Abstract

   Mesenchymal’ stem cells (MSC) are widely used for transplantation to
   treat various diseases due to their strong immune regulatory and tissue
   repair abilities. Magnetic nanoparticles (MNPs) can be used to track
   transplanted MSC. However, the potential impact of the MNPs we
   developed on MSC function remains unclear. In this study, we treated
   MSC with poly-L-lysine (PLL)-modified MNPs (MSC-MNPs) at a
   concentration of 0.1 µg/µL to assess their effects on MSC. The results
   showed that there were no significant effects on cell morphology,
   differentiation potential, proliferation, apoptosis, or the cell cycle
   after MSC were treated with MNPs. Interestingly, further experiments
   revealed that MNPs significantly enhanced the migratory capacity of
   MSC. Bio-Plex analysis revealed that MNPs promoted the expression of
   anti-inflammatory cytokines while inhibiting the secretion of
   pro-inflammatory factors. Flow cytometry also detected a significant
   increase in the number of MSC subpopulations, including CD184⁺MSC,
   CD106⁺MSC, and CD55⁺MSC, after MNPs labeling, which was further
   supported by proteomic analysis. Moreover, MSC-MNPs promoted the
   phenotypic transition of reactive astrocytes from A1 to A2 after
   coculture with activated astrocytes. In conclusion, MNPs have no
   cytotoxic effects on MSC labeling and significantly enhance their
   anti-inflammatory functions, offering new possibilities for the
   clinical application of MSC.

Supplementary Information

   The online version contains supplementary material available at
   10.1038/s41598-025-13083-3.

   Keywords: Mesenchymal stem cells, Magnetic nanoparticles, Migration,
   Cytokines, Anti-inflammatory

   Subject terms: Immunology, Neuroscience, Stem cells, Nanoscience and
   technology

Introduction

   Personalized therapies utilizing allogeneic or autologous cells are
   becoming increasingly common in the medical domain^[32]1–[33]3.
   Notably, mesenchymal stem cells (MSC) have been extensively employed in
   the treatment of various diseases because of their broad availability,
   ease of procurement, low immunogenicity, robust proliferation capacity,
   and roles in immune regulation, anti-inflammation and tissue
   repair^[34]4,[35]5. These unique properties make them highly promising
   candidates for the treatment of autoimmune diseases and central nervous
   system injuries^[36]6. Nevertheless, the clinical application of cell
   transplantation therapies is challenging: At present, it is difficult
   to noninvasively monitor the migration, homing, implantation efficiency
   and functional ability of transplanted cells in vivo. In recent years,
   magnetic nanomaterials (MNPs) have been increasingly used in the
   biomedical field. MNPs are renowned for their magnetism,
   biocompatibility, and biodegradability, and have a variety of potential
   applications, such as drug delivery, specific cell labeling, and
   magnetic resonance imaging (MRI), especially for cell tracking via
   MRI^[37]7,[38]8. The distinctive physicochemical properties of MNPs
   confer considerable advantages in medical imaging. Labeling mesenchymal
   stem cells with MNPs and conducting magnetic resonance imaging
   detection can significantly enhance the success rate of stem cell
   transplantation. Currently, histopathological techniques are limited in
   providing information about the fate of implanted cells after animal
   euthanasia or via biopsy surgery. To facilitate real-time tracking of
   the cellular position, viability, and functional status, recent
   advancements have introduced imaging techniques that involve the
   internalization of MNPs by cells^[39]9. These internalized MNPs
   function as imaging agents and are monitored via MRI. The utilization
   of MRI for cell tracking presents several significant advantages,
   including high spatial resolution, excellent anatomic background
   contrast, and the absence of ionizing radiation exposure^[40]10.
   Additionally, MRI enables long-term cell tracking, with the ability to
   monitor cells over periods of months. Some MNPs have already been
   employed to investigate the fate of labeled cells in animal models and
   specific clinical contexts^[41]11–[42]13. For instance, fluorescent
   magnetic nanoparticles have been used to label B10 cells, facilitating
   continuous magnetic resonance imaging for up to four weeks
   post-injection and transplantation to assess the survival and
   differentiation of labeled cells. Current research indicates that cells
   labeled with MNPs generally exhibit non-cytotoxic behavior and do not
   affect cell proliferation and differentiation abilities^[43]14.
   However, a few studies have reported that MNPs-labeled stem cells may
   partially lose their ability to differentiate in a
   concentration-dependent manner^[44]15. Therefore, before utilizing
   MNPs-labeled MSC, it is imperative to confirm that these cells continue
   to fulfill the established criteria for MSC and to evaluate whether
   MNPs have any impact on the biological functions of mesenchymal stem
   cells. MSC show great therapeutic potential in regenerative
   medicine^[45]16,[46]17. The International Society for Cell Therapy has
   established the following minimum criteria for MSC^[47]18: (1) MSC must
   adhere to uncoated plastic culture dishes under standard culture
   conditions; (2) MSC must simultaneously express CD105, CD73, and CD90,
   but not CD45, CD34, CD14, CD11b, CD79, CD19, or lineage markers such as
   HLA-DR; (3) MSC must be able to differentiate into osteoblasts,
   adipocytes, and chondrocytes in vitro. MNPs-labeled MSC must adhere to
   the aforementioned standards. The biological function of MSC may be
   influenced by the size, concentration, particle shape and coating type
   of MNPs^[48]15,[49]19,[50]20. Therefore, the impact of each type of
   MNPs on the biological functions of MSC should be assessed prior to
   their application. Currently, research on the biological effects of
   MNPs on MSC primarily focuses on aspects such as cell viability,
   proliferation, apoptosis, migration, and cell cycle regulation.
   However, in central nervous system diseases, transplanted MSC exert
   their effects mainly by secreting inflammation-related factors to
   improve the inflammatory microenvironment^[51]21. Yet, whether the
   uptake of MNPs by MSC affects the secretion of inflammatory cytokines
   and the underlying mechanisms remains unknown.

   In this study, we treated MSC with PLL-modified MNPs and
   comprehensively evaluated the effects of MNPs on the biological
   functions of MSC, including cell viability, cell phenotype, directed
   differentiation, proliferation, apoptosis, cell cycle, migration, and
   cytokine secretion. Our results indicate that PLL-modified MNPs can not
   only serve as tracers for biological imaging of MSC, but also
   significantly enhance the migration and anti-inflammatory capabilities
   of MSC. This dual functionality can greatly improve the transplantation
   success rate and overall functional performance of MSC.

Results

MNPs did not affect the differentiation potential or the expression of
intrinsic surface markers of MSC

   Our previous studies demonstrated that MSC transplantation can
   effectively promote spinal cord injury repair^[52]21. However, the
   survival and migration of transplanted MSC cannot be tracked in vivo.
   Magnetic nanomaterials, known for their excellent biocompatibility,
   have been widely used in the medical field, particularly as tracers in
   clinical applications^[53]22. MNPs possess strong magnetic resonance
   imaging capabilities and can effectively track cells after being
   ingested by MSC^[54]13,[55]23. Nevertheless, the potential effects of
   MNPs on MSC need to be thoroughly tested before their clinical use. We
   developed zinc-doped magnetic nanoparticles (Zn[0.4]Fe[2.6]O[4]), which
   exhibited a cubic shape with an average diameter of approximately 60 nm
   (Fig. [56]1A). These MNPs were modified with poly-L-lysine (PLL) to
   enhance their biocompatibility. The PLL-modified MNPs (PLL-MNPs)
   exhibited a high saturation magnetization of 76 emu g⁻¹ (Fig. [57]1B),
   indicating their rapid response to external magnetic fields. After
   surface modification, the MNPs showed excellent hydrophilicity and
   dispersibility. The hydrodynamic diameters of the MNPs and PLL-MNPs
   were 490 ± 50 nm and 253 ± 10 nm, respectively (Fig. [58]1C). The zeta
   potentials of the MNPs and PLL-MNPs were − 6.14 ± 0.32 mV and
   34.82 ± 0.52 mV, respectively (Fig. [59]1D). The PLL coating
   significantly improved the dispersibility of the MNPs through
   electrostatic repulsion and steric hindrance generated by the high
   density of amino groups. Additionally, PLL provided MNPs with the
   ability to interact with MSC.

Fig. 1.

   [60]Fig. 1
   [61]Open in a new tab

   Characterization of MNPs and identification of MSC. (A) Morphology of
   MNPs under TEM. (B) Curve displaying the magnetization of dry MNPs in
   the magnetic field. (C, D) Hydrodynamic diameter and zeta potential of
   MNPs and MNPs-PLL (n = 3). (E) Morphology of MSC and MSC-MNPs. (F)
   MSC-MNPs can induce differentiation into bone, cartilage and fat
   through alizarin red, alcian blue, oil red O staining. (G, H)
   Identification of the expression of MSC markers via flow cytometry.

   In accordance with the International Society for Cellular Therapy
   (ISCT) criteria for MSC identification, we assessed MSC that had
   ingested MNPs. We found that the morphology of MSC did not
   significantly change after uptake of PLL-MNPs (MSC-MNPs, 0.1 µg/µL,
   Fig. [62]1E). Moreover, the capacity of the MSC-MNPs for lipogenesis,
   osteogenesis, and chondrogenesis was the same as that of the untreated
   MSC(Fig. [63]1F). We also examined cell surface markers and found that
   MSC-MNPs expressed CD90 and CD105, but did not express CD34, CD11b,
   CD19, CD45, or HLA-DR, similar to untreated MSC(Fig. [64]1G, H). These
   results indicated that MNPs uptake by MSC does not affect their
   differentiation potential or the expression of intrinsic cell surface
   markers.

The migration ability of MSC was enhanced after MNPs uptake

   Normally, tracers are not expected to affect normal cell functions. To
   determine whether the uptake of MNPs by MSC has any effect on their
   function, we examined cell proliferation, apoptosis, the cell cycle,
   and migration. Prussian blue and Eosin staining revealed that iron
   oxide-containing blue nanoparticles were taken up by MSC (Fig. [65]2A).
   We assessed cell proliferation via the CCK-8 assay and found no
   significant differences between the MSC-MNPs group and the MSC group at
   6,12,24,48 h (Fig. [66]2B). These findings suggest that the uptake of
   MNPs by MSC did not induce cytotoxicity. To investigate whether MNPs
   influence cell apoptosis, we performed FACS analysis on MSC treated
   with MNPs for 24 h. There were no significant differences in PI and
   Annexin V double-positive populations between the MSC-MNPs group and
   the MSC group (Fig. [67]2C, D), indicating that MNPs uptake did not
   affect cell apoptosis. We also examined the cell cycle distribution of
   MSC after 24 h of MNPs treatment. Similarly, no significant changes
   were observed between the MSC-MNPs group and the MSC group
   (Fig. [68]2E, F). Subsequently, we utilized DCFH-DA assays to detect
   reactive oxygen species (ROS). Our findings indicated that MNPs did not
   augment ROS production, as evidenced by both immunofluorescence
   staining and flow cytometry (Supplementary materials Figure [69]S1).
   Interestingly, to assess whether MNPs are involved in MSC migration, we
   conducted a wound scratch assay. After 12 h, the coverage of migrating
   MSC in lesion areas was significantly greater in the MSC-MNPs group
   than in the MSC group (Fig. [70]2G, I). In the transwell assay,
   cultured macrophages were stimulated with LPS to produce chemokines in
   the lower chamber, while MSC or MSC-MNPs were seeded in the upper
   chamber. Crystal violet staining showed that a greater number of MSC
   migrated through the membrane pores in the MSC-MNPs group than in the
   MSC group (Fig. [71]2H, J). These results suggest that the uptake of
   MNPs promotes the migration of MSC.

Fig. 2.

   Fig. 2
   [72]Open in a new tab

   Effects of MNPs uptake by MSC on cell proliferation, apoptosis, the
   cell cycle and migration. (A) Prussian blue and Eosin staining revealed
   MNPs uptake by MSC. All MSC in the MSC-MNPs group were blue. (B) The
   proliferation of MSC after uptake of MNPs was detected via CCK-8 assay.
   (C, D) Using annexin V and propidium iodide (PI) staining, FACS
   analysis revealed a comparable percentage of apoptotic cells in the
   MSC-MNPs group and the MSC group. (E, F) FACS analysis of PI-stained
   cells showed no significant change in the cell cycle in the MSC-MNPs
   group or the MSC group. (G, I) MSC migration was studied via a wound
   scratch assay. At 12 h after scratching, the coverage of migrating MSC
   in lesion areas was significantly greater in the MSC-MNPs group than in
   the MSC group. (H, J) Transwell assays showed that more MSC migrated
   through the membrane pores in the MSC-MNPs group than in the MSC group.
   The data are represented as the mean ± SEM. *P < 0.05, **P < 0.01,
   Two-way repeated measures ANOVA with Bonferroni’s post hoc correction
   in B and Student’s t-test in others, n = 3 in B, F, I, J; n = 5 in D.

MNPs uptake by MSC promoted their anti-inflammatory function and affected the
number of cell subtypes related to migration and immune regulation

   To investigate the effects of MNPs uptake on the anti-inflammatory
   function of MSC, we collected the supernatant from MSC after 24 h of
   MNPs treatment and measured the cytokine levels using the Bio-plex
   assay (Fig. [73]3A). Compared with those in the MSC group, we observed
   significant decreases in TNF-α and MCP-1 in the MSC-MNPs group
   (Fig. [74]3A, B; MSC and MSC-MNPs in pg/ml: 271.4 ± 12.83 and
   150.7 ± 8.708 for TNF-α, 115.1 ± 6.836 and 85.46 ± 5.722 for MCP-1,
   P < 0.01 and P < 0.05, respectively). Conversely, the levels of the
   anti-inflammatory cytokines IL-10, IL-13, RANTES and VEGF were
   significantly increased in the MSC-MNPs group (Fig. [75]3A, B; MSC and
   MSC-MNPs in pg/ml: 14.55 ± 0.98 and 24.39 ± 1.33 for IL-10,
   47.62 ± 1.546 and 74.25 ± 7.341 for IL-13, 41.97 ± 3.13 and
   63.51 ± 6.49 for RANTES, 14.66 ± 1.86 and 22.88 ± 0.95 for VEGF,
   P < 0.01, P < 0.05, P < 0.05, P < 0.05). These findings suggest that
   MNPs uptake by MSC promote the secretion of anti-inflammatory factors
   while reducing the secretion of pro-inflammatory factors.

Fig. 3.

   Fig. 3
   [76]Open in a new tab

   MNPs affect the cytokine secretion and the number of cell subsets of
   MSC. (A) Heatmap of the Bio-plex analysis showing the relative
   expression of 27 cytokines in the MSC and MSC-MNPs groups at 24 h after
   MNPs treatment. (B) Quantitative analysis indicates that, compared with
   the MSC group, the level of TNF-α and MCP-1 decreased, and the level of
   IL-10, IL-13, RANTES and VEGF increased in the MSC-MNPs group. (C-E)
   FACS analysis showed that the numbers of CD184^+ MSC, CD106^+ MSC and
   CD55^+MSC were significantly greater in the MSC-MNPs group than in the
   MSC group. *P < 0.05; **P < 0.01; Student’s t-test, n = 3 in each
   group.

   Single-cell sequencing revealed the presence of distinct functional
   subgroups of MSC^[77]24,[78]25. Specifically, CD184⁺ MSC exhibited
   stronger migration capabilities, facilitating cell movement to injury
   sites. The CD106⁺ MSC subgroup expressed high levels of
   immunosuppressive factors, regulated cytokine secretion, and
   demonstrated robust immunomodulatory abilities. CD55, a complement
   activation suppressor, protected MSC from cell death caused by
   complement activation. Given that MNPs uptake by MSC enhances migration
   and anti-inflammatory cytokine secretion, we used flow cytometry to
   detect the expression of CD184, CD106, and CD55 on the surface of MSC.
   Under normal conditions, MSC express low levels of CD184 and CD55, with
   only a small amount of CD55 detected. After MNPs uptake, the number of
   CD184⁺, CD106⁺, and CD55⁺ subsets significantly increased (MSC and
   MSC-MNPs percentage of cell subtypes: 0.8800 ± 0.1818 and 7.367 ± 1.691
   for CD184⁺, 3.81 ± 0.53 and 13.20 ± 2.65 for CD106⁺, 2.39 ± 1.07 and
   30.40 ± 5.39 for CD55⁺, P < 0.05, P < 0.05, P < 0.01, Fig. [79]3C-E).

MNPs uptake by MSC alters proteomic profiles

   To further explore the potential molecular mechanisms by which MNPs
   affect cytokine secretion and the number of cell subsets of MSC, we
   performed data-independent acquisition (DIA) proteomics using protein
   extracts from cultured MSC treated with or without MNPs. We identified
   237 differentially expressed proteins (DEPs; FDR < 0.05,
   |FoldChange|>1.5) between the MSC and MSC-MNPs groups, including 93
   upregulated and 144 downregulated DEPs (Fig. [80]4A). The expression of
   multiple proteins related to the cytoskeleton and axon orientation was
   significantly modified in the MSC-MNPs groups. These include RhoU
   GTPase (RHOU), Slit Guidance Ligand 2 (SLIT2), and Slit Guidance Ligand
   3 (SLIT3). Additionally, the upregulation of tissue repair-related
   proteins, such as Midkine(MDK), Hepatocyte Growth Factor
   Activator(HGFAC); cell migration related proteins, such as Wnt Family
   Member 5 A(WNT5A), Semaphorin 3 C(SEMA3C); and anti-inflammatory and
   antioxidant stress-related proteins, such as Apolipoprotein A1(APOA1),
   IL-13, Ferritin Light Chain(FTL) was observed. In contrast,
   pro-inflammatory associated proteins, such as Chitinase 3-like protein
   1(CHI3L1), Bromodomain Containing 3(BRD3), and TNF Receptor Associated
   Factor 3(TRAF3) were downregulated. These changes were identified in
   the volcano diagram of 237 DEPs(Fig. [81]4A). We also found that the
   expression of many cytokines, such as connective tissue growth
   factor(CCN2), insulin like growth factor(IGF), CD55, was significantly
   elevated in the MSC-MNPs group(Fig. [82]4B). To confirm DIA results, we
   performed western blotting with antibodies against SLIT2, HGFAC, MDK,
   APOA1 and CD55. These proteins were confirmed to be upregulated in the
   MSC-MNPs group (Fig. [83]4C, D, original blots are presented in
   Supplementary materials Figure S2). Additionally, Western blot results
   confirmed the downregulation of CHI3L1 and TRAF3 in the MSC-MNPs group
   (Fig. [84]4C, D). We performed KEGG pathway enrichment analysis of the
   237 DEPs and found that these DEPs were clustered in signaling pathways
   related to complement and coagulation cascades and axon guidance
   pathway (Fig. [85]4E). Specifically, ten DEPs were identified in the
   complement and coagulation cascades pathway (F2, F3, F5, F10, uPAR,
   TPA, C3, Clusterin, DAF, fibrinogen), and seven DEPs were related to
   the axon guidance pathway (SLIT2, SLIT3, SEMA3A, WNT5A, EPHA, RGS3,
   PAR3). To further explore the relevant mechanisms, we used Western blot
   to detect the expressions of PD-L1 and FasL and found that MNPs did not
   affect the expressions of PD-L1 and FasL in MSCs(Figure S3 A-C).

Fig. 4.

   [86]Fig. 4
   [87]Open in a new tab

   MNPs alters proteomic profiles of MSC. (A) Volcano plot showing the
   proteins that were upregulated and downregulated in the MSC-MNPs group
   compared to the MSC group. DIA proteomics revealed 93 upregulated DEPs
   and 144 downregulated DEPs in the MSC-MNPs group compared with the MSC
   group (FDR < 0.05, |FoldChange|>1.5), including polarization and axon
   guidance related proteins, pro-inflammatory related proteins and
   anti-inflammatory related proteins. (B) Heatmap showing that MNPs
   uptake by MSC promoted the expression of cytokine-related proteins. (C,
   D) Western blot analysis verified the upregulation of SLIT2, HGFAC,
   CD55, APOA1 and MDK and the downregulation of TRAF3 and CHI3L1 in the
   MSC-MNPs group compared with those in the MSC group. The first three
   lanes correspond to three samples from the MSC group, while the last
   three lanes correspond to three samples from the MSC + MNPs group, as
   indicated in the figure. (E) KEGG pathway enrichment analysis revealed
   that these DEPs were clustered in the top 15 signaling pathways related
   to Complement and coagulation cascades and Axon guidance pathway.
   *P < 0.05, **P < 0.01, ***P < 0.001, triplicate experiments in each
   group, Student’s t-test.

The inflammatory regulatory function of MSC-MNPs promotes the phenotypic
switch of reactive astrocytes to alleviate neuroinflammation

   The above results revealed that MSC-MNPs possess anti-inflammatory and
   axonal orienting functions. Typically, MSC-MNPs exert their effects
   through paracrine mechanisms. To verify the functional impact of
   MSC-MNPs, we designed an experiment involving the coculture of MSC-MNPs
   with reactive astrocytes. Reactive astrocytes undergo complex
   morphological, molecular, and functional changes in response to
   injury^[88]26. In earlier studies, astrocytes were categorized into A1
   and A2 phenotypes based on their gene expression, secreted cytokines
   and function^[89]27. Using a similar classification, we further
   investigated the effect of MSC-MNPs on the astrocyte response in vitro,
   focusing on phenotype, synapse length, and angle. Cultured reactive
   astrocytes were subjected to scratch injury and coculture with MSC for
   24 h (Fig. [90]5A). The scratch induced the formation of astrocyte
   protrusions (Fig. [91]5B), but there was no significant difference in
   protrusion length between the MSC-MNPs group and the MSC group
   (Fig. [92]5C). Additionally, there was no significant change in the
   percentage of astrocytes with protrusions oriented perpendicularly to
   the lesion (45°−90°) in the MSC-MNPs group compared with the MSC group
   (Fig. [93]5D). Reactive astrocytes were cocultured with MSC for 24 h,
   after which their mRNA was collected for RT-qPCR analysis. In the
   MSC-MNPs group, the levels of A1-phenotype genes C3 (MSC: 1.00 ± 0.10,
   MSC-MNPs: 0.35 ± 0.14, P < 0.01) and Serping1 (MSC: 1.00 ± 0.13,
   MSC-MNPs: 0.34 ± 0.08, P < 0.01) were significantly lower, while the
   levels of A2-phenotype genes Tm4sf1 (MSC: 1.00 ± 0.15, MSC-MNPs:
   1.85 ± 0.23, P < 0.05) and Ptx3 (MSC: 1.00 ± 0.73, MSC-MNPs:
   4.27 ± 0.65, P < 0.05) were significantly higher compared to the MSC
   group (Fig. [94]5E). Meanwhile, we measured the levels of neurotrophic
   factors—GDNF, NGF, CTNF, and BDNF—in the supernatant of astrocytes
   using ELISA experiments (Fig. [95]5F-I). We found that, compared with
   the MSC group, the levels of GDNF in MNP-MSC were significantly
   increased (MSC and MSC-MNPs in pg/ml: 85.46 ± 5.72 and 115.1 ± 6.836,
   P < 0.05). These results suggest that MSC-MNPs promote the phenotypic
   switch of reactive astrocytes, thereby alleviating neuroinflammation.

Fig. 5.

   [96]Fig. 5
   [97]Open in a new tab

   Effects of MSC-MNPs on the polarization and phenotype of reactive
   astrocytes. (A) Schematic diagram of the coculture of MSC-MNPs and
   astrocytes. (B, C) Astrocyte reactivity was studied in cultured cells
   via a wound scratch assay. At 24 h after scratching, GFAP
   immunostaining revealed protrusions of polarized astrocytes, and there
   was no significant difference in protrusion length between the MSC-MNPs
   group and the MSC group. (D) Protrusion orientation was evaluated by
   the angles between the protrusion axes and the scratch line (indicated
   in B), which revealed that the number of astrocytes with protrusions
   preferentially oriented perpendicular to the lesion (45°−90°) was
   comparable between the MSC-MNPs group and MSC group. (E) RT-qPCR
   quantification of C3, Serping1, Clcf1, B3gnt5, Tm4sf1 and Ptx3
   transcripts in cultured spinal cord astrocytes. The mRNA levels are
   normalized to that of GAPDH. (F-I) The levels of neurotrophic factors,
   including GDNF, NGF, CNTF, and BDNF, in the supernatant of astrocytes
   were detected using an ELISA experiment. *P < 0.05, **P < 0.01; n = 100
   in C and D, n = 4 in E, n = 3 in F-I; Student’s t-test in C and E;
   Chi-square test in D.

Discussion

   In this study, we selected umbilical cord-derived mesenchymal stem
   cells (UCMSC) because of their robust immune regulatory and
   neuroprotective properties^[98]21. Additionally, UCMSC help to
   alleviate the limitations associated with the difficulty of generating
   sufficient cell numbers, making them an important source for
   transplantation^[99]28. Animal model studies have demonstrated that
   UCMSC transplantation holds great potential for improving the local
   microenvironment following central nervous system injury, as well as
   for promoting neural remodeling and functional recovery^[100]21. At
   present, a few types of MSC transplants have advanced to Phase I/II
   clinical trials^[101]29,[102]30. However, monitoring the spatiotemporal
   migration and homing of MSC, as well as assessing their implantation
   efficiency and functional capacity, remain challenging problems that
   need to be addressed. Currently, the use of the magnetic resonance
   imaging (MRI) capabilities of MNPs to track the fate and distribution
   of transplanted cells has been applied in animal models and some
   clinical settings^[103]7,[104]31,[105]32. From the perspective of
   future clinical applications, MNPs offer several significant
   advantages. However, detailed information regarding their impact on the
   biological function of MSC at the cellular level remains limited.

   Unmodified MNPs are difficult for MSC to take up and may adversely
   affect the proliferation and migration of MSC. To address this issue,
   various strategies have been developed to enhance the uptake of MNPs by
   MSC. These strategies include coating MNPs with a variety of
   substances, such as surfactants, precious metals, silica, and chitosan.
   Currently, the three main types of SPION coatings—polyethylene glycol
   (PEG), starch, and dextran—appear to be the most beneficial and have
   been approved for human use^[106]33. However, different MNPs or MNPs
   with different coatings exert varying effects on MSC. For example,
   Ferucarbotran—a SPION with a crystalline nonstoichiometric Fe^2+ and
   Fe^3+ iron oxide core—stimulates in vitro MSC proliferation^[107]33.
   Meanwhile, MNPs coated with polyglucose-sorbitol-carboxymethylether
   (PSC) have been shown to enhance osteogenic differentiation in human
   bone marrow-derived mesenchymal stem cells (hBMSC) in vitro^[108]34. In
   contrast, no significant effects were observed when MSC were treated
   with silica-coated MNPs^[109]20. Clearly, more research is needed in
   this area to establish the effects of different MNPs at various
   concentrations on specific MSC functions and differentiation pathways.

   Studies have shown that the concentration of MNPs can influence
   cellular biological functions, which may be related to the oxidative
   stress induced by high concentrations of MNPs^[110]15. For instance,
   the optimal concentration for MNPs@SiO₂-RITC in vitro is 0.1 µg/µL,
   while 1.0 µg/µL represents the plateau concentration for cellular
   uptake. When MSC were treated with 1.0 µg/µL MNPs@SiO₂-RITC their
   migration was significantly inhibited, ROS production was promoted, and
   cell viability was affected. In contrast, no significant cytotoxic
   effects were observed when MSC were treated with 0.1 µg/µL
   MNPs@SiO₂-RITC^[111]20. Ideally, the use of MNPs as tracers should have
   no significant effect on these functions of MSC, indicating that they
   have no cytotoxic effect. In our study, we developed zinc-doped
   magnetic nanoparticles (Zn[0.4]Fe[2.6]O[4]), which were confirmed to
   possess excellent capabilities for MRI^[112]35. We utilized MNPs at a
   concentration of 0.1 µg/µL, which is suitable for normal MRI^[113]32to
   evaluate their effects on MSC morphology, phenotype, targeted
   differentiation ability, proliferation, apoptosis, the cell cycle, and
   migration. Additionally, the MNPs used in our study were only modified
   with PLL, without any other coatings, to avoid the possible effects of
   surface coating on cellular biological functions^[114]36.
   Interestingly, we found that MNPs can enhance the migratory function of
   MSC, which is beneficial for their homing.

   Given that transplanted MSC primarily exert their effects through
   exocrine mechanisms and can improve the inflammatory microenvironment
   by regulating cytokine secretion^[115]21we investigated the impact of
   MNPs on MSC cytokine secretion. Specifically, through multi-cytokine
   analysis, we discovered that the expression of pro-inflammatory
   cytokines was reduced in the supernatant of MSC treated with MNPs, and
   the expression of anti-inflammatory factors was significantly
   increased. These finding suggest that MNPs uptake by MSC may alter
   their exocrine function, thereby enhancing their anti-inflammatory
   capabilities. Single-cell RNA sequencing (scRNA-seq) datasets have
   revealed the presence of distinct subpopulations of
   MSC^[116]24,[117]25. These MSC subpopulations exhibit specific surface
   proteins, different biological activities, and corresponding
   therapeutic effects. In our study, flow cytometry revealed that after
   MSC internalized MNPs, the number of CD184^+ MSC subpopulations
   associated with migration increased significantly. This finding is
   consistent with those reported in previous studies^[118]35. Similarly,
   the numbers of CD106^+ and CD55^+ MSC subpopulations, which are related
   to immunosuppression, also increased. These results further support the
   potential role of MNPs in promoting MSC migration and improving the
   inflammatory microenvironment.

   In our DIA proteomics analysis, we discovered that MNPs not only
   enhanced the anti-inflammatory function of MSC but also affected
   proteins associated with cell polarization and axon guidance. To
   further validate these findings, we conducted coculture experiments
   with MSC-MNPs and activated astrocytes. These results indicate that
   MSC-MNPs can indeed promote the conversion of astrocytes from the A1
   phenotype to the A2 phenotype and enhance the secretion of the
   neurotrophic factor GDNF, thereby mitigating neuroinflammation^[119]37.
   However, MSC-MNPs did not influence the synaptic orientation of
   astrocytes, which may be attributed to the fact that the axon-guiding
   functional proteins of MSC do not directly act on astrocytes. However,
   further experimental verification is still needed to elucidate the
   axon-guiding function of MSC.

   In summary, this study demonstrated that the MNPs(Zn[0.4]Fe[2.6]O[4])
   we developed at a concentration of 0.1 µg/µL had no cytotoxic effect on
   MSC and could serve as effective tracers for these cells. Furthermore,
   the MNPs were found to enhance the migration function of MSC and
   increase their anti-inflammatory capabilities. However, the
   comprehensive impact of MNPs on the biological functions of mesenchymal
   stem cells still requires further verification through animal model
   experiments. Therefore, in the future, we plan to investigate the
   effects of MNPs on the biological functions of MSC from different
   sources and to evaluate the in vivo MRI imaging efficacy of MSC-MNPs,
   thereby ensuring their safety and effectiveness.

Materials and methods

Synthesis of PLL-Modified magnetic nanoparticles

   Zinc acetylacetonate hydrate (97%) and oleic acid were purchased from
   Aladdin. Iron (III) acetylacetonate (97%) and dibenzyl ether (> 98%)
   were obtained from Sigma. ε-Poly-L-lysine (PLL, Mw = 4000) was sourced
   from MACKLIN (Shanghai). The zinc-doped magnetic nanoparticles
   (Zn[0.4]Fe[2.6]O[4], MNPs) were synthesized via high-temperature
   pyrolysis, based on a previously published method^[120]38. To impart a
   positive surface charge to the MNPs, they were modified with PLL
   through a surface double-exchange process, which enables the MNPs to
   interact with the MSC. Briefly, 5 mg of MNPs were dispersed in 10 ml of
   ethanol. Then, 50 mg of PLL was dissolved in 5 ml of deionized water.
   The ethanol solution of MNPs was exposed to an ultrasonic probe at an
   amplitude of 30%. The PLL aqueous solution was added dropwise to the
   MNPs solution and sonicated for 90 min in an ice bath. The resulting
   product (PLL-MNPs) was washed three times with deionized water and
   collected by magnetic separation.

Identification of MNP characterization

   The morphology of the MNPs was examined via transmission electron
   microscopy (TEM, JEM-1230, JEOL Ltd.). The particle size was
   statistically analyzed with Nano Measurer 1.2 software
   ([121]https://nano-measurer.software.informer.com/1.2/). The zeta
   potential and hydrodynamic size of both the MNPs and PLL-MNPs were
   assessed via dynamic light scattering (DLS, Mastersizer 2000, Malvern,
   Worcestershire, UK). The static magnetic properties of the dry MNPs
   were measured via a vibrating sample magnetometer (VSM, Lakeshore 7407,
   US).

Mesenchymal stem cell culture and identification

   Human umbilical cord-derived MSC were obtained from Saliai Stem Cell
   Science and Technology (Guangzhou, China). The MSC were cultured in
   Dulbecco’s Modified Eagle Medium-F12 (DMEM-F12) supplemented with 10%
   fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin (Gibco).
   The cells were maintained in standard incubators at 37 °C with 5% CO₂.
   To assess cell identity, surface antigens were analyzed via flow
   cytometry using the Human MSC Analysis Kit (BD Biosciences, NJ, USA).
   To evaluate the differentiation potential of MSC, cells at passages 3–4
   were treated with MNPs (0.1 µg/µL) overnight and then induced to
   differentiate into osteogenic, adipogenic, and chondrogenic lineages
   using complete induction and differentiation media (OriCell, CA, USA).
   Differentiation was confirmed by staining with Alizarin Red (for
   osteogenesis), Oil Red O (for adipogenesis), and Alcian Blue (for
   chondrogenesis).

Flow cytometry analysis

   Flow cytometry was employed to monitor cell surface antigens, cell
   apoptosis, cell cycle distribution, and ROS levels. Cell surface
   antigens were detected via the Human MSC Analysis Kit (BD Biosciences,
   NJ, USA), mouse anti-human CD184-PE antibodies (eBioscience,
   California, USA), mouse anti-human CD106-PE antibodies (eBioscience),
   and mouse anti-human CD55-FITC antibodies (eBioscience). Briefly, MSC
   treated with MNPs (0.1 µg/µL) overnight were collected (5 × 10⁵ cells)
   and incubated for 30 min with the added antibodies at room temperature
   in the dark. After washing three times, the cells were analyzed via a
   flow cytometer (Beckman Coulter, CA, USA). Cell apoptosis and the cell
   cycle were assessed via the Apoptosis and Cell Cycle Analysis Kit
   (Invitrogen). For apoptosis detection, 1 × 10⁶ cells were collected and
   stained with Annexin V-FITC and propidium iodide (PI). For cell cycle
   analysis, 1 × 10⁶ cells were fixed with 70% ethanol and stained with
   PI. The detection of ROS levels was performed using the ROS Assay Kit
   (Beyotime, S0033S), following the provided instructions. The data were
   analyzed via FlowJo software v11
   ([122]https://www.flowjo.com/flowjo/download).

MSC chemotaxis assay

Transwell assay

   Transwell chambers with 8-µm-pore membranes (Beyotime, Shanghai, China)
   were used in 6-well plates. MSC (5 × 10³ cells/ml), treated with or
   without MNPs, were seeded in the upper chambers in 10% FBS-DMEM/F12
   medium (Gibco). The lower chambers contained the supernatant of the
   macrophages treated with LPS and maintained for 12 h at 37 °C with 5%
   CO₂. After incubation, the membranes were fixed with 4%
   paraformaldehyde for 10 min and then stained with Crystal Violet
   solution (Beyotime). The cells on the upper side of the membrane were
   removed via a cotton swab, and the cells on the lower side were imaged
   using a microscope equipped with a 5x objective.

Wound scratch assay

   MSC were seeded in 6-well plates at a density of 5 × 10⁴ cells/well.
   After reaching 80% confluence, the cells were treated with MNPs
   (0.1 µg/µl) overnight. The monolayers were then scratched with a 200 µL
   pipette tip, and the debris was removed by washing. To assess cell
   migration, images were taken with a microscope equipped with a 5x
   objective lens in the bright field every 4 h after scratching. The
   images were analyzed via ImageJ V1.8.0 software
   ([123]https://imagej.net/downloads).

Cell proliferation assay

   The Cells (1 × 10³) were seeded in 96-well plates. Cell proliferation
   activity was assessed at 0, 6, 12, 24, and 48 h via the Cell Counting
   Kit-8 (CCK-8, Dojindo Molecular Technologies, CK04-11), following the
   manufacturer’s protocol.

RT-qPCR

   Total RNA was extracted from cells via TriZol Reagent (Invitrogen,
   Carlsbad, CA, USA) according to the manufacturer’s instructions. cDNA
   was synthesized from 1 µg of total RNA using the Reverse Transcription
   System (Promega). Subsequently, 1 µL of the cDNA was used for PCR
   amplification via the Eco™ Real-Time PCR System (Illumina). The
   sequences of the primers used are listed in Supplementary materials.
   The expression levels of the target genes were evaluated via the 2⁻ΔΔCt
   method.

Cell staining

MSC staining

   MSC treated with MNPs (0.1 µg/µL) overnight were stained with a
   Prussian Blue Stain Kit (Abcam, ab150674), Eosin (Solarbio, G1100) or
   ROS Assay Kit (Beyotime). The staining procedures followed the
   manufacturers’ instructions.

Astrocyte staining

   Astrocytes were incubated with rat anti-glial fibrillary acidic protein
   (GFAP; 13–0300, Thermo Fisher Scientific) overnight at 4 °C. Next, the
   cells were incubated with the appropriate secondary antibody for 1 h at
   room temperature.

Cytokine analysis

   The supernatant of MSC treated with MNPs for 24 h was collected for
   cytokine analysis. The levels of cytokines were measured via the
   Bio-Plex system (Bio-Rad) with a 27-plex cytokine kit (M500KCAF0Y,
   Bio-Rad, CA, USA). This kit includes antibodies against the following
   cytokines: FGF basic, Eotaxin, TNF-α, MCP-1, MIP-1α, RANTES, IFN-γ,
   G-CSF, GM-CSF, IL-1ra, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8,
   IL-9, IL-10, IL-12 (p70), IL-13, IL-15, IL-17 A, IP-10, PDGF-BB, and
   VEGF. The detection of neurotrophic factors was performed using ELISA
   kits (BDNF ELISA Kit: Beyotime, PB070; GDNF ELISA Kit: Thermo Fisher
   Scientific, EEL117; NGF ELISA Kit: Epizyme, HJ265; CNTF ELISA Kit:

   Cusabio, CSB-E07312m,) in accordance with the reagent instructions.
   Three samples were analyzed in each experimental group.

Data-Independent acquisition (DIA) proteome analysis

   Total protein was extracted from MSC treated with or without MNPs, via
   Radio Immunoprecipitation Assay Lysis buffer (RIPA). Protein sequencing
   and the construction of protein libraries were carried out by Guangzhou
   Jidiao. Data analysis was performed via OmicShare and Omicsmart
   tools([124]https://www.omicshare.com/). The criteria for identifying
   differentially expressed proteins included coefficient of
   variation < 0.5, average ratio-fold change ≥ 1.5 or ≤ 0.67, and P
   value < 0.05, as determined by Student’s t-test.

Western blots

   Proteins extracted from MSC-MNPs or MSC, were separated on 10% sodium
   dodecyl sulfate polyacrylamide gels and transferred to 0.45 μm
   polyvinylidene fluoride membranes. The following primary antibodies
   were used: anti-APOA1 (1:1000, Zenbio, 680033), anti-HGFAC (1:1000,
   Zenbio, 670841), anti-CHI3L1 (1:1000, Zenbio, 820410), anti-TRAF3
   (1:1000, Zenbio, 160776), anti-MIDKINE (1:1000, Zenbio, 381110),
   anti-SLIT2 (1:1000, Zenbio, 862267), anti-β-tubulin rabbit polyclonal
   antibody (1:1000, Abcam, ab18207), and anti-GAPDH mouse polyclonal
   antibody (1:1000, Abcam, ab8245), anti-PD-L1 rabbit monoclonal antibody
   (1:1000, CST, #13684), and anti-FasL rabbit monoclonal antibody
   (1:1000, Abcam, ab302905). The secondary antibodies used were
   peroxidase anti-rabbit IgG (1:10000, Abcam, ab6721) and peroxidase
   anti-mouse IgG (1:10000, Vector Laboratories). Immunoreactivity was
   detected by a chemiluminescence detection kit (Epizyme, SQ202). Images
   were captured via the ChemiDoc™ Touch Imaging System (Vilber, France),
   and the signals were quantified using ImageJ V1.8.0 software
   ([125]https://imagej.net/downloads). All the data were normalized to
   the corresponding control average ratios.

Animals

   Animal experiments were approved by the Laboratory Animal Ethics
   Committee at the University of Health and Rehabilitation Sciences,
   China (Approval No. 2023 − 2012), and all methods are reported in
   accordance with ARRIVE guidelines. C57BL/6 mice obtained from Beijing
   Huafukang Biotechnology Co., LTD. During the experiment, animals were
   housed at a temperature of 23 ± 1 °C and maintained on a 12-h
   dark/light cycle. Isoflurane was used to anesthetize the animals when
   required. The method of animal euthanasia was cervical dislocation
   after isoflurane anesthesia.

Primary spinal astrocyte culture and activation

   Astrocyte cultures were prepared from neonatal C57BL/6 mice at
   postnatal days 1–3, based on the previously published method^[126]37.
   Briefly, spinal cord tissues were dissected from neonatal mice, and
   then subjected to trypsinization (0.15% trypsin, 7 min, 37 °C). The
   dissociated spinal cord cells were suspended in astrocyte culture
   medium, which consisted of DMEM-F12(Gibco) supplemented with 10%
   FBS(Gibco). When the cells reached 90% confluence, the flasks were
   shaken at 200 rpm for 20 h to detach the neurons, microglia, and
   oligodendrocytes. The purity of the astrocytes was assessed by GFAP
   immunostaining and DAPI nuclear staining, and only astrocytes with a
   purity greater than 95% were used in subsequent experiments.

   Astrocytes were activated via the supernatant from macrophages treated
   with LPS, as previously described^[127]37. Briefly, macrophages (RAW
   264.7) were treated with lipopolysaccharide (LPS, 200 ng/ml, Thermo
   Fisher Scientific) for 24 h. The supernatant was then collected and
   added to the astrocyte cultures to induce activation.

Astrocyte phenotype and polarization assay

   Transwell chambers with 0.4-µm-pore membranes (Beyotime, Shanghai,
   China) were used in 6-well plates. MSC (5 × 10^3 cells/ml) treated with
   MNPs or not, were seeded into the upper chambers in 10% FBS-DMEM/F12
   medium (Gibco). Purified astrocytes were plated in poly-D-lysine-coated
   6-well plates at a density of 5 × 10⁴ cells/well and served as the
   lower chambers. Prior to adding the upper chamber, the astrocytes were
   pre-cultured for 12 h with the supernatant from the macrophages treated
   with LPS. After coculture for 24 h, some astrocytes were collected for
   RNA extraction. The remaining astrocytes were used for further analysis
   of the protrusion length and angle via a wound scratch assay. To study
   the protrusion length and angle of extension relative to the scratch
   line, astrocytes were fixed in 4% paraformaldehyde and subjected to
   anti-GFAP immunostaining. Images were captured with microscope (Zeiss,
   Germany) and analyzed by ImageJ V1.8.0 software
   ([128]https://imagej.net/downloads).

Statistical analysis

   The data are presented as mean ± SEM. For comparisons of cell
   proliferation at different time points, two-way repeated measures ANOVA
   with Bonferroni’s post-hoc correction was employed. The chi-square test
   was used for the statistical analysis of protrusion orientation. Other
   data were analyzed via Student’s t-test. All the statistical analyses
   were performed using GraphPad Prism 7.04
   ([129]https://www.graphpad.com/features). The significance level was
   set at P < 0.05. *P < 0.05, **P < 0.01, and ***P < 0.001.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [130]Supplementary Material 1^ (2.5MB, docx)

Author contributions

   A. L. designed the research project. X. L. performed studies and data
   analysis. S. Q., X. L. and Y. L. performed data analysis. A. L. and X.
   L. drafted the first version of the manuscript with input from S. Q.
   and X.L. A.L. reviewed and edited the manuscript. All authors read and
   approved the final manuscript.

Funding

   The work is supported by: National Natural Science Foundation of China
   (82401622, A. Liu), Natural Science Foundation of Shandong Province
   (ZR2023QH131, A. Liu), Natural Science Foundation of Qingdao
   Municipality (23-2-1-148-zyyd-jch, A. Liu), and Guangdong Provincial
   Key Laboratory of Laboratory of Spine and Spinal Cord
   Reconstruction(NO. 2023B121203001, A. Liu). The funding sources had no
   role in study conception and design, data analysis or interpretation,
   paper writing or the decision to submit this paper for publication.

Data availability

   The datasets generated and/or analysed during the current study are
   available in the PRIDE repository and project accession: PXD062622.

Declarations

Competing interests

   The authors declare no competing interests.

Ethical approval

   Animal experiments were approved by the Laboratory Animal Ethics
   Committee at the University of Health and Rehabilitation Sciences,
   China (approval No. 20232012), and all methods are reported in
   accordance with ARRIVE guidelines.

Footnotes

   Publisher’s note

   Springer Nature remains neutral with regard to jurisdictional claims in
   published maps and institutional affiliations.

   Xinyue Li and Sheng Qin contributed equally to this work.

References