Development and characterization of antacid microcapsules to buffer the acidic intervertebral disc microenvironment

During intervertebral disc (IVD) degeneration, microenvironmental challenges such as decreasing levels of glucose, oxygen, and pH play crucial roles in cell survival and matrix turnover. Antacids, such as Mg(OH) 2 and CaCO 3 , entrapped in microcapsules are capable of neutralizing acidic microenvironments in a controlled fashion and therefore may offer the potential to improve the acidic niche of the degenerated IVD and enhance cell-based regeneration strategies. The objectives of this work were, first, to develop and characterize antacid microcapsules and assess their neutralization capacity in an acidic microenvironment and, second, to combine antacid micro-capsules with cellular microcapsules in a hybrid gel system to investigate their neutralization effect as a potential therapeutic in a disc explant model. To achieve this, we screened five different pH-neutralizing agents (Al(OH) 3 , Mg(OH) 2 , CaCO 3 , and HEPES) in terms of their pH neutralization capacities, with Mg(OH) 2 or CaCO 3 being carried forward for further investigation. Antacid-alginate microcapsules were formed at different concentrations using the electrohydrodynamic spraying process and assessed in terms of size, buffering kinetics, cell compatibility, and cytotoxicity. Finally, the combination of cellular microcapsules and antacid capsules was examined in a bovine disc explant model under physiological degenerative conditions. Overall, CaCO 3 was found to be superior in terms of neutralization capacities, release kinetics, and cellular response. Specifically, CaCO 3 elevated the acidic pH to neutral levels and is estimated to be maintained for several weeks based on Ca 2 + release. Using a disc explant model, it was demonstrated that CaCO 3 microcapsules were capable of increasing the local pH within the core of a hybrid cellular gel system. This work highlights the potential of antacid microcapsules to positively alter the challenging acidic microenvironment conditions typically observed in degenerative disc disease, which may be used in conjunction with cell therapies to augment regeneration


| INTRODUCTION
Cell-based therapies for intervertebral disc (IVD) regeneration may hold significant potential to rejuvenate disc tissue and provide a longterm solution for low back pain.2][3] In particular, the acidic pH evolving due to lactate accumulation and impaired metabolite diffusion during disc degeneration plays a crucial role for cell survival and matrix turnover. 4This is critical, not only for resident disc cells, but also for any of the cell types proposed for tissue regeneration.Many researchers have investigated the use of nucleus pulposus (NP) cells for regenerating the IVD, which have been shown to delay degenerative disc changes after reimplantation in animal models. 5,68][9] Therefore, alternative cell types including articular chondrocytes (ACs) have been proposed for NP regeneration. 10,11As fully differentiated primary cells, ACs exhibit some phenotypical differences to NP cells.However, they are easily accessible and capable of matrix deposition similar to NP tissue, which makes them attractive candidates to be considered for IVD repair.
Within a healthy IVD, the average pH has been reported to be 7.1. 3,12With the onset of degeneration, the pH within the NP decreases to approximately 6.7-6.9 and can decrease further to 6.5 in a severely degenerated IVD. 3,13Wuertz et al. demonstrated that matrix acidity has detrimental effects on bone marrow stem cells (BMSCs) inhibiting cell proliferation, diminished cell viability, and inhibited anabolic gene expression. 14This has also been observed for disc-chondrocytes cultured in nutrient-deprived conditions at a pH of 6.2 15 with the accumulation of sulfated glycosaminoglycans (sGAG), a key component of disc tissue, found to be inhibited below a pH of 6.8. 16,17In our previous work when investigating pH effects on cells cultured in 3D hydrogels, we also observed increased death and diminished matrix accumulation in lower pH microenvironments for both alginate encapsulated BMSCs 18 and NP cells cultured in extracellular matrix (ECM) hydrogels. 191][22][23] Our laboratory and others have demonstrated that inducing BMSCs toward a discogenic phenotype using the growth factor TGF-β improves matrix deposition and enhances cell viability in a disc-like microenvironment. 24,25 have also made similar findings using ACs, which deposited similar levels of sGAG and collagen compared to BMSCs in acidic conditions after 2 weeks of priming using TGF-β3, and exhibited greater viability compared with BMSCs at a pH of 6.5. 22Nevertheless, the harsh nutrient microenvironment of low oxygen, glucose, and pH remains, affecting resident cells and inhibiting the regeneration process.Acidity appears to be one of the main challenges, whereby it reduces viability and matrix anabolism.Therefore, a possible solution is to increase the local pH of the tissue toward neutral levels, which may improve cellular function and aid in the enhancement of ECM deposition.
Antacids are multicomponent salts with pH neutralizing properties and have been widely used to treat symptoms such as heartburn and dyspepsia, which are associated with hyper-acidic gastric fluids. 26ey are a group of salts whose active component is often based on aluminum hydroxide (Al(OH) 3 ), magnesium hydroxide (Mg(OH) 2 ), sodium bicarbonate (NaHCO 3 ), calcium carbonate (CaCO 3 ), or combinations thereof.CaCO 3 has previously been used for its acid-sensitive properties in drug release vehicles for the manipulation of cancer microenvironments. 27,28Specifically, a drug is encapsulated into a core of CaCO 3 , which dissolves within the acidic carcinogenic environment and releases its content.For instance, Zhao et al. fabricated a vehicle of amorphous calcium carbonate/doxorubicin@Silica, which released doxorubicin in mildly acidic conditions (pH 6.5), resulting in efficient cell death of cancer cells. 29HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), a zwitterionic sulfonic agent also has acidic buffering capacities and is commonly used to supplement cell culture media to maintain its pH within a neutral range (pH 6.8-8.2).
Microencapsulation of pH neutralizing salts has been shown to improve the release kinetics for prolonged functionality. 30For example, Chen et al. encapsulated Mg(OH) 2 into alginate spheres to investigate release kinetic patterns depending on sphere size and alginate concentration.They demonstrated a correlation of higher Mg(OH) 2 release in smaller alginate spheres as well as lower alginate concentration, indicating the ability to fine-tune the antacid delivery in low pH environments. 30In terms of hydrogels, fibrin has been widely employed in clinical settings as a sealant and adhesive.Previous work in our laboratory demonstrated that the incorporation of hyaluronic acid (HA) can increase the bioactivity of fibrin-based hydrogels resulting in enhanced sGAG accumulation by ACs. 31 Therefore, a combination of antacid microcapsules with a fibrin-based gel may provide promise for minimally invasive disc repair.
The overall objective of this work was to develop and characterize pH neutralizing microcapsules and investigate their potential to buffer the typical acidic conditions that exist in vivo.Specifically, different antacids (Al(OH) 3 , Mg(OH) 2 , CaCO 3 , or HEPES) were initially compared in terms of their pH neutralization capacities, with Mg(OH) 2 or CaCO 3 being carried forward for further investigation and characterization.Specifically, we examined the impact of microcapsules containing Mg(OH) 2 or CaCO 3 antacids on both cell viability and matrix accumulation in vitro and in an ex vivo disc explant model under physiological degenerative-like conditions.

| Experimental design
An overview of the experimental design is illustrated in Figure 1.
Overall, this study was divided into three phases.In the first phase, different types of buffering salts and agents (Al(OH) 3 , Mg(OH) 2 , CaCO 3 , and HEPES) were assessed in terms of their pH neutralization capacities using a pH sensor foil system (PreSens, Germany).Next the size of sprayed antacid microcapsules (Mg(OH) 2 or CaCO 3 ) with various concentrations was analyzed and the release kinetics from microcapsules into the local environment at low pH was investigated.The second phase examined the effect of the developed antacid microcapsules on therapeutic cells (primed AC cells) and resident NP cells in terms of viability and matrix accumulation capacities.Finally, the third phase assessed the combination of cellular microcapsules and antacid capsules in a bovine disc explant model under physiological degenerative conditions.

| Fabrication of antacid microcapsules
The desired amount of various antacids (Al(OH) 3 , Mg(OH) 2 , CaCO 3 , or HEPES) was mixed with 1% alginate (Pronova UP LVG, FMC NovaMatrix, Norway).For homogeneous distribution of the nanoparticles, the suspension was blended using a homogenizer for 2Â 1 min cycles followed by 3Â 1 min cycles of ultrasonication.The final solution was diluted using 1% alginate to obtain different concentrations (25-500 mg/mL).For microcapsule fabrication, an electrohydrodynamic (EHD) sprayer was used as described in our previous studies. 20,25,32,33iefly, the homogenized alginate-antacid solution was loaded into a syringe and passed through a 30G needle at a flow rate of 50 μL/mL and working distance of 50 mm, with an applied voltage of 10-15 kV to achieve a stable Taylor cone.Microcapsules were collected in a dish containing 100 mM CaCl 2 to ionically crosslink the alginate polymer.

| Preparation of low pH media
Low pH media (pH 6.5) was prepared by adding an appropriate amount of acid to low glucose DMEM (lg-DMEM).Briefly, physiological levels of lactic acid (4 mM), typical of those found in the IVD, 1 were added to the media and pH levels were adjusted by adding 3M hydrochloric acid (10 μL/mL).The pH-adjusted media was incubated overnight in a humidified incubator (5% CO 2 ) to allow buffer equilibrium (CO 2 -dependent).The desired pH values were obtained after equilibration and were maintained for up to 72 h.

| pH mapping and measurements
The pH concentration maps were generated using a USB pH detector unit-microscope device and corresponding software (VisiSens; Pre-Sens GmbH, Regensburg, Germany).Briefly, a circular section measuring 8 mm in diameter was cut from a pH sensor foil while ensuring minimal exposure to light (Product code SF-HP5R, Pre-Sens GmbH, Regensburg, Germany).The circular sensor foil was carefully attached to the bottom of a standard cell culture dish using a silicone glue.The foil was calibrated using a six-point calibration following the manufacturer's instructions.For the experiment, antacid microcapsules within a fibrin (50 mg/mL)-hyaluronic acid (HA) (5 mg/ mL) hydrogel was placed on the surface. 31The construct was cultured in low pH media (pH 6.5) and images were taken from below using a camera-based detector unit (DU02, VisiSens; PreSens GmbH, Regensburg, Germany) at various time points to assess local pH change.
F I G U R E 1 Experimental design.Phase 1 investigated physicochemical parameters of different antacids in terms of local pH change, size distribution, and release kinetics.Phase 2 examined the effect of pH buffering microcapsules on cells (therapeutic cells and resident native cells).In phase 3, the combination of cellular microcapsules and antacid microcapsules was assessed in a bovine disc explant model under physiological degenerative-like conditions.
Control wells with known pH levels (7.1 and 6.5) were used to ensure stability of the pH over the time course investigated (72 h).The output file generated by the VisiSens™ AnalytiCal 2 Software (PreSens GmbH, Regensburg, Germany) was analyzed using a custom written MATLAB ® code (MathWorks ® , Version R2017a, Massachusetts, US).
Briefly, every pixel within the produced image file represents a pH value derived from the calibration image, enabling the creation of a pH map across the sensor foil.For explant cultures, pH levels were measured using fiber-optical fluorescent microsensors (PreSens, Regensburg, Germany).

| Size analysis
For size analysis, antacid microcapsules (n $350-1000) of each group were imaged using light microscopy, and microcapsule diameters were determined using image analysis software (ImageJ, National Institutes of Health, and Bethesda, Maryland).

| Antacid salt release kinetics
The amount of calcium and magnesium retained in microcapsules and released into the media was quantified using the Sentinel Calcium and Magnesium Liquid (Alpha Laboratories Ltd., UK) assay kits in accordance with the manufacturer's instructions.At specific timepoints in culture, antacid microcapsules were dissolved in 1 M HCl solution for 1 h under constant rotation.Media samples were also analyzed directly without 1 M HCl treatment.

| Isolation and expansion of primary cells
All animal tissue used in this work was obtained from a local abattoir and did not require ethical approval.IVDs were harvested from the lumbar region of the spine of porcine donors (3-4 months, 20-30 kg) within 3 h.Under aseptic conditions, IVDs were carefully exposed and the gelatinous NP tissue removed from the central section of the disc. 34To confirm the absence of bacterial growth, dissected tissue was cultured overnight at 37 C, 5% CO 2 in a humidified atmosphere in serum-free low-glucose Dulbecco's Modified Eagles Medium (lg-DMEM, 1 g/L D-glucose) supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin.NP tissue was enzymatically digested in 2.5 mg/mL pronase solution for 1 h followed by 2 h in 0.5 mg/mL collagenase solution at 37 C under constant rotation.The digest was subjected to physical agitation cycles at the start and every 30 min thereafter using a gentlemacs tissue dissociator (Miltenyi Biotech).
Digested tissue/cell suspension was passed through a 100-μm cell strainer to remove tissue debris followed by 70 and 40 μm cell strainers to separate notochordal cells from the desired NP cells as previously described. 35Cells were then washed three times by repeated centrifugation at 650 g for 5 min.NP cells were cultured to confluence in T-75 flasks with lg-DMEM, supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, 100 μg/mL streptomycin, 0.25 μg/mL amphotericin B (AmpB), and 5 ng/mL FGF-2 (PeproTech, UK) at 37 C in a humidified atmosphere with 5% CO 2 , and expanded to passage 2 (P2) with medium exchanges performed every 3 days.Juvenile porcine ACs were isolated from cartilage of the knee joint.Tissue was washed with phosphate buffered saline (PBS) and finely minced.For cell isolation, minced tissue was digested with concentrations of 3000 U/mL collagenase type II (Gibco, Ireland) for $2.5 h under constant rotation at 37 C in serum-free hg-DMEM, (4.5 mg/mL D Glucose, 200 mM L-Glutamine;) containing antibiotic/ antimycotic (100 U/mL penicillin, 100 mg/mL streptomycin) (all Gibco, Invitrogen, Ireland) and AmpB (0.25 μg/mL, Sigma-Aldrich).
The digest was subjected to physical agitation cycles at the start and every 30 min thereafter using a gentlemacs tissue dissociator (Miltenyi Biotech, UK). 36The digested tissue/cell suspension was passed through a 100-μm cell strainer to remove tissue debris and washed three times by repeated centrifugation at 650 g for 5 min.

| Cellular microencapsulation
ACs were trypsinized and resuspended in media and combined with sterile 2% alginate solution (Pronova UP LVG, FMC NovaMatrix, Norway) at a 1:1 ratio to yield a final alginate concentration of 1% and a seeding density of 10 Â 10 6 cells/mL.The alginate/cell solution was electrosprayed using an EHD sprayer, 20,25,32,33 with constant processing parameters (30G needle, 10-15 kV applied voltage, 100 μL/mL flow rate, 50 mm working distance).After ionically crosslinking for 5 min inside a 100 mM CaCl 2 bath (pH 7.2), microcapsules were rinsed thoroughly with PBS before resuspension in the appropriate medium for culture.
Microcapsules were cultured at 37 C and low oxygen (5% O 2 ) conditions for 14 days under constant agitation.Half media exchanges were performed twice weekly.

| In vitro culture of AC microcapsules and NP cells with antacids
After the priming regime, AC cellular microcapsules and antacid microcapsules were mixed into a fibrin hydrogel (100 mg/mL) and cultured for 24 h for cytotoxicity (n = 4 per group).NP cells were directly encapsulated into fibrin hydrogels (4 Â 10 6 cells/mL) in combination with antacid capsules for 24 h to evaluate any cytotoxicity effects (n = 4 per group).

| Assessment of cell viability
Cell viability was assessed using the LIVE/DEAD ® viability/ cytotoxicity assay kit (Invitrogen, Ireland).Constructs were removed from culture and incubated in phenol free DMEM media containing 2 μM calcein AM and 4 μM ethidium homodimer-1 (EthD-1) (both from Cambridge Bioscience, UK) for 1 hour at 37 C. Following incubation, constructs were imaged with a Leica SP8 scanning confocal microscope at 515 and 615 nm channels and analyzed using Leica Application Suite X (LAS X) software.Cell viability was determined by counting calcein AM positive cells (live) and EthD-1 positive cells (dead) followed by calculation of the percentage (%) of viable cells relative to the total cell population.

| Disc explant isolation and culture
Skeletally mature bovine tails were obtained from a local abattoir.
Discs were isolated with a custom-made guillotine and two parallel blades separated by a distance of 4.5 mm.From each obtained 'disc slab', three to five 6 mm biopsies were taken from the AF-region.AFcylinders (6 mm) were cored using a 3-mm biopsy punch to generate an annular disc explant, which was inserted into a 3D printed cage to provide confinement and prevent swelling of the tissue.Briefly, cages were 3D printed with a Formlabs 2 printer and dental resin SG V1 (Formlabs, UK).After printing, cages were washed in isopropanol to remove excess resin and cured for 1 hour at 60 C under ultraviolet (UV) light.Subsequently, cages underwent air drying and UV light exposure for surface sterilization prior to use.Based on results acquired during in vitro experiments, the 500 mg/mL CaCO 3 group was found to be superior in terms of long-term buffering capacities and cell survival compared with all other experimental groups.Therefore, two groups were compared in an IVD explant model: primed AC without CaCO 3 microcapsules (ÀCaCO 3 , control) and primed AC with 500 mg/mL CaCO 3 microcapsules (+CaCO 3 ) (n = 5 for each group, 3 for biochemical analysis and 2 for viability and histological assessment).Briefly, both empty (ÀCaCO 3 , control) and +CaCO 3 (500 mg/ mL) containing alginate microcapsules were prepared; 25 μL of a fibrin (100 mg/mL)-HA (5 mg/mL) blend containing blank or +CaCO 3 (500 mg/mL) alginate microcapsules and cellular microcapsules were combined and pipetted into the 3 mm core of disc explants (Figure 1, "Phase 3").Constructs were cultured in low pH (pH 6.5), high osmolarity ($450 mOsm, adjusted using 150 mM sucrose), chemically defined medium containing lg-DMEM supplemented with 1 mg/mL Primocin (Invivogen, France), 5% FBS (GIBCO, Invitrogen, Dublin, Ireland), 100 KIU/mL aprotinin solution (Nordic Pharma, Sweden), 0.25 μg/mL AmpB, 1.5 mg/mL BSA, 1Â ITS, 40 μg/mL L-proline, 100 nM dexamethasone, 50 μg/mL L-ascorbic acid 2-phosphate, and 4.7 μg/mL linoleic acid (all Sigma-Aldrich, Ireland).The total culture period was 21 days at 37 C in physioxic (5% oxygen) conditions, with medium exchanges performed twice weekly.

| Quantitative biochemical analysis
Three samples of each group were separated (AF ring from hydrogelcore) and both parts digested separately using papain (125 μg/mL) in 0.1-M sodium acetate, 5 mM L-cysteine HCl, 0.05 M EDTA, and sodium citrate (55 mM) (Sigma-Aldrich, Ireland) at 60 C under constant agitation for 18 h followed by an additional incubation with 1-M sodium citrate under constant rotation for 1 h to disrupt the alginate-calcium crosslinks.DNA content of each sample was quantified using the Hoechst Bisbenzimide 33258 dye assay, with a calf thymus DNA standard and expressed in terms of the total amount of DNA (ng) per sample.Proteoglycan content was estimated by quantifying the amount of sGAG in constructs using the dimethylmethylene blue dye-binding assay (DMMB Blyscan, Biocolor Ltd., Northern Ireland, UK), with a chondroitin sulfate standard and expressed as total amount of sGAG measured (μg).
Total collagen content was determined for each sample (μg) by measuring the hydroxyproline content.Samples were hydrolyzed at 110 C for 18 h in concentrated HCl (38%) and assayed using a chloramine-T assay 37 and a hydroxyproline-to-collagen ratio of 1:7.69. 38

| Histological Analysis
Two samples of each group were fixed in 4% paraformaldehyde (PFA) overnight at 4 C, followed by repeated washings in PBS.Fixed samples were dehydrated in a graded series of ethanol (70% to 100%), embedded in paraffin wax, sectioned at 6 μm, and affixed to microscope slides.Sections were stained with 1% alcian blue 8GX in 0.1 M HCL (alcian blue/aldehyde fuchsin for alginate-based samples) to assess sGAG content and picrosirius red to assess collagen distribution (all from Sigma-Aldrich, Ireland).

| Statistical analysis
Statistical analyses were performed using GraphPad Prism (version 10.1.1)software with three to four samples analyzed for each experimental group.One-way ANOVA or two-way ANOVA was used for the analysis of variance where appropriate, with Tukey post-tests to compare between groups.The results are displayed as mean ± standard deviation (SD) and significance was accepted at a level of p < .05.

| Antacid microcapsules can increase the local pH within an acidic environment
Different antacids, namely Mg(OH) 2 , CaCO 3 , the zwitterion HEPES, and Al(OH) 3 at a concentration of 50 mg/mL were compared in terms of buffering capacities when exposed to a low pH environment (pH 6.5) (Figure 2).Results revealed no discernible differences in pH concentration maps between HEPES, Al(OH) 3 , and the negative control group (pH 6.5 without pH buffering).In contrast, both Mg(OH) 2 and CaCO 3 increased the pH in the center of the construct to 7.5 and 7.0, respectively (Figure 2A).Comparing pH as a function of distance along the x-axis of each construct, it is clear that the main effect of the buffering microcapsules occurs in the center with decreasing pH values toward the periphery (Figure 2B).Comparing the acidity change over time of each group, it is evident that Mg(OH) 2 and CaCO 3 have the highest pH neutralization capacities.However, Mg(OH) 2 was found to increase the pH level to a very basic level (pH 8.47 ± 0.58) within the first 8 h before plateauing toward neutral levels, which may be detrimental for cell viability (Figure 2C).CaCO 3 was capable of increasing the local pH to a neutral level for the duration of 72 h (Figure 2C).The desired pH values were obtained after equilibration and were maintained for up to 72 h. Figure 2C also shows the stability of the positive (pH 7.4) and negative (pH 6.5) control wells over the course of 72 h.

| Increasing antacid concentration increases local pH levels
Based on the previous results (testing different pH buffering agents), Mg(OH) 2 and CaCO 3 were taken forward to investigate the effect of salt concentration.The temporal effect of Mg(OH) 2 and CaCO 3 on the local environment within the construct and the surrounding media was evaluated for different initial concentrations (25, 50, 100, 250, and 500 mg/mL) (Figure 3).Mg(OH) 2 was observed to alter the pH of the acidic environment to a larger extent depending on the initial salt concentration (Figure 3A).At the highest concentration (500 mg/mL), the local pH of the construct was found to exceed the range of the measuring limits (pH >10) and the pH of the media plateaued at pH 7.9 after 24 h (Figure 3A,B).The local pH of the constructs showed an initial peak within the first 5 h after acidic exposure but decreased toward a neutral pH over time (Figure 3A  did not exert any significant effect on pH levels in either the construct or surrounding media (Figure 3C).

| Mg(OH) 2 and CaCO 3 microcapsule size and release kinetics
After investigating different concentrations of Mg(OH) 2 and CaCO 3 in terms of pH neutralization limits, the highest Mg(OH) 2 group (500 mg/mL) was excluded as it exceeded the upper pH limit, while the lowest Mg(OH) 2 concentration (25 mg/mL) did not exhibit a sustained neutralization effect after 24 h and was therefore omitted.The lower concentrations of CaCO 3 (25 and 50 mg/mL) were also excluded as they did not demonstrate a noticeable impact on pH neutralization.Further characterization of the concentrations 50, 100 and 250 mg/mL of Mg(OH) 2 and 100, 250 and 500 mg/mL of CaCO 3 in terms of microcapsule size and release kinetics was performed.As seen in Figure 4A, CaCO 3 produced larger microcapsules (303.1 homogeneous microcapsule size for all groups except the highest concentration of CaCO 3 , which showed a wider range of microcapsule sizes (Figure 4B).Investigating the release of antacids from the microcapsules into the media demonstrated a rapid drop of Mg 2+ ions inside the microcapsules within the first 24 h, with a concomitant increase in Mg 2+ concentration in the media (Figure 4C).With respect to Ca 2+ release from microcapsules into the surrounding media, a decrease of 12.3% ± 4.5% of Ca 2+ within the microcapsules was observed over the time period investigated (72 h) (Figure 4D).There was no notable distinction in the release kinetics rate for any of the investigated concentrations of Mg(OH) 2 and CaCO 3 .(herein termed primed, microencapsulated AC) and on the host cells (NP cells), respectively.Results showed that culturing primed AC microcapsules in the presence of Mg(OH) 2 had no significant effect on cell viability with values as high as 92.9% ± 1.8% (Figure 5A,B).

| Buffering the local pH using Mg(OH) 2 negatively affects NP cells
In contrast, for NP cells, a significant decrease was observed with increasing concentration of Mg(OH) 2 with considerably lower cell viability at concentrations of 250 mg/mL (19.8% ± 12.4%) and 100 mg/mL (31.1% ± 26.5%) compared with AC ( p < .0001)(Figure 5A,B).In contrast, when culturing cells with CaCO 3 loaded microcapsules, a high number (>83%) of viable cells (AC and NP) were observed (Figure 5C,D).Quantification of cell viability revealed levels greater than 80% in all groups with significantly higher cell viability of NP cells compared with AC in all groups (Figure 5D).A significant yet minor reduction in NP cell viability was observed with the highest (500 mg/mL) CaCO 3 concentration ( p < .01)(Figure 5D).

| CaCO 3 significantly increases the pH level within the core of a disc explant cultured under acidic conditions
Buffering capacities were further assessed in a disc explant study, whereby a ring explant from a bovine AF tissue was taken, cored using a 3-mm biopsy punch and filled with a hybrid hydrogel with (+CaCO 3 ) and without (ÀCaCO 3 ) buffering microcapsules at a concentration of 500 mg/mL.This explant was further cultured inside a 3D printed cage to prevent excessive tissue swelling (Figure 6A).After 21 days in culture under disc-like acidic conditions of pH 6.5, pH values within the core were measured using a fiberoptic pH microsensor (Figure 6B).Results demonstrated a significantly higher pH level within the core when using primed AC cells and +CaCO 3 microcapsules in a hybrid gel compared to a control gel without (ÀCaCO 3 ) microcapsules ( p < .0001)(Figure 6C).Within the core gel, DNA levels were reduced relative to day 0 for the ÀCaCO 3 group (p < .05),with no significant differences observed for gels containing buffering microcapsules (+CaCO 3 ) at day 21 (Figure 7A).DNA content appeared to be maintained in the explant tissue for both ±CaCO 3 groups (Figure 7B).Densely packed viable AC microcapsules were observed within the core of both ±CaCO 3 groups, indicated with green calcein staining (Figure 7C).Semi-quantitative analysis of cell viability within the core was found to be high for both groups, with higher viability (96.2% ± 1.7%) in the ÀCaCO 3 group relative to the +CaCO 3 (89.3%± 3.5%) (Figure 7 D).

| CaCO 3 buffering enhances core sGAG levels and curtails depletion within disc tissue
To evaluate the change in sGAG levels within the hybrid gel (core) and the disc tissue (ring), the two compartments were separated and analyzed individually (Figure 8).Gels containing buffering microcapsules (+CaCO 3 ) resulted in higher levels of sGAG at day 21 (Figure 8A).Similar findings were observed for the explant tissue, with CaCO 3 curtailing the loss of sGAG which was significantly lower in ÀCaCO 3 group compared to day 0 (Figure 8B).Deep purple-stained primed AC microcapsules were visible within the gel core for both ±CaCO 3 explant groups (Figure 8C).

| CaCO 3 buffering maintains collagen content in explant culture
In terms of total collagen, no significant differences were observed for explants cultured with (+) or without (À) CaCO 3 microcapsules for either the core (Figure 9A) or the ring (Figure 9B) domains.No appreciable differences in picrosirius red collagen staining were observed between the groups, which correlated with the biochemical quantification (Figure 9C).

| DISCUSSION
The acidic microenvironment of the IVD plays a critical role in the imbalance of matrix anabolism and catabolism, which is believed to be a primary reason for the degeneration of the NP and subsequent low back pain. 15,39The overall aim of this study was to develop and characterize pH buffering microcapsules to alter the low pH disc microenvironment and support host cells to either maintain or even rejuvenate disc tissue composition.Finally, these microcapsules were evaluated in vitro and ex vivo to verify their potential when subjected to native disc-like microenvironmental conditions.
During the first phase of this study, different salts and buffering agents were investigated in terms of their pH neutralization capacities, which showed a limited effect of HEPES and Al(OH) 3 compared with Mg(OH) 2 and CaCO 3 .HEPES is capable of maintaining the pH within a neutral range (pH 6.8-8.2),which is typically desired in standard cell culture media.A pH of 6.5 was used to represent degenerative disc conditions, which appeared beyond the range to be adequately affected by HEPES.The weak base Al(OH) 3 is mostly insoluble within a pH range of pH 6-pH 8 and thus resulted in poor neutralization capacities.In contrast, Mg(OH) 2 and CaCO 3 were found to raise the pH, even to an alkaline range in the case of Mg(OH) 2 .Mg(OH) 2 is a strong base, which dissociates into Mg 2+ (aq) and 2OH À (aq) under acidic conditions.The increase of OH À ions within the surrounding solution creates an imbalance between H 3 O + and OH À causing a shift toward an alkaline pH, which is accelerated due to the high diffusion coefficient of OH À (5.27 Â 10 À5 cm 2 /s, 40 ).CaCO 3 is a mostly insoluble weak base, which dissociates into Ca 2+ (aq) , The pH measurements within explants after 21 days of acidic culture.(A) Explants were biopsied from bovine discs to create annular rings, which were cultured in 3D printed cages.Scale bar = 5 mm (B) schematic of pH measurement procedure (C) pH values measured within the core of cultured explants with (+CaCO 3 ) and without (ÀCaCO 3 ) buffering microcapsules.**** indicates significant difference ( p < .0001).pH was measured in n = 5 replicates, with measurements taken in triplicate per sample using a fiber optic pH sensor.properties. 32th respect to release kinetics, a faster diffusion of Mg 2+ ions from the microcapsules into the media was demonstrated compared to Ca 2+ ions for all concentrations investigated.Both antacids are considered insoluble in water.However, as pH decreases, solubility increases with a 3-fold higher solubility of CaCO 3 at pH 6.0 compared with a neutral pH. 42However, Ca 2+ ions have a higher affinity toward alginate compared with Mg 2+ ions, retarding their rapid release from the alginate network while Mg 2+ is more readily released into the media.Mg 2+ ions were found to be fully released within 72 h, whereas 80% of Ca 2+ ions remained within the microcapsules with a predicted complete release after several weeks.While alginate relies on calcium crosslinking to form a stable hydrogel, it is unlikely that the additional Ca 2+ contributed by the CaCO 3 had any noticeable impact on the alginate stiffness as ionic crosslinking sites are well saturated due to the reaction with CaCl 2 as part of the crosslinking step.There are several alternative natural (fibrin, collagen, chitosan, and pectin) and synthetic (PLGA, PCL) biopolymers compatible with the electrohydrodynamic process 33 that could be used for CaCO 3 encapsulation, which provides ample opportunity for optimization of microcapsule size and release kinetics.
However, extracellular calcium concentrations can affect biological processes.Previous work has shown that levels above 20 mM can induce apoptosis in osteoclasts, 43 while levels of 0.5 mM can activate calcium signaling receptors in cartilage endplate cells, leading to suppressed matrix deposition. 44In this study, within a 72-h timeframe, approximately 0.08 mM of calcium was released into the media for the CaCO 3 (500 mg/mL) loaded microcapsules, which is also below the normal range of calcium serum levels (1.9-2.5 mM).This concentration falls below toxic levels and is therefore unlikely to induce apoptosis or other forms of cellular damage.However, whether accumulated levels of calcium within the disc itself will exceed a critical threshold and negatively impact cells warrants further investigation.The mild buffering behavior and the slower release kinetics of CaCO 3 could be considered major advantages compared to Mg(OH) 2 .due to poor neutralization capacities, the effect of Mg(OH) 2 and CaCO 3 on cell bioactivity was investigated.Examining cell viability of primed AC, no difference was observed between groups.A powerful resilience of primed AC when exposed to low pH conditions has previously been demonstrated in our laboratory, 22 which may be equally resilient when exposed to a basic environment created by Mg(OH) 2 .
NP cells, on the other hand exhibited significant cell death within 24 h when cultured with Mg(OH) 2 microcapsules which was amplified with increasing concentration.This was found to occur predominantly in the peripheral region around antacid microcapsules, suggesting that the basic pH around the microcapsule boundaries is detrimental for these cells.Extreme extracellular pH conditions have previously been reported to increase the expression of calcium-sensing receptors (CaSR) in some cells, 45 through binding of Ca 2+ and Mg 2+ ions, thereby activating intracellular pathways involved in apoptosis, cell proliferation, survival, and cell maturation. 46,47Increased activation of this receptor due to basic pH and increased Mg 2+ levels may overstimulate NP cells resulting in cell death.
In phase 3 of this study, the potential of a hybrid hydrogel containing CaCO 3 microcapsules and microcapsules containing primed AC within a fibrin-HA bulking gel was explored using a disc explant model.In the first instance, the neutralization capacities of CaCO 3 observed in vitro were validated, demonstrating significantly higher pH levels within the core gel of the explants.After separating the core gel from the disc explant ring for individual analysis, higher DNA levels were found in the gel containing CaCO 3 microcapsules emphasizing the protective effects of CaCO 3 against the acidic environment.DNA levels within the disc tissue ring of CaCO 3 group did not increase compared with day 0 levels, however, levels could be maintained over 21 days of low pH culture most likely due to the pH elevating properties of CaCO 3 .
Higher levels of sGAG were observed in gel cores containing +CaCO 3 buffering microcapsules demonstrating the positive effect of maintaining a neutral pH.9][50] Within the disc ring, diminished levels of sGAG after 21 days in simulated degenerative disc-like conditions (i.e., low pH, low glucose, and high osmolarity) were demonstrated for both groups, with the ÀCaCO 3 group being significantly lower than day 0 levels.sGAG is readily eluted from disc tissue, even swelling, and loss of proteoglycans (PGs) from the tissue when using sucrose, 51 which correlates with findings observed in this study.This concomitant depletion of PGs from the disc tissue may be masking any beneficial effect of the CaCO 3 microcapsules on resident disc cells.Nevertheless, higher levels of total sGAG were observed in the +CaCO 3 group, which was not significantly different to day 0 and may be due to higher pH levels as reported previously. 14,17,18Interestingly, this beneficial effect was not observed for collagen content in either the gel core or within the explant ring.
Explant models offer many of the inherent advantages of full organ culture models, such as the preservation of cells within their native tissue structure, cellular composition, and cell-cell arrangements.In this study, an important objective was to be able to segregate the core and annular regions to evaluate the impact of the pH buffering system in a simplified and high-throughput manner.However, it is important to acknowledge that the fibrin core utilized in the explant model differs significantly from native NP tissue, resulting in distinct diffusion kinetics that may affect the effectiveness of the antacid microcapsules.Furthermore, this study utilizes an AF explant model instead of an NP explant model.This alternative AF explant model was chosen due to the natural swelling properties and the loss of proteoglycans from NP tissue typically experienced in an in vitro system, which is driven by insufficient osmotic pressure compared with its native environment. 51An AF explant model therefore is more appropriate for this type of study.However, some microcapsules may have been lost due to the fibrin-HA hydrogel degradation, increasing the variability in the biochemical measurements.Future studies should consider employing full organ culture or animal models to represent the in vivo scenario more closely.This will further address the limitation of the lack of integrity of the full organ and biomechanical stimulation.The IVD naturally experiences complex biomechanical stimuli including compression, torsion, flexion, and shear in different intensities and frequencies, which have differing effects on cell fate. 52For instance, dynamic compression loading has been found to enhance nutrient transport, including growth factors such as TGF-β and FGF, which in turn affects NP cells. 53Moreover, cyclic mechanical loading can induce chondrogenic differentiation of BMSC and enhance ECM deposition. 54These aspects were not considered in this work but may provide important insights into the matrix repair capacities for NP tissue repair.

| CONCLUSION
Overall, this study developed and characterized pH buffering microcapsules that can neutralize the acidic microenvironment that typically exists in degenerative disc disease.Comparing five different pH neutralizing agents, CaCO 3 was found to be superior in terms of neutralization capacities, release kinetics, and positive cellular response.
Specifically, CaCO 3 elevated the acidic pH to neutral levels, which is estimated to be maintained for several weeks based on Ca 2+ release kinetics.Additionally, CaCO 3 was found to be cytocompatible for both AC and NP cell types cultured in low pH conditions in vitro.Using a disc explant model, it was demonstrated that CaCO 3 microcapsules were capable of increasing the local pH within the core of a hybrid gel containing microcapsules of CaCO 3 and cells.This study underscores the proof-of-concept that pH-neutralizing agents possess the potential to beneficially modify the challenging acidic microenvironment associated with degenerative disc disease.Furthermore, these findings may have implications for addressing the early stages of disc degeneration in conjunction with cell therapies to augment regeneration.
solid line).Also, the media pH increased over time and reached values between 6.8 and 7.5 after 3 days depending on the initial Mg(OH) 2 concentration (Figure3A).In contrast, changing the concentration of CaCO 3 did not exert such large variations on the surrounding pH compared with Mg(OH) 2 (Figure 3C,D).Similar to Mg(OH) 2 , the initial peak occurred within the first 5 h of low pH exposure (Figure 3C), but never exceeded levels above pH 8.The concentration of 500 mg/mL appeared to equilibrate at a neutral pH of 7 locally as well as within the culture media.The lowest concentration (25 mg/mL of CaCO 3 ) F I G U R E 2 Local pH mapping of 50 mg/mL Mg(OH) 2 , CaCO 3 , HEPES, and Al(OH) 3 microcapsules exposed to low pH media in comparison to controls with constant pH media (7.4 representing positive control and pH 6.5 representing negative control).(A) pH concentration maps at 24 h.(B) pH profile in the x direction after 24 h, and (C) change in local pH evaluated over 72 h.Data shown from one representative sample from a total of n = 3 samples evaluated.

F I G U R E 3
Buffering capacity of different concentrations (25-500 mg/mL) of Mg(OH) 2 and CaCO 3 .(A) Temporal effect of Mg(OH) 2 on pH level within the hydrogel construct (solid line) and culture media (dashed line).(B) pH concentration map showing local pH distribution for Mg(OH) 2 at 24 h.(C) Temporal effect of CaCO 3 on pH level within the hydrogel construct (solid line) and culture media (dashed line).(D) pH concentration map showing local pH distribution for CaCO 3 at 24 h.Data shown from one representative sample from a total of n = 3 samples evaluated from each group.± 48.2 μm), whereas Mg(OH) 2 exhibited significantly smaller sized microcapsules (181.4 ± 11.1 μm, p < .0001).Moreover, comparing size distribution of both antacids at different concentrations reveals a To evaluate the effect of local pH change in an acidic environment on cell viability, two different aspects were investigated; the impact pH-buffering microcapsules exert on the potential therapeutic-cells F I G U R E 4 Characterization of microcapsules containing different initial concentrations, 50, 100, and 250 mg/mL of Mg(OH) 2 and 100, 250, and 500 mg/mL of CaCO 3 .(A) Average diameter of Mg(OH) 2 and CaCO 3 microcapsules, **** indicates significant difference ( p < .0001).(B) Size distribution of Mg(OH) 2 (top row) and CaCO 3 microcapsules (bottom row) at different concentrations.(C) Relative amount of Mg 2+ release (%) over 72 h from microcapsules (solid line) into media (dashed line) for different initial concentrations.(D) Relative amount of Ca 2+ release (%) over 72 h from microcapsules (solid line) into media (dashed line) for different initial concentrations.n >350 capsules of each group were analyzed for their size distribution and >100 capsules per sample were analyzed in duplicates for their respective Mg 2+ or Ca 2+ release.

F I G U R E 5
Concentration effects of Mg(OH) 2 and CaCO 3 buffering microcapsules on AC (primed) and NP cells, respectively.(A) Live/Dead assessment of cells after 24 h exposure to media with a pH of 6.5 in the presence of Mg(OH) 2 buffering microcapsules.(B) Semi-quantitative image analysis of cell viability when cultured at pH 6.5 with Mg(OH) 2 loaded microcapsules at various concentrations (50, 100, and 250 mg/mL).(C) Live/Dead assessment of cells after 24 h exposure to media with a pH of 6.5 in the presence of CaCO 3 buffering microcapsules.(D) Semiquantitative image analysis of cell viability when cultured at pH 6.5 with CaCO 3 loaded microcapsules at various concentrations (100, 250, and 500 mg/mL).A sample size of n = 4 biological replicates per group were analyzed.**, ***, and **** indicate significant differences between groups ( p < .01,p < .001,and p < .0001,respectively).AC, articular chondrocyte; NP, nucleus pulposus.

3. 6 |
CaCO 3 buffering curtails DNA loss within the core while preserving DNA content within disc tissue

HCO 3 À
(aq) , and H 2 O.CO 2(aq) within an acidic solution.The diffusion coefficient of HCO 3 À is 4.4 times lower than that of OH À explaining the slower increase in pH using CaCO 3 compared with Mg(OH) 2 .This neutralization capacity is mostly dependent on the local acidity, and not significantly influenced by other microenvironmental factors of the disc such as CO 2 , O 2 , or glucose levels.However, cell culture media contains sodium bicarbonate, which is commonly used to maintain a neutral pH under 5% CO 2 conditions, which may increase the rate of neutralization while both neutralizers react with the free H + ions.The media used was supplemented with acids capable of maintaining a low pH of 6.5 for at least 72 h, suggesting that the bicarbonate from the media alone is not sufficient to increase the pH level.Hence, the buffering observed in this study is most likely attributed to the antacid used.Microcapsule size and associated release kinetics also affect neutralization capacities of antacids.Evaluating different concentrations F I G U R E 7 DNA content and cell viability of cultured bovine annular ring explants containing a gel core with encapsulated primed articular chondrocyte cells with (+) and without (À) CaCO 3 buffering microcapsules after 21 days.(A) Total DNA (ng) content within the gel core.(B) Total DNA content (ng) of the explant ring compared to day 0. * indicates significant difference ( p < .05).(C) Confocal images of viable (green) and dead (red) cells.Sectioning was performed in the transverse direction.(D) Semi-quantitative image analysis of cell viability within the core region at day 21.A sample size of n = 3 was analyzed per group. of Mg(OH) 2 and CaCO 3 entrapped in alginate microcapsules revealed significantly larger microcapsules using CaCO 3 with the largest diameter observed when fabricating 500 mg/mL microcapsules (358.6 ± 100.3 μm).There are two possible reasons for this; the CaCO 3 nanoparticles used were two to three times bigger than those for Mg(OH) 2 microcapsules, which can influence the viscosity of alginateantacid slurries.Second, CaCO 3 can induce partial crosslinking of alginate, 41 which also increases the viscosity of the solution.Both influence the rheologic properties and hence EHD spraying

F
I G U R E 8 sGAG accumulation of cultured bovine annular ring explants containing a gel core with encapsulated primed AC cells with (+) and without (À) CaCO 3 buffering microcapsules after 21 days.(A) Total sGAG (μg) within the gel core.(B) Total sGAG (μg) of the explant ring relative to day 0. * indicates significant difference (p < .05).(C) Alcian blue/aldehyde fuchsin staining indicating sGAG within core and ring regions.Sectioning was performed in the transverse direction.A sample size of n = 3 was analyzed per group.

F I G U R E 9
Collagen accumulation of cultured bovine annular ring explants containing a gel core with encapsulated primed AC cells with (+) and without (À) CaCO 3 buffering microcapsules after 21 days.(A) Total collagen (μg) within the gel core.(B) Total collagen (μg) of the explant ring relative to day 0. (C) Picrosirius red staining indicating collagen within core and ring regions.Sectioning was performed in the transverse direction.A sample size of n = 3 was analyzed per group.in high osmolarity conditions.Previous work by van Dijk et al. examined sucrose and polyethylene glycol to match osmotic pressure present in native disc tissue.They revealed increased osmolarity, tissue