Our study identified that ADSCs predominantly clustered within clusters 5, 8, and 11. Cells stimulated to differentiate for 1 h were primarily located in cluster 2, while clusters 7 and 9 were dominant at 3 h. At 5 h, cluster 1 contained the majority of differentiated cells, and clusters 0, 3, and 10 were predominant at 6 h and 8 h. Pseudo-chronological trajectory analysis revealed that clusters 5, 8, and 11 of ADSCs were positioned at the initiation points of the three differentiation branches. These clusters predominantly differentiated along two trajectories, with a few cells from clusters 5 and 8 located at the termini of the first and second differentiation tracks, while cluster 11 remained along the first trajectory. Among the top 20 genes with significantly elevated expression at various time points during ADSCs induction, COL3A1, TGFBI, and COL5A2 emerged as potential markers of immature neurons. This insight lays the groundwork for identifying novel cell marker genes and advancing cell type characterization. At the 6 h, 8 h, and prei-1d time points, clusters 0, 3, 4, 6, and 10 were predominantly distributed along the bifurcation and differentiation trajectories of the first and second branches. DEGs analysis highlighted specific upregulation of MT1G, SERPINB4, IL1RN, MT1X, IL33, PTGS2, IL6, SERPINB7, IL24, CXCL1, CXCL3, IL36B, MYO10, and other genes in ADSCs, designating these as significant ADSCs signature genes. Similarly, genes such as COL3A1, FN1, TGFBI, COL1A1, COL1A2, IGFBP4, THY1, and COL5A2 were consistently up-regulated across time points, suggesting their association with neural stem cells. Furthermore, MYL9, TIMP3, CCDC80, and ZNF90 exhibited specific upregulation in the 1 h and 3 h groups, implying their roles as potential markers of immature neurons. Conversely, genes exclusively up-regulated in the 5 h, 6 h, and 8 h groups, such as COL6A3, FBN1, and others, may serve as markers of mature neurons. These genes were primarily involved in biological processes including protein metabolism, cell adhesion, endocytosis, cell migration, lipid metabolism, and cholesterol metabolism. Previous studies have confirmed the involvement of genes like TGFBI, COL1A1, COL1A2, IGFBP4, TIMP3, CCDC80, and COL6A3 in neuronal development and maturation (Wang et al. 2013; Das et al. 2024; Deng et al. 2019; Fang et al. 2023; Brusegan et al. 2012; Lam et al. 2022). However, the specific roles of COL3A1, FN1, THY1, COL5A2, MYL9, ZNF90, and FBN1 in neuronal differentiation remain unexplored, warranting further investigation. Single-cell transcriptomics analysis demonstrated that ADSCs stimulated for 5 h reached a terminal state, forming mature differentiated cells. These cells exhibited essential physiological functions such as adhesion, endocytosis, and migration, alongside active participation in lysosomal function, glutathione metabolism, and endoplasmic reticulum protein processing pathways. However, prolonged induction led to degenerative changes in these differentiated cells, resulting in diminished physiological activity.

According to single-cell transcriptome sequencing data, cells undergoing ADSC-induced neuronal differentiation were visually classified into 14 distinct clusters, with considerable specificity in the DEGs among these clusters. ADSCs predominantly belonged to clusters 5, 8, and 11, while cells at 1 h post-induction were mainly in cluster 2, cells at 5 h were primarily in cluster 1, and cells at 6 h and 8 h were concentrated in clusters 0, 3, and 6. Notably, recent studies have shown that by the fifth hour of ADSCs induction, cells exhibit the morphological, ultrastructural, and electrophysiological characteristics of mature neurons. However, extending the induction period beyond this point leads to degenerative changes in the differentiated neurons, accompanied by reduced electrophysiological function (Ye et al. 2010; Lu et al. 2012, 2013; Cai et al. 2011; Wang et al. 2016; Sun et al. 2014; Ou et al. 2011). Regarding DEGs, the top 20 genes positively expressed in ADSCs but negatively expressed in other time groups primarily included MT1G, SERPINB4, IL1RN, MT1X, IL33, PTGS2, IL6, SERPINB7, IL24, CXCL1, CXCL3 in cluster 5; IL36B, IL33, IL1RN, MT1X, MT1G, IL6, MYO10 in cluster 8; and MT1G, SERPINB7, IL24, and others in cluster 11. These genes, particularly IL33, MT1G, MT1X, and IL1RN, which belong to the top 10 genes with the lowest cell stemness and strong gene stability, can be considered hallmark genes of ADSCs. Additionally, the top 20 DEGs positively expressed in each time group during ADSCs induction and differentiation included genes such as COL1A2, COL1A1, SPARC, SERPINH1, P4HA1 in cluster 0; FN1, IL6ST, ITGAV, C1R, COL6A3 in cluster 1; SERPINE1, NQO1 in cluster 2; CKS2, GDF15, TUBA1B in cluster 3; ZYX, TPM1, FERMT2, FHL2 in cluster 4; MT1X, MT2A in cluster 5; HMOX1, C16orf72, GPX4, MSX1, MMP14, CTSK, CYGB, OAF, HSPA5, DNAJB9, OLFML3, TXNRD1 in cluster 6; HSPB1, SLC3A2, HSPA8, CAMTA2, XBP1, ATF4, PPP1R15A, CREBRF in cluster 7; MT2A, PID1 in cluster 8; FHL2, RPL18, RPS27L, GPX1, SLC3A2, FDXR, RPS13, GDF15, MDM2, S100A6, RPS16, RPL13 in cluster 9; AKR1C1, SERPINE2, LAPTM4A, CLU, CALM1, MRPS6, IFI27, S100A10 in cluster 10; CANX, HSPA5, DNAJB9, SEC11C, SLC3A2, UFM1 in cluster 11; CAVIN1, PKM, HMGA1, ACTB, CFL1, PFN1, ANXA2, S100A11, CCDC80, FAM129B, VIM, PSMD2, PPIA, RAB3B, CAPNS1, FKBP1A, CCND1, EIF5A, TLN1 in cluster 12; and MALAT1, LGALS1, S100A11, ATP5E, TMSB4X, CFL1, RPS20, RPS29, S100A6, RPL31, RPL27, NEAT1, RPL37A, RPL37, RPL30, PFN1, RPL38, RPS18, RPS8, RPS14 in cluster 13. Notably, LGALS1, RPS8, ATP5E, RPS14, RPS29, and S100A6 from this group belong to the top 10 genes with the strongest cell stemness, indicating their relevance as marker genes for ADSCs as pluripotent stem cells. Similarly, the top 20 DEGs that were exclusively negative in ADSCs but positive in differentiated cells at other time points included genes like COL3A1, COL5A1, TGFBI, LOXL2, PSD3, COL5A2, SLC6A15, BGN in cluster 0; CD36, BGN, LAMA4, REV3L, COL12A1 in cluster 1; TRNP1, TIMP3 in cluster 2; HMGB2, SMC4 in cluster 3; PTX3, MYL9 in cluster 4; WISP2 in cluster 6; HMOX1, MKNK2, CREBRF in cluster 7; FDXR, APOBEC3C, MKNK2 in cluster 9; PRG4, AKR1C2, FBN1, WISP2, ADD3, IGFBP6 in cluster 10. These genes, including COL3A1, TGFBI, and COL5A2, can serve as marker genes for immature neurons.

Numerous studies have highlighted the effectiveness of ScRNA-Seq combined with pseudo-chronological analysis in modeling cellular developmental lineages. This approach captures dynamic gene expression changes across different cell subpopulations over time, allowing for the inference of cellular transformations and sequential transitions along developmental trajectories (Treutlein et al. 2016; Hook et al. 2018; Hammond et al. 2019; Li et al. 2019, 2022; Mukherjee et al. 2020; Zhang et al. 2021; Gerrits et al. 2022; Sadick et al. 2022; Tang et al. 2022; Wei et al. 2022; Xing et al. 2022; Yang et al. 2022; Zhou et al. 2022).

Based on the findings of this study, clusters 5, 8, and 11, composed of ADSCs, are predominantly located near the initiation points of the three branches in the early differentiation trajectory. These clusters undergo division and maturation until reaching the bifurcation point. Subsequently, differentiated cells follow distinct branching pathways. A small subset of cells from cluster 5 persists at the terminal regions of the first and second branches, while cluster 8 is located at the end of the second branch, and cluster 11 is confined to the end of the first branch. These observations suggest that different ADSC clusters can differentiate into target cells through specific evolutionary trajectories. For instance, cluster 11 primarily follows the first branch, cluster 8 predominantly progresses along the second branch, and cluster 5 exhibits the flexibility to develop into target cells via both pathways. Previous studies have demonstrated that cells differentiated 1 h and 3 h post-induction exhibit characteristics of immature neurons, as evidenced by their morphology, ultrastructure, and immunohistochemical profiles (Ye et al. 2010; Lu et al. 2012; Cai et al. 2011; Ou et al. 2011). The analysis revealed that cells from the 2nd and 9th clusters, observed at 1 h and 3 h post-ADSCs induction, were distributed across the differentiation growth trajectory, while cells from the 7th cluster were predominantly concentrated at the bifurcation point and along the initial two differentiation trajectories. This gene-level information provides valuable insights into the early differentiation processes during the induction of ADSCs into neurons. At 5 h, when ADSCs had differentiated into mature neurons, cells from the 1st cluster were primarily located near the endpoints of the first and second differentiation trajectories. This finding suggests that, although ADSCs differentiation progresses through distinct branching pathways, the final target cells exhibit a consistent gene expression state. This indicates that multiple cell state pathways can converge to produce the same terminal cell phenotype. These results are consistent with previous research demonstrating that ADSC-derived neurons at this stage exhibit morphological, structural, and electrophysiological characteristics typical of mature neurons (Ou et al. 2011). However, the 0th, 3rd, and 10th clusters, representing the 6th and 8th hours of ADSCs differentiation, were predominantly distributed along the bifurcation and development tracks of the first two branches. This pattern suggests a decline in gene expression levels and the onset of degenerative changes. KEGG pathway annotation further revealed that differentiated cells at these time points exhibit heightened activity in metabolic and signaling pathways associated with neurodegenerative disorders, including Parkinson's disease, Huntington's disease, Alzheimer's disease, and amyotrophic lateral sclerosis.

We conducted an in-depth analysis of the top 20 genes with significantly increased expression during each time group of ADSCs induction and differentiation. Notably, in the 1 h and 3 h time groups, the up-regulated genes were primarily MYL9, TIMP3, CCDC80, and ZNF90. Further examination of the ADSC-induced differentiation pathway revealed that genes such as COL3A1, FN1, TGFBI, COL1A1, COL1A2, IGFBP4, THY1, and COL5A2 exhibited upregulation across all time groups. In contrast, MYL9, TIMP3, CCDC80, and ZNF90 were specifically up-regulated in the 1 h and 3 h groups. These genes were associated with key biological processes such as mitochondrial ATP synthesis coupled to proton transport, ATP biosynthesis, and NADH to ubiquinone activities. Regarding molecular functions, they demonstrated proton transporter activity, cytochrome c oxidase activity, transforming growth factor binding, and protein labeling. Cellular structure analysis highlighted significant mitochondrial activity during these stages. These findings underscore the pivotal role of mitochondrial activity in the early phases of ADSC-induced differentiation, aligning with observations from previous research (Wang et al. 2016). Similarly, the key up-regulated genes COL6A3 and FBN1 were specifically observed in the 5 h, 6 h, and 8 h groups. Notably, COL6A3 was identified as a significant DEG during the 5 h differentiation phase, indicating its potential as an essential marker gene for mature neurons. Cells differentiated at the 5 h time point exhibited diverse biological activities, including protein metabolism, cell adhesion, endocytosis, cell migration, lipid metabolism, and cholesterol metabolism. Additionally, these cells were involved in molecular functions such as collagen binding and glycosaminoglycan binding. Structurally, their cellular components included the endoplasmic reticulum, Golgi apparatus, and melanosome, reflecting their advanced functional specialization at this stage (Cai et al. 2011). Additionally, KEGG enrichment analysis revealed that these genes were associated with pathways including ribosome, ECM-receptor interaction, focal adhesion, lysosome, glutathione metabolism, endoplasmic reticulum protein processing, protein digestion and absorption, fatty acid metabolism, histidine metabolism, β-alanine metabolism, and other related metabolic or signaling pathways. Further investigation into the gene regulatory network activity in terminally differentiated cells indicated that cells at 6 h and 8 h post-ADSC differentiation exhibited diverse biological functions, such as protein metabolism, cell division, cell cycle regulation, fibroblast growth factor binding, electron transfer activity, transforming growth factor binding, protein kinase binding, protease binding, calcium binding, and collagen binding. Structurally, these cells demonstrated active roles for the endoplasmic reticulum and cytoskeleton. These findings suggest that endoplasmic reticulum functionality is critical during the late stages of ADSC-induced differentiation. Moreover, cells at various differentiation stages exhibit distinct activation of specific genes, with their biological roles and associated signaling pathways varying accordingly.