Laura
Rodriguez-Arco
*ab,
Ismael A.
Rodriguez
bc,
Victor
Carriel
bc,
Ana B.
Bonhome-Espinosa
ab,
Fernando
Campos
bc,
Pavel
Kuzhir
d,
Juan D. G.
Duran
ab and
Modesto T.
Lopez-Lopez
*ab
aDepartment of Applied Physics, University of Granada, Faculty of Science, Campus de Fuentenueva, 18071 Granada, Spain. E-mail: l_rodriguezarco@ugr.es; modesto@ugr.es
bInstituto de Investigación Biosanitaria ibs.GRANADA, Granada, Spain
cDepartment of Histology (Tissue Engineering Group), University of Granada, Faculty of Medicine, Avenida de la Investigación, 11, 18016 Granada, Spain
dLaboratory of Condensed Matter Physics, UMR No. 7336, University of Nice–Sophia Antipolis, CNRS, 28 Avenue Joseph Vallot, 06100 Nice, France
First published on 11th March 2016
The inclusion of magnetic nanoparticles into biopolymer matrixes enables the preparation of magnetic field-responsive engineered tissues. Here we describe a synthetic route to prepare biocompatible core–shell nanostructures consisting of a polymeric core and a magnetic shell, which are used for this purpose. We show that using a core–shell architecture is doubly advantageous. First, gravitational settling for core–shell nanocomposites is slower because of the reduction of the composite average density connected to the light polymer core. Second, the magnetic response of core–shell nanocomposites can be tuned by changing the thickness of the magnetic layer. The incorporation of the composites into biopolymer hydrogels containing cells results in magnetic field-responsive engineered tissues whose mechanical properties can be controlled by external magnetic forces. Indeed, we obtain a significant increase of the viscoelastic moduli of the engineered tissues when exposed to an external magnetic field. Because the composites are functionalized with polyethylene glycol, the prepared bio-artificial tissue-like constructs also display excellent ex vivo cell viability and proliferation. When implanted in vivo, the engineered tissues show good biocompatibility and outstanding interaction with the host tissue. Actually, they only cause a localized transitory inflammatory reaction at the implantation site, without any effect on other organs. Altogether, our results suggest that the inclusion of magnetic core–shell nanocomposites into biomaterials would enable tissue engineering of artificial substitutes whose mechanical properties could be tuned to match those of the potential target tissue. In a wider perspective, the good biocompatibility and magnetic behavior of the composites could be beneficial for many other applications.
In the particular case of biomedical applications, a precise temporospatial control of the nanocomposites in a minimally invasive way is tremendously advantageous. In this regard, magnetic nanoparticles (MNPs) can be integrated into the nanostructures to provide this feature, located either at the core of the core–shell architecture or forming the external shell.6,7 Because of their magnetic nature, MNPs can be guided by non-contact forces (i.e., external magnetic field gradients), and allow, at the same time, in situ monitoring by magnetic resonance imaging (MRI) or computerized axial tomography scanning (CT). Furthermore, their particulate character and high surface area make it possible to bind molecules such as tissue or tumor-specific antibodies, drugs, diagnostic molecules, growth factors, peptides, etc.7–9 As for tissue engineering applications, MNPs have been previously dispersed in biopolymer matrixes to synthesize innovative artificial magnetic scaffolds without affecting cell adhesion, proliferation or differentiation.10–26 Such artificial tissues also possess the unique feature of being effectively magnetized by the application of external fields, and thus, they may attract functionalized MNPs carrying growth factors, drugs or cells.11,17,22
In all these previous studies reporting magnetic scaffolds, the size of the MNPs was around 10 nm. Nevertheless, the use of MNPs larger than 50–100 nm has additional advantages from the magnetic point of view, because the magnetic interaction energy between large MNPs dominates over Brownian motion.27 Actually, the mechanical properties of dispersions of large MNPs typically used in materials science (e.g., suspensions, gels, foams) can be changed in a reversible way by the application of an external magnetic field.27–29 In a recent work we have explored such a phenomenon in the case of novel magnetic substitutes generated by tissue engineering consisting of solid MNPs of around 100 nm, to show that it is possible to tune, in a reversible way, their mechanical response by external magnetic forces.30
However, the colloidal stability of large magnetic particles is usually rather poor because they tend to settle due to gravitational forces. Such a problem may be partially overcome if magnetic core–shell nanocomposites are employed instead. Indeed, if the material of the core has low density (e.g., polymeric core), gravitational settling is much slower because of the reduction of the composite average density. In addition, the distribution of the MNPs around a non-magnetic core also has advantages in terms of the magnetic response of the composite particle. Indeed, in a previous study we have shown that for a given volume fraction of magnetic material, such a distribution leads to higher magnetic susceptibility at low to medium magnetic fields than in the case of solid particles, which means stronger responses to external fields.31
In this work we report new biocompatible magnetic core–shell nanostructures (diameter ∼500 nm) which are tested for the generation of magnetic bio-artificial tissue-like constructs. These nanocomposites consist of a polymeric core and a shell of MNPs. In addition, biocompatibility is achieved by an additional layer of polyethylene glycol. We characterize the composites from the physicochemical point of view (morphology, composition and magnetic properties) and compare them with solid particles (i.e., no core–shell structure). We also analyze their ex vivo and in vivo biocompatibility and use them to generate engineered magnetic tissues, which also exhibit an active cell proliferation and good in vivo biocompatibility. We finally evaluate the mechanical properties of the prepared engineered tissues and test if it is possible to tune them by the application of external magnetic fields. In addition, we compare the mechanical properties with those of the engineered tissues prepared with solid particles of our previous work,30 to show that using a core–shell architecture is highly advantageous to obtain stronger magnetic field-responsiveness in these novel biomaterials.
Fig. 1 Sketch of the procedure for magnetic core–shell nanostructure synthesis. In a first step, spheres of acrylic copolymers functionalized with COO− groups (Poly) were used as cores for electrostatic adsorption of Fe2+ and Fe3+ cations, and subsequent surface deposition of MNPs via co-precipitation under alkaline conditions. The resulting nanocomposites (Poly@Mag) were covered with polyethylene glycol (PEG) in a second step, using a previous procedure35 based on emulsion formation and yielding Poly@Mag@PEG composites. |
Stage (i) of the synthetic procedure resulted in the polymer cores fully and homogeneously coated with MNPs. More specifically, the smooth surface of the core (Fig. 2a and b) appeared completely covered with MNPs of a diameter of 10–50 nm as shown by scanning (SEM) and transmission (TEM) electron microscopy (see Fig. 2c and S1 of the ESI,† and 2d respectively). Energy-dispersive X-ray spectroscopy (EDX) revealed iron as the predominant material of the outer shell (not shown). The Fourier transform infrared spectroscopy (FTIR) spectra of Poly@Mag composites also exhibited the characteristic Fe–O band at 590–630 cm−1 confirming the presence of iron oxide in the nanocomposites (see Fig. S2 of the ESI†).36 Similarly, thermogravimetric analysis (TGA) showed a residual weight of inorganic material at high temperature associated with the MNPs of the shell (see Fig. S3 of the ESI†).
The obtained core–shell structures were uniform in size. The average diameter varied depending on the employed concentration of iron precursors with respect to polymer cores: higher concentrations of precursors yielded thicker MNP coatings. For example, doubling the concentration of iron cations from [Fe3+] = 0.3 mM and [Fe2+] = 0.6 mM to [Fe3+] = 0.6 mM and [Fe2+] = 1.2 mM resulted in an increase of Poly@Mag diameter from 830 ± 100 nm to 950 ± 70 nm (Fig. 2e). This result confirms the possibility of tuning the thickness of the magnetic shell, as shown in previous studies.37,38 However, a further double increase of the concentration did not yield a higher diameter, but a rather broader size distribution (Fig. 2e).
The synthesized composites were ferromagnetic with a tunable response to the external magnetic field. The co-precipitation reaction yielded magnetite (Fe3O4), which likely oxidized to maghemite (γ-Fe2O3) in the presence of oxygen,39 because the X-ray diffraction (XRD) spectrum of the nanocomposites could be fitted by both patterns (see Fig. S4 of the ESI†). The saturation magnetization of γ-Fe2O3 and Fe3O4 is similar (380 and 480 kA m−1 respectively).40 The saturation magnetization of our composites was as high as Ms = 150 kA m−1 for the samples with a higher content of the magnetic material (40 vol%) (Fig. 3a). The magnetization curves were reversible with almost negligible values of coercivity, remanence, or hysteresis, suggesting nearly superparamagnetic behaviour (Fig. 3a).
Fig. 3 Special features of magnetic core–shell nanostructures. (a) Magnetization, M, plotted against the magnetic field, H, for core–shell composites with different concentrations of the magnetic material, ϕ (indicated). As observed, samples with higher magnetic contents displayed higher values of the saturation magnetization. (b) Normalized magnetization (M/Ms) vs. the magnetic field, H, for core–shell nanocomposites of different diameters (indicated). Thicker magnetic coatings gave rise to lower values of the magnetic susceptibility at low-medium fields (slope of the curve). The susceptibility for solid magnetite nanoparticles with diameter of the same order of magnitude was even lower (reprinted from S. A. Gómez-Lopera, R. C. Plaza and A. V. Delgado, Synthesis and Characterization of Spherical Magnetite/Biodegradable Polymer Composite Particles, J. Colloid Interface Sci., 240, 40–47, Copyright (2001), with permission from Elsevier43). (c) Absorbance at 550 nm, A, normalized by the value at the beginning of the test, A0, as a function of time for core–shell nanocomposites and for solid particles (adapted with permission from S. A. Gómez-Lopera, J. L. Arias, V. Gallardo and A. V. Delgado, Langmuir, 2006, 22, 2816. Copyright (2006) American Chemical Society44). A significant reduction of the absorbance (i.e., A/A0 = 0.3) was only obtained after 24 h for the core–shell structures, in comparison with the fast reduction observed for solid particles. |
However, the most significant advantage of using core–shell nanostructures with respect to solid magnetic particles is the possibility of customizing the initial magnetic susceptibility of the particles (χi) by changing the thickness of the magnetic shell. Actually, classical electromagnetism predicts higher χi as the thickness of the shell becomes thinner, for a given amount of magnetic material.41,42 To test this for our nanocomposites we first normalized the magnetization of the samples by dividing it by the saturation magnetization (i.e., M/Ms), so we could compare them in conditions of an equal concentration of the magnetic material. We then plotted this normalized magnetization against the magnetic field and compared the initial slope of the curves, (i.e., the magnetic susceptibility). We confirmed that composites with smaller diameters (i.e., thinner magnetic shells) showed higher χi (Fig. 3b). Furthermore, the susceptibilities at low to medium fields of the core–shell nanostructures were always higher than the susceptibility of solid magnetite particles of diameter of the same order of magnitude, previously prepared in our group43 (Fig. 3b).
In addition to higher χi, our nanocomposites showed better stability against gravitational settling than solid particles. Indeed, for Poly@Mag samples, the absorbance (A) was reduced because of settling to a third of its initial value (A0) only after 24 h. For solid magnetite nanoparticles of similar size, such a reduction of absorbance took place in less than 1 h (see Fig. 3c).44 Such a better colloidal stability for our nanocomposites was due to their lower average density because of the lighter polymer core (density ∼1 g cm−3). This feature is highly advantageous for biomedical applications (e.g., preparation of magnetic artificial tissues), for which a good dispersion of the magnetic composites in the continuous medium is essential.
The magnetic nanostructures were successfully coated with PEG (stage (ii) in the sketch of Fig. 1), showing excellent chemical stability and biocompatibility. The successful PEG coating was evidenced by the appearance of new bands in the FTIR spectra corresponding to the in-plane C–H and O–H deformations and of the combination bands of O–C–H and C–O–H (see Fig. S2 of the ESI†). It also gave rise to an additional loss of weight in comparison with Poly@Mag composites in TGA (see Fig. S3 of the ESI†). The PEG shell appeared as a translucent thin film (thickness of a few nanometers) around the magnetic shell in TEM images (Fig. 4a and S5 of the ESI†). In spite of its small thickness, the PEG layer protected the magnetic core from mild acidic attack. Actually, Poly@Mag nanospheres lost their magnetic character after immersion in HCl (0.1 M) for 7 days (Fig. 4b), while Poly@Mag@PEG composites were not significantly affected (Fig. 4c). In addition to better chemical stability, the PEG coating improved the biocompatibility of the composites. Human gingival fibroblasts cultured in the presence of Poly@Mag composites showed substantial morphology alterations. While some of them maintained the typical elongated spindle shape of fibroblasts, others appeared to have a more irregular and rounded morphology (Fig. 4f). These latter results were similar but not comparable to those observed in the negative control group of cells cultured in the presence of malign agents (Fig. 4e). Actually, quantitative analysis of DNA release showed no significant differences (p > 0.05) for the Poly@Mag group as compared to the positive control group (cells growth under normal conditions) (Fig. 4h). These results suggested that Poly@Mag composites just altered cell attachment to the culture flask, but not the cell membrane permeability (cytoplasmic and nuclear), which evidenced the absence of irreversible cell damage. Interestingly, the WST-1 cell viability assay revealed significant differences (p < 0.05) in the cells exposed to Poly@Mag as compared to the positive control group (Fig. 4i). However, both quantitative analyses (DNA analysis and WST-1) showed no significant differences (p > 0.05) between the fibroblasts cultured in the presence of Poly@Mag@PEG particles and those cultured under normal conditions, i.e. positive control (Fig. 4h and i). Actually, these cells preserved the elongated morphology (Fig. 4g) and were almost identical to those of the positive control sample of natural proliferation (Fig. 4d). These results support the high ex vivo biocompatibility associated with the PEG coating, which makes Poly@Mag@PEG composites excellent candidates to be included in engineered tissue substitutes.
Fibrin is a natural polymer frequently used in tissue engineering. Combined with agarose, the biomechanical properties of the resulting engineered tissues are considerably enhanced and match the mechanical response of several native soft human tissues.45–47 Furthermore, fibrin-agarose matrixes have been successfully employed to generate substitutes of human tissues such as the cornea, oral mucosa, skin and peripheral nerves, which proved to be effective in vivo.45,46,48 In our case, the core–shell nanocomposites appeared homogeneously distributed over the network of fibrin fibers as shown by SEM images (Fig. 5a). Interestingly, they frequently acted as connectors between the fibrin fibers (see the inset of Fig. 5a). Very likely, the negatively charged Poly@Mag@PEG nanocomposites at physiological pH (zeta potential of −37.1 ± 0.8 mV at pH = 7.4) attracted the positively charged E domains of fibrin monomers (generated after thrombin cleavage of the fibrinopeptides of fibrinogen).49 The oxygen atoms of the PEG layer, which can play the role of hydrogen bond acceptors,50 may have formed hydrogen bonds with the hydrogen bond donors of fibrin monomers. As a result, fibrin monomers anchored on Poly@Mag@PEG composites, which acted as condensation sites for the subsequent polymerization of fibrin fibers. However, future investigations are needed to confirm these hypotheses.
Cell morphology and proliferation in the magnetic tissue-like constructs were normal as shown by histological analyses. The engineered tissues with H&E showed a random distribution of fibroblasts in the FA matrix (see Fig. 5b and S6 of the ESI†). Interestingly, fibroblasts were found alone or forming small cell clusters (see Fig. S6 of the ESI†), with prominent nuclei in both cases. The nuclei of both individual and cluster-forming fibroblasts showed intense positive immunoreaction for PCNA immunohistochemical analysis (see Fig. 5c and S6 of the ESI†), which confirmed the active proliferation of fibroblasts in the magnetic tissue substitutes. Fibroblasts preserved their characteristic orthotopic elongated shape with large filopodia along the fibrin fibers as shown by SEM (Fig. 5a) and by histological analyses (Fig. 5b–d and S6 of the ESI†), indicating cell-biomaterial interactions. Although homogeneously distributed over the biopolymer, Poly@Mag@PEG composites were frequently observed near the cell surface, around the perinuclear cytoplasm and the large filopodia of the fibroblasts as shown by the intense Prussian blue color of the composites after staining with Perls’ histochemical method (see Fig. 5d and S6 of the ESI†). Ex vivo Live/Dead™ assays revealed a high number of metabolically active cells in the engineered tissues containing Poly@Mag@PEG composites, which was similar to the values observed for the fibrin-agarose control group (cellular construct without composites). In both cases, fluorescence microscopy only showed live, green-stained, cells (dead cells are stained with red color in this technique) as shown in Fig. 5e. Remarkably, there were no statistically significant differences (p > 0.05) between cell viability in the artificial tissues prepared with Poly@Mag@PEG composites (98.3 ± 2.4%) and the fibrin-agarose control (98.1 ± 2.4%) (Fig. 5f), which demonstrated the high ex vivo biocompatibility of Poly@Mag@PEG nanocomposites (Fig. 5f).
The incorporation of magnetic nanocomposites into the artificial tissues transformed them into magnetic field-responsive engineered tissues. Indeed, the engineered tissues prepared with magnetic composites moved when exposed to the magnetic field gradient created by a magnet (see Video 1 of the ESI†). From magnetization measurements (see Fig. S7 of the ESI†) we estimated the volume fraction of the magnetic material in the engineered tissues, which turned out to be ∼0.3 vol%. Such a magnetized state of the engineered tissue would allow attraction of functionalized MNPs carrying growth factors, drugs or cells.11,17,22
In addition to field-responsiveness, the engineered tissues showed magnetic field-tunable mechanical properties. More specifically, we measured an increase of the storage modulus (G′) of the engineered tissue of ∼10% when the intensity of the external field was increased from 0 to 25.6 kA m−1 (Fig. 6a). G′ is connected to the elastic, solid-like, response of the biomaterial when subjected to an oscillatory mechanical stimulus, and therefore, is a measure of its mechanical strength.51 Furthermore, the sole presence of the nanocomposites in the fibrin network already increased the engineered tissue mechanical strength, even without magnetic field application. Actually, we observed a twofold increase of G′ for Poly@Mag@PEG engineered tissues with respect to the FA control at zero field (Fig. 6a). Such an increase was much higher than the one predicted by the classical theory of mechanics of composite materials for a continuous matrix with spherical, completely rigid inclusions,52 and could only be explained by microstructural changes in the pattern of the fibrin-agarose network due to the Poly@Mag@PEG nanocomposites. As seen in Fig. 5a and d, Poly@Mag@PEG composites were homogeneously distributed around the fibrin network and acted as connectors between the fibrin fibers. This may have enhanced adhesion between fibers and have had a failure retardation effect which could explain the higher values of G′ even in the absence of a field.53 Similar increases with the field were obtained for the loss modulus, G′′, related to the dissipation of energy upon the oscillatory mechanical stimulus (see Fig. S8 of the ESI†).51 The increase of the viscoelastic moduli with the magnetic field was maintained over a broad range of frequencies of the oscillatory strain (see Fig. S9 of the ESI†). Although not huge, such an increase proves that it is possible to control the mechanical behavior of these artificial tissues by the application of an external field. Stronger magnetic fields or higher composite loadings would give rise to stronger mechanical changes.
Fig. 6 Mechanical properties of magnetic tissue substitutes. (a) Storage modulus, G′, as a function of the amplitude of the oscillatory strain for four values of the applied magnetic field, H (indicated). The mechanical strength of the engineered tissue increased with the magnetic field as evidenced by the increase of G′. (b) Normalization of the relative increment of G′ with the field, [G′(H) − G′(0)]/G′(0), by the saturation magnetization of the nanocomposites, Ms for engineered tissues consisting of Poly@Mag@PEG composites and solid magnetic particles of similar size (adapted from M. T. Lopez-Lopez, G. Scionti, A. C. Oliveira, J. D. G. Duran, A. Campos, M. Alaminos and I. A. Rodriguez, PLoS One, 2015, 10, e013387830). As observed, the increment with the field is higher for core–shell nanostructures as a result of the enhancement of the initial magnetic susceptibility observed in Fig. 3. |
However, the most interesting result here was the significant enhancement of the mechanical properties when using core–shell magnetic nanostructures in comparison with the solid magnetic particles of similar size reported in ref. 30. Indeed, a normalization of the field-induced increase of G′ by the engineered tissue magnetic content, revealed a stronger magnetic response for Poly@Mag@PEG nanocomposites (Fig. 6b). Such an enhancement would be connected to the improvement of the magnetic susceptibility in core–shell nanostructures observed in Fig. 3. This latter result confirms the excellent suitability of these composites to be used in the preparation of novel magnetic tissue substitutes.
The subcutaneously injected Poly@Mag@PEG nanocomposites remained in the interscapulum during the duration of the experiment (i.e., 21 days), without migration to the distal organs of the body. The injected nanocomposites formed an irregular and dense mass consisting of nanocomposite aggregates (Fig. 7b–g). The host response around this mass was a moderate acute inflammatory reaction mainly composed of neutrophils, some mononuclear cells and predominantly macrophages (Fig. 7b–g). The inflammatory reaction was more evident after the first week, but it progressively decreased from the second to the third weeks. The inflammatory response was only localized around the region where Poly@Mag@PEG composites were injected. Cells progressively reabsorbed the mass starting from the external part to the inner part. Perls’ staining confirmed the presence of phagosomes with Poly@Mag@PEG composites inside the macrophages (Fig. 7d and f) and the picrosirius method revealed a progressive encapsulation of the composite mass by a collagen-rich extracellular matrix and a blood vessel network (Fig. 7c and e). The mass was not fully reabsorbed after 21 days. The histological and histochemical analyses of the liver, kidney, spleen and lungs did not show any inflammatory reaction. There were no macrophages with Poly@Mag@PEG composites inside either. Indeed, all organs were histologically normal during the 21 day follow-up period (Fig. 7n–q). Tissue samples taken from these organs did not exhibit any magnetic response, in contrast to the samples obtained from the interscapular region, which were magnetic because of the presence of the nanocomposites (see Table S1 of the ESI†).
The histological analysis of the magnetic tissue-like constructs evidenced their successful subcutaneous implantation. The constructs showed a regular and compacted morphology (Fig. 7h–m). The nanocomposites appeared homogeneously distributed over the thin FA hydrogel network, either individually or forming small aggregates, in contrast to the bigger aggregates found when the nanocomposites were injected in suspension (insets in Fig. 7c and i respectively). Therefore, the FA hydrogel prevented from the aggregation of the nanocomposites and favored their consistent and homogeneous individual distribution. The use of the FA hydrogel also eased the interaction with the host tissue: host cells were able to invade the implanted constructs from the first week (Fig. 7i and inset), in contrast to the few cells observed in the case of the injected nanocomposite suspension (Fig. 7c and inset). The constructs were not fully reabsorbed after 21 days. The host response to the constructs was a mild to moderate acute inflammatory reaction which progressively decreased over time (Fig. 7h–m). As in the case of the suspension of Poly@Mag@PEG composites, a thin connective tissue capsule composed of collagen fibers and a vascular network was formed around the implants (Fig. 7k and m). The moderate local acute inflammatory reaction did not affect the surrounding connective tissues and distal organs (Fig. 7h–m and r–u). Similarly to what happened to the nanocomposites in suspension, only the samples from the interscapulum showed a magnetic response (see Table S1 of the ESI†), which indicated no migration of the Poly@Mag@PEG composites to other organs during the duration of the experiment.
The prepared composites were uniform in size, the thickness of the magnetic shell increasing with the concentration of iron precursors in the co-precipitation reaction, which allowed tuning of the magnetic properties. The core–shell architecture improved the magnetic response of the composites with respect to solid magnetic particles of similar size. More specifically, the magnetic susceptibility at low to medium fields was higher, and increased when the magnetic shell became thinner. In addition to enhanced magnetic properties, core–shell nanocomposites exhibited remarkably good stability against gravitational settling because of the lower density of the polymer core.
Because of the PEG outer layer, the nanocomposites exhibited excellent chemical stability and biocompatibility (i.e., no significant cell damage and good cell growth and attachment). For this reason, the composites were successfully loaded into fibrin-agarose hydrogels containing human oral mucosa fibroblasts to obtain magnetic field-responsive engineered tissues with metabolically active cells. The fibrin-agarose matrix promoted the interaction with the host tissue when the magnetic tissue-like constructs were implanted in vivo. It also prevented colloidal aggregation of the nanocomposites, in contrast to the situation when the nanocomposites were subcutaneously injected in suspension. In both cases (i.e., implantation and injection), there was only a localized and transitory acute inflammatory reaction, without affecting the distal organs. Therefore, both the nanocomposites and the magnetic engineered tissues showed good in vivo biocompatibility. After 21 days, the nanocomposites and the constructs were not fully reabsorbed and still maintained their magnetic response. Remarkably, we measured an enhancement of the artificial tissue mechanical properties when exposed to an external magnetic field. This latter result evidences the suitability of these composites for being used in the preparation of new engineered magnetic tissues whose mechanical properties can be controlled by external, non-contact, forces. Future work will focus on the functionalization of the nanocomposites for specific purposes in regenerative medicine and tissue engineering. The magnetic tissue-like constructs could be used for cartilage tissue engineering (subjected to strong mechanical forces at the joint surface) and for the generation of biodegradable, functionalized, and mechanically stable tubes for peripheral nerve repair. However, the use of these nanocomposites should not only be restricted to tissue engineering, but could also be extended to other applications such as smart magnetic materials, biosensing and bioseparation, MRI, drug delivery or hyperthermia. In addition, the inner core may provide functionalities different than those of the shell (e.g., attachment of fluorescent molecules, proteins or drugs), which can be beneficial for some of these applications.
The experimental procedure for preparing Poly@Mag particles was as follows: 50 mg of PolymP–H particles were dispersed in 3 mL of a 0.1 M aqueous solution of NH4OH to promote deprotonation of the particle carboxylic groups and to charge the surface of the particles negatively. After this step, solutions of FeCl3·6H2O and FeSO4·7H2O with molar ratio [Fe3+]/[Fe2+] = 0.5 were added, and the mixture was mechanically stirred. We tested different concentrations of iron ions to determine how they affected the final thickness of the magnetic layer. To allow co-precipitation of the iron hydroxides, a solution of NH4OH (28%) was added until pH = 12 (a black precipitate appeared). The mixture was aged for 1 hour under strong mechanical stirring, followed by heating up to 95 °C to promote magnetite formation. After cooling, the black precipitate was magnetically separated and washed with distilled water until neutral pH.
We report here the mean values ± standard deviations of 8 independent experiments for each experimental group and each analysis. The Kruskal–Wallis test was used to identify statistical differences among the study groups, and the Mann–Whitney test was used to identify significant differences between two groups. Values of p less than 0.05 were considered statistically significant in two-tailed tests.
We conducted two types of oscillatory shear tests. (i) Amplitude sweep tests: we fixed the frequency, f = 1 Hz, of the oscillatory strain, γ = γ0cos(2πft) and increased the strain amplitude, γ0, in a logarithmic ramp. The sinusoidal strain at each step was applied over 8 periods of oscillations. We recorded the resulting viscoelastic moduli (i.e., the storage, G′, and loss, G′′, moduli) over the last 5 periods to discard transients. (ii) Frequency sweep tests: we fixed γ0 at a value belonging to the viscoelastic linear region (i.e., the region for which the viscoelastic moduli are independent of γ0) and varied f. Again, we maintained the oscillatory strain for 8 periods and recorded the last 5. The obtained values for all the quantities in this work are the average of at least 3 repetitions.
(i) in vivo kinetic evaluation of Poly@Mag@PEG nanocomposites (S-PMPc). We evaluated the time-dependent biodistribution and biocompatibility of the Poly@Mag@PEG composites by subcutaneously injecting 500 μl of a sterile physiological suspension of the composites of concentration 11 mg mL−1 in the interscapulum of each mouse.
(ii) In vivo kinetic evaluation of engineered magnetic tissue (FA-PMPc). We evaluated the time-dependent biocompatibility of the engineered magnetic tissues by a subcutaneous implantation of a 500 μl construct (concentration of 11 mg mL−1) in the interscapulum.
(iii) Control group (CTR) of healthy animals without any surgical intervention.
After injection or subcutaneous implantation of the constructs, all the animals were housed in a temperature-controlled room (21 ± 1 °C), provided with a 12 h light/dark cycle and ad libitum access to tap water and standard mice chow.
The in vivo histological biocompatibility was studied after 7, 14 and 21 days (3 mice of each group were analyzed each week, n = 3). Animals of each experimental group were euthanized by cervical dislocation and the interscapulum (skin with hypodermis), liver, kidneys, spleen and lungs were harvested for tissue processing and histological analyses.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6nr00224b |
This journal is © The Royal Society of Chemistry 2016 |