A biomimetic therapeutic nanovaccine based on dendrimer–drug conjugates coated with metal–phenolic networks for combination therapy of nasopharyngeal carcinoma: an in vitro investigation

Abstract

Developing a minimally invasive and potent therapy for nasopharyngeal carcinoma is still challenging. In this study, we report a photothermal nanovaccine based on phenylboronic acid (PBA)-modified poly(amidoamine) dendrimers of generation 5 (G5) attached with indocyanine green (ICG) as a photothermal agent, toyocamycin (Toy) as an endoplasmic reticulum stress (ERS) drug, and Mn²⁺-coordinated metal–phenolic networks. The developed nanocomplexes are camouflaged with homologous apoptotic cancer cell membranes, leveraging membrane proteins as an antigenic reservoir and incorporating the immune adjuvant cytosine–guanine (CpG) oligonucleotide to obtain the final nanovaccine formulation. The prepared nanovaccine with a size of 72.4 nm displays satisfactory colloidal stability and photothermal conversion efficiency (36.7%), and is capable of targeting cancer cells and inducing apoptosis under laser irradiation through combined ICG-mediated photothermal therapy, Toy-enabled chemotherapy and Mn²⁺-mediated chemodynamic therapy. Meanwhile, the combined therapeutic effects can elicit immune responses to mature dendritic cells through the immunogenic cell death of cancer cells and the inserted CpG adjuvant/apoptotic cancer cell membranes, and polarize tumor-associated macrophage cells to the antitumor M1 phenotype. The antitumor efficacy of the nanomedicine platform was proven by the test of the penetration and therapeutic inhibition of 3-dimensional tumor spheroids in vitro. The developed functional nanomedicine integrated with different therapeutic modes may be developed as a biomimetic therapeutic nanovaccine for nasopharyngeal carcinoma treatment.

---

Introduction

Nasopharyngeal carcinoma (NPC) is a common malignant tumor of the head and neck, and around half of the new cases worldwide emerge in China each year.¹ The traditional treatments for NPC are mainly radiotherapy and chemotherapy, which often result in damage to healthy tissues, leading to a poor prognosis and diminished quality of life for patients.^2,3 Hence, the development of a novel minimally invasive therapeutic strategy to tackle NPC is urgent.

Photothermal therapy (PTT) converts light energy into heat to ablate tumor tissue with minimal adverse effects.^4,5 Among many photothermal agents, indocyanine green (ICG) has reliable biosafety and excellent photothermal conversion ability,⁶ and an increasing number of studies have applied ICG for photothermal or photodynamic therapy.⁷ However, the instability of ICG and its lack of tumor targeting specificity significantly restrict its therapeutic effectiveness. To overcome these limitations, ICG has been conjugated with nanoplatforms to improve its stability and targeting specificity.⁸ Among diverse nanoplatforms, poly(amidoamine) (PAMAM) dendrimers are a family of nanoscale macromolecules that are highly branched, biocompatible after surface modification and functionalizable through periphery covalent conjugation.^9–11 In our earlier work, we developed an ICG-linked generation 5 (G5) PAMAM nanoplatform possessing good stability as well as photothermal conversion efficiency.⁶ Overall, PAMAM dendrimers have been used as effective carriers to deliver different chemotherapeutics, enabling efficient drug delivery within the tumor microenvironment (TME).^12,13 It has been reported that small molecule drugs with 1,2- or 1,3-dihydroxyl groups can be conjugated to PAMAM dendrimers through dual pH- and ROS-responsive boronic ester bonds to achieve precise release at the tumor site with improved bioavailability.¹⁴

In addition to PTT, chemodynamic therapy (CDT) operating based on Fenton or Fenton-like reactions with the H₂O₂ in the TME¹⁵ has been emerging due to its high selectivity and low invasiveness.^16–18 Metal–polyphenol networks (MPNs), known for their exceptional biocompatibility and chemical stability, have been developed as potent drug delivery carriers.^19,20 In particular, chelated metal ions such as Fe(II), Mn(II), or Cu(II) demonstrate significant potential in CDT by catalyzing the conversion of endogenous H₂O₂ in the TME to the highly toxic hydroxyl radicals (∙OH), which also depletes the intracellular glutathione (GSH) to induce oxidative stress (OS) for cancer cell apoptosis.²¹ Recent studies have revealed that Mn²⁺ not only facilitates the process of CDT but also activates the cyclic GMP–adenosine monophosphate (AMP) synthase (cGAS)–stimulator of interferon gene (STING) pathway,^22–25 thereby enhancing dendritic cell (DC)-mediated antigen presentation and activating the antitumor immune response.²⁶ Oxidative stress can further promote the endoplasmic reticulum stress (ERS) state of tumor cells, and as one of the ERS-amplification drugs for tumor therapy, toyocamycin (Toy) could act on the adaptive regulatory pathway of ERS to kill cancer cells.²⁷

For effective cancer therapy, it is necessary to create a nanoplatform that can specifically target cancer cells with prolonged blood circulation by preventing phagocytosis by macrophages and evading clearance by the reticuloendothelial system (RES).^28–33 Cancer cell membranes have been widely used to camouflage nanodrugs for efficient tumor delivery due to their homologous targeting specificity.³⁴ Recent studies have shown that nanocomplexes camouflaged by apoptotic cancer cell membranes (aCM) exhibit enhanced immunogenicity.³⁵ In addition, aCM can also act as a tumor antigen library to activate immune responses, thereby boosting the efficacy of cancer vaccines.³⁵

Considering the complexity of NPC treatment and in order to develop an advanced nanomedicine formulation with improved treatment efficacy that can generate immune responses, we attempted to combine the advantages of dendrimer/MPN carrier systems and develop an integrated nanomedicine platform for combined PTT/CDT/chemotherapy of NPC and explored its therapeutic efficacy and the induced immune responses in vitro. In this study, we developed a photothermal nanovaccine integrated with the chemical drug Toy, the photothermal agent ICG, Mn²⁺-based MPNs, the immune adjuvant cytosine phosphate guanine (CpG), and the antigen library aCM with homologous cancer cell targeting specificity (Scheme 1). First, G5 PAMAM dendrimers were partially surface-attached with Toy through pH/ROS-responsive boronic ester bonding and to ICG via amide bonding, then coated with Mn²⁺–tannic acid (TA) MPNs, and further camouflaged with aCM, followed by final insertion of cholesterol-modified CpG into the aCM. The prepared GIT@TM/C-aCM (G, G5; I, ICG; T, Toy; TM, Mn²⁺–TA; C, CpG) nanovaccine containing many dendrimers within one particle was systematically characterized in terms of its physicochemical properties, and its therapeutic potential for combined targeted therapy including PTT/CDT/chemotherapy was investigated in detail in vitro along with the induced immune responses. Lastly, its therapeutic effect was also proven by the test of the penetration and inhibition of 3-dimensional (3D) tumor spheroids in vitro.

Scheme 1 Preparation of GIT@TM/C-aCM nanomedicine for combined PTT/CDT/chemotherapy of NPC with elicited immune responses in vitro.

Results and discussion

Synthesis and characterization of the GIT@TM/C-aCM nanoparticles (NPs)

According to similar protocols in the literature,¹² amine-terminated G5 PAMAM dendrimers were partially functionalized with phenylboronic acid (PBA) and subsequently conjugated with indocyanine green (ICG), resulting in the formation of G5.NH₂-PBA-ICG dendrimers. ¹H NMR data show that there are around 16.5 PBA molecules attached to each G5 dendrimer (Fig. S1, ESI†) through comparison of the integration of PBA proton peaks at 7.0–8.0 ppm and the G5 methylene protons at 2.4–3.4 ppm. The loading of ICG onto G5.NH₂-PBA-ICG was quantified to be 7.1 ICG moieties per G5 dendrimer by UV-vis spectroscopy based on the ICG absorbance at 808 nm/concentration calibration curve (Fig. S2, ESI†). The residue G5 dendrimer terminal amines were then acetylated to form the G5.NHAc-PBA-ICG dendrimer, as evidenced by the NMR spectrum that clearly indicates the appearance of an acetyl proton signal at 1.9 ppm after the acetylation reaction (Fig. S1, ESI†). Through the reaction of Toy hydroxyl groups with the boric acid groups of G5.NHAc-PBA-ICG to form boronic ester bonds, Toy was attached to the dendrimers to form the G5.NHAc-ICG-Toy product. Fourier transform infrared (FTIR) spectroscopy was used to analyze the formation of G5.NHAc-PBA-ICG and G5.NHAc-ICG-Toy conjugates (Fig. 1(a)). The peaks at 1644 and 1557 cm⁻¹ in both G5.NHAc-PBA-ICG and G5.NHAc-ICG-Toy correspond to the amide bonds formed between the carboxyl groups of PBA and the G5 terminal amines. For the G5.NHAc-ICG-Toy conjugate, the absorption at 1009 cm⁻¹ can be ascribed to the sulfonic acid group of ICG, and the new peaks at 2161 cm⁻¹ can be attributed to the C≡N of Toy, suggesting the formation of G5.NHAc-ICG-Toy conjugates. Subsequently, we analyzed the encapsulation efficiency (EE) and loading capacity (LC) of Toy in GIT by UV-vis spectroscopy (Fig. S3, ESI†). The EE and LC of Toy in GIT were calculated to be 78.8% and 11.2%, respectively. Afterward, the G5.NHAc-ICG-Toy conjugates were coated with the TA–Mn MPNs to obtain G5.NHAc-ICG-Toy@TM (for short, GIT@TM) NPs. The Mn content was estimated to be 3.3 ± 0.2 wt% with inductively coupled plasma-optical emission spectroscopy (ICP-OES).

Fig. 1 (a) FTIR spectra of G5.NHAc-PBA-ICG and G5.NHAc-ICG-Toy. (b) UV-vis absorption spectra of ICG, CpG-Cy3, GIT@TM/aCM and GIT@TM/C-aCM. (c) TEM image of GIT@TM/C-aCM NPs and (d) the corresponding size distribution histogram. (e) SDS-PAGE analysis of protein markers, GIT@TM, aCM, and GIT@TM/aCM. (f) UV-vis spectra of MB, MB after reaction with H₂O₂, MB reacted with H₂O₂ + GIT@TM/C-aCM, or GIT@TM/C-aCM alone; each mixture was kept at 37 °C for 3 h ([MB] = 10 μg mL⁻¹ and [H₂O₂] = 10 mM). (g) Cumulative release of Toy from GIT@TM/C-aCM at pH 7.4 or 6.5 in the presence or absence of H₂O₂ (10 mM). (h) The temperature change of GIT@TM/C-aCM aqueous suspension at varying concentrations ([ICG] = 0, 2.5, 5, 10, or 20 μg mL⁻¹) under 808 nm laser irradiation (1.0 W cm⁻²) for 10 min. (i) Heating and cooling curves of GIT@TM/C-aCM ([ICG] = 10 μg mL⁻¹) for five cycles by turning the laser on/off.

Then, the aCM were camouflaged on the surface of GIT@TM NPs. To extract the aCM, we utilized free ICG and a laser to induce the apoptosis of FAT 7 cells. First, to optimize the timing for laser irradiation, the cellular uptake of free ICG in FAT7 cells at different incubation time points was evaluated by flow cytometry (Fig. S4, ESI†). At the same ICG concentration, the intracellular fluorescence intensity of ICG gradually increases to the maximum after 6-h incubation. Therefore, we decided to perform the laser irradiation on FAT 7 cells after treating them with ICG for 6 h. Next, the apoptosis rates induced by different laser power densities were assessed using flow cytometry (Fig. S5, ESI†). We found that a laser power of 1.5 W cm⁻² induced an apoptosis rate close to 50%. To prevent excessive necrosis, we selected this power density for inducing apoptosis. Under this condition, the expression of CRT (calreticulin, an essential immunogenic cell death (ICD) marker) on the surface of cell membranes was observed by confocal laser scanning microscopy (CLSM), showing the strongest green fluorescence (Fig. S6, ESI†). Based on these experimental results, FAT 7 cells were incubated in T75 culture flasks, treated with free ICG for 6 h, and irradiated using an 808 nm laser (1.5 W cm⁻²) for 5 min. Then, the cells were incubated for an additional 12 h. Next, the apoptotic cells were collected and the aCM was extracted for coating onto the GIT@TM NPs. Lastly, CpG-cholesterol (labeled with cyanine 3 (Cy3) dye) was inserted onto the surface of GIT@TM to obtain the final GIT@TM/C-aCM product.

The generated GIT@TM/aCM and GIT@TM/C-aCM were first characterized by UV-vis spectroscopy (Fig. 1(b)). Both particles show a characteristic peak at 800 nm that can be attributed to the ICG peak and a characteristic peak at 550 nm associated with the Cy3 dye, verifying the successful loading of ICG and CpG-cholesterol in the GIT@TM/aCM NPs. Next, the size and morphology of the GIT@TM/C-aCM NPs were examined by transmission electron microscopy (TEM; Fig. 1(c) and (d)). The GIT@TM/C-aCM NPs show a round shape with a mean size of 72.4 ± 0.19 nm and a relatively uniform size distribution. The cell membrane thickness was measured to be 10.3 nm. The hydrodynamic sizes and zeta potentials of different NPs are given in Table S1 (ESI†). After wrapping TA–Mn MPNs onto the surface of dendrimer conjugates, the surface potential of GIT NPs shifts from positive to negative (−15.4 mV), and the hydrodynamic size decreases from 110.6 nm to 78.1 nm, suggesting the tightened wrapping of MPNs onto the dendrimer particle surface. Furthermore, after aCM camouflage, the hydrodynamic size of GIT@TM/C-aCM slightly increases to 93.3 nm. Next, the stability of GIT@TM/C-aCM NPs was tested by measuring changes in their hydrodynamic size over a week after they were exposed to water, phosphate buffered saline (PBS), or culture medium (Ham's F12K containing 10% fetal bovine serum, FBS). As shown in Fig. S7 (ESI†), the hydrodynamic diameter of the GIT@TM/C-aCM NPs has no distinct changes and the dispersion of GIT@TM/C-aCM is stable for at least one week, demonstrating their good colloidal stability.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was conducted to verify the aCM protein retention on the surface of the GIT@TM/C-aCM NPs (Fig. 1(e)). The GIT@TM/C-aCM NPs show similar protein bands to the aCM, verifying the aCM coating onto the particles. To check the ability of GIT@TM/C-aCM NPs to generate ∙OH through an Mn²⁺-mediated Fenton-like reaction, a methylene blue (MB) degradation assay was performed (Fig. 1(f)). There is no significant change in the absorbance of MB when incubated with only H₂O₂. In contrast, the absorbance of MB decreases significantly after 3 h of treatment with GIT@TM/C-aCM and H₂O₂. This confirms the Mn²⁺-mediated Fenton-like reaction that can be induced by the GIT@TM/C-aCM NPs to produce a large amount of reactive oxygen species (ROS). In an acidic environment, the release rate of Mn²⁺ is much higher (Fig. S8, ESI†), indicating its great potential for CDT of tumors in the TME.

As shown in Fig. 1(g), at acidic pH (pH 6.5), 34.9% of Toy can be rapidly released within 48 h, while only 14.6% of Toy can be released under pH 7.4 at the same time point. Furthermore, GIT@TM/C-aCM can release 42.3% and 57.9% Toy in the presence of H₂O₂ under pH 7.4 and pH 6.5, respectively, within 48 h, which should be due to dual pH-/H₂O₂-responsive cleavage of the boronic ester bonds used to link Toy with dendrimers. These results reveal that under the TME (acidic pH and high levels of H₂O₂), the prepared GIT@TM/C-aCM NPs can responsively release Toy for efficient chemotherapy and Mn²⁺ for CDT of tumors.

The photothermal properties of GIT@TM/C-aCM particles were next assessed using an 808 nm laser. First, we irradiated the GIT@TM/C-aCM solution with different laser power densities. As shown in Fig. S9 (ESI†), at 1 W cm⁻², the solution shows a gradual temperature increase with increasing irradiation time, and at 10 min, the temperature increases to 48 °C. Hence, this power density was selected for subsequent experiments. Next, we monitored the temperature changes of GIT@TM/C-aCM aqueous solution with varying concentrations under laser irradiation for 10 min (Fig. 1(h)). The temperature increases progressively with the increase of the GIT@TM/C-aCM concentration and irradiation duration. Notably, at an ICG concentration of 20 μg mL⁻¹, the temperature leaps from 26.4 °C to 51.3 °C and then stabilizes after reaching the peak temperature. In contrast, the negative control of water shows no significant temperature change under the same conditions. As depicted in Fig. 1(i), after five heating–cooling cycles, the GIT@TM/C-aCM NPs do not seem to have noticeable changes in the peak temperature, demonstrating its excellent photothermal stability. Based on the time-constant curve, the photothermal conversion efficiency (η) was calculated to be 36.7% (Fig. S10, ESI†).

In vitro cytotoxicity and cellular uptake assays

The cell counting kit-8 (CCK-8) assay was utilized to evaluate the in vitro anticancer activity of GIT@TM/C-aCM (Fig. 2(a)). The cytotoxicity of the particles gradually increases after coating with MPNs and then camouflage with the aCM, which should be due to the Mn²⁺-mediated CDT effect generating ROS after TM coating as well as the aCM-rendered homologous cancer cell targeting. The cell viability was significantly inhibited after laser irradiation, indicating a favorable additional PTT effect of GIT@TM/C-aCM induced by the integrated ICG (p < 0.001). Furthermore, to examine the targeting role played by the aCM for the GIT@TM/aCM NPs, the therapeutic activity of GIT@TM/NM camouflaged with normal L929 cell membranes was also tested (Fig. 2(b)). Clearly, the GIT@TM/aCM NPs exhibit significantly stronger cytotoxicity than GIT@TM/NM NPs (p < 0.05 at a Toy concentration of 10 μg mL⁻¹ or above). This verifies the role played by the aCM to endow the NPs with homologous cancer cell targeting specificity.

Fig. 2 (a) Viability of FAT 7 cells treated with GIT, GIT@TM, GIT@TM/C-aCM, or GIT@TM/C-aCM + laser (for short, L) at different Toy/ICG concentrations for 24 h. (b) Viability of FAT 7 cells treated with GIT@TM/NM or GIT@TM/aCM at different Toy concentrations for 24 h. (c) Flow cytometry histograms and (d) mean fluorescence intensity of FAT 7 cells treated with PBS and GIT@TM/C-aCM at different concentrations for 6 h. (e) Intracellular ROS levels in FAT 7 cells after different treatments (PBS, GI, Toy, GIT, GIT@TM, GIT@TM/C-aCM, or GIT@TM/C-aCM + L) for 4 h, as observed by flow cytometry analysis. (f) Quantification of ROS levels corresponding to panel (e). (g) CLSM images of FAT 7 cells stained with DCFH-DA after the above treatments. (h) Relative intracellular GSH levels in FAT 7 cells after the above treatments for 4 h. In (g), the scale bar for each panel represents 20 μm. For panels (a), (b), (d), (f) and (h), n = 3 for each sample or measurement, NS stands for no significance, *p < 0.05, **p < 0.01, and ***p < 0.001.

The cellular uptake of GIT@TM/C-aCM in FAT 7 cells was quantitatively assessed via flow cytometry to measure the fluorescence intensity of ICG (Fig. 2(c) and (d)). The fluorescence intensity of FAT 7 cells gradually increases with the increase of GIT@TM/C-aCM concentration, confirming the concentration-dependent cellular uptake of the NPs. To further prove this, the intracellular distribution of GIT@TM/C-aCM was visualized by CLSM (Fig. S11, ESI†). The red fluorescence signals of cells associated with ICG are concentration-dependent, reaching the highest intensity at the highest GIT@TM/C-aCM concentration of 600 μg mL⁻¹ investigated, in accordance with the flow cytometry results. Furthermore, the aCM coating renders GIT@TM/C-aCM with much stronger cellular internalization than the counterpart material GIT@TM without CM coating at the same ICG concentration (Fig. S12, ESI†). In addition, we also observed the cellular uptake of different nanomedicines after co-incubation with FAT-7 cells via CLSM imaging (Fig. S12, ESI†). Apparently, the GIT@TM/C-aCM group exhibits a much stronger ICG-associated red fluorescence intensity than the GIT@TM/NM group, indicating that the aCM coating confers the ability for homologous targeting of nanomedicines.

GIT@TM/C-aCM-induced CDT of cancer cells

The main indicators of CDT are intracellular ROS generation and changes in the expression of lipid peroxidation (LPO) in cells. Considering the Mn component of GIT@TM/C-aCM that can catalyze a Fenton-like reaction to generate ROS, 2′,7′-dichlorofluorescein diacetate (DCFH-DA), which can be oxidized by ROS, was used as a fluorescent indicator to detect ROS production in FAT 7 cells via CLSM observation (Fig. 2(g)). Cells treated with GIT@TM show obvious green fluorescence signals associated with the intracellular ROS generation, and the ROS level in the GIT@TM/C-aCM group is much higher than that in the GIT@TM group, strengthening the role played by aCM coating in cancer cell targeting. After laser irradiation, the ROS level was further elevated in the GIT@TM/C-aCM group, probably due to the production of singlet oxygen (¹O₂) by the integrated ICG dye under 808 nm laser irradiation (Fig. S13, ESI†). Quantitative flow cytometry assays (Fig. 2(e) and (f)) reveal that GIT@TM/C-aCM enables significant ROS generation in FAT 7 cells, and the addition of laser irradiation further increases the intracellular ROS level. These results are consistent with the CLSM images. Furthermore, excessive levels of ROS disrupt the normal redox balance of the GSH and glutathione peroxidase (GPX) systems, leading to the down-regulation of GSH levels. As shown in Fig. 2(h), the cellular GSH levels are down-regulated after different treatments, with the most obvious depletion observed in the GIT@TM/C-aCM + L group among all groups (p < 0.01).

To examine the excessive intracellular ROS-caused LPO in cancer cells, C11-BODIPY581/591 with oxidation-sensitive fluorophores was used as a probe for CLSM imaging of the LPO level in FAT 7 cells (Fig. 3(a)). Compared to the PBS control group, the GIT@TM, GIT@TM/C-aCM, and GIT@TM/C-aCM + L treatments all induce decreased red fluorescence signals (non-oxidation level) along with increased green fluorescence signals (oxidation level) in FAT 7 cells, indicating enhanced LPO expression. In particular, the GIT@TM/C-aCM + L treatment leads to the brightest green fluorescence and the darkest red fluorescence, suggesting that combined PTT and CDT can augment the intracellular oxidative stress for enhanced LPO production. Overall, the developed GIT@TM/C-aCM NPs can aggravate the oxidative stress through GSH depletion and ROS generation and this process can be exacerbated when combined with PTT.

Fig. 3 (a) CLSM imaging of LPO expression levels in FAT 7 cells after different treatments (PBS, GI, Toy, GIT, GIT@TM, GIT@TM/C-aCM, or GIT@TM/C-aCM + L) for 4 h. (b) Flow cytometry analysis of mitochondrial membrane potential in FAT 7 cells after incubation with PBS, GI, Toy, GIT, GIT@TM, GIT@TM/C-aCM, or GIT@TM/C-aCM + L for 12 h (n = 3). (c) Red/green fluorescence intensity ratio of the JC-1 indicator in FAT 7 cells after the above treatments. Legend: 1, PBS; 2, GI; 3, Toy; 4, GIT; 5, GIT@TM; 6, GIT@TM/C-aCM; and 7, GIT@TM/C-aCM + L (808 nm, 1.0 W cm⁻² for 5 min). In (a), the scale bar for each panel represents 20 μm. In (c), *p < 0.05, **p < 0.01, and ***p < 0.001.

Mitochondrial dysfunction

The excessive accumulation of ROS leads to oxidative stress and mitochondrial dysfunction with decreased mitochondrial membrane potential (MMP). We next used the JC-1 probe to detect changes in MMP after different treatments by flow cytometry since the fluorescence of the JC-1 probe gradually changes from red to green as the MMP decreases (Fig. 3(b) and (c)). After GIT treatment, the red/green fluorescence intensity ratio decreases, indicating decreased MMP and increased mitochondrial dysfunction. Compared to the GIT group, the GIT@TM treatment leads to a more significant decrease of MMP (p < 0.001), which should be ascribed to excessive ROS catalyzed by Mn²⁺. Obviously, GIT@TM/C-aCM + L induces the lowest MMP owing to the synergistic effects of all components, Toy, Mn²⁺ and PTT, in combined chemotherapy/CDT/PDT, thus leading to the most significant mitochondrial dysfunction among all groups (p < 0.001).

Amplification of ERS in vitro

Toy has the potential to aggravate ERS by blocking the splicing of the XBP1u mRNA, thereby inhibiting the adaptive regulation of ERS. GSH depletion and ROS accumulation could induce cellular oxidative stress, thereby amplifying ERS. To investigate the ERS amplification effect, the mRNA expression levels of representative ERS-related factors in cancer cells were analyzed by real-time quantitative polymerase chain reaction (RT-PCR) assays. As an ERS marker, the cellular mRNA level of GRP78 shows a significant increase after the cells were treated with free Toy, verifying that Toy exacerbates ERS (Fig. 4(a)). The GRP78 mRNA level in the GIT@TM/C-aCM + L group is the highest among all groups (p < 0.01), likely attributed to the synergistic effect of combined Toy-mediated chemotherapy and CDT/PTT-induced ROS generation. Next, the expression levels of XBP1u and XBP1s mRNA were analyzed to verify the mechanism by which Toy regulates ERS. As expected, the expression of XBP1s shows a significant decrease after being treated with free Toy and Toy-containing materials (Fig. 4(b)), while the expression of XBP1u increases significantly in the GIT@TM/C-aCM + L group due to the combined therapeutic effects (Fig. S14a, ESI,†p < 0.01) when compared to the other groups. Besides, the expression levels of CHOP, as a marker of ERS-induced cell apoptosis, were also analyzed (Fig. S14b, ESI†). Again, the CHOP expression level is the highest in the GIT@TM/C-aCM + L group (p < 0.001) among all groups, confirming the excellent efficacy of the combination therapy.

Fig. 4 (a) and (b) RT-PCR results of ERS-related factors GRP78 (a) and XBP1s (b) after incubation with PBS, GI, Toy, GIT, GIT@TM, GIT@TM/C-aCM, or GIT@TM/C-aCM + L for 6 h (n = 3). (c) Western blot analysis of the expression of ERS-related proteins in FAT 7 cells after the above treatments for 24 h. (d) Flow cytometry analysis and (e) quantification of the apoptotic and necrotic percentages of FAT 7 cells after the above treatments for 12 h. (f) CLSM images of CRT expression on FAT 7 cells after the above treatments for 12 h. Detection of (g) ATP release from FAT 7 cells after the above different treatments for 24 h. In (f), the scale bar for each panel represents 20 μm. Legend: 1, PBS; 2, GI; 3, Toy; 4, GIT; 5, GIT@TM; 6, GIT@TM/C-aCM; and 7, GIT@TM/C-aCM + L (808 nm, 1.0 W cm⁻² for 5 min). In (a), (b), (e) and (g), n = 3, *p < 0.05, **p < 0.01, and ***p < 0.001.

Western blot analysis of different markers was performed to further verify the expression of related factors at a protein level (Fig. 4(c) and Fig. S15, ESI†). The GIT@TM/C-aCM + L treatment significantly up-regulates the GRP78, XBP1u and CHOP protein expression levels and down-regulates the XBP1s protein expression, in agreement with the RT-PCR results. The expression level of pIRE1α, as a major component of the IRE1α–XBP1 pathway, was also analyzed. Due to the amplification of ERS, pIRE1α shows increased expression. Overall, oxidative stress- and Toy-amplified ERS are the main mechanisms promoting cell apoptosis.

Cancer cell apoptosis in vitro

The cell apoptosis caused by combined PTT/CDT/chemotherapy was evaluated by flow cytometry (Fig. 4(d)). The GI, free Toy or GIT treatment leads to different degrees of cell apoptosis, while the percentage of apoptotic/necrotic cells in the GIT@TM group significantly increases compared to the GIT group, reflecting that oxidative stress induces cell apoptosis via the above-mentioned mechanisms. It should be noted that the treatment with GIT@TM/C-aCM NPs induces a higher cell apoptosis rate than GIT@TM NPs due to the further CM coating-mediated targeting and enhanced cellular uptake of the particles (p < 0.05). After laser irradiation, the apoptosis/necrotic rate in the GIT@TM/C-aCM group is the highest due to the synergistic effects contributed by PTT. Furthermore, we examined the efficacy on FAT 7 cells by live/dead staining assays (Fig. S16, ESI†). Consistent with the apoptosis assay results, the GIT@TM/C-aCM + L treatment leads to the most significant red fluorescence signals, causing the highest level of cell death among all groups.

ICD effect, STING pathway activation and maturation of DCs

To investigate the ICD effect caused by the developed GIT@TM/C-aCM NPs, we next examined the changes in representative damage-associated molecular patterns (DAMPs) following treatment of cells with different materials, including CRT exposure, Adenosine triphosphate (ATP) secretion and High Mobility Group Protein B1 (HMGB-1) release. As illustrated in Fig. 4(f), the expression of CRT on the surface of cancer cells can be clearly seen, as indicated by the appearance of strong green fluorescence signals. Apparently, the GIT@TM/C-aCM + L group shows the highest CRT exposure with the strongest green fluorescence signal on the cell surface, which should be attributed to the combination of chemotherapy/CDT/PTT, as well as the aCM-rendered targeting specificity and enhanced cellular uptake of the particles. The GIT@TM/C-aCM + L treatment also induces the release of highest amounts of ATP (Fig. 4(g)) and HMGB-1 (Fig. S17, ESI†) in cancer cells among all groups (p < 0.001), suggesting an active role of the GIT@TM/C-aCM + L treatment in mediating enhanced cancer cell ICD in vitro. Besides, in comparison with the PBS control group, free Toy and GIT can partially stimulate ATP release due to the respective chemotherapy and chemotherapy/CDT effects that can kill cancer cells to trigger ICD. Taken together, the integrated components Toy, Mn²⁺, and ICG within the nanovaccine formulation enabled effective combination therapy to trigger the most efficient ICD effect on cancer cells.

For effective maturation of DCs to promote antitumor responses, besides the DAMPs produced by ICD of cancer cells, the STING pathway activation in cancer cells has also played an important role.^35,36 As is known, free Mn²⁺ can directly activate and sensitize cGAS to tumor-derived cytoplasmic dsDNA, thus activating the STING pathway of cells.^37,38 Next, the STING pathway activation in FAT 7 cells was analyzed by western blot assays (Fig. 5(a)). The phosphorylated STING (p-STING), downstream phosphorylated TANK-binding kinase 1 (p-TBK1) and interferon regulatory factor 3 (p-IRF3) are all stimulated after treatment with GIT@TM and GIT@TM/C-aCM when compared to the PBS treatment due to the incorporated Mn²⁺ component. In addition, the GIT@TM/C-aCM + L group shows the most significant expression of the p-STING, p-TBK1 and p-IRF3 proteins (Fig. 5(b), p < 0.001) among all groups. This could be due to the PTT-induced DNA damage contributing to additional activation of the STING pathway.

Fig. 5 (a) Western blot analysis of the expression levels of STING, p-STING, TBK1, p-TBK1, IRF-3 and p-IRF3 in FAT-7 cells after different treatments and (b) the quantitative analysis of relative protein expression levels (n = 3, and *** represents p < 0.001). β-Actin was used as a reference.

DCs, as representative antigen-presenting cells, can be matured by DAMPs released during the process of ICD, the activation of the STING pathway of cancer cells,^39,40 as well as the tumor-associated antigen and immune adjuvant.^35,36 To analyze the maturation of DCs through all the above factors, a transwell system was set up to coculture immature DCs with FAT 7 cancer cells after different treatments. Flow cytometric analysis of CD80/CD86 expression on DCs was used to assess maturation of DCs (Fig. 6(a)). The GI, free Toy, and GIT treatments lead to slight maturation of DCs at rates of 1.9%, 2.6%, and 2.5%, respectively, primarily due to the ICD induced by the Toy-mediated chemotherapy effect in the last two groups. In contrast, due to the additional Mn-mediated STING activation and STING activation/antigen (aCM)-immune adjuvant (CpG) delivery mediated, respectively, by the GIT@TM and GIT@TM/C-aCM treatments that cause enhanced ICD through combined chemotherapy/CDT, the DC maturation rates can be increased to 14.4% and 30.6%, respectively. Importantly, due to additional enhanced ICD and improved STING activation through the added PTT effect, the GIT@TM/C-aCM + L treatment leads to the most significant maturation rate of DCs (up to 37.4%) among all groups (p < 0.001). IFN-β, as an important immune activator, was next analyzed through the enzyme linked immunosorbent assay (ELISA) of the DC culture medium after different treatments (Fig. 6(b)). Clearly, the GIT@TM/C-aCM + L group triggers the generation of the highest IFN-β level among all groups (p < 0.001), further suggesting the best anticancer immune response induced by combination therapy-mediated ICD of cancer cells, STING pathway activation, and the antigen (aCM)/immune adjuvant (CpG) delivery.

Fig. 6 (a) Flow cytometry analysis of CD80/CD86 expression on DCs stimulated by FAT 7 cells with different treatments for 24 h. (b) Quantification of CD80⁺CD86⁺ DCs to estimate the DC maturation rates. (c) IFN-β release from DCs after different treatments. (d) Flow cytometry analysis of CD206 and CD86 expressions on RAW264.7 cells after different treatments. (e) Quantification of the M1/M2 ratio of RAW264.7 cells after different treatments. Legend: 1, PBS; 2, GI; 3, Toy; 4, GIT; 5, GIT@TM; 6, GIT@TM/C-aCM; and 7, GIT@TM/C-aCM + L (1 W cm⁻², 5 min). In panels (b), (c) and (e), **p < 0.01 and ***p < 0.001 (n = 3).

TAMs polarization assay

TAMs, an important immune cell type, can be polarized to the M1 phenotype once stimulated (e.g., by lipopolysaccharides, LPS) to activate antitumor immune responses by secreting pro-inflammatory cytokines and presenting antigens.⁴¹ Since the nanovaccine developed in this work may also act on TAMs in the TME, we next checked whether the GIT@TM/C-aCM NPs + L treatment can polarize the TAMs to the M1 phenotype to reverse the immune suppressive TME with abundant M2 TAMs. The CD86/CD206 expression on RAW264.7 macrophages after different treatments was assessed by flow cytometry and the secreted pro-inflammatory cytokines were measured by ELISA (Fig. S18, ESI†). Macrophages treated with PBS and LPS were used as negative and positive controls, respectively. The M1/M2 ratio follows the order of PBS < GI < Toy < GIT < GIT@TM < GIT@TM/C-aCM < GIT@TM/C-aCM + L < LPS. Typically, compared to the GIT group, the M1/M2 ratio of the GIT@TM group significantly increases (p < 0.001) due to the Mn incorporation and the ratio further increases after the encapsulation of the CpG-engineered aCM due to the antigen/adjuvant delivery. After laser irradiation, the M1/M2 ratio further increases to a maximum among all groups, except the LPS positive control (p < 0.001), indicating that the PTT effect can induce further polarization of TAMs towards the M1 phenotype, demonstrating the best potential for immunotherapy.

The pro-inflammatory cytokine levels of TNF-α and IL-6 secreted by M1-like macrophages in the GIT@TM/C-aCM + L group are much higher than those in all other groups (p < 0.001, Fig. S18, ESI†), in agreement with the above flow cytometry results. Overall, the developed GIT@TM/C-aCM can effectively activate anticancer immune responses and promote inflammation, which can be further enhanced under laser irradiation.

Penetration and growth inhibition of FAT 7 tumor spheroids

To prove the combined therapeutic effect of the GIT@TM/C-aCM NPs, 3D FAT 7 multicellular tumor spheroids (MCTSs) were constructed. The FAT 7 MCTSs were incubated with GIT@TM, GIT@TM/C-aCM or free ICG at an ICG concentration of 10 μg mL⁻¹ for 6 h before CLSM imaging (Fig. 7(a)) according to the literature.^42,43 It can be seen that, compared to free ICG, GIT@TM and GIT@TM/C-aCM show a much higher red fluorescence intensity (Fig. 7(b)), suggesting their greater ability to penetrate the MCTSs than free ICG. Furthermore, the best penetration effect was observed in the GIT@TM/C-aCM group, indicating that the homologous aCM enables deep penetration of the NPs inside the MCTSs, which is consistent with the cellular uptake results.

Fig. 7 (a) CLSM images showing the in vitro penetration of free ICG, GIT@TM and GIT@TM/C-aCM in FAT 7 MCTSs (ICG concentration = 10 μg mL⁻¹). The surface of the spheroids was defined as 0 μm (scale bar: 50 μm). (b) Fluorescence intensity profile of FAT 7 MCTSs treated with free ICG, GIT@TM and GIT@TM/C-aCM. (c) Representative optical micrographs of FAT 7 MCTSs treated with NS, GIT@TM or GIT@TM/C-aCM with or without laser irradiation for 5 min at different time points. The scale bar in each panel represents 50 μm. (d) 3D MCTS diameter changes in each group at various time points (n = 3, and *p < 0.05, **p < 0.01, and ***p < 0.001).

To explore the combined therapeutic activity of GIT@TM/C-aCM, the MCTSs were incubated with different materials for different time periods, and phase contrast microscopy was used to capture the representative images (Fig. 7(c)). In the normal saline control group, there was no significant change in MCTS diameter over time, and the black shade became slightly deeper with incubation time, indicating cell proliferation during this period (Fig. 7(c)). By contrast, it can be found that, after laser irradiation, the diameters of MCTSs in different groups significantly decrease, indicating that PTT has a good proliferation inhibitory effect on MCTSs. Under laser irradiation, the diameter of MCTSs in the GIT@TM/C-aCM group decreases from 162 to 53 μm, while the diameter of MCTSs in the GIT@TM group decreases from 160 to 79 μm. Obviously, the GIT@TM/C-aCM + L group shows a more pronounced inhibitory effect on MCTSs than the GIT@TM/C-aCM group due to the additional PTT effect (Fig. 7(d), p < 0.001) and the GIT@TM + L group due to aCM-rendered enhanced cellular uptake and penetration of MCTSs. These findings imply that the developed GIT@TM/C-aCM NPs possess deep penetration ability into tumor spheroids and could induce excellent combination therapeutic effects within tumor cells under laser irradiation.

Conclusions

We developed a dendrimer-based GIT@TM/C-aCM nanomedicine incorporated with the chemical drug Toy, Mn²⁺, ICG, and aCM/CpG for combination therapeutic cancer vaccine applications. We show that the ICG- and Toy-conjugated dendrimers can be coated with TM MPNs and camouflaged with aCM, which is subsequently engineered by CpG. The formed GIT@TM/C-aCM NPs with a size of 72.4 nm possess good colloidal stability, can efficiently release the Toy drug as well as Mn²⁺ in the TME, and display good photothermal conversion efficiency (36.7%). Under laser irradiation, the NPs are able to exert combination chemotherapy effects due to the Toy-mediated ERS amplification, CDT due to the Mn²⁺-mediated Fenton-like reaction, and ICG-mediated PTT of cancer cells. These combined therapies lead to ICD of cancer cells, along with the Mn²⁺-mediated STING pathway activation of cancer cells and aCM antigen/CpG immune adjuvant delivery to DCs to achieve significant DC maturation, M1 polarization of TAMs and secretion of pro-inflammatory factors, thus leading to enhanced anticancer immune responses. Furthermore, the developed GIT@TM/C-aCM NPs can efficiently penetrate and inhibit 3D MCTSs in vitro. Moreover, owing to the fluorescence imaging ability of ICG, coupled with the active targeting ability of homologous cell membranes, the designed GIT@TM/C-aCM holds promise to be developed as a versatile theranostic nanovaccine to tackle NPC, which is currently underway in our ongoing efforts.

Data availability

The datasets supporting this article are available as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (U23A2096, 52350710203, W2433053 and W2421104), the Science and Technology Commission of Shanghai Municipality (24490711000, 23520712500 and 20DZ2254900), the National Key R&D Program (2024YFE0108100), and the Research Fund of Shanghai Tongren Hospital, Shanghai Jiao Tong University School of Medicine (2023DHYGJC-ZDA01). This project was also supported by Researchers Supporting Project Number (RSP2025R65), King Saud University, Riyadh, Saudi Arabia.

Notes and references

1. Y.-P. Chen, A. T. C. Chan, Q.-T. Le, P. Blanchard, Y. Sun and J. Ma, Lancet, 2019, 394, 64–80.
2. L. Zhang, Q. Y. Chen, H. Liu, L. Q. Tang and H. Q. Mai, Drug Des., Dev. Ther., 2013, 7, 37–52.
3. H. Huang, Y. Yao, X. Deng, Z. Huang, Y. Chen, Z. Wang, H. Hong, H. Huang and T. Lin, Int. J. Oncol., 2023, 63, 97.
4. J. Y. Kim, W. I. Choi, M. Kim and G. Tae, J. Controlled Release, 2013, 171, 113–121.
5. P. Xu and F. Liang, Int. J. Nanomed., 2020, 15, 9159–9180.
6. Z. Ouyang, Y. Gao, S. Shen, B. Jia, H. Yu, H. Wang, M. Shen and X. Shi, Nano Today, 2023, 51, 101936.
7. L. Yang, B. Huang, S. Hu, Y. An, J. Sheng, Y. Li, Y. Wang and N. Gu, Nano Res., 2022, 15, 4285–4293.
8. Q. F. Song, Z. Y. Ding, C. Y. Li, Y. R. Fan, X. L. Tang and Y. C. Han, Crystals, 2023, 13, 1387.
9. Y. Cheng, L. Zhao, Y. Li and T. Xu, Chem. Soc. Rev., 2011, 40, 2673–2703.
10. D. A. Tomalia, A. M. Naylor and W. A. Goddard, Angew. Chem., Int. Ed. Engl., 1990, 29, 138–175.
11. X. Wang, M. Zhang, Y. Li, H. Cong, B. Yu and Y. Shen, Small, 2023, 19, 2304006.
12. Z. Wang, Y. Guo, Y. Fan, J. Chen, H. Wang, M. Shen and X. Shi, Adv. Mater., 2022, 34, 2107009.
13. C. Song, M. Shen, J. Rodrigues, S. Mignani, J.-P. Majoral and X. Shi, Coord. Chem. Rev., 2020, 421, 213463.
14. H. Kang, W. Wei, L. Sun, R. Yu, E. Yang, X. Wu and H. Dai, Chem. Mater., 2023, 35, 2408–2420.
15. Y. Xu, Y. Guo, C. Zhang, M. Zhan, L. Jia, S. Song, C. Jiang, M. Shen and X. Shi, ACS Nano, 2022, 16, 984–996.
16. Z. M. Tang, Y. Y. Liu, M. Y. He and W. B. Bu, Angew. Chem., Int. Ed., 2019, 58, 946–956.
17. Y. Ding, P. Zhang, H. Xiong, X. Sun, A. Klyushin, B. Zhang, Z. Liu, J. Zhang, H. Zhu, Z.-A. Qiao, S. Heumann and S. Dai, Green Chem., 2020, 22, 6025–6032.
18. L.-S. Lin, J. Song, L. Song, K. Ke, Y. Liu, Z. Zhou, Z. Shen, J. Li, Z. Yang, W. Tang, G. Niu, H.-H. Yang and X. Chen, Angew. Chem., Int. Ed., 2018, 57, 4902–4906.
19. Y. Shi, M.-S. Jang, C. Liu, Y. Fu, J. H. Lee, Y. Li and H. Y. Yang, Eur. Polym. J., 2023, 194, 112130.
20. X. Tan, R. Sheng, Z. Liu, W. Li, R. Yuan, Y. Tao, N. Yang and L. Ge, Small, 2024, 20, 2305325.
21. Y. Wang, F. Gao, X. Li, G. Niu, Y. Yang, H. Li and Y. Jiang, J. Nanobiotechnol., 2022, 20, 69.
22. M. Lv, M. Chen, R. Zhang, W. Zhang, C. Wang, Y. Zhang, X. Wei, Y. Guan, J. Liu, K. Feng, M. Jing, X. Wang, Y.-C. Liu, Q. Mei, W. Han and Z. Jiang, Cell Res., 2020, 30, 966–979.
23. J. Yan, G. Wang, L. Xie, H. Tian, J. Li, B. Li, W. Sang, W. Li, Z. Zhang and Y. Dai, Adv. Mater., 2022, 34, 2105783.
24. Z. Zhao, Z. Ma, B. Wang, Y. Guan, X.-D. Su and Z. Jiang, Cell Rep., 2020, 32, 108053.
25. X. Sun, Y. Zhang, J. Li, K. S. Park, K. Han, X. Zhou, Y. Xu, J. Nam, J. Xu, X. Shi, L. Wei, Y. L. Lei and J. J. Moon, Nat. Nanotechnol., 2021, 16, 1260–1270.
26. Y. Gao, Z. Ouyang, S. Shen, H. Yu, B. Jia, H. Wang, M. Shen and X. Shi, ACS Nano, 2023, 17, 23889–23902.
27. Y. Guo, Y. Fan, Z. Wang, G. Li, M. Zhan, J. Gong, J.-P. Majoral, X. Shi and M. Shen, Adv. Mater., 2022, 34, 2206861.
28. L. Rao, L.-L. Bu, B. Cai, J.-H. Xu, A. Li, W.-F. Zhang, Z.-J. Sun, S.-S. Guo, W. Liu, T.-H. Wang and X.-Z. Zhao, Adv. Mater., 2016, 28, 3460–3466.
29. Y. Liu, U. K. Sukumar, M. Kanada, A. Krishnan, T. F. Massoud and R. Paulmurugan, Adv. Funct. Mater., 2021, 31, 2103600.
30. G. Li, Y. Guo, C. Ni, Z. Wang, M. Zhan, H. Sun, G. Choi, J.-H. Choy, X. Shi and M. Shen, J. Mater. Chem. B, 2024, 12, 7429–7439.
31. Y. Guo, Y. Fan, G. Li, Z. Wang, X. Shi and M. Shen, ACS Appl. Mater. Interfaces, 2021, 13, 55815–55826.
32. H. Maeda, H. Nakamura and J. Fang, Adv. Drug Delivery Rev., 2013, 65, 71–79.
33. I. Altinbasak, Y. Alp, R. Sanyal and A. Sanyal, Nanoscale, 2024, 16, 14033–14056.
34. Q. Jiang, M. Xie, R. Chen, F. Yan, C. Ye, Q. Li, S. Xu, W. Wu, Y. Jia, P. Shen and J. Ruan, Front. Immunol., 2022, 13, 973601.
35. Y. Guo, Z. Wang, G. Li, M. Zhan, T. Xiao, J. Wang, J. C. M. van Hest, X. Shi and M. Shen, Bioact. Mater., 2025, 43, 129–144.
36. T. Huang, Y. Guo, Z. Wang, J. Ma, X. Shi, M. Shen and S. Peng, ACS Appl. Mater. Interfaces, 2024, 16, 27187–27201.
37. C. Wang, Y. Guan, M. Lv, R. Zhang, Z. Guo, X. Wei, X. Du, J. Yang, T. Li, Y. Wan, X. Su, X. Huang and Z. Jiang, Immunity, 2018, 48, 675–687.
38. L. Hou, C. Tian, Y. Yan, L. Zhang, H. Zhang and Z. Zhang, ACS Nano, 2020, 14, 3927–3940.
39. R. M. Steinman, K. Inaba, S. Turley, P. Pierre and I. Mellman, Hum. Immunol., 1999, 60, 562–567.
40. L. Deng, H. Liang, M. Xu, X. Yang, B. Burnette, A. Arina, X.-D. Li, H. Mauceri, M. Beckett, T. Darga, X. Huang, T. F. Gajewski, Z. J. Chen, Y.-X. Fu and R. R. Weichselbaum, Immunity, 2014, 41, 843–852.
41. J. Gao, Y. Liang and L. Wang, Front. Immunol., 2022, 13, 888713.
42. P. Xue, M. Hou, L. Sun, Q. Li, L. Zhang, Z. Xu and Y. Kang, Acta Biomater., 2018, 81, 242–255.
43. S. S. Kim, H. K. Kim, H. Kim, W. T. Lee, E. S. Lee, K. T. Oh, H.-G. Choi and Y. S. Youn, Arch. Pharmacal Res., 2021, 44, 182–193.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and additional results. See DOI: https://doi.org/10.1039/d5tb00226e

---