Está en la página 1de 8

Colloids and Surfaces A: Physicochem. Eng.

Aspects 481 (2015) 367–374

Contents lists available at ScienceDirect

Colloids and Surfaces A: Physicochemical and


Engineering Aspects
journal homepage: www.elsevier.com/locate/colsurfa

Influence of main whey protein components on the mechanism of


complex coacervation with Acacia gum
Delphine Ach a , Stéphanie Briançon a , Vincent Dugas b , Jocelyne Pelletier a , Guy Broze c ,
Yves Chevalier a,∗
a
Université Lyon 1, Laboratoire d’Automatique et de Génie des Procédés, CNRS UMR 5007, 43 bd du 11 Novembre, 69622 Villeurbanne, France
b
Université de Lyon, Université Claude Bernard Lyon 1, Institut des Sciences Analytiques, CNRS UMR 5280, 5 rue de la Doua, 69100 Villeurbanne, France
c
Prayon S.A., rue J. Wauters 144, 4480 Engis, Belgium

h i g h l i g h t s g r a p h i c a l a b s t r a c t

• The two main components of whey Quantification of ␣-lac and ␤-lg in whey proteins/Acacia gum coacervate by capillary gel electrophoresis.
protein isolate, ␤-lactoglobulin and
␣-lactalbumin, undergo complex
coacervation with Acacia gum.
• ␤-lactoglobulin has stronger binding
to Acacia gum than ␣-lactalbumin.
• A residual concentration of proteins
remains in the aqueous phase in all
instances.

a r t i c l e i n f o a b s t r a c t

Article history: Complex coacervation between whey proteins isolate (WPI) and Acacia gum was investigated in order
Received 18 March 2015 to disclose the roles and the contributions of each component of WPI to the formation of the complex
Received in revised form 1 June 2015 coacervate. The main aim was to establish the balance of the main components of WPI, ␤-lactoglobulin
Accepted 2 June 2015
and ␣-lactalbumin, during the phase separation of complex coacervate. The complex coacervation of Aca-
Available online 10 June 2015
cia gum and pure ␤-lactoglobulin, pure ␣-lactalbumin, and WPI have been investigated and compared
together by means of macroscopic observations and capillary gel electrophoresis analyses performed
Keywords:
along pH scans from basic to acidic medium. Coacervate composition was influenced by the pro-
Complex coacervation
Phase separation
tein/polysaccharide (Pr:Ps) ratio and pH. An optimum pH for best coacervation yield was found for each
Capillary electrophoresis Pr:Ps ratio. Non-negligible concentrations of ␤-lactoglobulin and ␣-lactalbumin remain in solution in
Whey proteins isolate most instances. ␤-lactoglobulin undergoes complex coacervation stronger than ␣-lactalbumin in their
Acacia gum competitive coacervation of WPI and Acacia gum.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction products [1–3]. Two different biopolymers mixed together may


either form a one-phase or a two-phase system. Biopolymers
Proteins and polysaccharides are widely used polymers in the mixtures often undergo a phase separation. Two types of phase
food industry especially in emulsions and encapsulation based separation can be observed in biopolymers mixtures: the segrega-
tive phase separation and the associative phase separation. The
segregative phase separation occurs when the protein and the
polysaccharide are repelling each other because of their thermo-
∗ Corresponding author. Fax: +33 472431682. dynamic incompatibility, whereas the associative phase separation
E-mail address: chevalier@lagep.univ-lyon1.fr (Y. Chevalier).

http://dx.doi.org/10.1016/j.colsurfa.2015.06.006
0927-7757/© 2015 Elsevier B.V. All rights reserved.
368 D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374

occurs in case of attractive interaction between a protein and In WPI/AG complex coacervation system three transitions have
a polysaccharide [4–6]. An associative phase separation of elec- been observed in pH scan down to acidic medium. The formation of
trostatic origin between the two polyelectrolytes of opposite soluble complexes was observed first, thereafter phase separation
electrical charges takes place in the case of complex coacerva- took place in a second stage, and the last event was the dissoci-
tion. The complexation of protein and polyelectrolytes is driven ation of the complexes as the isoelectric point of Acacia gum is
by electrostatic interactions; it is controlled by a charge regulation approached [20]. A more detailed picture of complex coacervation
process [7]. has been obtained with the ␤-lg/AG system. It is described as a
The main origin of complex coacervation is an associative nucleation and growth type process where the phase separation is
phase separation phenomenon induced by electrostatic interac- delayed with respect to the formation of the insoluble complexes
tions between at least two macromolecules, typically a protein because a high enough supersaturation is required for nucleation to
and a polysaccharide of opposite charges [8]. But other weak inter- operate [25]. Evidence of six pH-induced structural transitions has
actions, such as hydrogen bonding and hydrophobic interactions, been given during complex coacervation between ␤-lg and Acacia
can contribute to the formation of complex coacervates [9]. The gum at a 2:1 ratio. Characteristic stages are the initiation of the
associative phase separation induces the formation of neutral sol- interactions at pH 4.9, the aggregation of complexes at pH 4.6, the
uble complexes which exhibit an attractive interaction [10]. Such nucleation of phase separation at pH 4.4, the massive phase sepa-
complexes undergo precipitation into liquid droplets that possi- ration at pH 4.2, the observation of large spherical coacervates by
bly settle to form the coacervated phase [11]. Though complex optical microscopy at pH 4.0 and morphological changes of coacer-
coacervation is a phase separation that can be considered as a pre- vates at pH 3.8 [26]. The thermodynamic mechanisms of ␤-lg/AG
cipitation, the complex coacervate particles are extensively swollen binding processes at pH 4.2 include two different binding steps, a
by water and better look like microgel particles in suspension in first exothermic step and a second endothermic step [27].
water. After complexation each biopolymer can be found both in The main objective of the present work was to get a definite bal-
the coacervates and to a small extent, in solution in the aque- ance of the coacervation of the main components of WPI with an
ous phase [12]. Gelatin/Acacia gum polymer pair is one of the anionic polysaccharide, Acacia gum. Commercial WPI was mainly
first and most widely used complex coacervation system [13,14]. composed of ␤-lg and ␣-lac. The precise composition of the com-
Gelatin/Acacia gum coacervates exhibit a wide range of useful func- mercial WPI used for this study was firstly determined. Then the
tionalities for the development of food products such as thickening, formation of complex coacervates was investigated in the WPI/AG
gelling, foaming and emulsifying abilities [15]. Coacervates are also system and compared to that of its pure components in the respec-
used for nano- or microencapsulation of various lipophilic com- tive ␤-lg/AG and ␣-lac/AG systems. The fraction of ␤-lg and ␣-lac
pounds, food ingredients and pharmaceuticals [9,16–18]. involved in WPI/AG complex coacervates as a function of pH and
Whey proteins isolate/Acacia gum (WPI/AG) is another biopoly- biopolymer ratio was determined by capillary electrophoresis.
mer system used for complex coacervation, where WPI is used
to replace gelatin. It was previously shown that the interaction
between WPI and AG is mainly driven by the interaction of ␤- 2. Material and methods
lactoglobulin (␤-lg) with Acacia gum [19]. Indeed ␤-lg is the major
component of WPI on the one hand; the second major component 2.1. Materials
is ␣-lactalbumin (␣-lac). In addition, the higher isoelectric point of
␤-lg (pH 5.2) compared to ␣-lac (pH 4.1) leads to the formation of Whey protein isolate (Prolacta® 95) was a gift from Lac-
␤-lg/AG complexes earlier than the ␣-lac/AG complexes as the pH talis Ingredients (Bourgbarré, France). The powder contained 95%
is lowered from above the isoelectric point [20]. protein with respect to full dry matter. Acacia gum (AG), ␤-
The effect of pH, protein to polysaccharide ratio (Pr:Ps), ionic lactoglobulin (␤-lg), ␣-lactalbumin (␣-lac), and hydrochloric acid
strength and total biopolymer concentration on complex coacer- were purchased from Sigma–Aldrich. Deionized water of 18 M cm
vates formation has been studied in WPI/AG and ␤-lg/AG systems. was used.
For the WPI:AG system, complex coacervation has been observed
in a pH range from 3 to 5 for mixtures of 10 wt% biopolymers and a 2.2. Preparation of polymers solutions
3:1 Pr:Ps ratio. Only very weak interactions were observed between
the two polymers at higher pH (between pH 5 and 7) [21]. The opti- Whey proteins isolate (WPI), ␤-lg, ␣-lac and Acacia gum were
mum pH of complex coacervation was found at 4.0 for a 2:1 Pr:Ps dispersed in deionized water and mixed under gentle stirring at
ratio. At this pH, phase separation occurred the fastest and the final room temperature until complete dissolution. The concentrations
coacervate volume was the largest. Varying the protein to polysac- of WPI solutions were 10 mg mL−1 for capillary electrophoresis
charide ratio (Pr:Ps) shifted the optimum pH to higher values when assays and 100 mg mL−1 for SDS-PAGE analyses. Complex coacerva-
Pr:Ps was increased. Such shift of optimum pH was probably due to tion studies were carried out with polymer solutions at 10 mg mL−1 .
variations of the concentrations of residual soluble species as the
Pr:Ps ratio was varied. It was also shown that increasing the ionic
strength led to a less concentrated and poorly structured coacervate 2.3. SDS-PAGE
phase, induced by the screening of the electrostatic interactions
[22]. The electrostatic nature of the interactions between ␤-lg and The identification of ␤-lg in WPI was carried out by poly-
Acacia gum has also been pointed out according to the pH depen- acrylamide gel electrophoresis (PAGE) in the presence of sodium
dence of the phase separation, together with the influence of the dodecyl sulfate (SDS). 5 samples of 1–3 ␮L of WPI solution were
protein to polysaccharide ratio and the total biopolymer concentra- deposited in five wells and 10 ␮L of bromophenol blue were added
tion [23]. Finally, changes in ␤-lg/AG complex coacervation were to each samples. Molecular weights were estimated using Fer-
observed in the presence of ␤-lg aggregates. The simultaneous for- mentas electrophoresis protein standard (PageRulerTM Unstained
mation of spherical vesicular coacervates and precipitates were Protein Ladder). The stacking gel and the separation gel were
observed when the ␤-lg solution contained aggregates, whereas 10% acrylamide. After electrophoresis, gels were fixed for 30 min,
only coacervates were observed in aggregate-free ␤-lg solution. destaining was carried out overnight until the background color
Removal of protein aggregates can be performed by centrifugation has completely disappeared, and images of the destained gels were
(RCF = 10,000 × g for 1 h at pH 4.75) [24]. taken.
D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374 369

2.4. Capillary gel electrophoresis (CGE)

Capillary gel electrophoresis was performed using the Agi-


lent HP CE 3D system equipped with a diode array detector and
computer controlled via the Agilent 3D-CE ChemStation software.
Buffers were from Proteome Lab SDS-MW Analysis Kit (Beckman
Coulter). A fused-silica capillary (35 cm total length with a detec-
tion window at 8.5 cm and 50 ␮m I.D.) was used with a SDS-Gel
Separation Buffer. Samples were diluted in SDS Sample Buffer
(100 mM Tris–HCl, pH 9.0/1% SDS) and then injected electroki-
netically (−8 kV, 20 s). The separation voltage was −17.5 kV; UV
detection at 220 nm wavelength was used. Analyses were per-
formed at 20 ◦ C. The measurement uncertainty was 4% (n = 6).

2.5. Elisa quantitative analysis of casein


Fig. 1. SDS-PAGE electrophoresis gel of whey protein isolate (␤-lg: ␤-lactoglobulin,
A microtiter plate coated with anti-casein antibodies was used. ␣-lac: ␣-lactalbumin, BSA: bovine serum albumin, LF: lactoferrin).
A solution of WPI was prepared at 1% in deionized water. Half of the
solution was purified as described in the next paragraph. Purified under moderate magnetic stirring. Samples were aged at room
and not purified solutions of WPI were diluted to 1/1000 before temperature for 24 h to ensure full phase separation and sedimen-
analysis following the assay procedure given in the test instruc- tation of complex coacervates. Pictures were taken at this stage.
tion (Test Instruction – Enzyme Immunoassay for the Quantitative Supernatants were recovered (cloudy samples were centrifuged)
Determination of bovine Casein in Food, Libios). The absorbance and analyzed for ␤-lg and ␣-lac.
at 450 nm was measured with a microplate reader (TECAN Infi-
nite 200 PRO). Casein Standards (0, 0.2, 0.6, 2, 6 ppm of casein) 2.8. ˇ-Lactoglobulin and ˛-lactalbumin complex coacervation
were prepared and analyzed simultaneously to the samples to
obtain a standard curve, which logarithm regression was y = 0.348 ␤-lg was mixed at a ratio of 2:1 with Acacia gum, and the pH was
ln(x) + 0.7788 (R2 = 0.9511). The limit of detection of the casein was lowered from 5.5 to 3.0 by addition of hydrochloric acid under mod-
0.04 ppm and the limit of quantification was 0.2 ppm. erate magnetic stirring. Thereafter the same protocol as for WPI was
applied. ␣-lac was mixed at a ratio of 1:1 with Acacia gum, and the
2.6. Whey protein isolate composition pH was lowered from 7.0 to 2.0 by addition of hydrochloric acid
under moderate magnetic stirring. Thereafter the same protocol as
WPI composition was studied by four different techniques: for WPI was applied.
SDS-PAGE electrophoresis, capillary electrophoresis, casein ELISA
quantitative analysis and macroscopic observations. Analyses were 3. Results and discussion
performed on two kinds of samples. One sample consisted in
a solution of WPI as described in Section 2.2. The other sam- 3.1. Whey protein isolate composition
ple was acidified at pH 4.75 with hydrochloric acid, centrifuged
at 7830 rpm for 30 min, as described in Weinbreck et al. [20] to 3.1.1. SDS-PAGE
remove whey protein aggregates. The supernatant was recovered SDS-PAGE patterns of WPI solutions of three different concen-
and analyzed. Solutions of ␤-lg and ␣-lac were analyzed by capil- trations are shown in Fig. 1. The results indicate the presence of
lary electrophoresis to compare their electrophoregrams to those of two broad bands between 10 and 20 kDa which can be identified
whey protein isolate. Standard solutions of ␤-lg and ␣-lac (between as ␤-lg (18.4 kDa) and ␣-lac (14.2 kDa). These molar masses were
2 and 10 mg mL−1 ) were used to determine the ␤-lg and ␣-lac previously determined by Léonil et al. [28]. Capillary electrophore-
concentrations in WPI sample. sis was performed to separate these proteins and to determine their
concentrations as reported in section 3.1.3. A band corresponding to
2.7. Whey protein isolate complex coacervation a molar mass around 20 kDa seemed to correspond to caseins which
molar masses determined were between 19 and 25 kDa [28]. Such
WPI solution was acidified at pH 4.75 and centrifuged at 7830 proteins were not presumed being present in the WPI according
rpm for 30 min. The supernatant was recovered and mixed with to the supplier’s information. Quantitative ELISA analysis of casein
Acacia gum at Pr:Ps ratios 4:1, 2:1, and 1:1. Then pH was progres- was performed to confirm the presence of casein in the samples as
sively lowered from 5.5 to 2.0 by addition of hydrochloric acid reported in Section 3.1.4. At least two bands were isolated between

Fig. 2. WPI solution (5.5%) as a function of pH. Formation of precipitates between pH 5.2 and 4.1.
370 D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374

␣-lac peaks were measured on WPI samples before and after


acidic treatment. So it was determined that a WPI solution of Pro-
lacta at 10 mg mL−1 concentration contained 5.67 mg mL−1 of ␤-lg
(56.7%) and 1.12 mg mL−1 of ␣-lac (11.2%). The concentration of ␤-
lg decreased to 5.35 mg mL−1 after the acidic treatment, while the
concentration of ␣-lac remained constant. A slight decrease of the
␤-lg concentration was noticed after acidic treatment, which could
be explained by the formation of some ␤-lg oligomers at pH 4.75
and their elimination by centrifugation [24] together with casein.

3.1.4. Quantitative Elisa analysis of casein


This analysis was performed on WPI before and after the acidic
treatment. The mean absorbances at 450 nm of WPI before and
after the acidic treatment were 0.52 and 0.14, respectively, cor-
Fig. 3. WPI electrophoregrams obtained by capillary gel electrophoresis (SDS-MW). responding to concentrations of 0.48 mg mL−1 and 0 mg mL−1 of
Evidence of two peaks disappearance after acidic treatment. casein respectively, according to the calibration curve. Indeed there
is a 0.14 background absorbance of the kit. So it can be concluded
that the WPI samples contained casein that was totally removed
60 and 80 kDa, which were identified as bovine serum albumin
by the acidic treatment followed by centrifugation. The precipitate
(66 kDa) and lactoferrin (80 kDa) according to Alomirah and Alli
that was removed contained the full casein content of WPI and a
[29] and Lönnerdal and Iyer [30].
small fraction of ␤-lg.

3.1.2. Macroscopic observations


3.2. Complex coacervation of whey protein isolate
As the WPI solution was acidified from pH 6.5 to pH 3.1, it
became cloudy at pH 5.4 and this turbid aspect lasted until pH
After ageing at room temperature for 24 h, macroscopic observa-
3.6 (Fig. 2). Moreover a white precipitate was observed between
tions of the mixed WPI/AG solutions revealed a turbidity appearing
pH 5.2 and pH 4.1. The turbidity and the precipitation were pH-
at a pH depending on the Pr:Ps ratio, pH 5.0 for 4:1 ratio and pH
dependent. These phase transitions were reversible, as the turbidity
4.5 (× in Fig. 5) for 1:1 and 2:1 ratios. The raising turbidity could
and the precipitate disappeared if the pH was increased again. The
correspond to the formation of soluble complexes, which made
first appearance of the precipitate matched with the pH of caseins
the solutions cloudy without precipitation. Indeed solutions stayed
precipitation which occurs at pH 4.6 according to Léonil et al.
cloudy after 24 h ageing while little or no sedimentation occurred.
[28]. Moreover, according to Stading and Hermansson [31], protein
These complexes are presumed electrically neutral, the pH being
aggregates are formed in ␤-lg solutions between pH 4 and pH 6. The
close to the isoelectric point (iep) of whey proteins (isoelectric
solubility curve of ␤-lg revealed that it was less soluble around its
points are: ␣-lac iep = 4.1, and ␤-lg iep = 5.2). Phase separation of
isoelectric point. The lowest solubility of ␤-lg was obtained at pH
a coacervate phase settling to the bottom of the vials was clearly
4.75 and protein aggregates were observed in the solution at this
observed between pH 3.4 and 3.1 for a Pr:Ps ratio 1:1, and between
pH [24]. Thus, in order to remove protein aggregates from the WPI
pH 4.0 and pH 3.1 (o in Fig. 5) for Pr:Ps ratios 2:1 and 4:1. Inter-
solution, the pH was decreased until 4.75 to precipitate insoluble
actions between whey proteins and Acacia gum were previously
materials, which were then removed by centrifugation at 7830 rpm
observed between pH 3 and pH 5 [21]. Three species have been
for 30 min at 25 ◦ C.
identified by Weinbreck et al. [20]: soluble polymers, soluble com-
plexes, and association of complexes. The formation of soluble
3.1.3. Capillary electrophoresis
complexes was observed at pH 5.3 and the association of com-
Capillary electrophoresis was performed on WPI, ␤-lg and ␣-lac.
plexes driving phase separation was observed at pH 4.8 for a total
Analysis of WPI solutions before and after removal of precipitate
concentration of biopolymers of 1 mg mL−1 and a Pr:Ps ratio of 2:1.
was performed in order to disclose the composition of the pre-
The coacervate phase decreased in volume and finally vanished at
cipitate. Electrophoregrams of these analyses (Fig. 3) shows the
the most acidic pH (below pH 3.1). This was probably due to the
disappearance of two peaks after the removal of precipitate. The
decrease of electrostatic interactions because the negative charges
locations of such peaks corresponded to molar masses between 20
of Acacia gum decrease when pH is approaching the iep of aca-
and 35 kDa. Accordingly, two hypotheses could be put forward.
cia gum at pH 2. Acacia gum is negatively charged above pH 2
These two peaks could either correspond to the dimeric form of
␤-lg or to casein. The dimer form of ␤-lg is only present in solution
between pH 5.1 and pH 8.0 [29]. However the pH of the sample
buffer being 9.0, only monomeric form should be present in the
sample. Therefore these two peaks probably revealed the presence
of casein considering that casein precipitation occurs at pH 4.6 [28].
Casein aggregates have been removed by the acidic treatment fol-
lowed by centrifugation. This later point has been confirmed by the
quantitative analysis of casein.
Capillary gel electrophoresis was also used to identify ␤-lg and
␣-lac in the WPI sample (Fig. 4). These proteins were identified
properly after the analysis of pure samples of ␤-lg and ␣-lac. Stan-
dard solutions of ␤-lg and ␣-lac (between 2 and 10 mg mL−1 ) were
analyzed and standard curves were established by plotting the
mean peak areas versus concentrations in mg mL−1 . Linear regres-
sion of ␤-lg and ␣-lac standard curves are y = 1164.1 x (R2 = 0.9914) Fig. 4. WPI electrophoregram obtained by capillary gel electrophoresis (SDS-MW).
and y = 1710 x (R2 = 0.9934), respectively. Areas under ␤-lg and Identification of the peaks of ␤-lg and ␣-lac.
D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374 371

Fig. 5. WPI/AG complex coacervation as a function of the Pr:Ps ratio (1:1, 2:1 and
4:1) and pH (between 5 and around 2.5). The onset of turbidity is marked by × and
the beginning of phase separation by o.

[32]. The pH for coacervate phase appearance increased with the


Pr:Ps ratio. This observation is in accordance with the results of
Weinbreck et al. [22], who showed that increasing the Pr:Ps shifted
the optimum pH to higher values. The authors explained this phe-
nomenon by the requirement for a higher positive charge of the
Fig. 6. Protein concentration in the supernatant as a function of pH for the Pr:Ps
protein (lower pH) to balance the higher amount of negatively
ratio 1:1 (A), 2:1 (B) and 4:1 (C) [blue diamonds: ␤-lg concentration; red trian-
charged Acacia gum at low Pr:Ps ratio. gles: ␣-lac concentration; dashed lines: initial proteins concentrations]. pH values
␤-lg and ␣-lac were quantified by capillary gel electrophoresis in are presented in descending order because experimentations were performed by
the WPI/AG solutions coexisting with the coacervate for Pr:Ps ratio decreasing the pH upon addition of hydrochloric acid.
of 4:1, 2:1 and 1:1 after solution ageing at room temperature for
24 h (Fig. 6). For the Pr:Ps ratio of 1:1 (Fig. 6A), concentrations of ␤-
stronger than that of ␣-lac (Fig. 6). ␤-lg was the predominant pro-
lg and ␣-lac were close to the initial concentrations of each protein
tein in the complex coacervation phenomenon observed between
in the mixture (2.6 and 0.54 mg mL−1 , respectively) when the pH
WPI and Acacia gum as already highlighted by Weinbreck et al. [20].
was varied from 4.9 to pH 4.5. Upon a further decrease of pH to 3.5,
The decrease of ␣-lac concentration was more important at the pH
the concentration of both proteins progressively decreased. Indeed
of optimum coacervation when the Pr:Ps ratio was low. Decreasing
the complexation of the proteins with Acacia gum precipitated and
the Pr:Ps ratio resulted in more free Acacia gum molecules avail-
formed a pellet, so that the protein concentration remaining in the
able for association with ␣-lac molecules. The presence of a larger
supernatant decreased. Minimum ␤-lg and ␣-lac concentrations
amount of free acacia gum molecules after complexation with ␤-
were 0.5 and 0.07 mg mL−1 respectively. They were observed at pH
lg at low Pr:Ps ratio promoted their association with ␣-lac as the
3.5 which corresponded to the maximum yield of complex coac-
pH approached its isoelectric point. The variation of proteins con-
ervation. Finally the protein concentrations increased again below
centrations in the supernatant against pH gave evidence of a shift
pH 2.3. At this pH ␤-lg and ␣-lac concentrations were 1.5 mg mL−1
in the pH value inducing the decrease of protein concentrations
and 0.4 mg mL−1 respectively. The increase of the protein content
when the Pr:Ps ratio increased. Such shift toward lower pH values
in the supernatant at acidic pH proved the dissociation of the com-
was observed for both proteins (␤-lg and ␣-lac) and illustrated that
plexes. Similar trends were observed for the 2:1 ratio (Fig. 6B). The
the pH of complex coacervation depended on the Pr:Ps ratio. These
behavior was the same for the 4:1 Pr:Ps ratio (Fig. 6C); but the
results were in accordance with the macroscopic observations.
decrease of protein concentration appeared at a higher pH, i.e. at
pH 4.5. These results showed that complex coacervation occurred
at a pH value depending on the Pr:Ps ratio, as previously pointed 3.3. ˇ-Lactoglobulin complex coacervation
out by Weinbreck et al. [22]. In their study the optimum condi-
tions for coacervation were reached at pH 3.5 for a Pr:Ps ratio of Macroscopic changes observed in ␤-lg/AG mixed solutions were
1:1, at pH 4 for a Pr:Ps ratio of 2:1, and at pH 4.5 for a Pr:Ps ratio induced by acidification followed by ageing at room temperature
of 8:1. The decrease of ␤-lg concentration as a function of pH was for 24 h (Fig. 7). No changes were observed at pH 6.0 where both
372 D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374

Fig. 7. Top: Macroscopic observations of ␤-lg/AG complex coacervation. Solutions


at 1% polymers with Pr:Ps ratio of 2:1, between pH 6.0 and pH 2.7, after 24 h of Fig. 8. Top: Macroscopic observations of ␣-lac/AG complex coacervation. Solutions
phase separation. Bottom: ␤-lg concentration in the supernatant, after 24 h of phase at 1% polymers with Pr:Ps ratio of 1:1, between pH 6.0 and pH 2.7, after 24 h of
separation of ␤-lg/AG solutions, as a function of pH for the Pr:Ps ratio 2:1. The dashed phase separation. Bottom: ␣-lac concentration in the supernatant, after 24 h of phase
line marks the full concentration of ␤-lg. separation of ␣-lac/AG solutions, as a function of pH for a Pr:Ps ratio of 1:1. The
dashed line marks the full concentration of ␣-lac.

␤-lg and AG were negatively charged. The solution turned turbid


at pH 4.8 for Pr:Ps ratio of 2:1; this was followed by phase sepa- tion between pH 2.8 and pH 2.3. Such changes corresponded to the
ration between pH 4.1 and pH 3.4. Turbidity corresponded to the formation of soluble complexes at pH 3.5 followed by the associ-
formation of soluble complexes. As the pH decreased further, the ation of complexes, leading to sedimentation between pH 2.8 and
complexes formed aggregates that grew and settled. The initiation pH 2.3. Phase separation disappeared at pH 1.6. Formation of com-
of interactions and the formation of coacervates between ␤-lg and plex coacervates appeared at lower pH with ␣-lac than with ␤-lg.
AG have previously been described at pH 4.90 and pH 4.24 [25,26]. Such pH shift was due to the more acidic isoelectric point of ␣-lac.
Redissolution took place at pH 2.7 as in the case of the WPI/AG Complex coacervate formation has also been observed in solutions
system. Complex coacervate formation has also been observed in with a ␣-lac/AG ratio of 1:2 between pH 2.8 and pH 2.0.
solutions with a ␤-lg/AG ratio of 1:1 between pH 4.6 and pH 2.7. Quantitative analysis of ␣-lac by capillary gel electrophoresis
Quantification of ␤-lg in solution was performed by capillary was also performed as for ␤-lg and WPI. The results were in full
gel electrophoresis as a function of pH after acid addition and 24 h accordance with the macroscopic observations and similar to that
ageing (Fig. 7). The concentration of ␤-lg was close to initial concen- of ␤-lg but shifted with respect to pH (Fig. 8). Starting with a con-
tration (6.7 mg mL−1 ) at pH 5.4, as previously described with WPI. centrations of ␣-lac of 5 mg mL−1 mixed with Acacia gum at a Pr:Ps
Then a decrease of the ␤-lg concentration, down to 1 mg mL−1 , was ratio of 1:1, this concentration was observed at pH 4.7, 4.3 and 2.1. A
observed in the supernatant of mixtures between pH 4.9 and pH decrease of ␣-lac concentration down to a minimum of 2 mg mL−1
3.7. The decrease of the concentration of ␤-lg in the supernatant was observed in the supernatant of mixtures between pH 3.9 and
matched the formation of complex coacervates between ␤-lg and pH 3.0.
Acacia gum. Complex coacervates settled after 24 h ageing, so that
the protein concentrations which were measured in the super- 3.5. Comparison of the behaviors of ˇ-lactoglobulin and
natant corresponded to proteins which have not formed complexes ˛-lactalbumin
with Acacia gum. When pH was further decreased to 3.0, the con-
centration of ␤-lg increased and finally reached 4 mg mL−1 . This ␤-lg underwent coacervation at a higher pH than ␣-lac because
increase of ␤-lg concentration in the supernatant corresponded to its isoelectric point was higher. Maximum coacervation was
the dissociation of polymer complexes. observed at pH 4.5 for ␤-lg anf pH 3.4 for ␣-lac, 0.7 pH units lower
than the pie of both proteins. The behaviors of ␤-lg and ␣-lac were
3.4. ˛-Lactalbumin complex coacervation quite similar with that respect. A noticeable difference between ␤-
lg and ␣-lac was the residual concentration of protein in the bulk
The behavior of ␣-lac and the corresponding macroscopic obser- aqueous phase at the pH of maximum coacervation: 1 mg mL−1
vations (Fig. 8) were fully similar to that of ␤-lg, but shifted along for ␤-lg against 2 mg mL−1 for ␣-lac. Stronger interactions were
the pH range. For Pr:Ps ratio of 1:1 according to the same proto- operating between Acacia gum and ␤-lg than for ␣-lac. It has often
cols as for ␤-lg and WPI, no change was observed between pH 5.0 been inferred that such stronger interactions came from the higher
and 4.1, turbidity appeared at pH 3.5 followed by phase separa- molar mass of ␤-lg. Indeed it is well-established that binding to
D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374 373

polymers and polymer precipitation are stronger for polymers of supernatant. Whether there remain non-vanishing concentrations
higher molar mass. But the molar masses of ␤-lg (18.4 kDa) and of anionic and cationic free polymer species seems a matter of a
␣-lac (14.2 kDa) are not so different. In the present case, the elec- large variability. The same competitive coacervation should also
trical charges of the proteins also obviously matter. The charges take place for the anionic partner Acacia gum that contains several
come from anionic and cationic residues of the proteins. According components, arabinogalactan and arabinogalactan-protein com-
to their chemical structures, ␤-lg contains 26 anionic residues (9 plex [36]. It has indeed been disclosed that the negatively charge
aspartic acids, 16 glutamic acids, and one terminal carboxylic acid) polysaccharide arabinogalactan was the main component respon-
and 23 cationic residues (2 tryptophans, 2 histidines, 15 lysines, 3 sible for complex coacervation with ␤-lg [37]. As an outcome, the
arginines, and one terminal amino group) out of 162 residues [33] present results could be used for a better control over the encap-
and ␣-lac contains 16 anionic residues (7 aspartic acids, 8 glutamic sulation process by complex coacervation [38].
acids, and one terminal carboxylic acid) and 21 cationic residues (4
tryptophans, 3 histidines, 12 lysines, 1 arginine, and one terminal Acknowlegments
amino group) out of 123 residues [34]. The densities of ionizable
groups are quite similar for both proteins. At the pie, the proteins Thanks to Prayon S.A. for the financial support to present work.
contain the same linear densities of cationic and anionic residues We are grateful to Mr. Olivier Marcillat (Université Lyon 1, ICBMS
Cpie . Upon shifting the pH from the pie, the linear density of cationic – UMR 5246) for his efficient assistance in SDS-PAGE experimen-
groups increases by x = 10(pie−pH) and the linear density of anionic tations and Mrs. Sandra Balvay (Université Lyon 1, MATEIS – UMR
groups decreases by x = 10(pie−pH) . The total linear charge density 5510) for her help in microplate reading.
was therefore the same for both polymers because the pH shift from
the pie at maximum coacervation were the same and the densities
References
of charged groups along the polypeptide backbone were the same.
However, the total charge of ␤-lg is larger because its molar mass [1] A. Matalanis, O.G. Jones, J. McClements, Structured biopolymer-based delivery
is higher on one hand and because the pH at maximum coacerva- systems for encapsulation, protection, and release of lipophilic compounds,
tion is higher on the other hand. Both effects are additive, but it is Food Hydrocolloids 25 (2011) 1865–1880.
[2] E. Bouyer, G. Mekhloufi, I. Le Potier, T. du Fou de Kerdaniel, J.-L. Grossiord, V.
clear that the molar mass of ␤-lg being higher than ␣-lac by only Rosilio, F. Agnely, Stabilization mechanism of oil-in-water emulsions by
25%, the supplementary charge coming from the difference of pH at ␤-lactoglobulin and gum arabic, J. Colloid Interface Sci. 354 (2011) 467–477.
maximum coacervation for ␤-lg and ␣-lac makes obviously a larger [3] O. Jones, E.A. Decker, J. McClements, Thermal analysis of ␤-lactoglobulin
complexes with pectins or carrageenan for production of stable biopolymer
contribution. As a conclusion, the origin of the stronger electrostatic
particles, Food Hydrocolloids 24 (2010) 239–248.
interactions of ␤-lg with Acacia gum causing a larger coacervation [4] J.-L. Doublier, C. Garnier, D. Renard, C. Sanchez, Protein-polysaccharide
yield (lower residual concentration at maximum coacervation) is interactions, Curr. Opin. Colloid Interface Sci. 5 (2000) 202–214.
primarily the higher pie of ␤-lg. A large difference of pie between [5] S.L. Turgeon, M. Beaulieu, C. Schmitt, C. Sanchez, Protein-polysaccharide
interactions: phase-ordering kinetics, thermodynamic and structural aspects,
the partners of the coacervate cause a high coacervation yield. Curr. Opin. Colloid Interface Sci. 8 (2003) 401–414.
[6] D.J. Mc Clements, Non-covalent interactions between proteins and
polysaccharides, Biotechnol. Adv. 24 (2006) 621–625.
[7] F.L.B. da Silva, B. Jönsson, Polyelectrolyte-protein complexation driven by
4. Conclusion
charge regulation, Soft Matter 5 (2009) 2862–2868.
[8] C.L. Cooper, P.L. Dublin, A.B. Kayitmazer, S. Turksen, Polyelectrolyte-protein
A full balance of the chemical composition of the coexisting complexes, Curr. Opin. Colloid Interface Sci. 10 (2005) 52–78.
[9] S.L. Turgeon, C. Schmitt, C. Sanchez, Protein-polysaccharide complexes and
complex coacervates and solution phases was achieved during
coacervates, Curr. Opin. Colloid Interface Sci. 12 (2007) 166–178.
the complex coacervation of WPI and Acacia gum by means of [10] C.G. de Kruif, F. Weinbreck, R. de Vries, Complex coacervation of proteins and
lowering the pH. Indeed WPI is a complex mixture of several anionic polysaccharides, Curr. Opin. Colloid Interface Sci. 9 (2004) 340–349.
components. Its analysis showed that the two main components [11] C. Schmitt, C. Sanchez, S. Desobry-Banon, J. Hardy, Structure and
technofunctional properties of protein-polysaccharide complexes: a review,
were ␤-lactoglobulin (56.7%) and ␣-lactalbumin (11.2%). The pres- Crit. Rev. Food Sci. Nutrition 38 (1998) 689–753.
ence of casein caused the formation of a precipitate that disturbed [12] A. Veis, A review of the early development of the thermodynamics of the
the complex coacervation process. Therefore it is wise to per- complex coacervation phase separation, Adv. Colloid Interface Sci. 167 (2011)
2–11.
form a purification of the WPI solution by precipitation of casein [13] D.J. Burgess, J.E. Carless, Microelectrophoretic studies of gelatin and Acacia for
before used. Macroscopic observations coupled with capillary elec- the prediction of complex coacervation, J. Colloid Interface Sci. 98 (1984) 1–8.
trophoresis analyses allowed reaching an improved view over the [14] H. Jizomoto, Phase separation induced in gelatin-base coacervation systems
by addition of water-soluble nonionic polymers II: effect of molecular weight,
balance of ␤-lg and ␣-lac during the course of the complex coac- J. Pharm. Sci. 74 (1985) 469–472.
ervation process. The chemical composition of the coacervates [15] C. Schmitt, S.L. Turgeon, Protein/Polysaccharide complexes and coacervates in
depends on the Pr:Ps ratio and pH. An optimum coacervation pH food systems, Adv. Colloid Interface Sci. 167 (2011) 63–70.
[16] H. Jizomoto, Phase separation induced in gelatin-base coacervation systems
was determined for each Pr:Ps ratio. Residual soluble species are
by addition of water-soluble nonionic polymers I: microencapsulation, J.
present in aqueous solution is most cases; full phase separation of Pharm. Sci. 73 (1984) 879–882.
either ␤-lg or ␣-lac took place in very scarce instances. Pure ␤-lg [17] V.B. Junyaprasert, A. Mitrevej, N. Sinchaipanid, P. Boonme, D.A. Wurster,
Effect of process variables on the microencapsulation of vitamin A palmitate
and ␣-lac species are able to form complex coacervates with Aca-
by gelatin-Acacia coacervation, Drug Develop. Ind. Pharm. 27 (2001) 561–566.
cia gum. In the WPI mixture, ␤-lg is more involved than ␣-lac in [18] M. Mellema, Complex coacervate encapsulate comprising lipophilic core, PCT
the complex coacervation phenomenon because ␤-lg is the main Int. Appl. WO2004054702A1, 20,040,701 (2004).
species of WPI and mainly because the isoelectric point of ␤-lg [19] E. Bouyer, G. Mekhloufi, N. Huang, V. Rosilio, F. Agnely, ␤-Lactoglobulin, gum
arabic, and xanthan gum for emulsifying sweet almond oil: Formulation and
is higher than that of ␣-lac. But this observation depends on the stabilization mechanisms of pharmaceutical emulsions, Colloids Surfaces A:
Pr:Ps ratio. The present observations and analyses give a quan- Physicochem. Eng. Aspects 433 (2013) 77–87.
titative account of the fate of the two main components of WPI [20] F. Weinbreck, R. de Vries, P. Schrooyen, C.G. de Kruif, Complex coacervation of
whey proteins and gum arabic, Biomacromolecules 4 (2003) 293–303.
in the complex coacervation process. Such trends have already [21] M. Klein, A. Aserin, P. Ben Ishai, N. Garti, Interactions between whey protein
been reported by Serov et al. [35] although they neither reported isolate and gum Arabic, Colloids Surfaces B: Biointerfaces 79 (2010) 377–383.
a systematic investigation nor they performed quantitative analy- [22] F. Weinbreck, R.H. Tromp, C.G. de Kruif, Composition and structure of whey
protein/gum Arabic coacervates, Biomacromolecules 5 (2004) 1437–1445.
ses. Interestingly, they reported instances of full depletion of ␣-lac [23] C. Schmitt, C. Sanchez, F. Thomas, J. Hardy, Complex coacervation between
upon complex coacervation with apple pectin, whereas a signif- ␤-lactoglobulin and Acacia gum in aqueous medium, Food Hydrocolloids 13
icant residual concentration of free ␤-lg was remaining in the (1999) 483–496.
374 D. Ach et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 481 (2015) 367–374

[24] C. Schmitt, C. Sanchez, S. Despond, D. Renard, F. Thomas, J. Hardy, Effect of [32] T.P. Kravtchenko, Application of Acacia gum as a natural source of soluble
proteins aggregates on the complex coacervation between ␤-lactoglobulin dietary fibre, food ingredients europe, in: Conference Proceedings, Miller
and Acacia gum at pH 4.2, Food Hydrocolloids 14 (2000) 403–413. Freeman Plc, Maarssen, Netherlands, 1997, pp. 56–60.
[25] C. Sanchez, G. Mekhloufi, D. Renard, Complex coacervation between [33] L. Sawyer, ␤-Lactoglobulin, in: P.L.H. McSweeney, P.F. Fox (Eds.), Advanced
␤-lactoglobulin and Acacia gum: a nucleation and growth mechanism, J. Dairy Chemistry Vol 1A. Proteins, Basic Aspects, 4th ed., Springer, New York,
Colloid Interface Sci. 299 (2006) 867–873. 2013, pp. 211–259.
[26] G. Mekhloufi, C. Sanchez, D. Renard, S. Guillemin, J. Hardy, pH-induced [34] K. Brew, ␤-Lactalbumin, in: P.L.H. McSweeney, P.F. Fox (Eds.), Advanced Dairy
structural transitions during complexation and coacervation of Chemistry Vol 1A. Proteins, Basic Aspects, 4th ed., Springer, New York, 2013,
␤-lactoglobulin and Acacia gum, Langmuir 21 (2005) 386–394. pp. 261–273.
[27] L. Aberkane, J. Jasniewski, C. Gaiani, J. Scher, C. Sanchez, Thermodynamic [35] A.V. Serov, A. Yu. Antonov, V.B. Tolstoguzov, Isolation of lactic whey proteins
characterization of Acacia gum-␤-lactoglobulin complex coacervation, in the form of complexes with apple pectin, Die Nahrung 1 (1985) 19–30.
Langmuir 26 (2010) 12523–12533. [36] M.-C. Vandevelde, J.-C. Fenyo, Macromolecular distribution of Acacia senegal
[28] J. Léonil, D. Mollé, F. Gaucheron, P. Arpino, P. Guénot, J.L. Maubois, Analysis of gum (gum arabic) by size-exclusion chromatography, Carbohydr. Polym. 5
major bovine milk proteins by on-line high-performance liquid (1985) 251–273.
chromatography and electrospray ionization-mass spectrometry, Lait 75 [37] C. Costa, C. Liz, V. Petkova, J.J. Benattar, M. Michel, M.E. Leser, R. Miller, X-ray
(1995) 193–210. reflectivity studies of liquid films stabilized by mixed ␤-lactoglobulin—Acacia
[29] H.F. Alomirah, I. Alli, Separation and characterization of ␤-lactoglobulin and gum systems, Colloids Surfaces A: Physicochem. Eng. Aspects 282-283 (2006)
␣-lactalbumin from whey and whey protein preparations, Int. Dairy J. 14 109–117.
(2004) 411–419. [38] T.F. Vandamme, D. Poncelet and P. Subra-Paternault, Microencapsulation: Des
[30] B. Lönnerdal, S. Iyer, Lactoferrin: molecular structure and biological function, Sciences aux Technologies. Paris: Lavoisier Tec & Doc (2007).
Annu. Rev. Nutrition 15 (1995) 93–110.
[31] M. Stading, A.M. Hermansson, Viscoelastic behaviour of ␤-lactoglobulin gel
structures, Food Hydrocolloids 4 (1990) 121–135.

También podría gustarte