Mesenchymal stem cells with an enhanced antioxidant capacity integrate as smooth muscle cells in a model of diabetic detrusor underactivity

Dear Editor,

Diabetic cystopathy, particularly when it progresses to detrusor underactivity (DUA), poses significant clinical management challenges and affects a substantial number of individuals with diabetes mellitus (DM).¹ Despite its prevalence, the etiology of diabetic DUA is poorly understood, and effective treatments are lacking. Our study addressed these gaps by investigating the mechanisms, tumorigenic risks, and optimal protocols of mesenchymal stem cell (MSC)² transplantation in a preclinical model of diabetic DUA. Molecular signature of the transplanted cells in the pathological micro-environments was characterised by single-cell transcriptome analysis,³ emphasising the importance of the hepatocyte growth factor (HGF)–mesenchymal-epithelial transition factor (MET) pathway and PD-L1 in the mechanism for muscle regeneration and immunomodulation.

We reported the first clinical study of multipotent-MSCs (M-MSCs) derived from human embryonic stem cells (hESCs) for treating Hunner-type interstitial cystitis, characterised by defective urothelium integrity and chronic inflammation.⁴ The hESC-derived M-MSCs were effective in a streptozotocin (STZ)-induced diabetic DUA (STZ-DUA) rat model.⁵ Transcriptomes of these preclinical samples were analysed to gain molecular insight into the pathogenesis of DM-associated DUA and the mechanism of the MSC therapy (Figure 1A). Transcriptomes of STZ-DUA bladders were distinct from those of sham-operated bladders and also from those of STZ-DUA rats administered M-MSCs (Figure 1B), with 525 and 112 differentially expressed genes in the STZ-DUA group relative to the sham and M-MSC groups, respectively (Figure S1A–C).

Gene networks/pathways analysis by MetaCore indicated altered expression of genes involved in oxidative-stress, inflammatory, and immune responses in the STZ-DUA group (Figures 1C and S1D–F). Gene-set enrichment analysis (GSEA) supported the significance of glutathione (GSH) metabolism and inflammatory responses in the pathogenesis of DM-associated DUA, with gene-sets related to muscle contraction and cardiomyopathy being downregulated in the STZ-DUA group (Figure 1D and Table S1). Four gene clusters were observed in transcriptome changes following the M-MSC therapy (Figure 1E,F). Cluster-3 (131) genes were upregulated in diabetic DUA, but their expression was normalised by the M-MSC therapy. The cluster-3 genes predominantly involved in GSH-related metabolic processes (Figures 1G and S1G–I).

For biomarkers from gene-network (MetaCore) and leading-edge (GSEA) analyses by comparing STZ-DUA with M-MSC groups, the M-MSC therapy effectively coordinated the regulation of genes associated with GSH synthesis and metabolism (Gclc and Gpx2), activation of NADPH oxidase (Noxa1, Noxo1, and Nox3), nitric oxide synthesis (Nos2), and immune responses (Figure S2A–D and Table S2). The alternation in these biomarkers was validated by quantitative-PCR (Figures 1H and S3) and immunofluorescence-staining (Figures 1I and S4A–E) assays. Consistently, the levels of carbonylated proteins, a validated biomarker of oxidative-stress were elevated in diabetic DUA (Figure S4F). The M-MSC therapy alleviates these oxidative-injuries in diabetic DUA. In vivo significance of these findings was validated by the beneficial outcomes of a GSH precursor/antioxidant N-acetylcysteine⁶ alone and combination of the sub-optimal dosage of M-MSCs in the STZ-DUA rat model (Figures 2A and S5–7). Collectively, these results provide in vivo proof of concept for the significance of oxidative-injury in the pathogenesis of diabetic DUA and the mode of action of the MSC therapy.

Accordingly, we hypothesised that adult-tissue derived MSCs with enhanced antioxidant capacity would ameliorate the pathological micro-environment, have a high in vivo engraftment capacity, and show superior therapeutic efficacy.^{7, 8} Importantly, they are safer than hESC-derivatives, which could provide advantages in clinical studies. Therefore, we investigated the benefits of human umbilical-cord derived MSCs (hUC-MSCs) using Primed/Fresh/OCT4 (PFO) procedure for treating diabetic DUA. As previously reported,^{9, 10} PFO-MSCs, characterised by small size and high GSH dynamics, exhibited the reduced level of reactive oxygen species and cell death by oxidative-stress (Figure S8).

Compared with naïve-cultured hUC-MSCs, animals injected with PFO-MSCs demonstrated significant enhancements of bladder function parameters (Figure 2B,C), restoring histological injuries (Figures 2D,E, and S9A), and the alterations in expression of GSH-related proteins in diabetic DUA (Figure 2F), validating their improved therapeutic efficacy. All these beneficial effects were sustained for 2 or 4 weeks after a single transplantation of PFO-MSCs (Figure S10), proving their long-lasting therapeutic effects on diabetic DUA. Longitudinal µ-PET/MRI bio-imaging analysis over a 9-month period revealed little tumorigenic potential of PFO-MSCs following injection (Figure S9B).

PFO-MSCs exhibited superior in vivo engraftment capacity and retention kinetics (Figures 2G,H and S11A). These findings were confirmed by immunofluorescence-staining of human  β2-microglobin (hB2M), with more hB2M⁺ cells detected after PFO-MSC transplantation (Figure 2I and S11B). The engrafted hB2M⁺ PFO-MSCs integrated as NG2 expressing pericytes around muscle fibres (Figures S11C and S12A) and directly differentiating into myocytes (Figure 3A), actively participating in muscle repair. The hB2M⁺/α-SMA⁺ cells persisted for 2 or 4 weeks following PFO-MSC transplantation (Figure S12B,C).

Single-cell transcriptome profiling revealed the molecular characteristics of engrafted PFO-MSCs within the pathological micro-environment (Figure S13A,B). The engrafted PFO-MSCs had distinct molecular characteristics from cultured cells (Figure 3B–D), with alterations in genes related to muscle progenitor cells, HGF-signalling, cell-adhesion, and immune responses (Figure S13C), with distinct nine gene clusters (Figure 3E,F). Cluster-1 was downregulated in engrafted cells by enriching apoptosis and cell division genes (Figure S13D,E). Cluster-5, up-regulated in engrafted cells, showed an enrichment of genes associated with skeletal muscle development and immunomodulation (Figures 3E–H and S13F), elucidating the mode of action of the PFO-MSC therapy.

Gene-network and leading-edge biomarker analyses identified HGF–MET and PD-L1 as key genes representing muscle regeneration and immunomodulatory processes, respectively (Figure 4A,B, S14, Table S3, and S4). In immunostaining results, the hB2M⁺/MET⁺ cells were mainly located within muscle bundles (Figures 4C and S15A) and hB2M⁺ cells, situated within or nearby muscle bundles, robustly expressed RHOA (Figures 4D and S15B) and PAX7 or MYOD1 muscle markers (Figures S15C,D), indicating the direct roles of the engrafted cells in muscle regeneration. Furthermore, hB2M⁺/PD-L1⁺ engrafted cells were found in various bladder locations, including the stroma around blood vessels and within muscle bundles, supporting their immunosuppressive capacity (Figure 4E,F).

In summary, comprehensive understanding of the key pathological mechanisms of diabetic DUA can guide the selection of optimal stem cells for therapeutic efficacy and safety. Our study demonstrates that hUC-derived PFO-MSCs, with enhanced GSH dynamics and engraftment capacity, effectively alleviate tissue-injury and contribute to muscle regeneration and immunomodulation in diabetic DUA, providing a promising approach for clinical translation of stem cell therapies (Figure S16). The significance and limitations of this study are discussed in detail in the Supporting information.

Chae-Min Ryu, YongHwan Kim, and Jung-Hyun Shin contributed equally to this work. Conceptualisation: Dong-Myung Shin, S.H.K., Juhyun Park, and Myung-Soo Choo. Methodology: Dong-Myung Shin, Chae-Min Ryu, Yong Hwan Kim, and Jung-Hyun Shin. Investigation: Chae-Min Ryu, Yong Hwan Kim, Jung-Hyun Shin, Seungun Lee, Hyein Ju, Yun Ji Nam, Hyungu Kwon, Min-Young Jo, Jinah Lee, Hyun Jun Im, and Min Gi Jang. Writing—original draft: Dong-Myung Shin, Chae-Min Ryu, Yong Hwan Kim, and Jung-Hyun Shin. Writing—review & editing: Dong-Myung Shin, Juhyun Park, Chae-Min Ryu, Yong Hwan Kim, Jung-Hyun Shin, Seong Who Kim, and Sang Hoon Song. Funding Acquisition: Dong-Myung Shin, Chae-Min Ryu, Jung-Hyun Shin, and Seong Who Kim. Resources: Ki-Sung Hong and Hyung-Min Chung. Data curation: Dong-Myung Shin, Chae-Min Ryu, Yong Hwan Kim, Seong Who Kim, and Jung-Hyun Shin. Supervision: Dong-Myung Shin, Juhyun Park, and Seong Who Kim.

D-.M.S. cofounded Cell2in, a company focused on developing FreSHtracer-based GRC assays. The other authors declare that no conflicts of interest exist.

All activities were conducted in compliance with the guidelines of the Ethics Committee on the Use of Human Subjects at Asan Medical Center (IRB#: 2015-0303). Approval for animal experiments was granted by the Institutional Animal Care and Use Committee of the University of Ulsan College of Medicine (IACUC-2020-12-160). All procedures adhered to the applicable regulations and guidelines.