Imaging-guided deep tissue in vivo sound printing

Abstract

Three-dimensional printing offers promise for patient-specific implants and therapies but is often limited by the need for invasive surgical procedures. To address this, we developed an imaging-guided deep tissue in vivo sound printing (DISP) platform. By incorporating cross-linking agent-loaded low-temperature-sensitive liposomes into bioinks, DISP enables precise, rapid, on-demand cross-linking of diverse functional biomaterials using focused ultrasound. Gas vesicle-based ultrasound imaging provides real-time monitoring and allows for customized pattern creation in live animals. We validated DISP by successfully printing near diseased areas in the mouse bladder and deep within rabbit leg muscles in vivo, demonstrating its potential for localized drug delivery and tissue replacement. DISP's ability to print conductive, drug-loaded, cell-laden, and bioadhesive biomaterials demonstrates its versatility for diverse biomedical applications.

Fig. 1.. Imaging-guided deep tissue in vivo sound printing (DISP).

(A) Schematic of the DISP platform. The DISP system utilizes an ultrasound-responsive bioink (US-ink) composed of non-crosslinked prepolymer, crosslinking agent-loaded low temperature sensitive liposomes (LTSLs), and gas vesicles (GVs). The US-ink is injected into the body to noninvasively fabricate a precise functional biostructure in vivo. Integrated GV-based ultrasound imaging is employed to monitor the target organ, detect the presence of the prepolymer, and ensure accurate targeting and successful ultrasound-induced gel (US-gel) formation. (B) In vivo printing setup for focused ultrasound (FUS) generation and monitoring. RF, radio frequency; T/R, transmitter/receiver. (C) Transmission electron microscopy (TEM) image of the crosslinking agent-loaded LTSLs. Scale bar, 100 nm. (D) Scanning electron microscopy (SEM) image of freeze-dried 3D printed alginate US-gel. Scale bar, 20 μm. (E) Functional hydrogel structures printed with sound in vivo printing. Scale bars, 5 mm. (F to H) DISP-based in vivo printing of bioelectronic devices for sensing and recording (F), biocarriers for drug delivery and tissue regeneration (G), and bioadhesives for wound sealing and device/tissue interfaces (H).

Fig. 2.. Synthesis and characterization of low temperature sensitive liposomes for controlled release of crosslinking agents.

(A) Schematic illustrating the formation of nanopores in lipid bilayers of LTSLs due to a phase transition from solid to liquid induced by mildly elevated temperatures. (B) Mass production of crosslinking agent (e.g., Ca²⁺)-loaded LTSLs through the extrusion process. (C) Dynamic light scattering (DLS) analysis of LTSLs before and after extrusion. (D) Fluorescent imaging of Ca²⁺-loaded LTSLs using fura-2-acetoxymethyl ester as an intracellular calcium indicator. Scale bar, 3 μm. (E) Stability study showing Ca²⁺ release from LTSLs stored at 4 °C and 25 °C after one month and six months. (F) Ultraviolet-visible analysis of LTSLs subjected to 43 °C for various durations. (G) Temperature-dependent Ca²⁺ release from LTSLs at 43 °C and 37 °C. (H) Crosslinking time for alginate US-inks under various heating temperatures when LTSLs concentration is fixed at 50 wt.%. (I) Ionic crosslinking of alginate US-inks with varying LTSL concentrations, evaluated by storage (Gʹ) and loss (Gʹʹ) modulus. Inset: images of the gelation status of alginate US-inks containing 0%, 15%, and 50% LTSLs after 30 s of exposure to 43 °C. Scale bars, 5 mm. (J) Crosslinking time measurements for alginate US-inks with different LTSL concentrations. (K) Live/Dead staining images of human dermal fibroblast cells cultured for 7 days with the alginate, alginate US-ink containing 50% LTSLs, and alginate US-gel. Scale bar, 100 μm. The error bars in the figures indicate the standard deviation from the mean (n=3).

Fig. 3.. Characterization of focused ultrasound-induced 3D printing.

(A) Schematic of FUS wave propagation, illustrating precise targeting of US-ink. (B) Comparison of tissue penetration depths for ultrasound waves versus various light sources, highlighting ultrasound’s superior penetration. UVA, ultraviolet A; UVB, ultraviolet B; NIR, near-infrared. Note the inverse relationship between the ultrasound frequency with penetration depth. (C) Thermal simulations showing the temperature distribution at the focal point under different frequencies and exposure times. Scale bar, 2 mm. (D) Temperature profile at the focal point of FUS at 8.75 MHz during and after 10 s of ultrasound exposure. (E to H) Normalized pressure maps at the focal point using a 2.65 MHz transducer: experimental measurements using a hydrophone in the X-Z plane (E) and in the X-Y plane (F), and simulation results in the X-Z plane (G) and in the X-Y plane (H). (I) DISP-printed US-gel patterns. Scale bar for inset, 400 μm. Scale bars for patterns on the right, 4 mm. (J) Printability of the alginate US-ink with an 8.75 MHz transducer at various power levels and printing speeds. (K) Printing resolution in terms of line width for alginate US-ink using an 8.75 MHz transducer at different power levels and printing speeds. Scale bar, 5 mm. (L) Printing resolution of alginate US-ink, measured as line width, when printed under 15 mm thick pork loin tissue at 18 W, with varying frequencies and printing speeds. Inset, a deep tissue printed pattern on pork tissue. Scale bar, 5 mm. (M) Dissociation of alginate US-gels patterned on tissue using DISP, achieved by 5 min treatment with 0.025 M EDTA solution. The error bars in the figures indicate the standard deviation from the mean (n=3).

Fig. 4.. Deep tissue in vivo sound printing-based 3D printing of functional biomaterials for various medical applications.

(A) Schematic of conductive US-ink composed of carbon nanotube (CNT) additives entangled within alginate US-ink, crosslinked using FUS. (B) Conductive US-gel patterns maintain stable electrical properties under cyclic bending deformations. (C) Temperature sensing using printed conductive US-gels. Inset, consistent and reversible temperature sensor response upon contact with human skin. RT, room temperature. (D) DISP-printed conductive US-gel sensors for electrocardiogram (ECG) and electromyography (EMG) recording in a human participant. (E) Integration of therapeutic biomolecules within US-ink, forming biocarrier US-gels for potential drug delivery applications. (F) Continuous and sustainable release of a model drug rhodamine B from US-gels. (G) Cell-encapsulated US-gels prepared by integrating cells within biocompatible US-inks followed by printing using an 8.75 MHz transducer at 7 W and a printing speed of 10 mm min⁻¹. (H) Live/Dead staining images of C2C12 mouse myoblast cells encapsulated within alginate US-gels on Day 1 and Day 3 post-printing. Scale bars, 100 μm. (I) Metabolic activity of cells assessed from Day 1 to Day 7 post-printing. Insets, images of the cell-laden US-gel pattern, printed with an 8.75 MHz transducer at 7 W and 10 mm min⁻¹, showing live cells 3 days post-print. Scale bar, 200 μm. (J) Catechol-modified gelatin-caffeic acid conjugates (GelCA) US-inks mixed with NaIO₄ liposomes for bioadhesive applications. (K) Adhesion strength of GelCA US-ink before and after crosslinking. Inset, images of the GelCA US-ink before and after mild heating and crosslinking. (L) Ex vivo adhesion testing of GelCA US-gel for sealing the punctured heart tissue. Scale bar, 5 mm. (M) In vivo US-induced adhesion, where FUS facilitates prepolymer jetting towards tissue, followed by in situ crosslinking of alginate US-ink to achieve mechanical interlock. (N) Alginate US-gels printed in vivo following intradermal injection of US-ink and crosslinking using a 2.65 MHz transducer at 7 W, with a printing speed of 20 mm min⁻¹ on live animals. The strong interfacial adhesion is observed between the alginate US-gel and tissue, with blue dye applied for visibility. Scale bars, 6 mm. The error bars in the figures indicate the standard deviation from the mean (n=3).

Fig. 5.. Imaging-guided deep tissue sound printing in vivo.

(A) Setup for sound printing in vivo in live animals, employing ultrasound imaging for precise targeting. Inset: a linear pattern printed in vivo in a mouse. Scale bar, 4 mm. (B and C) Schematic of AM-mode ultrasound imaging with a GV contrast agent used to monitor US-ink distribution in vivo (B) and to ensure precise targeting (C). Ultrasound image inset in (C): A line of GV-integrated alginate US-ink printed and imaged in cross-section. GVs in areas not exposed to FUS remained intact, while those exposed to FUS collapsed. (D) In vivo printing of US-gels on a tumor site in the bladder of an anesthetized mouse. Successful targeting confirmed by GV collapse. After printing, the mouse bladder was extracted to verify successful printing. Scale bar, 4 mm. (E) In situ Ca²⁺ sensing using GV Ca²⁺ sensors integrated into alginate US-inks, designed to activate upon exposure to Ca²⁺. A line was printed and imaged in cross-section using AM-mode ultrasound imaging. Higher pressures in the center of the printed line led to partial collapse of GV Ca²⁺ sensors, while GV Ca²⁺ sensors at the boundary of the printed US-gel were activated, confirming the shape. (F) US-gel line printed using a 2.65MHz FUS at 11 Watt and 15 mm min⁻¹ on the abdominal muscle in a rabbit model. (G) The US-gel line printed deep into the adductor muscle and below the biceps femoris muscle using 2.65 MHz FUS at 20 Watt and 10 mm min⁻¹. (H) In vivo biocompatibility study of US-ink injected intradermally and ultrasound-printed US-gel in mice, assessed through hematoxylin and eosin (H&E) staining of skin tissues at one week and four weeks post-printing. Scale bars, 200 μm.