The role of lipids in promoting hair growth through HIF-1 signaling pathway

Abstract

Understanding the underlying mechanisms regulating hair regeneration is crucial, especially given the increasing demand for effective drugs to treat hair loss, which remain not fully elucidated. In the present study, we found that lipid metabolism was attenuated in the scalp tissues of patients with androgenetic alopecia. Lipid supplementation in the culture medium upregulated hair growth-related genes and promoted the proliferation of human dermal papilla cells (DPCs). By using RNA-sequencing analysis and HIF-1α knockdown in DPCs, we found that HIF-1α is a potential candidate that governs lipid-reinforced upregulation of trichogenic genes. Finally, we assessed the hair growth-promoting effects of lipids using in vitro hair follicle organoids and found that lipids accelerated the elongation of hair-shaft-like structures. Our results highlight the importance of lipids in promoting hair growth through HIF-1 signaling, suggesting that this may be a promising target for the treatment of hair loss.

Introduction

Cells maintain homeostasis by regulating complex metabolic reactions, mainly through glucose, amino acid, and lipid metabolism. Recently, lipids have emerged as key players because of their diverse roles as energy sources, building blocks for cell components, and signaling molecules^1–3. Thus, fine-tuning of lipid metabolism may be indispensable for sustaining the dynamic equilibrium between cells and tissues in a continually changing environment. The hair follicle, which is one of the most dynamic tissues in the human body, undergoes loss and regeneration even after birth in the hair cycle, which comprises the growing (anagen), regression (catagen), and quiescent (telogen) phases. Histological analysis of hair follicles have shown that lipid-filled mature adipocytes surround the hair follicle during the transition from the early anagen to anagen phase, while the number of adipocytes decreases as the hair follicle regresses^4–6. Given the synchrony between dermal adipocytes and hair follicles, the detailed mechanism of their regulation could be a promising therapeutic target for hair loss; however, it remains poorly understood.

To elucidate the link between hair follicle generation and lipid metabolism, dermal papilla cells (DPCs), which are specialized mesenchymal cells in hair follicles, have been investigated because of their role in controlling hair follicle morphogenesis and cyclic hair growth by modulating signaling pathways⁷. Several potent pathways that regulate hair growth have been reported, including Wnt and sonic hedgehog (Shh)⁸. However, few reports on abnormal metabolic reprogramming and metabolic pathways have been published despite their potential to modulate hair growth. Choi et al. reported that glucose metabolism via histone acetylation is essential for the expression of hair-inducing genes in DPCs⁹. In particular, DPCs treated with low glucose, or a glycolysis inhibitor showed a decrease in trichogenic gene expression. In the view of lipid metabolism in DPCs, linoleic acid reportedly increases the cell proliferation rate of DPCs and the mRNA expression levels of hair growth-related genes¹⁰. Moreover, disruption of genes involved in lipid metabolism, such as elongation of very long-chain fatty acids 3 (Elovl3) and fatty acid transport protein 4 (Fatp4), results in abnormal hair follicle morphology and defects in hair growth^11,12. In humans with androgenic alopecia (AGA), an analysis of sebum revealed that the composition of free fatty acids (FAs) was lower in the AGA group than in the control group, indicating a positive correlation between hair growth and free FAs¹³.

Considering the growing interest in the treatment of hair loss and the elucidation of lipid metabolism in hair follicles, this study investigated the general downregulation of fatty acid metabolism in the scalp tissues of patients with AGA using informatics analysis. Interestingly, treatment with lipids accelerated the growth rate and hair growth-related gene expression in DPCs. Using gene chip analysis, the involvement of HIF-1α in regulating lipid-treated DPCs was suggested, and the lipid-dependent upregulation of trichogenic gene levels was further confirmed with HIF-1α knockdown (KD) DPCs. Finally, the hair-promoting effects of these lipids were investigated using hair follicle organoids.

Results

Downregulation of fatty acid metabolism in the human scalp tissues of patients with androgenetic alopecia

To investigate the alterations in lipid metabolism in the scalp tissues of patients with AGA, we conducted gene set enrichment analysis (GSEA). The results revealed that the enrichment of fatty acid metabolism-related genes was significantly attenuated in AGA scalp tissues compared with healthy controls (Fig. 1a and b, along with Supplementary tables 1 and 2). Given the fact that the scalp tissues of patients with AGA are deficient in hair growth and regeneration capabilities^14,15, these results suggest that lipid metabolism may be positively related to hair growth and regeneration.

Fig. 1.

Fatty acid metabolism is disrupted in scalp tissues of individuals with AGA. (a) Scalp tissues from individuals with AGA (n = 14) and healthy controls (n = 14) were obtained from GEO database (GSE90594) were utilized for gene set enrichment analysis (GSEA). The GSEA plot shows fatty acid metabolism-related gene signatures that are downregulated in AGA scalp tissues. (b) A heat map depicting the expression levels of core enriched genes within Hallmark fatty acid metabolism gene set.

Enhancement of trichogenic gene expression and cell growth in DPCs by lipids

As a positive correlation between lipids and hair growth was demonstrated (Fig. 1), we explored the effect of lipids in human DPCs. A lipid mixture (LM) was added to the DPC culture and enhanced the expression of trichogenic genes, including alkaline phosphatase (ALP), versican (VCAN), and lymphoid enhancer binding factor 1 (LEF1) in DPCs (Fig. 2a). Furthermore, LM supplementation activated the proliferation of DPCs (Fig. 2b and c). The protein levels of Ki-67 along with the gene expression levels of cell proliferation markers (MKI67, CCNA, CCND, and CCNE) and fatty acid metabolism-related genes (ACADS, CPT2, PDHA1, ECI1, PPARA, and ACAT2) were upregulated in LM-supplemented DPCs (Fig. 2d and e, Supplementary Fig. 1 and 2). These results suggest that lipid supplementation is closely associated with DPC activation and the subsequent promotion of hair growth.

Fig. 2.

Fatty acids are involved in improving the function and growth of DPCs. (a) DPCs were treated with 1% or 2% of lipid mixture (LM) for 24 h and lysed for RT-qPCR to determine ALP, VCAN, and LEF1 mRNA levels. Relative mRNA levels are presented as the means ± SD (n = 3) with one-way ANOVA. (b and c) DPCs were treated 2% of LM for 24 h. Phase-contrast microscopy images of LM treated DPCs are presented (b; scale bar, 500 μm), and the relative cell numbers were calculated relative to the values at 0 h (c). (d) DPCs were treated with 2% of LM for 24 h and subsequently harvested for Western blotting. Signal quantification was performed using Image J, normalized to GAPDH expression levels (mean ± SD, n = 3). (e) 2% LM-supplemented DPCs were subjected to qRT-PCR, and the relative gene expression levels of cell proliferation markers are presented as the means ± SD (n = 3).

Dependence of lipid-mediated upregulation of trichogenic gene levels on HIF-1α expression levels

To identify the molecular pathways associated with the effects of lipids on DPCs, microarray was performed in the presence and absence of LM supplementation. The analysis revealed that LM supplementation upregulated 2,329 genes, which were used for further Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis (Fig. 3a). We found that the PI3K-Akt and Ras signaling pathways were significantly upregulated by LM supplementation (Fig. 3b). As these pathways are known to regulate HIF-1α protein translation, we subsequently examined HIF-1α protein levels in LM-treated DPCs. Interestingly, LM supplementation elevated expression levels of HIF-1α in a dose-dependent manner (Fig. 3c, Supplementary Fig. 3). Furthermore, trichogenic gene expression upregulated by LM supplementation was completely reversed when HIF-1α was silenced with siRNA (Fig. 3d). Taken together, the lipid supplement in DPCs is closely linked to trichogenic gene expression via the HIF-1α pathway.

Fig. 3.

The upregulation of trichogenic gene levels mediated by lipids is dependent on HIF-1α expression levels. (a) Microarray analysis was conducted on LM treated DPCs. Differentially expressed genes between the groups were selected and are presented as a heatmap (threshold of Log2FC > 1). (b) The list of top enriched KEGG pathways identified genes that were upregulated in LM-treated DPCs. (c) DPCs were incubated with LM at the indicated concentrations for 24 h and subjected to western blotting. Signal quantification was performed based on GAPDH expression levels using Image J (mean ± SD, n = 3) with one-way ANOVA. (d) si-Con or si-HIF-1α transfected DPCs were treated with 2% LM for 24 h and analyzed using RT-qPCR. Relative expression levels of ALP and VCAN are presented as the means±SD (n = 3) with one-way ANOVA.

Lipids promote hair shaft sprouting in hair follicle organoids

In our previous study, hair follicle organoids prepared using human hair follicle-derived epithelial cells and DPCs showed aggregate formation and subsequent hair shaft-like sprouting filled with keratin proteins¹⁶. The elongation rate of hair shaft sprouting in organoids was accelerated by supplementation with minoxidil and other substances in the culture medium, which was equivalent to the results of hair follicle organ culture ^16–20. In this study, we examined whether LM supplementation accelerates hair shaft sprouting in organoids because lipids affect cell proliferation and trichogenic gene expression in DPCs. The LM supplementation significantly promoted the elongation of hair shaft-like structures, compared to the control without LM (Fig. 4a and b). These results suggest that lipid modulation in DPCs promotes epithelial cell-mediated hair shaft formation, providing a new perspective for the treatment of alopecia (Fig. 5).

Fig. 4.

The sprouting of hair shaft-like structures in hair follicle organoids is reinforced by lipid treatment. (a and b) Representative microscope images of hair follicle organoids treated with 2% of LM (or vehicle) for 10 days. The length ratios on days 6, 8, and 10, compared to that on day 4, are shown in the bar graph with means ± SEM (n = 56, bar: 500 μm).

Fig. 5.

A schematic representation of the experimental procedure and the mechanism proposed in this study. The mechanism by which the lipid supplement enhances the trichogenic potential and promotes hair growth is through upregulating trichogenic gene expression through the activation of the HIF-1α signaling pathway.

Discussion

Abnormalities of lipid metabolism have been reported to be associated with various diseases, including obesity, metabolic syndrome, cancer, and cardiovascular disease^21,22. Recently, the lipid content in the sebum of patients with AGA revealed that abnormal lipid content is involved in hair loss¹³. Moreover, a study conducted in the 1950s suggested that lipids are essential for hair follicle regeneration and the hair cycle. During the transition from telogen to anagen, the number and size of lipid-filled mature adipocytes increase via the activation of immature precursor adipocytes, whereas they decrease during the transition from anagen to catagen. Considering the positive correlation between the hair cycle and lipid-filled adipocytes, the detailed mechanisms need to be clarification but remain elusive^23,24. Adipocytes store fat as triglycerides (TGs) and release it in the form of FAs through sequential lipolysis²⁵. Despite the complex and essential activities of FAs being considered important, little is known about their role in hair growth. To the best of our knowledge, one study revealed that linoleic acid treatment results in cell growth of DPCs through Wnt/B-catenin signaling pathway activation¹⁰. However, FAs that mainly exist in scalp tissues are palmitic acid, palmitoleic acid, myristic acid, oleic acid, pentadecaneoic acid, stearic acid, and others, in order of composition in sebum from the scalps¹³. Among the types of FAs, it is important to study those that are abundantly present in scalp tissues to determine their effects on hair growth. In this study, we utilized commercially available, chemically defined LM, including palmitic acid, myristic acid, oleic acid, and stearic acid, to enhance the hair inductivity of hair follicle organoids, thereby mimicking the dermal microenvironment with less lipotoxicity. Treatment with LM increased the expression levels of trichogenic genes along with the growth of DPCs and elongated the hair shafts sprouting from organoids. Although a comparative analysis of free fatty acid components between non-AGA and AGA subjects could provide further evidence supporting that the dermal microenvironment influences hair growth, this was not feasible in the present study. Nevertheless, we hypothesize that a specific free fatty acid, oleic acid, could play a role in promoting hair growth, as oleic acid has been identified as a HIF-1α activator in hepatocellular carcinoma²⁶. We intend to investigate this hypothesis in detail in subsequent studies.

LM is reported as a HIF-1α activator in various cancer types, increasing the rates of tumor metastasis through LM/HIF-1α axis²⁷. Based on the RNA-sequencing data in Fig. 4a and b, we found that LM-treated DPCs also significantly exert PI3K-Akt and Ras signaling pathways, which are well known for regulating HIF-1α translation^28,29. Thus, we checked the enhanced HIF-1α protein levels in LM-treated DPCs and found that LM promotes trichogenic gene levels in an HIF-1α-dependent manner. Numerous studies have suggested that FAs are an energy source for cells to meet their energy demand³⁰. As evidenced in previous studies, FAs serve as an energy source for cellular proliferation and act as a trigger for HIF-1α signaling in DPCs. Although we could not explain the detailed metabolic changes via HIF-1α signaling pathways, we suggest that HIF-1α mediates the reprogramming of FA metabolism, as discovered in cancer³.

In summary, our results indicate that lipids trigger hair inductivity in DPCs and elongate the hair shaft-like structures of hair follicle organoids. HIF-1α-mediated trichogenic gene activation via lipid treatment could be a relevant therapeutic target for treating hair loss, and the supplementation of lipids could be applied in vitro preparation of hair follicles with shafts for hair transplantation.

Methods

Cultivation of dermal papilla cells (DPCs)

Human DPCs were purchased from PromoCell (Heidelberg, Germany) and cultured in follicle-derived dermal papilla cell growth medium (DPCGM; PromoCell, Heidelberg, Germany). DPCs at passages 3 to 5 were used in the in vitro studies, maintained at a gas tension of 21% O₂ and 5% CO₂ at 37 °C.

Treatment of lipid mixture and short interfering RNAs (siRNAs)

The lipid mixture (lipid mixture 1; #L0288) was purchased from Sigma-Aldrich (St. Louis, MO, USA). The sequences targeting HIF-1α and the control were synthesized as described in a previous study³¹ (Integrated DNA Technologies Japan, Japan).

Western blotting

Cultured cells were lysed using RIPA lysis buffer (EzRIPA Lysis Kit; ATTO, Japan), incubated at 4 °C for 15 min, and centrifuged for 10 min at 14,000 × g. The supernatant was mixed with 2 × sodium dodecyl sulfate (SDS) sample buffer. Proteins were separated on sodium dodecyl sulfate–polyacrylamide gels (Bio-Rad, Hercules, CA, USA) and sequentially transferred to Immobilon-P membranes (Merck KGaA). The membranes were blocked with 3% BSA dissolved in Tris-buffered saline containing 0.1% Tween-20 (TBST) solution for 1 h, cut with the adequate size for inhibiting non-specific detection, and incubated with primary antibodies (HIF-1α and Ki-67, 1:1000 dilution, Abcam, UK; GAPDH, 1:2000 dilution, Cell Signaling Technology, USA) at 4 °C overnight. The membranes were incubated with a horseradish peroxidase-conjugated secondary antibody (1:5000 dilution; Cell Signaling Technology, USA) for 1 h at room temperature and visualized using chemiluminescent reagents (ECL Prime; GE Healthcare, UK) with FUSION SOLO.7S.EDGE (Vilber Bio Imaging, France).

Quantitative reverse transcription polymerase chain reaction (RT-qPCR)

To analyze gene expression levels, total RNA was extracted from cultured cells using an RNeasy Mini Kit (Qiagen, Germany). cDNA was synthesized using a ReverTra Ace qPCR RT kit (TOYOBO, Japan), and quantitative real-time PCR was performed using a StepOne Plus RT-PCR system (Applied Biosystems, USA) with SYBR Premix EX Taq II (Takara Bio, Japan). The primer sequences used for qRT-PCR are listed in supplementary table 3. All mRNA values of the target genes were normalized to the GAPDH expression level.

Lipid treatment of hair follicloids

To prepare hair follicloids, DPCs (5 × 10³ cells) and epithelial cells (5 × 10³ cells, adult human follicular keratinocytes, Scientific Cell Research Laboratories, USA) were suspended in 0.2 mL advanced Dulbecco’s modified Eagle’s medium/nutrient mixture F-12 (DMEM/F-12; Thermo Fisher Scientific, USA) containing 2% (v/v) Matrigel (Corning Inc., USA) and seeded into a non-cell-adhesive round-bottom 96-well plate (Primesurface® 96U plate; Sumitomo Bakelite Co., Ltd., Japan). DPCs were pretreated with 2% LM before organoid culture and cultured for 4–10 days. Half of the spent medium per well (0.1 mL) was replaced with Advanced DMEM/F12 without Matrigel every two days. The hair length was measured using an all-in-one fluorescence microscope (BZ-X810; Keyence, Japan).

Informatics analysis

Gene expression in scalp tissues with AGA was analyzed using the public NCBI Gene Expression Omnibus (GEO) dataset. Gene set enrichment analysis (https://www.gsea-msigdb.org/gsea/index.jsp) was performed to analyze the enriched genes in the dataset.

Microarray analysis

Total RNA was extracted from DPCs using the RNeasy Mini Kit (Qiagen, Germany) according to the manufacturer’s instructions. Gene expression profiling was conducted using GeneChip Human Genome U133 Plus 2.0 arrays (Thermo Fisher Scientific, USA), and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was performed using the R package clusterProfiler. The upregulated genes of DPCs in accordance with LM treatment were selected with a threshold of Log₂FC > 1 and analyzed using the Database for Annotation, Visualization, and Integrated Discovery (http://david.abcc.ncifcrf.gov/).

Statistical analysis

All data are expressed as means and standard deviations. A two-tailed paired Student’s t-test or one-way ANOVA was performed to compare the differences between groups using GraphPad Prism 8 software. Statistical significance was defined as * P < 0.05, ** P < 0.0001.

Author contributions

J.S.: Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Writing-Original Draft, Visualization, Funding Acquistion; K.M.: Validation, Formal Analaysis, Writing-Original Draft; A.N. and S.T.; Investigation, Resources; D.W.J. and Y.S.C.: Conceptualization, Resources, L.Y. and T.K.: Investigation, Resources, Data curation; J.F.: Conceptualization, Writing-Reviewing & Editing, Supervision, Project Administration, Funding Acquisition.

Data availability

All Data analyzed in this study are included within the manuscript.

Declarations

Competing interests

The authors declare no competing interests.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-88697-8.