Serum protein-hyaluronic acid complex nanocarriers: Structural characterisation and encapsulation possibilities

A protein-polysaccharide-based potential nanocarrier system have been developed via a simple, one-step preparation protocol without the use of long-term heating and the utilization of hardly removable crosslinking agents, surfactants, and toxic organic solvents.

Available online 7 September 2020

0144-8617/© 2020 The Authors Published by Elsevier Ltd This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

Serum protein-hyaluronic acid complex nanocarriers: Structural

characterisation and encapsulation possibilities

aDepartment of Physical Chemistry and Materials Science, University of Szeged, H-6720, Rerrich B Square 1, Szeged, Hungary

bMTA-SZTE Biomimetic Systems Research Group, Department of Medical Chemistry, Faculty of Medicine, University of Szeged, H-6720 D´om Square 8, Szeged, Hungary

A R T I C L E I N F O

Keywords:

Hyaluronic acid

Bovine serum albumin

Protein-polymer nanoconjugates

Charge neutralisation

Encapsulation capacity

Drug release

A B S T R A C T

A protein-polysaccharide-based potential nanocarrier system have been developed via a simple, one-step

prep-aration protocol without the use of long-term heating and the utilization of hardly removable crosslinking agents, surfactants, and toxic organic solvents To the best of our knowledge, this article is the first which summarizes in detail the pH-dependent quantitative relationship between the bovine serum albumin (BSA) and hyaluronic acid (HyA) confirmed by several physico-chemical techniques The formation of colloidal complex nanoconjugates

with average diameter of ca 210–240 nm is strongly depend on the pH and the applied BSA:HyA mass ratio

Particle charge titrations studies strongly support the core-shell type structure, where the BSA core is covered by

a thick HyA shell Besides the optimization of these conditions, the drug encapsulation capacity and the disso-lution profiles have been also studied for ibuprofen (IBU) and 2-picolinic acid (2-PA) as model drugs

1 Introduction

Due to the outstanding biocompatible and biodegradable nature, the

utilization of natural polysaccharides as sustained-release carriers is

beneficial for pharmaceutical fields (Mohamed, El-Sakhawy, &

El-Sakhawy, 2020; Turcs´anyi, Varga, & Csap´o, 2020) The targeted drug

delivery is possible by using HyA, which is a negatively charged

glycosaminoglycan, but the chemical modification of its hydroxyl-

and/or carboxyl-group is required (Yamanlar, Sant, Boudou, Picart, &

Khademhosseini, 2011) The HyA has a prominent role at biomedical

applications as well (Huerta-´Angeles et al., 2020) The HyA-based

nanohydrogels/nanoparticles (NPs) and films are implied as a

prom-ising area of the cancer treatment, tissue engineering, gene delivery etc

(Graça, Miguel, Cabral, & Correia, 2020) BSA is a water soluble

glob-ular protein consist of 583 amino acid residues (Ghosh & Dey, 2015);

wide ranges of active compounds are able to bind at the appropriate

binding sites of the protein (D¨om¨ot¨or et al., 2018) Depending on the pH,

the charge of the BSA is shifted from the positive to the negative value

reaching the isoelectric point (pI⁓5.1), that regulates the interaction

between the BSA-polymers and BSA-drugs via electrostatic interactions

(Varga, Hornok, Sebok, & D´ek´any, 2016)

Polysaccharide-protein conjugates may represent a new dimension

in the design of drug delivery systems The utilization of these complex

conjugates enhances the colloid stability, targeted efficiency, biocom-patibility or the reduced drug toxicity (Gaber et al., 2018), which is

published previously for e.g BSA/Chitosan (Karimi, Avci, Mobasseri, Hamblin, & Naderi-Manesh, 2013), Ovalbumin/Chitosan (Yu, Hu, Pan, Yao, & Jiang, 2006), Protamine/HyA (Mok, Ji, & Tae, 2007) or Lyso-zyme/Alginate (Fuenzalida et al., 2016) nanocarriers The polysaccharide-protein nanoconjugates are generally prepared by elec-trostatic complexation (Antonov et al., 2019), chemical conjugation (Martínez, Iglesias, Lozano, Teij´on, & Blanco, 2011) and electrospinning techniques (Torres-Giner, Ocio, & Lagaron, 2009) In some cases, the Maillard reaction is also implied, but the reaction requires long term heating (60 ◦C; for at least 3 h) (Edelman, Assaraf, Levitzky, Shahar, & Livney, 2017) which is specifically unfavourable for heat-sensitive drugs The fabrication of carrier NPs is usually carried out by chemical coupling in the presence of crosslinking agents, like glutaraldehyde (Chen et al., 2013), or tripolyphosphate (Turcs´anyi et al., 2020), which components can be difficult to remove during the purification process Based on these facts, the demand for the development of an effective polysaccharide-protein nanoconjugates with tuneable-size is strongly required, where the long-term heating and the utilization of hardly removable crosslinkers and surfactants is excluded

In this work we first demonstrate a preparation possibility of BSA/ HyA conjugates by a simple, controllable charge neutralization

* Corresponding author at: Department of Physical Chemistry and Materials Science, University of Szeged, H-6720, Rerrich B Square 1, Szeged, Hungary

E-mail address: juhaszne.csapo.edit@med.u-szeged.hu (E Csap´o)

Contents lists available at ScienceDirect Carbohydrate Polymers

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

https://doi.org/10.1016/j.carbpol.2020.117047

Received 15 July 2020; Received in revised form 1 September 2020; Accepted 1 September 2020

and infrared spectroscopy studies, but the interaction between serum

protein and HyA was not investigated Two articles focus on the

inter-pretation of the HyA- serum protein interaction by NMR (Filippov,

Artamonova, Rudakova, Gimatdinov, & Skirda, 2012) and integrated

computer modeling (Grymonpr´e, Staggemeier, Dubin, & Mattison,

2001), but only the presence and strength of the electrostatic interaction

was mentioned Moreover, we also considered to demonstrate the

suc-cessful encapsulation of two drugs with slightly different hydrophilicity

(2-PA: logP = -0.1; IBU: logP = 1.74) at the optimized weight ratio Both

molecules have one aromatic ring and one carboxylic group, but the

charge of the molecules at pH = 4.5 is different The IBU was selected for

our experiments as a model drug because this molecule is often studied

non-steroidal anti-inflammatory compound and its binding mechanism

to serum albumin binding sites is well-known 2-PA is a neuroprotective,

immunological and antiproliferative compound, it has not been

encap-sulated as active component in any drug delivery system Drug loading

(DL%) and the release mechanism of the encapsulated molecules are

examined, and the results are compared with other nanocapsule-based

carrier systems containing same drugs

2 Experimental

2.1 Materials

BSA (~66,000 Da), HyA sodium salt (1.5–1.8⋅106 Da), IBU sodium

salt (≥98 %), and 2-PA (≥99 %) were purchased from Sigma-Aldrich

The disodium hydrogen phosphate (Na2HPO4; ≥99 %), the sodium

dihydrogen phosphate monohydrate (NaH2PO4⋅H2O; ≥99 %), the

so-dium acetate 3-hydrate (CH3COONa⋅3H2O; ≥99 %), and sodium

hy-droxide (NaOH; ≥96 %) pastilles and the hydrochloric acid (HCl, ≥99

%) were bought from Molar Chemicals Acetic acid (AcOH, ≥99 %) was

purchased from Erd˝ok´emia Ltd Company Highly purified water was

obtained by deionisation and filtration with a Millipore purification

apparatus (18.2 MΩ cm at 25 ◦C) All reagents and solvents used were of

analytical grade without further purification

2.2 Methods

2.2.1 Synthesis of the unloaded BSA/HyA conjugates

The BSA/HyA conjugates were prepared by charge neutralization

method The 1.6 mg mL− 1 HyA and the 2 mg mL− 1 BSA stock solutions

were prepared in acetate buffer (0.010 M acetic acid/0.0057 M sodium-

acetate; pH = 4.5) and in MilliQ water, respectively In the first step, the

HyA stock solution was stirred at 350 rpm for 1 h and stored overnight at

4 ◦C During the synthesis, several BSA/HyA conjugates were prepared

within the range of mBSA/mHyA =0.25–5.00 mass ratios Namely, 2 mL

of 2 mg mL− 1 BSA was added dropwise into the 10 mL of 0.08–1.6 mg

mL− 1 HyA solutions under 1000 rpm magnetic stirring at 25 ◦C After

mixing the appropriate amounts of BSA to HyA, the samples were

further stirred for 2 h under 500 rpm Finally, the samples were

cen-trifugated at 5000 rpm for 5 min; the supernatant was removed, and the

samples were redispersed in acetate buffer The cleaning method was

repeated three times The cleaned products were freeze-dried (Christ

source with 6.8 mW output power, while the reflected light intensity is monitored in the 574–1000 nm spectrum range using an IPE AS CR S2010 spectrometer The sensorgrams were registered by SPR UP 1.1.11.3 (2014 IPE AS CR) control software Firstly, the protein solution (cBSA =10 μM) was flowed under a constant flow rate (50 μL min− 1) above the gold-coated SPR chip in order to immobilize the protein to the

surface of the sensor via Au-S covalent bond (Csap´o et al., 2016) On the next step, the HyA solution was flowed across the protein-functionalized surface with 50 μL min-1 flow rate During studies, the following con-ditions were used: T = 15− 30 ◦C, cHyA =2.5–10.0 μg mL− 1 and the pH range of 3.6− 5.5

2.2.4 Particle charge detector (PCD)

The PCD measurements were performed by a PCD-04 Particle Charge Detector (Mütek Analytic GmbH, Germany) with manual titration ac-cording to our previously detailed technique (Turcs´anyi et al., 2020) Firstly, the HyA was dissolved in acetate buffer at four different pH values (pH = 3.6; 4.0; 4.5; 5.0), while the concentration kept constant (20 mL of cHyA=0.36 mg mL− 1) The BSA solution (10 mg mL− 1) was added dropwise in 100− 100 μL portions to the HyA solution at 25 ◦C and the streaming potential values (mV) were registered The acquired re-sults were analyzed and fitted with the modified version of the sigmoidal Boltzman equation

2.2.5 Rheology

Anton Paar Physica MCR 301 Rheometer (Anton Paar, GmbH, Ger-many) equipped with cylinder geometry (CC27-SN12793) was used; the changing of the viscosity was followed at 25 ± 0.1 ◦C and at 37 ± 0.1 ◦C

at different pH values using acetate buffer (pH = 3.6; 4.0; 4.5) 10 mg

mL− 1 BSA solution was added drop by drop in 19 mL of 0.1 mg mL− 1

HyA solution at 40 μL/3 min dosing speed The effect of solvent dilution was also considered as described in our previous article (Csap´o et al.,

2018)

2.2.6 Thermal behaviour

The thermal behaviour of the BSA, HyA, and the lyophilized powders

of BSA/HyA conjugates at 6 different mass ratios were studied with thermogravimetric (TG) and DSC The TG studies were performed with the use of a Mettler-Toledo TGA/SDTA851e instrument with 5 ◦C min− 1

between the range of 25− 1000 ◦C, under constant air flow (50 mL min− 1) The DSC studies were performed with the use of a Mettler- Toledo DSC822e calorimeter with 5 ◦C min− 1 in the range of 25− 500

◦C under nitrogen stream (50 mL min− 1)

2.2.7 Fourier transformed infrared spectroscopy

The FT-IR spectra were registered with a Jasco FT/IR-4700 spec-trometer with the use of an ATR PRO ONE Single-reflection accessory (ABL&E-JASCO, Hungary) The spectra were recorded at a resolution of

2 cm− 1 between 4000 and 500 cm− 1 by accumulating 128 interfero-gram The samples prepared in the same method discussed in the pre-vious section

2.2.8 Circular dichroism spectroscopy

CD spectra were recorded by using Jasco J-1100 CD spectrometer

(ABL&E-JASCO, Hungary) at 25 ± 1 ◦C using 1 cm optical pathlength

quartz cuvette All spectra were recorded at 100 nm min− 1 scanning

speed in the middle UV-region (200− 300 nm) under N2 flow (3 L min− 1)

and represents the average of three scans The light source was a water-

cooled, high-energy xenon lamp (450 W) The raw data was converted

into mean residue ellipticity (MRE) using the Eq (1), and the ratio of the

α-helix content was calculated from the Eq (2)

MRE208=observed CD(mdeg)

α− helix (%) =− MRE208− 4000

33000 − 4000 ×100 (2)

where the C p is the molar concentration of the protein, n is the number of

amino acid residues, and l is the pathlength of the cuvette

2.2.9 Characterisation of the BSA/HyA NPs

The average size, size distribution, morphology, polydispersity index

and the Zeta-potential values were measured by DLS using a HORIBA SZ-

100 NanoParticle Analyzer (Retsch Technology GmbH, Germany) The

light source was a semiconductor laser (λ = 532 nm, 10 mW) and

photomultiplier tubes (PMT) were used as detector at 90◦ scattering

angle For registration of TEM images a Jeol JEM-1400plus equipment

(Japan) at 120 keV accelerating voltage was applied To determine the

encapsulation efficiency (EE%) and DL%, the absorbance spectra of the

supernatants of the centrifuged drug-loaded BSA/HyA conjugates were

registered by Shimadzu UV-1800 UV–vis double beam

spectrophotometer in a 1 cm quartz cuvette The measurements were carried out at room temperature in 200− 500 nm wavelength range The exact concentration of the non-encapsulated free IBU and 2-PA was calculated from the calibration curves, where the characteristic absor-bance band of the IBU and 2-PA were appeared at λ = 222 nm and λ =

264 nm in acetate buffer (pH = 4.5) medium, respectively (Fig S1) The DL% and EE% values were defined by Eqs (3) and (4)

DL% = encapsulatedmassofdrug

EE% = encapsulatedmassofdrug

2.2.10 In vitro release study

The in vitro dissolution profiles of the IBU- and 2-PA-containing BSA/

HyA conjugates were measured by UV–vis The release measurements were performed in phosphate buffer solution (pH = 7.4 ± 0.1) at 37 ± 0.5 ◦C and a semipermeable cellulose membrane (avg flat width = 25 mm; Mw cut-off = 14,000; Sigma-Aldrich) was used The data points were registered for 250 min The concentrations of the IBU and 2-PA in the release medium were determined by calibration curves (Fig S2) The possible release kinetics and the proposed mechanism can be defined from the fitting of the Weibull kinetic models (Varga, Turcs´anyi, Hor-nok, & Csap´o, 2019; Veres et al., 2017)

Fig 1 (A) Representative SPR reflectance curves before (black) and after (blue) addition of HyA solution at pH = 3.6 and (B) the registered sensorgrams at different

pH values (cHyA =2.5 μg mL− 1, 50 μL min− 1 flow rate, t = 25 ± 0.1 ◦C) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig 2 (A) Change of the streaming potential of pure BSA and HyA as a function of pH (B) Change of the streaming potential of HyA titrated with BSA at different pH

values (starting concentrations: cHyA =0.36 mg mL− 1, cBSA =10.0 mg mL-1, VBSA =100-100 μL)

3 Results

3.1 Surface plasmon resonance spectroscopy

The pH-dependence SPR studies have been performed at five

different pH using acetate buffers (pH = 3.6; 4.0; 4.5; 5.0; 5.5) at 25 ±

0.1 ◦C The concentration of HyA solution was fixed at 2.5 μg mL− 1 in

every cases The registered sensorgrams are presented in Fig 1

The decrease in the pH of the HyA solutions causes a greater shift of

the signal of the sensor response which suggests a pH-dependent

inter-action between the HyA and BSA If the pH of the HyA solution exceeds

the pH = 5.0, no significant interaction can be observed (pH = 5.5) The

concentration- and the temperature-dependence of the interaction of the

two studied macromolecules was also investigated, while the pH of the

HyA solution was remain the same (pH = 4.5; acetate buffer) At this pH

=4.5, the HyA are in fully deprotonated form and it is well-known that

no measurable structural change occurs in the secondary structure of

BSA at this slightly acidic conditions (Csap´o et al., 2016) For both series

of measurements, merely a slight shift can be observed in the signal of

the SPR gold-coated biosensor (Fig S3) Based on these results, it can be

concluded that the interaction between the polysaccharide and the

protein is strongly depends on pH, and the effect of the temperature and

the concentration is negligible under the studied conditions

3.2 PCD measurements

The interaction between HyA and BSA was also confirmed in detail

by PCD titrations, where the neutralization points were determined at

different pH values However, the acid-base property of HyA is well-

known (pKa = 2.83 (Turcs´anyi et al., 2020)), but for quantitative

interpretation of the results the isoelelectric points of BSA and HyA were

determined by PCD (Fig 2A)

The titration clearly proved that the BSA has positive charge below

pH = 5.0; the neutralization point is obtained at pH = 5.10 which value

is in good correspondence with the pI of BSA (Varga et al., 2016) For

HyA, the negative surface charge is dominant in wide pH range (pH =

2–11) In case of BSA/HyA system the titrations were carried out in the

pH range of pH = 3.5–5.0, where the macromolecules have well-defined

opposite charges Fig 2B shows that the initial negative charge of the HyA is shifted to higher values by the addition of BSA At pH = 3.6 and 4.0 the course of the curves is steeper, which is much longer at pH = 4.5

In accordance with SPR results, we found that there is no measurable change in the streaming potential values at pH = 5.0 The following neutralization points (where the streaming potential is 0 mV) are ob-tained by fitting the measured points by the modified Boltzmann equation: mBSA/mHyA =2.04 ± 0.01 (pH = 3.6), 2.69 ± 0.01 (pH = 4.0) and 5.05 ± 0.01 (pH = 4.5) It is also observed that the inflection points

of the titration curves (1.97, 2.51 and 4.46, respectively) appear before the neutralization points, which suggests structural changes between the macromolecules and the possible formation of BSA/HyA colloidal NPs before charge neutralization To confirm this observation rheological studies were also performed

3.3 Rheological studies

By the addition of the BSA stock solution to HyA the viscosity values are continuously decreased to a given point and then a slightly constant values are measured The intersection point of the fitted lines gives a breaking point This trend is observed at all the studied pH values (Fig 3)

The breaking points can be given at the following mBSA/mHyA ratios

at 25 ◦C: 1.43 (pH = 3.6), 2.26 (pH = 4.0) and 4.14 (pH = 4.5) The determined breaking points can be obtained at nearly similar mBSA/

mHyA ratios than the inflection points of the PCD curves (1.97 (pH = 3.6), 2.51 (pH = 4.0) and 4.46 (pH = 4.5)) The rheological studies have been carried out at 37 ◦C and similar trend was observed that at 25 ◦C, but the breaking points shifted towards the smaller values because of the different solvatated states of the macromolecules (breaking points at 37

◦C: 0.91 (pH = 3.6), 1.79 (pH = 4.0) and 3.63 (pH = 4.5) It can be concluded that, before neutralization, a structural change is occurred,

which strongly indicates the possible formation of colloidal NPs via

electrostatic interaction of BSA and HyA

Fig 3 Apparent viscosity values of BSA/HyA conjugates (marked with (⸰)) and the calculated streaming potential curves (grey continuous lines) as a function of

mBSA/mHyA at pH = 3.6 (A); pH = 4.0 (B) and pH = 4.5 (C) at 25 ◦C (starting concentrations: cHyA =0.10 mg mL− 1, cBSA =10.0 mg mL-1, VBSA =40-40 μL)

3.4 Characterisation of the drug-free BSA/HyA nanocarrier systems

3.4.1 DLS and TEM investigations

To prove the formation of BSA/HyA colloidal NPs, DLS and turbidity

studies have been also performed The measured hydrodynamic

di-ameters (Z-average) and the turbidity values of the BSA/HyA

conjugates-containing aqueous dispersion at pH = 4.5 are presented in

Fig 4 The parallel registered Zeta-potential values are seen in Fig S4

Fig 4A clearly indicates the formation of colloidal NPs according to

the increasing turbidity values within the range of mBSA/mHyA =

0.25–3.75 mass ratios If the mass ratio exceeds the mBSA/mHyA =4.0

value, the aggregation of NPs can be observed, as the inserted photo also

represents This observation is confirmed by DLS At the above

mentioned mBSA/mHyA =0.75–3.50 mass ratios the average diameter of

240 – 210 nm is obtained depending on the mass ratios For small BSA

content (mBSA/mHyA <0.50), larger diameters (300–400 nm) can be

measured because of the still large excess of HyA, while at high BSA

content (mBSA/mHyA >3.75) both the adhesion and the aggregation of

NPs is feasible The results of DLS and turbidity studies are in good

agreement with the main conclusions of both PCD and rheological

measurements The change of the Zeta-potential values as a function of

mBSA/mHyA also shows similar trend The values continually decrease

with increasing mBSA/mHyA (ζ = − 50.2 ± 1.2 mV (mBSA/mHyA =2); ζ =

− 37.4 ± 1.5 mV (mBSA/mHyA =4)) Fig 4B represents the DLS curve of

BSA/HyA NPs at mBSA/mHyA =2.0, where the average diameter is the

smallest The representative TEM image of this system also supports the

formation of NPs with nearly 200 nm average diameter and a well-

defined core-shell structure The negative surface charge may indicate

the formation of core-shell structure, where the BSA is covered by

negatively charged HyA To confirm these theory, further PCD studies have been performed The pure HyA solution and the probably HyA- coated BSA-based particles-containing dispersion were titrated with the same CTABr solution under similar conditions We hypothesized that, if HyA functions as a thick shell, nearly the same amount of CTAB will compensate the negative surface charge as would be expected on the free macromolecule as well For pure HyA solution and for composites the charge compensation points are obtained at nCTABr /nHyA(monomer) =

0.88:1.0 and at 0.93:1.0, respectively (Fig S5) This means that ca one

CTABr compensates one HyA monomer unit This observation is in good agreement with our previous result (Csap´o et al., 2018) The presence of core-shell-type NPs instead of “alloy-like” structure is more preferred The adsorption of the BSA on the surface of HyA can be ruled out

3.4.2 CD, FT-IR and thermal behaviour

Detailed structural studies of the BSA/HyA conjugates at several

mBSA/mHyA ratios have been also carried out by CD, FT-IR as well as TG/ DSC The main text contains the results of BSA/HyA NPs prepared at

mBSA/mHyA =2.0 ratio; other data are summarized in (Figs S6–S7)

Fig 5A, B represents the CD curves registered for the pure BSA and BSA/ HyA conjugates

In both cases the characteristic negative bands of the pure BSA are occurred at 208 and 220 nm (Zhou, Wu, Zhang, & Wang, 2017) In MilliQ water (Fig 5A), by the presence of HyA, only a slight shift is observed in the intensity of the negative band at 208 nm, which confirms that, there is no significant change in the secondary structure of BSA According to Eqs (1) and (2), the calculated α-helix content: 54.49 % for pure BSA and 57.76 % for BSA/HyA nanoconjugates This is in good correspondence with previous values (Zhou et al., 2017) At nearly

Fig 4 (A) The hydrodynamic diameter (●, left y-axis) and the turbidity values ( , right y-axis) of the BSA/HyA system as a function of increasing mBSA/mHyA (cBSA

=2 mg mL− 1) with the representative photos of the samples at mBSA/mHyA =2 and mBSA/mHyA =4 (B) DLS curve of BSA/HyA NPs using mBSA/mHyA =2 with the TEM images of the sample

Fig 5 CD curves of BSA (continuous red line) and BSA/HyA conjugates (dotted grey lines) in MilliQ water (A) and in acetate buffer (pH = 4.5) (B) t = 25 ◦C, cBSA = 2.78 μg mL− 1 (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

neutral pH, both macromolecules have negative charge and thus the

potential for electrostatic interactions is low In contrast, the α-helix

content of the BSA in the BSA/HyA conjugates is drastically decreased in

the presence of HyA at acidic conditions (Fig 5B) The proportion of the

α-helix content in acidic conditions is: 41.52 % for pure BSA and 13.56

% for BSA/HyA nanoconjugates Parallel with the decreasing α-helix

content the ratio of the β-sheets is increased to ca 75 % supported by the

fitting of the measured CD curve by Reed model (Reed & Reed, 1997) At

pH = 4.5 the macromolecules have well-defined opposite charge and

most probably the serum protein chains are charge compensated with

the approx 20-fold larger HyA and the protein chains are partially

unfolded and arranged to form a core(BSA)-shell(HyA) structure

sup-ported by TEM, Zeta-potential and PCD studies (Fig S5) The FT-IR

measurements clearly indicate that the presence of both HyA and the

protein in the composite; the determinative bands of amide I and II as

well as the vibrations of COO− and C-O(H) are presented in Fig 6A

The data are in good agreement with previously published values for

same macromolecules (Zhou et al., 2017) The FT-IR results did not

confirm obviously the observations of CD studies, which can be

explained by the fact that the FT-IR spectra were recorded in solid

powder form, while the CD studies were measured in aqueous solution

The degradation temperature (Tg) and the composition of the BSA/HyA conjugates were determined by TG and DSC (Fig 6B, C) Based on

Fig 6B, it can be seen that the intensive exotherm peak of the HyA (230

◦C) and the endotherm peak of the BSA (222 ◦C) does not appear in the conjugates which indicates the effective washing procedure and the composite does not contain macromolecules in free form By fitting of the DSC curves, the Tg values of the BSA and HyA are 202 ◦C and 222 ◦C, while for conjugates is 181 ◦C The decrease is presumably due to the formation of electrostatic interaction between the macromolecules The

Tg value was also determined by TG, where similar data can be obtained:

204 ◦C (BSA), 224 ◦C (HyA) and 185 ◦C (BSA/HyA conjugates) Considering the weight changes and the shape of the curves, the com-posite contains both BSA and HyA, but the BSA content is more domi-nant and the presence of only physical mixture can be excluded (Figs S8, S9)

3.5 Characterisation of the drug-loaded BSA/HyA nanocarrier systems

After the comprehensive study of the drug-free BSA/HyA NPs, the encapsulation of two model drugs is performed using mBSA/mHyA =2.0 mass ratio The average size, the size distributions and the morphology

Fig 6 FT-IR spectra (A), DSC (B) and TG (C) curves of BSA, HyA and the lyophilized powder of BSA/HyA nanoconjugates

Fig 7 Size distribution curves of IBU- and 2-PA-loaded BSA/HyA colloidal particles by DLS with the representative TEM images of these particles

of the drug-loaded BSA/HyA conjugates are presented in Fig 7

It can be stated that the capsulation was successful; the size of the

drug-loaded NPs is greater than the size of the drug-free NPs (dDLS =210

± 56 nm) and spherical morphology is observed The results of DLS

(dDLS, IBU-loaded = 250 ± 80 nm, ζ = -38.9 ± 1.4 mV; dDLS, 2-PA-loaded

=276 ± 74 nm; ζ = -42.0 ± 1.1 mV) and TEM (dTEM, IBU-loaded = 247

±92 nm; dTEM, 2-PA-loaded = 264 ± 80 nm) are in good agreement

However, the TEM images do not present core-shell structure but based

on strongly supported structure of unloaded NPs and the preparation

conditions, most probably the BSA-drug conjugates form the inner core

and the outer shell contains dominantly HyA The measured Zeta-

potential values (presented above) also indicate this supposition The

EE% and the drug loading also calculated by the Eqs (3)− (4) For IBU-

loaded BSA/HyA NPs the EE % is 40 % and the DL % is 6 % In case of the

2-PA-loaded BSA/HyA NPs the EE % is 14 % and the DL% is 2 %

Although no previously published data on the encapsulation of 2-PA

were found, but it can be clearly stated for IBU that the presence of

HyA slightly increases the drug content from DL% = 4–4.5 % (Csap´o

et al., 2016) to 6 % compared the DL% of our previously published pure

BSA-based systems After encapsulating the active substance, the

dissolution profiles of the drugs are also investigated at pH = 7.4 (in

phosphate buffer solution) at 37 ◦C The registered curves are presented

in Fig 8

For the 2-PA-loaded BSA/HyA conjugates almost the 28 % of the

encapsulated drug is released in the examined period (t = 240 min),

while for IBU-loaded BSA/HyA conjugates ca 52 % of the encapsulated

IBU is liberated The primer data points are fitted by different kinetic

models (First-Order, Second-Order, Weibull, Korsmeyer–Peppas,

Higu-chi), but for IBU- and 2-PA-loaded particles the measured data fit well to

the Weibull expression To compare the rate of dissolutions, the

disso-lution data of several IBU-containing nanocomposites synthesized in our

lab were considered and the corresponding half-time (t1/2) data were

compared We can compare the t1/2 values because similar technique

was used for the registration of the dissolution profiles and same kinetic

models were applied for fitting In case of mesoporous SiO2 (Varga et al.,

2015) and pure BSA (Csap´o et al., 2016), the change of t1/2 of the IBU is

not determinative in the presence of these carriers (t1/2(IBU) =0.11 h, t1/2

(SiO2/IBU) =0.08 h, t1/2(BSA/IBU) = 0.13 h) In case of our BSA/HyA

complex NPs the following t1/2 values are calculated under the applied

conditions: t1/2(IBU) =0.6 h, t1/2(BSA/HyA/IBU) =2.2 h These data clearly

confirm that the combination of HyA with BSA strongly facilitates the

prolonged release of IBU at pH = 7.4 (in phosphate buffer solution),

where nearly fourfold drug retention is achieved

4 Conclusion

We first demonstrated the quantitative characterization of HyA and

BSA by several physico-chemical techniques The studies by PCD,

rheology, turbidity, DLS, TEM and CD proved that the optimized fabri-cation as well purififabri-cation protocols resulted in the formation of

colloidal drug carriers with average diameter of ca 200− 210 nm via the

partially charge compensation of these macromolecules The commonly used long-term heating as well as the application of crosslinking agents, surfactants and toxic organic solvents were eliminated It was confirmed that the pH as well as the applied BSA/HyA mass ratios strongly influ-ence the size, the size distribution, and the proposed core-shell structure

of the potential drug carrier particles This work clearly highlights the importance of detailed physico-chemical characterization of the pure carriers and drug-containing carriers to design effective nanosystem for encapsulation Our results may successfully contribute to the develop-ment of promising drug delivery and controlled drug release colloidal systems in the future

CRediT authorship contribution statement Alexandra N Kov´acs: Methodology, Investigation, Writing -

orig-inal draft Norbert Varga: Methodology, Investigation, Visualization

´

Ad´am Juh´asz: Validation, Formal analysis, Visualization Edit Csap´o:

Conceptualization, Resources, Writing - review & editing, Supervision

Acknowledgements

This research was supported by the National Research, Development and Innovation Office -NKDIH through GINOP-2.3.2-15-2016-0034, GINOP-2.3.2-15-2016-0060 and FK131446 The research is supported

by the J´anos Bolyai Research Fellowship of the Hungarian Academy of Sciences (E Csap´o) The authors thank the registration of TEM images for Vikt´oria Hornok (University of Szeged, Department of Physical Chemistry and Materials Science)

Appendix A Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.carbpol.2020.117047

References

Antonov, Y A., Emanuel, N M., Celus, M., Kyomugasho, C., Hendrickx, M., & Cardinaels, R (2019) Complexation of pectins varying in overall charge with

lysozyme in aqueous buffered solutions Food Hydrocolloids, 94, 268–278 Chen, Z., Chen, J., Wu, L., Li, W., Chen, J., Cheng, H., et al (2013) Hyaluronic acid- coated bovine serum albumin nanoparticles loaded with brucine as selective

nanovectors for intra-articular injection International Journal of Nanomedicine, 8,

3843–3853 Csap´o, E., Juh´asz, Varga, N., Seb˝ok, D., Hornok, V., Janov´ak, L., et al (2016) Thermodynamic and kinetic characterization of pH-dependent interactions between

bovine serum albumin and ibuprofen in 2D and 3D systems Colloids and Surfaces A:

Physicochemical and Engineering Aspects, 504, 471–478

Fig 8 Dissolution profiles of the 2-PA (A) and IBU (B) molecules before (⸰) and after loading ()•

at pH = 7.4 ± 0.1 (in phosphate buffer solution) at 37 ◦C (the dotted

lines represents the fitting of the primer data via Second-order (free molecules) and Weibull kinetic models (drug-loaded particles)

Gaber, M., Mabrouk, M T., Freag, M S., Khiste, S K., Fang, J Y., Elkhodairy, K A., et al

(2018) Protein-polysaccharide nanohybrids: Hybridization techniques and drug

delivery applications European Journal of Pharmaceutics and Biopharmaceutics, 133,

42–62

Ghosh, S., & Dey, J (2015) Binding of fatty acid amide amphiphiles to bovine serum

albumin: Role of amide hydrogen bonding The Journal of Physical Chemistry B, 119

(25), 7804–7815

Graça, M F P., Miguel, S P., Cabral, C S D., & Correia, I J (2020) Hyaluronic

acid—Based wound dressings: A review Carbohydrate Polymers, 241, 116364

Grymonpr´e, K R., Staggemeier, B A., Dubin, P L., & Mattison, K W (2001)

Identification by integrated computer modeling and light scattering studies of an

electrostatic serum albumin-hyaluronic acid binding site Biomacromolecules, 2(2),

422–429

Huang, D., Chen, Y S., & Rupenthal, I D (2017) Hyaluronic acid coated albumin

nanoparticles for targeted peptide delivery to the retina Molecular Pharmaceutics, 14

(2), 533–545

Huerta-´Angeles, G., Ondreas, F., Brandejsov´a, M., Kopeck´a, K., Vagnerov´a, H.,

Kulh´anek, J., et al (2020) Formulation of hyaluronan grafted with dodecanoic acid

as a potential ophthalmic treatment Carbohydrate Polymers, 246, 116578

Karimi, M., Avci, P., Mobasseri, R., Hamblin, M R., & Naderi-Manesh, H (2013) The

novel albumin-chitosan core-shell nanoparticles for gene delivery: Preparation,

optimization and cell uptake investigation Journal of Nanoparticle Research, 15(5),

1651

Martínez, A., Iglesias, I., Lozano, R., Teij´on, J M., & Blanco, M D (2011) Synthesis and

characterization of thiolated alginate-albumin nanoparticles stabilized by disulfide

bonds Evaluation as drug delivery systems Carbohydrate Polymers, 83(3),

1311–1321

Biochemistry, 254(1), 36–40 Shen, Y., & Li, W (2018) HA/HSA co-modified erlotinib–albumin nanoparticles for lung

cancer treatment Drug Design, Development and Therapy, 12, 2285–2292 Torres-Giner, S., Ocio, M J., & Lagaron, J M (2009) Novel antimicrobial ultrathin

structures of zein/chitosan blends obtained by electrospinning Carbohydrate

Polymers, 77(2), 261–266 Turcs´anyi, ´A., Varga, N., & Csap´o, E (2020) Chitosan-modified hyaluronic acid-based

nanosized drug carriers International Journal of Biological Macromolecules, 148,

218–225 Varga, N., Benk˝o, M., Seb˝ok, D., Bohus, G., Janov´ak, L., & D´ek´any, I (2015) Mesoporous silica core–shell composite functionalized with polyelectrolytes for drug delivery

Microporous and Mesoporous Materials, 213, 134–141 Varga, N., Hornok, V., Sebok, D., & D´ek´any, I (2016) Comprehensive study on the structure of the BSA from extended-to aged form in wide (2-12) pH range

International Journal of Biological Macromolecules, 88, 51–58 Varga, N., Turcs´anyi, ´A., Hornok, V., & Csap´o, E (2019) Vitamin E-Loaded PLA- and PLGA-based core-shell nanoparticles: Synthesis, structure optimization and

controlled drug release Pharmaceutics, 11(7), 357 Veres, P., K´eri, M., B´anyai, I., L´az´ar, I., F´abi´an, I., Domingo, C., et al (2017) Mechanism

of drug release from silica-gelatin aerogel—Relationship between matrix structure

and release kinetics Colloids and Surfaces B, Biointerfaces, 152, 229–237 Yamanlar, S., Sant, S., Boudou, T., Picart, C., & Khademhosseini, A (2011) Surface functionalization of hyaluronic acid hydrogels by polyelectrolyte multilayer films

Biomaterials, 32(24), 5590–5599

Yu, S., Hu, J., Pan, X., Yao, P., & Jiang, M (2006) Stable and pH-sensitive nanogels

prepared by self-assembly of chitosan and ovalbumin Langmuir, 22(6), 2754–2759 Zhou, L., Wu, F., Zhang, X., & Wang, Z (2017) Structural and functional properties of

Maillard reaction products of protein isolate (mung bean, Vigna radiate (L.)) with dextran International Journal of Food Properties, 20, 1–13