Engineering CpG‐ASO‐Pt‐Loaded Macrophages (CAP@M) for Synergistic Chemo‐/Gene‐/Immuno‐Therapy

Abstract

Adoptive cell therapy by natural cells for drug delivery has achieved encouraging progress in cancer treatment over small‐molecule drugs. Macrophages have a great potential in antitumor drug delivery due to their innate capability of sensing chemotactic cues and homing toward tumors. However, major challenge in current macrophage‐based cell therapy is loading macrophages with adequate amounts of therapeutic, while allowing them to play a role in immunity without compromising cell functions. Herein, a potent strategy to construct a macrophage‐mediated drug delivery platform loaded with a nanosphere (CpG‐ASO‐Pt) (CAP) composed of functional nucleic acid therapeutic (CpG‐ASO) and chemotherapeutic drug cisplatin (Pt) is demonstrated. These CAP nanosphere loaded macrophages (CAP@M) are employed not only as carriers to deliver this nanosphere toward the tumor sites, but also simultaneously to guide the differentiation and maintain immunostimulatory effects. Both in vitro and in vivo experiments indicate that CAP@M is a promising nanomedicine by macrophage‐mediated nanospheres delivery and synergistically immunostimulatory activities. Taken together, this study provides a new strategy to construct a macrophage‐based drug delivery system for synergistic chemo‐/gene‐/immuno‐therapy.

A multifunctional self‐assembled DNA nanosphere through adoptive macrophages delivery is developed for cancer treatment. The DNA nanosphere loaded macrophages not only are employed as carriers to deliver nanosphere toward tumor sites, but also simultaneously play a role in immune regulation. This study offers a promising strategy to construct a macrophage‐based drug delivery nanoplatform for synergistic therapy.

1. Introduction

Adoptive cell therapy is becoming a novel clinical approach for antitumor treatment via directly employing therapeutic immune cells, including T cells, NK cells and macrophages, to eradicate cancer cells.^{[ 1} , ² , ^{3 ]} The first cell therapeutics approved by FDA for clinical use is chimeric antigen receptor (CAR)‐T therapies, which utilize the engineered T cells to express CARs.^{[ 4 ]} As an alternative to conventional anti‐cancer treatments, (CAR)‐T cell therapies have demonstrated great achievements in clinical treatment of leukemia and lymphoma, which have caused an exponential increase in the research of cell therapy.^{[ 5 ]} Nevertheless, the use of (CAR)‐T technology for solid tumor therapy remains challenging, due to the difficulty of targeting and infiltrating tumor tissues.^{[ 6 ]} As natural antigen presenting cells, macrophages can stimulate immunogenicity as well as have the specific binding ability with tumor tissues, which are expected to be a prospective type in cell therapy to treat solid tumors.^{[ 7 ]} To a large extent, the phenotypic plasticity of a macrophage determines their immunostimulatory ability, which plays a significant role in regulating tumor progression and an immune microenvironment.^{[ 8 ]} Specifically, a macrophage can differentiate into an proinflammatory M1 phenotype or an anti‐inflammatory M2 phenotype in response to chemokine cues.^{[ 9 ]} M1‐phenotype macrophages produce immunogenic cytokines and inhibit tumor growth, while M2‐phenotype macrophages secrete immunosuppressive cytokines and promote tumor growth.^{[ 10 ]} Recently, inhibition of M2 phenotypes and polarization of M1 phenotypes at tumor tissues have been performed in the treatment of solid tumors.^{[ 11 ]} However, cells may alter their functions in response to immunosuppressive tumor microenvironments, which tremendously and negatively influenced the treatment efficacy.^{[ 12 ]} Hence, it is essential to control the phenotype of macrophages in macrophage‐based cell therapy.

Due to their intrinsic phagocytotic capability, macrophages are able to engulf particles, which allows a living macrophage to be loaded with therapeutic agents by incubating it with therapeutic agents in vitro.^{[ 13 ]} Additionally, macrophages can sense chemokine cues and home toward tumors with high efficiency, indicating that the employment of macrophages in chemo‐drug delivery could increase the drug accumulation in tumors. Based on these characteristics, macrophages can therefore act as mediators to directly phagocytose drugs, and subsequently transfer the drugs to tumor sites.^{[ 14 ]} It has been reported that macrophages can uptake Pt(IV) therapeutic nanoparticles, and then gradually release them at tumor tissues, which demonstrated that macrophages possess the ability of transferring the nanoparticles via cell‐to‐cell transmission.^{[ 15 ]} So far, a limited number of macrophage‐based cell therapeutic studies have been developed to deliver synthetic nanoparticles, such as liposomes, silica complex and inorganic particles.^{[ 16} , ¹⁷ , ^{18 ]} However, due to lower bioavailability, high concentrations of exogenously synthetic nanoparticles would induce immediate death and dysfunction of macrophages. Whereas, sublethal amounts of synthetic nanoparticles could lead to insufficient loading dosage and treatment failure. Collectively, it is urgent to develop a synergistic strategy by using endogenous materials to conquer the aforementioned hurdles for macrophage antitumor therapies.^{[ 19 ]}

Herein, we demonstrate a potent and facile strategy using novel CpG‐ASO‐Pt (CAP) nanospheres loaded macrophages(CAP@M) for antitumor therapy. As shown in Scheme  1 , this CAP nanosphere is composed of chemotherapeutic (cisplatin), endogenous nucleic acid therapeutic (unmethylated cytosine‐phosphate‐guanine [CpG] motif and anti‐P‐gp [P‐glycoprotein] antisense oligonucleotide [ASO]). Anti‐P‐gp ASO is able to down‐regulate the expressions of P‐glycoprotein, which is the main factor responsible for pumping chemo‐drugs out and consequently impairing the effectiveness of anti‐cancer drugs.^{[ 20 ]} CpG motif can induce immunostimulatory effects through M1 polarization, causing secretion of the proinflammatory cytokines.^{[ 21 ]} We propose a simple approach to assemble this CAP nanosphere, using the coordination‐driven interaction between cisplatin and nucleic acids.^{[ 22 ]} The obtained CAP nanospheres are incubated and loaded into macrophage‐like RAW 264.7 cells, in which the CpG motif could induce the immunostimulatory effects and guide the M1‐phenotype differentiation, followed by the adoptively transferred delivery of CAP nanospheres to tumor tissues. Then, CAP nanospheres simultaneously released cisplatin molecules and ASOs, dramatically intensifying the cytotoxicity toward the tumor cells through the cisplatin‐based irreversible damage and the P‐gp protein inhibition. Overall, this CAP nanosphere has the characteristics of favorable biocompatibility, excellent cellular internalization, good combinatorial effect, and high drug‐loading efficiency in macrophages. Considerably, our designed macrophage‐mediated delivery strategy facilitates in vivo antitumor activity for NSCLCs.

Scheme 1.

Schematic illustration of the macrophages loaded with CAP nanospheres for non‐small‐cell lung cancer (NSCLC) treatment. A) The design and synthetic route of the self‐assembled CAP nanospheres loaded macrophages (CAP@M). B) CAP@M as the antitumor drug delivery strategy to combat NSCLC.

2. Results and Discussion

2.1. Synthesis and Characterization of CAP Nanospheres

To synthesize this new type of DNA‐based Pt nanoparticle, a typical approach of cisplatin driven coordination interactions self‐assembly was applied, which can be utilized to prepare DNA‐metal hybrids in nanoscale under the mild conditions. Based on our previous report, we first attempted to prepare DNA‐Pt nanoparticles using 40‐mer DNA strands with random sequence.^{[ 23 ]} In details, a random DNA solution was incubated with an aqueous cisplatin solution at 95 °C for 2 h. The mixture solution was purified by centrifugation and the product of DNA‐Pt hybrids was dissolved in distilled water, which exhibited the gel‐like characteristics under room temperature (Figure S1, Supporting Information). Transmission electron microscopy (TEM) imaging was used to inspect the spherically shaped DNA‐Pt nanoparticles, with the average diameter ≈650 nm (Figures S1 and S3, Supporting Information). Further analysis evinced that the random DNA‐Pt nanospheres with favorable morphology and size distribution can be produced in the presence of 30 µm DNA and 1 mm cisplatin (Figures S3–S5, Supporting Information). To illustrate the impact of sequence of DNA on the self‐assembly, various DNA oligonucleotide strands were employed to synthesize this DNA‐Pt nanosphere under the same conditions. As a result, a series of products with similar morphology and size can be obtained, which indicated that this synthetic method of DNA‐based Pt hybrids has a wide scope of DNA substrate sequences (Figure S8, Supporting Information).

Inspired by the successful construction, we next attempted to introduce a nucleic acid therapeutic sequence to substitute the random sequences, which contained the CpG motif and anti‐P‐gp ASO. Similarly, a representative TEM image of the corresponding CpG‐ASO‐Pt product was performed, and the comparable morphology and diameter were shown, which was named as CAP nanosphere (Figure  1B). A scanning electron microscopy (SEM) image confirmed that the morphology of this CAP product is spherical (Figure 1A). Dynamic light scattering (DLS) analysis validated that the average hydrodynamic diameter (D _h ) of CAP nanospheres is ≈630 ± 58 nm, in agreement with the TEM and scanning electron microscopy (SEM) characterizations (Figure 1D). High‐Resolution‐TEM (HRTEM) imaging revealed a lack of diffraction contrast spots and lattice fringes, indicating the amorphous attribute of this CAP nanosphere, which was further verified by the amorphous diffuse halo in the selected area electron diffraction (SAED) image (Figure 1C). High‐angle annular dark‐field STEM spectroscopy (HAADF‐STEM) and elemental mapping determined that platinum and marked elements in DNA (nitrogen, oxygen, and phosphorus) are homogeneously distributed in the CAP nanospheres (Figure 1F,G). In addition, the uniform distribution of each element can be validated by energy dispersive X‐ray spectroscopy (EDS) line scanning (Figure 1H). Quantitative analysis of relative content for each element in the CAP and random DNA‐Pt nanospheres was determined by EDS spectrum, elucidating the atom percentage of platinum element in this CAP nanosphere was ≈10% (Figure 1J; and Figures S6 and S7, Supporting Information). Moreover, due to the component of DNA in nanospheres, zeta potential measurements showed both the CAP and random DNA‐Pt nanospheres are negatively charged with −26.4 and −24.1 mV, which suggests that the coordination interaction between DNA and metal ions would not impact the DNA characteristics (Figure 1E; Figure S2, Supporting Information). The DNA encapsulation efficiency of this CAP was ≈65%, and the UV–vis absorbance peak slightly shifted to ≈265 nm (Figure S9, Supporting Information). The aforementioned results showed the successful synthesis of CAP nanospheres by incorporating cisplatin molecule with CpG‐ASO agent through a simple one‐pot method, which provided a universal methodology to construct the DNA‐based Pt nanospheres.

Figure 1.

Characterization of CAP nanospheres and the Pt release behavior of CAP nanospheres. A) SEM image, B) TEM image, and C) High‐Resolution TEM image with its SAED pattern (inset image) of CAP nanospheres. D) Hydrodynamic diameters of CAP nanospheres and the solution of CAP at room temperature. E) Zeta potential of the CAP nanospheres. F) HAADF‐STEM image, and G) elemental mapping images of CAP with corresponding elements. H) EDS line scanning profiles of the CAP nanospheres (inset image) for Pt, N, O, and P elements along the line. I) In vitro Pt release behavior of CAP nanospheres in PBS (pH = 7.0), PBS (pH = 6.0) and PBS (pH = 4.4) at 37 °C for different time points. J) Quantitative analysis of CAP nanospheres in EDS spectrum.

As a new drug delivery system, the stability of CAP nanospheres in physiological conditions was first examined. After incubation of CAP nanospheres in a solution containing 10% fetal bovine serum (FBS) at 37 °C for 2–24 h, TEM imaging revealed that this CAP nanosphere was relatively stable at least 12 h, which is indispensable for a potential drug delivery system (Figure S10, Supporting Information). Then, we examined the disassembly behavior of CAP nanospheres through dialysis methods.^{[ 24 ]} In this assay, CAP nanospheres were incubated with phosphate buffered saline (PBS, pH = 7.0), PBS (pH = 6.0), and PBS (pH = 4.4), then dialyzed through dialysis bags (MWCO 1000 Da), respectively.^{[ 25 ]} The dialysates were collected at different time points and detected by inductively coupled plasma mass spectrometry (ICP‐MS). As shown in Figure 1I, CAP nanospheres remained relatively stable in PBS (pH = 7.0) buffer solution with only 20% Pt released after 24 h. The higher amount (30%) of released Pt was obtained in PBS (pH = 6.0) solution. In the presence of an acidic environment with PBS (pH = 4.4) buffer, the released amount of Pt was enhanced up to 40% within 20 h, then reached ≈45% after 48 h incubation. These results evinced that the CAP nanospheres have favorable bioavailability, as well as that they are able to be disassembled in acidic environments, and specifically release the therapeutic drugs after being internalized by cancer cells.^{[ 26 ]}

2.2. Cellular Uptake Efficiency and In Vitro Antitumor Effect of CAP Nanospheres

We then investigated the cellular uptake efficiency of CAP nanospheres by flow cytometry and confocal laser scanning microscopy (CLSM). A Cy5 fluorescent dye labeled CpG‐ASO strand was used to synthesize the CAP‐Cy5 nanospheres, followed by the incubation with A549 cells for different lengths of time (4, 8, 12, and 24 h). Flow cytometry analysis exhibited that the CAP‐Cy5 nanospheres could efficiently enter A549 cells, with the fluorescence intensity increased along with incubation time (Figure  2E,F). CLSM images validated that the CAP‐Cy5 nanospheres can be internalized by A549 cells and distributed throughout the cytoplasm, in which strong red fluorescence signals can be detected and gradually increased, consistent with the flow cytometry characterization (Figure 2A). Further, we studied the subcellular localization of CAP‐Cy5 nanospheres in cancer cells. As the incubation continued, CLSM images showed that the red signals (CAP‐Cy5) and the green signals (endo‐ and lysosomes) were not overlapped after incubation until 24 h (Figure 2B). Spatial fluorescence intensity analysis of CLSM images clearly verified that there was no obvious co‐localization between CAP‐Cy5 nanospheres and endosomes/lysosomes since there was no overlapping of fluorescence signals at each position in the cells (Figure 2C). To quantitatively characterize the association between the CAP‐Cy5 nanospheres and endosomes/lysosomes, the Pearson's coefficients (PCs) were calculated. The PCs of nanosphere/endo‐ and lysosomes ranged from 0.37 to 0.28, which is far below the correlation threshold of 0.5 (Figure 2D).^{[ 27 ]} All of these results demonstrated that this CAP nanosphere can be internalized by cancer cells effectively and would not be captured in endo‐ and lysosomes, which satisfies the requirement of gene therapy for CpG‐ASO delivery.

Figure 2.

The cellular uptake of CAP nanospheres and the subcellular distribution of CAP nanospheres in A549 cells. A) CLSM images of A549 cells incubated with Cy5 labeled CAP‐Cy5 nanospheres (pseudo‐colored in red) at different time points (4, 8, 12, and 24 h). Scale bars: 10 µm. B) Subcellular distribution of CAP‐Cy5 nanospheres (pseudo‐colored in red) at different time points (4, 12, and 24 h). The cells were co‐stained with Hoechst (pseudo‐colored in blue) and LysoTracker (pseudo‐colored in green), respectively. Scale bars: 2 µm. C) Spatial fluorescence intensity analysis of A549 cells along the white arrows in the images. D) Pearson's coefficients (PCs) of CAP nanospheres/endo‐ and lysosomes in corresponding CLSM images. E) Flow cytometry analysis of cellular uptake efficiency of CAP‐Cy5 nanospheres by A549 cells at different time points (4, 8, 12, and 24 h). F) Mean fluorescence intensity of the corresponding results in flow cytometry analysis. Data were presented as mean ± SD. Not significant = NS. Statistical significance: ***p < 0.001.

Having confirmed the suitable physiochemical properties and effective cellular uptake efficiency, the cytotoxicity of CAP nanospheres toward A549 cells was evaluated by MTT assay, utilizing free cisplatin, free DNA strands, cisplatin and CpG‐ASO (physical mixture) and random DNA‐Pt nanospheres (40‐mer DNA strands without CpG‐ASO sequence) as the control groups (Figure  3A). As a result, free cisplatin , cisplatin and CpG‐ASO groups exhibited the similarly lower cytotoxicity, which indicated that the free cisplatin could not be internalized into cancer cells effectively, as well as that the naked DNA strands have no obvious cytotoxicity toward cancer cells (Figure S11, Supporting Information). At low dosages, the random DNA‐Pt nanospheres group presented similar cytotoxicity compared to the CAP group. In contrast, CAP nanospheres induced evidently enhanced cytotoxicity relative to other groups at high dosages. The major reason for this is that the nucleic acid therapeutic of ASO promotes the retention of cisplatin molecules by down‐regulation of P‐gp protein at a high concentration, resulting in the enhanced cytotoxicity through a combinatorial effect. According to the MTT assay, IC₅₀ values toward A549 cells are 20.38, 21.78, 16.89, and 7.73 µm for free cisplatin, free cisplatin and CpG‐ASO, random DNA‐Pt, and CAP nanospheres, respectively. As a of broad‐spectrum anti‐cancer molecule, the cytotoxicity of different drug formulations toward MGC‐803, MCF‐7, HeLa and Helf (human embryonic lung fibroblast) cell lines were also studied, in which the CAP nanospheres also showed the remarkable ability to inhibit the proliferation of cancer cells (Figure S12, Supporting Information).

Figure 3.

In vitro antitumor performance for the CAP nanospheres. A) In vitro cytotoxicity of different drug formulations toward A549 cells assessed by MTT assays. B) Flow cytometry‐based analysis of P‐gp protein expression in A549 cells treated with different drug formulations. C) P‐gp protein expression in A549 cells treated with different drug formulations assessed by western blotting assay. D) Calcein–AM (green, live cells) and propidium iodide (red, dead cells) co‐staining experiments in A549 cells treated with different drug formulations. E) Flow cytometry‐based apoptosis analysis of A549 cells induced by different drug formulations. Data were presented as mean ± SD. Statistical significance: *p < 0.1, **p < 0.01.

To further verify the in vitro antitumor performance of the CAP nanospheres toward A549 cells, live/dead cells co‐staining assay was carried out by Calcein–AM/PI (Figure 3D). As expected, strong green fluorescence signals were obtained in the free cisplatin , cisplatin and CpG‐ASO groups, signifying the cells were relatively normal and mainly interacted with Calcein–AM. By contrast, the groups treated with random DNA‐Pt and CAP nanospheres revealed obvious red fluorescence and less green fluorescence signals, which suggests that nanospheres could elicit the greater cytotoxicity. As a P‐gp inhibitor, ASO is able to down‐regulate the expression of P‐gp and restore the sensitivity of cancer cells to cisplatin. To investigate whether our designed CAP nanospheres can suppress the P‐gp protein expression, flow cytometry, and western blotting assay were conducted to evaluate the P‐gp expressions on the protein level. A549 cells were incubated with CAP nanospheres, and free ASO, cisplatin and random DNA‐Pt were used as the control groups. As shown in Figure 3B, naked ASO strands and free cisplatin failed to knock down the P‐gp expressions in A549 cells, which is lower than the knockdown (42%) in the random DNA‐Pt treated group. In comparison, over 59% down‐regulation of P‐gp expression was observed after incubation with CAP nanospheres (Figure S13, Supporting Information). Meanwhile, western blotting assay was also used to confirm this molecular mechanism (Figure 3C). As a consequence, the P‐gp protein expressions were down‐regulated by ≈30% in naked ASO and free cisplatin treated groups, owing to the poor cellular uptake efficiency. Meanwhile, the P‐gp protein knockdown efficiency in CAP nanospheres treated group was determined to be up to 67%, which was higher than the knockdown efficiency (49%) induced by random DNA‐Pt nanospheres, indicating that the P‐gp protein down‐regulation by ASO was sequence‐specific. The results above corroborated that the presence of ASO could enhance the capability of inhibiting the cancer cells’ proliferation by down‐regulating the P‐gp protein expression. Furthermore, the ability of CAP nanospheres to induce cell apoptosis was investigated by flow cytometry analysis (Figure 3E). A549 cells were incubated with a series of drug formulations at an equivalent Pt concentration. Flow cytometry analysis results revealed that the total apoptotic cells rate reached over 81.0% after incubation with CAP nanospheres, which was higher than the rate induced by free cisplatin (47.3%), cisplatin and CpG‐ASO (50.8%) and random DNA‐Pt nanospheres (59.8%), respectively. These results demonstrated that CAP nanospheres indeed facilitate the cell apoptosis, owing to their excellent cellular internalization and a good combinatorial effect of the chemo‐drug cisplatin and the gene therapeutic ASO.

2.3. Synthesis and Characterization of CAP@M

Encouraged by the dramatic antitumor performance of CAP nanospheres in vitro, we next incubated CAP nanospheres with macrophage‐like RAW 264.7 cells (a murine macrophage cell line) and tried to prepare the nanospheres‐laden macrophage for the anti‐cancer strategy in vivo. Before loading experiments, the cellular uptake behavior of CAP nanospheres in RAW 264.7 cells was first evaluated. As revealed by the CLSM imaging in Figure  4A, the intracellular fluorescence signals of CAP‐Cy5 increased along with the incubation time, elucidating the macrophages can effectively engulf CAP‐Cy5 nanospheres, which is attributed to the innate phagocytic capacity of the macrophage.^{[ 7 ]} Then, the subcellular distribution of CAP nanospheres was also studied by CLSM. As depicted in Figure 4B, the CAP‐Cy5 nanospheres were greatly internalized and distributed throughout the cytoplasm after 2 h incubation, evidenced by the fact that the most of the red fluorescence signals of the CAP‐Cy5 nanospheres were not co‐localized with the green fluorescence signals of the endosomes/lysosomes. Spatial fluorescence intensity scanning of CLSM images validated that there were no obvious overlapping fluorescence signals between CAP‐Cy5 nanospheres and endosomes/lysosomes along the selected positions (Figure 4C). Since the CAP nanospheres elicited strong cytotoxicity in A549 cells, the macrophages’ viability after incubation with various concentrations of CAP nanospheres was assessed by MTT assay, to prepare the CAP loaded macrophages under the appropriate concentration (Figure S15, Supporting Information). When the total concentration of CAP was less than 20 µm, the cell viability of macrophages was maintained over 80% after 24 h incubation. Otherwise, the cell viability was less than 60% when the concentration of CAP was higher than 40 µm. To quantitatively evaluate the loading capacity of CAP nanosphere in macrophages, RAW 264.7 cells were incubated with CAP nanospheres at 20 µm Pt concentration for different periods, and the amount of intracellular Pt was measured by ICP‐MS (Figure 4E). We found that 12 h incubation resulted in efficient CAP nanosphere loading capacity with 11.7 ± 1.5 pg cell⁻¹, while extending the incubation time slightly increased the intracellular Pt content. Compared with the control group (incubation with free cisplatin), the intracellular Pt amount in the CAP incubation group was obviously higher, again demonstrating the CAP nanospheres can effectively be uptaken and loaded by macrophage cells. Based on the results above, the 20 µm Pt concentration of CAP nanospheres and 12 h incubation time were considered as the optimized condition to construct the CAP nanospheres loaded macrophages, namely CAP@M.

Figure 4.

Characterizations of CAP@M. A) CLSM images of RAW 264.7 cells incubated with CAP‐Cy5 nanospheres (pseudo‐colored in red) at different time points (2, 4, 8, and 12 h). B) Subcellular distribution of CAP‐Cy5 nanospheres after 12 h incubation. The RAW 264.7 cells were co‐stained with Hoechst (pseudo‐colored in blue) and LysoTracker (pseudo‐colored in green), respectively. C) Spatial fluorescence intensity analysis of RAW 264.7 cells along the white arrows in image. D) Representative gene expressions of M1‐phencotpye (TNF‐α, CD86) and M2‐phencotype (CD206, Arg‐1) in RAW 264.7 cells induced by various drug formulations. E) The amount of Pt in RAW 264.7 cells after incubation with cisplatin and CAP nanospheres, respectively. F) Schematic illustration of in vitro CAP‐Cy5 cell‐to‐cell transferring delivery toward A549 cells. G) CLSM images of A549 cells at the bottom well (CAP‐Cy5@M in the top transwell insert). H) Release behavior of CAP@M at different times in vitro. I) Fluorescence images of A549 cells at the bottom well (CAP@M in the top transwell insert), co‐stained with Calcein–AM/PI. Data were presented as mean ± SD. Not significant = NS. Statistical significance: *p < 0.1, **p < 0.01, and ***p < 0.001.

We also assessed whether our CAP nanosphere could differentiate the macrophage toward M1‐phenotype and stimulate the immunological activity. A series of cytokines, including tumor necrosis factor alpha (TNF­α), CD86, CD206, and Arg‐1 were examined by quantitative reverse transcription polymerase chain reaction analysis (qRT‐PCR) after treatment with various drug formulations. In the group treated with CAP nanospheres, the expressions of TNF­α and CD86, which are the proinflammatory markers of the M1‐phenotype, were shown to be obviously elevated both in the (CpG equivalent to 0.5 and 2.0 µm) CAP groups, when compared to other groups (Figure 4D). In particular, treatment of CAP nanospheres (CpG equivalent to 2.0 µm) resulted in a 4.2 fold increases of TNF­α expressions, and a 5.3 fold increases of CD86 expressions when compared to the naked CpG, which was attributed to the poor cell uptake efficiency of free DNA strands. Besides, the expressions of CD206 and Arg‐1, which are markers of anti‐inflammatory (M2) phenotypes, showed comparable expression levels with the control groups (Figure 4D). Meanwhile, after incubation of CAP nanospheres, macrophages changed their morphology and exhibited an elongated status, which is the characteristic state of the M1‐phenotype, further confirming the successful internalization and polarization (Figure S14, Supporting Information).^{[ 12} , ^{14 ]} These results corroborated that the macrophage cells trend to differentiate toward the proinflammatory M1‐phenotype after being loaded with CAP nanospheres.

Another crucial concern for this CAP@M strategy is whether the loaded CAP nanospheres could be released and transferred from the macrophage. Thus, we studied the CAP laden macrophages’ ability to transfer CAP nanospheres via a transwell experiment.^{[ 28 ]} As depicted in Figure 4F, the pore size of transwell insert membrane is 3.0 µm, which allows the diffusion of CAP nanospheres by passing through the membrane, while the macrophage cells are unable to translocate. Specifically, CAP‐Cy5 loaded macrophages (CAP‐Cy5@M) were seeded in a transwell insert, and set at the top of the bottom culture wells. While, the untreated A549 cells were cultured in the bottom wells. As a result, the fluorescence signals in the A549 cells were observed and increased with the incubation time, indicating that the CAP‐Cy5 nanospheres can be cumulatively released from the macrophage, then diffused through the transwell membrane, and is subsequently uptaken by A549 cells (Figure 4G). These results demonstrated that our CAP@M drug delivery strategy has the ability to efficiently transfer CAP nanospheres. Next, the quantitative release behavior of CAP nanospheres from CAP@M was also studied.^{[ 17 ]} As shown in Figure 4H, CAP nanospheres can be sustainably released from CAP@M, with a cumulative amount of nearly 50% within 36 h. Then, we evaluated the therapeutic efficacy of the released drug from CAP@M and whether the intracellular CAP nanospheres could maintain their cytotoxicity. Similarly, CAP@M were seeded in the transwell insert at the top of the bottom culture wells, untreated A549 cells were cultured in the bottom wells. After incubation with various times, the dead A549 cells were determined through Calcein–AM/PI co‐staining assay, which indicated that the CAP nanospheres can maintain the good cytotoxicity after being transferred by macrophages (Figure 4I). It is found that the rate of dead cells also increased with the incubation time, in agreement with the CLSM and release behavior characterizations. These results demonstrated that we successfully designed a macrophage‐mediated adoptive cell transfer therapy.

Then, we analyzed whether our CAP@M delivery strategy could maintain the intrinsic tumor tropic property of macrophages.^{[ 14} , ^{28 ]} Specifically, the migratory ability of CAP@M was also investigated via transwell experiments, in which untreated RAW 264.7 cells (free macrophage) or CAP@M laden with different concentrations of CAP (5 or 20 µm) were seeded in the top transwell insert (8.0 µm pore sized), and A549 cells were cultured in the bottom wells (Figure  5A). There was no observable RAW 264.7 cells transmigrated through the membrane without A549 cells seeded at the bottom wells that only contained cell culture medium, which was set as the negative control. In contrast, both RAW 264.7 cells and CAP@M showed excellent migratory ability when the A549 cells existed in the bottom wells (Figure 5B). Over 85% CAP@M can transmigrate through the membrane, whether CAP@M contained a moderate or a high concentration (5 or 20 µm) of CAP nanospheres inside (Figure 5C). The results above demonstrated that the loading of CAP nanospheres would not affect the intrinsic tumor tropic and homing ability of macrophages.

Figure 5.

Characterizations of tumor tropic ability and antitumor efficacy of CAP@M. A) Schematic illustration of transwell migration assay of CAP@M toward A549 cells. B) Bright‐field images of migrated RAW 264.7 cells and CAP@M (with an equivalent CAP concentration of 5 or 20 µm) after incubated with a) cell‐culture medium solution or b–d) A549 cells. C) The relative migratory ability of macrophages toward A549 cells ((a) was set as negative control). D,E) Flow cytometry‐based apoptosis analysis of A549 cells induced by macrophages (untreated RAW 264.7), CAP@M (5 µm), and CAP@M (20 µm), respectively, via transwell experiments. F,G) The cell cycle arrest analysis of A549 cells after treated with different formulations via transwell experiments. H) Fluorescence microscopy images of A549 cells stained with caspase‐3 tracker after being treated with different formulations via transwell experiments. I) Flow cytometry analysis of caspase‐3 activity of A549 cells after treated with different formulations. Data were presented as mean ± SD. Not significant = NS. Statistical significance: *p < 0.1, **p < 0.01.

2.4. In Vitro Antitumor Efficacy of CAP@M

Further, we quantitatively evaluated the in vitro antitumor performance of the CAP@M strategy. After the similar treatment of CAP@M via transwell experiments, the apoptotic A549 cells were examined by flow cytometry analysis. As depicted in Figure 5D, the free macrophages did not show the obvious cytotoxicity toward the A549 cells. Whereas, after incubation with CAP@M at an equivalent amount of CAP (5 or 20 µm) for 24 h, the apoptotic percentage of A549 cells significantly increased up to 24.8% and 57.3%, which indicated that CAP@M is capable of inducing cell apoptosis through the transferring CAP nanospheres (Figure 5E). In addition, since the cisplatin related drugs can induce cell cycle arrest and perturbation, the blocking effects of CAP@M on cell cycle progression were further investigated.^{[ 25 ]} As a result, the free macrophages treated group failed to block the cell cycle progression, which showed the same level in each cell cycle phase with the control group. Not unexpectedly, CAP@M was able to simultaneously block the cancer cell cycle at S and G2/M phase, in which the CAP@M (20 µm) exhibited the stronger ability to arrest the cancer cell cycle progression, owing to the higher amount of CAP nanosphere contained (Figure 5F,G). Moreover, to verify that the cell apoptosis is indeed induced by CAP nanosphere, the expression of caspase‐3 in A549 cells was further determined by caspase‐3 tracker imaging, which has been identified as a major mechanism in cisplatin‐induced cell death.^{[ 24 ]} As expected, the detectable caspase‐3 activation signals were obtained both in the treatments of CAP@M (5 and 20 µM) groups (Figure 5H). Particularly, the higher concentration CAP@M treatment group exhibited the stronger caspase‐3 activation signals. The consistent trend was also determined by the flow cytometry analysis (Figure 5I). Together, the aforementioned results demonstrated that our CAP@M strategy has the ability to facilitate the cell apoptosis, enhance the cell cycle blockage, and activate the caspase‐3 signals.

To robustly validate the capability of macrophages to uptake and transfer the CAP nanospheres, we applied an in‐situ visualized method, through 3D microscopy (3D Cell Explorer, NanoLive, Lausanne, Switzerland) to track the movement behaviors of CAP nanospheres in macrophages and A549 cells in real‐time (Figure  6A).^{[ 29 ]} After incubation of CAP nanospheres for 17 min, the uptake process of CAP nanospheres can be visualized, and nanospheres were swallowed at 34 min (Figure 6B; Figure S16, Movie S1, Supporting Information). Then, the intercellular transferring process of nanospheres that were released out from the macrophage cells can be clearly observed after 1 h incubation, which directly demonstrated the successful transferring and delivery of CAP nanospheres by macrophages.

Figure 6.

In‐situ imaging of cellular uptake and intercellular transferring process of CAP nanospheres in macrophages by 3D Cell Explorer (NanoLive). A) Schematic illustration of in‐situ imaging assay. B,C) Time‐lapse images tacked the movement of CAP nanospheres in RAW 264.7 cells, the enlarged regions showing the typical behaviors. Blue arrows indicated the CAP nanospheres. The exact time (hours:minutes) after incubation of CAP nanospheres was labeled on the table. B) Representative images showed the intercellular transferring of CAP nanospheres between RAW 264.7 cells. C) Representative images showed the cellular uptake process of CAP nanospheres in RAW 264.7 cells.

Besides, we can also visualize good ability of macrophage to uptake the nanospheres and the large number of nanospheres inside the macrophage (Figure 6C; Movie S1, Supporting Information). After incubation with nanospheres for ≈20 min, a mass of nanospheres moved around the cells, which were then captured and phagocytosed by the macrophage. Within ≈1 h 45 min after incubation, abundant nanospheres were captured and loaded into the macrophage. Notably, the captured CAP nanospheres remained relatively stable with no obvious changes in their size and morphology within 3 h after incubation. In addition, similar experiments were implemented in A549 cells, in which the excellent uptake effect can be also visualized within a 2 h incubation (Figure S17 and Movie S2, Supporting Information). Thereby, it is demonstrated that macrophage cells can be employed as carrier to effectively load this nanosphere, as well as efficiently deliver them through cell‐to‐cell transferring, distinguishing this strategy from conventional drug delivery systems.

2.5. In Vivo Biodistribution and Pharmacokinetic Evaluation

Inspired by the impressive therapeutic performance in vitro, we further assessed the tumor accumulating ability and biodistribution for various drug formulations in vivo. Specifically, nude mice bearing subcutaneous A549 tumors were intravenously injected with free CpG‐ASO‐Cy5 strands, CAP‐Cy5 nanospheres, and CAP‐Cy5 loaded macrophages (CAP‐Cy5@M), respectively (Figure  7A). The fluorescent signals were captured by optical imaging at 2, 4, 6, 8, 12, and 24 h post‐injection. As depicted in Figure 7B, free CpG‐ASO‐Cy5 strands were not accumulated at the tumor site and were rapidly cleared from the body, which indicates that naked DNA strands can be easily decomposed in vivo.^{[ 25 ]} The CAP‐Cy5 nanospheres group showed the moderate fluorescent signals due to the passive diffusion of nanospheres at the tumor site.^{[ 30 ]} In comparison, owing to the tumor homing ability of macrophages and the passive absorption of nanosized CAP by tumor blood vessels, the strongest accumulation of fluorescence at the tumor area was observed even after 24 h injection in the CAP‐Cy5@M treated group. In addition, to examine the biodistribution of each drug formulation, the major organs and tumors were excised and their fluorescent signals were quantitatively determined through ex vivo imaging after euthanizing the mice at 24 h post‐injection (Figure 7C,F). As a result, the free CpG‐ASO‐Cy5 group revealed no obvious drug accumulation in the tumors, while the CAP treated group displayed moderate fluorescence intensity at the tumor tissues and kidneys, indicating that the DNA‐based nanosphere can be accumulated, and metabolized by kidneys.^{[ 25 ]} Notably, CAP‐Cy5@M group showed the highest fluorescence intensity at the tumor site, which proved that macrophage‐mediated delivery is able to enhance the targeting efficiency through their tumor homing capability.^{[ 7 ]} We also analyzed the distribution of Cy5 fluorescence in the tumor slices of each group. As expected, the CAP‐Cy5@M treated group showed a large intensity and deep penetration of Cy5 fluorescence in the tumor, in agreement with the ex vivo imaging results (Figure 7D). To further verify tumor tropic and homing ability of CAP@M in vivo, we stained the macrophages’ cell membrane with Dio, then studied the tumor slices by CLSM after intravenous injection of Dio‐CAP‐Cy5@M for 24 h (Figure 7E).^{[ 31 ]} As a consequence, major Cy5 red fluorescence signals were co‐localized with the Dio green fluorescence signals at the tumor area, indicating that CAP@M is able to maintain the tumor homing ability of macrophages and target to tumor tissues in vivo. Besides, the CAP‐Cy5 red signals were also separately observed in other sites of the tumor, demonstrating that CAP nanospheres can be released from CAP@M in vivo, as well as the macrophage can act as a vehicle to deliver CAP in vivo. Meanwhile, the pharmacokinetics of each drug formulation were assessed through determination of Pt content in the blood of mice at different time points post‐injection. Free cisplatin was rapidly cleared from the body, and the CAP had the prolonged blood circulation ability, with the half‐life time (t _1/2) of 0.87 and 2.16 h, respectively. Not unexpectedly, the CAP@M had the longest blood circulation time with a half‐life time (t _1/2) of 3.92 h, which demonstrate that the macrophage can improve the circulation of CAP in vivo (Figure 7G).^{[ 31} , ^{32 ]}

Figure 7.

In vivo biodistribution and pharmacokinetics of CAP nanospheres and CAP@M in A549 tumor‐bearing mice. A) Schematic illustration of the preparation of CAP‐Cy5 nanospheres loaded macrophages (CAP‐Cy5@M). B) In vivo fluorescent images of A549 tumor‐bearing mice after intravenous administration of free CpG‐ASO‐Cy5 strands, CAP‐Cy5 nanospheres, and CAP‐Cy5@M at 2, 4, 6, 8, 12, and 24 h post‐injection. Black‐dashed circles indicate the tumor areas. C) Fluorescent images of major organs and tumors harvested from the mice treated with different drug formulations at 24 h after injection. 1 heart, 2 lungs, 3 livers, 4 kidneys, 5 spleens, 6 tumors. D) Fluorescence images of tumor slices after treated with different drug formulations at 24 h. Scale bars: 500 µm. E) CLSM images of tumor slices after intravenous injection of Dio labeled Dio‐CAP‐Cy5@M at 24 h. Scale bars: 100 µm. F) Quantitative analysis of fluorescence distribution of free CpG‐ASO‐Cy5 strands, CAP‐Cy5 nanospheres, and CAP‐Cy5@M in different organ tissues at 24 h after injection. G) In vivo pharmacokinetics of cisplatin, CAP nanospheres, and CAP@M in A549 tumor‐bearing mice, with the Pt dosage of 1.5 mg kg⁻¹. Data were presented as mean ± SD. n = 3. Statistical significance: *p < 0.1, **p < 0.01.

2.6. Antitumor Efficacy and Immunomodulation Studies In Vivo

Further, to systematically investigate the in vivo antitumor efficacy of CAP nanospheres and CAP@M, subcutaneous A549 tumor‐bearing nude mice were randomly divided into six groups and intravenously injected with saline, free cisplatin, macrophage cells (untreated RAW 264.7), random DNA‐Pt nanospheres, CAP nanospheres and CAP@M at an equivalent dosage of Pt (1 mg kg⁻¹) every three days (Figure  8A,B). As a result, the tumor volume curves of saline group radically increased, and the average tumor volume finally reached 300 mm³. Macrophages, free cisplatin and random DNA‐Pt groups only elicited limited antitumor efficacy, in which the average tumor volumes reached 239, 207 and 149 mm³ after the treatment, respectively. In contrast, the CAP nanospheres group exhibited favorable tumor‐inhibiting ability, which might be induced by the co‐delivery of cisplatin and ASO agent. In particular, the CAP@M group showed the stronger therapeutic performance. The mean tumor volume of the CAP@M group reached only 57 mm³ after 15 days treatment, which was smaller than the CAP treated group (96 mm³) and other control groups (Figure 8D). In addition, the changes of mice body weight were monitored during the experiments, and the slight increase of body weight can be detected in both the CAP and CAP@M groups, which indicated that the CAP@M strategy has lower toxicity and side effects than other control groups (Figure 8F). The overall survival curve revealed that the survival condition of mice can be improved after the treatments with CAP nanospheres and CAP@M, proving the competitive advantages of CAP@M treatment in prolonging the survival rate and antitumor activity (Figure 8G). Moreover, histological analysis of the major organs for all groups revealed that there was no observable damage induced by the CAP nanospheres and CAP@M (Figure S18, Supporting Information). These results determined the significant antitumor performance as well as the well‐tolerated biosafety of our designed CAP@M drug delivery strategy.

Figure 8.

In vivo antitumor efficacy in A549 tumor‐bearing mice. Schematic illustration of A) the preparation of CAP loaded macrophages (CAP@M) as nanomedicine and B) the treatment process. C) Representative photographs of excised tumors in each group after the treatments. D) Tumor volume curves of A549 tumor‐bearing mice after intravenous injection of saline, free cisplatin, macrophage, random DNA‐Pt, CAP nanospheres, and CAP@M at the dosage of Pt (1 mg kg⁻¹) every 3 days for 5 times (n = 5 for each group). E) Tumor weight of excised tumors after the treatments. F) Body weight change curves of mice during the treatments. G) The overall survival rate of mice in each group. H) H&E staining, TUNEL staining, immunohistochemical staining of P‐gp, caspase‐3 and MHC‐II for the tumor sections excised from A549 tumor‐bearing mice after the treatments. Scale bars: 50 µm. Data were presented as mean ± SD. Statistical significance: *p < 0.1, **p < 0.01, and ***p < 0.001.

We also investigated the level of inflammation related cytokine in blood samples taken from the mice at the day 3rd, 9th, and 15th during the treatments, including TNF­α and interleukin‐12 (IL‐12).^{[ 33 ]} As shown in Figure  9A,B, no clear changes of both TNF­α and IL‐12 levels were observed in CAP@M treated group, which had the comparable levels within the normal ranges, compared with other control groups. The results verified that our CAP@M strategy would not induce the severe immune rejection and systematic toxicity in vivo. At the end of the antitumor assay, mice were euthanized, then the tumors were excised and photographed (Figure 8C). The tumor weight in each group was calculated, which indicated that the CAP@M treated group presented the highest inhibition rate, consistent with the tumor volume curves results (Figure 8E). To further investigate whether the CAP@M could be accumulated in tumor tissues and maintained as the proinflammatory M1 phenotype, we separated tumor tissues and dissociated them into single cell suspensions, then stained the cells with fluorescence‐labeled antibodies of CD11b, F4/80, CD86, and CD206.^{[ 34 ]} As depicted in Figure 9C–F, the proportion of M1 marker CD86 in tumors markedly increased after the CAP@M treatments in comparison with other groups, which validated the M1 differentiation and tumor homing capability of CAP@M. Meanwhile, there is no significant increase in the proportion of M2 marker CD206 among each group, again evincing that CAP@M can maintain the M1 phenotype during the in vivo therapy.

Figure 9.

In vivo immunomodulatory effect in A549 tumor‐bearing mice. A,B) Cytokine level of TNF­α and IL‐12 in sera collected form mice during the treatments, evaluated by ELISA. C,D) Representative flow cytometry‐based analysis and relative quantification of M1 macrophages (CD86^hi) (gating by F4/80⁺ CD11b⁺) in the tumor tissues after the treatments. E,F) Representative flow cytometry‐based analysis and relative quantification of M2 macrophages (CD206^hi) (gating by F4/80⁺ CD11b⁺) in the tumor tissues after the treatments. Data were presented as mean ± SD. n = 3. Not significant = NS. Statistical significance: *p < 0.1, **p < 0.01.

Moreover, histological and immunohistochemical studies were performed to analyze the tumor tissues and verify the antitumor mechanisms (Figure 8H).^{[ 11} , ^{20 ]} Hematoxylin and eosin (H&E) staining showed the obvious nuclear fragments and shrinkage in the CAP@M treated group. Transferase‐mediated dUTP nick‐end labeling (TUNEL) staining validated that the CAP@M treated group induced more tumor cells apoptosis, as evidenced by the highest green fluorescent signals. P‐gp and caspase‐3 immunohistochemical assays were further employed to verify the down‐regulation of P‐gp expression and caspase‐3 activation in vivo. As expected, both CAP and CAP@M treated groups showed the decreased expression of P‐gp and the elevated expression of caspase‐3, demonstrating the successful combinatorial therapeutics were obtained in vivo, which was in accordance with the results in vitro. To evaluate the enrichment of M1‐phenotype macrophages in tumor tissues, major histocompatibility complex II (MHC‐II) was selected as the marker of M1 phenotype, and the immunofluorescence staining was conducted in each group. As a consequence, the CAP@M treated group exhibited the intensive MHC‐II fluorescence in the tumor, which confirmed that the immunotherapeutic effect was successfully combined in CAP@M strategy. These results demonstrated that our designed CAP@M drug delivery strategy has remarkable in vivo therapeutic efficacy through the synergistic chemo‐/gene‐/immuno‐therapy.

Finally, to verify the universality of our CAP@M strategy in enhancing the therapeutic performance and immunostimulatory efficacy, we employed the CAP@M strategy on the immunogenic Lewis lung carcinoma (LLC) tumor‐bearing BALB/c mice model. Similarly, the mice were intravenously injected with saline, macrophage, cisplatin, random DNA‐Pt, CAP nanospheres, and CAP@M with an equivalent Pt dosage (1 mg kg⁻¹) every three days (Figure  10A). As displayed in Figure 10B, saline, macrophage and cisplatin treated groups failed to suppress the tumor growth, with the average tumor volume of 1616.7, 1292.6 and 791.5 mm³, respectively. For random DNA‐Pt and CAP nanospheres treated group, we observed the moderate antitumor effect, which had the average tumor volume of 510.1 and 359.2 mm³. Comparatively, CAP@M treated group exhibited the significant tumor growth inhibiting efficacy, in which the mean tumor volume merely reached 202.1 mm³. After the treatments, mice were euthanized, the tumors were separated and photographed. The tumor weight graph and the corresponding photo indicated that the best tumor growth‐inhibiting efficacy was achieved in CAP@M treated group, in consistency with the result of tumor volume curves (Figure 10C,D).

Figure 10.

In vivo antitumor efficacy, immunotherapeutic performance in LLC tumor‐bearing BALB/c mice. A) Schematic illustration of the treatment process in LLC tumor‐bearing BALB/c mice. B) Tumor volume curves of LLC tumor‐bearing BALB/c mice after intravenous injection of saline, free cisplatin, macrophage, random DNA‐Pt, CAP nanospheres, and CAP@M at the dosage of Pt (1 mg kg⁻¹) every 3 days for 7 times (n = 5 for each group). C) Tumor weight of excised tumors after the treatments. D) Representative photographs of excised tumors in each group after the treatments. E) Flow cytometry‐based analysis of M1 macrophages (CD80^hi) and M2 macrophages (CD206^hi) in the tumor tissues after the treatments. F) Relative quantification of M2 macrophages (CD206^hi). G) Relative quantification of M1 macrophages (CD80^hi). H) Relative ratio of M2 to M1. Data were presented as mean ± SD. n = 3. Statistical significance: *p < 0.1, **p < 0.01, and ***p < 0.001.

We continued to assess the immunostimulatory activity induced by the different drug formulations.^{[ 34 ]} Specifically, the tumor tissues were collected and dissociated into single cell suspensions, then cells were stained with fluorescence‐labeled antibodies to identify the intratumoral macrophages’ phenotype by flow cytometry. As a result, there were no obvious difference in the percentage of M1 and M2 between the macrophage, cisplatin and random DNA‐Pt treated groups, which had the similar level of M1 and M2 phenotype compared to the saline group. For CAP treated group, the proportion of M1 phenotype (18.7%) was slightly increased, while the proportion of M2 phenotype (34.7%) showed the modest decline, suggesting that only the moderate immunoregulation can be obtained after the CAP therapy. Strikingly, the CAP@M treated group exhibited the strongest immunostimulatory performance, as evidenced by the highest M1 percentage of 29.8% and the lowest M2 percentage of 22.3% in the tumor tissues. Overall, these results comprehensively demonstrated that our CAP@M strategy have the excellent antitumor efficacy and effective immunotherapeutic capability (Figure 10E–H).

3. Conclusion 

In summary, we demonstrated a novel strategy to construct a multifunctional self‐assembled DNA nanosphere through adoptive macrophages delivery for NSCLC therapy. These CAP nanospheres can be easily manufactured through self‐assembly with the introduction of the chemo‐drug cisplatin and gene therapeutic CpG‐ASO under moderate conditions, which realized the multifunction over one single nanostructure. Additionally, this biocompatible DNA nanosphere displayed efficient cellular uptake and satisfied subcellular distribution, as well as enhanced antitumor efficacy via a combinatorial effect by cisplatin‐based chemotherapeutics and ASO‐based blockage of P‐gp drug efflux. Importantly, the CpG motif in CAP nanosphere can differentiate the macrophage toward the M1‐phenotype and stimulate immunoregulation. After the loading in macrophages, CAP@M showed prolonged circulation life‐time, favorable tumor homing ability, effectively adoptive delivery, and enhanced intratumoral immunotherapeutic capability via the differentiation to pro‐inflammatory M1 phenotype. Comprehensively, owing to high biocompatibility, synergistic effect of cisplatin, ASO agent, CpG motifs and macrophage‐based cell delivery system, the dramatically enhanced antitumor efficacy was obtained both in vitro and in vivo. Encouraged by our results, other gene therapeutic sequences can also be considered to self‐assemble other functional nanospheres by a similar methodology. Meanwhile, macrophages can be engineered as an amenable mediator for safe and effective cancer therapies. We believe that this CAP@M provides a promising strategy with diverse advantages for cancer therapy in the future.

4. Experimental Section

Synthesis of CAP Nanospheres

The CAP nanospheres were prepared via the facile one‐pot self‐assembly method.^{[ 23 ]} Briefly, 5 µL an aqueous solution of 20 mm cisplatin was added to 95 µL an aqueous solution of 32 µm CpG‐ASO DNA strands. Then the mixture was heated at 95 °C for 2 h. The crude product of CAP nanospheres was washed by ultrapure water through the centrifugation at 3000 rpm. The collected CAP nanospheres were dissolved in ultrapure water to form a stock solution. The cisplatin concentration was detected by ICP‐MS. To determine DNA encapsulation efficiency, the unassembled DNA strands were collected after purification, which were further determined by UV–vis spectrometry at 260 nm. Then, the DNA encapsulation efficiency was calculated by the formula below:^{[ 22 ]}

$$\begin{array}{ccc}& & \mathrm{encapsulation}\mathrm{efficiency}\left(\mathrm{EE}\right)\hfill \\ & & =\left(\left({\mathrm{DNA}}_{\mathrm{total}}-{\mathrm{DNA}}_{\mathrm{unassembled}}\right)/{\mathrm{DNA}}_{\mathrm{total}}\right)\times 100\%\hfill \end{array}$$ (1)

Preparation of CAP Nanospheres Loaded Macrophages (CAP@M)

The macrophage cells were seeded in 6‐well plates and cultured for 24 h. Then, the CAP nanospheres were incubated, and the FBS‐free Dulbecco's modified Eagle's medium (DMEM) medium was used for the cell culture. After a 12 h incubation, the cells were washed three times with ice‐cold PBS through the centrifugation at 500 rpm. The CAP@M suspension was then collected for the subsequent experiments. To determine the amount of CAP nanospheres loaded into macrophage cells, the intracellular platinum content was tested by ICP‐MS measurements after being digested with HNO₃ (65%).^{[ 24 ]}

Characterization of CAP Nanospheres

TEM images were obtained by using the TALOS L120C G2 electron microscope at 120 kV. High‐Resolution‐TEM (HRTEM) imaging, HADDF‐STEM imaging, Elemental mapping and EDS line scanning were carried out on the TALOS F200X TALOS F200X electron microscope (FEI, USA) at 200 kV. SEM images were performed using the JSM‐7800F electron microscope. The size measurements and zeta potential were conducted by the Zetasizer Nano ZS90 (Malvern Instruments Ltd., UK). UV–vis spectrum was performed by the microplate reader (TECAN, InfiniteM200, Switzerland). ICP‐MS was performed by using the iCAPQ ICP‐MS spectrometer (Thermo Fisher, Germany).

The Stability of CAP Nanospheres

CAP nanospheres solution was mixed with FBS containing culture medium, with the final FBS concentration of 10%. The mixture solution was incubated at 37 °C for predetermined time (from 0 to 24 h). After incubation, all samples were analyzed by TEM imaging.^{[ 25 ]}

In Vitro Drug Release Behaviors

CAP nanospheres solutions were transferred into dialysis bags (MWCO: 1000 Da) and then dialyzed in different buffer solutions PBS (pH = 4.4), PBS (pH = 6.0), and PBS (pH = 7.0), respectively. The release process was carried out at 37 °C with gentle shaking. At predetermined time points (from 0 to 72 h), the dialysate of each buffer was taken out, and equal volumes of corresponding fresh buffer solutions were added. The release contents of cisplatin were evaluated by ICP‐MS.^{[ 20 ]}

Cell Culture

A549 cells (human lung adenocarcinoma cell line), Helf, mouse LLC cell line were cultured in Roswell Park Memorial Institute 1640 (RPMI 1640) medium, RAW 264.7 (murine macrophages), HeLa (cervical cancer cell line), MCF‐7 (pancreatic cancer cell line), and MGC‐803 (gastric cancer cell line) were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% FBS and antibiotics (50 units mL⁻¹ penicillin and 50 units mL⁻¹ streptomycin) in 5% CO₂ atmosphere at 37 °C.

Cellular Uptake Studies of CAP Nanospheres

CLSM and flow cytometry were used to investigate the cell uptake behaviors of CAP nanospheres. For CLSM imaging, A549 and RAW 264.7 cells were incubated with Cy5 labeled CAP‐Cy5 nanospheres at the concentration of Cy5 with 300 nm for predetermined time points. Subsequently, the cells were stained with Hoechst fluorescence dye and LysoTracker Green DND‐26. Thereafter, the culture medium was removed, then the cells were washed with PBS three times and visualized using confocal microscopy (Leica TCS SP8 STED 3×). For flow cytometry analysis, A549 cells were incubated with CAP‐Cy5 nanospheres at a concentration of Cy5 with 300 nm for 4, 8, 12, and 24 h, respectively. Thereafter, the culture medium was removed and the cells were washed with PBS by three times. The amounts of intracellular fluorescent signal were quantified using flow cytometry system (LSRFortessa, BD, USA).

In Vitro Cytotoxicity Assay

The cytotoxicity of different drug formulations against A549, HeLa, MCF‐7, MGC‐803, RAW 264.7, and Helf cell lines was investigated by MTT assay. Cells were seeded in 96‐well plates and incubated for 24 h. The different formulations of drug were adjusted to equal Pt concentrations and equal DNA concentrations before they were added, then the cells were incubated under the appointed concentrations for 44 h. After incubation, 10 µL MTT solution (5 mg mL⁻¹ in PBS) was added to each well and incubated for 4 h. Finally, the medium was removed, then the DMSO was added into each well (100 µL per well). The absorbance (490 nm) was measured by microplate reader (TECAN, InfiniteM200, Switzerland) to calculate cell viability. Each experiment was repeated at least three times.

Cell Live/Dead Staining Assay

A549 cells were seeded in 24‐well plates and cultured for 24 h. Then the cells were incubated with different drug formulations at the equal concentration of Pt (15 µm) for 48 h. Then the cells were co‐stained by using Calcein–AM/PI detection kit (Invitrogen, China) according to the manufacturer's instructions. Then the samples were imaged by using fluorescence microscopy.

Cell Apoptosis Assay

A549 cells were seeded in 24‐well plates and cultured for 24 h. Then the A549 cells were incubated with different drug formulations at the equal concentration of Pt (15 µm) for 48 h. The apoptosis assays were co‐stained by using Annexin V‐FITC/propidium iodide apoptosis detection kit (Invitrogen, China) according to the manufacturer's instructions. Briefly, cells were suspended in binding buffer, with the addition of Annexin V‐FITC (10 µL) and PI (5 µL), and incubated at 4 °C for 15 min. Then the samples were analyzed by BD LSRFortessa flow cytometry.

Cell P‐Glycoprotein Expression Assay

A549 cells were seeded in 12‐well plates and cultured for 24 h. Then the A549 cells were incubated with different drug formulations at the equal concentration of Pt (15 µm) for 48 h. The P‐glycoprotein (P‐gp) expressions were evaluated by flow cytometry analysis and western blot assay. For flow cytometry analysis, the cells were collected and stained with PE‐conjugated mouse antihuman P‐gp antibody (1:50, ab93590, Abcam, UK). The amounts of fluorescent signal were quantified using flow cytometry system (LSRFortessa, BD, USA). For western blot assay, the cell lysates were separated by SDS‐PAGE gel, and extracted proteins in the gel were transferred by the polyvinylidene difluoride (PVDF) membranes. Then, the membranes were incubated with primary antibodies against P‐gp and GAPDH for 12 h at 4 °C and secondary antibodies for 2 h at room temperature. The binding of antibody was evaluated using ECL Prime Western Blotting Detection Reagent (GE Healthcare UK Ltd., UK) and the images were obtained by ChemiDoc XRS system.

Macrophage Phenotype Change Assay

The RAW 264.7 cells were seeded in 24‐well plates and cultured for 24 h. The different formulations of drug were adjusted to equal Pt concentration and equal DNA concentration (free CpG [2 µm], CAP nanosphere [CpG equivalent to 0.5 and 2 µm]) before they were added, then the cells were incubated for 24 h. The gene expression of TNF‐α, CD86, Arg‐1, and CD206 were evaluated by the quantitative real‐time PCR. For quantitative reverse transcription polymerase chain reaction (qRT‐PCR), the total mRNAs were extracted from RAW 264.7 cells and cDNAs were produced through 5× PrimeScript RT Master Mix (Takara, China) according to the manufacturers’ instructions.^{[ 10b ]}

Cell Transwell Assay

A transwell polycarbonate membrane cell culture insert (3.0 µm pore sized, Corning) was employed for the transwell assay, which allows the free diffusion of CAP nanospheres between the transwell insert and the bottom culture well, but not cell translocations.^{[ 28 ]} To quantitatively evaluate the release behavior of CAP@M, the CAP@M were seeded in the transwell inserts, with the CAP content of 20 µm, and placed on the top of the bottom wells, then incubated with fresh cell culture medium. At predetermined time points (from 0 to 36 h), cell culture medium (200 µL) in the bottom wells were collected, and equal volumes of fresh cell culture medium were replenished. The contents of released CAP nanospheres were determined by ICP‐MS. To investigate the ability of CAP@M to induce cell apoptosis, block cell cycle and induce caspase‐3 activation, the CAP@M was seeded in the transwell inserts, with the CAP amount of 5 and 20 µm, and then A549 cells were cultured in the bottom wells. After 24 h incubation, the A549 cells in the bottom wells were analyzed by Annexin V‐FITC/propidium iodide apoptosis detection kit (Invitrogen, China), cell‐cycle detection kit (Invitrogen, China), and Caspase‐3 Activity Assay kit (Invitrogen, China) respectively, according to the manufacturer's instructions, then analyzed by flow cytometry and fluorescence microscopy.

Cell Migration Assay

A transwell polycarbonate membrane cell culture insert (8.0 µm pore sized, Corning) was employed for the tumor tropic migration assay, which allows the RAW 264.7 cells and CAP@M to migrate through membrane.^{[ 28 ]} For the negative control group, the bottom wells only contained the cell culture medium. For other groups, the bottom wells were seeded with A549 cells. Then CAP@M were seeded in the transwell inserts, with the CAP amount of 5 and 20 µm, and the untreated RAW 264.7 (free macrophages) served as the control. After 12 h of incubation, the transwell inserts were fixed in 4% paraformaldehyde for 30 min. Then the transwell inserts were washed with PBS by three times and stained by crystal violet agent for 30 min. The macrophage cells that failed to transmigrate through the membrane were scraped off by swabs, then the migrated macrophage cells were imaged and counted via the fluorescence microscopy.

Cellular In‐Situ Imaging Assay

RAW 264.7 cells and A549 cells were seeded in confocal plates and cultured for 24 h. Then the cells were incubated with CAP nanospheres and in‐situ imaged by a 3D tomographic microscopy (3D Cell Explorer, NanoLive, Lausanne, Switzerland) to record videos.^{[ 29 ]} Representative images were taken with the appropriate field of view at different time points.

Animal Tumor Model

Animal studies were conducted in accordance with the guidelines for the care and use of laboratory animals approved by the Animal Ethics Committee of Shanghai Jiao Tong University. The suspensions of A549 cells were subcutaneously injected in the flank of BALB/c female nude mice with 200 µL (1 × 10⁶ cells) to established A549 tumor‐bearing model. The suspensions of LLC cells were subcutaneously injected in the flank of BALB/c female mice with 200 µL (1 × 10⁶ cells) to the established LLC tumor‐bearing model.

In Vivo Tumor Targeting and Biodistribution Studies

In vivo biodistribution imaging was performed using Bruker In‐Vivo F PRO imaging system (Billerica, MA, USA) with 630 nm excitation wavelength and 680 nm emission wavelength. The A549 tumor‐bearing nude mice were anesthetized with isoflurane and images were taken at predetermined time points with 2, 4, 6, 8, 12, and 24 h after intravenously injected with Cy5 labeled CpG‐ASO‐Cy5 strands, CAP‐Cy5 nanospheres, and CAP‐Cy5@M. For the ex vivo biodistribution imaging studies, mice were euthanized at 24 h post‐injection. Then the major organs and tumor tissues were harvested and imaged using the same system. For the analysis of Dio‐labeled Dio‐CAP‐Cy5@M, CAP‐Cy5@M was stained with Dio (Invitrogen, China) according to the manufacturer's instructions. Mice were euthanized at 24 h post‐injection, and tumor tissues were excised, then the tumor slices were analyzed by CLSM.

In Vivo Pharmacokinetic Studies

A549 tumor‐bearing nude mice were randomly divided into 3 groups. Mice were intravenously injected with free cisplatin, CAP nanospheres and CAP@M, respectively with an equivalent Pt dosage of 1.5 mg kg⁻¹. The 150 µL orbital venous blood samples were collected from each mouse at the predetermined time points by 0.5, 2, 4, 8, 12 and 24 h, and then the Pt content of blood was evaluated by ICP‐MS measurements after digested with HNO₃ (65%).

In Vivo Antitumor Efficacy and Immunomodulation Evaluation

The A549 tumor‐bearing nude mice were randomly divided into six groups of five mice when the tumor volume reached 50 mm³. Saline, free cisplatin, macrophages (untreated RAW 264.7 cells), random DNA‐Pt nanospheres, CAP nanospheres, and CAP@M were intravenously injected via the tail vein at the dosage of Pt (1 mg kg⁻¹, 200 µL) once every three days for 15 days. The tumor size was calculated according to the formula: V (mm³) = 1/2 × a (mm) × b (mm) × b (mm), in which a and b were the length and width of the tumors. The orbital venous blood samples were collected on the day of 3rd, 9th and 15th, then were tested the level of TNF‐α and IL‐12 through ELISA according to the manufacturer's instructions. Finally, mice were euthanized on the day 15th, then the tumor tissues and major organs were harvested for further flow cytometry‐based analysis, hematoxylin and eosin (H&E) staining, terminal deoxynucleotidyl transferase‐mediated dUTP nick‐end labeling (TUNEL) staining and immunohistochemical analysis of P‐gp, Caspase‐3 and MHC‐II according to the manufacturer's instructions. Similarly, the LLC tumor‐bearing BABL/c mice were performed under the similar experimental conditions. Mice were euthanized on day 21th, then the tumor tissues were separated for further flow cytometry‐based analysis

Flow Cytometry Analysis

For analysis of intratumoral M1 and M2 phenotype macrophages, tumor tissues were separated after the antitumor experiments. Then, tumor tissues were cut into small pieces, and digested into single cell suspensions. The tumor cells were stained using fluorescence‐labeled antibodies of CD11b, F4/80, CD80, CD86, and CD206 (BioLegend, USA) according to the manufacturer's instructions. All samples were analyzed using BD LSR Fortessa analyzer (BD Biosciences). Data were analyzed by FlowJo.^{[ 34 ]}

Statistical Analysis

The experimental data were statistically analyzed by using the Student's t‐test. p‐values < 0.1 were considered with the statistical significance. Besides, experimental data were reported as the mean ± standard deviations (SD) from at least 3 individual experiments.

Conflict of Interest

The authors declare the following conflict of patent “ZL 2020 1 0318950.4”. W. Cai is a scientific advisor, stockholder, and grantee of Focus‐X Therapeutics, Inc.

Supporting information

Supporting Information

Supplemental Movie 1

Supplemental Movie 2

Wang Y., Zhang L., Liu Y., Tang L., He J., Sun X., Younis M. H., Cui D., Xiao H., Gao D., Kong X.‐Y., Cai W., Song J., Engineering CpG‐ASO‐Pt‐Loaded Macrophages (CAP@M) for Synergistic Chemo‐/Gene‐/Immuno‐Therapy. Adv. Healthcare Mater. 2022, 11, 2201178. 10.1002/adhm.202201178

Data Availability Statement

Research data are not shared.