In vivo tumor-targeted dual-modal fluorescence/CT imaging using a nanoprobe co-loaded with an aggregation-induced emission dye and gold nanoparticles

Abstract

As an intensely studied computed tomography (CT) contrast agent, gold nanoparticle has been suggested to be combined with fluorescence imaging modality to offset the low sensitivity of CT. However, the strong quenching of gold nanoparticle on fluorescent dyes requires complicated design and shielding to overcome. Herein, we report a unique nanoprobe (M-NPAPF-Au) co-loading an aggregation-induced emission (AIE) red dye and gold nanoparticles into DSPE-PEG₂₀₀₀ micelles for dual-modal fluorescence/CT imaging. The nanoprobe was prepared based on a facile method of “one-pot ultrasonic emulsification”. Surprisingly, in the micelles system, fluorescence dye (NPAPF) efficiently overcame the strong fluorescence quenching of shielding-free gold nanoparticles and retained the crucial AIE feature. In vivo studies demonstrated the nanoprobe had superior tumor-targeting ability, excellent fluorescence and CT imaging effects. The totality of present studies clearly indicates the significant potential application of M-NPAPF-Au as a dual-modal non-invasive fluorescence/X-ray CT nanoprobe for in vivo tumor-targeted imaging and diagnosis.

1. Introduction

The past decade has witnessed the rapid development of imaging nanoprobes, which are able to provide physiological and pathological information with high sensitivity and specificity for disease diagnosis [1–5]. However, single imaging techniques are only able to supply limited information, which is sometimes insufficient for accurate imaging diagnosis [6–9]. Due to this drawback, dual-modal nanoprobes, which combine the advantages of each imaging modality, have attracted great attention in recent years [10–12].

As a clinically approved imaging modality, X-ray computed tomography (CT) possesses the incomparable advantages of high spatial resolution and unlimited penetration depth [13, 14]. Among various CT contrast agents, gold nanoparticle owns an extremely high X-ray absorption coefficient, regardless its preparation method, shape, diameter, etc. [15]. However, CT imaging modality shows an inherent disadvantage of low sensitivity [13]. Hence, in order to develop a complementary dual-modal imaging probe, some imaging modalities with high sensitivity need to be combined with CT.

Among all other imaging modalities, red to near-infrared (Red-NIR, 600 – 900 nm) fluorescence imaging is highly attractive for early non-invasive detection of cancers because it has lots of excellent advantages such as high sensitivity, low consumption and facile operation [16, 17]. Fluorescence probes with long wavelength emission are highly desirable for detection, because biological tissues show relatively low absorption and autofluorescence in this region [18, 19]. Therefore, it is advisable to fabricate sensitive fluorescence nanoprobes using Red-NIR dyes. However, fluorescence imaging suffers from its own shortcomings, including low spatial resolution and limited penetration depth even at Red-NIR wavelengths [7].

Based on the analysis above, it would be ideal to combine CT and fluorescence imaging together to develop a complementary imaging modality with high spatial resolution and high sensitivity. However, conventional fluorescence dyes always suffer from low brightness due to a notorious phenomenon known as aggregation caused quenching (ACQ) [20]. In addition, gold nanoparticles (Au NPs), an intensely explored CT contrast agent, are well known as a strong quencher of fluorescence dyes [21]. Thus, the fabrication of fluorescence/CT dual-modal nanoprobes has been a great challenge [22, 23]. In 2001, Tang et al. unveiled a fluorescent molecule which emitted strong fluorescence when aggregated and thus presented an early example of aggregation-induced emission (AIE) [16]. These unique AIE fluorescent molecules are resistant to self-quenching and provide a promising solution for fabricating new imaging probes with superior performance both in vitro and in vivo. Furthermore, if the size of a nanoprobe is appropriate, it is ideal for tumor-targeted imaging because of the enhanced permeation and retention (EPR) effect [24]. To date, however, there is rare report in the literatures of whether the AIE dyes could overcome the quenching effect of gold nanoparticles to fabricate highly efficient dual-modal fluorescence/CT nanoprobes.

Herein, we fabricated a unique nanoprobe by co-loading the AIE red dye (NPAPF) and gold nanoparticles into the well-known FDA-approved material 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (DSPE-PEG₂₀₀₀) micelles. The nanoprobe was prepared by a facile method of “one-pot ultrasonic emulsification”. Unexpectedly, the utilized AIE dye (NPAPF) showed relatively enhanced emission in the as-prepared micelles system despite the existence of gold nanoparticles, which guaranteed its efficient fluorescence imaging effect. In vitro and in vivo results demonstrate the nanoprobe has good biocompatibility, long blood circulation half-life, superior tumor-targeting ability, and excellent fluorescence and CT imaging effects. To our knowledge, this is the rarely reported fluorescence/CT dual-modal micelles system in which red fluorescence dye (NPAPF) reserves its AIE feature in the presence of shielding-free gold nanoparticles.

2. Materials and Methods

2.1. Materials and instruments

1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (DSPE-PEG₂₀₀₀) was purchased from Avanti Polar Lipids (Alabaster, AL). Bis(4-(N-(2-naphthyl) phenylamino) phenyl)-fumaronitrile (NPAPF) was synthesized according to a previous report [25]. 4-bromophenylacetonitrile, N-phenyl-substituted amine, Pd(OAc)₂, Cs₂CO₃ and P(t-Bu)₃ were purchased from J&K Scientific Ltd. Diethyl ether, sodium, dichloromethane, petroleum, methanol, toluene, chloroform, metal-oxide-semiconductor (MOS) grade nitric acid, hydrochloric acid and hydrogen peroxide were purchased from the Beijing Chemical Reagents Institute (Beijing, China). Au NPs were prepared according to the literature [26]. Gold chloride trihydrate (99.9%, HAuCl₄•3H₂O) and oleylamine (70%) were purchased from Energy-Chemical (Beijing, China). Au standard solution (1,000 µg/mL) was obtained from the National Analysis Center for Iron and Steel (Beijing, China). All of the glasswares used for the preparation and storage of Au NPs were pre-cleaned with aqua regia (HCl: HNO₃ = 3:1, v/v). All of the chemicals were used without further purification, and Milli-Q water (18.2 M_) was used throughout this study.

The micelles were prepared on a KQ-100DE ultrasonic cleaner (Kunshan, China). Particle size was determined by dynamic light scattering (DLS) using a Malvern Zeta sizer ZS90 instrument (Worcestershire, U.K.). TEM images of all micelles were obtained through a FEI Tecnai G2 F20 S-Twin TEM (Hillsboro, OR). Emission spectra were characterized by a Fluoromax4 spectrometer (Horiba Jobin Yvon, Edison, NJ). UV-Vis-NIR spectra were obtained by a LAMBDA 950 UV/Vis/NIR spectrometer (PerkinElmer, U.S.A.). Cellular uptake was characterized with a Zeiss LSM510 confocal laser microscope (Carl Zeiss Shanghai Co. Ltd, Shanghai, China) and an Attune® acoustic focusing cytometer (Applied Biosystems, Life Technologies, Carlsbad, CA). The biodistribution of Au in tumor tissues and organs was determined by NexlON 300X inductively coupled plasma mass spectrometry (ICP-MS), (PerkinElmer, U.S.A.). In vitro and in vivo fluorescence images were collected by a Maestro 2 multi-spectral imaging system (Cambridge Research & Instrumentation, U.S.A.). In vitro and in vivo CT images were obtained by SPECT/CT scanning system (Triumph X-SPECT/X-O CT, GMI Company, U.S.A.).

2.2. Preparation and characterization of M-NPAPF-Au

Micelles loaded with NPAPF, or Au NPs, or NPAPF and Au NPs together (shortened as M-NPAPF, M-Au, M-NPAPF-Au, respectively) were prepared by “one-pot ultrasonic emulsification”. Briefly, DSPE-PEG₂₀₀₀ (8 mg) powder was placed into a round-bottom flask, and then Au NPs in chloroform (2 mg/L, 0.5 mL) and NPAPF in chloroform (1 mg/mL, 1 mL) were added into the flask and mixed thoroughly until the DSPE-PEG₂₀₀₀ was completely dissolved. After that, 10 mL Milli-Q water (18.2 MΩ) was added into the mixture, then the flask was placed in an ultrasonic bath cleaner for about 10 min ultrasonic emulsification at 100 W power. The organic solvent was then removed by evaporation while stirring in a fume hood overnight at room temperature. Following similar procedures, M-NPAPF was prepared without using Au NPs and M-Au was prepared without using NPAPF.

The morphology of M-NPAPF, M-Au and M-NPAPF-Au was examined using a Tecnai G2 20 STWIN transmission electron microscope with a 200 kV acceleration voltage. The distribution of hydrodynamic particle size was measured by a Malvern Zeta sizer ZS90.

2.3. Analysis of AIE properties and quantum yield determination

The absorption spectra of M-NPAPF, M-Au and M-NPAPF-Au were determined using a UV/Vis spectrometer and the photoluminescence (PL) spectrum was measured with a luminescence spectrometer. The concentration of CH₃CN in the CH₃CN/H₂O mixture ranged from 0% to 99.9%. Fluorescence quantum yields (QY) were determined according to the published reports in the literature [27]. Rhodamine B was chosen as a standard molecule. All UV-Vis absorption values were measured at a wavelength of 500 nm, and all PL spectra data were obtained with an excitation wavelength of 500 nm. QY values were calculated according to the following equation [28]:

$${\mathrm{\phi }}_{\mathrm{u}}={\mathrm{\phi }}_{\mathrm{s}}*({\mathrm{F}}_{\mathrm{u}}/{\mathrm{F}}_{\mathrm{s}})*({\mathrm{A}}_{\mathrm{s}}/{\mathrm{A}}_{\mathrm{u}})$$ (1)

(φ_u: quantum yield of test substance; φ_s: quantum yield of standard substance; F_u: integrated fluorescence intensity of test substance; F_s: integrated fluorescence intensity of standard substance; A_u: UV absorption of test substance; A_s: UV absorption of standard substance.)

2.4. Cytotoxicity studies

The cytotoxicity of M-NPAPF-Au was evaluated by MTT assay. BALB/c mice colon adenocarcinoma cells (CT26) incubated in RPMI 1640 medium, human hepatocellular carcinoma cells (HepG2) and normal human liver cells (L02) incubated in DMEM medium, which were all supplemented with 10% FBS, 100 U/mL penicillin, and 100 U/mL streptomycin. The cells were seeded at 5 × 10³ per well into a 96-well plate overnight, respectively, then incubated with 100 µL medium containing various concentrations of M-NPAPF-Au ranging from 37.5 to 600 µg/mL for 24 h. After that, the medium was replaced with medium containing MTT (0.5 mg/mL, 100 µL) for 4 h, and then the MTT medium was replaced with 100 µL DMSO. The absorbance was measured at 570 nm with a reference wavelength of 630 nm using an Infinite M200 microplate reader (Tecan, Durham, U.S.A.). Untreated cells were used as control. All experiments were carried out with five replicates.

2.5. Flow cytometric analysis

CT26 cells were seeded at 1 × 10⁵ cells per well into 6-well plates, then incubated with 2 mL M-NPAPF-Au (200 µg/mL). Cells were harvested at 6, 12, 24 h, and then analyzed using an Attune® acoustic focusing cytometer.

2.6. Confocal laser scanning microscopy (CLSM) imaging

CT26 cells were seeded with a density of 5 × 10⁴ per dish in 35 mm glass microscopy dishes and incubated overnight at 37°C. After washing with PBS, cells were incubated with 1 mL M-Au, M-NPAPF or M-NPAPF-Au (in each case the concentration of micelles was 200 µg/mL) for 24 h. Excess micelles were removed by washing three times with PBS, then cells were stained with 200 µL DAPI (20 µg/mL) for 1 min, washed three times with PBS (100 µL) and observed by confocal laser scanning microscopy (CLSM) with laser excitation at 488 nm. Fluorescence was collected at wavelengths from 660 nm to 750 nm.

2.7. Measurement of the Hounsfield units of M-NPAPF-Au

The CT numbers (called Hounsfield units, HU) of M-NPAPF-Au at different concentrations (50, 100, 200, 400, 800 µg/mL) were determined using a SPECT/CT scanning system (Triumph X-SPECT/X-O CT, GMI Company, U.S.A.). The measurement parameters were set as follows: effective pixel size, 50 µm; 80 kVp, 500 µA; field of view, 91.07 mm × 91.07 mm × 91.07 mm; fly, 360; binning, 2. Hounsfield units were acquired and analyzed using SPECT/CT scanning system workplace software.

2.8. In vivo fluorescence imaging, biodistribution and semi-quantitative pharmacokinetics studies

Female BALB/c mice aged 4 – 6 weeks were purchased from Vital River Company. All protocols for this animal study conformed to the Guide for the Care and Use of Laboratory Animals. All animal experiments were performed in accordance with guidelines approved by the ethics committee of Peking University. Tumor-bearing mice models were established by subcutaneous injection of 2 × 10⁷ CT26 cells into each mouse. Two weeks later, CT26 tumor-bearing mice were intravenously injected with 200 µL of M-NPAPF-Au (1 mg/mL) and imaged using the Maestro in vivo spectrum imaging system. The peak excitation wavelength was 488 nm and multispectral imaging was from 600 to 850 nm (in 10 nm steps). The exposure time was 1000 ms for all fluorescence images and 200 ms for all bright field images.

CT26 tumor-bearing mice injected with M-NPAPF-Au were sacrificed by cervical dislocation at 6, 12 and 24 h after injection. The tumor tissues and organs including brain, heart, liver, spleen, lung, kidney and intestine were collected and imaged immediately after sacrifice. Semi-quantitative biodistribution analysis of the average (n = 3) fluorescence intensity of each tissue and organ was calculated using Maestro 2 software.

For semi-quantitative pharmacokinetic studies, free NPAPF (in DMSO) and M-NPAPF-Au solution were intravenously injected into different groups of BALB/c mice [29]. About 10 – 20 µL blood samples were drawn from the tail vein at 5, 15 and 30 min and 1, 2, 4, 6, 8, 12 and 24 h after intravenous injection. The blood samples were then solubilized with lysis buffer (RIPA) and analyzed for NPAPF with a fluorescence spectrometer at 640 nm (λ_ex = 488 nm). The fluorescence of NPAPF was calculated by deducting the blank control blood sample.

2.9. In vivo CT imaging and biodistribution

For CT imaging of animals treated by direct injection into tumors, CT26 tumor-bearing mice were intratumorally injected with M-NPAPF-Au (30 µL, 70 mg/mL), then anesthetized and imaged using a SPECT/CT scanning system 15 min post-injection.

For CT imaging of animals treated by injection into the bloodstream, CT26 tumor-bearing mice were intravenously injected with M-NPAPF-Au (200 µL, 70 mg/mL), anesthetized by chloral hydrate (3.5%, w/v, 100 µL/10 g body weight), then imaged using a SPECT/CT scanning system at 6, 12 and 24 h after injection.

The CT imaging parameters were set as follows: effective pixel size, 50 µm; 80 kVp, 500 µA; field of view, 91.07 mm × 91.07 mm × 91.07 mm; fly, 360; binning, 2. Images were acquired and analyzed using the SPECT/CT scanning system workplace software. For biodistribution evaluation of intravenously injected M-NPAPF-Au, tumor tissue and organs including brain, heart, liver, spleen, lung, kidney and intestine were collected and digested by aqua regia, and then the Au concentration was determined by ICP-MS measurement according to a previously published method [30].

3. Results and Discussion

3.1. Synthesis and characterization of M-NPAPF-Au

NPAPF, with a hydrophobic molecular structure (Fig. 1a), was synthesized according to a previous report [25]. Hydrophobic Au NPs with diameters of ~ 9 nm were prepared in oil phase [26]. TEM confirmed the particle size and morphology of the obtained Au NPs (Fig. S1). Due to the hydrophobic properties of NPAPF and Au NPs, they could both be encapsulated inside the hydrophobic core of DSPE-PEG₂₀₀₀ micelles (Fig. 1a). Micelles loaded with NPAPF or Au NPs, or co-loaded with NPAPF and Au NPs (called M-NPAPF, M-Au, M-NPAPF-Au, respectively) were prepared based on a “one-pot ultrasonic emulsification” method. A schematic diagram of the preparation procedure of M-NPAPF-Au is shown in Fig. 1b. The concise optimization were conducted and showed in Table S1. The 1:1:8 ratio of Au/NPAPF/DSPE-PEG₂₀₀₀ (w/w/w) was chose to prepare the micelles.

Fig. 1.

(a) Molecular structures of DSPE-PEG₂₀₀₀ and NPAPF. (b) Scheme for preparation of M-NPAPF-Au. (c) TEM image of M-NPAPF-Au. (d) Size distribution of M-NPAPF-Au.

The size and morphology of M-NPAPF-Au, M-NPAPF and M-Au were characterized through TEM and dynamic light scattering (DLS) (Fig. 1c and S2). The average diameter of M-NPAPF-Au is ~ 65 nm. DLS studies showed that the micelles had good mono-dispersity with an average diameter of ~ 120 nm (Fig. 1d), PDI = 0.200. It is known that particles with sizes between 30 and 150 nm are particularly favorable for tumor-targeted in vivo imaging due to the EPR effect [24].

3.2. AIE properties of M-NPAPF-Au

The photophysical properties of M-NPAPF-Au were studied. Firstly, NPAPF was dissolved in CH₃CN, and water was added subsequently. It was obvious that the fluorescence emission intensity of NPAPF is weak in pure CH₃CN, but surprisingly it increased when the water fraction (f_w) in the CH₃CN/water mixture was higher than 50%. At f_w > 50%, the fluorescence emission intensity continuously increased as more water was added. This phenomenon is called aggregation-induced emission (AIE) (Fig. S3). It is attributed to the aggregation-induced restriction of intramolecular rotation and consequential suppression of non-radiative pathways. At a water fraction of approximately 100%, the fluorescence emission intensity is ~ 750 times higher than that in pure CH₃CN. As Au NPs can easily quench the fluorescence emission of dyes, it is crucial to verify whether the fluorescence of NPAPF is quenched when Au NPs are co-loaded into the micelles. The fluorescence emission intensity of a mixture of DSPE-PEG₂₀₀₀, NPAPF and Au NPs in pure CH₃CN is extremely low (Fig. S4a). When f_w in the CH₃CN/water mixture was less than 50%, the fluorescence of M-NPAPF-Au was extremely weak, but as f_w rose above 50%, the fluorescence was gradually enhanced. M-NPAPF-Au became highly emissive in pure aqueous solution, a phenomenon that is typical of AIE (Fig. S4a and Fig. S4b). This means that the AIE property of NPAPF is reserved in M-NPAPF-Au. UV-induced fluorescence photographs (Fig. S5) also showed that the fluorescence intensity of M-NPAPF-Au in water was obviously stronger than NPAPF in pure organic solvent (CH₃CN). This indicated that NPAPF retained its AIE properties in M-NPAPF-Au, and overcame the fluorescence quenching caused by Au nanoparticles [21]. As presented in Table S2, the fluorescence quantum yield of NPAPF in the micelles in the presence of Au NPs (QY = 8%) is much higher than that in pure CH₃CN (QY = 0.024%), suggesting that the as-prepared M-NPAPF-Au might be suitable for in vivo fluorescence imaging. The UV-Vis spectrum showed no impurity peak (Fig. 2a), indicating that the micelles did not aggregate. Therefore, we could affirm that the preparation method was very successful. The optimum emission of NPAPF is at ~ 640 nm (Fig. 2b), which is beneficial for non-invasive imaging because longer wavelength light (600 – 900 nm) has deeper penetration through biological tissues and weaker autofluorescence [17]. In addition, the Stokes shift of the as-prepared M-NPAPF-Au is as large as ~ 120 nm, which will help to reduce the optical interference from the excitation light source. M-NPAPF-Au combine the advanced properties of strong Red-NIR fluorescence, large Stokes shift and appropriate size, and might therefore be applicable for in vivo non-invasive and tumor-targeted fluorescence imaging.

Fig. 2.

(a) Normalized UV-Vis absorption spectra of M-Au, M-NPAPF and M-NPAPF-Au. (b) Normalized excitation and emission spectra of M-NPAPF and M-NPAPF-Au.

3.3. Cytotoxicity assay

Before M-NPAPF-Au can be applied as a dual-modal fluorescence/CT imaging nanoprobe, it is crucial to investigate its biocompatibility in vitro. After CT26, HepG2, L02 cells were incubated with M-NPAPF-Au at different concentrations (from 37.5 to 600 µg/mL) for 24 h, an MTT colorimetric assay was performed to assess cell viability (Fig. 3). Compared with the group treated with PBS, cells treated with M-NPAPF-Au were highly viable (> 80%) even at M-NPAPF-Au concentrations up to 600 µg/mL. Our results indicate that the as-prepared M-NPAPF-Au have superior biocompatibility with both tumor cells and normal cells. In the subsequent cell studies, the M-NPAPF-Au concentration was kept under 600 µg/mL, within the safe concentration range.

Fig. 3.

Viability of CT26, HepG2, L02 cells incubated with different concentrations of M-NPAPF-Au for 24 h.

3.4. Fluorescence properties in vitro

The in vitro fluorescence imaging performance of M-NPAPF-Au was next examined. As shown in Fig. 4a, higher nanoprobe concentrations resulted in stronger fluorescence intensity. Subsequently, we quantified the cellular uptake of M-NPAPF-Au by flow cytometry. The intracellular fluorescence intensity of NPAPF gradually increased when the incubation time was extended from 6 h to 24 h (Fig. 4b), which indicated that M-NPAPF-Au (200 µg/mL) was gradually taken up by CT26 cells in a time-dependent manner (within 24 h). The intracellular imaging performance of M-NPAPF-Au was then examined. CT26 cells were incubated for 24 h with M-Au, M-NPAPF and M-NPAPF-Au (micelle concentrations were 200 µg/mL in each case), and then cells were stained with DAPI prior to CLSM imaging. As shown in Fig. 4c, blue and red regions represent DAPI (which labels nuclei) and NPAPF, respectively. For CT26 cells incubated with M-NPAPF-Au and M-NPAPF, strong red fluorescence could be observed, while no signal could be detected from cells treated with M-Au, which demonstrated that the red fluorescence came from NPAPF. In addition, the fluorescence of NPAPF in M-NPAPF-Au was not strongly quenched by Au NPs. In summary, M-NPAPF-Au has excellent fluorescence emission properties and superior biocompatibility, and therefore shows great promise for use in in vivo fluorescence imaging studies.

Fig. 4.

(a) Change in the fluorescence intensity of M-NPAPF-Au with increasing concentration (µg/mL). (b) Quantitative analysis of M-NPAPF-Au uptake by flow cytometry. (c) Confocal images of CT26 cells after incubation with M-Au, M-NPAPF and M-NPAPF-Au for 24 h at micelle concentrations of 200 µg/mL. The images were recorded under excitation at 488 nm with 660 – 750 nm bandpass filters. Scale bars are 10 µm.

3.5. X-ray attenuation property in vitro

Theoretically, a good CT contrast agent must have a high X-ray attenuation coefficient [13]. Traditional CT imaging agents have high X-ray absorption coefficients but also have several inherent shortcomings such as rapid clearance by the kidney (preventing long-term imaging), and serious renal toxicity [31, 32]. Recently, some nanoparticles containing Au NPs have been actively used for CT imaging, and have given beautiful results [33, 34], attributable to the higher X-ray absorption coefficient and superior biocompatibility of Au NPs. In our study, the Hounsfield units (HU) of M-NPAPF-Au at different concentrations (50, 100, 200, 400 and 800 µg/mL) were evaluated using a SPECT/CT scanning system. As demonstrated in Fig. 5a, samples of higher concentration had a deeper color and are more radiodense than more dilute samples, which corresponds with previously reported results that higher concentration leads to a stronger X-ray CT attenuation intensity [35]. The HU of M-NPAPF-Au was evaluated using a SPECT/CT scanning system, and was described by the following equation:

$$\text{HU}=0.2242*\mathrm{X}+3.2139$$ (2)

(HU: CT numbers; X: concentration of M-NPAPF-Au, µg/mL)

Fig. 5.

(a) Digital photos (top) and corresponding CT images (bottom) of M-NPAPF-Au samples of increasing concentration (µg/mL). (b) Standard curve of CT values (Hounsfield units, HU) of M-NPAPF-Au at different concentrations.

Further, there is a well-correlated linear relationship (R² = 0.9983) between M-NPAPF-Au concentration and HU (Fig. 5b). These results suggest that M-NPAPF-Au is an ideal candidate for a positive CT imaging nanoprobe.

3.6. In vivo fluorescence imaging, biodistribution and semi-quantitative pharmacokinetics studies

Before evaluating the in vivo imaging, we used hemolysis analysis to study the blood compatibility of M-NPAPF-Au. As shown in Fig. S6, no visible hemolytic effects were seen even at the highest M-NPAPF-Au concentration of 10 mg/mL, indicating that M-NPAPF-Au had good hemocompatibility (< 3.5%) [36].

The performance of M-NPAPF-Au in in vivo tumor-targeted and non-invasive fluorescence imaging was examined in CT26 tumor-bearing mice. After intravenous injection of M-NPAPF-Au (200 µL, 1 mg/mL), the mice were scanned with a multi-spectral imaging system at different time points (up to 24 h). Fig. 6a showed how the in vivo tumor-targeted fluorescence imaging effects changed over time. There were basal fluorescence signals from the skin and hair before injection. After intravenous injection of M-NPAPF-Au solution, the signal at the tumor site gradually became more intense and could be distinguished more readily from the autofluorescence of the mice as the time increased. At 24 h post-injection, the tumor site was much brighter than any other body part. This was attributed to passive enrichment of the as-prepared nanoprobe at the tumor site by the EPR effect [18].

Fig. 6.

(a) Non-invasive fluorescence images of CT26 tumor-bearing mice and their dissected tumors and organs 6, 12 and 24 h after intravenous injection. The white arrows indicate tumor sites and the red circles indicate dissected tumors. 1 - Liver, 2 - Spleen, 3 - Kidney, 4 - Heart, 5 - Lung, 6 - Tumor, 7 - Brain, 8 - Intestine. (b) Semi-quantitative biodistribution of M-NPAPF-Au in mice determined by the averaged fluorescence intensity of each tumor and organ (after subtraction of the fluorescence intensity before injection). Error bars are based on three mice per group. (c) Blood circulation curves of free NPAPF (black) and M-NPAPF-Au (red) determined by measuring the fluorescence intensity of NPAPF in the blood at different time points post-injection. The y-axis shows the percentage of injected dose per gram tissue (%ID/g).

Next, semi-quantitative biodistribution analysis was performed. A series of dissected organs and tumor tissues of mice (n = 3) were freshly collected at various time points (6, 12 and 24 h) post-injection, and fluorescence images were immediately taken with the multi-spectral imaging system. The fluorescence intensity of each image was subsequently calculated to give average intensity values as well as standard errors. Background autofluorescence values, obtained from tumors and organs from a control mouse without M-NPAPF-Au treatment, were subtracted. Fig. 6b showed that the as-prepared M-NPAPF-Au accumulated at higher levels in the tumor than any other organ except the liver, which was in agreement with the images of the dissected tumors and organs. The accumulation level of nanoprobe in liver is higher than that of other organs, this is consistent with other reports [31, 37], which suggested that nanoparticles tended to accumulate in the reticuloendothelial system (RES). Hence, the high intensity of nanoprobe in liver is due to the uptake nature of liver.

In order to study the pharmacokinetics profile, free NPAPF (in DMSO) and M-NPAPF-Au solution were intravenously injected in two different groups of BALB/c mice [29]. 10 – 20 µL blood was drawn from the tail vein at different time points after injection and solubilized in lysis buffer, and then the NPAPF concentration was determined by measuring the fluorescence intensity of NPAPF and subtracting the blank blood sample from an untreated control mouse. As shown in Fig. 6c, the blood circulation half-life of free NPAPF and M-NPAPF-Au was 0.5 h and 8 h, respectively. The greatly prolonged half-life of M-NPAPF-Au was largely attributed to the perfect stability of the carrier material DSPE-PEG₂₀₀₀ [38, 39]. The long half-life explains why the M-NPAPF-Au nanoprobes gradually accumulated at the tumor site throughout the test period (24 h).

Overall, these results clearly indicate that the M-NPAPF-Au has significant potential for in vivo tumor-targeted fluorescence imaging.

3.7. In vivo CT imaging and biodistribution

Computed tomography is one of the most useful clinical diagnostic tools, and has been widely explored for its potential in tumor detection [13, 14]. The distinguished in vitro performance of M-NPAPF-Au as a potential CT contrast agent encouraged us to test its applicability to in vivo CT imaging by two different methods. Initially, one group of CT26 tumor-bearing mice (n = 3) were intratumorally injected with M-NPAPF-Au (20 µL, 70 mg/mL). Pre-injection mice were used as reference (Movie. S1) and the average HU of the tumor site was estimated to be 65. After intratumoral injected for 15 min (Movie. S2), the average HU of the tumor site increased to 201. It was clear that after 15 min, the tumor region injected with M-NPAPF-Au showed an obvious enhancement of CT signal compared with the pre-injection reference (Fig. S7), indicating that the nanoprobe is suitable for in vivo CT imaging. Secondly, another group of CT26 tumor-bearing mice (n = 3) were intravenously injected with M-NPAPF-Au (200 µL, 70 mg/mL). The average pre-injection and post-injection (6, 12, 24 h) HU values were 73 and 94, 110, 149, respectively (Fig. 7a, Movie S3–S6). We concluded that as the post-injection time increased, the nanoprobe diffused into tumor regions and gradually accumulated there via the EPR effect [24]. These results corresponded well with the multi-spectral fluorescence imaging data. Together, the results clearly demonstrate that M-NPAPF-Au has tumor-targeting properties and can be suitably used as a CT imaging agent.

Fig. 7.

(a) CT images of mice bearing transplanted CT26 tumors before (pre-injection) and after (6, 12 and 24 h) intravenous injection of M-NPAPF-Au. The white circles indicate tumor regions. The top row shows stereo images; the bottom row shows sectional images. (b) Biodistribution of Au by ICP-MS in tumor tissues and major organs including brain, heart, liver, spleen, lung, kidney, and intestine.

In addition, we also analyzed the biodistribution of M-NPAPF-Au after intravenous injection. Elemental Au accumulation in the major organs including brain, heart, liver, spleen, lung, kidney, intestine and tumor was examined by ICP-MS (Fig. 7b). It was clear that at 6, 12 and 24 h post-injection, tumor, liver and spleen had a gradually increased Au uptake. Comparing to the biodistribution analyzed by semi-quantitative fluorescent analysis, the general biodistribution data consisted with each other, in spite of minor difference. This maybe contributed by the intrinsic different characteristics of organic molecule and inorganic nanoparticle, as well as the distinction of detecting techniques.

It is well known that surface PEGylation can increase the blood circulation time of nanoparticles, reduce their clearance by the reticuloendothelial system (RES) and facilitate their accumulation at the tumor site. With the prolonged circulation time resulting from DSPE-PEG₂₀₀₀ encapsulation, M-NPAPF-Au gradually accumulated at tumor sites, while in other organs except liver and spleen, Au NPs uptake was quite low.

In order to evaluate the in vivo toxicity of M-NPAPF-Au, especially toward the liver and spleen, we investigated body weight variation and histology analysis of mice injected with M-NPAPF-Au (200 µL, 70 mg/mL). No obvious body weight variation was observed from the micelles injection group (Fig. S8) compared to the control group. 8 days post injection, the mice were sacrificed, livers and spleens were sectioned and stained by hematoxylin and eosin (H&E) for histology analysis (Fig. S9). In spite of relatively high uptake of M-NPAPF-Au in liver and spleen, no apparent histopathological abnormalities were observed in comparison with the control mice. These demonstrated that M-NPAPF-Au had unconspicuous in vivo toxicity at the concentration we used.

Above all, these results indicate that the as-prepared M-NPAPF-Au might be a unique and promising nanoprobe for potential in vivo tumor-targeted CT imaging. Notably, the signal intensity and nanoprobe accumulation in the tumor 24 h post intravenous injection are much higher than at the other time points. This is beneficial for future clinical applications because the contrast agents can be administered to patients during the daytime and the imaging procedure can be carried out at the same time the next day.

4. Conclusions

In summary, we have successfully developed a unique and promising dual-modal fluorescence/CT nanoprobe for tumor-targeted imaging. The nanoprobe was facile to be fabricated, via “one-pot ultrasonic emulsification”. Most important of all, the utilized AIE dye (NPAPF) showed relatively enhanced emission and retained the crucial AIE feature in the as-prepared micelles system despite the existence of gold nanoparticles, which guaranteed its efficient fluorescence imaging capacity. In vitro and in vivo results demonstrate that the nanoprobe has good cytocompatibility and hemocompatibility, long blood circulation half-life, superior tumor-targeting ability, unconspicuous in vivo toxicity, and excellent fluorescence and CT imaging effects. To our knowledge, this is the rarely reported fluorescence/CT dual-modal micelles system in which red fluorescence dye (NPAPF) reserves its AIE feature in the presence of shielding-free gold nanoparticles. Overall, the studies assuredly indicate the significant potential application of M-NPAPF-Au as a dual-modal non-invasive fluorescence/X-ray CT nanoprobe for in vivo tumor-targeted imaging and diagnosis.