Long-term multimodal imaging of tumor draining sentinel lymph nodes using mesoporous silica-based nanoprobes

Abstract

The imaging of sentinel lymph nodes (SLNs), the first defense against primary tumor metastasis, has been considered as an important strategy for noninvasive tracking tumor metastasis in clinics. In this study, we report the development and application of mesoporous silica-based triple-modal nanoprobes that integrate multiple functional moieties to facilitate near-infrared optical, magnetic resonance (MR) and positron emission tomography (PET) imaging. After embedding near-infrared dye ZW800, the nanoprobe was labeled with T₁ contrast agent Gd³⁺ and radionuclide ⁶⁴Cu through chelating reactions. High stability and long intracellular retention time of the nanoprobes was confirmed by in vitro characterization, which facilitate long-term in vivo imaging. Longitudinal multimodal imaging was subsequently achieved to visualize tumor draining SLNs up to 3 weeks in a 4T1 tumor metastatic model. Obvious differences in uptake rate, amount of particles, and contrast between metastatic and contralateral sentinel lymph nodes were observed. These findings provide very helpful guidance for the design of robust multifunctional nanomaterials in SLNs’ mapping and tumor metastasis diagnosis.

1. Introduction

Sentinel lymph node (SLN) is the hypothetical first lymph node or group of nodes reached by metastasizing cancer cells from a primary tumor [1–3], and continues to be used as an important parameter in tumor staging and therapeutic decision-making. Thus, lymph node imaging can be applied to evaluate the metastatic status of a tumor. SLN imaging is based on an injected contrast agent near the primary tumor that is taken up by the adjacent lymphatic system and then transported to the SLN. Currently, vital dyes and radionuclide-labeled sulfur colloids are the most common imaging agents for SLN imaging [4]. However, these methods have a number of drawbacks. For instance, SLNs need to be dissected to observe the blue dye staining and lymphoscintigraphy requires radiation exposure with relatively low resolution. Therefore, various groups have performed studies to develop lymphatic imaging probes and imaging methods that would exceed the capabilities of the established “blue dye” procedure, and to improve identification and mapping of lymph nodes, especially sentinel lymph nodes during surgery [5–7].

With the emergence of nanotechnology, several categories of nanoprobes have been developed to locate SLNs in living organisms, including quantum dots (QDs) [5,8–11], iron oxide [12], gold nanoparticles (NPs) [13], rare-earth-based NPs [14], carbon nanotubes [15], and perfluorocarbon-based NPs [16]. Based on their inherent properties, QDs and iron oxide can be detected and visualized by optical imaging or magnetic resonance imaging (MRI), respectively. When labeled with positron emitting radioisotopes, nanoprobes can be imaged with positron emission tomography (PET). However, each imaging modality has its own strengths and limitations. For example, MRI can provide three-dimensional tomography but is limited by low target sensitivity, whereas PET and optical imaging have good sensitivity but suffer from low spatial resolution or tissue penetration. To harness the strengths of different imaging methods, multimodality imaging has become attractive for both small animal and human studies [17,18]. It has emerged as a strategy that combines the strengths of different modalities and yields a hybrid imaging platform with characteristics superior to those of any of its constituents considered alone [19,20]. Development of imaging probes with multiple functions is the key for successful multimodal imaging.

With large surface area-to-volume ratios, unique mesoporous structure and excellent biocompatibility, mesoporous silica nanoparticles (MSNs) are considered an ideal matrix to integrate imaging tags for the development of either single functional [21,22] or multi-functional nanoprobes [23–25]. In several recently reported studies, silica-based nanoprobes have been designed to image the SLNs [26–28]. However, very few studies on the multimodal imaging of tumor metastatic SLNs (T-SLNs) have been reported so far. Herein, we designed a mesoporous silicabased triple-modal imaging nanoprobe (MSN-probe) that possesses the long-term imaging ability to track tumor metastatic SLNs. In this system, three imaging tags including near-infrared (NIR) dye ZW800, T₁ contrast agent Gd³⁺ and positron emitting radionuclide ⁶⁴Cu were integrated into MSNs by different conjugation strategies. We also applied these MSN-probes to visualize T-SLNs in a 4T1 tumor metastasis model. Due to their high stability and long intracellular retention time, signals from tumor draining SLNs are detectable up to 3 weeks. More importantly, obvious differences in uptake rate, amount of particles and contrast between metastatic and normal contra-lateral SLNs (N-SLNs) were observed.

2. Materials and methods

2.1. Materials

Cetyltrimethylammonium bromide (CTAB), tetraethyl orthosilicate (TEOS), aqueous ammonia, 3-aminopropyltriethoxysilane (APTES), bromoacetic acid, 3-(trimethoxysilylpropyl) diethylene triamine, and fluorescein isothiocyanate (FITC) were obtained from Sigma. Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from Gibco. The centrifugal filter (30k cutoff) was bought from Millipore. RAW 264.7 cells were purchased from ATCC. Phallotoxins and 4′, 6-diamidino-2-phenylindole (DAPI) were obtained from Invitrogen and Vector Laboratories, respectively.

2.2. Synthesis and characterization of dye doped MSNs

Dye (FITC or ZW800) doped MSNs were synthesized according to previous reports [29,30]. Briefly, 3-aminopropyl triethoxysilane-dye (APTES-dyes) was firstly conjugated by stirring dyes in ethanolic APTES solution (wAPTES/wd_ye = 50) for 4 h. Separately, CTAB (0.27 mmol) was dissolved in 70 ml of H2O, and 14.29 mmol NH3H2O (28%–30%) was added with magnetic stirring for 10 min at room temperature. Half of TEOS (0.72 mmol) was then added with vigorous stirring for 30 min. 50 µl APTES-dye in ethanol solution (v_APTES/v_ethanol = 1:3) was added, and the additional half of TEOS was added with vigorous stirring for 4 h. The resulting particles were collected by centrifugation and then washed three times with deionized water and ethanol. The mesoporous structures of MSNs were obtained by removing CTAB in acidic ethanol (1 ml of concentrated HCl in 50 ml of ethanol) for 24 h. The particles were washed three times with deionized water and then stored at 4 °C. After sonication with a probe sonicator, the final particles were obtained, including FITC doped MSNs (MSN-FITC) and ZW800 doped MSNs (MSN-ZW800). The resulting particles were observed by transmission electron microscopy (TEM). The UV–Vis and fluorescence spectra of particles were recorded on a Genesys 10s UV–Vis spectrophotometer (Thermo, IL) and a F-7000 fluorescence spectrophotometer (HiTachi, Japan), respectively.

2.3. Conjugation of Si-Gd-DTTA with dye doped MSNs (Dye@MSN@Gd)

Si-DTTA was prepared by conjugating bromoacetic acid with 3-(trimethoxysilylpropyl)diethylene triamine according to a previously reported method [23]. Briefly, 10 mmol bromoacetic acid and 2.5 mmol 3-(trimethoxysilylpropyl)diethylene triamine were mixed with 1 ml distilled water and 5 ml 2 M NaOH. The reaction solution was heated to 50 °C and then an additional 3 ml of 2 M NaOH was added into the solution. After stirring for 2 h, the resulting Si-DTTA was isolated by precipitation with EtOH, and subsequent drying in vacuo. The Si-Gd-DTTA was synthesized by chelating Si-DTTA with Gd³⁺. 0.5 mmol Si-DTTA was dissolved in 10 ml deionized water at room temperature. 0.5 mmol GdCl₃ was added into the solution and then the solution was adjusted to pH 9 by 2 M NaOH. After stirring the reaction for 2 h, Chelex 100 was added to remove excess Gd³⁺ and the resulting Si-Gd-DTTA was collected by centrifugation.

0.01 mmol Si-Gd-DTTA solution was dispersed in 1 ml deionized water containing dye doped MSNs (3 mg/ml). The mixture was stirred for 12 h. The particles were subsequently washed three times with water by re-dispersing via sonication and isolating via centrifugation at 15000 rpm for 10 min.

2.4. Radiolabeling of Dye@MSN@Gd with ⁶⁴Cu (Dye@MSN@Gd@⁶⁴Cu)

The Dye@MSN@Gd was firstly modified by 1µg/ml APTES. After washing twice, DOTA-NHS in DMSO was added and incubated for 2 h. The conjugated particles were purified by centrifugation. Details regarding ⁶⁴Cu labeling have been reported elsewhere [31,32]. Briefly, ⁶⁴CuCl₂ was converted to Cu(OAc)₂ by adding 0.5 ml of 0.4 M ammonium acetate solution (pH = 5.5) to 20 µl ⁶⁴CuCl₂. Cu(OAc)₂ (1 mCi) was added into a solution of DOTA labeled particles and incubated for 1 h with constant shaking. The labeled particles were purified by centrifugal filter (30k cutoff). To make sure free ⁶⁴Cu was removed completely, the filtered solution was monitored by radiation dosimeter until radiation was not measured. The labeled efficiency was calculated based on the radiation dosimeter readings before and after purification.

2.5. Cellular uptake and retention ability of MSN-FITC

RAW 264.7 cells were plated 24 h before the start of the experiment in chamber slides at a density of 5 × 10³ cell/cm². After incubation with 0.1 mg/ml MSN-FITC for 2 h, the cells were washed twice with PBS and incubated with Z-fix solution for 20 min. Subsequently, the cells were incubated with 0.1% Triton X-100 in PBS at room temperature for 5 min and subsequently incubated with Phallotoxins for staining F-actin for 20 min, followed by 1.5 µg/ml DAPI staining at room temperature. The slides were washed twice with PBS and then observed with an Olympus FV10i confocal microscope.

For the cell retention study, the procedures were similar with cellular uptake of particles. Briefly, the particles were incubated with cells for 2 h and subsequently removed by PBS washing three times. The cells were left in culture for different time points, and then were stained similar with the cellular uptake procedures. The images were acquired by the confocal microscopy (Olympus FV10i).

2.6. 4T1 tumor lymph node metastasis model

The tumor model was established in 4–6 week-old female BALB/c mice by subcutaneous injection of 5 × 10⁶ of 4T1 cells transfected with firefly luciferase to the hock of the right leg of mice [33]. Tumor metastatic progression of sentinel lymph node was evaluated after i.v. injection of luciferase substrate for 15 min in an IVIS imaging system. Animal procedures were performed according to a protocol approved by the National Institutes of Health Clinical Center Animal Care and Use Committee (NIH CC/ACUC).

2.7. Small animal imaging

For near-infrared fluorescence (NIRF) imaging, mice were anesthetized with isoflurane and were subcutaneously injected in the sole of the foot with 50 µl particles (2 mg/ml). NIRF imaging was observed on a Maestro all-optical imaging system at different time points. At the end of each time point, the mice were sacrificed, and the major organs were collected and subjected to ex vivo imaging with a Maestro imaging system.

Small animal MRI was performed based on our previous studies [34]. 50 µl particles (2 mg/ml) were injected through the soles of the foot into the anesthetized mice. Three-dimensional gradient-echo scan (FLASH) images were acquired on a 7.05 T small-animal MR scanner (Bruker Biospin) before injection and at different time points post injection.

The details of small animal PET imaging and the region-of-interest (ROI) analysis have been reported in our previous studies [32]. 50 µl particles (2 mg/ml) with 20 µCi ⁶⁴Cu were injected in the sole of the foot of the lymphatic metastasis model. PET scans and image analysis were performed using an Inveon microPET scanner (Siemens) at different post injection time points.

2.8. Statistical analysis

The level of significance in all statistical analyses was set at a probability of p < 0.05. Data are presented as means ± SD. Analysis of variance and t tests was used to analyze the data.

3. Results and discussion

3.1. Synthesis and characterization of multimodal MSN-probes

Silica matrix is optically transparent that allows excitation and emission light to pass through the matrix efficiently. This property has obvious advantages over other NPs with quenching ability, such as gold NPs [35], iron oxide NPs [36] and carbon nanotubes [37]. The most common strategy to label MSNs is to conjugate dyes on the surface of the particles. However, in this exposed way fluorescent molecules are easily quenched or detached from the particle, which may comprise in vivo application and imaging quantification. Alternatively, we doped NIR ZW800 dyes into MSNs to prevent quenching and falloff of the dye molecules in vivo [38]. As shown in Fig. 1A, 3-aminopropyl triethoxysilane-ZW800 (APS-ZW800) was firstly prepared by the reaction between -NH₂ of APS and -NHS of ZW800. APS-ZW800 was then added into the reaction of the particle fabrication under diluted TEOS and low surfactant conditions with aqueous ammonia as a catalyst. Consequently, APS-ZW800 was embedded into particles by the binding of APS to silicon oxide. After removing surfactant CTAB of particles in acidic condition, the mesoporous structure was formed, resulting in high Brunauer, Emmet and Teller (BET) surface area and large pore volume of the particles [39].

Fig. 1.

Schematic illustration of tri-modal imaging MSN-probes. (A) Diagram of ZW800 doped MSN fabrication; (B) Diagram of Gd³⁺ and ⁶⁴Cu integration.

To further prepare multi-functional nanoprobes, high payloads of Gd³⁺ and ⁶⁴Cu were integrated to ZW800 doped MSNs by chelating reaction. First, Gd³⁺ was chelated with Si-DTTA and then conjugated to Si-O bonds on the particles based on a previously reported procedure [23]. We chose ⁶⁴Cu as the positron emitting radioisotope for PET imaging since it has a relatively long physical half-life (t1/2 = 12.6 h) [40]. After functionalization of particles with DOTA-NHS, ⁶⁴Cu was also chelated on the surface and in the mesoporous channels of MSNs (Fig. 1B). Once integrated into mesoporous channels of particles, both Gd³⁺ and ⁶⁴Cu were protected from the influence of the in vivo environment. Thus, the MSN-probes were endowed with enough stability and robustness to meet the requirement for long-term in vivo imaging.

Before proceeding to in vivo application for multiple modal imaging, the major features of the MSN-probes were firstly characterized in vitro, including particle size, NIR fluorescence properties, T1 relaxation time and label efficiency of ⁶⁴Cu (Fig. 2). The diameter of particles is about 60 nm determined by transmission electron microscopy (TEM) imaging (Fig. 2A). A clear mesoporous structure of the particles is observed by an enlarged TEM image (boxed area). The hydrodynamic diameter of MSN-probes was 76.8 ± 8.3 nm by a dynamic light scattering (DLS) measurement (Supplementary Figure S1). It was reported that particles smaller than 5 nm leak into blood vessels and enter into the blood circulation system very quickly, whereas those >1000 nm of particles remain at the injection sites [41–43]. Thus, these particles at the nanoscale will be ideal for relatively long-term lymph node imaging.

Fig. 2.

Characterization of MSN-nanoprobes. (A) TEM of nanoprobes, and (B) excitation and emission of MSN-nanoprobes. Fluorescence imaging of particles was directly acquired before (left) and after (right) excitation by Maestro imaging system. (C) Phantom images and quantitative analysis of MSN-nanoprobes by MRI. (D) ⁶⁴Cu labeling efficiency at different ratios between ⁶⁴Cu and particles. The images were shown before (left) and after (right) PET imaging.

Excitation and emission peaks of MSN-probes were 770 nm and 795 nm (Fig. 2B), respectively, which are similar to those of ZW800 dye (Supplementary Figure S2). Strong fluorescence signal from MSN-probes in deionized water were also identified with a Maestro optical imaging system, validating the potential ability of the probes for optical imaging. As shown in Fig. 2C, phantom MRI revealed a linear relationship between the concentration of Gd³⁺ and R1 (1/T₁). Fig. 2D shows successful labeling of ⁶⁴Cu and an increased label efficiency of ⁶⁴Cu is seen with an increase of the ratio between ⁶⁴Cu and particles. Since PET imaging is more sensitive than MRI and optical imaging, we did not pursue high specific activity. With 4 mCi per mg of MSN-probes, the labeling efficiency is 75.9%, which provides strong enough signal for in vivo imaging. For optimal in vivo SLN imaging, the loading values of ZW800, Gd and ⁶⁴Cu were controlled and were quantified as 30 µg, 16 µg and 0.2 mCi per mg of MSNs, respectively.

3.2. Stability of MSN-probes

Maintaining stability is one of the prerequisites for long-term in vivo imaging. At different time points after i.v. injection of MSN-probes, the mice were imaged by a Maestro optical imaging system. Consistent with previous reports [30], the particles were located mainly in the liverand spleen and the signal gradually decreased over time (Fig. 3A). Ex vivo biodistribution study confirmed liver and spleen localization of the MSN-probes (Fig. 3B). Even at 2weeks after injection, the MSN-probes maintained an obvious optical signal. These results imply that the fluorescence of dyes embedded in MSN-probes is not quenched in vivo. In addition, it indicates that the MSN-probes are stable and slow to be cleared or metabolized in vivo.

Fig. 3.

Long-term imaging ability and biodistribution of MSN-nanoprobes in vivo. (A) In vivo optical imaging of nanoprobes over time. The images were normalized.(B) Ex vivo imaging of MSN-nanoprobe biodistribution in different organs after injection for 7 days (left) and 14 days (right).

The high and localized liver and spleen signal is usually caused by uptake of MSN-probes through the reticuloendothelial system (RES). To further elaborate on the in vivo distribution and stability of MSN-probes, we chose RAW 264.7 macrophage cells as a model to evaluate the retention ability of MSN-probes in vitro. For this purpose, FITC embedded MSNs (MSN-FITC) were synthesized and cellular uptake of particles was verified (Fig. 4). A microscopic co-localization study demonstrated that the particles (green) distributed in cytoplasm and mainly localized in the peripheral area of the nucleus. The fluorescence intensity of particles was gradually reduced in cells over time (Supplementary Figure S3A–D), but still observable after 7 days of culture. At that time, the proliferating dilution was over 40 folds (Supplementary Figure S3E). It is apparent that after being embedded into the MSNs, the doped fluorescent dyes are protected from the intracellular environment including oxidative agents, enzymes and pH. Moreover, the particles did not easily efflux from cells and maintained a long retention time in the cells. The longterm intracellular retention and stability of MSNs in biological environment definitely facilitate long-term in vivo imaging. Moreover, it has been demonstrated that the current dose of MSN particles does not induce significant toxicity in vivo [30].

Fig. 4.

Cellular uptake of MSN-probes in macrophage. MSN-nanoprobes were incubated with macrophages for 2 h and free MSN-nanoprobes in the medium were removed by washing. The nucleus and cytoskeleton of cells were stained and subsequently imaged by a confocal microscopy. (A) Nucleus staining of cells with DAPI. (B) Imaging of MSN-FITC. (C) F-actin staining with phallotoxins. (D) The merged imaging of nucleus, particles and F-actin.

3.3. Tumor metastasis model development and optical imaging

Regional lymph node metastasis represents the first step of tumor dissemination for a variety of common human cancers, such as carcinomas of the breast, colon, and prostate as well as melanoma [2]. Herein, we developed a T-SLN model using 4T1 murine breast cancer cells and performed long-term multimodal imaging with the MSN-probes for T-SLN mapping and indirect tumor metastasis tracking. Before hock inoculation in the right leg of Balb/C mice, 4T1 cells were stably transfected with firefly luciferase.After two weeks, an obvious localized bioluminescent signal was observed by both in vivo (Fig. 5A) and ex vivo (Fig. 5B) imaging using an IVIS imaging system, confirming that tumor cells metastasized to co-lateral (right side) draining SLNs. No signal was detected in the contralateral (left side) N-SLNs. Moreover, the T-SLNs enlarged in size, mainly due to tumor cell growth and tumor induced lymphangiogenesis [33].

Fig. 5.

Optical imaging of sentinel lymph nodes in a 4T1 tumor metastatic model. (A) Tumor metastasis was tracked before (left) and after (right) injection with substrate of luciferase by IVIS imaging system. The tumor metastasis model was established by right hock injection of 4T1 cells. Right sentinel lymph node: tumor metastasis (T-SLN). Left opposite sentinel lymph node: normal lymph node (N-SLN) as control. (B) Ex vivo imaging of T-SLN (left) and N-SLN (right) by IVIS imaging system. (C) Optical imaging of tumor metastatic lymph nodes after injection of MSN-nanoprobes at different time points (1 h, 6 h, 1 day, 5 days, 10 days, 15 days and 21 days). Square dot, T-SLN; solid line, N-SLN. (D) Quantitative analysis of the fluorescence signal of T-SLN and N-SLN at different time point. (E) The co-localization of particles and nucleus (DAPI) in T-SLN (E) and N-SLN (F) were imaged by microscopy.

To mimic the administration route of other SLN imaging agents, the MSN-probes were subcutaneously injected in the sole of the foot and then in vivo optical imaging was acquired at different time points by Maestro imaging system. The fluorescence signal of T-SLNs was rapidly increased and reached maximum at 1 h p.i. (Fig. 5C). Positive signal was also observed over the liver region at 1 h p.i., implying that the injected MSN-probes entered into blood circulation and were subsequently captured by the liver. At 6 h after injection, the signal of T-SLNs was still very strong and gradually decreased over time, which might be due to macrophage migration and slow degradation of MSN-probes. T-SLNs had a more rapid uptake rate compared with N-SLNs (p < 0.01). More importantly, at all the time points examined, T-SLNs showed significantly stronger signals than N-SLNs (p < 0.01), confirmed by the quantification analysis of fluorescence signals presented in Fig. 5D. The strong fluorescent signal observed in T-SLNs could be explained by several reasons. One of them is increased number of macrophages induced by tumor mediated inflammation-like reaction [1,44]. In addition, lymphangiogenesis induced by tumor [3,45] also enhances the delivery and uptake of MSN-probes in T-SLNs. Finally, the uptake ability of macrophages to foreign materials may also be improved by the activation of tumor cells.

At 3 weeks after MSN-probe administration, an obvious signal of T-SLNs is still observable (Fig. 5C), showing the long-term imaging ability of the MSN-probes. Ex vivo imaging confirmed the distribution of MSN-probes in both liver and T-SLNs (Supplementary Figure S4A and B), which is consistent with in vivo results. Moreover, T-SLNs have strong heterogeneously distributed fluorescence signal, while almost no signal is detected in N-SLNs at both 6 h and 21 day time points (Supplementary Figure S4C and D). An unevenly aggregated distribution of fluorescence signal was evidently observed by microscopy in tissue sections from T-SLNs (Supplementary Figure S5A and B), but almost no signal was identified in sections from N-SLNs (Fig. 5E and F). The nanoprobes were mainly localized in the subcapsular sinus (Supplementary Figure S5C), which is the main distribution zone of macrophages [46]. These data support our speculation that the increased number of activated macrophages and lymphangiogenesis are responsible for increased accumulation of MSN-probes in T-SLNs. However, the enlarged volume of T-SLNs itself may not be helpful for the uptake of the MSN-probes. These results indicate that the T-SLN uptake of MSN-probes is related to tumor metastasis to some extent, implying SLN mapping by MSN-probes might be useful for longterm tumor metastasis tracking.

3.4. MR and PET imaging

As mentioned above, optical imaging has its own limitations such as photon scattering and poor tissue penetration [47]. Gd³⁺ chelated by Si-DTTA in the MSN-probes provides high T₁ signal with MRI [23]. To investigate the feasibility of MRI of MSN-probes, three-dimensional gradient-echo scan (FLASH) images were acquired with a 7.05 T small animal MR scanner. As shown in Fig. 6, enhanced T1 signal in T-SLNs (circle) was observed after MSN-probe injection. The enlarged images of T-SLNs (inlet square) show that MSN-probes were mainly distributed in the margin of T-SLNs (arrow), which is consistent with the pattern observed with ex vivo optical imaging and fluorescent microscopy of tissue sections (Fig. 5). These data further confirm that different imaging tags within the MSN-probes localize at the same sites, as revealed by different imaging techniques. Up to 21 days, we still observed contrast of MSN-probes in T-SLNs. However, the signal enhancement effect of MSN-probes in T-SLNs with MRI was weaker compared to optical imaging. The phenomenon may be partially due to the low sensitivity of contrast enhanced MRI. Besides, MRI only showed a tomographic single layer of SLNs, while optical imaging reflected a 2D projection of the signal from the entire SLN. However, high spatial resolution and clear anatomic references from MRI images make this imaging modality highly complementary to visualize and map tumor draining SLNs.

Fig. 6.

MRI of sentinel lymph nodes in a 4T1 tumor metastatic model. MRI of lymph nodes were shown (A) before and after injection of MSN-nanoprobes for (B) 1 h, (C) 6 h, (D) 1 day, (E) 5 days and (F) 15 days. Arrow, the accumulation area of particles.

Compared with optical imaging and MRI, PET imaging is highly sensitive and can be used to evaluate whole-body distribution of imaging probes quantitatively [48]. After local injection of ⁶⁴Cu labeled MSN-probes, PET imaging were performed using an Inveon microPET scanner (Siemens). Fig. 7A displays the PET images of MSN-probes at 1 h, 6 h, 1 day and 2 days after injection. The radioactivity accumulation in T-SLNs was very high and reached nearly 80 %ID/g at 1 h p.i., while no obvious signal in N-SLNs was observed. The signal intensity in liver was low at early time point and increased gradually over time. After decay correction, the quantitative analysis of signal in T-SLNs and N-SLNs is shown in Fig. 7B. Within 2 days, the MSN-probe uptake in T-SLNs was slowly decreased from 76.7 ± 2.21 %ID/g to 61.3 ± 5.41 %ID/g, while the uptake in N-SLNs was increased from 2.3 ± 0.12%ID/g to 8.5 ± 0.58% ID/g. The signal intensity in T-SLNs was about 35 and 7 folds higher than that in N-SLNs at 1 h and 2 days, respectively. These results were consistent with that from optical imaging, indicating a successful establishment of a multiple imaging nanoplatform. Since the half-life of ⁶⁴Cu is 12.6 h, it is difficult to acquire PET images beyond 2 days, as we did for optical and MR imaging. Based on all these imaging data, a possible clinically related imaging strategy should be as followed. After injection of tri-modal MSN-probes, PET imaging would be performed first to identify any skeptical “hot spots”. Then MRI should be taken for better anatomic localization. Last, on-spot optical imaging can be used to provide imaging guidance during surgery.

Fig. 7.

PET imaging of sentinel lymph nodes in a 4T1 tumor metastatic model. (A) PET imaging of T-SLN (square dot) and N-SLN (solid line) after injection of particles for 1 h, 6 h, 1 day and 2 days. Arrow denotes bladder. Top, cross section; bottom, transverse section. (B) Quantitative comparison of signal between T-SLN and N-SLN.

4. Conclusion

We have successfully developed a mesoporous silica nanoparticle based tri-modal imaging probe and accomplished nonin-vasive, in vivo long-term imaging of tumor draining SLNs using triple imaging modalities with optical, MR and PET imaging. Due to the high stability and robustness of the imaging probe, the imaging results from different modalities are consistent and complementary. Our findings indicate that the changes of lymph nodes are closely related to tumor metastasis. A faster uptake rate and higher uptake of the multifunctional MSN-probes were observed in T-SLNs compared with N-SLNs, confirming the feasibility of these MSN-probes as contrast agents to map SLNs and identify tumor metastasis. High stability and intracellular retention ability of the MSN-probes make it possible to track SLN changes induced by tumor cell migration over a long time. Further studies will focus on optimization of particle size, ratios of three imaging tags, and delivery route for potential clinical translation. We believe that multimodal imaging of tumor draining SLNs will better our understanding of initiation, process and development of tumor metastasis to local lymph nodes.

Appendix A. Supplementary Data

Supplementary data related to this article can be found online at doi:10.1016/j.biomaterials.2012.02.060.