Breast tumor, characterized by high morbidity and mortality, is the primary factor threatening worldwide women's health [1–3]. Despite the great efforts made in the last decades, surgical excision remains the main treatment because of its simplicity of operation[4]. However, it is still challenging to completely eliminate solid tumors, leading to inevitable tumor recurrence. Furthermore, surgical excision and postoperative medications could cause primary damage to normal tissues[5–7]and organs[8, 9]. Consequently, effective synergistic strategies should be elaborated to improve the therapeutic efficacy and alleviate the side effects of breast tumor treatment[10].
Photothermal therapy (PTT) has received increasing attention in breast tumor therapy due to its advantages of minor damage to non-target tissues and low invasiveness[11, 12]. However, although PTT could achieve rapid and effective ablation of tumors; the released intracellular components would trigger adverse inflammatory responses and hamper the therapeutic efficacy[13]. In addition to pro-inflammatory cytokines and proteins, the accumulation of highly reactive oxygen species (ROS) also plays a crucial role in stimulating tumor recurrence and damaging peripheral normal cells[14–16]. Molybdenum (Mo)-based polyoxometalate (POM) clusters have emerged as promising mediators of PTT due to their small size, efficient light-to-heat conversion, good biosafety, remarkable ROS-scavenging activity and anti-inflammatory effects[17–19]. Unfortunately, conventional intravenous administration of POM would decrease their tumor accumulation, thereby weakening the POM-induced photothermal effect at the tumor site and metabolizing more rapidly[20–22]. Therefore, it is of great importance to develop novel photothermal materials for tumor-targeting treatment with ideal safety to peripheral normal cells.
Cancer immunotherapy has emerged as an revolutionary therapeutic approach to generate specific and durable anti-tumor responses by harnessing the natural immune system of the host[23, 24]. The stimulator of interferon genes (STING) is a type of intracellular signaling receptor that activates innate immune cells downstream[25–27]. Cyclic dinucleotides (CDN) are specially designed and synthesized as non-nucleotide STING-activating agonists[28–30]. Tumor-bearing CDN is a promising candidate in clinical antitumor trials (NCT02675439), and it has demonstrated efficacy in tumor-bearing mice in promoting specific inhibition of various tumors [31–33]. Nevertheless, the effectiveness of agonists is significantly reduced due to metabolic instability and rapid clearance from the body after systemic administration [31, 34, 35]. Therefore, novel strategies to improve the efficacy of CDN in tumor treatment are urgently needed.
In this study, we developed a NIR-responsive and injectable hybrid hydrogel (denoted as CDN-POM), which could reduce PTT-induced inflammation for photothermally ablating tumors and releasing CDN to activate innate antitumor immunity (Fig. 1)[36]. Compared to POM, CDN-POM hydrogels exhibited superior photothermal conversion capability of rapid heating under the same near-infrared irradiation. Apart from this, the biocompatible CDN-POM hydrogels were capable of showing remarkable ROS-scavenging activity to suppress PTT-induced inflammatory responses. Meanwhile, the CDN-POM hydrogels enabled continuable release of CDN for a prolonged time, resulting in a rapid, sustained and effective activation of the STING pathway. Thus, CDN-POM hydrogels that integrate the advantages of higher photothermal effect, excellent ROS-scavenging ability and effectively sustained activation of STING pathway could become a prospective medical composite for combined cancer therapy.
Preparation and characterization of CDN-POM hybrid hydrogels
As illustrated in Fig. 2A, the injectable CDN-POM hybrid hydrogels were successfully constructed via a DNA mediated self-assembly method. Firstly, POM was prepared by a redox method and then used as the cross linker to form hybrids with CDN, salmon sperm DNA (smDNA, 2000 bp) and polylysines (PLL,1000 bp). The transmission emission microscopy (TEM) images showed that POM exhibited an average diameter of 1–2 nm. Ultimately, the CDN-POM hydrogel was formed by electrostatic complexation of POM and smDNA and with cationic PLL (Fig. 2A). The scanning electron microscopy showed that the hydrogels proved a porous interconnected structure by adding CDN and PLL. (Fig. 2B). This suggested that PLL could electrostatically attract POM and CDN to form a compact and confined network. Then we conducted in vivo photoacoustic imaging experiments of POM and CDN-POM hybrid hydrogels in 4T1 tumor-bearing mice. Before the injection, relatively weak photoacoustic signals of tumor tissues were detected due to the background signal produced by intrinsic hemoglobin and melanin. Within 8 h after the injection of POM, the PA signal gradually disappeared, indicating that POM had been rapidly excreted from the tumor tissue (SI Fig. 1). However, at 24 h post-injection, the PA signal of the CDN-POM hybrid hydrogel was still evident. PA images showed that CDN-POM hybrid hydrogel could prolong tumor retention for the long-acting release of CDN. The rheometer results indicated that the CDN-POM hydrogel had a low viscosity (Fig. 2C), making it possible to be injected via syringe (Fig. 2D). In the measured frequency range (0.1–100 Hz), the storage modulus (G′) was larger compared to the loss modulus (G″), which proved the typical viscoelastic property of CDN-POM hydrogels. (Figure.2D).
The photothermal performance of the CDN–POM hydrogels
Subsequently, we compared the photothermal performance of POM and CDN–POM hydrogels. First, various CDN–POM hydrogels with different concentrations of POM were fabricated. The CDN–POM hydrogel (0.2 wt% POM) exhibited a fast-heating rate, and it could reach 54°C under 808 nm laser irradiation (1 W cm− 2) for 30 s. After irradiation for 3 min, the temperature of hydrogel was up to 89°C. In contrast, the POM with the same concentration exhibited less photothermal efficiency (Fig. 3A), which was confirmed by the thermal images of samples with different concentrations (Fig. 3B). In the CDN–POM hydrogels, the photothermal conversion efficiency (η) was calculated to be 43.4%, which was much higher than that of the pristine POM (32.3%) (Fig. 3D and SI Fig. 3). Those results indicated the confined network formed by the compounding of POM and CDN with PLL played an important role in the extraordinary photothermal effect, which was superior to conventional photothermal systems. Additionally, in the hydrogels, the change of the photothermal effect was negligible after several cycles of laser irradiation, well demonstrating the extraordinary photothermal stability of these hybrid hydrogel materials (Fig. 3C).
Investigation of POM-based photothermal cell-killing ability and ROS scavenging effect.
Inspired by the impressive photothermal conversion ability and negligible cytotoxicity of POM, we therefore evaluated their localized photothermal ablation effect on 4T1 cells under 808 nm laser irradiation (Fig. 4A). No obvious cell necrosis was detected in 4T1 cells when incubated with gradient concentrations of POM for different periods, further indicating the good biocompatibility of POM (Fig. 4A and 4B). The safety of the 808 nm laser irradiation was demonstrated by the fact that no obvious dead cells were observed under the power density of 2.0 W/cm2 for 5 min without adding POM. After co-incubating 4T1 cells with POM (400 µg/mL, 100 µL) and irradiating the co-incubation sample with 808 nm laser for 5 min, obvious cell necrosis (propidium iodide staining showed red fluorescence) was observed in the area irradiated by NIR light. The area of dead cells could be changed by either the concentration of POM or the power density of NIR laser irradiation.
These results demonstrated the NIR light-induced cell-killing ability of POM. Next, the ROS-scavenging activity of POM in 293T cells was further investigated through ROS threated by using H2O2. The intense green fluorescence in 293T cells, resulting from the oxidation of ROS probe (DCFH-DA) by H2O2, was gradually weakened by incubating POM with various concentrations (Fig. 4B). The successful elimination of ROS by POM significantly increased cell viability, as confirmed by typical live/dead cell staining (Fig. 4C). Therefore, the above data could demonstrate that the POM has extraordinary ability in NIR photothermal conversion and ROS scavenging.
Investigation of the in vivo anticancer effect of CDN–POM hydrogels
To investigate the anticancer effect in vivo, female BALB/c mice bearing 4T1 tumors with a volume of about 100 mm3 on the dorsum of the right hind leg were used as animal models. Five randomly grouped mice (n = 5) were treated with PBS + NIR laser, POM + NIR laser, CDN–POM hydrogels, free CDN and the CDN–POM hydrogels + NIR laser, injected around the tumors (Fig. 5A). An infrared thermal camera was applied to simultaneously record the real-time temperature changes. Compared to the slight warming (∼1.9°C) of the PBS + NIR laser group, the tumor temperature of the CDN-POM hydrogel reached 58.3°C, compared to 50.6°C of POM, which could normally lead to tumor cell necrosis. No obvious noticeable bodyweight body weight during the treatment course in all groups (SI Fig. 4), indicating that all of the therapeutic agents were non-toxic for 4T1 tumor-bearing mice. After monitoring the tumor volume for 3 weeks, significant tumor inhibition could be observed with the coexistence of the free CDN, POM + NIR and the CDN–POM hydrogel + NIR laser irradiation (Fig. 5B). Under NIR laser irradiation, the group with POM treatment showed similar rapid tumor growth performance compared to the PBS-treated mice, revealing the low toxicity of POM, which was further confirmed by body weight (SI) and pathological features of major organs (Fig. 5D).
Determination of the levels of inflammation-related cytokines
The immunostimulatory activity of the CDN–POM hydrogels in 4T1 tumor-bearing mice was evaluated. In light of the excellent activity of the CDN in stimulating the STING pathway, we further investigated whether the stimulation of the CDN–POM hydrogels could induce DC maturation in vivo. The assays focused on the expression of major histocompatibility complex II (MHC- II) and CD11c in BMDCs (splenocytes, flow cytometry). As shown in Fig. 6A, free CDN and CDN–POM hydrogels + NIR significantly upregulated the expression of MHC II and costimulatory markers of CD11c in BMDCs (SI Figure.5). Next, we performed flow analysis to study the expression of CD4+ and CD8+ T cells in the spleen after different treatments. Mice treated with PBS + NIR, POM + NIR, CDN–POM hydrogels induced an average of 4.51% CD8 + T-cells on day 8 (7 days after the injection) (Figure.6B). In contrast, mice treated with the CDN–POM hydrogels + NIR elicited 17.20% CD8+ T-cells, which was 3.81-fold higher than the PBS group (P < 0.01), and elicited 13.60% CD8+ T-cells treated with free CDN (SI Figure.6). These experimental results indicated that CDN–POM hydrogels enabled the controlled liberation of CDN, which would eventually facilitate the activation of the STING pathway and increase the tumor infiltration of CD4+ T cells and CD8+ T cells, suggesting the CDN–POM hydrogel is a promising platform to elicit potent T-cell response for efficient tumor immunotherapy.
Immunoregulatory efficacy of CDN–POM hydrogels
Although PTT is effective in inhibiting primary tumors, the adverse release of intracellular constituents caused by high-temperature PTT would induce acute inflammation that may increase the probability of tumor recurrence and metastasis. Inspired by the considerable ROS scavenging activities of POM in vitro, the inflammatory-related cytokine levels in tumors of mice after various treatments were determined. As shown in Fig. 7Aand 7B, due to the ROS-eliminating function of POM, no significant increase of inflammatory-related cytokine levels could be observed in POM + NIR group and the CDN–POM hydrogel + NIR group compared with the PBS + NIR gruop. Therefore, these findings proved that the CDN-POM hydrogels could be used for photothermal cancer treatment under NIR laser irradiation with minimal adverse inflammatory responses.