Int. J. Med. Sci. 2020, Vol.

17 903

Ivyspring
International Publisher International Journal of Medical Sciences
2020; 17(7): 903-911. doi: 10.7150/ijms.44188
Research Paper

Tranexamic Acid Inhibits Angiogenesis and

Melanogenesis in Vitro by Targeting VEGF Receptors
Jian-Wei Zhu1, Ya-Jie Ni1, Xiao-Yun Tong1, Xia Guo1, Xiao-Ping Wu1, Zhong-Fa Lu2
1. Department of Dermatology, Zhejiang Hospital, No. 12, Lingyin Rd., Hangzhou, 310013, Zhejiang Province, China.
2. Department of Dermatology, Second Affiliated Hospital, Zhejiang University School of Medicine, No. 88, Jiefang Rd., Hangzhou, 310009, Zhejiang Province,
China.

 Corresponding author: Zhong-Fa Lu, E-mail: [email protected]

© The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/).
See http://ivyspring.com/terms for full terms and conditions.

Received: 2020.01.22; Accepted: 2020.03.06; Published: 2020.03.25

Abstract
Melasma is a common but complex skin condition concerning cosmetic problems. Tranexamic acid (TA)
has been proved to be effective in treatment of melasma with still unclear mechanisms. Here, we show
that VEGF165 enhanced the expression of VEGF receptors (VEGFRs, including VEGFR-1, VEGFR-2 and
NRP-1) in human umbilical vein endothelial cells (HUVECs), which was attenuated by TA. VEGF165 also
promoted tyrosine phosphorylation of VEGFR-1 and VEGFR-2 in HUVECs, which was again abolished by
TA. TA further showed similar effects to neutralization of VEGFR-1 and VEGFR-2 in inhibiting cell
proliferation, migration, invasion and tube formation of HUVECs induced by VEGF165, suggesting that
TA could inhibit angiogenesis by targeting VEGFRs in HUVECs. In addition, VEGF165 enhanced the
expression of VEGFRs and promoted tyrosine phosphorylation of VEGFR-1 and VEGFR-2 in normal
human melanocytes, which were also attenuated by TA. Furthermore, TA showed similar effects to
neutralization of VEGFR-1 and VEGFR-2 in inhibiting tyrosinase activity, melanin production and even
melanogenic proteins induced by VEGF165, suggesting that TA could reduce melanogenesis via inhibiting
activation of VEGFRs and subsequent expression of melanogenic proteins in melanocytes. Taken
together, we demonstrate that TA can inhibit angiogenesis and melanogenesis in vitro at least in part by
targeting VEGFRs, which may offer a new understanding of the pathogenesis of melasma as well as the
molecular mechanism for TA in treatment of the disease.
Key words: tranexamic acid, VEGF receptors, human umbilical vein endothelial cells, angiogenesis, melanocytes,
melanogenesis

Introduction
Melasma, formerly known as chloasma, is a Localized microinjection of TA also improved
common acquired dermatological condition melasma in vivo [4]. However, it still remains unclear
characterized with hyperpigmentation and increased how TA exactly works. TA can prevent the binding of
dermal capillaries in the lesional skin, typically plasminogen originating from endothelial cells to
occurring symmetrically on the face, with higher keratinocytes, which is thought to be a possible
prevalence in females and darker skin types [1]. mechanism for melasma treatment [5], implicating
Although common, the management of this disorder those interactions between the altered cutaneous
remains challenging given the incomplete vasculature and the overlying epidermis may have an
understanding of the pathogenesis, its chronicity and influence on the development of hyperpigmentation
recurrence rates [2]. In addition to traditional of melasma.
treatments for melasma, there are also promising new Keratinocytes are the main epidermal cells in the
treatments, including topical, oral and procedural skin. Vascular endothelial growth factor (VEGF)
therapies, among which oral tranexamic acid (TA) is constitutes an important keratinocyte-derived factor
now proved to be effective for melasma and in response to UV irradiation and acts in both
ultraviolet (UV)-induced hyperpigmentation [2, 3]. autocrine and paracrine manner affecting epidermal

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 904

cells and dermal tissues through binding with VEGF Materials and Methods
receptors (VEGFRs) [6-9]. Epidermal keratinocytes
had been demonstrated to express VEGFRs and Cell culture
co-receptors, and VEGF/VEGFR-2 autocrine signal- HUVECs are purchased from the China
ing had been detected in keratinocytes [10, 11], even Infrastructure of Cell Line Resources (Beijing, China)
in epidermal appendages [12-16]. UV could activate and were cultured in endothelial basal medium
VEGFRs in normal keratinocytes, and the activated containing 1% fetal bovine serum (FBS) and
VEGFR-1 and VEGFR-2 signaling was involved in the endothelial cell growth medium 2 Bullet Kit in a
pro-survival mechanism [17]. Moreover, the over- humidified atmosphere of 5% CO2 at 37°C. Passages 3
expressed VEGFRs in psoriatic epidermis could be and 5 were used in all experiments.
attenuated by narrow-band UVB therapy [18]. Thus, For melanocytes, adolescent foreskin was
VEGFRs are believed to participate in physiological obtained from urinary surgery and handled asepti-
and pathological conditions in the epidermis. cally. After removing subcutaneous elements, the
Although keratinocytes account for the main specimen was cut into 0.5 cm2 pieces and put into
epidermal cells in the skin, melanocytes, situated at 0.5% dispase (Gibco, Invitrogen, Auckland, USA) for
the basal layer of the epidermis, are responsible for overnight incubation at 4°C. The epidermis was
melanin synthesis and skin color. One of the most peeled off from the dermis. The epidermis was further
specific and potent environmental factors associated incubated in 0.25% trypsin for 10 min at 37°C. Trypsin
with skin color is UVB [8], which induces pigmenta- activity was neutralized by adding FBS. The mixed
tion by controlling several parameters of melanocyte cell suspension was filtered through nylon gauze (100
differentiation such as melanin synthesis, dendrite μm mesh) and cells were washed twice with 0.1 M
outgrowth, and melanosome transport required for phosphate-buffered saline (PBS) prior to resuspension
melanin distribution within the epidermis [19]. in M2 medium (PromoCell, Heidelberg, Germany)
Melanogenesis is a common cellular process, whereas supplemented with penicillin (100 units/ml) and
hyperfunction of melanocytes can also result in streptomycin (100 mg/ml). Cells were seeded onto 10
pigmentary disorders such as melasma [20]. Melano- cm culture dishes (Corning, Corning NY, USA),
cytes in the lesional melasma epidermis are generally maintained at 37°C in a humidified atmosphere
enlarged, have prominent dendrites and increased containing 5% CO2. Passages 3 to 5 were used in all
melanosomes, resulting in a major clinical feature as experiments, including cell identification by Dopa
hyperpigmented patches, but patients have additional staining and S-100 protein immunohistochemical
distinguishing features like pronounced telangiectatic staining.
erythema basically confined to the melasma lesional
skin [21]. These features were confirmed by signifi- Western blot analysis
cant increase in vascularity in the dermis and VEGF Cells were treated with VEGF165 and TA when
over-expression in the epidermis in melasma lesions grown to 80% confluent. Total cellular protein was
compared with those in perilesional normal skin [22], extracted with lysis buffer (Beyotime Biotechnology,
suggesting that a connection between vessels and Beijing, China). An aliquot of proteins were separated
cutaneous pigmentation could exist. Human melano- by 10% sodium dodecyl sulfate-polyacrylamide gel
cytes may respond to angiogenic factors because and transferred onto the polyvinylidene difluoride
normal human melanocytes express functional membranes (Millipore, Bedford, MA). After blocking
VEGFRs, and the expression of VEGFR-2 was for 30 min, the membrane was incubated with
up-regulated by UVB [23]. However, the influence of primary antibodies for 2 h at room temperature, and
VEGFRs on physiological function of melanocytes then conjugated with horseradish peroxidase
especially melanogenesis needs further clarification. (HRP)-conjugated secondary antibodies (Jackson
Therefore, in this report, we investigated the Laboratories, West Grove, PA, USA) (1:5000 dilution)
functional significance of VEGFRs in human umbilical for 1 h. Immunoreactivity was detected by ECL plus
vein endothelial cells (HUVECs) and epidermal luminal solution (Amersham Biosciences, Piscataway,
melanocytes in response to TA. Our results NJ, USA). Immunoreactive bands were detected using
demonstrate that VEGFRs signaling is involved in an enhanced chemiluminescence system (Millipore).
angiogenesis and melanogenesis, and TA potently Protein density was measured by a Bio-Rad XRS
suppresses both angiogenic and melanogenic chemiluminescence detection system (Bio-Rad, USA).
processes via inhibiting activation of VEGFRs The blots shown are representative of at least three
respectively in HUVECs and normal human repeats. The primary antibodies used in this study
melanocytes, which may offer a new therapeutic were antibodies against VEGFR-1, VEGFR2, NRP-1
target for hyperpigmented diseases like melasma. (1:500-1:1000 dilution) (Santa Cruz, CA, USA),

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 905

phospho-VEGFR-1 (Y1213), phospho-VEGFR-2 the underside were stained with 1% crystal violet and
(Tyr1175) (1:500 dilution) (Cell Signaling Technology, then counted under microscope.
Danvers, MA, USA), MITF, Tyrosinase, Trp-1, Trp-2
(1:1000 dilution) and β-actin (1:5000 dilution) (Abcam, Tube Formation Assay
Shanghai, China). Plates with 24 wells were first coated with
Matrigel (BD Biosciences). Unpolymerized Matrigel
Cell proliferation assay was placed in the wells (300 μl/well) and allowed to
Cell proliferation was measured by MTT assay. polymerize for 1 h at room temperature. HUVECs in
Briefly, cells were plated in collagen-coated 96-well 500 μl medium were seeded onto the polymerized
plates (5×104 cells/ml, 100 μl/well) and cultured Matrigel at a density of 5 ×104 cells/well. After incu-
overnight. Cells were then stimulated with 0 or 10 bation with or without 10 ng/ml VEGF165, 1 mg/ml
ng/ml VEGF165 for 48 h, with or without incubation TA, 5 μg/ml VEGFR-1 and/or VEGFR-2 neutralizing
of 1 mg/ml TA, 5 μg/ml VEGFR-1 and/or VEGFR-2 antibody for 48 h, images of tube formation were
neutralizing antibody. 20 μl of 3-[4,5-dimethylthiazol- acquired with an inverted phase- contrast light
2-yl]2,5-diphenyltetrazolium bromide (MTT; Sigma, microscope (Olympus Corporation, Tokyo, Japan).
St. Louis, MO, USA) at a final concentration of 5 The degree of tube formation was quantified in 5
mg/ml in PBS was incubated with the analyzed cells random fields from each well using ImageJ software.
for 4 h at 37°C. Dark blue formazan crystals that
formed in the living cells were dissolved by adding Tyrosinase activity
100 μl of dimethylsulfoxide. The optical absorbency Tyrosinase enzyme activity was estimated by
was measured at 570 nm as A value with a measuring the rate of L-DOPA oxidation. Briefly,
spectrophotometric reader Elx808 (Bio-Tek, Winooski, melanocytes were treated with 10 ng/ml VEGF165 for
VT, USA). 48 h with or without incubation of 1 mg/ml TA,
5 μg/ml VEGFR-1 and/or VEGFR-2 neutralizing
Wound Healing Assay antibody. The cells were solubilized with phosphate
HUVECs were seeded in 6-well plates until buffer (pH 6.8) containing 1% Triton X-100. The cells
confluent. A mechanical wound was created by gently were then sonicated, and the lysates were clarified by
scratching the cells with the tip of a pipette (time 0 centrifugation at 18,000 × g for 15 min. After protein
hour). The cells were then washed with serum-free quantification and adjustment of protein concentra-
medium and cultured in endothelial cell growth tions with lysis buffer, 170 μL of 2 mM L-DOPA were
medium containing 0.1% FBS with or without added to each 96-well plate. The absorbance was
incubation of 10 ng/ml VEGF165, 1 mg/ml TA, 5 measured spectrophotometrically at 470 nm following
μg/ml VEGFR-1 and/or VEGFR-2 neutralizing a 15-min incubation period at 37 °C. Tyrosinase
antibody. Images were captured after 48 hours, and activity was expressed as a fold change of the treated
the relative migration distance was calculated using group vs. the control group.
the following formula: the relative migration distance
(%) =100 (AX-BX)/(A blank-B blank), where A is the Melanin content assay
width of the cell wound before incubation and B is the To determine the melanin content, melanocytes
width of the cell wound after incubation. were treated with 10 ng/ml VEGF165 for 48 h with or
without incubation of 1 mg/ml TA, 5 μg/ml VEGFR-1
Transwell Invasion Assay and/or VEGFR-2 neutralizing antibody, after which
The assay was performed using chambers they were washed with PBS and harvested. After
containing polycarbonate filters with pore size 8 μm centrifugation at 18,000 × g for 5 min, the harvested
(Merck Millipore, Darmstadt, Germany). The upper cell pellets were solubilized in boiled 1 N NaOH (65
side of a polycarbonate filter was coated with Matrigel °C) for 2 h, and color was analyzed at 405 nm using a
(BD Biosciences, Franklin Lakes, New Jersey) to form microplate reader. Melanin content was expressed as
a continuous, thin layer. HUVECs (1× 105) were a fold change of the treated group vs. the control
resuspended in 300 μl of 0.1% FBS medium and then group.
added to the upper chamber. The lower chamber was
filled with 200 μl endothelial cell growth medium Statistical analysis
containing 0.1% FBS, with or without incubation of 10 Results are expressed as means ± SD. All
ng/ml VEGF165, 1 mg/ml TA, 5 μg/ml VEGFR-1 experiments were performed at least in triplicate.
and/or VEGFR-2 neutralizing antibody. After a 48-h Statistical analysis was performed using one-way
incubation and removal of the cells from the upper analysis of variance (ANOVA) and Student’s t-test.
chamber of the filter with a cotton swab, the cells on The differences were considered statistically
significant when p value was less than 0.05.

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 906

Figure 1. Normal human melanocytes were cultured and proved to be functionally active, and 1 mg/ml TA inhibited proliferation of both HUVECs and
melanocytes without obvious toxicity. (A) The growth of melanocytes was observed by inverted microscope at 7 days of culture (×40). Melanocytes were identificated by
(B) DOPA staining (×40) and (C) S-100 protein immunohistochemical staining (DAB, ×200). (D) Cell counting of HUVECs at 0 h, 24 h, 48 h and 72 h of culture respectively, with
or without incubation of 1 mg/ml TA. (E) Cell counting of melanocytes at 0 h, 24 h, 48 h and 72 h of culture respectively, with or without incubation of 1 mg/ml TA. TA,
tranexamic acid. * P < 0.05, Control vs. TA group at the same time points. Data are presented as the mean ± SD of three independent experiments.

Results examined the regulation of VEGFR-1, VEGFR-2 and

NRP-1 protein expression in HUVECs by VEGF165
TA inhibited cell growth of HUVECs and with or without incubation of 1 mg/ml TA for 48 h.
melanocytes without cellular toxicity As shown in Fig. 2A, a significant induction of
Both HUVECs and normal human melanocytes VEGFR-1, VEGFR-2 and NRP-1 expression was
were successfully cultured. The melanocytes were detected 48 h after 10 ng/ml VEGF165 stimulation.
multi-dendritic, without contamination of epidermal However, this effect was attenuated by 1 mg/ml TA,
keratinocytes or dermal fibroblasts (Fig. 1A). Dopa suggesting that TA could inhibit the enhanced
staining (Fig. 1B) and immunohistochemical staining expression of all the above VEGFRs induced by
for S-100 protein (Fig. 1C) were positive, proving VEGF165 in HUVECs.
them to be functionally active. Then, cell counting of We next examined if TA could affect the tyrosine
HUVECs and melanocytes was performed at 0 h, 24 h, phosphorylation of VEGFR-1 and VEGFR-2 induced
48 h and 72 h of culture respectively, with or without by VEGF165 in HUVECs. As expected, 10 ng/ml
incubation of 1 mg/ml TA. As shown in Fig. D and E, VEGF165 promoted tyrosine phosphorylation of
compared with their control groups, both HUVECs VEGFR-1 and VEGFR-2 in HUVECs 5 min after
and melanocytes that were treated by 1 mg/ml TA stimulation, while no similar effect was observed in
showed similar but slower growth curves, and after cells who were pre-incubated with 1 mg/ml TA for 48
48-72 h of culture, TA significantly inhibited cell h (Fig. 2B), suggesting that TA could abolish the
growth of both HUVECs and melanocytes without activation of VEGFR-1 and VEGFR-2 induced by
obvious cellular toxicity. VEGF165 in HUVECs.

TA inhibited VEGF165-induced TA inhibited cell proliferation, migration,

over-expression and activation of VEGF invasion and tube formation of HUVECs
receptors in HUVECs It is well known that VEGF, via VEGFRs in
As VEGF is natural ligand for VEGFRs, of which endothelial cells, acts as an essential factor for normal
the most abundant form is VEGF165 [24], we then and aberrant angiogenesis [24]. We next examined

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 907

whether TA further interferes with the classical roles

of VEGF-VEGFR signaling in blood vessel formation
in vitro. As shown in Fig. 3A, 10 ng/ml VEGF165
promoted cell proliferation of HUVECs, while
neutralization of VEGFR-1 and/or VEGFR-2
significantly decreased this effect. Interestingly, TA
showed a similar effect to that by neutralization of
VEGFR-1 and VEGFR-2 in HUVECs, as it also
significantly decreased cell proliferation of HUVECs
induced by VEGF165. VEGF165 also promoted cell
migration of HUVECs, and TA exerted a similar effect
to that by neutralization of VEGFR-1 and VEGFR-2,
because both of them significantly decreased cell
migration of HUVECs induced by VEGF165 (Fig. 3B).
In addition, VEGF165 accelerated cell invasion of
HUVECs, which was inhibited by TA as well as by
neutralization of VEGFR-1 and/or VEGFR-2 (Fig. 3C).
Tube formation of HUVECs was also greatly
facilitated by VEGF165, while TA again exerted a
similar effect to that by neutralization of VEGFR-1
Figure 2. TA inhibited VEGF165-induced over-expression and activation
and/or VEGFR-2 which significantly attenuated tube of VEGF receptors in HUVECs. (A) The expression of VEGFR-1, VEGFR-2 and
formation of HUVECs (Fig. 3D), suggesting that TA NRP-1 in HUVECs in response to 0 or 10 ng/ml VEGF165 with or without incubation
of 1 mg/ml TA for 48 h. (B) The tyrosine phosphorylation of VEGFR-1 and VEGFR-2
could inhibit angiogenesis via targeting VEGFR in HUVECs in response to 0 or 10 ng/ml VEGF165 for 5 min with or without
signaling in HUVECs in vitro. pre-incubation of 1 mg/ml TA for 48 h. TA, tranexamic acid. P-VEGFR-1,
phospho-VEGFR-1. P-VEGFR-2, phospho-VEGFR-2. β-actin was served as loading
control for protein normalization.

Figure 3. TA showed similar effects to neutralization of VEGFR-1 and VEGFR-2 in inhibiting cell proliferation, migration, invasion and tube formation of
HUVECs. (A) The A value of HUVECs 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation of 1 mg/ml TA, VEGFR-1 and/or VEGFR-2 neutralizing antibody
(5 µg/ml, respectively). (B) The migration rate of HUVECs 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation of 1 mg/ml TA, VEGFR-1 and/or VEGFR-2
neutralizing antibody (5 μg/ml, respectively). (C) The number of invaded HUVECs 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation of 1 mg/ml TA,
VEGFR-1 and/or VEGFR-2 neutralizing antibody (5 μg/ml, respectively). (D) Tube formation of HUVECs 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation
of 1 mg/ml TA, VEGFR-1 and/or VEGFR-2 neutralizing antibody (5 µg/ml, respectively). * P < 0.05. TA, tranexamic acid. VEGF, VEGF165. A-R1, VEGFR-1 neutralizing antibody.
A-R2, VEGFR-2 neutralizing antibody. A-(R1+R2), VEGFR-1 & VEGFR-2 neutralizing antibodies.

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 908

and skin color. Thus, we investigated if TA further

interferes with cell proliferation, tyrosinase activity
and melanin synthesis of melanocytes via
VEGF-VEGFR signaling in vitro. As shown in Fig. 5A,
10 ng/ml VEGF165 failed to significantly promote cell
proliferation of melanocytes, and neutralization of
VEGFR-1 and/or VEGFR-2 showed non-effective for
melanocyte proliferation induced by VEGF165.
However, TA could still significantly decrease cell
proliferation of melanocytes independent of VEGFRs.
VEGF165 promoted tyrosinase activity of
melanocytes, and TA exerted a similar effect to that by
neutralization of VEGFR-1 and VEGFR-2, as both of
them significantly decreased tyrosinase activity of
melanocytes induced by VEGF165 (Fig. 5B). In
addition, VEGF165 enhanced melanin production of
melanocytes, which was reduced by TA as well as by
neutralization of VEGFR-1 and VEGFR-2 (Fig.5C),
suggesting that TA could inhibit melanogenesis via
Figure 4. TA inhibited VEGF165-induced over-expression and activation
targeting VEGFR signaling in melanocytes in vitro.
of VEGF receptors in melanocytes. (A) The expression of VEGFR-1, VEGFR-2
and NRP-1 in melanocytes in response to 0 or 10 ng/ml VEGF165 with or without TA inhibited the expression of melanogenic
incubation of 1 mg/ml TA for 48 h. (B) The tyrosine phosphorylation of VEGFR-1 and
VEGFR-2 in melanocytes in response to 0 or 10 ng/ml VEGF165 for 5 min with or
proteins in melanocytes
without pre-incubation of 1 mg/ml TA for 48 h. TA, tranexamic acid. P-VEGFR-1, Melanocytes are responsible for melanin
phospho-VEGFR-1. P-VEGFR-2, phospho-VEGFR-2. β-actin was served as loading
control for protein normalization. synthesis, which is a complex process requiring
melanogenic factors including microphthalmia-
associated transcription factor (MITF), tyrosinase,
TA inhibited VEGF165-induced
tyrosinase related protein-1 (Trp-1), and Trp-2 [25]. To
over-expression and activation of VEGF
elucidate if they are downstream proteins of VEGF
receptors in melanocytes
receptors, melanocytes were stimulated with 10
We next turned to examine the regulation of ng/ml VEGF165 in the absence or presence of the
VEGFR-1, VEGFR-2 and NRP-1 protein expression by VEGFR-1 and/or VEGFR-2 neutralizing antibody. As
VEGF165 in normal human melanocytes with or shown in Fig. 6, VEGF165 induced a strong
without incubation of 1 mg/ml TA for 48 h. As shown over-expression of MITF, tyrosinase, Trp-1, and Trp-2
in Fig. 4A, an enhanced expression of all the above in melanocytes, which could be reduced by
VEGF receptors was detected 48 h after 10 ng/ml neutralization of VEGFR-1 and/or VEGFR-2. TA also
VEGF165 stimulation, but this effect was again significantly decreased the over-expression of MITF,
attenuated by 1 mg/ml TA, suggesting that TA could tyrosinase, Trp-1 and Trp-2 induced by VEGF165.
also inhibit VEGF165-induced over-expression of This suggested that over-expression of melanogenic
VEGFRs in melanocytes. proteins induced by VEGF165 involved activation of
We then studied if TA could affect VEGFR-1 and VEGFR-2, and TA could decrease
VEGF165-induced tyrosine phosphorylation of tyrosinase activity and melanin production via
VEGFR-1 and VEGFR-2 in melanocytes. As shown in inhibiting activation of VEGF receptors and
Fig. 4B, tyrosine phosphorylation of VEGFR-1 and subsequent expression of melanogenic proteins in
VEGFR-2 was also promoted 5 min after 10 ng/ml melanocytes.
VEGF165 stimulation in melanocytes, while no similar
effect occurred in cells that were pre-incubated with 1 Discussion
mg/ml TA for 48 h, suggesting that TA could also Melasma is a common but complex skin
abolish VEGF165-induced activation of VEGFR-1 and condition. Histologically, melasma can display
VEGFR-2 in melanocytes. increased epidermal and/or dermal pigmentation,
TA inhibited cell proliferation, tyrosinase enlarged melanocytes, increased melanosomes, and
activity and melanin production of an increase in dermal blood vessels [1, 2, 21]. Multiple
melanocytes etiologies including UV exposure and family history
have been implicated in the pathogenesis of this
Situated at the basal layer of the epidermis,
disorder [21]. VEGF, an angiogenic factor released
melanocytes are responsible for melanin synthesis

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 909

from keratinocytes after UV damage, can sustain and melanocytes. Meanwhile, in order to elucidate
human melanocytes in tissue culture, which is how TA acts on melasma, we performed our
proposed as one of the mechanisms for the increased experiments to study if TA could regulate VEGFRs
activity of melanocytes in melasma [23]. Family and their possible related cell functions. We found
history is another important risk factor for melasma, that although 1 mg/ml TA inhibited cell growth of
no genome-wide study has been performed to HUVECs and melanocytes compared to control cells,
examine associated genes, but current findings would no obvious cellular toxicity was detected as both of
suggest that the genes responsible involve pigment- the two cells were still in a good proliferation state,
ary, inflammatory, hormonal, and vascular responses suggesting it is a safe concentration for next
[26]. With increased recognition that angiogenesis experiments. We then found that TA could inhibit
may play a role in the pathogenesis of melasma, a VEGF165-induced over-expression and activation of
copper bromide anti-angiogenesis laser was found to VEGF receptors in HUVECs. As VEGF acts as an
significantly decrease the MASI score and expression essential factor for normal and aberrant angiogenesis
of endothelin-1 and VEGF on immunohistochemistry via VEGFRs in endothelial cells, we also investigated
[27]. The plasmin inhibitor, TA, has been proved to be whether TA further interferes with the classical roles
effective in treatment of melasma with still unclear of VEGF-VEGFR signaling in blood vessel formation
mechanisms, one study showed that tranexamic acid in vitro. As expected, cell proliferation, migration,
may also decrease VEGF and entothelin-1, both of invasion and tube formation of HUVECs, which
which are responsible for increasing vascularity in represent the capacity to form blood vessels, were all
affected lesions of melasma [28]. significantly decreased by VEGFR-1 and/or VEGFR-2
Based on the above knowledge, we speculated neutralizing antibodies. Interestingly, TA exerted a
that a connection between dermal vessels and similar effect to that by neutralization of VEGFR-1
cutaneous pigmentation could exist in the and VEGFR-2 in blood vessel formation of HUVECs,
pathogenesis of melasma through VEGF receptors suggesting that TA could inhibit angiogenesis via
(VEGFRs) that are expressed both in endothelial cells targeting VEGFRs in HUVECs in vitro.

Figure 5. TA showed similar effects to neutralization of VEGFR-1 and VEGFR-2 in inhibiting tyrosinase activity and melanin production of melanocytes.
(A) The A value of melanocytes 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation of 1 mg/ml TA, VEGFR-1 and/or VEGFR-2 neutralizing antibody (5 μg/ml,
respectively). (B) The tyrosinase activity of melanocytes 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation of 1 mg/ml TA, VEGFR-1 and/or VEGFR-2
neutralizing antibody (5 μg/ml, respectively). (C) The melanin production of melanocytes 48 h after 0 or 10 ng/ml VEGF165 stimulation with or without incubation of 1 mg/ml
TA, VEGFR-1 and/or VEGFR-2 neutralizing antibody (5 μg/ml, respectively). * P < 0.05. TA, tranexamic acid. VEGF, VEGF165. A-R1, VEGFR-1 neutralizing antibody. A-R2,
VEGFR-2 neutralizing antibody. A-(R1+R2), VEGFR-1 & VEGFR-2 neutralizing antibodies.

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 910

our patients in the clinic declined biopsy of melasma

lesions for experimental use, so we had only to
conduct cell studies and present our in vitro data in
this report. Our findings indicate that VEGFRs are
involved in the angiogenesis and melanogenesis in
response to VEGF. VEGFRs might be one of the
melanogenic factors, and VEGFRs-related melanocyte
hyperfunction might lead to development of
melasma. TA has been extensively used for clinical
treatment of melasma and other hyperpigmentary
Figure 6. TA showed similar effects to neutralization of VEGFR-1 and skin conditions for over 30 years without severe side
VEGFR-2 in inhibiting the expression of melanogenic proteins in
melanocytes. Western blotting detection of MITF, tyrosinase, Trp-1 and Trp-2 in effects [28, 29], this may be due to the low treatment
melanocytes 48 h after 10 ng/ml VEGF165 stimulation with or without incubation of 1 dosage compared to that for hemostatic purpose,
mg/ml TA, VEGFR-1 and/or VEGFR-2 neutralizing antibody (5 μg/ml, respectively).
TA, tranexamic acid. A-R1, VEGFR-1 neutralizing antibody. A-R2, VEGFR-2 which might be sufficient to relieve activation of
neutralizing antibody. A-(R1+R2), VEGFR-1 & VEGFR-2 neutralizing antibodies.
β-actin was served as loading control for protein normalization.
VEGFRs in the lesional skin. Although it is well
known that TA inhibits plasmin and TA could reduce
melanocyte tyrosinase activity by suppressing the
We next turned to examine the regulation of production of prostaglandins and UV-induced
VEGFRs by VEGF165 and TA in normal human melanogens through the suppression of the
melanocytes. TA also inhibited VEGF165-induced UV-induced increase in epidermal plasmin activity
over-expression and activation of VEGF receptors in [3], there was hardly any plasmin or plasminogen
melanocytes, a result similar to that in HUVECs. As detected in the supernatants throughout our in vitro
melanocytes are responsible for melanin synthesis cell studies (data not shown), implicating that TA did
and distribution within the epidermis, which may not act on VEGFRs indirectly through plasmin in vitro.
involve cell proliferation, tyrosinase activity and Other previous studies showed that TA could
melanin production, we investigated if TA could suppress UVB-induced melanocyte activation by
further interfere with these activities of melanocytes decreasing the levels of prohormone convertase-2 and
via VEGFRs in vitro. TA significantly decreased cell α-MSH [30], and TA inhibited melanogenesis by
proliferation of melanocytes independent of VEGFRs, activating the autophagy system in cultured
because neither VEGF165 promoted cell proliferation, melanoma cells [31], suggesting that TA can affect
nor neutralization of VEGFR-1 and/or VEGFR-2 melanocyte function in other multiple pathways
showed any effect to melanocyte proliferation. without traditional inhibition of plasmin. In addition,
VEGF165 promoted tyrosinase activity and melanin TA did not seem to act through binding with VEGF,
production of melanocytes, which was significantly as TA alone could inhibit the expression and
decreased by VEGFR-1 and VEGFR-2 neutralizing phosphorylation of VEGFRs without VEGF
antibodies. TA exerted a similar effect to that by stimulation (Fig. 2 and 4), and each experiment group
neutralization of VEGFR-1 and VEGFR-2, suggesting for phosphorylation of VEGFRs was performed by
that TA could also inhibit melanogenesis via targeting pretreatment of TA before VEGF stimulation, there
VEGFRs in melanocytes in vitro. As melanin synthesis was no chance for TA binding directly with VEGF. So,
is a complex process requiring melanogenic proteins we imagined that TA might act directly through
including MITF, tyrosinase, Trp-1 and Trp-2, we also binding with VEGFRs or indirectly through other
asked if they are downstream factors of VEGFRs. molecular signaling. We tried to seek several possible
VEGF165 induced a strong over-expression of these molecular mechanisms or signaling pathways
proteins, which could be reduced by neutralization of concerning how TA directly interacted with VEGFRs,
VEGFR-1 and/or VEGFR-2. TA again exerted a but we failed to find any positive result. In our
similar effect to that by neutralization of VEGFR-1 opinion, TA largely interacts with VEGFRs indirectly
and VEGFR-2, suggesting that over-expression of through other undefined molecular signaling, and
melanogenic proteins induced by VEGF165 involved importantly of course, further work will be needed to
activation of VEGFR-1 and VEGFR-2, and TA could make a more definite conclusion. Anyway, our
decrease tyrosinase activity and melanin production findings can offer new understanding of the
via inhibiting activation of VEGFRs and subsequent pathogenesis of melasma, which may help in the
expression of melanogenic proteins in melanocytes. development of future treatments for this common,
Melasma is a complex condition involving yet challenging condition.
multiple etiologies, and no animal model is available
nowadays for our further in vivo study, and almost all

http://www.medsci.org
Int. J. Med. Sci. 2020, Vol. 17 911

Acknowledgements 19. Buscà R, Ballotti R. Cyclic AMP a key messenger in the regulation of skin
pigmentation. Pigment Cell Res. 2000; 13: 60-9.
20. Passeron T. Melasma pathogenesis and influencing factors-an overview of the
This work is supported by grant 2017KY003 and latest research. J Eur Acad Dermatol Venereol. 2013; 27(Suppl 1): S5-S6.
2020KY006 from the Health Bureau of Zhejiang 21. Ogbechie-Godec OA, Elbuluk N. Melasma: an Up-to-Date Comprehensive
Review. Dermatol Ther. (Heidelb) 2017; 7: 305-18.
Province, grant 81773350 and 81972959 from Natural 22. Kim EH, Kim YC, Lee ES, et al. The vascular characteristics of melasma. J
Science Foundation of China, and grant KN20170158 Dermatol Sci. 2007; 46: 111-6.
23. Kim EJ, Park HY, Yaar M, et al. Modulation of vascular endothelial growth
from Natural Science Foundation of Zhejiang factor receptors in melanocytes. Exp Dermatol. 2005; 14: 625-33.
province. 24. Ferrara N. Vacular endothelial growth factor: basic science and clinical
progress. Endocr Rev. 2004; 25: 581-611.
25. Bertolotto C, Abbe P, Hemesath TJ, et al. Microphthalmia gene product as a
Author contributions signal transducer in cAMP-induced differentiation of melanocytes. J Cell Biol.
1998; 142: 827-35.
Conceptualization: ZFL and JWZ. Methodology: 26. Ortonne JP, Arellano I, Berneburg M, et al. A global survey of the role of
YJN, XYT, and JWZ. Statistics: XG and XPW. Writing: ultraviolet radiation and hormonal influences in the development of melasma.
J Euro Acad Dermatol Venereol. 2009; 23: 1254-62.
JWZ and ZFL. Funding acquisition: JWZ. 27. Lee HI, Lim YY, Kim BJ, et al. Clinicopathologic efficacy of copper bromide
plus/yellow laser (578 nm with 511 nm) for treatment of melasma in Asian
Competing Interests patients. Dermatol Surg. 2010; 36: 885-93.
28. Kim SJ, Park JY, Shibata T, et al. Efficacy and possible mechanisms of topical
tranexamic acid in melasma. Clin Exp Dermatol. 2016; 41: 480-5.
The authors have declared that no competing 29. Hiramoto K, Yamate Y, Sugiyama D, et al. The gender differences in the
interest exists. inhibitory action of UVB-induced melanocyte activation by the administration
of tranexamic acid. Photodermatol Photoimmunol Photomed. 2016; 32: 136-45.
30. Hiramoto K, Yamate Y, Sugiyama D, et al. Tranexamic acid suppresses
References ultraviolet B eye irradiation-induced melanocyte activation by decreasing the
levels of prohormone convertase 2 and alpha-melanocyte-stimulating
1. Sanchez NP, Pathak MA, Sato S, et al. Melasma: a clinical, light microscopic,
hormone. Photodermatol Photoimmunol Photomed. 2014; 30: 302-7.
ultrastructural, and immunofluorescence study. J Am Acad Dermatol. 1981; 4:
698-710. 31. Cho YH, Park JE, Lim DS, et al. Tranexamic acid inhibits melanogenesis by
2. Sarkar R, Bansal A, Ailawadi P. Future therapies in melasma: What lies ahead? activating the autophagy system in cultured melanoma cells. J Dermatol Sci.
Indian J Dermatol Venereol Leprol. 2020; 86: 8-17. 2017; 88: 96-102
3. Maeda K, Naganuma M. Topical trans-4-aminomethylcyclohexanecarboxylic
acid prevents ultraviolet radiation-induced pigmentation. J Photochem
Photobiol B. 1998; 47: 136-41.
4. Lee JH, Park JG, Lim SH, et al.. Localized intradermal microinjection of
tranexamic acid for treatment of melasma in Asian patients: a preliminary
clinical trial. Dermatol Surg. 2006; 32: 626-31.
5. Maeda K, Tomitab Y. Mechanism of the inhibitory effect of tranexamic acid on
melanogenesis in cultured human melanocytes in the presence of
keratinocyte-conditioned medium. J Health Sci. 2007; 53: 389-96.
6. Mildner M, Weninger W, Trautinger F, et al. UVA and UVB radiation
differentially regulate vascular endothelial growth factor expression in
keratinocyte-derived cell lines and in human keratinocytes. Photochem
Photobiol. 1999; 70:674-9.
7. Kosmadaki MG, Yaar M, Arble BL, et al. UV induces VEGF through a
TNF-alpha independent pathway. FASEB J. 2003; 17: 446-8.
8. Hirakawa S, Fujii S, Kajiya K, et al. Vascular endothelial growth factor
promotes sensitivity to ultraviolet B-induced cutaneous photodamage. Blood.
2005; 105: 2392-9.
9. Supp DM, Supp AP, Bell SM, et al. Enhanced vascularization of cultured skin
substitutes genetically modified to overexpress vascular endothelial growth
factor. J Invest Dermatol. 2000; 114: 5-13.
10. Man XY, Yang XH, Cai SQ, et al. Immunolocalization and expression of
vascular endothelial growth factor receptors (VEGFRs) and neuropilins
(NRPs) on keratinocytes in human epidermis. Mol Med. 2006; 12: 127-36.
11. Yang XH, Man XY, Cai SQ, et al. Expression of VEGFR-2 on HaCaT cells is
regulated by VEGF and plays an active role in mediating VEGF induced
effects. Biochem Biophysic Res Commun. 2006; 349: 31-8.
12. Man XY, Yan XH, Cai SQ, et al. Expression and localization of vascular
endothelial growth factor and vascular endothelial growth factor receptor-2 in
human epidermal appendages: a comparison study by immunofluorescence.
Clin Exp Dermatol. 2009; 34: 396-401.
13. Li W, Man XY, Li CM, et al. VEGF induces proliferation of human hair follicle
dermal papilla cells through VEGFR-2-mediated activation of ERK. Exp Cell
Res. 2012; 318: 1633-40.
14. Wu XJ, Zhu JW, Jing J, et al. VEGF165 modulates proliferation, adhesion,
migration and differentiation of cultured human outer root sheath cells from
central hair follicle epithelium through VEGFR-2 activation in vitro. J
Dermatol Sci. 2014; 73: 152-60.
15. Wu XJ, Jing J, Lu ZF, et al. Expression and localization of VEGFR-2 in hair
follicles during induced hair growth in mice. Arch Dermatol Res. 2018: 310:
591-98.
16. Wu XJ, Jing J, Lu ZF, et al. VEGFR-2 is in a state of activation in hair follicles,
sebaceous glands, eccrine sweat glands, and epidermis from human scalp: an
in situ immunohistochemistry study of phosphorylated VEGFR-2. Med Sci
Monit Basic Res. 2019; 25: 107-12.
17. Zhu JW, Wu XJ, Luo D, et al. Activation of VEGFR-2 signaling in response to
moderate dose of ultraviolet B promotes survival of normal human
keratinocytes. Int J Biochem Cell Biol. 2012; 44: 246-56.
18. Zhu JW, Wu XJ, Lu ZF, et al. Role of VEGF receptors in normal and psoriatic
human keratinocytes: evidence from irradiation with different UV sources.
PLoS One. 2013; 8: e55463.

http://www.medsci.org