Soft Matter View Article Online

PAPER

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Cite this: DOI: 10.1039/c5sm00864f

View Journal

Controlled assembly of magnetic nanoparticles on microbubbles for multimodal imaging† Lei Duan,ab Fang Yang,a Lina Song,a Kun Fang,a Jilai Tian,a Yijun Liang,a Mingxi Li,a Ning Xu,c Zhongda Chen,b Yu Zhanga and Ning Gu*a Magnetic microbubbles (MMBs) consisting of microbubbles (MBs) and magnetic nanoparticles (MNPs) were synthesized for use as novel markers for improving multifunctional biomedical imaging. The MMBs were fabricated by assembling MNPs in different concentrations on the surfaces of MBs. The relationships between the structure, magnetic properties, stability of the MMBs, and their use in magnetic resonance/ ultrasound (MR/US) dual imaging applications were determined. The MNPs used were NPs of 3-aminopropyltriethoxysilane (APTS)-functionalized superparamagnetic iron oxide g-Fe2O3 (SPIO). SPIO was assembled on the surfaces of polymer MBs using a ‘‘surface-coating’’ approach. An analysis of the underlying mechanism showed that the synergistic effects of covalent coupling, electrostatic adsorption, and aggregation of the MNPs allowed them to be unevenly assembled in large amounts on the surfaces of the MBs. With an increase in the MNP loading amount, the magnetic properties of the MMBs improved significantly; in this way, the shell structure and mechanical properties of the MMBs could be modified. For surface densities ranging from 2.45  10

7

mg per MMB to 8.45  10

7

mg per MMB, in vitro MR/US imaging

experiments showed that, with an increase in the number of MNPs on the surfaces of the MBs, the MMBs Received 13th April 2015, Accepted 28th May 2015

exhibited better T2 MR imaging contrast, as well as an increase in the US contrast for longer durations.

DOI: 10.1039/c5sm00864f

modality image signals could be obtained for mouse tumors. Therefore, by adjusting the shell composition

www.rsc.org/softmatter

of MBs through the assembly of MNPs in different concentrations, MMBs with good magnetic and acoustic properties for MR/US dual-modality imaging contrast agents could be obtained.

In vivo experiments also showed that, by optimizing the structure of the MMBs, enhanced MR/US dual-

1. Introduction In recent years, magnetic nanoparticles (MNPs) and microbubbles (MBs) have come to be used as effective imaging contrast agents and drug carriers, respectively.1–12 MBs loaded with MNPs can act as novel carrier materials that exhibit the strengths of both MBs and MNPs. Given that they have both excellent magnetic and acoustic properties, they are increasingly being used as an integrated multifunctional platform for dual-mode ultrasound (US)/magnetic resonance imaging (MRI) as well as for the targeted delivery of genes and various drugs.13–19 However, the optimization of the structure of these magnetic microbubbles (MMBs), which are capable of significantly enhancing the results of dualmode US/MRI, remains a challenge. a

State Key Laboratory of Bioelectronics, Jiangsu Key Laboratory for Biomaterials and Devices, School of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, P. R. China. E-mail: [email protected] b Department of Biomedical Engineering, Nanjing Medical University, 210029, P. R. China c Department of Pathology, Nanjing Medical University, 210029, P. R. China † Electronic supplementary information (ESI) available: The optical microscopy images of MMBs coated different SPIO concentrations. See DOI: 10.1039/c5sm00864f.

This journal is © The Royal Society of Chemistry 2015

There are two main methods currently available for loading MNPs onto MBs: embedding the MNPs directly into the membrane shells of the MBs or coating them onto the surfaces of the shells of the MBs.20–24 Compared to embedding MNPs in the shells of MBs, coating the surfaces of the MB shells with MNPs has three main advantages. First, studies have shown that, for the same MNP type, size, and concentration, when MNPs are coated on the surfaces of MB shells, the quality of the US imaging and that of the MR imaging are both enhanced.21 Second, since different approaches for loading MNPs on the MBs may have different effects on the release performance of the MNPs as well as on their interactions with cells,25 when MNPs are coated on the shell surfaces of MBs, they can directly interact with cells, resulting in greater uptake by the cells. Ensuring that many MNPs can enter specific cells is beneficial for subsequent image tracking. Finally, the surfaces of MNPs are often modified with a large number of amino or carboxyl groups, which act as sites for the binding of the MNPs with antibiotics and genes.26 Therefore, coating MNPs onto MB surfaces allows the MBs to be coupled to additional active substances. We believe that MMBs with MNPs coated on their surfaces have a greater potential for use in biomedical applications. Methods for coating MNPs on the surfaces of MBs include a chemical

Soft Matter

View Article Online

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Paper

Soft Matter

approach and a physical method, each of which has advantages and disadvantages.21–23 The physical method is usually simpler than the chemical approach; however, the binding force of the physical effect is weaker than the degree of chemical coupling. If one could integrate the advantages of the above methods in a synergistic manner, it would allow for the easy preparation and optimization of MMBs. In addition, the assembly of MNPs on the shells of MBs can affect the magnetic and acoustic properties of the MBs. On the one hand, modifying the shells of MBs with MNPs is an effective method for improving MRI quality. On the other hand, the MNPs are likely to change the ultrasonic properties of the shells. The composition and microstructure of MBs significantly affect their mechanical and transport properties in the quasistatic state, which, in turn, will influence their response in the insonified state as well. In particular, the covalent bonding or chain entanglement of MNPs on the shells of MBs may significantly influence their acoustic properties, which, in turn, would affect the US imaging results.22,23,27 Yang et al. studied the effects of embedding Fe3O4 NPs in varying concentrations in the shells of MBs on MR and US imaging results. The obtained results showed that the concentration of the Fe3O4 NPs in the shells had to be kept at a certain threshold to ensure good US and MR imaging results.13,28 Therefore, a quantitative analysis of MNPs coated on the surfaces of MBs needs to be performed and their effects on the magnetic and acoustic properties of the MBs need to be studied in detail, in order to obtain optimized MR/US imaging contrast agents. In this study, we investigated polymer MBs onto whose surfaces NPs of 3-aminopropyltriethoxysilane (APTS)-functionalized superparamagnetic iron oxide g-Fe2O3 (SPIO) were assembled. To begin with, we characterized the properties of the MBs covered with MNPs in different concentrations. Next, the mechanism underlying the ‘‘surface-coating’’ approach was improved through an in-depth study of the structure of the MMBs. We found that the synergistic effects of covalent coupling, electrostatic adsorption, and the aggregation of the MNPs resulted in an uneven loading of a great number of the MNPs on the MB surfaces. Then, we studied the relationship between the quantity of MNPs loaded onto the MBs; the structure, magnetic properties, and stability of the MMBs; and their effects on the suitability of the resulting MMBs in MR/US dual-mode imaging. The results indicated that, by adjusting the quantity of MNPs loaded, optimized MMBs could be synthesized such that they improved both MR and US imaging quality. Lastly, the optimized MMBs were tested through in vitro MR/US dual-modality imaging experiments.

2. Materials and methods 2.1.

Materials

The APTS-coated SPIO g-Fe2O3 (APTS/g-Fe2O3) NPs were provided by the Jiangsu Key Laboratory for Biomaterials and Devices (China).29 Polyvinyl alcohol (PVA) (molecular weight (MW) = 31 000) was obtained from Sigma-Aldrich and poly L-lactic acid (PLLA) (MW = 30 000) from Shandong Daigang Company (China).

Soft Matter

Sodium periodate and sodium chlorite were purchased from Shantou Xilong Chemical Company (China). 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), N-hydroxysulfosuccinimide (NHS) sodium salt, and 2-(N-morpholino)ethanesulfonic (MES) acid were procured from Shanghai Medpep Company (China). We also used the surfactants Spans20 and Tweens80, all of which were reagent grade. N2 was also been used. 2.2.

Fabrication of MBs

Three kinds of MBs were prepared: uncoated MBs, MMBs and MMBs-phys. The term MBs-phys refers to the MBs synthesized on the basis of physical method alone. The method used for fabricating the MMBs has been described previously.21 Briefly, PVA was modified to obtain telechelic PVA with carboxylic groups at the chain ends according to the method described by Gaio Paradossi et al.30 The modified PVA and PLLA were used to fabricate multiple emulsions of the microcapsules.20 After being separated from free telechelic PVA, microcapsules were suspended in MES buffer (50 mM, pH = 5.4). The suspension was activated by EDC (0.4 mg mL 1) and NHS (0.6 mg mL 1) at room temperature, and then incubated with a certain amount of APTS/g-Fe2O3 NPs. After 48 h, products were collected and washed 3 times with distilled water. Products were stored in vials and lyophilized (FreeZone freeze dryer, LABCONCO, USA) using mannitol as a protective agent. After the drying cycle was completed, N2 was introduced into the vials. The method mentioned above of assembling APTS/g-Fe2O3 NPs on the surfaces of MBs is called ‘‘surface-coating’’ in this article. A series of MMB samples were prepared by adding 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 mL, respectively, of 4.6 mg mL 1 APTS/g-Fe2O3 NPs to 109 MBs. The same SPIO was used to fabricate MMBs-phys, using a method similar to that mentioned above—but without adding EDC and NHS to activate carboxylic groups on the microcapsule surfaces, and directly mixing the microcapsule suspension and the APTS/g-Fe2O3 NPs for 48 hours. The uncoated MBs were synthesized using the same method without adding the APTS/g-Fe2O3 NPs. 2.3.

Characterization

The MBs were examined using optical microscopy (BM1000, Jiang Nan Optical Company, China) to determine the particle size distribution. Images were obtained at 400 magnification. Samples were taken from each of the 3 batches, and 10 images were taken from each sample. The size distribution of the MBs was then obtained using purpose-written image analysis software in MatLab.31 The concentration of the MBs was measured with a hemocytometer. We evaluated all types of MBs at the same range of concentrations: 1  109 to 2  109 MBs per mL. The zeta-potential changes of the MMBs were determined at 25 1C using a photon correlation spectroscopy instrument (Malvern Zetasizer 3000, Malvern Instruments Company, Worcestershire, UK). The morphology of the MBs was observed using scanning electron microscopy (SEM) (Ultra Plus, Carl Zeiss, Germany). For a more detailed view of the assembled structure of MNPs on MBs, thin sections of about 50 nm to 60 nm, produced using a microtome (EM UC7, Leica Mikrosysteme Vertrieb GmbH, Germany), were imaged using a transmission electron microscope (TEM) (FEI Tecnai G2 Spirit Bio TWIN, US) at 80 kV.

This journal is © The Royal Society of Chemistry 2015

View Article Online

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Soft Matter

Dehydrated samples for TEM were prepared by mixing the microcapsule solution in acetone. The dried samples were embedded in liquid epoxy resin. After hardening, the resin blocks were sectioned. The Fe content of the APTS/g-Fe2O3 NPs assembled onto the shells of the MBs was determined from the corresponding absorbance versus iron concentration (r 2 = 0.9999) calibration curve, which was determined using the 1,10-phenanthroline spectrophotometric method;32 an ultraviolet-visible (UV-vis) spectrophotometer (UV-3600, Shimadzu, Japan) was employed for this purpose. The magnetization properties of the MMBs were studied using a vibrating sample magnetometer (VSM) (Model 7407, Lake Shore Cryotronics, Inc., USA). The samples included MBs coated with various concentrations of SPIO as well as MBs not coated with SPIO. All the samples were tested in dry powder form. The structure of the polymer shell (PVA and PLLA) of MMBs was investigated using a DSC204 (NETZSCH, Germany) differential scanning calorimeter (DSC). About 3–5 mg of lyophilized MBs was sealed in an aluminum pan. The scans were performed from 50 1C to 250 1C at a heating rate of 10 1C min 1 under a flux of 50 mL min 1 of dry N2. Data were collected after the first reference thermal cycle. 2.4.

In vitro US imaging

In vitro acoustic imaging was performed using a laboratorymade agar power phantom which had the similar ultrasonic parameters to human soft tissue, there was a cellulose tube pipeline in the phantom for sample loading. The test samples were imaged using a digital B-mode diagnostic ultrasonic instrument (Belson 3000A, Belson Imaging Technology Co., Ltd, China) and a 3.5 MHz R60 convex array probe. A suspension of the MBs was injected into the pipeline of the phantom imaged by the US instrument. The mean grayscale values of the US images were measured by using the imaging analysis software Image J. When a region of interest (ROI) was selected, its mean grayscale value was calculated automatically. Three scanned segments (ROIs) were processed for each sample, and the average mean grayscale value was used. As mentioned above, we tested MBs coated with SPIO in different amounts as well as uncoated MBs. The concentrations of all the MB samples were 1  109 MBs per mL to 2  109 MBs per mL. Distilled and degassed water was used as the control sample. All the measurements were performed under the same conditions, namely, at the same temperature using the same US exposure parameters and solvents with the same compositions. 2.5.

Paper

The T2 relaxation times were calculated using the post-processing software ParaVision 5.0. To determine the optimal amount of the SPIO APTS/g-Fe2O3 NPs to load onto the shells of the MBs, MBs coated with different amounts of the SPIO as well as uncoated MBs were studied. All the T2 values were measured for concentrations of 1  109 MBs per mL to 2  109 MBs per mL, in order to calculate the transverse relaxation rate (R2) as a function of the iron oxide content. 2.6.

In vivo tumor US imaging

The human colon cancer cell line COLO-205 was obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). Female BALB/C nude mice, aged 4–6 weeks, were obtained from the Center of Laboratory Animal Sciences of Nanjing Medical University (China). Approximately 106 to 107 COLO-205 cells were inoculated subcutaneously into the right flanks of the nude mice, and tumors were allowed to grow for 10–14 days prior to analysis. All the experiments were performed in compliance with the guidelines set by the Animal Care Committee of Southeast University. The test animals were kept anesthetized with 2% isoflurane in oxygen-enriched air using a facemask during the entire imaging process. In vivo US imaging was performed using a small animal US imaging system (Visualsonics Vevo2100, Canada). The tumors were scanned in three dimensions prior to injection to confirm a clear background signal. The US tumor images were obtained in real time after 200 mL of the MMB dispersion ([MMB] = 1  109 MBs per mL) had been injected intratumorally. The imaging settings for the US system were a center frequency, f, of 30 MHz and a gain of 20 dB. The mean grayscale values in the ROIs of the tumors were analyzed and normalized to the grayscale values before the injection of the MMBs. 2.7.

In vivo tumor MR imaging

For in vivo MRI, MMBs were injected intratumorally in the same dose, which was 200 mL per mouse ([MMB] = 1  109 MBs per mL). Using a 7.0 T scanner (Micro-MRI, PharmaScan, Brukers, Germany) equipped with a 3.8 cm circular coil, dynamic susceptibility imaging was performed both before and after the injection of the MMBs; a two-dimensional T2 fast low-angle shot sequence with respiratory gating control was employed. The parameters were TR/TE = 100 ms/8 ms, flip angle = 3501, FOV = 10  10 mm, slice thickness = 2 mm, number of excitations = 2, in-plane resolution = 0.78  0.78 mm2, and temporal resolution = 21 s. To evaluate the contrast, the signal intensities in the ROIs covering the tumors were measured before and after the injection of the MMBs by using the imaging analysis software Image J.

In vitro MRI experiments

In vitro MRI was performed on a 7.0 T system (Micro-MRI, PharmaScan, Bruker, Germany). To obtain the absolute T2 relaxation times, a multislice, multiecho T2 map sequence was used. The scan parameters were as follows: TR of 3000 ms, TE of 12–192 ms in steps of 12 ms, field of view (FOV) of 60  60 mm, matrix size of 256  256, slice thickness of 1 mm, and total scan time of 8 min.

This journal is © The Royal Society of Chemistry 2015

3. Results and discussion 3.1. Characterization of MBs loaded with MNPs at different concentrations SEM pictures and the size distribution of the MBs before and after they were loaded with the APTS/g-Fe2O3 NPs are shown in Fig. 1.

Soft Matter

View Article Online

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Paper

Fig. 1 SEM images of (A) the uncoated MBs and (B) the MMBs and (C) their size distribution.

One can see clearly that the APTS/g-Fe2O3 NPs were distributed on the surfaces of the MBs. Further, there was no significant difference in the diameters of the magnetic and nonmagnetic MBs. The mean diameters of the MBs and MMBs were 2.48 mm and 2.68 mm, respectively. Therefore, it can be concluded that the ‘‘surface-coating’’ approach changes the shell composition and microstructure of MMBs without significantly affecting their size. As mentioned previously, the SPIO APTS/g-Fe2O3 NPs were assembled on the surfaces of the polymer MBs at different concentrations using the ‘‘surface-coating’’ approach. Fig. 2A shows SEM images of MBs coated with different concentrations of the SPIO NPs. Fig. 2B shows the Fe contents of the various

Soft Matter

groups of MBs. The samples are (a) the uncoated MBs and (b–g) the MMBs with Fe contents of 135.1, 163.1, 232.4, 310.6, 393.2, and 466.2 mg mL 1, respectively. For a dispersion concentration of 1  109 MBs per mL, the SPIO APTS/g-Fe2O3 concentrations on a single MB were (b–g) 2.45  10 7, 2.95  10 7, 4.20  10 7, 5.63  10 7, 7.13  10 7, and 8.45  10 7 mg per MMB, respectively. It can be seen that, as the quantity of the SPIO APTS/g-Fe2O3 NPs added was increased, the MMB suspension became darker, and the Fe content of the MMBs increased accordingly. Moreover, even in sample g (which had the highest SPIO concentration, which was 8.45  10 7 mg per MMB, a value much larger than those reported previously13,14), the amount of SPIO coated on the MBs was lower than the saturation limit. This means that a greater number of the MNPs can be assembled on the MB by the ‘‘surface-coating’’ approach. Fig. 2C shows the magnetization characteristics of the MMBs coated with different concentrations of SPIO. After the SPIO NPs had been assembled on the shells of the MBs, no remnant magnetization was observed in the corresponding VSM curve. All the samples exhibited good superparamagnetic properties. With an increase in the SPIO concentration on the shells of the MBs, the saturation magnetization of the MMBs also increased. As for the uncoated MBs, a magnetic hysteresis loop was not observed. Therefore, the magnetic properties of the MMBs could be adjusted by controlling how many MNPs coated on them. 3.2.

Mechanism underlying the ‘‘surface-coating’’ approach

Previous studies have shown that the assembling of MNPs on MBs can influence the mechanical properties of the polymeric shells of the MBs.23 In our current study, the following synergistic steps (Fig. 3A) were hypothesized to cause the coating of the surfaces of the MBs with MNPs. (1) During the activation of EDC and NHS, chemical coupling takes place between the carboxyl groups added to the surfaces of the MBs and the amino groups

Fig. 2 (A) SEM images and (B) Fe concentrations of the (a) uncoated MBs and (b–g) the MMBs coated with SPIO at different concentrations. (b–g) are the MMBs with Fe contents of 135.1, 163.1, 232.4, 310.6, 393.2, and 466.2 mg mL 1, respectively. (C) (a) VSM curves and (b–g) the saturation magnetizations of the MMBs coated with SPIO at different concentrations. In the case of the uncoated MBs, a magnetic hysteresis loop was not observed.

Soft Matter

This journal is © The Royal Society of Chemistry 2015

View Article Online

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Soft Matter

Fig. 3

Paper

(A) Schematic showing the underlying principle of the ‘‘surface-coating’’ method. (B) TEM images of the MMBs-phys and MMBs.

present on the APTS/g-Fe2O3 NPs. (2) The MBs, which are covered with carboxyl groups, carry a negative charge, whereas the APTS/g-Fe2O3 NPs carry a positive charge. The effects of the neutralization of the positive and negative charges lead to the mounting of more SPIO on the MBs. (3) Since, during the assembly of the SPIO NPs, any changes in the system environment cause the NPs to undergo agglomeration to some degree, they accumulate on the MBs, increasing the coating capacity. Therefore, the cumulative effect of several factors acting together is that the ‘‘surface-coating’’ method results in the loading of many SPIO APTS/g-Fe2O3 NPs. In order to verify the above hypothesis, the shell microstructure was characterized and differential scanning calorimetry (DSC), and zeta-potential measurements were made on the MMBs and MMBs-phys. The term MMBs-phys refers to the MMBs synthesized on the basis of the carboxyl groups present on the surfaces of the MBs without requiring the activation of EDC and NHS. That is to say, in this case, the MNPs were formed only via physical interactions and not through chemical bonding. In contrast, the MMBs were generated through a combination of physical adsorption and chemical coupling. Fig. 3B shows TEM pictures of an MMB-phys and an MMB. As shown in the figure, the different assembly methods affected the structures of the MNPs assembled on the surfaces of the MBs. The MNPs loaded by physical adsorption were distributed relatively evenly on the surfaces of the MBs, whereas the MNPs generated by the combination of physical adsorption and chemical coupling tended to be unevenly cross-linked and distributed unevenly in the form of aggregates. To investigate this phenomenon further, we used an acid–base titration method to determine the carboxyl group content on a single MB. This value was found to be 1.03  10 10 mol per MB. Further, using the ninhydrin-staining reaction method, the amino group content on the MNPs was determined to be 4.22 mmol mg 1. Therefore, the theoretical MNP loading capacity of a single MB

This journal is © The Royal Society of Chemistry 2015

was determined to be 2.40  10 5 mg per MB, which was far greater than 8.45  10 7 mg per MB, the actual measured maximum MNP load capacity of the MMBs (sample g in Fig. 2B). The question that needed to be answered was why did a large number of carboxyl groups remain on the MBs even after several MNPs were loaded on their surfaces. Given the structure of the self-assembled MNPs present on the surfaces of the MMBs, as indicated by TEM, the above-mentioned question can be answered as follows. The carboxyl groups on the surfaces of the MBs could be divided into two categories: those activated by EDC and NHS and which could therefore undergo chemical coupling with the amino groups, and those not activated and which thus did not participate in the chemical coupling process. Of the activated carboxyl groups, only some of them underwent chemical coupling with the amino groups present on the surfaces of the MNPs; the remaining did not get to bond with the amino groups because of steric hindrance. Consequently, many carboxyl groups remained on the MBs, as was demonstrated by the negative zeta potential under this condition (Fig. 4A). However, as mentioned in our hypothesis, in addition to the chemical coupling between the MBs and MNPs, physical interactions also occurred, including opposite-charge adsorption and MNP accumulation, resulting in more MNPs being loaded on the surfaces of the MBs through physical adsorption. Because of the combined effects of chemical coupling and physical adsorption, the distribution of the MNPs assembled on the MBs under this condition differed significantly from that of the MNPs loaded via physical adsorption only, with the MNPs in the former case forming uneven cross-linkages and agglomerates and being loaded at a higher rate. Fig. 4A shows the differences in the zeta electrical potentials of the MMB-phys and MMB samples. The uncoated MBs were negatively charged with a zeta potential of 46.7 mV, while MNPs were positively charged with a zeta potential of 38.6 mV. With the continuous addition of MNPs, the zeta potential of the MBs increased, regardless of whether MMBs-phys or MMBs were formed. However, the change in the zeta potential of the MMBs

Soft Matter

View Article Online

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Paper

Soft Matter

Fig. 4 (A) Changes in the zeta electrical potentials of the MMBs-phys and MMBs. (B) Differential scanning calorimetry results of the uncoated MBs, MMBs-phys, and MMBs.

was slower than that for the MMBs-phys. The aforementioned TEM results could explain this phenomenon: physical adsorption allowed the MNPs to be distributed relatively evenly on the surfaces of the MBs (Fig. 3B), yielding a neutralization of the negative charges on the surfaces of the MBs. On the other hand, the combination of physical adsorption and chemical coupling changed the distribution of the MNPs on the MBs, with the MNPs tending to be unevenly cross-linked and agglomerated (Fig. 3B). Consequently, the degree of neutralization of the negative charge on the surfaces of the MBs by the MNPs was affected, and the rate of change in the zeta potential was relatively low. The shell structure plays a crucial role with respect to the mechanical properties of MBs. Fig. 4B shows the DSC curves of the uncoated MBs, MMBs-phys, and MMBs. As shown in the figure, the glass transition temperature (Tg), crystallization positions, and melting peaks of the uncoated MBs were close to those of MMBsphys, indicating that the polymer structure of the membrane shell of MB did not change significantly during the physical adsorption of the MNPs. However, there were two differences between the DSC curve of the MMBs and the other DSC curves. For one, the glass transition temperature (Tg) was observed to increase by B5 1C: the polymer on the surfaces of the MBs was connected to the MNPs through covalent bonds, which played a role similar to that of the side chain or cross-linking agent, reducing the activities of the molecular bonds and increasing the Tg. The second difference was the appearance of a molten doublet peak: as the temperature was increased, the covalent bond formed by the PVA on the surfaces of the MBs and MNPs was destroyed before the melting of the membrane shell material of the MBs, resulting in the endothermic peak. In summary, the membrane shell structure of the MMBs differed significantly from that of the MMBs-phys. The aforementioned TEM results, the results of the amino and carboxyl group content measurements, and the results of the zeta potential and DSC measurements demonstrated that the distribution mode of the MNPs on the surfaces of the MMBs, the change in the zeta potential, and the membrane shell structure of the MMBs prepared using the ‘‘surface-coating’’ approach differed significantly from those of MMBs-phys, which were

Soft Matter

prepared via physical coupling only. The mechanism through which the MNPs were assembled in the former case leveraged both chemical coupling and physical adsorption. We could even control the chemical/physical assembly ratio by controlling the degree of carboxylation and the degree of activation of the carboxyl groups on the surfaces of the MBs. The MMBs assembled using this combined approach had an extremely high loading capacity for MNPs, and their magnetic and mechanical properties could be modified by controlling the quantity of MNPs, which was shown to affect the US/MRI imaging quality. 3.3.

In vitro MR and US imaging

In vitro MR imaging was performed using the MBs coated with SPIO at different concentrations. The results are shown in Fig. 5A. With an increase in the SPIO concentration, the MR image gradually became darker, indicating that the higher the amount of SPIO that assembled on the MBs, the lower the T2 signal from the MR imaging process. That is to say, the properties of the MBs as a T2 MRI agent improved. This result is closely related to the VSM curves of the MMBs shown in Fig. 2C, which suggested that the higher the amount of SPIO that assembled onto the surfaces of the MBs, the greater the degree of saturation magnetization. Furthermore, the impact of the surrounding protons on the transverse section time, R2, during MRI was also greater. The functional relationship between the number of APTS/ g-Fe2O3 NPs coated on the surfaces of the MBs and the transverse relaxation rate, R2, is shown in Fig. 5B. The data could be fitted well with a linear function (r 2 = 0.9619). The linear relationship between R2 and the amount of SPIO coated was also applicable in the case of MBs with SPIO NPs embedded in their membrane shells.13 However, the SPIO NP-coated MBs exhibited a much higher degree of saturation magnetization and a higher R2 value than did the SPIO NP-embedding MBs for the same SPIO concentration.21 A possible explanation is that the SPIO NPs on the surfaces of the MBs are exposed directly to magnetic fields; as a result, they preserve their superparamagnetic properties to a greater extent than do those embedded in the polymeric shells. Therefore, the former vary the R2 value of the protons in the

This journal is © The Royal Society of Chemistry 2015

View Article Online

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Soft Matter

Paper

Fig. 5 (A) In vitro MRI images of uncoated MBs (a) and MMBs coated with different concentrations of SPIO. (b–g) are images of MMBs with SPIO contents of 2.45  10 7, 2.95  10 7, 4.20  10 7, 5.63  10 7, 7.13  10 7, and 8.45  10 7 mg per MMB. (B) Relationship between R2 and the SPIO concentration. All samples were diluted 10 times. (C) In vitro ultrasound images and (D) the mean grayscale–time curve of the (a) uncoated MBs and (b–g) the MMBs coated with different concentrations of SPIO.

surrounding water to a greater degree. Thus, in order to increase R2 during MRI, more SPIO APTS/g-Fe2O3 NPs should be assembled onto the surfaces of the MBs. By adjusting the quantity of SPIO added, we could regulate the MRI results obtained using the MMBs. We obtained US images of the samples at 0, 1, and 5 min after injection; these images are shown in Fig. 5C. In contrast to the images of the degassed and deionized water sample, those of the MBs were brighter. The brightness of the uncoated MBs was lower than that of the MMBs. In the case of the MMBs, as the concentration of SPIO was increased, the image brightness increased too. The increase in image brightness was calculated quantitatively from the average grayscale values of the ROIs of the images, which were recorded after intervals of 20 s. The plotted curves can be seen in Fig. 5D. As the concentration of the SPIO NPs coated onto the MB surfaces was increased, the duration for which the US image could be observed increased was well. This phenomenon may be explained in a couple of ways. (1) The dynamic behavior of the coated MBs in a US field is affected by the viscoelastic characteristics of their membrane shells, which, in turn, depend on the thickness and composition of the shells.33 The assembly of APTS/g-Fe2O3 NPs onto the surfaces of the MBs at different concentrations caused the membrane shell structure, composition, and thickness to change. These changes affected the degree of scattering of the coated MBs, leading to variations in the US images. On the other hand, the coating and stacking of NPs on the surfaces of the MBs was uneven and led to complex linear and nonlinear signals, which resulted in improved image brightness.34 (2) The assembly of MNPs on the surfaces of the MBs trapped the air within the MBs and prevented it from expanding and escaping from the shells. The hard shells also prevented the US waves from rapidly destroying the MBs. Thus, the acoustic properties could be maintained for a long time during the imaging process.35 Besides, even after the MMBs had

This journal is © The Royal Society of Chemistry 2015

broken down, the MNPs coated on their surfaces continued to be released. They may have aggregated locally and thus continued to scatter the US echo. The greater the number of MNPs that assembled, the greater the number of MNPs that will be released after the breakdown of the MMBs, resulting in a longer duration of US imaging. Thus, by controlling the number of MNPs assembled on the surfaces of the MBs, the quality of the US images and the imaging duration time can be improved. 3.4.

In vivo US and MR imaging

The in vitro MR/US imaging results mentioned above indicated that the greater the number of SPIO NPs that assembled on the surfaces of the MBs, the greater the improvement in the quality of the MR images and the longer the duration of US imaging. However, we found that, as the amount of SPIO loaded was increased further (especially when the concentration was increased to a value greater than 8.45  10 7 mg per MMB), the stability of the MMBs was affected (Fig. S1 in ESI†), resulting in the formation of agglomerates and deposits; such formations cannot be injected intravenously and thus have poor clinical applicability. Hence, a balance must be maintained between the quantity of SPIO coated on the MBs and the stability of the resulting MMBs. Therefore, MMBs coated with SPIO at a concentration of 5.63  10 7 mg per MMB were chosen for the US/MR imaging of tumors in nude mice. Fig. 6A shows US images of the MMBs before and after they were injected into the tumor sites. It can be observed that the injection of MMBs at the tumor sites improved the image contrast. By calculating the average grayscale values of the ROIs in the US tumor images, the change in the average grayscale values could be plotted over time (Fig. 6B). It can be seen that there was an enhancement in the intensity of the US signal of the tumors immediately after the injection of the MMBs. However, the US signals began to decrease gradually in the following 10 minutes due to the breakdown of the MMBs.

Soft Matter

View Article Online

Paper

Soft Matter

and subsequent treatments. Under external fields such as magnetic or US fields, it should be possible to use the MMBs for targeted delivery of drugs and genes. It should also be possible to use the MMBs in other biomedical applications such as the evaluation of treatment outcomes using multimodal imaging techniques.36–39

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

Acknowledgements

Fig. 6 In vivo US/MR imaging. (A) Contrast-enhanced US images before and after the injection of the MMBs. (B) Grayscale–time curve of the US images. (C) T2 MR images. (D) The average signal intensity before and after the injection of the MMBs.

The in vivo MRI results (Fig. 6C and D) showed an immediate decrease in the intensity of the T2 signals of the tumor sites following the injection of the MMBs. The T2 signals at the tumor sites were maintained in the 10 min observation period that followed. A possible reason for this is that the injection of the MMBs into the tumor sites caused the MMBs to form clusters, such that even when the MMBs broke, the MNPs were released from the surface of the MBs and subsequently aggregated and accumulated at the injection sites for a while. Thus, it was found that MMBs coated with MNPs in the optimized concentration simultaneously improved both US and MR imaging results in vivo.

4. Conclusions In this study, we prepared MMBs by assembling SPIO APTS/ g-Fe2O3 NPs at different concentrations on the surfaces of MBs by the ‘‘surface-coating’’ approach. By studying the structure and physicochemical properties of the MMBs, it was found that the ‘‘surface-coating’’ approach could be used to coat MBs with large quantities of MNPs through the synergistic effects of both chemical coupling and physical adsorption. The assembly process does not significantly affect the size of the MBs, but their shell structure is changed, their magnetic properties are improved, and their stability is affected, resulting in changes in their usability and performance as MR/US imaging agents. In vitro MR/US imaging experiments demonstrated that an increase in the number of MNPs loaded onto the MBs resulted in MR image enhancement and increased the duration for which the US image could be observed. Therefore, by adjusting the loading amount of the MNPs, we could control the magnetic and mechanical properties of the MMBs to improve the MR and US imaging at the same time. In vivo MR/US imaging experiments also proved that the MMBs could improve the results of both US and MR imaging simultaneously. Furthermore, the MNP-coated MMBs exhibited several other advantages. For example, the MNPs, which could interact with lesion tissues directly, were released and phagocytized in large quantities even when the MBs broke down; this was conducive for MRI detection

Soft Matter

This investigation was financially funded by the project of National Key Basic Research Program of China (2011CB933503, 2013CB733804), National Natural Science Foundation of China (31370019, 61127002), Natural Science Foundation of Jiangsu Province (NBK2011036) and Jiangsu Provincial Special Program of Medical Science (BL2013029). The funding partially also comes from the Author of National Excellent Doctoral Dissertation of China (No. 201259), as well as the Fundamental Research Funds for the Central Universities.

Notes and references 1 M. Mahmoudi, S. Sant, B. Wang, S. Laurent and T. Sen, Superparamagnetic iron oxide nanoparticles (SPIONs): development, surface modification and applications in chemotherapy, Adv. Drug Delivery Rev., 2011, 63, 24–46. 2 K. Ferrara, R. Pollard and M. Borden, Ultrasound microbubble contrast agents: fundamentals and application to gene and drug delivery, Annu. Rev. Biomed. Eng., 2007, 9, 415–447. 3 E. C. Unger, E. Hersh, M. Vannan, T. O. Matsunaga and M. McCreery, Local drug and gene delivery through microbubbles, Prog. Cardiovasc. Dis., 2001, 44, 45–54. 4 S. Tinkov, R. Bekeredjian, G. Winter and C. Coester, Microbubbles as Ultrasound Triggered Drug Carriers, J. Pharm. Sci., 2009, 98, 1935–1961. 5 S. Hernot and A. L. Klibanov, Microbubbles in ultrasoundtriggered drug and gene delivery, Adv. Drug Delivery Rev., 2008, 60, 1153–1166. 6 K. Kooiman, H. J. Vos, M. Versluis and N. D. Jong, Acoustic behavior of microbubbles and implications for drug delivery, Adv. Drug Delivery Rev., 2014, 72, 28–48. 7 M. M. Yallapu, S. F. Othman, E. T. Curtis, B. K. Gupta, M. Jaggi and S. C. Chauhan, Multi-functional magnetic nanoparticles for magnetic resonance imaging and cancer therapy, Biomaterials, 2011, 32, 1890–1905. 8 S. Fokong, B. Theek, Z. J. Wu, P. Koczera, L. Appold, S. Jorge, U. R. Genger, M. V. Zandvoort, G. Storm, F. Kiessling and T. Lammers, Image-guided, targeted and triggered drug delivery to tumors using polymer-based microbubbles, J. Controlled Release, 2012, 163, 75–81. 9 S. M. Janib, A. S. Moses and J. A. MacKay, Imaging and drug delivery using theranostic nanoparticles, Adv. Drug Delivery Rev., 2010, 62, 1052–1063. 10 C. Y. Ting, C. H. Fan, H. L. Liu, C. Y. Huang, H. Y. Hsieh, T. C. Yen, K. C. Wei and C. K. Yeh, Concurrent blood brain barrier opening and local drug delivery using drug-carrying

This journal is © The Royal Society of Chemistry 2015

View Article Online

Soft Matter

11

12

Published on 28 May 2015. Downloaded by Purdue University on 11/06/2015 07:57:29.

13

14

15

16

17

18

19

20

21

22

23

microbubbles and focused ultrasound for brain glioma treatment, Biomaterials, 2012, 33, 704–712. C. H. Fan, C. Y. Ting, H. L. Liu, C. Y. Huang, H. Y. Hsieh, T. C. Yen, K. C. Wei and C. K. Yeh, Antiangiogenic-targeting drug-loaded microbubbles combined with focused ultrasound for glioma treatment, Biomaterials, 2013, 34, 2142–2155. S. P. Qin, C. F. Caskey and K. W. Ferrara, Ultrasound contrast microbubbles in imaging and therapy: physical principles and engineering, Phys. Med. Biol., 2009, 54, 27–57. F. Yang, Y. X. Li, Z. P. Chen, Y. Zhang, J. R. Wu and N. Gu, Superparamagnetic iron oxide nanoparticle-embedded encapsulated microbubbles as dual contrast agents of magnetic resonance and ultrasound imaging, Biomaterials, 2009, 30, 3882–3890. Z. Liu, T. Lammers, J. Ehling, S. Fokong, J. Bornemann, ¨tjens, Iron oxide nanoparticle-containing F. Kiessling and J. Ga microbubble composites as contrast agents for MR and ultrasound dual-modality imaging, Biomaterials, 2011, 32, 6155–6163. C. C. Niu, Z. G. Wang, G. M. Lu, T. M. Krupka, Y. Sun, Y. F. You, W. X. Song, H. T. Ran, P. Li and Y. Y. Zheng, Doxorubicin loaded superparamagnetic PLGA-iron oxide multifunctional microbubbles for dual-mode US/MR imaging and therapy of metastasis in lymph nodes, Biomaterials, 2013, 34, 2307–2317. C. H. Fan, C. Y. Ting, H. J. Lin, C. H. Wang, H. L. Liu, T. C. Yen and C. K. Yeh, SPIO-conjugated, doxorubicin-loaded microbubbles for concurrent MRI and focused-ultrasound enhanced brain-tumor drug delivery, Biomaterials, 2013, 34, 3706–3715. H. Y. Huang, S. H. Hu, S. Y. Hung, C. S. Chiang, H. L. Liu, T. L. Chiu, H. Y. Lai, Y. Y. Chen and S. Y. Chen, SPIO nanoparticle-stabilized PAA-F127 thermosensitive nanobubbles with MR/US dual-modality imaging and HIFU-triggered drug release for magnetically guided in vivo tumor therapy, J. Controlled Release, 2013, 172, 118–127. ¨thel, H. Mannell, J. Pircher, F. Fochler, Y. Stampnik, T. Ra ¨rnle, B. Gleich, C. Plank, O. Mykhaylyk, C. Dahmani, M. Wo ¨tz, Site directed vascular gene A. Ribeiro, U. Pohl and F. Kro delivery in vivo by ultrasonic destruction of magnetic nanoparticle coated microbubbles, Nanomedicine, 2012, 8, 1309–1318. B. Xu, H. J. Dou, K. Tao, K. Sun, J. Ding, W. B. Shi, X. S. Guo, J. Y. Li, D. Zhang and K. Sun, ‘‘Two-in-One’’ Fabrication of Fe3O4/MePEG–PLA Composite Nanocapsules as a Potential Ultrasonic/MRI Dual Contrast Agent, Langmuir, 2011, 27, 12134–12142. F. Yang, A. Y. Gu, Z. P. Chen, N. Gu and M. Ji, Multiple emulsion microbubbles for ultrasound imaging, Mater. Lett., 2008, 62, 121–124. W. He, F. Yang, Y. H. Wu, S. Wen, P. Chen, Y. Zhang and N. Gu, Microbubbles with surface coated by superparamagnetic iron oxide nanoparticles, Mater. Lett., 2012, 68, 64–67. ¨rmark, T. B. Brismar, D. Grishenkov, B. Gustafsson, J. Ha Å. Barrefelt, S. V. V. N. Kothapalli, S. Margheritelli, L. Oddo, K. Caidahl, H. Hebert and G. Paradossi, Magnetite Nanoparticles Can be Coupled to Microbubbles to Support Multimodal Imaging, Biomacromolecules, 2012, 13, 1390–1399. M. Poehlmann, D. Grishenkov, S. V. V. N. Kothapalli, ¨rmark, H. Hebert, A. Philipp, R. Hoeller, M. Seuss, J. Ha

This journal is © The Royal Society of Chemistry 2015

Paper

24

25

26

27

28

29

30 31

32

33

34

35

36

37 38

39

C. Kuttner, S. Margheritelli, G. Paradossi and A. Fery, On the interplay of shell structure with low- and high- frequency mechanics of multifunctional magnetic microbubbles, Soft Matter, 2014, 10, 214–226. H. Mulvana, R. J. Eckersley, M. X. Tang, Q. Pankhurst and E. Stride, Theoretical and Experimental Characterisation of Magnetic Microbubbles, Ultrasound Med. Biol., 2012, 38, 864–875. F. Yang, M. Zhang, W. He, P. Chen, X. W. Cai, L. Yang, N. Gu and J. R. Wu, Controlled Release of Fe3O4 Nanoparticles in Encapsulated Microbubbles to Tumor Cells via Sonoporation and Associated Cellular Bioeffects, Small, 2011, 7, 902–910. D. E. Lee, H. Koo, I. C. Sun, J. H. Ryu, K. Kim and I. C. Kwon, Multifunctional nanoparticles for multimodal imaging and theragnosis, Chem. Soc. Rev., 2012, 41, 2656–2672. J. I. Park, D. Jagadeesan, R. Williams, W. Oakden, S. Chung, G. J. Stanisz and E. Kumacheva, Microbubbles Loaded with Nanoparticles: A Route to Multiple Imaging Modalities, ACS Nano, 2010, 4, 6579–6586. F. Yang, L. Li, Y. X. Li, Z. P. Chen, J. R. Wu and N. Gu, Superparamagnetic nanoparticle inclusion microbubbles for ultrasound contrast agents, Phys. Med. Biol., 2008, 53, 6129–6141. M. Ma, Y. Zhang, W. Yu, H. Shen, H. Zhang and N. Gu, Preparation and characterization of magnetite nanoparticles coated by amino silane, Colloids Surf., A, 2003, 212, 219–226. G. Paradossi, F. Cavalieri, E. Chiessi, V. Ponassi and V. Martorana, Biomacromolecules, 2002, 3, 1255. C. A. Sennoga, V. Mahue, J. Loughran, J. Casey, J. M. Seddon, M. Tang and R. J. Eckersley, On sizing and counting of microbubbles using optical microscopy, Ultrasound Med. Biol., 2010, 36, 2093. A. E. Harvey, J. A. Smart and E. S. Amis, Simultaneous spectrophotometric determination of iron(II) and total iron with 1,10-phenanthroline, Anal. Chem., 1955, 27, 26–29. J. C. Machado and J. S. Valente, Ultrasonic scattering cross sections of shell-encapsulated gas bubbles immersed in a viscoelastic liquid: first and second harmonics, Ultrasonics, 2003, 41, 605–613. E. Strid, K. Pancholi and M. J. Edirisinghe, Increasing the nonlinear character of microbubble oscillation at low acoustic pressures, J. R. Soc., Interface, 2008, 5, 807–811. M. Fosnaric, A. Iglic and S. May, Influence of rigid inclusions on the bending elasticity of a lipid membrane, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2006, 74, 051503. F. Yang, Z. X. Gu, X. Jin, H. Y. Wang and N. Gu, Magnetic microbubble: a biomedical platform co-constructed from magnetics and acoustics, Chin. Phys. B, 2013, 22, 104301. T. Hussain and Q. T. Nguyen, Molecular imaging for cancer diagnosis and surgery, Adv. Drug Delivery Rev., 2014, 66, 90–100. F. Kiessling, S. Fokong, J. Bzyl, W. Lederle, M. Palmowski and T. Lammers, Recent advances in molecular, multimodal and theranostic ultrasound imaging, Adv. Drug Delivery Rev., 2014, 72, 15–27. N. Bertrand, J. Wu, X. Y. Xu, N. Kamaly and O. C. Farokhzad, Cancer nanotechnology: the impact of passive and active targeting in the era of modern cancer biology, Adv. Drug Delivery Rev., 2014, 66, 2–25.

Soft Matter

Controlled assembly of magnetic nanoparticles on microbubbles for multimodal imaging.

Magnetic microbubbles (MMBs) consisting of microbubbles (MBs) and magnetic nanoparticles (MNPs) were synthesized for use as novel markers for improvin...
3MB Sizes 1 Downloads 9 Views