Preparation and evaluation of MRI detectable poly (acrylic acid) microspheres loaded with superparamagnetic iron oxide nanoparticles for transcatheter arterial embolization

Accepted Manuscript Title: Preparation and evaluation of MRI detectable poly (acrylic acid) microspheres loaded with superparamagnetic iron oxide nanoparticles for transcatheter arterial embolization Author: Huan Wang Xiao-Ya Qin Zi-Yuan Li Li-Ying Guo Zhuo-Zhao Zheng Li-Si Liu Tian-Yuan Fan PII: DOI: Reference:

S0378-5173(16)30658-5 http://dx.doi.org/doi:10.1016/j.ijpharm.2016.07.028 IJP 15925

To appear in:

International Journal of Pharmaceutics

Received date: Revised date: Accepted date:

16-4-2016 6-7-2016 14-7-2016

Please cite this article as: Wang, Huan, Qin, Xiao-Ya, Li, Zi-Yuan, Guo, Li-Ying, Zheng, Zhuo-Zhao, Liu, Li-Si, Fan, Tian-Yuan, Preparation and evaluation of MRI detectable poly (acrylic acid) microspheres loaded with superparamagnetic iron oxide nanoparticles for transcatheter arterial embolization.International Journal of Pharmaceutics http://dx.doi.org/10.1016/j.ijpharm.2016.07.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Preparation and evaluation of MRI detectable poly (acrylic acid) microspheres loaded with superparamagnetic iron oxide nanoparticles for transcatheter arterial embolization

HuanWanga,b,1, Xiao-YaQina,b,1, Zi-Yuan Lia,b, Li-Ying Guoa,b, Zhuo-Zhao Zhengc, Li-Si Liud, Tian-Yuan Fana.b*

a

The State Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences,

Peking University, Beijing 100191, China b

Beijing Key Laboratory of Molecular Pharmaceutics and New Drug Delivery Systems, School of

Pharmaceutical Sciences, Peking University, Beijing 100191, China c

Department of Nuclear Medicine , Beijing Tsinghua Changgung Hospital, Beijing 100044, China

d Department

of Radiology, Peking University Third Hospital, Beijing100191 , China

Graphical abstract

*

Corresponding author. Tel.: +86-010-82805123.

E-mail address: [email protected]. Address: Department of Pharmaceutics, Peking University, 38 Xueyuan Road, Haidian District, Beijing 100191, People’s Republic of China. 1

These authors contributed equally to this work.

Abstract: To monitor the spatial distribution of embolic particles inside the target tissues during and after embolization, blank poly (acrylic acid) microspheres (PMs) were initially prepared by inverse suspension polymerization method and then loaded with superparamagnetic iron oxide (SPIO) nanoparticles by in situ precipitation method to obtain magnetic resonance imaging (MRI) detectable SPIO-loaded poly (acrylic acid) microspheres (SPMs). The loading of SPIO nanoparticles in SPMs was confirmed by vibrating sample magnetometer, transmission electron microscopy, X-ray diffraction, X-ray photoelectron spectroscopy and infrared spectrum, respectively. The results showed that SPMs exhibited excellent superparamagnetism and the SPIO embedded in SPMs were proved to be inverse spinel magnetite. The content of SPIO loaded in wet SPMs of subgroups of 100-300, 300-500, 500-700 and 700-900 μm was measured to be 11.84 ± 0.07, 10.20 ± 0.05, 9.98 ± 0.00 and 8.79 ± 0.01 mg/ml, corresponding to the weight percentage in freeze-dried SPMs to be 18.07 ± 0.28%, 18.54 ± 0.13%, 18.66 ± 0.01% and 18.50 ± 0.07%, respectively. The SPMs were spherical in shape, had smooth surface, and were within the range of clinical demands for embolization. The compression tests indicated that SPMs were more rigid than PMs and commercially used Embospheres (P < 0.01). The MRI detectability of SPMs was evaluated with the SPMs embedded in gel phantom in vitro and injected subcutaneously into the back of mice in vivo. Both the results demonstrated that the SPMs could provide distinct negative contrast enhancement and be sensitively detected by T2-weighted MR imaging. All the results show that SPMs are potential MRI detectable embolic microspheres for the future embolotherapy.

Keywords: Microspheres, Superparamagnetic iron oxide, Magnetic resonance imaging, Detectability, Embolization

1. Introduction

In recent years, trancatheter arterial embolization (TAE) has emerged as an important noninvasive interventional therapy for the treatment of inoperable tumors (Bendszus et al., 2000; Skalicky et al., 2010;Treska et al., 2010), symptomatic uterine fibroids (Bendszus et al., 2000), arteriovenous malformations (Fleetwood and Steinberg, 2002; Guziński et al., 2010; Khan et al., 2010) and hemorrhage (Murakami et al., 2000). During the procedure of TAE, embolic materials

are introduced through a microcatheter to selectively occlude blood vessels that feed a tumor or other target tissues to obtain therapeutic benefits (Negussie et al., 2015). There is a variety of embolic agents available for clinical uses, including coils, liquids, nonspherical particles and hydrogel microspheres, etc (Negussie et al., 2015). Unfortunately, none of these embolic agents can be visualized with computed tomography (CT) or magnetic resonance imaging (MRI). The spatial distribution of embolic agents inside the target tissues can only be estimated indirectly by iodinated contrast agents mixed with embolic agents being assessed the devascularization area before and after embolization through X-ray angiography (Stampfl et al., 2012). The main disadvantage of this method is that the contrast agents can easily dissociate with the embolic agents mixed before, leading to fuzzy imaging and misdiagnosis as well as causing systemic toxicity (Lewis et al., 2006). Thus, it is of great importance to develop detectable embolic agents by CT or MRI for embolotherapy. Compared with CT, MRI has relatively high spatial and temporal resolution and superior soft tissue contrast properties (Oerlemans et al., 2013). Besides, MRI detection can avoid ionizing radiation exposure and reduce the need for used of iodinated contrast agents, preventing serious complications and decreasing costs (Chung et al., 2012). Therefore, MRI detectable embolic microspheres have gained increasing interests in recent decade years(Chung et al., 2012; Chen et al., 2014; Ferreira et al., 2012; Oerlemans et al., 2015; Pouponneau et al., 2009; van Elk et al., 2015). The most common method for the preparation of MRI-visualized embolic microspheres is incorporating superparamagnetic iron oxide (SPIO) nanoparticles into the matrix of microspheres (Chung et al., 2012; Chen et al., 2014; Ferreira et al., 2012; Kim et al., 2007).SPIO nanoparticles are an important class of multifunctional particles with unique superparamagnetism, high saturation magnetization and excellent biocompatibility (Ramimoghadam et al., 2015).In tissues, the large magnetic moments associated with SPIO nanoparticles will result in local magnetic field inhomogeneities. Diffusion of water through these local filed disturbances will produce rapid proton dephasing, which results in preferential shortening of transverse relaxation time and finally generates detectable changes in the MR signals (Ferrucci and Stark, 1990).SPIO agent, like Ferumoxides, had been widely used as MR contrast agent for detection of tumor in liver and spleen (Arnold et al., 2003). In most studies to prepare SPIO loaded microspheres, the SPIO

nanoparticles are synthesized or purchased commercially beforehand and after that loaded on microspheres by various methods. For example, by mixing ferrofluid (EMG 304) with an organic solution of poly (lactide-co-glycolide) (PLG) polymer, ferrofluid encapsulated PLG microspheres could be obtained by emulsion evaporation method (Chen et al., 2014). Moreover, after suspending SPIO in acetic solution of chitosan, SPIO-embedded chitosan microspheres were prepared by emulsion crosslinking method (Chung et al., 2012; Kim et al., 2007). Moreover, Fe3O4 nanoparticles-incorporated poly (vinyl acetate) microparticles were produced by dispersing oleic acid-modified Fe3O4 nanoparticles in vinyl acetate and subsequent polymerization and alcoholysis of the vinyl acetate (Ferreira et al., 2012). In 1980s, Ugelstad et al. proposed an in situ precipitation method to prepare magnetic polymer particles. Briefly, particles were immersed in iron salts solution and the iron salts would then penetrate into the particles. By raising the pH value with or without heating, magnetic iron oxide would be formed in the particles. This method has been used in the fields of magnetic-targeted drug delivery, magnetic separation, and immobilization of enzymes by now (Ugelstad et al., 1988). Compared with the methods mentioned above, in situ precipitation method will be more time-saving and simpler to prepare SPIO-loaded microspheres. Moreover, the property of SPIO nanoparticles will not be affected by the next process of preparing microspheres. In our previous report (Cui et al., 2012), poly (acrylic acid) microspheres (PMs) have been prepared and proved to be an ideal embolic agent with excellent morphology, size distribution, elasticity, catheter deliverability, cytocompatibility and drug loading ability. Thus, PMs were considered promising microspheres for the loading of SPIO nanoparticles by in situ precipitation. Besides, iron ions were supposed to penetrate into the network of PMs with most bounding to the carboxyl groups, and magnetic particles would take shape and disperse uniformly in the microspheres after alkalization. Additionally, poly (acrylic acid) was reported to be used as a surfactant to modify the surface of SPIO nanoparticles to prevent aggregation (Ge et al., 2007; Xuan et al., 2009). Therefore, the magnetic particles were speculated to be dispersed stably in the PMs and present long-term MRI detectable property. In this study, PMs were initially prepared by inverse suspension polymerization method and then loaded with SPIO nanoparticles by in situ precipitation method to develop long-term MRI detectable microspheres. The iron oxide in the SPIO-loaded poly (acrylic acid) microspheres

(SPMs) was characterized by vibrating sample magnetometer, transmission electron microscopy, X-ray diffraction, X-ray photoelectron spectroscopy and infrared spectrum. A series of tests was also performed to evaluate the properties of SPMs for embolization and MRI detectability, including morphology, size distribution, elasticity and MRI visibility in vitro and in vivo.

2. Materials and methods

2.1. Materials

Acrylic acid (AA, C.P.), potassium persulfate (KPS, A.R.) and Span 80 (C.P.) were supplied by Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). N, N’-methylenebisacrylamide (MBA, A.R.) was purchased from Junyao Albert Biotechnology Company (Beijing, China). Liquid paraffin (C.P.) was obtained from Xilong Chemical Co., Ltd. (Shantou, China). Ferric chloride (FeCl3•6H2O, A.R.) and ferrous sulfate (FeSO4•7H2O, A.R.) were purchased from Tianjin Guangfu Fine Chemical Research Institute (Tianjin, China). Sodium hydroxide (NaOH, A.R.) was purchased from Beijing Chemical Plant (Beijing, China). Embospheres® were supplied by Biosphere Medical Inc. (USA). Agarose, pentobarbital sodium and sodium carboxylmethyl cellulose were purchased from Sigma Aldrich (USA). All the other chemicals and solvents were analytical agents and used without further purification.

2.2. Preparation of PMs

PMs were prepared by the inverse suspension polymerization method with small modification from our previous report (Cui et al., 2012). Briefly, 40 ml of liquid paraffin was mixed with Span 80 and used as a continuous phase (oil phase). Required amount of AA (monomer, 3.6 g), MBA (cross-linking agent, 0.139 g) and KPS (initiator, 0.135 g) were dissolved in water and used as a dispersed phase (water phase). Then, the oil phase was poured into a three-neck flask equipped with a mechanical stirrer and incubated in a thermostatic water bath. After the water phase was added into the flask under stirring, the polymerization reaction was carried out at 55ºC for 3 h under nitrogen atmosphere. After cooling to room temperature, the

resultant particles were washed successively with 0.5% Tween 80 (w/v) solution and deionized water, and subsequently sieved to the final size range of 100-900 m. Then, some of the PMs were further separated into different subgroups with the size ranges of 100-300, 300-500, 500-700, 700-900m by wet sieving. Finally, all PMs were stored in distilled water for further analysis.

2.3. Preparation of SPMs

SPMs were prepared by loading SPIO nanoparticles into PMs using in situ precipitation approach (Ugelstad et al., 1988). In a typical reaction, 2.360 g of FeCl3•6H2O and 1.390 g of FeSO4•7H2O were dissolved in 50 ml of deionized water and poured into a three-necked flask equipped with a mechanical stirrer. A certain amount of wet PMs (i.e., the water on surface of microspheres was blotted, in the size range of 100-900 m) was added in and mixed with the iron salts solution for 30 min. Afterwards, 10 ml of NaOH solution (8 mol/L) was put in to start an precipitation reaction which was continued at 60ºC for 1 h under nitrogen atmosphere. After cooling to room temperature, the brown-black SPMs were washed repeatedly with deionized water until no chloride ion could be detected in the supernatant. For special use in later experiments, some SPMs were lyophilized and some were separated into four subgroups of 100-300, 300-500, 500-700, and 700-900 m by wet sieving and stored in distilled water for further use.

2.4. Magnetic property

In order to validate the loading of SPIO nanoparticles in SPMs, the magnetization curve for freeze-dried SPMs was measured at room temperature using a vibrating sample magnetometer (VSM, HH-15, Nanjing NanDa Instrument Plant, China) with magnetic field in the range of -6 kOe to 6 kOe. The superparamagnetism was evaluated from the magnetic hysteresis, residual magnetization and coercivity (Frounchi and Shamshiri, 2015; Hagit et al., 2010).To investigate the particle size of SPIO loaded in SPMs quantitatively, the magnetization curve of SPMs was fitted to a Langevin function based on the magnetic property of SPIO loaded in the SPMs and the assumption that the magnetic phase of SPIO was magnetite (Fe3O4) (Trohidou, 2015).

2.5. Transmission electron microscopy

The morphology and particle size of SPIO nanoparticles loaded in SPMs were investigated by transmission electron microscopy (TEM, JEM-2010HR, JEOL Ltd., Japan) operating at 100 kV. Samples were prepared by cutting the freeze-dried SPMs into slices and suspending them in alcohol. A drop of suspension was then deposited on a carbon-coated copper grid and left to dry under the ambient air (Chen et al., 2014; Xue et al., 2016).

2.6. X-ray diffraction

The X-ray diffraction (XRD) patterns of the freeze-dried SPMs was analyzed by an X-ray diffractometer (Mini Flex 600 X, Rigaku, Japan) with a Cu Kα radiation source (λ=1.54 Å) at 40 kV and 15 mA scanning from 10 ° to 80 ° at a scan rate of 4 °/ min. By comparing the resultant XRD spectrum with standard JCPDS cards, the crystalline phase of SPIO loaded in SPMs could be confirmed (Frounchi and Shamshiri, 2015; Kim et al., 2007).

2.7. X-ray photoelectron spectroscopy

To confirm the magnetite phase of SPIO loaded in SPMs, X-ray photoelectron spectroscopy (XPS) analysis of SPMs was further performed on an X-ray photoelectron spectrometer (Axis Ultra, Kratos, England) using monochromatic Al Kα radiation. Pass energy of 160 eV and step size of 1.00 eV were employed for survey spectrum of SPMs. For Fe 2p high-resolution spectrum, pass energy of 40 eV and step size of 0.10 eV were adopted (Jiang et al, 2012; Yang et al. 2014).

2.8. Fourier transform infrared spectrum

To illustrate the infrared characteristics of PMs and SPMs, freeze-dried PMs and SPMs were mixed with KBr and compressed into pellets, respectively (Ferreira et al., 2012; Xue et al., 2016). A Nicolet model NEXUS670 spectrometer (USA) was used to record the fourier transform

infrared (FT-IR) spectra in the scanning range of 4000-400 cm-1.

2.9. SPIO content

The SPIO content of SPMs in whole size and in different subgroups (100-300, 300-500, 500-700, and 700-900 μm) was determined separately by measuring the iron concentrations using ο-phenanthroline method (Oerlemans et al., 2013; Zhang et al., 2009). In brief, 0.5 ml of wet SPMs or 20 mg of freeze-dried SPMs was put in a mixed solution of 4 ml of concentrated hydrochloric acid (37%, w/w) and 1 ml of nitric acid (65%, w/w). The suspension was further ultrasonicated for 30 min allowing a complete digestion of the iron oxide and finally diluted to 100 ml. Then 3 ml of the diluted solution was mixed with 1 ml of hydroxylamine hydrochloride solution (10%, w/w), followed by adding 2 ml of ο-phenanthroline (0.15%, w/w) to form an orange-red complex. Subsequently, the pH of the mixture was adjusted to about 4 by adding 5 ml of sodium acetate solution (1 mol/L). The sample was finally diluted to the 100 ml and the absorbance at 510 nm was determined using a UV-visible spectrometer (UV-1100, Mapada Instruments Co., Ltd. China). The iron concentrations in SPMs were calculated from a calibration curve constructed from the absorbance and the iron solution with known concentrations from 0.8 to 2.8 μg/ml. The SPIO content was expressed as the weight of magnetite per volume of wet SPMs or the weight percentage of magnetite in freeze-dried SPMs.

2.10. Morphology and size distribution

The morphology of PMs and SPMswere observed by both optical microscope (XTZ-D/T, Shanghai Optical Instrument Factory No. 6) and environmental scanning electron microscope (ESEM) (FEI Quanta 200F, EDAX/AME-TEK, USA). The particle size of PMs and SPMs were determined respectively by measuring the diameters of at least 1000 individual microspheres from optical micrographs. The size distribution graphs of PMs and SPMs were drawn and compared. The number-average diameter (Dn) of microspheres was expressed as Eq. (1) (Zhou et al., 2012).

Dn   ni d i /  ni

(1)

Where di is the diameter of individual microsphere, and ni is the number of microspheres with di.

2.11. Elasticity

The PMs and SPMs both in the subgroup of 700-900 m were further sieved to the size range of 700-750 m for the determination of elasticity. All microspheres were chosen within this range. A series of tests was performed using a texture analyzer (TA. XTPlus, Stable Micro Systems, UK) and carried out according to the method described by Cui et al (Cui et al., 2012). The analyzer was equipped with a 6 mm cylindrical probe and a 5 N load cell. The initial force was set at 0.001 N. Both compression force and displacement of probe were collected 100 times per second by a computer connected to the analyzer. PMs and SPMs were used as samples and Embospheres were used as a contrast. All investigations were performed at least three times. In general, the following four tests were conducted, including compression test for monolayer of microspheres, stress relaxation test for monolayer of microspheres, compression test for single microspheres, and repeated compression test for single microspheres. Based on the results, the compression curves was recorded, the Young’s modulus and relaxation half time (RHT) were calculated, and the resisting capability as well as the recovery properties of individual microspheres against great stress and deformation were evaluated.

2.12. In vitro MRI study in gel phantom

To determine the MRI visibility of SPMs in vitro and investigate the effects of particle sizes and concentrations on the MRI signals, a gel phantom study was carried out using a clinical 3.0 T MRI scanner (Magnetom Trio Tim, Siemens, Garmany)(Choi et al., 2014; Franklin-Ford et al., 2012). In brief, 0.5 ml of 1% agarose solution (w/v) was pipetted into 1.5 ml Eppendorf tubes and allowed to gel to a foundation. Then, a certain volume of SPMs in the different size ranges (100-300, 300-500, 500-700 and 700-900 μm) were evenly suspended in preheated 1% agarose solution to the final concentrations of 0%, 2.5%, 5%, 10% and 20% (v/v), and respectively transferred to the Eppendorf tubes mentioned above, cooling to form a gel with SPMs suspended in. The samples of PMs in different subgroups with the concentration of 20% were also prepared

as controls. To obtained the T2-weighted MR images of the gel phantom within the Eppendorf tubes, a fast spin-echo (TSE) sequence was used with the following parameters: repetition time (TR) of 3000 ms, echo time (TE) of 88 ms, slice thickness of 3.0 mm, field of view of 120×120 mm, flip angle of 150°, matrix of 179×256, and number of excitation of 2.

2.13. In vivo MRI study in mice

The MRIdetectabilityof SPMs in vivo was evaluated in mice with an approval of the Animal Care Committee of Peking University Health Science Center. Three male adult Kun Ming mice with body weight of about 25 g were purchased from the Peking University Experimental Animal Center (Beijing, China) and housed under standard conditions with free access to food and water. The MRI measurement was performed with a 3.0 T MR scanner (Magnetom Trio Tim, Siemens, Garmany) with a wrist coil. Each mouse was anesthetized by intraperitoneal injection of pentobarbital sodium (1%, w/v, 75 mg/kg body weight). The baseline scans for mice were performed immediately prior to the injection of SPMs. Then, 0.2 ml of SPMs suspension composed of SPMs (100-300 μm) and sodium carboxylmethyl cellulose solution (1%, w/v) at a volume ratio of 1:7 was injected subcutaneously into the back of each mouse. Subsequently, follow-up MRI scans were performed immediately and at 14 d after the injection using multi-slice T2-weighted TSE sequence with TR of 4710 ms, TE of 88 ms, slice thickness of 3.0 mm, field of view of 120×120 mm, flip angle of 150 °, matrix of 179×256, and number of excitation of 2.

2.14. Statistical analysis

All experiments were performed at least in triplicate. All quantitative data were expressed as mean value and standard deviation (SD). Student’s t-test was used for the comparison of means and nonparametric Wilcoxon signed-rank test was used for the comparison between two groups. A SPSS statistical package (version 20.0; IBM Corp., Armonk, New York, USA) was used for the statistical analysis. Differences were considered not significant if P value was greater than 0.05 and highly significant if P value was less than 0.01.

Results and discussion

3.1. Preparation of SPMs

PMs were prepared initially by inverse suspension polymerization method as our previous report (Cui et al., 2012). Then PMs were immersed in the solution composed of di- and trivalent iron slat at a molar ratio of 1:2 to produce SPMs by in situ precipitation. Under this condition, the magnetic particles resulted in PMs were supposed usually to be in the form of Fe3O4 (Kim et al., 2007). In this way, the amount of SPIO loaded in SPMs could be readily adjusted by varying the concentration of iron salts solution and the incubation time with PMs within a reasonable range.

3.2. Magnetic property

The magnetization curve of SPMs studied by VSM was presented in Fig. 1. The superparamagnetic behavior of SPMs was clearly demonstrated with no hysteresis loop shown in the figure and both remanence magnetization and coercivity approximating zero. The superparamagnetism of magnetic particles has been reported to be size-dependent and generally arise when the particle size is reduced to nanoscale (Ramimoghadam et al., 2015; Wahajuddin and Arora, 2012). Thus, it was speculated that the magnetic particles loaded in SPMs were nano-sized. By fitting the magnetization curve of SPMs to a Langevin function, the magnetic moment of the SPIO nanoparticles loaded in SPMs was calculated to be 1.56 ×10-20 Nβ (where N was Avogadro number and β was bohrmagneton). With the assumption that the magnetic phase of loaded SPIO was magnetite, the effective particle size of SPIO was reckoned to be 6.92 nm. The saturation magnetization of SPMs at room temperature was determined to be 1.66 emu/g. In other literatures, the saturation magnetization of magnetic γ-Fe2O3/P (MAOETIB-GMA) microparticles (Hagit et al., 2010), Fe3O4/PLGA microcapsules (Sun et al., 2012) and Fe3O4-GMH microspheres (Xue et al., 2016) were reported to be 0.67, 1.00 and 5.44 emu/g, respectively. Comparing with these results, it was indicated that the magnetic property of SPMs was in a relatively moderate level. Besides, it was reported that the saturation magnetization of magnetic microspheres was proportionate with the concentration of SPIO within the microspheres (Frounchi

and Shamshiri, 2015). In our case, the saturation magnetization of SPMs could be further increased by increasing the content of SPIO embedded in the SPMs.

3.3. Transmission electron microscopy

The TEM image of SPMs was shown in Fig. 2. The SPIO loaded in SPMs was represented as dark regions in the figure. It could be seen clearly that SPIO nanoparticles distributed uniformly in the SPMs with diameters ranging from 10 to 15 nm. The results of TEM confirmed the speculations above that the SPIO loaded in SPMs was nano-sized based on its superparamagnetism and distributed throughout the SPMs presumably due to the stable effect of poly (acrylic acid).

3.4. X-ray diffraction

The XRD pattern of SPMs was shown in Fig. 3. Although the diffraction peaks in the XRD spectrum were not obvious due to the existence of large amount of amorphous polymers, the iron oxide particles loaded in SPMs could be confirmed in inverse spinel structure with the characteristic peaks at 2θ of 30.3°, 35.7°, 43.2°, 57.3° and 62.9°, which could be indexed to the (220), (311), (400), (511) and (440) lattice planes of Fe3O4 (JCPDS card No. 85-1436), respectively (Peng et al., 2015). Among various phases of the iron oxide, Fe3O4 and γ-Fe2O3 were reported to be in inverse spinel structure and both of them could be used as contrast agent for MRI (Lee et al., 2015; Ramimoghadam et al., 2015). Moreover, the superparamagnetism of SPIO nanoparticles was associated with its crystalline structure which could facilitate the alignment of adjoining spins to the external magnetic fields (Ferrucci and Stark, 1990).

3.5. X-ray photoelectron spectroscopy Due to the similarity in the XRD patterns of Fe3O4 and γ-Fe2O3 (Maity and Agrawal, 2007), XPS spectra were further performed to verify the compositions of SPIO loaded in SPMs. The XPS survey spectrum of SPMs was shown in Fig. 4a. The main peaks could be ascribed to Fe, C, O and S elements as labeled in the figure. The Fe 2p spectrum of SPIO loaded in SPMs was shown in Fig.

4b. The two broad peaks located at 710.4 eV and 723.8 eV were respectively ascribed to the Fe 2p3/2 and Fe 2p1/2 of Fe3O4. Usually, the Fe3+ in γ-Fe2O3 exhibits a shakeup satellite peak around 719.0 eV. The absence of this characteristic peak in the XPS spectra confirmed the pure magnetite phase in SPMs (Gao and Chambers, 1997; Yang et al., 2014).

3.6. Fourier transform infrared spectrum

The FT-IR spectra of PMs and SPMs were respectively shown in Fig. 5. For PMs (Fig. 5A), the absorption peaks at 3456 cm

-1

and 1723 cm

-1

were respectively attributed to the O-H

stretching vibration and the C=O bond of carboxyl group. Furthermore, the peak at 2932 cm -1 was assigned to the stretching vibration of -CH2-, which suggested the polymerization of double bonds (-C=C-) of acrylic group. All of these absorption peaks were fully in accordance with our previous report (Cui et al., 2012). In the spectrum of SPMs (Fig. 5B), the peaks at 3434, 2929, and 1720 cm -1

were respectively assigned to the O-H stretching vibration, -CH2- stretching vibration and the

C=O bond as PMs. Besides, a characteristic absorption peak assigned to Fe-O stretching vibration was observed at 626 cm -1, which was consistent with the reported spectrum of SPIO nanoparticles (Xue et al., 2016). However, the other characteristic absorption of Fe-O bond at 568 cm -1 around was not observed due to the relatively low intensity (Anbarasu et al., 2015; Xue et al., 2016). Combining the results of magnetization curve, TEM, XRD pattern, XPS and FT-IR spectrum of SPMs, it was demonstrated that SPIO nanoparticles was successfully loaded into PMs by in situ precipitation method.

3.7. SPIO content

In practical clinical embolization, embolic microspheres are generally calibrated to various size ranges for occlusion of vessels with different diameters. Therefore, we measured the SPIO content in different subgroups aiming to provide more information for potential clinical applications. Table 1 showed the SPIO content of SPMs in different subgroups. The SPIO content in wet SPMs of 100-300 μm, 300-500 μm, 500-700 μm and 700-900 μm was 11.84 ± 0.07, 10.20 ± 0.05,

9.98 ± 0.00 and 8.79 ± 0.01 mg/ml, respectively. The content of SPIO in wet SPMs decreased with the particle size increasing. This phenomenon could be explained by the existence of interspace among the individual microsphere, i.e. with the same packing volume of SPMs, bigger size of SPMs left more interspace than the smaller ones (Meng et al., 2015). The SPIO content per volume of wet SPMs could provide valuable reference for clinical use, because the radiologists concern more about the volume of wet microspheres for embolization and SPMs in blood vessels for MRI detection were wet and in stacking state. In Table 1, the weight percentage of SPIO content in freeze-dried SPMs of 100-300 μm, 300-500 μm, 500-700 μm and 700-900 μm was 18.07 ± 0.28, 18.54 ± 0.13, 18.66 ± 0.01 and 18.50 ± 0.07 %, respectively. The results demonstrated that the SPIO contents in one batch of SPMs prepared by in situ precipitation were almost the same between different subgroups. Compared with SPIO content of per volume of wet SPMs, the weight percentage of SPIO content in freeze-dried SPMs presented more information for chemical characterization of SPMs than for clinical application. The SPIO content of SPMs in whole size was measured to be 18.25 ± 0.37 % (per weight of freeze-dried SPMs). In the previous studies, the weight percentage of SPIO in SPIOs-embedded chitosan microspheres (Kang et al., 2009), iron oxide nanoparticles-encapsulated PLGA microspheres(Chen et al., 2014) and magnetic nanoparticles-loaded PLA/PEG microspheres (Frounchi and Shamshiri, 2015) were reported to be 0.26-0.32%, 0.89% and 5-25%, respectively. Hence, the SPIO content of SPMs in this study indicated a relatively high level, which would ensure a successful MRI detectability of SPMs.

3.8. Morphology and size distribution

Fig. 6A and 6B showed the optical macrographs of PMs and SPMs. The pictures indicated that both PMs and SPMs were spherical in shape and highly dispersed.The spherical microspheres have been proved to allow more distal vascular penetration and produce more homogeneous and complete occlusion (Flandroy et al., 1990), consequently minimize the undesirable inflammation of the embolized vessel walls and vascular recanalization(Andrew et al., 2003).SPMs were observed to be orange under optical microscope while to be dark-brown by macroscopy. The

appearance of SPMs different from PMs was attributed to the loading of SPIO in the matrix of the microspheres. The ESEM images (Fig. 6C and 6D) demonstrated that the surface of PMs and SPMs were both smooth, which enabled PMs and SPMs more readily to be calibrated than irregular particles (Laurent, 2007). The size distribution of PMs and SPMs were shown in Fig. 7. Both PMs and SPMs had a broad size distribution from 100 to 900 μm and mostly distributed in the range of 100-300 μm. No significant difference was found between the size distribution of PMs and SPMs in this experiment by nonparametric Wilcoxon signed-rank test (P>0.05). The number-average diameters of PMs and SPMs were measured and calculated as248±104 μm and 242±132 μm, respectively.

3.9. Elasticity

Fig. 8 showed the compression curves of PMs, SPMs and Embospheres, respectively. At the same deformation, SPMs showed greater force than PMs, implying the loading of SPIO nanoparticles could lead to the increase of rigidity of microspheres. SPMs had a significantly higher Young’s modulus (477.13 ± 57.11 kPa) than Embospheres (158.27 ± 13.54 kPa) and PMs (95.38 ± 12.07 kPa) as shown in Table 2. The reason for this result may be attributed to the stiff mechanical property of SPIO nanoparticles in SPMs. The stiffer microspheres is considered to occlude more proximally than soft ones, however, regarding the comprehensive property of different embolic materials, it is not sufficient to determine the in vivo degree of occlusionjust based on the rigidity of microspheres (Hidaka et al., 2011). As shown in Table 2, when the probe stopped and held still in an equilibrium state, SPMs showed a greater residual force (59.11 ± 0.37%) than PMs (41.25 ± 2.08%) and Embospheres (39.7 ± 1.6%). It was speculated that the movement of polymer chains in SPMs were greatly hinder by the SPIO nanoparticles loaded. The RHT for PMs, SPMs and Embospheres were respectively calculated to be 20.69 ± 1.73 s, 10.53 ± 1.75 s and 62.13 ± 1.06 s, suggesting that SPMs achieved equilibrium state much faster than PMs and Embospheres. Results of compression test and repeated compression test for single microspheres were also shown in Table 2. During the compression test, none of the PMs and SPMs was compressed to be broken, suggesting both PMs and SPMs were safe during embolization with low risk to be broken

into small particles. There was no significant difference in springiness, cohesiveness and resilience between PMs, SPMs and Embospheres (P > 0.05), indicating the total recovery properties of PMs, SPMs and Embospheres were similar. The arterial delivery behaviors of microspheres are closely related to their elasticity(Kim et al., 2013). However, as far as we know, the systematic study related to the elasticity of magnetic embolic microspheres has not yet been reported. In this study, it was indicated that the loading of SPIO nanoparticles would affect the elasticity of PMs and make them more rigid. Thus, it reminds us that if increasing the SPIO content, the elasticity of SPMs should be balanced in the future.

3.10. In vitro MRI study in gel phantom

Fig. 9 showed the T2-weighted MR images of PMs and SPMs in agar gels. The SPMs in all subgroups could be detected clearly with low signal intensity in circular areas, while the PMs could not be detected due to their similar signal intensity with blank gel. The decrease of signal intensity in T2-weighted images demonstrated that SPMs could provide negative contrast enhancement on transverse proton relaxation time-weighted sequence, which originated from the dipolar interaction between the magnetic moment of SPIO nanoparticles loaded in SPMs and the protons in the water (Sun et al., 2012). The MR signal intensity of SPMs in all subgroups decreased with the increasing of SPMs concentration. Even at a relatively low concentration of 2.5%, distinct signal voids representing the clusters of SPMs could be identified on MR images. Moreover, at the same concentration, the MR signal intensity of smaller size of SPMs was darker than that of the bigger ones, which was consistent with the result of the SPIO content of wet SPMs in section 3.7. In some reports, the MRI detectability of microspheres was usually not identical in different size ranges and especially the small-sized microspheres could only be vaguely detected (Choi et al., 2014; Namur et al., 2007). While in this study, the small-sized SPMs exhibited more clear MRI detectability than the bigger ones, which was superior to the previous reported studies. Thus, showing excellent detectability in vitro phantom study, the SPMs were supposed to be potential visible embolic agent by MRI for clinical applications.

3.11. In vivo MRI study in mice

Since the PMs were proved to be undetectable by T2-weighted MR imaging through in vitro phantom study, only SPMs were injected into the back of mice to investigate the in vivo MR detectability. According to the previous studies, the in vivo MRI detectability of microspheres is usually investigated by comparing the MR images of the same tissues or organs scanned before and after the injection of microspheres. If the signal intensity of MR images became significantly darker (for SPIO) after injection, it indicated that the microspheres could be detected by MRI (Chen et al., 2014; Franklin-Ford et al., 2012). Fig. 10 showed the typical T2-weighted MR images of a mouse before and after subcutaneous injection of SPMs. Dark signal area (highlighted by a white circle) was detected around the subcutaneous tissue of the experimental mouse immediately and at 14 d after injection of SPMs, owing to the transverse relaxation time shortening effect of SPIO in the SPMs. Thus, the result demonstrated that SPMs could be efficiently detected in vivo under T2-weighted MR imaging and the MRI detectability could maintain at least for 14 days.

4. Conclusions

In this study, MRI detectable SPMs were successfully prepared by loading SPIO nanoparticles into PMs using in situ precipitation method. The SPIO loaded in SPMs were proved to be magnetite exhibiting superparamagnetic behavior with inverse spinel structure. The resultant SPMs possessed excellent roundness with smooth surface and the particle size was within the range of clinical demands for embolization. It was showed that due to the loading of SPIO nanoparticles, SPMs were more rigid than PMs and commercially used Embospheres. Both in vitro and in vivo MRI study showed that SPMs could induce strong hypointense signal in T2-weighted MR images and be sensitively detected by MRI. On the basis of these study results, SPMs are shown to be a potential MRI detectable embolic agent to be used for TAE and exactly monitor the spatial distribution of embolic microspheres inside the target tissues during and after embolization. Further evaluation of in vivo intravascular embolization and MRI detectability of SPMs will be performed in the near future.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (Grant No 81571779). The authors thank Xiao-nan Yao of College of Chemistry and Molecular Engineering of Peking University a lot for his kind help in calculating the effective particle size of SPIO based on the magnetization curve.

References

Andrews, R.T., Binkert, C.A., 2003.Relative rates of blood flow reduction during transcatheter arterial embolization with tris-acryl gelatin microspheres or polyvinyl alcohol: quantitative comparison in a swine model. J. Vasc. Interv. Radiol. 14, 1311-1316. Arnold, P., Ward, J., Wilson, D., Ashley Guthrie, J., Robinson, P.J., 2003. Superparamagnetic iron oxide (SPIO) enhancement in the cirrhotic liver: a comparison of two doses of ferumoxides in patients with advanced disease. Magn. Reson. Imaging 21, 695-700. Anbarasu, M., Anandan, M., Chinnasamy, E., Gopinath, V., Balamurugan, K., 2015.Synthesis and characterization of polyethylene glycol (PEG) coated Fe3O4 nanoparticles by chemical co-precipitation method for biomedical applications. Spectrochim. Acta A Mol. Biomol. Spectrosc. 135, 536-539. Bendszus, M., Martin-Schrader, I., Warmuth-Metz, M., Hofmann, E., Solymosi, L., 2000.MR imaging- and MR spectroscopy-revealed changes in meningiomas for which embolization was performed without subsequent surgery. AJNR Am. J. Neuroradiol. 21, 666-669. Chung, E.Y., Kim, H.M., Lee G.H., Kwak, B.K., Jung, J.S., Kuh, H.J., Lee, J., 2012.Design of deformable chitosan microspheres loaded with superparamagnetic iron oxide nanoparticles for embolotherapy detectable by magnetic resonance imaging. Carbohydr. Polym. 90, 1725-1731. Cui, D.C., Lu, W.L., Sa, E.A., Gu, M.J., Lu, X.J., Fan, T.Y., 2012. Poly(acrylic acid) microspheres loaded with lidocaine: preparation and characterization for arterial embolization. Int. J. Pharm. 436, 527-535. Chen, J., Sheu, A.Y., Li, W., Zhang, Z., Kim, D.H., Lewandowski, R.J., Omary, R.A., Shea, L.D., Larson, A.C., 2014. Poly(lactide-co-glycolide) microspheres for MRI-monitored transcatheter

delivery of sorafenib to liver tumors. J. Control. Release 184, 10-17. Choi, S.Y., Kwak, B.K., Shim, H.J., Lee, J., Hong, S.U., Kim, K.A., 2014. MRI traceability of superparamagnetic iron oxide nanoparticle-embedded chitosan microspheres as an embolic material in rabbit uterus. Diagn. Interv. Radiol. 21, 47-53. Ferrucci, J.T., Stark, D.D., 1990. Iron oxide-enhanced MR imaging of the liver and spleen: review of the first 5 years. AJR Am. J. Roentgenol. 155, 943-950. Flandroy, P., Grandfils, C., Collignon, J., Thibaut, A., Nihant, N., Barbette, S., Jerome, R., Teyssie, P., 1990. (D, L) Polylactide microspheres as embolic agent. Neuroradiology 32, 311-315. Fleetwood, I.G., Steinberg, G.K., 2002. Arteriovenous malformations. Lancet. 359, 863-873. Ferreira, G.R., Segura, T., de Souza Jr., F.G., Umpierre, A.P., Machado, F., 2012. Synthesis of poly(vinyl acetate)-based magnetic polymer microparticles. Eur. Polym. J. 48, 2050-2069. Franklin-Ford, T., Shah, N., Leiferman, E., Chamberlain, C.S., Raval, A., Vanderby, R., Murphy, W.L., 2012.Tracking injectable microspheres in dynamic tissues with encapsulated superparamagnetic iron oxide nanoparticles. Macromol. Biosci. 12, 1615-1621. Frounchi, M., Shamshiri, S., 2015.Magnetic nanoparticles-loaded PLA/PEG microspheres as drug carriers. J. Biomed. Mater. Res. A 103, 1893-1898. Gao, Y., Chambers, S.A., 1997. Heteroepitaxial growth of α-Fe2O3, γ-Fe2O3 and Fe3O4 thin films by oxygen-plasma-assisted molecular beam epitaxy. J. Ctyst. Growth 174, 446-454. Ge, J., Hu, Y., Biasini, M., Beyermann, W.P., Yin, Y., 2007.Superparamagnetic magnetite colloidal nanocrystal clusters. Angew. Chem. Int. Ed. Engl. 46, 4342-4345. Guziński, M., Kurcz, J., Bereza, S., Garcarek, J., Sasiadek, M., 2010. Application of cyanoacrylate glue and ethylene vinyl alcohol copolymer for the treatment of vascular malformations of the central nervous system. Polim. Med. 40, 57-62. Hagit, A., Soenke, B., Johannes, B., Shlomo, M., 2010.Synthesis and characterization of dual modality (CT/MRI) core-shell microparticles for embolization purposes. Biomacromolecules 11, 1600-1607. Hidaka, K., Moine, L., Collin, G., Labarre, D., Grossiord, J.L., Huang, N., Osuga, K., Waka, S., Laurent, A., 2011.Elasticity and viscoelasticity of embolization microspheres. J. Mech. Behav. Biomed. Mater. 4, 2161-2167. Jiang, H.H., Han, X.Y., Li, Z.L., Chen, X.C., Hou, Y.H., Gai, L.G., Li, D.C., Lu, X.R., Fu, T.L.,

2012. Superparamagentic core-shell structured microspheres carrying carboxyl groups as adsorbents for purification of genomic DNA. Colloids Surf. A 401, 74-80. Kim, E.H., Ahn, Y., Lee, H.S., 2007. Biomedical applications of superparamagnetic iron oxide nanoparticles encapsulated within chitosan. J. Alloys Compd. s434-435, 633-636. Kang, M.J., Oh, I.Y., Choi, B.C., Kwak, B.K., Lee, J., Choi., Y.W., 2009. Development of superparamagnetic iron oxide nanoparticles (SPIOs)-embedded chitosan microspheres for magnetic resonance (MR)-traceable embolotherapy. Biomol.Ther.17, 98-103. Khan, S.U., Rahman, K.M., Siddiqui, M.R., Hoque, M.A., Mondol, B.A., Hussain, S., Mohammad, Q.D., 2010. Endovascular embolization of life threatening intracranial arterio-venous malformation. Mymensingh. Med. J. 19, 438-441. Kim, H., Lee, G.H., Ro, J., Kuh, H.J., Kwak, B.K., Lee, J., 2013. Recoverability of freeze-dried doxorubicin-releasing chitosan embolic microspheres. J. Biomater. Sci. Polym. Ed. 24, 2081-2095. Lewis, A.L., Gonzalez, M.V., Lloyd, A.W., Hall, B., Tang, Y., Willis, S.L., Leppard, S.W., Wolfenden L.C., Palmer, R.R., Stratford, P.W., 2006. DC bead: in vitro characterization of a drug-delivery device for transarterial chemoembolization. J. Vasc. Interv. Radiol. 17, 335-342. Laurent, A., 2007. Microspheres and nonspherical particles for embolization. Tech. Vasc. Interv. Radiol. 10, 248-256. Lee, N., Yoo, D., Ling, D., Cho, M.H., Hyeon, T., Cheon, J., 2015. Iron oxide based nanoparticles for multimodal imaging and magnetoresponsive therapy. Chem. Res. 115 (19): 10637-10689. Murakami, R., Ichikawa, T., Kumazaki, T., Kobayashi, Y., Oqura, J., Kurokawa, A., 2000.Transcatheter arterial embolization for postpartum massive hemorrhage: A case report. Clin. Imaging 24, 368–370. Namur, J., Chapot, R., Pelage, J., Wassef, M., Langevin, F., Labarre, D., Laurent, A., 2007.MR imaging

detection

of

superparamagnetic

iron-oxide-loaded

tris-acryl

embolization

microspheres. J. Vasc. Interv. Radiol. 18, 1287-1295. Negussie, A.H., Dreher, M.R., Johnson, C.G., Tang Y.Q., Lewis, A.L., Storm, G., Sharma, K.V., Wood, B.J. 2015.Synthesis and characterization of image-able polyvinyl alcohol microspheres for image-guided chemoembolization. J. Mater. Sci.: Mater. Med. 26, 198.

Oerlemans, C., Seevinck, P.R., van de Maat, G.H., Boulkhrif, H., Bakker, C.J., Hennink, W.E., Nijsen, J. F., 2013. Alginate-lanthanide microspheres for MRI-guided embolotherapy. Acta Biomater. 9, 4681-4687. Oerlemans, C., Seevinck, P.R., Smits, M.L., Hennink, W.E., Bakker, C.J., van den Bosch, M.A., Nijsen, J.F., 2015. Holmium-lipiodol-alginate microspheres for fluoroscopy-guided embolotherapy and multimodality imaging. Int. J. Pharm. 482, 47-53. Pouponneau, P., Leroux, J., Martel, S., 2009. Magnetic nanoparticles encapsulated into biodegradable microparticles steered with an upgrated magnetic resonance imaging system for tomor chemoembolization. Biomaterials 30, 6327-6332. Peng, M.L., Li, H.L., Luo, Z.Y., Kong, J., Wan, Y.S., Zheng, L.M., Zhang, Q.L., Niu, H.X., Vermorken, A., Van de Ven, W., Chen, C., Zhang, X.K., Li, F.Q., Guo, L.L., Cui, Y.L., 2015. Dextran-coated superparamagnetic nanoparticles as potential cancer drug carriers in vivo. Nanascale 7, 11155-11162. Ramimoghadam, D., Bagheri, S., AbdHamid, S.B., 2015.Stable monodisperse nanomagnetic colloidal suspensions: An overview. Colloids Surf. B Biointerfaces133, 388-411. Skalicky, T., Treska, V., Sutnar, A., Liska, V., Duras, P., Slauf, F.,2010.Chemo-embolization of inoperable liver tumors. Bratisl. Lek. Listy. 111. 676-679. Stampfl, U., Sommer, C.M., Bellemann, N., Holzschuh, M., Kueller A., Bluemmel, J., Gehrig, T., Shevchenko, M., Kenngott, H., Kauczor, H.U., Radeleff, B., 2012. Multimodal visibility of a modified polyzene-F-coated spherical embolic agent for liver embolization: feasibility study in a porcine model. J. Vasc. Interv. Radiol. 23, 1225-1231. Sun, Y., Zheng, Y.Y., Ran, H.T., Zhou, Y., Shen, H.X., Chen, Y., Chen, H.R., Krupka, T.M., Li, A., Li, P., Wang, Z.B., Wang, Z.G., 2012. Superparamagnetic PLGA-iron oxide microcapsules for dual-modality US/MR imaging and high intensity focused US breast cancer ablation. Biomaterials 33, 5854-5864. Treska, V., Skalický, T., Sutnar, A., Liska, V., Ferda, J., Mírka, H., Slauf, F., Duras, P., Kreuzberg, B., 2010. Portal vein branch embolization in patients with primary inoperable liver tumors. Rozhl. Chir. 89, 456-460. Trohidou, K.H., 2015. Magnetic nanoparticles assemblies. Boca Raron: CRC Press. Ugelstad, J., Ellingsen, T., Berge, A., Helgee, O.B., 1988. Process for preparing magnetic polymer

particles. United State Patent No. 4774265. Van Elk, M., Ozbakir, B., Barten-Rijbroek, A.D., Storm, G., Nijsen, F., Hennink, W.E., Vermonden, T., Deckers, R., 2015.Alginate microspheres containing temperature sensitive-liposomes (TSL) for MR-guided embolization and triggered release of doxorubicin. Plos one 10, e0141626. Wahajuddin, Arora, S., 2012. Superparamagnetic iron oxide nanoparticles: magnetic nanoplantforms as drug carriers. Int. J. Nanomedicine. 7, 3445-3471. Xuan S.H., Wang, Y.J., Yu, J.C., Leung, K.C., 2009. Tuning the grain size and particle size of superparamagnetic Fe3O4 microsparticles. Chem. Mater. 21, 5079-5087. Xue, P., Gu, Y.H., Su, W.G., Shuai, H.H., Wang, J.L., 2016. In situ one-pot preparation of superparamagnetic hydrophilic porous microspheres for covalently immobilizing penicillin G acylase to synthesize amoxicillin. Appl. Surf. Sci. 362, 427-433. Yang, J.H., Zou, P., Yang, L.L., Cao, J., Sun, Y.F., Han, D.L., Yang, S., Wang, Z., Chen, G., Wang, B.J., Kong, X.W., 2014.A comprehensive study on the synthesis and paramagnetic properties of PEG-coated Fe3O4 nanoparticles. Appl. Surf. Sci. 303, 425-432. Zhang, Q., Wang, C.H., Qiao, L., Yan, H.S., Liu, K.L., 2009. Superparamagnetic iron oxide nanoparticles coated with a folate-conjugated polymer. J. Mater. Chem. 19, 8393-8402. Zhou, C., Cui, D.C., Zhang, Y., Yuan, Y.H., Fan, T.Y., 2012.Preparation and characterization of ketoprofen-loaded microspheres for embolization. J. Mater. Sci.: Mater. Med. 23, 409-418.

Figure Captions:

Fig.1. The magnetization curve of SPMs.

Fig.2. TEM image of SPMs.

Fig.3. The XRD pattern of SPMs.

Fig.4. The XPS spectra of SPMs. (A) XPS survey spectrum; and (B) Fe 2p XPS spectrum.

Fig.5. FT-IR spectra of (A) PMs and (B) SPMs.

Fig.6. Morphology of microspheres.(A) PMs under optical microscope (64×); (B) SPMs under optical microscope (64×); (C) PMs under ESEM; and (D) SPMs under ESEM.

Fig.7. Size distribution of PMs (blank column) and SPMs (black column).

Fig.8. Compression curves of different microspheres. Symbol: (open circle) PMs; (filled circle) SPMs; (filled rhombus) Embospheres.

Fig.9. T2-weighted MR images of in vitro gel phantom from left to right: PMs with the concentration of 20% (v/v) (as controls) and SPMs with concentrations of 0, 2.5%, 5%, 10% and 20% (v/v).

Fig.10. The typical T2-weighted MR images of the same mouse before, immediately after and 14 d after subcutaneous injection of SPMs into the back. The circle and arrow denote the dark signal area induced by SPMs.

Table 1 The content of SPIO in different subgroups of SPMs (n = 3). Subgroup (μm)

100-300

300-500

500-700

700-900

Content (mg/ml)

11.84 ± 0.07

10.20 ± 0.05

9.98 ± 0.00

8.79 ± 0.01

Content (wt%)

18.07 ± 0.28

18.54 ± 0.13

18.66 ± 0.01

18.50 ± 0.07

Table 2 Young’s modulus, residual force, relaxation half time (RHT), failure deformation, failure stress, springiness, cohesiveness and resilience of different microspheres (n = 3). Microspheres

PMs

SPMs

Embospheres

Young’s modulus (kPa)

95.38 ± 12.07

477.13 ± 57.11**

158.27 ± 13.54**

Residual force (%)

41.25 ± 2.08

59.11 ± 0.37**

39.7 ± 1.6

RHT (s)

20.69 ± 1.73

10.53 ± 1.75**

62.13 ± 1.06**

Failure deformation (%)

N.A.a

N.A.

82.1 ± 2.3

Failure stress (N)

N.A.

N.A.

0.45 ± 0.12

Springiness

0.73 ± 0.05

0.73 ± 0.04

0.71 ± 0.15

Cohesiveness

0.74 ± 0.03

0.74 ± 0.10

0.57 ± 0.12

Resilience

0.52 ± 0.03

0.52 ± 0.02

0.60 ± 0.15

Data was expressed as the mean ± SD. Statistical significance: **P < 0.01, *P < 0.05. a

Not applicable.