Comparative Metabolomic and Transcriptomic Analysis of 2D and 3D Mesenchymal Stem Cell Cultures for Improved Therapeutic Applications

Background and Purpose: Mesenchymal stem cells (MSCs) are widely utilized in regenerative medicine due to their multipotency and immunomodulatory properties. Compared to conventional two-dimensional (2D) monolayer cultures, three-dimensional (3D) spheroid cultures better mimic the in vivo microenvironment, influencing the metabolic activity and therapeutic efficacy of MSCs. This study aims to evaluate how 2D and 3D culture conditions affect the behavior, proliferation, and functional properties of MSCs. Experimental Approach: Metabolomic and transcriptomic analyses were conducted on MSCs cultured under both 2D and 3D conditions. To assess metabolic differences between 2D and 3D cultured MSCs, polar metabolites were extracted and analyzed using ¹H-NMR spectroscopy. The data was processed with Chenomx and subjected to multivariate statistical analysis. For transcriptomic analysis, RNA sequencing was performed, followed by differential gene expression and genesetenrichmentanalysis.KeyResults:ThefindingsrevealthatMSCsin3Dspheroidsexhibitreducedproliferation,enhanced stemness, and distinct metabolic adaptations, including increased glycolysis and altered nutrient metabolism. Additionally, genes associated with ribosome biogenesis and cell cycle progression were downregulated in 3D MSCs. These changes promote a more quiescent state, favoring its applications on tissue repair and immune modulation. Conclusion and Implications: Understanding these metabolic adaptations offers valuable insights for optimizing culture conditions, improving MSC-based therapies, and identifying novel therapeutic targets and biomarkers. Comparative Metabolomic and T ranscriptomic Analysis of 2D and 3D Mesenchymal Stem Cell Cultures for Improved Therapeutic Applications Manju Shresthaa,*, Yun-Seo Kilb,c,*, Yunju Jod,*, Simmyung Yooke,f, Ki Hyun Kimf, Dongryeol Ryud,**, Joo-Won Namc,**, Jee-Heon Jeonga,** aDepartment of Precision Medicine, School of Medicine, Sungkyunkwan University, Suwon 16419, Republic of Korea bCollege of Pharmacy and Inje Institute of Pharmaceutical Sciences and Research, Inje University, Gimhae, Gyeongnam 50834, Republic of Korea cCollege of Pharmacy, Yeungnam University, Gyeongsan, Gyeongbuk 38541, South Korea 1 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. dDepartment of Biomedical Science and Engineering, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of Korea. eDepartment of Biopharmaceutical Convergence, Sungkyunkwan University, Suwon, Gyeonggi 16419, Repu- blic of Korea fSchool of Pharmacy, Sungkyunkwan University, Suwon, Gyeonggi 16419, Republic of Korea *These authors contributed equally to this work **Corresponding authors Jee-Heon Jeong, Ph.D. Department of Precision Medicine, School of Medicine, Sungkyunkwan University, Suwon, Gyeonggi 16419, Republic of Korea Email: [email protected], Tel: +82-31-299-6165 Joo-Won Nam, Ph.D. College of Pharmacy, Yeungnam University, Gyeongsan, Gyeongbuk 38541, Republic of Korea Email: [email protected], Tel: +82-53-810-2818 Dongryeol Ryu, Ph.D. Department of Biomedical Science and Engineering, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of Korea Email: [email protected], Tel: +82-62-715-5374

Background and Purpose: Mesenchymal stem cells (MSCs) are widely utilized in regenerative medicine due to their multipotency and immunomodulatory properties. Compared to conventional two-dimensional (2D) monolayer cultures, three-dimensional (3D) spheroid cultures better mimic thein vivo microenviron- ment, influencing the metabolic activity and therapeutic efficacy of MSCs. This study aims to evaluate how 2D and 3D culture conditions affect the behavior, proliferation, and functional properties of MSCs. Experimental Approach: Metabolomic and transcriptomic analyses were conducted on MSCs cultured under both 2D and 3D conditions. To assess metabolic differences between 2D and 3D cultured MSCs, polar metabolites were extracted and analyzed using¹H-NMR spectroscopy. The data was processed with Chenomx and subjected to multivariate statistical analysis. For transcriptomic analysis, RNA sequencing was performed, followed by differential gene expression and gene set enrichment analysis. Key Results: The findings reveal that MSCs in 3D spheroids exhibit reduced proliferation, enhanced stemness, and distinct metabolic adaptations, including increased glycolysis and altered nutrient metabolism. Additionally, genes associated with ribosome biogenesis and cell cycle progression were downregulated in 3D MSCs. These changes promote a more quiescent state, favoring its applications on tissue repair and immune modulation.

and Implications: Understanding these metabolic adaptations offers valuable insights for optimizing culture conditions, improving MSC-based therapies, and identifying novel therapeutic targets and biomarkers.

Mesenchymal stem cells, Metabolomics, Transcriptomics, 3D spheroids, 2D culture Abbreviations: 2 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. MSCs, mesenchymal stem cells; 2D, two-dimensional; 3D, three-dimensional; ECM, extracellular ma- trix; NMR, nuclear magnetic resonance; DMEM, Dulbecco’s Modified Eagle Medium; P/S, penicil- lin/streptomycin; AO, acridine orange; PI, propidium iodide; CCK, cell counting kit; DSS, 2,2-dimethyl- 2-silapentane-5-sulfonic acid; AQ, acquisition time; SW, spectral width; RG, receiver gain; number of scans; DS, dummy scans; OPLS-DA, orthogonal partial least squares-discriminant analysis; GSEA, gene set en- richment analysis;ΑςΑ, αςετατε· ΑΔΠ, αδενοσινε διπηοσπηατε· Αλα, αλανινε· ΑΜΠ, αδενοσινε μονοπηοσπηατε· Ασπ, ασπαρτατε· ΑΤΠ, αδενοσινε τριπηοσπηατε· ΒΑ, βενζοατε· ἣο, ςηολινε· ῝ρ, ςρεατινε· ῝ρΠ, ςρεατινε πηο- σπηατε· Γλς, γλυςοσε· Γλν, γλυταμινε· Γλυ, γλυταματε· Γλψ, γλψςινε· ΓΠ῝, γλψςεροπηοσπηοςηολινε· ΗψπΤαυ, ηψποταυρινε· Ιλε, ισολευςινε· ΛΑ, λαςτατε· Λευ, λευςινε· ΜΙ, μψο-ινοσιτολ· ΠΑ, προπιονατε· Π῝, πηοσπηοςηολι- νε· ΠΓ, προπψλενε γλψςολ· Πηε, πηενψλαλανινε· ΣΑ, συςςινατε· Ταυ, ταυρινε· Τηρ, τηρεονινε· Τψρ, τψροσινε· Υ ΔΠ-Γλς, υριδινε διπηοσπηατε γλυςοσε· Υ ΔΠ-ΓλςΝΑς, υριδινε διπηοσπηατε Ν-αςετψλ γλυςοσαμινε· ἅλ, vαλινε· βΑλα, β-αλανινε· GOBP, gene ontology biological process; GOMF, gene ontology molecular function; NES, normalized enrichment score; Trpm2, transient receptor potential cation channel subfamily m member 2; Kcnb1, potassium voltage-gated channel subfamily b member 1; Txnrd2, thioredoxin reductase 2; Nfe2l2, nuclear factor, erythroid 2 like 2; Prdx1-4, peroxiredoxin 1-4; GO, gene ontology; IL-10, Interleukin-10; IDO, indoleamine 2,3-dioxygenase; PEG2, prostaglandin E2; TGF- β, transforming growth factor beta Bullet point summary: • Culture environments have a high impact on MSC functionality. • 2D and 3D cultured MSCs exhibit distinct metabolic and gene expression profiles. • Metabolic reprogramming in 3D MSC spheroids enhances the therapeutic potential of MSCs.

Mesenchymal stem cells (MSCs) are multipotent stromal cells widely distributed throughout the body and can be isolated from various tissues, including bone marrow, adipose tissue, umbilical cord, placenta, dental pulp, tendon, skin, amniotic fluid, and peripheral blood [1, 2]. These plastic-adherent, fibroblast-like cells possess remarkable self-renewal capacity and the ability to differentiate into multiple lineages [3, 4]. These unique characteristics have made MSCs an ideal cell source and highly promising candidates for regenera- tive medicine and tissue engineering [5, 6]. MSCs secrete a diverse array of bioactive molecules, including cytokines, growth factors, extracellular vesicles, and metabolic products, which play crucial roles in im- munomodulation, tissue repair, remodeling, and homeostasis [7, 8]. Their immunosuppressive effects make them a potential therapeutic tool for treating autoimmune and inflammatory diseases [9, 10]. However, while the intrinsic properties of MSCs make them highly suitable for various therapeutic applications, their behav- ior and functionality are significantly influenced by their surrounding microenvironment [11, 12]. Different microenvironmental factors can have varying effects on MSCs proliferation, differentiation, metabolism, and therapeutic efficacy [13, 14]. Understanding the complex and dynamic relation between MSCs and their microenvironment, involving cell–cell interactions, soluble signaling molecules, extracellular matrix compo- nents, nutrient availability, and oxygen tension—is essential for optimizing MSC-based therapies [15, 16]. Metabolism has also emerged as a key regulator of stem cell properties, fate, function, and clinical effective- ness. Investigating the metabolic characteristics of MSCs can help identify optimal therapeutic strategies [17-19]. Metabolomics, the comprehensive study of small-molecule metabolites, can provide valuable insights into metabolic shifts that can reveal MSCs’ therapeutic potential [20, 21]. Traditionally, MSCs are typically grown and cultured as monolayers on flat two-dimensional (2D) surfaces [22]. While 2D culture systems are simple, cost-effective, and highly scalable, they fail to accurately mimic the in vivo behavior and have notable limitations [23]. Studies have shown that prolonged in vitro culture under 2D conditions can lead to the loss of MSCs intrinsic self-renewal properties, differentiation poten- tial, and paracrine activity [3, 24-26]. Moreover, 2D cultures do not replicate the dynamic cell–cell and cell-extracellular matrix (ECM) interactions that naturally occur in vivo , which reduces their translational relevance. In contrast, three-dimensional (3D) spheroid cultures enable MSCs to form clusters, facilitating enhanced cell–cell communication and ECM deposition. These interactions in 3D cultures enhance the se- cretion of growth factors, creating a microenvironment that closely mimics in vivo conditions. As a result, 3 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. 3D cultures better simulate the complex native tissue microenvironment while preserving or even enhancing MSC phenotypes [27, 28]. Consequently, 3D culture systems effectively regulate key cellular behaviors, in- cluding proliferation, differentiation, migration, and survival. The transition from 2D to 3D culture systems represents a significant advancement in MSC research, improving MSC-based therapeutics. MSC spheroid cultures have demonstrated numerous advantages over 2D monolayers, including better maintenance of stem- ness and quiescence, increased secretion of paracrine factors such as cytokines and growth factors, enhanced antioxidative, antiapoptotic, and anti-inflammatory properties, and reduced stress-induced senescence un- der prolonged culture conditions [29]. Preclinical studies have demonstrated the efficacy of MSC spheroids in various disease models, including transplantation, tissue engineering, and regenerative medicine [30-32]. More importantly, 3D culture systems preserve the core properties of MSCs, enhancing metabolic adaptabil- ity and strengthening resistance to environmental challenges [33, 34]. These metabolic adaptations improve cell viability and functionality, ultimately increasing MSCs paracrine activity and promoting a tissue-repair microenvironment through the secretion of bioactive metabolites [7, 35]. In this study, we aim to investigate the distinct metabolic profiles of MSCs cultured in 2D monolayers versus 3D spheroidal environments using nuclear magnetic resonance (NMR)-based metabolomic and transcriptomic profiling (Fig. 1). By identifying these metabolic differences, we seek to justify how metabolic reprogramming in 3D spheroids influences critical cellular behaviors, such as enhanced stemness, regulated proliferation, and increased therapeutic efficacy. This research provides valuable insights into the metabolic regulation of MSCs across different culture conditions and broadens the understanding of the optimization of culture conditions to improve MSC quality and functionality for clinical applications in disease treatment. Furthermore, these findings offer promising avenues for the development of targeted and tailored MSC-based therapies and the identification of metabolic biomarkers for therapeutic monitoring and personalized medicine. Fig. 1. Schematic illustration for the metabolomics and transcriptomics analysis of mesenchy- mal stem cells (MSCs) cultured under 2D and 3D conditions. (Created with BioRender.com). This integrative approach offers a comprehensive understanding of the metabolic reprogramming and their underlying molecular mechanisms that drive differences between 2D and 3D culture environments.

Isolation of MSCs All animal-related procedures comply with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. MSCs were isolated by harvesting the subcutaneous adipose tissues from 8–10 -week -old male C57BL/6 mice (Orient Bio, Seongnam, Gyeonggi-do, Republic of Korea) as describedpreviously [36]. In brief, the euthanized mice were disinfected by immersion in 70% ethanol. Subcutaneous adipose tissues were exposed and collected after excising the inguinal lymph nodes. The harvested tissues were minced and digestedwith 0.1% collagenase P solution (Sigma-Aldrich, MA, USA) at 37 °C for 30 min. The enzymatic reaction was neutralized using Dulbecco’s Modified Eagle Medium (DMEM, Bylabs, Hanam, Gyeonggi-do, Republic of Korea) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) and 1% 4 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. penicillin/streptomycin (P/S) 100X (Bylabs, Hanam, Gyeonggi-do, Republic of Korea), followed by cen- trifugation at 400 × g for 5 min. To remove residual fat cells, the cell suspension was filtered through a 40-μm cell strainer and cultured at 37 °C. The next day, unattached cells and debris were washed away using phosphate-buffered saline (PBS), and the culture was maintained until the cells reached 90% confluence. MSCs from passage 3 were used for the subsequent experiments. Fabrication of 3D MSC spheroids AggreWellTM 400 plates (STEMCELL Technologies, Cambridge, MA, USA) were pretreated with an anti- adherence rinsing solution (STEMCELL Technologies, Cambridge, MA, USA). A total of 1.2 x 10 6 MSCs in 2 ml of complete DMEM were added to each well and centrifuged at 300 x g for 3 min to facilitate cell capture within the microwells. The plate was then incubated at 37 °C with 5% CO 2 and 95% humidity for 5 h. After 5 h of incubation, spheroids were collected and transferred to non-adherent culture plates for an additional 24 h of incubation. Cell sample collection For the 2D culture, 1 x 10 6passage 3 MSCs were seeded onto 100 mm culture plates with complete DMEM containing 10% FBS and 1% P/S. The cells were maintained at 37 °C with 5% CO 2 and 95% humidity, with media replacement after 5 h, followed by further incubation for 24 h. For the 3D culture, spheroids were prepared in AggreWell TM 400 using MSCs of the same passage. Cell samples were collected at two time points: 0 h and 24 h. The 0 h time point corresponded to the complete adhesion of 2D MSCs to the plate and the formation of 3D spheroids. Viability assays Cell viability was evaluated by a live/dead staining test. For 2D MSCs, 0.05 M MSCs were cultured with 900 μL of complete DMEM in a 24-well plate. For 3D spheroids, 50 spheroids were collected in 900 μL of RPMI medium in a 1.5 mL E-tube. A 100 μL staining solution containing acridine orange (AO) (0.67 μM) and propidium iodide (PI) (75 μM) (Sigma-Aldrich, USA) was added to both samples, mixed, and incubated in the dark for 10 min at room temperature. Stained cells were then visualized using a fluorescence microscope (Eclipse Ti; Nikon Instruments Inc., Melville, NY, USA), where viable cells appeared green, and dead cells appeared red. In addition, cell viability was also assessed by colorimetric assay using the Cell Counting Kit-8 (CCK-8) (Dojindo Laboratories Co. Ltd, Japan). Briefly, fifty spheroids were first collected in 100 μL of DMEM in a 1.5 mL E-tube, followed by the addition of 10 μL of CCK-8 solution. The mixture was incubated at 37ºC for 1.5 h, after which the supernatant was collected, and absorbance was measured at 450 nm using a microplate reader. The intensity of formazan dye production, catalyzed by cellular dehydrogenases, was directly proportional to the number of viable cells. Quantification of DNA DNA quantification was performed using the PicoGreen dsDNA reagent (Invitrogen, Thermo Fisher Scienti- fic, MA, USA) following the manufacturer’s instructions. Briefly, a working solution was prepared by diluting the PicoGreen stock solution 200-fold in Tris-EDTA buffer. For sample preparation, single cells and sphero- ids were lysed using RIPA buffer with the aid of a probe sonicator, followed by centrifugation to collect the supernatant. A 10 μL aliquot of the supernatant was mixed with 140 μL of the PicoGreen working solution in a black 96-well plate and incubated in the dark for 5 min at RT. Fluorescence intensity was then measured using a microplate reader at an excitation wavelength of 480 nm and an emission wavelength of 520 nm. DNA concentrations were determined by comparing fluorescence values against a standard curve. Sample preparation for metabolomics Metabolite extraction was conducted as previously described [37], but with some modifications specifically for spheroids samples. Briefly, cells were first washed twice with PBS after 24 h of culture. To extract metabolites, 3 mL of cold methanol (Daejung Chemical and Metals, Republic of Korea) was added to scrape 5 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. adherent cells from the culture plates, and the detached cells were transferred to a tube. An additional 1 mL of methanol was used to ensure complete collection. For spheroid samples, 4 mL of methanol was directly added to the tube after washing with PBS. The cells were then lysed via ultrasonication for 30 s. Next, 4 mL of chloroform (Daejung Chemical and Metals, Republic of Korea) and 3.6 mL of distilled water were added to the lysed cells. The mixture was vortexed and then centrifuged at 5,000 rpm for 15 min at 4 °C. Following centrifugation, the upper aqueous layers were carefully collected and transferred into glass vials for speed vacuum drying at 2,000 rpm, maintaining a chamber temperature of 30 °C. Chemicals for NMR experiments, D 2O (99.9 atom% D), 2,2-dimethyl-2-silapentane-5-sulfonic acid (DSS) solution (1 wt% in D 2O, 99.9 atom% D) as an internal calibrant, and PBS tablets, were all purchased from Sigma-Aldrich. For sample preparation, 230 μL of D2O-prepared PBS containing 0.002 wt% DSS was added to each dried sample. The mixture was then subjected to ultrasonication for 10 min, with a short vortexing to facilitate dissolution. The samples were subsequently centrifuged at 13,500 rpm for 10 min at RT. A 200 μL aliquot of the resulting clear supernatant was transferred into a 3 mm NMR tube for NMR spectroscopy. Metabolomic data acquisition NMR analysis of the cell extracts was conducted using a Bruker AVANCE NEO spectrometer (1H, 600 MHz, Oxford magnet, Bruker Switzerland AG, Fallanden, Switzerland) at the Core Research Support Center for Natural Products and Medical Materials (CRCNM). The system was operated using Bruker TopSpin 4.1.3 software (Billerica, MA, USA). To suppress the water signal, a standard 1D NOESY sequence with presat- uration (Bruker pulprog: noesypr1d ) was applied, using a presaturation frequency of δ H 4.70 ppm. During data acquisition, the probe temperature was maintained at 298.0 K, along with the following parameter settings: calibrated 90 ° pulse (P1), relaxation delay (D1) of 4 s, acquisition time (AQ) of 4 s, spectral width (SW) of 20 ppm, receiver gain (RG) of 64, number of scans (NS) of 128, and number of dummy scans (DS) of 4. Post-acquisition processing was carried out using MestReNova software (Mestrelab Research SL, Santiago de Compostela, Spain). An optimized weighting function with a line broadening of -0.30 Hz and a Gaussian factor of 0.05 was applied, and a fifth-order polynomial fit was used for baseline correction [38]. Spectral referencing was performed using the methyl singlet of DSS at δ H 0.00 ppm. Metabolomic processing and multivariate statistical analysis Polar metabolite identification and spectral binning for quantification were performed using Chenomx NMR Suite 8.4 software (Edmonton, AB, Canada). First, the pre-processed 1H NMR spectra of the cell extract samples were imported, and metabolites in the samples were identified by comparison with the embedded library based on chemical shift values and peak patterns. Spectral binning was conducted with bin sizes ofδ H 0.001 ppm at spectral widths between δ H -0.001 and 9.000 ppm. A total of 8,660 bins were obtained, and bin area values were adjusted between samples by taking the area value of the DSS methyl group as 1. Negative bin area values, attributed to metabolite concentrations below the limit of quantification (resulting in high signal-to-noise ratios), were assigned a value of zero to prevent misinterpretation as negative metabolite content. To extract quantitative data for each of the identified metabolites, the relevant peaks were selected, and bin areas corresponding to their δ H ranges were determined (Table S1). Normalization with DNA quantity was also performed prior to statistical analysis. For the comparative analysis of MSC metabolites cultured in 2D monolayers vs. 3D spheroidal environments, statistical interpretation using an orthogonal partial least squares-discriminant analysis (OPLS-DA) model was applied on SIMCA 17.0.2 (Umetrics, Malmo, Sweden). The quality of the model with respect to model fitness and prediction ability was explained by R 2 and Q2parameters, respectively. A permutation test with 200 iterations was used to validate the OPLS-DA model. The heat map visualization was performed using R 4.4.2 and RStudio 2024.12.0 with the packages circlize (version 0.4.16), ComplexHeatmap (version 2.22.0), and normalize (version 0.1.0). Additionally, bar graphs of each quantified metabolite were generated using GraphPad Prism (GraphPad Software, Inc., USA). Transcriptome data analysis and Gene Set Enrichment Analysis (GSEA) 6 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. Transcriptome data analysis was performed using previously published protocols [39]. The process encom- passed RNA extraction, library preparation, and sequencing, with FASTQ files generated by Theragen Bio (South Korea). Data quality control was assessed using FastQC to confirm the integrity of the sequencing data. The reads were aligned to the mouse reference genome (GRCm38, release 102, October 2020) using STAR to ensure precise mapping. Transcript abundance was quantified using RSEM, and the data was normalized using a log2 transformation of transcripts per million (TPM) + 1. This normalization approach accounts for differences in sequencing depth and transcript length, enabling robust comparisons between samples. All data processing and visualization were performed using R and R Studio. Differential expression analysis was carried out using the DESeq2 package (version 1.42.0). Gene Set Enrichment Analysis (GSEA) and visualization were conducted using clusterProfiler (version 4.10.0), DOSE (version 3.28.2), msigdbr (version 7.5.1), and enrichplot (version 1.22.0) packages. Additional data processing and visualization tasks employed biomaRt (version 2.58.0), stringr (version 1.5.1), dplyr (version 1.1.4), scales (version 1.3.0), ggplotify (version 0.1.2), ggpubr (version 0.6.0), ggvenn (version 0.1.10), pheatmap (version 1.0.12), and RColorBrewer (version 1.1-3). Statistical analysis Statistical interpretation and graphical representations were performed using GraphPad Prism (version 8.4.2, GraphPad Software, Inc., USA). One-way ANOVA, followed by Tukey’s multiple comparisons test, was used to assess variations between groups. A p- value <0.05 was considered statistically significant.

Isolation and characterization of mouse adipose MSCs MSCs were successfully isolated from the subcutaneous adipose tissues of 8-week-old C57BL/6 mice and characterized throughdifferentiation assays and flow cytometric analysis for surface markers, following our previously established protocol [36]. The isolated MSCs exhibited a characteristic fibroblast-like morphology and demonstrated their ability to differentiate into multiple lineages. Specifically, osteogenic differentiation was evidenced by dark blue staining in MSCs, indicating the presence of alkaline phosphatase activity; adi- pogenic differentiation was indicated by reddish lipid vesicles stained with Oil Red O, appearing as red globules within the cells; and chondrogenic differentiation was confirmed by Alcian Blue-stained glycosaminoglycans. Additionally, these MSCs expressed high levels of MSC-specific markers, including CD90, CD44, CD29, and Sca-1, while showing minimal expression of CD11b and CD45, consistent with our previous findings [36]. Morphological and proliferative differences of MSCs in 2D and 3D cultures MSCs were cultured on adherent and non-adherent culture plates for creating 2D and 3D microenvironments. In 2D culture, MSCs exhibited a typical spindle-shaped fibroblast-like morphology, whereas in 3D spheroids, elongated MSCs became more compact with increased cell–cell contacts mediated by cadherin proteins [27], transitioning into a rounded shape (Fig. 2A). The spheroids measured 180 –220 μμ in diameter, with each containing approximately 1,000 cells. The three-dimensional structure of MSC spheroids resembled natural tissue organizationin vivo, providing a more physiologically relevant microenvironment [40, 41]. Cell viability analysis showed high viability in both culture conditions at 0 h and 24 h, confirmed by both live/dead assay using AO/PI staining and CCK-8 assay (Figs. 2B and 2C). Some dead cells were seen in 3D spheroids after 24 hours but were non-significant. Proliferation analysis revealed significantly higher growth rates in 2D cultures, whereas 3D culture showed no proliferation within 24 h (Fig. 2D). DNA quantification using PicoGreen confirmed a higher overall DNA content in 2D cultures, indicating greater cell density and proliferation over time. Collectively, these results demonstrate the distinct differences between 2D and 3D MSC cultures in terms of morphology, viability, and proliferation rate, emphasizing the impact of culture conditions on MSC behavior. 7 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. Fig. 2. Evaluation of mesenchymal stem cells (MSC) morphology, viability, and proliferation. (A) Representative optical images of MSCs in 2D and 3D culture conditions. (B) Live/dead assay showing cell viability in 2D and 3D cultures at 0 h and 24 h, with live cells marked in green and dead cells in red. (C) CCK-8 assay results indicating cell viability in both culture conditions over time (n = 3). (D) DNA quantification estimating MSCs proliferation based on DNA content at 0 h and 24 h (n = 3). Data in (C) and (D) are expressed as the mean ± SD and one-way ANOVA with Tukey’s multiple comparisons test was used for statistical differences ( ***p < 0.001, ****p < 0.0001 and ns = not significant ). Metabolomic analysis reveals metabolic reprogramming in 3D spheroids compared to 2D mo- nolayers Metabolomic analysis was performed to evaluate the metabolic activity of MSCs cultured under 2D and 3D conditions. NMR is as versatile as MS in metabolomics studies, enabling direct metabolite quantification with high reproducibility. This reproducibility allows reliable identification of content metabolites by means of external source libraries in NMR-based metabolomics studies. For the present NMR-based metabolomics study, 1H NMR spectra of the cell extracts were acquired using the 1D NOESY presat pulse sequence, which is considered ideal for metabolomic studies due to metabolite signal enhancement through water signal suppression [42]. As a result, significantly stronger signals were observed throughout the NMR data from the 2D MSC cultures compared to 3D cultures (Figs. 3, S1 and S2). For further comparative analysis, metabolite identification was performed using the external library, and metabolites with distinct differences in metabolic profiles between 2D and 3D MSC cultures were successfully identified. 8 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. Φι γ. 3. Ρεπρεσεντατιvε 1Η ΝΜΡ σπεςτρα οφ ςελλ εξτραςτς οβταινεδ φρομ ΜΣ῝ς ατ 24 η ςυλτυρεδ ιν 2Δ ανδ 3Δ ςονδιτιονς. Τηε 1Η ΝΜΡ δατα ωερε αςχυιρεδ ωιτη 1Δ ΝΟΕΣΨ πρεσατ πυλσε σεχυενςε (νοεσψπρ1δ, α πρεσατυρατιον φρεχυενςψ ατ δ Η 4.70 ππμ) ατ 600 ΜΗζ ιν Δ 2Ο-πρεπαρεδ ΠΒΣ containing 0.002% ΔΣΣ (Ι῝). Αββρεvιατιονς: ΑςΑ, αςετατε· ΑΔΠ, αδενοσινε διπηοσπηατε· Αλα, αλανινε· ΑΜΠ, αδενοσινε μονοπηοσπηατε· Ασπ, ασπαρτατε· ΑΤΠ, αδενοσινε τριπηοσπηατε· ΒΑ, βενζοατε· ἣο, ςηολινε· ῝ρ, ςρεατινε· ῝ρΠ, ςρεατινε πηοσπηατε· Γλς, γλυςοσε· Γλν, γλυταμινε· Γλυ, γλυταματε· Γλψ, γλψςινε· ΓΠ῝, γλψςεροπηοσπηοςηολινε· ΗψπΤαυ, ηψποταυρινε· Ιλε, ισολευςινε· ΛΑ, λαςτατε· Λευ, λευςινε· ΜΙ, μψο-ινοσιτολ· ΠΑ, προπιονατε· Π῝, πηοσπηοςηολινε· ΠΓ, προπψλενε γλψςολ· Πηε, πηενψλαλανινε· ΣΑ, συςςινατε· Ταυ, ταυρινε· Τηρ, τηρεονινε· Τψρ, τψροσινε· Υ ΔΠ-Γλς, υριδινε διπηοσπηατε γλυςοσε· Υ ΔΠ-ΓλςΝΑς, υριδινε διπηοσπηατε Ν-αςετψλ γλυςοσαμινε· ἅλ, vαλινε· βΑλα, β-αλανινε. Α σινγλετ σι γναλ οφ αν υνκνοων ςονταμιναντ ις μαρκεδ ωιτη αν αστερισκ (*). In an effort to derive as many quantitative values as possible for the identified metabolites, spectral binning was performed, yielding quantitative data for 27 metabolites (Table S1; Fig. S3). Multivariate statis- tical analysis is essential for interpreting the multi-variable data matrices that are inevitably present in metabolomics studies. To visualize the differences between 2D and 3D MSCs for the quantified metabolites, multivariate analysis of the OPLS-DA model was performed, resulting in clearly separated clusters with cumulative R2X, R2Y, and Q 2values of 0.901, 0.981, and 0.965, respectively (Fig. 4A). The loadings scatter plot provides a visual representation of the variables that contributed to the sample clustering in the scores scatter plot, with placements skewed in the same direction as a particular sample cluster indicating that the variables contributed to that clustering. In the present analysis, the majority of metabolites were biased towards the 2D MSC group in the loadings scatter plots (Fig. 4B), which is consistent with the visual observation of stronger signals in 2D than 3D MSCs throughout the NMR data (Fig. 3). The OPLS-DA model was validated through a permutation test with 200 iterations (Fig. 4C). Additionally, a heat map analysis was performed for an intuitive visualization of the metabolomic data matrix. The resulting heat map clearly demonstrated that the identified metabolites were quantitatively much more abundant in 2D than in 3D MSCs (Fig. 4D). 9 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. Fig. 4. Multivariate statistical analysis of targeted metabolites in MSCs cultured for 24 h under 2D and 3D conditions. (A) OPLS-DA score scatter plot and (B) corresponding loadings scatter plot. (C) Statistical validation of the OPLS-DA model using 200 permutation tests, with regression lines represented by dashed lines. (D) Heatmap visualization of metabolomic profiles. Transcriptomic analysis reveals enhanced viability and functionality of mesenchymal stem cells in 3D culture conditions compared to 2D cultures. To further explore the metabolomic differences, we sought to delineate the molecular signatures that dis- tinguish 2D and 3D cell culture system through transcriptomic profiling. RNA sequencing was performed to generate comprehensive gene expression datasets, followed by differential gene expression analysis using DESeq and gene set enrichment analysis (GSEA) to identify biologically significant gene sets. Differential expression analysis revealed substantial transcriptional divergence between 2D and 3D cultures, with 1,438 genes upregulated and 936 genes downregulated at 0 h, and 966 genes upregulated and 1,336 downregulated at 24 h in the 3D condition (Figs. 5A and 5B). Furthermore, a comparative analysis of shared gene sets revealed 379 commonly upregulated and 330 downregulated genes across both time points, indicating the presence of temporally conserved transcriptional programs modulated by the culture environment (Fig. 5B). GSEA identified distinct enrichment patterns between the culture systems, underscoring the functional di- vergence conferred by the 3D microenvironment. Key pathways upregulated under 3D conditions included cation channel activity (GOMF Cation Channel Activity, NES = 1.68, p -value = 1.21e-4), regulation of ion transmembrane transport (GOBP Regulation Of Ion Transmembrane Transport, NES = 1.40, p -value = 1.27e-3), G-protein coupled receptor activity (GOMF G Protein Coupled Receptor Activity, NES = 1.45, p -value = 1.34e-3), and cellular redox homeostasis (GOBP Cell Redox Homeostasis, NES = 1.48, p - value = 0.030), emphasizing the impact of the three-dimensional microenvironment on cellular signaling and metabolic pathways (Fig. 5C and 5D). To further elucidate these observations, heatmap analyses of gene clusters associated with signal transduction, metabolism, and redox regulation demonstrated a pronounced 10 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. upregulation of key genes in 3D cultures. Specifically, genes such as Trpm2 ,Kcnb1 , Txnrd2 , Nfe2l2, and Prdx1-4 , integral to membrane signaling, metabolic processes, and oxidative stress regulation, exhibited higher expression in 3D conditions (Fig. 5E). These findings highlight the superior ability of 3D culture systems to better recapitulate the physiological complexity of in vivo environments, thereby modulating gene expression profiles in a more biologically relevant manner. Taken together, this study underscores the enhanced physiological relevance of 3D culture systems compared to conventional 2D systems, offering a robust platform for investigating cellular functions and molecular mechanisms under conditions that closely approximate in vivo biology. Fig. 5. Transcriptomic analysis reveals enhanced viability and functionality of mesenchymal stem cells in 3D culture conditions compared to 2D cultures. (A) Volcano plot comparing the transcriptomic profiles of mesenchymal stem cells cultured under 2D and 3D conditions at two time points (0 h and 24 h). The x-axis represents the log 2 fold change in gene expression, where positive values indicate upregulated genes, and negative values indicate downregulated genes in 3D culture. The y-axis represents -log(p -value), highlighting statistically significant differentially expressed genes between the two culture 11 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. conditions. (B) Venn diagram illustrating the overlap of upregulated and downregulated genes in 3D culture conditions relative to 2D conditions at two time points. (C) Bubble plot showing the results of gene set enrichment analysis (GSEA), highlighting the gene sets enriched in 3D conditions. These include gene sets associated with membrane proteins, signal transduction, metabolism, and cellular redox homeostasis. Bubble size corresponds to the gene set size, and the color scale indicates the magnitude of statistical significance (-log(p -value)). Representative GO terms include GO:0001653 (peptide hormone processing), GO:0034703 (calcium channel complex), GO:0090087 (regulation of calcium ion transmembrane transport via high-voltage-gated calcium channels), GO:0051591 (response to cAMP), and GO:0045454 (cell redox homeostasis). (D) Enrichment plots for specific gene sets associated with membrane protein and signal transduction, metabolism, and cellular redox homeostasis. Each plot displays the enrichment score (ES) curve for a particular gene set, with the peak ES indicating the core genes driving the enrichment. (E) Heatmaps depicting the expression patterns of selected genes within the enriched gene sets. The color scale represents relative gene expression levels, highlighting the differential expression of key genes driving the enrichment of these pathways in 3D conditions.

MSC-based therapies have emerged as promising treatment strategies for a wide range of diseases, including inflammatory disorders, neurodegenerative diseases, and tissue regeneration [5, 43]. The therapeutic effects of MSCs are primarily mediated through the secretion of paracrine signaling molecules, such as cytokines, growth factors, immunomodulatory molecules, and extracellular vesicles, all of which are directly influenced by their metabolic state [44, 45]. Variations in culture conditions can significantly alter the metabolic activity of MSCs, thereby affecting their fate and functions [17, 46]. Numerous studies have explored the metabolic properties of MSCs, particularly in relation to their differentiation potential [47-49], and the advantages of 3D culture systems are well recognized [3, 29, 40]. However, few studies have comprehensively examined the overall metabolic differences between MSCs cultured in 2D and 3D environments. To address this gap, we conducted metabolic profiling of MSCs under both culture conditions using NMR-based metabolomic and transcriptomic analysis. 2D-cultured MSCs are commonly used to study cellular responses; however, this culture condition does not accurately recapitulate the physiological microenvironment of most cell types. To overcome this limitation, 3D culture systems have been developed to more closely mimic native cell–cell interactions. In our study, 3D spheroids exhibited a slight decrease in viability compared to 2D cultures after 24 h, though this difference was not statistically significant. Conversely, cell proliferation was significantly lower in 3D spheroids, suggesting that the 3D microenvironment induces a quiescent state. Metabolite analysis revealed a reduction in key metabolites associated with cellular metabolism in 3D spheroids, suggesting a metabolic shift as an adaptation to the 3D culture environment. In contrast, 2D MSCs exhibited higher metabolic activity, with significantly elevated levels of most of the metabolites at both 0 h and 24 h, indicating a more active metabolic state. Both culture conditions showed distinct changes in targeted metabolite concentrations over the 24-h period. Notably, 2D cultures exhibited significantly higher levels of adenine nucleotides (ATP/ADP/AMP), creatine (Cr), and creatine phosphate (CrP) at both time points, reflecting an elevated energy charge associated with their proliferative state [50, 51]. Conversely, 3D spheroids exhibited a lower yet more stable energy state, indicative of a tightly regulated metabolic environment. The higher availability of oxygen and nutrients in 2D cultures supports oxidative phosphorylation, while the rapid proliferation of cells can induce a metabolic shift toward anaerobic glycolysis despite oxygen availability, a phenomenon known as the “Warburg effect” [52]. In contrast, the hypoxic cores of 3D spheroids necessitate a greater dependence on glycolysis for ATP production, resulting in lower overall ATP levels [53, 54]. This metabolic shift toward glycolysis in 3D cultures is linked to enhanced stemness, as it promotes an energy- efficient state that supports cellular survival under hypoxic or nutrient-deprived conditions. However, 3D spheroids exhibit metabolic heterogeneity, with cells in the outer layers exposed to higher oxygen levels, thereby engaging in oxidative phosphorylation [52, 54]. Additionally, the controlled metabolic environment 12 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. of 3D spheroids may contribute to regulated proliferation and improved therapeutic potential by optimizing the balance between energy production and cellular function. Glycolytic MSCs have been shown to secrete higher levels of anti-inflammatory and regenerative factors, which enhance their immunomodulatory potential in wound healing, autoimmune diseases, and organ repair [27, 55, 56]. Similarly, the elevated levels of myo-inositol (MI), phosphocholine (PC), and glycerophospho- choline (GPC) in 2D cultures suggest an increased capacity for cell membrane synthesis and signaling, in- dicative of active proliferation and growth [57]. A significant rise in taurine (Tau) and hypotaurine (HypTau) after 24 h in both 2D and 3D MSC cultures suggests their role in oxidative stress regulation, cellular home- ostasis, and mitochondrial function [58, 59]. In 2D cultures, their upregulation may serve as a compensatory response to high reactive oxygen species levels generated by elevated metabolic activity, whereas in 3D cultures, it is likely associated with hypoxic adaptation and autophagy, which enhance MSC survival and function [60, 61]. The lower glucose levels and reduced accumulation of lactate (LA) and acetate (AcA) in 3D spheroids suggest tightly regulated glucose utilization, likely due to hypoxic adaptation, a quiescent metabolic state, and lower energy demand compared to 2D cultures [62, 63]. Additionally, several amino acids, including glutamine (Gln), alanine (Ala), aspartate (Asp), glutamate (Glu), glycine (Gly), isoleucine (Ile), leucine (Leu), phenylalanine (Phe), threonine (Thr), tyrosine (Tyr), and valine (Val), were elevated in 2D cultures, suggesting increased involvement in biosynthetic pathways such as protein synthesis and cell signaling [64, 65]. The elevated levels of uridine diphosphate sugars (UDP-Glc/UDP-GlcNAc) in 2D cultures suggest enhanced glycosylation activity, which supports cell growth and extracellular matrix remodeling [66]. Moreover, the increased levels of butyrate (BA) and propionate (PA) over time indicate active lipid metabolism and high metabolic turnover [67, 68]. The higher succinate (SA) levels in 2D cultures, followed by a significant decline over 24 h, reflect increased tricarboxylic acid (TCA) cycle activity, whereas lower levels in 3D cultures indicate reduced mitochondrial respiration and a more quiescent metabolic state [52, 69]. Additionally, RNA sequencing analysis revealed a downregulation of genes associated with ribosome biogen- esis and cell cycle progression in 3D spheroids compared to 2D monolayers, further supporting a transition toward a less proliferative state [70, 71]. These metabolic and transcriptomic variations carry significant im- plications for in vivo applications and therapeutic effectiveness. The diminished proliferation and metabolic activity observed in 3D spheroids indicate a quiescent state, which can be beneficial for sustained tissue regeneration and immune modulation [31, 72, 73]. Furthermore, the metabolic profile of 3D spheroids more closely mirrors that of in vivo tissues, establishing them as a physiologically relevant model for disease modeling research and therapeutic advancements [40, 74]. The lower metabolic activity of 3D MSC spheroids enhances the secretion of anti-inflammatory cytokines and immunomodulatory factors, including IL-10, IDO, PEG2, and TGF- β [27, 75]. Furthermore, 3D MSCs demonstrate superior immunosuppressive capabilities, such as enhanced suppression of lymphocyte activation and increased Treg cell expansion, as their energy resources are primarily allocated to maintaining immune modulation rather than promoting inflammation or cell proliferation [36, 76-78]. These properties underscore the potential of 3D MSC spheroids for treating autoimmune and chronic inflammatory diseases [43, 79].

This study integrates metabolomics and transcriptomics to compare mesenchymal stem cells (MSCs) cultured under 2D and 3D conditions, revealing key differences in metabolism and gene expression. The results highlight the significant influence of culture conditions on MSCs metabolic behavior and functionality, with 3D spheroids demonstrating superior therapeutic potential due to metabolic reprogramming. Future research should aim to validate these metabolic and transcriptomic differences in in vivo models to confirm their translational relevance and therapeutic applicability. Data Availability Statement All relevant data for this study are included within this paper and supplementary file. Other 13 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. additional data will be available upon request from the corresponding authors. Acknowledgment This research was supported by multiple grants from the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT (MSIT) (grant Nos. RS-2023-00272815, RS-2022-NR074857, RS- 2023-NR077276, and 2022R1A2C1009496) and the Ministry of Education (grant No. RS-2023-00240669). Additionally, it was supported by the Korean Fund for Regenerative Medicine (KFRM) grant funded by the MSIT and the Ministry of Health & Welfare (grant No. 23A0205L1), as well as the Korean ARPA-H Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare (grant No. RS-2024-00507183). This work was also supported by the Korea Basic Science Institute (National Research Facilities and Equipment Center) grant funded by the Ministry of Education (grant No. 2019R1A6C1010046). CRediT author statement Manju Shrestha: Writing – original draft, Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization. Yun-Seo Kil: Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Visualization, Validation, Writing – original draft. Yunju Jo: Data curation, Formal analysis, Investigation, Methodology, Vali- dation, Visualization, Writing – original draft, Simmyung Yook: Writing – review and editing. Ki Hyun Kim: Resources. Dongryeol Ryu: Funding acquisition, Supervision, Resources, Writing – review and editing. Joo-Won Nam: Funding acquisition, Supervision, Resources, Writing – review and editing. Jee-Heon Jeong: Conceptualization, Funding acqui- sition, Project administration, Visualization, Resources, Supervision, Writing – review and editing. Declaration of Interest Statement The authors declare that they have no competing interests. Ethics approval statement All animal-related procedures comply with the Reporting of In Vivo Experiments (ARRIVE) guidelines and were conducted as per the Institutional Animal Care and Use Committee (IACUC)/Ethics Com- mittee of Sungkyunkwan University (Suwon-si, Gyeonggi-do, Republic of Korea); with approval number: SKKUIACUC2022-12-48-2. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work the authors used ChatGPT to improve language and readability. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.

1. Pharoun, J., et al., Mesenchymal stem cells biological and biotechnological advances: Implications for clinical applications. European Journal of Pharmacology, 2024. 977 : p. 176719.2. Wu, X., et al., Mes- enchymal stromal cell therapies: immunomodulatory properties and clinical progress. Stem Cell Research & Therapy, 2020.11 (1): p. 345.3. Regmi, S. and J.-H. Jeong, Superiority of three-dimensional stem cell clusters over monolayer culture: An archetype to biological application. Macromolecular Research, 2016.24 (12): p. 1037-1046.4. Dominici, M., et al., Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy, 2006. 8 (4): p. 315-317.5. Margiana, R., et al., Clinical application of mesenchymal stem cell in regenerative medicine: a narrative review. Stem Cell Research & Therapy, 2022. 13 (1): p. 366.6. Dehghanifard, A., et al., The emerging role of mesenchymal stem cells in tissue engineering. Int J Hematol Oncol Stem Cell Res, 2013. 7 (1): p. 46-7.7. Han, Y., et al., The secretion profile of mesenchymal stem cells and potential applications 14 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. in treating human diseases. Signal Transduction and Targeted Therapy, 2022. 7 (1): p. 92.8. Gonz´ alez- Gonz´ alez, A., et al.,Mesenchymal stem cells secretome: The cornerstone of cell-free regenerative medicine. World J Stem Cells, 2020.12 (12): p. 1529-1552.9. Ghannam, S., et al., Immunosuppression by mesenchymal stem cells: mechanisms and clinical applications. Stem Cell Research & Therapy, 2010.1 (1): p. 2.10. Zhou, J. and Y. Shi, Mesenchymal stem/stromal cells (MSCs): origin, immune regulation, and clinical applications. Cellular & Molecular Immunology, 2023. 20 (6): p. 555-557.11. Atiya, H., et al., Mesenchymal Stem Cells in the Tumor Microenvironment. Adv Exp Med Biol, 2020. 1234 : p. 31-42.12. Xue, Z., Y. Liao, and Y. Li, Effects of microenvironment and biological behavior on the paracrine function of stem cells. Genes Dis, 2024. 11 (1): p. 135-147.13. Liu, J., et al., Mesenchymal stem cells and their microenvironment. Stem Cell Research & Therapy, 2022. 13 (1): p. 429.14. Almeida-Porada, G., A.J. Atala, and C.D. Porada, Therapeutic Mesenchymal Stromal Cells for Immunotherapy and for Gene and Drug Delivery. Mol Ther Methods Clin Dev, 2020. 16 : p. 204-224.15. Redondo, J., et al.,The Bone Marrow Microenvironment in Immune-Mediated Inflammatory Diseases: Implications for Mesenchymal Stromal Cell-Based Therapies. Stem Cells Transla- tional Medicine, 2023. 13 (3): p. 219-229.16. Xue, Z., Y. Liao, and Y. Li, Effects of microenvironment and biological behavior on the paracrine function of stem cells. Genes & Diseases, 2024. 11 (1): p. 135-147.17. Barilani, M., et al., Central metabolism of functionally heterogeneous mesenchymal stromal cells. Scientific Reports, 2019. 9 (1): p. 15420.18. Coller, H.A., The paradox of metabolism in quiescent stem cells. FEBS Letters, 2019. 593 (20): p. 2817-2839.19. Ross, E.A., et al., Nanotopography reveals metabolites that maintain the immunomodulatory phenotype of mesenchymal stromal cells. Nature Communications, 2023. 14 (1): p. 753.20. Gonzalez-Covarrubias, V., E. Mart´ ınez-Mart´ ınez, and L. Del Bosque-Plata,The Potential of Metabo- lomics in Biomedical Applications. Metabolites, 2022.12 (2).21. Clish, C.B., Metabolomics: an emerging but powerful tool for precision medicine. Cold Spring Harb Mol Case Stud, 2015. 1 (1): p. a000588.22. Kusuma, G.D., et al., Effect of 2D and 3D Culture Microenvironments on Mesenchymal Stem Cell-Derived Extra- cellular Vesicles Potencies. Front Cell Dev Biol, 2022. 10 : p. 819726.23. Jin, H., et al., Enhanced survival of human mesenchymal stem cells following co-delivery with glucagon-like peptide-1 analogue in fibrin gel. Journal of Pharmaceutical Investigation, 2015. 45 (2): p. 143-149.24. Frith, J.E., B. Thomson, and P.G. Ge- never, Dynamic three-dimensional culture methods enhance mesenchymal stem cell properties and increase therapeutic potential. Tissue Eng Part C Methods, 2010. 16 (4): p. 735-49.25. Alvites, R., et al., Mesenchymal Stem/Stromal Cells and Their Paracrine Activity-Immunomodulation Mechanisms and How to Influence the Therapeutic Potential. Pharmaceutics, 2022.14 (2).26. Vilar, A., et al., Substrate mechanical properties bias MSC paracrine activity and therapeutic potential. Acta Biomaterialia, 2023. 168 : p. 144-158.27. Kouroupis, D. and D. Correa, Increased Mesenchymal Stem Cell Functionalization in Three-Dimensional Manufactu- ring Settings for Enhanced Therapeutic Applications. Frontiers in Bioengineering and Biotechnology, 2021.9 .28. Yun, C., et al., Advantages of Using 3D Spheroid Culture Systems in Toxicological and Pharmacological Assessment for Osteogenesis Research. Int J Mol Sci, 2024. 25 (5).29. Jensen, C. and Y. Teng, Is It Time to Start Transitioning From 2D to 3D Cell Culture? Frontiers in Molecular Biosciences, 2020. 7 .30. Han, Y., et al., Mesenchymal Stem Cells for Regenerative Medicine. Cells, 2019. 8 (8).31. Kim, W., et al.,Therapeutic strategies of three-dimensional stem cell spheroids and organoids for tissue repair and regeneration. Bioactive Materials, 2023. 19 : p. 50-74.32. Petrenko, Y., E. Sykov´ a, and ˇS. Kubinov´ a,The therapeutic potential of three-dimensional multipotent mesenchymal stromal cell spheroids. Stem Cell Research & Therapy, 2017. 8 (1): p. 94.33. Urz` ı, O., et al., Three-Dimensional Cell Cultures: The Bridge between In Vitro and In Vivo Models. International Journal of Molecular Sciences, 2023.24 (15): p. 12046.34. Edmondson, R., et al.,Three- dimensional cell culture systems and their applications in drug discovery and cell-based biosensors. Assay Drug Dev Technol, 2014.12 (4): p. 207-18.35. Giacomini, C., et al., The critical role of apoptosis in mesenchy- mal stromal cell therapeutics and implications in homeostasis and normal tissue repair. Cellular & Molecular Immunology, 2023. 20 (6): p. 570-582.36. Nguyen, T.T., et al., Engineering of hybrid spheroids of mesen- chymal stem cells and drug depots for immunomodulating effect in islet xenotransplantation. Sci Adv, 2022. 8 (34): p. eabn8614.37. Nguyen, T.T., et al., Fabrication of stem cell heterospheroids with sustained-release chitosan and poly(lactic-co-glycolic acid) microspheres to guide cell fate toward chondrogenic differentiation. International Journal of Biological Macromolecules, 2024. 263 : p. 130356.38. Pauli, G.F., et al., Importance of Purity Evaluation and the Potential of Quantitative 1H NMR as a Purity Assay. Journal of Medicinal 15 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. Chemistry, 2014. 57 (22): p. 9220-9231.39. Jo, S., et al., Bioengineered cell-constructs using decellularized fish skin-based composite bioink for regenerating muscle tissue. Applied Physics Reviews, 2024. 11 (2).40. Yen, B.L., et al., Three-Dimensional Spheroid Culture of Human Mesenchymal Stem Cells: Offering Therapeutic Advantages and In Vitro Glimpses of the In Vivo State. Stem Cells Transl Med, 2023. 12 (5): p. 235-244.41. Smolinska, V., et al., Exploring the Three-Dimensional Frontier: Advancements in MSC Spheroids and Their Implications for Breast Cancer and Personalized Regenerative Therapies.Biomedicines, 2024.12 (1): p. 52.42. Kil, Y.S., et al., Combining NMR and MS to Describe Pyrrole-2-Carbaldehydes in Wheat Bran of Radiation. J Agric Food Chem, 2022. 70 (40): p. 13002-13014.43. Regmi, S., et al., Mesenchymal stem cell therapy for the treatment of inflammatory diseases: Challenges, opportunities, and future perspectives. European Jour- nal of Cell Biology, 2019. 98 (5): p. 151041.44. Ferreira, J.R., et al., Mesenchymal Stromal Cell Secretome: Influencing Therapeutic Potential by Cellular Pre-conditioning. Front Immunol, 2018. 9 : p. 2837.45. Zhang, X., et al., Immunoregulatory paracrine effect of mesenchymal stem cells and mechanism in the treatment of osteoarthritis. Frontiers in Cell and Developmental Biology, 2024. 12 .46. Yuan, X., T.M. Logan, and T. Ma,Metabolism in Human Mesenchymal Stromal Cells: A Missing Link Between hMSC Biomanufacturing and Therapy? Frontiers in Immunology, 2019.10 .47. Shum, L.C., et al., Energy Metabolism in Mesenchymal Stem Cells During Osteogenic Differentiation. Stem Cells Dev, 2016. 25 (2): p. 114-22.48. Bispo, D.S.C., et al.,NMR Metabolomics Assessment of Osteogenic Differentiation of Adipose-Tissue-Derived Mesenchymal Stem Cells. J Proteome Res, 2022. 21 (3): p. 654-670.49. Bispo, D.S.C., et al., An Intracellular Metabolic Signature as a Potential Donor-Independent Marker of the Osteogenic Differentiation of Adipose Tissue Me- senchymal Stem Cells. Cells, 2022. 11 (23): p. 3745.50. Deldicque, L., et al., Creatine enhances differentiation of myogenic C2C12 cells by activating both p38 and Akt/PKB pathways. American Journal of Physiology-Cell Physiology, 2007. 293 (4): p. C1263-C1271.51. Papalazarou, V., et al., The creatine-phosphagen system is mechanoresponsive in pancreatic adenocarcinoma and fuels invasion and metastasis. Nat Metab, 2020. 2 (1): p. 62-80.52. Rybkowska, P., et al., The Metabolic Changes between Monolayer (2D) and Three-Dimensional (3D) Culture Conditions in Human Mesenchymal Stem/Stromal Cells Derived from Adipose Tissue. Cells, 2023.12 (1).53. Kapa lczy´ nska, M., et al.,2D and 3D cell cultures - a comparison of different types of cancer cell cultures. Arch Med Sci, 2018. 14 (4): p. 910-919.54. Tse, H.M., et al., The Importance of Proper Oxygenation in 3D Culture. Front Bioeng Biotechnol, 2021. 9 : p. 634403.55. Li, X., et al., Glycolytic repro- gramming in macrophages and MSCs during inflammation. Frontiers in Immunology, 2023. 14 .56. Zhou, C. and X.-Y. Bai, Strategies for the induction of anti-inflammatory mesenchymal stem cells and their application in the treatment of immune-related nephropathy. Frontiers in Medicine, 2022. 9 .57. Smucny, J., C.S. Carter, and R.J. Maddock, Greater Choline-Containing Compounds and Myo-inositol in Treatment-Resistant Versus Responsive Schizophrenia: A (1)H-Magnetic Resonance Spectroscopy Meta-analysis. Biol Psychiatry Cogn Neurosci Neuroimaging, 2024. 9 (2): p. 137-145.58. Surai, P.F., K. Earle-Payne, and M.T. Kidd, Taurine as a Natural Antioxidant: From Direct Antioxidant Effects to Protective Action in Various Toxicological Models. Antioxidants (Basel), 2021. 10 (12).59. Hou, X., et al., Taurine Attenuates the Hypotaurine-Induced Progression of CRC via ERK/RSK Signaling. Front Cell Dev Biol, 2021. 9 : p. 631163.60. Homma, K., et al., Taurine rescues mitochondria-related metabolic impairments in the patient-derived induced pluripotent stem cells and epithelial-mesenchymal transition in the retinal pigment epithelium. Redox Biol, 2021. 41 : p. 101921.61. Lou, J., et al., ῝ψτοπροτεςτιvε Εφφεςτ οφ Ταυρινε αγαινστ Ηψδρογεν Περοξιδε-Ινδυςεδ Οξιδατιvε Στρεσς ιν ΥΜΡ-106 ἓλλς τηρουγη τηε Ωντ/β-ἃτενιν Σι γναλινγ Πατηωαψ. Biomol Ther (Seoul), 2018. 26 (6): p. 584-590.62. Deorosan, B. and E.A. Nauman, The role of glucose, serum, and three-dimensional cell culture on the metabolism of bone marrow-derived mesenchymal stem cells. Stem Cells Int, 2011. 2011 : p. 429187.63. Fuentes, P., et al., Dynamic Culture of Mesenchymal Stromal/Stem Cell Spheroids and Secretion of Paracrine Factors. Front Bioeng Biotechnol, 2022. 10 : p. 916229.64. Higuera, G.A., et al., Patterns of amino acid metabolism by proliferating human mesenchymal stem cells. Tissue Eng Part A, 2012. 18 (5-6): p. 654-64.65. Li, X., O. Jiang, and S. Wang, Molecular mechanisms of cellular metabolic homeostasis in stem cells. International Journal of Oral Science, 2023. 15 (1): p. 52.66. Sheikh, M.A., B.S. Emerald, and S.A. Ansari, Stem cell fate determination through protein O-GlcNAcylation. J Biol Chem, 2021. 296 : p. 100035.67. Sch ¨onfeld, P. and L. Wojtczak, Short- and medium-chain fatty acids in energy metabolism: the cellular perspective. J Lipid Res, 2016. 57 (6): p. 943-54.68. Mansuy-Aubert, V. and Y. Ravussin, Short 16 Posted on 12 Apr 2025 — The copyright holder is the author/funder. All rights reserved. No reuse without permission. — https://doi.org/10.22541/au.174446284.47691145/v1 — This is a preprint and has not been peer-reviewed. Data may be preliminary. chain fatty acids: the messengers from down below. Frontiers in Neuroscience, 2023. 17 .69. Zhang, W. and R. Lang,Succinate metabolism: a promising therapeutic target for inflammation, ischemia/reperfusion inju- ry and cancer. Front Cell Dev Biol, 2023. 11 : p. 1266973.70. Bijonowski, B.M., et al., Aggregation-induced integrated stress response rejuvenates culture-expanded human mesenchymal stem cells. Biotechnology and Bioengineering, 2020. 117 (10): p. 3136-3149.71. Yuan, X., et al., Engineering extracellular vesicles by three- dimensional dynamic culture of human mesenchymal stem cells. Journal of Extracellular Vesicles, 2022. 11 (6): p. e12235.72. Liu, K.-C., et al., Scaffold-free 3D culture systems for stem cell-based tissue regenerati- on. APL Bioengineering, 2024. 8 (4).73. Ohori-Morita, Y., et al., Novel Mesenchymal Stem Cell Spheroids with Enhanced Stem Cell Characteristics and Bone Regeneration Ability. Stem Cells Translational Medici- ne, 2022. 11 (4): p. 434-449.74. Bidarra, S.J., et al., A 3D in vitro model to explore the inter-conversion between epithelial and mesenchymal states during EMT and its reversion. Scientific Reports, 2016. 6 (1): p. 27072.75. Kamprom, W., et al., Enhanced potent immunosuppression of intracellular adipose tissue-derived stem cell extract by priming with three-dimensional spheroid formation. Scientific Reports, 2024. 14 (1): p. 9084.76. Chen, L.C., H.W. Wang, and C.C. Huang, Modulation of Inherent Niches in 3D Multicellular MSC Spheroids Reconfigures Metabolism and Enhances Therapeutic Potential. Cells, 2021. 10 (10).77. Silva- Carvalho, A., et al., Regulatory T-Cell Enhancement, Expression of Adhesion Molecules, and Production of Anti-Inflammatory Factors Are Differentially Modulated by Spheroid-Cultured Mesenchymal Stem Cells. Int J Mol Sci, 2022.23 (22).78. Sarsenova, M., et al., Recent advances to enhance the immunomodulatory poten- tial of mesenchymal stem cells. Front Immunol, 2022. 13 : p. 1010399.79. Li, P., et al., Immunomodulatory properties of mesenchymal stem cells/dental stem cells and their therapeutic applications. Cell Mol Immunol, 2023.20 (6): p. 558-569. 17