Fig. 2: ARF-dominated acoustic trapping of genetically engineered bacteria.

a Schematic of the acoustic trapping of GVs@E. coli. b Phase-contrast microscopy and c representative TEM images of uninduced control E. coli. d Phase-contrast microscopy and e representative TEM images of GVs@E. coli. The black parts in (b, d) are the cytoplasm of the bacteria, and the white areas in (d) represent the GVs. Images are representative of three experiments in (b–e). f 3D simulated schematic diagram of the focal acoustic field (focal point: (0, 0, 6) mm). Simulated acoustic field (x–y plane, z = 6 mm) and ARF vectors of the g control E. coli and h GVs@E. coli at the focal point, respectively. i ARF exerted on E. coli and GVs@E. coli by the focal beam in the x–axis. Microscopic fluorescence photographs of j control E. coli and k GVs@E. coli in a silicone tube being exposed to ultrasound (focal beam) for 30 s, respectively. Only GVs@E. coli can aggregate at the focal beam centre and form clusters under the dominance of the ARF. Bacterial cluster formation times and cluster sizes of the GVs@E. coli under different l input voltages and m duty cycles. **** means P < 0.0001 compared to the 5 V group in (l) or 1% duty cycle group in (m). n Bacterial cluster size of the GVs@E. coli with various concentrations under the same ultrasound conditions. **** means P < 0.0001 compared to the 12.5 × 10⁷ cell mL⁻¹ case in (n). Scale bar: 100 μm. Data in (l–n) are presented as mean  ±  s.d. (n  =  3 biologically independent samples per group). Statistical analysis was multiple comparisons by using the one-way analysis of variance with Sidak’s test. Source data are provided as a Source Data file. The ARF-dominated acoustic trapping of GVs@E. coli is shown in Supplementary Movie 1.