Dual mechanisms of green tea extract (EGCG)-
induced cell survival in human epidermal
keratinocytes1
JIN HO CHUNG, JI HYUN HAN, EUN JU HWANG, JIN YOUNG SEO,
KWANG HYUN CHO, KYU HAN KIM, JAI IL YOUN, AND HEE CHUL EUN2
Department of Dermatology, Seoul National University College of Medicine, and Laboratory
of Cutaneous Aging Research, Clinical Research Institute, Seoul National University Hospital,
Seoul, Korea
SPECIFIC AIMS
Recently, epigallocatechin-3-gallate (EGCG), a major
constituent polyphenol in green tea, was found to have
a pronounced growth inhibitory effect on cancer cells,
but not on normal cells. In this study, we investigated
the effects of EGCG on the proliferation and UV-
induced apoptosis of normal human epidermal kera-
tinocytes and the signaling pathways through which
EGCG acts in vitro and in vivo.
PRINCIPAL FINDINGS
1. EGCG induced keratinocyte proliferation in human
skin in vivo and in vitro
The topical application of 10% EGCG to aged human
skin three times a week for 6 wk increased the epider-
mal thickness. The number of Ki-67 positive keratino-
cytes in the basal cell layer also increased. These results
indicate that the topical application of EGCG stimu-
lates keratinocyte proliferation, which increases the
epidermal thickness of human skin in vivo (Fig. 1A).
Treatment of cultured keratinocytes with EGCG for
5 days significantly increased their proliferation in a
dose-dependent manner up to 1.0 ␮M of EGCG
(Fig. 1B).
2. EGCG stimulated keratinocyte proliferation
through the phosphorylation of Bad by activating
Erk- and Akt-dependent pathways
Treatment with EGCG increased the phosphorylation
of Erk and Akt 1 h post-treatment in cultured normal
keratinocytes, while the level of total Erk and Akt was
unchanged. EGCG increased phosphorylation at
Ser112 and Ser136 of Bad protein, while the total Bad
protein was unchanged (Fig. 2).
The MEK inhibitor PD98059 (20 ␮M) inhibited the
EGCG-induced phosphorylations of Erk and Ser112 of
Bad, but the phosphorylation of Ser136 of Bad was
0892-6638/03/0017-1913 © FASEB
unaffected (Fig. 2A). The PI-3 kinase inhibitor
LY294002 (20 ␮M) inhibited the EGCG-induced phos-
phorylation of Akt and of Ser136 but did not affect the
phosphorylation of the Ser112 (Fig. 2B). When cul-
tured keratinocytes were treated with PD98059 or
LY294002, EGCG-induced proliferation was inhibited.
Therefore, our data suggest that EGCG activates both
Erk- and Akt-dependent pathways and that this results
in phosphorylation of Ser112 and Ser136 of Bad pro-
tein, respectively. Moreover, this increased phosphory-
lation of Bad protein plays roles in the stimulation of
keratinocyte proliferation by EGCG. We confirmed
these effects of EGCG on Bad phosphorylation in
human skin in vivo.
3. EGCG increased Bcl-2 expression and decreased
Bax expression
EGCG significantly increased Bcl-2 protein expression
in a dose-dependent manner 24 h post-treatment in
cultured keratinocytes. Conversely, treatment of cul-
tured keratinocytes with EGCG inhibited Bax protein
expression in a dose-dependent manner. These results
indicate that EGCG increased the ratio of Bcl-2 to Bax,
resulting in the increased proliferation of epidermal
keratinocytes. We confirmed that EGCG had the same
effect on Bcl-2 and Bax expression in human skin in
vivo.
4. ECCG inhibited UV-induced apoptosis by inducing
Bad phosphorylation and increasing the ratio of Bcl-2
to Bax
EGCG prevented UV-induced apoptosis in epidermal
keratinocytes both in vitro and in vivo. UV irradiation
increased Erk and Akt phosphorylation levels in cul-
tured keratinocytes and induced the phosphorylation
1
To read the full text of this article, go to http://www.fasebj.
org/cgi/doi/10.1096/fj.02-0914fje; doi: 10.1096/fj.02-0914fje
2
Correspondence: Department of Dermatology, Seoul Na-
tional University Hospital, 28 Yongon-dong, Chongno-Gu,
Seoul 110-744, Korea. E-mail: hceun@snu.ac.kr
1913

Figure 1. EGCG increased the epidermal thickness of aged
human skin in vivo and stimulated the proliferation of human
epidermal keratinocytes. A) 10% EGCG or its vehicle (70%
propylene glycol, 30% ethanol) was applied to the buttock
skin of 5 elderly men, 3 times a week for 6 wk, under
occlusion. EGCG- or vehicle-treated skin samples were ob-
tained by punch biopsy. The skin sections were stained with
H&E and immunostained with anti-Ki-67 antibody. The aver-
age epidermal thickness was measured using an image ana-
lyzer and the number of Ki-67 positive keratinocytes per mm
of the basal cell layer was calculated. The pictures are
representatives of the 5 individuals; values shown are the
means ⫾ se of the 5 individuals. B) Cultured human epider-
mal keratinocytes were treated with various concentrations of
EGCG (0 –100 ␮M) every 2 days for 5 days and an MTT assay
performed. Values are means ⫾ se from 6 wells and the graph
is a representative result from triplicate experiments. *P ⬍
0.05, **P ⬍ 0.01 compared with vehicle-treated group.
of Bad at both Ser112 and Ser136. Pretreatment with
EGCG prior to UV irradiation augmented the UV-
induced phosphorylation of Erk and Akt and the
UV-induced phosphorylation of both Ser112 and
Ser136 of Bad. The MEK 1 inhibitor PD98059 (20 ␮M)
inhibited the augmentation of the Ser112 phosphory-
lation, but not of the Ser136, of Bad protein by EGCG.
The PI-3 kinase inhibitor LY294002 (20 ␮M) inhibited
the augmentation of the Ser136 phosphorylation, but
not that of Ser112 by EGCG. These results indicate that
EGCG prevents UV-induced apoptosis by augmenting
Bad phosphorylation through Erk and Akt activation.
On the other hand, UV irradiation significantly de-
creases the ratio of Bcl-2 to Bax in cultured keratino-
cytes; EGCG can reverse this UV-induced decrease,
1914 Vol. 17 October 2003 The FASEB Journal
which resulted in the prevention of UV-induced apo-
ptosis. We demonstrated that these effects are present
in human skin in vivo.
5. Differential effects of EGCG on normal
keratinocyte and squamous carcinoma cell
proliferation
In normal keratinocytes, EGCG enhanced proliferation
at 0.5 ␮M but did not affect it significantly at 50
␮M.However, in squamous carcinoma cells, EGCG de-
creased proliferation at both concentrations dose-de-
pendently. Treatment with EGCG also modulated the
activities of Erk and Akt and expression of Bcl-2 and
Bax differently in normal keratinocytes and squamous
carcinoma. The phosphorylations of Erk and Akt were
increased by EGCG at 0.5 ␮M, but not at 50 ␮M, in
normal keratinocytes, which led to increased phosphor-
ylation of Bad at Ser112 and Ser136, respectively. On
the other hand, phosphorylation of Erk was increased,
but the phosphorylation of Akt and Bad at Ser112 and
at Ser136 was decreased by EGCG in a dose-dependent
manner in squamous carcinoma cells. EGCG increased
the ratio of Bcl-2 to Bax in normal keratinocytes, but
decreased it in squamous carcinoma cells. Taken to-
gether, these results suggest that the observed differ-
ences between normal keratinocytes and squamous
cancer cells, in terms of the proliferation induced by
treatment of EGCG, may be due to the different
signaling events induced by EGCG.
CONCLUSIONS AND SIGNIFICANCE
We found, for the first time, that topical EGCG appli-
cation to aged human skin induces epidermal keratin-
ocytes proliferation and results in increased epidermal
thickness. We also found that EGCG inhibits the UV-
induced apoptosis of human epidermal keratinocytes
in vivo and in vitro and that EGCG affects the prolifer-
ation of normal keratinocytes and squamous carcinoma
cells differently.
EGCG promotes the survival of keratinocytes and
inhibits UV-induced apoptosis by a dual mechanism
(Fig. 3): 1) increased phosphorylation at Ser112 and
Ser136 of Bad protein through Erk-dependent and
Akt-dependent pathways, respectively, and 2) by in-
creasing the ratio of Bcl-2 to Bax. The suppression of
Bad-mediated cell death caused by EGCG occurs rela-
tively early, 1 h post-EGCG treatment, whereas the
contribution made by the Bcl-2/Bax ratio is detected
significantly later, 24 h after treatment. Therefore, these
two proposed molecular mechanisms associated with
EGCG-induced cell proliferation promote the survival of
keratinocytes with different kinetics. ECGC acts differ-
ently on cancer cells, and may protect against cancers by
inhibiting proliferation and inducing apoptosis even at
physiological concentrations. Our data suggest that
EGCG can be used topically to induce an anti-aging effect
and an anti-cancer effect in human skin.
CHUNG ET AL.
Figure 2. EGCG phosphorylated Ser112 and Ser136 of Bad protein through Erk and Akt pathways, respectively. Cultured
keratinocytes were treated with EGCG (0 – 0.5 ␮M) for 1 h. Western blot analysis was performed. All bands shown are
representative of triplicate experiments. A) The MEK 1 inhibitor PD98059 (20 ␮M) inhibited EGCG-induced phosphorylation
at Ser112 of Bad. Cultures were incubated with MEK 1 inhibitor 1 h before treatment with EGCG until harvesting. B) The PI-3
kinase inhibitor LY294002 (20 ␮M) inhibited EGCG-induced phosphorylation at Ser136 of Bad. The cultures were incubated
with LY294002 1 h before treatment with EGCG until harvesting. *P ⬍ 0.05 compared with vehicle-treated group, §P ⬍0.05
compared with each EGCG-treated group, respectively.

Figure 3. Proposed molecular mechanisms of EGCG-induced

cell proliferation and survival in human epidermal keratino-
cytes.

MECHANISMS OF EGCG-INDUCED KERATINOCYTES SURVIVAL 1915