DOI: 10.1111/exd.12278

www.wileyonlinelibrary.com/journal/EXDLetter to the Editor

Effluent syntaxin3 from dying cells affords protection againstapoptosis in epidermal keratinocytes

Takafumi Miyazaki, Nanako Kadono, Yoshihiro Konishi, Natsumi Hagiwara, Kenji Maekubo and Yohei Hirai

Department of Bioscience, Kwansei Gakuin University, Sanda, Japan

Correspondence: Yohei Hirai, PhD, Department of Bioscience, Kwansei Gakuin University, 2-1, Gakuen, Sanda 669-1337, Japan, Tel./Fax: +81-79-

565-7234, e-mail: y-hirai@kwansei.ac.jp

Abstract: Ultra-violet B (UVB)-induced oxidative stress crucially

perturbs the epidermal homeostasis, and the skin is endowed with

protective mechanisms to take action against such damage. Here,

we show the possible involvement of t-SNARE protein syntaxin3,

a membrane fusion mediator of cytoplasmic vesicles, and which is

released from dying keratinocytes, to play a role in this response.

UVB irradiation, which generates reactive oxidative stress in cells,

was shown to lead to the keratinocyte cell death accompanied by

a release of cytoplasmic syntaxin3. We found that such

extracellularly sourced syntaxin3 completely blocked the

processing of a crucial effector for apoptotic cell death, caspase-3,

and thus facilitated the survival of keratinocytes damaged by

oxidative stress. These results demonstrate the latent prosurvival

function of syntaxin3 and underline the importance of

intracellular molecular elements for the maintenance of

homeostasis in epidermal keratinocytes.

Key words: epidermis – epimorphin – survival – syntaxin3 – ultra-violet B

Accepted for publication 28 October 2013

BackgroundThe epidermis is a unique type of tissue that is continuously and

directly exposed to the external environment. To confront the var-

ious stimuli encountered and maintain epidermal homeostasis, the

stratified epidermal cell layer has developed so as to serve as a

dynamic physical barrier, where nascent keratinocytes in the basal

layer deliver daughter cells upwards, along with a gradual mature

keratinization process and the formation of a sealing compartment

between cells (1–3). Nevertheless, such an epidermal barrier is not

enough by itself to completely protect cells from harmful stimuli,

for example UVB irradiation is known to penetrate through the

layer to the nascent keratinocytes, generating reactive oxidative

species (ROS) that in turn causing DNA damage, thereby inducing

apoptotic cell death (4).

Syntaxin3, one of the t-SNARE syntaxin family members that

function as cytoplasmic vesicle fusion mediators, resembles two

other members of the plasmalemmal family, epimorphin and syn-

taxin4, in terms of molecular size, amino acid sequence and over-

all domain structure (5). In addition to the cytoplasmic t-SNARE

function, many studies have demonstrated that epimorphin exerts

an extracellular signal in the course of being translocated across

the membrane, processed at the membrane-proximal domain and

secreted in response to various stimuli (5–7). Recently, the

secreted forms of epimorphin and syntaxin4 were reported to con-

fer a protective effect against oxidative stress upon certain epithe-

lial cell types, including the epidermal keratinocyte (8,9).

Questions addressedThe aim of this study is to clarify whether cytoplasmic syntaxin3

in the epidermal keratinocyte possesses a latent extracellular func-

tion as has been shown for other syntaxin family members and,

what this function is and how it is exerted.

Experimental designFor a detailed description of our research, please see the section

Appendix S1.

ResultsExpression profile of syntaxin3 in the skinIn addition to confirming epimorphin’s reported stromal localiza-

tion (10), we observed that syntaxin3 is abundantly expressed in

the epidermal compartment in the mouse skin (Fig. 1a). Consis-

tent with this, the expression of syntaxin3 was detectable in

human primary keratinocytes and the keratinocyte cell line Ha-

CaT, but not primary dermal fibroblasts (Fig. 1a). UVB irradia-

tion led to keratinocyte cell death, a process in which the early

apoptotic and late apoptotic/necrotic cell populations were clearly

detected in the adherent cells (Fig. 1b, upper; Fig. S1). To investi-

gate whether syntaxin3 is endowed with an extracellular function

similar to epimorphin, that is, secreted from the keratinocytes so

as to protect them from UVB-inflicted damage, we tested for a

possible extracellular release and prosurvival effect in response to

UVB irradiation. To carry out this aim, we generated HaCaT cells

that were stably introduced with T7-tagged syntaxin3 for the pur-

pose of an efficient collection as well as sensitive detection of syn-

taxin3 secreted in the medium. Upon treatment with direct UVB

irradiation (~15 mJ/cm2), which is considered to be equivalent to

or slightly weaker than exposure to normal sunlight (11), the

extracellular release of T7-syntaxin3 from HaCaT keratinocytes

became evident (Fig. 1b, middle). As cytoplasmic b-actin was also

detected in the medium in response to this irradiation, the secre-

tion of syntaxin3 might be primarily a consequence of UVB-

induced cell death. We assume that the release of cytoplasmic syn-

taxin3 is attributed to an effluent from the dying cells, and not to

physiological secretion, as has been shown for epimorphin, given

that a universal blocker of the physiological secretion of leaderless

proteins had no effect on this syntaxin3’s release (Fig. S2).

Intriguingly, we found that the ratio of the cells that do not suc-

cumb to apoptosis was evidently higher in HaCaT cells that stably

expressed syntaxin3 than did not (Fig. 1b). While the difference in

the cell survival ratio between these two cell types was consistently

ª 2013 John Wiley & Sons A/S. Published by John Wiley & Sons LtdExperimental Dermatology, 2013, 22, 832–849 845

reproducible, the actual cell number varied between experiments,

due to a difficulty in controlling the dose of direct UVB-irradia-

tion.

Anti-apoptotic function of extracellularly released syntaxin3We thus investigated whether extracellular syntaxin3 would have an

impact on the survival of keratinocytes treated with H2O2, in substi-

tution for ROS (12), a key hazardous apoptosis-trigger induced by

UVB irradiation (13). As the membrane-spanning and SNARE

domains in epimorphin and syntaxin4 have been demonstrated to

be dispensable for their extracellular activity (9), a recombinant

form of syntaxin3 lacking these domains was directly added to the

culture. We found that recombinant syntaxin3 as well as epimor-

phin conferred a survival effect upon not only HaCaT cells, but also

primary keratinocytes damaged by H2O2 (Fig. 2a). In these cells, the

processing of the crucial effector for apoptotic cell death, caspase-3

was completely blocked in accord with a dramatic reduction in the

cell population exhibiting apoptotic cell death (Fig. 2b and c). This

demonstrated that the extracellularly supplied syntaxin3 counter-

acted the H2O2-induced apoptosis in these keratinocytes. Taking

these results together, we propose a novel self-protection mecha-

nism in the epidermal keratinocyte against external stimuli, which is

that the extracellular syntaxin3 released by damaged keratinocytes

attenuates the ROS-triggered apoptotic cell-death programme in

adjacent cells (Fig. 2d).

ConclusionMany previous investigations have focused on a small number of

well-known extracellular signalling molecules that exert a prosurvival

effect on keratinocytes (14,15), cytoplasmic proteins may also need to(a)

(b)

sup

cell

UVBUVB0 5 15 0 5 15

sup

IP : anti-T7 WB: anti-T7-HRP

cell(mJ/cm2 )

- T7-STX3(35 kD)

- β-actin(44 kD)

2nd Ab onlySTX3

E

D

E

D

HaCaT

HaCaT-T7-stx3

UVB 0 UVB 15 (mJ/cm2)

UVB 15 (mJ/cm2)

HaCaT-T7-STX3HaCaT- empty

Healthy

early apoptosis

late apoptosisor necrosis

UVB 15 (mJ/cm2)

- STX3(34 kD)

- β-actin(44 kD)

IP : anti- β-actin WB: anti- β-actin -HRP

Figure 1. Expression profile of syntaxin3 in the keratinocyte. (a) Expression ofsyntaxin3 in the mouse dorsal skin (left) and human skin cells, including primarydermal fibroblasts (DF), primary epidermal keratinocytes (NHEK) and HaCaT cells(right). E, epidermis. D, dermis. The secondary antibodies gave non-specific signalsat the outermost layer of the epidermis (2nd Ab only). Bar, 100 lm. b-actin,loading control. (b, upper), ultra-violet B (UVB) irradiation (15 mJ/cm2) resulted inthe appearance of apoptotic, and possibly necrotic, cell populations as judged bythe double staining with Hoechst 33258 and membrane-impermeable PI. Bar,20 lm. Middle, UVB irradiation (5 or 15 mJ/cm2) led to the extracellular release ofcytoplasmic syntaxin3 (STX3) and b-actin from HaCaT cells that are stablyexpressing T7-tagged syntaxin3 (HaCaT-T7-STX3). Lower, the cell staining withHoechst 33258 and PI revealed that the number of apoptotic, and possiblynecrotic, cells decreased when the cells stably expressed syntaxin3. Bar, 20 lm.The circular graphs represent the ratio of healthy (white), early apoptotic (blue) andlate apoptotic/necrotic (red) cells in a typical experiment.

(a)

(b)

+ r-GFP

+ r-STX3+ r-EPM

H2O2

+ r-protein

Caspase3(latent)(35 kD)

Caspase3(active)

(17/19 kD)

β-actin -

NT -

H2O2

24 hours

(c)

+ r-GFP

+ r-STX3

+ r-GFP

+ r-STX3

H2O2

*

UVB

syntaxin3ROS

apoptosis

50 -

40 -

30 -

20 -

10 -

0 -H2O2 0 (mM) 1.0

Ap

op

toti

c ce

lls (

%) + r-GFP

+ r-STX3

**

syntaxin3

necrotic cells

(d)

ROSROS

Figure 2. Anti-apoptotic effect of syntaxin3 in HaCaT cells and normalkeratinocytes. (a, left), H2O2 (1 mM) led to apoptotic cell death, which was blockedby recombinant syntaxin3 (r-STX3) or epimorphin (r-EPM), but not GFP (r-GFP;20 lg/ml). Bar, 500 lm. Right, the quantification of the surviving keratinocytes.The results are expressed as the mean � SD n = 6, *P < 0.01. (b, left), HaCaT cellsunderwent apoptosis in response to the addition of 1 mM H2O2, which wasblocked by recombinant syntaxin3 (+r-STX3), but not by GFP control (+GFP). Right,typical images of cells stained with Hoechst 33258 and PI. Chromosomefragmentation/condensation in many cells was evident in the control (+GFP;arrows), but not in the STX3-treated cells (+STX3). In STX3-treated cells, evenmitotic cells were observed (*). The enlarged images of the dotted areas in themiddle panels are shown in the right. (c) the activation of caspase3 by H2O2 wascompletely blocked by r-EPM or r-STX3, but not by r-GFP. Bar, 20 lm. (d)Schematic model. Ultra-violet B (UVB) irradiation induces reactive oxidative speciesproduction in keratinocytes, leading to cell death and effluent syntaxin3, the latterexerting an intrinsic anti-apoptotic function.

846ª 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Experimental Dermatology, 2013, 22, 832–849

Letter to the Editor

be taken into account, considering that dying cell populations dis-

charge certain cytosolic components extracellularly (16). We propose

to add cytoplasmic syntaxin3, which is discharged in this manner

from the dying cells, as a powerful protector against apoptosis in epi-

dermal keratinocytes. These results lead to new insight into clinical

prevention and/or treatment of UVB-damaged skin.

AcknowledgementsPart of this work was supported by Grant-in Aid for Scientific Research

(KAKENHI 24590365).

Authors’ contributionsTM, NK, YK, and YH performed the research. YM and TM designed the

research. KM and NH contributed essential reagents and reviewed the

data. YH wrote the manuscript. All authors approved this paper.

Ethics approvalExperiments with mice skin were approved by the animal care committee

of Kwansei Gakuin University (2013-09).

Conflicts of InterestThe authors have declared no conflicting interest.

References1 Kirschner N, Brandner J M. Ann N Y Acad Sci

2012: 1257: 158–166.2 Morita K, Miyachi Y, Furuse M. Eur J Dermatol

2011: 21: 12–17.3 Proksch E, Brandner J M, Jensen J M. Exp Der-

matol 2008: 17: 1063–1072.4 Lippens S, Hoste E, Vandenabeele P et al.

Apoptosis 2009: 14: 549–569.5 Hirai Y, Nelson C M, Yamazaki K et al. J Cell

Sci 2007: 120: 2032–2043.6 Chen C S, Nelson C M, Khauv D et al. J Biol

Chem 2009: 284: 6877–6884.7 Lehnert L, Lerch M M, Hirai Y et al. J Cell Biol

2001: 152: 911–922.

8 Iizuka M, Sasaki K, Hirai Y et al. Am J PhysiolGastrointest Liver Physiol 2007: 292: G39–G52.

9 Kadono N, Miyazaki T, Okugawa Y et al. Bio-chem Biophys Res Commun 2012: 417: 1200–1205.

10 Okugawa Y, Hirai Y. J Invest Dermatol 2008:128: 1884–1893.

11 Ahn B N, Kim J A, Kong C S et al. Exp Derma-tol 2013: 22: 155–157.

12 Dhumrongvaraporn A, Chanvorachote P. J Cos-met Sci 2013: 64: 207–217.

13 Chang H, Oehrl W, Elsner P et al. Free RadicalRes 2003: 37: 655–663.

14 Muller E J, Williamson L, Kolly C et al. J InvestDermatol 2008: 128: 501–516.

15 Shirakata Y. J Dermatol Sci 2010: 59: 73–80.16 Kim H J, Song S B, Yang Y et al. Exp Dermatol

2011: 20: 920–925.

Supporting InformationAdditional Supporting Information may be found inthe online version of this article:Figure S1. Assessment of cell types by a double stain-

ing with Hoechst33258 and membrane-impermeable PI.Figure S2. The secretion profile of epimorphin or

syntaxin3 expressed in HaCaT cells.Appendix S1. Methods.

DOI: 10.1111/exd.12279

www.wileyonlinelibrary.com/journal/EXDLetter to the Editor

Fructose 1, 6-diphosphate regulates desmosomal proteins andcollagen fibres in human skin equivalents

Hyun Choi1, Seung Ha Yang1, Il-Hong Bae2, Ju-Yearl Park1, Hyoung-June Kim1, Minsoo Noh3,Tae Ryong Lee1 and Dong Wook Shin1

1Bioscience Research Institute, Amorepacific Corporation R&D Center, Yongin, Gyeonggi-do, Korea; 2Medical Beauty Research Institute,

AmorePacific Corporation R&D Center, Yongin, Gyeonggi-do, Korea; 3College of Pharmacy, Seoul National University, Seoul, Korea

Correspondence: Dongwook Shin, PhD, AmorePacific Corporation R&D Center, 314-1, Bora-dong, Giheung-gu, Yongin-si, Gyeonggi-do 446-729,

Korea, Tel.: 82-31-280-5968, Fax: 82-31-899-2595, e-mail: biopang@amorepacific.com; and

Tae Ryong Lee, PhD, Amorepacific Corporation R&D Center, 314-1, Bora-dong, Giheung-gu, Yongin-si, Gyeonggi-do 446-729, Korea, Tel.: 82-31-

280-5850, Fax: 82-31-899-2595, e-mail: TRLee@amorepacific.com

Abstract: We previously reported that fructose 1,6-diphosphate

(FDP), a glycolytic metabolite, alleviates ultraviolet B-induced

oxidative skin damage. Here, we further examined the effects of

FDP on skin. FDP decreased the number of desmosomes, whereas

it increased collagen fibres in skin equivalents (SEs). FDP

significantly decreased the expression of corneodesmosomal

components such as desmoglein 1 (DSG1), desmocollin 1 (DSC1)

and corneodesmosin (CDSN), and desquamation-related

proteases, kallikrein 5 (KLK 5) and kallikrein 7 (KLK7) in normal

human epidermal keratinocytes (NHEKs). In addition, FDP

treatment increased the phosphorylation of p38 MAPK, but the

decreased expression of corneodesmosomal components is not

recovered by the treatment of p38 MAPK inhibitors. Interestingly,

FDP diminished the amplitude of Ca2+ fluxes through down-

regulation of SERCA2. Taken together, these results suggested that

FDP induced a decrease in desmosomes and an increase in

collagen fibres similar to the process of chemical peeling, the most

common treatments for ageing skin.

Abbreviations: CDSN, corneodesmosin; DSC1, desmocollin 1; DSG1,

desmoglein 1; FDP, fructose 1,6-diphosphate; KLK5, kallikrein 5; KLK7,

kallikrein 7; NHEKs, normal human epidermal keratinocytes; NHDFs,

normal human dermal fibroblasts; SC, stratum corneum; SE, skin

equivalents; UVB, ultraviolet B.

Key words: collagen – desmosome – fructose 1,6-diphosphate – normal

human epidermal keratinocytes – skin equivalents

Accepted for publication 27 October 2013

ª 2013 John Wiley & Sons A/S. Published by John Wiley & Sons LtdExperimental Dermatology, 2013, 22, 832–849 847

Letter to the Editor