Tranexamic Acid Ameliorates Skin Hyperpigmentation by Downregulating Endothelin-1 Expression in Dermal Microvascular Endothelial Cells

Abstract

Background

Although reports suggest that tranexamic acid (TXA) has clinical benefits for melasma patients by oral, intralesional and topical treatment, the optimal route of TXA therapy and the underlying mechanism involved remain poorly defined.

Objective

To compare the skin lightening effect between oral TXA and topical TXA and to dissect the molecular mechanisms using ultraviolet B (UVB)-induced hyperpigmentation mouse model, ex vivo cultured human skin explant, and cultured melanocytes (MCs) and endothelial cells.

Methods

Melanin content and cluster of differentiation 31 (CD31)-positive cell numbers were measured in tail skins from UVB-irradiated mice treated by intragastral or topical TXA using immunofluorescent and Fontana-Masson staining. The conditioned medium (CM) was harvested from human umbilical vein endothelial cells treated with or without 3 mM TXA and was used to treat MCs for 48 hours. mRNA and protein levels of tyrosinase and microphthalmia-associated transcription factor were measured using quantitative real-time reverse transcription polymerase chain reaction and western blotting assays. HMB45- and CD31-positive cell numbers as well as melanin content were also examined in ex vivo cultured human skin explants.

Results

The hyperpigmented phenotype were significantly mitigated in UVB-irradiated tail skin plus intragastral TXA-treated mice compared with mice treated with UVB only or with UVB plus topical TXA. CD31-positive cell numbers correlated with the anti-melanogenic activity of TXA therapy. The data from cultured cells and skin tissues showed that suppression of endothelin-1 (ET-1) in vascular endothelial cells by TXA reduced melanogenesis and MC proliferation.

Conclusion

Oral TXA outperforms topical TXA treatment in skin lightening, which contributes to suppression of ET-1 in dermal microvascular endothelial cells by TXA.

INTRODUCTION

Melasma (chloasma) is a highly prevalent and cosmetically disfiguring hyperpigmented skin disease that mainly manifests as bilateral irregular brown macules and patches on sun-exposed areas of the face¹. Clinically, there are three predominant facial patterns commonly seen: centrofacial, malar and mandibular². The incidence of melasma is as high as 30%–40% in darker skinned females with Fitzpatrick skin types III–IV, especially in Asian or Hispanic women who are of childbearing age¹,³. Although none of the symptoms causes discomfort, the distressing cosmetic appearance of melasma can negatively affect the quality of life and can pose a significant psychological burden for affected individuals⁴. The precise mechanisms underlying melasma remain obscure, however, there is well documented evidence showing that sun exposure, genetic predisposition and female sex hormones are the leading predisposing factors for the onset and development of this disorder²,³,⁴,⁵. Histologically, a sustained epidermal hypermelanization is prominently observed in lesional melasma skin that arises from the overproduction of melanin by hyperactive melanocytes (MCs). Information from recent studies points out that melasma is now understood to be a much-referenced photo-aging disorder. Beyond hyperactive MCs, this disease also involves the functional abnormalities of dermal fibroblasts and microvascular endothelial cells as well as mast cell-mediated inflammation in the affected skin⁶,⁷,⁸. Hydroquinone represents a prototypic skin-lightening agent that is used widely to treat melasma through a mechanism that competitively inhibits tyrosinase (TYR), the rate limiting enzyme in melanin synthesis, which results in the suppression of active MCs⁹. Despite the fact that hydroquinone-based formulations are prescribed as a first-line therapy for melasma, recommended by multiple expert consensus and guidelines¹⁰,¹¹, this disease is still a challenging clinical problem to solve with hydroquinone monotherapy because of its chronic rebounding and even drug resistance.

Increasing evidence suggests that tranexamic acid (TXA), a synthetic antifibrinolytic drug, has clinical benefits for melasma patients¹²,¹³. However, the skin lightening activity of TXA therapy varies greatly depending on its route of oral, intralesional and topical administration¹⁴. It has been reported that oral TXA therapy seems to be the most effective, especially in cases of refractory melasma coexisting with subtle telangiectatic erythema¹⁵. The question has arisen as to why TXA therapy is improved by oral administration. In this study, we employed a mouse model of ultraviolet B (UVB)-induced tail skin hyperpigmentation and treated animals with intragastral or topical TXA administration to compare the anti-melanogenic effects between oral TXA versus topical TXA therapy. The results show that oral TXA outperforms topical TXA administration in skin lightening, and reveals a mechanism whereby TXA preferentially suppresses endothelin-1 (ET-1) expression in dermal microvascular endothelial cells and eventually mitigates melanin production by MCs in the epidermis.

MATERIALS AND METHODS

Chemicals and antibodies

All chemical reagents were of analytical grade and were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless indicated otherwise. 4-(aminomethyl)cyclohexane-1-carboxylic acid (tranexamic acid) was commercially supplied by Aladdin Biotechnology (Cat#: A111900, Shanghai, China). Bosentan, an endothelin receptor antagonist for ET-A and ET-B, was purchased from Selleck Chemicals (Cat#: S3051, Houston, TX, USA). The following antibodies were used in this study: anti-ET-1 rabbit polyclonal antibody (Cat#: A0686) was obtained from ABclonal (Shanghai, China), anti-microphthalmia-associated transcription factor (MITF) antibody (Cat#: ab12039) and anti-cluster of differentiation 31 (CD31) antibody (Cat#: ab28364) were purchased from Abcam (Woburn, MA, USA). Secondary antibodies conjugated with Alexa fluor 488 (green) or cy3 (red) or horseradish peroxidase (HRP) were purchased from Servicebio (Wuhan, China).

Animal experiments

The Dct‐LacZ transgenic mice on a C57BL/6J background were generously provided by Dr. Ian Jackson (Queen’s Medical Research Institute, The University of Edinburgh, Edinburgh, UK)¹⁶. Dct-LacZ mice were housed in cages under specific pathogen-free (SPF) conditions at the experimental animal facility of the Renmin Hospital of Wuhan University (WDRM-20221211A). All animal experiments were approved by the Institutional Animal Care and Use Committee of the Renmin Hospital of Wuhan University. The mice were randomly divided into four groups of 5 female mice each: mock control, UVB-irradiated, UVB-irradiated and intragastral TXA-treated, as well as UVB-irradiated and topical 3% TXA-treated groups. For the UVB radiation, the tail skin of each mouse was exposed daily for 1 week to 180 mJ/cm² UVB from the UVB phototherapy device (SS-01; Sigma High-Tech Co., Ltd., Shanghai, China) that produced a light with wavelengths between roughly 290 and 320 nm and a peak emission of 312 nm. The dose of UVB irradiation was calibrated using a digital radiometer (Sigma High-Tech Co., Ltd.) before each exposure. In the intragastric administration group, each mouse was given a TXA suspension at 750 mg/kg each day for 1 month¹⁶. In the topical administration group, the tail skin of each mouse was treated twice a day with 3% TXA cream for 1 month.

Human skin explant culture

The ex vivo model of human skin explant culture was prepared according to previous reports¹⁷,¹⁸. In brief, human sun-exposed eyelid skin specimens were obtained from three healthy female patients who were undergoing upper eyelid blepharoplasty in the Department of Plastic Surgery. Written informed consent was obtained from each participant after full information disclosure prior to participation in the study. This study was approved by the Ethics Committee of the Renmin Hospital of Wuhan University (WDRM2022-D056) in accordance with the Declaration of Helsinki Principle. Circular full-thickness skin explants (5-mm in diameter) were obtained using a Kai sterile disposable biopsy punch (Kai Industries Co. Ltd., Gifu, Japan) and were placed on the bottom of a Transwell insert (epidermis upwards) and then transferred into 24-well plates. The wells were filled gently with Dulbecco's Modified Eagle Medium (DMEM; Sigma-Aldrich) containing 10% fetal bovine serum (FBS) (Hangzhou, China) and antibiotics to ensure that the epidermis was not submerged in the medium and was kept at the air-liquid interface. Tissues were maintained in constant temperature incubators at 37°C and 5% CO₂. For TXA treatment, the fresh culture medium containing 3 mM TXA was changed every 48 hours.

Cell culture, treatment and preparation of conditioned medium (CM)

Immortalized human PIG1 MCs were kindly provided by Dr. Caroline Le Poole (Department of Dermatology, Microbiology and Immunology, Northwestern University at Chicago, IL, USA). All MCs were cultured in M254 medium with the addition of human melanocyte growth supplement (HMGS) (all from Gibco, Invitrogen, Carlsbad, CA, USA), 5% FBS, 100 U/ml streptomycin and 100 U/ml penicillin. MCs were used between passages 2 and 4 for these experiments.

Human umbilical vein endothelial cells (HUVECs) were generously supplied by the Department of Neurosurgery, Wuhan University Zhongnan Hospital. The cells were cultured in DMEM containing 10% FBS. To prepare the CM from HUVECs, the cells were seeded in 6-well plates and were cultured until they reached up to 70% confluence. The cells in fresh medium were treated with or without 3 mM TXA for 48 hours, after which the cell-free supernatants were harvested and further centrifuged and filtered using a 0.22 µm filter to remove cell debris¹⁹. For treatment of HUVEC-derived CM (HUVEC-CM), the MCs were incubated with the media consisting of a mixture of HUVEC-CM and complete M254 (1:1) for 48 hours.

Cell viability assay

A cell counting kit-8 (CCK-8) reagent (Vazyme Biotech Co. Ltd., Nanjing, China) was used to measure cell viability according to the manufacturer’s instructions. In short, HUVECs were seeded at 1×10³ per well in 96-well plates, allowing the cells to attach to the bottom of the wells. The cells were then treated with various concentrations of TXA. After 48 hours, 10 μl CCK-8 reagent was added to each well and incubated in a dark, humid chamber at 37°C for 2 hours. The optical density value was recorded at 450 nm using a microplate reader (PerkinElmer, Waltham, MA, USA).

5-Ethynyl-2′-deoxyuridine (EdU) incorporation assay

EdU is a nucleoside analog to thymidine that is incorporated into DNA during active DNA synthesis. Cell proliferation was measured by testing the incorporation of EdU into DNA. In brief, the MCs were treated with the indicated amount of HUVEC-CM for 48 hours, and then incubated with diluted EdU working solution (Beyotime Biotechnology, Nanjing, China) for 12 hours. Afterwards, the cells were fixed in 4% paraformaldehyde and were then permeabilized with 0.5% Triton X-100, followed by incubation of the Click Reaction Buffer (Beyotime Biotechnology) for 30 minutes. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (Servicebio) for 5 minutes. Cells were visualized using an upright BH2 fluorescent microscope (Olympus, Tokyo, Japan). A minimum of three randomly selected high-power fields were examined per sample to count EdU-positive cells. The experiments were repeated 3 times.

Fontana-Masson melanin staining

The skin tissues were biopsied and fixed in 4% paraformaldehyde. Paraffin sections were processed, and then deparaffinized, hydrated and then placed into a 2.5% silver nitrate solution for 16 hours in a dark chamber at room temperature. After being rinsed in ultrapure water, the slides were placed in 0.2% gold chloride for 10 minutes, rinsed again in ultrapure water and then incubated in 5% sodium thiosulfate at room temperature for an additional 10 minutes. Cell nuclei were counterstained with hematoxylin. Image J software (National Institutes of Health, Bethesda, MD, USA) was used to quantify the gray value of melanin granules.

Immunofluorescent staining

Tissue sections were prepared from formalin-fixed embedded paraffin tissue blocks for routine immunofluorescent staining, as described previously²⁰. The sections were incubated with anti-HMB45 antibody (1:200) and anti-CD31 antibody (1:200) overnight at 4°C, and then incubated with the Cy3/FITC-labeled goat anti-rabbit IgG (H+L) secondary antibody (1:1,000) for 1 h at 37°C. Nuclei were stained using DAPI solution for an additional 10 min at room temperature. Imaging was performed using an FV1200 (Olympus) confocal microscope.

Isolation of total RNA and quantitative real-time reverse transcription polymerase chain reaction (PCR)

The RNA extraction reagent (Servicebio, Wuhan, China) was used to extract total RNA from cultured cells according to the manufacturer’s instructions. The concentration and purity of RNA samples were determined using a NanoDrop ND-1000 spectrophotometer (Thermo Scientific). cDNAs were synthesized from total RNAs using a SweScript All-in-One First-Strand cDNA Synthesis SuperMix Kit (Servicebio).

Quantitative real-time PCR was performed in triplicate with the use of ChamQ Universal SYBR qPCR Master Mix (Q711-02; Vazyme Biotech Co. Ltd.). The reaction mixture for quantitative real-time PCR was 10 μl and included 5 μl 2×ChamQ Universal SYBR qPCR Master Mix, 0.5 μl cDNA, 0.2 μl forward and reverse primers (10 mmol/L) and 4.1 μl double-distilled water. Real-time PCR was carried out using an Applied Biosystems 7500 Fast Real-Time PCR System (Foster City, CA, USA) with cycle parameters as follows: denaturation at 95°C for 5 minutes; 40 cycles of 95°C for 15 seconds, 60°C for 30 seconds. The primers were purchased from Sangon Biotech Co. Ltd. (Shanghai, China). The primer sequences used are as follows: hMITF, forward: 5’-CAAATACGTTGCCTGTCTCGG-3’ and reverse: 5’-TGGCCAGTGCTCTTGCTTCA-3’; hTYR, forward: 5’-GCTATCTACAAGATTCAGACCCAGA-3’ and reverse: 5’-TGACGACACAGCAAGCTCAC-3’; hEDN1, forward: 5’-CCCGTTAAAAGGGCACTTGGG-3’ and reverse: 5’-CTCAGCGCCTAAGACTGCTGTT-3’; hGAPDH, forward: 5’-CAATGAATACGGCTACAGCA-3’ and reverse: 5’-AGGGAGATGCTCAGTGTTGG-3’. The relative expression of each gene was calculated relative to the expression of the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) with the formula 2^−ΔΔCt. Amplification of specific transcripts was confirmed by melting curve profiles at the end of each PCR.

Western blot analysis

The total cell lysate was obtained by lysing cells in RIPA lysate buffer (Servicebio) containing 1× protease inhibitor cocktail (Servicebio). Protein content was determined with a BCA assay kit (Beyotime Biotechnology). Equal amounts of each protein extract (10 µg per lane) were then resolved using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. After transfer to polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA) and blocking with 5% non-fat milk in Tris buffered saline, the membranes were incubated with anti-ET-1antibody, anti-TYR antibody or anti-MITF antibody at a dilution of 1:2,000, overnight at 4°C; GAPDH was used as a loading control. The membranes were then washed and incubated with HRP-conjugated anti-rabbit (Cat#: GB23204; Servicebio) or anti-mouse (Cat#: GB23301; Servicebio) IgG at a dilution of 1:2,000 for 1 hour at room temperature. Each membrane was then washed again. Immunoreactive bands were visualized and detected by enhanced chemiluminescence (Cat#: GB15004; Servicebio). The intensity of each band was quantified using Image J software and was normalized against GAPDH.

Statistical analysis

Data were analyzed in at least three independent experiments and expressed as mean ± standard deviation (SD). All statistical analyses were performed with SPSS version 19.0 software (IBM SPSS, Armonk, NY, USA) and GraphPad Prism version 6.0 (San Diego, CA, USA) software. The two-tailed Student’s t-test was used to compare the means of 2 groups. One-way analysis of variance was used to evaluate differences between 3 or more groups followed by the Tukey post-hoc test. A p<0.05 is considered to be statistically significant.

RESULTS

Treatment with oral TXA outperforms topical TXA to alleviate UVB-induced hyperpigmentation in mouse tail skin

Although there are reports that TXA might alleviate hyperpigmentation in melasma by oral, intralesional and topical administration¹⁴, the optimal TXA delivery approach was not well defined. To address this issue, we established an in vivo model of UVB-induced hyperpigmentation in mouse tail skin, and compared the melanin content and number of CD31-positive cells in tail skin specimens from mice treated by intragastral TXA or by topical TXA. The results showed that melanin content and the pigmented phenotype were significantly mitigated in the tail skin from intragastral TXA-treated mice compared with those treated with UVB alone and with UVB plus topical TXA (Fig. 1A). Meanwhile, immunofluorescent staining analysis of CD31, a specific marker for renascent vascular endothelial cells²¹, expression in mouse skin showed that the number of CD31-positive cells correlated closely with the anti-melanogenic activity of TXA therapy (Fig. 1B). Our results imply that TXA therapy exerts a potent inhibition on both melanin production and neovascularization, especially in the groups treated with intragastric TXA.

Fig. 1. Changes of pigmentation phenotype, melanin content and CD31+ cell number in UVB-exposed mouse tail skin after intragastrical or topical TXA treatment. (A) C57BL/6J background Dct-LacZ transgenic mice (n=20) were randomly divided into 4 groups: mock control, UVB-irradiated, UVB-irradiated and intragastral TXA-treated, as well as UVB-irradiated and topical 3% TXA-treated groups. The protocols of TXA intragastric and topical administration were detailed in the MATERIALS AND METHODS. Representative images of the changes in skin pigmentation examined using a FotoFinder dermoscope (Fotofinder System, Bad Birnbach, Germany) are shown (5×, upper panels). The skin specimens were sectioned for Fontana-Masson staining (middle panels). The neovascularization of mouse tail skin was examined using CD31 immunofluorescence staining. Representative images of CD31 (red) in mouse tail skin are shown (lower panels). Nuclei were stained with DAPI (blue). White arrows indicate the CD31+ microvascular lumen. (B) Skin pigmentation intensity, melanin content and the number of CD31+ cells were analyzed using Image J software (National Institutes of Health, Bethesda, MD, USA) and are shown in the histograms. The data represent means ± standard deviation from 3 independent experiments (scale bars=50 µm).

CD31: cluster of differentiation 31, UVB: ultraviolet B, TXA: tranexamic acid, DAPI: 4′,6-diamidino-2-phenylindole.

^**p<0.01.

TXA inhibits UVB-induced hyperpigmentation by downregulating ET-1 expression in dermal microvascular endothelial cells

The results described above demonstrated that the anti-melanogenic activity of TXA therapy was found to be closely correlated with anti-angiogenesis. We therefore hypothesized that a mechanistic link potentially exists between vascular endothelial cell-derived cytokines and melanogenesis, which could help to explain why oral TXA therapy is more efficacious than topical therapy of TXA to treat melasma. ET-1 secreted by endothelial cells became a target of interest since it activates MCs through binding with its cognate endothelin type B receptor (EDNRB) on the cell surface²². We first examined the cell viability of HUVECs exposed to TXA at varying concentrations (0–100 mM). CCK-8 analysis showed that there were no significant changes in cell viability of HUVECs at TXA concentrations of 10 mM or less but was compromised obviously at concentrations of 30 mM or higher (Fig. 2A). qPCR analysis and western blotting assays showed that the expression levels of ET-1 mRNA and protein were suppressed in a dose-related manner (Fig. 2B). In addition, we harvested the CM from HUVECs treated or untreated with 3 mM TXA, and then treated MCs with HUVEC-CM for 48 hours. The mRNA and protein expression levels of TYR and MITF were significantly decreased in MCs treated with the CM from HUVECs exposed to 3 mM TXA compared with MCs treated with the CM from HUVECs-unexposed TXA (Fig. 2C and D). To further validate the role of endothelial cell-derived ET-1 in UVB-induced melanogenesis, MCs were treated with 2 µM Bosentan (a specific antagonist for endothelin type A and B receptor)²³ and/or were co-treated with the CM from HUVECs exposed to 3 mM TXA in the presence or absence of 30 mJ/cm² UVB. Bosentan significantly abrogated the UVB-induced expression of TYR mRNA and protein in MCs treated with the CM from HUVECs-unexposed to TXA compared with MCs treated with the CM from HUVECs-exposed TXA (Fig. 3). In addition, we also found that Bosentan mitigated cell proliferation, as shown by EdU incorporation assays (Fig. 4). These results demonstrate that ET-1 in the CM from HUVECs may stimulate MC proliferation and upregulate expression of the critical melanin synthesizing enzyme TYR. Thus, the suppression of ET-1 secretion by vascular endothelial cells is implicated in the anti-melanogenic effect of the TXA therapy.

Fig. 2. Suppression of ET-1 in vascular endothelial cells by TXA downregulates melanogenesis in MCs. (A) HUVECs were treated with TXA at varying concentrations (0–100 mM), after which the cell viability was measured using the CCK-8 assay. ns: not significant. (B) HUVECs were treated with (0–10 mM) TXA for 48 hours, after which the mRNA expression level of ET-1 was measured by qPCR (left). Data are representative of 3 independent experiments and are normalized to the housekeeping gene GAPDH. Western blotting (middle) was performed to detect the protein level of ET-1; a representative blot is shown. Histograms (right) showing the densitometric quantification of data with means ± SD from 3 independent experiments. (C) CM was harvested from HUVECs treated or untreated with 3 mM TXA for 48 hours, after which PIG1 MCs were treated with the HUVEC-CM for 48 hours. The mRNA expression levels of TYR and MITF were measured using qPCR. Data are representative of 3 independent experiments and are normalized to the housekeeping gene GAPDH. (D) Western blotting was performed to detect the protein levels of TYR and MITF in HUECV-CM-treated or -untreated MCs; representative blots are shown on the left. The histograms show the densitometric quantification of data with means ± SD from 3 independent experiments.

ET-1: endothelin-1, TXA: tranexamic acid, MC: melanocyte, HUVEC: human umbilical vein endothelial cell, CCK-8: cell counting kit-8; qPCR: quantitative polymerase chain reaction, GAPDH: glyceraldehyde 3-phosphate dehydrogenase; SD: standard deviation, CM: conditioned medium, TYR: tyrosinase, MITF: microphthalmia-associated transcription factor, ns: not significant.

^*p<0.05, ^**p<0.01 vs. control.

Fig. 3. Decreased UVB-induced TYR expression in MCs treated with the CM from HUVECs exposed to TXA and Bosentan. (A) MCs were treated with 2 µM Bosentan (a specific antagonist for endothelin type A and B receptor) or were co-treated with the CM from HUVECs treated with (w/) or without (w/o) 3 mM TXA; cells were irradiated with 30 mJ/cm² UVB as a positive control. The mRNA expression level of TYR was measured by qPCR. Data are representative of 3 independent experiments and are normalized to the housekeeping gene GAPDH. (B) Western blotting was performed to detect the protein level of TYR in HUECV-CM-treated or -untreated MCs; representative blots are shown on the left. Histograms (right) show the densitometric quantification of data with means ± SD from 3 independent experiments.

UVB: ultraviolet B, TYR: tyrosinase, MC: melanocyte, CM: conditioned medium, HUVEC: human umbilical vein endothelial cell, TXA: tranexamic acid, TYR: tyrosinase, qPCR: quantitative polymerase chain reaction, GAPDH: glyceraldehyde 3-phosphate dehydrogenase, SD: standard deviation.

^*p<0.05 vs. the Bosentan-treated group.

Fig. 4. Decreased UVB-induced proliferation of MCs treated with the CM from HUVECs exposed to TXA and Bosentan. MCs were treated with 2 µM Bosentan or were co-treated with the CM from HUVECs treated with (w/) or without (w/o) 3 mM TXA. After 48 hours of treatment, the MCs were labeled with EdU for 12 hours. Proliferating cells were identified by EdU incorporation into nuclear DNA (red); the cell nuclei were counterstained with DAPI (blue). For each EdU experiment, three random fields were imaged at 100× magnification and the numbers of EdU-positive cells were counted; representative images of EdU+ cells are shown (at the top). The histogram (at the bottom) shows the EdU+ cell count of data with means ± SD from 3 independent experiments.

UVB: ultraviolet B, MC: melanocyte, CM: conditioned medium, HUVEC: human umbilical vein endothelial cell, TXA: tranexamic acid, EdU: 5-ethynyl-2′-deoxyuridine, DAPI: 4′,6-diamidino-2-phenylindole, SD: standard deviation.

^*p<0.05 vs. the Bosentan-treated group.

The presence of decreased melanogenesis and reduced angiogenesis in ex vivo cultured human skin explants treated with 3 mM TXA

To exclude the possibility that certain factors in vivo could also influence MC function, we established an ex vivo cultured human skin explant model treated or untreated with 3 mM TXA for 48 hours²⁰. The results showed that 3 mM TXA significantly inhibited melanin production (Fig. 5A). Immunofluorescent staining demonstrated that the number of CD31-positive cells was much lower in human skin explants exposed to 3 mM TXA than in the untreated control. Meanwhile, the decreased number of HMB45-positive MCs was also seen in the human skin explants exposed to 3 mM TXA (Fig. 5B). These observations reveal that melanogenesis and angiogenesis are closely related in UVB-exposed skin, which further supports that TXA inhibits melanogenesis via the suppression of endothelial cell-derived ET-1.

Fig. 5. Decreased melanogenesis and reduced angiogenesis are observed in ex vivo cultured human skin explants treated with 3 mM TXA. (A) An ex vivo cultured human skin explant model was established at the air-liquid interface, then treated with or without 3 mM TXA for 48 hours. Melanin content and distribution were examined using Fontana-Masson staining; representative images are shown on the left. The histograms (right) shows the densitometric quantification of melanin content with means ± SD from 3 independent experiments. (B) The dermal vascularity and MC distribution in the TXA-treated (upper-row) or -untreated (lower row) mouse tail skin were examined using CD31 or HMB45 immunofluorescence staining. Representative images of CD31 (red) and HMB45 (green) in mouse tail skin are shown. Nuclei were stained with DAPI (blue). White arrows indicate the CD31+ microvascular lumen. Dashed white lines represent the dermal-epidermal junction. The histograms (at the bottom) show the quantification of CD31+ or HMB45+ cells with means ± SD from 3 independent experiments.

TXA: tranexamic acid, SD: standard deviation, MC: melanocyte, CD31: cluster of differentiation 31, DAPI: 4′,6-diamidino-2-phenylindole.

^**p<0.01 vs. the untreated group.

DISCUSSION

It has been reported that sub-clinical telangiectatic erythema confined to melasma lesions in some patients usually implies a risk factor for disease rebound after treatment²⁴. Recent studies have shown that vascular-targeted therapies such as oral or topical TXA as well as pulsed dye laser treatment to mitigate angiogenesis has proven to be effective treatments for recalcitrant facial melasma²⁵. A retrospective analysis conducted by Lee et al.²⁶ showed that 561 melasma patients who received oral 250 mg TXA twice a day for 4 months, the majority (89.7%) improved, 10% had no improvement, and 0.4% worsened. Topical TXA alone was found to be least effective method but could be combined with microneedling and fractional carbon dioxide laser to improve outcomes²⁷,²⁸. Although there is increasing evidence for the off-label use of TXA to treat melasma, the exact mechanisms involved are not yet fully understood. A competitive inhibitory mechanism whereby TXA antagonistically inhibits TYR with its substrate tyrosine seems to be highly questionable²⁶. A more plausible explanation is that TXA blocks the conversion of plasminogen to decrease the interaction between keratinocytes and MCs, also suppressing the production of promelanogenic substances such as arachidonic acid and prostaglandins by keratinocytes²⁵,²⁹.

In this study, we provide evidence that TXA inhibits TYR activity and MC proliferation mainly through the downregulation of ET-1 expression in dermal microvascular endothelial cells. Our in vivo study testing the anti-melanogenic effects of oral or topical TXA therapy revealed that melanin content and the pigmented phenotype were significantly attenuated in the tail skin from UVB-irradiated mice that were treated with intragastral TXA, thus intragastral treatment seems to be superior to topical application (Fig. 1). Mechanistic analysis indicated that TXA therapy preferentially represses ET-1 expression in microvascular endothelial cells, which indirectly inhibits melanogenesis (Figs. 2 and 3). EdU incorporation assay also showed that the CM from HUVECs exposed to TXA significantly inhibited MC proliferation as compared with those from HUVECs unexposed to TXA (Fig. 4). An ex vivo cultured human skin explant model was used to assess the anti-melanogenic effects of TXA treatment; the data showed that decreased cell numbers of HMB45-positive MCs and CD31-positive vascular endothelial cells were observed in cultured human skin explants exposed to TXA (Fig. 5). It is conceivable that a myriad of TXA molecules is accumulated quickly in dermal microvascular plexus by oral administration more efficiently than by topical treatment, and that TXA suppresses the secretion of ET-1 from vascular endothelial cells, which eventually reduces ET-1-related melanogenesis. Although ex vivo cultured human skin explant offers an opportunity to study the interaction between epidermis and dermis, the major limitation of this model is their inability to resist radiation of UVB. In addition, this animal experiment is only a pigmentation model, not a melasma model. In the future, the clinical trials are needed to investigate the therapeutic effects of oral and topical administration of TXA on patients with melasma. The vascularized human skin equivalent with hypodermis seems to be an ideal model system used for further dissecting the role of TXA on UVB-induced hyperpigmentation in vitro ³⁰.

Taken together, to our knowledge, this is the first comparative study of oral TXA versus topical TXA to inhibit UVB-induced hyperpigmentation. To demonstrate this, we used a UVB-induced hyperpigmentation mouse tail skin model and an ex vivo cultured human skin explant model to assess the skin lightening effects by the 2 TXA delivery approaches. Suppressing the vascular endothelial cell-derived promelanogenic cytokine ET-1 represents an effective strategy for treating melasma, especially in refractory cases that coexist with subtle telangiectatic erythema.