[Biological activity of biomimetic dermal papilla spheres prepared by culture of dermal papilla cells of mice based on hanging drops of gelatin methacrylate and its hair-inducing function in nude mice].

Objective: To investigate the biological activity of biomimetic dermal papilla spheres (DPSs) prepared by three-dimensionally cultured dermal papilla cells (DPCs) of mice based on the biomimetic microenvironment of gelatin methacrylate (GelMA) and hanging drop method and its hair-inducing function in nude mice. Methods: Experimental research methods were adopted. DPCs from the vibrissa of male C57BL/6J mice aged 5 to 6 weeks and keratinocytes (KCs) from the skin of 1 d old C57BL/6J mice were obtained by enzymatic digestion method. A stable expression of DPCs markers such as nerve cell adhesion molecules, alkaline phosphatase (ALP), β-catenin, and α-smooth muscle actin were identified by immunofluorescence method in the third passage of the former cells, while the latter primary cells stably expressed keratin 15, a marker of KCs. The 8th passage of DPCs were re-suspended with GelMA and inoculated on the bottom surface of the Transwell plate insert, and then the GelMA drops were photocrosslinked and cultured upside down later. The DPCs aggregation in GelMA drops after in cultures of 0 (immediately) and 3 day (s) was observed under an optical microscope (the DPCs aggregates were the biomimetic DPSs). The cell viability of 3 day biomimetic DPSs culture was detected by live/dead staining kit. The primary DPCs and the 8th passage of DPCs derived from traditionally two-dimensional cultures, and the biomimetic DPSs prepared by the above-mentioned method were set as primary DPCs group, the 8th passage of DPCs group, and biomimetic DPSs group, respectively. Transcriptome sequencing was performed using the high-throughput sequencing technology platform, with 3 samples in each group analyzed after three days in culture. The principal component analysis, Pearson similarity analysis, and screening of differentially expressed genes were performed using OmicShare Tools based on the transcriptome data. Cluster analysis of expression patterns of differentially expressed genes was performed using time series trend analysis software. The Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment analyses of differentially expressed genes with specific expression patterns were performed using the OmicShare Tools. The cells were grouped as before, and the sex determining region Y-Box 8 (SOX8), matrix metallopeptidase 9 (MMP-9), collagen type ΧΧⅥ alpha 1 chain (COL26A1), and wingless-type MMTV integration site family member 6 (Wnt6) were screened out from the differentially expressed genes according to the random number table, which were determined by real-time fluorescent quantitative reverse transcription polymerase chain reaction (RT-PCR) to verify the consistency between mRNA expression of differentially expressed genes and sequencing results (n=9); the mRNA expressions of DPCs biological function markers fibroblast growth factor 7 (FGF7), Wnt10a, lymphoid enhancement factor 1 (LEF1), ALP, β-catenin, versican, and SOX2 were determined by real-time fluorescent quantitative RT-PCR (n=9). Three male BALB/c nude mice aged 5-6 weeks were divided into primary DPCs group, the 8th passage of DPCs group, and biomimetic DPSs group. The primary DPCs, the 8th passage of DPCs, and the biomimetic DPSs were mixed with primary KCs at a ratio of 2∶1 in cell number and then injected subcutaneously into mice of corresponding groups, with 6 injection regions for each mouse. Two weeks after the injection, the full-thickness skin of the injection region was taken, the regenerated hair was counted, and the regenerated hair follicle was observed after hematoxylin-eosin staining. Data were statistically analyzed with one-way analysis of variance, Tukey test, and Bonferroni correction. Results: After 3 days of culture, DPCs aggregated into biomimetic DPSs in GelMA hanging drops from the dispersed state on culture day 0, and the cells in the biomimetic DPSs had good cell activity. After 3 days of culture, principal component analysis showed that compared with that of the 8th passage of DPCs group, the degree of variation among samples within primary DPCs group and biomimetic DPSs group was relatively low, and the degree of variation was the lowest between samples of biomimetic DPSs group and primary DPCs group; the variation of gene profile data of more than 90% of DPCs samples in 3 groups could be explained by the first and second principal components. Pearson similarity analysis showed that samples within each group had good repeatability. The similarity was 0.84-0.95 between samples of primary DPCs group and biomimetic DPSs group, and was 0.72-0.87 between samples of primary DPCs group and the 8th passage of DPCs group. The differentially expressed genes among the three groups were analyzed and 642 differentially expressed genes with group intersection were screened out. Clusters of expression patterns showed that two gene expression patterns had a significant trend (P<0.05), the first pattern showed that gene expression of the 8th passage of DPCs group was significantly lower than that of primary DPCs group or biomimetic DPSs group, and the second pattern showed that gene expression of the 8th passage of DPCs group was significantly higher than that of primary DPCs group or biomimetic DPSs group, including a total of 411 differentially expressed genes. KEGG enrichment analysis showed that the 411 differentially expressed genes were significantly enriched in Wnt signaling pathway and phosphatidylinositol 3 kinase/protein kinase B pathway (P<0.05), while GO enrichment analysis showed that GO terms such as extracellular matrix, classical Wnt signaling pathway, and cell differentiation were significantly enriched (P<0.05). The mRNA expressions of genes SOX8, MMP-9, COL26A1 and Wnt6 of cells in the 8th passage of DPCs group were significantly decreased compared with those in primary DPCs group and biomimetic DPSs group (q=15.950, 8.854, 11.890, 11.050, 9.851, 5.884, 7.418, 4.870, P<0.01), consistent with the sequencing data. Compared with those in primary DPCs group and biomimetic DPSs group, the mRNA expressions of biological function markers FGF7, Wnt10a, LEF1, ALP, β-catenin, versican, and SOX2 of cells in the 8th passage of DPCs group were significantly decreased (q=11.470, 9.795, 4.165, 9.242, 10.970, 10.570, 8.005, 7.472, 4.976, 3.651, 4.784, 5.236, 6.825, 5.214, P<0.05 or P<0.01). Two weeks after injection, nude mice in the 8th passage of DPCs group failed to regenerate hair, while the numbers of hair regenerated in nude mice in biomimetic DPSs group and primary DPCs group were close (q=1.852, P>0.05) and both were significantly higher than the number in the 8th passage of DPCs group (q=18.980, 17.130, P<0.01). In the 8th passage of DPCs group, only necrotic foci were found in the injection region of the skin of nude mice, while regenerated hair follicles were observed in the injection region of the skin of nude mice in both biomimetic DPSs group and primary DPCs group, and melanosis was observed in the cross section of hair follicles. Conclusions: Based on GelMA biomimetic microenvironment and hanging drop method, the biomimetic DPSs culturing model prepared by three dimensional culture of DPCs of mice can restore the hair-inducing ability of high passage of DPCs in nude mice to a certain extent, and its biological characteristics are more similar to those of the primary DPCs, which can restore the characteristics of DPCs after amplification.