Carbohydrate Polymers 154 (2016) 8695

Contents lists available at ScienceDirect

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

Polyelectrolyte complexes via desalting mixtures of hyaluronic acid

and chitosanPhysicochemical study and structural analysis
G. Laleve a,b , G. Sudre a , A. Montembault a , J. Meadows b , S. Malaise b , A. Crpet a ,
L. David a , T. Delair a,
a
b

Univ. Lyon, Universit Claude Bernard Lyon 1, CNRS, IMP, UMR 5223, 15 Boulevard Latarjet, F-69622, Villeurbanne, France
Laboratoire ObvieLine, Sinclair IS Pharma, 8 Chemin du Jubin, F-69570, Dardilly, France

a r t i c l e

i n f o

Article history:
Received 24 March 2016
Received in revised form 20 July 2016
Accepted 3 August 2016
Available online 3 August 2016
Keywords:
Chitosan
Hyaluronic acid
Polyelectrolyte complexes
Coacervate
Hydrogel
Nanostructure

a b s t r a c t
Polyelectrolyte complexes (PECs) were prepared from Chitosan (CS) and Hyaluronic Acid (HYA) homogeneous mixtures of aqueous solutions. The method consisted of preparing a homogeneous mixture of the
two polysaccharides via charge screening at high salt concentrations. Then, the mixture was dialyzed,
leading to the controlled self-assembly of the two polyelectrolytes. Critical parameters like the chitosan
degree of acetylation (DA) and molar mass (Mw), the residual salt concentration and the molar charge
ratio r = nNH3 + (CS)/nCOO (HYA) accounted for the transition from homogeneous aqueous solutions to
colloidal suspensions (r = 0.1) or gel coacervates (r = 0.5). The inuence of the DA and Mw of CS was evaluated by visual observations, light scattering and rheological measurements. For low values of r, Small
Angle X-ray Scattering (SAXS) experiments revealed that the HYA nanostructure was weakly affected
by the presence of PECs. On the contrary, the structure was impacted when increasing r, revealing a
heterogeneous aggregate morphology with ladder-like chain interactions.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Polysaccharide biomaterials have a wide potential of applications in health sciences, for drug delivery (Hamman, 2010; Luo &
Wang, 2014) or tissue engineering (Li, Ramay, Hauch, Xiao, & Zhang,
2005; Tan, Chu, Payne, & Marra, 2009; Yamane et al., 2005) because
polysaccharides are generally regarded as safe, biocompatible
and exhibit a variety of suitable properties for biomedical applications. Polyelectrolytes are polymers bearing ionizable groups.
These groups dissociate in aqueous solutions leading to charged
polymer chains and counterions, dispersed or nanostructured in
the solution. This specicity of polysaccharide polyelectolytes has
been widely used since they can form complexes with salts (Hori,
Winans, & Irvine, 2009), proteins (Water et al., 2014) or other polyelectrolytes (Coimbra et al., 2011; Delair, 2011; Li et al., 2005; Wu
et al., 2007).
Among them, chitosan (CS) receives a growing attention, as
the only cationic and naturally occurring polysaccharide at acidic
pH (pKa = 6.36.7). It is obtained from the partial deacetylation

Corresponding author.
E-mail address: Thierry.Delair@univ-lyon1.fr (T. Delair).
http://dx.doi.org/10.1016/j.carbpol.2016.08.007
0144-8617/ 2016 Elsevier Ltd. All rights reserved.

of chitin, its parent polymer extracted from the exoskeleton of

crustaceans, endoskeleton of cephalopods or cell walls of fungi.
Chitosan is a copolysaccharide of N-acetyl-d-glucosamine and
d-glucosamine. It is dened by its degree of acetylation (DA) corresponding to the molar fraction of acetylated residues and is
known for its bioactivity, biodegradability and mucoadhesion as
well as a low toxicity. It has been associated to negatively charged
polysaccharides like alginates (Li et al., 2005), carboxymethyl cellulose (Chen & Fan, 2007), hyaluronic acid (Al-Qadi, Alatorre-Meda,
Zaghloul, Taboada, & Remunn-Lpez, 2013; Tan et al., 2009),
dextran sulfate (Drogoz, David, Rochas, Domard, & Delair, 2007),
heparin (Costalat, Alcouffe, David, & Delair, 2015; Costalat, David,
& Delair, 2014), to form a wide variety of biomaterials such as
nanoparticles (Costalat, David et al., 2014; Delair, 2011), scaffolds
(Coimbra et al., 2011), bers (Desorme et al., 2013; Ma et al., 2012),
hydrogels (Martnez-Ruvalcaba, Chornet, & Rodrigue, 2007), etc.
Hyaluronic acid (HYA) is an anionic polysaccharide at pH values above its pKA = 2.9. Its repeat disaccharide unit is composed
of d-glucuronic acid and N-acetyl-d-glucosamine, linked by -1,4
and -1,3 glycosidic bonds. HYA is widely present in the extracellular matrix of various tissues of the human body, such as in skin,
cartilage, synovia and vitreous humor. HYA exhibits some interesting features such as anti-inammatory properties (Foschi et al.,

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

1990) and plays an important role in tissue regeneration (Chen

& Abatangelo, 1999). Due to the outstanding properties of each
polysaccharide, biomaterials containing hyaluronic acid and chitosan are actually widely studied. Recently, Kaderli et al. published
a study about the efcacy of HYA-CS hydrogel for osteoarthritis
therapy on rabbit model which provided promising results (Kaderli
et al., 2015).
Polyelectrolyte complexes are generally formed by mixing
aqueous solutions of two oppositely charged polymers and the
entropic gain generated by the liberation of the counterions drives
the association reaction. The main limitation when associating
HYA and CS is the spontaneous association leading to precipitates.
Hence, there is a need to control the complexation to develop materials with high potential for biomedical applications.
For this purpose, we investigated a controlled complexation
between HYA and CS by slowing down the process of association. In particular, we used the dialysis method developed in our
team to induce and control the self-assembly of the two polyelectrolytes (Costalat, Alcouffe, David, & Delair, 2014; Costalat et al.,
2015; Costalat, David, et al., 2014). In the case of polysaccharides,
additional interchain interactions like hydrophobic interactions
and/or H-bonding may play a role. Complexation depends on various parameters either intrinsic, like the polymer charge density
(Dautzenberg, 1997; Webster, Huglin, & Robb, 1997) and chain stiffness (Narambuena, Leiva, Chvez-Pez, & Prez, 2010), or extrinsic
such as the molar charge ratio (Mrquez-Beltrn et al., 2012),
pH (Kayitmazer, Koksal, & Iyilik, 2015), temperature (Chollakup,
Smitthipong, Eisenbach, & Tirrell, 2010), ionic strength (Webster
et al., 1997) and the polymer concentration (Schatz, Domard,
Viton, Pichot, & Delair, 2004). Macroscopically, polyelectrolyte
complexes can be divided into three categories: soluble complexes
(Dautzenberg, 1997; Kabanov & Zezin, 1984), colloidal suspensions (Delair, 2011; Schatz et al., 2004) and biphasic systems
ranging from precipitates (Chollakup et al., 2010) to coacervates
(de Kruif, Weinbreck, & de Vries, 2004; Water et al., 2014) composed of a complex-rich phase and a polymer-poor phase. Recently,
Kayitmazer et al. (2015) published an in-depth study of the complex coacervation between hyaluronic acid and chitosan, in the case
of rather diluted systems and HYA chain lengths with Mw between
50 kg mol1 and 750 kg mol1 .
Here, we show that the intrinsic parameters of chitosan like
the DA and the Mw, as well as the residual salt concentration led
to the formation of materials of different nanostructures. These
nanostructures were investigated by Static Light Scattering and
Small Angle X-ray Scattering (SLS and SAXS) and they were compared with results obtained in our team for chitosan-heparin and
chitosan-dextran sulfate systems.

87

Table 1
Macromolecular characterization of homologous series of chitosans. The degree of
acetylation (DA) was obtained by 1 H NMR; The molar mass (Mw) and dispersity ()
by SEC-MALLS.
DA (%)

Mw (kg mol1 )

56 3
85 5
150 8
55 3
75 4
160 8
28 1
100 5
160 8
19 1
107 5
170 9
17 1
120 6
185 9

1.5 0.2
1.7 0.2
1.9 0.2
1.5 0.2
1.6 0.2
1.6 0.2
1.4 0.1
1.7 0.2
1.9 0.2
1.3 0.1
1.6 0.2
1.9 0.2
1.3 0.1
1.2 0.1
1.6 0.2

14

28

49

solution. Repeated washings with deionized water were necessary

to remove the excess of ammonia and recover neutral pH. Then,
chitosan was freeze-dried.
The puried chitosan was N-reacetylated in controlled conditions in order to obtain a homogeneous series of samples of
different DAs. Chitosan was dissolved in a water/1,2-propanediol
mixture (1/1 (v/v)). Then, acetic anhydride was dissolved in
propanediol and added dropwise. The propanediol minimized the
hydrolysis of acetic anhydride. These homogenous conditions allow
a statistical distribution of N-acetyl residues within the chains with
no O-acetylation (Vachoud, Zydowicz, & Domard, 1997). The Nreacetylated chitosan was isolated by precipitation with aqueous
ammonia. Washings were achieved by centrifugation and/or dialysis with deionized water.
The N-reacetylated chitosans were nally depolymerized in
controlled conditions by nitrous deamination to produce chitosans
with lower molar masses. Chitosan was dissolved at 0.5% (w/v) in
a 0.2 M acetic acid/0.15 M sodium acetate buffer. A 1 g L1 sodium
nitrite aqueous solution was added to chitosan solutions to obtain
a nitrite/glucosamine units molar ratio of 0.1 (for DAs from 1% to
15%), and 0.2 (for DAs from 30% to 50%) to optimize the nitrous
deamination reaction kinetics. The reaction was performed under
high mechanical stirring for different times in order to reach the
desired molar mass. Low molar mass chitosans were recovered by
precipitation with aqueous ammonia, then washed repeatedly with
deionized water until neutrality and nally freeze-dried.

2.2. Methods
2. Experimental
2.1. Materials
Sodium hyaluronate produced by fermentation of Streptococcus Equi was purchased from HTL-biotechnology (France) with a
molar mass Mw 1000 kg mol1 measured by SEC-MALLS. Chitosan obtained from shrimp shell chitin with a medium molar
mass and low Degree of Acetylation was purchased from Mahtani
chitosan Pvt. Ltd. India (batch type 243, DA 1% measured by 1 H
NMR, Mw = 150 kg mol1 and dispersity = Mw/Mn = 1.9 measured
by SEC-MALLS). The chitosan raw material was further puried
by dissolution at 0.5% (w/v) in diluted acetic acid aqueous solution and by ltering the resulting solution on successive cellulose
Millipore membranes with decreasing porosity ranging from 3 m
to 0.22 m, allowing the elimination of all insolubles. The puried chitosan was precipitated with a 37% ammonium hydroxide

2.2.1. Chitosan characterization

The degree of acetylation was calculated by the Hirai et al.
method by 1 H NMR spectroscopy on a Brucker Avance III 400 MHz
5 mm at 300 K (Hirai, Odani, & Nakajima, 1991).
The weight average molar mass Mw and the dispersity () were
measured by size exclusion chromatography (2500 and 6000 PW
TSK gel columns from Tosohaas) coupled online with a differential
refractometer (Wyatt Optilab T-rEx) and a multi angle laser light
scattering detector (Wyatt Dawn EOS) operating at  = 633 nm. A
degassed 0.2 M acetic acid/0.15 M ammonium acetate buffer with
a pH 4.5 was used as the eluent. The ow rate was maintained
at 0.5 mL/min. The refractive index increments (dn/dc) were indicated independently for each degree of acetylation according to a
previous study (Schatz, Viton, Delair, Pichot, & Domard, 2003). The
physicochemical parameters of the chitosan samples are reported
in Table 1.

88

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

The residual water contents of HYA and CS were determined by

thermogravimetric analysis (TA Instrument TGA Q500).
2.2.2. Preparation of polyelectrolyte solutions and homogeneous
mixtures by charge screening method
The hyaluronic acid and chitosan solutions were prepared separately before mixing. Sodium hyaluronate was directly dissolved
in a 0.65 M sodium chloride solution to obtain a 4.5% (w/v) HYA
solution. Mechanical stirring at room temperature was performed
overnight to obtain a homogeneous and transparent viscous solution. Chitosans were solubilized in a 0.65 M sodium chloride
solution, acetic acid was added to achieve the stoichiometric protonation of the NH2 sites in glucosamine residues. CS concentration
depended on the expected charge ratio r = nNH3 + /nCOO (molar mixing ratios between CS positive charges and HYA negative charges)
while HYA concentration in the nal mixture was maintained at
3% (w/v). Mechanical stirring at room temperature was performed
overnight.
CS solutions were then added to HYA solution respecting a
1:2 vol mixing ratio and mechanical stirring was maintained for
at least 2 h. The selected sodium chloride concentration induced
a charge screening for all the DAs, sufcient to prevent the
spontaneous association of the two polyelectrolytes. The nal
homogeneous mixture, containing HYA, CS and NaCl had a pH
around 5.
The resulting mixtures were nally placed into Spectra/Por
1 membranes with a molar mass cut-off MWCO = 3500 g mol1
purchased from Spectrumlabs. They were then dialyzed against
deionized water (V 100 times the volumes of the polymer solution), containing or not 0.15 M sodium chloride, for a minimum of
6 h under gentle mechanical stirring (the bath was changed every
2 h). This dialysis procedure was dened according to Costalat et al.
and allowed the system to reach the desired salt concentration
(Costalat, David et al., 2014). The progressive elimination of sodium
chloride induced the self-assembly of the two polysaccharides.
2.2.3. Measurements of colloid diameters
PECs colloidal suspensions were diluted about 500 folds in
Versol sterile water and characterized by dynamic light scattering using a Malvern Nanosizer SZ equipped with a 5 mW He/Ne
laser beam ( = 633 nm) operating at 25 C with a scattering angle
of 173 . The self-correlation function was expanded in a power
series (Cumulants method). 3 samples of the same formulation
were tested and measurements were repeated at least 10 times.
2.2.4. Rheological measurements
The viscosity of the systems before and after dialysis was measured using an AR2000 rheometer from TA instruments. We used
a 25 mm cone-plate geometry with an angle of 4 and a truncation of 116 m. Alternatively, a Couette device was used for low
viscosity materials. The temperature was xed at 298 K thanks
to a Peltier plate and the shear rate varied between 103 s1 and
100 s1 . The viscosity 0 was then determined from the Newtonian plateau present at low shear rate. Viscoelastic characterization
of coacervates was performed by frequency sweep measurements
using a 25 mm steel plate geometry equipped with a solvent trap
to avoid the drying of the samples. The strain was xed at 1% and
the angular frequency varied from 100 to 0.1 rad/s.
2.2.5. Small angle X-ray scattering
SAXS experiments were performed at the European Synchrotron
Radiation Facility (Grenoble, France) on the BM2-D2AM beamline.
First, SAXS analysis was performed on HYA and CS solutions.
The roles of the salt and polymer concentrations were investigated.
The HYA and CS samples were placed and analyzed in silica tubes

(external diameter 3 mm, wall thick-ness 0.2 mm, 76 mm long,

from Deutero GmbH) with rubber caps to avoid water evaporation.
For the PECs, according to their physical forms, either solidlike gels or colloids, they were placed either in a home-made
sample holder of 2 mm thickness trapped in Kapton tape or silica tubes respectively. The incident photon energy was set to
17 keV. We used a 2D CCD X-ray detector from Ropper scientic.
The sample-to-detector distance was about 1.15 m and the beam
stop had a diameter of 3 mm. The q-calibration (q = 4 sin()/;
2 = scattering angle) was realized thanks to a silver behenate powder standard. The scattering contribution of the empty cell and of
the solvent (glass tube lled with deionized water) was subtracted
from the scattering intensity of the samples. The image data treatments took into account the distortion of the camera tapper, the
at eld response and the dark image. The intensity was further
normalized by the incident ux and the sample thickness. The scattering proles of the intensity I as a function of the scattering vector
q are obtained by azimuthally averaging the corrected images.

3. Results and discussion

3.1. Study of HYA CS complexation
Since polyelectrolyte complexation is entropically driven, the
direct mixing of HYA and CS solutions in water led to a fast association and appearance of white precipitates. This observation
was true for every chitosan precursor solution, irrespective of the
degree of acetylation and molar mass. However, the whitening
effect was less pronounced when the DA was high and the Mw
was low.
A method for the control of the complexation of two polyelectrolytes was recently developed in our team (Costalat, David et al.,
2014). This was achieved by adding sodium chloride to the polyelectrolyte solutions in order to screen the electrostatic charges of
both polyelectrolytes. This method was also investigated by Kaderli
et al. for the development of HYA-CS hydrogels for osteoarthrosis
therapy where a minimum salt concentration was added to avoid
precipitation (Kaderli et al., 2015).
We prepared a series of samples combining HYA with each chitosan sample by mixing the two solutions in the presence of salts.
The investigated parameters were the chitosan DA, the molar mass
and the charge ratio r. DA = 1% and DA = 5% were eliminated from
the study, because they could not be dissolved in NaCl solutions
irrespective of the Mw. Thus, we focused on chitosans with DAs
higher than 10%. A visual assessment was performed on the different solutions and mixtures before and after the dialysis process in
order to screen a variety of experimental conditions. So we could
select the appropriate polymer and salt concentrations to avoid the
CS precipitation and the spontaneous precipitation due to uncontrolled complexation between the two polysaccharides prior the
dialysis step. After dialysis, we could observe the inuence of the
charge ratio, the salt concentration in the dialysis bath as well as
the degree of acetylation and the molar mass of chitosan on the
formed polyelectrolyte complexes.

3.1.1. Polyelectrolyte complexes obtained at low charge ratio

(r = 0.1)
3.1.1.1. In the absence of salt in the dialysis bath, i.e. dialysis against
water, the polyelectrolyte assembly led to insoluble PECs dispersed
in the aqueous phase. For each DA, the complexation led to turbid
and viscous dispersions i.e. dispersions with a high light scattering
(Fig. 1a [1][6]) whatever the molar mass of chitosan. However,
we could notice a slight decrease of the light scattering with higher
DAs and lower molar masses.

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

89

Fig. 1. Visual observations ([a] and [b]) and average diameter by DLS ([c] and [d]) of colloidal suspensions containing HYA at 3% (w/v) and CS at ratio r = 0.1 and [NaCl] in the
dialysis bath = 0 mol L1 (left) or [NaCl] in the dialysis bath = 0.15 mol L1 (right). CS characteristics are Mw = 160 kg mol1 and DA = 14% [1], Mw = 28 kg mol1 and DA = 14%
[2], Mw = 170 kg mol1 and DA = 28% [3], Mw = 19 kg mol1 and DA = 28% [4], Mw = 185 kg mol1 and DA = 48% [5], Mw = 17 kg mol1 and DA = 49% [6].

The average diameters of the colloids were highly impacted by

the DA of chitosan and ranging between 300 nm and 900 nm. For
DA = 14%, large objects were obtained with a low polydispersity
whatever the molar mass. For DA = 28% and DA = 49%, the colloids
were smaller and the polydispersity higher (Fig. 1c). These differences may be attributed to the charge density of CS. At low
DA, a high charge density on the chitosan macromolecule allowed
interactions with many HYA molecules, leading to larger and stable objects. Conversely, at high DAs, i.e. low charge densities,
the chitosan macromolecules interacted with few HYA molecules,
resulting in small objects of varying sizes. Interestingly, the molar
mass of chitosan had no major effect on the nal particle size, which
is consistent with preceding works that showed the molar mass
of the polymer in default had a limited inuence on particle size
(Schatz et al., 2004).

ation would occur, leading to loose and hydrated chain aggregates

responsible for the observed high colloidal sizes.
In conclusion, in the absence of salt in the dialysis bath, insoluble
complexes were formed whatever the molar mass and DA of chitosan (Fig. 2a). In the presence of salts in the dialysis baths, the
partial desalting of the homogenous mixture of polysaccharides
yielded various polyelectrolyte complexes either precipitated or
water soluble PECs depending on the DA and molar mass of chitosan (Fig. 2b). As a general trend, lower molar masses associated to
higher DAs led to water soluble complexes as described by Kabanov
et al. for synthetic polymers (Kabanov & Zezin, 1984). The formation
of these soluble complexes required conformational adaptation of
the macromolecules and/or rearrangements within the complexes
that are only possible in the presence of salt.

3.1.1.2. In the presence of salt, i.e. 0.15 M sodium chloride during

dialysis. For DA = 14%, the results were similar to those obtained
without salt. The PECs dispersions were turbid and viscous, whatever the Mw of chitosan (Mw = 28 kg mol1 and Mw = 170 kg mol1
in Fig. 1b [1] and [2] respectively).
For DA = 28% and Mw = 170 kg mol1 , the complexation led also
to a turbid and viscous system (Fig. 1b [3]). When Mw was
decreased to 18 kg mol1 , the system was not turbid anymore but
opalescent (Fig. 1b [4]), i.e. with a lower light scattering intensity.
For DA = 49% and Mw = 185 kg mol1 the system was opalescent (Fig. 1b [5]) while for Mw = 17 kg mol1 , soluble PECs were
obtained and the solution was totally transparent (Fig. 1b [6]).
The presence of 0.15 M NaCl as residual salt concentration after
dialysis led to larger diameters and a bi-populated particle dispersion (Fig. 1d). The highest polydispersity was observed for the
opalescent and transparent systems. The bi-populated dispersions
might originate from an alteration of the colloidal stability of the
smaller particles that agglomerated into larger ones, as a result of
the decrease in intensity of the stabilizing electrostatic forces in the
presence of salt. Alternatively, we could also consider that, in the
presence of the screening salt, a lower polycation/polyanion associ-

3.1.2. Polyelectrolyte complexes obtained at higher charge ratios

(r = 0.5)
For charge ratios r = 0.5, we could not clearly point out the
inuence of the salt concentration since at the end of the dialysis
process, either coacervates or turbid solutions were already formed
in both the absence and presence of salt in the dialysis bath.
The association of HYA and CS at r = 0.5 led to two types of
materials. For the higher molar masses, we observed the formation of a biphasic system with a dense white gel phase, also called
coacervate and a transparent supernatant, irrespective of DA (see
Fig. 3a). Coacervation resulted from a syneresis phenomenon. During the dialysis process, associations were progressively formed via
electrostatic attractive interactions, as a result of the de-salting process. As the complexes formed, additional short-range interactions
could develop such as hydrogen bonding or hydrophobic interactions. Water molecules were expulsed out of the complex structure
leading to the contraction of the system. The supernatant is a liquid phase of low viscosity containing mainly water and a small
quantity of polyelectrolyte (the concentration of polymer in the
supernatant was measured by weight residue after freeze-drying
and found lower than 1% (w/v)). For lower molar masses, turbid sus-

90

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

[a]

[b]

Fig. 2. Behavior of HA-CS associations at r = 0.1 as a function of the DA and the molar mass of CS in absence of salts [a] or in presence of 0.15 M NaCl [b] residual concentration
in the dialysis bath.

Fig. 3. Mixture containing HYA and CS (Mw = 170 kg mol1 ; DA = 14%) before and after the dialysis process against deionized water at charge ratio r = 0.5 (a). Turbid suspensions
obtained after dialysis for chitosans of low molar masses at r = 0.5 (b).

pensions were obtained, except for DA = 48%, for which a solution

containing opalescent gel-like aggregates was obtained (Fig. 3b).

(compare Fig. 4a and b). This result is also consistent with our visual
observations showing that the turbidity of the complex dispersion
was always higher for the higher molar masses of the polycation.

3.2. Rheological measurements

The objective of this section is to correlate the visual observations with the rheological properties of the solutions and gels.
For this purpose, we investigated the impact of the charge ratio,
the DA and the Mw of chitosan on the Newtonian viscosity or the
viscoelastic shear moduli (G and G ).
3.2.1. At low charge ratio (r = 0.1)
As seen in Fig. 4a and b, the viscosities of the blends containing
HYA and CS before the dialysis process were mainly determined by
the HYA component, and were independent of the DA of chitosan
and slightly impacted by CS molar mass. Indeed, the Newtonian
viscosities 0 of the blends ranged between 800 Pa s and 900 Pa s for
chitosans of Mw about 170185 kg mol1 , and between 650 Pa s and
700 Pa s for chitosans of Mw around 1728 kg mol1 , irrespective of
DA.
As a general trend, the formation of the PECs led to a decrease
in the viscosities of the solutions. As expected, this decrease was
stronger when the dialysis was carried out against water than
against 0.15 mol L1 sodium chloride. Indeed, as seen in Section
3.1, in the absence of salt, the complexes always precipitated out,
whereas in the presence of salt, more material remained in solution. Finally, the DA of chitosan also impacted the variations in
viscosities: the lower the DA, the higher the loss of viscosity due to
increased complexation. This is consistent with the strong solubility of reacetylated chitosans at a DA > 40% in acidic sodium chloride
aqueous solutions (Costalat, David et al., 2014). The molar mass of
chitosan also impacted the extent of the viscosity loss upon complexation: the higher molar mass induced the larger viscosity loss

3.2.2. At higher charge ratio (r = 0.5)

Again, the viscosities of the blends before the dialysis process
were not impacted by the DA of chitosan but by its Mw, similarly to
the low charge ratio systems. Newtonian viscosities were around
1000 Pa s and around 750 Pa s for respectively the highest and lowest molar masses of CS (Fig. 4c), when the viscosity of a HYA solution
in the same concentration (i.e 3% w/v) without chitosan was around
250 Pa s. After dialysis, for the low molar masses, as the viscosities
were too low to be measured with a cone-plate system, we measured the viscosity of the turbid suspensions using a Couette device.
For DAs 14% and 28%, the decrease in viscosity was drastic and the
Newtonian viscosity 0 was found about 1.8 Pa s and 4 Pa s respectively (Fig. 4d). This was due to the presence of a large quantity of
polyelectrolyte complexes, colloids or aggregates, with no soluble
residual polyelectrolytes, reducing the viscosity of the system. For
DA = 49%, there was no Newtonian plateau and 0 was assimilated
to the viscosity at 0.1 s1 . Similarly to the charge ratio r = 0.1, we
pointed out that viscosity loss was reduced with an increase in the
DA of chitosan.
The coacervates, obtained with the high molar masses were
retrieved by elimination of the supernatant. The storage and loss
moduli of these coacervates were then evaluated from a constantstrain frequency sweep within frequency ranges of 1000.1 rad/s.
A piece of coacervate was cut and the residual supernatant was
removed with absorbing paper before analysis. G and G values
are in that case only qualitative because they highly depended on
the percentage of syneresis which we were not able to control from
an experiment to another. G and G were found similar for DA = 15%
and DA = 30% and the trend for DA = 49% was an increase of the mod-

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

[a]

[b]

[c]

[d]

91

Fig. 4. Comparison of the viscosity loss for high Mw ([a]) and low Mw ([b]) depending on the DA and the sodium chloride concentration for PECs with r = 0.1. Diagrams
representing the Newtonian viscosity of the blends before the dialysis process ([c]) and after dialysis ([d]) as a function of the CS molar mass and DA at charge ratio r = 0.5.
High molar masses yielded coacervates hydrogels.

3.3. SAXS analysis of HYA and CS solutions

3.3.1. Structure of HA in solution
In a rst step, we studied the conformation of the chains of HYA
in solution. Two important parameters were the polymer concentration and the residual salt concentration in the solution. Since
HYA is the excess polymer, knowing the structure prior the dialysis process was important for the analysis of the SAXS data we
obtained for PECs.

Fig. 5. Curves representing the storage modulus (G ) and the loss modulus (G )
moduli as a function of the angular frequency . Symbols , and for G , G and
tan for DA = 14% and Mw = 170 kg mol1 respectively; Symbols ,  and for G ,
G and tan for DA = 28% and Mw = 175 kg mol1 respectively; Symbols ,  and
for G , G and tan for DA = 49% and Mw = 185 kg mol1 .

uli. The most important result was that every coacervates exhibited
a gel viscoelastic behavior dominated by elasticity with values of
tan = G /G comprised between 0.2 and 0.4 (Fig. 5).

3.3.1.1. Inuence of the concentration. We prepared HYA solutions

at concentrations varying between 0.3% and 4%. The corresponding
SAXS proles I(q) are presented in Fig. 6a for a q-range extending
from 0.01 to 0.32 1 .
For each concentration, we observed the so-called polyelectrolyte peak, characteristic of a periodic organization of the
polyelectrolyte chains in solution, resulting from the repulsive electrostatic interactions between the polymer chains (Essa, Spiteri,
Williams, & Boue, 2009). The evolution of the location of this
peak with the polymer concentration is related to the hydrophobicity/hydrophilicity of the polyelectrolyte in aqueous solutions
(Popa-Nita, Rochas, David, & Domard, 2009). On increasing the HYA
concentration, the SAXS diagrams revealed that the polyelectrolyte
peak was shifted to higher q values, corresponding to smaller interchains distances (or correlation distances). At high q values, we
used a law I(q) = C/q giving coefcients (Table 2): slightly
increased with the concentration until 3% with values close to

92

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

Fig. 6. SAXS analysis of HYA in solution at various concentrations in% (w/v) [a]. Evolution of the position of the polyelectrolyte peak qmax as a function of HYA concentration
Cp [b]. Drawings are adapted from Ref. Dobrynin and Rubinstein (1999). (The molar mass of the averaged residue of Sodium Hyaluronate is taken to be 194 g mol1 ).

Table 2
Values of coefcients determined from the slope at high q values depending on
the HYA concentration.
HA% (w/v)

0.3%

0.5%

1%

2%

3%

4%

1.20

1.31

1.32

1.35

1.37

1.84

1.21.4 reecting a rod-like conformation probably coming from

the helical structure of HYA in solution (Heatley & Scott, 1988).
underwent an abrupt change at 4% corresponding to a conformational change and the apparition of a more aggregated structure.
We performed a tting on the different curves using a lorentzian
function, which led to the position qmax of the maximum intensity
of the peak. Then, the position qmax was plotted as a function of the
HYA concentration (Fig. 6b) and compared with previous results of
Popa-Nita et al. (2009).
For polyelectrolytes in solution, the scaling model proposed by
Dobrynin and Rubinstein (1999) consisted in considering the polyelectrolyte chains conformation as beaded necklaces. Depending
on the polymer concentration, it predicts a transition from a stringcontrolled conformation (where the correlation length is larger
than the string length; qmax Cp 1/2 ) to a beaded necklace model
where the chain conformation is governed by the bead size (where
is of the order of the distance between the beads; qmax Cp 1/3 ).
In our system, i.e. for HYA salt-free solutions, the concentration
corresponding to this transition was found about 0.06 mol L1 , in
agreement with previous results, and showing that hyaluronate is
more hydrophilic than the CS of same acetylation degree (i.e. 50%).
In the following complexation study, the HYA concentration was
3% (w/v) (i.e. a residue molar concentration of Cp = 0.15 mol L1 )
so the conformation of HYA chains was in the beads-controlled
regime. The corresponding correlation length , i.e. the distance
between the beads can be determined with equation  = 2/q and
was found close to 76 in absence of added salts.
3.3.1.2. Inuence of the salt concentration. As we prepare the HYA
and CS mixture in presence of salts, we studied the evolution of the
HYA chain conformation in the presence of salts.
We prepared HYA solutions at 3% (w/v) with various amounts
of sodium chloride from 0 to 0.65 mol L1 . The synchrotron SAXS
diagrams (Fig. 7) revealed an outstanding reproducibility. In the
salt-free solution, we found again the well-dened polyelectrolyte
peak. As the salt concentration was increased, the scattered intensity increased and the polyelectrolyte peak disappeared for sodium
chloride concentration higher than 0.1 mol L1 . These diagrams also

Fig. 7. Inuence of the salt concentration on the chains conformation of HYA in

solution at 3% (w/v).

Table 3
Values of Gyration radius determined with a Guinier law at low q values >13Rg
depending on the salt concentration.
HA% (w/v)

Concentration on NaCl (mol L1 )

Rg ()

3
3
3

0.100
0.150
0.650

20
26
38

revealed the presence of a pseudo-isosbestic point at q 0.083 1 ,

reecting the existence of mainly two possible structural states.
The rst state corresponds to the classical polyelectrolyte conformation of a salt-free polyelectrolyte solution, characterized by
the presence of the polyelectrolyte peak. The other state was
observed in the presence of high concentration of salts. The scattering intensity increased in the low q region, showing the presence
of bigger objects which was attributed to the presence of aggregates at the nanoscale due to a more hydrophobic behavior. The
gyration radii (Rg ) were determined using a Guinier law I(q) = I0
exp (q2 Rg 2 /3) and results are reported in Table 3. The results
showed an evolution to higher values of Rg when the salt concentration was increased. We also noted that for a salt concentration
of 0.65 mol L1 , a Rg value of 38 corresponding to the half of
the correlation length at [HYA] = 3% was found. Starting from a

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

[a]

93

[b]

[c]

Fig. 8. Inuence of the DA for PECs containing HYA at 3% (w/v) and CS at a constant charge ratio r = 0.1 ([a] and [b]) and r = 0.5 ([c]) and comparison to a salt-free HYA
solution. In [a] and [b], the CS concentration (w/v) was 0.16%, 0.21% and 0.30% for DAs 14%, 28% and 49% respectively. In [c], two curves showing the differences observed for
coacervates at DA = 14% from two different batches. The CS concentration (w/v) was 0.82%.

beaded-controlled regime, nanoscale aggregates are growing with

increasing the salt concentration until entering in contact. Since
[HYA] and [NaCl] were maintained at 3% (w/v) and 0.65 mol L1
respectively in the mixtures before dialysis, this conformational
state corresponded to the starting point of the complexation for
HYA in our study.

3.3.2. Study of the structure of PECs

The HYA concentration was maintained at 3% (w/v) for the
whole study. The nanostructure of a series of PECs was then studied
varying the charge ratio r. The CS concentration was varied depending on the expected ratio and the DA. This part of the work focuses
on the comparison of the PECs nanostructure at two charge ratios
(r = 0.1 and r = 0.5).

3.3.2.1. Charge ratio r = 0.1. Polyelectrolyte complexes were

formed by desalting a homogeneous mixture of HYA and CS
with a charge ratio r = 0.1. The starting salt concentration in the
HYA and CS solutions was 0.65 mol L1 and desalting was then
achieved by dialysis against deionized water. The resulting PECs
(opalescent/turbid systems) were analyzed by SAXS. Here, we

compared the structure of PECs with different DAs. The diagram of

a 3% (w/v) salt-free HYA solution was added for comparison.
Whatever the molar mass of the chitosan, the SAXS diagrams
revealed the existence of the polyelectrolyte peak (Fig. 8a and b).
The position of this peak was almost independent on the variation of the DA of CS. As the CS concentration was low compared
to the HYA concentration, we can assume that every CS macromolecule was complexed to HYA. Thus, the polyelectrolyte peak can
be attributed to uncomplexed or partially-complexed HYA chains.
Compared to the 3% HYA solution, qmax was slightly shifted to lower
q values, i.e. to higher values of the correlation length . Such evolution of the correlation length could be due to an increase in the
persistence length by polyelectrolyte association and/or by dilution
effect of HYA. The position qmax of the PECs was found at 0.071 1
for every DA and corresponds to that obtained with a 2% (w/v)
salt-free HYA solution.
At high q values, a law I(q) = C/q gave = 1.32 whatever the
DA. This behavior was similar to HYA solutions and corresponds to
the HYA nanostructure. At low q values, an increase of the upturn
of scattered intensity was observed and can be interpreted as a
signature of aggregation with decreasing DA. These observations
at the nanoscale were consistent with the macroscopic observa-

94

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

tions, which showed a decrease in the light scattering contrast with

increasing DA.

3.3.2.2. Charge ratio r = 0.5. Polyelectrolyte complexes were also

formed by completely desalting a homogeneous mixture of HYA
and CS at a charge ratio r = 0.5. The starting salt concentration was
0.65 mol L1 and desalting was achieved by dialysis against deionized water. The resulting PECs (coacervates) obtained for the higher
molar masses were also analyzed by SAXS. On Fig. 8c, different
scattering diagrams are obtained for samples prepared in similar
conditions (pH, ionic strength). The discrepancy of these results
illustrates the strong sensibility of the structure of the PECs towards
the experimental preparation parameters.
However, the SAXS diagrams did not exhibit the existence of
the classical polyelectrolyte peak. HYA repulsive interactions were
highly reduced due to the polyelectrolyte associations obtained at
high charge ratio. As a consequence, the neutralized hyaluronan
chains no longer exhibited a patterned polyelectrolyte organization
anymore. The pattern could be divided in three scattering domains.
At high q values, the slope was found close to q1 , which is similar
to the behavior already described by our team for chitosan-heparin
and chitosan-dextran sulfate complexes and can be attributed to
a stable morphology constituted of large rod-like objects corresponding to ladder-like chain associations (Costalat et al., 2015).
At intermediate q values, the slope was comprised between q0.5
and q0.8 ; the absence of plateau probably originates from the contribution of larger objects which contribution is characterized by
an increase of the slope in the low q region.

4. Conclusions
Polyelectrolyte complexation with a slow kinetic process using
the desalting method was performed to control the hyaluronan/chitosan assembly. Sodium chloride was used to screen the
electrostatic charges of the polyions, at a suitable concentration
to avoid CS precipitation. The assembly via attractive electrostatic
interactions was restored by dialysis to reduce the salt concentration. We investigated the parameters impacting the assembly
process. We showed a direct correlation between the visual observations, to be better quantied with light scattering and the
rheological properties of the PECs. The nanostructure was also
investigated by Small Angle X-Ray Scattering.
Mainly colloidal suspensions resulted from HYA/CS assembly
at low charge ratio r, whereas coacervates were obtained when
increasing r. The properties of the nal materials were highly
dependent on the chitosan molar mass and degree of acetylation,
as well as on the residual salt concentration.
Prior to the dialysis, aggregated systems were already formed
due to the high HYA concentration which impacted the nanostructure of HYA in solution. On adding a low quantity of CS,
the SAXS diagrams revealed that the conformation of HYA chains
remained mainly unchanged. The PECs nanostructure was close to
the nanostructure of an HYA solution. This was also demonstrated
by rheological measurement since the PECs have a typical viscous
behavior.
When the charge ratio was increased, the PECs were either colloidal suspensions or coacervates, depending on CS molar mass. In
both cases, the PEC formation led to very aggregated systems, also
revealed by SAXS analysis. Because both hyaluronic acid and chitosan exhibit interesting biological properties, the study and the
understanding of such associations is one of the main goals in the
polyelectrolyte domain. The controlled production of either colloidal suspensions or coacervates could break open developments
in the eld of biomaterials.

Acknowledgments
The authors thank the Laboratoire ObvieLine a Sinclair Pharma
company and the ANRT Association nationale de la recherche et de
la technologie through the Cifre program for the nancial support.
They also want to thank the CRG group at ESRF (France), in particular Cyrille Rochas (CERMAV, CNRS France) for his assistance during
SAXS experiments. The authors would like to acknowledge the Centre for the Characterization of Polymers by Liquid Chromatography
of the Institut de Chimie de Lyon for their expertise and assistance
in molar mass determination by SEC measurements.

References
Al-Qadi, S., Alatorre-Meda, M., Zaghloul, E. M., Taboada, P., & Remunn-Lpez, C.
(2013). Chitosan-hyaluronic acid nanoparticles for gene silencing: The role of
hyaluronic acid on the nanoparticles formation and activity. Colloids and
Surfaces B: Biointerfaces, 103, 615623. http://dx.doi.org/10.1016/j.colsurfb.
2012.11.009
Chen, W. Y. J., & Abatangelo, G. (1999). Functions of hyaluronan in wound repair.
Wound Repair and Regeneration, 7(2), 7989. http://dx.doi.org/10.1046/j.1524475X.1999.00079.x
Chen, H., & Fan, M. (2007). Chitosan/carboxymethyl cellulose polyelectrolyte
complex scaffolds for pulp cells regeneration. Journal of Bioactive and
Compatible Polymers, 22(5), 475491. http://dx.doi.org/10.1177/
0883911507081329
Chollakup, R., Smitthipong, W., Eisenbach, C. D., & Tirrell, M. (2010). Phase behavior
and coacervation of aqueous poly(acrylic acid)-poly(allylamine) solutions.
Macromolecules, 43(5), 25182528. http://dx.doi.org/10.1021/ma902144k
Coimbra, P., Alves, P., Valente, T. A. M., Santos, R., Correia, I. J., & Ferreira, P. (2011).
Sodium hyaluronate/chitosan polyelectrolyte complex scaffolds for dental
pulp regeneration: Synthesis and characterization. International Journal of
Biological Macromolecules, 49(4), 573579. http://dx.doi.org/10.1016/j.
ijbiomac.2011.06.011
Costalat, M., Alcouffe, P., David, L., & Delair, T. (2014). Controlling the complexation
of polysaccharides into multi-functional colloidal assemblies for
nanomedicine. Journal of Colloid and Interface Science, 430, 147156. http://dx.
doi.org/10.1016/j.jcis.2014.05.039
Costalat, M., David, L., & Delair, T. (2014). Reversible controlled assembly of
chitosan and dextran sulfate: A new method for nanoparticle elaboration.
Carbohydrate Polymers, 102, 717726. http://dx.doi.org/10.1016/j.carbpol.
2013.10.098
Costalat, M., Alcouffe, P., David, L., & Delair, T. (2015). Macro-hydrogels versus
nanoparticles by the controlled assembly of polysaccharides. Carbohydrate
Polymers, 134, 541546. http://dx.doi.org/10.1016/j.carbpol.2015.07.071
Dautzenberg, H. (1997). Polyelectrolyte complex formation in highly aggregating
systems. 1. Effect of salt: Polyelectrolyte complex formation in the presence of
NaCl. Macromolecules, 30(25), 78107815.
Delair, T. (2011). Colloidal polyelectrolyte complexes of chitosan and dextran
sulfate towards versatile nanocarriers of bioactive molecules. European Journal
of Pharmaceutics and Biopharmaceutics, 78(1), 1018. http://dx.doi.org/10.
1016/j.ejpb.2010.12.001
Desorme, M., Montembault, A., Lucas, J.-M., Rochas, C., Bouet, T., & David, L. (2013).
Spinning of hydroalcoholic chitosan solutions. Carbohydrate Polymers, 98(1),
5063. http://dx.doi.org/10.1016/j.carbpol.2013.04.070
Dobrynin, A. V., & Rubinstein, M. (1999). Hydrophobic polyelectrolytes.
Macromolecules, 32(3), 915922. http://dx.doi.org/10.1021/ma981412j
Drogoz, A., David, L., Rochas, C., Domard, A., & Delair, T. (2007). Polyelectrolyte
complexes from polysaccharides: Formation and stoichiometry monitoring.
Langmuir, 23(22), 1095010958. http://dx.doi.org/10.1021/la7008545
de Kruif, C. G., Weinbreck, F., & de Vries, R. (2004). Complex coacervation of
proteins and anionic polysaccharides. Current Opinion in Colloid & Interface
Science, 9(5), 340349. http://dx.doi.org/10.1016/j.cocis.2004.09.006
Essa, W., Spiteri, M.-N., Williams, C., & Boue, F. (2009). Hydrophobic
polyelectrolytes in better polar solvent. Structure and chain conformation As
seen by SAXS and SANS. Macromolecules, 42(24), 95689580. http://dx.doi.org/
10.1021/ma9003874
Foschi, D., Castoldi, L., Radaelli, E., Abelli, P., Calderini, G., Rastrelli, A., . . . &
Trabucchi, E. (1990). Hyaluronic acid prevents oxygen free-radical damage to
granulation tissue: A study in rats. International Journal of Tissue Reactions,
12(6), 333339.
Hamman, J. H. (2010). Chitosan based polyelectrolyte complexes as potential
carrier materials in drug delivery systems. Marine Drugs, 8(4), 13051322.
http://dx.doi.org/10.3390/md8041305
Heatley, F., & Scott, J. E. (1988). A water molecule participates in the secondary
structure of hyaluronan. Biochemical Journal, 254(2), 489493.
Hirai, A., Odani, H., & Nakajima, A. (1991). Determination of degree of
deacetylation of chitosan by 1H NMR spectroscopy. Polymer Bulletin, 26(1),
8794. http://dx.doi.org/10.1007/BF00299352
Hori, Y., Winans, A. M., & Irvine, D. J. (2009). Modular injectable matrices based on
alginate solution/microsphere mixtures that gel in situ and co-deliver

G. Laleve et al. / Carbohydrate Polymers 154 (2016) 8695

immunomodulatory factors. Acta Biomaterialia, 5(4), 969982. http://dx.doi.
org/10.1016/j.actbio.2008.11.019
Kabanov, V. A., & Zezin, A. B. (1984). A new class of complex water-soluble
polyelectrolytes. Die Makromolekulare Chemie, 6(S19841), 259276. http://dx.
doi.org/10.1002/macp.1984.020061984120
Kaderli, S., Boulocher, C., Pillet, E., Watrelot-Virieux, D., Rougemont, A. L., Roger, T.,
. . . & Jordan, O. (2015). A novel biocompatible hyaluronic acidchitosan hybrid
hydrogel for osteoarthrosis therapy. International Journal of Pharmaceutics,
483(12), 158168. http://dx.doi.org/10.1016/j.ijpharm.2015.01.052
Kayitmazer, A. B., Koksal, A. F., & Iyilik, E. K. (2015). Complex coacervation of
hyaluronic acid and chitosan: Effects of pH, ionic strength, charge density,
chain length and the charge ratio. Soft Matter, 11(44), 86058612. http://dx.
doi.org/10.1039/C5SM01829C
Li, Z., Ramay, H. R., Hauch, K. D., Xiao, D., & Zhang, M. (2005). Chitosan-alginate
hybrid scaffolds for bone tissue engineering. Biomaterials, 26(18), 39193928.
http://dx.doi.org/10.1016/j.biomaterials.2004.09.062
Luo, Y., & Wang, Q. (2014). Recent development of chitosan-based polyelectrolyte
complexes with natural polysaccharides for drug delivery. International Journal
of Biological Macromolecules, 64, 353367. http://dx.doi.org/10.1016/j.ijbiomac.
2013.12.017

Mrquez-Beltrn, C., Castaneda,

L., Enciso-Aguilar, M., Paredes-Quijada, G.,

H., Maldonado-Arce, A., & Argillier, J.-F. (2012). Structure and

Acuna-Campa,
mechanism formation of polyelectrolyte complex obtained from PSS/PAH
system: Effect of molar mixing ratio, base-acid conditions, and ionic strength.
Colloid and Polymer Science, 291(3), 683690. http://dx.doi.org/10.1007/
s00396-012-2775-y
Ma, G., Liu, Y., Fang, D., Chen, J., Peng, C., Fei, X., & Nie, J. (2012). Hyaluronic
acid/chitosan polyelectrolyte complexes nanobers prepared by
electrospinning. Materials Letters, 74, 7880. http://dx.doi.org/10.1016/j.
matlet.2012.01.012
Martnez-Ruvalcaba, A., Chornet, E., & Rodrigue, D. (2007). Viscoelastic properties
of dispersed chitosan/xanthan hydrogels. Carbohydrate Polymers, 67(4),
586595. http://dx.doi.org/10.1016/j.carbpol.2006.06.033
Narambuena, C. F., Leiva, E. P. M., Chvez-Pez, M., & Prez, E. (2010). Effect of
chain stiffness on the morphology of polyelectrolyte complexes. A Monte Carlo

95

simulation study. Polymer, 51(14), 32933302. http://dx.doi.org/10.1016/j.

polymer.2010.04.065
Popa-Nita, S., Rochas, C., David, L., & Domard, A. (2009). Structure of natural
polyelectrolyte solutions: Role of the hydrophilic/hydrophobic interaction
balance. Langmuir, 25(11), 64606468. http://dx.doi.org/10.1021/la900061n
Schatz, C., Viton, C., Delair, T., Pichot, C., & Domard, A. (2003). Typical
physicochemical behaviors of chitosan in aqueous solution. Biomacromolecules,
4(3), 641648. http://dx.doi.org/10.1021/bm025724c
Schatz, C., Domard, A., Viton, C., Pichot, C., & Delair, T. (2004). Versatile and
efcient formation of colloids of biopolymer-based polyelectrolyte complexes.
Biomacromolecules, 5(5), 18821892. http://dx.doi.org/10.1021/bm049786+
Tan, H., Chu, C. R., Payne, K. A., & Marra, K. G. (2009). Injectable in situ forming
biodegradable chitosan-hyaluronic acid based hydrogels for cartilage tissue
engineering. Biomaterials, 30(13), 24992506. http://dx.doi.org/10.1016/j.
biomaterials.2008.12.080
Vachoud, L., Zydowicz, N., & Domard, A. (1997). Formation and characterisation of
a physical chitin gel. Carbohydrate Research, 302(34), 169177. http://dx.doi.
org/10.1016/S0008-6215(97)00126-2
Water, J. J., Schack, M. M., Velazquez-Campoy, A., Maltesen, M. J., van de Weert, M.,
& Jorgensen, L. (2014). Complex coacervates of hyaluronic acid and lysozyme:
Effect on protein structure and physical stability. European Journal of
Pharmaceutics and Biopharmaceutics, 88(2), 325331. http://dx.doi.org/10.
1016/j.ejpb.2014.09.001
Webster, L., Huglin, M. B., & Robb, I. D. (1997). Complex formation between
polyelectrolytes in dilute aqueous solution. Polymer, 38(6), 13731380. http://
dx.doi.org/10.1016/S0032-3861(96)00650-7
Wu, J., Wang, X., Keum, J. K., Zhou, H., Gelfer, M., Avila-Orta, C.-A., . . . & Chu, B.
(2007). Water soluble complexes of chitosan-g-MPEG and hyaluronic acid.
Journal of Biomedical Materials Research Part A, 80A(4), 800812. http://dx.doi.
org/10.1002/jbm.a.30972
Yamane, S., Iwasaki, N., Majima, T., Funakoshi, T., Masuko, T., Harada, K., . . . &
Nishimura, S. (2005). Feasibility of chitosan-based hyaluronic acid hybrid
biomaterial for a novel scaffold in cartilage tissue engineering. Biomaterials,
26(6), 611619. http://dx.doi.org/10.1016/j.biomaterials.2004.03.013