Melanin-dot–mediated delivery of metallacycle for NIR-II/photoacoustic dual-modal imaging-guided chemo-photothermal synergistic therapy

Significance

Clinical applications of Pt(II)-based anticancer agents have been hampered because of drug resistance, severe side effects, and lack of precise guidance for the therapeutic procedure. To address these issues, we use molecular-dye–modified melanin dots as a multifunctional drug delivery platform to effectively deliver a Pt(II) metallacycle to tumor sites via an enhanced permeability and retention effect. In vivo studies show that nano-agent 1 displays strong NIR-II fluorescence and photoacoustic signals from tumors with a high signal-to-background ratio and successfully guides chemo-photothermal therapy with a superior antitumor performance and reduced side effects. These promising results will provide an opportunity for the design of novel multimodal and synergistic therapeutic agents for biomedical applications.

Abstract

Discrete Pt(II) metallacycles have potential applications in biomedicine. Herein, we engineered a dual-modal imaging and chemo-photothermal therapeutic nano-agent 1 that incorporates discrete Pt(II) metallacycle 2 and fluorescent dye 3 (emission wavelength in the second near-infrared channel [NIR-II]) into multifunctional melanin dots with photoacoustic signal and photothermal features. Nano-agent 1 has a good solubility, biocompatibility, and stability in vivo. Both photoacoustic imaging and NIR-II imaging in vivo confirmed that 1 can effectively accumulate at tumor sites with good signal-to-background ratio and favorable distribution. Guided by precise dual-modal imaging, nano-agent 1 exhibits a superior antitumor performance and less severe side effects compared with a single treatment because of the high efficiency of the chemo-photothermal synergistic therapy. This study shows that nano-agent 1 provides a promising multifunctional theranostic platform for potential applications in biomedicine.

Over the last decade, discrete Pt(II) metallacycle-based agents have attracted extensive attention in many scientific fields such as supramolecular chemistry, molecular imaging, and biomedicine (1–6). Well-defined Pt(II) metallacycles can enhance their cellular uptake and binding affinities with biomolecules. Poor cellular and in vivo stability, low tumor uptake, weak fluorescence properties, and poor selectivity for cancer cells are major obstacles to the widespread use of macrocyclic Pt(II)-based agents in biomedicine (7–9). Various drug-carrier nanosystems, e.g., liposomes, have been actively explored to improve the in vivo stability and pharmacokinetics of Pt(II)-based agents (10–12). However, some potential issues must be taken into consideration. Single chemotherapy cannot simultaneously balance the needs for efficiency and safety because serious Pt(II) drug resistance and continuous administration during the entire therapy period will lead to inefficient therapy and potentially long-term systemic toxicity (13, 14). In addition, the lack of precise guidance for therapeutic procedures can reduce therapeutic efficiency. Engineered nanosystems with intrinsic multitherapeutic and imaging properties are therefore needed for precision medicine.

Inspired by the intrinsic multifunctionality of melanin biopolymers in nature, highly biocompatible and water-soluble melanin dots have been used as active nanoplatforms for aromatic structures for drug delivery (15–18). Compared with inorganic platforms, melanin exists in nature as an endogenous substance and has a better biocompatibility and clearance, which leads to reduced immunogenicity. Melanin also can be used as a theranostic platform without complicated procedures. Additionally, compared with traditional methods such as vesicle carriers, melanin dots can load more drugs through π–π stacking on the high-volume surface (16, 18). Recent studies reported that melanin dots can absorb near-infrared (NIR) optical energy and convert it into heat for photothermal therapy (PTT). The heat released by melanin dots can stimulate acoustic waves of the surrounding medium and these are finally converted to photoacoustic signals; this can provide a pretherapy imaging guidance tool. Fluorescence imaging in the second near-infrared channel (NIR-II) region is at the forefront of biomedical research because of its clear-cut advantages, such as relatively lower tissue autofluorescence and higher spatiotemporal resolution (19–24). The NIR-II modality enables diverse biological procedures to be achieved with high signal-to-background (S/B) levels and this is beneficial in biomedicine (25–30). Considering the compatibility and flexibility of NIR-II fluorescence imaging and photoacoustic (PA) imaging, a combination of these 2 promising modalities can acquire complementary information that synergistically improves cancer diagnosis and enables precise image-guided therapy with both a superior S/B ratio and high tissue penetration depth (31, 32).

Herein we describe a multifunctional theranostic nano-agent 1 which is obtained by incorporating a discrete Pt(II) metallacycle 2 and NIR-II molecular dye 3 into melanin dots (Fig. 1). Nano-agent 1 has the following advantages. Because of its NIR-II molecular dye unit, 1 can achieve good photostability and a better fluorescence performance in vivo, which results in more precise diagnosis of cancer and monitoring of tumor development (33–35). The melanin dot platform provides 1 good solubility and stability in vivo and high preferential passive accumulation at tumor sites via the enhanced permeability and retention (EPR) effect, which enables more precise cancer detection and efficient treatment. More importantly, the native PA and PTT properties of melanin dots can endow 1 with NIR-II/PA dual-modal imaging and chemo-photothermal synergistic therapeutic functions for theranostic cancer. In vivo results indicate that 1 has a high level of nonspecific tumor uptake along with a superior S/B ratio, which enables the precise guidance of the therapeutic process and evaluation of the treatment efficacy. Compared with cisplatin, Pt(II) metallacycle 2 in 1 shows more efficient antitumor activity and decreased side effects in U87MG tumor-bearing nude mice models.

Fig. 1.

(A) Structures of discrete Pt(II) metallacycle 2 and NIR-II molecular dye 3. (B) Schematic diagram of nano-agent 1 in chemo-photothermal synergistic therapy.

Results and Discussion

Preparation and Characterization of Nano-Agent 1.

The discrete Pt(II) metallacycle 2 was readily synthesized via the coordination-driven self-assembly. Multinuclear NMR analysis was used to verify the structure of 2. The ¹H NMR signals from the pyridine H_a and H_b shifted downfield after coordination-driven self-assembly to form rhomboid 2 (SI Appendix, Fig. S2). The ³¹P{¹H} NMR of rhomboid 2 showed an upfield shift of ∼6.16 ppm and a sharp singlet peak at ca. 15.68 ppm with concomitant ¹⁹⁵Pt satellites; this indicates a single phosphorus environment (Fig. 2A). Electrospray ionization time-of-flight mass spectrometry (ESI-TOF-MS) confirmed the stoichiometry of rhomboid 2; i.e., m/z = 1,184.15 for [M − 3OTf]³⁺ and m/z = 850.58 for [M − 4OTf]⁴⁺. These peaks were in good agreement with the calculated theoretical distributions and further support metallacycle formation (Fig. 2B). The NIR-II molecular dye 3 was synthesized in several steps from commercially available chemicals (Fig. 1A). Compound 3 was characterized by multinuclear NMR analysis and MALDI-TOF-MS (SI Appendix, Figs. S20–S22). PEGylated (poly(ethylene glycol) [PEG]) melanin nanoparticles were synthesized from commercial melanin granules according to a previously reported method (16–18).

Fig. 2.

(A and B) Partial ³¹P{¹H} NMR spectra of Pt(II) metallacycle 2 (A) before and (B) after coordination-driven self-assembly. (C) ESI-TOF mass spectra of discrete Pt(II) metallacycle 2. Inset shows experimental (red) and calculated (blue) results.

The theranostic probe 1 was prepared by functionalizing melanin dots with NIR-II molecular dye 3 to form 4 dots via the traditional coupling conditions and then loading Pt(II) metallacycle 2 via π–π binding to enable both imaging and therapy (Fig. 1B and SI Appendix, section A). Transmission electron microscopy (TEM) and dynamic light scattering (DLS) showed that 1 has high monodispersity in aqueous solution (∼110 nm) (Fig. 3A and SI Appendix, Fig. S23). Nano-agent 1 tends to aggregate and form nanoparticles that are larger than the free melanin dots. This is due to the conjugation with the hydrophobic NIR-II dye 3 and binding with hydrophobic drugs that both decreased the hydrophilicity of the melanin dots and induced melanin dots’ self-aggregation (16–18). Further experiments showed that the nano-agent 1 particle size was unchanged after incubation for 24 h in PBS containing 10% (vol/vol) FBS at 37 °C. This suggests that 1 has good stability in vitro (SI Appendix, Fig. S24). The maximum fluorescence emission wavelength of 1 was ∼1.0 µm, and it exhibited a bright fluorescence in the NIR-II channel (Fig. 3B). UV-visible spectrophotometry indicated that the conjugated number of 3 per melanin dot was about 5 (SI Appendix, Fig. S25). In addition to a maximum fluorescence wavelength shift to the NIR-II channel, 1 also showed better photostability than that of the traditional NIR-I theranostic agent indocyanine green (ICG) (Fig. 3C) and minimum decay in various mediums under continuous irradiation at 808 nm for 60 min (SI Appendix, Fig. S26).

Fig. 3.

(A) TEM image of melanin dots and nano-agent 1. (B) UV/vis absorbance and NIR-II fluorescence emission of nano-agent 1. Inset shows fluorescent solution of nano-agent 1. (C) Fluorescence intensity ratios of nano-agent 1 and ICG at different times under 808 nm irradiation. (D) Release of Pt(II) metallacycle 2 from nano-agent 1 at various pH values. (E) PA intensity of nano-agent 1 at various concentrations. (F) Photothermal heating curves for nano-agent 1 at different concentrations in water. (G) Photothermal heating and natural cooling cycles of nano-agent 1 (1 W/cm²).

Inductively coupled plasma mass spectrometry (ICP-MS) showed that the melanin dot platform has a high Pt(II) metallacycle loading capacity (∼20%, wt/wt). The high loading capacity can be attributed to the high volume of melanin dot surfaces. The release of Pt(II) metallacycle 2 from the melanin dots in vitro was investigated by suspending nano-agent 1 in a dialysis bag and performing diffusion under neutral or acidic conditions (pH 7.2 or 5.5). The Pt(II) metallacycle 2 content was quantified by ICP-MS. The result indicated that the release rate was similar at different pH values and ∼50% of the Pt(II) metallacycle 2 was released from 1 over 24 h (Fig. 3D). This gradual and linear drug release, which was unaffected by pH, can be attributed to the π–π interactions, as opposed to the usual electrostatic interactions between the drug and the delivery system (16). The gradual release of the Pt(II) metallacycle 2 from the melanin dot platform at tumor locations can enhance the therapeutic effect.

The PA spectrum has a broad peak extending over a wide NIR-I region (SI Appendix, Fig. S27). The intensity of the PA signals from 1 showed good linearity with increasing concentration from 31.25 to 250 µM (Fig. 3E). This verifies that 1 could be used as a PA probe. The photothermal properties of nano-agent 1 were investigated by laser irradiation at various powers (0.5 to 1.5 W/cm²). A significant temperature increase was detected, which suggests that nano-agent 1 has potential applications in PTT (SI Appendix, Fig. S28). The temperature of nano-agent 1 increased more rapidly with the increasing concentration (5 to 50 mg/mL) when exposed to an 808-nm laser at 1 W/cm²; there was no obvious temperature change in a control (PBS) group subjected to laser irradiation (Fig. 3F). Nano-agent 1 can be reversibly photothermally heated and naturally cooled for 5 cycles without any significant change, indicating good photothermal stability of 1 (Fig. 3G).

In Vitro Cell Imaging and Anticancer Efficacy Study.

Nano-agent 1 was investigated as a fluorescent probe for imaging of cancer cells via its endocytosis behavior. As shown in Fig. 4A, after incubation with nano-agent 1 (containing ∼100 nM 3) for 2 h, U87MG cells showed strong intracellular NIR-II fluorescence because of an efficient endocytosis pathway, and the corresponding average fluorescence intensity in the U87MG cells increased with increasing incubation time (SI Appendix, Fig. S29). The cellular uptake of 1 and release of Pt(II) metallacycle 2 from 1 were quantified by ICP-MS analysis. The cellular uptake of 1 gradually increased over the period from 1 to 4 h (Fig. 4B). We also normalized the biodistribution of Pt(II) metallacycle 2 in nano-agent 1 vs. the protein content, as in ref. 36. In terms of the subcellular distribution, nearly 95.2% of the Pt(II) content was within the nucleus (143 ± 12.3 ng Pt/mg protein), which was significantly higher than that in the cytoplasm (7.08 ± 0.66 ng Pt/mg protein; SI Appendix, Table S1). This may be attributed to the π–π interactions between the Pt(II) metallacycle 2 and melanin dots being disrupted under in vivo conditions; after internalization the released Pt(II) metallacycle 2 therefore entered the cell nuclei and coordinated with DNA (13, 37). The in vitro anticancer efficacies of the Pt(II) metallacycle 2 in 1 and cisplatin over the range 5 to 50 µM and 4 over the range 0.375 to 3.75 µM, with or without laser irradiation, were evaluated by using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay. As shown in Fig. 4 C and D, the photothermal antitumor activity of 4 dots with laser irradiation or the chemotherapeutic antitumor activity of 1 or cisplatin showed observable cellular toxicity even at the relatively low dose of up to 5 µM for 1 and cisplatin or 0.375 µM for 4 dots. The better anticancer efficiency achieved with 1 plus laser irradiation can be attributed to the synergistic effects of the photothermal activity from melanin dots and chemotherapeutic activity of the released Pt(II) metallacycle 2.

Fig. 4.

(A) NIR-II fluorescence images of U87MG cells incubated with nano-agent 1. NIR-II images were taken at an exposure time of 200 ms and 808 nm excitation wavelength with an NIR-II fluorescence microscope. (B) Time-dependent quantitative analysis of the amount of Pt in U87MG cells after incubation. (C) Cellular toxicities of PBS, cisplatin, and nano-agent 1 toward U87MG cells. (D) Cellular toxicities of PBS, 4, and nano-agent 1 with laser irradiation toward U87MG cells.

In Vivo NIR-II Fluorescence/PA Dual-Modal Imaging and Theranostic Study.

To evaluate the NIR-II fluorescence imaging performance of nano-agent 1 in vivo, imaging of the lymphatic system at the rear paws of C57BL/6 mice was performed. The lymphatic node was clearly distinguished from the surrounding background with a high S/B ratio and sharp images (∼6.06 and 161 µm) (SI Appendix, Fig. S30). We also noninvasively observed strong NIR-II signals in the vertebrae and thigh bone within 5 min after i.v. injecting 1 in C57BL/6 mice (SI Appendix, Fig. S31). Ex vivo NIR-II imaging also provided shape images of ribs, vertebrae, and other bones (SI Appendix, Fig. S32). These results confirm that nano-agent 1 has advantages over traditional visible and NIR-I theranostic agents. The pharmacokinetics of the Pt content on 1 were investigated by injecting C57BL/6 mice (n = 3) with 1 and determining the Pt content in the plasma at various times by ICP-MS. The results show that the blood circulation time of 1 is longer than that of cisplatin (SI Appendix, Fig. S33), thus partly resulting in the enhanced efficacy over cisplatin.

The U87MG tumor-bearing nude mice (n = 3) were i.v. injected with 1 (2 mg Pt/kg) in PBS, and the NIR-II fluorescence and PA imaging signals of the living mice were monitored to investigate the dual-modal images in vivo. In the NIR-II fluorescence and PA images at 2, 4, 6, 12, and 24 h postinjection, the tumor was clearly distinguished from the surrounding background tissues (Fig. 5A). Semiquantitative analysis of the NIR-II and PA signals showed that the uptake of 1 by the tumor gradually increased with time and reached a maximum at 6 h (Fig. 5B). The S/B ratio for NIR-II and PA both increased steadily in the first 2 h after injection and reached a maximum value at 6 h postinjection (Fig. 5C). To verify the in vivo imaging results, ex vivo imaging of tumors and other major organs or tissues was performed at 24 h postinjection. The NIR-II fluorescence signals from the tumor after injection of 1 were much brighter than those from other organs (SI Appendix, Fig. S34); this is consistent with the in vivo imaging results. The Pt distributions in major organs and at the cancer sites were quantitatively determined by injecting 1 or cisplatin into U87MG tumor model animals (n = 3 per group). At 12 or 24 h postinjection, the amounts of Pt in the tumor and normal tissues were determined by ICP-MS to evaluate the accumulation of 1. Fig. 5 D and E shows a high accumulation in the tumor of Pt(II) from 1 and a lower Pt(II) uptake by normal organs, which can be attributed to the EPR effect. In contrast, cisplatin displayed a significantly higher accumulation of Pt(II) in the normal organs than in the tumor. All these findings indicate that nano-agent 1 has favorable in vivo distribution and high tumor accumulation.

Fig. 5.

(A) NIR-II fluorescence and PA images of U87MG tumor mice (n = 3) at different times after tail vein injection of 1. (B) Semiquantitative analysis of NIR-II/PA signals from tumor regions on U87MG tumor-bearing nude mice at various times. (C) S/B ratios for NIR-II and PA images at various times. (D and E) The distributions of (D) cisplatin and (E) nano-agent 1 in main organs at 12 and 24 h after tail vein injection. Li, liver; Ki, kidney; Sp, spleen; Lu, lung; He, heart; Tu, tumor.

Guided by the NIR-II fluorescence/PA imaging in vitro and in vivo, the chemo-photothermal synergistic therapeutic efficacy of nano-agent 1 (2 mg Pt/kg) against U87MG tumors in mice was then evaluated under irradiation with an 808-nm NIR laser. In the experiments, 4 dots (800 mg/kg) with 808 nm NIR laser irradiation, nano-agent 1 (2 mg Pt/kg), cisplatin (2 mg Pt/kg), PBS with 808 nm NIR laser irradiation, and PBS were set as controls. PBS (200 µL) and the various therapeutic agents were injected into U87MG tumor-bearing nude mice via the tail vein in a single dose. The in vivo imaging data showed that maximum accumulation of 1 was achieved at around 6 h; PTT was therefore performed at 6 h postinjection. Fig. 6 A and B shows that after 60 s irradiation, the temperature at the tumor sites treated with 1 and 4 dots rose to ∼45.0 °C; this is much higher than the temperature increase observed for the group treated with PBS under 808 nm NIR laser radiation (∼30.0 °C). After administration of the various treatments, the body weights and tumor growth of all of the mice were recorded every 2 d up to 20 d. The group treated with cisplatin showed obvious body weight losses for several days after a single administration, whereas no significant body weight losses were observed for all of the other groups (Fig. 6C). These results indicate that 1-based therapy has better in vivo biocompatibility and decreased systemic side effects. Compared with the rapid growths observed for the PBS and PBS with laser groups, the tumor growths for groups treated with a single chemotherapy (1 and cisplatin) were limited (Fig. 6 D and E). The antitumor efficiency of 1 was higher than that of cisplatin because of the EPR effect. Tumor growth in mice treated by PTT alone (4 dots with laser) and chemo-photothermal therapy (1 with laser) was clearly inhibited for all groups, but the tumors treated with PTT alone eventually recurred after 12 d. This indicates incomplete ablation of the tumor by single PTT. Additionally, the survival rate of the U87MG tumor-bearing mice treated with 1 under laser irradiation was clearly higher than those of the other groups after 60 d (Fig. 6F), which indicates effective chemo-photothermal synergistic therapy with 1. Hematoxylin and eosin (H&E) stains of organs excised from all groups at 20 d showed no obvious adverse effect with 1 dose administration, indicating the negligible histological toxicity and biocompatibility of 1 (SI Appendix, Fig. S35). The H&E stains showed significant tumor cell damage with nuclear membrane fragmentation and nuclei pyknosis for the group treated with 4 under laser irradiation. For the other 4 groups, no obvious tumor cell damage was observed (SI Appendix, Fig. S36). Furthermore, TUNEL (transferase-mediated dUTP nick end-labeling) stains indicated that PTT caused the highest tumor cell apoptosis. On the basis of the passive tumor-targeting capacity, the number of apoptotic cells in tumors from mice treated with 1 was higher than that in the case of mice treated with cisplatin. The tumors in mice treated with 1 under laser irradiation were ablated completely; therefore there is no result of H&E and TUNEL stains for 1 with laser irradiation. These results suggest that chemo-photothermal therapy displays the highest in vivo antitumor efficiency.

Fig. 6.

(A) IR thermal images of mice treated with PBS, nano-agent 1, and 4 dots with laser irradiation. (B) Temperature changes (T ‒ T₀) of mice treated with PBS, nano-agent 1, and 4 during laser irradiation at various times. (C) Relative body weights of mice treated with different formulations. (D) Representative photographs of tumors in mice treated with different formulations. (E) Relative tumor volumes for mice treated with different formulations. (F) Kaplan–Meier survival curves for tumor-bearing mice treated with different formulations.

Conclusion

We prepared a Pt(II) metallacycle-based multifunctional theranostic nano-agent 1 by incorporating Pt(II) metallacycle 2 and NIR-II molecular dye 3 into melanin dots. Nano-agent 1 showed good stability, optical properties, and passive targeting ability for tumors. NIR-II fluorescence imaging with molecular dye 3 and complementary PA imaging with melanin dots strengthened the signals and enabled guided therapy. A combination of the antitumor activity of Pt(II) metallacycle 2 and inner photothermal properties of melanin dots enabled synergistic chemo-photothermal therapy to be precisely performed and gave the desired treatment effects with lower side effects and lengthened lifetimes. These results indicate that integrating PTT with chemotherapy can greatly decrease the possibility of drug resistance and tumor recurrence under the guidance of intrinsic signal feedback from theranostic agents. Nano-agent 1 therefore provides a promising multifunctional theranostic platform for biomedical applications.

Materials and Methods

UV-Vis absorbance was recorded on a PerkinElmer Lambda 25 UV-Vis spectrophotometer. The NIR-II fluorescence microscope and NIR-II in vivo imaging system were purchased from Suzhou NIR-Optics Technologies Co., Ltd. Hydrodynamic diameter was measured using a Malvern Zetasizer Nano ZS. HPLC was performed on a Dionex HPLC System (Dionex Corporation) and a reversed-phase C18 column was used for analysis (Phenomenax, 5 μm, 4.6 mm × 250 mm) and semipreparation (Agilent, 5 μm, 10 mm × 250 mm). TEMs were recorded on a Hitachi system and the measurements were conducted with transmission electron microscopy operated at 200 kV. All animal experimentation was performed under the approval of Central China Normal University Administrative Panel on Laboratory Animal Care.