Alendronate/cRGD-Decorated Ultrafine Hyaluronate Dot Targeting Bone Metastasis

In this study, we report the hyaluronate dot (dHA) with multiligand targeting ability and a photosensitizing antitumor model drug for treating metastatic bone tumors. Here, the dHA was chemically conjugated with alendronate (ALN, as a specific ligand to bone), cyclic arginine-glycine-aspartic acid (cRGD, as a specific ligand to tumor integrin αvβ3), and photosensitizing chlorin e6 (Ce6, for photodynamic tumor therapy), denoted as (ALN/cRGD)@dHA-Ce6. These dots thus prepared (≈10 nm in diameter) enabled extensive cellular interactions such as hyaluronate (HA)-mediated CD44 receptor binding, ALN-mediated bone targeting, and cRGD-mediated tumor integrin αvβ3 binding, thus improving their tumor targeting efficiency, especially for metastasized MDA-MB-231 tumors. As a result, these dots improved the tumor targeting efficiency and tumor cell permeability in a metastatic in vivo tumor model. Indeed, we demonstrated that (ALN/cRGD)@dHA-Ce6 considerably increased photodynamic tumor ablation, the extent of which is superior to that of the tumor ablation of dot systems with single or double ligands. These results indicate that dHA with multiligand can provide an effective treatment strategy for metastatic bone tumors.


Introduction
Extremely small-sized particles with a few nanometers in diameter have been extensively utilized to control or visualize biological functions in the field of biology and improve the disease-targeting ability of drugs in the field of pharmaceutics [1][2][3][4]. Indeed, these particles such as quantum dots, carbon dots, and polymeric dots, can efficiently interact with the cell receptors, cell proteins, genes, and cytokines owing to their surface functionality and tailor-made extremely small-sized configuration [5][6][7][8][9]. However, the innate toxicity of such dots should be considered in the course of accompanying various biomedical applications [10]. Especially, the fabrication of biofunctional dots with low cytotoxicity can offer various possibilities for realizing successful dot-based therapeutics and diagnosis [11][12][13].
Recently, our groups developed organic hyaluronate dots with biocompatible/biodegradable properties [4,11,14]. These dots (3-10 nm in diameter) conjugated with an antitumor model drug have exhibited increased in vitro cellular interaction and in vivo tumor targeting ability, resulting in a significant improvement of in vitro and in vivo tumor ablation [4,14]. Based on these studies, we here designed multifunctional hyaluronate dots with more tumor-specific targeting ligands to aggressively attack tumor cells with bone metastasis phenotype, and use them to improve the treatment efficacy of metastatic tumors.
It is known that bone metastasis originating from the relocation of circulating tumor cells usually causes pathogenic infiltration and multiple fractions, which create complicated in vivo conditions [15][16][17][18]. Actually, conventional antitumor drugs are significantly limited in penetration into bone-metastasized tumor tissues owing to the conventional densely packed metastatic tumor tissues in bone regions [19][20][21][22]. However, it is interesting to note that bisphosphonate drugs inducing apoptosis in osteoclasts and inhibiting osteolysis have shown their potential in high interaction with bone tissues owing to their high affinity (e.g., electrostatic interaction) for bone hydroxyapatite, revealing their attractive role as bone-specific ligands [23][24][25][26][27]. Furthermore, it is also known that hyaluronic acid (HA), a linear biodegradable polysaccharide, can specifically interact with CD44 receptors overexpressed on various metastatic tumors [28][29][30][31][32]. The tumor-targeting peptide, cyclic arginine-glycine-aspartic acid (cRGD) can also bind to α v β 3 integrin receptors overexpressed on metastatic breast tumors [33][34][35][36]. Therefore, in this study, the hyaluronate dot (dHA) with multiligand targeting abilities was developed to treat in vivo metastatic bone tumors. First, dHA was synthesized using the chemical conjugating method by using C 60 and HA [4,14]. Then, the dHAs were chemically conjugated with alendronate (ALN), cRGD, and chlorin e6 (Ce6, as a photodynamic antitumor agent), denoted as (ALN/cRGD)@dHA-Ce6. We expect that the dHA with multiple ligands (HA, ALN, and cRGD) facilitates multiple receptor-mediated cellular internalizations and provides multiple routes for efficient drug uptake in bone-metastasized tumors, resulting in improved photodynamic tumor ablation (Figure 1a). In particular, we investigated the in vitro/in vivo tumor targeting ability and antitumor efficacy of (ALN/cRGD)@dHA-Ce6 against MDA-MB-231 bone-metastasized tumors.

Hydroxyapatite Binding Analysis
The binding affinity of each dHA sample (1 mg/mL) and (ALN/cRGD)@HDOC-Ce6 NP (1 mg/mL) to bone-like hydroxyapatite particles (10 mg) in PBS solution was analyzed using a Cary 1E UV/visible spectrophotometer (Varian Inc., Palo Alto, CA, USA) by analyzing the light absorbance of solution [41,44,45]. Here, each sample (in pH 7.4 150 mM PBS) was mixed with hydroxyapatite particles (≈2.5 µm in diameter) and incubated with mechanical shaking (100 rpm) at 37 • C. At the specified time point, the incubated sample was centrifuged at 5000 rpm for 5 min, and the supernatant was collected. The light absorbance of the collected supernatant was measured at a wavelength of 670 nm. Consequently, the binding affinity was calculated by the following formula: binding affinity (%) = (A 0 − A)/A 0 × 100 (%), where A 0 is initial absorbance and A is absorbance at the specified time [46].

In Vitro Phototoxicity
MDA-MB-231, A549, BT-474, and NIH3T3 cells were used to verify the phototoxicity of each dHA sample and (ALN/cRGD)@HDOC-Ce6 NP under light irradiation. Here, the cells were incubated in type I collagen solution without or with hydroxyapatite particles (mimicking the live in vivo bone environment) and then incubated at 37 • C for 2 h. Next, the collagen gel containing cells in RPMI-1640 or DMEM medium was flipped to expose the cells on the surface as shown in Figure 4 of [46]. These cells were incubated with each dHA sample (equivalent Ce6 10 µg/mL), (ALN/cRGD)@HDOC-Ce6 NP (equivalent Ce6 10 µg/mL), and free Ce6 (10 µg/mL) at 37 • C for 4 h, washed three times with PBS (pH 7.4, 150 mM), and then were irradiated using a 670 nm laser source (5.2 mW/cm 2 for 10 min). The treated cells were further incubated at 37 • C for 12 h. Subsequently, we measured cell viability using a CCK-8 assay [14,35,36,41,42]. In addition, the original toxicity of each dHA samples and (ALN/cRGD)@HDOC-Ce6 NP without light irradiation was evaluated after 24 h of treatment [14,35,36,41,42].

Animal Care
All animal experiments were performed using 6-8 weeks old female BALB/c nude mice (Orient Bio Inc., Seongnam, Korea) and progressed under the guidelines of an approved protocol from the Institutional Animal Care and Use Committee (IACUC, the project identification code: 2018-016, July, 09, 2018) of the Catholic University of Korea [4,41,42].

Ex Vivo Photodynamic Tumor Therapy Using a Bone Metastasis Model
To evaluate the photodynamic antitumor efficacy of each sample using a bone metastasis model, BALB/c nude mice were euthanized using carbon dioxide asphyxiation, and their tibias were extracted under sterile conditions. The separated tibias were placed into 6-well plates and incubated with MDA-MB-231 tumor cells (1 × 10 6 cells/mL) in RPMI-1640 medium at 37 • C for 48 h [27,47]. The tibias with MDA-MB-231 tumor cells were incubated with each dHA sample (equivalent Ce6 10 µg/mL), (ALN/cRGD)@HDOC-Ce6 NP (equivalent Ce6 10 µg/mL), and free Ce6 (10 µg/mL) at 37 • C for 4 h, washed three times with PBS (pH 7.4, 150 mM), and then irradiated using a 670 nm laser source (5.2 mW/cm 2 for 10 min). The treated cells on tibias were further incubated at 37 • C for 12 h and then fixed using glutaraldehyde, ethanol, and HMDS. The resulting tibias were monitored using a scanning electron microscopy (SEM, S-4800, Hitachi, Tokyo, Japan) [27,47].

In Vivo Biodistribution
We prepared in vivo bone metastasized tumor model using BALB/c nude mice intraosseously injected with MDA-MB-231 tumor cells (1 × 10 7 cells in pH 7.4, 150 mM PBS) [48]. After 21 days, the in vivo transplanted tumors were confirmed using a micro CT imaging scanner (CLS140083, PerkinElmer Inc., Waltham, MA, USA) with parameters of 90 kV, 88 µAs, and 4 min scan time [41,42]. Next, each dHA sample (equivalent Ce6 2.5 mg/kg) and free Ce6 (2.5 mg/kg) were intravenously administered to the BALB/c nude mice to treat the MDA-MB-231 metastasized tumor. Importantly, fluorescence images of each sample (with fluorescent Ce6) at the tumor site were obtained at 1, 4, and 8 h post-injection using a Fluorescence-labeled Organism Bioimaging Instrument (FOBI, Neo-Science, Seoul, Korea) [4,49]. In addition, the tumors (in the right leg) and major organs (heart, lung, liver, kidney, and spleen) were harvested from the BALB/c nude mice (at 24 h post-injection) euthanized by carbon dioxide asphyxiation, and then assessed using FOBI analysis [4,49].

In Vivo Photodynamic Tumor Therapy
Each dHA sample (equivalent Ce6 2.5 mg/kg), free Ce6 (2.5 mg/kg), and control (saline) was intravenously injected into the tumor-bearing BALB/c nude mice through their tail vein. At 12 h post-injection, the metastasized tumor sites (tibias) of the BALB/c nude mice were locally irradiated for 40 min at a light intensity of 5.2 mW/cm 2 with a 670 nm laser source. The tumor volume was calculated using the following formula: tumor volume = length × (width) 2 /2. The relative tumor volume change (V/V 0 ), where V is the tumor volume at a given time and V 0 is the initial tumor volume, was plotted to evaluate the photodynamic tumor ablation of (ALN/cRGD)@dHA-Ce6 [14,35,36]. Furthermore, the micro CT images of tumor sites at 14 days post-injection were captured using a micro CT imaging scanner [41,42].

Statistics
All of the experimental results were analyzed using Student's t-test or ANOVA at a significance level of p < 0.01 (**) [4,14,35].
Next, MDA-MB-231 tumor cells were cultured on the surface of ex vivo tibias (as an ex vivo bone metastasized tumor model) to determine the phototoxicity of the dHA samples under ex vivo conditions [27,47]. After 48 h, the cells growing on tibias were treated with dHA samples (equivalent Ce6 10 µg/mL), (ALN/cRGD)@HDOC-Ce6 NP (equivalent Ce6 10 µg/mL), and free Ce6 (10 µg/mL) for 4 h, irradiated using a 670 nm laser source (5.2 mW/cm 2 for 10 min), and then further incubated for 12 h. Figure 5b shows the SEM images of tibias treated with each sample. In particular, the tibia treated with (ALN/cRGD)@dHA-Ce6 exhibited a decreased population of tumor cells, indicating highly increased tumor ablation ability of (ALN/cRGD)@dHA-Ce6. However, the tibia treated with the other samples showed partial proliferation and distribution of tumor cells on their surface, revealing their limited therapeutic activities.
The in vivo tumor-targeting ability of each dHA sample was also investigated using a tumor-bearing mouse model with MDA-MB-231 tumors implanted on the right tibia. The dHA sample (equivalent Ce6 2.5 mg/kg) or free Ce6 (2.5 mg/kg) was intravenously injected to the tail vein of mice, and their fluorescence images were obtained for 24 h using a FOBI (Figure 6a,b) [4,49]. As shown in Figure 6a, (ALN/cRGD)@dHA-Ce6 exhibited a strong Ce6 fluorescence signal at the tumor sites after 8 h post-injection. However, (ALN)@dHA-Ce6, (cRGD)@dHA-Ce6, and free Ce6 exhibited relatively weak Ce6 fluorescence because of the relatively weak tumor-binding ability. Furthermore, we verified the Ce6 fluorescence intensity in excised organs (heart, lung, liver, kidney, spleen, and tumor) at 24 h post-injection (Figure 6b) [4,49]. As a result, (ALN/cRGD)@dHA-Ce6 and (ALN)@dHA-Ce6 exhibited strong Ce6 fluorescence in the tumor site on the right tibia. In contrast, (cRGD)@dHA-Ce6 without ALN and free Ce6 exhibited relatively weak Ce6 fluorescence in the tumor site. It was observed that the fluorescence intensity of (ALN/cRGD)@dHA-Ce6 in normal organs (heart, lung, liver, kidney, and spleen) was relatively weak, indicating reduced accumulation to normal organs.
To confirm the in vivo photodynamic antitumor efficacy of the dHA samples, the dHA samples (equivalent Ce6 2.5 mg/kg), free Ce6 (2.5 mg/kg), and control (saline) were intravenously injected to MDA-MB-231 tumor-bearing nude mice, and tumor sites were irradiated using a 670 nm laser source (5.2 mW/cm 2 for 10 min) (Figure 6c). Importantly, the (ALN/cRGD)@dHA-Ce6 resulted in a significant tumor volume regression after 14 days post-injection ( Figure 6d); the relative tumor volume in the nude mice injected with the (ALN/cRGD)@dHA-Ce6 was approximately 2.9, 1.9, 9.5, and 13.4 times smaller than those of the nude mice injected with the (cRGD)@dHA-Ce6, (ALN)@dHA-Ce6, free Ce6, and saline (control), respectively. Furthermore, the micro CT images (Figure 6e) support that (ALN/cRGD)@dHA-Ce6 was highly effective in inhibiting MDA-MB-231 tumor cells, presenting no difference from the CT image of a normal right leg.
Overall, the results of in vitro/in vivo studies demonstrate that (ALN/cRGD)@dHA-Ce6-mediated photodynamic antitumor therapy can provide an efficient strategy to treat metastatic bone tumors. Furthermore, we confirmed that the targeting efficiency of ALN moieties to bone-metastasized tumors was quite high. Nevertheless, since the in vivo antitumor activity of (cRGD)@dHA-Ce6 with cRGD moieties is not bad, we think that (ALN/cRGD)@dHA-Ce6 with multiple tumor-targeting ability will have various advantages for targeting complicated in vivo tumors. Of course, more clear proof should be clarified through various in vivo antitumor experiments in the future. targeting complicated in vivo tumors. Of course, more clear proof should be clarified through various in vivo antitumor experiments in the future.

Conclusions
In this study, (ALN/cRGD)@dHA-Ce6 was successfully synthesized for highly efficient photodynamic therapy of metastatic tumors. The in vitro/in vivo results demonstrated that the multiple tumor-targeting ability of (ALN/cRGD)@dHA-Ce6 with HA, cRGD, and ALN as tumor-binding ligands enhanced tumor cell binding affinity and improved photodynamic tumor ablation. Based on the results of this study, it is concluded that the extremely small-sized dot system with multiple ligands can be effective in treating metastatic tumors, although in vivo pharmacokinetics evaluation should be performed in the future.

Conclusions
In this study, (ALN/cRGD)@dHA-Ce6 was successfully synthesized for highly efficient photodynamic therapy of metastatic tumors. The in vitro/in vivo results demonstrated that the multiple tumor-targeting ability of (ALN/cRGD)@dHA-Ce6 with HA, cRGD, and ALN as tumor-binding ligands enhanced tumor cell binding affinity and improved photodynamic tumor ablation. Based on the results of this study, it is concluded that the extremely small-sized dot system with multiple ligands can be effective in treating metastatic tumors, although in vivo pharmacokinetics evaluation should be performed in the future.