release of active bio(macro)molecules

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Versatile oligo(ethylene glycol)-based biocompatible microgels for loading/ release of active bio(macro)molecules

T

Garbine Aguirrea,b, Elise Deniaua, Annie Brûletc, Kamel Chougranid, Valérie Alardd, ⁎ Laurent Billona,b, a

Université de Pau & Pays Adour, CNRS, IPREM UMR 5254, Equipe de Physique et Chimie des Polymères, 2 avenue du Président Angot, Pau F-64053, France Bio-inspired Materials Group: Functionality & Self-assembly, Université de Pau & Pays Adour, 2 avenue du Président Angot, Pau F-64053, France c UMR12 CEA CNRS CEA Saclay, Laboratoire Léon Brillouin, F-91191 Gif Sur Yvette, France d LVMH Recherche Parfums et Cosmétiques, 185 Av. De Verdun, St Jean de Braye F-45804, France b

A R T I C LE I N FO

A B S T R A C T

Keywords: Stimuli-responsive microgels Cosmetic active molecules Microstructure Delivery systems

The present study aims in the understanding of the effect of oligo(ethylene glycol)-based biocompatible microgels inner structure on the encapsulation/release mechanisms of different types of cosmetic active molecules. For that, multi-responsive microgels were synthesized using three types of cross-linkers: ethylene glycol dimethacrylate (EGDMA), oligo(ethylene glycol) diacrylate (OEGDA) and N,N-methylenebisacrylamide (MBA). The inner morphology of the microgels synthesized was studied by 1H-nuclear magnetic resonance (1H NMR) and small-angle neutron scattering (SANS) techniques and no effect of cross-linker type on microgel microstructure was observed in the case of analysing purified microgel dispersions. Moreover, all the microgels synthesized presented conventional swelling/de-swelling behavior as a function of temperature and pH. Two hydrophobic, one hydrophilic, and one macromolecule as cosmetic active molecules were effectively loaded into different microgel particles via hydrophobic interactions and hydrogen-bonding interactions between −OH groups of active molecules and ether oxygens of different microgel particles. Their release profiles as a function of cross-linker type used and encapsulated amounts were studied by Peppas-Sahlin model. No effect of the crosslinker type was observed due to the similar inner structure of all the microgels synthesized.

1. Introduction In recent years, among the broad research field in materials science, microgel particles have received considerable attention as controlled delivery systems [1]. Microgel particles are environmentally responsive cross-linked colloidal particles able to swell rapidly in a thermodynamically good solvent, responding to different external stimuli such as temperature, pH, or ionic strength, among others [2,3]. In addition, due to their porous structure microgel particle are able to contain different types of small molecules, such as hydrophobic [4], hydrophilic [5,6], or even macromolecules [7,8] inside and release them by changing their volume. These unique properties make them an outstanding potential choice to be used as effective smart delivery systems. From the point of view of delivery applications, to date low interest has been devoted to the study of controlled cosmetic active molecules loading and release using stimuli-responsive microgels. However, the key parameter for achieving the most effective cosmetic and personal

care products is the controlled release of active cosmetic molecules to the target site of action on/in the skin [9,10]. Therefore, the design of new sophisticated and effective targeted delivery systems is considered to be the panacea of modern cosmetic. Nowadays in an effort to obtain more effective delivery systems, multi-responsive microgels that respond simultaneously to a combination of several stimuli are being developed [11–15]. Although poly(Nisopropylacrylamide) (PNIPAM) is the building block of a plethora of them in terms of thermo-responsiveness [16–19], the use of poly(oligo ethylene glycol) methacrylates (POEGMA) could be a better alternative thanks to their biocompatibility in the design of multi-responsive microgels for bio-applications [20–22]. In this regard, our group has recently reported the synthesis and characterization of new colloidally stable, biocompatible and multi-responsive oligo(ethylene glycol)based microgels able to encapsulate high contents of cosmetic active molecules as well as preformed magnetic nanoparticles [23–25]. In addition, the increasing interest that microgel particles have

⁎ Corresponding author at: Université de Pau & Pays Adour, CNRS, IPREM UMR 5254, Equipe de Physique et Chimie des Polymères, 2 avenue du Président Angot, Pau F-64053, France. E-mail address: [email protected] (L. Billon).

https://doi.org/10.1016/j.colsurfb.2018.12.019 Received 28 June 2018; Received in revised form 17 November 2018; Accepted 10 December 2018 Available online 11 December 2018 0927-7765/ © 2018 Elsevier B.V. All rights reserved.

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

throughout work.

gained as controlled delivery systems is mostly related to their swelling/de-swelling behavior. It is well known that the swelling properties of the microgels are strongly dependent on their microstructure [26,27]. With the aim of controlling the inner morphology of microgels and thereby their swelling properties, different polymerization techniques have been used to synthesize them such as batch, semibatch and controlled/living radical emulsion polymerizations [28–30]. Despite the growing interest in microgel particles as delivery systems, the effect of their inner morphology on encapsulation/release processes has been scarcely investigated. In this sense, with the aim of better exploiting the usefulness of stimuli-responsive microgels as smart delivery systems, an improved understanding of the underlying morphologies impact on the different molecules encapsulation/release mechanisms is required. Up to now, microgels inner morphology has been studied using different techniques such as static light scattering [31], 1H-nuclear magnetic resonance (1H NMR) transverse relaxation measurements [26,32,33], and small-angle neutron scattering (SANS) [34,35]. Recently, Gawlitza and coworkers reported the effect of poly (ethylene glycol) (PEG)-based microgels composition on particle size, volume phase transition temperature (VPTT) and inner structure. They observed that the microgels morphology changed below, close and above the VPTT contributing to the investigations on the use of these microgel particles for controlled uptake and release of active molecules [36]. The novelty of this contribution comes from the study of the inner morphology of different oligo(ethylene glycol)-based microgels by 1H NMR and SANS techniques together with the understanding of the inner morphology effect on the encapsulation/release processes of different types of active cosmetic molecules. In this work, the inner morphology and application as delivery systems of different multi-responsive oligo(ethylene glycol)-based microgels are reported. For this purpose, three different cross-linkers were used to control the cross-linking density distribution inside microgel particles. Colloidal characteristics such as the change of the average hydrodynamic diameters at different pH and temperatures were measured by Dynamic Light Scattering (DLS). Then, the microgels microstructure was determined by following the kinetics of the polymerization reactions using 1H NMR and also by SANS measurements. In addition, the encapsulation of different types of active cosmetic molecules, i.e. hydrophobic, hydrophilic, and macromolecules, was studied combining conventional techniques together with Nuclear Overhauser Enhancement Spectroscopy (NOESY NMR) technique. The in vitro release profiles of the different cosmetic active molecules in response to microgels inner morphology and loading amounts were analysed using dialysis method and in situ ATR/FTIR technique.

2.2. Synthesis of the microgels The synthesis of P(MeO2MA-OEGMA-MAA)-based microgels was carried out following the procedure described previously by Boularas et al. [23], using in this case a 2 L jacketed glass reactor. Briefly, 83.90 mmol of MeO2MA, 9.36 mmol of OEGMA, 1.92 mmol of crosslinker (OEGDA, EGDMA or MBA) and 930 g of “Milli-Q” grade water were placed into a 2 L jacketed glass reactor. The reactor content was stirred at 150 rpm and purged with nitrogen for 45 min to remove oxygen at room temperature. Then, 5 mmol of MAA dissolved in 30 mL of “Milli-Q” grade water were added to the jacketed glass reactor and the mixture was heated up to 70 °C. After adding the initiator (0.89 mmol of KPS dissolved in 40 mL of degassed water), the polymerization reaction was allowed to continue under nitrogen atmosphere while stirring for 6 h. The reaction mixture was subsequently cooled to 25 °C maintaining the stirring, and the final dispersion was purified by 5 centrifugation ∕ redispersion cycles (10 000 rpm, 30 min) with “Milli-Q” grade water. After the centrifugation step, the separated supernatant was dried and the contents of water soluble polymers (WSP) formed during the polymerization process were determined gravimetrically. 2.3. Polymeric characterization of microgels The evolution of the partial conversions of MeO2MA, OEGMA, MAA and cross-linkers (OEGDA, EGDMA or MBA) in the synthesis of multiresponsive microgels were determined by proton nuclear magnetic resonance spectroscopy (1H NMR). 1H NMR spectra were recorded at 400 MHz on a Bruker spectrometer, using trioxane (10% mol M) as internal standard and D2O/H2O as a solvent. The partial conversions of MeO2MA, OEGMA, MAA and cross-linkers were determined by the following expression:

xM = 1 −

IM (t) /ITrioxane (t) IM (0) /ITrioxane (0)

(1)

where IM and ITrioxane are the values of the peak integration of monomer M and internal standard (trioxane) at initial time t0 and sampling time t. 2.4. Colloidal characterization of microgels 2.4.1. Dynamic light scattering Colloidal characteristics of the microgels synthesized, such as the average hydrodynamic particle diameters at different temperatures and pHs, were measured with a Vasco-w Particle Size Analyzer from Cordouan Technologies (Pessac, France) working at an angle of 135° and a wavelength of 658 nm. Autocorrelation functions were recorded using a multi-acquisition mode and apparent diffusion coefficients were determined via the Pade-Laplace inverse algorithm of a large number of measurements. Hydrodynamic diameters were determined by the Stokes-Einstein equation (Eq. 2) and all calculations were performed using the NanoQ software. In all the measurements, ionic strength and pH of the medium were controlled using different buffered mediums depending on the pH of the solution.

2. Experimental part 2.1. Materials Di(ethylene glycol) methyl ether methacrylate (MeO2MA 95%, Aldrich), oligo(ethylene glycol) methyl ether methacrylate (OEGMA, monomethyl terminated with 8 EG repeat units, number average weight Mn = 475 g mol−1, Aldrich), methacrylic acid (MAA, Aldrich), oligo (ethylene glycol) diacrylate (OEGDA, number average weight Mn = 250 g mol−1, Aldrich), N, N’-methylenebisacrylamide (MBA, Aldrich), (ethylene glycol) diacrylate (EGDMA, Aldrich), potassium persulfate (KPS, 99% ABCR), and ethanol (VWR Chemicals) were used as received. Hydrochloric acid (HCl, 36 w/w ABCR) and potassium hydroxide (KOH, Aldrich) were used to control the pH of dispersions. Citric acid (Sigma-Aldrich) and sodium phosphate dibasic (Na2HPO4, Sigma-Aldrich) were used to prepare the buffers. Deuterium oxide (D2O, EurosiTop) was used for NMR characterization. Diethylamino hydroxybenzoyl hexyl benzoate (Uvinul-A), salicylic acid and hyaluronic acid were supplied by LVMH. Citronellol was provided by Dérivés Résiniques et Terpeniques DRT. “Milli-Q” grade water was used

Dz =

kBT 6Πƞr

(2)

where kB is the Boltzmann constant, T is the absolute temperature, ƞ is the viscosity of the medium, r is the hydrodynamic radius, and Dz is the apparent diffusion coefficient. 2.5. Small angle neutron scattering (SANS) The microgels were dissolved in deuterated water (D2O, 999% D, 446

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

After that, the mixed dispersion was stirred overnight at 50 °C to remove the organic solvent. Then, the dispersion was filtered to remove the unloaded cosmetic active molecule precipitate and the filter was washed with ethanol thrice, obtaining a solution containing unloaded cosmetic active molecules. Unloaded active molecule concentration was determined by UV–vis. In the case of citronellol, as it is an oil, after overnight incubation the samples were centrifuged during 30 min at 10 000 rpm. Then, the water solution of the supernatant containing free citronellol molecules was analyzed by 1H NMR to determine unloaded citronellol concentration. The second protocol used, renamed as “hydrophilic protocol” was used to encapsulate Hyaluronic acid. Microgel particles lyophilized were resuspended in different aqueous Hyaluronic acid solutions at particle concentration of 1 mg/mL. Then, the microgel particles were allowed to rehydrate for 12 h at room temperature while shaking. After separating microgel particles from the aqueous medium containing free Hyaluronic acid molecules through centrifugation, the equilibrium active molecule concentration was determined by ATR/FTIR. Entrapment efficiency (E.E.) was calculated as follows:

Eurisotop) at the desired concentration. SANS measurements were performed on PAXY spectrometer at Laboratoire Léon Brillouin (LLB) (Saclay, France). The scattered intensity was measured over a wide range of scattering vectors q, between 0.002 Å−1 and 0.4 Å−1, using three configurations (D = 1 m, λ = 6 Å ; D = 3 m, λ = 6 Å ; D = 7 m, λ = 15 Å; where D and λ are the sample-to-detector distance and the neutron wavelength, respectively, q = (4π/λ) sin(θ/2), and θ is the scattering angle). All samples were measured in 1 mm pathlength quartz cells at 20 °C and 50 °C. H2O measured in a 1 mm pathlength quartz cells was used as calibration standard. The scattered intensity was corrected for the empty cell scattering and the electronic background and set on an absolute unit (cm-1) according to standard procedures of the PASINET sofware [37]. The adjustments were performed using the SASVIEW software (http://www.sasview.org). SANS curves recorded at T = 20 °C were fitted with a Fuzzy Sphere Model [38] at low q-range (∼0.002-0.1 Å-1) and Lorentz (Ornstein-Zernicke Model) at high q-range (∼ 0.1–0.4 Å-1) while the SANS curves measured at 50 °C were fitted only with a Sphere model [39]. - In the Fuzzy Sphere Model, the scattering intensity I(q) is calculated as:

I(q) =

E.E. %=

scale (Δρ)2A2 (q)S(q) + bkg V

(3)

where the amplitude A(q) is given as the typical sphere scattering convoluted with a Gaussian to get a gradual drop-off in the scattering length density. It writes:

−(σFuzzy q)2 ⎞ 3[sin(qR) − qR cos(qr)] exp ⎜⎛ A(q) = ⎟ 3 2 (qR) ⎝ ⎠

The interactions between different cosmetic active molecule and microgel particles were studied by Diffusion Ordered Spectroscopy NMR (DOSY-NMR) and Nuclear Overhauser Enhancement Spectroscopy (NOESY-NMR) [25]. All NMR experiments were performed at 25 °C and 50 °C on a Bruker Avance 400 spectrometer equipped with a Bruker 5 mm BBFO probe and a gradient amplifier, which provides a z-direction gradient strength of up to 47.5 G/cm. The temperature was maintained constant within ± 0.1 °C by means of the BCU 05 unit. The proton chemical shifts were referenced to the HOD signal at 4.70 ppm. All DOSY experiments were performed using the bipolar longitudinal eddy current delay pulse sequence (BPLED). Typically, a value of 5 ms was used for the gradient duration (δ) 250 ms for the diffusion time (δ), and the gradient strength (g) was varied from 2.3 G/cm to 45.0 G/cm in 40 steps. The pulse repetition delay (including acquisition time) between each scan was larger than 2 s. Data acquisition and analysis were performed using the Bruker Topspin software (version 2.1). The T1/T2 analysis module of Topspin was used to calculate the diffusion coefficients. The diffusion measurement was carried out by observing the attenuation of the NMR signals during a pulsed field gradient experiment. In DOSY experiments using the Bipolar Longitudinal Eddy current Delay (BPLED) pulse sequence [40], this intensity change is described by:

(4)

In Eq. (3), A (q) is the form factor of fuzzy spheres. S(q) is the structure factor. From the reference [38] “the “fuzziness” of the interface is defined by the parameter σfuzzy. The particle radius R represents the radius of the particle where the scattering length density profile decreased to 1/2 of the core density. σfuzzy is the width of the smeared particle surface; i.e., the standard deviation from the average height of the fuzzy interface. The inner regions of the microgel that display a higher density are described by the radial box profile extending to a radius of approximately Rbox ∼ R - 2σ. The profile approaches zero at Rsans ∼ R + 2σ. Rsans is thus an approximated value of the overall size of scattering particles. - The Ornstein-Zernicke model is defined by Eq. (5):

scale + bkg 1 + (qξ)2

(5)

where ξ is the screening length, approximated in our case to a mesh size - In the Sphere Model, the scattering intensity I(q) is calculated as: 2

I(q) =

scale ⎡ 3V (Δρ)[sin(qR) − qR cos(qR)] ⎤ + bkg ⎥ V ⎢ (qR)3 ⎣ ⎦

(7)

2.7. Characterization of loaded-microgels

2

I(q) =

weight of active molecule in microgel dispersion × 100 weight of feeding active molecule

−Dγ 2g 2δ2 ⎛Δ − δ 3 ⎞ ⎝ ⎠

(6)

I= I 0 e

In Eq. (3), (4), and (6), scale is the volume fraction, V is the volume of the scatterer, R is the radius of the sphere, bkg is the incoherent background, Δρ is the difference between the scattering length densities (SLD) of the scatterer and the one of the solvent.

(8)

where I is the observed intensity, I0 is the reference intensity (unattenuated signal intensity), D the diffusion coefficient, g the gyromagnetic ratio of the observed nucleus, g the gradient strength, d the length of the gradient, and D the diffusion time. Diffusion Ordered Spectroscopy (DOSY) provides a mean for “virtual separation” of compounds, by providing a 2D map in which one axis is the chemical shift while the other is the diffusion coefficient [41]. All 2D NOESY spectra were recorded using a 1H spectral width of 12 ppm and acquired with the Bruker standard pulse program (noesygpph) of the Avance 400 spectrometer. For each NOE spectrum, 256 slices were recorded in the t1 dimension, and the number of scan was 32 (2048 data points for each scan). The FID was treated by square-shifted sine bell weighting functions in both dimensions using TopSpin 2.1. The mixing time and the relaxation time were set to 400 ms and 2 s,

2.6. In vitro cosmetic active molecules loading Active molecules were encapsulated by following the two protocols described in our previous work [25]. The first one, renamed as “hydrophobic protocol” was used to encapsulate Uvinul A, Salicylic acid and citronellol. Briefly, microgel dispersion (1 mg∕mL) was heated to and incubated at 50 °C (above the volume phase transition temperature, VPTT) for 30 min. To this microgel dispersion different concentrations of cosmetic active molecules in ethanol and preheated were added under magnetic stirring. The mixture was stirred for 30 min at 50 °C. 447

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

dependence of cross-linker consumption rate on WSP amount was not clear. In this regard, in the case of EGDMA, as it reacted in the early stage of polymerization, higher WSP amount than in the case of using OEGDA and MBA was expected. However, similar amount of WSP than in the case of MBA was obtained. In addition, the samples for NMR analyses were taken directly from the reactor. Therefore, the presence of WSP could affect on NMR macroscopic results obtained, i.e. WSP plus microgels. This point will be further analysed and discussed by screening microstructure of purified microgel by SANS measurements. The stimuli-responsiveness of the purified microgels synthesized was investigated by dynamic light scattering at different temperatures and pHs using different buffered media at an ionic strength of 1 mM. The thermal behavior of microgels was studied by DLS at pH 6 and they showed the conventional thermal behavior: upon increasing the temperature of the medium, the hydrodynamic average diameters decreased, in all the cases (Table S1). These results were corroborated by Diffusion Ordered Spectroscopy (DOSY-NMR). Increasing the temperature the diffusion constants of all the microgels decreased meaning a decrease of their size, i.e. the collapse of microgel particles (Fig. S2). In order to analyse the response to pH of the microgel particles, the final average hydrodynamic diameters at different pHs and at swollen state (25 °C) were measured by DLS. The pH-sensitiveness was the expected one: below the VPTpH microgel particles were collapsed and above it, particles were swollen due to the charge repulsion in the polymer network caused by the ionization of carboxylic groups (Table S1) [23,25]. Therefore, these results confirmed the thermo- and pH-responsiveness of different oligo(ethylene glycol)-based microgels synthesized using 2 L reactor, hence opening the possibility to scale-up their synthesis.

respectively. 2.8. In vitro cosmetic active molecules release The in vitro release of Uvinul A and Salicylic acid from microgel particles was evaluated by a dialysis method. Loaded-microgel particles with different active molecule concentrations were placed inside a dialysis tube and dialyzed in buffered media at pH 6 and 25 °C. The free active molecule concentration in the release medium was determined by UV–vis. In the case of Uvinul A, the in vitro release was also followed by realtime ATR/FTIR spectroscopy. For that, loaded-microgel particles with 1000 μg/mgmicrogel concentration were placed directly in buffered media at pH 6 and 25 °C. In situ ATR/FTIR monitoring was performed using a ReactIR 15 with a diamond Attenuated Total Reflection DiComp probe and equipped with a liquid nitrogen cooled MCT detector. Spectra were collected directly in the release medium at dif Moreover, drug release profiles were fitted to Peppas-Sahlin model [42], following the equation:

Mt

n 2m M∞ = k1t + k2t

(9)

where Mt and M∞ represented the cosmetic active molecules amount released at time t and that initially contained in the formulation, respectively, and k, k1 and k2 were release rate coefficients. 3. Results and discussion 3.1. Polymeric and colloidal characterization of microgels Recently, Boularas et al. have reported kinetics of the surfactant-free precipitation copolymerization of MeO2MA, OEGMA, MAA and different cross-linkers (OEGDA, EGDMA, and MBA) [23]. Regarding the synthesis, the novelty in this work comes from its up-scaling with the use of a 2 L reactor and over 100 g of microgels per batch. An impact of the type of cross-linkers was observed on the kinetic data suggesting a homogeneously cross-linked microgel using OEGDA, highly crosslinked shell and slightly cross-linked core microgel using MBA, and slightly cross-linked shell and highly cross-linked core microgel using EGDMA (Fig. S1). Similar results were obtained by Boularas et al. [23]. However, a heterogeneous microstructure in which core and shell have different cross-linking densities was expected in all the cases, due to the polymerization/cross-linking processed involved [26,43,44]. In Fig. 1, TEM images are in line with the dimensions determined by DLS in collapse state (Table S1). Indeed, the size of both EGDMA and MBA cross linker-modified microgels is homogeneous and around 400 nm whereas it increases to 600 nm of diameter for OEGDA. Moreover, similar spherical morphologies seem to be obtained without significant variation of the inner microstructure due to the low electronic contrast of the polymeric network. At this point it is important to remark that during the synthesis 30, 13 and 10 wt% of water soluble polymers (WSP) were formed in the case of OEGDA, EGDMA and MBA cross-linkers, respectively [45]. The

3.2. Encapsulation of cosmetic active molecules The loading of Uvinul A (hydrophobic molecule) and Salicylic acid (hydrophilic molecule) was analysed at different active molecule/microgel ratios and using the “hydrophobic protocol”. In Fig. 2 the encapsulated amounts and E.E. for both active molecules as a function of concentration are shown. As can be observed, encapsulated amount increased linearly as the cosmetic active molecules concentrations increased and a high loading amount of 750 μg/mgmicrogel was achieved, in all the cases. It seems that the type of cross-linker used had not any effect on the both active molecules loaded amounts, at least in the range of concentrations studied. In addition, E.E. increased slightly as the active molecule concentration increased indicating that the microgels had no adsorption interaction with Uvinul A and Salicylic acid [25]. Similar behavior was observed by Peng et al. [46] and Gui et al. [47] in the case of Doxorubicin (DOXO) encapsulation into different nanogels. With the aim of ensuring the encapsulation efficiency of both cosmetic active molecules into different microgel particles and in order to understand the internal interactions in cosmetic active molecules loaded-microgel particles, NOESY-NMR measurements were used to determine the short-distance local interactions between Salicylic acid and different microgel particles. As can be observed in Fig. 3, cross correlation peaks of the ethylene/methylene groups of the different

Fig. 1. TEM micrographs in dried state of Poly(MEO2MA-co-OEGMA-co-MAA) microgels crosslinked with EGDMA (left), MBA (center) and OEGDA (right). 448

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

Fig. 2. Encapsulated amounts (a) and entrapment efficiencies (E.E.) (b) as a function of active molecules concentration for different microgels particles. OEGDMA ■ Uvinul A and □ Salicylic acid, EGDMA ● Uvinul A and ○ Salicylic acid, MBA ▲ Uvinul A and Δ Salicylic acid. Fig. 3. NOESY-NMR spectra of Salicylic acid loaded-microgel particles crosslinked with a) OEGDA and b)MBA at pH 7 and 25 °C.

microgel particles (about 3–4 ppm) and aromatic ring of Salicylic acid (about 6.5–8 ppm) were visible, in all the cases. It is important to note that Uvinul A has two aromatic rings in their structure and therefore, we assumed that the same interactions could be given. Same interactions were observed in our previous work between microgel particles cross-linked with OEGDA and Benzophenone-4 and Salicylic acid active molecules [25]. In addition, Zhou and coworkers reported the hydrogen bonding interactions between the hydroxyl groups of dipyridamole (hydrophobic drug) and the ether oxygen of the ethylene glycol units of core-shell microgels [48]. Therefore, it could be concluded that the driving forces to encapsulate Salicylic acid and Uvinul A into different microgel particles were hydrophobic together with H-bonding interactions. In order to study the ability of oligo(ethylene glycol)-based microgels to encapsulate a broad spectrum of molecules, the encapsulation of citronellol (saturated aliphatic 10 Carbons-terpenoid) and macromolecular hyaluronic acid into microgel particles cross-linked with OEGDA was studied. Citronellol is a natural acyclic monoterpenoid (without aromatic ring) and hyaluronic acid is a hydrophilic diposysaccharide macromolecule which structure contains repeating units of D-glucoronic acid and N-acetyl-D-glucosamine (Scheme S1). Although the synthesis of Hyaluronic acid-based microgels has been reported [49–51], its encapsulation into microgel particles has not been studied yet. However, the ability of stimuli-responsive microgels to encapsulate macromolecules such as siRNA has been confirmed already [52]. Citronellol was encapsulated following the “hydrophobic protocol” and hyaluronic acid by “hydrophilic protocol” (see Experimental part for more information). As can be seen in Fig. 4, the encapsulated amounts of both molecules increased linearly with their concentration. On the one hand, in the case of citronellol it seems that hydrogen bonds

between OH group of active molecule and the ether oxygen of the ethylene glycol moieties of microgel were strong enough to obtain a good (E.E. > 70%). On the other hand, in the case of hyaluronic acid the E.E. values were lower than those observed in the case of small cosmetic active molecules (Uvinul A, salicylic acid, and citronellol). At this point it is important to point out that at pH 6 (encapsulation pH) hyaluronic acid molecules were negatively charged (pKa = 3) [53] as microgel particles and therefore, one of the reason could be the presence of an electrostatic repulsion between hyaluronic acid molecules and microgel particles being more difficult its encapsulation. However, in the case of salicylic acid, also negatively charged at encapsulation pH (pKa = 2.98), no effect of electrostatic repulsion on E.E. was observed. The second reason could be the larger size of this macromolecule (∼2 nm). This point will be further discussed in the next section. Nevertheless, the total amount of macromolecule loaded was much higher than that observed in the case of using other polymeric delivery systems to encapsulate macromolecules. For example, Cun et al. studied the encapsulation of siRNA molecules into poly(D,L-lactide-co-glycolide acid) (PLGA) nanoparticles obtaining E.E. values of 70% and siRNA encapsulated amounts of ∼2000 ng/mgmicrogel [54]. In summary of this part, the ability of multi-responsive oligo(ethylene glycol) microgels synthesized using different cross-linkers to encapsulate high amounts of different molecules such as hydrophobic, hydrophilic and macromolecules has been demonstrated. 3.3. In vitro cosmetic active release kinetics The effect of the type of cross-linker on Uvinul A and salicylic acid 449

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

Fig. 4. Encapsulated amounts (a) and entrapment efficiencies (E.E.) (b) as a function of active molecules concentration for microgel synthesized with OEGDA crosslinker. ■ Citronellol ● Hyaluronic acid.

was followed by real-time FTIR spectroscopy with an ATR probe, for the first time as far as we know. Microgel particles were loaded with Uvinul A at 1000 μg/mgmicrogel concentration and the spectra were collected directly in the release medium (25 °C and pH 6) at different incubation times. As can be observed, the release was the slowest one in the case of using MBA as cross-linker (Fig. 6). This could be due to the higher crosslinking density of the shell in the case of this microgel. The influence of the cross-linking density of hydrogels on the release kinetics of different active compounds has already been observed by other authors [56,57]. In this sense, Aguirre et al. observed that increasing the cross-linking density of poly(2-diethylaminoethyl) methacrylate-poly(N-vinylcaprolactam)- based (PDEAEMA-PVCL) core-shell nanogels the release of siRNA was more difficult and thereby, slower [52]. However, no difference was observed surprisingly in the case of microgels cross-linked with OEGDA and EGDMA. To investigate the effect of the encapsulated amount on release kinetics, different amounts of salicylic acid were encapsulated into the different microgel particles synthesized and the release kinetics was studied by dialysis at 25 °C and pH 6. As can be observed in Fig. S4, decreasing the encapsulated amount of salicylic acid more sustained release of it was obtained, in all the cases. It is known that the burst release of drug is typically linked to the surface concentration of the encapsulated drugs [58]. Common, strategies used to reduce the burst release are surface coating methods, surface cross-linking techniques, and surface extraction (removing drug molecules from the outer layers of controlled release delivery systems) [59]. Therefore, decreasing the encapsulated amount of salicylic acid, less active molecules were located near the surface of microgel particles and the diffusional distance was increased diminishing the release rates. Moreover, no effect of cross-linker type was observed decreasing the encapsulated amount of salicylic acid. In order to better understand the release mechanisms, the obtained release profiles were fitted to Peppas-Sahlin model [42]. A

Fig. 5. Cosmetic active molecules release as a function of cross-linker type. OEGDA ■ Uvinul A and □ Salicylic acid, MBA● Uvinul A and ○ Salicylic acid, EGDMA ▲ Uvinul A and Δ Salicylic acid.

release kinetics was studied by dialysis method. For that, Uvinul A and salicylic acid were encapsulated at a concentration of 1000 μg/ mgmicrogel using the “hydrophobic protocol” in both cases. Then, the release kinetics were studied at 25 °C and pH 6 being different microgel particles swollen (Fig. 5). In the case of salicylic acid, a burst release was observed in the case of microgel cross-linked with OEGDA. It seems that the expected homogeneous distribution of the cross-linker enhanced the release of salicylic acid. In this sense, Hsu et al. observed that bare poly(lactic-co-glycolic acid) (PLGA) nanoparticles presented a higher burst release of Ibuprofen than PLGA nanoparticles incorporated within poly(ethylene glycol(PEG) microgels [55]. They explain this behavior suggesting that the PEG matrix possessed barrier properties reducing the diffusion of ibuprofen. Therefore, we could assume that in the case of expected core-shell morphologies (microgels cross-linked with EGDMA and MBA), the presence of shell could hinder the release of active molecules obtaining more controlled release rates. However, although the release kinetics were slower at the beginning of the incubation in the case of microgel particles cross-linked with EGDMA and MBA, after 160 h the same amount of active molecule was released. In addition, similar release kinetics were observed for microgels crosslinked with EGDMA and MBA. This was an unexpected result since NMR results obtained suggested inverse core-shell microstructure for these microgels, i.e. slightly and highly cross-linked shell in the case of using EGDMA and MBA, respectively. Regarding the release of Uvinul A, no release of it was observed in the case of microgel particles crosslinked with MBA and EGDMA. The reason behind these unexpected results could be the precipitation of Uvinul A inside dialysis tube (Fig. S3). Solid Uvinul A were not able to cross the dialysis membrane and therefore, the dialysis method was not the suitable one to study the release of it from different microgel particles. With the aim of tackle this issue, Uvinul A in vitro and in situ release

Fig. 6. Uvinul A release by ATR/FTIR as a function of cross-linker type. ■ OEGDA, ● MBA, ▲ EGDMA. 450

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

superposition of two apparently independent mechanisms of active molecule transport was observed (Tables S2 & S3). On the one hand a Fickian controlled active molecule release mechanism, and on the other hand a Case II transport, which was associated with the swelling of microgel particles. In all the cases, the k1 values were higher than k2 ones indicating that the predominant release mechanism was the Fickian diffusion due to the non-covalent interactions between Uvinul A and salicylic acid and different microgel particles [60]. In the case of Uvinul A, the n values were below 0.43, indicating a Fickian diffusion mechanism. On the other hand, in the case of salicylic acid, the n values were closer to 0.85 for all the microgel particles, indicating a swelling controlled mechanism. The reason could be related with the hydrophobic/hydrophilic nature of active molecules. At pH 6, all the microgel particles were swollen, as explain above, and as salicylic acid is a hydrophilic active molecule the swelling of the microgel particles was enhanced. Regarding the effect of the active molecule encapsulated amount, decreasing it the n values were below 0.43, indicating a release mechanisms controlled by Fickian diffusion (Table S3).

Table 1 Structural properties of the network of the microgels prepared at a concentration of 10 g.L−1 and at pH6a. Microgels crosslinked with

R20°C (nm)

R50°C (nm)

RSANS (nm)

σfuzzy

ξ (nm)

σ

Φ

OEGDA EGDMA MBA

110 97 118

87 73 86

160 131 140

23 17 12

2 3 2

0.25 0.27 0.17

2.1 2.4 2.6

a

R20°C and R50°C are the size of the microgel particles at 20 and 50 °C, respectively, RSANS is the overall size of the microgel particles determined by SANS, σFuzzy is the fuzziness of the interface, ξ is the mesh size of the network, Swelling ratio: Φ = (R20°C/ R50°C)3 and σ is the polydispersity of the spheres.

those obtained by DLS (Table S1). At the collapsed state (T = 50 °C in Fig. 6a), the increase of the scattering intensity at low q-range varying according to a Porod’s law (in q−4) revealed the presence of dense and large heterogeneities of the network as already observed for other microgels [61–64]. SANS data recorded at 20 °C were fitted using two simple models: the fuzzy sphere model at low q-range describing the scattering from spherical particles with a “fuzzy” interface (Eqs. (3) and (4)) and the Lorentz model in the high q-range to characterize the mesh size of the microgel network at lower scale and thus in the core region of the microgel (Eq. (5)). The SANS data recorded at 50 °C were fitted using a polydisperse sphere model (Eq. (6)) only, which also indicates that collapsed spheres are formed at 50 °C. Though more complex models were reported in the literature to describe the structure of microgels [61–63], the models applied in the present work were sufficient to describe the parameters of interest like the overall size of the microgel (RSANS), the polydispersity (σ) of the sphere at 50 °C and the mesh size (ξ) (Table 1). The swelling ratios of the three types of microgels were calculated from their average sizes obtained in SANS at 20 °C and 50 °C and surprisingly similar values (Φ ∼ 2) were obtained, in all the cases. It has been reported that the microstructure of the microgels has an effect on the swelling ability of the microgel particles [65]. Therefore, similar swelling values obtained suggest that the variation of the type of cross-linker has no effect on the microgels microstructure. Indeed, SANS curves obtained for the different microgels have a similar trend (Fig. 7b) and all the parameters of interest obtained

3.4. Insight into microgels microstructure In order to better understand the results obtained in terms of no effect of the microstructure on encapsulation and/or release of different active molecules, small angle neutron scattering (SANS) measurements were carried out on different microgels. This technique is well suited to investigate the inner structure at nanoscales of microgels [61–64] In the present work, SANS measurements were undertaken in order to probe both the effect of the cross-linker type and the temperature on their inner structure. Typical SANS curves of microgels cross-linked with OEGDA were measured at two temperatures, below and above the VPTT, in order to probe the temperature-induced transitions (Fig. 7a). The change of the scattering curves with temperature confirmed the morphological transition of the thermo-responsive microgels. Indeed the increasing intensity observed at low q-range is shifted towards the higher q values when increasing the temperature from 20 °C to 50 °C. The size of the scattering objects thus decreased, i.e. microgel particles collapse at high temperature (above the VPTT). These results are in accordance with

Fig. 7. Small angle neutron scattering (SANS) curves of microgels dispersed in D2O at 10 g.L−1: a) Influence of the crosslinker type on SANS measured at T = 20 °C: OEGDA (empty blue circle), EGDMA (empty green triangle) and MBA (empty brown square); b) Influence of the measurement temperature on microgels cross-linked with OEGDA (empty red square for T = 50 °C and empty blue circle for T = 20 °C). Black lines correspond to the fits of experimental data (see experimental part) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 451

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

from the fits (Table 1) are very close. The mesh size values are peculiarly similar, around 2–3 nm: these values can thus explain similar swelling capabilities. In addition, it is important to note that the mesh size of the core is close to the size of hyaluronic acid and this could be the reason of the lower E.E. obtained. Nevertheless, as explained before, obtained E.E. values were higher than those reported for other macromolecules. Results obtained by SANS are completely different than those obtained by NMR (Fig. S1). As mentioned above, during the synthesis WSP were formed and their presence could affect NMR measurements. By contrast, SANS measurements were carried out using purified microgel particles, i.e. without WSP, and this could be reason of the discrepancy between the results obtained by two techniques. Therefore, it seems that the type of cross-linker has no effect on the microgel microstructure and hence, similar encapsulation/release trend are observed for all the microgel particles. In addition, it should be mentioned that the range of polydispersity values (σ < 0.3) is low compared to values generally reported for gels, which confirms that microgels synthesized in the present work are particularly well-defined. This result is very interesting from the point of view of delivery applications, since well-defined and monodisperse particles, in terms of size, are needed to ensure the reproducibility of the loading. Moreover, functional films by trigger-free self-assembly at skin temperature of such adhesive soft microgels were recently described. These microgels exhibiting simultaneously spontaneous film-forming with colored photonic properties [65] and unusually high encapsulation of active molecules [45] open up the development of a unique easy-to-handle material for cosmetic applications, but also smart patches for healthcare or medicine.

Acknowledgments The authors are grateful to LVMH Recherches for funding G. Aguirre postdoc position. A. Khoukh and T. Pouget are thanked for their support with the NMR technique and for TEM imaging, respectively. L. Billon acknowledges the CNRS for his “Délégation” fellowship. Gradient NMR probe is supported by public grants overseen by the French National Research Agency (ANR), ANR-10-EQPX-16 XYLOFOREST. 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.colsurfb.2018.12.019. References [1] D. Klinger, K. Landfester, Stimuli-responsive microgels for the loading and release of functional compounds: fundamental concepts and applications, Polymer 53 (2012) 5209–5231. [2] J. Ramos, A. Imaz, J. Callejas-Fernandez, L. Barbosa-Barros, J. Estelrich, M. Quesada-Perez, J. Forcada, Soft nanoparticles (thermo-responsive nanogels and bicelles) with biotechnological applications: from synthesis to simulation through colloidal characterization, Soft Matter 7 (2011) 5067–5082. [3] M. Das, H. Zhang, E. Kumacheva, Microgels: Old materials with new applications, Annu. Rev. Mater. Res. 36 (2006) 117–142. [4] J. Liu, J. Li, Z. Zhang, Y. Weng, G. Chen, B. Yuan, K. Yang, Y. Ma, Encapsulation of hydrophobic phthalocyanine with poly(N-isopropylacrylamide)/Lipid composite microspheres for thermo-responsive release and photodynamic therapy, Materials 7 (2014) 3481–3493. [5] A. Khanal, M.-P. Ngoc Bui, S.S. Seo, Microgel-encapsulated methylene blue for the treatment of breast cancer cells by photodynamic therapy, J. Breast Cancer 17 (2014) 18–24. [6] M. Dadsetan, K.E. Taylor, C. Yong, Z. Bajzer, L. Lu, M.J. Yaszemski, Controlled release of doxorubicin from pH-responsive microgels, Acta Biomater. 9 (2013) 5438–5446. [7] D. Costa, A.J.M. Valente, M.G. Miguel, J. Queiroz, Plasmid DNA microgels for drug/ gene co-delivery: a promising approach for cancer therapy, Colloids Sur. A: Physicochem. Eng. Asp. 442 (2014) 181–190. [8] Y. An, L. Zhang, S. Xiong, S. Wu, M. Xu, Z. Xu, Fluorine-containing thermo-sensitive microgels as carrier systems for biomacromolecules, Colloids Surf. B 92 (2012) 246–253. [9] I.M. Martins, M.F. Barreiro, M. Coelho, A.E. Rodrigues, Microencapsulation of essential oils with biodegradable polymeric carriers for cosmetic applications, Chem. Eng. 245 (2014) 191–200. [10] G. Cevc, U. Vierl, Nanotechnology and the transdermal route: a state of the art review and critical appraisal, J. Controlled Release 141 (2010) 277–299. [11] G. Aguirre, E. Villar-Alvarez, A. Gonzalez, J. Ramos, P. Taboada, J. Forcada, Biocompatible stimuli‐responsive nanogels for controlled antitumor drug delivery, J. Polym. Sci. Par A: Polym. Chem. 54 (2016) 1694–1705. [12] Y. Wang, J. Nie, B. Chang, Y. Suns, W. Yang, Poly(vinylcaprolactam)-based biodegradable multiresponsive microgels for drug delivery, Biomacromolecules 14 (2013) 3034–3046. [13] Y. Hu, W. Liu, F. Wu, Novel multi-responsive polymer magnetic microgels with folate or methyltetrahydrofolate ligand as anticancer drug carriers, RSC Adv. 7 (2017) 10333–10344. [14] W.-H. Chiang, W.-C. Huang, Y.-J. Chang, M.-Y. Shen, H.-H. Chen, C.-S. Chern, H.C. Chiu, Doxorubicin‐loaded nanogel assemblies with pH/Thermo‐triggered payload release for intracellular drug delivery, Macromol. Chem. Phys. 215 (2014) 1332–1341. [15] X. Chen, L. Chen, X. Yao, Z. Zhang, C. He, J. Zhang, X. Chen, Dual responsive supramolecular nanogels for intracellular drug delivery, Chem. Commun. 50 (2014) 3789–3791. [16] T. Hoare, R. Pelton, Impact of microgel morphology on functionalized microgel−drug interactions, Langmuir 24 (2008) 1005–1012. [17] P. Li, R. Xu, W. Wang, X. Li, Z. Xu, K.W.K. Yeung, P.K. Chu, Thermosensitive poly (N-isopropylacrylamide-co-glycidyl methacrylate) microgels for controlled drug release, Colloids Surf. B 101 (2013) 251–255. [18] W. Zhang, Z. Mao, C. Gao, Preparation of TAT peptide-modified poly(N-isopropylacrylamide) microgel particles and their cellular uptake, intracellular distribution, and influence on cytoviability in response to temperature change, J. Colloid Interface Sci. 434 (2014) 122–129. [19] J. Ramos, A. Imaz, J. Forcada, Temperature-sensitive nanogels: poly(N-vinylcaprolactam) versus poly(N-isopropylacrylamide), Polym. Chem. 3 (2012) 852–856. [20] T. Cai, M. Marquez, Z. Hu, Monodisperse thermoresponsive microgels of poly (ethylene glycol) analogue-based biopolymers, Langmuir 23 (2007) 8663–8666. [21] C. Chi, T. Cai, Z. Hu, Oligo(ethylene glycol)-based thermoresponsive core−shell microgels, Langmuir 25 (2009) 3814–3819. [22] T. Cai, P.D. Hu, M. Sun, J. Zhou, Y.-T. Tsai, D. Baker, L. Tang, Novel thermogelling dispersions of polymer nanoparticles for controlled protein release, Nanomedicine: NBM 8 (2012) 1301–1308. [23] M. Boularas, E. Deniau-Lejeune, V. Alard, J.-F. Tranchant, L. Billon, M. Save, Dual

4. Conclusions Multi-responsive oligo(ethylene glycol)-based microgels were synthesized by precipitation polymerization using three different crosslinkers. With respect to the swelling/de-swelling behavior, all the final microgel particles presented the expected pH- and thermo-responsive swelling-collapse transition in buffered media. An impact of the crosslinker type was observed on kinetic data by NMR due to the influence of the WSP formed during the synthesis. However, when purified samples were analysed by SANS, no influence of the type of cross-linker used on microgels microstructure was observed. Therefore, same heterogeneous microstructure was obtained for the three cross-linkers used. From the point of view of cosmetic applications, all the microgels synthesized were able to encapsulate high amounts of different molecules such as hydrophobic (Uvinul A and citronellol), hydrophilic (salicylic acid), and even macromolecules (Hyaluronic acid). It was demonstrated by NOESY-NMR that the driving forces to encapsulate them were the hydrophobic interactions together with the H-bonding interactions between −OH groups of active molecules and ether oxygen of different microgel particles. Regarding the in vitro release of Uvinul A and salicylic acid from different microgel particles, it was observed that the active molecules release was not influenced by the type of crosslinker used since the microgels microstructure was similar as demonstrated by SANS measurements. Moreover, decreasing the amount of active molecule encapsulated its burst release was reduced. In summary, a better understanding on the encapsulation/release of high amounts of versatile active molecules was achieved through a deep insight into the microstructure of different multi-responsive microgels.

Author contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

452

Colloids and Surfaces B: Biointerfaces 175 (2019) 445–453

G. Aguirre et al.

[24]

[25]

[26]

[27]

[28] [29]

[30] [31]

[32]

[33]

[34] [35]

[36]

[37] [38]

[39] [40]

[41] [42] [43] [44]

[45] G. Aguirre, A. Khoukh, P. Taboada, K. Chougrani, V. Alard, L. Billon, Smart selfassembled microgel films as encapsulating carriers for UV-absorbing molecules, Polym. Chem. 9 (2018) 1155–1159. [46] J. Peng, T. Qi, J. Liao, M. Fan, F. Luo, H. Li, Z. Qian, Synthesis and characterization of novel dual-responsive nanogels and their application as drug delivery systems, Nanoscale 4 (2012) 2694–2704. [47] R. Gui, Y. Wang, J. Sun, Embedding fluorescent mesoporous silica nanoparticles into biocompatible nanogels for tumor cell imaging and thermo/pH-sensitive in vitro drug release, Colloids Surf. B 116 (2014) 518–525. [48] T. Zhou, W. Wu, S. Zhou, Engineering oligo(ethylene glycol)-based thermosensitive microgels for drug delivery applications, Polymer 51 (2010) 3926–3933. [49] C. Luo, J. Zhao, M. Tu, R. Zeng, J. Rong, Hyaluronan microgel as a potential carrier for protein sustained delivery by tailoring the crosslink network, Mat. Sci. Eng. C. 36 (2014) 301–308. [50] C.A. Busatto, H. Labie, V. Lapeyre, R. Auzely-Velty, A. Perro, N. Casis, J. Luna, D.A. Estenoz, V.J. Ravaine, Colloid Interface Sci. 493 (2017) 356–364. [51] X. Jia, Y. Yeo, R.J. Clifton, T. Jiao, D.S. Kohane, J.B. Kobler, S.M. Zeitels, R. Langer, Biomacromolecules 7 (2006) 3336–3344. [52] G. Aguirre, J. Ramos, J. Forcada, Advanced design of t and pH dual‐responsive PDEAEMA–PVCL core–shell nanogels for siRNA delivery, J. Polym. Sci. Part A: Polym. Chem. 54 (2016) 3203–3217. [53] M.B. Brown, S.A. Jones, Hyaluronic acid: a unique topical vehicle for the localized delivery of drugs to the skin, JEADV 19 (2005) 308–318. [54] D. Cun, D.K. Jensen, M.J. Maltesen, M. Bunker, P. Whiteside, D. Scurr, C. Foged, H.M. Nielsen, High loading efficiency and sustained release of siRNA encapsulated in PLGA nanoparticles: quality by design optimization and characterization, Eur. J. Pharm. Biopharm. 77 (2011) 26–35. [55] M.N. Hsu, R. Luo, K.Z. Kwek, Y.C. Por, Y. Zhang, C.-H. Chen, Sustained release of hydrophobic drugs by the microfluidic assembly of multistage microgel/poly (lactic-co-glycolic acid) nanoparticle composites, Biomicrofluidics 9 (2015) 052601. [56] K.D.F. Vlugt-Wensink, T.J.H. Vlugt, W. Jiskoor, D.J.A. Crommelin, R. Verrijk, W.E. Hennink, Modeling the release of proteins from degrading crosslinked dextran microspheres using kinetic Monte Carlo simulations, J. Controlled Release 111 (2006) 117–127. [57] S.R. Van Tomme, W.E. Hennink, Biodegradable dextran hydrogels for protein delivery applications, Exp. Rev. Med. Devices 4 (2007) 147–164. [58] R. Luo, B. Neu, S.S. Venkatraman, Biodegradable dextran hydrogels for protein delivery applications, Small 8 (2012) 2585–2594. [59] X. Huang, C.S. Brazel, On the importance and mechanisms of burst release in matrix-controlled drug delivery systems, J. Controlled Release 73 (2001) 121–136. [60] I. Altimari, U.G. Spizzirri, F. Iemma, M. Curcio, F. Puoci, N. Picci, pH‐sensitive drug delivery systems by radical polymerization of gelatin derivatives, J. Appl. Polym. Sci. 125 (2012) 3006–3013. [61] S.V. Ghugare, E. Chiessi, R. Fink, Y. Gerelli, A. Scotti, Structural investigation on thermoresponsive PVA/Poly(methacrylate-co-N-isopropylacrylamide) microgels across the volume phase transition, Macromolecules 44 (2011) 4470–4478. [62] S. Maccarrone, C. Scherzinger, O. Holderer, P. Lindner, M. Sharp, W. Richtering, D. Richter, Cononsolvency effects on the structure and dynamics of microgels, Macromolecules 47 (2014) 5982–5988. [63] F. Schneider, A. Balaceanu, A. Feoktystov, A. Pipich, W. Vitaliy, Y. Wu, J. Allgaier, W. Pyckhout-Hintzen, A. Pich, G.J. Schneider, Monitoring the internal structure of poly(N-vinylcaprolactam) microgels with variable cross-link concentration, Langmuir 30 (2014) 15317–15326. [64] L. Etchenausia, E. Deniau, A. Brûlet, J. Forcada, M. Save, cationic thermoresponsive poly(N-vinylcaprolactam) microgels synthesized by emulsion polymerization using a reactive cationic macro-RAFT agent, Macromolecules 51 (2018) 2551–2563. [65] M. Boularas, S. Radji, E. Gombart, J.-F. Tranchant, V. Alard, L. Billon, Functional film by trigger-free self-assembly of adhesive soft microgels at skin temperature, Mater. Des. 147 (2018) 19–27.

stimuli-responsive oligo(ethylene glycol)-based microgels: insight into the role of internal structure in volume phase transitions and loading of magnetic nanoparticles to design stable thermoresponsive hybrid microgels, Polym. Chem. 7 (2016) 350–363. M. Boularas, E. Gombart, J.F. Tranchant, L. Billon, M. Save, Design of smart oligo (ethylene glycol)‐based biocompatible hybrid microgels loaded with magnetic nanoparticles, Macromol. Rapid Commun. 36 (2015) 79–83. G. Aguirre, A. Khoukh, K. Chougrani, V. Alard, L. Billon, Dual-responsive biocompatible microgels as high loaded cargo: understanding of encapsulation/release driving forces by NMR NOESY, Polym. Chem. 9 (2018) 757–768. A. Pikabea, G. Aguirre, J.I. Miranda, J. Ramos, J. Forcada, Understanding of nanogels swelling behavior through a deep insight into their morphology, J. Polym. Sci. Part A: Polym. Chem. 53 (2015) 2017–2025. R. Acciaro, T. Gilanyi, I. Varga, Preparation of monodisperse poly(N-isopropylacrylamide) microgel particles with homogenous cross-link density distribution, Langmuir 27 (2011) 7917–7925. A. Imaz, J. Forcada, Optimized buffered polymerizations to produce N-vinylcaprolactam-based microgels, Eur. Polym. J. 45 (2009) 3164–3175. T. Still, K. Chen, A.M. Alsayed, K.B. Aptomicz, A.C. Yodh, Synthesis of micrometersize poly(N-isopropylacrylamide) microgel particles with homogeneous crosslinker density and diameter control, J. Colloid Interface Sci. 405 (2013) 96–102. N. Sanson, J. Rieger, Synthesis of nanogels/microgels by conventional and controlled radical crosslinking copolymerization, Polym. Chem. 1 (2010) 965–977. J. Clara-Rahola, A. Fernandez-Nieves, B. Sierra-Martin, A.B. South, A.L. Lyon, Structural properties of thermoresponsive poly(N-isopropylacrylamide)-poly(ethyleneglycol) microgels, J. Chem. Phys. 136 (2012) 214903. A. Balaceanu, D.E. Demco, M. Moller, A. Pich, Microgel heterogeneous morphology reflected in temperature-induced volume transition and 1H high-resolution transverse relaxation NMR. The case of poly(N-vinylcaprolactam) microgel, Macromolecules 44 (2011) 2161–2169. S. Schachschal, A. Balaceanu, C. Melian, D.E. Demco, T. Eckert, W. Richtering, A. Pich, Polyampholyte microgels with anionic core and cationic Shell, Macromolecules 43 (2010) 4331–4339. B.R. Saunders, On the structure of poly(N-isopropylacrylamide) microgel particles, Langmuir 20 (2004) 3925–3932. M. Stieger, W. Richtering, J.S. Pedersen, P. Linder, Small-angle neutron scattering study of structural changes in temperature sensitive microgel colloids, J. Chem. Phys. 120 (2004) 6197–6207. K. Gawlitza, A. Radulescu, R. von Klitzing, S. Wellert, On the structure of biocompatible, thermoresponsive poly(ethylene glycol) microgels, Polymer 55 (2014) 6717–6724. A. Brulet, D. Lairez, A. Lapp, J.P. Cotton, Improvement of data treatment in smallangle neutron scattering, J. Appl. Crystallogr. 40 (2007) 165–177. M. Stieger, J.S. Pedersen, P. Lindner, W. Richtering, Are thermoresponsive microgels model systems for concentrated colloidal suspensions? A rheology and smallangle neutron scattering study, Langumir 20 (2004) 7283–7292. A. Guinier, G. Fournet, Small-Angle Scattering of X-Rays, John Wiley and Sons, New York, 1995. D. Wu, A. Chen, C.S. Johnson, An improved diffusion-ordered spectroscopy experiment incorporating bipolar-gradient pulses, J. Magn. Reson. 115 (1995) 260–264. C.S. Johnson, Diffusion ordered nuclear magnetic resonance spectroscopy: principles and applications, J. Prog. NMR Spectrosc. 34 (1999) 203–256. N.A. Peppas, J.J. Sahlin, A simple equation for the description of solute release. III. Coupling of diffusion and relaxation, Int. J. Pharm. 57 (1989) 169–172. G. Aguirre, J. Ramos, J. Forcada, Synthesis of new enzymatically degradable thermo-responsive nanogels, Soft Matter 9 (2013) 261–270. A. Balaceanu, D.E. Demco, M. Moller, A. Pich, Microgel heterogeneous morphology reflected in temperature-induced volume transition and 1H high-resolution transverse relaxation NMR. The case of poly(N-vinylcaprolactam) microgel, Macromolecules 44 (2011) 2161–2169.

453