The microfluidic chip device is consisted of PDMS construction, capillary tubes, dispensing needles, and rubber tubes. The rubber tubes are used as microchannels for the dispersed phase and continuous phase. One end of the capillary tube is inserted into the PDMS body, and the other end is connected to the rubber tubes. In experiment, the fluid flows into the PDMS construction through the rubber tubes and capillary tube in sequence, where the micro droplets are sheared, and then flows out collectively through the rubber tube at the other end. All openings on the device surface are sealed with leak-proof oil sealant and left to stand for 15 min for the sealant to cure. The device is then stored in a cool and dry place.

A three-degree-of-freedom Helmholtz coil, a multi-function data acquisition device, and three single-channel output power amplifiers jointly constitute the external rotating uniform magnetic field. By controlling the current and voltage of the Helmholtz coil, a circular external rotating uniform magnetic field can be generated on any plane in three-dimensional space to drive the microrobots to move in different ways (Wang et al., 2023). In the experiment, the cell-culture dish was placed in the center of the three-dimensional Helmholtz coil, where the magnetic field intensity was evenly distributed. After adding the drug-loaded microrobot, turning on the magnetic field can drive the microrobot to move towards the cells.

CRC cell lines DLD-1 and LoVo, as well as human intestinal epithelial cell line NCM460, are provided by Fuheng Cell Centre (Shanghai, China). LoVo cells were cultured in RPMI 1640 medium (Gibco; Thermo Fisher Scientific, Inc.), while DLD-1 and NCM460 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Inc., Waltham, MA, United States). Both media were supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and 1% antibiotic (Sigma-Aldrich, United States). All cells were cultured in a standard humidified incubator at 37°C under an atmosphere with 5% CO₂. LH (cat. no. L101559) was purchased from Aladdin Industrial Corporation (Shanghai, China) and was dissolved in dimethyl sulfoxide (DMSO) to prepare a stock solution of 77.16umol/L.

The cytotoxicity of LH was conducted by MTT assay. Cells were incubated in 96-well plate (5×10³ per/well) for 24 h, then treated with different concentrations of LH (0–100 μM) for 48 h. After staining with MTT solution (5 mg/mL, 20 μL/well, 4 h) (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany), 150 μL of DMSO was added to each well to solubilize the formazan crystals. A microplate reader (ELx808, BioTek Instruments, Winooski, VT, United States) was employed to detect the plate at the wavelength of 490 nm. Three biological experiments were performed.

CRC cells (0.5 × 10³) were cultured in 6-well plates for 24 h. Three groups were set up, including a negative control group (NC), LH-alone (20 μM), and a magnetic microrobot drug-loading group (LH-robot) (20 μM). Treatments were conducted on days 2, 4. After 10 days, the cells were fixed with methanol and stained with a 0.5% crystal violet solution. Three biological experiments were performed.

DLD-1 and LoVo cells (90% confluence) were scratched with a sterile 200-µL pipette tip and then treated with LH and LH-robots (20 μM) separately. After 48 h, the width of the scratch was observed. The scratch was imaged under a microscope (magnification, ×100). The widths of the scratches were analysed with ImageJ V1.8.0 (NIH, Bethesda, MD, United States). Three biological experiments were performed.

The invasion experiment was conducted using a Transwell chamber. Before cells seeding, 50 μL of Matrigel was added to the upper chamber to coat the polycarbonate membrane. DLD-1 and LoVo cells were treated with LH alone and LH-robots (20 μM) separately and then cultured for 24 h. After resuspension in FBS-free medium, they were inoculated into the upper chamber (1×10⁴ cells/200 µL medium per well). Then, the lower chamber containing 700 µL of medium with 10% FBS was used as a hemoattractant. After 48 h of incubation, cells on the polycarbonate membrane were wiped off, and then methanol was used to fix the cells penetrating to the dorsal side, followed by staining with a 0.5% crystal violet solution. Under a microscope (magnification, ×100), 5 randomly selected fields were quantitatively analysed. Three biological experiments were performed.

DLD-1 and LoVo cells were seeded in 6-well plates (2.5×10⁵ per/well) and allowed to adhere overnight. Then, LH alone and LH-robots (20 μM) were added separately, and the cells were incubated for 48 h before being harvested and stained with an Annexin V-FITC/PI Apoptosis Detection kit (catalog no. FXP018; 4A Biotech). FACS DiVa 6.1.3 (BD Biosciences, Franklin Lakes, NJ, United States) was used to analyze apoptosis. Three biological experiments were performed.

All data were shown as means ± SD via at least triplicate samples. A two-tailed, Student’s t test was used for testing the significance between two groups. Statistical analyses were performed using GraphPad PRISM 9 (GraphPad Software, Inc.) A one-way analysis of variance (ANOVA) with Dunnett’s test was performed to test the significance for multiple comparisons. A statistical significance was assumed at p < 0.05.

Autonomous micro/nano-robots can propel and navigate in various liquid media, and are expected to provide revolutionary technological advancements for drug delivery, microsurgery, and micro/nano-engineering (Xiao et al., 2022; Yang et al., 2022; Wang et al., 2021; Gao et al., 2021; Zhang et al., 2023). Magnetic nanorobots show great potential in practical biomedical applications due to their wireless fuel-free actuation, strong propulsion, precise motion control, and high biocompatibility (Wang and Zhang, 2021; Xie et al., 2020; Zhou et al., 2021; Yang et al., 2023). This study describes the preparation of magnetic-driven hydrogel microrobots. It has the capability to transport drug LH through the digestive channel and into the intestine. Subsequently, the microrobot, driven by an external magnetic field, precisely maneuvers to the target lesion location for drug release and treatment (Figure 1A). This application demands that the microrobot is capable of loading drugs, exhibiting magnetic responsiveness, and possessing a biocompatible structure. Additionally, the microrobot should have a disintegrable body structure for drug release. Figure 1B illustrates the fabrication process of the magnetic-driven hydrogel microrobot. The hydrogel microrobot is fabricated using a microfluidic chip based on the principle of flow focusing. In the flow convergence device, the dispersed phase fluid and continuous phase fluid pass through a narrow region under pressure. At the micro-droplet generation site, there are three fluid streams, with the continuous phase fluid symmetrically distributed on both sides. The dispersed phase fluid in the middle is focused and sheared by the continuous phase fluids on both sides, forming micro-droplets. The micro-droplets are subsequently cured and solidified on the surface under UV light exposure, thus transforming into magnetic-driven hydrogel microrobots. In this process, the continuous phase is plant oil, and the dispersed phase is a gelatin aqueous solution containing photoinitiator, along with Fe₃O₄ particles and the LH.

The optical microscopy image shown in Figure 1C displays the spherical geometry of the microrobots. The clearly visible yellow microparticles inside the microrobot indicate the successful loading of Fe₃O₄ particles. The fluorescent microscope image (Figure 1D) illustrates that the desired drug loading has been achieved. The loading of drugs and Fe₃O₄ particles decreased in the homogeneity of the gelatin solution as the dispersed phase, leading to a lack of strict uniformity in the size of the generated microrobots. Figure 1E presents the statistical analysis of the size distribution of the prepared microrobots, indicating that the size distribution follows a generally normal distribution. The average diameter is approximately 90 μm, predominantly distributed within the range of 30∼120 μm, exhibiting a high degree of monodispersity. In the microfluidic chip, the generation rate and size of microrobots can be modulated by altering the flow rate ratio, viscosity, and channel dimensions of the continuous and dispersed phase fluids.

Then the effect of frequency of the rotating magnetic field on the velocity of magnetic-driven hydrogel microrobot was investigated experimentally as well. Figure 2A shows the trajectories of different-sized microrobots over a period of 14 s under the conditions of a magnetic force of 15 mT and a driving frequency of 5 Hz. It can be seen that larger microrobots exhibit smaller movement speeds. To further reveal the magnetic field-driven motion performance of microrobots, we explore the variation of microrobots’ velocity with the magnetic frequency increased from 2 to 40 Hz and shown in Figure 2B, for a 30 μm microrobots, the speed increased linearly with the driving frequency and reached a maximum velocity of 8.4 μm/s at 5 Hz, further increasing the frequency reduced the velocity. Such a maximum synchronized frequency is called step-out frequency (Xie et al., 2019; Yu et al., 2019). The occurrence of the out-of-step phenomenon and the increase in drag caused by the increasing speed are the reasons we speculate for this variation. After the step-out phenomenon occurred, the speed of the microrobot fluctuates within a certain range as the frequency of the magnetic frequency increase. Furthermore, the 60 and 90 μm microrobots showed the same movement performance and obtained the highest velocities of 6 μm/s and 3.7 μm/s, respectively. This magnetically actuated motility of magnetic-driven hydrogel microrobots of different sizes can provide guidance for the customization of microrobots for different working conditions.

For the application of micro/nano-scale robots in precision medical procedures, the ability of remote driving has very attractive characteristics (Chirarattananon et al., 2014; Xie et al., 2019; Aram et al., 2022; Zhang et al., 2022). Here, we demonstrate the remote locomotion of Janus magnetic-driven hydrogel microrobots. Figure 3A illustrates the control strategy of three-dimensional rotating magnetic field generated by the three degrees of freedom Helmholtz coil and corresponding movement of microrobots. First, a circularly polarized rotating magnetic field applied in the X-Z plane excited the microrobot rolled along X-axis. When the rotating magnetic field was changed and applied in the Y-Z plane, the locomotion direction of microrobot changed to the Y-axis. The propulsion direction of the microrobot could be altered by changing the direction of rotating magnetic field manually. Based on this rule, we realized the controllable trajectory movement of microrobot by modulating the magnetic field. As shown in Figure 3B, the microrobot walked along letter “H”-, “P”-, and “H”-shaped trajectory. The microrobot realized flexible direction switching under the drive of magnetic field of 15 mT and 5 Hz. This remote controllable movement capability provides support for targeted locomotion of magnetic-driven hydrogel microrobots on the complex surface of the intestine.

• The chemical structure formula and 3D structure of LH are shown in respectively. Cell proliferation was assessed using MTT and colony formation assays (Chan et al., 2023). CRC cells were exposed to multiple concentrations (0–100 µM) of LH for 24 and 48 h. MTT assay results showed that LH significantly inhibited CRC cells proliferation in a dose and time-dependent manner (Figure 4B). Notably, LH showed minimal cytotoxicity towards NCM460 cells (Supplementary Figure S1). We chose a non-lethal concentration of 20 µM of LH and compared the effects of single drug administration with microrobot-based drug delivery on CRC cells. A large number of previous studies (Zhang et al., 2021; Zhou et al., 2021) have shown that Fe₃O₄ as a component of magnetic drive micro-nano robot will not cause biological tissue and cell damage, so the blank control group in this study containing a small dose of Fe₃O₄ will not cause additional effects. The colony formation experiment showed that the drug under the control system of magnetic microrobots had a more significant ability to inhibit the proliferation of CRC cells (Figure 4C).

Metastasis is the process by which cancer cells grow in organs far away from their primary organ, and is the deadliest manifestation of cancer (Xia et al., 2023). The vast majority of cancer patients die from metastatic disease, rather than primary tumors (Chen et al., 2019). Tumour cell mobility is essential for metastasis and is typically assessed by wound healing and Transwell assays (Jin et al., 2019). As shown in Figure 5A the wound healing speed in the cells treated with LH-robot was significantly slower. The Transwell assay results showed a significant decrease in the invasion ability of the experimental group cells, especially in the magnetic microrobot-based drug delivery group (Figure 5B).

Apoptosis is a programmed cell death that balances the ratio of cell survival and death. Cancer cells have the ability to escape apoptosis, so aimed at inducing cancer apoptosis is a very important direction for treating cancers (Elbanna et al., 2021). Cytotoxic chemotherapy and radiotherapy attempt to trigger apoptosis through endogenous pathways by acting on cell division and/or directly damaging DNA, and are currently the main methods for treating cancer through the induction of apoptosis mechanisms (Wanner et al., 2020; Barroso et al., 2023). In our experiments, we observed that LH has the ability to induce apoptosis in CRC cells, and the magnetic propelled hydrogel microrobots significantly enhance this ability (Figure 6).