Hyaluronic acid–alginate hydrazone crosslinked hydrogels support the generation and maturation of V2a interneurons Authors: Alycia N. Galindo, Madison E. McLaren, Alyssa K. Chi, Jenna D. Khachatourian, Jessica C. Butts, Marian H. Hettiaratchi Published: 2026 ABSTRACT Injury to the central nervous system (CNS) causes inflammation, cell death, and glial scar formation that inhibits tissue repair. Injectable hydrogels modified with extracellular matrix (ECM)-derived peptides can provide biochemical cues to promote neural tissue repair and serve as a vehicle to deliver therapeutics across the blood-spinal/blood-brain barrier in a minimally invasive manner. We developed an injectable hydrazone crosslinked hyaluronic acid–alginate (HA–Alg) hydrogel for neural tissue repair. We fabricated hydrogels with a range of polymer concentrations and evaluated their physicochemical properties to identify formulations that mimic the stiffness and viscoelastic properties of the CNS tissue environment. Hyaluronic acid was further modified with ECM-derived, cell-adhesive peptides (RGD and IKVAV) to enhance neuronal adhesion and viability. To evaluate the therapeutic potential of our hydrogel platform, we embedded mouse embryonic stem cell aggregates and differentiated them toward mature V2a interneurons. These interneurons are critical for relaying motor signals and represent a promising therapeutic cell population for treating spinal cord injuries. We demonstrated successful enrichment for V2a interneurons in HA–Alg hydrogels containing ECM-derived peptides, with a 2-fold increase in Chx10 expression compared to laminin controls. Interestingly, both our newly described HA–Alg and established crosslinked HA–HA hydrogels containing IKVAV peptides demonstrated significantly increased neurite length in interneuron-enriched cultures compared to hydrogels without peptides, averaging approximately 197.2 ± 53.5 µm for HA–IKVAV + HA and 212.4 ± 75.2 µm for HA–IKVAV + Alg. This study demonstrates that a biomimetic hydrogel platform containing ECM-derived peptides can support the neuronal adhesion and viability required for future applications in functional CNS tissue repair. Introduction The spinal cord transmits sensory information to the brain while relaying motor commands back to the peripheral nervous system.1 Spinal cord injury (SCI) is characterized by degeneration of neurons that can result in motor, sensory and autonomic dysfunction.2 Following injury, the rapid invasion of immune cells and activation of radial glia create an inflammatory environment that causes secondary tissue damage, a cavity of lost tissue, and the formation of a glial scar that physically and biochemically inhibits tissue regeneration.3 Various therapeutic approaches have been developed to attempt to restore neural tissue function after SCI. These approaches include targeting the neuroinflammatory environment, reconstituting damaged tissue by delivering cells that support neural tissue repair,4,5 removing biomolecules from the injury environment that inhibit tissue repair,6 and delivering neurotrophic factors to promote neuronal growth and remyelination.7,8 While some of these approaches have had success in promoting neural tissue regeneration in rodent models, these successes rarely translate into meaningful motor or sensory gains in human clinical trials. This translational failure likely stems from the different anatomical scales, the heterogeneity of clinical injuries,9 and the development of multi-centimeter cystic cavities that create growth barriers that are magnitudes more severe than those modeled in typical rodent contusions.9,10 Biomaterials such as hydrogels, nanoparticles, and aligned scaffolds have the potential to overcome these hurdles because they can be engineered to support local delivery of viable cells,11 sustained release of bioactive proteins,8,12 and incorporation of other biomolecules that modify the injury environment to support axonal regeneration.13,14 Biomaterials can be tailored for SCI treatment to be injected through the blood-spinal cord/blood-brain barrier, provide directional topographical cues that guide axons across the lesion, and act as reservoirs that protect therapeutic molecules from rapid degradation in vivo.15 Hydrogels that mimic mechanical properties of CNS extracellular matrix (ECM), including its stiffness that ranges from 100–1000 Pa,16–19 cell adhesion, and viscoelastic properties, can support cell proliferation and differentiation and guide axonal extension.20,21 Hydrogels made from modified ECM molecules, such as hyaluronic acid (HA), collagen, and gelatin, are particularly attractive biomaterials for treating SCI because they can also mimic the biochemical composition of neural tissue,22 providing neuroprotection and stabilizing neuronal networks.23 HA, which is the most abundant ECM molecule in the CNS, maintains neural tissue homeostasis by enhancing migration, proliferation, and differentiation of neural progenitor cells and oligodendrocyte progenitor cells to drive repair after injury.24,25 HA can be modified with functional groups such as aldehydes, maleimides, methacrylates, or hydrazides that allow it to be crosslinked into a hydrogel. For these reasons, HA hydrogels have been used extensively in preclinical models to deliver cells, proteins, and enzymes to the CNS and are a favored biomaterial for CNS tissue engineering applications.20,26,27 Although HA hydrogels have shown promising results in preclinical studies, further research is required to facilitate their translation into clinical practice. Alginate, a naturally occurring polysaccharide derived from brown seaweed, is emerging in CNS applications for its antioxidant, anti-inflammatory, and neuroprotective properties.28,29 Although not a native component of the brain ECM, alginate hydrogels can be fabricated to match CNS tissue stiffness and viscoelasticity.30 Alginate's high carboxylic acid content allows it to sequester divalent cations, potentially mitigating the oxidative stress triggered by the release of redox-active ions during hemorrhage.31 Alginate hydrogels physically crosslinked with calcium ions have facilitated neurite outgrowth and neural stem cell transplantation.32,33 However, calcium-crosslinked alginate hydrogels are limited in CNS applications because they can undergo calcium ion exchange, resulting in an inflammatory response,34,35 and tend to swell or stiffen in the highly ionic CNS environments.36 Alternatively, modifying alginate with functional groups, such as aldehydes, ester linkages, or amide bonds,37 can enable chemical crosslinking schemes that do not require the use of calcium ions. Because functional tissue restoration after SCI depends on the directed regeneration of specific neuronal subpopulations at their natural target regions, we sought to investigate the development of a hydrogel to support the growth and viability of a specific subpopulation of neurons. V2a interneurons are prime candidates for this application, as they are excitatory cells that span the hindbrain and spinal cord, serving as a critical bridge for relaying sensorimotor information between intersegmental regions and contributing to the coordination of sympathetic and parasympathetic functions.38–40 Within the spinal cord, these neurons coordinate excitatory inputs that are critical for autonomic and voluntary motor function.41 V2a interneurons also play a role in spinal respiratory networks and facilitate the plasticity of neuronal networks in respiratory muscles as observed in models of amyotrophic lateral sclerosis and cervical SCI.42,43 The ablation of V2a interneurons in mice disrupts breathing patterns, forelimb reaching tasks, and left-right hindlimb coordination.40 The Sakiyama-Elbert lab pioneered initial protocols for V2a interneuron differentiation from mouse embryonic stem cells (mESCs) in 2014.44,45 Since then, V2a interneurons have been successfully differentiated from human pluripotent stem cells,40,41 and their transplantation in SCI has been identified as a promising therapeutic approach.46 V2a interneurons transplanted into rat cervical SCI models functionally integrate with the host spinal cord, improving electrophysical signaling and restoring diaphragm activity.39,43,47,48 However, the survival and engraftment of transplanted neurons in vivo can be a challenge in larger injuries, as chronic inflammation, the growth-inhibitory environment of the glial scar, and the lack of cell-adhesive substrates can all contribute to cell death.49 While HA hydrogels containing astrocyte-derived ECM have been successfully used for the transplantation of V2a interneurons,48 neuronal processes in the lesion area were not significantly enhanced in this system compared to HA hydrogels alone, which may be due to the fact that HA inherently lacks neuronal adhesion cues necessary to support neuron attachment and proliferation.8 Thus, there is a need to develop hydrogels with cell-adhesive cues to support the viability and engraftment of transplanted neurons. To address these challenges, here we describe a dual polymer, hyaluronic acid–alginate (HA–Alg) hydrazone crosslinked hydrogel to support V2a interneuron survival, differentiation, and maturation for neural tissue repair. We hypothesized that a hydrogel that combines the properties of both HA and alginate through dynamic covalent crosslinking could more closely mimic the biomechanical properties of CNS tissue and eliminate the need for calcium-based crosslinkers. HA was functionalized with adipic acid dihydrazide for crosslinking with oxidized alginate to form reversible hydrazone bonds. This dynamic covalent crosslinking enables the fabrication of injectable hydrogels with shear-thinning and self-healing properties that are ideal for CNS tissue repair as they can be easily injected through the blood-spinal/blood-brain barrier in a minimally invasive manner to fill the cavity of lost tissue.50–52 However, despite successfully mimicking the physical properties of CNS tissue, current hydrogels often fail to provide the cell-adhesive cues essential for the long-term survival and maturation of transplanted cells. Thus, we also hypothesized that chemically conjugating ECM-derived peptides, such as fibronectin-derived (RGD) and laminin-derived peptides (IKVAV), to HA could provide an adhesive substrate with the appropriate biochemical cues to support the culture of V2a interneurons. We evaluated the physicochemical properties of hydrogels fabricated using a range of polymer concentrations to identify a hydrogel formulation that could support neuronal culture. We targeted a hydrogel with a gelation time between 1 and 5 minutes to facilitate cell dispersion throughout the hydrogel, stiffness of approximately 100 Pa to mimic CNS tissue, and minimal degradation or swelling (±25% mass change) over 28 days to support long-term cell culture. We then investigated whether HA–Alg hydrogels modified with ECM-derived peptides could support the differentiation of mESCs into viable, mature V2a interneurons. HA–Alg hydrogels were compared to similar, previously developed HA–HA hydrazone-crosslinked hydrogels to evaluate the use of alginate in a biomaterial for V2a interneuron transplantation. The differentiation and maturity of V2a interneurons were assessed via cell viability, neurite outgrowth, and gene expression. We observed Chx10 expression across hydrogel conditions, demonstrating enrichment of V2a interneurons. We also observed robust neurite extension in IKVAV-modified hydrogels, with neurite lengths averaging approximately 197.2 ± 53.5 µm for HA–IKVAV + HA and 212.4 ± 75.2 µm for HA–IKVAV + Alg. We concluded that alginate can be combined with HA to form stable hydrogels that meet our criteria for a cell transplantation vehicle and that HA–Alg hydrogels modified with IKVAV can support V2a interneuron viability and maturation. Overall, we developed an HA–Alg hydrogel platform that can be used for future applications in neural tissue repair. Materials and methods Synthesis of norbornene–hyaluronic acid (HA–Nor) HA–Nor was synthesized as previously described.53,54 Briefly, an initial intermediate product, tetrabutylammonium–hyaluronic acid (HA–TBA) was created through an ion exchange. HA–TBA was made by dissolving sodium hyaluronic acid (Lifecore Biomedical LLC, Chaska, MN) (1 g, 2.64 mmol) for 20 minutes in 20 mL of double distilled water (ddH2O) at 2% w/v. Dowex (Amberlight™ MB) (Sigma Aldrich, Saint Louis, MO) was weighted at a 3 : 1 mass ratio and added to a flask for ion exchange for 5 hours followed by vacuum filtration. The pH of the filtrate was then adjusted to pH 7 using tetrabutylammonium hydroxide 30-hydrate (Millipore Sigma, Milwaukee, WI). The product was then frozen at −80 °C prior to being lyophilized. HA–TBA (1 g, 1.61 mmol) was dissolved in dimethyl sulfoxide (DMSO) (Fisher Scientific, Waltham, Massachusetts) at 2% w/v for 20 minutes followed by a purge with N2 and the addition of 5-norbornene-2-carboxylic acid (Millipore Sigma) at 0.5 molar equivalents (0.11 g, 0.81 mmol), 4-(dimethylamino)pyridine (DMAP) (Millipore Sigma) at 0.25 molar equivalents (0.05 g, 0.40 mmol), and di-tert-butyl dicarbonate (Boc2O) (Millipore Sigma) at 0.07 molar equivalents (0.03 g, 0.11 mmol). The reaction was stirred at 45 °C for 20 hours. The solution was quenched with equal volume cold water and dialyzed for 3 days against ddH2O. Following 3 days, the product was precipitated out with the addition of NaCl (1 g per 100 mL of solution) and cold acetone (1 L per 100 mL solution). Precipitate was then redissolved in ddH2O, frozen, and lyophilized. Synthesis of adipic acid dihydrazide norbornene hyaluronic acid (HA–Nor–ADH) HA–Nor (0.13 g, 0.24 mmol) was then modified a second time with an adipic acid dihydrazide (ADH) group for crosslinking. HA–Nor was dissolved in ddH2O at 2% w/v followed by the addition of ADH (Spectrum Chemical Mfg. Corp., Gardena, CA) at 10 molar equivalents (0.418 g, 2.40 mmol) and hydroxybenzotriazole hydrate (Chem-Impex, Wood Dale, IL) at 1 molar equivalent (0.03 g, 0.24 mmol). The pH was adjusted to 4.75, then 1-1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride (EDC) (G-Biosciences, St. Louis, MO) EDC was added at 1 molar equivalent (0.05 g, 0.24 mmol). pH was monitored at 4.75 for 2 hours. The reaction stirred for 48 hours, then dialyzed against 0.1 M NaCl for 2 days followed by dialysis against ddH2O for an additional 2 days prior to being frozen and lyophilized. Synthesis of oxidized alginate (Alg–Ox) Sodium alginate (low viscosity <20 mPa s and molecular weight <75 kDa) purchased from NovaMatrix (Sandvika, Norway) (1 g, 2.51 mmol) was dissolved at 2% w/v in ddH2O. Sodium periodate (NaIO4) (Millipore Sigma) was dissolved in ddH2O at 3% w/v at 0.6 molar equivalents (0.32 g, 1.51 mmol) and then added to alginate in a round bottom flask covered with aluminum foil to protect from light. After letting the solution react for 4 hours, the reaction was quenched with 5 molar excess propylene glycol (Avantor, Center, PA) (0.96 g, 12.56 mmol) and rotated for 2 hours. The solution was then dialyzed against ddH2O for 2 days, frozen, then lyophilized.55 HA was oxidized similarly, as previously described.54 Peptide conjugation and quantification A PBS solution was made containing 1% w/v HA–Nor–ADH (10 mg, 0.01 mmol, 0.0016 mmol norbornene), 2 molar excess of RGD (CGRGDSG) peptide (GenScript, Piscataway, NJ) (2 mg, 0.0032 mmol) or IKVAV (CGKIKVAVG) peptide (GenScript) (2.8 mg, 0.0032 mmol) to norbornene, and 0.05% w/v Irgacure 2959 from Advanced Biomatrix (Carlsbad, California). A thiol-ene reaction between HA–Nor–ADH and the terminal cysteine on the peptides was performed by exposing the solution to 365 nm UV light at 5 mW cm−2 for 5 minutes. The solution was dialyzed against ddH2O for 1 day before being sterile-filtered, frozen at −80 °C, and lyophilized. The disappearance of thiols after the photoinitiated reaction was detected using Ellman's reagent as previously described.54 The Ellman's reagent (Fisher Scientific) consisted of 2 mM of 5,5-dithio-bis-(2-nitrobenzoic acid) (DTNB) and 50 mM sodium acetate in 1 M tris(hydroxymethyl)aminomethane (TRIS) buffer. Samples of the reaction before and after UV exposure were diluted to fall within the range of a standard curve of serially diluted peptides. 50 µL of each standard and sample were added to a 96-well plate and mixed with 250 µL of Ellman's reagent. The plate was developed at room temperature for 5 minutes prior to reading absorbance at 412 nm on a microplate reader (BioTek Synergy Neo2). Degree of modification (DOM) determination of HA–Nor and HA–Nor–ADH Proton nuclear magnetic resonance spectroscopy (1H NMR, 500 Hz, BrukerUSA) was used to quantify the degree of modification of both HA–Nor and HA–Nor–ADH as previously described.53,54 A representative spectrum of HA–Nor–ADH can be found in the SI (Fig. S1). A hydroxylamine hydrochloride titration was performed to quantify the degree of modification of Alg–Ox. The volume of NaOH used in the titration was used in eqn (1) to equate the degree of modification. Eqn (1): hydroxylamine hydrochloride titration equation for oxidation degree of modification.1where 198 g mol−1 is the monomeric molecular weight of alginate, 0.1 M is the molarity of NaOH used to adjust the solution pH, Vf is the final volume of NaOH in the burette after titration recorded in liters, V0 is the initial volume of NaOH in the burette before titration recorded in liters, and mAlg–Ox is the mass in g of Alg–Ox used in the titration. Preparation of hyaluronic acid alginate (HA–Alg) hydrogels To form HA–Alg hydrogels to evaluate the gelation time, compressive modulus, and mass change, 50 µL of HA–Nor–ADH and 50 µL Alg–Ox at varying concentrations (1–3% w/v) were mixed (Table 1). Hydrogels used for mass change and gelation time optimization experiments were prepared in 2 mL microcentrifuge tubes. Hydrogels for compression testing were prepared in cylindrical silicon molds with an 8 mm diameter and depth of 4 mm. Table 1Table of hydrogel formulations for physicochemical characterizationHA–Nor–ADH (% w/v)Alg–Ox (% w/v)Hydrogel polymer total (% w/v)1.01.01.01.02.01.51.03.02.02.01.01.52.02.02.02.03.02.53.01.02.03.02.02.53.03.03.0 Table of hydrogel formulations for physicochemical characterization Physiochemical characterization of hydrogels Gelation time Due to the rapid nature of hydrazone crosslinking, in which bonds form too quickly to measure the gelation point using rheology, the gelation time of the hydrogels was quantified using the inverted tube test.53,56–58 50 µL of HA–Nor–ADH and 50 µL of Alg–Ox at desired concentrations were added to a 2 mL tube, mixed, and inverted once every minute to assess viscosity until gelation. Determination of gelation time began at initial mixing of the two solutions, and the time at which there were no visual signs of fluid flow was recorded. Compressive modulus Hydrogel compression modulus was evaluated using a Discovery Hybrid Rheometer-2 (DHR-2; TA Instruments, New Castle, DE, USA). 100 µL HA–Alg hydrogels were created in 8 mm cylindrical molds and allowed to crosslink at 4 °C overnight. Hydrogels were then placed on a Peltier plate maintained at 37 °C and compressed using an 8 mm aluminum parallel plate at a constant linear rate of 10 µm s−1. The stress–strain curves were used to determine the compressive modulus of the hydrogels compressed to 15% of their original height. A representative stress-strain curve can be found in the SI (Fig. S2). Mass change We determined the mass change of the hydrogels over 28 days in PBS and artificial cerebrospinal fluid (aCSF) containing NaCl (149 mM), KCl (3 mM), MgCl2·6H2O (0.8 mM), CaCl2·2H2O (1.4 mM), Na2HPO4·7H2O (1.5 mM), and NaH2PO4 (0.2 mM). 100 µL hydrogels were formed in 2 mL tubes and allowed to crosslink for 24 hours. Initial crosslinked hydrogels were weighed before the addition of 500 µL of 0.05% sodium azide in PBS or 500 µL of aCSF in their respective tubes. At 0 hours, 24 hours, 7, 14, 21, and 28 days, the solution was carefully removed from the tubes, and the hydrogel was weighed. The mass change compared to the initial hydrogel mass was calculated at each time point as a percentage using eqn (2). Eqn (2). Percent mass change2 Injectability 250 µL of either 3% w/v HA–Nor–ADH or Alg–Ox was added to each syringe. A female-to-female syringe Luer Lock was then used to attach both syringes, and the solutions were mixed between syringes before being collected in a single syringe. The solution was allowed to crosslink overnight at 4 °C. The following day, images of hydrogels extruded through a 22-gauge needle were collected to visualize hydrogel injectability. Shear-thinning Hydrogel shear-thinning was evaluated using a Discovery Hybrid Rheometer-2. 100 µL 3% w/v HA–Alg or HA–HA hydrogels were formed in 8 mm cylindrical molds and allowed to crosslink at 4 °C overnight. Hydrogels were then placed on a Peltier plate maintained at 37 °C, and an increasing shear rate from 0 to 50 s−1 was applied using an 8 mm aluminum parallel plate. Porosity Scanning electron microscopy (SEM) images were collected as previously described using a ThermoFisher Apreo 2 SEM.53 100 µL 3% w/v HA–Alg or HA–HA hydrogels were formed in an 8 mm mold, frozen at −80 °C, and lyophilized. Hydrogels were then sputter-coated using a gold sputter coater (Au–Pd, Cressington, Watford, UK) and imaged at 120 times magnification at 0.40 nA and 5 kV. ImageJ was used to calculate 2D porosity. Briefly, images were converted to 8-bit images, and the scale bar was cropped. Image contrast was enhanced to 0.35% saturated pixels. An Otsu threshold was performed and adjusted from 0–80 to include pores. “Analyze particles,” including pores ranging from 20-infinity pixels with a 0–1 circularity, was used to determine the pore area. The percentage of porosity was then calculated using eqn (3).59 Eqn (3). Percent porosity3 Mouse embryonic stem cell culture AB2.2 mouse embryonic stem cells (mESCs), an established male mESC line, were a gift from the Baylor College of Medicine Mouse Embryonic Stem Cell Core. mESCs were cultured on 0.1% gelatin-coated (Millipore) T25 flasks. The complete culture medium consisted of high-glucose Dulbecco's Modified Eagle's medium (DMEM; Gibco) supplemented with 10% newborn calf serum (NBCS; Gibco), 10% fetal bovine serum (FBS; GeminiBio), 1× nucleosides (Sigma-Aldrich), 1000 U mL−1 leukemia inhibitory factor (LIF; Millipore), and 100 µM beta-mercaptoethanol (BME; Gibco). Cells were passaged every 2 days and seeded at a density of 1.75 × 106 cells per flask in fresh complete medium. Preparing embryoid bodies (EBs) Polydimethylsiloxane (PDMS) microwell inserts (400 × 400 µm inverted pyramids) were used to form 3D embryoid body (EB) aggregates of mESCs. Inserts were sterilized by sonicating twice in 100% isopropyl alcohol, placed in fresh 24-well plates, washed again with 100% isopropyl alcohol, then rinsed three times in excess 1× Dulbecco's phosphate-buffered saline (dPBS) followed by UV sterilization for 30 minutes. Prior to seeding cells, 500 µL of DFK5 medium was added to each insert and centrifuged at 2000 × g for 5 minutes to remove bubbles. DFK5 consisted of a DMEM/F12 base supplemented with 5% Knockout Serum Replacement (KSR; Gibco), 50 µM non-essential amino acids (NEAA; Gibco), 100 µM BME, 1 : 100 100× Insulin–Transferrin–Selenium (ITS; Gibco), 1 : 200 1× Nucleosides. 160 000 cells were seeded into each microwell insert then centrifuged at 200 × g for 5 minutes to achieve approximately 130 cells per EB. EBs were allowed to form undisturbed for 2 days. Embedding embryoid bodies (EBs) into hydrogels Solutions of HA–Nor–ADH, HA–IKVAV, and HA–RGD were dissolved at 3% w/v in 1× dPBS, while HA–Ox and Alg–Ox were dissolved at 6% w/v in dPBS. In a 48-well plate, 250 µL of gelatin were added to control wells and 100 µL of HA–Nor–ADH, HA–IKVAV, or HA–RGD were added to their respective wells. For each plate, 2 mL of EB suspension were collected from two microwells, centrifuged at 200 × g for 5 minutes, and resuspended in 400 µL of DFK5 medium. 100 µL of HA–Nor–ADH, HA–IKVAV, or HA–RGD were added to the well. Next, 50 µL of EBs were gently mixed with HA hydrogels for an estimated seeding density of 300 EBs (∼6.4 × 105 cells per hydrogel). 50 µL of crosslinker (HA–Ox or Alg–Ox) were added to their respective wells and gently mixed, which resulted in a final hydrogel volume of 200 µL with a polymer concentration of 3% w/v. To avoid EB breakage, P200 pipette tips were trimmed using a sterile blade to increase the bore diameter. The hydrogels were allowed to crosslink for 30 minutes before adding 400 µL of DFK5 medium supplemented with 10 nM retinoic acid (RA; Sigma-Aldrich) and 1 µM purmorphamine (Pur; Millipore). For the control wells, 50 µL of EB suspension were added to a gelatin-coated 48-well plate containing DFK5 media supplemented with 10 nM RA and 1 µM Pur. Cells were then cultured for 2 days at 37 °C and 5% CO2. Differentiation and maturation of V2a interneurons Cells were enriched for V2a interneurons following previously established protocols in mESCs.41,44,45 After 2 days of culture in hydrogels, the medium was replaced with DFK5 medium supplemented with 10 nM RA, 1 µM Pur, and 5 µM N-[N-(3,5-difluorophenacetyl)-l-alanyl]-S-phenylglycine t-butyl ester (DAPT; Sigma-Aldrich). DAPT was included to inhibit the Notch signaling pathway and initiate differentiation to V2a interneurons. The EBs were cultured for an additional 2 days. On Day 6, the control wells cultured on gelatin were trypsinized for 5 minutes at 37 °C using 0.25% w/v trypsin-EDTA and transferred to laminin-coated wells. To prepare these wells, 150 µL of 0.01% v/v poly-l-ornithine (Millipore Sigma) solution was added to each well and incubated at 37 °C for 1 hour. The poly-l-ornithine was aspirated, and the wells were washed five times with 250 µL Tris-buffered-saline (TBS; Fisher Scientific). 100 µL of 0.01 mg mL−1 laminin (Thermo Fisher Scientific) diluted in HEPES buffered saline (Millipore Sigma) were added to each well and incubated overnight at 4 °C. Prior to plating cells, the laminin solution was aspirated and washed once with HEPES buffered saline. Approximately 175 000 V2a interneurons (∼184 210 cells per cm2) were plated in maturation medium (DFK5NB) in each laminin-coated well. DFK5NB consisted of a 1 : 1 ratio of DFK5 and neurobasal media (Gibco), supplemented with 1× B-27 (Gibco), and 1× GlutaMAX (Gibco). The differentiation medium in the hydrogel-containing wells was similarly exchanged for 400 µL of DFK5NB maturation medium on Day 6. All cultures were then maintained until Day 12 of the protocol, with replacement of 50% of the medium performed every 2 days between Day 6 and Day 12 to yield mature V2a interneurons. Live/dead assay Live/dead staining was performed using the LIVE/DEAD Viability/Cytotoxicity Kit for mammalian cells (Invitrogen). Briefly, 1 µM of Calcein AM and 4 µM of Ethidium Homodimer-1 (EthD-1) were mixed in 1× dPBS, 300 µL of solution were added to each well and incubated at 37 °C for 1 hour protected from light. Cells were then imaged using the Zeiss AxioObserver. RNA extraction and purification Cells were collected for RNA extraction by adding 500 µL of TRIzol reagent (Thermo Fisher) to each well and incubating at room temperature for 5 minutes. Following incubation, 200 µL of chloroform (Fisher Scientific) were added, and tubes were vortexed for 15 seconds. The tubes were incubated on ice for 10 minutes followed by centrifugation at 12 000 × g for 15 minutes at 4 °C to induce phase separation. 150 µL of the upper aqueous phase were collected, mixed with 250 µL of 70% ethanol, and purified using the RNeasy Mini Kit (Qiagen) according to the manufacturer's instructions. To elute the RNA, 30 µL of RNase-free water were added directly to the column membrane and centrifuged for 1 minute at 23 000 × g. To maximize RNA yield, the eluate was pipetted back onto the spin column and centrifuged for additional minute. cDNA synthesis The purified RNA samples underwent cDNA conversion using the High-Capacity RNA-to-cDNA kit (Applied Biosystems). Briefly, 10 µL of RT Buffer and 1 µL of RT Enzyme Mix were added to an 8-tube PCR strip. For each sample, 80 ng of RNA were added, and the final reaction volume was brought up to 20 µL using Ultrapure DNase/RNase free water (Invitrogen). The reverse transcription reaction was performed in a thermocycler with an incubation at 37 °C for 60 minutes, followed by enzyme inactivation at 95 °C for 5 minutes. Quantitative real-time PCR (qPCR) For qPCR, a master mix was prepared for each target gene (Chx10, Slc17a6, Mapt, Ppia) by adding 5 µL of PowerTrack SYBR Green Master Mix (Applied Biosystems), 1 µL of a 100 µM primer mix (Table 2) and 3 µL of ultrapure RNase/DNase free water per reaction. Next, 9 µL of the respective master mix and 1 µL of cDNA were added to the wells of an optical 96-well reaction plate (Applied Biosystems). The plates were sealed, vortexed, and centrifuged to remove bubbles. qPCR was performed using the BioRad CFX96 Real-Time PCR Detection System with the following thermal cycling conditions: an initial denaturation step at 95 °C for 2 minutes, followed by 40 cycles of denaturation at 95 °C for 15 seconds and annealing/extension at 60 °C for 1 minute. Following amplification, a continuous melt curve analysis was performed (95 °C for 15 seconds, 60 °C for 1 minute, followed by a temperature ramp to 95 °C) to verify product specificity. Table 2Table of primer pairs for qPCR listed 5′ to 3′Target geneGene functionForward primerReverse primer Ppia Housekeeping gene that catalyzes protein foldingCATACAGGTCCTGGCATCTTGTCAGACCACATGCTTGCCATCCAG Chx10 Transcription factor that acts as the primary marker for V2a interneuron identityGGCATAATGGATTCCTGTGCTACAGTCCCCAGAACCTTGG Slc17a6 Encodes the vesicular glutamate transporter 2 (VGLUT2) and indicates the mature, excitatory phenotype characteristic of V2a interneuronsCCTATGCTGGAGCAGTCATTGCGGCTCTCATAAGACACCAGAAGC Mapt Serves as a general marker for post-mitotic neuronal maturationCCTGAGCAAAGTGACCTCCAAGCAAGGAGCCAATCTTCGACTGG Table of primer pairs for qPCR listed 5′ to 3′ Table 3Table of antibodies used for immunostainingAntibodySpeciesDilutionCatalog no.Anti-βTubbIIIRabbit1 : 1000Biolegend 802001Anti-Rabbit 555Goat1 : 500Invitrogen A-21428 Table of antibodies used for immunostaining Relative gene expression was calculated by using the comparative cycle threshold (2−ΔΔCq) method. Target gene expression was normalized to Ppia as the housekeeping gene, and final gene expression levels were reported as log2 fold change relative to the control group (laminin). Immunocytochemistry After 12 days of culture, cells were washed with 250 µL of 1× dPBS and fixed with 200 µL 4% w/v paraformaldehyde (Thermo Fisher Scientific) for 30 minutes at room temperature. The wells were then washed three times with 250 µL of 1× dPBS. To permeabilize cells, 200 µL of 0.1% v/v Triton X-100 in 1× PBS were added to each well and incubated at room temperature for 15 minutes. The wells were washed three times with 200 µL of washing buffer (0.1% v/v Tween-20 in 1× PBS). To prevent non-specific binding, 200 µL of blocking buffer, consisting of 1% w/v bovine serum albumin (BSA; Sigma-Aldrich), 22.5 mg mL−1 glycine (Alkali Scientific), 5% v/v normal goat serum (NGS; Sigma-Aldrich) and 0.1% v/v Tween-20 (Sigma-Aldrich) in PBS, were added to each well. The plates were incubated overnight at 4 °C with shaking at 150 rpm. The following day, wells were washed three times for 15 minutes each with 200 µL of washing buffer. Cells were then incubated with 200 µL of primary antibody solution, consisting of rabbit anti-βIII-tubulin (1 : 1000; BioLegend) in 1% w/v BSA and 2% v/v NGS in 1× PBS overnight at 4 °C with shaking at 150 rpm (Table 3). After primary antibody incubation, the wells were washed three times for 15 minutes each with 200 µL of washing buffer. Next, 200 µL of secondary antibody solution containing goat anti-rabbit 555 (1 : 500; Invitrogen) and DAPI (1 : 1000; Invitrogen) in 1% w/v BSA and 2% v/v NGS in 1× PBS were added to each well. Plates were incubated overnight at 4 °C with shaking at 150 rpm. Following the secondary antibody incubation, the wells were washed three times with washing buffer with gentle agitation at 20 rpm for 30 minutes at room temperature. Finally, 200 µL of 1× PBS were added to each well, and the cells were imaged using the Zeiss Observer microscope. Live/dead image analysis All image analysis was performed with ImageJ. For each condition, one image per biological replicate (n = 3) was analyzed per hydrogel, capturing approximately five EBs per hydrogel. Images were converted to 8-bit image type, and z stacks were projected to maximum intensity. Channels were split for Calcein AM (green) and EthD-1 (red). The background was subtracted using a rolling ball radius of 600 pixels, and image contrast was enhanced for saturated pixels of 0.35%. To ensure visual consistency across all hydrogel conditions, fixed display intensity thresholds were uniformly applied to all images (Green channel = [15–120]; Red channel = [10–90]) for Fig. 3. For viability quantification, an Otsu threshold was performed and manually adjusted for each image to account for variations in dye penetration and light scattering within the 3D hydrogel matrix.59,60 Binary images were then cleaned by filling holes, and open commands were used to remove debris. The area and integrated density were measured, and particles larger than 100 µm2 were analyzed. The percentage viability was calculated using eqn (4).4 Neurite length analysis All image processing for immunostaining was performed with ImageJ. For neurite length analysis, a total of three images per biological replicate (n = 3) were analyzed and averaged, capturing approximately 18 EBs per hydrogel. Images were converted to 8-bit and z stacks were projected to maximum intensity. Channels were split for DAPI (blue), TUBB3 (red), and Brightfield (gray). The background was subtracted using a rolling ball radius of 600 pixels, and image contrast was enhanced for saturated pixels of 0.35%. An Otsu threshold was calculated for each individual image, and a global threshold was calculated and applied across all images for Fig. 4. For neurite length quantification, image brightness and contrast were adjusted using the set display range to oversaturate the red channel (50–86) and make neurites more visible. Neurite lengths from the aggregate were manually measured using the neuroanatomy plug in and the simple neurite tracer software in ImageJ. Results Polymer modifications for the synthesis of hydrazone crosslinked HA–Alg hydrogels To form hydrazone-crosslinked HA–Alg hydrogels (Fig. 1), alginate was oxidized resulting in a ring break, exposing aldehydes for crosslinking (Alg–Ox). A hydroxylamine hydrochloride titration was performed to calculate average percent degree of modification (%DOM) or the percentage of aldehydes per chain of alginate, which was 61.26 ± 8.82% with an average percent yield of 72.65 ± 10.93%. HA was then modified with norbornene functional groups as previously described and optimized for a low degree of modification to favor the downstream peptide conjugation.54,61 The average %DOM, or the number of norbornene functional groups per chain, of HA was 4.75 ± 2.21% with an average percent yield of 42.77 ± 1.93%. HA–Nor was then functionalized with adipic acid dihydrazide (HA–Nor–ADH) for crosslinking with Alg–Ox. An average %DOM of adipic acid dihydrazide groups per chain of HA determined by 1H NMR was 37.50 ± 5.71% with an average percent yield of 65.19 ± 5.29%. HA–Nor–ADH was then used to conjugate cysteine terminal peptides IKVAV and RGD using a photoinitiated thiol–ene click chemistry reaction (HA–IKVAV or HA–RGD). The average concentration of peptides conjugated to the backbone of HA was determined using an Ellman's assay, which detects the disappearance of free thiol groups before and after UV exposure and conjugation. The average concentration of IKVAV on the polymer was determined to be 12.69 ± 6.10 nmol mg−1 HA, and the average concentration of RGD on the polymer was determined to be 14.27 ± 7.55 nmol mg−1 HA. A summary of polymer modification quantifications can be found in Table S1. Polymer modifications for the fabrication of HA–Alg hydrogels. Alginate was oxidized to reveal aldehyde groups for crosslinking (Alg–OX). An EDC/HoBt coupling reaction was performed with HA–Nor to form HA functionalized with adipic acid dihydrazide and norbornene (HA–Nor–ADH). A photoinitiated reaction between HA–Nor–ADH and the terminal cysteine on the IKVAV or RGD peptide was then performed with UV light at a wavelength of 365 nm. The final products of these reactions, HA–Nor–ADH and Alg–Ox, were mixed at different polymer concentrations to produce hydrazone crosslinked hyaluronic acid–alginate (HA–Alg) hydrogels. HA–Alg hydrogels exhibit a range of physicochemical properties HA and alginate were mixed at different polymer concentrations (1–3% w/v) to form hydrogels. We confirmed hydrogel injectability by passing the hydrogel through a needle, and the physiochemical properties of the HA–Alg hydrogels were evaluated to determine the optimal hydrogel formulation for neuronal culture. Our target gelation time was between 1 to 5 minutes to produce a workable hydrogel for embedding mESCs for V2a interneuron differentiation and maturation. Increasing the polymer concentrations of the hydrogel decreased the gelation time, likely due to the increase of functional groups available for crosslinking. All hydrogels gelled within 5 minutes except for the HA1.0Alg1.0, HA1.0Alg2.0, and HA2.0Alg1.0 hydrogels, which had lower polymer concentrations (Fig. 2A). We targeted a hydrogel stiffness that would match the stiffness of CNS tissue (100–1000 Pa). We observed an overall increase in compressive modulus with increasing polymer concentrations (Fig. 2B). The HA3.0Alg3.0 hydrogel was the only hydrogel to display a compressive modulus above 100 Pa. Lastly, we measured the mass change of the hydrogels in both PBS and aCSF over 28 days to observe swelling and degradation over time (Fig. 2C and D). We targeted a stable hydrogel, defined as a mass change of no more than ± 25% of its original mass over 28 days, to enable long-term neuronal culture. Hydrogel stability in both PBS and aCSF decreased as polymer concentrations deviated from a 1 : 1 stoichiometric ratio. Additionally, hydrogel formulations with higher alginate concentrations were less stable in aCSF compared to PBS due the increased concentration of ions in the solution. Multivalent cations in highly ionic environments such as aCSF can interact with the negatively charged carboxyl groups on the alginate via electrostatic interactions, which has been previously reported to change alginate hydrogel stability and other properties.36 Physiochemical properties of HA–Alg hydrogels. HA–Alg hydrogels with various concentrations of HA and alginate denoted by subscripts (HA%w/vAlg%w/v). Shaded regions indicate desired hydrogel properties. (A) The gelation time of HA–Alg hydrogels decreased with increasing polymer concentration. (B) The compressive modulus of HA–Alg hydrogels increased with increasing polymer concentration. (C) Mass change of HA–Alg hydrogels in phosphate-buffered saline (PBS) over 28 days. (D) Mass change of HA–Alg hydrogels in artificial cerebrospinal fluid (aCSF) over 28 days. One-way ANOVA with Tukey's post hoc test was performed. Statistically significant differences in (A) and (B) are summarized using the compact letter display, where differing letters represent groups that are statistically different (p < 0.05) and the same letter represents groups with no statistical differences (p > 0.05) for three technical replicates (n = 3). We determined that the hydrogel formulation HA3.0Alg3.0 displayed physicochemical properties that best aligned with our desired properties for neural cell culture. This hydrogel exhibited an average gelation time of 1.2 ± 0.3 minutes and a compressive modulus of 130.3 ± 9.6 Pa. This compressive modulus fell within the range of stiffness reported for neural tissue (100–1000 Pa),62,63 although it was significantly lower than the Young's modulus of spinal cord tissue, which varies between 0.012 and 1.37 MPa depending on the region of the spinal cord.64,65 Although the HA3.0Alg3.0 hydrogel lost more than 25% of its original volume in the first 14 days in both PBS and aCSF, the hydrogel stabilized within 25% of its original mass by day 28. A previously characterized HA–HA hydrogel with a similar formulation (3% w/v) displayed a gelation time of 1.4 ± 0.3 minutes, a compressive modulus of 25.91 ± 5 Pa, and a relatively stable mass change within ±25% over 28 days in PBS and was used as a comparison for subsequent experiments.53 We demonstrated that the HA3.0HA3.0 hydrogel formulation was injectable through a 22-gauge needle (Fig. S3A), and that both 3% w/v HA–Alg and HA–HA hydrogels exhibited shear-thinning properties (Fig. S3B). Using SEM imaging, we also determined the average porosity of HA–HA and HA–Alg hydrogels to be 17.0 ± 8.3% and 22.8 ± 6.2%, respectively, with no statistically significant differences between hydrogels (Fig. S4). Conjugating IKVAV and RGD peptides to HA–Nor–ADH to form peptide-containing hydrogels resulted in approximately 0.38 ± 0.18 mM of IKVAV per hydrogel and 0.43 ± 0.23 mM of RGD per hydrogel, which met the requirements for neural cell adhesion.66,67 HA–Alg hydrogels support V2a interneuron viability To evaluate the ability of the HA–Alg hydrogels to support neuronal culture, V2a interneurons were differentiated from mESCs, cultured for 12 days in HA–Alg hydrogels, and compared to cells cultured in previously established HA–HA hydrogels and 2D gelatin/laminin-coated control wells (Fig. S5).41,44,45 Following 2 days of forced aggregation, EBs were mixed with either unmodified or peptide-modified HA–ADH, either HA–Ox or Alg–Ox was added, and the hydrogels were crosslinked for 30 minutes. After 12 days of culture, quantitative live/dead analysis demonstrated that the incorporation of the laminin-derived IKVAV peptide was essential for sustaining 3D cell viability at levels comparable to standard 2D laminin controls (Fig. 3A and B). While viability did not differ between any of the hydrogels themselves, cells embedded in the unmodified HA + Alg hydrogels (54.9 ± 2.6%) and RGD-modified HA–RGD + HA and HA–RGD + Alg hydrogels (50.0 ± 7.7% and 52.7 ± 1.4%, respectively) exhibited significantly lower viability than cells in the 2D laminin control (91.4 ± 6.0%). In contrast, IKVAV-modified HA–IKVAV + HA and HA–IKVAV + Alg hydrogels supported cell survival, yielding viabilities (77.1 ± 7.2% and 72.4 ± 20.9%) that were comparable to the 2D laminin control. V2a interneurons remain viable and express neural markers in HA–Alg and HA–HA hydrogels. (A) Representative live/dead fluorescent images of cells encapsulated within hydrogels after 12 days in culture. Live cells are stained green (Calcein-AM), and dead cells are stained red (EthD-1). Scale bars = 250 µm. (B) Quantification of cell viability expressed as the percentage of live cells over total cells. Data are presented as mean ± standard deviation. One-way ANOVA, post hoc Tukey's multiple comparisons was performed for individual biological replicates (N = 3). Statistics are summarized using the compact letter display, where differing letters represent groups that are statistically different (p < 0.05) and the same letter represents groups with no statistical differences (p > 0.05). (C) Relative gene expression (log 2 fold change) of neural differentiation markers Chx10, Mapt, and Slc17a6 quantified via RT-qPCR. Bars represent the mean log 2 fold change relative to the 2D laminin control, and error bars represent the standard error of the mean (SEM). Individual data points represent independent biological replicates (N = 3). No statistically significant differences were found between hydrogel formulations (p > 0.05) using a one-way ANOVA with Tukey's post hoc test. Evaluation of V2a interneuron differentiation in HA–Alg hydrogels To assess if the culture had been successfully enriched for V2a interneurons after 12 days, quantitative real-time PCR (qPCR) was performed. Gene expression profiles of cells embedded in all hydrogel formulations were compared to those of cells cultured on standard 2D laminin coatings (Fig. 3C). Primer specificity and reaction efficiencies were confirmed via melt curve analysis and amplification plots (Fig. S3 and S4), and the stable expression of the endogenous control gene Ppia was verified across all samples. Overall, embedding cells within the hydrogels promoted upregulation of the V2a interneuron marker Chx10 and general neuronal maturation markers (Mapt and Slc17a6) compared to the 2D laminin control (Fig. 3C). Expression of Chx10, a transcription factor expressed in V2a interneurons, was upregulated across all hydrogel formulations. Interestingly, despite their restricted morphological development, unmodified hydrogel formulations demonstrated some of the highest relative expression levels of Chx10. To assess the maturation of the differentiated neurons, the expression of Mapt, a general marker of mature neurons, and Slc17a6, the gene that encodes for VGLUT2, a marker of excitatory glutamatergic neurons, were evaluated. Similar to the Chx10 profiles, Mapt and Slc17a6 were increased across most of the hydrogel conditions compared to laminin. Importantly, all hydrogel formulations supported comparable expression of Chx10 and Slc17a6 compared to 2D laminin controls, confirming successful enrichment of V2a interneurons. The comparable gene expression profiles observed across the different hydrogel formulations indicate that the small molecules driving differentiation were efficiently diffusing through the 3D polymer networks, regardless of the polymer type, crosslinking chemistry, and peptide modification. These transcriptional profiles further suggest that early lineage specification was primarily driven by soluble media cues rather than matrix-bound adhesive signals. IKVAV-modified HA–Alg hydrogels enhance V2a interneuron neurite outgrowth The surrounding hydrogel network heavily influenced the morphology and structural maturation of the cells in 3D. To characterize this development, we evaluated EB morphology alongside immunostaining for the neuron-specific microtubule marker βIII-Tubulin (TUBB3) after 12 days of culture. As expected, cells cultured on 2D laminin controls exhibited characteristic planar cellular adhesion and dispersed TUBB3 positive neurite networks (Fig. S5). In 3D cultures, the biochemical makeup of the microenvironment dictated the structural fate of the V2a interneurons (Fig. 4A and Fig. S5). EBs encapsulated within the unmodified hydrogels (HA + HA and HA + Alg) formed dense, DAPI-positive spheroids that maintained tightly restricted morphologies. These structures were characterized by distinct smooth borders and an absence of outward neurite extension. Interestingly, modifying hydrogels with the fibronectin-derived RGD peptide (HA–RGD + HA and HA–RGD + Alg) failed to overcome this physical confinement. TUBB3 staining indicated a lack of macroscopic cytoskeletal extensions with minimal expression restricted to the EB surface. While a small fraction of EBs within these unmodified and RGD functionalized environments occasionally exhibited small projections, overall structural development was negligible. Immunostaining assessing V2a interneuron maturation and neurite length. (A) Representative maximum intensity projections of V2a interneurons cultured for 12 days in HA–HA and HA–Alg crosslinked hydrogels (unmodified, RGD, and IKVAV). Cells were immunostained for the neuronal microtubule marker βIII-Tubulin (TUBB3, red) and nuclei (DAPI, blue). Scale bars = 250 µm. (B) Zoomed in images for the HA–IKVAV + HA and HA–IKVAV + Alg hydrogels, represented by the dotted boxes in (A). Scale bars = 50 µm. (C) Quantification of average neurite length (µm) extending from the embryoid bodies in various hydrogel formulations and the 2D laminin control. One-way ANOVA, post hoc Tukey's multiple comparisons was performed for individual biological replicates with three technical replicates (N = 3, n = 3). Error bars represent the standard error of the mean (SEM). Statistics are summarized using the compact letter display, where differing letters represent groups that are statistically different (p < 0.05) and the same letter represents groups with no statistical differences (p > 0.05). Conversely, functionalization with the laminin-derived-peptide IKVAV (HA–IKVAV + HA and HA–IKVAV + Alg) established a permissive microenvironment that facilitated extensive structural development. EBs within IKVAV-modified hydrogels demonstrated clear morphological maturation, marked by TUBB3 positive neurites projecting radially into the surrounding 3D hydrogel space. These robust microtubule networks demonstrate that IKVAV not only permits physical cell spreading but actively supports the cytoskeletal assembly required for mature neuronal process extension. While the presence and length of extending neurites naturally exhibited some heterogeneity within samples, outgrowth within the IKVAV-containing hydrogels was overwhelmingly consistent. Quantitative analysis of neurite length corroborated these visual observations (Fig. 4C). IKVAV-modified hydrogels supported substantial neurite outgrowth, averaging approximately 197.2 ± 53.5 µm for HA–IKVAV + HA and 212.4 ± 75.2 µm for HA–IKVAV + Alg hydrogels that was comparable standard 2D laminin controls (232.3 ± 25.6 µm) where robust neurite extension would be expected. Conversely, the unmodified and RGD-modified hydrogels yielded negligible extensions. This morphological divergence demonstrates that incorporating laminin-derived peptides like IKVAV effectively breaks the physical restrictions around the EBs and drives the maturation and neurite outgrowth of V2a interneurons in 3D. Taken together with the gene expression profiles, while chemical differentiation proceeded efficiently across all groups, the physical architecture and biochemical cues provided for the IKVAV-modified hydrogels was critical for structural maturation. Discussion The successful maturation of neurons within defined 3D hydrogel microenvironments is a critical step toward developing effective cell-based regenerative therapies for SCI. In this study, we developed hydrazone crosslinked HA–Alg hydrogels modified with fibronectin- and laminin-derived peptides (RGD and IKVAV, respectively) and evaluated their capacity to support the viability, lineage specification, and morphological development of mESC-derived V2a interneurons. A primary challenge in neural tissue engineering is designing scaffolds that mimic the environment of the CNS without sacrificing 3D architecture. By optimizing the polymer concentration to the HA3.0Alg3.0 formulation, we achieved a compressive modulus of 130.3 ± 9.6 Pa, aligning with the properties of native CNS tissue (100–1000 Pa). Previous literature has demonstrated that stiffer substrates (∼1000–10 000 Pa) drive neural progenitor cells toward a glial lineage and exacerbate astrogliosis, whereas compliant matrices (∼100–500 Pa) are permissive to neurogenesis and neurite outgrowth.68 Consistent with this, primary cortical and spinal neurons have been shown to grow preferentially on soft matrices of 200 Pa and 230 Pa, respectively.69,70 Additionally, neural stem cells encapsulated within alginate hydrogels with a low stiffness of 183 Pa, similar to our hydrogel formulation, exhibited high TUBB3 expression.71 Thus, we hypothesize maintaining a hydrogel stiffness ∼100 Pa provides a soft matrix that prioritizes mESC neurogenesis while suppressing glial commitment. In addition to matching target stiffness, the gelation kinetics of our HA–Alg system were optimized to address the practical challenges of embedding cells in 3D. Slower-gelling systems frequently suffer from gravity-induced cell sedimentation before crosslinking is complete, leading to heterogeneous cell distributions, localized nutrient deprivation, and inconsistent cell–matrix interactions. We observed that the HA3.0Alg3.0 formulation gels within 1.2 ± 0.3 minutes, which is rapid enough to ensure uniform 3D cell suspension.72 The most significant physicochemical advantage of our system is its long-term stability in physiological fluids. Conventional alginate hydrogels rely on transient ionic crosslinks, which rapidly dissociate in sodium-rich or chelator-containing media due to ion exchange.73 This premature dissolution limits their utility in neural cultures, as neurons require extended timelines for functional maturation. By using covalent hydrazone bonds between oxidized alginate and ADH-modified HA, our hydrogels overcame this limitation, stabilizing within 25% of their original mass over 28 days in ion-rich aCSF. We compared HA–Alg hydrogels to a previously established HA–HA hydrogel to specifically evaluate the effects of alginate on cell viability and maturation. While HA is a ubiquitous component of the neural ECM, our results demonstrated that an HA–Alg hybrid system is equally cytocompatible compared to HA–HA hydrogels. Similar hydrazone crosslinked HA–HA and HA–Alg hydrogels have previously displayed high biocompatibility, as demonstrated by low cytotoxicity and high cell viability, and have been used to support tissue repair including myocardial and nucleus pulposus tissue repair.74,75 Our findings are significant as they validate the use of alginate as a tunable and clinically relevant copolymer for neural cell culture that does not compromise the ability of the scaffold to support neural differentiation and neurite outgrowth. Our findings reveal a notable divergence between differentiation and morphological maturation within these 3D environments. While V2a interneuron differentiation was efficiently driven by soluble cues across hydrogel conditions, we observed that structural maturation and neurite outgrowth required specific ECM-derived peptides (Fig. 4A). Specifically, our findings highlight that, while both the HA–Alg and HA–HA backbones provided a stable structural system for supporting cell viability, the presence of laminin-derived IKVAV was the essential driver for neurite extension, thereby dictating functional cellular behavior. Gene expression profiling revealed that both peptide-modified and unmodified hydrogels supported the upregulation of the V2a interneuron transcription factor Chx10 and mature neuronal markers (Mapt, Slc17a6) (Fig. 3C). The success of this differentiation across all hydrogel formulations indicates that the small molecules added for differentiation were equally effective across the different hydrogel networks and were not hindered by the crosslinking chemistry or presence of peptides. The successful upregulation of Chx10 is a critical benchmark for V2a interneuron differentiation, as Chx10+ V2a interneurons play an integral role in driving the excitatory pathways of motor neurons and are essential for locomotor initiation. In fact, recent studies have demonstrated that V2a interneurons are explicitly required for the recovery of critical motor functions following cervical SCI, making them a primary therapeutic target for regenerative medicine.38,43,76,77 Interestingly, the morphology of V2a interneurons varied drastically across hydrogel formulations. While unmodified hydrogels and hydrogels modified with RGD exhibited tightly confined EB morphology and lacked meaningful neurite projections, IKVAV-modified hydrogels facilitated radial neurite extension (Fig. 4A and Fig. S5). Furthermore, immunostaining revealed dense, TUBB3+ microtubule networks (Fig. 4A and Fig. S5). This structural divergence highlights that, while the basic physical properties of the hydrogels may permit cell survival, tissue-specific biochemical cues were required to drive structural maturation. Universal cell-adhesive motifs like RGD have been shown to successfully support neural differentiation and neurite extension in specific defined 3D microenvironments, including HA and polyethylene glycol-based hydrogels.78,79 However, although few V2a interneurons exhibited spontaneous neurite extension in RGD-modified hydrogels similar to unmodified HA hydrogels, the morphological maturation within RGD networks proved highly variable and less robust than IKVAV-functionalized hydrogels. These findings reflect established in vitro standards, as laminin is often used in neural differentiation protocols. Furthermore, laminin plays a central role in the native CNS ECM, targeting specific integrin pathways that are essential for neural maturation. By integrating the IKVAV peptide into our 3D scaffold, we successfully recapitulated the biochemical cues traditionally provided by 2D laminin coatings, thereby better mimicking the native CNS environment. Notably, our peptide conjugation strategy yielded relatively low peptide concentrations compared to conventional hydrogels, resulting in approximately 0.38 ± 0.18 mM IKVAV and 0.43 ± 0.23 mM RGD per hydrogel compared to 2–6 mM observed in literature.80–82 Given our low peptide concentrations, it is likely that our RGD hydrogels lacked the ligand density required to reach the critical threshold for neural adhesion, aligning with findings from neurospheres cultured on HA–RGD-modified hydrogels where neurite density was directly dependent on RGD concentration.79 In contrast, IKVAV mimics the native laminin-rich ECM of the central nervous system, engaging neural specific receptors to promote true microtubule assembly, possibly requiring a lower concentration of conjugated peptide for a robust morphological response. While fibronectin-binding integrins are expressed in early stages of stem/progenitor cell neuronal differentiation, they are reported to be downregulated as neurons mature,83 and late-stage neurons predominantly express laminin-binding integrins (α1β1 and α6β1).69 These differences in integrin expression further support the hypothesis that higher concentrations of RGD are needed to activate cell adhesion while lower concentrations of IKVAV may be needed to elicit similar responses. Native brain and spinal cord ECM heavily rely on complex proteoglycans and structural proteins to regulate the adhesion of neurons and drive subsequent neurite extension.84 Our results align with recent studies demonstrating that polymers such as alginate or HA lack the requisite motifs to engage neural integrins, thereby acting as physical barriers rather than permissive scaffolds unless modified. For instance, alginate hydrogels functionalized with laminin and gelatin significantly enhanced the neurite outgrowth and migration of human induced pluripotent stem cell (hiPSC)-derived neurospheres, suggesting that ligand–receptor interactions are required to trigger the cytoskeletal remodeling necessary for axonal extension.85 Similarly, hydrogels containing alginate and collagen supported robust neural network formation.86 The success of these hybrid systems is largely attributed to the fact that gelatin and collagen naturally present a high density of cell-adhesive sequences, including RGD, which provide the necessary biochemical anchors for neural attachment. However, while these general motifs support broad neural network formation, V2a interneurons appear to require more specialized signaling for functional development. Specifically, one study found that the incorporation of astrocyte-derived protoplasmic ECM into HA hydrogels promoted V2a interneuron process extension, with the ECM being abundant in fibronectin, collagen, perlecan, and laminin.48,87 In another study, laminin was required to support the viability of V2a interneurons embedded in HA hydrogels and treated with neurotrophin-3.8 These findings support our observation that V2a interneurons are highly sensitive to their microenvironment and require laminin-derived ECM recognition cues for maturation. Our use of precise peptide conjugation offers a defined approach for engaging the cell's mechanotransducive machinery. By utilizing IKVAV peptides, we successfully bypassed the restrictive physical boundaries of the hydrogel, allowing mESC-derived interneurons to achieve maturation. The heterogeneity in neurite length observed within these functionalized groups likely reflects the variability of mESC differentiation kinetics and intra-EB spatial positioning, which could be further optimized in future studies by tuning the peptide density or the degradation rate of the polymer network. Furthermore, we believe that these results could easily be translated to the differentiation of hiPSCs to interneurons. The protocol for V2a interneuron differentiation has already been optimized in hiPSCs,40,41 and successful neuronal differentiation of hiPSCs has previously been demonstrated in HA and alginate hydrogels.88–90 However, applying 3D peptide-modified hydrogels to drive the maturation and neurite extension of human V2a interneurons remains unexplored. Conclusion We successfully engineered an injectable, hydrazone crosslinked HA–Alg hydrogel designed to meet the complex physicochemical and biochemical requirements of neuronal cell culture with eventual applications in CNS tissue repair. We discovered a fundamental decoupling of biochemical differentiation and morphological maturation of V2a interneurons within 3D microenvironments. While small molecules for neuronal differentiation efficiently permeated all hydrogel formulations to drive the robust transcriptional differentiation of Chx10-expressing V2a interneurons, structural maturation required biochemical cell-adhesive cues. Modification of the hydrogels with the laminin-derived IKVAV peptide transformed the matrix into a highly permissive scaffold that maintained cell viability and drove extensive neurite growth. Furthermore, the HA–Alg composite hydrogel outperformed the similarly crosslinked HA–HA network in facilitating this structural maturation, highlighting the advantages of using blended polymer systems. Ultimately, the HA–IKVAV + Alg hydrogel represents a highly tunable, biomimetic platform capable of overcoming the physical restrictions of encapsulating stem cell aggregates in hydrogels. By successfully supporting the survival, differentiation, and structural maturation of therapeutically critical V2a interneurons, this injectable biomaterial offers a promising, minimally invasive strategy to transplant cells to promote functional regeneration following SCI in the future. Conflicts of interest There are no conflicts to declare. Supplementary Material