Matrix-free cryo-microneedles array patch for 3D hair follicle organoids delivery and rapid hair regeneration

Abstract

Hair follicle (HF) organoid transplantation has emerged as a significant advancement for hair regeneration and wound repair. This study developed an efficient and minimally invasive cryo-microneedles array patch (cryo-MAP) technique to transplant self-assembled 3D HF organoids from matrix-free, serum-free culture and low-DMSO cryopreservation. This method successfully avoids the growth suppression caused by hydrogels, allowing HFs to germinate within 3 days. Furthermore, proteomics confirmed that cryopreservation not only maintained high cell viability but also enhanced the organoids' capacity for directional differentiation. The transplanted organoids regenerated complex skin structures, achieving successful hair growth penetrating the skin in approximately 15 days with an over 86% success rate. These results open up new possibilities for organoid transplantation platforms in the fields of tissue engineering and regenerative medicine.

Graphical abstract

1. Introduction

Organoid transplantation, which provides innovative methods for the study of organ regeneration, repair, and disease modeling, has progressed significantly in recent years [[1], [2], [3], [4]]. This technology offers promising therapeutic alternatives for conditions requiring organ transplantation [[5], [6], [7], [8]] and has demonstrated significant potential in regenerating diverse tissues, including the colon, hair follicle, and liver [[9], [10], [11], [12], [13], [14], [15], [16]]. Among these, the transplantation of HF organoids is of crucial value in hair regeneration and skin wound repair. A typical method for HF organoids transplantation is to implant in vitro-cultured HF organoids into the skin directly by manual implantation. Although the manual method enables fully functional hair regeneration in nude mice, it presents challenges of time consumption, potential infection risks and precise batch transplantation. Alternatively, microneedles array patch (MAP) offers minimally invasive, efficient, and safe transdermal delivery of cargos [17]. Recently, the technology has been gradually introduced into living ‘drugs’ delivery, such as cells and even organoids [[18], [19], [20]]. In vivo cultured mesenchymal stem cells (MSCs) were encapsulated in gel matrix and prepared as MAP for delivery to enhance the therapeutic efficacy of MSCs in tissues such as skin and heart [[21], [22], [23], [24], [25]]. In addition, cryo-microneedles array patch (cryo-MAP), which can be easily inserted into skin and dissolve after deployment of the cells, was proposed to deliver cells and preserve their viability and proliferative capacity for a long time [26]. Zheng [18] et al. proposed the feasibility of GelMA cryo-MAP in HF organoid transplantation, demonstrating that GelMA cryo-MAP loaded with dermal cells (DCs) and epidermal cells (ECs) aggregates could successfully form HF organoids in vivo and induce hair growth in the skin of nude mice. However, the cells in the GelMA required additional time (over 7 days) to form HFs due to the inhibition of hydrogel network on cell migration and differentiation, which contrasts with the observation that HF organoids typically sprout within 3 days in traditional in vitro culture systems. Consistent with these in vitro findings, in vivo results indicated that the GelMA system required 14 days to achieve hair growth post-implantation, whereas the direct transplantation of a matrix-free cell suspension completed this process in just 11 days [27]. This observed delay further suggests that the hydrogel network imposes physical constraints on early cellular self-assembly and morphogenesis. More importantly, this result demonstrated that the primary aggregate of two types of stem cells could form HFs in vivo after cryopreservation and transplantation, but no evidence was provided to confirm that in vitro pre-formed 3D organoids with self-organized complex structure could be effectively transplanted by cryo-MAP. The potential impact of cryopreservation on organoid growth and development necessitates further investigation.

In this study, we developed an approach of transplanting HF organoids, which were pre-cultured in vitro to form 3D structures, with matrix-free, serum-free, and low-DMSO cryo-MAP. Before implantation, HF organoids have tightly aggregated and self-organized in vitro and have begun to differentiate into HF stem cells, vascular endothelial cells, and neuronal precursor cells, which can avoid the risk of cell translocation and abnormal differentiation after implantation. The hydrogel-free cryo-MAP avoids the inhibition of hydrogel network on the growth and development of HF organoids (forming HF germination in 3 days). And the proteomics analysis revealed that the cryopreservation did not significantly decrease the cell viability of HF organoids, but enhanced the ability of directional differentiation into HF-related cells. After implantation, the reconstructed HFs exhibited the complex structure of a natural skin attachment, and black hairs broke through the skin surface after about 15 days with a success rate of more than 86 %. These results provide a new approach to organ transplantation in tissue engineering and regenerative medicine.

2. Results and discussion

2.1. Establishment of the transplantation system and construction of HF organoids

A cryo-MAP-assisted organoid transplantation system was constructed to investigate the feasibility of cryopreservation and transplantation of organoids. The developed system is applicable to both cells and organoids and is capable of simultaneously preserving and transplanting organoids. Using HF organoids cultured in vitro for 3 days, cell viability and multilineage differentiation potential were preserved by freezing the microspheres. After transplantation, the neat graft area exhibits a complex structure of natural skin appendages, with hair growing and penetrating the surface of the skin (Fig. 1).

Fig. 1.

Schematic diagram of cryo-MAP transplantation of HF organoids. A) Keratinocytes and mesenchymal stem cells are mixed in vitro to spontaneously form aggregates. B) The formed HF organoids are incorporated into cryo-MAP, creating organoid-loaded cryo-MAP. C)The organoids are implanted into nude mice through cryo-MAP.

To construct HF organoids, we modified the protocol previously reported [28]. The Matrigel concentration was reduced to 0.5 % (v/v) during the first three days, which did not affect the subsequent growth of the HF organoids. Over the first 1–3 days of culture, the HF organoids gradually aggregated and formed spherical structures, where mesenchymal stem cells surrounded the central aggregate of keratinocytes [15,18]. This self-assembled core-shell structure is consistent with previous reports on HF organoid formation, driven by the differential adhesion properties between epithelial and mesenchymal cells [29,30]. These structures transitioned from a loose to a dense configuration, with their diameter decreasing until stabilization. After 3-6 days of culture, hair fiber-like structures, termed hair fibers, began to appear. By day 12, the organoid spheres exhibited substantial growth (Fig. 2A). To confirm the cellular organization and identity, we performed immunofluorescence staining. The results showed that the organoids spontaneously assembled into a core-shell structure, where Vimentin-positive mesenchymal stem cells wrapped around the inner K14-positive keratinocytes (Fig. 2B and Supporting Video 1) [31]. As development progressed, markers associated with HF differentiation, such as CD34 and SOX9, were highly expressed, indicating the maturation of stem cell niches within the organoids (Fig. 2C) [32,33]. Structurally, after 18 days of in vitro culture, the HF organoids demonstrated significant axial extension. Scanning electron microscopy (SEM) revealed that their surface ultrastructure closely resembled that of naturally developed hair (Fig. 2D). Functionally, the maturation of sebaceous gland-like structures was evaluated. Typical lipid droplets were observed in the generated HF-like structures using Oil Red O staining (Fig. 2E) [34], confirming their lipid secretion capability. Finally, we constructed HF organoids with varying cell densities. The results confirmed that organoids across a range of diameters could spontaneously form spheres and exhibit appropriate differentiation markers (Fig. 2F).

Fig. 2.

Characteristics and analysis of HF organoids. A) Representative images of HF organoids captured by bright-field microscopy (Days 3, 6) and confocal microscopy (Day 12). Scale bar = 200 μm. B) Confocal Laser Scanning Microscopy (CLSM) analysis of HF organoids on Day 5. Left: Representative optical cross-sections at the superficial plane (outer region) and central plane (inner core region). Right: 3D volumetric reconstruction (Z-stack) of the organoid. Immunofluorescence staining highlights the core-shell architecture: K14-positive keratinocytes (green) aggregate in the center, enclosed by a Vimentin-positive mesenchymal stem cell shell (red). Scale bar: 100 μm. C) Analysis of stem cell properties of HF organoids. After 3 and 15 days of culture, HF organoids were stained with CD34 and SOX9, and observed by laser confocal microscope. The bottom images show magnified views of the dashed box in the top images. Bar = 100 μm. D) Electron microscopic view of a representative hair generated after prolonged culture. Bar₁ = 100 μm, Bar₂ = 20 μm, Bar₃ = 100 μm. E) The whole-mounted HFs were stained with Oil Red O and observed using a stereomicroscope. Bar = 200 μm. F) Differentiation capacity of HF organoids at different diameters. Bar₁ = 100 μm, Bar₂ = 800 μm.

2.2. Low-DMSO cryopreservation preserves enhanced viability and functionality of HF organoids

We established three cryogenic media to identify the most optimal cryopreservation conditions. The cryogenic medium was prepared using dimethyl sulfoxide (DMSO) at concentrations of 2 %, 6 %, and 10 %, while the sucrose concentration in the solvent was maintained at 100 mM. Various concentrations of the DMSO cryogenic medium were employed to cryopreserve HF organoids cultured in vitro for three days, after which the organoids were thawed. To simulate the post-MAP implantation process, the resuscitated HF organoids were further cultured in vitro for three days. Finally, the changes in each group of samples were assessed through optical microscopy and proteomics analysis, with three replicates for each group. Differentially expressed proteins were identified with a $|{\log }_2\left(\text{fold}\text{change}\right)|$ greater than 1.0 and a false discovery rate (FDR) less than 0.05 between the groups. As illustrated in Fig. 3A, hierarchical clustering analysis revealed three distinct clusters of cell characteristic proteins (p-value <0.05, t-test S0 = 0). Principal component analysis (PCA) revealed significant variability among the three groups. The first principal component accounted for 66.7 % of the variation and effectively distinguished the 2 % DMSO group, the 6 % DMSO group, and the 10 % DMSO group (Fig. 3B). The Venn diagram in Fig. 3C illustrates a total of 3323 common proteins across the three groups, indicating that despite varying cryopreservation media concentrations, HF organoidsmaintained stability during the continued culture process post-resuscitation. However, the unique intergroup protein differences may reveal changes in HF organoids under different cryopreservation conditions. To further analyze the differences exhibited by HF organoids during development, gene ontology (GO) biological process analysis revealed that the 2 % DMSO group enriched GO terms related to cell activity, skin development, cell differentiation and migration, cell stability, and cell morphology regulation. These included keratinocyte differentiation, fibroblast proliferation and differentiation, angiogenesis, neural precursor cell proliferation, and differentiation of mesoderm and endoderm cells. The down-regulated GO terms in the 2 % DMSO group were associated with processes such as oxidative stress response, apoptosis regulation, cellular stress response, and neuronal apoptosis (Fig. 3D and Fig. S1A–B). In total, the 2 % DMSO group exhibited stronger intercellular connections and enhanced cell proliferation and differentiation capabilities. Meanwhile, the high expression of angiogenesis and neural precursor cell proliferation suggested that the HF organoids in this group possessed a greater potential for differentiation into functionally intact HF organoids. In contrast, the GO analysis results of the 6 % DMSO group and 10 % DMSO group revealed that the HF organoids in these groups were more inclined to repair cell metabolism and generate inflammatory responses, with intensity increasing as DMSO concentration increased (Fig. S1C–E). KEGG pathway enriched in the function of the phosphatidylinositol 3-kinase (PI3K)-Akt signaling pathway in the 2 % DMSO group differed significantly from that in the other two groups (Fig. 3E). Existing literature suggests that the PI3K-Akt signaling pathway may be a key mechanism in HF morphogenesis [35,36]. Furthermore, optical microscopy and CCK-3D detection revealed that the HF germination rate was higher in the 2 % DMSO group, with significantly enhanced cell activity (Fig. 3F–H and Fig. S3B). Moreover, studies have shown that low concentrations of DMSO do not induce skin burns, erythema, or desquamation [37,38]. Sulfur atoms in DMSO can capture hydroxyl radicals (OH⁻) and superoxide anions (O²⁻), thereby reducing oxidative stress damage. Regarding metabolism, DMSO is metabolized by oxidation to dimethyl sulfone (DMSO₂) or reduction to dimethyl sulfide (DMS). DMSO₂ is excreted through urine and feces, while DMS is excreted through respiration and via the skin [39].

Fig. 3.

Proteomics analysis of DMSO-mediated effects on 3 Day HF organoids. A–C) Heat maps, principal-component analysis (PCA) and venn diagram of differentially expressed proteins among 2 % DMSO, 6 % DMSO and 10 % DMSO groups. D) GO enrichment (biological process) analysis of the over-expressed proteins of interest in 2 % DMSO group. E) KEGG pathway enrichment analysis of the over-expressed proteins of interest in 2 % DMSO group. F) Comparison of cell activity of HF organoids among three groups. G) Light field observations of HF organoids. H) Statistical analysis of HF budding rate in three groups. Bar = 100 μm.

2.3. Cryopreservation enhances the directed differentiation potential of HF organoids

To further investigate the development of HF organoids in the 2 % DMSO-3 Day group, this study compared it with HF organoids that developed under normal conditions for 3 and 6 days in vitro (Fig. 4A). The results of the clustering heatmap and PCA of differentially expressed proteins demonstrated that the three groups were successfully clustered according to their respective states (Fig. 4B–C). GO analysis revealed that the 2 % DMSO-3 Day group was significantly upregulated in key mechanisms of HF development, including cell differentiation, angiogenesis, HF development, the Wnt signaling pathway, and the NF-κB signaling pathway (Fig. 4D) [40,41]. Conversely, the down-regulated GO terms in the analysis revealed that proteins and pathways associated with muscle cell development, cardiomyocyte differentiation, and contraction were significantly enriched (Fig. 4E). Under normal culture conditions, HF organoids can generate heartbeat activity of cardiomyocytes (Supporting Video 2) [41]. This phenomenon may suggest that cryopreservation reduces the differentiation of cells into unrelated cell types during the differentiation process [42]. To further validate the reliability of the proteomic findings, we selected representative upregulated and downregulated proteins for verification. As shown in Fig. 4F and G, the AUC (Area Under the Curve) analysis indicated high predictive accuracy for these potential markers. Additionally, the expression distributions shown in the violin plots (Fig. S6) were consistent with the proteomic data, confirming that 2% DMSO treatment effectively maintains the specific proteomic signature required for HF development. Furthermore, to ensure cell viability was not compromised post-thawing, we performed Live/Dead staining. The fluorescence intensity analysis (Fig. 4H and I) indicated no significant difference in the signal of living cells (green) between the 2% DMSO-3 Day group and the Control-3 Day group, while the fluorescence intensity of dead cells in the 2 % DMSO-3 Day group was slightly higher than that in the Control-3 Day group, although the difference was not statistically significant. The cryo-MAPs made with 2 % DMSO cryopreservation solution also have sufficient hardness to pierce the skin of mice, extend into the dermis, and deliver cells to specific areas (Fig. S2) [26].

Fig. 4.

Proteomic analysis of thawed and normally cultured HF organoids. A) Schematic diagram of the culture process of the cryopreservation group (2 % DMSO-3 Day) and the control group. B–C) Heat maps and GO enrichment (biological process) analysis of differentially expressed proteins among 2 % DMSO, Control-3 Day and Control-6 Day groups. D–E) GO enrichment (biological process) analysis of the up-expressed and down-expressed proteins of interest in 2 % DMSO group. F–G) Validation of proteomics data: AUC (Area Under the Curve) analysis of representative upregulated and downregulated proteins. H–I) Representative Live/Dead staining images (Green: Live; Red: Dead) and quantitative analysis of fluorescence intensity. Bar = 100 μm.

2.4. Cryo-MAP-assisted transplantation promotes HF organoid regeneration

After transplanting the cryo-MAP loaded with HF organoids into the dorsal skin of nude mice, approximately 15 days later, we observed black hair emerging from the skin surface and growing vertically in the grafted area (Fig. 5A). Further magnification of the transplanted area revealed that HF growth distribution on the back of nude mice was similar to MAPs. HFs first form black vesicular protrusions under the skin, followed by the appearance of black hairs through the skin surface, a process consistent with the morphological stages of natural hair eruption [43]. The average incidence of HF regeneration was 86 %, with a maximum of 94 % (Fig. 5B and Fig. S4). A complete structural integrity, including the root sheaths and sebaceous glands, is essential for the functional regeneration of HFs [43]. After sectioning the skin of nude mice with hair growth and performing HE staining, the results showed that 15 days after transplantation, the regrown hair on the dorsal skin of the nude mice exhibited significant morphological differences compared to the nude mice without implants. It was observed that the hair growing from the reconstructed HFs resembled normal hair on the dorsal skin of C57BL/6 mice, growing vertically and emerging from the dermis through the stratum corneum (Fig. 5D). Immunofluorescence staining results indicated that the outer root sheath (ORS) and inner root sheath (IRS) structures were enveloped by vacuolated and loose connective tissue, with the outer root sheath connected to the host skin [44]. The cytoplasmic hyaline sac, representing sebaceous glands, was also visible. Additionally, on the left side of the lower part of the outer root sheath and within the isthmus of the HF, between the sebum duct and the insertion point of the arrector pili muscle, the presence of the arrector pili muscle can be observed. Detection of the dermal papillary marker Versican and the follicular protrusion marker SOX9 in the HF showed that the protruded area is the primary location of HF stem cells, which possess multi-directional differentiation potential and high proliferative capacity [31]. Rapidly dividing Ki67-positive epithelial cells extended the hair bud region, indicating active anagen-phase growth [45]. The positive expression of K15 confirmed the presence of the stem cell niche in the bulge region, which serves as the attachment point for the arrector pili muscle, further confirming the association between the raised area and the base of the arrector pili muscle in the HE staining results (Fig. 5E–H) [46]. This was the same as the structure of the skin of C57BL/6J mice with naturally growing HFs (Fig. 5C). The transplantation approach using cryo-MAP allows precise control over the location and quantity of the reconstructed hair, while ensuring that the transplanted HF organoids can penetrate the skin and successfully generate hair [47]. The comparative results across in vitro morphogenesis, in vivo regeneration performance, and safety profiles, as detailed in Table S1–S3, provide further evidence for the clinical potential of our Matrix-free cryo-MAP platform.

Fig. 5.

Histology and immunofluorescence of reconstructed hair follicles in nude mice. A) Observation of the cryo-MAP transplantation area. Bar = 250 μm. B) Application of the organoids-loaded cryo-MAP to the mouse dorsal skin at day 15 in vertical view. Bar = 3 mm. C) Immunofluorescence detection of normal-growing HF. Bar = 200 μm. D) HE staining of the nude mice and cryo-MAP groups. Bar = 200 μm. E–H) Immunofluorescence of reconstructed HF organoids, including the structure of the bulge, HF matrix, arrector pili muscle, and sympathetic nerve area. Bar = 200 μm.

3. Conclusion

In this study, we successfully constructed a method for the transplantation of HF organoids based on hydrogel-free cryo-MAPs, and realized the use of preformed 3D structures of HF organoids in vitro for hair regeneration under cryopreservation conditions. Through proteomics analysis, we found that 2 % DMSO was the most effective treatment to maintain the cell viability and differentiation ability of HF organoids. Compared with the non-cryopreserved control group, the optimized cryopreservation protocol not only maintained the cell viability of HF organoids at non-significant levels of reduction, but also enhanced their potential for directional differentiation into HF-associated cell lineages. In vivo transplantation experiments further confirmed that the cryo-MAP loaded with HF organoids could successfully guide the formation of neatly arranged hair structures on the skin of nude mice that break through the skin surface. This study provides an innovative and effective strategy for hypothermic transplantation of HF organoids to achieve hair regeneration.

4. Methods

4.1. Preparation of epithelial cells and mesenchymal cells

Mouse epithelial cells and mesenchymal cells were made to dissociate from each other using a previously described method [28]. Briefly, the skin of C57BL/6 mice (E16–18) was harvested under a stereomicroscope (OLYMPUS, USA), and incubated overnight at 4 °C in Dispase II (4.8 U/mL) (Thermo Fisher Scientific, USA). The following day, epidermal and dermal layers were separated using ophthalmic tweezers under the microscope and then digested with Collagenase Type I (100 U/mL) (Keygen BioTECH, China) at 37 °C for 40 min. The epidermal layer was digested with 0.25 % trypsin (Thermo Fisher Scientific, USA) at 37 °C for 20 min. The digested tissue was filtered through a 40 μm strainer (Corning, USA) to remove debris, followed by centrifugation at 1200 rpm for 5 min. The resulting keratinocytes and mesenchymal stem cells were suspended in a 1:1 ratio in DMEM/F12 medium (Gibco, USA).

4.2. Spontaneous mouse HF organoids

To enable the spontaneous aggregation of the two cell types in vitro to form HF organoids, epithelial cells and mesenchymal stem cells were seeded at 0.2 mL per well in 96-well U-bottom plates (Corning, USA). The medium contained 0.5 % (v/v) Matrigel (Corning, USA) and 1 % (v/v) Gluta-max (Thermo Fisher Scientific, USA) in DMEM/F12. After incubation at 4 °C for 30 min, the plates were transferred to an incubator and cultured at 37 °C with 5 % CO₂, changing 0.1 mL of medium per well every two days. After three days of culture in vitro, HF organoids were mixed with 100 % (v/v) Matrigel in a 6-well plate and cultured in DMEM/F12 for an additional 20 days for in vitro characterization.

4.3. LC-MS/MS analysis

The details were performed as described previously by Sun [48] et al. Each group has three replicates, and each replicate contains only one HF organoid. All samples were washed with PBS (Gibco, China) and then freeze-dried. Then, HF organoid samples are mixed with the lysis buffer. After heating at 95 °C for 10 min and cooling to room temperature, the samples are digested overnight. Trifluoroacetic acid was used to terminate the digestion. Subsequently, twice the volume of ethyl acetate was added, and the mixture was vortexed for 3 min and centrifuged at 15,000 g to separate the aqueous and organic phases. Finally, the upper layer was discarded. The aqueous phase was then purified using EVTOP beads with 95 % of ACN in H₂O, followed by elution with 400 mM NH₄OH in H₂O. After lyophilization, the product was dissolved in 0.1 % formic acid for subsequent analysis.

4.4. LC-MS/MS data processing and bioinformatics analysis

The raw files search was conducted by PEAKS Studio X+ software (Bioinformatics Solutions Inc.) against the Uniprot database (obtained in December 2020) without repetitive entries. The tolerance of the initial precursor was set to 15 ppm, the final tolerance to 6 ppm, and MS/MS tolerance at 0.05 Da. The heatmap and PCA were generated using R Studio. (p-value <0.05, t-test S0 = 0, |log₂(Foldchange)| >1 were regarded as differential expressed proteins).

4.5. HF organoid-loaded Cryo-MAPs fabrication

To realize the matrix-free design, the Cryo-MAPs were fabricated via a cryo-casting strategy without the use of crosslinkable hydrogels. All Cryo-MAPs were fabricated using PDMS molds, with each needle cavity having a diameter of 1000 μm and a height of 1200 μm. These cavities were arranged in a 4∗4 array with a tip spacing of 3000 μm. The organoid-loaded Cryo-MAPs were prepared using a casting process, with the PDMS needle molds acquired from Shiling Laike Die Business Company (Guangzhou City, China). Before fabricating the cell-loaded needles, the PDMS molds were plasma-treated and then sterilized with UV irradiation for 30 min to remove air bubbles and facilitate filling. The molds were then vacuum-processed in a vacuum oven for 30 min. A solution of 1 mL organoid-loaded freezing medium (PBS with defined concentrations of DMSO and 200 mM sucrose, free of any gel-forming polymers) was poured into the mold cavities and settled for 5 min. The molds were then centrifuged at 4000 rpm for 5 min to sediment the organoid spheroids to the needle tips. An additional 0.8 mL of cryogenic medium was then added to fill the PDMS mold cavities, and any excess freezing medium was removed using a pipette. The molds were frozen at −20 °C for 4 h to induce the phase transition from liquid to solid ice, and the needles were gently demolded and transferred to −80 °C for long-term storage.

4.6. SEM analysis of HF organoids

To prepare samples for SEM, the HF organoids were washed with PBS, fixed for 1 h with a mixed solution of 2.5 % glutaraldehyde (Keygen BioTECH, China) and 2 % formaldehyde (Keygen BioTECH, China) in PBS, washed again with PBS, and dehydrated via a 30 %–90 % ethanol gradient (Keygen BioTECH, China), before being washed three times with absolute ethanol. This final solution was then substituted with 100 % t-butyl alcohol (Gibco, USA), and the HF organoids were cooled at 4 °C for 1 h and 20 °C for 1 h. The HF organoids were finally lyophilized, and SEM images were taken at 5–10 kV.

4.7. Skin penetration ability of Cryo-MAPs

Cryo-MAPs loaded with fluorescently labeled (CEA, Beyotime, China) HeLa cells were removed from liquid nitrogen and placed on a precooled metal block. HELA cells at the tip of the needle were observed under a fluorescence microscope (Leica, USA). The same samples were implanted into the skin of nude mice by thumb pressure. The skin changes were observed at 0 s, 30 s, 120 s and 300 s after implantation. HF organoid-loaded cryo-MAPs were implanted subcutaneously into nude mice, and then the mouse skin was cryo-sectioned to directly observe the distribution of cells in the skin under a fluorescence microscope. All images were analyzed using ImageJ software (version 1.54 f, without plugins).

4.8. Cell viability test of Cryo-MAPs

HF organoid-loaded cryo-MAPs were placed into the 96-well plates without supplementation of culture medium and the live/dead cell viability assay (Beyotime, China) was conducted according to the protocol provided by the manufacturers. All fluorescent images were analyzed by ImageJ software (version 1.54 f, without plugins). Quantitative data for cell viability were calculated according to live (green)/dead (red) staining by dividing the number of living cells (green) by the total number of cells.

4.9. Detection of stem cell properties in HF organoids

After thawing, HF organoid-loaded cryo-MAPs were cleaned with PBS for three times and fixed directly with 4 % paraformaldehyde (Beyotime, China), and then immunofluorescence staining was performed to detect the stem cell properties of HF organoids. All fluorescent images were analyzed using the ImageJ software (version 1.54f, without plugins).

4.10. Intracutaneous transplantation of HF organoid-loaded Cryo-MAPs

BALB/c nude mice were anesthetized with isoflurane, and the HF organoids loaded with cryo-MAPs were inserted into the back skin of the nude mice and melted by pressing a cylinder block to the base. Mice were maintained under pathogen-free conditions and fed AD libitum for up to 25 days. The transplanted sites were observed every 2 to 3 days, and the resulting hair images were taken using Sony (DSC-RX100M6) and Sony (ILCE-7M2) digital cameras, respectively.

4.11. Hematoxylin-eosin staining

Skin samples of nude mice with regenerated HFs were obtained and placed in 4 % paraformaldehyde for 24 h at 4 °C. After gradient dehydration with different concentrations of graded ethanol, the tissues were embedded in paraffin. After sectioning, paraffin tissue sections were rehydrated with gradient ethanol. Sections were stained using a hematoxylin-eosin staining kit (Beyotime, China). The results were observed under a microscope (Zeiss, AxioImager D2, Germany).

4.12. Immunofluorescence staining

For HF organoid samples or tissue immunofluorescence staining, samples were fixed overnight at 4 °C with 4 % Paraformaldehyde solution, before the preparation of frozen (8 μm) sections. For cellular immunofluorescence staining, samples were fixed for 40 min at room temperature with 4 % Paraformaldehyde solution. All the samples were blocked for 1 h at room temperature with a solution containing 1 % BSA (Gibco, USA) and 0.01 % Triton X-100 (Gibco, USA) in PBS to prevent non-specific binding. Subsequently, the samples were incubated with the following primary antibodies overnight at 4 °C: anti-Cytokeratin 14 (K14, 1: 500, 10143-1, Proteintech, USA), anti-Vimentin (1: 1000, 60330-5, Proteintech, USA), anti-CD34 (1: 200, Y70757, Immunoway, USA), anti-SOX9 (1: 500, 67439-1, Proteintech, USA), anti-Ki67 (1: 400, 9129T, Cell Signaling, USA), anti-Versican (1: 200, YT4874, Immunoway, USA), and anti-Keratin 15 (K15, 1: 500, 10137-1, Proteintech, USA). After washing three times with PBS to remove unbound antibodies, the samples were incubated with the corresponding fluorescence-conjugated secondary antibodies for 2 h at room temperature in the dark. The secondary antibodies used were: Goat Anti-Rabbit IgG H&L (Alexa Fluor® 488) (1:500, ab150077), Goat Anti-Mouse IgG H&L (Alexa Fluor® 555) (1:500, ab150114), Goat Anti-Rabbit IgG H&L (Alexa Fluor® 555) (1:500, ab150078), or Goat Anti-Mouse IgG H&L (Alexa Fluor® 488) (1:500, ab150113) (all acquired from Abcam, UK). Nuclei were counterstained with Antifade Mounting Medium with DAPI (Beyotime, China). A confocal microscope (FV3000, Olympus, Japan) and microscope (CKX31, Olympus, Japan) were used for imaging.

4.13. Statistical analysis

All graphing and statistical analysis were analyzed and calculated with GraphPad Prism (Version 8.4.2). All statistical results were reported as mean ± standard deviation (SD). Statistical analyses were performed by using a two-tailed Student's t-test or one-way analysis of variance (ANOVA). Results with p-values <0.05 were considered statistically significant.

CRediT authorship contribution statement

Shu Wang: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization. Jie Sun: Data curation, Formal analysis, Funding acquisition, Methodology. Yu Li: Data curation, Formal analysis. Yifan Zhou: Data curation. Yuhan Cai: Data curation. Zulihabire Simayijiang: Data curation, Formal analysis. Zhongze Gu: Funding acquisition. Zhuoying Xie: Funding acquisition, Software, Writing – original draft, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.