Abstract
Oral drug delivery systems had natural potential for colorectal cancer drug therapy. While the drug delivery efficiency is severely hindered by the complex intestinal barriers, especially mucus and epithelium barriers, resulting in unsatisfactory therapeutic effects and limited clinical translation. In this work, a bioactive self-thermophoretic and gas dual-driven nanomotor is developed for colorectal cancer therapy through efficient intestinal mucus and epithelial barrier penetration. The nanomotor shows intestinal mucus barrier penetration and the paracellular pathway reversibly opening properties of intestinal epithelium barrier, increasing the delivery efficiency of cisplatin by 3.5 folds. Owing to the targeted delivery of cisplatin and the reduced side effects on normal intestinal tissues, the therapeutic efficiency of the nanomotor for colorectal cancer in vivo is as high as 98.6%. With autonomous and reversible intestinal barriers penetration property, the nanoplatform may innovate the current oral drug delivery.
Similar content being viewed by others
Introduction
Colorectal cancer (CRC) is a highly malignant digestive tract tumor, causing 1 million of incidence worldwide every year with a mortality rate of 33%1,2. Oral administration is commonly used for the clinical drug therapy of CRC, whereas the extreme gastric acid and enzymes in the gastrointestinal hinder the clinical translation of most oral drugs for CRC therapy3,4. With the development of nanotechnology, oral drug delivery systems (ODDS) are discovered to protect the oral drugs from the harsh environment of stomach5,6. However, the delivery efficiency of current ODDS is fundamentally hindered by the formidable gastrointestinal barriers, especially mucus and epithelial barriers, resulting in unsatisfactory therapeutic effects and limited clinical conversion7,8.
The intestinal mucus in colorectal tracts had dense and viscous structure, and fast clearance time, setting limited time-window for drug delivery of ODDS. Researchers constructed polyethylene glycol encapsulations9, mucus adhering patch10, and zwitterionic nanoparticles11 to promote the mucus crossing of nanoparticles. While these innovations still had inefficient mucus barrier penetration to improve the bioavailability of ODDS. Recently, nanomotors were widely applied in biomedical field, especially in targeted drug delivery12,13. Nanomotors can transform energy into motion, which have prominent motion performance in liquids14,15. Nanomotors are expected to breaking through the complex mucus barriers. For example, Zhang et al. reported self-propelled nanomotors had a 4.0-fold higher penetration of gastric mucosa than that of passive nanoparticles16. However, the single and limited functional activities of most nanomotors result in relatively low drug delivery and therapeutic efficiency due to the complexity of physiological barriers.
The gastrointestinal barrier system comprises multiple protective layers, with the epithelial lining serving as a critical secondary defense mechanism following the mucus layer. Intestinal epithelium had complex drug absorption pathways, such as transcytosis and paracellular diffusion, tremendously limiting the drug delivery of ODDS. Epithelial transcytosis had complex mechanism, prone to degradation of ODDS during the drug transport process. Interestingly, paracellular transport is a promising drug delivery pathway. Recently, several researchers explored chemical permeation enhancers to increase paracellular diffusion to intestinal epithelium and bioavailability in preclinical and clinical trials by weakening tight junctions of intestinal epithelium barrier17,18. While these strategies improve drug permeation efficiency by enhancing paracellular pathway diffusion. Long-term and non-specific opening of tight junctions between intestinal epithelial channels may lead to intestinal inflammation and even blood infection5. Therefore, the accurate and in-situ regulation of epithelial tight junction is the key for the clinical conversion for epithelial paracellular permeation enhancing strategy. To address these problems, we created an orally dual-driven nanomotor for colorectal cancer therapy through efficient mucus barrier penetration and reversible intestinal epithelial paracellular permeation.
In this work, the Janus nanomotor adopts mesoporous silica nanoparticle (MSN) as nanocarrier and polydopamine (PDA) nanoparticle loaded with nitric oxide (NO) donor BNN6 (abbreviated as PB) as propelling element in a single nanodevice (MSN/PB). After loading with chemotherapy drug cisplatin (CP), the Janus nanomotor is camouflaged with the cell well of Lactobacillus rhamnosus GG (CWL) to prepare bioactive self-propelled nanoplatform (CP@MSN/PB@CWL, abbreviated as CMPBC). After oral administration, the nanoplatform can localize to the intestinal site of CRC. This is because the glycan structure abundant in CWL can specifically recognize immunoglobulin A (IgA), which is overexpressed in the CRC - associated intestinal segment19,20. Under near infrared (NIR) laser irradiation, the dual-driven nanoplatform not only produces self-thermophoretic propulsion force, but also generates gas propulsion force through asymmetric generation of NO. The dual-driven force promotes nanoplatform to efficiently cross the intestinal mucus barrier. Meanwhile, the NO activates cGMP/PKG signaling pathway by inducing the expression of Ras homologous protein family member A (RhoA), Rho-associated protein kinase (ROCK) and myosin light chain (MLC) proteins. This leads to downregulation of the tight junction related proteins Claudin-1, Occludin, and Zona Occludens-1 (ZO-1). Therefore, the NO can reversibly and in situ open the intestinal epithelial paracellular pathway to benefit the epithelial penetration of the nanoplatform (Fig. 1). Furthermore, in a mouse model bearing CRC, we investigated the intestinal barrier penetration capacity and underlying mechanism of the nanomotor, the delivery efficiency of CP, and the in vivo therapeutic efficacy of the nanomotor. Thanks to its autonomous and reversible intestinal barrier penetration characteristic, the nanomotor has the potential to revolutionize the existing oral drug therapy of colorectal cancer.
Results
Characterization and motion analysis of CMPBC
The main preparation process of CMPBC was shown in Fig. 2a. Firstly, MSN and PDA with excellent drug loading capacity and biocompatibility were selected as the carrier and propelling element of the nanoplatform21. Then MSN was confined to the Pickering emulsion between paraffin and water-containing surfaces at the interface in order to achieve its partial functionalization22. It was found that MSN monolayer could be achieved when the mass ratio was 1: 100 (MSN/wax) by varying the molar ratios of two structural domains attached to paraffin spheres as shown in the scanning electron microscope image (SEM) (Supplementary Fig. 1). Given that the rigorous and successful preparation of bisphere structures relies on the molar ratios of MSN and PDA, the molar ratios were evaluated to determine a molar ratio of 1: 1 (MSN/PDA) for the Janus bisphere combination (Supplementary Fig. 2). The transmission electron microscopy (TEM) images confirmed that MSN and PDA exhibited uniform spherical morphology, both with a diameter of ~90 nm. PDA was successfully attached to the MSN side, forming a Janus MSN/PDA structure with a diameter of ~180 nm (Fig. 2b). Both elemental mapping images and energy dispersive X-ray (EDX) spectra demonstrated the successful preparation of the MSN/PDA bisphere structure (Fig. 2c and Supplementary Fig. 3). In addition, the characteristic peaks of Fourier transform infrared (FTIR) spectra verified the successful preparation of the MSN/PDA bisphere structure (Supplementary Fig. 4).
a A schematic of the synthetic procedure of CMPBC. b TEM images of PDA, MSN, MSN/PDA, and negative staining TEM images of CWL and CMPBC. The experiment was repeated independently with similar results. c Element mapping image of MSN/PDA. d Schematic diagram of the recognition of nanomotor to IgA-labeled magnetic beads. e The fluorescence intensity of magnetic beads following their incubation with various nanoparticles (n = 3 independent experiments). f NO release rate of different nanoparticles in simulated artificial colon fluid (n = 3 independent experiments). g Schematic diagram of the MPBC movement direction under NIR (808 nm) irradiation. Motion speed (h) and MSD (i) of different nanoparticles in mucin solution (n = 3 independent experiments). Data are presented in the form of mean values ± standard deviation (SD). Significance was assessed via one-way analysis of variance (ANOVA) with Tukey’s post hoc test. Source data are provided as a Source Data file.
The hydrodynamic diameters of different nanoparticles were measured using dynamic light scattering and were 103.0 ± 8.2 nm for PDA and 93.3 ± 14.1 nm for MSN, and the hydrodynamic diameters of MSN/PDA bisphere structure increased to about 179.13 ± 15.9 nm with a good polydispersity index (0.2 ± 0.1) (Supplementary Fig. 5 and Supplementary Fig.6). In addition, the zeta potential values provided further evidence for the successful construction of the nanoplatforms, with zeta values of −33.7 ± 0.6 mV and −19.8 ± 1.3 mV for the MSN and PDA monolayers, respectively. The negative surface charge of MSN/PDA decreased to −27.4 ± 1.3 mV (Supplementary Fig. 7). The mesoporous structures of MSN and PDA were then confirmed by nitrogen adsorption-desorption isotherms (Supplementary Fig. 8), and the specific surface areas of MSN and PDA were calculated to be 241.4 m2/g and 30.8 m2/g, with pore sizes of 5.8 nm and 19.6 nm, respectively (Supplementary Fig. 9). Therefore, their mesoporous structure laid the foundation for subsequent drug loading.
Notably, BNN6, as a photothermally activatable NO donor, was first synthesized via the replacement of two hydrogen atoms of p-phenylenediamine (BPA) with NO moieties, which was then characterized by 1H NMR spectra (Supplementary Fig. 10). Then it was easily co-assembled with PDA into nanoparticles (PDA-BNN6) using a simple π-π conjugation. Subsequently, the chemotherapeutic drug CP was loaded into MSN, and it could be demonstrated by the elemental mapping image that CP was mostly located on the MSN side (Supplementary Fig. 11). The drug loading capacity of CP and BNN6 in CMPBC was 27.5 ± 1.3% and 46.0 ± 0.9%, respectively.
IgA is overexpressed in the CRC intestinal segment (Supplementary Fig. 12). Due to the interaction of Lactobacillus with IgA in the intestinal microenvironment20, CWL was used to camouflage nanoplatforms for the chemotaxis of inflamed intestinal sites. CWL was encapsulated on CP@MSN/PB by liposome extrusion method. According to the TEM images in Fig. 2b, it was shown that CMPBC had a shell of ~20 nm after CWL coating. The particle size of CWL was 198.2 ± 2.6 nm, and the zeta potential of CWL was −9.7 ± 1.1 mV (Supplementary Fig. 5). Due to the CWL coating on the surface, the particle size of CMPBC increased significantly to 182.1 ± 9.6 nm, while the zeta potential of CMPBC increased to −9.4 ± 1.0 mV. In addition, thermogravimetric analysis verified the component content of CMPBC (Supplementary Fig. 13). In addition, glycan content analysis verified that the extraction and encapsulation processes did not prominently affect the glycan on CWL (Supplementary Fig. 14). Based the fluorescence colocalization of MSN/PB and CWL (Supplementary Fig. 15) and the negative staining TEM images CMPBC (Fig. 2b), CWL was successfully coated on the surface of nanomotors. Then CP@MSN/PB and CMPBC were placed in PBS and 10% FBS (fetal bovine serum) for 7 days, and the changes of particle size and potential were detected and recorded every other day. It was observed that the particle size and potential of the nanoparticles did not change significantly during the experiment, indicating the good stability of CP@MSN/PB and CMPBC (Supplementary Fig. 16 and Supplementary Fig. 17).
The outcomes of the adsorption experiments demonstrated that the adsorption of nanoparticles to magnetic beads following CWL coating outperformed that of other groups (Fig. 2d, e). Additionally, when the CWL-coated nanoparticles were pre-incubated with IgA, the adsorption effect vanished. Encapsulating with CWL to construct biological chemotactic nanoplatforms is conducive to the identification function of IgA and the realization of localization effects in the intestinal sites of CRC.
To delve into the protective function and detachment behavior of CWL on the nanomotor, CMPBC was dispersed in various physiological simulation fluids for 2 h (Supplementary Fig. 15). Most of the CWL were not detached from the nanoplatforms after 2 h in artificial gastric and artificial small intestinal fluids. This finding suggests that CWL has the ability to withstand gastric acid digestion. Then CWL gradually fell off after dispersed in artificial colonic fluids, which laid a foundation for subsequent across intestinal barrier. Following this, to evaluate the in vitro drug release behavior of nanomotor, CP@MSN/PB and CMPBC were each dispersed in various artificial solutions (Supplementary Fig. 18). The release rate of CMPBC in phosphate buffer solution (PBS) for 12 h was 26.3 ± 0.8%, which was markedly lower than that of CP@MSN/PB (45.0 ± 0.4%). This disparity can be attributed to the biomimetic membrane structure which kept the major drug within the nanomotor. Subsequently, the release of CP from CMPBC was assessed after incubation in artificial mimics. The drug release rate of CMPBC in artificial gastric fluid for 12 h was only 22.5 ± 0.5%, which was not significantly accelerated compared with PBS group. It showed that CMPBC has a significant protective effect on CP in a complex gastrointestinal environment. Next, the effect of NIR laser irradiation on drug release in vitro was evaluated. It could be observed that the drug release rate of CP@MSN/PB + NIR (1.5 W cm−2) within 12 h was 49.2 ± 0.4%. The relatively slow drug release also paves the way for nanomotor to transport most of the CP across mucus barrier.
The successful preparation and favorable properties of CMPBC prompted us to further explore its NO generation and autonomous motility properties under NIR laser irradiation. When PDA-BNN6, MSN/PB and MPBC were irradiated by an NIR laser, NO release could be detected up to 9.3 ± 1.0 μM, 8.9 ± 1.0 μM and 8.1 ± 0.2 μM, respectively (Fig. 2f). After turning off the laser, the NO concentration increased slowly over 10 min. Once exposed to the NIR laser again, the amount of NO released increased significantly. The results demonstrated the NIR laser-triggered NO release behavior of PDA-BNN6. In contrast, only a small amount of NO was released from pure BNN6 with or without NIR laser irradiation, suggesting that the NO release behavior could be modulated by NIR light after BNN6 was loaded onto PDA.
Second, we verified the motility performance of the nanoplatforms within the intestinal mucus. By combining the nanomotor with high temporal and spatial resolution NIR laser (Fig. 2g), CMPBC was able to absorb the NIR laser asymmetrically. This absorption led to the generation of a self-thermophoretic effect and a dual driving force sourced from gaseous NO. Representative tracking trajectories of PDA-BNN6, MSN/PDA, MSN/PB, and CMPBC in the mucin solution at different power of NIR laser are shown in Supplementary Fig. 19. When only NO served as the driving force, PDA-BNN6 (a single PDA sphere loaded with BNN6) exhibited only slow motion in the mucin solution. Its motion speed was measured at 2.2 ± 0.3 μm s−¹ (Fig. 2h, supplementary movie 1). Whereas the motion of MSN/PDA with thermophoretic driving double sphere structure was 3.1 ± 0.5 μm s−1 when only NIR laser irradiation produced self-thermophoretic as the driving force (Supplementary Movie 2). The motion of gas-thermophoretic dual-driving MSN/PB was significantly accelerated to 3.6 ± 0.3 μm s−1 under the simultaneous action of the dual driving force (Supplementary Movie 3). In contrast, the motion velocity of final formulation CMPBC after NIR laser irradiation was 3.9 ± 0.6 μm s−1 (Fig. 2i, Supplementary Movie 4), which was not significantly different from that of MSN/PB. As the laser power increased, the motion velocity and mean-square displacement (MSD) of the nanomotor showed a remarkable increase (Fig. 2i and Supplementary Fig. 20). This enhanced motion velocity enabled the nanomotor to autonomously penetrate the mucus barrier. Such an ability has the potential to substantially boost the drug delivery efficiency.
The intestinal mucous penetration performance in vitro of CMPBC
To validate the in vitro crossing performance of the nanoplatforms through the mucus barrier, the nanomotor loaded with fluorescein isothiocyanate (FITC) were injected into the left ventricle of the microfluidic chip23 through a matched syringe, and 2% mucin solution was slowly injected into the right ventricle (Fig. 3a). Subsequently, a high-time-resolved NIR laser was employed to conduct spot irradiation on the nanoplatforms. Confocal laser scanning microscopy (CLSM) was utilized to detect the travel distance of the nanoplatforms as they traversed the intestinal mucus barrier. As depicted in the CLSM image in Fig. 3b, PDA-BNN6 (PB), MSN/PDA, and MSN/PB were able to rapidly penetrate the mucus following NIR laser irradiation. This observation indicated that the combined effects of NIR laser irradiation and the release of NO could propel the nanomotor to penetrate the mucus barrier. When NO was used as the driving force alone, the movement distance of the PB group was only 272 ± 25 μm. After NIR laser irradiation of the asymmetric nanoplatforms, the MSN/PDA group could reach 575 ± 56 μm. When NIR irradiation combined with NO as the dual driving force, the MSN/PB group could reach 754 ± 23 μm (Fig. 3c). Subsequently, the ability of different nanoparticles to penetrate the mucus layer was quantified using the transwell system (Fig. 3d). A mixture of gel and mucin was used to simulate the mucus layer, and FITC-loaded nanoparticles were added from the apical chamber to detect the fluorescence intensity of nanoparticles in the basal chamber after a certain time. The results showed that pure MSN/PB nanoparticles exhibited minimal mucus penetration without any external force, while NIR irradiation combined with NO as a dual driving force could reach 90% in the MSN/PB group. This rate was 2.9 times higher compared to that of the PB + NIR group and 1.5 times higher than that of the MSN/PDA + NIR group, respectively (Fig. 3e).
a Schematic diagram of the microfluidic channel. b CLSM images of the motility of nanomotors in a two-outlet microfluid channel. c The corresponding movement distance (n = 3 independent experiments). d Schematic diagram of transwell assay composed of mucus and gels. e The ratio of nanoparticles penetrated into basal chamber after 5 min of incubation in apical chamber (n = 3 independent experiments). f Schematic diagram of nanomotor penetration intestinal mucus barrier. g CLSM fluorescent images nanoparticle penetration intestinal mucus barrier for 5 min. h MFI of red fluorescence in the dotted box in (g) (n = 3 independent experiments). Data are presented in the form of mean values ± SD. Significance was assessed via one-way ANOVA with Tukey’s post hoc test. Source data are provided as a Source Data file.
In an effort to observe the distribution of nanoparticles in the mucus layer and to simulate the mucus environment in vivo, we incubated the nanomotor with rat colorectum tissue. Subsequently, NIR laser irradiation was applied. Confocal laser scanning microscopy (CLSM) was utilized to determine the penetration depth of the nanoplatforms into the intestinal mucus barrier (Fig. 3f). As shown in the CLSM images presented in Fig. 3g, upon exposure to NIR laser irradiation at a power density of 1.5 W cm−², the MSN/PB nanomotor were capable of actively traversing the intestinal mucus barrier. This finding strongly suggested that the combination of NIR laser irradiation and the action of NO as a dual-drive mechanism was more conducive to the autonomous crossing the mucus barrier. To quantify the mucus permeability of the nanomotor, we calculated the fluorescence intensity of the nanoparticles in the sub-mucus layer, which was only 6.1 ± 1.1 for MSN/PB in the absence of NIR irradiation in the sub-mucus layer. In contrast, after 5 min of NIR laser irradiation, the fluorescence intensity of MSN/PB + NIR group was 61.3 ± 3.6, which was 1.8, 2.6, 10.0 times higher than that of MSN/PDA + NIR group, PB + NIR group and MSN/PB group, respectively. It indicated that MSN/PB + NIR could effectively cross the mucus barrier (Fig. 3h).
To demonstrate that the nanomotor crossed the above epithelial barrier due to the autonomous motion of the asymmetric nanoparticles, we examined the effect of NO on the viscosity changes of the mucus. The results showed that there was no statistically significant difference in the viscosity of mucus of the NO-producing nanoparticles in each group as compared to control (Supplementary Fig. 21). Therefore, the mucus penetration effect was mainly attributed to the accelerated movement speed of the nanoparticles.
The intestinal epithelial barrier permeability properties in vitro of CMPBC
In addition to the intestinal mucus mentioned above, the intestinal epithelial cell barrier also affects the delivery efficiency of ODDS. Therefore, we next evaluated the intestinal epithelial cell-spanning ability of the nanoplatforms. Caco-2 cells, which were widely used as an intestinal cell monolayer model, exhibited features similar to those of normal intestinal epithelium, including the expression of tight junctions (TJs), microvilli, and brush border enzymes24. TJs in the apical junction complex improve cell adhesion and act as channels in the paracellular space25. Therefore, Caco-2 cells were an ideal model for studying the intestinal barrier and drug transport across the intestinal epithelial barrier.
First, we utilized a hydrogel-derived artificial tumor model. Briefly, a channel gel system consisting of Caco-2 cell-loaded hydrogel and different nanoparticles loaded with FITC was constructed in the presence of an external field (Fig. 4a). Caco-2 cells were embedded in the hydrogel matrix to form an artificial intestinal epithelial tissue model. The CLSM images in Fig. 4b showed the movement process of nanomotor in the artificial intestinal epithelial model. The fluorescence signal was not observed for MSN/PB in the artificial intestinal epithelial model geometry in the absence of any external field, which indicated that nanoparticles alone could not penetrate the intestinal epithelial cell barrier. Subsequently, it was found that both the PB + NIR group with only NO as the driving force and the MSN/PDA group with only NIR laser irradiation as the driving force were able to migrate slowly along the channel toward the cell side and penetrate into the inner layer of the artificial intestinal epithelial cells, with travel distances of 284.3 ± 0.7 nm and 401.7 ± 1.6 nm, respectively. In contrast, the MSN/PB group, which was driven by NIR laser irradiation and gaseous NO as a dual driving force, had a longer penetration depth of about 637.1 ± 1.3 nm in the artificial intestinal epithelial model, which was 2.2- and 1.6- times longer than that of the PB + NIR group and the MSN/PDA + NIR group, respectively (Fig. 4c). These findings suggested that dual-drive significantly improved the penetration efficiency of nanoparticles in the artificial intestinal epithelial model. Due to PDA showed striking photothermal conversion ability under NIR light irradiation, the effect of temperature change to the integrity of right artificial intestinal epithelial model was examined. The maximum temperature on the nanoparticle side reached 35.4 °C under NIR laser irradiation for 5 min, while the maximum temperature on the cell side was only 30.0 °C, demonstrating that the temperature did not have a prominent effect on the cell of artificial intestinal epithelial model (Supplementary Fig. 22).
a Schematic representation of the movement of different FITC-loaded nanoparticles in artificial intestinal epithelial tissue containing hydrogel and Caco-2 cells under under NIR (808 nm) irradiation for 5 min. b Time-lapse fluorescence imaging (Hoechst-33342-labelled Caco-2 cells, blue, FITC-nanoparticles, green, 5 min). c The distance traveled by different nanoparticles through the artificial intestinal epithelium (n = 3 independent experiments). d Schematic diagram of the Transwell model of monolayer intestinal epithelium and the measurement of resistance values. e TEER resistance changes at 2 h intervals after 24 h incubation with different nanoparticles in the apical chamber (n = 3 independent experiments). f Schematic diagram of trafficking pathway experiment. After incubating with different endocytosis inhibitors for 1 h, the Transwell intestinal epithelial models were incubated with nanoparticles for 4 h. g Mean fluorescence intensity (MFI) measured in Caco-2 cells by CLSM after incubation with different nanoparticles (n = 3 independent experiments). h Fluorescence intensity of nanoparticles in the pathway paracellular pathway (n = 3 independent experiments). i Fluorescence images showing ZO-1 expression and j corresponding fluorescence intensity quantification in mouse colon organoids after incubation with different nanoparticles for 4 h (n = 3 independent experiments). k Fluorescence images of colonic organoids incubated with nanoparticles loaded with different Rhodamine B (RhB) for 4 h. l Signal plots of different nanoparticles (red) colocalized with the cytoskeleton (green). m The MFI of different nanoparticles inside the organoid (n = 3 independent experiments). n The basal chamber fluorescence intensity changes every 2 h after incubation with different nanoparticles in the apical chamber (n = 3 independent experiments). Data are presented in the form of mean values ± SD. Significance was assessed via one-way ANOVA with Tukey’s post hoc test. Source data are provided as a Source Data file.
The above results and previous studies revealed that NO, as a signaling molecule, had a role in specifically regulating the intestinal epithelial cell barrier in addition to acting as a driving force as described above. Therefore, we examined the release of NO in monolayer Caco-2 cells after the action of different nanoparticles, and found that after NIR irradiation, MSN/PB group and PB group could release NO, as revealed by CLSM images (Supplementary Fig. 23). To investigate whether the released NO enhances the transport of nanoparticles through tight junctions of intestinal epithelial cells and the effect on intestinal epithelial barrier function, we constructed a monolayer Caco-2 enterocyte transwell model and applied different nanoparticles to the tips of the Caco-2 monolayer. Within 24 h after particle incorporation, we measured the trans-epithelial electrical resistance (TEER) using a resistivity meter (Fig. 4d). The results showed that both the PB + NIR group and the MSN/PB + NIR group reduced the TEER values to some extent, and after 8 h, a gradual recovery of TEER values in the monolayer of enterocytes was observed, and the barrier function was restored within 24 h (Fig. 4e).
ZO-1 is an Occludin protein found in epithelial TJs, which is crucial for regulating paracellular permeability26. To observe the effect of nanoparticles on tight junction arrangement, ZO-1 immunofluorescence staining was performed on nanoparticle-treated Caco-2 cell monolayers. The CLSM images revealed that, after 4 h of different nanoparticle treatments, compared without the NO treated group, the staining of cellular ZO-1 in the treated group was reduced and discontinuous (Supplementary Fig. 24). These results suggested the increased penetration depth in the MSN/PB + NIR group compared with that of the MSN/PDA + NIR group. Since NO reduced translocation of ZO-1 at the intercellular contact site, it led to increased intestinal epithelial permeability and transient opening of tight junctions. In contrast, ZO-1 staining of cell interstitial spaces was again visible after 24 h of co-incubation of nanoparticles with Caco-2 cell monolayers, further confirming that NO induced a reversible increase in intestinal epithelial permeability without permanently disrupting the cell monolayer structure. Reversibly increasing intestinal epithelial penetration could safely and efficiently improve the delivery efficiency of ODDS.
NO could induce an increase in intestinal epithelial permeability, but whether nanoparticles were able to cross the intestinal epithelium via the paracellular pathway and the exact manner in which different nanoparticles cross the intestinal epithelium remain elusive. Subsequently, we verified the mechanism of nanoplatform crossing intestinal epithelium after NO action. Currently, the main ways for nanoparticles to cross the intestinal epithelial barrier were transcytosis as well as the paracellular pathway, where the transcytosis pathway consists of cytochalasin-mediated endocytosis (CME), clathrin-dependent endocytosis (CDE) and macrocytosis27. To investigate whether the released NO enhanced the nanoparticle translocation via the paracellular pathway, we used three typical inhibitors corresponding to the transcytosolic pathway, chlorpromazine (CPZ), M-β-cyclodextrin (M-β-CD), and EIPA, to act on a monolayer of Caco-2 cell model (Fig. 4f). The CLSM results showed that the fluorescence intensity of the nanoparticles in the cell layer significantly reduced after the co-action of the inhibitors. It indicated that the internalization of nanoparticles in cells reduced and the transcytosis process of nanoparticles in intestinal epithelial cells was inhibited (Fig. 4g, Supplementary Fig. 25). Subsequent analysis of the fluorescence intensity by transwell lower chamber showed that the fluorescence intensity of the PB + NIR group and the MSN/PB + NIR group, which were able to release NO, was significantly higher than that of the other two groups (Fig. 4h). The fluorescence intensity of the MSN/PB + NIR group being 3.5-fold higher than that of the MSN/PDA + NIR group, which proved NO was able to enhance the translocation of nanoparticle through the paracellular pathway.
To further visualize the passage location of nanoparticles in the cell layer, we analyzed the co-localization of different nanoparticles with the cytoskeleton in the Caco-2 monocytic cell layer. CLSM images and fluorescence co-localization curves demonstrated that for the MSN/PB as well as the MSN/PDA + NIR groups, only a small portion of the two co-localized, and most of the nanoparticles were intracellular (Supplementary Fig. 26). Those nanoparticles capable of releasing NO, the colocalization rate of PB + NIR group and MSN/PB + NIR group was significantly higher than that of the other two groups. These results further provided direct evidence that NO enhances the transport of nanoparticles through the paracellular pathway, thereby increasing the efficiency of crossing the intestinal epithelial barrier.
Nowadays, intestinal organoids have been developed as potential molecular tools for studying host-microbe interactions, nutrient uptake and drug screening. They share close structural and functional similarities with primitive organs in vivo, which makes it more authentic for us to utilize colonic organoids to study whether the nanomotor affect the intestinal epithelial barrier function28. The expression of tight junction closure protein ZO-1 was significantly reduced in the NO-releasing PB + NIR group and MSN/PB + NIR group after different nanoparticles were acted with intestinal-like organs (Fig. 4i, Supplementary Fig. 27). It is further demonstrated that NO has the ability to transiently open intestinal tight junctions and enhance intestinal epithelial permeability (Fig. 4j). Subsequently, we further analyzed the specific pathway of nanoparticles through the intestinal epithelial barrier. The colocalization of nanoparticles with the cytoskeleton during entry into organoids was observed by CLSM images (Fig. 4k, Supplementary Fig. 28). It can be seen that the strongest green fluorescence marked by the black box represented the tight junctions between cells. The corresponding red fluorescence was significantly higher in the MSN/PB + NIR group than in the MSN/PB and MSN/PDA + NIR groups (Fig. 4l). In addition, the analysis results of the red fluorescence intensity inside the organoid likewise proved that the intestinal epithelial spanning ability of the MSN/PB + NIR group was 1.3 and 3.6-fold higher than that of the MSN/PDA + NIR and PB + NIR groups, respectively (Fig. 4m). Consistent with the above results for Caco-2 monolayers, colocalization was much higher in the PB + NIR and MSN/PB + NIR groups than in the other two groups. This series of results demonstrated that NO was able to open intestinal tight junctions, thereby enhancing intestinal epithelial permeability and consequently nanoparticle translocation through the paracellular layer.
Finally, the trans-epithelial translocation efficiency of different nanoparticles was further tested by measuring the apparent permeability coefficient, defining as the crossing rate of micelles to the Caco-2 monolayer on the transwell membrane (Fig. 4d). Cell monolayers with a trans-epithelial electrical resistance (TEER) of more than 400 ω. cm2 were used in the experiments. As shown in Fig. 4n, MSN/PB + NIR was transported toward the basolateral side over time. The transport efficiency at 24 h was 1.5-and 2.3-fold higher than that of MSN/PDA + NIR and PB + NIR groups, respectively. The above results suggested that NO was more favorable for nanoparticles to autonomously penetrate the intestinal epithelial barrier by safely and reversibly enhancing the intestinal epithelial permeability.
Study on intestinal barrier penetration mechanism and anti-tumor ability of CMPBC in vitro
To further illustrate the specific mechanism by which NO reduces tight junction expression, we investigated the effect of MSN/PB + NIR on the transcriptional profile of Caco-2 cells. The volcano plots revealed differentially expressed genes in Caco-2 after co-incubation (Fig. 5a). In addition, the KEGG pathway analysis of DEGs in Caco-2 cells revealed enrichment of genes regulated by the cyclic guanosine monophosphate-protein kinase G (cGMP-PKG) signaling pathway and tight junctions (Fig. 5b). NO promoted the cGMP/PKG signaling pathway through the activation of guanylate cyclase (sGC). Moreover, the expression of Ras homolog protein family member A (RhoA), Rho-associated protein kinase (ROCK) and myosin light chain (MLC) was down-regulated in the pathway. This resulted in the downregulation of the expression of tight junction related proteins Claudin-1, Occludin and Zona Occludens-1 (ZO-1), which opened the epithelial barrier (Fig. 5c).
a Differential Gene Visualization Scatter Map and Volcano Map Display (FPKM). MSN/PB was applied to Caco-2 cells, irradiated under 808 nm laser for 5 min, and incubated for 4 h. b Column diagram of KEGG annotation of up-regulated and down-regulated differentially expressed genes. c Tightly connected signal path scheme. NO upregulates the cGMP/PKG signaling pathway by activating sCG. This step, in turn, inhibits RhoA/ROCK, resulting in downregulation of tight junction related proteins Claudin-1, Occludin, and ZO-1, thus opening the intestinal epithelial barrier. d, e Western blot and quantification of the expression levels of RhoA, ROCK, Claudin-1, Occludin and ZO-1 in Caco-2 cells incubated with different nanoparticles for different times (n = 3 independent experiments). 1, 4 h: control; 2, 4 h: MSN/PB; 3, 4 h: PB + NIR; 4, 4 h: MSN/PDA + NIR; 5, 4 h: MSN/PB + NIR; 6, 24 h: MSN/PB + NIR. f A diagram illustrating the mucus-producing transwell model, which is composed of Caco-2 and HT29-MTX cells. g CLSM images showing the CT26 cells in the basal chamber following a 4 h incubation period with various nanoparticles located in the apical chamber. h The MFI of CT26 cells in the basal chamber after being incubated with different nanoparticles in the apical chamber for 4 hours (n = 3 independent experiments). i The viability of CT26 cancer cells after being incubated with various nanoparticles for 24 hours (n = 3 independent experiments). j The proportion of dead CT26 cells after incubation with different nanoparticles for 24 h (5 μg mL−1, based on CP content; n = 3 independent experiments). k Fluorescence images of live/dead cell staining of CT26 cells stained with Calcein AM/propidium iodide (PI) after incubation with different nanoparticles for 24 hours. The experiment was repeated independently and yielded similar results. l Flow cytometric analysis of CT26 cells in the mucus-producing transwell model after 24 h of incubation with various nanoparticles. Data are presented in the form of mean values ± SD. Significance was assessed via one-way ANOVA with Tukey’s post hoc test. Source data are provided as a Source Data file.
We next assessed the expression of the associated proteins by Western blotting experiment. These NO-releasing PB + NIR and MSN/PB + NIR groups significantly up-regulated PKG and down-regulated the levels of RhoA, ROCK, Claudin-1, and Occludin. The levels of the corresponding proteins in the MSN/PB + NIR group were all restored to the normal level after 24 h of co-incubation of different nanoparticles with Caco-2 cells (Fig. 5d, e). This demonstrated that NO transiently and reversibly acted on the relevant pathways, thus affecting intestinal epithelial permeability.
Due to the favorable cellular uptake on monolayer cell models, the nanomotor was tested the crossing properties through the mucus-producing transwell assay. A mixture of Caco-2 and HT29-MTX cells was placed in the top chamber of the transwell plate to construct a transwell model for mucus secretion in vitro (Fig. 5f). Figure 5g showed that after NIR laser irradiation, MSN/PB could effectively cross the intestinal barrier that produces mucus in vitro and was internalized by CT26 cells in the basal compartment. The MFI of the MSN/PB + NIR group was 1.2-fold, 2.5-fold, and 2.9-fold higher than that of the MSN/PDA + NIR group, the PB + NIR group, and the MSN/PB group, respectively (Fig. 5h). The results indicated that MSN/PB + NIR enhanced the mucus permeation and endocytosis of transwell model. Moreover, flow cytometry analysis also verified that a large number of nanoparticles were internalized by CT26 cells in the basal compartment in the MSN/PB + NIR group (Supplementary Fig. 29).
Having validated the outstanding in vitro intestinal epithelial barrier penetration capabilities of CMPBC, our subsequent focus shifted to exploring the in vitro anticancer properties of the drug loaded nanoplatforms. For nanoparticles to exhibit effective anticancer performance, ensuring good biosafety is an essential prerequisite. Thus, we initiated the research by first assessing the in vitro biosafety of the nanomotor. When Caco-2 and CT26 cells were co-cultured with different nanoparticles for 24 h, the cell viability of Caco-2 and CT26 cells still exceeded 90% even when the concentration of the nanoplatform reaches 200 μg mL−1 and the NIR laser irradiation power reached 1.5 W cm−2. This indicated that the nanoplatform has good biosafety in vitro (Supplementary Fig. 30, Supplementary Fig. 31). Afterward, the drug-loading nanomotor was co-incubated with CT26 cells (Fig. 5i). When the drug concentration was set at 5 μg mL−1, the cell viability of CT26 cells in the CP@MSN/PB + NIR group was 46.6 ± 0.7%. This indicated that the nanoparticles exerted the most significant cytotoxic effect at this condition. At the drug concentration of 20 μg mL−1, the cell viability of CP@MSN/PB + NIR is as low as 21.5 ± 1.8%, which was substantially lower compared to that of the free CP group and the CP@MSN/PB group without NIR irradiation. Following this, the transwell experiment was employed to evaluate the anticancer activity of the nanoparticles after they penetrated the in vitro intestinal barrier. After adding the drug loaded nanoparticles to the upper chamber of the transwell and incubating for 24 hours, the Calcein acetoxymethyl ester/Propidium Iodide (Calcein AM/PI) staining method was utilized to measure the death rate of CT26 cells in the basal chamber. After penetrating the in vitro mucus-producing intestinal barrier, CP@MSN/PB + NIR exhibited a 74.4 ± 2.6% of killing effect on cancer cells, much higher than that of other groups (Fig. 5j, k). Also, flow cytometric analysis confirmed a 74.1% of necrosis and apoptosis rate in CP@MSN/PB + NIR group, dramatically higher than that of free CP group (30.6%, Fig. 5l). Such an increased necrosis and apoptosis rate of CP@MSN/PB + NIR validated the improved anticancer activity of drug-loading nanoplatform after crossing the in vitro mucus-producing transwell barrier.
In vivo biodistribution and intestinal barrier crossing ability of CMPBC
Leveraging their exceptional ability to cross transwell barriers, the targeting efficiency of nanomotor to intestinal segments associated CRC was assessed through in vivo distribution experiments (Fig. 6a). After oral administration of different nanoparticles to the mice, a small animal imaging system captured fluorescence (Fig. 6b). MSN/PDA-BNN6@CWL (MPBC) showed stronger and more persistent fluorescence at tumor sites. The results of fluorescence quantitative analysis of colorectal cancer intestinal segments showed that due to the encapsulation of CWL, MPBC showed significant accumulation of CRC-related intestinal segments at 6 h after oral administration, which was 5.7-fold higher than that of MSN/PB (Fig. 6c). However, after preincubation with IgA, the fluorescence of MPBC in CRC intestinal segments decreased sharply, suggesting that its biological-chemotactic ability was dependent on IgA recognition by MPBC. In addition, the distribution of MPBC in normal organs was not significant compared to the other groups (Fig. 6d). In addition, inductively coupled plasma mass spectrometry (ICP-MS) showed 2.4-fold higher nanoparticle concentration at the tumor site in the MPBC group than in the MSN/PB group at 24 h (Fig. 6e). The improved drug delivery efficiency of the bioactive nanomotor on unresectable primary tumors and human colorectal cancer model were also conducted (Supplementary Fig. 32 and Supplementary Fig. 33). The nanoparticle concentration had 1.4-fold and 1.3-fold increase after NIR irradiation in unresectable primary tumors and human colorectal cancer model, respectively. MPBC tended to accumulate at CRC sites after oral administration, while also reducing accumulation to major tissues such as liver and spleen, which may reduce the side effects of chemotherapy on normal tissues and organs (Fig. 6f). The superior ability of CWL camouflage strategy to identify CRC-associated intestines laid a solid foundation for the efficient penetration of the CRC intestinal barrier by the dual-drive nanoplatform.
a Schematic illustration of nanoparticle adsorption to the CRC-intestinal segment. b Representative fluorescence images of digestive tracts and major tissues at various time points post-oral administration of different nanoparticles. The white circles idenote the CRC location. S means stomach; Li means liver; H means Heart; Sp means spleen; Lu means lung; K means kidney. c Quantitative analysis of fluorescence intensity in the CRC intestinal segment of mice at different time points after oral administration (n = 3 independent experiments). d Quantitative analysis of fluorescence intensity in major tissues 24 h post-oral administration (n = 3 independent experiments). e Analysis of the Si element at the CRC sites of mice at different time points after oral administration (n = 3 independent experiments). f Analysis of the Si element in major tissues 24 hours after oral administration (n = 3 independent experiments). g TEM images showing the distribution of different nanoparticles in the intestinal loop and villi after 5 min of intestinal circulation. h Average number of nanoparticles in TEM images of intestinal loop and villi (n = 3 independent experiments). i Fluorescence images displaying changes in ZO-1 protein expression (j) and nanoparticle fluorescence intensity (k) in mouse colon tissue after 5-min incubation with different nanoparticles (n = 3 independent experiments). l Changes in the width of tight junctions in mouse colon tissues after 5-min incubation with different nanoparticles (n = 3 independent experiments). m Fluorescence intensity changes in the basal chamber after different nanoparticles were incubated with Caco-2 cells and then co-incubated with GFP-Escherichia coli 1917 in the apical chamber for 2 h (n = 3 independent experiments). Data are presented in the form of mean values ± SD. Significance was assessed using one-way ANOVA with Tukey’s post hoc test. Source data are provided as a Source Data file.
We further evaluated the multiple intestinal barrier crossing capability of the nanoplatform in orthotopic CRC mice8. We continued to investigate the tumor penetration of the nanomotor using an in situ intestinal loop model29. After 5 min of intestinal circulation, the intestinal tube at the tumor site was taken and examined by transmission electron microscope. The mean number of nanoparticles reaching intestinal epithelial cells in the MSN/PB + NIR group was 1.5-, 5.0- and 8.3-fold higher than that in the MSN/PDA + NIR, PB + NIR and MSN/PB groups, respectively (Fig. 6g, h). Furthermore, we examined the crossing ability of the nanoplatforms to the intestinal epithelial barrier in vivo. In order to observe the effect of nanoparticles on tight junction arrangement, 5 min after intestinal circulation, the intestinal tubes from the tumor site were taken and examined by confocal microscopy and transmission electron microscopy. CLSM images revealed that after 5 min of treatment with different nanoparticles, the staining of ZO-1 in the NO-treated group was discontinuous compared with the NO-treated group (Fig. 6i, j). The red fluorescence intensity of the MSN/PB group was 1.3-fold higher than that of the MSN/PDA group after NIR laser irradiation for 5 min (Fig. 6k). This demonstrated the increased efficiency of nanoparticles to penetrate the intestinal epithelial barrier after a brief opening of tight junctions. The results of TEM showed that PB + NIR group and MSN/PB + NIR group, which could release NO, could increase the gap of tight junctions (Supplementary Fig. 34). The proportion of nanoparticles passing through tight junctions was significantly increased, compared with MSN/PB and MSN/PDA + NIR group, which could not produce NO (Fig. 6l). These results suggested that the strategy of NO opening the intestinal epithelium used in this work reduced the transfer of ZO-1 at the intercellular contact site. Then the intestinal epithelial permeability increased and the tight junctions were transiently opened, thereby increasing the fraction of nanoparticles passing through the paracellular pathway. It could safely and efficiently improve the intestinal barrier crossing ability of nanoparticles in vivo.
Subsequently, we examined whether bacteria in the gut would internalize and cross the intestinal barrier after NO opened it. CLSM results showed that the fluorescence intensity of NO-releasing PB + NIR and MSN/PB + NIR groups was not significantly different from that of the control group (Supplementary Fig. 35). This indicated that the NO opening intestinal epithelium strategy used in this work did not cause obvious bacterial internalization (Fig. 6m). Then we examined the integrity and function of the epithelial barrier after the action of NO producing nanoparticles by treating with FITC-dextran 4 kDa. The fluorescence results showed that at the time point of 4 h, luminal translocation of FITC-dextran occurred in PB + NIR group and MSN/PB + NIR group, which could release NO, and the blood glucose content increased significantly. These results indicated that NO induced an increase in intestinal permeability. Then, considerable traces of FITC-dextran were not observed after 24 h of treatment, and the glucose content in the blood did not change significantly (Supplementary Fig. 36). This indicated that the intestinal barrier function has been significantly restored and the increase in intestinal permeability induced by NO is reversible and relatively safe.
Given the above targeting properties of MPBC, we performed hemodynamic clearance experiments to evaluate the pharmacokinetic properties of the nanomotor. As exhibited in Supplementary Fig. 37, and Supplementary Table 1, the peak concentration (Cmax) of MPBC + NIR was 1.6 and 1.3 times higher than that of MPBC and MSN/PB + NIR, respectively. The half-life of MPBC + NIR in plasma was 1.3 and 1.2 times higher than that of MPBC and MSN/PB + NIR, respectively. In addition, the bioavailability of MPBC + NIR was also increased by 1.6 and 1.3-fold compared with MPBC and MSN/PB + NIR, respectively.
For the subsequent anti-cancer in vivo, we investigated the content of chemotherapeutic drug at the site of the CRC sites. Inductively coupled plasma mass spectrometry (ICP-MS) showed that the CP content of CMPBC + NIR was 1.4, 2.2 and 3.5-fold that of CP@MSN/PB + NIR, CP@MSN/PDA + NIR and CP, respectively (Supplementary Fig. 38). These results indicated that MPBC could significantly penetrate the intestinal barrier and quickly reach the CRC-related intestinal segments after NIR laser irradiation, and significantly increase the content of chemotherapeutic drugs in CRC intestinal segments. This paved the way for improving drug delivery efficiency in subsequent anticancer experiments in vivo.
Subsequently, the biosafety of the nanomotor in vivo was evaluated by a 3-week gavage experiment. The results in Supplementary Fig. 39 indicated that the changes in mouse blood biochemical indexes and hematological parameters were negligible in mice treated with MPBC and MPBC + NIR compared to the control group. Hematoxylin-eosin (H&E) staining of the main tissues was performed to observe the physiological changes of the tissues, which did not cause obvious tissue damage (Supplementary Fig. 40). These results demonstrated that the nanomotor had unconspicuous toxicity. The good biosafety of this nanoplatform laid a solid foundation for oral targeted therapy of colorectal cancer.
In vivo tumor inhibition experiments
The in vivo antitumor experimental protocol was performed as the schedule showed in Fig. 7a. Thirteen days in advance, an orthotopic colorectal tumor-bearing mouse model was construct by injecting CT26-luc cancer cells into the cecum of mice. Tumor-bearing mice were orally administered the drug-loaded nanomotor twice a week for 3 weeks starting from day 1. The CP group had a significantly decreased body weight in the treatment period (Fig. 7b). Symptoms of body weight loss in the chemotherapy period was also occurred in the literature30. While the targeted delivery of CMPBC and CMPBC + NIR reduced the weight loss in mice.
a Schedule of the in vivo anticancer experiment. b Body weight change of mice during the treatment period (n = 5 mice). c In vivo bioluminescence images of mice at specified time points. d In vivo bioluminescence curves of mice during the treatment period (n = 5 mice). e Tumor weight in different groups following 3 weeks of treatment (n = 5 mice). f H&E staining and C-caspase (Cleaved-caspase) 3 staining analysis of tumor after treatment. g Incidence of tumor metastasis across different groups after treatment. h Survival analysis of tumor-bearing mice treated with saline, CP, and CMPBC + NIR over a 3-week period (n = 5 mice). Data are presented in the form of mean values ± SD. Significance was assessed using one-way ANOVA with Tukey’s post hoc test. Source data are provided as a Source Data file.
The tumor growth of mice was monitored by bioluminescence imaging technology. The bioluminescence images in Fig. 7c showed that CP and CP@MSN/PB failed to delay tumor growth compared with saline. CP@MSN/PDA + NIR, CP@MSN/PB + NIR and CMPBC had a certain in vivo anticancer effect (Fig. 7d), while CMPBC + NIR had the best anti-tumor effect, with a significant difference compared with other groups. In addition, after 3 weeks of treatment, the anatomical photographs of the tumors obtained in each group showed that the mice in the CMPBC + NIR group had the smallest tumor burden and one mouse had tumor disappearance (Supplementary Fig. 41). In addition, the tumor weight of mice in the CMPBC + NIR group was significantly lighter than that in the other groups after 3 weeks of treatment (Fig. 7e). The tumor inhibition rate of CMPBC + NIR group was 98.6%, which confirmed that the dual-driven nanoplatform improved the treatment efficiency against CRC and had a good anticancer effect.
The anticancer effect of the nanoplatform in vivo was examined by H&E staining and immunohistochemistry of cleaved caspase-3 (c-caspase-3) (Fig. 7f). H&E staining results showed that CMPBC combined with NIR laser irradiation could cause a large number of cancer cell death, proving that CMPBC + NIR had a significant anti-tumor effect. The images of c-caspase-3 immunohistochemistry showed that CMPBC + NIR had the highest cancer cell apoptosis-inducing activity. The therapeutic potential of the bioactive nanomotor on unresectable primary tumors and human colorectal cancer models were also evaluated (Supplementary Fig. 42 and Supplementary Fig. 43). CMPBC + NIR also showed prominent anticancer activity on unresectable primary tumors and human colorectal cancer models.
The orthotopic CRC mouse model in this study can spontaneously form distant metastasis31. After treatment, major organs and the entire gut of the mice in each group were examined for metastasis by means of bioluminescence imaging (Supplementary Fig. 44). The bioluminescence image showed that distant metastases in the saline group occurred mostly in the liver and other parts of the intestine. The metastatic sites in the CP and CP@MSN/PB groups were similar with control (Fig. 7g). In addition, pathological images confirmed liver and colorectal metastases in each group (Supplementary Fig. 45). While the mice had no distant metastases of CRC in CMPBC and CMPBC + NIR groups, indicating the effect of anti-tumor metastasis of the nanoparticles.
Also, gut microbiota is closely correlated with CRC32. To investigate the CRC-therapeutic effect on the impact of gut microbiota, 16S ribosomal DNA sequencing technology was used to evaluate the change of gut bacterial composition (Supplementary Fig. 46, Supplementary Fig. 47). The β-diversity index showed that the gut microbiota in the CMPBC + NIR group was similar to that of normal mice (Supplementary Fig. 48), indicating that the nanoplatform had better in vivo therapeutic effect. In conclusion, the nanoplatform not only exhibited excellent in vivo therapeutic effect on CRC, but also rescued the disordered gut microbiota caused by CRC.
Furthermore, we conducted an in vivo therapeutic assay with the nanomotor for 3 weeks and observed for 2 months to explore the effect of nanomotor on the long-term survival and the tumor recurrence. The tumor bioluminescence image was shown in Supplementary Fig. 49. At the end of the experiment, all mice in the saline and CP groups died within 60 days. In stark contrast, CMPBC + NIR significantly delayed orthotropic CRC tumor growth. Furthermore, CMPBC + NIR prolonged the survival of CRC-bearing mice (Fig. 7h). At Day 60 CMPBC + NIR group had a survival rate of 60%. Therefore, this nanomotor can extend the survival time of CRC-bearing mice and reduce tumor recurrence, indicating the good tumor inhibition activity in vivo of the nanoplatform.
Discussion
ODDS had natural merits for the drug therapy of CRC. However, the physiology of the gastrointestinal tract prevents its implementation for decades. Recently, PEG and zwitterionic modification of nanoparticles was discovered to enhance the crossing of the formidable intestinal mucus barrier. PEG modified nanoparticles may have accelerated Brownian motion in mucus. He group reported the PEG modified nanoparticles had a ~ 2% penetration efficiency on intestinal mucus barrier33. Veider et al. constructed zwitterionic modified nanoparticles to overcome the mucus barrier of the intestinal mucosa with a ~ 4% penetration efficiency after 1 h of incubation with mucus layer34. Unfortunately, the clearance rate of intestinal mucus is as high as 30 ~ 300 μm/s, setting a limited time-window for oral drug delivery35. The oral nanoparticles may be cleared by intestinal mucus during long detention in mucus layer. Thus, successful oral drug delivery for CRC therapy requires rapid permeabilization of the intestinal mucus barrier. In sharp contrast, the bioactive dual-driven nanomotor in this work proposed a concept of enhancing mucus penetration through autonomous movement. The nanoplatform could improve the mucus penetration efficiency by 8.3-fold only within 5 min. With self-propelled ability, the nanoplatform improved the therapeutic efficiency of CRC and reduced the off-target risk of chemotherapy drugs.
Except for mucus barrier, intestinal epithelium tremendously limits the drug delivery of ODDS. The main rout pathways for drug absorption through the colon epithelium are passive transcellular and paracellular diffusion. Current methods for cellular diffusion are mainly receptor and transcytosis-mediated endocytosis. However, the modification of receptors and transporters may lead to bacterial phagocytosis ODDS due to the presence of similar receptors on the surface of some native intestinal bacteria. This may lead to an imbalance of gut microbiota and reduce delivery efficiency9,36. In addition, the mechanism of cell transcytosis is complex and ODDS are prone to degraded by intestinal epithelium, resulting in toxic and side effects of chemotherapy. For paracellular channels, several previous studies explored chemical permeation enhancers to increase paracellular diffusion to intestinal epithelium and bioavailability in preclinical and clinical trials17,18. Those permeation enhancers reduce the tight junctions between intestinal epithelial cells to enhance the permeability of intestinal epithelium. While there are large numbers of bacteria in the colon microenvironment. Long-term and non-specific opening of tight junctions between intestinal epithelial channels may lead to intestinal inflammation and even blood infection5. Therefore, the current permeation enhancers may be not suitable for crossing the intestinal epithelial barrier in the colorectum. Encouragingly, the dual-driven nanomotor described in this work real-time, in-situ and reversibly opened the epithelial tight junction at the intestinal segment site of CRC with superior biosafety. The in-situ generated NO by the nanomotors activated cGMP/PKG signaling pathway to downregulate the tight junction related proteins such as Claudin-1, Occludin, and ZO-1. Therefore, the nanomotors can reversibly and in situ open the intestinal epithelial paracellular pathway. Therefore, the ODDS had a significant 3.5-fold increase drug delivery efficiency by crossing the paracellular pathway of epithelium.
The physiology of the colorectum prevents the clinical translation of most ODDS for CRC therapy. Successful ODDS for CRC therapy require permeabilization of the viscous mucus and complex intestinal epithelial barrier in colorectum. For effective permeation of physiology in the colorectum, an efficient therapeutic strategy was developed using real-time, in-situ and reversibly opening the intestinal barrier in the intestinal segment of CRC. We anticipate that this strategy will make a valuable contribution for clinical therapy of CRC and numerous gastrointestinal diseases. In non-human primates, more work is needed to evaluate the translatability between species and the potential for clinical research in the future. The application prospect of the nanomotor to different types of solid CRC tumors also need further to be assessed.
Methods
This research complies with all relevant ethical regulations. The animal experimental operations were in accordance with the Guidelines for Care and Use of Laboratory Animals of Zhengzhou University, and all experimental procedures and protocols were approved by Life Sciences Ethical Review Committee of School of Pharmaceutical Sciences, Zhengzhou University (approval number: yxyllsc20240124, Zhengzhou, China).
Materials
Pluronic F-127 (F127), 1, 3, 5-trimethylbenzene (TMB), Dopamine hydrochloride, trithanolamine (TEA), Hexadecyltrimethylammonium bromide (CTAB), Sodium salicylate (NaSal), tetraethyl orthosilicate (TEOS), and fluorescein isothiocyanate (FITC) were obtained from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Mucin from porcine gastrointestinal tract were obtained from Yuanye Bio-Technology Co., Ltd (Shanghai, China). Loading buffer and Cell Counting Kit-8 (CCK-8) was purchased from Solarbio Technology Co., Ltd. (Beijing, China). Calcein/PI Live/Dead Viability Assay Kit was obtained from Beyotime (Shanghai, China). Goat Anti-rabbit IgA/HRP, ZO-1 Polyclonal antibody, Claudin-1 Polyclonal antibody, Occludin Polyclonal antibody and ROCK Polyclonal antibody were obtained from Proteintech Group, Inc. (Chicago, USA). IgA Polyclonal antibody was obtained from Absin Bioscience Inc. (Shanghai, China). Azoxymethane was purchased from Sigma-Aldrich Merck KGaA. (Darmstadt, Germany). from Dextran sulfate sodium salt (MW: 36000 ~ 50000) was purchased from Yisheng Biotechnology Co., Ltd. (Shanghai, China).
Instruments
TEM images and EDS analysis were studied with a JEM-ARM300F field emission transmission electron microscope. Dynamic light scattering and zeta potential were studied with a Malvern Zetasizer Instrument (Mastersizer 3000, Britain). Nanoparticle movement and cell uptake were studied by confocal laser scanning microscope (CLSM, Leica SP-8, Germany). Preparation of the preparations was carried out using a collector constant temperature magnetic stirrer (DF-101S, China) and a centrifuge (Thermo Scientific Multifuge X1R, USA). NO release, magnetic bead adsorption and cell viability were measured using a microplate reader (Synergy™ H1, Bio Tek, USA). Cell uptake and apoptosis were measured using a flow cytometer (BD LSRFortessa™ X−20, USA). Changes in protein expression were detected using an electrophoretic apparatus (DYCZ-24DN, China) and a transmembrane apparatus (DYCZ-TRANS2, China). Drug loading and drug release were determined using high performance liquid chromatography (HPLC, 1260 Infinity II, Agilent, USA). In vivo drug content was determined using inductively coupled plasma mass spectrometry (ICP-MS, Thermo Fisher, USA). The in vivo bioluminescence of mice was detected by small animal in vivo multimodal imaging analyzer (Lumina XRMS Series III, America).
Cells and animals
Caco-2 cells (human colon carcinoma cell line), and CT26 cells (murine colon carcinoma cell line) was purchased from the cell bank of Chinese Academy of Science (Shanghai, China). Mucus-producing HT29-MTX cells was obtained from Jinyuan Biotechnology Co., Ltd. (Shanghai, China). CT26 cell transfected with luciferase (CT26-luc) and HT29 cells (human colon carcinoma cell lines) transfected with luciferase (HT29-luc) were obtained from Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China). The absence of mycoplasma contamination was routinely tested, and the cells were authenticated by Short Tandem Repeat test. Caco-2, CT26 cells, and CT26-luc cells were maintained at RPMI−1640 supplemented with 10% fetal bovine serum (FBS, Gibico, USA). HT29-luc cells and HT29-MTX cells were maintained at McCOY’s 5 A medium supplemented with 10% FBS. Male Balb/c mice (6 to 8 weeks, 18 to 20 g) were obtained from the experimental animal center of Henan province (Zhengzhou, China). All animals were grown at the condition of 25.0 ± 2.0°C and 55% humidity with 12 h light/dark cycle. The mice adapted the environment with 10 days before experiment. The mice were closely monitored daily and euthanized if a mouse with more than 20% body weight loss. The maximal tumor burden in the experiments is less than 10% body weight of mice, which was with approval from the Life Sciences Ethical Review Committee of School of Pharmaceutical Sciences, Zhengzhou University. During the experiments, the tumor burden of mice did not exceed 10% body weight.
Preparation of mesoporous polydopamine loaded N, N’-Di-sec-butyl-p-phenylenediamine (PDA-BNN6)
First, mesoporous polydopamine (PDA) nanoparticles were prepared according to previous literature with minor modification21. Briefly, 0.6 g of F127 and 0.64 mL of TMB were first dissolved in a mixture of H2O (20 mL) and ethanol (20 mL), and 1.5 mL ammonia was added. After stirring well, 120 mg of dopamine hydrochloride was added to the mixture. The reaction mixture was stirred at room temperature for 12 h, and then the product nanoparticles were separated by centrifugation. Finally, the samples were treated with ultrasound (three times for 30 min each) in a mixture of ethanol and acetone (2: 1, v/v) to remove the template.
BNN6 was synthesized according to previous literate with some modifications37. In details, N, N’-di-sec-butyl-p-phenylenediamine (2.34 mL, 10 mmol) was diluted to 18 mL with ethanol. Under N2 atmosphere, degassed sodium nitrite solution (6 M, 20 mL) was added and then stirred for 30 min. Then degassed hydrochloric acid aqueous solution (6 M, 20 mL) was introduced dropwise with a constant pressure separatory funnel. After stirring for 4 h, the reaction was over, and the precipitation was collected through centrifugation. Then the precipitation was washed with ethanol aqueous solution (50%, 100 mL) to remove the excess reactants.
PDA nanoparticles (2 mg) and BNN6 (2 mg) were suspended in 4 mL dimethyl sulfoxide (DMSO), respectively. In constant agitation, BNN6 solution was added to the PDA solution and the mixture was stirred at room temperature for 12 h. Afterward, PDA-BNN6 nanoparticles was collected by centrifugation and washed with water for 3 times.
Preparation of mesoporous silica nanoparticles/PDA-BNN6 (MSN/PB)
The MSN was synthesized according to previous literate with slight modifications38. First, 0.136 g of triethanolamine (TEA) was added to 50 mL of water. The resulting mixture was placed in an oil bath maintained at 80 °C and stirred for 30 min. Subsequently, 760 mg of cetyltrimethylammonium bromide (CTAB) and 80 mg of sodium salicylate (NaSal) were introduced into the solution, and the stirring was continued for 1 h. Next, while stirring, a mixture containing 8 mL of tetraethyl orthosilicate (TEOS) was added to the solution, and the reaction was allowed to proceed for 12 h. After the reaction was complete, the MSN were collected by centrifugation and thoroughly washed several times with ethanol. To remove the template, the MSN were then extracted three times with a hydrochloric acid - methanol solution at 60 °C for 6 hours each time. Finally, the purified MSN were dried under vacuum at room temperature overnight.
Janus MSN/PB was synthesized following a previously reported method22. MSN (40 mg) was evenly dispersed by ultrasound in an aqueous solution (6.7% ethanol, 20 mL), and CTAB was added (416 μL, 1 μM). The temperature of the solution was gradually elevated to 75 °C, after which 2 g of paraffin wax was incorporated. Once the paraffin had completely melted, we used a DLAB D−160 (manufactured by DLAB SCIENTIFIC CO., LTD., China) to homogenize the mixture at 12000 g for 10 minutes to ensure the MSN were in a monodisperse state. Subsequently, the reaction mixture underwent magnetic stirring at 75 °C for an additional hour to generate a Pickering emulsion. After the emulsion was formed, the mixture was cooled down to room temperature, diluted with 20 mL of methanol, and 400 μL of (3-mercaptopropyl) trimethoxysilane was added. The reaction was then magnetically stirred for 3 h. The resulting solid was separated by centrifugation and washed several times with methanol. Next, the partially thiol-functionalized MSN nanoparticles were dispersed into a solution of PB nanoparticles in a 10 mM Tris-HCl buffer with a pH of 8.5. This reaction system was stirred overnight at room temperature. Finally, the solid was filtered and washed twice with chloroform to eliminate the remaining paraffin, thus obtaining Janus MSN/PB.
Characterization of particle size and zeta potential
The zeta potential and particle size of the nanoparticles were tested using a Malvern Zetasizer Instrument.
Drug loading of MSN/PB
CP (10 mg) was dissolved in DMF (2 mL), and MSN/PB (10 mg) was added. After stirring in the dark for 12 h, the mixture was centrifuged at 10000 g for 10 min to get the precipitate. The precipitate was CP loading MSN/PB (CP@MSN/PB). The amount of free CP present in the supernatant was quantified using high performance liquid chromatography. The loading content and encapsulation efficiency was calculated by formula 1 and formula 239.
CWL camouflage of CP@MSN/PB
The procedure was conducted according to our previous method9. Lactobacillus rhamnosus (American Type Culture Collection, 53103) was cultured in Modified MRS Medium Base. Begin by taking 2 mL of Lactobacillus rhamnosus culture medium and centrifuging it at 4000 g. Post-centrifugation, wash the resulting sediment with PBS of pH 7.4. Then, introduce 200 μL of a lysozyme solution, which has a concentration of 1 mg mL−¹ in a tris buffer with a pH of 8.0, into the washed sediment. Shake this mixture at 37 °C for 30 minutes. After the shaking process, centrifuge the decomposed bacteria at 4000 g by employing an ultrafiltration centrifuge tube (100 kDa, from Corning). Wash the product obtained from this centrifugation step with PBS (pH 7.4) to isolate the CWL. Next, blend CP@MSN/PB with the isolated CWL. Use an mini extruder (manufactured by Avanti Polar Lipids) to sequentially pass this mixture through polycarbonate porous membranes, repeating each pass five times. Finally, centrifuge the CWL-coated nanomotor using an ultrafiltration centrifuge tube at 4000 g for 10 minutes. This will yield CMPBC. Then, a certain amount of powder of nanoparticles were resuspend in PBS to calculate the concentration when using.
Detection of glycan content
First, 20 mg of glucan was weighed into a 500 mL volume bottle, and water was added to the scale. 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6 and 1.8 mL were absorbed respectively, and each was supplemented with distilled water to 2.0 mL. Then, phenol (6%, 2.0 mL) and concentrated sulfuric acid (10.0 mL) were added to the mixture, shaken and left for 30 min. The standard curve of absorbance measured at 490 nm was obtained. Then LGG, CWL and MPBC were taken 1.0 mL respectively, 1.0 mL distilled water was added, then 1.0 mL 6% phenol was added, and 5.0 mL concentrated sulfuric acid was added quickly. The absorbance was determined at 490 nm after being shaken on a vortex tester, fully mixed, and left for 30 min. The measured absorbance was put into the standard curve to obtain the concentration of polysaccharide.
Investigation of bacterial bionic membrane coating and shedding
MSN/PB@CWL (MPBC) nanoparticles were double-dyed, then 4 mg MSN/PB and 4 mg FITC were ultrasonically dispersed in 4 mL ultra-pure water for 12 h overnight, centrifuged at 10000 g for 10 min, and washed in ultra-pure water for 3 times to obtain MSN/PB@FITC. 1 mg mL−1 bacterial bionic membrane (CWL) and cell membrane fluorescent probe (DiI) were incubated in a metal shock bath for 0.5−1 h, centrifuged at 6000 g for 10 min, and washed in ultra-pure water for 3 times to obtain CWL@DiI. The stained MSN/PB and bacterial bionic membrane were coated with liposome extrusion method, and the double-dyed nano preparation was obtained, and incubated in PBS, artificial gastric juice, artificial intestinal juice, and artificial colon liquid for 2 h, respectively. After 10 microliter drops were placed on the slide, and the fluorescence of nanomotor was detected by CLSM.
Analysis of the movement behavior of nanoMOTOR
The PDA-BNN6, MSN/PDA, MSN/PB and MPBC were dispersed in mucin solution (10 μg mL−1, in terms of PDA) and placed in a glass slide. After that, the trajectory of the nanomotor was traced under a bright field condition by an inverted microscope fitted with a 100× oil objective lens. The mean - squared displacement (MSD) was determined based on formula 39.
Magnetic beads adsorption experiments
After fluorescence labeling, MSN/PB and MPBC were mixed with magnetic beads (Yeasen Biotechnology Co. Ltd., Shanghai, China) labeled with IgA for 30 min. Then the magnets were used to precipitated the magnetic beads. After washing three times by PBS (pH 7.4), the magnetic beads were detected the fluorescence intensity by fluorescence spectrophotometer.
Drug release experiments
CP@MSN/PB, CMPBC were separately suspended in PBS (pH 7.4), and CMPBC were suspended in PBS (pH 7.4) and different artificial fluid. Artificial gastric fluid is an acidic fluid with a pH of ~2. Artificial colonic fluid is with a pH of ~8 and with colonic microflora. At set times, the mixture was centrifuged at 120000 g for a duration of 10 min to take the supernatant, and then measured by HPLC.
NO release experiments
BNN6, PDA-BNN6, and MSN/PB were dispersed in artificially simulated colon fluid, irradiated with a NIR laser at 10 min intervals over 45 min, and released NO was detected using a commercial Griess detection kit (China Beyotime). Then the absorbance was measured at 540 nm. The NaNO2 was used as standard solution; the absorbance of solution without Griess reagent was deducted as background. To test the in vivo NO release, the colorectum of mice after treatment was cut into small pieces, After tissue homogenate and centrifugation at 10000 g at 4 °C for 10 min, NO was detected as previously described40, The content of NO2- in the supernatant was determined using the Griess detection kit.
In vitro biosafety experiments
Caco-2 cells and CT26 cells were seeded in 96 well plates (5000 cells per well). Ater incubation for 12 h, the wells were added with different concentrations of MSN/PDA, MSN/PB, MPBC were added. After the wells were continued to incubate for 24 h, the medium was added with of CCK-8 (10 μL) and incubate for another 2 h. Then microplate reader was applied to examine the absorbance at 450 nm.
In vitro intestinal mucous penetration experiment
The motion of nanoparticles in microfluidic chip was recorded by inverted microscope and 10× objective lens. PDA-BNN6, MSN/PDA, and MSN/PB loaded with FITC and dispersed in ultra-pure water (50 μg mL−1, calculated based on the amount of PDA) were injected into the left ventricle of the microfluidic chip with matching syringes, and 2% mucin solution was injected into the right ventricle. It was then irradiated with an 808 nm laser to record the distance travelled by the nanoparticles in 5 min.
In vivo intestinal mucous penetration experiment
PDA-BNN6, MSN/PDA and MSN/PB were loaded with Rhodamine B (RhB). Sprague Dawley rats, weighing between 220-280 g, were humanely sacrificed following established ethical protocols. Immediately after sacrifice, the colorectal tissues were carefully excised. Subsequently a solution containing RhB-labeled PDA-BNN6, MSN/PDA and MSN/PB (50 μg mL−1, calculated based on the amount of PDA) were added to colorectum for a period 5 min. After the incubation time elapsed, Fluorescein Isothiocyanate (FITC)-labeled wheat germ agglutinin was utilized to selectively stain the intestinal mucus present within the colorectal samples. This staining step was crucial for visualizing the mucus layer. Then the fluorescence of the colorectum was detected by CLSM after fixing colorectum on the glass slide.
Transepithelial transport experiments
Caco-2 and HT29-MTX cells were combined at a ratio of 7:3. This cell mixture was then carefully seeded into the apical chamber of a transwell system equipped with polycarbonate membranes (24 - well format, pore size 0.4 μm; sourced from Corning). The mixed cells were incubated until the transepithelial electrical resistance surpassed 400 ohms cm−2. Subsequently, CT26 cells were seeded into the basal chamber of the transwell plate. After seeding, the entire transwell setup was further incubated for an additional 12 h period to allow the CT26 cells to adhere and interact with the established cell layer in the apical chamber. Then PDA-BNN6, MSN/PDA and MSN/PB were loaded with RhB. RhB loaded PDA-BNN6, MSN/PDA and MSN/PB (50 μg mL−1, in terms of PDA) in 10% FBS RAPI 1640 medium were added to the apical chamber of a transwell plate and allowed to incubate for 4 h.
Following nanoparticles addition, transwell-cultured cells were exposed to NIR irradiation for 5 min. Post-irradiation processing involved three PBS (pH 7.4) washing cycles to remove extracellular nanomotor. Cellular uptake dynamics were subsequently evaluated through dual-modal analysis: initial qualitative visualization was performed via confocal laser scanning microscopy (CLSM, Nikon A1R), followed by quantitative assessment using a BD LSRFortessa flow cytometer (BD Biosciences, San Jose, CA).
In vitro intestinal barrier penetration experiment
The artificial intestinal epithelium was constructed by mixing 3.0 × 106 caco-2 cells with 1 mL 1% HA solution according to the method described in the literature41, and the artificial intestinal epithelium was prepared with Hoechst 33342 incubated in darkness at 37 °C for 1 h and washed the cells twice with PBS. A 1 cm cellulose tube was cut, washed with PBS, and 1 µL 1% HA solution was added to the center of the tube with a syringe. Then, 3.5 μL artificial intestinal epithelial tissue and PDA-BNN6, MSN/PDA and MSN/PB loaded with FITC dispersed in mucin (50 μg mL-1, calculated by the amount of PDA) were added at both ends of the test tube. Subsequently, the penetration depth of different groups of nanoparticles in the artificial intestinal epithelium was determined after exposure to an 808 nm laser (intensity: 1.5 W cm-2) for a duration of 5 minutes.
Epithelial resistance (TEER) and epithelial permeability measurement
Caco-2 cells were inoculated with a polycarbonate membrane (24 wells, 0.4 μm; Corning), when the transepithelial resistance surpassed 400 ohm cm−2, a single intestinal epithelial barrier was formed. Then, FITC-loaded PDA-BNN6, MSN/PDA and MSN/PB were added to the top chamber and incubated for 24 h. After the nanoparticles were added, they were irradiated by a laser at 808 nm (1.5 W cm−2) for 5 min. Subsequently, the resistance value and fluorescence intensity of the basement side chamber were measured every 2 h with a resistance meter and a fluorescence spectrophotometer. The epithelial permeability was expressed as the percentage of leakage of nanoparticles from the apex chamber to the basal lateral chamber.
The ratio of nanoparticles through the paracellular pathway
Caco-2 cells were inoculated with a polycarbonate membrane (24 wells, 0.4 μm; Corning). When the transepithelial resistance surpassed 400 ohm cm−2, a single intestinal epithelial barrier was formed. Inhibition strategies of the endocytosis mechanism were studied through pharmacology. Caco-2 cells were incubated with three major inhibitors (CPZ, M-β-CD, EIPA) for 1 h, then the medium was removed and PDA-BNN6, MSN/PDA, and MSN/PB loaded with RhB were added (50 μg mL−1, in terms of PDA). After incubation for 4 h, fluorescence intensity in basal lateral chamber and apical chamber was detected by fluorescence spectrophotometer. The uptake of nanoparticles in the cell layer was detected by CLSM.
Mechanism of paracellular pathway
MSN/PB was applied to Caco-2 cells, irradiated under NIR laser for 5 min, and incubated for 4 h. Total RNA was extracted from the cell-treated tissues using TRIzol® reagents. To ensure the purity of the RNA sample, genomic DNA was removed using DNase I (purchased from TaKara). To precisely assess the quantity and quality of the extracted RNA, two distinct techniques were utilized. The 2100 Bioanalyzer (manufactured by Agilent) was first used to evaluate the integrity and concentration of the RNA. Additionally, the ND-2000 (NanoDrop technology) was applied to accurately measure the RNA mass. Subsequently, library preparation and Illumina Hiseq sequencing were performed. To pinpoint DEGs (differential expression genes) between the two different samples, the expression level for each gene was determined by employing the fragments per kilobase of exon per million mapped reads (FPKM) approach. For the differential expression analysis, the R statistical package edgeR (Empirical analysis of Digital Gene Expression in R) was utilized. The selection criteria for DEGs between the two samples were as follows: the logarithm of fold change should be greater than 2, and the false discovery rate (FDR) was required to be less than 0.05. Subsequently, to gain insights into the functions of these differentially expressed genes, KEGG pathway analysis was performed using Goatools.
Cell viability assay
CT26 cells were plated in 96-well plates. Then the cells were incubated for 12 h. Various concentrations of CP and drug-loading nanomotor (5 μg mL−1, as the amount of CP) were introduced into the 96-well plates and continued to incubate for 24 h. Following this, the cells were rinsed with PBS (pH 7.4) and then added to medium containing 10 μL of CCK-8 for 2 hours. Thereafter, the absorbance at 450 nm was measured for each well using a microplate reader.
Transepithelial transport anticancer experiments
After the construction of the in vitro transwell model, CP, CP@MSN/PDA, and CP@MSN/PB (5 μg mL−1, based on CP content) were introduced into the apical chamber and incubated for 4 h. Post-incubation, the CT26 cells were stained using an annexin V-FITC apoptosis detection kit. Subsequently, the cells were analyzed via flow cytometry using a BD LSRFortessa instrument.
In vivo biosafety experiments
A total of 12 mice were randomly assigned to four experimental groups (n = 3 per group). The mice were administered intragastrically with saline (control) or MPBC (6 mg kg−1) twice weekly for a duration of 3 weeks. Following 3 weeks of treatment, the mice were euthanized, and blood samples were collected for serum biochemical and hematological analysis. Additionally, tumor tissues and major organs were collected for subsequent H&E staining.
Construction of orthotopic CRC mice model
The method was as previously described methods with some adjustments8. After disinfecting the abdominal area with an alcohol swab, a midline incision was made in the lower abdomen to expose the cecum. Meanwhile, CT26-luc cells were resuspended to a concentration of 4 × 106 cells per milliliter by combining Matrigel (10 μg μL−1, Corning) and RAPI 1640 medium in a 3:1 volume ratio. Subsequently, the cell suspension was administered into the cecal wall using a 30-gauge needle. Following this, the cecum was placed back into the abdominal cavity, and the incision was sutured closed with a 5 - 0 absorbable suture.
In vivo intestinal barrier penetration assay
PDA-BNN6, MSN/PDA, and MSN/PB were labeled with RhB. Following the inoculation of orthotopic tumor cells for two weeks, the mice were anesthetized with sodium pentobarbital. Subsequently, the intestinal loop of mice was exposed. After that, a latex tube was inserted into the upper part of the intestine. Then it was connected to a peristaltic pump. The pump was used to circulate a PBS (pH 7.4) solution containing RhB-labeled PDA-BNN6, MSN/PDA, and MSN/PB (50 μg mL−1, based on PDA content). The other end of the latex tube was inserted into the lower part of the intestine to form a closed circuit. Five minutes later, the intestines at the tumor site were removed and then analyzed.
In vivo intestinal permeability experiment
FITC-glucan (FD-4) permeability assay was used to evaluate intestinal permeability in mice. After 12 and 24 h of intragastric administration of PDA-BNN6, MSN/PDA, and MSN/PB, the mice were given 0.2 mL FD-4 (60 mg/100 g) orally. The whole blood was collected dynamically 4 h later and serum was separated at room temperature. The serum was diluted with equal volumes of PBS and the test was repeated. Serum FD-4 levels were measured by fluorescence spectrophotometer and colonic site fluorescence was measured by CLSM.
In vivo biodistribution experiments
The mice bearing orthotopic tumors were given Cy5.5-labeled MSN/PDA, MSN/PB, MPBC, and MPBC + IgA via intragastric administration at a dose of 1 mg kg−1 (based on Cy5.5 content). At predetermined time points, the mice were intraperitoneally injected with D-luciferin (150 mg kg−1, J&K Chemical, Ltd., China). Following injection, the mice were anesthetized and subjected to imaging using a small animal imaging system (IVIS Spectrum, PerkinElmer, USA). Post-imaging, the mice were sacrificed, and their gastrointestinal tracts were exposed for analysis. Subsequently, the concentration of Si in various tissues was measured using inductively coupled plasma-mass spectrometry (ICP-MS) after the tissues were dissolved in aqua regia.
Blood clearance kinetic experiments
The mice were administered intragastrically with Cy5.5-labeled MSN/PDA, MSN/PB, and MPBC at a dose of 1 mg kg−1 (based on Cy5.5 content) 14 days after the inoculation of orthotopic tumor cells. Two hours post-intragastric administration, the NIR group received NIR laser irradiation. To quantify the fluorescence of nanoparticles in the plasma, 200 µL of blood was collected from the orbital vein at predetermined time points.
In vivo antitumor experiments
On the fourteenth day following the inoculation of orthotopic tumor cells, the mice were intraperitoneally injected with D-luciferin to assess the tumor’s bioluminescence intensity. Mice exhibiting a bioluminescence intensity of approximately 1 × 106 photons (p) s−1 cm−2 sr−1 were selected for the following experiments. Subsequently, the mice were randomly assigned to seven groups (5 mice per group) and administered intragastrically with saline (control), CP, CP@MSN/PDA, CP@MSN/PB, and CMPBC (3 mg kg−1, based on CP content) twice weekly for a period of 21 days. Twelve hours post-intragastric administration, the mice in the CP@MSN/PDA + NIR group, CP@MSN/PB + NIR group, and CMPBC + NIR group were exposed to NIR laser irradiation for 5 minutes. Moreover, the mice in each group were intraperitoneally administered D-luciferin. Following the experimental treatments, the mice were first anesthetized. Then the mice were captured bioluminescence images with a small animal imaging system to observe tumor bioluminescence.
After a 3-week treatment period elapsed, the mice in each experimental group were humanely euthanized in accordance with ethical guidelines. Immediately after euthanasia, samples of blood, tumor tissue, and major organs such as the liver, kidneys, and spleen were carefully collected. The collected blood samples were then transferred to centrifuge tubes and subjected to centrifugation at a relative centrifugal force of 2500 g for a duration of 5 minutes. This centrifugation step separated the serum from the blood cells, and the resulting serum was subsequently utilized for comprehensive chemical analysis. In addition to the above-mentioned sample collection, mesenteric draining lymph nodes were also carefully dissected from the mice. These lymph nodes were then imaged using the same bioluminescent imaging technique. For the evaluation of bone marrow cellularity, femur samples were collected and processed for Hematoxylin and Eosin (H&E) staining. The stained femur sections were then digitized, and the resulting images were analyzed using ImageJ software. Finally, to comprehensively evaluate the antitumor efficacy of the nanoplatform, both the tumor tissues and lung samples were subjected to H&E staining and c-caspase-3 (a key marker of apoptosis) immunohistochemistry.
Long-term survival experiments
Fourteen days after the inoculation of cancer cells, 15 mice were randomly divided into three distinct groups, with each group consisting of 5 mice. Starting from day 1 and continuing until day 21, the groups were administered different substances via intragastric route. The control group received normal saline, while the second group was given CP (3 mg kg−1). The third group was administered CMPBC at an equivalent dosage of 3 mg kg−¹, calculated based on the CP content within CMPBC. This intragastric administration was carried out twice a week. Upon the completion of the treatment period, the mice in all groups were closely monitored for an additional month. By tracking the survival status over this period, we could assess the overall effectiveness of the treatments in prolonging the lives of the tumor-bearing mice. Moreover, to further assess the tumor status within the mice, all groups of mice were intraperitoneally injected with D-luciferin. After the injection, the mice were anesthetized to ensure their immobility during the imaging process. Once anesthetized, they were placed in a small animal imaging system. This system was used to track the tumor bioluminescence. The bioluminescence signals emitted by the tumors, which were activated by the D-luciferin, allowed us to monitor the growth, location, and potential spread of the tumors over time.
On day 60, the mice were humanely euthanized following established ethical procedures. Immediately after euthanasia, their tumors and major tissues, such as the liver, lungs, and kidneys, were carefully collected. These collected samples were then subjected to bioluminescent imaging. This final imaging step was performed to comprehensively check for any signs of tumor recurrence in the collected tissues, providing a more in-depth understanding of the long-term impact of the treatments on tumor growth and recurrence.
Statistics and reproducibility
All of the experiments were repeated at least three times independently with similar results. No data were excluded from the analyses. Statistical analysis between groups was analyzed by GraphPad Prism 6.0 software with one-way or two-way analysis of variance (ANOVA) followed by post hoc Tukey’s tests and 95% confidence intervals for categorical variables. n.s. means not significant.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
All data are available within the Article, Supplementary Information. The next-generation sequencing data produced in this study have been submitted and can be accessed in the BioProject database under accession code SRX23911018, and SRX23911019. Source data are provided with this paper.
References
Siegel, R. L., Wagle, N. S., Cercek, A., Smith, R. A. & Jemal, A. Colorectal cancer statistics, 2023. CA Cancer J. Clin. 73, 233–254 (2023).
Euliano, E. M., Sklavounos, A. A., Wheeler, A. R. & McHugh, K. J. Translating diagnostics and drug delivery technologies to low-resource settings. Sci. Transl. Med. 14, eabm1732 (2022).
Bustos, N. A., Ribbeck, K. & Wagner, C. The role of mucosal barriers in disease progression and transmission. Adv. Drug Deliv. Rev. 200, 115008 (2023).
Kaur, P. et al. Bacterioboat-A novel tool to increase the half-life period of the orally administered drug. Sci. Adv. 8, eabh1419 (2022).
Ouyang, J. et al. Oral drug delivery platforms for biomedical applications. Mater. Today 62, 296–326 (2023).
Srinivasan, S. S. et al. RoboCap: Robotic mucus-clearing capsule for enhanced drug delivery in the gastrointestinal tract. Sci. Robot. 7, eabp9066 (2022).
Kim, K. S., Na, K. & Bae, Y. H. Nanoparticle oral absorption and its clinical translational potential. J. Control. Release 360, 149–162 (2023).
Wang, Z.-H. et al. Biological chemotaxis-guided self-thermophoretic nanoplatform augments colorectal cancer therapy through autonomous mucus penetration. Sci. Adv. 8, eabn3917 (2022).
Wang, Z.-H. et al. Self‐thermophoretic nanoparticles enhance intestinal mucus penetration and reduce pathogenic bacteria interception in colorectal cancer. Adv. Funct. Mater. 33, 2212013 (2023).
Bej, R. & Haag, R. Mucus-inspired dynamic hydrogels: Synthesis and future perspectives. J. Am. Chem. Soc. 144, 20137–20152 (2022).
Ma, Y. et al. pH-mediated mucus penetration of zwitterionic polydopamine-modified silica nanoparticles. Nano Lett. 23, 7552–7560 (2023).
Zhang, S. et al. Recent progress of micro/nanomotors to overcome physiological barriers in the gastrointestinal tract. J. Control. Release 360, 514–527 (2023).
Liu, T. et al. Controlled propulsion of micro/nanomotors: Operational mechanisms, motion manipulation and potential biomedical applications. Chem. Soc. Rev. 51, 10083–10119 (2022).
Yong, J., Mellick, A. S., Whitelock, J., Wang, J. & Liang, K. A biomolecular toolbox for precision nanomotors. Adv. Mater. 34, e2205746 (2022).
Zhou, H., Mayorga-Martinez, C. C., Pane, S., Zhang, L. & Pumera, M. Magnetically driven micro and nanorobots. Chem. Rev. 121, 4999–5041 (2021).
Zhang, L. et al. A dual-biomineralized yeast micro-/nanorobot with self-driving penetration for gastritis therapy and motility recovery. ACS Nano 17, 6410–6422 (2023).
Maher, S. & Brayden, D. J. Formulation strategies to improve the efficacy of intestinal permeation enhancers. Adv. Drug Deliv. Rev. 177, 113925 (2021).
Kim, J. C., Park, E. J. & Na, D. H. Gastrointestinal permeation enhancers for the development of oral peptide pharmaceuticals. Pharmaceuticals 15, 1585 (2022).
Abokor, A. A. et al. Immunoglobulin A, an active liaison for host-microbiota homeostasis. Microorganisms 9, 2117 (2021).
Huus, K. E. et al. Commensal bacteria modulate Immunoglobulin A binding in response to host nutrition. Cell Host Microbe 27, 909–921 (2020).
Chen, F. et al. Nanoscale polydopamine (PDA) meets pi-pi interactions: An interface-directed coassembly approach for mesoporous nanoparticles. Langmuir 32, 12119–12128 (2016).
Diez, P. et al. Ultrafast directional Janus Pt-mesoporous silica nanomotors for smart drug delivery. ACS Nano 15, 4467–4480 (2021).
Sontheimer-Phelps, A. et al. Human colon-on-a-chip enables continuous in vitro analysis of colon mucus layer accumulation and physiology. Cell Mol. Gastroenterol. Hepatol. 9, 507–526 (2020).
Lamson, N. G., Berger, A., Fein, K. C. & Whitehead, K. A. Anionic nanoparticles enable the oral delivery of proteins by enhancing intestinal permeability. Nat. Biomed. Eng. 4, 84–96 (2019).
Wu, J. et al. How Nanoparticles Open the Paracellular Route of Biological Barriers: Mechanisms, Applications, and Prospects. ACS Nano 16, 15627–15652 (2022).
Horowitz, A., Chanez-Paredes, S. D., Haest, X. & Turner, J. R. Paracellular permeability and tight junction regulation in gut health and disease. Nat. Rev. Gastroenterol. Hepatol. 20, 417–432 (2023).
Yang, D. et al. Transferrin functionization elevates transcytosis of nanogranules across epithelium by triggering polarity-associated transport flow and positive cellular feedback loop. ACS Nano 13, 5058–5076 (2019).
Rallabandi, H. R. et al. Evaluation of intestinal epithelial barrier function in inflammatory bowel diseases using murine intestinal organoids. Tissue Eng. Regen. Med. 17, 641–650 (2020).
Song, Q. et al. A probiotic spore-based oral autonomous nanoparticles generator for cancer therapy. Adv. Mater. 31, 1903793 (2019).
Fukui, J. A., White, K., Frankland, T., Oshiro, C. & Wilkens, L. Weight changes according to treatment in a diverse cohort of breast cancer patients. J. Clin. Oncol. 39, e12513 (2021).
Datta, J., Narayan, R. R., Kemeny, N. E. & D’Angelica, M. I. Role of hepatic artery infusion chemotherapy in treatment of initially unresectable colorectal liver metastases: A review. JAMA Surgery 154, 768–776 (2019).
Temraz, S. et al. Gut microbiome: A promising biomarker for immunotherapy in colorectal cancer. Int. J. Mol. Sci. 20, 4155 (2019).
Hortelão, A. C. et al. LipoBots: Using liposomal vesicles as protective shell of urease‐based nanomotors. Adv. Funct. Mater. 31, 2002767 (2020).
Veider, F., Knoll, P., Jorgensen, A. M., Stengel, D. & Bernkop-Schnurch, A. Oral drug delivery: Influence of mucus on cellular interactions and uptake of lipid-based nanocarriers in Caco-2 cells. Acta Biomater. 167, 416–424 (2023).
Murgia, X., Loretz, B., Hartwig, O., Hittinger, M. & Lehr, C.-M. The role of mucus on drug transport and its potential to affect therapeutic outcomes. Adv. Drug Deliv. Rev. 124, 82–97 (2018).
Spanogiannopoulos, P. et al. Host and gut bacteria share metabolic pathways for anti-cancer drug metabolism. Nat. Microbiol. 7, 1605–1620 (2022).
Fan, J. et al. A novel self-assembled sandwich nanomedicine for NIR-responsive release of NO. Nanoscale 7, 20055–20062 (2015).
Yang, Y. et al. Anion assisted synthesis of large pore hollow dendritic mesoporous organosilica nanoparticles: understanding the composition gradient. Chem. Mater. 28, 704–707 (2016).
Zhou, X. et al. Targeted delivery of cisplatin-derived nanoprecursors via a biomimetic yeast microcapsule for tumor therapy by the oral route. Theranostics 9, 6568–6586 (2019).
Wu, H. et al. A Bacteria-Inspired Morphology Genetic Biomedical Material: Self-Propelled Artificial Microbots for Metastatic Triple Negative Breast Cancer Treatment. ACS Nano 15, 4845–4860 (2021).
Tang, X. et al. Magnetic-acoustic sequentially actuated CAR T cell microrobots for precision navigation and in situ antitumor immunoactivation. Adv. Mater. 35, e2211509 (2023).
Acknowledgements
This work was supported by the National Natural Science Foundation of China [nos. 82473863 (J.S.), 82222067 (J.S.), and 82102936 (Z.W.)], Outstanding Youth Foundation of Henan Province [no. 222300420020 (J.S.)], China Postdoctoral Science Foundation [no. 2023M743232 (Z.W.)], the Postdoctoral Fellowship Program of CPSF under Grant Number [no. GZB20230676 (Z.W.)], and Scientific and Technological Project of Henan Province [no. 242102310450 (Z.W.)]. We feel grateful to the staff at Modern Analysis and Computing Center of Zhengzhou University for their technical assistance. The manuscript of Figs. 1, 2a, 2g, 3d, 3f, 4d, 4f, 5f, 6a, and Supplementary Fig. 35 were created with BioRender.com.
Author information
Authors and Affiliations
Contributions
Z.-H. Wang, X. Zeng, W. Huang, and Y. Yang performed experimental measurements. X. Zeng, S. Zhang, M. Yang, and H. Liu drafted the paper. A. Li, F. Zhao, and J. Liu contributed to the data interpretation. Z.-H. Wang, Z. Zhang, J. Liu, and J. Shi revised the paper. Z.-H. Wang, and J. Shi provided financial support. All authors approved the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Ana Jaklenec, Fuwu Zhang, Piotr Rychahou and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Wang, ZH., Zeng, X., Huang, W. et al. Bioactive nanomotor enabling efficient intestinal barrier penetration for colorectal cancer therapy. Nat Commun 16, 1678 (2025). https://doi.org/10.1038/s41467-025-57045-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-025-57045-9