j ourna l ho me pa g e: www.elsev ier .com/ locate /carbpol
-(furfural) chitosan hydrogels based on Diels–Alder cycloadditionsnd application as microspheres for controlled drug release
arcelino Montiel-Herreraa, Alessandro Gandinib, Francisco M. Goycooleac,e,eil E. Jacobsend, Jaime Lizardi-Mendozae, Maricarmen Recillas-Motaa,aldo M. Argüelles-Monala,∗
Laboratorio de Polímeros Naturales, Centro de Investigación en Alimentación y Desarrollo A.C., Carretera al Varadero Nacional km 6.6, Colonia Laslayitas, Guaymas CP 85480, Sonora, MexicoMaterials Engineering Department, Engineering School of São Carlos, University of São Paulo, 13566-590 São Carlos, BrazilInstitut für Biologie und Biotechnologie der Pflanzen, Westfälische Wilhelms-Universtät Münster, Hindernburgplatz 55, 48143 Münster, GermanyDepartment of Chemistry and Biochemistry, University of Arizona, 1306 E. University Blvd, Tucson, AZ 85721, USALaboratorio de Biopolímeros (CTAOA), Centro de Investigación en Alimentación y Desarrollo A.C., Carretera a La Victoria km 0.6, Ejido La Victoria,P 83000 Hermosillo, Sonora, Mexico
r t i c l e i n f o
rticle history:eceived 1 October 2014eceived in revised form 13 March 2015ccepted 15 March 2015vailable online 6 April 2015
eywords:hitosan
a b s t r a c t
In this study, chitosan was chemically modified by reductive amination in a two-step process. The syn-thesis of N-(furfural) chitosan (FC) was confirmed by FT-IR and 1H NMR analysis, and the degrees ofsubstitution were estimated as 8.3 and 23.8%. The cross-linkable system of bismaleimide (BM) and FCshows that FC shared properties of furan–maleimide chemistry. This system produced non-reversiblehydrogel networks by Diels–Alder cycloadditions at 85 ◦C. The system composed of BM and FC (23.8%substitution) generated stronger hydrogel networks than those of FC with an 8.3% degree of substitution.Moreover, the FC–BM system was able to produce hydrogel microspheres. Environmental scanning elec-
tron microscopy revealed the surface of the microspheres to be non-porous with small protuberances.In water, the microspheres swelled, increasing their volume by 30%. Finally, microspheres loaded withmethylene blue were able to release the dye gradually, obeying second-order kinetics for times less than600 min. This behavior suggests that diffusion is governed by the relaxation of polymer chains in theswelled state, thus facilitating drug release outside the microspheres.
Biopolymers can be chemically modified by the insertion ofolecules into their structures to react with specific molecules
nd respond to particular targets. Thus, novel modified-polymersan exhibit new properties, allowing them to display specificesponses to one or several stimuli (Recillas et al., 2009). For exam-le, click chemistry and biopolymers such as chitosan, carrageenannd cellulose have been used for these purposes in the recentast (Bertoldo, Nazzi, Zampano, & Ciardelli, 2011; Ifuku, Wada,orimoto, & Saimoto, 2011; Ifuku, Wada, Morimoto, & Saimoto,
012).Click chemistry is the name given to a group of reactions that
roceed rapidly under simple experimental conditions, producing
high yield of stereospecific molecules that are easily recovered, etc.(Kolb, Finn, & Sharpless, 2001; Kolb & Sharpless, 2003; Crescenzi,Cornelio, Di Meo, Nardecchia, & Lamanna, 2007; Hein, Lui, &Wang, 2008). One of those reactions is the well-known Diels–Aldercycloaddition, in which a conjugated diene and a substituted alkene(dienophile) react to generate a substituted cyclohexene (Gandini,Hariri, & Nest, 2003). Currently, the use of click chemistry inpolymer science is a strategy to design polymer-based hydrogels,drug and gene delivery systems, scaffolds for tissue engineeringand toxic substance and mineral chelation, among other appli-cations (Lee & Mooney, 2001; Crescenzi et al., 2007; Gao, Zhang,Chen, Gu, & Li, 2009). For example, 6-N,N,N-trimethyltriazolechitosan has been tested for gene delivery; N-carboxymethylchitosan and dithiocarbamate chitosan have been used as ion
scavengers in water treatment; and cross-linked chitosan withcollagen–glycosaminoglycans derivatives have been used for tis-sue engineering and wound healing (Peniche, Argüelles-Monal, &Goycoolea, 1998; Gao et al., 2009).
Polymer-based hydrogels are macromolecular networks thatwell rather than dissolve in a good solvent. They show usefuliscoelastic properties for a wide range of biomedical and tech-ological applications (Lee & Mooney, 2001; Ross-Murphy, 2003).hitosan itself can form highly viscous solutions at high concen-rations in acidic media but does not form true macromolecularetworks. However, specific chemical changes in the structure ofhitosan can lead to the formation of hydrogels. These chemicalhanges may occur on its amine moieties and hydroxyl groupss a result of several reactions (Bertoldo et al., 2011; Ifuku et al.,011, 2012; Kurita, 2006; Pastor de Abraham, 2004; Peniche et al.,998; Yalpani & Laurance, 1984). However, when these modifica-ions are made by amine reduction, usually the reaction is simplend occurs in one or two chemical steps. Nonetheless, in some cases,he chitosan imine product must react with a crosslinker to produce
acromolecular networks (Azevedo & Kumar, 2012).Furan chemistry can be a versatile route for the prepara-
ion of novel materials from renewable sources (Gandini, 2008).uran and maleimide derivatives can react through Diels–Alderycloaddition above 60 ◦C and this reaction sometimes reversest temperatures higher than 100 ◦C. As a consequence of theseeatures furan–maleimide Diels–Alder cycloaddition has been con-eived as a strategy to generate polymeric materials with potentialroperties of mendability, recyclability and thermal reversibil-
ty (Gandini, 2008; Gandini, Coelho, & Silvestre, 2008). Likewise,urfural is a furan derivative produced at major scale annu-lly from renewable sources. Its furan moiety is likely to beinked to maleimide derivatives to yield several architecturesf polymeric materials with the benefits of furan chemistry. Inddition, furfural can also produce imine products by amine reduc-ion.
In the present work, we report for the first time that aurfural–chitosan derivative designed and synthesized to be ableo click to a maleimide cross-linker leads to the generation of ahemical polymer gel network. The properties of this material wereharacterized and it was subsequently used to obtain hydrogelicrospheres. This study presents proof-of-principle that renew-
ble resources such as furfural and modified-chitosan, togetherith click chemistry, can be used to design and produce novelaterials.
. Materials and methods
.1. Materials
Chitosan (Fluka, with a degree of N-acetylation of 0.23 cal-ulated from 1H NMR and a viscosity-average molecular weightf 1.3 × 105 estimated from intrinsic viscosity measurements at5 ◦C in 0.3 M acetic acid/0.2 M sodium acetate, Rinaudo, Milas,
Dung, 1993) and Milli-Q grade water with conductivity valuesess than 2 �S/cm were used through all procedures. All othereagents were purchased from Sigma Aldrich unless otherwisendicated.
.2. Purification of chitosan
Chitosan particles were dissolved in 0.33 M acetic acid and thenequentially filtered through sintered glass filters (pore diameters00–160, 16–40 and 10–16 �m) and nitrocellulose filters (poreiameters 3, 1.2 and 0.8 �m). Next, the solution was precipitated
y the addition of 3 M NH4OH, and the chitosan was washed withater until the conductivity of the supernatant was less than
�S/cm. Finally, the product was dried at room temperature underacuum.
e Polymers 128 (2015) 220–227 221
2.3. Synthesis of N-(furfural) chitosan (FC)
The free amine groups of chitosan were covalently linked tofurfural. Briefly, 1 g chitosan (4.51 -NH2 mmol) was dissolved in50 mL of 2% aqueous acetic acid. To synthesize derivatives withtwo degrees of substitution, chitosan was reacted at room tem-perature with different stoichiometric amounts of furfural, 25 mg(0.26 mmol) and 125 mg (1.3 mmol) each for 2 h. Then, freshly pre-pared aqueous 10 mM NaH3BCN (10 mL) was gradually added to thereaction mixture at 1 mL/6 min intervals by means of a peristalticpump (Minipuls 3, Gilson, France). The reaction exhibited smallbubbles during this process. FC was precipitated with 3 M NH4OHand then was successively washed with water, water:ethanol 50:50and 25:75 (v:v) and ethanol. After drying, the product was purifiedwith ethyl ether by Soxhlet extraction during 48 h. Finally, the puri-fied FC (0.94 g; 94% yield) was dried at room temperature undervacuum.
2.4. Synthesis of bismaleimide
Bismaleimide (BM) was prepared in a two-step processwith 4,7,10-trioxa-1,13-tridecanediamine and maleic anhydride asdescribed by Gandini et al. (2008) with few modifications. Thefirst step consisted of the drop-wise addition of two equivalentsof maleic anhydride (previously dissolved in diethyl ether) to oneequivalent of 4,7,10-trioxa-1,13-tridecanediamine. This mixturewas agitated for 3 h in a magnetic stirring device at 25 ◦C. Then, thissolution was refluxed for 3 h at 50 ◦C to give a white solid material,which was filtered, washed with diethyl ether and dried. The sec-ond step consisted of the cyclization of the amic acid end-groupsby stirring anhydrous sodium acetate and acetic anhydride refluxedfor 5 h at 100 ◦C with a mild stream of N2. The mixture obtained waspoured onto cold water, and the precipitate was filtered, washedwith water and dried over sodium sulfate before removing the sol-vent. The final product, BM, was purified in a SiO2 chromatographiccolumn using a mixture of petroleum ether and ethyl acetate 1:2(v:v) as the eluent.
2.5. Preparation of N-(furfural) chitosan–bismaleimide hydrogels
A fresh mixture of FC dissolved in 2% acetic acid and BM dis-solved in DMSO, in a ratio of 2:1 (w:w), was used in all experiments.Before cross-linking, this mixture was handled as follows: first, themixture was stirred and sonicated gently for 5 min. Then, it wassubmitted to five cycles of vacuum/normal pressure to remove airbubbles.
Infrared spectra were recorded on a Nicolet Protege (System 460E.S.P) FT-IR spectrometer (Madison WI, USA) in pellet form withKBr, by the accumulation of 64 scans with a resolution of 4 cm−1.
2.7. Nuclear magnetic resonance (1H NMR)
High-resolution liquid 1H NMR spectroscopy was carried out ona Varian Inova-600 (599.7 MHz) equipped with a triple-resonance(HCN) cryogenic probe. All spectra were recorded at 25 ◦C. Sep-arately, chitosan and samples of FC with different degrees ofN-furfural substitution were solubilized in DCl/deuterium oxidesolution (0.5 N) and tested.
2.8. Dynamic rheology studies
Oscillatory viscoelastic measurements were performed usinga highly sensitive stress controlled Rheometer AR-G2 (TA
222 M. Montiel-Herrera et al. / Carbohydrate Polymers 128 (2015) 220–227
-(furfu
IgtvpamwmoApfi
2
so1msbdbcsfw(m
iwp
2
e(dtocdi
2
iw
Fig. 1. Synthetic scheme for the preparation of N
nstruments, New Castle, DE) equipped with stainless steel plateeometry (diameter 40 mm, gap 1000 �m) and a Peltier system foremperature control. After loading the sample, a thin layer of low-iscosity silicone oil was added around the sample’s periphery torevent evaporation. First, a mechanical spectrum was recordedt 25 ◦C to obtain the variation of the storage (G′) and loss (G′′)oduli with frequency (ω = 0.1–100 rad s−1). Second, the sampleas rapidly heated to 85 ◦C and measurements of G′ and G′′ wereade at three frequencies between 1 and 10 rad s−1 over the course
f 5 h. Finally, another mechanical spectrum was recorded at 85 ◦C. strain of 5% was used, thus ensuring that measurements wereerformed within the linear viscoelastic region, as previously con-rmed by strain sweep experiments (ω = 10 rad s−1) at 85 ◦C.
To obtain N-(furfural) chitosan–bismaleimide hydrogel micro-pheres, it was prepared a fresh mixture of 2.5 wt.% FC (8.3% degreef substitution, 150 mg dissolved in 6 mL of 2% acetic solution) and% BM dissolved in DMSO, in a ratio of 2:1 (w:w). Thereafter, theixture was stirred and sonicated gently for 5 min. Then, it was
ubmitted to five cycles of vacuum/normal pressure to remove airubbles. Next 3 mL of mixture was loaded into a 5 mL syringe torop-wise it onto vegetable oil at 65 ◦C, under continuous agitationy a helical stirrer. It was important to maintain the microspheresircling into the oil during the procedure (6 h), followed by 1 h ofedimentation at the same temperature. Then the oil was care-ully removed, and the formed microspheres were washed out firstith hexane (three times) and then with a solution of 2% Contrex
Decon Labs, Inc. King of Prussia, PA, USA) and water. Finally, theicrospheres were dried at room temperature.To measure the diameter of the microspheres the samples were
maged using an optical inverted microscope (AmScope) equippedith a previously calibrated ruler at 25× (0.4 AN). With this pur-ose, ToupView 3.7 AmScope software was employed.
.10. Morphological studies
The morphology of the microspheres was analyzed using annvironmental scanning electron microscope, model EVO LS10Carl Zeiss). Dried samples were mounted on aluminum stubs withouble-sided sticky carbon tape. Secondary and backscattered elec-ron detectors were used to visualize the samples. First, imagesf dried microspheres were taken, and then the humidity in thehamber was increased to 100%. Then, samples were dried imme-iately under high vacuum, and images were recorded again. Other
nstrument settings are described in the text and figure captions.
.11. Swelling of microspheres
To study the swelling process, a dried microsphere wasmmersed in water at 25 ◦C until equilibrium, and its diameter
as measured as described in Section 2.9. From the diameter of
ral) chitosan, FC, and Diels–Alder cycloaddition.
the dried and swelled microsphere, its volume was calculated. Thisprocedure was also conducted at 5 and 55 ◦C (n = 5) to evaluate theinfluence of temperature on the swelling process. The swelling wascalculated as follows:
S (%) = (Vsm − Vdm)Vdm
× 100 (1)
where Vsm and Vdm are the volume of swollen and dried micro-spheres, respectively, as calculated by the change in diameter. Theresults are given as the means ± SE.
2.12. Release of methylene blue by microspheres
To study the release of methylene blue (MB) by the micro-spheres, they were first loaded with MB. Briefly, a group of 30microspheres was immersed in 1 mL of aqueous MB (4.5 �M) andleft for 24 h. Then, the supernatant was removed, and the micro-spheres were rapidly rinsed with water to remove the excess MBfrom their surfaces. Finally, the microspheres were dried for 24 hat room temperature.
Loaded microspheres were placed inside a quartz cell contain-ing 3 mL of deionized water to measure the release of MB basedon absorbance (� = 665 nm) in a UV–vis spectrometer (UNICAMUV500) at different time intervals. Periodically, the quartz cell wasgently inverted a few times to mix. All measurements were car-ried out at room temperature. The concentration of MB released bythe microspheres into the aqueous medium was calculated from acalibration curve.
3. Results and discussion
3.1. Synthesis and characterization of N-(furfural) chitosan: FT-IRand 1H NMR analysis
Reductive amination is widely used to generate different iminecompounds by the reaction between primary or secondary aminesand aldehydes (Morrison & Boyd, 1998; Abdel-Magid & Mehrman,2006). This reaction has been used to synthesize aldehyde-chitosanderivatives in the past to develop polysaccharide derivatives suit-able for biomedical applications (Muzzarelli, Weckx, Filippini, &Lough, 1989; Muzzarelli et al., 1993; Kumar, Dutta, & Koh, 2011;Kumar, Koh, Kim, Gupta, & Dutta, 2012). In this sense, it was decidedto synthesize FC by reductive amination (Fig. 1). The synthesis ofFC was simple and carried out in two steps.
Furfural was covalently linked to free amine groups on chitosan,as indicated by the FT-IR and 1H NMR spectra. The FT-IR spectrumexhibits a band (1483 cm−1) associated with the secondary amineson chitosan, as well as several bands related to the unsaturatedcarbons (1654, 946 and 823 cm−1) and ether moieties (1130 and
1083 cm−1) of furan rings (Fig. 2A) (Kumar & Koh, 2012; Kumaret al., 2012). The 1H NMR spectrum shows the expected chemicalshifts generated by protons bonded to furan rings (Fig. 2B) (Martín-Matute, Nevado, Cárdenas, & Echavarren, 2003). In addition, all
M. Montiel-Herrera et al. / Carbohydrat
Fig. 2. A: FT-IR spectra of chitosan (C) and N-(furfural) chitosan (FC) (in KBr). B:1H NMR spectra of chitosan and N-(furfural) chitosan (8.3% degree of substitution).Bi
: Letters a, b and c, are the protons of furfural as represented in Fig. 1. Asterisksndicate unreacted furfural.
f these signals were absent in the FT-IR and 1H NMR spectraf chitosan. These results corroborated the synthesis of FC. Otherinor signals were also observed (Fig. 2B) and may correspond to
nreacted furfural, even when the final product was exhaustivelyurified by precipitation and 48 h extraction with ethyl ether (see
ection 2.3). Some degradation of furfural units on the DCl/D2Oolution (0.5 N) used as solvent for NMR experiments could not beiscarded as well.
ig. 3. Evolution of elastic modulus (closed symbols), G′ , and viscous modulus (open symf substitution) in 2% acetic acid and bismaleimide dissolved in DMSO at different concen.7%; (C and D) 1.0%; (E and F) 1.5%. (A, C and E) Time sweeps at ω = 1, 3.14 and 10 rad s−1
e Polymers 128 (2015) 220–227 223
The ratio of the 1H NMR signal integrals of the protons of fur-fural linked to chitosan (Brugnerotto, Desbrieres, Heux, Mazeau,& Rinaudo, 2001) to the integrals of those at C2 chitosan protonsgave degrees of substitution of 8.3 and 23.8%, respectively, for bothderivatives, calculated according to:
Furfural D.S. (%) =(∫
Ha +∫
Hb +∫
Hc)
/3∫H2
where Ha, Hb, Hc, and H2, are integral intensities corresponding toprotons of furan or C2-piranose ring (as indicated in Fig. 1).
These degrees of substitution provided the possibility to per-form reactions with different stoichiometric ratios between thepolymer and cross linker, to produce macromolecular networks.
3.2. Rheological studies
According to the literature, dienes such as fulvene-maleimidecompounds, maleimide-furan derivatives and cyclopentadienes bythemselves can react through Diels–Alder cycloaddition at specifictemperatures (Peterson & Palmese, 2009). That is, one compoundacts as a diene and the other as a dienophile. For this reason andbecause the FC in our study shared some of the chemical prop-erties of dienes, the capability of a FC solution alone to generatemacromolecular networks was tested. It was found that solutions of2.5 wt.% FC (8.3% degree of furfural substitution) dissolved in aceticacid (2%) did not produce any apparent change in the resultingphysical and rheological characters at temperatures up to 85 ◦C.
Macromolecular networks of diverse chitosan derivatives havebeen achieved by different methodologies (Argüelles-Monal,
Stokke, 1998; Montembault, Viton, & Domard, 2005a,b; Azevedo& Kumar, 2012; Kumar & Koh, 2012). In this case, BM was usedas a cross-linker between FC chains to form hydrogel networks
bols), G′′ , with time and frequency for N-(furfural) chitosan 2.5 wt.% (8.3% degreetrations, in a ratio of 2:1 (w:w). Bismaleimide concentration as follows: (A and B)
. (B, D and F) Frequency sweeps. All measurements at 85 ◦C.
224 M. Montiel-Herrera et al. / Carbohydrat
A
B
Fig. 4. Evolution of elastic modulus (closed symbols), G′ , and viscous modulus(open symbols), G′′ , with time and frequency for the system of N-(furfural) chitosan2.5 wt.% FC (23.8% degree of substitution) in 2% acetic acid and 1% bismaleimideiF
biosuBmo8w
show that G′ and G′′ were independent over frequency (Fig. 4B).
Fsi
n DMSO, in a ratio of 2:1 (w:w). A: Time sweeps at ω = 1, 3.14 and 10 rad s−1. B:requency sweeps. Measurements at 25 and 85 ◦C, as shown in the figure.
ased on Diels–Alder cycloaddition. The gelation process was stud-ed isothermally by rheology at 85 ◦C. It was found that mixturesf FC and BM at concentrations less than 2% FC (8.3% degree ofubstitution) and 0.5% BM were not able to generate macromolec-lar networks (data not shown), probably because the amount ofM is not enough to generate a polymer network, and/or the poly-er concentration is lower than the percolation threshold. Based
n these results, all viscoelastic measurements were performed at5 ◦C on mixtures of 2.5 wt.% FC and up to 1.5% BM in the sameeight relationship.
ig. 5. Micrographs of hydrogel microspheres for the system of 2.5 wt.% FC (8.3% degree
canning electron microscopy. Images were taken at spot size 500, 660× and 20 kV. In (Bnside the chamber as detailed in the text.
e Polymers 128 (2015) 220–227
Fig. 3 presents the variation of viscoelastic moduli during thecross-linking process using 2.5 wt.% FC (8.3% degree of substitu-tion) and bismaleimide solutions at concentrations between 0.7and 1.5 wt.%. The mechanical spectra at the end of the gelationare also included. The gelation process of the mixture with 0.7%BM does not show any significant change in the values of bothmoduli (Fig. 3A). In fact, this system generated a very weak hydro-gel network after 5 h of reaction, as G′ shows some dependencyover frequency and higher values than G′′ before the polymericnetwork collapsed at frequencies over 30 rad s−1 (Fig. 3B). Increas-ing the BM concentration to 1 wt.% results in a firmer hydrogel.The effect of increasing BM in the mixture was observed after100 min of reaction, when G′ = G′′ (Fig. 3C). In this case, the hydro-gel formation was faster in comparison with that of 0.7% BM. Inaddition, G′ showed almost no dependence on frequencies below60 rad s−1 and also showed higher values than G′′ (Fig. 3D). More-over, when the BM concentration was increased to 1.5%, hydrogelformation occurred even faster. After 60 min of reaction, G′ sur-passed G′′, and both moduli showed their highest values comparedwith the other two systems (Fig. 3E). Indeed, G′ showed values 10times higher than G′′, and both moduli exhibited very little depend-ence on frequency (Fig. 3F). It is well known that stronger hydrogelnetworks display a significant parallel distance between G′ andG′′ over frequency. All mixtures containing 1.5% BM showed thisdistinction.
These results indicate that macromolecular networks formedbetween FC and BM are influenced by reaction stoichiometry. Forthis reason, another mixture using FC with a higher degree of sub-stitution was prepared (23.8% degree of substitution). As expected,the mixture of 2.5 wt.% FC and 1% BM generated a stronger polymernetwork, characterized by quicker and significant increases of bothmoduli, G′ and G′′ (Fig. 4A). This macromolecular network displayedvalues of G′ and G′′ that were several orders of magnitude higherthan those displayed by previous systems. The mechanical spectra
Combined, these results showed that mixtures of FC with a higherdegree of substitution and greater BM concentrations producedfaster and stronger polymer networks.
of substitution) in 2% acetic acid and 1% BM in DMSO visualized by environmental) arrows indicate the pores produced at 100% humidity. Temperature and pressure
hydrat
nton2vi
ucl
�
a
�
wtshoe5
mmgoML&Titt(anmH
3
hsc(gaoic2igaK
rtw
M. Montiel-Herrera et al. / Carbo
It is evident that the elastic behavior of these networks displaysoticeable differences as the degree of substitution on chitosan andhe amount of BM are varied. In particular, by changing the degreef furan substitution on chitosan from 8.3 to 23.8%, the value ofetwork storage elastic modulus is increased by two decades (from0 to 2400 Pa; Figs. 3D and 4). This is in marked contrast to con-entional flexible polymer materials, where the elasticity is rathernsensitive to the density of chemical cross-links.
Then, provided that rheological measurements were conductednder the linear viscoelastic regime, allow us to use the basiconcepts of the Flory rubber elasticity theory to estimate the cross-inking density, �x, according to (Pescosolido et al., 2012):
x = G′
RT
nd the network average mesh size, �a:
a = 3
√6
��xNa
here R is the universal gas constant, T is the absolute tempera-ure and Na is the Avogadro number. The network average meshize could allow characterizing the cross-linking process for a stiffydrogel crosslinked in solution (Gardel et al., 2004). From the dataf the storage modulus shown in Figs. 3D, F and 4, it is possible tostimate network mesh-sizes of 123, 78 and 16 nm (for G′ values of, 20 and 2400 Pa, respectively).
Hydrogels produced by chitosan derivatives are considered softaterials as they show large deformation responses, and theiroduli have slight frequency dependence by contrast with strong
els that have no frequency dependence. Moreover, the valuesf G′ and G′′ can reach up to a few hundred Pascal (Argüelles-onal et al., 1998; Guo et al., 1998; Argüelles-Monal, Goycoolea,
izardi, Peniche, & Higuera-Ciapara, 2003; Grassi, Grassi, Lapasin, Colombo, 2007; Azevedo & Kumar, 2012; Kumar & Koh, 2012).he viscoelastic characteristics of some chitosan networks are ofnterest for the development of new materials, as shown by cer-ain polysaccharide derivatives obtained by click chemistry andheir hydrogels used for biological and biomedical applicationsSuch, Johnston, Liang, & Caruso, 2012). In addition, chitosan hasdvantageous properties (e.g., biocompatibility, mucoadhesive-ess, degradation into amino sugars, positive charges, etc.) thatake it attractive for drug delivery systems (Betancourt, Doiron,oman, & Brannon-Peppas, 2009).
.3. Preparation and characterization of hydrogel microspheres
In this work, FC–BM hydrogel system was utilized to prepareydrogel microspheres with interesting properties. The micro-pheres were produced by means of simple experimentalonditions. Dried microspheres presented radii of 66 ± 4 �mn = 43). When in contact with water, the microspheres swelled,radually reaching their swollen state after 15 min of immersion,nd maintained their morphological characteristics for long periodsf time (optical micrograph of a dried and swollen microspheres given in Supplementary material, Fig. S1). The swelling pro-ess was reversible and independent of the temperature at 5,5 or 55 ◦C. At either of those temperatures, the microspheres
ncreased their volume by 30%. This feature is important for hydro-els that are required to exhibit volume conservation (swellingnd shrinkage) over wide ranges of temperatures (Kamata, Akagi,ayasuga-Kariya, Chung, & Sakai, 2014).
The morphology of the microspheres was studied with envi-onmental scanning electron microscopy (Fig. 5). The surfaceopology of dried microspheres were visualized as non-porousith small protuberances (Fig. 5A). These types of structures have
e Polymers 128 (2015) 220–227 225
been described in several chitosan derivatives (Kumar et al., 2011;Kumar et al., 2012). The FC–BM microspheres showed collision-like formations in their surfaces, most likely due to the centripetalforces to which they were subjected during production as wellas to impacts incurred during processing. Nonetheless, despitetheir compact surface morphology, the microspheres graduallyswelled and changed in volume as the humidity in the cham-ber increased to 100%. Fig. 5 presents a sequence of images fromthe same microsphere as the humidity was increased inside themicroscope. Initially, at 4.6% humidity (1 ◦C, 30 Pa), this micro-sphere showed a compact structure with small bumps on its surface(Fig. 5A). Then, while the humidity in the chamber was periodi-cally increased from 15 to 100% (1 ◦C, 659 Pa), small pores appearedon some areas of the microsphere’s surface (Fig. 5B). These poresremained visible for another 30 min while the humidity levels weredecreased to 4.6%. After less than 5 min, the surface of the micro-sphere returned to its initial state (Fig. 5C and D). This behaviorsuggests the existence of dynamic viscoelastic properties on themicrospheres during the process of swelling in high humidity con-ditions. Such viscoelastic behavior may generate pores throughoutthe surface of the microsphere, allowing the diffusion of waterthroughout the macromolecular network. These pores could formappropriate pathways inside the microspheres, giving rise to thenecessary conditions for use as delivery vehicles.
The release of drugs involve several mechanisms, principallybased on diffusion, and sometimes occur due to syneresis, ero-sion or signals produced by the environment (Schwartz, 2002).These mechanisms are influenced by the chemical characteristicsof each material. Usually, drugs are released from the core of thehydrogel by diffusion processes; however, interactions betweenthe polymer, the drug and the environment are not absent in theseprocesses. Combined, these factors regulate diffusion and thus drugrelease. These features must be taken into account in the designof biocompatible materials for specific applications. For example,drug nanocarriers should maintain their sizes and remain unaggre-gated when injected intravenously, to avoid thrombus formationin blood vessels during distribution throughout the body (Mora-Huertas, Fessi, & Elaissari, 2010).
The capability of these microspheres as a drug deliverysystem was studied. Methylene blue was used as a model sub-stance (Cárdenas, Argüelles-Monal, Goycoolea, Higuera-Ciapara, &Peniche, 2003). For simplicity, microspheres were loaded with MBto study their release behavior. By the naked eye, it was possibleto observe the loading of MB on the microspheres because theychanged in color from yellow to dark blue (not shown). Once themicrospheres were swollen in water, they slowly released the MB,reaching equilibrium after 8 h (Fig. 6). If it is assumed that eachmicrosphere released an equal amount of MB (based on their homo-geneous radii), then each released approximately 200 pg/mL everyhour until becoming empty. The release of MB by the microspheresbehaved as most diffusion-drug delivery systems. Initially, the sys-tem presented an exponential growth release of MB (burst) thatreached a plateau phase, as a result of the diffusion out of MB fromthe microspheres (Mora-Huertas et al., 2010).
From Fig. 6, it is evident that the release of MB from the FC–MBmicrospheres obeys second-order kinetics at times less than 10 hbecause the release data were readily fit with Schott’s equation(Schott, 1992):
t
W= A + Bt (2)
where W is the release value at time t, B = 1/W∞ is the inverse ofmaximum release and A = 1/(dW/dt)o is the reciprocal of the initialrelease rate.
226 M. Montiel-Herrera et al. / Carbohydrat
Fig. 6. Plot showing the dependence of the release of methylene blue (W) and thereciprocal of the average release rate (t/W) on time at 25 ◦C by hydrogel microspheresfit
ir
w(
k
wimnt
b&icttnm
4
nsNwtbDwvwpb
or the system of 2.5 wt.% FC (8.3% degree of substitution) in 2% acetic acid and 1% BMn DMSO. Filled symbols correspond to experimental values. Dotted line representshe linear regression of the corresponding t/W vs. t plot [Eq. (2)].
It was shown that this equation describes a second-order kinet-cs process regarding the remaining swelling, in this case, theemaining amount of MB, expressed as:
dW
dt= k(W∞ − W)2 (3)
here the specific rate constant, k, is related to parameter A in Eq.2) as follows:
= 1AW2∞
(4)
This kinetic profile confirms the suggested elastic behavior inhich diffusion is governed by the relaxation of polymer chains
n the swollen state, thereby facilitating drug release outside theicrospheres. Moreover, as a positively charged molecule, MB is
ot electrostatically attracted to chitosan, giving rise to this mono-onic second-order kinetics release.
Chitosan micro- and nanovehicles have been proposed for noveliological and biomedical purposes (Betancourt et al., 2009; Kumar
Koh, 2012). In this sense, the FC–BM hydrogel system exhib-ted the potential use of reductive amination and Diels–Alderycloaddition to generate materials with remarkable viscoelas-ic and diffusion properties that could be taken as candidates forhe development of easy-to-produce new materials for biotech-ological applications. Further studies are needed to test theseicrospheres as drug carriers for biological systems.
. Conclusions
Through simple experimental conditions, the synthesis of aovel chitosan derivative based on the use of renewable resourcesuch as furfural, has been achieved. The FT-IR and NMR analysis of-(furfural) chitosan suggest the possibility to produce derivativesith different degrees of furfural substitution. This characteris-
ic opens opportunities to investigate the chemical interactionsetween furan–chitosan derivatives and maleimide compounds viaiels–Alder reactions. In this sense, N-(furfural) chitosan reactedith bismaleimide yielding hydrogel networks with remarkable
iscoelastic properties. Thus, non-porous hydrogel microspheresere produced, which exhibited interesting controlled releaseroperties appropriate for the development of biological andiomedical applications.
e Polymers 128 (2015) 220–227
Acknowledgements
We are grateful to MSc. Karla Guadalupe Martínez Robinson andMSc. Luisa Lorena Silva Gutiérrez for their technical support. Thisresearch was financed by Centro de Investigación en Alimentación yDesarrollo, A.C. and by Fondo de Infraestructura, CONACYT, Mexico(Grant 226082). MM-H acknowledges a stipend from CIAD AC.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.carbpol.2015.03.052
References
Abdel-Magid, A. F., & Mehrman, S. J. (2006). A review of sodium triacetoxyboro-hydride in the reductive amination of ketones and aldehydes. Organic ProcessResearch & Development, 10, 971–1031.
Argüelles-Monal, W., Goycoolea, F. M., Lizardi, J., Peniche, C., & Higuera-Ciapara, I.(2003). Chitin and chitosan in gel network systems. In H. B. Bohidar, P. Dubin,& Y. Osada (Eds.), Polymer gels. Fundamentals and applications (pp. 102–121).Washington, DC: American Chemical Society. ACS symposium series 833
Argüelles-Monal, W., Goycoolea, F. M., Peniche, C., & Higuera-Ciapara, I. (1998). Rhe-ological study of the chitosan/glutaraldehyde chemical gel system. Polymer Gelsand Networks, 6, 429–440.
Azevedo, E. P., & Kumar, V. (2012). Rheological, water uptake and controlled releaseproperties of a novel self-gelling aldehyde functionalized chitosan. CarbohydratePolymers, 90, 894–900.
Bertoldo, M., Nazzi, S., Zampano, G., & Ciardelli, F. (2011). Synthesis and pho-tochromic response of a new precisely functionalized chitosan with clickedspiropyran. Carbohydrate Polymers, 85, 401–407.
Betancourt, T., Doiron, A., Homan, K. A., & Brannon-Peppas, L. (2009). Controlledrelease and nanotechnology. In M. M. de Villers, P. Aramwit, & G. S. Kwon (Eds.),Nanotechnology in drug delivery (pp. 267–312). USA: Springer.
Brugnerotto, J., Desbrieres, J., Heux, L., Mazeau, K., & Rinaudo, M. (2001). Overview onstructural characterization of chitosan molecules in relation with their behaviorin solution. Macromolecular Symposia, 168, 1–20.
Cárdenas, A., Argüelles-Monal, W., Goycoolea, F. M., Higuera-Ciapara, I., & Peniche, C.(2003). Diffusion through membranes of the polyelectrolyte complex of chitosanand alginate. Macromolecular Bioscience, 3, 535–539.
Crescenzi, V., Cornelio, L., Di Meo, C., Nardecchia, S., & Lamanna, R. (2007). Novelhydrogels via click chemistry: Synthesis and potential biomedical applications.Biomacromolecules, 8, 1844–1850.
Gandini, A., Coelho, D., & Silvestre, A. J. D. (2008). Reversible click chemistry at theservice of macromolecular materials. Part 1: Kinetics of the Diels–Alder reac-tion applied to furan–maleimide model compounds and linear polymerizations.European Polymer Journal, 44, 4029–4036.
Gandini, A., Hariri, S. L., & Nest, J.-F. (2003). Furan–polyether-modified chitosans asphotosensitive polymer electrolytes. Polymer, 44, 7565–7572.
Gandini, A. (2008). Polymers from renewable resources: A challenge for the futureof macromolecular materials. Macromolecules, 41, 9491–9504.
Gao, Y., Zhang, Z., Chen, L., Gu, W., & Li, Y. (2009). Synthesis of 6-N,N,N-trimethyktriazole chitosan via click chemistry and evaluation for gene delivery.Biomacromolecules, 10, 2175–2182.
Gardel, M. L., Shin, J. H., MacKintosh, F. C., Mahadevan, L., Matsudaira, P., & Weitz, D.A. (2004). Elastic behavior of cross-linked and bundled actin networks. Science,304, 1301–1305.
Grassi, M., Grassi, G., Lapasin, R., & Colombo, I. (2007). Understanding drug releaseand absorption mechanisms. In A physical and mathematical approach (1st ed.).Boca Raton, FL, USA: CRC Press Taylor & Francis Group. Chapter 3.
Guo, B., Elgsaeter, A., & Stokke, B. T. (1998). Gelation kinetics ofscleraldehyde–chitosan co-gels. Polymer Gels and Networks, 6, 113–135.
Hein, C. D., Lui, X. M., & Wang, D. (2008). Click chemistry, a powerful tool for phar-maceutical sciences. Pharmaceutical Research, 25, 2216–2230.
Ifuku, S., Wada, M., Morimoto, M., & Saimoto, H. (2011). Preparation of highly regios-elective chitosan derivatives via click chemistry. Carbohydrate Polymers, 85,653–657.
Ifuku, S., Wada, M., Morimoto, M., & Saimoto, H. (2012). A short synthesis of highlysoluble chemoselective chitosan derivatives via click chemistry. CarbohydratePolymers, 90, 1182–1186.
Kamata, H., Akagi, Y., Kayasuga-Kariya, Y., Chung, U.-I., & Sakai, T. (2014). Non-swellable hydrogel without mechanical hysteresis. Science, 343, 873–875.
Kolb, H. C., Finn, M. G., & Sharpless, K. B. (2001). Click chemistry: Diverse func-tion from a few good reactions. Angewandte Chemie International Edition, 40,2004–2021.
Kolb, H. C., & Sharpless, K. B. (2003). The growing impact of click chemistry on drugdiscovery. Drug Discovery Today, 8, 1128–1137.
Kumar, S., Dutta, P. K., & Koh, J. (2011). A physic-chemical and biological study ofnovel chitosan–chloroquinoline derivative for biomedical applications. Interna-tional Journal of Biological Macromolecules, 49, 356–361.
umar, S., Koh, J., Kim, H., Gupta, M. K., & Dutta, P. K. (2012). A new chitosan-thymineconjugate: Synthesis, characterization and biological activity. International Jour-nal of Biological Macromolecules, 50, 493–502.
umar, S., & Koh, J. (2012). Physicochemical, optical and biological activity ofchitosan–chromone derivative for biomedical applications. International Journalof Molecular Sciences, 13, 6102–6116.
urita, K. (2006). Chitin and chitosan: Functional biopolymers from marine crus-taceans. Marine Biotechnology, 8, 203–226.
ee, K. Y., & Mooney, D. J. (2001). Hydrogels for tissue engineering. Chemical Reviews,101, 1869–1879.
artín-Matute, B., Nevado, C., Cárdenas, D. J., & Echavarren, A. M. (2003). Intra-molecular reactions of alkynes with furans and electron rich arenes catalyzedby PtCl2: The roles of platinium carbenes as intermediates. The Journal of theAmerican Chemical Society, 125, 5757–5766.
ontembault, A., Viton, C., & Domard, A. (2005a). Rheometric study of the gelationof chitosan in aqueous solution without cross-linking agent. Biomacromolecules,6, 653–662.
ontembault, A., Viton, C., & Domard, A. (2005b). Physico-chemical studiesof the gelation of chitosan in a hydroalcoholic medium. Biomaterials, 26,933–943.
ora-Huertas, C. E., Fessi, H., & Elaissari, A. (2010). Polymer-based nanocapsules fordrug delivery. International Journal of Pharmaceutics, 385, 113–142.
orrison, R. T., & Boyd, R. N. (1998). Organic chemistry (5th ed.). USA: Allyn & BaconInc (Chapter 21).
uzzarelli, R., Weckx, M., Filippini, O., & Lough, C. (1989). Characteristic propertiesof N-carboxybutyl chitosan. Carbohydrate Polymers, 11, 307–320.
uzzarelli, R. A., Biagini, G., Bellardini, M., Simonelli, L., Castaldini, C., & Fratto,G. (1993). Osteoconduction exerted by methylpyrrolidinone chitosan in dentalsurgery. Biomaterials, 14, 39–43.
astor de Abram, A. (2004). Quitina y Quitosano: obtención, caracterización y aplica-ciones. Perú: Pontificia Universidad Católica del Perú.
e Polymers 128 (2015) 220–227 227
Peniche, C., Argüelles-Monal, W., & Goycoolea, F. M. (1998). Chitin and chitosan:Major sources, properties and applications. In M. N. Belgacem, & A. Gan-dini (Eds.), Monomers, polymers and composites from renewable resources (pp.517–542). Amsterdam: Elsevier B.V.
Pescosolido, L., Feruglio, L., Farra, R., Fiorentino, S., Colombo, I., Coviello, T., Matri-cardi, P., Hennink, W. E., Vermondend, T., & Grassi, M. (2012). Mesh sizedistribution determination of interpenetrating polymer network hydrogels. SoftMatter, 8, 7708–7715.
Peterson, A. M., & Palmese, G. R. (2009). Reversible Diels–Alder cycloaddition for thedesign of multifunctional network polymers. In E. Lahann (Ed.), Click chemistryfor biotechnology and materials science (pp. 195–215). Chichester, West Sussex,UK: John Wiley & Sons, Ltd.
Recillas, M., Silva, L. L., Peniche, C., Goycoolea, F. M., Rinaudo, M., & Argüelles-Monal,W. M. (2009). Thermoresponsive behavior of chitosan-g-N-isopropylacrylamidecopolymer solutions. Biomacromolecules, 10, 1633–1641.
Rinaudo, M., Milas, M., & Dung, P. L. (1993). Characterization of chitosan. Influenceof ionic strength and degree of acetylation on chain expansion. InternationalJournal of Biological Macromolecules, 15, 280–285.
Ross-Murphy, S. B. (2003). Thermoreversible and irreversible physical gels frombiopolymers. In H. B. Bohidar, P. Dubin, & Y. Osada (Eds.), ACS symposium series833 Polymer gels. Fundamentals and applications (pp. 51–69).
Schott, H. (1992). Swelling kinetics of polymers. Journal of Macromolecular Science,B, 31, 1–9.
Schwartz, M. (2002). Encyclopedia of smart materials. New York, USA: John Wiley &Sons, Inc.
Such, G. K., Johnston, A. P. R., Liang, K., & Caruso, F. (2012). Synthesis and functional-
ization of nanoengineered materials using click chemistry. Progress in PolymerScience, 37, 985–1003.
Yalpani, M., & Laurance, D. H. (1984). Some chemical and analytical aspects ofpolysaccharide modifications. Formation of branched-chain, soluble chitosanderivatives. Macromolecules, 17, 272–281.
![Tandem [4+2]/[3+2] Cycloadditions of Nitroalkenes](https://static.documents.pub/doc/80x56/6344c77adf19c083b107c0e7/tandem-4232-cycloadditions-of-nitroalkenes.jpg)
![Acyclic stereocontrol in [3+2]-cycloadditions of amino acid-derived isomünchnone dipoles](https://static.documents.pub/doc/80x56/635c2ff59576d778510a08e1/acyclic-stereocontrol-in-32-cycloadditions-of-amino-acid-derived-isomuenchnone.jpg)
![Origin of diastereoselectivity in the thermal [4+2] cycloadditions of dienophiles derived fromOppolzer's sultams: Stericvs. Stereoelectronic Influences](https://static.documents.pub/doc/80x56/634942b909e11fdd740b31db/origin-of-diastereoselectivity-in-the-thermal-42-cycloadditions-of-dienophiles.jpg)
