ObjectiveTo evaluate the feasibility of using near-infrared-Ⅱ (NIR-Ⅱ) rare-earth down-conversion nanoparticles (DCNPs) as a novel contrast agent for labeling adipose-derived mesenchymal stem cells (ADSCs), and to explore their potential application for in vivo tracking in a mouse model of subacute liver injury. MethodsCore-shell DCNPs (NaYF₄:Yb³⁺,Er³⁺@NaYF₄) were synthesized via a high-temperature thermal decomposition method. Water-soluble DCNPs were prepared through surface modification with DSPE-PEG2000, and further conjugated with fluorescein isothiocyanate (FITC) to observe cellular uptake. The CCK-8 assay was used to detect cell viability after co-incubation of ADSCs with DCNPs at 24 h and 48 h. Confocal laser scanning microscopy was employed to evaluate the labeling efficiency of DCNPs in ADSCs across various probe concentrations and incubation periods. Flow cytometry and differentiation assays were employed to evaluate the effects of DCNPs labeling on ADSCs surface antigen phenotype and multidirectional differentiation potential. NIR-Ⅱ fluorescence imaging was adopted to evaluate the dynamic changes of NIR‑Ⅱ fluorescence signals of labeled ADSCs during continuous in vitro culture. The in vivo toxicity of DCNP-labeled ADSCs transplanted via a single tail vein injection into SPF-grade male Sprague-Dawley (SD) rats was evaluated one month after transplantation through complete blood count, serum biochemical analysis, and multi-organ histopathological examination. A subacute liver injury model was induced in SPF-grade male BALB/c mice using carbon tetrachloride (CCl₄) administration. DCNP-labeled ADSCs were administered via tail vein injection, and the homing of ADSCs in vivo was monitored using NIR-Ⅱ fluorescence imaging. The effects of tracking on ADSC therapy were assessed using biochemical analysis of liver function indicators and hematoxylin and eosin (HE) staining. ResultsThe NaYF₄:Yb³⁺/Er³⁺ nanoparticles synthesized in this study exhibited a spherical crystal structure with particle sizes ranging from approximately 20 to 30 nm, and demonstrated good dispersion stability in water. CCK-8 results showed that DCNPs labeling exerted no cytotoxic effect on ADSCs viability after 24 and 48 h of labeling within the concentration range of 0 to 500 μg/mL used in this study. The cell labeling efficiency of DCNPs increased with dosage, while the cellular uptake remained essentially stable within 16 h. The NIR-Ⅱ fluorescence signal of DCNPs remained stable during the early stage of in vitro ADSCs culture. Although the fluorescence intensity decreased on day 3 of culture (P=0.003), detectable fluorescence signals were still observed in the cells on day 8. Flow cytometry and differentiation assays confirmed that DCNPs labeling had no significant adverse effects on the phenotype and differentiation capacity of ADSCs. One month after the transplantation of DCNP-labeled ADSCs into normal SD rats, blood routine tests, serum biochemical analysis, and multi-organ histopathological examination all revealed no obvious toxic reactions in the body. After the DCNP-labeled ADSCs were injected via the tail vein into mice with subacute liver injury, the serum levels of the liver function indicators alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were significantly reduced (both P<0.05), and the infiltration of inflammatory cells in the liver tissue was markedly decreased. Moreover, in vivo tracing results showed that the labeled ADSCs could home and migrate to the liver within 1 h, with the fluorescence signal gradually weakening over time. By 24 h, the fluorescence signal enriched in the liver had significantly decreased. ConclusionDCNPs possess good biocompatibility for ADSC labeling without affecting stem cell therapy. They may enable real-time, high signal-to-noise ratio in vivo tracking of stem cell homing and migration in ADSC therapy for subacute liver injury.