknowledgedfor grant No. ER074 ‘‘Materiales Biomoleculares’’ and forstudentship support to one of us (Y. L. Lopez-Franco).

[1] R. Felger, M. B. Moser, Econ. Bot. 1974, 28, 414.[2] J. R. Balderrama, ‘‘Caracterizacion Fisicoquımica y Anali-

sis del Aprovechamiento de la Goma ‘Chucata’ y Galacto-manana del Mezquite (Prosopis spp.) como PosiblesHidrocoloides Alimentarios’’, BSc Thesis, Universidad deSonora. Hermosillo, Sonora 1998.

[3] J. E. Vernon-Carter, C. L. Beristain, R. Pedroza-Islas,‘‘Mesquite Gum (Prosopis Gum)’’, in: Novel Macromole-cules in Food Systems, Developments in Food Science,G. Doxastakis, V. Kiosseoglou, Eds., Elsevier, Amsterdam2000, p. 217.

[4] C. I. Beristain, E. J. Vernon-Carter, Drying Technol. 1994,12, 727.

[5] C. I. Beristain, H. S. Garcıa, E. J. Vernon-Carter, Lebensm.Wiss. Technol. 2001, 34, 398.

[6] C. I. Beristain, E. Azuara, E. J. Vernon-Carter, J. Food Sci.2002, 67, 206.

[7] F. M. Goycoolea, A. M. Calderon de la Barca, G. Hernandez,J. R. Valenzuela, J. R. Balderrama, ‘‘Immunochemical,Structural and Functional Properties of Mesquite Gum inComparison with Gum Arabic’’, in: Cell and DevelopmentalBiology of Arabinogalactan-Proteins, E. A. Nothnagel, A. E.Clarke, A. Bacic, Eds., Kluwer Academic/Plenum Publish-ers, New York 2000, pp. 263–276.

[8] G.O.Aspinall,C.C.Whitehead,Can.J.Chem.1970,48,3850.[9] G.O.Aspinall,C.C.Whitehead,Can.J.Chem.1970,48,3840.

[10] D. M. W. Anderson, G. K. Farquhar, Int. Tree Crops J. 1982,2, 15.

[11] P. Jurasek, G. O. Phillips, Food Hydrocolloids 1993, 7, 337.[12] S. Biswas, B. Biswas, G. O. Phillips, Food Hydrocolloids

1995, 9, 151.[13] M. E. Osman, P. A. Williams, A. R. Menzies, G. O. Phillips,

J. Agric. Food Chem. 1993, 41, 71.

[14] C. A. Street, D. M. W. Anderson, Talanta 1983, 30, 887.[15] J. Defaye, E. Wong, Carbohydr. Res. 1986, 150, 221.[16] A. M. Stephen, ‘‘Structure and Properties of Exudate

Gums’’, in: Gums and Stabilisers for the Food IndustryVol. 5, G. O. Phillips, P. A. Williams, D. J. Wedlock, Eds.,IRL Press on Oxford University, 1990, p. 3.

[17] G. B. Fincher, B. A. Stone, A. E. Clarke, Annu. Rev. PlantPhysiol. 1983, 34, 47.

[18] F. M. Goycoolea, E. R. Morris, R. K. Richardson, A. E. Bell,Carbohydr. Polym. 1995, 27, 37.

[19] F. J. Miskiel, J. H. Pazur, Carbohydr. Polym. 1991, 16, 17.[20] F. M. Goycoolea, A. M. Calderon de la Barca, J. R. Balderrama,

J. R. Valenzuela, Int. J. Biol. Macromol. 1997, 21, 29.[21] F. M. Goycoolea, M. A. Valdez, G. A. Alvarez, F. Soto,

J. Hernandez, G. Hernandez, J. Lizardi, A. M. Calderon de laBarca, J. R. Valenzuela, A.Cardenas, Manuscript in preparation.

[22] G. J. van Holst, A. E. Clarke, Plant Physiol. 1986, 80, 786.[23] R. C. Randall, G. O. Phillips, P. A. Williams, Food

Hydrocolloids 1989, 3, 65.[24] P. A. Williams, G. O. Phillips, A. M. Stephen, Food

Hydrocolloids 1990, 4, 305.[25] J. Orozco-Villafuerte, F. Cruz-Sosa, E. Ponce-Alquicira,

E. J. Vernon-Carter, Carbohydr. Polym. 2003, 54, 327.[26] S. Connolly, J. C. Fenyo, M. C. Vandevelde, Food

Hydrocolloids 1987, 1, 477.[27] S. Connolly, J. C. Fenyo, M. C. Vandevelde, Carbohydr.

Polym. 1988, 8, 23.[28] W.Qi,C.Fong,D.T. A.Lamport,PlantPhysiol.1990,96, 848.[29] A. K. Ray, P. B. Bird, G. I. Iacobucci, B. J. Clark, Food

Hydrocolloids 1995, 9, 123.[30] R. A. Buffo, G. A. Reineccius, G. W. Oehlert, Food

Hydrocolloids 2001, 15, 53.[31] M. L. Fauconnier, C. Blecker, J. Groyne, H. Razafindra-

lambo, E. Vanzeveren, M. Marlier, M. Paquot, J. Agric. FoodChem. 2000, 48, 2709.

[32] E. J. Vernon-Carter, S. A. Gomez, C. I. Beristain,G. Mosqueira, R. Pedroza-Islas, R. C. Moreno-Terrazas,J. Texture Stud. 1996, 27, 625.

[33] C. I. Beristain, E. Azuara, H. S. Garcıa, E. J. Vernon-Carter,Int. J. Food Sci. Technol. 1996, 31, 379.

[34] D. M. W. Anderson, W. Weiping, Food Hydrocolloids 1989,3, 235.

[35] D. M. W. Anderson, D. M. B. Douglas, N. A. Morrison,W. Weiping, Food Addit. Contam. 1990, 7, 303.

[36] M. Dubois, K. A. Gilles, J. K. Hamilton, P. A. Rebers,F. Simth, Anal. Chem. 1956, 28, 350.

[37] [37a] M. C. Vandevelde, J. C. Fenyo, Carbohydr. Polym.1985, 5, 251; [37b] A. Veis, D. N. Eggenberger, J. Am. Chem.Soc. 1954, 76, 1560; [37c] H. A. Swenson, H. M. Kaustinen,O. A. Kaustinen, N. S. Thompson, J. Polym. Sci. 1968, A-2, 6,1593; [37d] D. M. W. Anderson, I. C. M. Dea, J. Soc. Cosmet.Chem. 1971, 22, 61.

[38] S. F. Sun, ‘‘Physical Chemistry of Macromolecules: BasicPrinciples and Issues’’, Wiley, New York 1994, p. 188.

[39] M. E. Osman, A. R. Menzies, P. A. Williams, G. O. Phillips,T. C. Baldwin, Carbohydr. Res. 1993, 246, 303.

[40] A. M. Islam, G. O. Phillips, A. Sljivo, M. J. Snowden, P. A.Williams, Food Hydrocolloids 1997, 11, 493–505.

[41] P. A. Williams, M. J. Langdon, Chromatogr. Anal. 1995, 5,October/November.

[42] E. Buhler, M. Rinaudo, Macromolecules 2000, 33, 2098.[43] D. E. Graham, M. C. Phillips, J. Colloid Interface Sci. 1980,

76, 227.[44] M. Cornec, D. A. Kim, G. Narsimhan, Food Hydrocolloids

2001, 15, 303.

874 Y. L. Lopez-Franco, M. A. Valdez, J. Hernandez, A. M. Calderon de la Barca, M. Rinaudo, F. M. Goycoolea

Macromol. Biosci. 2004, 4, 865–874 www.mbs-journal.de � 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim