Docetaxel-loaded pH/ROS dual-responsive nanoparticles for the targeted treatment of gastric cancer

Abstract

Gastric cancer (GC), characterized by its high incidence and mortality, poses a great threat to public health worldwide. Although various advanced treatments have been developed for GC, the cure rate remains poor. Docetaxel (DTX), a broad-spectrum antitumor drug, has been widely used for the treatment of GC. However, its use in clinical practice is limited by its low water solubility and severe side effects. Our previous work developed a pH/ROS dual-responsive nanoplatform to deliver DTX (DTX/FA-CA-Oxi-αCD NPs) for the targeted treatment of breast cancer. These nanotherapeutics displayed desirable therapeutic effects for breast cancer without obvious adverse effects. On the basis of the treatment potential of DTX/FA-CA-Oxi-αCD NPs, these nanoparticles (NPs) were used for the targeted treatment of GC. In vitro experiments demonstrated that DTX/FA-CA-Oxi-αCD NPs can be efficiently internalized by HGC-27 cells and deeply penetrate tumor spheroids. Moreover, DTX/FA-CA-Oxi-αCD NPs effectively hindered GC cell migration by approximately 60.1% and decreased GC cell invasion. In vivo experiments revealed that DTX/FA-CA-Oxi-αCD NPs obviously accumulated at the tumor site and that the released DTX significantly blocked tumor growth by approximately 79.6%. In conclusion, DTX/FA-CA-Oxi-αCD NPs exhibited promising therapeutic outcomes for GC treatment.

Background

According to the survey data from the International Agency for Research on Cancer (IARC) in 2022, gastric cancer (GC) ranks as the fifth most common malignant, which brings great mortality rate for patients (Bray et al. 2024). In East Asia, including South Korea, Japan and China, the incidence of GC is notably high (Huang et al. 2023; Teng et al. 2024), which seriously affects public health. Owing to the unclear early clinical symptoms of GC and the limited adoption of gastroscopy examination (Zhang et al. 2023), many patients are diagnosed at an advanced stage (Smyth et al. 2020). Although advanced treatments have been used for GC treatment, the overall long-term survival rate for patients with GC remains low (Alsina et al. 2023; Bando et al. 2019; Shridhar et al. 2013). Currently, surgery combined with chemical drugs is an important approach for GC treatment (Baba et al. 2023). DTX, 5-fluorouracil, and oxaliplatin are widely used chemotherapy drugs for the treatment of GC (Al-Batran et al. 2019; Chabner and Roberts 2005). DTX blocks the cell cycle in the G2/M phase, inhibits tumor cell proliferation and promotes tumor cell apoptosis (Nehme et al. 2001). However, the application of DTX in GC chemotherapy is restricted by its nonspecific distribution, low solubility, fast degradation, and obvious side effects (Nagaraju et al. 2021).

Nano-drug delivery systems (NDDSs) represent promising therapeutic approaches that have gained increased attention from researchers owing to their numerous advantages, including enhanced drug solubility, prolonged drug circulation time within the body, improved drug biocompatibility, and reduced drug toxicity. The targeted delivery of chemotherapy drugs via NDDSs for the treatment of GC has been widely investigated. For example, paclitaxel (Fernandes et al. 2019; Yu et al. 2020), 5-fluorouracil (Qu et al. 2015; Sun et al. 2017), and cisplatin (Yang et al. 2018) nanotherapeutics have demonstrated significant antitumor efficacy against GC as well as reduced systemic toxicity. In addition, DTX-loaded NPs have also been used for the treatment of GC. For example, Xu et al. employed DTX-loaded NPs modified with antibodies against programmed death-ligand 1 (PD-L1) for the targeted treatment of GC cells which are highly express PD-L1 (Xu et al. 2019). Cui et al. (2014) synthesized DTX-loaded and gelatinase-responsive NPs as a radiosensitizer, thereby increasing the efficacy of radiotherapy for GC. Liang et al. (2021) synthesized π electron-stabilized polymeric micelles to increase the efficacy of DTX in the treatment of advanced GC. In conclusion, chemical drug-based nanotherapeutics have promising prospects for GC treatment. Like in most cancer types, high levels of reactive oxygen species (ROS) (Xu et al. 2017) and acidic microenvironment are typical features of GC (Liu et al. 2024; Wang et al. 2024), which provides an opportunity to control drug release for GC treatment. In addition, tumor cells overexpress various receptors on their cell membrane surface, including folate receptors (FR), arginine-glycine-aspartic acid (RGD) receptors, and hyaluronic acid (HA) receptors (Zhao et al. 2020). Some ligands, such as folic acid (FA), RGD, and HA, exhibit strong affinity for these receptors. Consequently, these ligand-modified nanotherapeutics can actively target GC via ligand–receptor interactions (Fig. 1).

Fig. 1

(Source materials were permitted from https://www.figdraw.com/)

Schematic illustration of DTX-loaded NPs for GC treatment

In previous work, we prepared a pH/ROS dual-responsive material by using cinnamaldehyde and 4-hydroxyphenylboronic acid pinacol ester to modify α-cyclodextrin (CA-Oxi-αCD), which has desirable pH and ROS responsiveness. This material was used to as a carrier to encapsulate dexamethasone and prednisolone for the targeted treatment of rheumatoid arthritis and chronic pelvic pain syndrome (Lu et al. 2023; Yang et al. 2023). Importantly, using this material as a carrier, we have fabricated DTX-loaded nanoparticles (DTX/FA-CA-Oxi-αCD NPs) which significantly blocked breast cancer growth and inhibited metastasis (Wang et al. 2023). On the basis of these interesting results, herein, we investigated the efficacy of DTX-loaded pH/ROS dual-responsive NPs for GC treatment. In vitro experiments demonstrated that DTX-loaded NPs can be efficiently internalized by HGC-27 cell, hindering GC cell migration, decreasing GC cell invasion and promoting GC cell apoptosis. In vivo experiments confirmed that DTX-loaded NPs can target the tumor site and significantly block tumor growth.

Methods

Reagents

Roswell Park Memorial Institute Medium 1640 (RPMI-1640) was purchased from Gibco Co. Ltd. (New York, USA). Fetal bovine serum (FBS) was purchased from BioChannel Biotechnology Co., Ltd (Nanjing, China). The streptomycin–penicillin solution was provided by Biosharp Technology Co., Ltd. (Anhui, China). 4ʹ,6-Diamidino-2-phenylindole (DAPI), Cell Counting Kit-8 (CCK-8), and serum-free cell freezing medium were obtained from Beyotime Biotechnology (Shanghai, China).

Cells and animals

The human GC cell lines HGC-27 and BGC-823, authenticated by short tandem repeat (STR) analysis, were generously supplied by the Department of Pathology, Southwest Hospital of Army Medical University. The cell lines were cultured in RPMI-1640 medium supplemented with 10% FBS and 1% penicillin/streptomycin solution at 37 °C in a 5% CO₂ atmosphere.

Six-week-old female BALB/c nude mice weighing 18–20 g were purchased from Huafukang Co., Ltd. (Beijing, China) and housed in an SPF-level laboratory animal room. The animal experiments were approved by the Laboratory Animal Welfare and Ethics Committee of Army Medical University (AMUWEC20237049, Chongqing, China).

Fabrication and characterization of the NPs

The DTX-loaded NPs were fabricated as described in our previous work (Wang et al. 2023). In brief, 6 mg of lecithin and 8 mg of DSPE-PEG₂₀₀₀ were dispersed in deionized water containing ethanol and heated at 65 °C for 30 min. 50 mg of CA-Oxi-αCD and 10 mg of DTX were dissolved in 200 μL of DMSO and 400 μL of methanol. The drug and carrier solution were slowly added dropwise to the lipid dispersion, followed by 2 h of self-assembly at room temperature. The NPs were harvested by centrifugation and stored in deionized water. FA-modified NPs were prepared via a similar method except that 4 mg of FA-DSPE-PEG₃₄₀₀ was used. In addition, the preparation of Cy5-labeled CA-Oxi-αCD NPs and Cy5-labeled FA-CA-Oxi-αCD NPs was described in detail in our previous work (Wang et al. 2021).

The size, polydispersity index (PDI), and zeta potential of the NPs were measured via dynamic light scattering (DLS) and laser Doppler velocimetry (Nano ZS, Malvern, UK). The morphology of the NPs was imaged via transmission electron microscopy (TEM, JEM-1400, Tokyo, Japan). The concentration of DTX loaded in the NPs was detected via high-performance liquid chromatography (HPLC).

Cell uptake of NPs

HGC-27 cells (1 × 10⁵ cells/dish) were seeded in confocal dishes. After overnight incubation, the cells were cocultured with Cy5, Cy5-labeled CA-Oxi-αCD NPs, or Cy5-labeled FA-CA-Oxi-αCD NPs at 1 μg/mL Cy5 for 2 or 6 h. After treatment, the cells were gently washed three times with cold PBS, fixed with 4% (v/v) paraformaldehyde, and then washed with PBS. All procedures were carried out in the dark. Finally, nuclear staining was performed by using DAPI for 10 min, and the cellular uptake of the NPs was captured via confocal laser scanning microscopy (CLSM) (Carl Zeiss, Baden-Württemberg, Germany).

Penetration of NPs into tumor spheroids

For tumor spheroid formation, 3 × 10² HGC-27 cells were seeded in a 96-well ultralow attachment (ULA) round-bottom plate (Corning Inc. USA) with 200 μL of RPMI-1640 medium supplemented with 20% FBS at 37 °C and 5% CO₂ for 4 days (Wang et al. 2016). After formation, the tumor spheroids were cocultured with Cy5, Cy5-labeled CA-Oxi-αCD NPs, or Cy5-labeled FA-CA-Oxi-αCD NPs at 1 μg/mL Cy5 for 6 h. Then, the tumor spheroids were washed three times with cold PBS, fixed with 4% paraformaldehyde for 30 min, and washed three times with cold PBS again. After the nuclei were stained with DAPI for 30 min, the tumor spheroids were transferred to a confocal dish, and CLSM layer scanning mode was used to observe the penetration of the NPs.

In addition, the treated tumor spheroids were fixed and dehydrated with 30% (w/v) sucrose solution before being cryosectioned into 10-μm-thick sections. After the nuclei were stained with DAPI for 10 min, the fluorescence intensity in the tumor spheroids was observed via CLSM.

Wound healing assay

HGC-27 cells were seeded in 6-well plates with RPMI-1640 medium containing 10% FBS. After the cells grew to almost 95% confluence, the cell monolayer was scratched for lesion formation. Then, the medium and cell debris were aspirated, and the cells were treated with DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs at a concentration of 10 ng/mL DTX in RPMI-1640 supplemented with 2% FBS for 48 h. Subsequent measurements and calculations of wound closure were performed at 0, 24, and 48 h.

Transwell invasion assay

HGC-27 cells were seeded into a 24-well plate Transwell inserts precoated with Matrigel matrix (Corning, USA) and then treated with DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs at 5 ng/mL DTX for 24 h. After being washed with PBS twice, the cells on the Transwell insert membrane were fixed in 4% (v/v) paraformaldehyde solution and then stained with 0.1% (w/v) crystal violet. After incubation for 30 min at room temperature, the inserts were washed with PBS three times, and the adherent cells were imaged and counted.

Cell cycle arrest assay

To test the cell cycle distribution, HGC-27 cells (2 × 10⁵) were seeded in a 6-well plate and cultured for 24 h, followed by incubation with DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs at 2 ng/mL of DTX for 24 h. The cell cycle distribution was determined via Cell Cycle and Apoptosis Analysis Kit (Beyotime, Shanghai, China) following the manufacturer’s protocol.

CCK-8 assay

Cell viability was determined by CCK-8 assay following the manufacturer's protocol. In brief, HGC-27 cells were seeded and cultured at a density of 1 × 10⁴ cells per well with 100 μL of medium in 96-well plates (Corning, USA). The cells were subsequently treated with different concentrations of DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs at 5 or 10 ng/mL DTX. After 24 h or 48 h of incubation, 10 μL of CCK‐8 reagent was added to each well, and the cells were then cultured for 1 h. The absorbance of the suspension was measured at 450 nm via a microplate reader (Bio‐Rad, USA). Cell proliferation was calculated on the basis of the absorbance values.

Cell apoptosis assay

HGC-27 cells (6 × 10⁴ cells/well) were grown in RPMI-1640 medium supplemented with 10% FBS for 24 h in 6‐well plates. The cells were subsequently treated with DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs at 5 ng/mL DTX for 24 h. After treatment, the cells were harvested and washed with precooled PBS. The cells were resuspended in binding buffer and stained with Annexin V‐FITC and propidium iodide (Beyotime, Shanghai, China) at room temperature for 30 min in the dark. The percentage of apoptotic cells was analyzed via flow cytometry within 30 min.

Inhibition of tumor spheroid growth

After the tumor spheroids formed, DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs were cocultured with 2 ng/mL or 5 ng/mL DTX, respectively. The diameter of the tumor spheroids was measured daily by capturing images with an inverted microscope.

The processed tumor spheroids were sectioned into 10-μm-thick frozen sections for ki67 immunofluorescence (IF) staining to observe inside tumor spheroid growth. A rabbit monoclonal anti-ki67 (1:300 dilution; Proteintech Technology Inc., Wuhan, China) antibody was used for IF analysis.

Western blot analysis

HGC-27 cells were treated with DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs with 5 ng/mL DTX for 24 h. The cells were subsequently lysed and centrifuged, and the concentration of the supernatant was detected using a BCA kit (Beyotime, Shanghai, China). The expression of cleaved-caspase-3, Bcl-2, and TUBB3 was analyzed by Western blot. A mouse monoclonal anti-cleaved-caspase-3 antibody (1:1500 dilution, Affinity Biosciences, OH, NJ, USA), rabbit monoclonal anti-Bcl-2 antibody (1:1000 dilution, Affinity Biosciences, OH, NJ, USA), rabbit monoclonal anti-TUBB3 antibody (1:500 dilution, Affinity Biosciences, OH, NJ, USA), and mouse monoclonal anti-GAPDH antibody (1:10,000 dilution, Proteintech Technology Inc., Wuhan, China) were used for the Western blot analysis.

In vivo biodistribution of NPs

Xenograft BALB/c nude mouse models were established by injecting 100 μL of 3 × 10⁶ BGC-823 cells into the backs of the mice. Free Cy5, Cy5-labeled CA-Oxi-αCD NPs, and Cy5-labeled FA-CA-Oxi-αCD NPs were administered intravenously at a dose of 1 mg/kg Cy5 when the tumor volume reached approximately 200 mm³. The fluorescence intensity in the mice was measured via an in vivo imaging system (Perkin Elmer, Waltham, MA, USA) at 2, 4, 8, 12, 24, and 48 h post-injection. After 24 and 48 h of treatment, the mice were killed, and the tumors and major organs of the xenograft mice were harvested and imaged.

The tumor tissues were cryosectioned into 10-μm-thick sections after being fixed with 4% (v/v) paraformaldehyde and dehydrated with 30% (w/v) sucrose solution, and the cell nuclei were stained with DAPI. The NPs distribution in the tumor tissue was observed via CLSM.

In vivo antitumor efficacy

After the xenograft models were established as described above, the mice were randomly divided into four groups when the tumor volume reached 80 mm³. DTX, DTX/CA-Oxi-αCD NPs, or DTX/FA-CA-Oxi-αCD NPs were administered at a dose of 2 mg/kg DTX once every four days for a total of three injections. The control group received the same volume of normal saline. Tumor growth was evaluated by measuring 1/2 (length × width²) every 2 days. After 10 days of treatment, tumor-bearing mice were received T2 magnetic resonance imaging before the mice were killed, and the harvested tumor tissues were imaged and fixed in 4% paraformaldehyde for immunohistochemical analysis. The major organs, including the heart, liver, spleen, lungs, and kidneys, of the mice were harvested for hematoxylin–eosin (H&E) staining. A rabbit monoclonal anti-ki67 (1:300 dilution, Proteintech Technology Inc., Wuhan, China) antibody and a colorimetric TUNEL apoptosis assay kit (Beyotime, Shanghai, China) were used for tumor tissue immunohistochemical analysis.

Statistical analysis

The results are expressed as the mean ± standard deviation (SD) of at least three independent experiments. Statistical analysis was performed via GraphPad Prism 9 (GraphPad Software, Inc.). The data were analyzed via Student’s t test for two groups or one-way variance (ANOVA) for more than two groups. Statistical significance was determined at p < 0.05.

Results and discussion

Cellular uptake and tumor spheroid penetration

The physicochemical properties of DTX/CA-Oxi-αCD NPs and DTX/FA-CA-Oxi-αCD NPs are similar to our reported results (Wang et al. 2023). Dynamic light scattering (DLS) analysis revealed that the diameters of the DTX/CA-Oxi-αCD NPs and DTX/FA-CA-Oxi-αCD NPs were 214.0 ± 7.6 nm and 220.0 ± 5.1 nm, respectively (Table S1). With FA modification, the drug loading and encapsulation efficiency of NPs were not obviously changed (Table S1). In addition, these NPs exhibited a negative zeta potential (about − 30 mV), and the polymer dispersity index (PDI) of all NPs were less than 0.2 (Table S1 and Figure S1), indicating that these NPs are suitable for in vivo application. TEM images confirmed that these NPs have spherical morphology (Figure S1). Moreover, these NPs can smartly release their payloads under pH and/or ROS microenvironment (Wang et al. 2023).

To investigate whether FA-CA-Oxi-αCD NPs can be internalized by GC cells, Cy5-labeled FA-CA-Oxi-αCD NPs were coincubated with HGC-27 cells and imaged via CLSM. As shown in Fig. 2A, after coincubation of Cy5 or its nanoformulations with HGC-27 cells for 2 h, the fluorescence intensities of the Cy5-labeled CA-Oxi-αCD NPs and Cy5-labeled FA-CA-Oxi-αCD NPs groups were significantly greater than those of the free Cy5 group (Fig. 2A), indicating that both Cy5-labeled CA-Oxi-αCD NPs and Cy5-labeled FA-CA-Oxi-αCD NPs can be efficiently internalized by HGC-27 cells. As expected, the fluorescence intensity of the Cy5-labeled FA-CA-Oxi-αCD NPs group was significantly greater than that of the Cy5-labeled CA-Oxi-αCD NPs group, indicating that FA modification can enhance the cellular uptake of NPs by GC cells via FA-mediated endocytosis. Similar results were obtained after 6 h of incubation of the cells with the NPs. Interestingly, semiquantitative analysis indicated that the fluorescence intensity of the NPs in the cells did not obviously increase with prolonged incubation time (Fig. 2C), indicating that the cellular uptake of the NPs reached saturation after 2 h of incubation.

Fig. 2

Cellular uptake of Cy5-labeled NPs by HGC-27 cells and distribution of Cy5-labeled NPs in HGC-27 3D tumor spheroids. A Cellular uptake of Cy5-labeled NPs by HGC-27 cells after 2 or 6 h of treatment. CLSM images of HGC-27 cells incubated with Cy5 (1 μg/mL, red), Cy5-labeled CA-Oxi-αCD NPs or Cy5-labeled FA-CA-Oxi-αCD NPs (containing 1 μg/mL Cy5) for 2 or 6 h. Cell nuclei were stained with DAPI (blue). The scale bar represents 50 μm. B Distribution of Cy5-labeled NPs (red) in HGC-27 tumor spheroids. After 6 h of incubation, the tumor spheroids were observed in CLSM layer scanning mode. The cell nuclei were stained with DAPI (blue). The scale bar represents 200 μm. C Semiquantitative analysis of the corresponding Cy5 fluorescence intensity in A. D Distribution of Cy5-labeled NPs (red) in frozen tumor spheroid sections. Frozen tumor spheroid sections were observed via CLSM. The cell nuclei were stained with DAPI (blue). The scale bar represents 200 μm. E Semiquantitative analysis of the corresponding Cy5 fluorescence intensity in B. F Semiquantitative analysis of the corresponding Cy5 fluorescence intensity in D *, significantly different at p < 0.05; **, significantly different at p < 0.01; ***, significantly different at p < 0.001. ****, significantly different at p < 0.0001

Three-dimensional (3D) tumor spheroids are widely used to evaluate the antitumor efficacy of therapeutics because they exhibit heterogeneity in the tumor microenvironment (Busse et al. 2013). To further assess the ability of the NPs to target GC cells, we constructed 3D tumor spheroids from HGC-27 cells. As shown in Fig. 2B, compared with those in the Cy5 group, both Cy5-labeled CA-Oxi-αCD NPs and Cy5-labeled FA-CA-Oxi-αCD NPs obviously accumulated in 3D tumor spheroids after 6 h of incubation. Interestingly, the semiquantitative results suggested that the accumulation of Cy5-labeled FA-CA-Oxi-αCD NPs was greater than that of Cy5-labeled CA-Oxi-αCD NPs in the tumor spheroids (Fig. 2E), further confirming that FA modification can increase the ability of the NPs to target GC cells. To further investigate the penetration of the NPs into 3D tumor spheroids, the NPs-treated tumor spheroids were frozen and sliced for imaging. As shown in Fig. 2D and F, considerable red fluorescence was observed in the center of the sliced 3D tumor spheroids in the Cy5-labeled CA-Oxi-αCD NPs and Cy5-labeled FA-CA-Oxi-αCD NPs groups, indicating that both CA-Oxi-αCD NPs and FA-CA-Oxi-αCD NPs can reach the interior of the tumor. In addition, the fluorescence intensity and semiquantitative analysis of the sliced 3D tumor spheroids in the Cy5-labeled FA-CA-Oxi-αCD group was obviously greater than that in the other groups (Fig. 2D and F), which further demonstrated that FA-modified NPs can increase the ability to target GC. In conclusion, FA-modified CA-Oxi-αCD NPs can be efficiently internalized by HGC-27 cells and subsequently penetrate the center of 3D GC tumor spheroids, which may increase the efficacy of GC treatment.

Cell migration and invasion assessment

As shown in Fig. 3A, compared with the control and the DTX/CA-Oxi-αCD NPs, the DTX/FA-CA-Oxi-αCD NPs significantly inhibited tumor cell migration after 24 h of treatment, and this trend became more prominent as the treatment time increased to 48 h. The semiquantitative results also revealed that compared with the other treatments, the DTX/FA-CA-Oxi-αCD NPs obviously inhibited HGC-27 cell migration after 24 and 48 h (Fig. 3C). These results revealed that DTX/CA-Oxi-αCD NPs obviously inhibited HGC-27 cell migration, which may inhibit GC metastasis. In addition, compared with the other treatments, the FA-modified NPs displayed greater cell migration inhibition because of their enhanced ability to target HGC-27 cells. Transwell assay results revealed that fewer HGC-27 cells passed through the Transwell insert membrane after DTX/FA-CA-Oxi-αCD NPs treatment than after other treatments (Fig. 3B and D). These results further demonstrated that DTX/FA-CA-Oxi-αCD NPs may inhibit GC cell invasion and metastasis. In conclusion, FA-modified DTX/CA-Oxi-αCD NPs significantly inhibited HGC-27 cell migration via FA-mediated targeting, which may increase the efficacy of GC treatment via the inhibition of GC invasion and metastasis.

Fig. 3

Influence of DTX and its nanoformulations on the migration and invasion of HGC-27 cells. A Wound healing images of HGC-27 cells treated with DTX or DTX nanoformulations; the scale bar represents 200 μm. B Images of HGC-27 cells that crossed the Transwell membrane with DTX or the corresponding nanoformulations; the scale bar represents 400 μm. C Semiquantitative analysis of the migration rate shown in A. D Number of HGC-27 cells in random fields of view shown in B. *, significantly different at p < 0.05; **, significantly different at p < 0.01; ***, significantly different at p < 0.001

Cell cycle arrest

DTX has an antitumor effect by blocking the tumor cell cycle in the G2/M phase and promoting cell apoptosis (Nehme et al. 2001). Herein, flow cytometry analysis of the cell cycle was employed to evaluate the antitumor effects of DTX nanotherapeutics. As shown in Fig. 4A and B, compared with DTX alone, both DTX/CA-Oxi-αCD NPs and DTX/FA-CA-Oxi-αCD NPs significantly blocked HGC-27 cells in the G2/M phase. Importantly, the proportion of cells in the G2/M phase was greater in the DTX/FA-CA-Oxi-αCD NPs group than in the DTX/CA-Oxi-αCD NPs group (Fig. 4B). These results suggest that DTX nanotherapeutics may exert their antitumor effects by blocking the tumor cell cycle in the G2/M phase.

Fig. 4

Influence of DTX and its nanoformulations on cell cycle distribution of HGC-27 cells. A Flow cytometry images of the HGC-27 cell cycle after DTX or DTX nanoformulations treatment. B Proportions of HGC-27 cells in each cell cycle stage shown in A. C Western blot analysis of the TUBB3 protein in HGC-27 cells subjected to different treatments. D Semiquantitative analysis of TUBB3 protein expression in HGC-27 cells subjected to various treatments. *, significantly different at p < 0.05; **, significantly different at p < 0.01; ***, significantly different at p < 0.001

The TUBB3 protein, a key component of the cytoskeleton and spindle, plays a crucial role in the proper progression of the cell cycle (Xu et al. 2019). The TUBB3 protein is also the site of action associated with DTX chemotherapy. As shown in Fig. 4C and D, the protein expression of TUBB3 was strongly suppressed in the DTX/CA-Oxi-αCD NPs and DTX/FA-CA-Oxi-αCD NPs groups compared with the DTX group. Furthermore, DTX/FA-CA-Oxi-αCD NPs significantly decreased TUBB3 protein expression compared with DTX/CA-Oxi-αCD NPs (Fig. 4C and D), since FA modification enhanced the ability of the NPs to target tumor cells. In conclusion, DTX/FA-CA-Oxi-αCD NPs obviously block HGC-27 cells in the G2/M phase compared with free DTX, which may have better antitumor effects on GC cells than free DTX does.

In vitro antitumor efficacy

The cytotoxicity of DTX and its nanoformulations to HGC-27 cells was assessed via a CCK-8 assay. As shown in Figure S2A, the viability of HGC-27 cells was obviously lower in the DTX/FA-CA-Oxi-αCD NPs group than in the other groups (Figure S2A). As expected, the cell viability decreased with increasing drug concentration and prolonged incubation time (Figure S2A). Moreover, flow cytometry analysis was also used to detect the apoptosis of HGC-27 cells treated with DTX or various nanoformulations. As shown in Fig. 5A and B, the percentage of apoptosis cells in the nanotherapeutic groups was obviously greater than that in the DTX group, indicating that the antitumor efficacy of DTX was enhanced when it was loaded into the NPs. Importantly, DTX/FA-CA-Oxi-αCD NPs caused significant death in HGC-27 cells compared with the other groups (Fig. 5A and B), which further verified that FA modification can increase the antitumor efficacy of DTX nanotherapeutics through the overexpression of the folate receptor (FR) on the HGC-27 cell membrane.

Fig. 5

In vitro antitumor efficacy of DTX and its nanoformulations. A Flow cytometry analysis of HGC-27 cell apoptosis after treatment with DTX or its nanoformulations. (i) Control, (ii) DTX, (iii) DTX/CA-Oxi-αCD NPs, (iv) DTX/FA-CA-Oxi-αCD NPs. The same as in the following description. B Percentage of HGC-27 cell apoptosis shown in A. C Images of 3D tumor spheroids after incubation with DTX and its nanoformulations; the scale bar represents 200 μm. a, 2 ng/mL DTX; b, 5 ng/mL DTX. D Growth curves of tumor spheroids after treatment with DTX or its nanoformulations. a, 2 ng/mL DTX; b, 5 ng/mL DTX. E Western blot analysis of the Bcl-2 and cleaved-caspase-3 protein in HGC-27 cells subjected to different treatments. F Semiquantitative analysis of Bcl-2 and cleaved-caspase-3 protein expression in HGC-27 cells subjected to various treatments. *, significantly different at p < 0.05; **, significantly different at p < 0.01; ***, significantly different at p < 0.001; ****, significantly different at p < 0.0001

HGC-27 tumor spheroids were also used to evaluate the antitumor efficacy of DTX nanotherapeutics. As shown in Fig. 5C and D, with prolonged incubation time, HGC-27 tumor spheroids grew rapidly without treatment. However, the growth of HGC-27 tumor spheroids was obviously blocked on day 8 after treatment with 2 ng/mL or 5 ng/mL DTX (Fig. 5C and D). Interestingly, on day 8, the growth of HGC-27 tumor spheroids was further inhibited by DTX nanotherapeutic treatment, and compared with the other groups, the DTX/FA-CA-Oxi-αCD NPs group displayed the smallest tumor spheroid diameter (Fig. 5C and D). Ki67 is an antigen associated with cell proliferation that is closely related to mitosis and is indispensable for cell proliferation (Miller et al. 2018). Consequently, the expression of ki67 in HGC-27 tumor spheroids was used to evaluate tumor cell proliferation. As shown in Figure S2B, bright red ki67-positive fluorescence was observed in the outer layer of the tumor spheroids in the control group, indicating that the tumor cells strongly promoted cell proliferation. However, the fluorescence intensity clearly decreased in the DTX group compared with the control group (Figure S2B), indicating that ki67 expression decreased with DTX treatment. Importantly, the lowest fluorescence intensity was observed in the DTX/FA-CA-Oxi-αCD NPs group, indicating that the lowest ki67 expression was observed in HGC-27 tumor spheroids treated with DTX/FA-CA-Oxi-αCD NPs. These results further demonstrated that DTX/FA-CA-Oxi-αCD NPs obviously inhibited HGC-27 proliferation in 3D tumor spheroids. In conclusion, in vitro experiments demonstrated that DTX/FA-CA-Oxi-αCD NPs can decrease HGC-27 cell viability, inhibit HGC-27 cell proliferation and block HGC-27 tumor spheroid growth, which may be suitable for in vivo GC treatment.

To investigate antitumor mechanism of NPs, western blot assay concerning apoptosis protein was performed. As shown in Fig. 5E and F, the expression of Bcl-2 protein was extremely inhibited due to the treatment of various DTX nanoformulations compared with DTX group. However, the expression of cleaved-caspase-3 protein was dramatically increased after treated with various DTX nanoformulations (Fig. 5E and F). Significantly, the highest expression level of cleaved-caspase-3 was found in DTX/FA-CA-Oxi-αCD NPs group. The results demonstrated that DTX-loaded NPs significantly arrest cell cycle in G2/M phase. Previous studies have indicated that prolonged cell cycle arrest in the G2/M phase precipitates mitotic catastrophe and induce a cascade of downstream events, including apoptosis (Mc Gee 2015), which inhibit the anti-apoptotic protein expression and increase the apoptotic protein expression, in turn lead to the GC cells apoptosis.

In vivo biodistribution

Furthermore, HGC-27 cells are not suitable for in vivo experiments because of their long subcutaneous tumor formation time (> 17 days), high failure rate, and difficulty in achieving a uniform initial tumor volume. However, BGC-823 human GC cells, which resemble HGC-27 cells and have high FR expression (Deng et al. 2019), are more prone to forming subcutaneous tumors but are less effective at generating 3D tumor spheroids. Therefore, BGC-823 cells were used solely for the in vivo experiments.

To investigate the in vivo biodistribution of the NPs, Cy5, Cy5-labeled CA-Oxi-αCD NPs, and Cy5-labeled FA-CA-Oxi-αCD NPs were injected into nude mice bearing BGC-823 xenografts via the tail vein. As shown in Fig. 6A and C, little fluorescence intensity was detected in the free Cy5 group because of its rapid metabolism in the body. Cy5-labeled CA-Oxi-αCD NPs or Cy5-labeled FA-CA-Oxi-αCD NPs exhibited sustained fluorescence intensity in the body because of their prolonged circulation time with polyethylene glycol modification (Fig. 6A) (Zhang and Zhang 2020). The fluorescence intensity in the NPs group decreased with prolonged administration (Fig. 6C). Interestingly, the fluorescence intensity of the Cy5-labeled FA-CA-Oxi-αCD NPs group was greater than that of the Cy5-labeled CA-Oxi-αCD NPs group (Fig. 6C), indicating that FA modification can enhance the ability of the NPs to target GC tissues. The excised tumor images and semiquantitative analysis at 24 and 48 h post-injection also revealed that, compared with nontargeted NPs, Cy5-labeled FA-CA-Oxi-αCD NPs effectively accumulated in tumor tissues (Fig. 6B and D). Fluorescence imaging of frozen tumor sections also revealed that the Cy5-labeled FA-Oxi-αCD NPs group presented stronger fluorescence intensity in tumor tissues than the Cy5-labeled CA-Oxi-αCD NPs group did (Fig. 6E and F), which further verified that FA modification can enhance the ability of NPs to target tumor tissues in vivo. In conclusion, NPs, especially FA-CA-Oxi-αCD NPs, can effectively accumulate in BGC-823 tumor tissues as a result of the high FR expression in BGC-823 cells.

Fig. 6

In vivo biodistribution of free Cy5, Cy5-labeled CA-Oxi-αCD NPs and FA-CA-Oxi-αCD NPs in BGC-823 tumor-bearing mice. A In vivo images of mice 2, 4, 8, 12, 24 and 48 h after injection. The tumor area is within the red circle. B Ex vivo fluorescence image of the excised tumors and major tissues at 24 and 48 h post-injection. C Semiquantitative analysis of the fluorescence intensity of the tumor site 2, 4, 8, 12, 24 and 48 h after injection. D Semiquantitative analysis of fluorescence intensity in excised tumors at 24 and 48 h. E CLSM images of frozen tumor sections from mice after different Cy5-labeled NPs (red) treatments. Cell nuclei were stained with DAPI (blue). The scale bar represents 100 μm. F Semiquantitative analysis of CLSM images of tumor tissues from mice subjected to different treatments at 24 and 48 h. (i) Free Cy5, (ii) Cy5-labeled CA-Oxi-αCD NPs, (iii) Cy5-labeled FA-CA-Oxi-αCD NPs. *, significantly different at p < 0.05; **, significantly different at p < 0.01; ***, significantly different at p < 0.001

In vivo antitumor efficacy

As mentioned above, DTX/FA-CA-Oxi-αCD NPs significantly inhibited GC cell proliferation and blocked 3D tumor spheroid growth, which inspired us to investigate their in vivo antitumor efficacy. Tumor-bearing nude mice were administered DTX or its nanoformulations at a dose of 2 mg/kg DTX via the tail vein (Fig. 7A), whereas the control group received PBS as a placebo. As shown in Fig. 7B and C, tumors grew rapidly in the control group after 10 days of inoculation. DTX inhibited tumor growth to a certain extent because the dose decreased and the dosing frequency prolonged (Fig. 7B and C). Compared with that after DTX treatment, tumor growth was reduced after DTX/CA-Oxi-αCD NPs treatment, since DTX/CA-Oxi-αCD NPs can passively target tumor tissues and control the release of DTX in the presence of ROS and an acidic microenvironment. Furthermore, DTX/FA-CA-Oxi-αCD NPs displayed the best in vivo antitumor efficacy compared with the other treatments (Fig. 7B and C) because of their strong ability to target GC tissues. T2 magnetic resonance imaging (MRI) was also used to determine the tumor volume in vivo. As shown in Fig. 7B (b), the DTX/FA-CA-Oxi-αCD NPs group presented the smallest tumor size and high signal intensity, which may imply drug-induced edema and tumor necrosis.

Fig. 7

In vivo antitumor efficacy of DTX, DTX/CA-Oxi-αCD NPs, and DTX/FA-CA-Oxi-αCD NPs in BGC-823 tumor-bearing mice. A Administration time after tumor inoculation. The mice received 2 mg/kg DTX or 2 mg/kg DTX nanoformulations every four days. B Images of excised tumors treated with free DTX or different nanoformulations (a) and representative T2 MR images of tumors (b). C Tumor growth curve after intravenous injection of DTX or different nanoformulations, n = 4. D Immunohistochemistry assay for ki67 and TUNEL detection in tumor tissues; scale bar represents 100 μm. E Semiquantitative analysis of ki67 expression (a) and TUNEL-positive areas in tumor tissues (b). *, significantly different at p < 0.05; **, significantly different at p < 0.01; ***, significantly different at p < 0.001; ****, significantly different at p < 0.0001

To further investigate the efficacy of DTX nanotherapeutics for GC treatment, immunohistochemical staining and TUNEL assays were conducted on paraffin-embedded tumor sections. Immunohistochemical staining revealed minimal ki67-positive staining in the DTX/FA-CA-Oxi-αCD NPs group compared with the other groups (Fig. 7D and E), suggesting that DTX/FA-CA-Oxi-αCD NPs significantly inhibited BGC-823 cell proliferation in tumor tissues. In addition, TUNEL staining confirmed that more apoptotic cells were found in the DTX/FA-CA-Oxi-αCD NPs group than in the other groups (Fig. 7D and E), which further verified that DTX/FA-CA-Oxi-αCD NPs can block GC growth through the inhibition of tumor cell proliferation and the induction of tumor cell apoptosis. The reason may be that FA-modified NPs have a special affinity for the FR on BGC-823 cells, which increases the penetration depth of nanotherapeutics in tumor tissues. The accumulated nanotherapeutics can release DTX in response to ROS/pH changes in the microenvironment of GC. The long-term maintenance of high concentrations of DTX in tumor tissues causes tumor cell apoptosis, and tumor cell proliferation is inhibited.

Conclusion

To overcome the adverse effects and poor solubility of DTX, DTX-loaded and FA-modified pH/ROS dual-responsive NPs were employed for GC treatment. In vitro experiments verified that FA-modified NPs can be efficiently internalized by HGC-27 cells, decreasing HGC-27 cell viability and inducing cell apoptosis. In addition, compared with free DTX, DTX/FA-CA-Oxi-αCD NPs can penetrate 3D tumor spheroids more deeply and obviously inhibit spheroid growth. Moreover, DTX nanotherapeutics can effectively hinder tumor cell migration and decrease cell invasion. In vivo experiments confirmed that DTX/FA-CA-Oxi-α CD NPs effectively accumulate at the tumor site, leading to significant inhibition of tumor growth via inhibition of tumor cell proliferation and induction of tumor cell apoptosis. In conclusion, DTX-loaded and FA-modified pH/ROS dual-responsive NPs exhibited promising therapeutic outcomes for GC treatment.