Cation-crosslinked κ-carrageenan sub-microgel medium for high-quality embedded bioprinting

Three-dimensional (3D) bioprinting embedded within a microgel bath has emerged as a promising strategy for creating intricate biomimetic scaffolds. However, it remains a great challenge to construct tissue-scale structures with high resolution by using embedded 3D bioprinting due to the large particle size and polydispersity of the microgel medium, as well as its limited cytocompatibility. To address these issues, novel uniform sub-microgels of cell-friendly cationic-crosslinked kappa-carrageenan (κ-Car) are developed through an easy-to-operate mechanical grinding strategy. These κ-Car sub-microgels maintain a uniform submicron size of around 642 nm and display a rapid jamming-unjamming transition within 5 s, along with excellent shear-thinning and self-healing properties, which are critical for the high resolution and fidelity in the construction of tissue architecture via embedded 3D bioprinting. Utilizing this new sub-microgel medium, various intricate 3D tissue and organ structures, including the heart, lungs, trachea, branched vasculature, kidney, auricle, nose, and liver, are successfully fabricated with delicate fine structures and high shape fidelity. Moreover, the bone marrow mesenchymal stem cells encapsulated within the printed constructs exhibit remarkable viability exceeding 92.1% and robust growth. This κ-Car sub-microgel medium offers an innovative avenue for achieving high-quality embedded bioprinting, facilitating the fabrication of functional biological constructs with biomimetic structural organizations.


Introduction
The treatment of human tissue damage and organ loss urgently requires the development of organ analogs with the physiologically relevant architectures.Tissue engineering has the potential to create organ analogs by combining biomaterials, cells, and bioactive factors, but it often lacks the fine structural cues necessary to comprehensively mimic human tissues and organs.Consequently, three-dimensional (3D) bioprinting has been increasingly used to replicate the 3D geometry of organs and precisely deposit cells within these organ analogs.Meanwhile, polymer hydrogels are the preferred choice for cell scaffold and printing bioink owing to their similarity to the extracellular matrix and suitability for cellular encapsulation and bioprinting [1,2].Nonetheless, hydrogels inherently comprise a substantial water content and are soft, frequently leading to diminished resolution of printed layers and the rapid collapse of tissue and organ-scale constructs during conventional bioprinting.Ensuring the shape fidelity of hydrogel scaffolds throughout the printing process is crucial for the production of organ analogs with intricate 3D geometries.
Considerable efforts have been made to enhance 3D stability and shape fidelity of printed hydrogels, such as printing multi-material inks [3], employing photo-crosslinkable hydrogels [4], adding thickeners [5], and adopting embedded printing within sacrificial supporting materials [6].Particularly, embedded 3D bioprinting deposits bioink into a matrix composed of granular hydrogels, which provides structural support throughout the printing process [7][8][9][10].The granular hydrogels consist of jammed hydrogel microparticles that exhibit a shear-yielding response [11][12][13].During the embedded 3D bioprinting process, the jammed microgels behave as viscous fluids under shearing forces, facilitating the smooth movement of the printing nozzle within the bath, and enabling precise bioink deposition.Once the external load is released, the granular hydrogels immediately return to a viscoelastic solid state, providing crucial support and preventing the collapse or deformation of the extruded ink materials.Moreover, the granular hydrogel bath can be subsequently removed, leaving behind the desired construct after printing.A wide range of materials, including gelatin [14,15], agarose [16], cellulose [17,18], gellan gum (GG) [19,20], Pluronic F127 [21], polyvinyl alcohol [22], and carbomer [23], have been successfully convert to microgel bath to support the 3D embedded printing, greatly expediting the widespread application of the soft materials in the fields of tissue engineering and regenerative medicine.
Since the microgel bath serves as a critical component that directly interacts with the bioink and the encapsulated cells during the bioprinting process, two substantial considerations arise concerning granular hydrogel baths.Firstly, the size and dispersity of the microgels are pivotal determinants for achieving high-resolution bioprinting within the support medium [24].Numerous studies have demonstrated that larger and unevenly distributed microgels often result in the formation of enlarged gaps and voids amidst the microgel matrix [7,25,26].As a consequence, the fluid inks tend to diffuse into these spaces, inevitably leading to compromised print resolution and irregular morphology.Therefore, achieving a small-scale and uniform microgel bath becomes imperative for enhancing resolution in microgelsupported 3D printing.For example, the commonly used GG microgel bath, produced through mechanical grinding, typically showed hundreds of particle sizes and polydispersity due to the aggregation caused by hydrogen bonding and ionic interactions among GG microgels [27][28][29].The printing precision within this bath was constrained to a few hundred micrometers.To enhance print resolution, Xie et al introduced trisodium citrate (TSC) into GG granular gels to suppress particle aggregation based on the Hofmeister effect, resulting in the formation of GG-TSC microgels with a reduced average size (≈30 µm) and a uniform distribution [29].The print resolution was greatly enhanced, enabling the reliable printing of collagen filaments as fine as 25 µm.Therefore, granular hydrogels with small particle sizes and a homogeneous microgel morphology are conducive to achieving high printing precision.Secondly, the cytocompatibility of microgel components, including the hydrogel material itself, crosslinking agents, and any additives, is critical to preserve the viability of embedded cells during the cell-laden bioprinting.Evidently, the addition of TSC in the GG-TSC bath has an adverse effect on cell viability, rendering the GG-TSC microgel bath unsuitable for cellladen bioprinting.Therefore, these two aspects have been main challenges for a well-designed granular gel bath.Additionally, because of the requirement of removing the printed constructs from supporting medium after the embedded bioprinting process, the use of efficient extraction methods while minimizing any negative effects on the printed constructs is essential [19].
To address these challenges, we present a novel suspension medium of cationic-crosslinked kappacarrageenan (κ-Car) sub-microgels with a uniform morphology for high-quality 3D embedded bioprinting.κ-Car consists of alternating disaccharide repeating sulfate groups units of 3-linked-β-D galactopyranose and 4-linked-α-D-galactopyranose or 4-linked-3,6-anhydro-α-D-galactopyranose.κ-Car aqueous solution can undergo reversible sol-gel transition upon temperature stimuli or the addition of cations (such as K + and Ca 2+ ) through the formation of a 'double-helix' structure.In particular, κ-Car is dissolved in phosphate buffer saline (PBS, pH 7.4), a cell culture medium or a calcium chloride (CaCl 2 ) solution and then cooled to form the carrageenan-cation hydrogel (schemes 1(a) and (b)).Subsequently, the bulk κ-Car hydrogel is controllably broken into a sub-microgel slurry (scheme 1(c)).The sub-microgels could self-assemble into granular gels through extensive cationic interactions and hydrogen bonding at the particle interface (scheme 1(d)).Attributing to the reversible linkages between sub-microgels, jammed κ-Car gels behave as Bingham flow behavior with typical shear-thinning and rapid self-healing properties.Consequently, this new sub-microgel bath enables the complete suspension of printed bioinks, facilitating the fabrication of complex tissue structures with high precision and great shape fidelity through the freedom embedded 3D bioprinting (scheme 1(e)).Moreover, the κ-Car sub-microgel medium is devoid of any potentially toxic components, ensuring the high cyto-compatibility of cell-laden bioprinting.The bioprinted constructs can be easily released from the support bath through water washing (schemes 1(f) and (g)).This versatile support medium provides a novel avenue for achieving high-quality fabrication of large and complex tissues and organs.

Experimental section
Preparation and microstructure characterization of κ-Carrageenan support bath: κ-Carrageenan (κ-Car, Aladdin, China) was mixed into phosphate buffered saline (PBS, pH 7.4) or 0.1% (wt/vol) calcium chloride (CaCl 2 , Aladdin, China) solution at 70 • C until the polymer was fully dissolved.Subsequently, the κ-Car solution was incubated at 4 • C for 12 h to make gel.To create the slurry of κ-Car particles, κ-Car gels were disrupted at 25 • C by continuous stirring at 1200 rpm for 1 h using an overhead stirrer.The resulting κ-Car slurry was degassed by centrifugation at 1000 rpm for 3 min to produce support bath.To modulate the viscoelastic properties of floating bath, the concentrations of κ-Car were varied from 0.3% to 0.6% (wt/vol).As control groups, the suspension baths including carbomer, carboxymethyl cellulose, and gelatin microgels, were prepared following previous reports [14,30].
To measure size, uniformity, and distribution of microparticles, the compacted κ-Car slurry was diluted in a washing solution and stained with calcein (aladdin, China).Microparticle diameter distribution was analyzed using a laser nanometer particle size analyzer (Litesizer 500, Anton Paar, Austria), while confocal laser scanning microscope with a z stack scanning (CLSM, STELLARIS 8, Leica, Germany) was employed for imaging.The particle size of κ-Car sub-microgel was further determined by transmission electron microscope (TEM, HT7800, Hitachi, Japan).The optical transmission of compacted κ-Car slurry was analyzed via an ultraviolet-visible spectrophotometer (UV3200S, Jingke, China).Furthermore, the compacted κ-Car slurry was subjected to freezedrying and subsequently examined using scanning electron microscopy (SEM, TM4000 IIplus, Hitachi, Japan).
Rheological characterization of κ-Carrageenan floating bath: The rheological properties of the κ-Carrageenan floating bath were investigated by using a rheometer (Discovery HR-2, TA Instruments, USA), equipped with a 40 mm diameter plate with a 1000 µm separation at 25 • C. To measure the jammed behavior, a dynamic frequency sweep in the range of 0.1-100 rad•s −1 was performed at a strain of 1%.To analyze the viscosity and yield stress of κ-Car slurry, a shear rate sweep varied from 0.01 to 100 s −1 were carried out to measure viscosity and yield stress.To measure self-recovery behavior, the step shear tests were performed by using 0.1 s −1 and 10 s −1 shear rates for 60 s, respectively for three cycles.The thermo-sensitivity of the granular κ-Car gels was measured by using a temperature sweep ranging from 5 Bioink preparation: The composite bioink for high-precision bioprinting comprised gelatin methacrylate (GelMA, EFL, China) and sodium alginate (SA, Aladdin, China).GelMA and SA were dissolved sequentially in deionized water at 40 • C, with concentrations of 5% and 1.5% (wt/vol), respectively.In addition, two typical bioinks, SA and GelMA were used in this study.SA bioink was prepared by dissolving SA at a concentration of 2.0% (wt/vol) in deionized (DI) water.The 10% (wt/vol) GelMA was mixed with 0.2% (wt/v) photoinitiator lithium phenyl(2,4,6-trimethylbenzoyl) phosphinate (Aladdin, China) and phosphate buffer solution (pH 7.4) to achieve a GelMA concentration of 10% solution.The rhodamine grafted GelMA was employed to facilitate bioprinting visualization.To create the cell bioink, bone mesenchymal stem cells (BMSCs)derived rabbit spine of 10 million cells ml −1 were gentle mixed into sterilized GelMA solution.Notably, apart from printing for cell loading, all printing performance tests were conducted using rhodaminelabeled GelMA or GelMA/SA composite inks.
Embedded 3D bioprinting of bioinks: The 3D bioprinting process using a 0.6% (wt/vol) granular κ-Car gel, unless specified otherwise, involved embedding GelMA/SA, GelMA and SA.Our custom-designed 3D bioprinting system was utilized for this purpose.All digital models were either generated in Rhino software or downloaded from the 3D database (www.thingiverse.com;https://3d.nih.gov/.).The downloaded models included heart, lungs, trachea, spiral coil, branched vasculature, auricle, nose, kidney and liver.The dimensions of these digital models were adjusted using Pango software (Panowin, China) to enhance printing accuracy.Subsequently, all 3D models were exported in STL format and processed in Pango to generate G-code instructions for the printer.For the printing parameters, slicer settings were speeds of 10-90 mm•s −1 , 1 perimeter, and 30%-80% line infill.Prior to printing, the bioink was loaded into a syringe and attached to the bioprinter, while the support bath was filled in a container that could adequately contain the printed construct.Needles with inner diameters of 160 µm and 60 µm were fitted to the syringe, respectively.The needle was manually positioned at the center of the container in the xy plane and lowered to 1 mm above the bottom of the print container.All constructs were printed at room temperature (25 • C).Following the completion of the printing process, the constructs within the κ-Car bah underwent crosslinking by using a 405 nm blue light.Finally, the printed constructs were carefully released, and the κ-Car support material was removed through repeated washing with PBS (pH 7.4).
Cell culture: The rabbit BMSCs were cultured in a humidified incubator at 37 • C with 5% CO 2 .The culture medium was composed of low-glucose Dubecco's Modified Eagle Medium (DMEM, VivaCell, China) supplemented with 10% fetal bovine serum (FBS, VivaCell, China) and 1% penicillin/streptomycin (Solarbio, China).The medium was replaced every 2 d.To evaluate the cytocompatibility of κ-Car sub-microgel bath, a transwell co-culture assay was initially implemented.Transwell plates (LABSELECT® Tissue Culture Plate Inserts, 12-well Inserts/Plate) with 0.4 µm pore filters on the bottom of the upper compartment were used.κ-Car sub-microgels were placed in the upper compartment of the transwell, while BMSCs were seeded in the lower compartment.After incubation for 24 h, cells were assessed by using CCK-8 assay according to the protocol.In addition, the cell morphology was visualized using laser scanning confocal microscopy.
For the assessment of cell-laden embedded bioprinting, the BMSCs density encapsulated in the GelMA ink was 10 million cells/ml.3D grid constructs (designed size: 15 × 15 × 2 mm 3 ) and kidneylike constructs (designed size: 22 × 20 × 2 mm 3 ) were printed within the κ-Car sub-microgel bath, with sterile printing performed inside a biosafety cabinet.The layer thickness and interval between filaments were 200 µm and 400 µm, respectively.The printed constructs were collected on days 1 and day 3 for the cell viability evaluation and on days 5 and 7 for spreading observations.
Live/dead assay: To assess cell viability, the printed cell-laden constructs were incubated with calcein-AM/propidium iodide (Beyotime, China).After a 45 min incubation at 37 • C, the construct underwent three PBS washes and fluorescent images were captured using confocal laser scanning microscopy (CLSM).
Fluorescence staining: Cell-laden constructs cultured at days 2 and 3 were fixed with 4% paraformaldehyde (Beyotime, China) for 30 min, followed by permeabilization using 0.1% Triton X100 (Beyotime, China) for 30 min.After washing with PBS for three times, cells were stained with FITC-phalloidin to label the actin cytoskeleton (Solarbio, China) for 60 min.The samples were further washed by using PBS for 3 times.Subsequently, cells were stained with DAPI to visualize the cell nucleus (Solarbio, China).Finally, the constructs underwent three PBS washes to remove the residual stains.All procedures were done at room temperature.Fluorescent images were gathered by using a CLSM.

Preparation and rheological characterization of jammed κ-Car Sub-microgels
The jammed κ-Car sub-microgel bath was generated by mechanically breaking up the bulk hydrogels into a sub-microgel slurry.As an instance, the bulk κ-Car hydrogel was initially formed by dissolving κ-Car powder in PBS or CaCl 2 solution at 70 • C, followed by gelation at 4 • C via the formation of a double-helix structure with K + or Ca 2+ ions.Subsequently, the bulk κ-Car hydrogel was ground into sub-microgels using vigorous mechanical blending at 1000 rpm.Longer blending times resulted in smaller microparticle sizes, with a blending time of 60 min yielding sub-microparticles with an average diameter of 642 ± 65 nm (figure 1(a)), which was much smaller than those of microgels reported in previous research (table S1) [7,14,25,27,29,31].Notably, multiple factors, including concentrations of κ-Car and CaCl 2 , gelation time and temperature, stirring speeds and duration, as well as degassing speeds and duration, directly impact the diameter of κ-Car microgels (table S2).Following degassing through centrifugation at 1000 rpm, the particulates assembled into a granular gel through hydrogen bonding and cation interactions between the particles.The κ-Car submicrogels exhibit a submicron-level and homogenous morphology, as confirmed by CLSM and scanning electron morphology (SEM) (figures 1(b)-(d)).Such small particle size and uniform dispersity are expected to improve the print resolution.Additionally, the granular κ-Car gel bath displays high transparency, allowing real-time observation of the printing process (figures 1(e) and (f)).After photo-crosslinking using 405 nm irradiation, excess sub-microgels surrounding the construct could be gently pipetted away, followed by a PBS washing step to fully release the printed constructs (figures 1(g)-(j)).
Importantly, granular gel media used in highquality embedded bioprinting possess unique characteristics that enable them to fully suspend printed bioinks.Initially, granular gel media exhibit jammed solid-like behaviors when subjected to minimal or no stress.The application of stress surpassing a critical threshold, known as the yield stress, triggers a transition to an unjammed, liquid-like state and initiates flow within the media.Furthermore, after the disturbance of the microstructure of a suspension medium by a passing nozzle and its displacement by deposited material, the microstructure spontaneously recovers.This self-healing capacity permits the medium to transition from a fluidized state back to a solid-like state, effectively suspending the deposited material.To assess the suspending capability of the κ-Car sub-microgel bath, the jamming state of the κ-Car sub-microgel bath was initially evaluated by rheological measurement at 25 • C. Frequency sweep test shows that, the storage modulus (G ′ ) of κ-Car, ranging in concentrations from 0.3 to 0.6% (wt/vol) surpasses the loss modulus (G ′′ ) across the entire frequency spectrum from 0.1 to 100 rad•s −1 , indicating a jammed gel behavior (figure 2(a)).Moreover, the G ′ values corresponding to κ-Car concentrations of 0.3, 0.4, 0.5, and 0.6% (wt/vol) exhibit significant increments from 7.3 to 47.8, 103.1, and 116.5 Pa at 10 rad•s −1 , thereby providing a tunable window for embedded bioprinting.Meanwhile, the G ′′ values progressively increase in tandem with rising frequencies, implying that the sub-microgel bath could readily release energy to adapt to disruptions during rapid printing.
Granular gels designed for 3D embedded bioprinting enable facile flow under a needle shear once the yield stress threshold is exceeded.The representative yield behavior and shear thinning of the granular κ-Car hydrogels were further studied using a flow test.As shown in figure 2(b), with the increasing of shear rate from 0.01 to 100 s −1 , the shear stresses for all granular κ-Car hydrogels remain relatively constant (representing yield stress) when the shear rates are below ≈0.2 s −1 .Subsequently, they exhibit a linear increment, indicative of the flow behavior resembling that of a Bingham fluid.The yield stress of the granular κ-Car hydrogels incrementally elevates from 0.05 to 0.08, 0.09 and 2.22 Pa as the κ-Car concentrations ascend from 0.3 to 0.4, 0.5 and 0.6% (wt/vol), respectively.This augmentation in yield stress is attributed to inherent hydrogen bonding and cation interactions between particles.Similarly, the initial viscosity increases from 82.5 to 2778.1 Pa•s when increasing the κ-Car concentration from 0.3%  to 0.6% (wt/vol), but decreases by nearly three orders of magnitude as the shear rates increases from 0.01 to 100 s −1 , representing a typical shear-thinning behavior.This attribute is ascribed to the continuous shear that disrupts the interactions between the particles and could induce alignment of the κ-Car sub-microgels to facilitate flow.
After the bioink was dispensed, the flow should immediately discontinue to prevent the diffusion, deformation or collapse of the extruded bioinks.To quantitatively estimate the self-healing capacity of κ-Car sub-microgel bath, shear recovery tests upon continually cyclic shearing at alternate 10 s −1 and 0.1 s −1 rates were conducted (figure 2(c)).At low shear rate, the κ-Car sub-microgel bath exhibits a high viscosity (21 Pa•s −1 ) that sharply decreases to a low viscous state (0.5 Pa•s −1 ) under high shear rate.In contrast, as the shear rate is fallen to 0.1 s −1 , the viscosity recovers to about 20 Pa•s −1 within 5 s, which suggests that the granular κ-Car hydrogels could rapidly reform the physical cross-links between the micro-particulates.Such shear-thinning and self-recovery behaviors did not appreciably change for many cycles, which is important for embedded printing processes as it helps facilitate the smooth movement of the nozzle and the retention of shape following deposition.
Because of the inherent thermo-sensitivity nature of cationic-crosslinked κ-Car hydrogels, alterations in temperature have the potential to induce the disintegration of microgels, thereby compromising the stability and reliability of the printing process.Consequently, the thermal stability of the granular κ-Car gels was assessed via temperature sweeps ranging from 5 • C to 40 • C. As shown in figure 2(d), the lower temperatures induced more elastic κ-Car gels with higher G ′ values, while higher temperatures trigger a reduction in G ′ .This sensitivity to temperature variations diminishes as the concentration of κ-Car gels increases.For instance, the G ′ values of 0.3% (wt/vol) κ-Car decrease by over three orders of magnitude as the temperature increases from 5 • C to 40 • C, whereas those of 0.6% (wt/vol) κ-Car only drop by nearly one order of magnitude.Interestingly, we observed that this characteristic of decreasing stiffness with increasing temperature not only has a minor impact on the performance of embedded printing but also contributes to the easy removal of the supporting bath.

Embedded 3D printing in jammed κ-car sub-microgel medium
The small and uniform morphologies of the granular κ-Car gels, coupled with their shear-thinning, rapid self-healing, and yield stress, render them conducive to embedded 3D bioprinting.Considering that the low-viscosity support bath potentially disturbs the printed constructs during rapid printing, we opted to use the highest concentration of 0.6% (wt/vol) sub-microgels to showcase the fidelity of ink deposition within the support bath.A common bioink consisting of gelatin methacrylate of 5% (wt/vol) and sodium alginate of 1.5% (wt/vol) composites (GelMA/SA) was utilized for printing within the κ-Car medium [32,33].For better visualization, the GelMA was labeled by rhodamine [34].Microextrusion was performed using needles with inner diameter of 160 µm and 60 µm, while maintaining a constant bioink flow rate of 8 µl•min −1 .The fidelity of printing, influenced by the support medium, was evaluated by comparing the average filament diameter and cross angle across nozzle movement speeds ranging from 20 to 60 mm•s −1 .Upon the printing completion, the κ-Car supporting bath containing the printed cross-filaments was cured using a 405 nm blue light.Figure 3(a) presents the printed structures of GelMA/SA cross-filaments at printing speeds ranging from 20 to 60 mm•s −1 .It is noteworthy that all GelMA/SA filaments within the κ-Car bath maintain well-defined shapes without boundary diffusion.As the nozzle movement speed increases from 20 to 60 mm•s −1 , the printed line diameter through 160 µm and 60 µm micro-extrusion decreases from approximately 305 µm to 166 µm and from 98 µm to 33 µm, respectively (figure 3(b)).This reduction is attributed to the diminished ink extrusion at higher speeds when printing lines of equal length.Moreover, the acute angle between two filaments at the cross point is approaching 90 • , indicating high printing fidelity.These fine and vertical filaments were similar to those printed in carbomer and demonstrated significant improvement compared to those printed in cellulose and gelatin microgels (figure S1).These results demonstrate that it is possible to successfully fabricate high-resolution structures within the κ-Car bath by solely adjusting the printing parameters, without the need to change the needles.
To further illustrate the supportive capabilities of the κ-Car system, a high-resolution 3D grid construct (designed size: 15 × 15 × 2 mm 3 ) with a 150 µm layer thickness was subsequently printed by using the κ-Car sub-microgel bath.The speed of 40 mm•s −1 was adopted for printing.As illustrated in figures 3(c) and (d), the grid maintains high integrity and resolution, featuring individual filaments of approximately 46 µm and filament spacing of around 300 µm, as confirmed through CLSM imaging.Moreover, the 3D grid construct exhibits a network of pores and pore interconnectivity (figure 3(e)), critical factors that influence printability due to their impact on various cellular responses, including cell viability, growth, and even differentiation.
Compared to the relatively simplified arrangements of microfibers and grids, tissues and organs possess intricate architectures.Consequently, constructing biomimetic tissue-scale structures remains a significant challenge.It is particularly evident when dealing with complex organs such as the heart, which features highly curved surfaces and intricate internal architectures, thereby posing a formidable obstacle for bioprinting.To evaluate the capabilities of the κ-Car sub-microgel bath in facilitating the printing of complex organs with intricate details, we created a human heart model using the GelMA/SA ink at a high infill density of 50%.As shown in figure 4(a), the printed heart model faithfully replicates the 3D anatomical shape, complete with large veins and arteries.Moreover, this heart-like construct exhibits resilience against mechanical loading and could repeatedly revert to its original shape, simulating the behavior of a beating heart (figure S2).To highlight the internal structures, micro-computed tomography imaging was employed to verify the accurate reproduction of the left and right ventricles the 3D heart model (figure 4(b)).A cross-section of the heart reveals that the fine internal features of the print are accurately recreated (figures 4(d) and (e)).The print resolution was quantified using the CLSM, showing an average diameter of 45 µm and an inter-space of 270 µm (figure 4(f)).These fine features also include arteries (figure 4(g)) and tri-leaflet heart valve (figure 4(h)), which exhibits well separated leaflets, as demonstrated in microcomputed tomography imaging (figures 4(i) and (j)).Furthermore, a heart with a thickness of only 0.5 mm was successfully printed (figure 4(k)).When observed under 405 nm UV irradiation, this heart model revealed more essential anatomical structures and hollow chambers (figure 4(l)).Subsequent to heating to 37 • C and thorough washing with abundant PBS, the k-Car in the model was completely removed from the constructs (figure 4(m)).To enhance visualization, a blood-mimicking dye solution was injected into the right and left ventricles through the aorta, demonstrating both the fidelity of pattern replication and the patency of the channels (figures 4(n) and S3).Although the currently prepared model remains considerably distant from achieving the operational equivalence of an authentic human heart, the developed κ-Car sub-microgel bath could potentially offer a viable and promising strategy for the in-vitro fabrication of intricate tissues and organs.
The heart functions as a vital pump responsible for propelling blood throughout the entire body, while the kidneys play a continuous role in blood filtration, removing metabolic waste products, and regulating the body's fluid balance through hormone secretion.Kidneys are highly intricate organs, comprising approximately 1 million functional units referred to as nephrons.Consequently, the development of kidney-like constructs with both micro and macro precision and resolution presents a significant challenge.To further demonstrate the capabilities of high-resolution printing and the potential for engineering organ-scale scaffolds, we successfully printed a kidney-like construct using GelMA/SA inks through the κ-Car sub-microgel bath.As shown in figures 5(a) and (b), the printed kidney construct faithfully replicates the external surface structure characteristic of a natural kidney.Furthermore, the printed scaffold architecture features an intricate arrangement of interconnected networks, accurately mimicking the renal vasculature and tubular system found in the natural kidney, including arteries, ureter, veins, and pelvis renalis (figure 5(c)).Additionally, a biomimetic lung structure with high-precision trachea is successfully printed (figures 5(d)-(g)), further demonstrating the excellent support capacity of κ-Car bath.To accentuate the exceptional fidelity, we freely constructed a helical coil and branched vasculature (figures 5(h) and (i)).The branched vasculature distinctly showcases the patency of the channels through vessel perfusion using a blood-mimicking dye solution (figure 5(j)).
Besides employing the GelMA/SA composite as a representative bioink for high-resolution printing, both GelMA and SA were chosen as illustrative examples of photo-crosslinking and ionic crosslinking to highlight the facile and versatility of granular κ-Car hydrogel bath.Figure S4(a) presents the printed structures of 10% (wt/vol) rhodamine-labeled GelMA cross-filaments at printing speeds ranging from 10 to 90 mm•s −1 .Although these filaments show slight deformation caused by a combination of factors such as GelMA surface tension, diffusion, extrusion swell, and vibration during the rapid printing process, GelMA filaments within the κ-Car bath maintain well-defined shapes and vertical cross-angle (no significant difference, figure S4(b)), demonstrating high printing adaptability.As the nozzle movement speed increases from 10 to 90 mm•s −1 , the printed line diameter decreases from approximately 395 µm to 78 µm (figure S4(c)).Moreover, the printed GelMA grid features high integrity and resolution, with individual filaments measuring approximately 200 µm (figure S5).On the other hand, a 2.0% (wt/vol) SA ink was printed into the κ-Car bath containing 0.1% (wt/vol) CaCl 2 to form a partially cross-linked alginate hydrogel to avoid the diffusion of SA.As shown in figure 5(k), various exemplary complex shapes including the auricle, nose, kidney and liver are mimicked by using SA inks.The printed auricle-like structure accurately depicts the helix, antihelix, earlobe, and ear screen.The noselike construct displays clearly visible symmetrical nasal cavity.Excellent supporting property and fidelity were also observed in the printed kidney and the liver, whereby the printed feature each contains veins and arteries.Overall, the resolution and fidelity of these printed structures are well maintained, regardless of the nature of the ink polymer or its crosslinking mechanism.Such exceptional versatility holds significant promise for embedded bioprinting as it effectively caters to a wide range of soft and low-viscosity bioinks.

Jammed κ-car sub-microgels for high-viability cell-laden bioprinting
Herein, the newly developed granular κ-Car gel bath, free of any toxicity (figure S6), could be easily sterilized through precursor filtration and effortlessly removed by washing with PBS or cell culture medium after bioprinting, thereby enhancing its suitability for cell-laden bioprinting application.To demonstrate the feasibility of the κ-Car bath in cell-laden embedded bioprinting, GelMA/SA-loaded BMSCs grids with a cell density of 1 × 10 7 cells/ml and dimensions of 15 × 15 × 2 mm 3 were printed.Considering that small nozzles tend to cause significant cell damage during rapid printing, microextrusion was performed using a needle with an inner diameter of 160 µm and a bioink flow rate of 8 µl•min −1 .Cell viability was firstly investigated using a standard LIVE/DEAD staining assay.The resulting live (green)/dead (red) images show that BMSCs within the GelMA/SA constructs display an isolated growth distribution (figure S7).The cell viability is 91.5 ± 1.2% and 92.4 ± 2.3% on day 1 and day 3, robustly affirming the high safety of both the embedded bioprinting process using κ-Car and the subsequent crosslinking procedures.Moreover, parts of live cells have extended pseudopods on day 3 (figure S8).To further observe the cell morphology of the GelMA/SA constructs, phalloidin (green)/DAPI (blue) images were captured after 5 and 7 d of cell culture.As illustrated in figure 6(a), BMSCs started to spread on day 5 and became more significant after 7 d of culture.Connections with adjacent cells had already been established on day 7.Most cells even exhibit alignment along the microfibers owing to topographical cues, indicating the remarkable potential for generating organized tissues.
Kidney disease has emerged as a pressing public health issue, with a substantial number of individuals currently awaiting organ transplantation.Presently, the primary treatment options for these patients consist of dialysis and organ transplantation, although the latter is significantly constrained due to the limited availability of donor kidneys.To address the shortage of kidneys, 3D bioprinting has been employed to fabricate engineered kidneys.Herein, a kidney-like construct with dimensions of 22 × 20 × 2 mm 3 was printed in the κ-Car submicrogel bath (figure 6(b)).The BMSCs in the construct similarly show a high cell viability of 92.1% on day 1 and exhibit substantial spreading on day 3.The engineered kidneys hold great potential for applications in disease modeling and drug testing, addressing the pressing need for effective treatments for kidneyrelated conditions.Overall, the granular κ-Car gels provide an optimal microenvironment for embedded bioprinting, facilitating the fabrication of tissues and organ analogs.This advancement significantly enhances the prospects of the fields of tissue engineering and regenerative medicine.

Conclusion
In summary, we have developed a novel uniform κ-Car sub-microgel bath based on a mechanical grinding strategy for embedded bioprinting of tissue and organ analogs.The κ-Car sub-microgels exhibited a uniform morphology with a small particle size of 642 nm and a jammed state as low as 0.3% (wt/vol).Due to the reversible linkages in jammed particulates, the κ-Car sub-microgel bath featured the flow behavior of Bingham fluid with shear-thinningand rapid self-recovery capabilities, allowing for embedded bioprinting with remarkable fidelity and high resolution as fine as 33 µm.Organs with their sophisticated architectures were successfully engineered within the new bath, including the heart, kidney, ear, nose and liver.Especially, cellularized kidneys with a biomimetic architecture were nicely grown, demonstrating the potential of the approach for tissue and organ construction.Therefore, we envision that this innovative submicrogel bath can be used for engineering tissues and organs, holding immense promise for tissue engineering and medical applications.

Scheme 1 .
Scheme 1. Schematic illustration of self-healing kappa-carrageenan (κ-Car) sub-microgel bath preparation for 3D embedded bioprinting.(a), (b) The network (a) and chemical structures (b) of cationic-crosslinked κ-Car hydrogel.(c), (d) The bulk κ-Car hydrogel is mechanically ground into sub-microgels (c) that assemble into granular gels through hydrogen bonding and cation interactions (d).(e) Scheme of utilization of jammed κ-Car sub-microgels as a supportive bath for achieving high-quality embedded bioprinting.(f), (g) Removing the sub-microgel bath via water washing (f) to release the printed 3D constructs (g).

Figure 1 .
Figure 1.Characterization of κ-Car sub-microgels.(a) The diameter distribution of κ-Car microparticles measured by a laser nanometer particle size analyzer.(b) Confocal laser scanning microscopy images and (c), (d) scanning electron microscopy images showing the homogenous morphology of κ-Car sub-microgels.(e) Absorption analysis highlighting the high optical transparency of the 0.3% (wt/vol) κ-Car sub-microgel medium.(f)-(i)The basic process of embedded 3D printing using high-optical κ-Car sub-microgel bath, including (f) real-time printing observation, (g) photo-crosslinking by using 405 nm irradiation, (h) release of the printed construct using a pipette, and (i) the obtained construct exhibiting structural integrity.

Figure 3 .
Figure 3. High-resolution embedded 3D bioprinting of rhodamine-labeled GelMA/SA using jammed κ-Car sub-microgel bath.(a) Cross-filaments printed by using GelMA/SA composite inks within support bath under moving speeds ranging from 20 to 60 mms −1 .(b) The average diameter of the printed cross-filament as a function of nozzle movement speed.(c) Confocal laser scanning microscopy images of a 3D printed high-resolution grid construct, including (d) the perspectives of projection and (e) 3D reconstruction.

Figure 4 .
Figure 4. Embedded 3D bioprinting of human heart.(a) A photograph of a heart model printed using the rhodamine-labeled GelMA/SA ink.(b), (c) Micro-computed tomography images showing the (b) 3D reconstruction and (c) inner structure of the heart model.(d) A cross section of the 3D printed heart.(e), (f) CLSM images of the cross section of printed heart, showing (e) the recreation of the internal structure and (f) a high printing resolution as fine as 45 µm.(g), (h) Photographs of 3D printed (g) vasculature and (h) tri-leaflet heart valve.(i), (j) Micro-computed tomography images showing the (i) 3D reconstruction and (j) inner structure of the tri-leaflet heart valve.(k) A printed heart with 0.5 mm thickness within the k-Car suspension bath.(l) An observation photograph of the heart model under 405 nm UV irradiation.(m) A photograph of the heart model after extraction.(n) The injection of a blood-mimicking dye solution into the right and left ventricles through aorta.

Figure 5 .
Figure 5. Embedded 3D bioprinting of complex biomimetic tissues and organs using jammed κ-Car sub-microgel bath.(a) Front and (b) rear-view photographs capturing a kidney model printed with the GelMA/SA ink.(c) CLSM images showing the internal 3D reconstruction of the kidney model.(d) A photograph of a 3D printed lung featuring trachea.(e)-(g) Detailed photographs showcasing 3D printed trachea within the κ-Car bath (e), highlighting precise wall structure (f) and hollow airway (g).(h)-(j) Printed spiral coil structure (h) and branched vasculature (i), with (j) an injection of a blood-mimicking dye solution showing the patency of the channels in the printed vasculature model.(k) Photographs of 3D printed auricle, nose, kidney and liver shapes achieved by utilizing SA ink within the jammed granular κ-Car gel bath.

Figure 6 .
Figure 6.Embedded cell-laden bioprinting using jammed κ-Car medium.(a) Fluorescent microscopy images of BMSCs within the printed GelMA/SA hydrogels on day 5 and day 7. From left to right: the nucleus, actin and the merge view.(b) LIVE/DEAD staining of BMSCs grown in large-scale kidney model on day 1 and fluorescent microscopy images on day 3. From left to right: a low-magnification view and a high-magnification view.