Functional regeneration strategies of hair follicles: advances and challenges

Abstract

Hair follicles are essential appendages of human skin that function in protection, sensation, thermoregulation and social interactions. The multicellular components, particularly the dermal papilla, matrix and bulge housing stem cells, enable cyclic hair growth postnatally. However, miniaturization and loss of hair follicles can occur in the context of ageing, trauma and various alopecia-related diseases. Conventional treatments involve the redistribution of existing follicles, which may not be viable in patients lacking follicular resources. Recent progress in the comprehension of morphogenesis and the development of biomaterials has significantly advanced follicle reconstruction, incorporating organ germ assembling, stem cell induction and bioprinting techniques. Despite these advancements, fully restoring hair follicles remains challenging due to the complexities of replicating embryonic signals and sustaining growth cycles. Identifying suitable cell sources for clinical applications also presents a hurdle. Here, we retrospect the progress made in the field of hair follicle regeneration, aiming to offer an exhaustive analysis on the benefits and limitations of these methods, and to foster the development of innovative solutions.

Introduction

Hair follicles (HFs) are self-sustaining mini-organs with diverse stem cell populations and intricate niches. Driven by the periodic activation of hair follicle stem cells (HFSCs), HFs undergo cyclical growth throughout the lifetime. Despite their ability to regenerate, HFs are susceptible to the effects of ageing, stress and environmental stimuli, resulting in reduced hair thickness and density. Currently, the main approach used to combat hair loss relies on the autograft of HF units; however, this method is inadequate for treating patients with insufficient HF reserves. Studies of HF regeneration are increasingly significant due to the prevalence of alopecia and the recognized contribution of HFs to wound healing [1].

Hair follicle morphogenesis

Understanding the mechanism that governs HF morphogenesis is essential for the reconstruction of functional HFs. The development of HF during embryogenesis shares the same driving force with the kidney, lung and tooth, known as epithelial-mesenchymal interactions (EMIs) [2, 3]. Owing to ethical constraints in human studies, knowledge of HF organogenesis is based mainly on mice. Around embryonic day 13, the unspecified epidermis receives “first dermal signals” from the mesenchyme and forms a thickened epithelial layer called the placode, marking the early stage of HF morphogenesis [4, 5]. Placode formation involves the activation of ectodysplasin A/ectodysplasin A receptor (EDA/EDAR) pathways within the epithelium, followed by Wnt and bone morphogenetic protein (BMP) signaling [6, 7]. EDA/EDAR and Wnt pathways collaboratively facilitate placode fate, while BMP represses placode development in adjacent areas [8]. The underlying mesenchymal cells subsequently receive “first epithelial signals” from the placode and aggregate to form the dermal condensate, which is controlled by fibroblast growth factor (FGF) signaling [9–11]. The dermal condensate communicates with the placode to boost the downgrowth of the hair germ, and further evolves into the dermal papilla (DP) enveloped by matrix cells via Wnt/SHH signaling [11, 12]. This signifies the emergence of the HF rudiment, where the DP and matrix collaborate to generate the hair shaft [13]. As HFs mature, the bulge emerges as a vital epithelial pool alongside the matrix, harbouring CD34/CD49f⁺ HFSCs for self-renewal [14, 15].

Hair cycle dynamics

A HF is a sustainable system in which cells are replenished through tightly regulated specification, differentiation and proliferation. Hair growth is fuelled by HFSCs in the bulge area under the regulation of DP, the signaling center [16–18]. Dependent on the dynamic epithelial-mesenchymal crosstalk, HFs exhibit cyclic growth (anagen), regression (catagen) and rest (telogen) after birth [16–18] (Fig. 1, adapted from [2] with permission). During telogen, DPs release inhibitory cues, especially BMPs, to maintain the quiescent state of HFSCs [19–21]. When transitioning to anagen, FGFs and BMP inhibitors, such as Noggin and transforming growth factor-β2 (TGFβ2), accumulate in the DP, and simultaneously, Wnt levels rise in the hair germ. These signals trigger the activation of HFSCs and the formation of new shafts [22, 23]. As catagen ensues, TGFβ1 and Wnt antagonists from the DP induce the apoptosis of basal epithelial cells [24, 25]. The dermal sheath contracts to retract DP upwards along the epithelial strand and relocate DP beneath the bulge, thereby posing appropriate spatial arrangements for cyclic switching [26, 27].

Fig. 1.

Signaling molecules involved in the hair cycle. (a) In telogen, bone morphogenetic proteins (BMPs), such as BMP4 from endothelial cells, BMP2 from adipocytes, and BMP4 from fibroblasts, maintain the quiescence of hair follicle stem cells (HFSCs). Oncostatin M (OSM) released by TREM2⁺ macrophages, and fibroblast growth factor 18 (FGF18) from the dermal papilla (DP) and bulge, also contribute to the dormancy of HFSCs. (b) During the transition from telogen to anagen, factors like FGF7, FGF10, Norrin, and BMP inhibitors (Noggin and transforming growth factor-β2 (TGFβ2)) promote the activation of HFSCs. Other contributors to this shift include endothelin 1 (EDN1) from endothelial cells, platelet-derived growth factor subunit A (PDGFA) from adipocyte precursor cells, and Wnts (Wnt7b and Wnt10a) from apoptotic CD11b⁺ macrophages. Additionally, cutaneous transient receptor potential cation channel subfamily V member 1 (TRPV1) innervations accelerate hair growth via Spp1 from CD9⁺CD26⁺ fibroblasts. (c) In anagen, sonic hedgehog (SHH) from transit-amplifying cells (TACs) further enhances the proliferation of HFSCs and the differentiation of adipocyte precursor cells, leading to the extension of hair follicles. (d) During the anagen-catagen transition, FGF5 from the matrix and outer root sheath (ORS), and Wnt antagonists (dickkopf WNT signaling pathway inhibitor 2 (DDK2) and Notum) from the DP induce the apoptosis of follicular cells. (e) In catagen, epithelial cells release EDN1 to stimulate calcium influx and dermal sheath retraction to relocate the DP. (f) In the subsequent telogen, the DP and the new bulge are positioned adjacent to the old bulge. Fig. 1 was adapted from [2] (Zhang B, Chen T. Local and systemic mechanisms that control the hair follicle stem cell niche. Nat Rev Mol Cell Biol. 2024;25(2):87-100. https://www.nature.com/nrm/), which is not part of the governing OA license, with reproduction permission.

We can see that, EMIs play a pivotal role in HF morphogenesis and regeneration, although the identified signaling signatures are fragmentary. Recent breakthroughs in single-cell transcriptomics have propelled the exploration of a comprehensive regulatory network, but with so many factors involved, further research is needed to minimize redundancy. Benefitting from improved comprehension of underlying mechanisms and advanced bioengineering technologies, HF regeneration strategies have undergone a remarkable transformation, incorporating germ assembling, pluripotent stem cell (PSC) induction and bioprinting [28]. This review thoroughly traces the timeline of HF regeneration, providing novel insights into the optimization of HF bioengineering.

Follicle germ assembling

The pioneering work in HF regeneration dates back to 1970, established on the restoration of EMIs. Yuspa and colleagues grafted cultures derived from embryonic mouse skin onto lesions lacking epidermis and dermis and observed the formation of hair-bearing skin [29]. With the progress in cell isolation techniques, Worst and colleagues demonstrated that neogenesis of HFs occurred only when epidermal cells were transplanted alongside dermal cells [30]. Since then, HF restoration has become practicable through the codelivery of dissociated epithelial and mesenchymal progenitors via patch or chamber assays [31, 32]. While these engineered HFs largely recapitulated the natural anatomical features of HFs, their generative efficiency was limited, probably owing to insufficient EMIs.

The introduction of organ-germ culture revolutionized HF regeneration as an innovative delivery approach [33]. In this method, dissociated epithelial and mesenchymal cells self-organize into follicular primordia prior to transplantation (Fig. 2). Tsuji Lab reported that self-assembled hair follicle germs (HFGs) successfully produced pigmented HFs in vivo, with natural compartments and proper integration into the surrounding skin [34]. These HFs contained CD34/CD49f⁺ bulges and Sox2⁺ DPs, supporting cyclic growth (repeated at least 3 times). Nevertheless, protruding shafts occurred in < 1% of HFGs under Tsuji Lab’s protocol [35], necessitating improvements in trichogenicity. By adding 2% Matrigel before aggregation, Fukuda Team significantly enhanced the sprouting efficiency to nearly 100% [35]. This upswing was attributed to the morphological change in assemblages. Unlike the previous dumbbell-like conformation [34, 36], 2% Matrigel resulted in aggregates with a core‒shell structure, augmenting contacts between cells and reinforcing EMIs [35]. To enable clinical scalability, the high-throughput production of HFGs has been facilitated via culture on biomaterial substrates, such as polydimethylsiloxane and poly(ethylene-co-vinyl alcohol) [36, 37].

Fig. 2.

Methods for assembling hair follicle germs. (a) In initial attempts to regenerate HFs, mixed epithelial cells and mesenchymal cells from embryonic or neonatal mice were directly transplanted via patch or chamber assays. In the method of HFG formation, self-organized primordia are transplanted, with Matrigel supplement shown to enhance trichogenicity. (b) Strategies incorporating IGF2-VEGF, Wnt3a-Wnt10b and MMP14 or IFNγ/VEGF supplements effectively restore HF regeneration in adult cells. (c) Vibrissa DPCs and epithelial cells from adult mice serve as an alternative source for generating HFGs.

Appropriate cell sourcing

Studies on HFGs have focused on using progenitor cells extracted from embryonic or neonatal murine skin, with a few studies utilizing dermal papilla cells (DPCs) from adult HFs [34, 37] (Table 1). However, the use of embryonic or neonatal cells is not feasible clinically due to ethical concerns and immune rejection. Additionally, patient-derived DPCs tend to lose their intrinsic properties when cultured in vitro [38–41]. These limitations underscore the need to explore suitable cell sources, with reprogramming emerging as a potential strategy. Precursors induced through small molecule treatment and genetic engineering hold promise as viable sources [42–45]. It is noteworthy that, despite exhibiting phenotypes resembling natural progenitors, pluripotent stem cell (PSC)-derived precursors may fail to assemble into HFs due to inaccurate cell specification [46]. Therefore, the refinement of induction protocols from PSCs is crucial, offering the opportunity to generate abundant resources while ensuring safety.

Table 1.

Cells and biomaterials used in follicle germ assembling

Organism	Cell	Biomaterial
• Neonatal/embryonic mice [34–36, 47, 48, 96]	• Epithelial cells and mesenchymal cells	• Matrigel [35] • Polydimethylsiloxane [36] • Collagen 1 A [96]
• Adult mice [34]	• DPCs and bulge epithelial cells	/
• 6-week-old rats [37]	• DPCs and keratinocytes	• Poly (ethylene-co-vinyl alcohol)

Rejuvenating adult cells

Environmental reprogramming presents a direct solution to address reproducibility challenges in mature cells. Lei and colleagues identified differential expression profiles between neonatal and adult cells and devised a corresponding system to restore morphogenetic competence [47]. Sequential supplements with IGF2-VEGF/Wnt3a-Wnt10b/MMP14 and continuous administration of PKC inhibitors reinitiated self-organization in adult mouse cells, prompting HF generation from 0 to approximately 40% of newborn-derived germs [47]. An alternative strategy involving PKC/PKR inhibition and IFNγ/VEGF supplement also enabled HF reconstruction from adult cells [48], indicating the potential to revitalize mature cells through ambient cues.

Preserving DPCs for inducibility

During the hair cycle, DPCs act as the primary trichogenic dermal cells [49]. While implanting dissected DPs from rodents and humans successfully stimulated HF growth [50], attempts to replicate this process using human DPCs or late-passage rodent DPCs failed owing to the diminished inducibility during culture [39, 50]. Studies have linked Wnt, BMP and FGF signaling to the hair-inducing potential of DPCs [51–53]. Accordingly, a specialized medium containing agonists of these pathways was formulated (Table 2). This medium was proven to partially recover the expression of DP signature genes, and restore the characteristic function of DPs to communicate with epithelial components [53].

Table 2.

Approaches to inducible DPCs

Cell	Organism	Method
DPCs	• Human	• Three-dimensional culture [50, 54] • A cocktail composed of BIO, BMP2 and FGF [53] • Lef-1 overexpression [56] • Bioprinting calcium molybdate nanoparticles laden with DPCs and macrophages [57]
	• Mice	• Three-dimensional culture [55]
Dermal fibroblasts	• Human	• A cocktail composed of PDGF, FGFs and BIO [58] • Alginate-poly-L-lysine-alginate microencapsulation [60]
	• Mice	• Small molecule: TTNPB [59]

Considering the tendency of DPCs to aggregate, three-dimensional culture has been proposed to revive the trichogenic phenotype and foster subsequent folliculogenesis [50, 54, 55]. Christiano Team reported that approximately 22% of transcripts perturbed by planar culture were regained in spherical culture [50]. Lef-1 overexpression further restored hair lineage signatures, leading to a 70% rate of HF formation in vitro [56]. Furthermore, Wu Team introduced immune regulation into the inducibility-preserving strategies of DPCs [57]. Calcium molybdate nanoparticles loaded with DPCs and macrophages not only supported the survival of DPCs, but also facilitated hair regrowth in vivo by creating an anti-inflammatory microenvironment [57].

Additionally, transforming dermal fibroblasts into DP-like cells represents an alternative approach (Table 2). Studies have shown that a cocktail composed of PDGF, FGFs and BIO [58], the small molecule TTNPB [59], and microencapsulated alginate-poly-L-lysine-alginate [60] can stimulate trichogenicity in dermal fibroblasts, as evidenced in patch assays.

The events subsequent to the transplantation of DPCs have not been fully elucidated, especially with respect to the origin of follicular epithelia — whether they stem from preexisting epidermal progenitors or if there is a redefinition of cell fate. Delving into the interactions of transplanted DPCs with native cells is essential for a comprehensive understanding of HF regeneration and degeneration. Enhanced knowledge of these processes will, in turn, contribute to the establishment of HFs that persist in the long term.

Skin organoid induction from pluripotent stem cells

Achievements in stem cell research have facilitated the generation of skin organoids from PSCs (Table 3). PSCs, including induced pluripotent stem cells (iPSCs) and embryonic stem cells (ESCs), have the potential to differentiate into distinct cells and tissue patterns under particular cues. The initial construction of skin equivalents involves two main steps: first, guiding PSCs to differentiate into keratinocytes and fibroblasts, the predominant cell types in the skin, and second, stacking these cells into a bilayer structure [61, 62]. These structures serve as basic frameworks lacking HFs or sebaceous glands. As our understanding on morphogenesis deepens, there is optimism about generating authentic skin substitutes through the recapitulation of key signals.

Table 3.

Strategies for inducing skin organoids from PSCs

Organism	Cell	Induction strategy
• Mice [64]	• Mouse ESCs • Mouse iPSCs generated from embryonic fibroblasts	Sequential application of SB/BMP and LDN/FGF
• Human [67–70]	• Human ESCs: WA25 [67] • Human iPSCs: nciPS02, RC01001-B [68, 69]; LUMCi045-A1 and LUMCi046-A1 [70]	SB/BMP-LDN/FGF strategy
• Human [71]	• Human iPSCs: CMC003 and CMC011	SB/BMP-LDN/FGF/CHIR strategy and ALI culture

Murine skin induction from PSCs

In 2016, Tsuji Team successfully engineered an integumentary system containing HFs and sebaceous glands from murine iPSCs [63]. Embryoid bodies derived from iPSCs were transplanted as clusters, giving rise to integrated skin, HFs with properly arranged niches, and subcutaneous adipose tissue. Wnt10b supplement was proven to enhance the germination and maturation of HFs via EMIs [63]. This study represented a pioneering effort in regenerating skin with HFs using PSCs, albeit heavily dependent on the variables inherent in living organisms.

In 2018, Koehler Lab detailed a stepwise formula to form murine folliculogenic skin from PSCs [64], based on an established protocol for inducing cranial surface ectoderm [65, 66]. In this method, the administration of a TGFβ inhibitor (SB431542) and BMP4 specified surface ectoderm at the outermost region, and the subsequent treatment with FGF and a BMP inhibitor (LDN-193189) committed the pre-placodal fate [64]. This skin organoid surprisingly mimicked the differentiation aligning to embryonic development. Derived HFs shared critical features with natural HFs in embryonic stages 7–8, containing αSMA⁺ dermal sheath, KRT5⁺ p63⁺ outer root sheath, GATA3⁺ inner root sheath, Ki67⁺ p63⁺ matrix, and AE13⁺ shaft. However, these HFs exhibited limitations in long-term culture, as the matrix deteriorated around day 32 and the shaft failed to shed physiologically, impeding the transition into the next cycle [64].

Human skin induction from PSCs

In 2020, Koehler Lab extended their strategy to human skin regeneration by refining the timing of LDN/FGF administration to day 3 of differentiation (Fig. 3) [67]. This adjustment consistently led to the development of epithelial cysts enveloped by cranial neural crest cells (CNCCs). Hair germs emerged on ~ day 70, aligning with the organogenesis process during gestation. After more than 100 days in culture, these organoids were comparable to 18-week foetal skin, featuring a stratified epidermis, a fat-rich dermis and pigmented HFs. On day 140, the cystic organoids were implanted into nude mice, and planar skin, an extended vasculature and pilosebaceous units were successfully established. These results signify the nearly complete reconstruction of human skin through self-assembly. Moreover, this methodology has shown considerable reproducibility in subsequent studies [68–70] and has come to the forefront in disease modeling [68, 69].

Fig. 3.

Strategies for inducing skin organoids from PSCs. (a) In the classical protocol, the use of a TGFβ inhibitor (SB431542) and BMP4 facilitates the differentiation of surface ectoderm, followed by the administration of FGF and a BMP inhibitor (LDN-193189) to prompt the formation of CNCCs. This process results in the development of enclosed cysts with an inside-out morphology, which matures into planar skin following transplantation. (b) A modified approach incorporates a Wnt activator (CHIR99021) and ALI culture, transforming the enclosed aggregates into larger, open skin organoids.

Transforming enclosed organoids into an open skin model

An apparent defect of these organoids is their inside-out morphology, in which epidermal cells are encircled by dermal components [67]. The aberrant structure imposes a maximum culture duration of 150 days; otherwise, squamous cells would accumulate in the core. To rectify the inverted conformation, Jung and colleagues introduced Wnt activation and air-liquid interface (ALI) culture into the SB/BMP-LDN/FGF protocol, resulting in enlarged open skin models (Fig. 3) [71]. Nevertheless, the lifespan of these organoids remained approximately 150 days. It is advisable to investigate methods for artificial ripening or prolonged culture so that fully developed structures can be harvested for restoration. On the other hand, determining the opportune stage for grafting may hold greater significance than adjusting the culture formula. While it took 4–5 months to coax PSCs into skin organoids with requisite appendages, single-cell RNA sequencing on day 48 revealed diverse cell populations resembling those in embryonic skin [67]. It is speculated that in situ incubation may operate more efficiently than various cocktails, as it covers biochemical messengers and subtle mechanical forces.

Scalp skin induction to be investigated

Notably, since CNCCs are responsible for the development of facial dermis, the organoids described above recapitulate only facial skin. Scalp HFs may be attained through the inclusion of mesodermal cells, or not. This represents an obstacle to holistic skin reconstruction from PSCs.

Skin substitute bioprinting

3D printing, evolving alongside fabrication technologies, brings a new dimension to HF engineering. Initially, 3D printing was utilized to produce noncellular scaffolds resembling HFs, after which the cells were manually seeded [56, 72]. This process was later simplified with the introduction of bioprinting (Table 4; Fig. 4). Bioprinting is a computer-aided technology allowing the precise deposition of bioinks — biomaterials loaded with cells [73]. Multiple factors, including cell types, biomaterial properties, fabrication modalities, and post-printing maturation, need to be considered when printing target tissues [74]. While primary cells and commercial cell lines are commonly used, the application of stem cells is restricted owing to their sensitivity to mechanical shear and viscosity [75, 76]. Biomaterials, as carriers and supporters of living cells, are expected to exhibit flexibility, biocompatibility, and controllable biodegradability [77, 78]. Polymers widely used in bioprinting include collagen, hyaluronic acid, alginate, agarose, chitosan and fibrin, depending on the desired pattern [79, 80]. An appropriate platform is then employed to accurately position bioinks according to the predetermined geometry. Finally, post-printing procedures, such as ALI and bioreactor processing, help to mature and functionalize these constructs [74, 81].

Table 4.

Cells and biomaterials used in bioprinting

Product	Cell	Cell origin	Biomaterial	Modality	Deposition manner	Position
HF-bearing skin	Fibroblasts, HUVECs, DPCs, epithelial cells [84]	• Neonatal mice: fibroblasts, epithelial cells • 4 to 6-week-old mice: DPCs • Commercial: HUVECs	Gelatin alginate hydrogel	EBB	layer-by-layer	in vitro
HF-bearing skin	Fibroblasts, HUVECs, DPCs, keratinocytes, melanocytes [86]	• Neonates: fibroblasts, keratinocytes, melanocytes • Commercial: human DPCs, HUVECs	Collagen I, dermatan sulfate, collagen IV	EBB	layer-by-layer	in vitro
HF-bearing skin	Epidermal stem cells, skin-derived precursors [87]	Neonatal mice	Matrigel	EBB	one-step	in situ
HF-bearing skin	Epidermal stem cells, skin-derived precursors [88]	Neonatal mice	Gelatin methacrylate hydrogel	EBB	one-step	in situ
HFGs	Epithelial cells, mesenchymal cells [96]	Embryonic mice	Collagen IA	DBB	/	in vitro
DPC spheres	DPCs, macrophages [57]	• Human: DPCs • Mice: macrophages	Calcium molybdate	EBB	/	in vitro

Fig. 4.

Application of bioprinting in hair-bearing skin regeneration. (a) Layer-by-layer deposition of fibroblasts, DPC spheres, and epithelial cells generates skin substitutes with trichogenicity. (b) Robot-assisted bioprinting utilizing precursors from neonatal mice enables on-site HF-inclusive skin repair.

Layer-by-layer bioprinting

Bioprinting involves four main modalities: droplet-based bioprinting (DBB), extrusion-based bioprinting (EBB), laser-assisted bioprinting (LAB), and stereolithography-based bioprinting (SBB) [82, 83]. In the realm of skin constructs, EBB is the preferred method due to its versatility [75, 77]. Miao Lab managed to engineer murine skin capable of HF regeneration through EBB [84]. During printing, cells encapsulated in gelatin-alginate hydrogels were deposited in stratified layers, with fibroblasts and human umbilical vein endothelial cells (HUVECs) forming the dermis, DPCs dot-printed into the middle stratum, and epithelial cells comprising the epidermis [84]. These scaffolds enabled the self-aggregation of DPCs and the initiation of EMIs in vitro. After 7 days of culture, DPCs and epithelial cells generated HF-like elongate structures, leading to substantial hair growth in an appropriate orientation following transplantation [84]. Notably, the proper inclusion of DPCs is decisive in HF formation, as direct deposition without pro-aggravating measures would result in the loss of trichogenic properties [85].

In a recent study by Karande Team, HFs were successfully incorporated into human skin constructs via EBB [86]. Spheroids consisting of DPCs and HUVECs were bioprinted within a pre-gelled dermal layer containing fibroblasts. Subsequently, type IV collagen was deposited to establish the dermal-epidermal junction, and keratinocytes were paved to generate the epidermis. Within 48 h of spheroid printing, HF-like columns extended up to the epidermal layer [86]. The resulting HFs resembled natural tissues, with a CK14⁺ outer root sheath and a CK10⁺ inner root sheath surrounding a dense core akin to the DP. Markers associated with DP inducibility, such as versican and alkaline phosphatase, were also detected. However, the model was not transplanted to evaluate HF sprouting or cycling capabilities in vivo [86].

In situ bioprinting

The studies outlined above are grounded on the transplantation of bioprinted architectures. Remarkably, a bioprinting robot has realized in situ regeneration of HF-equipped skin, utilizing skin-derived precursors from neonatal mice [87, 88]. This innovative approach allows one-step printing onto skin defects, in contrast to conventional layer-by-layer deposition and subsequent transplantation. Matrigel [87] and gelatin methacrylate [88] were identified as competent substrates supporting the survival and differentiation of these sensitive progenitors. In addition, intraoperative bioprinting of human adipose-derived extracellular matrix and stem cells led to the formation of HF-like extensions, suggesting the involvement of adipocytes in matrix shaping and downgrowth formation [89]. This discovery promisingly sparks interest in incorporating adipocytes for HF regeneration.

Challenges in tailoring bioinks

Throughout these studies, we can see that reestablishing HFs by bioprinting shares the same tenet with organ germs, that is the inducibility restoration of DPCs and the fate commitment of progenitors. While reprogramming strategies have been investigated to recover the trichogenic potential of DPCs, challenges persist in accessing DPCs and other primary cells owing to donor variations and massive clinical demand. Millions of cells are required to achieve a physiological density (greater than 10⁷ cells/ml of bioink), making cell isolation and expansion an individualized and labor-intensive task [90]. Such inefficiency has prompted the exploration of alternative cell sources. Heartened by advancements in xenotransplantation, it is postulated that xenogeneic cells devoid of immunogenicity could serve as universal ingredients in bioinks [91–93]. The incorporation of genetically modified cells, for instance, porcine cells, holds promise for streamlining bioprinting processes and realizing immediate repair.

Another limitation of bioprinting is the restricted cell density permissible in bioinks to retain printability [94]. It is observed that bioprinted structures frequently undergo shrinkage while maintaining a relatively consistent conformation [75, 95]. The unexpected compaction necessitates adjustments to predefined parameters, but it also enables cell enrichment. Leveraging this idea, Fukuda Lab bioprinted paired collagen droplets containing mesenchymal and epithelial cells in close proximity, which allowed a spontaneous contraction through cell traction forces [96, 97]. This approach led to HFG-like constructs with cells enriched over tenfold and enhanced trichogenic potential in vivo. Consequently, the inherent shrinkage offers a promising solution to condense cells for future bioprinting projects.

Discussion

Organ loss and dysfunction caused by disease and trauma drives the development of regenerative therapies [98–100]. With the rising prevalence of hair loss, HF reestablishment has become one of the focal points in regenerative medicine [101, 102]. A genuine follicle is characterized by coordinative niches, including the distal papilla, bulged stem cell reservoir, and active transit-amplifying cells [18, 103]. Engineered HFs are expected to cycle stably, sprouting new hair shafts while extruding old ones, and stay synchronized with the native pattern. Current methods for HF regeneration are comprised of organ germs, PSC-derived organoids, and bioprinting, in which bioprinting is a facilitative instrument rather than a standalone approach [28, 104, 105]. Utilizing these strategies, the reconstruction of essential compartments and in vivo hair growth have been achieved, yet the cyclability throughout the lifespan remains a hurdle to fully functional HF substitutes [34, 64]. At the mechanism level, research efforts are focused primarily on the preparation of HF precursors or analogues, with a limited understanding on the interplay between implants and their surroundings. It is imperative to decipher the signaling crosstalk and cellular dynamics following transplantation. Clarifying whether the components of neogenetic HFs come solely from implants or are partially derived from existing or respecified progenitors in recipients will help find a way to sustainable regeneration.

Apart from unmet cycling goals, there are several intrinsic and technical challenges in HF reconstruction (Table 5). The organ-germ method, while appreciated as an early controllable approach, faces limitations due to its cut-and-paste nature. The progenitors extracted from embryonic or neonatal murine skin undergo transfer to generate hair in another individual [34–36], which is not suitable in clinical settings. Rather than utilizing allogeneic progenitors, revitalizing autologous cells holds potential for regenerating HFs in human. The restoration of trichogenicity in mature cells has been achieved through environmental signals like VEGF and PKC inhibition, although the durability of resulting HFs remains unexplored [47, 48]. Moreover, iPSCs are considered as a promising source for generating HF-equipped skin. iPSCs can be derived from various cell types, such as dermal fibroblasts, peripheral blood mononuclear cells, hepatocytes, pancreatic beta cells, and neural stem cells, offering benefits in terms of accessibility, expandability, and ethical compatibility [106]. Despite these advantages, several issues need to be addressed prior to their clinical application. Human iPSCs are “primed” and heterogeneous in differentiation potential, tumorigenicity and genome instability [107, 108]. These variations necessitate rigorous characterization and thorough selection during iPSC expansion and maturation, making customizing organoids extremely costly and time-consuming [109]. In order to maximize the utility of iPSCs, a practical strategy is to construct universal iPSC lines and to establish standardized differentiation protocols [110]. These iPSC lines, whether allogenic or xenogeneic, can be modified using CRISPR or other genome-editing tools to prevent immune rejection [111–113]. By producing standardized skin organoids, safety concerns can be minimized, and a timely clinical supply can be ensured. In addition, the accurate specification of iPSCs into epithelial and mesenchymal progenitors opens up the possibility of HFG transplantation as an alternative and convenient approach. All these prospects hinge on precise reprogramming and genetic editing techniques, requiring the collaborative efforts of researchers.

Table 5.

Summary of advantages and drawbacks of present methods

Method	Advantages	Drawbacks	Potential solutions
Follicle germ assembling	• Easy self-assembling procedures • Short culturing period	• Ethical issues and immune rejection • DPCs tend to lose trichogenicity in vitro.	• Progenitor cells may be specified from iPSCs. • 3D culture and genetic reprogramming helps to reserve the inducibility of DPCs.
Skin organoid induction	• No ethical issues • Abundant access to iPSCs • iPSCs induction can not only generate HFs, but also establish HF-bearing skin.	• Long period to coax PSCs into skin organoids • Concerning safety • Organoids described in current studies only recapitulate facial skin, excluding the induction of scalp skin.	• Optimized environmental conditioning and genetic modifications may induce more competent organoids. • In situ incubation may work more efficiently than chemical cocktails. • Scalp HFs may be attained through the induction of mesodermal cells.
Bioprinting	• Automated • High throughput • Generating HF-bearing skin • Realizing in situ restoration	• Primary cell sources restrain the standardized manufacture. • Limited cell density in the printing process	• The development of universal cells holds promise for standardizing a bioprinting streamline and achieving an immediate repair. • The spontaneous contraction may be utilized to achieve the desired density.

Bioprinting technology emerges as an avenue to fabricating organ-mimicking structures with high resolution [114]. Current studies concerning HF bioprinting are guided by the principle of EMI restoration. Both DPC-based and progenitor-based bioinks have succeeded in integrating HFs into bioengineered skin [84, 86–88]. Recent employment of robotic systems has ushered in new opportunities for expeditious treatment using universal bioinks [87, 88], which comes down to optimizing cell sourcing and biomaterial selection. The biomaterials commonly used for printing iPSC-derived cells include Matrigel, alginate, agarose, gelatin methacryloyl, and nano-fibrillated cellulose [115]. Matrigel, derived from EHS mouse sarcoma cells, simulates the natural extracellular environment and is frequently utilized as a substrate for iPSC proliferation and differentiation [116]. Nevertheless, its heterologous nature and undefined chemical composition have raised safety concerns and experimental variations. Biomaterials such as hyaluronic acid and vitronectin also demonstrate the capacity to support iPSC culture, whereas further modifications are needed to align them with the bioactivity and formability requirements in bioprinting. Additionally, engineering improvements are necessary to alleviate thermal and mechanical stresses during printing, and thus to guarantee high viability and functionality of constructs. Apart from these technical constraints, a critical issue that tends to be overlooked in bioprinting and HFG studies is the cellular dynamics and niche reconstruction following transplantation. Although the comprehension of these cellular and molecular events is currently inadequate, novel technologies like live imaging and tracing present an accessible and efficient platform to visualize and illuminate these processes [117, 118].

Conclusions

In summary, although notable progress has been made in HF regeneration, it remains a work-in-progress. Future investigations should emphasize the regenerative nature and produce HFs from viable sources to propel organ reconstruction beyond basic science and ultimately benefit patients. Understanding the dynamics of regenerated HFs and their interplay with surroundings may offer solutions to the constrained lifespan of HFs. With deeper insights into the underlying mechanism, we will be able to define authentic HF reconstruction, normalize engineering procedures, and thus update therapeutics for alopecia and cutaneous defects. More importantly, achievements in HF neogenesis will encourage de novo organogenesis on a greater scale.

Abbreviations

ALI

Air-liquid interface

BMP

Bone morphogenetic protein

CNCC

Cranial neural crest cell

DBB

Droplet-based bioprinting

DDK2

Dickkopf WNT signaling pathway inhibitor 2

DP

Dermal papilla

DPC

Dermal papilla cell

EBB

Extrusion-based bioprinting

EDA

Ectodysplasin A

EDAR

Ectodysplasin A receptor

EDN1

Endothelin 1

EMI

Epithelial-mesenchymal interaction

ESC

Embryonic stem cell

FGF

Fibroblast growth factor

HF

Hair follicle

HFG

Hair follicle germ

HFSC

Hair follicle stem cell

HUVEC

Human umbilical vein endothelial cell

iPSC

Induced pluripotent stem cell

LAB

Laser-assisted bioprinting

ORS

Outer root sheath

OSM

Oncostatin M

PDGFA

Platelet-derived growth factor subunit A

PSC

Pluripotent stem cell

SBB

Stereolithography-based bioprinting

SHH

Sonic hedgehog signaling molecule

TAC

Transit-amplifying cell

TGF

Transforming growth factor

TRPV1

Transient receptor potential cation channel subfamily V member 1

Author contributions

XC and ZZ conceptualized and wrote the manuscript. XQ performed the literature search. XC, HS and HC revised the manuscript. JZ supervised the writing process and reviewed the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Key Discipline Project of Hangzhou (grant number 0020200044), and Medicine and Health Research Project of Zhejiang Province (grant number 2025KY1037).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.