Ionically annealed zwitterionic microgels for bioprinting of cartilaginous constructs

Foreign body response (FBR) is a pervasive problem for biomaterials used in tissue engineering. Zwitterionic hydrogels have emerged as an effective solution to this problem, due to their ultra-low fouling properties, which enable them to effectively inhibit FBR in vivo. However, no versatile zwitterionic bioink that allows for high resolution extrusion bioprinting of tissue implants has thus far been reported. In this work, we introduce a simple, novel method for producing zwitterionic microgel bioink, using alginate methacrylate (AlgMA) as crosslinker and mechanical fragmentation as a microgel fabrication method. Photocrosslinked hydrogels made of zwitterionic carboxybetaine acrylamide (CBAA) and sulfobetaine methacrylate (SBMA) are mechanically fragmented through meshes with aperture diameters of 50 and 90 µm to produce microgel bioink. The bioinks made with both microgel sizes showed excellent rheological properties and were used for high-resolution printing of objects with overhanging features without requiring a support structure or support bath. The AlgMA crosslinker has a dual role, allowing for both primary photocrosslinking of the bulk hydrogel as well as secondary ionic crosslinking of produced microgels, to quickly stabilize the printed construct in a calcium bath and to produce a microporous scaffold. Scaffolds showed ∼20% porosity, and they supported viability and chondrogenesis of encapsulated human primary chondrocytes. Finally, a meniscus model was bioprinted, to demonstrate the bioink’s versatility at printing large, cell-laden constructs which are stable for further in vitro culture to promote cartilaginous tissue production. This easy and scalable strategy of producing zwitterionic microgel bioink for high resolution extrusion bioprinting allows for direct cell encapsulation in a microporous scaffold and has potential for in vivo biocompatibility due to the zwitterionic nature of the bioink.


Introduction
Progress in three-dimensional (3D) extrusion bioprinting in recent years, both technologically and in bioink development, has made it possible to bioprint complex and human-sized tissue and organ-like structures with high levels of complexity and cell organization [1].Different natural and synthetic polymers have been used to develop bioinks for high resolution bioprinting [2,3].However, besides the need to ensure proper in vitro functionality of the bioprinted tissue and optimal shape fidelity of the printed construct, the implant must survive the immediate and aggressive response from the innate and adaptive immune system upon implantation.An uncontrolled inflammatory response to implants results in fibrous encapsulation, which can lead to implant failure [4,5].Foreign body response (FBR) caused by implanted biomaterials seriously impedes the function of the implants and is a major obstacle to the development of implantable tissues and medical devices [6].
Adverse FBR response has been reported for many of the commonly used natural and synthetic biomaterials in tissue engineering.For example, polyethylene glycol (PEG) is one of the most used synthetic polymers in tissue engineering.However, adverse FBR has been reported in relation to PEG-based hydrogels upon in vivo implantation [7][8][9][10].Moreover, conjugation of macromolecules to PEG in order to increase in vivo stability and circulation time induced anti-PEG antibody formation, resulting in efficacy loss [11].Researchers have reported several strategies to regulate immune response toward implants and mitigate the FBR, such as adjusting the physical properties of the implanted biomaterials [12] or combining anti-inflammatory drugs or biological agents to suppress the inflammatory response [13].Yet these strategies are not versatile and applicable to all different biomaterials and implants, and they may impair biological function and induce unwanted side effects.Therefore, preventing FBR from the perspective of the biomaterial itself and fabricating implants based on materials with intrinsic anti-FBR properties is the safest and most reliable approach.
In this regard, zwitterionic hydrogels have attracted a lot of attention, as they are a unique class of materials showing exceptional anti-fouling properties that enable them to effectively inhibit FBR in vivo [14].Zwitterions have equal anionic and cationic groups and are therefore highly hydrophilic, antifouling, and biocompatible [15].In 2013, Zhang et al prepared zwitterionic hydrogels from poly(carboxybetaine methacrylate) and compared them to poly(2-hydroxyethyl methacrylate) (pHEMA) hydrogels implanted subcutaneously in mice for up to three months.They demonstrated that the fibrous capsule surrounding the zwitterionic hydrogels was much thinner and had lower collagen density compared to the pHEMA hydrogels [16].They recently reported a more mechanically robust zwitterionic hydrogel made of carboxybetaine acrylamide (CBAA) and sulfobetaine methacrylate (SBMA) monomers that effectively resists the fibrous capsule formation upon implantation in mice for up to 1 year [17].Liu et al extended this strategy to minimize the FBR to hydrogels containing transplanted islet cells for the treatment of type 1 diabetes.They showed less dense fibrotic layer and more vessels within the fibrotic layer and overall better performance in terms of diabetes correction for zwitterionic hydrogels compared to conventional alginate hydrogels [18].Also, zwitterionic polymers have been used to shield proteins from immunogenic responses in the bloodstream and have been shown to be more effective than the commonly used technology, 'PEGylation' , in terms of preserving protein bioactivity [19][20][21].To the best of our knowledge, there is only one recent study on extrusion bioprinting with zwitterionic hydrogels [22].This work exemplifies the challenges of stabilizing zwitterionic microgels due to their natural affinity for water.The study uses reactants such as SBMA, Nisopropylacrylamide (NIPAM), and PEG diacrylate to create annealed particulate matter, which is later treated with sugar crystals to introduce porosity for post-seeded cells.In contrast, our microgel system relies on a CBAA monomer as a primary component, known as one of the most efficient building block for non-immunogenic biomaterials.Notably, our method enables cell encapsulation through mixing and cell-compatible microgel annealing, a key distinction from the referenced study.
Despite these unique properties, the use of zwitterionic hydrogels has been limited to mostly photocrosslinked bulk hydrogels with limited versatility [21].In this project, our goal is to develop a bioink made of pure zwitterionic hydrogels that not only allows for extrusion printing with high shape fidelity but also is optimal for cell encapsulation.
Zwitterionic polymer solutions normally have very low viscosities, making them unsuitable for extrusion bioprinting.One solution to this is to add viscosity modifiers, such as nanofibers [23]; however, this may result in compromise of the antifouling properties and of the biocompatibility of the formulation.Our group has previously shown that mechanically fragmented microgels can be used as bioinks for extrusion bioprinting [24].Therefore, we chose to develop a zwitterionic microgel bioink, to have high viscosity formulation of pure zwitterionic hydrogels with inherent printability.The versatility of microgel bioinks permits diverse material compositions and printing designs [25][26][27][28][29][30][31][32].When microgels are packed closely together in a jammed state, they behave as a solid, but when external forces are applied, they display fluidic collective movement, i.e. shear-thinning behavior, allowing for material deposition and ensuring high viability of the encapsulated cells [33].Another advantage of microgel-based formulations is the creation of constructs with microscale porosity, which facilitates cell migration and remodeling as well as nutrition diffusion [34], thus addressing a key challenge in developing human-sized constructs [35].Moreover, microgels are modular and can be used to fabricate heterogeneous structures [36,37].Zwitterionic microgels have been described in literature and have been shown to have the shear thinning and shear recovery properties required for extrusion bioprinting [38].However, these microgels are not secondarily crosslinkable and lack long-term stability in aqueous environments.Therefore, a secondary crosslinking is required to anneal microgels and generate a microporous, stable scaffold that can be used for in vitro culture and in vivo implantation.
Overall, in this study, we have developed crosslinkable zwitterionic microgel bioink using alginate methacrylate (AlgMA) as a dual crosslinker, which allows for primary photocrosslinking of the bulk Figure 1.Illustration of zwitterionic microgel bioink preparation and bioprinting.(A) Zwitterionic bulk hydrogel is prepared by photocrosslinking of carboxybetaine acrylamide (CBAA) and sulfobetaine methacrylate (SBMA) monomer using alginate methacrylate as crosslinker.The bulk hydrogel is mechanically fragmented through µm-sized meshes to produce zwitterionic microgels.Microgels are mixed with cells and secondarily crosslinked using calcium ions resulting in a cell-laden microporous scaffold.(B) A cell-laden meniscus is bioprinted using zwitterionic microgel bioink.The bioprinted construct is quickly stabilized in calcium bath.The construct is further cultured in vitro in chondrogenic cell culture media.Illustration was created with BioRender.com.hydrogel as well as secondary ionic crosslinking of produced microgels.Zwitterionic bulk hydrogels are made by photocrosslinking zwitterionic CBAA and SBMA monomers; they are then mechanically fragmented using µm-sized meshes, to produce zwitterionic microgels.The microgels are next mixed with cells and secondarily crosslinked using calcium chloride (CaCl 2 ), resulting in cell-laden microporous scaffolds (figure 1(A)).The bioink composed of zwitterionic microgels and primary human chondrocytes is used for bioprinting of cm-scale meniscus structure.The printed construct is then submerged in a CaCl 2 bath to be stabilized by ionic crosslinking and then cultured in chondrogenic media to promote extracellular matrix expression (ECM) production and tissue-like development (figure 1(B)).The microporous zwitterionic scaffold showed high cell viability and chondrogenic potential for encapsulated human primary chondrocytes.Altogether, the novel and versatile strategy introduced in this study produces ionically annealable zwitterionic microgel bioink for high-resolution bioprinting, which allows for biocompatible cell encapsulation in a microporous scaffold.Gentamycin, Dulbecco's modified eagle medium (DMEM 31966) and fetal bovine serum (FBS) were obtained from Gibco.ITS + Premix Universal Culture Supplement was bought from Corning and fibroblast growth factor-2 (FGF-2) and transforming growth factor-β3 (TGF-β3) from PreproTech.

CBAA synthesis
The monomer was synthesized according to a literature procedure [39], with minor modifications.In a 100 ml one-neck round-bottom flask equipped with a magnetic stirring bar, DMAPA (8.9 g, 57.28 mmol, 1 eq) was dissolved in 60 ml anhydrous THF and the flask was sealed with dropping funnel and placed to −10 • C ethanol bath.Beta-Propiolactone (5 ml, 79.86 mmol, 1.4 eq) dissolved in 15 ml anhydrous THF was added to the dropping funnel and slowly dropped into the above solution under stirring for about 2 h.Afterwards, the reaction mixture was allowed to warm to room temperature and stirred overnight.Then the resulting white suspension was placed to the freezer for another 24 h at −20 • C to precipitate the product.Then the mixture was filtered through sintered glass funnel (S4 porosity) washed with dry diethyl ether and the product was obtained with vacuum filter, washed several times with cold ether, and dried overnight under high vacuum.

AlgMA synthesis
AlgMA was synthesized according to the modified procedure [40].In round bottom flask equipped with large magnetic stirrer, sodium alginate (PRONOVA UP MVG, NovaMatrix ® , MW > 200 kDa, 250 mg, 1.3 mmol repeating units, 2.6 mmol of hydroxyl groups) was dissolved in 25 ml of Milli-Q water to produce 1% (w v −1 ) solution.Methacrylic anhydride (Merck, distilled prior to use, 20 eq. to hydroxyl groups, 52 mmol, 7.7 ml) was added and formed emulsion was vigorously stirred for 24 h at room temperature.pH of the mixture was periodically measured and adjusted to the value 7-8 (measured with pH paper) using 5 M NaOH.After 24 h, the reaction mixture was transferred to Falcon tubes and centrifuged at 4200 RPM for 15 min to remove excess of methacrylic anhydride.Aqueous phase was precipitated into ethanol and the precipitate was filtered on fritted glass funnel (S4 porosity) and dried overnight at high vacuum.Dry AlgMA was dissolved in Milli-Q water (1% w v −1 ) purified by dialysis against Milli-Q water using Spectra/Por ® RC membrane MWCO 12 kDa for four days.Then, solution was filtered through 0.22 mm syringe filter, and freeze-dried to obtain 230 mg (92%) of white solid. 1 H-NMR spectra to determine degree of methacrylation degree of substitution (DS) was acquired using Bruker Avance III spectrometer operating at a 400 MHz proton frequency and can be found in figure S1.AlgMA was dissolved at concentration 1% (w v −1 ) in D 2 O. DS was calculated according to a previously reported formula [41] and was found to be 6.6%.

Zwitterionic bulk hydrogel preparation
Bulk zwitterionic hydrogels were produced by photopolymerization of zwitterionic monomers CBAA and SBMA using AlgMA as the crosslinker and LAP as photoinitiator.In a glass vial with stirring bar, crosslinker AlgMA (DS 6.6%, 123 mg, 0.25 mol% methacryloyl groups, 1.8 wt% of alginate), monomers CBAA (3 g, 1.875 M, 75 mol%) SBMA (1.2 g, 0.625 M, 25 mol%) were weighted and Milli-Q Water (6.65 ml) was added, and solution was let to dissolve overnight while stirring in the dark.Next, solution of the photoinitiator LAP (0.1% w v −1 .0.35 ml, 14 mg in 0.7 ml water) was added and mixed into the solution.Hydrogels were made in a simple glass apparatus (figure S2) composed of two microscope glass slides separated with folded Parafilm M spacer (1 mm).The thickness of the paraffin film is 130 µm, therefore placing 8 layers on each other, desired thickness can be obtained, and appropriate shape cut using cutter knife.Parafilm M is convenient because it seals and sticks well to the glass.This step is not critical, the spacer can be produced from Poly(dimethylsiloxane) (PDMS) or Teflon either.The solution was then injected between two glass slides and Photopolymerization was initiated by UV-VIS lamp for 40 min (405 nm, 8 mW cm −2 ).The resulting hydrogels were removed and washed in deionized (DI) water for at least 5 d to remove unreacted monomers and to allow the gels reach equilibrium.At the end of dialysis, 4 mm hydrogel discs were punched, weighted, and then freeze dried to measure equilibrium water content (EWC) of the hydrogels.EWC was measured as the ratio of water mass (swollen hydrogel weight minus dried hydrogel weight) to the swollen hydrogel mass.To prepare fluorescently labeled hydrogels, fluorescein o-acrylate comonomer was added to the starting monomer solution at a final concentration of 0.018 wt%.

Zwitterionic microgel preparation
Zwitterionic microgels were made by mechanical fragmentation.Equilibrated bulk zwitterionic hydrogels were cut into small pieces and transferred into a 10 ml custom-made extruder connected to a metal sieve with mesh width of 90 or 50 µm.The bulk gels were manually sieved five consecutive times with mesh resulting in microgel formulation.Microgels were then sterilized by precipitation in ethanol, aspirated excess of ethanol, dried overnight in vacuum oven, re-suspended in sterile water, and lyophilized.Fore secondary crosslinking, the lyophilized microgels were first resuspended as 6 wt% in sterile saline solution (0.9% NaCl) and CaCl 2 (100 mM) was added to crosslink microgels.

Microgel size distribution
Bulk zwitterionic hydrogels were made as previously described with 0.018% fluorescein o-acrylate in the hydrogel precursor solution.The fluorescently labelled zwitterionic microgels were then prepared using the mechanical fragmentation method described before.Lyophilized microgels were then resuspended in Milli-Q water at a low concentration of ∼0.5%, to permit microgel separation under the microscope.Microgel size was determined by dispersing the microgels into glass slides and imaging with a fluorescent microscope (SP8, Leica).Microgel diameter was evaluated using ImageJ software with the particle analysis tool.

Porosity measurement
The fluorescently labelled zwitterionic microgels were prepared and crosslinked as previously described to obtain fluorescent hydrogels.Samples were imaged by confocal microscopy (SP8, Leica).Porosity and pore size were determined by converting the stacks into single images and using a threshold to select the void spaces.Cross-sectional areas occupied by void spaces were determined for each image and averaged for the whole stack.

Rheological characterization
Rheological analysis was carried out on an Anton Paar MCR 301 rheometer equipped with 10 mm parallel plate geometry (PP10, Anton Paar) and a peltier element with thermal hood (H PTD 200, Anton Paar).All tests were performed at 25 • C. Humidity in the thermal hood was controlled by placing a wet tissue inside the chamber to prevent drying of the sample.Ramped shear rate (0.01-300 s −1 ) was performed to evaluate the shear-thinning behavior.To evaluate the shear-recovery properties, microgels were repeatedly exposed to cycles of alternating low (1 Hz, 1% strain) and high strain (1 Hz, 500% strain).All rheological characterizations were performed on 6 wt% microgel.Each test was repeated three times with a new sample.

Swelling measurement
Hydrogel discs were made by casting microgels into cylindrical PDMS molds, (6 mm in diameter and 2 mm in height) and were then ionically crosslinked using CaCl 2 (100 mM).Discs were then transferred to saline solution (0.9% NaCl) and incubated at 37 • C for 21 d.At regular intervals, samples were weighed, and the swelling ratio was determined as the ratio of hydrogel mass at a given time point divided by its initial mass and reported as a percentage.

Compression testing
Unconfined compression experiments were performed on a TA.XTplus texture Analyzer (Anton Paar) equipped with a 500 g load cell.For each sample (6 mm in diameter and 2 mm in height), a pre-load was applied to the sample until it reached full contact with the plate and was then allowed to relax completely.Samples were compressed at a rate of 0.01 mm s -1 until they reached 15% strain.The compressive modulus was extracted from the slope of the first linear part of the stress vs strain curve.

Primary human articular chondrocyte (hACh) isolation and culture
Primary hACh were collected from corrective surgeries of polydactyly patients.The research was conducted in accordance with the principles embodied in the Declaration of Helsinki and in accordance with local statutory requirements.Collection of human tissue was approved by the Kantonale Ethikkommission Zurich, license number PB_2017-00510.All participants (or their parent or legal guardian in the case of children under 16) gave written informed consent to participate in the study [42].Cartilage pieces were finely sliced (∼0.5 mm thickness), washed extensively in PBS with 50 µg ml -1 gentamicin (Gibco) and digested in collagenase solution (DMEM (Gibco), 1000 CDU ml -1 collagenase from Clostridium histolyticum, 2 v% FBS, Gibco, 1x Antibiotic-Antimycotic (Anti-Anti, Gibco)) overnight with gentle shaking at 37 • C. The resulting cell suspension was passed through a 40 µm cell strainer before collecting the cell pellet by centrifugation (500 rcf, 10 min).The cells were plated at ∼10 000 cells cm -2 and expanded in DMEM, 10 V% FBS, 1x Anti-Anti and 10 ng ml -1 FGF-2 at 37 • C, 5% CO2 and 95% humidity.After the first passage, the seeding density was reduced to ∼3000 cells cm -2 and Anti-Anti was exchanged for 10 µg ml -1 gentamicin.

Cell encapsulation and in vitro chondrogenesis
Cells at passage 2 were trypsinized and mixed with zwitterionic microgels (6 wt%) at a final density of 15 million cells ml -1 .The microgel-cell suspension was then gently mixed casted in cylindrical PDMS molds with 6 mm diameter and 2 mm height and crosslinked with sterile 100 mM CaCl 2 water solution.Scaffolds were cultured in chondrogenic medium containing DMEM, 10 µg ml -1 gentamycin, 1% ITS+, 50 µg ml -1 L-ascorbate-2-phosphate, 40 µg ml -1 L-proline, and 10 ng ml -1 TGF-β3.Medium was changed every second day.In vitro experiments were stopped after 21 d for viability, mechanical and histological analysis.

Filament dimeter, printability and ink spreading assessment
The filament diameter as well as microgel ink printability and spreading was assessed using quantitative analysis of printed grid structures.Grids were printed and imaged right after printing as well as after crosslinking and submerging in aqueous solution and the images were analyzed using ImageJ.The printability (Pr) and ink spreading (Sp) were determined using Equations ( 1) and ( 2), respectively: (1) A t and A p are the theoretical and printed areas of the squares in the grid structure, respectively and L is the perimeter of the printed square.For an ink with ideal shape fidelity, which forms perfectly square windows, Sp = 0 (i.e.A t = A p ) and Pr = 1.Values plotted represent measurements from the 36 windows of a grid for each microgel ink.

Bioprinting
Bioprinting was performed on a Biofactory bioprinter (regenHU, Switzerland) enclosed in a laminar flow hood.Gcode files were prepared in Slic3r (https:// slic3r.org)and post-processed in a custom written Matlab postprocessor (MATLAB 2019a, MathWorks).Bioinks were transferred into printing cartridges and briefly centrifuged at 250 rcf for 5 s before mounting them into the bioprinter.Bioprinting was performed with a tapered 410 µm nozzle at a speed of 10 mm s -1 and a pressure of ≈25-30 kPa.Postbioprinting, samples were crosslinked in 100 mM CaCl 2 for 1 h and afterward cultured in chondrogenic media.

Cell viability
To assess viability, samples were stained with a medium supplemented with 1 µm CalceinAM, 1 µm propidium iodide (PI) and 0.3 µma for 1 h.Imaging was performed on a Leica SP8 microscope equipped with a 10x and 20x objective.100 µm Z-stacks were acquired with 5 µm steps.Viability was assessed by counting viable (CalceinAM+) and dead (PI+) cells throughout the entire range and dividing the number of viable cells by the total number of viable cells plus dead cells.Quantification was performed using ImageJ.

Histology and Immunohistochemistry
Samples were fixed in 4% paraformaldehyde for 4 h, dehydrated in an ethanol sequence, embedded in paraffin wax (Milestone LogosJ) and cut into 5 µm sections on a microtome.Samples were progressively deparaffinized and rehydrated before staining.Brightfield images of stained sections were recorded on a 3DHistech Pannoramic 250-slide scanner and visualized with the case viewer 2.4 software.For Safranin O staining sections were first stained in Weigert's iron hematoxylin solution for 5 min, washed in DI water and differentiated in 1% acidalcohol for 2 s, washed again and stained in 0.02% Fast Green solution for 1 min and rinsed with 1% acetic acid for 30 s. Finally, sections were stained in 1% Safranin O for 30 min, dehydrated to xylene and mounted.For Collagen I and II immunohistochemistry, first antigen retrieval was first performed in hyaluronidase (1200 U ml -1 ) at 37 • C for 30 min.Sections were blocked with 5% BSA in PBS for 1 h.Primary antibody, rabbit anti-collagen I (1:1500, ab138492, Abcam) and mouse anti-collagen II (1:20, II-II6B3-s, DSHB Hybridoma) were dissolved in 1% BSA in PBS, and sections were incubated overnight at 4 • C. Sections were then incubated with the secondary antibody, goat anti-rabbit IgG-HRP for rabbit anti-collagen I (1:1000, ab6721, Abcam) or goat anti-mouse IgG-HRP for collagen II (1:1000, ab6789, Abcam) in 1% BSA in PBS for 1 h and developed with the DAB substrate kit (ab64238, Abcam) for 5 min.Sections were stained with Weigert's iron hematoxylin (Thermo Fisher Scientific) for 3 min, destained in 1% acid-alcohol, blued in 0.1% Na 2 CO 3 , dehydrated to xylene, and mounted.For alizarin red staining rehydrated sections were stained in 2% w v −1 Alizarin Red solution (2 mins), washed in 95% ethanol, dehydrated in 100% ethanol, cleared in xylene and airdried.Quantification of Safranin O, collagen type I and type II staining intensity were performed by ImageJ using a cartilage sample as a reference.Semiquantitative analyses of glycosaminoglycans (GAG), collagen I, and collagen II were performed following a published protocol [43].Color deconvolution was performed to isolate the hematoxylin and DAB stain (collagen I and II) or hematoxylin and Safranin O GAG respectively.To quantify the staining intensities, the mean gray value was taken for the deconvoluted collagen and GAG channels respectively.

THP-1 culture
THP-1 dualTM cell line (Cat.thpd.nfis) was purchased from InvivoGen, USA.Cells were cultured in suspension in T-150 cm2 cell culture flask at a subculture density of 1-5 × 10 5 cells ml -1 .Cells were cultured in RPMI-1640 medium supplemented with 10% FBS, 1% penicillin-streptomycin, 1% GlutaMAX 100-X (Gibco), 1% MEM Non-Essential Amino Acids Solution 100-X and 1% sodium pyruvate at 37 • C, 5% CO 2 and 95% humidity.Suspended cells were split by centrifugation and the cell pellet was resuspended in fresh medium.Cells were cultured on hydrogels (6 mm in diameter and 2 mm in height) at a density of 2.5 × 10 5 cells ml -1 without or with 100 ng ml -1 lipopolysaccharide (LPS) for 24 h.Afterwards, the supernatant was analyzed for nuclear factor kappa B (NF-κB) pathway activation.THP-1 dual cells have a stable integrator-secreted alkaline phosphatase (SEAP) reporter gene for monitoring NF-κB which is readily measurable in the cell culture supernatant using QUANTI-Blue™, a SEAP detection reagent.

Statistical analyses
All statistical analyses were performed in OriginPro.A one-way ANOVA with Tukey's multi comparison test was used to analyze the data.Data are represented as mean ± standard deviation.Statistical significance was determined using a one-way or two-way ANOVA with a Tukey's multiple comparisons test ( * p < .05,* * p < .01,* * * p < .001).

Zwitterionic microgels form microporous scaffolds with tunable porosity
We used mechanical fragmentation of a zwitterionic bulk hydrogel using µm-sized grids to form zwitterionic microgels.It is an easy and scalable microgel fabrication method which does not require using toxic reagents, and the resulting hydrogel has tunable porosity and high mechanical stability [44].The bulk hydrogels were produced by free-radical photocrosslinking of aqueous solution of zwitterionic monomers, using AlgMA as crosslinker, LAP as photoinitiator, and irradiation by a 405 nm light source for 40 min.The zwitterionic monomers CBAA and SBMA at 75:25 molar ratio were used.It has been shown that hydrogels produced with CBAA monomers possess the highest degree of hydration and tighter interactions with water molecules compared to other zwitterionic hydrogels, making them the most anti-fouling [45].Furthermore, it has been shown that hydrogels made of CBAA and SBMA monomers have long-term resistance to fibrous capsule formation after implantation in immunocompetent mice [17].Moreover, in our previous study we showed that this combination is favorable for chondrocyte proliferation and chondrogenesis compared to pure CBAA microgels [46].Hydrogels were made in a simple glass apparatus and the formed hydrogel sheets were then extensively washed with Milli-Q water to remove unreacted monomer and photoinitiator and also to reach equilibrium swelling.The swollen hydrogels were then mechanically pressed through µm-sized metal grids using a custom-made extruder, resulting in a viscous microgel slurry.We used two different grids with 50 µm or 90 µm aperture diameter to produce two different sets of hydrogels with different microgel sizes and thus different porosities (figure S2).We set the number of repetitive fragmentation iterations 5, based on the initial experiment where we investigated the impact of the number of sizing procedures on microgel size distribution.Although we observe a decrease in mean diameter and distribution range during sizing repetitions 3-7, these differences are not statistically significant (figure S3).After mechanical fragmentation, microgels were immersed in ethanol for sterilization, dried, and then reswollen in aqueous buffer.Dried microgels can also be stored at this step for later use.
The microgels can be visualized and analyzed using a confocal microscope, with the incorporation of fluorescein-o-acrylate comonomer.As evident in figure 2(A), the microgels have irregular and polygonal surfaces, which are characteristic of microgels produced by mechanical fragmentation techniques.Using ImageJ, we analyzed confocal images and evaluated microgel size.The average diameter for microgels is controlled by the size of the grid used for their fabrication and is 98 ± 40 µm for the 50 µm grid and 140 ± 82 µm for the 90 µm grid (figure 2(B)).Both microgel groups have wide size distributions, with polydispersity being slightly smaller for microgels made with the smaller grid (40% vs 59%).A broad microgel size distribution can be attributed to the fact that during the process of mechanical fragmentation, the grid's aperture affects the microgel size only in two dimensions.As a result, the produced formulation contains microgels with an average diameter larger than the grid's aperture.It was shown, that by employing a singlestep mechanical fragmentation, it is possible to create long, entangled microstrands with a high aspect ratio [24].Consistent with this work, mechanical fragmentation has been observed to produce microgels with an average diameter exceeding the grid's aperture.In anotherstudy, the bulk hydrogel was sized three-times using a 150 µm grid aperture, resulting in a size distribution ranging from 50 to 400 µm [46].In yet another study, hydrogels were successively extruded through grids with decreasing pore sizes, starting from 500 µm down to 25 µm.Micrographs from this study clearly showed that the resulting formulation contained microgels with dimension around 100 µm [38].These findings suggest that repeated extrusion offers only limited control over size distribution.Greater control can be achieved through alternative microgel production techniques, such as microfluidics [47], or lithographic templating [36,48].
To produce a stable microporous granular hydrogel using microgels, a secondary crosslinking (annealing) mechanism is required, to hold the microgels together.Several crosslinking methods, including enzymatic [49], light mediated [50] and click reactions [51][52][53] have been reported for secondary crosslinking of microgels to result in stable microporous scaffolds.Zwitterionic materials are highly water soluble and hydrated; therefore, the annealing of zwitterionic microgels is particularly challenging.Zwitterionic microgels injected into the water rapidly disperse with or without agitation (occurs immediately).Therefore, the crosslinking chemistry has to be fast enough to stabilize the construct and prevent microgel suspension.Ionic crosslinking of alginate is one of the faster crosslinking methods.The kinetics of sol-gel transition between bivalent calcium cations and sodium alginate for formation of hydrogels occurs over the millisecond time scale [54][55][56].When alginate is in contact with calcium electrolyte, a primary membrane is rapidly formed around the construct, which prevents the alginate sol inside the construct from deteriorating into the solution and allows for further diffusion of calcium ions and crosslinking.
Upon mechanical fragmentation of zwitterionic bulk hydrogels crosslinked with AlgMA, we obtained microgels that contain alginate within the structure and, importantly, on the surface.We hypothesized that alginate that is exposed on the surface of the microgels can be employed to ionically crosslink microgels and to rapidly form a stable construct due to inter-particle interaction.This could serve as a simple strategy that does not require any additional modification on microgels.It has been previously shown that as methacrylation takes place at hydroxyl groups, it does not impair ionic crosslinks formed between adjacent carboxyl groups on neighboring alginate chains, thus allowing AlgMA to sustain dual-cross-linking [57].
As AlgMA is used as a minor component in our structure and it acts as a dual crosslinker, governing both bulk hydrogel properties as well as secondary crosslinking robustness, it must be carefully designed, meaning that the DS of methacryloyl groups had to be optimized.We found that optimal methacryloyl DS of alginate is between 2%-8% to obtain bulk hydrogel that allows for mechanical fragmentation and sufficient alginate for microgel annealing.When AlgMA with a higher DS was used, the bulk hydrogel was highly crosslinked and too tough for mechanical fragmentation.Decreasing the amount of such AlgMA naturally produces a softer hydrogel that can be sized; however, the content of alginate would not be sufficient for the secondary crosslinking of microgels and stabilizations of the construct.When optimally methacrylated alginate is used as a crosslinker for photopolymerization, the resulting hydrogel is soft, and upon equilibrium swelling has water content ∼95%.This water content allows for the preparation of formulation >5 w% of microgel.Bulk gels with EWC >97% were typically very soft upon swelling and did not contain enough alginate for the subsequent annealing.It is important to note that the composition of zwitterionic microgels was purposely adjusted to minimize the content of AlgMA.After the bulk hydrogel is swollen to equilibrium, subjected to mechanical fragmentation and the resulting microgels are dried and resuspended in a saline solution, the composition becomes a 6 wt% formulation, with 5.82 wt% for the monomer part and 0.18 wt% for AlgMA.Therefore, in this final formulation, AlgMA constitutes only 3% of the total solid content.
It is shown in figure 2(C) that when microgels are injected into the water, they completely dispersed within a second.On the contrary, when calcium 100 mM CaCl 2 bath is used, microgels quickly form a stable construct due to ionic crosslinking between alginate and calcium ions.This observation confirms our hypothesis that the minor amount of AlgMA used in the bulk hydrogel production is enough to trigger ionic crosslinking of microgels in CaCl 2 solution, resulting in a stable construct.It is also important to note that ionic crosslinking has several advantages over covalent crosslinking.It has been shown that ionic bonds have better stress-relaxation properties compared to covalent F et al bonds [58], and this results in enhanced cell spreading, proliferation, and metabolism [59].In this study, the zwitterionic microgels are designed to be nonbiodegradable on a macromolecular level.The hydrogel network is constructed using zwitterionic polymer chains with a stable hydrocarbon backbone crosslinked with AlgMA.Both the hydrocarbon polyacrylic backbone and the glycosidic linkages of alginate are known to be stable under physiological conditions [60,61].On the contrary, calcium alginate bonds are known to weaken over time in the presence of cells, leading to a reduction in hydrogel strength [62].To assess the stability and swelling of the crosslinked granular hydrogels, we conducted tests in a saline solution at 37 • C for 21 d, as depicted in figure S4.Our results showed that granular hydrogels of both sizes remained stable over the 21 d period, with no significant signs of swelling or degradation.
The microstructure of the resulting microporous scaffold after ionic crosslinking was also evaluated by confocal microscopy (figure 2(A)).The confocal images show microgel packing and the void space between microgels for both microgel sizes, which is required for cell spreading and proliferation.Using ImageJ, we evaluated average void fraction or porosity of the scaffolds.Diverse ranges of scaffold porosity ranging from 2%-20% have been reported, with mechanical fragmentation technique depending on the biomaterial type, microgel size and annealing strategy [28,37,44].Our zwitterionic microporous scaffolds had overall high porosities, with 15 ± 2% total void space for hydrogels composed of 50 µmsized microgels and 20 ± 1.4% for 90 µm sized microgels (figure 2(D)) This is in line with previous studies showing that decreasing particle size results in reduced porosity [53,63].In summary, we produced novel microporous zwitterionic scaffolds with different microgel sizes and porosities through mechanical fragmentation technique and ionic secondary crosslinking.

Zwitterionic microgels enable high-resolution extrusion printing
Next, we evaluated the ability of the zwitterionic microgels to be used as ink for high resolution extrusion printing.Rotational and oscillatory measurements of microgels were conducted using a rheometer to characterize shear thinning and shear recovery properties, as two important characteristics required for extrusion printing.Shear-thinning fluids show a decrease of viscosity with increasing shear rates.In extrusion printing, shear-thinning is required to enable easy extrusion, with a decrease of viscosity during the extrusion phase where the shear forces dramatically increase.Shape preservation of the bioink occurs after extrusion, with increasing viscosity where the shear rate drops [64].Viscosity measurements showed similar shear-thinning behavior of zwitterionic microgels for both sizes (figure 3(A)).
A linear relationship between viscosity and shear rate was observed, showing decreased viscosity with increasing shear, indicating viscous flow under shear and thus optimal injectability and printability.
Shear recovery measures the timing and extent of recovery after a reduction in viscosity during extrusion.It is essential that the viscosity of inks quickly increases after exiting the nozzle to ensure high resolution printing and shape fidelity [65].Oscillatory strain sweep tests were performed by repeated cycles of low (1%) and high (500%) strain for shear recovery characterization.Under high shear stress, both microgel sizes changed from a solid (G ′ > G ′′ ) to a liquidlike (G ′′ > G ′ ) state.When returned to a low shear mode, they showed excellent and fast shear-recovery and self-healing properties, regardless of microgel size (figure 3(A)).
We first printed a one-layer grid structure to check printing resolution and continuity of the print.As shown in figure 3(B), and supplementary video 1, continuous lines were printed with no major ink flow and spreading with both microgel sizes.The grid structures were then crosslinked with immersion in CaCl 2 .The crosslinked structures did not show any macroscopically significant swelling and impaired resolution compared to uncrosslinked structures.Moreover, the crosslinked monolayer grids were stable for handling with tweezers in the CaCl 2 bath, showing robustness of the secondary crosslinking (supplementary video 2).
We also performed quantitative analysis on the printed grid structures.We first measured printed filament diameter for both inks.As shown in figure S4, the filament diameter for the ink with smaller microgels is slightly lower than that of the ink with larger microgels at 672 ± 78 and 772 ± 80 µm for 50 and 90 µm microgels respectively.Also, the diameter slightly increases after crosslinking and submerging in aqueous solution to 792 ± 61 and 1025 ± 62 µm respectively.This increase could also partially be due to some drying of the grids during the printing process and their recovery in the aqueous media.
Moreover, we performed quantitative analysis of ink printability and spreading using previously reported methods [66].Two metrics of shape fidelity were quantified from printed grid structures which is a common print geometry used to test shape fidelity as the square structure formed within the grid can be readily measured (figure 3(C)).We first explored shape fidelity by quantifying printability (Pr), which describes the similarity of the printed squares to a perfect square using the perimeter (L) and area (A p ) of the printed square in equation (1).Ideal squares have Pr = 1 which indicates well-defined filaments and ideal printability.Pr < 1 indicates a low viscosity ink that forms squares that are rounded or circular and Pr > 1 indicates a high viscosity ink that might form jagged or uneven filaments.Both microgel inks in this study had Pr value of 1.0-1.25 with no statistical difference for different microgel sizes, which indicates high shape fidelity for both of the inks (figure 3(C)).Also, this parameter was calculated for crosslinked structures which was decreased for both inks compared to non-crosslinked structures from 1.0-1.25-1.0-1.1.This is probably due to the fact that after crosslinking the rough edges are smoother resulting in smaller perimeter and thus smaller Pr values.
We also measured ink spreading (Sp) as the relative difference between printed and theoretical square area (equation ( 2)) which is an indicator of shape retention.A perfect print will have Sp = 0, however, in practice, the printed area of the square (A p ) differs from the theoretical area (A t ), as there is spreading due to gravity at the intersection of two printed filaments, making A p smaller than A t .In this study, our microgel inks had low Sp values of 22 ± 9 and 30 ± 7% for 50 and 90 µm microgel inks respectively which is low compared to previously reported microgel inks [63].These results show good shape fidelity for both inks and slightly better shape fidelity and resolution for the ink made up of the smaller 50 µm microgels.Also, the Sp values for grids after crosslinking were calculated and the results showed a slight increase for both inks to 30 ± 6 and 38 ± 5% for 50 and 90 µm microgels respectively.This observation is in line with the results observed for increased filament diameter after crosslinking (figure S4).It should be noted that the Sp values even after crosslinking are still relatively low and that this minor difference also shows that shape fidelity and resolution is not impaired by crosslinking.
Next, larger, more complex objects with hollow and overhang features were printed to evaluate and compare the printability and resolution of the zwitterionic microgel ink (figure 4(A)).Centimeter-size meniscus, nose, and ear shapes were printed with both microgel sizes and then secondarily crosslinked by submerging the structure in CaCl 2 bath.All structures were printed with good shape fidelity and printing precision.No anomalies were detected, especially on the sharp edges.Overhang structures were completely intact and stable after printing (figure 4(B)).As shown in figure 4(C) the printed and crosslinked objects are completely stable in aqueous media, with no defect in resolution and shape fidelity, making them suitable for long term culture.Overall, zwitterionic microgel bioink showed excellent printability and the secondary ionic crosslinking provided high stability of the printed objects.

Zwitterionic microgels allow for biocompatible cell encapsulation
Next, we encapsulated human primary chondrocytes to investigate the potential of zwitterionic microgels for cell encapsulation and cartilage tissue engineering.Cells are collected from corrective surgeries of polydactyly patients using a previously reported protocol [42].As these chondrocytes are obtained from young patients, they have higher proliferative and secretory properties compared to adult cells.Also, they have been reported to be non-immunogenic and immunosuppressive, making them a promising cell source for allogeneic treatments [67][68][69].
We investigated chondrocyte encapsulation in our zwitterionic microgel scaffold, made with two different grid sizes (50 and 90 µm) to study the effect of microgel size and porosity on cell response.Human chondrocytes were mixed with microgels and cast in cylindrical PDMS molds, 6 mm in diameter and 2 mm in height, and were then ionically crosslinked for 1 h at 37 • C by addition of CaCl 2 .This resulted in a microporous hydrogel with chondrocytes encapsulated in the micrometer-sized pores between microgels.Chondrocytes were encapsulated at 15 million cells ml -1 and cultured for up to 21 d.We used rhodamine-labeled microgels to track cell distribution in samples.
As shown in figure 5(A), there is a homogenous distribution of cells in the pores between microgels right after encapsulation.Representative livedead images of cells encapsulated in hydrogels with different microgel sizes and at different timepoints are shown in figure 5(B).Cells are homogenously distributed and spreading in the porous structure at all timepoints.It can also be appreciated that cell area is increasing over time, especially in the 90 µm sized microgels, showing cell proliferation.
Cells are highly viable at all timepoints and in both scaffolds as shown in figure 5(C).Cells showed high viability at day 1 in both scaffolds as 87 ± 10% for 50 µm and 77 ± 11% for 90 µm sized microgels, showing no cytotoxicity of the material.The viabilities were also not changed at day 7 as 88 ± 9% and 76 ± 7% for 50 and 90 µm sized microgels, respectively.Moreover, the cell viability at day 21 was slightly increased for 90 µm sized microgels, probably due to the higher porosity of the samples and better nutrition diffusion, with viabilities of 77 ± 11% and 82 ± 9% for 50 and 90 µm sized microgels, respectively.
Microgels offer greater support for cell proliferation and migration compared to bulk hydrogels [49].To investigate cell proliferation, we counted the number of cell nuclei per area stained with Hoechst at various time points.As expected, the calculated number increased for each hydrogel type compared to day 1, indicating continuous proliferation over the 21 day culture period.For the first 7 d of in vitro culture, cells proliferated with minimal differences between the conditions.Proliferation rate at day 21 was higher for hydrogels made from 90 µm microgels in comparison to those made from 50 µm microgels figure 5(D).This observation suggests that the increased porosity and larger pore size in these hydrogels contribute to an elevated proliferation rate.Overall, we showed that encapsulated primary human chondrocytes are highly viable and spreading in both zwitterionic microgel scaffolds over 21 d of in vitro culture.

Zwitterionic microgel scaffold supports chondrogenesis of encapsulated chondrocytes
After the 21 d of culturing samples in chondrogenic media, increased compressive modulus of the samples (figure 6(A)) and a progressive opacification (figure 6(B)) was observed, indicating matrix secretion by cells.An increase in compressive modulus was observed for both samples from ∼8-20 kPa at day 0 to ∼50-160 kPa at day 21, due to ECM secretion (compressive modulus at day 21: 119 ± 28 and 86 ± 31 kPa for 50 and 90 µm microgels respectively).The average compressive modulus of the cultured hydrogel samples from 50 µm microgels is indeed higher than that of 90 µm microgels, although the difference is not statistically significant.This result is consistent with another study where chondrocytes were cultured in hyaluronan microgels [28].In that study, at day 21, there was a decreasing trend in the elastic modulus of cultured crosslinked granular hydrogels as the grid aperture for fragmentation was increased from 40 µm to 500 µm.According to the literature, mechanical properties of crosslinked granular hydrogels are influenced by the size, shape, and packing density relative to the surrounding interstitial phase [44,70,71].Smaller microgels have a greater surface area and packing density, leading to more interlinking points, which enhances the scaffold's mechanical strength.Furthermore, the ECM formed between the microgels creates a denser secondary network, reinforcing the remaining interconnected structure, as some crosslinks might be disrupted by produced ECM.Also, the secreted ECM results in increased opacity of transparent samples at day 0 to a white cartilage-like disc at day 21 (figure 6(B)).It is also important to note that the cast discs showed no swelling and shrinkage over 21 d of culture and were fully stable, only turning opaque and stiff.
At the end of the 21 d, samples were fixed and prepared for histological analysis and stained for GAGs and collagen type II as major components of articular cartilage, as well as for collagen type I. Representative images of stained samples are shown in figures 6(C) and (D) where GAGs are stained red in safranin O staining, and collagens are stained brown in immunohistochemistry.The majority of the constructs still consist of zwitterionic microgels, but the constructs also showed deposition of ECM expression.Semiquantitative analyses of histological images were conducted to compare GAG, collagen I, and collagen II expression.The expression of both collagen I and II proteins exhibited statistically insignificant differences between the conditions.The Col I/Col II ratio was approximately 1, indicating that the collagen types were equally present.This suggests that encapsulated chondrocytes in zwitterionic microgels produce fibrocartilage tissue.As illustrated in figure 6(C), the intensity of GAGs was nearly equal for both types of microgels.
As seen in figure 6(D), deposition of GAGs, collagen type II, and collagen type I were observed in the void spaces between the microgels for both microgel sizes.A spatial distribution of ECM was observed, showing increased GAG and collagen type II content in the inner portion of the sample while regions of collagen type I were more confined to the periphery of the samples.In order to to examine calcification of tissue engineered construct, we conducted Alizarin Red staining to detect calcium deposits in the cultured hydrogel samples.However, we encountered a challenge with strong background staining in the zwitterionic microgels, making it difficult to identify calcium deposits definitively.Based on the staining results, we could not confirm calcification throughout the sample.Interestingly, we observed higher intensity of Alizarin Red staining at the periphery of the samples (figure S5).This may be attributed to calcium diffusion from the culture media and its binding to the zwitterionic hydrogel.An excessive concentration of calcium in the periphery could potentially lead to an ECM imbalance, favoring collagen type 1 production prevalence [72,73].Overall, we showed that zwitterionic microgel scaffolds support chondrogenesis of encapsulated chondrocytes, and secreted ECM by cells results in an ∼10-fold increase in compressive modulus of the samples.It is encouraging for future work in tissue engineering to observe cell encapsulation capability and chondrogenesis in zwitterionic materials, recognized as biologically inert.We believe that chondrogenesis in the presented microporous hydrogels is primarily driven by cell-cell contacts.The high porosity of microgel-based hydrogels and large pores allow cells to form clusters or migrate to micropores, fostering cell clustering similar to cell micropellet culture.High cell density in hydrogel micropores facilitates numerous cell-cell contacts, stabilizing the chondrocyte phenotype and promoting cartilage tissue production [74,75].After chondrocytes produce their initial ECM, negatively charged proteoglycans can electrostatically bind to carboxybetaine moieties.Additionally, sulfonic groups in sulfobetaine moieties have the capacity to electrostatically bind and sustain the release of growth factors [76,77].In our previous publication, we demonstrated that a mixed carboxybetaine and sulfobetaine composition in hydrogel enhances chondrogenesis compared to a purely carboxybetaine-based hydrogel [46].We hypothesize that the incorporation of sulfonic groups improves the retention of the chondrogenic growth factor TGF-β3, thereby enhancing chondrogenesis.Another contributing factor to chondrogenesis is the presence of alginate and the ionic crosslinking of the hydrogel [78,79].Soft hydrogels, allowing rapid stress relaxation and reversible ionic crosslinking, create a space that cells can remodel with their pericellular ECM [59,80].

Zwitterionic microgel scaffold does not induce an immune response in vitro
It has been shown in several studies that zwitterionic hydrogels have immunomodulatory effects on immune cells in vitro [18,[81][82][83][84].For example, macrophages seeded onto zwitterionic hydrogels showed inhibited inflammatory activation in response to LPS and Interferon gamma, compared to nonzwitterionic hydrogels [18,81].
To test immunogenicity of our scaffolds we used THP-1 monocytes seeded on our scaffolds with no further stimulation or in presence of 100 ng ml -1 LPS as a stimulatory factor.THP-1, an acute monocytic leukemia cell line is frequently used in immunomodulation studies [85], as they exhibit many similarities to primary monocytes with respect to cell morphology, expression of membrane receptors and antigens as well as secretory products [86].The THP-1 cells used in this study have an inducible cell-based reporter gene, which allows for the study of the NF-κB pathway, by monitoring the activity of SEAP in the media.As shown in figure 7(A), both microgel sizes had no stimulatory effect on cells compared to tissue culture plastic (TCP) when there was no stimulation.However, as expected, cells were highly stressed and the NF-κB pathway was activated in presence of LPS in the media.Yet, the presence of microgels had no further negative effect, and moreover there was a slight dampening of the activation compared to TCP especially for larger microgels.This preliminary result shows potential non-immunogenicity and an anti-inflammatory effect of our zwitterionic microgel inks, making them suitable for in vivo applications of biofabricated implants.

Zwitterionic microgel bioink enables bioprinting of large cellular constructs
We next performed bioprinting using bioink made of zwitterionic microgels mixed with human primary chondrocytes at 15 million cells ml -1 .All bioprinting was done using a 0.41 mm nozzle as in the printing of zwitterionic microgels without cells.We checked for the effect of the bioprinting process on cell viability to see whether the shear stress during the printing would cause any cell death.As shown in figures 7(B) and (C), there is no negative effect of bioprinting on cells and viability of the cells is >80% for both microgel sizes (89 ± 5% and 86 ± 4% for 50 and 90 µm sized microgels respectively).
Next, we evaluated bioprinting of a large, cellladen construct using the zwitterionic microgel bioink.A centimeter-scale meniscus was bioprinted with both microgel sizes and was cultured in chondrogenic media for 60 d.The macroscopic view of the meniscus after 60 d of culture is shown in figure 7(D), showing increased opacity and stability of the constructs.The histological sections of the constructs after in vitro culture are shown in figure 7(E), showing GAG and collagen secretion in the pores between microgels.There is less GAG deposition in the meniscus printed with 50 µm sized microgels compared to 90.Also, in both samples, there is much less collagen I expression than collagen II, similar to in the native tissue.
It is interesting to note that the ECM composition in the long-term meniscus culture is different from small discs, with less GAG and collagen I, probably due difference in culture time as well as difference in nutrition diffusion in large constructs.Also, it is worth mentioning that even though in culturing of small discs there was no significant difference in ECM deposition of samples made with different microgel sizes, it was not the case for large constructs.The large cell-laden meniscus printed with 90 µm sized microgels showed increased ECM production, as evident in histological stainings, and stability compared to the 50 µm sized microgels, probably due to the higher porosity of the scaffold.The maturation of large centimeter-scale cell-laden constructs into tissue grafts can be hindered by inadequate nutrient diffusion, resulting in inhomogeneous tissue development.To enhance biomaterials tissue-like constructs towards full maturation, bioreactors offer a promising solution by mimicking the tissue microenvironment.These bioreactors enable automatic exchange of culture media, promote flow-enhanced nutrient diffusion, and facilitate noninvasive monitoring of pH, glucose, and lactate concentration [87].

Conclusion
We have developed a novel strategy for producing zwitterionic microgels with excellent rheological properties for extrusion 3D bioprinting, as well as biocompatible crosslinking, showing strong potential for diverse tissue engineering applications.This study addressed challenges in the need for novel bioinks that do not induce an immune response upon implantation.the numerous natural and synthetic polymers available to make hydrogels, zwitterionic hydrogels have gained particular attention lately because of their unique antifouling properties in vivo upon implantation in animal models.Despite their remarkable non immunogenicity, no straightforward routes to a zwitterionic hydrogel bioink have been reported, perhaps because zwitterionic monomers and polymers solutions have very low viscosity and high swelling and are disadvantageous for precise additive manufacturing.
We have produced zwitterionic bulk hydrogels that are mechanically fragmented into microgels with tunable size and porosity.Using AlgMA as a dual crosslinker enables both primary photocrosslinking of the bulk hydrogel as well as biocompatible secondary crosslinking of microgels to rapidly stabilize the printed construct by ionic interactions.Our zwitterionic microgel bioink shows outstanding shear thinning and shear recovery properties and enables high resolution extrusion printing of complex objects with overhang features without the need for any support.The bioink also supports viability, spreading and ECM production of encapsulated human primary chondrocytes.Large cellular constructs can be printed with the bioink and can be cultured for long periods in vitro, resulting in cartilaginous biomaterial constructs.The importance of microgel size and scaffold porosity in long-term culture of large cell-laden constructs was demonstrated, with higher porosity resulting in better cartilage tissue maturation.Future work will be needed to enhance microgel degradation to further support biomaterial cellular constructs towards tissue maturation, which could result in a more homogenous tissue implant.Our strategy is highly versatile and can produce microgels and bioink formulations starting from any combination of acrylic monomers.Therefore, a variety of novel synthetic bioinks can be designed using the dual crosslinker AlgMA, which allows for mild and biocompatible annealing and stabilization of printed microporous structures in aqueous media.

Figure 3 .
Figure 3. Rheological and printability evaluation.(A) Shear-thinning behavior measured by rotational rheometry with ramped shear rate (0.01-300 s −1 ) and shear-recovery behavior measured by oscillatory rheometry with cycles of low (1%) and high (500%) strain for both microgel sizes.(B) Grid structures printed with zwitterionic ink for both microgel sizes, before and after crosslinking with CaCl2.(C) Quantitative analysis of inks printability (Pr) and ink spreading (Sp).

Figure 4 .
Figure 4. 3D printing of zwitterionic microgels.(A) 3D printed objects-human auricle, nose, and meniscus using zwitterionic microgel ink for both microgel sizes.Printing was performed with a tapered 410 µm nozzle at a speed of 10 mm s −1 and a pressure of ≈26 kPa.(B) Side view of the printed ear with 90 µm sized microgels showing the stability of the over-hanging feature.(C) Crosslinked printed ear with 90 µm sized microgels in CaCl2 bath showing stability in aqoeus media.

Figure 6 .
Figure 6.Mechanical properties and ECM quality evaluation.(A) Compressive elastic moduli calculated from the stress-strain curves obtained from in vitro samples after 0 and 21 d of culture.n = 5, one-way ANOVA with a Tukey's multiple comparisons test ( * p < 0.05).(B) Macroscopic view of the crosslinked hydrogels made of zwitterionic microgels and hACh cells casted as 6 mm discs at Day 0 and after 21 d of culture.(C) Semiquantitative evaluation of deposited GAG, collagen I and collagen II intensity (D) Representative histological and immunohistological staining for GAG, collagen I, collagen II after 21 d of in vitro culture in chondrogenic medium.

Figure 7 .
Figure 7.In vitro study with macrophages large construct bioprinting with zwitterionic microgel bioink.(A) SEAP reporter readout for NF-κB activation for THP-1 cells seeded on hydrogels after 24 h.(B), (C) Representative confocal images of viability assay of hACh mixed with zwitterionic microgels after bioprinting and calculated viability of cells, n = 3, one-way ANOVA with a Tukey's multiple comparisons test.(D) Macroscopic view of the bioprinted meniscus after 60 d of in vitro culture in chondrogenic medium.(E) Representative histological and immunohistological staining for GAGs, collagen I and collagen II in bioprinted meniscus structures after 60 d of in vitro culture.