Exploration of the co-structuring and stabilising role of flaxseed gum in whey protein isolate based cryo-hydrogels

Carbohydrate Polymers 289 (2022) 119424

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Exploration of the co-structuring and stabilising role of flaxseed gum in
whey protein isolate based cryo-hydrogels
Thierry Hellebois a, b, Claire Gaiani b, c, S´
ebastien Cambier a, Anaïs Noo a, Christos Soukoulis a, *
a

Environmental Research and Innovation (ERIN) Department, Luxembourg Institute of Science and Technology (LIST), 5 Avenue des Hauts Fourneaux, Esch-sur-Alzette,
L4362, Luxembourg
b
Universit´e de Lorraine, LIBio, Nancy, France
c
Institut Universitaire de France (IUF), France

A R T I C L E I N F O

A B S T R A C T

Keywords:
Mucilage
Cryotropic gelation
Microstructure
Phase separation
Colloidal stability
Molecular interactions

In the present work, the structuring and stabilising potential of flaxseed gum (FG) in whey protein isolate (WPI)
cryo-hydrogels was investigated. The FG presence (0.1–1% wt.) in the heat-treated WPI dispersions (10% wt.)
induced strong segregative phase separation phenomena, which were associated with a depletion flocculation
mechanism. The cryotropic processing of the WPI-FG solutions led to the formation of diverse macroporous
protein gel networks depending on the colloidal state of their biopolymeric precursors. Cryogel formation was
primarily mediated via covalent (thiol-disulphide bond) bridging, whilst to a lesser extent, non-covalent in­
teractions contributed to the overall stabilisation of the protein gel network. Although FG had a rather minor
contribution to the formation of elastically active crosslinks (FG was partitioning mainly into the serum phase
located in the macropores), its presence (at concentrations ≥0.75% wt.) improved the homogeneity and inter­
connectivity of the stranded protein network, whilst it reduced its colloidal instability and macroporosity.

1. Introduction
Biobased cryogels constitute hydrogel templates produced via the
cryogenic processing (i.e., freezing, incubation in the frozen state and
thawing) induced crosslinking of biopolymer precursors (Lozinsky &
Okay, 2014). Hitherto, polymeric precursors such as polysaccharides (e.
g. galactomannans (Hellebois, Gaiani, et al., 2021; Lozinsky et al.,
2000), glucomannans (Guo et al., 2022), glucans (Lazaridou & Bilia­
deris, 2004), xanthan (Giannouli & Morris, 2003), starch (Zou & Bud­
tova, 2020), sodium alginate (Gurikov & Smirnova, 2018) etc.) and
proteins (e.g. egg white (Balaji et al., 2019), whey protein isolate
(Hellebois et al., 2022; Shiroodi et al., 2015), and gelatine (Regand &
Goff, 2003; Savina et al., 2011) have been successfully employed in the
development of biodegradable, biocompatible and mechanically rein­
forced cryogels. It is well established that the cryogenic processing
conditions (e.g. ice nucleation and crystal growth, incubation temper­
ature and duration in the frozen state), the thawing rate, the amount and
thermophysical profile of the polymeric precursors, as well as the
presence of chaotropes or kosmotropes (e.g. sugars, aminoacids, salts

etc.) govern the mechanical and structural characteristics of the ob­
tained hydrogels (Lozinsky, 2020; Lozinsky & Okay, 2014). Cryo-

hydrogels exhibit a tangible industrial relevance, including biomed­
ical, tissue engineering, biotechnological, environmental, pharmaceu­
tical, cosmetics and food applications. In the latter case, cryo-hydrogels
have been successfully implemented for structuring (active fillers),
texturing and cryo-stabilising purposes (Coria-Hern´
andez et al., 2021;
Patmore et al., 2003; Yang et al., 2020; Zhao et al., 2018). In addition,
the feasibility of cryo-hydrogels as alternative biopolymer-based tem­
plates for controlled/sustained release of bioactive compounds has been
showcased (Lazaridou et al., 2015; Nakagawa & Nishimoto, 2011;
Sowasod et al., 2013).
From a mechanistic standpoint, the cryotropic gelation may involve
both non-covalent (e.g. hydrogen bonds, van der Waals forces or hy­
drophobic interactions) and covalent interactions, resulting in physical
(“reversible”) or permanent hydrogels (Gulrez et al., 2011; Lozinsky,
2018). For instance, galactomannans such as locust bean gum, alfalfa
gum, and fenugreek gum are known to interact primarily via interchain
polymer hydrogen bond bridging and intermolecular association of the
smooth regions (i.e. galactosyl depleted) via hydrophobic interactions
(Dea et al., 1977; Hellebois, Gaiani, et al., 2021; Lozinsky et al., 2000).
Similarly, oat β-glucans and xanthan gum based cryo-hydrogels are
stabilized via -intra- and interchain hydrogen bonds in the junction

* Corresponding author.
E-mail address: (C. Soukoulis).
/>Received 28 January 2022; Received in revised form 28 February 2022; Accepted 26 March 2022
Available online 29 March 2022

0144-8617/© 2022 Luxembourg Institute of Science and Technology. Published by Elsevier Ltd.
( />
This is an open access article under the CC BY license


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

zones of the polymeric network (Giannouli & Morris, 2003; Lazaridou &
Biliaderis, 2004). In the case of proteins, cryogenic processing induced
gelation is driven by both covalent and non-covalent molecular in­
teractions. For example, Hellebois et al. (2022) reported that the addi­
tion of 25 mM of n-ethylmaleimide (NEM) hampered the ability of whey
proteins to undergo cryogenic crosslinking, postulating that their cry­
ogelation ability is primarily associated with the occurrence of thioldisulphide (− SH/− S − S− ) bridging. It is well documented that the
cryogenic processing of heteropolymeric aqueous systems, i.e. protein polysaccharide blends may offer tangible cryo-structuring and cryostabilising benefits. For instance, the presence of soluble poly­
saccharides (i.e. alfalfa galactomannan, xanthan/curdlan hydrogel
complex) in denaturated whey protein solutions led to a significant
enhancement of the mechanical strength and colloidal stability (Helle­
bois et al., 2022; Shiroodi et al., 2015). In a similar manner, Goff et al.
(1999) and Patmore et al. (2003) demonstrated that locust bean gum
exerts a strong cryostabilising role in model ice creams (sucrose + skim
milk powder) due to its inherent cryogelling ability under dynamic
freezing conditions.
Flaxseed gum, i.e. a by-product of the flaxseed oil industry, has
received much attention as an emerging hydrocolloid due to its tech­
nofunctional versatility, environmental sustainability, low cost, as well
as high biocompatibility and biodegradability (Liu et al., 2018).
Although the chemical composition and structure conformational

properties of flaxseed gum can vary on its genotypic and phenotypic
characteristics, the extraction, isolation and purification conditions are
also well known for impacting its proximate and osidic composition and
consecutively, its technofunctional profile (Hellebois, Fortuin, et al.,
2021; Liu et al., 2018; Liu et al., 2021; Soukoulis et al., 2018). Hitherto,
flaxseed gum has been successfully employed in producing food com­
posites due to its prevalent thickening and gelling capacity (Liu et al.,
2018; Soukoulis et al., 2018). In protein-rich hydrogel-based food sys­
tems, the stabilising, structurising and texturising effectiveness of flax­
seed gum depends on its colloidal interactions with proteins as
influenced by extrinsic factors such as the ionic strength, pH and tem­
perature (Basiri et al., 2018; Chen et al., 2016; Kuhn et al., 2011; Li et al.,
2012; Liu et al., 2018; Soukoulis et al., 2019). For example, the extent of
segregative phase separation in flaxseed gum - whey protein binary
systems is inextricably associated with the mechanical properties and
the physical stability of the obtained cold or acid induced gels (Kuhn
et al., 2011; Soukoulis et al., 2019). Li et al. (2012) reported that the
presence of flaxseed gum in cold-set casein gels resulted in the propor­
tional elevation of the onset gelation temperature due to the adsorption
of flaxseed gum onto the micelles, facilitating their intermolecular
bridging.
In a previous work, Hellebois et al. (2022) demonstrated that the
presence of alfalfa gum, i.e., a non-ionic galactomannan extracted from
the endosperm of alfalfa seeds (Hellebois, Gaiani, et al., 2021), in whey
protein cryogels possess an active filler mediating role due to its inherent
ability to form soft cryogels. This was associated with a remarkable
improvement of the mechanical properties and colloidal stability of the
whey protein cryogels (Hellebois et al., 2022). In the present work we
aimed at exploring the functional role of an anionic natural poly­
saccharide, i.e. flaxseed gum in the cryo-gelling performance of whey

protein isolate. Hereby, the fundamental question is whether and to
which extent flaxseed gum can mediate the cryogelation mechanistic
action of whey proteins, and hence, the structural conformation, me­
chanical and physical properties of the obtained hydrogels.

minerals 2.5%) was kindly donated by Ingredia (Arras, France). The
flaxseed gum (FG) was extracted at mild alkaline conditions (pH = 8)
from golden flaxseeds and fully characterised as detailed in Hellebois,
Gaiani, et al. (2021). The FG was composed of 87.1% carbohydrates
(8.4% arabinose, 25.3% xylose, 24.2% rhamnose, 22.3% galacturonic
acid, 14.3% galactose, 4.2% fucose and 1.4% glucose), 7.2% proteins
and 5.7% ash, on a dry basis, while lipids were detected in traces. The
polysaccharidic populations corresponded to arabinoxylans (AX),
rhamnogalacturonan–I (RG-I), and two AX-RG-I composite fractions.
The gum extract exhibited a molecular weight of 1.3 × 106 Da, an
intrinsic viscosity of 6.52 dL g− 1, a critical coil overlap concentration
(c*) of 0.55% and an average z-diameter of 97.8 nm. All the chemicals
used were purchased from Sigma Aldrich (Leuven, Belgium), and they
were of analytical grade.
2.2. Preparation of the WPI-FG solutions and cryogels
WPI aqueous aliquots (10% wt. in protein matter) were prepared by
dispersing ca. 11.65 g 100 g− 1 of WPI powder into Milli-Q water (18 mΩ,
Merck-Millipore Inc., Burlington, US) and under gentle mechanical
stirring (IKA GmbH, Staufen, Germany) overnight to allow complete
hydration of the whey proteins. The pH of the WPI solutions was
adjusted at 7.00 ± 0.05 using NaOH 1 M. The insoluble particle impu­
rities were removed (<0.5% wt. of the total solids) by centrifuging the
WPI solutions at 10000g for 30 min (Multifuge X3R, Fiberlite F14–6,
ThermoFisher, Belgium). Then the WPI solutions were heat-treated at
80 ◦ C for 20 min using a shaking water bath (Julabo SW22, Seelbach,

Germany). The heat-treated WPI solutions underwent one-stage ho­
mogenisation at 700 bar (Panda plus 2000, GEA, Düsseldorf, Germany)
to prevent the formation of protein flocs. Flaxseed gum was then
dispersed into the WPI solution at varying concentrations, i.e. 0, 0.1,
0.25, 0.5, 0.75 and 1% wt., and the WPI-FG blends were kept under
magnetic stirring until complete gum dissolution. Sodium azide was
added (0.02% wt.) to prevent microbial spoilage.
The WPI-FG solutions were transferred into 5 mL Eppendorf tubes (in
duplicate per type of analysis and freeze-thaw cycle) and subjected to
sequential freezing-thaw cycles between 25 and − 28 ◦ C as described in
Hellebois et al. (2022). The WPI-FG systems were held at − 28 ◦ C for 20 h
between the freezing and thawing events.
2.3. WPI-FG phase separation measurements
Binary blends of WPI (0.1–10% wt.) and FG (0.025–1.0% wt.) were
prepared for the construction of the phase separation diagram. WPI-FG
solutions, pre-stained with Fast Green fluorophore, were microscopi­
cally assessed as described in paragraph 2.6.
The phase instability kinetics of the WPI-FG solutions used for the
cryogelation experiments were monitored with near infrared light (λ =
865 nm) every 10 s for 30 min at 2300 g, 25 ◦ C using LUMiSizer® (LUM
GmbH, Berlin, Germany). Data analysis was performed using the SEP­
View software (LUM GmbH, Berlin, Germany).
2.4. Syneresis of the cryogels

2. Materials and methods

The resistance of the final WPI-FG cryo-hydrogels against forced
serum exudation was assessed by monitoring the light transmittance (λ
= 865 nm) over the tube height at 2300g for 30 min with an interval of
10 s between measurements at 25 ◦ C using LUMiSizer®. The syneresis

rate representing the perpendicular velocity of the meniscus at the gel –
serum interface was expressed in μm s− 1.

2.1. Materials

2.5. Rheological measurements

PRODIET® 90S whey protein isolate (WPI) with a total protein
content of 85.8% (proximate composition on wet basis: whey proteins
85.8%, caseins <0.1%, moisture 5.9%, lipids 0.3%, lactose 5.5% and

The steady state flow behaviour of the WPI-FG solutions was
measured using a cone-plate geometry (CP50-1, 50 mm, angle 1◦ , gap of
0.101 mm) mounted on an oscillatory rheometer (MCR 302, Anton Paar,
2


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Graz, Austria).
The cryo-structuring performance of the WPI solutions in the absence
or presence of FG was assessed by recording the viscoelastic moduli G′
and G′′ at 1 Hz (in Pa, strain 0.1%, 25 ◦ C) at the end of each freeze-thaw
event. On the completion of the cryogenic processing, the final WPI-FG
hydrogels were stirred until achieving a homogenous gel paste and
subjected into frequency (0.1–100 Hz, strain 0.1%, 25 ◦ C, gap 1 mm)
and amplitude (0.01–1000% strain, 1 Hz, 25 ◦ C, gap 1 mm) sweep
measurements using a sandblasted plate – plate geometry (PP25/S, 25

mm). The obtained rheological spectra were analysed using the Rheo­
Compass software (Anton Paar, Graz, Austria).

centrifuged at 10000g for 10 min at ambient temperature and the
amount of soluble nitrogen in the supernatant in each solution was
quantified according to the Dumas method using a CHNS analyser
(Elementar Vario Macro Cube, Langensenbold, Germany). In addition,
the total nitrogen content of the final cryo-hydrogels was also
quantified.
2.10. Statistical analysis
The normal distribution of the data was verified by means of the
Shapiro-Wilk test and Q-Q plot representation statistical tests. To
determine the significance of FG addition on the rheological and phys­
icochemical properties of the solutions and cryogels, one-way ANOVA
was performed using Origin 2019b software (OriginLab Inc., Massa­
chusetts, USA). Tukey's multiple range test was used to separate mean
values when significant differences (p < 0.05) were detected.

2.6. Confocal Laser Scanning Microscopy (CLSM)
The WPI-FG solutions and the obtained cryo-hydrogels were micro­
structurally assessed using a CLSM microscope (LSM 880 with Airy scan,
Zeiss, Jena, Germany) at a 20× magnification. The fluorophore used
(Fast Green) was excited at 633 nm and the emitted fluorescence signal
was detected at 635–735 nm. In short, 10 μL of 0.05% wt. Fast Green was
added to 1 mL of the WPI-FG solution to stain the proteins noncovalently. The solutions were degassed by centrifugation for 1 min at
3000g. Then, aliquots of 300 μL were transferred into eight-well Nunc
Lab-Tek® II chamber microscope slides and were either tempered at
25 ◦ C for 30 min (for assessing the spontaneously occurring segregative
microphase separation) or immediately subjected to the aforementioned
cryogenic process. The WPI-FG cryo-hydrogels produced were assessed

at ambient temperature (25 ± 1 ◦ C).

3. Results and discussion
3.1. Characterisation of the WPI-FG solutions
3.1.1. Colloidal stability of the WPI-FG solutions
Aliquots of heat treated WPI aqueous solutions (0.5–10% wt.) con­
taining FG at concentrations varying from 0.005 to 1% wt. were pre­
pared in order to assess the thermodynamic compatibility of the
biopolymers. As a starting step the WPI-FG solutions were centrifuged at
3000g for 1 h (Kontogiorgos et al., 2009) and were visually assessed for
the occurrence of macroscopic segregative phase separation as indicated
by the presence of a well-distinguishing boundary between the proteinrich (lower) and polysaccharide-rich (upper) phases. As this was not
always achievable, the WPI-FG solutions that did not exhibit a clear
macroscopic phase separation were analysed by means of CLSM for
detecting microscopic phase separation.
As seen in the constructed phase diagram (Fig. 1), segregative
microphase separation took place at total biopolymer concentrations as
low as 1% wt. by cause of the intermolecular repulsive forces between
WPI and FG. From a colloidal perspective, phase separation is a kineti­
cally driven process, which arises from the thermodynamic in­
compatibility between the biopolymers as described by spinodal
decomposition models (Turgeon et al., 2003). It is well-documented that
segregative phase separation is related either to the Huggins-Flory or the
depletion flocculation theories (Doublier et al., 2000). The mechanistic
pathway of the segregative polymer-polymer interactions is dependent
on several parameters including the molecular characteristics (structure
conformation, molecular weight, surface charge density etc.) and total
concentration of the biopolymers (Tolstoguzov, 2006). In general,
depletion interactions are favoured in the case of colloidal suspensions,
i.e. spherical particles (e.g. casein micelles) dispersed in a semi-dilute

polysaccharide continuous phase (Turgeon et al., 2003). Although the
native whey proteins exhibit a good compatibility with polysaccharides,
their denatured (e.g. heat-treated) derivates e.g. oligomer soluble ag­
gregates, behave as colloidal spheres inducing depletion flocculation
mediated phase separation (Hellebois et al., 2022). The CLSM analysis of
the WPI-FG aqueous systems revealed the presence of diverse micro­
structural conformation features including fine and coarse water-inwater emulsions (i.e. polysaccharide-in-protein, protein-in-poly­
saccharide; e.g. see Fig. 1A,C,D), as well as bi-continuous interconnected
microstructures (near the critical composition), see Fig. 1B.
The colloidal stability kinetics of the WPI-FG aqueous systems as
associated with their microstructure conformational state are displayed
in Fig. 2. According to the acquired CLSM micrographs, WPI solutions
containing 0.1 and 0.25% wt. of FG exhibited a distinct water-in-water
emulsion and interconnected bi-continuous microstructure, respec­
tively. When cFG ≥ 0.5% wt. the formation of whey protein micro­
aggregates dispersed into the continuous semi-dilute polysaccharide
rich phase (for flaxseed gum: c* = 0.55% wt. according to Hellebois,

2.7. Particle size analysis
The mean size of the soluble protein aggregates in the initial WPI-FG
solutions, as well as of the protein particulates in the stirred cryohydrogels obtained was measured by static laser light scattering (SLS)
(Mastersizer 3000, Malvern Instruments Ltd., Worcestershire, UK). To
ensure a sufficient breakdown of the loosely interconnected protein
particulates, the stirred cryo-hydrogels were dispersed into MilliQ water
at the ratio of 1:10 prior to analysis. Then, all samples were dispersed
into MilliQ water in the measuring cell (Hydro MV) at a stirring speed of
3000 rpm for 90 s. The refractive index of whey protein and water was
set at 1.45 and 1.33, respectively.
2.8. Surface charge density
The zeta potential values of WPI and WPI-FG solutions were recorded

using dynamic light scattering (Zetasizer Nano, Malvern Instruments,
Worcestershire, UK). The zeta-potential of the solutions was determined
by diluting the initial solution one-hundred times into MilliQ water for
obtaining a final protein concentration of 0.1% wt. The pH of the so­
lutions obtained was adjusted to 7.0 using 0.1 M HCl and NaOH solu­
tions and filtered through a 0.2 μm cellulose acetate filter (VWR, Leuven,
Belgium). The refractive index of whey protein and water was set at 1.45
and 1.33, respectively.
2.9. Biopolymer molecular interactions
The relative contribution of the electrostatic and/or hydrogen (ES/
H), hydrophobic (Hyd) and disulphide (− S − S− ) molecular interactions
in the WPI-FG cryogels was determined according the method developed
by Tanger et al. (2021). The WPI-FG cryo-hydrogels were dissolved
(1:40) in different buffer solutions and shaken overnight using an
overhead rotator (GFL 3040, Burgwedel, Germany). The composition of
the buffers aiming at blocking the ES/H, ES/H + Hyd and ES/H + Hyd +
− S − S− interactions was as follows: buffer 1 (B1): 50 mM NaH2PO4, 50
mM Na2HPO4; buffer 2 (B2): B1 + 0.2% w/v sodium dodecyl sulfate
(SDS); buffer 3 (B3): B2 + 1.5% w/v dithiothreitol (DTT), respectively.
All the buffers were adjusted to pH = 7.5. Afterwards, the samples were
3


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Fig. 1. WPI-FG phase diagram and CLSM micrographs displaying different segregative phase separation conformation states obtained for: A) WPI 10% wt. + FG
0.05% wt., B) WPI 5% wt. + FG 0.1% wt., C) WPI 2.5% wt. + FG 0.138% wt. and D) WPI 2.5% wt. + FG 0.175% wt. solutions. Green squares represent single-phase
systems and tilted red squares two-phase systems. CLSM micrograph scale bar = 50 μm. (For interpretation of the references to colour in this figure legend, the reader

is referred to the web version of this article.)

Fortuin, et al. (2021)) was microscopically identified. In line with our
previous findings (Hellebois et al., 2022), the presence of FG led to a
concentration dependent increase in the mean size of the soluble whey
protein aggregates (Table 1). The changes in the mean size of the soluble
protein aggregates became more evident when cFG > 0.5% wt., which
implies that the transition of the aqueous polysaccharide rich contin­
uous phase from the dilute to the semi-dilute regime favours the selfassociation of the protein chains via weak non-covalent bonding such
as hydrogen bonds, hydrophobic interactions or electrostatic forces
between counter-charged patches in the polymeric backbone. Although
the protein-polysaccharide electrostatic interactions between opposite
charged side chain groups (e.g. − NH2 and − COOH) cannot be excluded,
it is considered unlike being the primary factor leading to the increase in
the protein aggregates mean size. This finds also support on the dynamic
light scattering measurements of the individual and binary biopolymer
solutions showing that the surface charge density was estimated at
− 33.2, − 32.2 and − 33.9 to − 35.5 mV for WPI, FG and WPI-FG solu­
tions, respectively. Under the hereby tested conditions, the FG-rich
aqueous phase appears to be far from the measured concentrated solu­
tion regime (c** > 2.5% wt., Hellebois, Fortuin, et al. (2021)) and thus,
it is assumed that the occurring segregative phenomena were not steri­
cally hindered. The viscosimetric analysis of the serum (polysaccharide
rich) samples obtained from the macroscopically phase separated WPIFG solutions (i.e. 0.1 ≤ cFG ≤ 0.5% wt.), revealed a reciprocal to cFG
increase in the apparent viscosity values (by 1.7, 73.4 and 93.4%,
respectively) compared to the FG solutions. Although it was not possible
to measure the macroviscosity of the serum phase when cFG > 0.5% wt.,
it is expected that the increase in the cFG elevates the osmotic pressure
between the polysaccharide rich and the colloidal suspension (protein
rich) microdomains, and hence, favours the competition between the

biopolymers for free volume occupancy (Doublier et al., 2000). Owing to
the higher molar mass, polymer chain flexibility and solvent affinity of
FG compared to whey proteins, the increase in the cFG impelled the WPI
oligomeric microspheres to supramolecular bridging via weak noncovalent interactions such as hydrogen or hydrophobic bonding, lead­
ing to the formation of whey protein flocs (as detected by SLS and CLSM)
when cFG > c* = 0.55% wt.
As displayed in Fig. 2, the colloidal stability of the WPI-FG solutions
was found to be significantly lower at 0.1 ≤ cFG ≤ 0.5% wt., exerting a
straightforward correlation with the intensity of the biopolymer

demixing phenomena. At cFG ≥ 0.75% wt., a substantial drop in the
instability index values was detected (Table 1), which suggests that the
biopolymers' demixing commences to be mass transfer driven.
3.1.2. Rheological behaviour of the WPI-FG solutions
The flow behaviour curves of the WPI-FG solutions are given in
Fig. 3A. Corroborating our recent findings (Hellebois et al., 2022), all
systems exhibited a weak to pronounced shear thinning behaviour
depending on the FG concentration (i.e. n ∝ e− 1.35c, where n and c
denote the flow behaviour index and FG concentration, respectively). It
has previously shown (Kontogiorgos et al., 2009), that the weak pseu­
doplastic behaviour of heat-treated WPI dispersions stems from the
disentanglement of the clusters of the soluble whey protein oligomeric
aggregates. On the other hand, the increase in the rheological behaviour
of the WPI-FG solutions can be mechanistically more complex as it can
be influenced not only by the intrinsic parameters of each biopolymer (e.
g. molecular weight, hydrodynamic radius, degree of polymerisation
and branching, solvent affinity etc.) but also their thermodynamic
compatibility (Fang, 2021). In the latter case, the concentration of the
biopolymers due to excluded volume effects, can enhance the intermo­
lecular crosslinking of the alike biopolymer species, which in turn may

affect not only the microrheology of the phase separated microdomains
but also the overall macroscopic rheological behaviour of the
biopolymer blend solution (Firoozmand et al., 2009). To evaluate the
apparent viscosity dependence on cFG, the experimental viscosimetric
data (logη; for comparative reasons η was measured at 50 s− 1) were
plotted as function of logcFG (Fig. 3B). As illustrated in Fig. 3B, a change
in the slope of the double logarithmic plot, referred here as breakpoint
(cFGbr), was identified. The calculated slopes for the lower and upper
linear plot parts were 1.74 and 3.53, respectively, which interestingly
are in keeping with the slopes of the dilute and semi-dilute regime of
flaxseed solutions, i.e. ∝ C1.72 and C3.92, respectively (Hellebois, Fortuin,
et al., 2021). In a like manner, similarities between cFGbr (0.51% wt.)
and c* (0.55% wt.) were found, which suggests that FG retains its
macromolecular solution properties in the WPI dispersions.
To unveil the synergism or antagonism between the co-existing
macromolecules (WPI and FG) in solution, the biopolymer interaction
coefficient R, was calculated as follows (Agoda-Tandjawa et al., 2012):

4


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Fig. 2. Dynamic light transmission changes and CLSM assisted microstructural visualisation of WPI-FG solutions for assessing the temporal evolution of the colloidal
stability of WPI-FG under accelerated storage conditions (25 ◦ C, 2300g). CLSM micrograph scale bar = 50 μm.

their thermodynamic compatibility. At the single-phase regime, the
rheological behaviour of the biopolymer solution obeys to the ideal

mixture rule. At biopolymer compositions in the proximity of the
binodal curve a progressive increase in the macroviscosity and visco­
elasticity of the aqueous systems may be observed owing to the polymerpolymer molecular interactions (Tolstoguzov, 2003). In segregated
biopolymer solutions, antagonistic or synergistic effects on their rheo­
logical behaviour have been experimentally justified (Bourriot et al.,
1999; Garnier et al., 1995; Sadeghi et al., 2018). According to our
findings, the synergistic interactions between WPI and FG can be
ascribed to either the enhancement of the alike polymer – polymer
molecular interactions in the depleted phases (in the dilute solution
state, i.e. at cFG ≤ cFGbr) or the aggregation of the soluble whey protein
clusters (cFG ≥ cFGbr).

Table 1
Physicochemical properties of WPI-FG solutions (0–1% wt.).
Gum content

ζ-potential

(% wt.)

(mV)

0
0.1
0.25
0.5
0.75
1

−

−
−
−
−
−

33.2 ±
33.8 ±
34.7 ±
34.4 ±
33.9 ±
35.2 ±

Instability index

Mean particle size d4,3
(μm)

a

1.6
1.5a
0.7a
1.4a
0.8a
1.8a

0.031 ±
0.196 ±
0.383 ±

0.276 ±
0.078 ±
0.024 ±

a

0.008
0.031b
0.064c
0.068bc
0.029a
0.024a

0.243 ± 0.01a
0.321 ± 0.01ab
0.369 ± 0.01b
0.564 ± 0.07c
7.7 ± 0.56d
15.0 ± 0.50e

a-d

The different letters between rows for each property indicate a significant
difference (p < 0.05).

R=

ηWPI−

− (ηFG + ηWPI )

ηFG + ηWPI

FG

(1)
3.2. Cryostructuration performance of the WPI-FG solutions

As displayed in Fig. 3C, for all WPI-FG solutions R > 0, which in­
dicates a synergistic interaction between the biopolymers in terms of
apparent viscosity. It is well known that the steady state and oscillatory
rheological behaviour of biopolymer aqueous systems is associated with

To understand the cryo-hydrogel formation performance of WPI as
influenced by the presence of FG, the elastic modulus G′ and particle size
(d4,3) values of the WPI-FG systems were recorded for five consecutive
5


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Fig. 3. Flow behaviour curves (A), double logarithmic plot of the viscosity vs. concentration (B) and biopolymer interaction coefficient R (C) of WPI solutions as
influenced by FG concentrations (0.1–1% wt.) a-dDifferent letters between bars denote a significant difference (p < 0.05).

freeze-thaw cycle events between 25 and − 28 ◦ C (Fig. 4). In keeping
with our previous findings (Hellebois et al., 2022), the sequential freezethaw processing of the WPI-FG solutions led to a progressive increase in
the G′ values. After a single freeze-thaw event, a significant improve­
ment of the elastic modulus of the WPI-FG systems − proportionally to
cFG− was observed. The cryogenic structuring was kept improving until

the end of the second freeze-thaw event and then, a pseudo-equilibrium
viscoelastic state was achieved (Fig. 4A). It should be pointed out that
the reinforcement of the elasticity of the formed cryogels was adversely
associated with the colloidal instability index of the polymeric precursor
systems (r = − 0.695, p < 0.05). Contrarily to the WPI – alfalfa gum
systems (Hellebois et al., 2022), which exerted a cycle-by-cycle

structuring effect owing to the concomitant cryogelling activity of the
biopolymer precursors, in the case of WPI-FG systems it was not possible
to identify any clear contribution of FG in the cryogelation process. The
negligible cryogelation ability of FG was also confirmed when individual
FG solutions (c* = 0.55% < cFG < 5% wt.) were cryogenically processed
at the same conditions (data not shown). Hence, it is dictated that FG
contributes to the viscoelastic build-up of the cryogels probably via an
inert filler role partitioning as a major solute of the cryo-concentrated
serum phase (see also section 3.3). Yet, the increasing concentration
of FG did not impair (0.1 ≤ cFG ≤ 0.75% wt.) or even led to an
improvement (at cFG = 1% wt.) of the elasticity of the final cryohydrogels. Mixed effects as concerns the impact of the polysaccharidic

Fig. 4. Dynamic changes in the elastic modulus (A) and mean particle size (B) of WPI-FG solutions occurring throughout cryogenic processing between 25 and
− 28 ◦ C. a-f, A-CDifferent letters between concentrations (lowercase) or among the cycles (uppercase) indicate a significant difference (p < 0.05).
6


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

precursor on the G′ values of WPI based cryogels have been previously
reported for galactomannans (i.e. active filler effect), (Hellebois et al.,

2022) and xanthan/curdlan co-polymer blend (i.e. adverse effect)
(Shiroodi et al., 2015) beyond a specific concentration have been
reported.
In Fig. 5 the microstructural features of the WPI-FG cryogels
following a single and five consecutive freeze-thaw cycles are illus­
trated. As seen in the CLSM acquired micrographs, cryogenic processing
of the WPI-FG precursor solutions led to the formation of macroporous
hydrogel templates consisting of irregularly interconnected stranded
WPI-based structures. Interestingly, the hydrogel templates developed
in the end of the first freeze-thaw event retained partially the micro­
phase separated structure conformation of the unprocessed polymeric
solutions; this was fairly evidenced in the case of the bicontinuous WPIFG solution (Fig. 5–0.25% wt.). Likewise, the presence of the clustered
whey protein aggregates was evidently seen in the case of the cryogels
comprising at least 0.75% of FG. The ability of arresting the micro­
structure of segregated biopolymer systems via sol-gel (cold, heat or acid
induced) or rubbery to glass physical state transitions is well-reported
(Tanaka, 2012; Turgeon et al., 2003).
Throughout freeze-thaw cycling, a re-organisation of the cryogels

microstructure was observed as represented by an overall increase in the
thickness of the protein strands and the macropores mean size. This
behaviour is suggestive of the inertness of FG as cryostructuring or
cryostabilising polymeric precursor. Indeed, it was recently shown that
when a cryogel forming polymeric precursor was added into the WPI
dispersions, no significant changes in the microstructural elements of
the cryogels could be detected throughout the cryostructuration process
(Hellebois et al., 2022).
It is well documented that the cryotropically induced hydrogels may
be obtained via either physical (non-covalent) or chemical (covalent)
crosslinking (Lozinsky, 2018, 2020; Lozinsky & Okay, 2014). As for food

biopolymers, i.e. polysaccharides or proteins, cryogelation may occur
via different mechanistic pathways. In general, polysaccharide based
cryogels are primarily of physical character i.e. the cryogelation process
is triggered through weak interchain polymer crosslinking owing to
hydrogen bonding and hydrophobic interactions (Giannouli & Morris,
2003; Hellebois, Gaiani, et al., 2021; Lazaridou & Biliaderis, 2004;
Lozinsky et al., 2000; Tanaka et al., 1998). On the other hand, food
proteins e.g. milk proteins, egg white or gelatine can be formed on the
basis of concomitant non-covalent and covalent (mainly disulphide
bonding) interactions (Balaji et al., 2019; Hellebois et al., 2022; Savina

Fig. 5. CLSM micrographs of the WPI-FG cryogel obtained after one and five freeze-thaw cycles influenced by the presence of FG (0–1% wt.). Scale bar = 50 μm.
7


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

et al., 2011).
To get a better overview of the mechanistic background of the cry­
ostructuration of the WPI-FG systems, the obtained hydrogels at the end
of the freeze-thaw process were dissolved in different solvents aiming at
cleaving specific biopolymer molecular interactions such as: a)
hydrogen bonding and electrostatic complexation (phosphate buffer), b)
hydrophobic interactions (SDS containing phosphate buffer), and c)
covalent bonding (DTT/SDS containing phosphate buffer (Tanger et al.,
2021). As illustrated in Fig. 6, the cryostructuration of the WPI-FG so­
lutions was governed by thiol-disulphide (− SH/− S − S− ) interchange
interactions favoured by the heat induced dissociation and unfolding of

the native whey protein molecules and the increased molecular mobility
of the peptides (Nicolai et al., 2011). To a lesser extent, nonspecific (i.e.
electrostatic and hydrogen bond) and hydrophobic interactions, ac­
counting for approx. 20 and 3% of the total protein solubility, respec­
tively, also contributed to the cryo-hydrogel formation (Fig. 6). As
concerns the contribution of FG to the cryotropically mediated molec­
ular interactions, a proportional to FG content dependence of the
nonspecific interactions (r = − 0.856, p < 0.001) and disulphide bridging
(r = 0.837, p < 0.001), respectively, was identified. As urea could not be
tested as a hydrogen bond blocking agent (due to method restrictions)
and taking into account the minor contribution of FG to the overall
surface charge density, hereby it is hypothesised that the nonspecific
interactions represent mainly hydrogen bond bridging. In this context, it
appears that the reduction of the hydrogen bonding prevalence on FG
addition stemmed from the segregative phase separation of the bio­
polymers. The latter hindered partially the protein – polysaccharide
hydrogen bond interactions, whilst at the same time arose the − SH/− S
− S− interchain protein crosslinking (r = − 0.908, p < 0.001). On the
other hand, the contribution of FG to the prevalence of the hydrophobic
interactions remained unclear. Similar effects have been also reported in
the case of microphase separated WPI-polysaccharide heat-set hydrogels
(Zhang et al., 2021).

Table 2
Viscoelastic properties of the WPI-FG cryogels obtained after five freeze-thaw
cycles.
Gum
content

Amplitude sweeps

(% wt.)

(Pa)

(%)

(Pa)

(Pa)

9.5 ±
0.7a
12.5
± 0.2a
8.0 ±
1.5a
12.2
± 2.5a
24.0
± 4.8b
26.9
± 1.9b

1.03 ±
0.09a
1.29 ±
0.12a
1.17 ±
0.21a

1.22 ±
0.18a
1.29 ±
0.12a
1.43 ±
0.00a

109.4 ±
20.5a
113.8 ±
1.6a
58.4 ±
0.5a
101.7 ±
7.6a
191.6 ±
61.6b
230.3 ±
61.9b

80.3 ±
13.5a
88.4 ±
8.2a
66.1 ±
9.9a
96.4 ±
8.3a
200.3 ±
35.3b

227.1 ±
5.5b

0
0.1
0.25
0.5
0.75
1

τy

γ˙ LVE

Frequency sweeps

τf

G'f

G'-f slope

tanδ at 1
Hz

0.081 ±
0.050a
0.093 ±
0.006ab
0.096 ±

0.009b
0.103 ±
0.003bc
0.109 ±
0.005cd
0.119 ±
0.006d

0.169 ±
0.009a
0.175 ±
0.008a
0.185 ±
0.006ab
0.199 ±
0.010bc
0.217 ±
0.013cd
0.236 ±
0.007d

a-d

Different letters between concentrations for each rheological property indi­
cate a significant difference (p < 0.05). Abbreviations used: γ˙ LVE : strain at the
end of the LVE boundary; τy: yield stress; τf: flow stress; G'f: Storage modulus at
flow stress. Measurements performed at 25 ◦ C, 1 Hz for amplitude sweeps and
0.1% strain for frequency sweeps.

(Fig. 7) and colloidal instability (Fig. 8) characteristics. For mapping the

mechanical profile of obtained cryogels, the normalised elastic modulus
(i.e. G′ /G′ 0) strain sweep rheological spectra were assessed (Suppl.
Fig. 1A). According to Ross-Murphy (1995), the strain dependence of the
reduced modulus can be used to distinguish weak from strong gels. As
illustrated in Suppl. Fig. 1A and Table 2, the G′ /G′ 0 values started to
˙
become strain dependent at γ≫0.05%
that is suggestive of a strong gel.
The yield point τy (i.e., the minimum required stress to induce irre­
versible internal structure conformational changes) values were signif­
icantly increased only when cFG ≥ 0.75% wt. Further increase in the
deformation stress (at τf) resulted in the structural collapse of the cry­
ogels as result of the increasing molecular motion of the polymeric
moieties that are not strongly fixated in the whey protein gel network.
However, only when cFG exceeded 0.75% wt., a significant increase in
the flow point values could be observed. The strain sweep data confirm
the hypothesis that FG does not possess an active filler role; instead, it
acted as a thickener of the serum phase present in the cryogel
macropores.
To further investigate their frequency dependent behaviour, the
WPI-FG cryogels were subjected to frequency sweep tests (Suppl.
Fig. 1B). Fitting the elastic modulus – frequency data to the power model
′
i.e., G′ = K′ ωn allowed the calculation of the slopes of the rheological
spectra. As given in Table 2, the slopes ranged from 0.08 to 0.12 indi­
cating the formation of strong physical gels that involve covalent
crosslinking. The obtained values are generally in keeping with the
literature data regarding cold or acid protein gels comprising flaxseed
gum (Chen et al., 2016; Kuhn et al., 2011; Soukoulis et al., 2019). The
progressive increase in the n' parameter and damping factor (tanδ)

values implies that the FG presence in the WPI cryo-hydrogels amplified
their viscoelastic character. Contrary to galactomannans, which exert an
active filler activity endowing composite-like characteristics to the WPI
cryogels (Hellebois et al., 2022), the FG did not improve remarkably the
elastic component of the cryo-hydrogels due to its inexistent cryogelling
ability.
For gaining a better insight into the structural elements of the WPIFG cryogels, the obtained CLSM micrographs at the end of the freezethaw processing were subjected to image analysis using the AngioTool
software adopting the computational analysis protocol described in
detail by (Zudaire et al., 2011). In Fig. 7A,B is given a schematic rep­
resentation of the CLSM acquired and the AngioTool computed micro­
structural conformation images of the WPI-FG cryogels. Selected

3.3. Characterisation of the final WPI-FG cryogels
The cryo-hydrogels obtained at the end of the 5th freeze-thaw event
were assessed for their viscoelastic (Table 2), structural conformational

Fig. 6. Normalised protein solubility indicative of the occurrence of noncovalent and covalent molecular interactions quantified in the WPI-FG cry­
ogels obtained in the end of the cryogenic processing. a-dDifferent letters be­
tween bars (within the same type of molecular interaction) denote a significant
difference (p < 0.05).
8


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Fig. 7. Indicative CLSM micrograph WPI cryogel
obtained after five freeze-thaw cycles (A) and its
analysed protein network using the AngioTool soft­

ware (B). The blue dots and red vessels represent the
network junctions and skeleton, respectively. The
obtained macropore area occupancy is depicted in (C)
and lacunarity in (D). a-cDifferent letters between
concentrations for each stage indicate a significant
difference (p < 0.05). (For interpretation of the ref­
erences to colour in this figure legend, the reader is
referred to the web version of this article.)

parameters such as the macropores space occupancy (Fig. 7C) and the
lacunarity (Fig. 7D) were calculated. As clearly depicted in Fig. 7C,D the
macroporosity and the heterogeneity (lacunarity) of the obtained cry­
ogels were positively (r = 0.651, p < 0.05; r = 0.679, p < 0.05,
respectively) associated with the colloidal instability of the bio­
polymeric precursor system. Thus, the hydrogels obtained from the
cryogenic processing of bicontinuous or smaller interconnected phase
separated systems were found as having a predominantly heterogenic
macroporous structure. Finally, it should be noted that at least 1% wt of
FG is required for attaining a significant improvement of the structural
uniformity and protein strands space occupancy. On this occasion a
significant increase (p < 0.001) in the number of the vessel intercon­
nection (junction) points was observed (Suppl. Fig. 2).
One of the major challenges in the design of mechanically and
physically resilient biopolymer cryogels is to prevent excessive solvent
losses due to the structural re-organisation of the gel during the ageing
process (Lucey, 2002). In general, milk protein gels are prone to syn­
eresis, that is the spontaneous exudation of the loosely held water found
in the interspace of the entangled protein network (Lucey, 2020). It is
well documented that parameters such as the coarseness, permeability
and stiffness of the protein gel network are major determinants of the

syneresis of milk protein gels (Urbonaite et al., 2015, 2016). To avoid
any bias in the computed syneresis kinetics due to the mechanical
disturbance of the hydrogel, aliquots of WPI-FG solutions (ca. 1.8 mL)
were transferred into the LUMiSizer cuvettes and processed cryogeni­
cally as above mentioned. As illustrated in Fig. 8, all cryo-hydrogels
exhibited significant syneresis rates ranging from 7.5 to 9.1 μm s− 1.
Although there were no significant differences in the syneresis rates of
the cryogels containing up to 0.5% wt. of FG, the exuded serums had
distinctively different turbidity. The quantification of the residual pro­
tein in the serum samples, revealed a progressively eminent reduction in
the proteinaceous matter content as function of cFG. Similar behaviour
has been reported in WPI heat-set gels containing acidic hetero­
polysaccharides (He et al., 2021). It is assumed that the residual protein

partitioning in the serum dictates a lower ability of the proteins to un­
dergo disulphide bonding. Interestingly, the rheological characterisa­
tion of the FG containing serum and the individual FG solutions (WPI
free) failed to unveil any significant differences in terms of their
apparent viscosity (Fig. 8C). This gives support to our hypothesis that FG
is found primarily as solute in the serum and therefore, it is ability to
enable the formation of elastically active network chains, as previously
showcased for FG-WPI acid gels (Soukoulis et al., 2019), is limited.
4. Conclusions
The co-structuring and stabilising role of flaxseed gum in cryo­
tropically produced whey protein isolate hydrogels was studied. Flax­
seed gum exhibited a strong thermodynamic incompatibility with whey
proteins resulting into water-in-water emulsion like, bicontinuous,
interconnected or aggregated conformational states depending on the
protein to polysaccharide content ratios. The segregative phase sepa­
ration phenomena were primarily ascribed to a depletion flocculation

mechanism. The cryogenic processing of the WPI-FG solutions resulted
in the formation of macroporous hydrogels. From a molecular stand­
point, the cryotropic gelation was mediated primarily via covalent
interchain protein crosslinking (cysteine thiol-disulphide bridging) and
to a lesser extent via non-covalent (i.e. hydrogen bonds and hydropho­
bic) polymeric interactions. Although FG had a modulating role on the
prevalence of the molecular interactions, it did not arrest the micro­
structural re-organisation of the protein gel network during cryogenic
processing. This was attributed to the inability of FG to undergo cryo­
tropic gelation and therefore, to enact an active filler role within the
fixated protein gel network. The presence of FG at concentrations
≥0.75% wt. improved the gel homogeneity and interconnectivity of the
whey protein strands while it reduced the macropores mean size. Hence,
at least 1% wt. of FG is required for constructing WPI cryo-hydrogels
that could endow tangible structuring, stabilising and texturing bene­
fits to complex food matrices.
9


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Fig. 8. LUMiSizer transmission spectra and macroscopic imaging of the 5-fold cycled WPI-FG cryogels after 30 min of centrifugation at 2300g, 25 ◦ C and syneresis
rate (A), protein content in the exuded serum (B) and viscosity of the exuded serum compared to FG (WPI free) solutions (dashed bars) (C). a-dDifferent letters
between the bars indicate a significant difference (p < 0.05).

CRediT authorship contribution statement

Acknowledgement

T. Hellebois: Conceptualisation, Investigation, Formal Analysis,
Writing Original Draft, Writing – Review and Editing.
S. Cambier: Investigation, Writing – Review and Editing.
A. Noo: Investigation, Resources.
C. Gaiani: Writing - Review and Editing, Project administration, PhD
student (TH) supervision.
C. Soukoulis: Conceptualisation, Writing - Review and Editing, PhD
student (TH) supervision, Project administration, Funding Acquisition.

This work was supported by the Luxembourg National Research
Fund (FNR) (Project PROCEED: CORE/2018/SR/12675439). Mrs.
Manon Hiolle (Ingredia SA, France) is thanked for generously providing
the whey protein isolate.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.carbpol.2022.119424.

Declaration of competing interest

References

The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.

Agoda-Tandjawa, G., Durand, S., Gaillard, C., Garnier, C., & Doublier, J.-L. (2012).
Rheological behaviour and microstructure of microfibrillated cellulose suspensions/

10

T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

low-methoxyl pectin mixed systems.Effect of calcium ions. Carbohydrate Polymers, 87
(2), 1045–1057. />Balaji, P., Murugadas, A., Shanmugaapriya, S., & Abdulkader Akbarsha, M. (2019).
Fabrication and characterization of egg white cryogel scaffold for three-dimensional
(3D) cell culture. Biocatalysis and Agricultural Biotechnology, 17, 441–446. https://
doi.org/10.1016/j.bcab.2018.12.019
Basiri, S., Haidary, N., Shekarforoush, S. S., & Niakousari, M. (2018). Flaxseed mucilage:
A natural stabilizer in stirred yogurt. Carbohydrate Polymers, 187, 59–65. https://doi.
org/10.1016/j.carbpol.2018.01.049
Bourriot, S., Garnier, C., & Doublier, J.-L. (1999). Phase separation, rheology and
microstructure of micellar casein–guar gum mixtures. Food Hydrocolloids, 13(1),
43–49. />Chen, C., Huang, X., Wang, L., Li, D., & Adhikari, B. (2016). Effect of flaxseed gum on the
rheological properties of peanut protein isolate dispersions and gels. LWT, 74,
528–533. />Coria-Hern´
andez, J., Mel´
endez-P´erez, R., M´endez-Albores, A., & Arjona-Rom´
an, J. L.
(2021). Effect of cryostructuring treatment on some properties of xanthan and
karaya cryogels for food applications. Molecules, 26(9), 2788. />10.3390/molecules26092788
Dea, I. C. M., Morris, E. R., Rees, D. A., Welsh, E. J., Barnes, H. A., & Price, J. (1977).
Associations of like and unlike polysaccharides: Mechanism and specificity in
galactomannans, interacting bacterial polysaccharides, and related systems.
Carbohydrate Research, 57, 249–272. />81935-7
Doublier, J.-L., Garnier, C., Renard, D., & Sanchez, C. (2000). Protein–polysaccharide
interactions. Current Opinion in Colloid & Interface Science, 5(3), 202–214. https://

doi.org/10.1016/S1359-0294(00)00054-6
Fang, Y. (2021). Chapter 5 - mixed hydrocolloid systems. In G. O. Phillips, &
P. A. Williams (Eds.), Handbook of Hydrocolloids (3rd ed., pp. 125–155). Woodhead
Publishing. />Firoozmand, H., Murray, B. S., & Dickinson, E. (2009). Microstructure and rheology of
phase-separated gels of gelatin+oxidized starch. Food Hydrocolloids, 23(4),
1081–1088. />Garnier, C., Schorsch, C., & Doublier, J.-L. (1995). Phase separation in dextran/locust
bean gum mixtures. Carbohydrate Polymers, 28(4), 313–317. />10.1016/0144-8617(95)00090-9
Giannouli, P., & Morris, E. R. (2003). Cryogelation of xanthan. Food Hydrocolloids, 17(4),
495–501. />Goff, H. D., Ferdinando, D., & Schorsch, C. (1999). Fluorescence microscopy to study
galactomannan structure in frozen sucrose and milk protein solutions. Food
Hydrocolloids, 13(4), 353–362. />Gulrez, S. K. H., Al-Assaf, S., & Phillips, G. O. (2011). Hydrogels: Methods of Preparation,
Characterisation and Applications. IntechOpen. />Guo, Y., Wu, M., Li, R., Cai, Z., & Zhang, H. (2022). Thermostable physically crosslinked
cryogel from carboxymethylated konjac glucomannan fabricated by freeze-thawing.
Food Hydrocolloids, 122, Article 107103. />foodhyd.2021.107103
Gurikov, P., & Smirnova, I. (2018). Non-conventional methods for gelation of alginate.
Gels, 4(1), 14. />He, Z., Ma, T., Zhang, W., Su, E., Cao, F., Huang, M., & Wang, Y. (2021). Heat-induced
gel formation by whey protein isolate-Lycium barbarum polysaccharides at varying
pHs. Food Hydrocolloids, 115, Article 106607. />foodhyd.2021.106607
Hellebois, T., Fortuin, J., Xu, X., Shaplov, A. S., Gaiani, C., & Soukoulis, C. (2021).
Structure conformation, physicochemical and rheological properties of flaxseed
gums extracted under alkaline and acidic conditions. International Journal of
Biological Macromolecules, 192, 1217–1230. />ijbiomac.2021.10.087
Hellebois, T., Gaiani, C., Fortuin, J., Shaplov, A., & Soukoulis, C. (2021). Cryotropic gelforming capacity of alfalfa (Medicago sativa L.) and fenugreek (Trigonella foenum
graecum) seed galactomannans. Carbohydrate Polymers, 267, Article 118190.
/>Hellebois, T., Gaiani, C., & Soukoulis, C. (2022). Freeze − thaw induced structuration of
whey protein − alfalfa (Medicago sativa L.) galactomannan binary systems. Food
Hydrocolloids, 125, Article 107389. />foodhyd.2021.107389
Kontogiorgos, V., Tosh, S. M., & Wood, P. J. (2009). Phase behaviour of high molecular
weight oat β-glucan/whey protein isolate binary mixtures. Food Hydrocolloids, 23(3),
949–956. />ˆ L. F., & da Cunha, R. L. (2011). Cold-set whey protein–flaxseed

Kuhn, K. R., Cavallieri, A.
gum gels induced by mono or divalent salt addition. Food Hydrocolloids, 25(5),
1302–1310. />Lazaridou, A., & Biliaderis, C. G. (2004). Cryogelation of cereal β-glucans: Structure and
molecular size effects. Food Hydrocolloids, 18(6), 933–947. />j.foodhyd.2004.03.003
Lazaridou, A., Kritikopoulou, K., & Biliaderis, C. G. (2015). Barley β-glucan cryogels as
encapsulation carriers of proteins: Impact of molecular size on thermo-mechanical
and release properties. Bioactive Carbohydrates and Dietary Fibre, 6(2), 99–108.
/>Li, X., Li, D., Wang, L., Wu, M., & Adhikari, B. (2012). The effect of addition of flaxseed
gum on the rheological behavior of mixed flaxseed gum–casein gels. Carbohydrate
Polymers, 88(4), 1214–1220. />Liu, J., Shim, Y. Y., Tse, T. J., Wang, Y., & Reaney, M. J. T. (2018). Flaxseed gum a
versatile natural hydrocolloid for food and non-food applications. Trends in Food
Science & Technology, 75, 146–157. />
Liu, Y., Liu, Z., Zhu, X., Hu, X., Zhang, H., Guo, Q., Yada, R. Y., & Cui, S. W. (2021). Seed
coat mucilages: Structural, functional/bioactive properties, and genetic information.
Comprehensive Reviews in Food Science and Food Safety, 20(3), 2534–2559. https://
doi.org/10.1111/1541-4337.12742. Scopus.
Lozinsky, V. I. (2018). Cryostructuring of polymeric systems. 50.† cryogels and
cryotropic gel-formation:Terms and definitions. Gels, 4(3), 77. />10.3390/gels4030077
Lozinsky, V. I. (2020). Cryostructuring of polymeric systems. 55. Retrospective view on
the more than 40 years of studies performed in the A.N.Nesmeyanov Institute of
Organoelement Compounds with respect of the cryostructuring processes in
polymeric systems. Gels, 6(3), 29. />Lozinsky, V. I., & Okay, O. (2014). Basic principles of cryotropic gelation. In O. Okay
(Ed.), Polymeric cryogels: Macroporous gels with remarkable properties (pp. 49–101).
Springer International Publishing. />Lozinsky, V. I., Damshkaln, L. G., Brown, R., & Norton, I. T. (2000). Study of
cryostructuring of polymer systems. XIX. On the nature of intermolecular links in the
cryogels of locust bean gum. Polymer International, 49(11), 1434–1443. https://doi.
org/10.1002/1097-0126(200011)49:11<1434::AID-PI525>3.0.CO;2-F
Lucey, J. A. (2002). Formation and physical properties of milk protein gels. Journal of
Dairy Science, 85(2), 281–294. />Lucey, J. A. (2020). Chapter 16 - Milk protein gels. In M. Boland, & H. Singh (Eds.), Milk
proteins (3rd ed., pp. 599–632). Academic Press. />Nakagawa, K., & Nishimoto, N. (2011). Cryotropic gel formation for food nutrients

encapsulation - A controllable processing of hydrogel by freezing. Procedia Food
Science, 1, 1968–1972. />Nicolai, T., Britten, M., & Schmitt, C. (2011). β-Lactoglobulin and WPI aggregates:
Formation, structure and applications. Food Hydrocolloids, 25(8), 1945–1962.
/>Patmore, J. V., Goff, H. D., & Fernandes, S. (2003). Cryo-gelation of galactomannans in
ice cream model systems. Food Hydrocolloids, 17(2), 161–169. />10.1016/S0268-005X(02)00048-6
Regand, A., & Goff, H. D. (2003). Structure and ice recrystallization in frozen stabilized
ice cream model systems. Food Hydrocolloids, 17(1), 95–102. />10.1016/S0268-005X(02)00042-5
Ross-Murphy, S. B. (1995). Structure–property relationships in food biopolymer gels and
solutions. Journal of Rheology, 39(6), 1451–1463. />1.550610
Sadeghi, F., Kadkhodaee, R., Emadzadeh, B., & Phillips, G. O. (2018). Phase behavior,
rheological characteristics and microstructure of sodium caseinate-Persian gum
system. Carbohydrate Polymers, 179, 71–78. />carbpol.2017.09.060
Savina, I. N., Gun'ko, V. M., Turov, V. V., Dainiak, M., Phillips, G. J., Galaev, I. Y., &
Mikhalovsky, S. V. (2011). Porous structure and water state in cross-linked polymer
and protein cryo-hydrogels. Soft Matter, 7(9), 4276–4283. />C0SM01304H
Shiroodi, S. G., Rasco, B. A., & Lo, Y. M. (2015). Influence of xanthan-curdlan hydrogel
complex on freeze-thaw stability and rheological properties of whey protein isolate
gel over multiple freeze-thaw cycle. Journal of Food Science, 80(7), E1498–E1505.
/>Soukoulis, C., Gaiani, C., & Hoffmann, L. (2018). Plant seed mucilage as emerging
biopolymer in food industry applications. Current Opinion in Food Science, 22, 28–42.
/>Soukoulis, C., Cambier, S., Serchi, T., Tsevdou, M., Gaiani, C., Ferrer, P., Taoukis, P. S., &
Hoffmann, L. (2019). Rheological and structural characterisation of whey protein
acid gels co-structured with chia (Salvia hispanica L.) or flax seed (Linum
usitatissimum L.) mucilage. Food Hydrocolloids, 89, 542–553. />10.1016/j.foodhyd.2018.11.002
Sowasod, N., Nakagawa, K., Tanthapanichakoon, W., & Charinpanitkul, T. (2013).
Cryogel based oil encapsulation for controlled release of curcumin by using a ternary
system of chitosan, kappa-carrageenan, and carboxymethylcellulose sodium salt.
Advanced Materials Research, 701, 98–102. />net/AMR.701.98
Tanaka, H. (2012). Viscoelastic phase separation in soft matter and foods. Faraday
Discussions, 158, 371–406. />Tanaka, R., Hatakeyama, T., & Hatakeyama, H. (1998). Formation of locust bean gum

hydrogel by freezing–thawing. Polymer International, 45(1), 118–126. https://doi.
org/10.1002/(SICI)1097-0126(199801)45:1<118::AID-PI908>3.0.CO;2-T
Tanger, C., Andlinger, D. J., Brümmer-Rolf, A., Engel, J., & Kulozik, U. (2021).
Quantification of protein-protein interactions in highly denatured whey and potato
protein gels. MethodsX, 8, Article 101243. />mex.2021.101243
Tolstoguzov, V. (2003). Thermodynamic considerations on polysaccharide functions.
Polysaccharides came first. Carbohydrate Polymers, 54(3), 371–380. />10.1016/S0144-8617(03)00210-8
Tolstoguzov, V. (2006). Phase behaviour in mixed polysaccharide systems. In Food
polysaccharides and their applications (2nd ed., pp. 590–628). CRC Press, Taylor and
Francis.
Turgeon, S. L., Beaulieu, M., Schmitt, C., & Sanchez, C. (2003). Protein–polysaccharide
interactions: Phase-ordering kinetics, thermodynamic and structural aspects. Current
Opinion in Colloid & Interface Science, 8(4), 401–414. />S1359-0294(03)00093-1
Urbonaite, V., de Jongh, H. H. J., van der Linden, E., & Pouvreau, L. (2015). Permeability
of gels is set by the impulse applied on the gel. Food Hydrocolloids, 50, 7–15. https://
doi.org/10.1016/j.foodhyd.2015.03.024

11


T. Hellebois et al.

Carbohydrate Polymers 289 (2022) 119424

Urbonaite, V., van der Kaaij, S., de Jongh, H. H. J., Scholten, E., Ako, K., van der
Linden, E., & Pouvreau, L. (2016). Relation between gel stiffness and water holding
for coarse and fine-stranded protein gels. Food Hydrocolloids, 56, 334–343. https://
doi.org/10.1016/j.foodhyd.2015.12.011
Yang, X., Li, A., Li, X., Sun, L., & Guo, Y. (2020). An overview of classifications,
properties of food polysaccharides and their links to applications in improving food

textures. Trends in Food Science & Technology, 102, 1–15. />tifs.2020.05.020
Zhang, M., Sun, H., Liu, Y., Wang, Y., Piao, C., Cai, D., Wang, Y., & Liu, J. (2021). Effect
of pullulan concentration and pH on the interactions between whey protein

concentrate and pullulan during gelation. Journal of the Science of Food and
Agriculture, 101(2), 659–665. />Zhao, Y., Chen, Z., & Wu, T. (2018). Cryogelation of alginate improved the freeze-thaw
stability of oil-in-water emulsions. Carbohydrate Polymers, 198, 26–33. https://doi.
org/10.1016/j.carbpol.2018.06.013
Zou, F., & Budtova, T. (2020). Tailoring the morphology and properties of starch aerogels
and cryogels via starch source and process parameter. Carbohydrate Polymers,
117344. />Zudaire, E., Gambardella, L., Kurcz, C., & Vermeren, S. (2011). A computational tool for
quantitative analysis of vascular networks. PLoS ONE, 6(11), Article e27385. https://
doi.org/10.1371/journal.pone.0027385

12