Abstract
Dual-modal imaging by combining magnetic resonance (MR) and near-infrared (NIR) fluorescence can integrate the advantages of high-resolution anatomical imaging with high sensitivity in vivo fluorescent imaging, which is expected to play a significant role in biomedical researches. Here we report a dual-modality imaging probe (NIR/MR-MSNs) fabricated by conjugating NIR fluorescent heptamethine dyes (IR-808) and MR contrast agents (Gd-DTPA) within highly aminated mesoporous silica nanoparticles (MSNs-NH2). The dual-modality imaging probes NIR/MR-MSNs possess a size of ca. 120 nm. The NIR/MR-MSNs show not only near-infrared fluorescence imaging property with an emission peak at 794 nm, but also highly MR relaxivity of 14.54 mM−1 s−1, which is three times more than Gd-DTPA. In vitro experiment reveals high uptake and retention abilities of the nanoprobes, while cell viability assay demonstrates excellent cytocompatibility of the dual-modality imaging probe. After intratumor injection with the NIR/MR-MSNs, MR imaging shows clear anatomical border of the enhanced tumor region while NIR fluorescence exhibits high sensitive tumor detection ability. These intriguing features suggest that this newly developed dual-modality imaging probes have great potential in biomedical imaging.
1. Introduction
Magnetic resonance (MR) and near-infrared (NIR) fluorescence have been widely used as powerful imaging tools in biomedical areas [1–6]. However, due to the inherent limitations of MR and NIR, it is still impossible to get molecular, functional, and anatomical information by signal imaging method [1, 2, 7, 8]. In order to take the synergistic advantages of each imaging modality, great efforts have been conducted to explore MR/NIR dual-modal imaging tools [3, 9–13]. For example, NIR dye has been modified onto MR-detectable nanomaterials such as superparamagnetic iron oxide nanoparticles (SPIONs) to build MR/NIR bimodal imaging probe [14–16]. However, the direction modification dyes on the SPIONs indeed do not have many disadvantages. First, the black SPIONs can directly quench the fluorescence of the dye. Second, the emitted fluorescence can also be absorbed by the SPIONs [14, 16]. In addition, surface modifications are commonly needed to connect functional molecules, which makes the SPION based dual-modality probes even more complicated [14]. Another method is conjugating MR agents with semiconductor quantum dots (QDs) [1, 17–19]. Yet, the drawback of the toxicity of heavy cadmium metal makes QD based imaging probes potential clinical usage limited [1].
Recently, mesoporous silica nanoparticles (MSNs) have attracted increasing attention as an ideal matrix to integrate imaging tags to develop imaging probes, due to the high surface area, uniform pore size, excellent biocompatibility, and no absorption of NIR light or interference with magnetic fields [11, 17, 20, 21]. There have been several reports about MSN based imaging probes [17, 21–23]. Taylor et al. [23] synthesized MSN based nanoparticulate weighted MR contrast agents with a high relaxivity by refluxing MSNs and Gd-Si-DTTA complex in toluene. Although rodent experiment confirmed the MR imaging ability of the particles, the authors failed to evaluate the optical fluorescent property in vivo [23]. Sharma et al. [22] produced gadolinium-doped silica nanoparticles by encapsulation indocyanine green with a reverse-microemulsion-based synthesis protocol; both in vitro and in vivo studies showed fine NIR and MR contrast capabilities. However, the protocol is complex and the relaxivity elevation was only 7.3 mM−1 s−1 on a 4.7 T MR. Till now, there very few reports on the high MR relaxivity and NIR imaging probes have been documented.
In this study, we designed and synthesized a novel NIR/MR dual-modality imaging probe (NIR/MR-MSNs) by facile conjugation of NIR fluorescent heptamethine dyes IR-808 and MR contrast agents Gd-DTPA within highly aminated MSNs (MSNs-NH2) via a one-step procedure. The NIR/MR-MSNs not only were easy to produce, but also possessed uniform particle size, high surface area, and excellent biocompatibility. In vitro cell experiment illustrated high cellular uptake, retention abilities, and excellent biocompatibility of the NIR/MR-MSNs. Excellent NIR fluorescence property and highly efficient MR contrast effect were also found in in vitro and in vivo imaging studies. Importantly, the NIR/MR-MSNs provide simultaneously clear anatomical border of enhanced tumor region and high sensitive fluorescence imaging. These all suggested that the NIR/MR-MSNs are suitable for potential applications as a useful dual-modality biomedical imaging probes.
2. Materials and Methods
2.1. Materials
Hexadecyltrimethyl ammonium bromide (CTAB), tetraethyl orthosilicate (TEOS), and N-hydroxysulfosuccinimide sodium salt (NHS) were purchased from Sinopharm Chemical Reagent Co. (China). Ethyl ethanoate (EtOAc), ammonium hydroxide in water (25 wt%), and ethyl alcohol were purchased from Nanjing Chemical Reagent Co. (China). 3-Aminopropyltriethoxysilane (APTES) was purchased from Sigma-Aldrich (USA). 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC·HCl) was purchased from Aladdin Industrial Inc. (China). 2-(N-Morpholino) ethanesulfonic acid (MES) sodium salt buffer solution was purchased from Thermo Fisher Scientific Inc. (USA). Gd-DTPA was purchased from Guangzhou Consun Pharmaceutical Group Ltd. (China). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was purchased from Sigma-Aldrich (USA). The near-infrared fluorescent heptamethine indocyanine dye (IR-808) was presented by Professor Chun-meng Shi of the Third Military Medical University.
2.2. Synthesis of MSNs-NH2
MSNs-NH2 were synthesized according to the previously reported method [24]. In brief, 7.04 mL of ethyl acetate, 24.24 mL of ammonium hydroxide, and 0.8 g of CTAB were added to 1135 mL deionized water and stirred at room temperature until CTAB completely dissolved. Afterward, 1.8 mL of TEOS and 2.2 mL of APTES were added under vigorous stirring. After 24 h, the products were centrifuged and washed with deionized water. Finally, the CTAB was removed by three extractions in 240 mL of ethanol and 480 μL of concentrated hydrochloric acid at 60°C for 3 h.
2.3. Synthesis of Gd/NIR-MSNs
1 mL of MSNs-NH2 solution (50 mg/mL), 0.7 mg IR-808, and 4 mL Gd-DTPA aqueous solution (0.1 M) were mixed in MES-NHS buffer and shattered by an ultrasonic cell disrupter for 30 min in dark. Then, EDC·HCl was added in the solution and stirred in dark overnight. The sample was washed with water and redispersed in deionized water.
2.4. Characterization of Gd/NIR-MSNs
Transmission electron microscopy (TEM) images were obtained with a Tecnai G2 F20 microscope (FEI, Hillsboro, OR) at 200 KV. Nitrogen sorption isotherms were obtained with a V-Sorb 2800P physisorption instrument (Gold APP Instrument Corporation, China). Surface areas were calculated according to the Brunauer-Emmett-Teller (BET) method. The pore size distributions were calculated from the adsorption branches of the isotherms according to the nonlocal density functional theory (NLDFT) method. The absorption curve of Gd/NIR-MSNs and loading efficiency of IR-808 were determined by UV-vis spectrum with a UV-vis Lambda 35 (PerkinElmer, USA). To confirm the presence of Gd-DTPA in Gd/NIR-MSNs, electron dispersive X-ray (EDX) was performed using a Tecnai G2 F20 electron microscopy with an EDX detector system. Gd concentration was determined with a Perkin-Elmer OPTIMA 5300 DV (PerkinElmer, USA) inductively coupled plasma atomic emission spectroscopy (ICP-AES). MR examination was performed with a 3 T clinical MR scanner (MAGNETOM Trio, SIMENS, Germany) at room temperature. After acquiring the map MR images, values were measured by manually drawn regions of interest for each sample. Relaxation rates () were calculated from values at different Gd concentrations.
2.5. Cellular Experiments
2.5.1. Cell Culture
Human embryonic kidney HEK 293T cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen, California, USA) supplemented with 1 mmol/L L-glutamine and 10% fetal bovine serum (FBS; Life Technologies, Inc. Burlington, Canada). Human glioblastoma cell line U87-MG-GFP cells were maintained in Dulbecco’s modified eagle medium (DMEM; KeyGEN Bio TECH, Nanjing, China) supplemented with 10% FBS. All media were added with antibiotics (100 U/mL of penicillin G and 100 μg/mL of streptomycin). All cells were incubated at 37°C in a humidified atmosphere of 95% air and 5% CO2.
2.5.2. In Vitro Fluorescent Microscopy and MR Imaging of U87-MG Cell Labeling with Gd/NIR-MSNs
Green fluorescent protein (GFP) labeling human glioblastoma cell line U87-MG were passaged 1 day before adding Gd/NIR-MSNs and inoculated in a 24-well chamber slides. After the cells attached to the chamber slides, stock solutions of Gd/NIR-MSNs were centrifuged and redispersed in fresh growth media and added to cultured cells to a final concentration of 0.1 mg/mL. Cells were incubated at 37°C overnight. The next day, label-containing media were aspirated carefully. Cells were washed five times with phosphate buffered saline solution (PBS) to remove excess nanoparticles and fixed with 10% formaldehyde at 4°C. After supernatant removed, 4′,6-diamidino-2-phenylindole (DAPI) was added to each chamber and incubated for 20 min. The slides were then washed three times with PBS and covered with glass coverslips.
Fluorescent images were recorded by a fluorescent inverted microscopy (Olympus, Japan). Imaging of DAPI was carried out with excitation of 330~385 nm and emission at 420 nm, while imaging of GFP was performed with excitation of 460~490 nm and emission of 510 nm. Gd/NIR-MSNs were excited by the external laser at 950 nm, while the emissions were collected in the ranges of 684 to 719 nm. To identify the Gd/NIR-MSNs uptaken by cells, fluorescent microscopy images were overlaid.
To further evaluate the dual-modality imaging ability of the Gd/NIR-MSNs, we designed an in vitro phantom experiment. In brief, U87-MG cells were incubated with Gd/NIR-MSNs or MSNs-NH2 containing medium at a density of 5 × 106 cells/mL in 6 cm culture dishes at 37°C for 4 h, respectively. And then, the labeled cells were harvested and washed five times by centrifuge to remove additional nanoparticles. NIR imaging was performed using an IVIS Lumina XR system (Xenogen, USA), with the emission collected 808 nm and excited by the external laser at 778 nm. MR map imaging was performed with a 3.0 T clinical MR scanner and a custom, premade four-channel mouse coil. The parameters were as follows: TR = 15.0 ms, TE = 1.73 ms, Average = 2, Slice Thickness: 2.0 mm, FOV = 113 mm, and Flip Angle = 5, 26.
2.5.3. Cell Cytotoxicity Evaluation of Gd/NIR-MSNs
HEK 293T cells were passaged when they are at confluence of 70% and inoculated to a 96-well plate (cell density 1 × 104 cells/well) in growth media (100 μL). Cells were cultured for 24 h before experiments. For the dose-dependent assay, Gd/NIR-MSNs were centrifuged and redispersed in fresh growth media and added to cultured cells to a final concentration of 0, 20, 40, 100, and 200 μg/mL. After 24 h of incubation at 37°C, supernatant was carefully removed and 100 μL medium and 20 μL MTT solution (5 mg/mL) were added and incubated for 4 additional hours. The supernatant in each well was aspirated and 150 μL dimethyl sulfoxide (DMSO) was added to solubilize cells and MTT crystals. After 4 h of shaking on a shaking table at room temperature and dark to dissolve all crystals, the blue color was read in a multiwall scanning spectrophotometer (ELx800, BioTek, USA) at a wavelength of 540 nm.
2.6. Animal Experiment
2.6.1. Animal Model
The animal study protocol was approved by the Animal Ethics Committee of Jinling Hospital. Female BALB/c nude mice (3-4 weeks old, 20–25 g) were purchased from the Comparative Medical Department of Jinling Hospital. U87-MG cells were harvested at 70% confluence and resuspended in PBS. Mouse tumor xenografts were established by U87-MG cells (2.0 × 106) inoculation into the right hind limbs of nude mice.
2.6.2. In Vivo Dual-Modal NIR and MR Imaging
U87-MG tumor bearing mice were anesthetized with 2% pentobarbital sodium and received an intratumor injection of 15 μL of Gd/NIR-MSNs PBS solution (5 mg/mL). MRI was conducted on a 3.0 T clinical MR scanner and a custom, premade four-channel mouse coil, using a coronal and transversal weighted sequence (TR = 670 ms, TE = 11 ms, Average = 4, Slice Thickness = 2.0 mm, FOV = 60 mm, and Flip Angle = 150). The nude mice were scanned before and after the intratumor injection of Gd/NIR-MSNs PBS solution (50 μL, 5 mg/mL). NIR imaging was performed using an IVIS Lumina XR system, with the emission collected at 808 nm and excited by the external laser at 778 nm. MR and NIR imaging were performed at different time points after injection (0, 4, 8, 24, and 48 h).
3. Results and Discussion
The synthesis procedures of the Gd/NIR-MSNs are shown in Figure 1. Firstly, MSNs-NH2 were synthesized via a CTAB-directed sol-gel procedure by using TEOS and APTES as coprecursors. Secondly, CTAB was removed with ethanol and concentrated hydrochloric acid. Thirdly, NIR dye IR-808 and MR contrast agents Gd-DTPA were conjugated to MSNs-NH2 simultaneously with condensation reaction between -NH2 moieties of MSNs-NH2 and -COOH groups of IR-808 as well as Gd-DTPA, resulting in the dual-modality imaging probes Gd/NIR-MSNs.
TEM images show that MSNs-NH2 have a truncated-octahedral shape with a mean diameter about 120 nm (Figures 2(a)–2(c)). It was also observed that the particles consist of highly ordered mesostructure (Figure 2(b)). To further determine the porous features, nitrogen adsorption-desorption isotherm assay was conducted and uncovered a type IV isotherm according to the IUPAC nomenclature (Figure 2(d)). The sharp capillary condensation step and large hysteresis loop in the range of 0.45–1.0 reveal characteristics of mesoporous materials with a narrow pore size distribution. The pore size distribution calculated based on the nonlocal density functional theory (NLDFT) illustrated that MSNs-NH2 have uniform mesopore size of 3 and 6 nm and a total pore volume of 0.53 cm3 g−1.
(a)
(b)
(c)
(d)
(e)
(f)
After conjugation of IR-808 dyes and Gd-DTPA to MSNs-NH2, BET results uncovered the nanoparticle surface area reduction from 473 m2/g to 200 m2/g, while the zeta potential is reduced from mV to mV. These features suggested the successful conjugation of the NIR and MR agents with MSN-NH2. UV-vis absorption spectroscope and EDX spectrum measurement were also employed to confirm the presence of IR-808 and Gd-DTPA on Gd/NIR-MSNs particles. The absorption curve of Gd/NIR-MSNs illustrated a peak at 779 nm (Figure 2(e)), which is similar to the absorption peak of IR-808 and suggested the successful connection of IR-808 dyes with MSN-NH2 [25]. Another peak documented by UV-vis absorption at 680 nm was also observed, which is attributed to Gd-DTPA on the particles. UV-vis absorption spectroscope analysis also showed that Gd/NIR-MSNs particles have an IR-808 dye loading amount of 0.3 wt%. EDX spectrum result provided direct evidence of Gd3+ presence on Gd/NIR-MSNs particles (Figure 2(f)), while ICP-AES measurements demonstrated 0.045 wt% conjugation amount for Gd-DTPA. The conjugation rate of Gd3+ is lower than previous report (3.42 wt%) [23, 26].
Strong fluorescence signals from NIR Gd/NIR-MSNs in deionized water were also identified (Figure 3(a)). The fluorescence emission spectra of the Gd/NIR-MSNs demonstrated a peak at 794 nm, which was similar with NIR dye IR-808 under the same condition [25]. These features suggesting the spectral properties of IR-808 are unchanged during the synthesis procedure, which will validate the potential ability of the bimodal probes for NIR imaging. As shown in Figure 3(b), in vitro phantom MRI study demonstrated a clear Gd3+ concentration dependent signal enhancement effect with a high value of 14.54 mM−1 s−1. The relaxivity value is much larger than Gd-DTPA (4.75 mM−1 s−1) under the same condition. The high magnitude relaxivities of Gd/NIR-MSNs are attributed to Gd3+ chelates attached to the mesochannels of the MSNs-NH2, which makes readily accession and efficient relax of water protons through a reduction in the tumbling rate [10, 23].
(a)
(b)
Before proceeding to in vivo applications for NIR and MR dual-modal imaging, the uptake and retention ability of Gd/NIR-MSNs were evaluated in vitro with human glioblastoma cells U87-MG, which were stably transfected with green fluorescence protein. Figures 4(a)–4(d) demonstrated that the particles (red) distributed in the cytoplasm of U87-MG cells. TEM (Figure 4(e)) image further confirmed the uptake of the Gd/NIR-MSNs by the U87-MG cells. Biocompatibility is one of the most important prerequisites of nanoprobers for in vivo imaging. We chose human embryonic kidney cells HEK 293T to assess the cytotoxicity of Gd/NIR-MSNs with a MTT method. The results showed that no significant differences in the proliferation of the HEK 293T cells after 24 h of incubation with 20–200 μg/mL of Gd/NIR-MSNs (Figure 4(f)), suggesting the Gd/NIR-MSNs, have good biocompatibility.
(a)
(b)
(c)
(d)
(e)
(f)
To further investigate the biomedical dual-modality imaging ability of the probes, U87-MG cells were incubated with the Gd/NIR-MSNs before unndergoing NIR fluorescent and MR map imaging. U87-MG cells labeled with MSNs-NH2 were treated as control. Strong NIR signal fluorescence was clearly visible in the optical image of tumor cells labeled with Gd/NIR-MSNs but completely absent in the image of cells incubated with MSNs-NH2 (Figure 5(a)). Moreover, significant longitudinal relaxation time reduction was observed of cells labeled with Gd/NIR-MSNs on MR map image compared with control (Figure 5(b)), which suggested that robust enhancement can be achieved with the bimodal probes. These results validated the potential NIR and MR in vivo imaging ability of the Gd/NIR-MSNs.
(a)
(b)
In vivo experiments were performed on a rodent subcutaneous U87-MG glioblastoma modal to evaluate the dual-modal imaging effects of Gd/NIR-MSNs. NIR fluorescence and MR imaging were acquired at different time points (0, 4, 8, 24, and 48 h) after 50 μL of Gd/NIR-MSNs were intratumor injected (Figure 6). At 4 h after injection, the enhanced region on MR images were slightly augmented and showed a more homogeneous distribution with a distinction margin of enhanced region, suggesting the excellent anatomic information provided by MR imaging (Figure 6(a)). And then, gradually MR signal reduction was observed over time (Figure 6(c)). At 48 h after injection, only slightly hyperintense signal remained in the tumor. The MR signal reduction is attributed to the diffusion of the nanoparticles. Figure S1 (in the Supplementary Material available online at http://dx.doi.org/10.1155/2016/6502127) shows the values (inversely proportional to signal intensity) at the injection region increase more quickly than the peripheral region, further confirming the diffusion of the nanoparticles in the tumor. For optical imaging, the NIR fluorescence signal showed a gradual augmentation over time and a poor signal margin due to the limitation of poor spatial resolution (Figures 6(b), 6(d) and Supplementary Figure S2 A. At the last time point of 48 h, there are still dramatic fluorescence signals that could be detected in the tumor region, which may be explained by the high sensitivity of optical imaging and the spreading of nanoparticles to the tumor surface. Hence, the imaging features revealed that MRI should be taken for high spatial resolution and anatomic images, while NIR fluorescent imaging should be used to provide high sensitivity lesion detection information. Combination of the two imaging modality could make excellent complementary for tumor detection and provide guidance for treatment.
(a)
(b)
(c)
(d)
4. Conclusions
In summary, we have successfully prepared a mesoporous silica nanoparticle based dual-modality probe (Gd/NIR-MSNs) by a facile protocol and carried out in vivo NIR fluorescence and MR imaging on a rodent tumor modal. The Gd/NIR-MSNs have uniform particle size about 120 nm and ordered mesostructure. Excellent biocompatibility and high intracellular retention ability of Gd/NIR-MSNs make it possible to be used as an in vivo imaging tool. Remarkable NIR fluorescence property and highly MR relaxivity observed both in vitro and in vivo confirm the feasibility of Gd/NIR-MSNs as dual-modality imaging contrast agents in tumor detection. In addition, high sensitive images with distinct anatomical information are obtained by the Gd/NIR-MSNs, making the nanoprobes more potential for biomedical research.
Competing Interests
No potential competing interests were reported by the authors of this paper.
Acknowledgments
The authors thank the financial support from the National Key Basic Research Program of China (2014CB744504 and 2014CB744501), the Major International (Regional) Joint Research Program of China (81120108013), the National Natural Science Foundation of China (30930028, 81371622, 81471632, and 81371611), and China Postdoctoral Science Foundation (2015M581837).
Supplementary Materials
Time-intensity Curve of MR imaging and NIR fluorescence.