3D-printed micrometer-scale wireless magnetic cilia with metachronal programmability

Abstract

Biological cilia play essential roles in self-propulsion, food capture, and cell transportation by performing coordinated metachronal motions. Experimental studies to emulate the biological cilia metachronal coordination are challenging at the micrometer length scale because of current limitations in fabrication methods and materials. We report on the creation of wirelessly actuated magnetic artificial cilia with biocompatibility and metachronal programmability at the micrometer length scale. Each cilium is fabricated by direct laser printing a silk fibroin hydrogel beam affixed to a hard magnetic FePt Janus microparticle. The 3D-printed cilia show stable actuation performance, high temperature resistance, and high mechanical endurance. Programmable metachronal coordination can be achieved by programming the orientation of the identically magnetized FePt Janus microparticles, which enables the generation of versatile microfluidic patterns. Our platform offers an unprecedented solution to create bioinspired microcilia for programmable microfluidic systems, biomedical engineering, and biocompatible implants.

Fig. 1.. Fabrication of programmable MAC arrays.

(A) Schematics of the fabrication platform consisting of a five-coil electromagnetic setup integrated inside a 2PP system (3D direct laser writing system) working with near-infrared (NIR) laser (see also fig. S1), showing that a MAC array composed of six cilia in a circular arrangement with a 60° orientation difference (i.e., phase difference) is being printed by cross-linking SF. The integrated electromagnetic setup enables the positioning and orienting of the premagnetized hard magnetic FePt JMPs whose magnetization direction is indicated by the red arrow (see fig. S2 and movie S1). (B) Schematics of the fabrication process of a representative MAC array with three cilia in a line with a phase difference of 45° (see movie S2). (C) Schematics of one representative programmable MAC array consisting of three cilia facing each other in a triangular configuration (phase difference = 120°) with each cilium containing a 3D flag-shaped additional structure, demonstrating the large design space of the platform. The geometry of the flag-shaped structure is shown in fig. S4. Illustrations are not to scale. Photo credit: Meng Li, Xinghao Hu, and Shuaizhong Zhang, Max Planck Institute for Intelligent Systems.

Fig. 2.. Magnetic, mechanical, and biological property characterization results of the MAC.

(A) Scanning electron microscopy (SEM) image of the FePt JMPs. (B) Magnetic hysteresis loop of the FePt JMPs. The inset shows the relative direction of the FePt JMPs and the applied magnetic field. (C) 3D oblique view of the confocal images of the SF blocks. (D) Volumetric swelling behavior of the printed SF blocks over time analyzed using the confocal images (see fig. S6). (E) Linear swelling behavior of the MAC over time. The insets show the swelling of one cilium printed with a laser power of 60%. (F) Young’s modulus of the SF blocks (see fig. S7). (G) Top-view optical microscopic images of the MAC under various temperatures, showing no visible shape change. (H) Measured linear swelling ratio of the MAC before and after baking at 60°C for 10 min. The insets show the microscopy images of the MAC before and after baking. (I) Measured maximal bending angle of the MAC under a uniform rotating magnetic field of 3 mT before and after baking at 60°C for 10 min. The insets show the images of a bending cilium printed with a laser power of 60% before and after baking. (J) Fluorescent images of human fibroblast cells after 96 hours cocultured with the MAC, showing no dead cells, which should exhibit a red color. The red dots are the printed SF beams and SF anchors. (K) Fluorescent and bright-field microscopic images of the MAC under a flow of protease XIV solution (1 U ml⁻¹) at a flow rate of 2 μl min⁻¹, showing quick biodegradation of the printed SF. Error bars indicate SDs for n ≥ 5 measurements. Photo credit: Shuaizhong Zhang, Meng Li, and Ugur Bozuyuk, Max Planck Institute for Intelligent Systems.

Fig. 3.. Magnetic actuation of the MAC using rotating uniform magnetic fields.

(A) Experimental (top) and numerically simulated (bottom) cilium motion at 1 Hz under a uniform magnetic field of 3 mT rotating in the CCW direction, showing a 2D reciprocal motion (see also movie S5). (i) Bottom view of one static cilium with the red arrow indicating the magnetic moment (m) direction of the FePt JMP. (ii to iv) Time-lapse images of the cilium motion during one beating cycle at 1 Hz. The images are composed of image sequences with an identical time interval of 0.002 s. The blue and green arrows indicate the direction of the magnetic stroke and the elastic stroke, respectively. (B) Cilia tip speed during one beating cycle at 1 Hz obtained experimentally and numerically. (C) Maximal bending angle θ of the MAC as a function of the amplitude of the applied magnetic field. The inset shows the definition of θ. (D) Maximal bending angle θ of the MAC printed over time under a uniform magnetic field of 3 mT rotating in the CCW direction. The insets show the bending of one cilium at times 0, 24, and 48 hours after developing with DI water, respectively. (E) Opening angle ϕ of the MAC as a function of their beating cycles, showing only slight change in the motion. The insets show the time-lapse image of a cilium during one beating cycle at 5 Hz at cycle number 1, 36,000, and 64,500, respectively. The definition of ϕ is indicated in the inset. The MAC reported in (A) and (B) are fabricated with a laser power of 60%. Error bars indicate SDs for n ≥ 5 measurements. Photo credit: Shuaizhong Zhang and Rongjing Zhang, Max Planck Institute for Intelligent Systems.

Fig. 4.. Programmable metachronal MAC array demonstrations.

(A) Programmable metachrony by changing the orientation of the FePt JMPs and the arrangement of the MAC array (see movie S6). Δψ indicates the orientation difference of the FePt JMPs between neighboring cilia. The black arrows indicate the direction of the propagating metachronal wave. (B) Programmable metachrony by changing the mechanical properties (i) and the geometry (ii) of the MAC while keeping the orientation of the FePt JMPs the same (see movie S7). The black arrows indicate the direction of the propagating metachronal wave. (C) 2D programmable metachrony with the fluorescent microscopic images (left) showing the SF structures and the bright-field microscopic images (right) showing the orientation of the FePt JMPs and the metachronal motion of the MAC arrays (see movie S8). All the MAC shown here, except for [(B), i], are printed with a laser power of 60%. The uniform rotating magnetic field is kept at 3 mT. Photo credit: Shuaizhong Zhang, Max Planck Institute for Intelligent Systems.

Fig. 5.. Fluid transporting capability of a single cilium.

(A) Time-lapse (left) and particle tracking (right) images of the generated flow by a single cilium over a period of 30 s (see movie S10). The overlapping images are composed of 150 consecutive images of 30-s videos. The observation layers for (i) to (iii) were 10, 40, and 40 μm above the substrate, respectively. The red arrows in (ii) and (iii) indicate the flow direction. The results of the static control experiments are shown in fig. S12. (B) Bottom-view microscopic images (top), oblique-view schematics (middle), and top-view modeling results (bottom) of the F-MAC beating at 5 Hz, showing the deflection of the flag-shaped structure during the magnetic (blue arrows) and elastic (green arrows) strokes and the induced instantaneous flow (white lines in the modeling results). The red dashed lines indicate the projected flag on the substrate plane. See also fig. S13 and movie S9. (C) F-MAC tip speed during a single beating cycle at 5 Hz. (D) Measured flow speed generated by both a single normal MAC and an F-MAC. (E) Maximal bending angle θ of the F-MAC as a function of the amplitude of the applied magnetic field immediately and after 24 hours of developing with DI water. The inset shows the microscopic image of an F-MAC bending at 10 mT and the definition of θ. All the MAC beams and the flag-shaped structures shown here are printed with a laser power of 60 and 40%, respectively. The applied uniform rotating magnetic field is 3 and 6 mT for the normal MAC and the F-MAC in (A) to (D), respectively. Error bars indicate SDs for n ≥ 5 measurements. Photo credit: Shuaizhong Zhang and Rongjing Zhang, Max Planck Institute for Intelligent Systems.

Fig. 6.. Programmable flow patterns.

(A) Bottom-view optical microscopic images (left) and particle tracking results (right) of the F-MAC arrays with different metachronal configurations. The phase difference of the laeoplectic and dexioplectic metachrony is π/3 and −π/3, respectively. The particle tracking images are composed of 150 images over a 30-s period. The black arrows indicate the flow direction, and the red arrows mark the rotating direction of a 6-mT uniform magnetic field (see movie S12). (B) Measured flow speed generated by laeoplectic, synchronous, and dexioplectic F-MAC arrays at different heights above the substrate. (C) Schematic of the circular F-MAC arrangement (left) and the time-lapse particle tracking results (right). The particle tracking image is composed of 300 images over a 60-s period. The black arrows indicate the flow direction, and the red arrow marks the rotating direction of a 6-mT uniform magnetic field. The results under a rotating magnetic field in the CW direction can be found in fig. S14 (see movie S13). (D) Measured flow speeds generated by the circular F-MAC array at different heights above the substrate under different rotating directions of the magnetic field. All the MAC beams and the flag-shaped structures shown here are printed with a laser power of 60 and 40%, respectively. Error bars indicate SDs for n ≥ 5 measurements. Photo credit: Shuaizhong Zhang, Max Planck Institute for Intelligent Systems.