Fractal evolution of dual pH- and temperature-responsive P(NIPAM-co-AA)@BMMs with bimodal mesoporous silica core and coated-copolymer shell during drug delivery procedure via SAXS characterization

2.1 Materials

The cetyltrimethyl-ammonium bromide (CTAB, A. R.), potassium persulfate (KPS, A. R.), tetraethyl orthosilicate (TEOS, A. R.), ammonium nitrate (NH₄NO₃, A. R.), sodiumdodecyl sulfate (SDS, A. R.), acrylic acid (AA, A. R.), IBU (A. R.) and hexane (A. R.), were obtained from Sinopharm Chemical Reagent Co. Ltd. Alfa Aesar company supplied other chemicals, including monomer N-isopropylacrylamide (NIPAM, A. R., 97%), N,N-methylenebisacrylamide (BIS, A. R., 97%), 3-methacryloxypropyltri-ethoxysilane (MPS, A. R., 98%). Before used, NIPAM and BIS were re-crystallized from hexane and methanol. Ammonia (NH₃·H₂O, A. R., 28%), anhydrous ethanol (C₂H₅OH, A. R., 99.7%), phosphoric acid, disodium hydrogen phosphate, sodium dihydrogen phosphate were purchased from Beijing chemical works. The deionized water used in the experiment was self-made in our laboratory.

2.2 Preparation and surface modification of BMMs-MPS nanoparticles

BMMs containing CTAB (t-BMMs) was synthesized partially on the basis of reported method (Sun et al., 2003). The synthesis procedure was as follows: The mixtures of 20.8 g of CTAB and 832 mL of deionized water was stirred well, then, 64 mL of TEOS was added dropwise, but 19.2 mL of ammonia was added quickly. The mixture was stirred continuously and became progressively white gel. After filtered, washed, and dried at 120 °C for 3 h, the resulting precipitates were denoted as t-BMMs. In order to modify the silane coupling agent of MPS on the outer surface of the BMMs, the MPS was used prior to remove the CTAB surfactant. BMMs-MPS was synthesized according to the literature (Jin et al., 2017, 2018, 2019). The obtained t-BMMs (0.1 g) was dispersed into the mixtures of 200 mL anhydrous ethanol and 1 mL MPS, which kept stirring for 4 h at 60 °C for 4 h. Then, the modified t-BMMs was collected after repeated washing with ethanol and drying for 12 h under vacuum. Finally, the CTAB was removed from the mesopores according to the method reported in the literature (Jin et al., 2017, 2018, 2019), the detailed process was as follows: 0.5 g of above product was refluxed for 2 h in 100 mL ethanol solution containing 0.6 g NH₄NO₃. After repeated for 3 times, then filtered, washed thoroughly with anhydrous ethanol, and dried at 60 °C under vacuum. The vinyl-modified materials (BMMs-MPS) were obtained.

2.3 Synthesis of P(NIPAM-co-AA)@BMMs composites

The synthesis of P(NIPAM-co-AA)@BMMs was carried out according to the method reported in the literature (Jin et al., 2017, 2018, 2019). The details of synthesis procedure of P(NIPAM-co-AA)@BMMs were as follows: dispersed 0.2 g of BMMs-MPS in 50 mL deionized water containing in 100 mL three-neck flask equipped with reflux under nitrogen protection, then NIPAM (50 mg), AA (3.3 μL), BIS (3.5 mg) and SDS (4.4 mg) were added into the mixture solution and stirred for 1 h at 70 °C. After the air was exhausted and nitrogen was passed, the initiator KPS was quickly added to the suspension. After the polymerization reaction was completed for a certain time, the composite nanoparticles were obtained by centrifugation, washing with deionized water, drying at 50 °C for 12 h, denoted as P@BMMs(n), where n represents the coating time (h) during polymerization, and is 1, 2, 4, and 7, respectively.

2.4 IBU adsorption

The experimental procedure of drug loading of P@BMMs was studied in the literature (Jin et al., 2017, 2018, 2019). The details of procedure were as follows: P@BMMs(n) (0.1 g) was dispersed in 10 mL solution of IBU hexane (40 mg/mL), and continuously stirred at 25 °C for 48 h. After that filtered, washed repeatedly with hexane, and dried in vacuum, the IBU-loaded P@BMMs(n) was denoted as I/P@BMMs(n)-48.

In order to measure the amount of drug loaded, the filtrate was collected and then diluted to 100 mL with hexane, whereas concentration was analyzed on a high performance liquid chromatography (HPLC). The corresponding calibrated curve of IBU were shown in Fig. S1 of Electronic Supporting Information (ESI) section, showing a good linear relationship (y = 0.0027x + 0.038) and high correlation coefficient (R² > 0.9995), in which, Y-axis represents their concentrations and X-axis represents peak areas deriving from HPLC chromatograms. Whereas, the peak area was integral value of signal value output by the detector and retention time when the maximum concentration of IBU was eluted from the chromatographic column. In this regard, the unknown concentration of the IBU-contained solution can be obtained by testing the peak area in the HPLC chromatogram. The similar determination method was also reported (Shan et al., 2013).

The drug loading capacity was calculated by the following equation (Jin et al., 2017): $$\mathit{LC}\text{\%}=\frac{m_0-m_1}{m_2}\times 100\text{\%}$$ where m₀ is the initial weight of IBU, m₁ is the weight of IBU left in the medium after equilibrium, m₂ is the weight of I/P@BMMs(n)-48.

2.5 In vitro release

In vitro drug release procedure was similar to that of our previous report (Jin et al., 2017, 2018, 2019). In details, firstly pressed I/P@BMMs(n)-48 into slice and placed in a dialysis bag with a molecular weight cut off 3500 Da, which was immersed in 100 mL erlenmeyer flask containing PBS solution (pH 2.0, or 7.4) with an ionic strength of 0.02 M at 25 °C or 37 °C, respectively. After a certain time interval, 1 mL of the liquid was taken out of the release solution, at the same time 1 mL of fresh PBS solution was added. The sample obtained after release for 24 h was named as I/P@BMMs(n)-48-24.

The IBU concentration of released solution was analyzed by UV–visible spectrophotometer at the wavelength of 272 nm, which were performed three times. The corresponding calibrated curve of IBU were shown in Fig. S2 of ESI section, showing a good linear relationship (y = 1.42671x + 0.0047) and high correlation coefficient (R² > 0.9997) in pH of 2.0, a good linear relationship (y = 1.44173x + 0.00769) and high correlation coefficient (R² > 0.9996) in pH of 7.4.

The drug release percent was calculated using the following formula (Jin et al., 2017): $$C_c=C_t+\frac{v_1}{v_2}\sum _0^{t-1}{C}_i$$ where C_c is the corrected concentration at time t, C_t is the apparent concentration at time t, V₁ is the volume of sample collected, and V₂ is the total volume of released fluids.

Based on the mentioned-above introductions, the drug-releasing percent (%) could be calculated by formula as follow: $$\mathit{Drug}-releasingpercent(\%)=\frac{C_c\times V}{\mathit{wt}\%\times m}\times 100\%$$ where C_c is the corrected concentration at time t, V is the volume of release system, wt% is the amount of IBU loaded, and m is the weight of I/P@BMMs(n)-48.

2.6 Characterizations

The SEM images were obtained using a Hitachi S-4300 electron microscope at an acceleration voltage of 15 kV. The TEM micrograph was recorded on a JEM-2100 at 200 kV. UV–vis absorbance spectra were obtained from Shimadzu UV-2600 spectrophotometer. The XRD pattern was measured by a D8 ADVANCE X-ray powder diffractometer using Cu Kα radiation (λ = 0.154 nm) source in 2θ ranges from 1.0 to 10.0° with a scanning speed of 1.0°/min at the tube voltage of 36 kV and the tube current of 20 mA. The FT-IR spectra were measured on Tensor-27 Fourier infrared spectrometer using a scanned range of 4000–400 cm⁻¹ and 4 cm⁻¹ resolutions and sample was pressed with potassium bromide to a disc prior to measurement. The particle size was measured from Marvin Zetasizer Nano ZS instrument. TGA profiles were measured from Perkin-Elmer STA 8000 instrument in the range of 50–800 °C under air atmosphere with the heating rate of 10 °C/min⁻¹. The N₂ adsorption-desorption isotherms of all the samples were determined by JW-BK 300 specific surface and pore size analyzer, and the isotherm data was analyzed by BET (Brunauer-Emment-Teller) and the pore size distributions were calculated using adsorption branches of the isotherm by the Barrett-Joyner-Halenda (BJH) method. The HPLC was used for analysis the drug loading content of samples by Agilent Technologies 1200 series.

The SAXS profiles were measured at Beamline 1W2A at the Beijing Synchrotron Radiation Facility. The wavelength of the instrument was kept to 0.154 nm, and the SAXS detector (Mar 165 CCD) was set to 1.55 m distance from the sample with the experimental condition is under room temperature (22 °C). The powder samples were supported on an adhesive tape for measurement which was used for background correction for the samples. The nanoparticles structure information was obtained from fractal dimension data and the scattering of fractals (mass or surface) obeyed the power law.

According to Hajji’ demonstration (Hajji et al., 2015), the fractal dimension belongs to mass fractal (D_m) in the range of 0 < α < 3, and its value is equal to α, while the surface fractal (D_s) in the range of from 3 to 4, D_s value = 6 − α (D_s ≠ 3). Recently, other methods of fractal calculation have been further developed (Cattani, 2018; Jafari et al., 2018; Liu and Zhang, 2018; Maitama and Zhao, 2019; Prakash and Kaur, 2018; Singh et al., 2016; Yang and Machado, 2017c; Yang et al., 2015, 2016a, 2016b, 2017a, 2017b, 2017d, 2018).

The particle size was estimated originating from distance distribution function P(r), as follows (Glatter, 1977):

On the basis of small angle scattering theory and Porod law, the relation between the characteristic function of a particle γ(r) and the scattering function is given by the following equation:

The distance distribution function P(r) was obtained by multiplying r²:

Essential information on the shape of the particle can be determined from the P(r) function. According to Glatter’ demonstration (Glatter, 1977), P(r) represents the distance distribution function, which is a numerical value, but does not have special statistical significance. The function P(r) thus differs from zero only in a certain limited region of real space, namely between 0 and r, where r corresponds to the maximum distance in the particle. Therefore, the r value for spherical nanocomposite in this work can be defined as the diameters of the particles.

2.7 Cytotoxicity measurement and pharmacokinetic experiment in mice

3-(4,5-Dimethyl-2-thiazoly) 2,5-diphenyl-2-H-tetrazolium bromide (MTT) assay was performed to evaluate cell viability by treating I/P@BMMs(7)-48 with conventional MTT protocols. HepG-2 cells (5 × 104 cells/well) were first incubated for 24 h in 96-well plate, and then 100 μL of I/P@BMMs(7)-48’s concentration in the range of 0.2–100 μg/mL was added. After incubation for 48 h, the medium was discarded and replaced by another medium containing DMSO (100 μL/well). The cell viability was determined using plate reader at 490 nm.

BALB/c mice were supplied by the Laboratory Animal Service Center, The Fourth Military Medical University. All animal experiments were approved by the Animal Experiment Administration Committee of the Fourth Military Medical University. Animals were housed under standard conditions of temperature, humidity and light. Food and water were provided ad libitum. Mice were randomly divided in 10 groups (n = 5 in each group) and injected intravenously with 0.2 mL I/P@BMMs(7)-48-conatined solution at a concentration of 3 mg/mL. Finally, drug dose was around 600 μg/mice through the tail vein. Blood (∼0.5 mL) was drawn from mice and put into heparinized Eppendorf tubes at 10 different time intervals after injection. The blood sample was centrifuged at 4 °C, and 8000 rpm immediately and 0.3 mL of plasma was harvested. Each plasma sample was mixed with 0.5 mL of ethanol, vortexed, and centrifuged at 15000 rpm for 5 min. The supernatant was collected to measure the drug concentrations using the UV–vis spectrophotometer (Shimadzu UV-2600 spectrophotometer). Standard equation of drug and its linear fitting was established by UV absorption at different drug concentrations. Y = 9.6584x − 0.0068 (R² = 0.9996, linear range: 0.0010–600 μg/mL).

3.1 Structural and textural characterizations

Fig. 1A showed the XRD patterns of before and after drug-loaded samples with different copolymer-coating time interval and drug-loaded amounts (inset). As can be seen, an obvious peak (100) in all samples suggested the presence of highly ordered mesopore structures, while, the diffraction peak of BMMs modified by silane coupling agent shifted from 1.94° of pure BMMs (Fig. 1A-a) to 2.08° of BMMs-MPS (Fig. 1A-b), implying the successful grafting of the organic functional groups on the mesoporous surface of BMMs (Jin et al., 2017). The peak (100) of each mesoporous hybrid sample presented not only decreasing intensity but also shifting from 2.1° of P@BMMs(1) (Fig. 1A-c) to 2.14° of P@BMMs(7) (Fig. 1A-d) with the increasing grafted amount of P(NIPAM-co-AA) copolymer, indicating a slight effect of grafting procedure on the ordered mesopore structures. After IBU-loading, a pounced reduction in peak (100) intensity of I/P@BMMs(n)-48 (Fig. 1A-e, and -f) was observed, implying the successful loading of IBU molecules in mesoporous channels of P@BMMs. The corresponding wide angle XRD patterns were shown in Fig. 1A (inset). As can be seen in Fig. 1A-g (inset), the strong diffraction peaks within 2 theta of 10–50° belonged to the crystalline phase of used IBU. The loaded-IBU samples of both P@BMMs(1)-48 and P@BMMs(7)-48 presented a wide diffraction peak in the range of 20–30°, suggesting the highly uniform distribution of IBU in mesoporous I/P@BMMs(n)-48 (seen in Fig. 1A-e and -f (inset)) (Li et al., 2016; Shen et al., 2011). Unfortunately, as shown in Fig. 1B, the drug-released samples obtained under different conditions, taking I/P@BMMs(7)-48-24 as an example, exhibited no characteristic diffraction peaks, indicating the disappearance of ordered mesopores structures.

Close

Fig. 1

A: XRD patterns of pre and after drug-loaded samples. (a) BMMs, (b) BMMs-MPS, (c) P@BMMs(1), (d) P@BMMs(7), (e) I/P@BMMs(1)-48, and (f) I/P@BMMs(7)-48, The corresponding wide angle XRD patterns of (e) I/P@BMMs(1)-48, (f) I/P@BMMs(7)-48, and (g) IBU (inset); B: XRD patterns of after drug-releasing I/P@BMMs(7)-48-24 under various media. (a) 25 °C, pH 2.0, (b) 25 °C, pH 7.4, (c) 37 °C, pH 2.0, and (d) 37 °C, pH 7.4.

Export to PPT

The SEM and TEM micrographs can provide the morphologies and microstructures of BMMs-MPS, P@BMMs(7), I/P@BMMs(7)-48, and I/P@BMMs(7)-48-24, as imaged in Fig. 2. Firstly, as seen in Fig. 2a and 2e, the BMMs-MPS exhibited a spherical nanoparticle of around 50 nm with the abundant worm-like pore channels of around 3 nm, which were found consistent with our previous report (Gao et al., 2011b). However, the SEM image of P@BMMs(7), as shown in Fig. 2b, presented a spherical morphology, but had a little larger size of around 60 nm than that of pure BMMs-MPS (Fig. 2a). Meanwhile, the TEM image (Fig. 2f) revealed the core (black color) in the center of copolymer shell (gray edges) and small mesopores with uniform size of 3 nm. These observations clearly indicated that the copolymer has been successfully wrapped onto the surface of BMMs, but did not significantly impact on the morphology and mesopore structure of BMMs. On the other hand, the SEM micrograph (Fig. 2c) of IBU-loaded sample (I/P@BMMs(7)-48) showed the spherical morphologies in size of 60 nm, almost same as that of copolymer-coated BMMs (Fig. 2b), Fig. 2g presented that the TEM image of I/P@BMMs(7) also had the core-shell structure. Taking a typical drug-released sample obtained at 25 °C, pH 2 as an example, the spherical morphologies (Fig. 2d) and the core-shell structure of I/P@BMMs(7)-48-24 after IBU releasing still remained intact (Fig. 2h), but the mesoporous structure may be collapsed, or even disappeared to somehow.

Close

Fig. 2

SEM images of (a) BMMs-MPS, (b) P@BMMs(7), (c) I/P@BMMs(7)-48, (d) I/P@BMMs(7)-48-24. The corresponding TEM images of (e) BMMs-MPS, (f) P@BMMs(7), (g) I/P@BMMs(7)-48, and (h) I/P@BMMs(7)-48-24 (the typical drug-released sample, taking I/P@BMMs(7)-48 after IBU releasing at 25 °C, pH 2.0 as example).

Export to PPT

TG profiles of pre and after drug-loaded samples were also carried out (Fig. S3 of ESI section). As can be seen in Fig. S3-A, there were two stages of weight loss procedures: one before 200 °C could be attributed to the removal of moisture absorbed or residual solvent (Jaroniec et al., 1998), another between 200 and 800 °C was assigned to the decomposition of the surface silanol groups and the residual organic components, corresponding to 4.7 wt% of pure BMMs (Fig. S3-A(a)), 16.2 wt% of BMMs-MPS (Fig. S3-A(b)), and 16.3 wt% to 39.4 wt% of P@BMMs(n) (Fig. S3-A(c), (d), (e), and (f). Taking BMMs-MPS as a reference, the copolymer-coated amount was about 0.1%, 4.6%, 11.5%, and 23.2%, respectively. The TG curves of IBU-loaded samples were present in Fig. S3-B, obviously showing three stages of weight loss procedure, which was different from that of copolymer-loaded samples (seen in Fig. S3-A): the weight loss profile for I/P@BMMs(n)-48 (n = 1, 2, 4, and 7, respectively) during the temperature of before 200 °C were corresponding to the removal of adsorbed water, the second one for decompositions of surface-coated copolymer and partial loaded-IBU in the mesoporous BMMs during the temperature of 200–400 °C, corresponding to 7.7, 4.6, 6.8, and 11.8 wt% for first stage, and 7.3, 12.0, 7.1, and 10.7 wt% for second stage, respectively. The last one of around 5.9, 8.7, 6.2, and 10.0 wt% in the temperature range of 400–800 °C was attributed to completely decomposition of organic component including residual copolymer and IBU. Taking P@BMMs(n) as a reference, as shown in Fig. S3-B, the drug loading of I/P@BMMs(n = 1, 2, 4, and 7, respectively)-48 was about 4.5, 4.7, 7.6, and 6.9 wt%, respectively, which proved that the drug loading capacity increased with increasing copolymer-coated time intervals.

The elemental analysis results of pre and after copolymer-coated, as well as IBU-loaded samples were collected in Table S1 of ESI section. It can be seen that the content of C and H in BMMs was 3.43 wt% and 1.01 wt%, respectively, which may be due to the unremoved CTAB completely, while, the content of C and H in BMMs-MPS corresponding to 11.65 wt% and 5.7 wt%, respectively, indicated the successful modification by 3-methacryloxypropyltri-ethoxy groups. After coating by copolymer, the content of C and H in P@BMMs(n = 1, 2, 4, and 7, respectively) obviously increased from 11.67 wt% and 5.83 wt% to 27.36 wt% and 10.82 wt%, respectively. Particularly, the content of N increased from 0 for BMMs-MPS to 4.35 wt% for P@BMMs(7) , but, the molar ratio of N/C kept almost unchanged with ca. 0.12–0.15, suggesting successful coating of copolymer on the mesoporous BMMs. After drug-loading, the content of C in I/P@BMMs (n = 1, 2, 4, and 7, respectively)-48 presented the increased tendency from 10.9 to 20.42 wt%, indicating that IBU was gradually loaded into the mesoporous channels with the increasing amounts of coated-copolymer.

Fig. 3 displayed the N₂ adsorption-desorption isotherms and corresponding pore size distributions of related samples, their textural parameters were collected in Table S2. As shown in Fig. 3A, the isotherms of all samples belonged to type IV, and the appearance of two distinct hysteresis loops in the relative pressures (P/P₀) range of 0.3–0.5 and 0.7–0.9, respectively, verified the existence of bimodal mesopores structures, which are found in good agreement with our previous report (Jin et al., 2017). However, the hysteresis loop in the low pressure region gradually disappeared along with the copolymer-coating time, while the other one in the high pressure region changed unobvious. The corresponding pore distributions (as shown in Fig. 3B-c and -d) exhibited that the bimodal mesopores size of pre and after copolymer-coated samples remained almost intact, similar to that of BMMs (Fig. 3B-a) or BMMs-MPS (Fig. 3B-b), suggesting that P(NIPAM-co-AA) copolymer as a shell was mainly coated on the outer surface of BMMs via polymerization between vinyl functional groups and acylamine groups. However, their specific surface area and pore volume decreased from 1019 cm²/g and 1.64 cm³/g for P@BMMs(1) to 554 cm²/g and 1.12 cm³/g for P@BMMs(7), respectively, as shown in Table S2 of ESI section.

Close

Fig. 3

N₂ adsorption–desorption isotherms (A) and their corresponding pore size distributions (B) of (a) BMMs, (b) BMMs-MPS, (c) P@BMMs(1), (d) P@BMMs(7), (e) I/P@BMMs(1)-48, and (f) I/P@BMMs(7)-48.

Export to PPT

After drug-loading, both the specific surface area and pore volume decreased along with copolymer-coated time from 669 m²/g and 1.29 cm³/g for I/P@BMMs(1)-48, to 385 m²/g and 0.69 cm³/g for I/P@BMMs(7)-48, respectively, while the mean primary pore diameter decreased from 2.54 to 2.27 nm (seen in Table S2). These results clearly indicated the successful IBU-loading into the mesoporous channels of P@BMMs. Interestingly, as can be seen in Fig. 3B-e and -f, the gradual disappearances of larger mesopores for I/P@BMMs(1)-48 and I/P@BMMs(7)-48 again proved that IBU was successfully loaded into the mesopores.

As shown in Fig. S4 of ESI section, FT-IR spectra presented the characteristic information of the functional groups in before and after IBU-loaded samples using IBU as a reference. Firstly, as can be seen in Fig. S4-a for BMMs, the absorption peaks centered at 1080 and 802 cm⁻¹ belonged to asymmetric stretching vibration of Si—O—Si, and another at 952 cm⁻¹ could be attributed to the stretching vibration of Si—OH (Gao et al., 2012). In case of BMMs-MPS (Fig. S4-b), the additional appearance of a vinyl characteristic peak at 1637 cm⁻¹ and —CH₂ stretching peal at 2952 cm⁻¹ suggested the successful modification by vinyl groups (Chang et al., 2013). The spectrum of P@BMMs(7) (Fig. S4-c) exhibited the comparatively new absorption peaks at 1560, 1467, and 1376 cm⁻¹, which were ascribed to the characteristic peaks of copolymer P(NIPAM-co-AA) (Bourgeat-Lami and Lang, 1998; Efthimiadou et al., 2014). Furthermore, I/P@BMMs(7)-48 (Fig. S4-d) showed a strong carboxyl band at 1718 cm⁻¹, indicating the successful IBU-loading in the mesoporous hybrid material (Brás et al., 2014). However, a characteristic peak of IBU (Fig. S4-g) was also observed at 1718 cm⁻¹ in I/P@BMMs(7)-48-24 (Fig. S4-e and -f) after-releasing under different conditions, which confirmed the presence of the residues of IBU in the mesoporous BMMs even after releasing for 24 h.

The SAXS patterns of BMMs, BMMs-MPS and P@BMMs(n) were shown in Fig. 4A, the straight line for each scattering pattern in low region (q < 1) implied the presence of typical fractal features in all samples with a good correlation (correlation coefficient (R²) > 0.99). Fig. 4B showed the effect of copolymer-coated time on the D_m values of hybrid P@BMMs by accompanying with their morphologies (as imaged in inset), it can be seen that the D_m value increased from 2.01 for BMMs to 2.02 for BMMs-MPS, 2.12 for P@BMMs(7), meaning that the copolymer-coated procedure has an effect on the microstructure transformation from regular to complex (Liu et al., 2015). Meanwhile, SEM images, as seen in Fig. 4B(inset), not only presented the spherical particles in the copolymer-coating duration, but also showed the increased particle size from 50 to 60 nm, which was proportional to D_m values of hybrid P@BMMs with the copolymer-coated time intervals. Interestingly, it is noted that the symmetric profiles of distance distribution curves (as shown in Fig. 4C) suggested the spherical morphologies of all samples (Glatter, 1979; Shang et al., 2013), in close agreement with the SEM observations in Figs. 2 and 4B (inset).

Close

Fig. 4

A: SAXS patterns of (a) BMMs, (b) BMMs-MPS, (c) P@BMMs(1), (d) P@BMMs(2), (e) P@BMMs(4), and (f) P@BMMs(7); B: Effect of copolymer-coating time intervals (1, 2, 4, and 7 h, respectively) on the D_m values of hybrid P@BMMs(n) along with their morphologies (inset); C: r-P(r) chart originating from description given in the Experimental section. (a) BMMs, (b) BMMs-MPS, (c) P@BMMs(1), (d) P@BMMs(2), (e) P@BMMs(4), and (f) P@BMMs(7); D: Estimated particles size of P@BMMs(n) on the basis of distance distribution function (equation (4) in the Experimental section) with their estimated shell thickness on the basis of estimated particle size by deducing particle size of BMMs-MPS (inset left up) and DLS measurements (inset right down) with different copolymer-coated time intervals (1, 2, 4, and 7 h, respectively).

Export to PPT

On the basis of SAXS patterns and the distance distribution curves (Fig. 4C), Fig. 4D presented the evolution of particle size with copolymer-coating interval (Becker et al., 1998), the estimated shell thickness of P@BMMs(n) (inset left up) and DLS measurements (inset right down). As can be seen, the increase in estimated particle size of P@BMMs from 30.2 to 32.2 nm was similar to the increasing tendency from 297.2 to 536.8 nm of DLS profiles (inset right down), but much smaller than that of results measured using DLS data. The main reason was due to both the swelling behavior of surface-coated P(NIPAM-co-AA) copolymer in the hybrid P@BMMs and the solvation effect under neutral aqueous solutions during the DLS measurements. These observations again demonstrated that the copolymer gradually wrapped to the surface of BMMs with prolonged coating time interval. Taking BMMs-MPS as a reference, the estimated shell thickness of P@BMMs(1), P@BMMs(2), P@BMMs(4), and P@BMMs(7) was about 0.07, 0.08, 0.38 and 0.99 nm, respectively (Fig. 4D (inset left up)). These results showed that the shell thickness was proportional to the copolymer-coated time interval. The increase phenomena of D_m values with copolymer-coating time implied that nanoparticles with mass fractal feature were gradually changed to dense structures. Furthermore, the shells of obtained nanocomposites after encapsulation of the polymers were thicker with copolymer-coating time, meaning the successful preparation of nanocomposites with core-shell structures.

3.2 IBU loading and in vitro releasing performance

The SAXS patterns of drug-loaded samples were shown in Fig. 5A, showing the straight line for each scattering pattern in low region (q < 1) with a good correlation efficient (R² > 0.99). However, Fig. 5B indicated that their fractal characteristics varied from mass fractal of pre IBU-loaded sample (seen in Fig. 4B) to surface fractal, suggesting the transformation of the mesoporous structures from high porosity to densely rough surface (Varga et al., 2015). The D_s value also revealed increasing shell thickness from 2.47 for I/P@BMMs(1) to 2.77 for I/P@BMMs(7) with copolymer-coating time interval ranges, one of main reasons may be due to the introduction of IBU onto the mesoporous channels of P@BMMs.

Close

Fig. 5

A: SAXS patterns of (a) I/P@BMMs(1)-48, (b) I/P@BMMs(2)-48, (c) I/P@BMMs(4)-48, and (d) I/P@BMMs(7)-48; B: Evolution of surface fractal dimensions (D_s) calculated from SAXS patterns for I/P@BMMs(n) along with their morphologies (inset); C: r-P(r) chart originating from description given in the Experimental section. (a) I/P@BMMs(1)-48, (b) I/P@BMMs(2)-48, (c) I/P@BMMs(4)-48, and (d) I/P@BMMs(7)-48; D: Estimated particles size of I/P@BMMs(n) on the basis of distance distribution function (Eq. (4) in the Experimental section) with the estimated shell thickness of I/P@BMMs(n)-48 on the basis of estimated particle size by deducing particle size of BMMs-MPS (inset left up) and their DLS measurements (inset right down) with different copolymer-coated time intervals (1, 2, 4, and 7 h, respectively).

Export to PPT

The similarity of symmetric profiles for pre and after IBU-loaded samples was comparable with distance distribution curves, as shown in Fig. 5C, demonstrating that the spherical morphologies of I/P@BMMs(n) were reserved (Glatter, 1979; Shang et al., 2013), which could be also confirmed by SEM images (Figs. 2 and 5B (inset)).

As can be seen in Fig. 5D, the variations in the particle sizes originating from the distance distribution curves (Fig. 5C) in the range of 36 to 37.4 nm were similar to the increasing particle sizes from 1028 to 1303 nm of I/P@BMMs(n) measured by DLS (Fig. 5D (inset right down)), but much smaller than that of DLS measurements. Taking BMMs-MPS as a reference, the shell thickness of I/P@BMMs(1), I/P@BMMs(2), I/P@BMMs(4), and I/P@BMMs(7) was calculated around 2.9, 3.3, 3.5, and 3.6 nm , respectively (Fig. 5D (inset left up)), clearly showing comparatively more thickness than the corresponding hybrid P@BMMs (Fig. 4D (inset left up)), and suggesting that the drug was successfully loaded into the nanocomposites.

Presently, the IBU-loading content of P@BMMs was about 23%, a little lower than that of pure BMMs (27%) (Gao et al., 2012), indicating the certain influence of copolymer shell thickness on the IBU drug loading behavior. The detailed discussions are given in the following section.

Taking I/P@BMMs(4)-48-24 as an example, Fig. 6 presented the SAXS patterns of after drug-released samples obtained under various conditions. As shown in Fig. 6A, the straight line for each scattering pattern in low region (q < 1) with a good correlation coefficient (R² > 0.99) also implied the presence of typical mass fractal features for I/P@BMMs(4)-48-24, showing D_m value of around 2.93, 2.91, 2.9, 2.85, respectively. Interestingly, Fig. 6A (inset) exhibited that the fractal features of after IBU-released samples reverted back from surface fractal for drug-loaded sample (seen in Fig. 5B) to mass fractal, but was similar to the copolymer-coated BMMs (Fig. 4B). Combined with mentioned-above XRD analysis (Fig. 1B), these phenomena may be due to disappearances of the mesoporous structures (Jin et al., 2017).

Close

Fig. 6

A: SAXS patterns of I/P@BMMs(4)-48-24 obtained under different conditions of drug releasing and corresponding D_m values (inset); B: r-P(r) chart of I/P@BMMs(4)-48-24 originating from description in the Experimental section under different conditions of drug releasing. (a) 25 °C, pH 2.0; (b) 25 °C, pH 7.4; (c) 37 °C, pH 2.0; and (d) 37 °C, pH 7.4.

Export to PPT

However, Fig. 6B presented the symmetric profiles for after IBU-released samples, similar to that of pre and after drug-loaded samples, also confirming the reservation of spherical morphologies. The estimated particle size of I/P@BMMs(4)-48-24 under various releasing conditions was 36.11 (25 °C, pH 2.0), 35.72 (25 °C, pH 7.4), 35.34 (37 °C, pH 2.0), and 36.09 nm (37 °C, pH 7.4), respectively, when BMMs-MPS was taken as a reference, the estimated shell thickness of coated copolymers could be deduced to 2.95 (25 °C, pH 2.0), 2.75 (25 °C, pH 7.4), 2.56 (37 °C, pH 2.0), 2.94 nm (37 °C, pH 7.4), showing much thicker than that of P@BMM(4) (around 0.4 nm, seen in Fig. 4D (inset left up)), but a little thinner than that of drug-loaded I/P@BMMs (4)-48 (around 3.5 nm, seen in Fig. 5D (inset left up)). The estimated particle size and corresponding shell thickness may be influenced by various factors, such as shrunken-swollen behaviors of copolymer-coated shell, the residues of IBU left in the mesopores, and nonuniform nanoparticle morphologies of mesoporous BMMs. Similar behavior was observed in the I/P@BMMs(1)-48-24, I/P@BMMs(2)-48-24, I/P@BMMs(7)-48-24 obtained under various releasing conditions, and the detailed information was shown in Figs. S5–S7 of ESI section.

Fig. 7 presented the in vitro IBU-release performance from I/P@BMMs(n)-48 under different releasing conditions. As can be seen in Fig. 7A and 7C, in acidic medium (pH 2.0), the IBU-releasing equilibrium amount at 37 °C reached up to about 43.5% for I/P@BMMs(1)-48, 58% for I/P@BMMs(2)-48, 66.5% for I/P@BMMs(4)-48, and 75% for I/P@BMMs(7)-48, respectively, higher than that measured at 25 °C (24%, 30.5%, 37%, 40.5%), showing the temperature-responsive properties. These phenomena could be further interpreted in terms of low temperature, the hydrogen bonding interactions between IBU and P@BMMs carrier caused a delay in drug molecule’s release. On the contrary, the hydrogen bonding interactions were broken at high temperature (Sousa et al., 2005), which can facilitate the release of IBU from P@BMMs(n) carrier. However, in the PBS solution of pH 7.4, the IBU-releasing amount of I/P@BMMs (n = 1, 2, 4, and 7, respectively) at 37 °C was 22.5%, 33%, 35.5%, 42% (Fig. 7D), more or less similar to that (26%, 29%, 30.5%, 32.5%) at 25 °C (Fig. 7B). On the other hand, the IBU-release equilibrium amounts of I/P@BMMs (n = 1, 2, 4, and 7) at pH 2.0 (Fig. 7C) were around 43.5%, 58%, 66.5%, 75%, respectively, higher than that at pH 7.4 (Fig. 7D) (22.5%, 33%, 35.5%, 42%, respectively), showing the pH-responsive behaviors. These observations may be related to the shrinkage networks of copolymer in the low pH release solution and the swollen state at pH of basic medium, which have been investigated in our previous work (Jin et al., 2017). In addition, the IBU-releasing amount at pH 2.0 (Fig. 7A) was similar to that at pH 7.4 (Fig. 7B). The IBU release profiles presented the copolymer has dual pH- and temperature- responsive performances. These results demonstrated that, the equilibrium release rate also increased with the increasing polymer-coated time interval, which proved that the release performance of the IBU was significantly affected with the increasing coated-copolymer thickness.

Close

Fig. 7

In vitro IBU-release behaviors of I/P@BMMs(n)-48 obtained with different copolymer coated time (1, 2, 4, and 7 h, respectively) under various releasing conditions. A: 25 °C, pH 2.0; B: 25 °C, pH 7.4; C: 37 °C, pH 2.0; and D: 37 °C, pH 7.4.

Export to PPT

Thereafter, based on the mentioned-above discussion, the relationships between estimated shell thickness and release behavior from I/P@BMMs(n)-48 under different conditions were displayed in Fig. 8. As can be seen in Fig. 8A and 8C, the results presented that under the acid media of pH 2.0, the estimated shell thickness all decreased (at 25 °C, from 3.2 to 2.6 nm, at 37 °C from 3.2 to 2.4 nm) with increasing drug release percent (at 25 °C from 24.0 to 40.5%, at 37 °C from 43.5 to 75%). On the contrary, Fig. 8B and 8D presented that the estimated shell thickness increased at pH 7.4 (at 25 °C from 2.2 to 2.8 nm, at 37 °C, from 2.3 to 2.9 nm) with the increasing drug release percent (at 25 °C from 26 to 32.5%, at 37 °C from 22.5 to 42%), which may be due to the shrinkage networks of copolymer in the low pH release solution, but the swollen state at pH of basic medium. These observations certified that the shell thickness has an obvious influence on the drug release system.

Close

Fig. 8

Relationship between estimated shell thickness (black line) and drug release behavior (red line) for I/P@BMMs(n)-48 with copolymer-coating time intervals (1, 2, 4, and 7 h, respectively) under different conditions. A: 25 °C, pH 2.0; B: 25 °C, pH 7.4; C: 37 °C, pH 2.0; and D: 37 °C, pH 7.4.

Export to PPT

In order to further explore the influence of the shell thickness of hybrid P@BMMs on the drug release procedure, the kinetic performances of amine-modified BMMs as a reference (Han et al., 2015) and I/P@BMMs(7) were compared on the basis of Korsmeyer-Peppas model. As seen in Fig. 9, all fitting curves had a high linear correlation coefficient (R² > 0.88), meaning that the IBU-releasing could be well described by the Korsmeyer-Peppas model, in which, the release rate (k) value for amine-modified BMMs under acidic conditions (Fig. 9A) and alkaline conditions (Fig. 9B) remained intact in size of around 0.55–0.56, further demonstrating that amine-modified BMMs had no obvious pH-responsive property, in good agreement of our previous report (Han et al., 2015). However, after copolymer coating, the release rate of I/P@BMMs(7) under acidic conditions (k = 0.31) was significantly higher than that under alkaline conditions (k = 0.12), indicating that the drug-releasing behaviors followed the synergistic mechanism of diffusion/swelling-based control (Siepmann and Peppas, 2001).

Close

Fig. 9

Korsmeyer-Peppas model fitting curve of IBU-released samples under different conditions. A: 37 °C, pH 2.0; B: 37 °C, pH 7.4.

Export to PPT

3.3 Cytotoxicity test and pharmacokinetic performance in mice

The MTT assay (Cao et al., 2015) result of I/P@BMMs(7)-48 was showed in Fig. 10A. As can be seen, HepG-2 cells were incubated in I/P@BMMs(7)-48 by different concentrations, such as 0.2, 0.4, 0.8, 1.6, 3.2, 6.3, 12.5, 25, 50 and 100 μg/mL. The IC50 is around 70 μg/mL, indicated the good biocompatibilty and non-toxic properties, which can be used in drug delivery (Sun et al., 2017).

Close

Fig. 10

A: Cell viability with different concentrations of I/P@BMMs(7)-48 (0.2, 0.4, 0.8, 1.6, 3.2, 6.3, 12.5, 25, 50, and 100 μg/mL, respectively); B: Plasma concentration-time profile of IBU releasing from I/P@BMMs(7)-48 in mice.

Export to PPT

An ideal drug delivery vehicle must be non-toxic, biocompatible, non-immunogenic, biodegradable, and maintaining a desired drug plasma concentration. The theory of this ideal drug delivery system was firstly introduced by Teorell in 1937 (Teorell, 1937). Accordingly, the ideal drug release profiles must be greater than the minimum effective concentration but less than the minimum toxic concentration. It is noteworthy to mention that sustained release systems have advantages of improving “efficacy” and “safety” with long drug duration, fewer doses, and less fluctuation in plasma concentration. In the present study, Fig. 10B presents the sustained drug release behavior from hybrid nanocomposite, taken I/P@BMMs(7)-48 as an example, showing a high drug concentration at the beginning of the treatment, and thereafter presence of the burst release characteristics was noticed, but still existence of the sustained release was confirmed, whereas almost all the drug captured by liver or lung in a very short period gradually entered in the blood circulation after administration by tail intravenous injection. However, the plasma drug concentration profile containing two peaks at different time intervals (0.75 h and 6 h) might be due to the drug accumulation.