3D necroprinting: Leveraging biotic material as the nozzle for 3D printing

Selection of biological micro dispense tip

To explore the potential of biological micro dispense tips for high-resolution dispensing applications, we began with a comprehensive survey of biological micro dispense tips found in nature. Figure 1A provides an overview of various dispense tip-like structures in nature that facilitate fluid delivery. These structures exhibit diverse morphologies and functionalities, making them intriguing templates for engineered dispensing tools. By examining specific characteristics such as curvature, length, diameter, stiffness, and strength, one can identify promising candidates for specific applications, such as drug delivery (25), biomedical analysis through body fluid collection, adhesive dispensing in automated manufacturing, 3D bioprinting for tissue engineering (26, 27), and biohybrid atomic force microscopy (28).

Biological micro dispense tips can be broadly classified into two main categories: fluid depositing and fluid withdrawing dispense tips. Fluid depositing dispense tips eject substances from the organism into the environment. Typically, they are used to inject toxins into prey to induce paralysis or digestion (29–31), aiding in feeding, self-defense (31), or territorial competition. Examples include stingers, as seen in bees, wasps, and scorpions (32, 33); fangs, common in venomous snakes (34); claws, such as centipede forcipules with varying poison duct structures (29, 35); and harpoons, such as those of cone snails, which use tethered, hollow harpoons to deliver paralyzing agents (36, 37). In comparison, fluid withdrawing micro dispense tips serve to absorb fluids from the environment, transporting them into the organism. This category includes plant xylem vessels, which transport water from roots to leaves (38), as well as insect proboscides that are elongated stylets for feeding. Proboscides vary in composition, structure, and mechanics, adapting to their target food sources. They can be categorized into four types: flexed under the head, retracted into the head, flexed in front of the head, and coiled (39). For instance, assassin bugs and long-tongued bees have proboscides flexed under the head, while butterflies and moths possess coiled proboscides optimized to reach deep within flowers (23, 39, 40).

A systematic analysis of biological micro dispense tips was conducted to identify the most suitable candidates for DIW 3D printing. Ideal dispense tips should exhibit minimal curvature, be composed of materials with relatively high stiffness and strength, ensuring minimal compliance during ink extrusion, criteria consistent with prior studies on DIW printability (41–43). Beyond these requirements, two additional parameters are critical: inner diameter and dispense tip length. Inner diameter is important for achieving high printing resolution, while the dispense tip length influences the feasibility of fabrication. Micro dispense tips must be sufficiently visible and manipulable for integration into the printing system; however, exceedingly long dispense tips can generate high backpressure, increasing the risk of mechanical failure.

Referring to the selection chart (Fig. 1A) and length/curvature plot (Fig. 1B), several biological micro dispense tips were identified as potential candidates for meeting our selection criteria, including the proboscides of mosquitos, assassin bugs, bed bugs, aphids, sandflies, and tsetse flies. Among them, the female mosquito proboscis is particularly noteworthy for its stiff and straight structure, capable of even piercing the epidermis and infiltrating blood vessels, despite of being composed of soft polymeric materials (44–47). Anatomically, it features a core-shell structure, with the labium forming the shell and a fascicle composed of stylets (e.g., labrum and hypopharynx) serving as the core (28). The labrum and the hypopharynx together form a single, sealed, hollow-tube structure that transports blood from the host to the female mosquito (44). This structural design makes the female mosquito fascicle a compelling candidate as a biological micro dispense tip.

We further examine the characteristics of the female mosquito proboscis against the established criteria of the DIW nozzles (Fig. 1C). The straight shape of proboscis gives nearly zero curvature and its inner diameter, averaging 20 to 25 μm (44), is smaller than the minimum diameter of commercially available metal and plastic dispense tips. While glass-pulled dispense tips can achieve diameters below 1 μm (48), they are difficult to fabricate (49, 50) and extremely brittle (51). The female mosquito proboscis length is approximately 2 mm (44), which is manageable for manipulation and can be adjusted during fabrication to tune backpressure. Mechanically, the stiffness of the female mosquito proboscis is approximately 200 MPa (44), comparable to common plastics (52–54). While its mechanical strength is not documented in the literature, we measured it through burst pressure tests, to be approximately 708 kPa (detailed below). Furthermore, mosquitoes are widely available, easy to rear, and present on a global scale, enhancing their accessibility for this application.

Given these attributes, we selected the female mosquito proboscis as a primary candidate for biohybrid DIW printing. We hypothesized that integrating female mosquito proboscides with a 3D printing platform could enable high-resolution 3D necroprinting. The structure and size of the female mosquito proboscis facilitate easy mounting onto a printer platform, while its mechanical properties provide resistance against rupture during the extrusion of DIW inks. The extensive evolutionary refinement of the female mosquito proboscis yields an optimized internal surface architecture designed for the efficient transport of non-Newtonian fluids under laminar flow conditions (fig. S18 and see the Supplementary Materials). To test this hypothesis, we will materialize the concept of 3D necroprinting and systematically investigated its operational parameters and performance below.

Design of 3D necroprinting

To harness biological micro dispense tips for 3D printing, we developed a custom 3D DIW printer. This setup included a high-resolution motion stage, allowing for the attachment of biological micro dispense tips, as well as a piston-driven extruder (fig. S1) capable of micrometer-level dispensing resolution, synchronized with the motion stage using an Arduino microcontroller and a DC signal switch (fig. S4). A power law model was used during the preliminary design phase to create a theoretical process window (fig. S2), providing insight into the feasibility of various extruder stepper motor options during the selection process. The validity of the simplified theoretical process window was authenticated via COMSOL simulations using a Herschel-Bulkley inelastic flow model (fig. S3). Figure 2A illustrates the overall setup, including the extruder and syringe assembly. Additional details on the design and implementation of the printer are available in the Supplementary Materials.

Figure 2B illustrates the working principle to use biological micro dispense tips in DIW printing applications. The integration of the female mosquito proboscis with the custom printer leverages the Luer-Lock mechanism, which connects the printer-mounted syringe to disposable engineered dispense tips. This approach involves bonding the biological micro dispense tip directly to the outlet of the engineered dispense tip, enabling a continuous fluid pathway from the syringe barrel, through the dispense tip conduit, and ultimately through the biological micro dispense tip to deposit ink onto the substrate. While engineered dispense tips are currently used to facilitate this process, customized reusable adapters could be developed to interface directly with syringes and biological micro dispense tips, eliminating the reliance on engineered tips. Figure 2C demonstrates the DIW biological micro dispense tip setup, showing the female mosquito proboscis mounted on the printer. A zoomed-in view highlights the presence of the female mosquito proboscis extending from the Luer-Lock dispense tip. To demonstrate the feasibility of extruding a 3D-printable ink, Fig. 2D presents a sequence of snapshots showcasing the successful extrusion of Cellink Start, an off-the-shelf bioink optimized for cell-supporting structural 3D printing (movie S1).

Characterization and analysis of mechanical failure

It is important to characterize the mechanical failure behavior of the female mosquito proboscis during function and its dependence on the operating conditions of 3D printing. We first tested the female mosquito proboscis dispense tip by extruding various bioinks, including Cellink Start bioink gel and Pluronic F-127 bioink gel, which are commonly used for bioprinting. When ramping up the pressure to extrude Cellink Start bioink gel mid-air, we discovered an occasional behavior where fracture occurred at the outlet and an axial crack merged and propagated rapidly in the direction of the inlet (Fig. 3A, fig. S7, and movie S2). When extruding Pluronic F-127 bioink gel at a high extrusion rate, this would result in the proboscis experiencing a burst-like phenomenon near its inlet (Fig. 3B and movie S3). Notably, similar to the rupture at the outlet, it was an axial crack that propagated along the axis of the proboscis.

To further investigate the initiation and propagation of axial cracking observed in the female mosquito proboscis dispense tip, we combined experimentation and analysis to study the rupture behavior. The mechanical failure behavior of the female mosquito proboscis has not been studied before. We built and implemented a customized setup for the burst pressure measurements. Figure 3C presents a schematic of the test apparatus, where a syringe, pressure transducer, standard dispense tip, and female mosquito proboscis are assembled and filled with deionized water. Once all air bubbles are expelled from within the system, the tip of the proboscis is sealed shut, converting the setup into a sealed pressure vessel. By displacing the plunger at a slow rate, we created a quasistatic pressure within the system, maintaining a constant pressure at any given period all throughout. The plunger was continuously loaded until the female mosquito proboscis failed, as manifested by a breach in the walls of the proboscis where the water escaped (movie S4). The pressure transducer recorded a pressure increase until a peak was attained at the moment of failure (Fig. 3D and fig. S8 for the raw data of all tested samples). Multiple burst pressure tests yielded an average burst pressure of 59.7 kPa (Fig. 3E).

To rationalize our results, we next performed an analysis to reveal the critical stress state within the female mosquito proboscis during rupture. Consider the structure of the female mosquito proboscis; the labrum is rolled up into a cylindrical tube with the two edges forming a seam covered by hypopharynx (Fig. 3F). Our imaging of its microstructure confirmed that the inner diameter is more than 10 times of the wall thickness (fig. S5). Therefore, as an idealization, we safely treated it as a uniform thin-walled tube, and so applied a thin-walled pressure vessel model. The following assumptions were also made: The proboscis is composed of an isotropic material, and the strains resulting from the applied internal pressure are small (55).

After revealing the axial crack during rupture, we next investigated the location of rupture and its relation to operating conditions. As described above, we found that the rupture would either occasionally occur at the outlet of the female mosquito proboscis (type 1 failure) or consistently near its inlet (type 2 failure). For type 1 failure, a shear-thinning ink is extruded by the biological micro dispense tip and left to accumulate at the outlet. As such, the shear stress is removed from the shear-thinning ink, so that the storage modulus in turn dominates the loss modulus (G″ < G′), inducing a solid-phase behavior of the ink. As the ink exits the outlet of the female mosquito proboscis dispense tip, it no longer experiences the shear-stress caused by the extruder backpressure, enabling the solid-phase property of the fluid to prevail (fig. S11). The accumulation of this solid-like ink forms a blockage at the female mosquito proboscis dispense tip outlet, impeding the flow in this location. On the basis of the Herschel-Bulkley principle (see the Supplementary Materials), due to the blockage-induced ink velocity decrease at the female mosquito proboscis dispense tip outlet, an elevation in internal pressure at the nozzle outlet is attained (Fig. 3G). As the pressure rises, a substantial hoop stress develops in the female mosquito proboscis inner walls, particularly causing localized stress developments near structural imperfections or geometric discontinuities, leading to the initiation of microfractures. These microfractures propagate under continuous stress loading and ultimately coalesce, resulting in the observation of type 1 failure. Given this, type 1 failure is driven by clog-induced overpressure at the proboscis tip.

When the apparent viscosity of a shear-thinning ink is too high, forcing the backpressure requirements to exceed the material strength of the female mosquito proboscis, the rupture occurs in the upper section of the female mosquito proboscis dispense tip close to its inlet (Fig. 3H). Since its occurrence is driven by the pressure demands imposed by the desired fluid flow rate, we termed the type 2 failure as uniform overpressure due to high viscosity–induced flow requirements. It can be understood as follows: When the apparent viscosity of an ink is high, the backpressure necessary to maintain a certain ink velocity increases, with the backpressure experiencing a steady drop from the inlet to the outlet as the ink flows. Consequently, if the backpressure is set too high, as the pressure reaches the female mosquito proboscis dispense tip inlet walls, it will instantly stimulate ${\mathrm{\sigma }}_{\mathrm{\theta }{\mathrm{\theta }}_{\text{inlet}}}\ge {\mathrm{\sigma }}_{\mathrm{\theta }{\mathrm{\theta }}_{\text{critical}}}$ , causing the female mosquito proboscis dispense tip to consistently rupture near the inlet.

Assuming the presence of microfractures along the female mosquito proboscis, and that these fractures govern failure behavior by dominating the intrinsic material strength of the biological matter, a fracture mechanics framework provides a more accurate method for quantifying material failure. To examine this fracture-driven behavior, a theoretical analysis was performed using a range of initial crack lengths to estimate the corresponding critical stress intensity factors (fig. S10). The results indicate that for specific crack lengths (2a = 8 and 12 μm), the calculated critical stress intensity factors (K_IC = 4.35 and 7.04 kPa m^1/2) closely approximate the reported value for chitin hydrogels (K_IC ~10 kPa m^1/2) (56). This alignment supports the anatomical composition of the female mosquito proboscis, wherein the labrum and hypopharynx consist of chitinized biological material and are connected by a less-chitinized membrane (57, 58).

Process window and performance of 3D Necroprinting

Beyond regulating flow and pressure to prevent fracture of the female mosquito proboscis dispense tip, it is crucial to balance the ink extrusion speed (v_ink) and nozzle movement speed (v_nozzle), or the draw ratio. An imbalance between these two critical parameters can result in overextrusion or underextrusion. To address this, parametric studies were conducted using a common shear-thinning ink [40% (w/v) Pluronic F-127], varying each speed independently within the instrument’s operational limits to define an optimal process window for synchronized printing (Fig. 4A, table S2, and fig. S12). Overextrusion is characterized by nonuniform, yet continuous, printed lines, typically accompanied by type 1 failure of the mosquito proboscis (as defined above). This failure mechanism is explained by the Bernoulli principle, wherein ink accumulates between the nozzle and the substrate, causing a decrease in extrusion speed at the outlet relative to the inlet velocity. This imbalance generates a pressure buildup at the nozzle outlet, ultimately leading to either gushing of ink or catastrophic rupture of the mosquito proboscis. As a general rule of thumb, the draw ratio (r = v_ink/v_nozzle) should be kept below 1 to alleviate the issue of pressure build up at the outlet to prevent gushing or fracture.

Good extrusion is the combination of ink speeds and nozzle speeds that produce ideal printed lines that are continuous and uniform (61). This section commences at r ≤ 1 where v_ink equals or is smaller than v_nozzle such that no apparent pressure buildup occurs to cause clogging/gushing or fracture of the female mosquito proboscis dispense tip (42). Printed lines in this regime demonstrate a consistent line width, approaching the inner diameter of the female mosquito proboscis of 20 to 30 μm (Fig. 4A, right). The lower boundary of the draw ratio for achieving good extrusion was determined experimentally and found to be at 0.25, below which under extrusion happened. Under extrusion is the regime when broken/noncontinuous filaments are printed. Because of the excessive stretching of the filament caused by the immense velocity contrast Δv = v_nozzle − v_ink, the ink fractures into discontinuous segments, as shown in Fig. 4A, right. On the basis of this governing ink fracture draw ratio, the experimental failure strain of 40% (w/v) Pluronic F-127 is 3000% (see the Supplementary Materials for more details).

With the guidance of the process window, 3D necroprinting achieved a printing resolution of approximately 20 μm, which is ~250% finer than the ~50 μm inner diameter of a standard 34G dispense tip, the smallest commonly used commercial dispense tip to our knowledge. While a few suppliers offer 36G dispense tips with an inner diameter of ~35 μm, capable of achieving a printing resolution of ~40 μm, these are specialized products costing approximately $80 per dispense tip.

To evaluate the shape fidelity of printed structures, we printed an array of high-resolution structures such as a honeycomb, a maple leaf and a cell-laden scaffold (Fig. 4, B to D, and fig. S14). The first demonstration is a honeycomb structure with overall dimensions of approximately 600 μm × 600 μm × 310 μm (Fig. 4B). Scanning electron microscopy (SEM) images, including an isotropic view and a zoomed-in section of the sidewall, highlight the high print fidelity achieved between successive layers. The extruded filaments have a printed line width of ~22 μm, enabling the fabrication of a microscopic structure that surpasses the resolution capabilities of standard metal and plastic dispense tips. Only glass-pulled dispense tips have demonstrated comparable capabilities for producing 3D designs at this scale (62, 63). The second demonstration, a maple leaf structure, exhibits even greater fidelity in printed lines between layers, with each filament clearly defined in the sidewall (Fig. 4C). The structure measures 900 μm × 870 μm × 310 μm, and its printed lines, measuring ~18 μm, exceed the resolution of the first demo, further demonstrating the potential of this biohybrid printing approach.

3D necroprinting also has the potential for advanced bioprinting applications. Demo 3 features a 600 μm × 600 μm × 310 μm grid scaffold fabricated using the female mosquito proboscis dispense tip nozzle, producing 28-μm-wide lines (Fig. 4D). The ink used in this demonstration was Pluronic F-127 containing B16 cancer cells suspended in solution (fig. S15A). The Pluronic F-127 ink was also used in a demonstration containing red blood cells (RBCs), demonstrating the capabilities of the dispense tip nozzle to print high-density cell-laden constructs (fig. S15, B to D). The printed structures expressed a post-printing cell viability of 86.1% ± 2.1% (n = 3), confirming a sufficient resistance against shear stress–induced cell-rupture susceptible during extrusion (fig. S15, E and F). This highlights the potential of 3D necroprinting for biological applications at the microscopic scale, offering high production rates and low input costs. Collectively, the three demonstrations underscore the high resolution, enhanced print fidelity, and straightforward implementation achievable with biological micro dispense tips in DIW 3D printing.

Beyond cell-laden bioprinting, we also investigated the female mosquito proboscis dispense tip’s potential for high-resolution drug delivery (fig. S16). Using hydrogel as a model drug carrier (64) and pig skin as a tissue analogue, the tip enabled both uptake and redeposition of material at the picoliter scale, mimicking therapeutic delivery in live tissue. Its elasticity and compliance reduce the risk of substrate damage compared to rigid nozzles, offering advantages for in vivo applications. Furthermore, the proboscis exhibits a defined burst pressure that passively limits extrusion forces, acting as a biological “fuse” to protect sensitive cell-laden bioinks from shear-induced damage, an intrinsic safeguard not found in synthetic micronozzles. Future studies can further investigate the benefits of using the female mosquito proboscis dispense tip for biomedical drug delivery application.

A preliminary simulation analysis has been carried out to quantify the impact of inner surface defects on the ink flow during printing (fig. S19). Surface defects, adopting the shape of a cavity/crevice, were modeled after the SEM results of the female mosquito proboscis internal structure (fig. S18) to more accurately emulate its natural state. Velocity surface profile results show the impact of surface defects and their promotion of inducing friction during fluid flow to be negligible on the ink’s velocity as it exits the female mosquito proboscis dispense tip. Simulations for perfect surface conditions and defect-ridden surface conditions show a marginal difference in outlet velocity of 0.1% (0.02 μm/s). While these simulation results offer valuable insight into the friction-induced flow behavior in the female mosquito proboscis dispense tip, future studies building on this framework could yield a more comprehensive understanding of how surface roughness influences fluid flow performance in this biological micro dispense tip.

Bio dispense tip fabrication

Laboratory-reared, frozen, uninfected, female mosquitoes (species Aedes Aegypti) were supplied from BEI resources and stored in a freezer at −20°C. At the time of fabrication, the female mosquitoes were removed from the freezer and dipped several times in a beaker containing 80% ethanol to ensure their sterilization.

First, the labium (i.e., the mosquito proboscis’ outer sheath) is detached from the female mosquito [according to the procedure outlined in literature (28)] and discarded. Second, the fascicle is coated with an ultraviolet (UV) curable resin (UV resin by Let’s Resin) by dipping the fine tip of a toothpick into the resin and stroking the ball of resin found at the tip of the toothpick along the female mosquito fascicle. Ensure that the female mosquito proboscis outlet is oriented downward to mitigate gravity-induced flow of liquid resin into the proboscis fluid channel. The resin aggregates together at the center of the fascicle, creating a type of back-stop, leaving the native female mosquito fascicle’s biological material exposed everywhere else along the biological micro dispense tip. Third, a 3-W, 395-nm UV flashlight (UV flashlight by COSMOING, ASIN: B0B49STT1G) is shined on the female mosquito fascicle for 10 s while rotating the fascicle along its long axis to ensure even curing all throughout the resin. The fascicle is then cut on both ends using a sharp razor blade, removing the proboscis from the female mosquito’s head on one end and exposing the opposite end in case of clogging during application of the resin. The fascicle is temporarily set aside; a custom-designed adapter, fabricated using SLA printing and whose function is to ensure concentricity between the engineered dispense tip and the female mosquito proboscis, is placed in the outlet of the 30G engineered dispense tip as schematized in fig. S20. The fascicle is then carefully placed in the outlet of the concentricity adapter. Using a toothpick, the UV resin is then applied on the metal barrel of the dispense tip and spread along its long axis to cover the sidewall of the adapter and the resin bulge present on the fascicle, creating a seal between the commercial dispense tip and the biological dispense tip. Last, the UV flashlight is shined on the freshly applied resin for 10 s while slowly rotating the dispense tip along its long axis, ensuring even curing throughout the resin once again. The scalability of the fabrication procedure has been evaluated and addressed in the Supplementary Materials.

Engineered dispense tips can be repurposed by soaking the metal end of the dispense tip in acetone, followed by injecting compressed air at the inlet of the engineered dispense tip to evacuate the mosquito proboscis from its sealed location. Wipe the outlet end of the 30G dispense tip with any tissue to remove residual acetone and epoxy remaining on the outer walls. Once the outlet of the engineered dispense tip has been cleared of obstruction (validate by extruding any fluid), the engineered dispense tip can be integrated into the next assembly for reuse.

Ethanol-based sterilization is sufficient for bioprinting research due to its effectiveness against viruses and bacteria (68). For biomedical applications, advanced sterilization methods that balance sterilization efficiency with preservation of structural integrity should be considered. In particular, ethylene oxide gas, low-dose ionizing radiation, and hydrogen peroxide vapor are promising options (69). Such treatments may also be applied directly to certain printed products, such as acellular scaffolds post-printing. Future studies are needed to assess the suitability of these methods for specific applications.

Fracture pressure testing

Burst pressure properties of the female mosquito proboscis were measured via the custom DIW printer mounted with a 30-psi pressure transducer. One-milliliter syringes were loaded with deionized water and connected to the pressure transducer using a three-way Luer-Lock adapter. Air was removed from the syringe and tubes to create a pressurized system loaded with a single, noncompressible fluid. Each female mosquito proboscis dispense tip was then mounted to the syringe and water was extruded through, once again ensuring no air was present in the system. Last, with the female mosquito proboscis dispense tip still connected to the syringe, the outlet end of the proboscis was sealed with a high-strength, two-part epoxy (5 Minute Epoxy by Permatex, model number 84111) and left to cure for 10 min (i.e., until a 90% cure was achieved). The extruder mechanism was then set to apply a linear displacement of 0.0167 mm/s on the syringe plunger. The internal pressure of the apparatus was measured until catastrophic rupture of the female mosquito proboscis was achieved.

Rheological characterizations

Rheological properties of inks were measured via rotational rheometer (MCR 302, Anton Paar) with a 25-mm-diameter disposable aluminum parallel plate set to a constant gap distance of 1 mm from the base. All inks were kept at room temperature (~25°C) for 15 min before testing. Oscillatory frequency sweep tests were performed from 0.01 to 1000 Hz to measure the apparent viscosity of the inks and corresponding shear stress as a function of shear rate. To measure the dynamic viscoelastic characteristics of the inks, oscillatory stress sweep tests were performed from 0.1 to 1000 Pa and at a constant frequency of 1 Hz, generating results for storage and loss moduli as a function of shear stress. All tests were conducted at room temperature (~25°C).

Process window characterization

Measurements of the nozzle speed were provided by the motion stage software (automation 1). Measurements of the ink speed were acquired through a calibration procedure, whereby the z-axis motion stage speed was matched by the ink speed. The z-stage speed was set to the desired ink velocity, and the extruder mechanism was dynamically altered until the ideal pressure was reached, creating an equilibrium between ink speed and z-stage speed. If the z-stage speed was greater than the ink speed, the extruded filament would thin and eventually fracture. If the z-stage speed was less than the ink speed, the extruded filament would begin to bend/flow out of plane. See the Supplementary Materials for more information (fig. S13). The printed lines were executed at a layer height of 20 μm ± 3 μm.

Inks and ink synthesis

The ink used for the proof of extrusion demonstration is a ready-to-use, polyethylene oxide–based training bioink purchased and used directly from the vendor (Cellink Start, Cellink). The ink was stored at room temperature, away from light. When preparing to print, the ink was loaded into a 1-ml syringe and centrifuged at 1000 RPM for 30 s to remove air bubbles. The ink used for the honeycomb demonstration and the maple leaf demonstration is a sacrificial, temperature-sensitive, 40% (w/v) Pluronic F-127 in deionized water bioink purchased and used directly from the vendor (Pluronic F-127, Allevi). The ink was stored in a refrigerator at 4°C. When preparing to print, the ink was removed from the refrigerator 10 min before printing and immediately loaded into a 1-ml syringe while the ink was still in liquid state. The ink was set aside to reach room temperature for 10 min.

The ink used for the first cell-laden grid demonstration is Pluronic F-127 bioink with B16 cancer cells suspended in solution. B16 murine melanoma cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, 0.1% amphotericin, and 0.1% gentamicin reagent. To maintain the cells, cells were cultured at 37°C in 5% CO₂. Cells were passaged at 80% confluency and cells of passages 5 to 8 were used in bioprinting experiments. Pluronic F-127 of molecular weight of 12,600 g/mol (Pluronic F-127, Sigma-Aldrich, product no. P2443) was dissolved in complete cell medium at 35 wt % and allowed to dissolve overnight at 4°C. On day of printing, cells were washed with phosphate-buffered saline (PBS), trypsinized, then labeled before use in printing. Cells were labeled with CellMask Plasma Membrane Stain Orange (CellMask Plasma Membrane Stains, Thermo Fisher Scientific, catalog no. C10045). For the staining procedure, cells were centrifuged at 1400 RPM for 5 min, resuspended in 1× staining solution in complete medium and incubated at room temperature for 10 min. Following this, cells were centrifuged, and washed in PBS twice, then resuspended in warmed 4% paraformaldehyde in PBS and fixed for 10 min. Cells are subsequently centrifuged and washed thrice, counted, and split to desired quantity for printing. Cells are pellets and resuspended in chilled Pluronic F-127/medium solution at a density of 4 million cells/ml. The cell solution is pipette-mixed on ice before being loaded into 1-ml syringes for printing.

The ink used for the second cell-laden grid demonstration is Pluronic F-127 bioink embedded with RBCs. Pluronic F-127 (Sigma-Aldrich) was dissolved in PBS at 29 wt % and allowed to dissolve overnight at 4°C. On the day of printing, RBCs were isolated from whole bovine blood (Lampire) by centrifugation at 1350 relative centrifugal force (RCF) for 5 min at 4°C. The plasma supernatant was removed, and the RBC pellet was washed three times by resuspending in ninefold excess chilled PBS, followed by centrifugation at 1350 RCF for 5 min at 4°C, and removal of the supernatant each time.

N-hydroxysuccinimide (NHS)–fluorescein (Thermo Fisher Scientific, 46410) was dissolved in anhydrous dimethyl sulfoxide at 25 mg/ml before use. Isolated RBCs were resuspended in PBS at 10% hematocrit and combined with the NHS-fluorescein stock solution to achieve a final concentration of 0.2 mg/ml fluorescein reagent. The cell suspension was incubated for 1 hour at room temperature under gentle agitation.

Labeled RBCs were then resuspended at a 10% volume fraction in the chilled 29% Pluronic F-127 solution. This corresponds to an estimated 2 billion cells/ml in the print, calculated with an observed hematocrit of 40% and assumed 8 billion RBC/ml in the bovine whole blood (70). The cell suspension was thoroughly mixed by pipetting on ice before being loaded into 1-cc syringes for subsequent printing.

The ink used for the cell viability experiments is Pluronic F-127 bioink with B16 cancer cells suspended in solution. Five microliters of calcein-AM (component A) and 20 μl of ethidium homodimer-1 (component B) were added to 10 ml of Dulbecco’s PBS to prepare a Live/Dead (Thermo Fisher Scientific, USA) fluorescent PBS solution. Pluronic F127 was dissolved in this Live/Dead PBS solution at a final concentration of 29% (w/v) and sterilized by filtration through a 0.22-μm membrane filter at 4°C. B16 cells were then mixed with the fluorescent F127 solution to achieve a final cell density of 5 × 10⁶ cells/ml. The hydrogel ink used to conduct the conceptual drug delivery experiment is 4% (w/v) low–molecular weight alginate solution (90 kDa, KIMICA, Japan) prepared in sterile DMEM (high glucose, Sigma-Aldrich).

Printing of bioinks

The paths used during the execution of the process window and the demos were manually created through the simultaneous integration of G-code and AeroScript programming. AeroScript is the custom software language used by Aerotech to control their XYZ gantry systems, which is also capable of accepting most G-code commands. Printing parameters (ink speed and nozzle speed) for all three demos were applied on the basis of the process window results. The print speed was 20 μm/s, the ink speed was approximately 14 μm/s and the layer height varied between 15 and 20 μm, depending on the variation of the proboscis in use. All demos were printed on standard 75 mm × 25 mm × 1 mm glass slides with their individual female mosquito proboscis dispense tips and constrained to nanoliter structures to minimize fabrication time (see the Supplementary Materials). All printing was executed in ambient conditions: temperature between 20° and 30°C and relative humidity between 30 and 70%.

Cell viability testing

Following printing of the Pluronic F-127 bioink with B16 cancer cells suspended in solution, the printed structures were immediately covered with DMEM high-glucose complete medium and incubated at 37°C with 5% CO₂ for 6 hours. After incubation, the medium was carefully removed, and the samples were imaged using an EVOS M5000 microscope (Thermo Fisher Scientific, USA).

Investigation of lifespan

Lifespan of the female mosquito proboscis dispense tips were conducted by storing the female mosquito proboscis dispense tips in ambient conditions (room temperature and standard atmospheric pressure) and verifying their extrusion capabilities at intervals of 24 hours. The tests included 10 samples, and extrusion validation was performed with deionized water as the ink and recorded using a video camera mounted with a microscope lens for additional magnification.

SEM characterization

SEM images were generated via a scanning electron microscope (SU3400, Hitachi). Once each print was complete, structures were flashed with liquid N₂ and immediately placed in a freeze drier for 2 to 3 days. After all humidity had been removed from the sample, the structure was mounted on a sputter coater (Q150TS, Quorum) and sputtered with platinum at 4-nm thickness.

Confocal imaging

Images were taken using a confocal laser scanning microscopy (LSM 800, ZEISS). B16 cancer cell sample imaging was conducted at a magnification of 10×. Z-stacks were taken to capture all cells within the gel, taken over a range of 30 μm with a distance of 10 μm between each layer. Z-stacks were collapsed to 2D images as maximum intensity z-projections using ImageJ software. RBC sample imaging was conducted at magnifications of 10×, 20×, and 63×.

Funding:

C.C., J.P., Z.Z., and L.Z. acknowledge funding support from New Frontiers in Research Fund Exploration program (NFRFE-2022-00384), Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery program (RGPIN-2021-02664), FRQNT New Academics program (FRQ-NT NC-298727), and Canada Foundation for Innovation John Evans Leaders Fund (CFI-G258039-41166). J.L. acknowledges the Canada Research Chair Program and NSERC-FRQNT Nova grant (FRQ-NT 2024-NOVA-344903).

Author contributions:

Conceptualization: J.P., A.A., J.L., and C.C. Methodology: J.P., Z.Y., E.J., Z.H., J.L., and C.C. Investigation: J.P., Z.Y., E.J., Z.Z., X.L., L.Z., H.H., J.L., and C.C. Visualization: J.P., E.J., Z.Z., X.L., H.H., J.L., and C.C. Resources: J.P., Z.H., A.A., M.A.C., J.L., and C.C. Supervision: J.L. and C.C. Validation: J.P., J.L., and C.C. Formal analysis: J.P., Z.Y., J.L., and C.C. Software: J.P. and Z.H. Project administration: J.P., J.L., and C.C. Funding acquisition: J.L. and C.C. Writing–original draft: J.P., E.J., J.L., and C.C. Writing–review and editing: J.P., Z.Y., E.J., L.Z., Z.H., A.A., J.L., and C.C.

Competing interests:

The authors declare that they have no competing interests.

Data and materials availability:

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

The PDF file includes:

Other Supplementary Material for this manuscript includes the following: