Sie sind auf Seite 1von 9

European Journal of Pharmacology 786 (2016) 128–136

Contents lists available at ScienceDirect

European Journal of Pharmacology


journal homepage: www.elsevier.com/locate/ejphar

Molecular and cellular pharmacology

Pharmacological activation of cannabinoid 2 receptor attenuates


inflammation, fibrogenesis, and promotes re-epithelialization during
skin wound healing
Lin-Lin Wang a,1, Rui Zhao a,1, Jiao-Yong Li a, Shan-Shan Li a, Min Liu a, Meng Wang a,
Meng-Zhou Zhang a, Wen-Wen Dong a, Shu-Kun Jiang a, Miao Zhang a, Zhi-Ling Tian a,
Chang-Sheng Liu b, Da-Wei Guan a,n
a
Department of Forensic Pathology, China Medical University School of Forensic Medicine, Shenyang, China
b
Institute of Forensic Science, Anshan Municipal People's Procuratorate, Anshan, China

art ic l e i nf o a b s t r a c t

Article history: Previous studies showed that cannabinoid 2 (CB2) receptor is expressed in multiple effector cells during
Received 11 January 2016 skin wound healing. Meanwhile, its functional involvement in inflammation, fibrosis, and cell pro-
Received in revised form liferation in other organs and skin diseases implied CB2 receptor might also regulate skin wound healing.
2 June 2016
To verify this hypothesis, mice excisional wounds were created and treated with highly selective CB2
Accepted 2 June 2016
receptor agonist GP1a (1-(2,4-dichlorophenyl)-6-methyl- N-piperidin-1-yl-4H-indeno[1,2-c]pyrazole-3-
Available online 3 June 2016
carboxamide) and antagonist AM630 ([6-iodo-2- methyl-1-(2-morpholin-4-ylethyl)indol-3-yl]-(4-
Keywords: methoxyphenyl)methanone) respectively. The inflammatory infiltration, cytokine expression, fibrogen-
Cannabinoid 2 receptor esis, and wound re-epithelialization were analyzed. After CB2 receptor activation, neutrophil and mac-
Skin wound healing
rophage infiltrations were reduced, and expressions of monocyte chemotactic protein (MCP)-1, stromal
Inflammation
cell-derived factor (SDF)-1, Interleukin (IL)-6, IL-1β, tumor necrosis factor (TNF)-α, transforming growth
Fibrogenesis
Re-epithelialization factor (TGF)-β1 and vascular endothelial growth factor (VEGF)-A were decreased. Keratinocyte pro-
liferation and migration were enhanced. Wound re-epithelialization was accelerated. Fibroblast accu-
Chemical compounds studied in this article: mulation and fibroblast-to-myofibroblast transformation were attenuated, and expression of pro-col-
GP1a (PubChem CID: 10252734) lagen I was decreased. Furthermore, HaCaT cells in vitro were treated with GP1a or AM630, which re-
AM630 (PubChem CID: 4302963) vealed that CB2 receptor activation promoted keratinocyte migration by inducing the epithelial to me-
senchymal transition. These results, taken together, indicate that activating CB2 receptor could ameliorate
wound healing by reducing inflammation, accelerating re-epithelialization, and attenuating scar for-
mation. Thus, CB2 receptor agonist might be a novel perspective for skin wound therapy.
& 2016 Elsevier B.V. All rights reserved.

1. Introduction The endocannabinoid system (ECS; comprising the en-


docannabinoids, G-protein-coupled cannabinoid receptors, bio-
Skin wound healing is a complicated process proceeding by synthetic and metabolism enzymes) in the mammals has attracted
three stages that overlap in time and space: inflammation, new great attention in the last decades for its multiple functions both in
tissue formation, and remodeling. During this process, a series of health and diseases, and the ECS of the skin is recognized as a
cytokines, growth factors, extracellular matrix molecules, and novel therapeutic target for cutaneous pathologies (Biro et al.,
various cells including leukocytes, fibroblasts, and keratinocytes 2009; Maccarrone et al., 2015). Cannabinoid 2 (CB2) receptor is an
coordinately interacted with each other to reform the epidermal important constituent of the ECS, which has been increasingly
barrier and dermal structure (Gurtner et al., 2008; Martin, 1997; emphasized in recent years because it doesn’t arouse psychotropic
Martin and Nunan, 2015). and cardiovascular side-effects (Yrjölä et al., 2015). CB2 receptor is
mainly distributed in peripheral and immune cells, including
lymphocytes, NK cells, monocytes, macrophages, fibroblasts, ker-
n
Corresponding author.
atinocytes, etc., and functionally involved in a series of pathologies
E-mail addresses: dwguan@mail.cmu.edu.cn,
dwguan@aliyun.com (D.-W. Guan). referring to these cells (Battista et al., 2012; Kupczyk et al., 2009;
1
These authors have contributed equally to the work. Munro et al., 1993; Witkamp and Meijerink, 2014). Previous

http://dx.doi.org/10.1016/j.ejphar.2016.06.006
0014-2999/& 2016 Elsevier B.V. All rights reserved.
L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136 129

studies have revealed that CB2 receptor interruption results in immunohistochemical staining of myeloperoxidase (MPO), F4/80,
enhanced allergic inflammation in cutaneous contact hypersensi- von Willebrand factor (vWF), proliferating cell nuclear antigen
tivity (Karsak et al., 2007), while activating CB2 receptor reduces (PCNA), E-cadherin, and vimentin; as well as triple immuno-
inflammation in murine pancreatitis (Michler et al., 2013), cystitis fluorescent staining of CD45/pro-collagen I/α-smooth muscle actin
(Wang et al., 2014), traumatic brain injury (Amenta et al., 2014), (SMA). mRNA expressions of monocyte chemotactic protein
and sepsis (Tschop et al., 2009). CB2 receptor activation also at- (MCP)-1, stromal cell-derived factor (SDF)-1, tumor necrosis factor
tenuates pathological liver fibrosis in mice (Guillot et al., 2014; (TNF)-α, transforming growth factor (TGF)-β1, epidermal growth
Mahmoud et al., 2014; Munoz-Luque et al., 2008), and exerts an- factor (EGF), and keratinocyte growth factor (KGF) were quantified
tifibrotic effect in bleomycin-induced dermal fibrosis (Akhmet- by qRT-PCR. Protein expressions of interleukin (IL)-6, IL-1β, in-
shina et al., 2009). In addition, absence of CB2 receptor improves terferon (IFN)-γ, vascular endothelial growth factor (VEGF)-A, and
keratinocyte differentiation and decreases proliferation in a tape- pro-collagen I were assayed by Western blotting. See details in
stripping-induced epidermal barrier abrogation model (Roelandt Supplementary materials and methods.
et al., 2012).
As aforementioned, CB2 receptor is involved in the regulation of 2.4. Cell culture and in vitro wound healing study
inflammation, fibrogenesis, and epidermal growth, which are the
key biological processes during skin wound healing. Taking into HaCaT cells were cultured as described previously (Zhao et al.,
account our previous finding that expression of CB2 receptor was 2013). Cells were treated with vehicle (0.1% DMSO), Gp1a (40 nM
up-regulated during skin wound healing in mice (Zheng et al., to 25 μM, dissolved in vehicle), and AM630 (40 nM to 25 μM,
2012), we presumed CB2 receptor might regulate skin wound dissolved in vehicle) respectively. Wound scratch assay was per-
healing, and pharmacomodulation of its activity might be a novel formed. Cell viability was measured by cell counting kit (CCK)-8
perspective for wound therapy. In the present study, we verified assay. Protein expressions of involucrin, E-cadherin, vimentin,
this hypothesis using Gp1a (highly selective CB2 receptor agonist) PCNA, and cleaved caspase 3 were measured by Western blotting.
and AM630 (highly selective CB2 receptor antagonist) in mice skin Co-localization of E-cadherin and vimentin was performed by
excisional wounds in vivo and HaCaT keratinocytes in vitro. Our immunofluorescent staining. See details in Supplementary mate-
results indicate that activating CB2 receptor could ameliorate skin rials and methods.
wound healing by attenuating inflammation and fibrogenesis, as
well as promoting re-epithelialization. 2.5. Statistical analysis

Data were expressed as means 7 standard deviation and ana-


2. Materials and methods lyzed using SPSS for Windows 13.0. The one-way ANOVA with a
Bonferroni post hoc test was used for data analysis. Difference
2.1. Reagents and Abs associated with Po 0.05 was considered statistically significant.

See details in Supplementary materials and methods.


3. Results
2.2. Wound model and experimental grouping
3.1. The general morphology changes after CB2 receptor modulation
Eight-weeks-old male BALB/c mice, each weighing 20–25 g,
were used in this study. Wound model was established according To investigate the regulatory function of CB2 receptor during
to the instruction of Birch et al. (2005) with slight modification skin wound healing, we firstly detected the general morphology
(see details in Supplementary materials and methods). Two 6 mm- changes after the agonist/antagonist treatment. Macroscopically,
diameter full-thickness dermal excisional wounds were created CB2 receptor activation by GP1a inhibited wound contraction after
symmetrically on the dorsal skin of each mouse. decrustation (Fig. 1A, B). The original wound sizes (measured by
After surgery, mice were randomly divided into three groups tracing the dermal border of the wound) in GP1a group were
and daily intraperitoneal injected with vehicle (5% DMSO/2% larger than vehicle group at 10–21 d post-injury. No significant
tween-80/physiological saline, 2.5 μl/g), Gp1a (3 mg/kg, dissolved difference was observed between AM630 and vehicle group. Mi-
in vehicle), or AM630 (3 mg/kg, dissolved in vehicle) respectively. croscopically, wound re-epithelialization was accelerated in GP1a
Mice were euthanized at 12 h, 1 d, 3 d, 5 d, 7 d, 10 d, 13 d, 17 d and group (Fig. 1C, D). The epithelial sheets were longer at 3–10 d, with
21 d post-injury (6 mice at each time point). The original wound larger cross-sectional areas at 5 d and 7 d post-injury. The in-
sizes (measured by tracing the dermal border of the wound) were flammatory infiltration and granulation formation in the wound
analyzed. Six mice without surgery were used as control. Two cavity were apparently decreased as compared with vehicle group.
1 cm  1 cm skin specimens centered on the wound were col- In AM630 group, wound re-epithelialization was delayed. The
lected from each mouse. One specimen was used for morpho- epithelial sheet length and cross-sectional area were both declined
metric, and another for Western blotting and real-time quantita- at 5 d and 7 d post-injury.
tive PCR (qRT-PCR) analyses. Keratinocyte proliferation continues for many days after com-
Experiments were conformed to the Guide for the Care and Use plete re-epithelialization, resulting in hypertrophic epidermis that
of Laboratory Animals (National Institutes of Health published is about four times as thick as unwounded skin (Gallant-Behm
no.86-23, revised 1985), and the Guidelines for the Care and Use of et al., 2011). Similar appearances were observed in the present
Laboratory Animals of China Medical University. study. At 21 d post-injury, the epidermal thicknesses in three
groups were all elevated as compared with the normal skin.
2.3. Morphological analyses of wound re-epithelialization, in- However, the epidermal hypertrophy in GP1a group was greatly
flammatory infiltration, fibrogenesis, and qRT-PCR and Western ameliorated when compared with vehicle group (Fig. 2A, C). In
blotting assays addition, the dermal scar in GP1a group was thinner and the
collagen fibers were much slenderer (Fig. 2B, D). AM630 group
For morphometrical analyses, wound sections were per showed similar epidermal and dermal morphologies with vehicle
formed with H&E-staining; Masson's trichrome-staining; group.
130 L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136

Fig. 1. General morphology changes after GP1a/AM630 treatment. (A) Representative macrophotographs of the wounds at different posttraumatic intervals. Bar ¼5 mm.
(B) Percentage of original wound size, N ¼6. *Po 0.05, GP1a versus vehicle; #Po 0.05, AM630 versus vehicle. (C) Microphotographs of H&E-stained wound sections at 3 d, 7 d
and 10 d post-injury. Bar¼ 1 mm. Arrows represent the dermal border; arrowheads represent the epidermal margin. (D) Rate of wound re-epithelialization, and mea-
surement of epithelial sheet length, cross section area and average thickness. N ¼ 6. *Po 0.05, GP1a versus vehicle; #Po 0.05, AM630 versus vehicle.

These morphology changes, taken together, highly indicated A were decreased in GP1a group and increased (VEGF-A un-
that CB2 receptor was functionally involved in re-epithelialization, changed) in AM630 group. Protein expression of IFN-γ showed a
inflammation, and fibrogenesis. Thus, further experiments were different pattern that was decreased after injury. This result was
carried out in these aspects. similar to the finding of Maeda et al. (2011) in a mouse excisional
wound study. IFN-γ is predominantly secreted by NK cells and T
3.2. CB2 receptor activation attenuates inflammation lymphocytes (Farrar and Schreiber, 1993). We presumed that a
large amount of other cytokines secreted by the massively in-
Inflammation was evaluated by neutrophil and macrophage filtrated neutrophils and macrophages resulted in relative de-
infiltrations, as well as cytokine expressions. Neutrophils and crease in quantity of IFN-γ at the early inflammation stage. The
macrophages were identified by MPO and F4/80 immunostaining relatively increased IFN-γ expression in GP1a group and decreased
(Fig. S1). In GP1a group, neutrophil content was decreased at 3– expression in AM630 group as compared with vehicle group
7 d, and macrophage content was decreased at 1–13 d (Fig. 3A). In (Fig. 3C) were consistent with their dynamics in neutrophil and
AM630 group, neutrophil and macrophage contents were in- macrophage infiltration. The aforementioned results, taken to-
creased at 3–5 d and 3–10 d respectively. gether, highly indicated that wound inflammation was attenuated
mRNA levels of MCP-1, SDF-1, TNF-α, TGF-β1, EGF, and KGF after CB2 receptor activation.
were measured by qRT-PCR (Fig. 3B), and protein expressions of IL-
6, IL-1β, IFN-γ, and VEGF-A were measured by Western blotting 3.3. CB2 receptor activation attenuates fibrogenesis
(Fig. S2, Fig. 3C). In general, mRNA expressions of MCP-1, SDF-1,
TNF-α and TGF-β1 were decreased in GP1a group, whereas in- Fibroblasts (pro-col I þ ), myofibroblasts (pro-col I þ /α-SMA þ ),
creased in AM630 group. mRNA expression of EGF was not sig- and fibrocytes (CD45 þ /pro-col I þ ) were simultaneously detected
nificantly changed, and that of KGF was decreased after GP1a by triple immunofluorescent staining (Fig. 4A). In GP1a group, fi-
treatment at 10–21 d. Protein expressions of IL-6, IL-1β, and VEGF- broblast content was significantly reduced at 5–21 d (Fig. 4B). The
L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136 131

Fig. 2. Wound morphology at 21 d post-injury. (A) Representative microphotographs of epidermis in H&E-stained wound sections. Bar ¼50 mm. (B) Representative mi-
crophotographs of Masson's-stained wound sections. Bar ¼1 mm. (C) Analysis of the epidermal thickness at 21 d post-injury & Po 0.05, wound epithelium versus normal
skin epithelium. *Po 0.05, GP1a versus vehicle; #P o0.05, AM630 versus vehicle. (D) Analysis of the dermal thickness. *Po 0.05, GP1a versus vehicle; #P o0.05, AM630
versus vehicle.

Fig. 3. Inflammatory infiltration and cytokine expression changes after GP1a/AM630 treatment. (A) Cell counting of neutrophils and macrophages in the wounds. NS
represents the results obtained from normal skin as control. (B) mRNA levels of MCP-1, SDF-1, TNF-α, TGF-β1, EGF, and KGF. N ¼6. *P o 0.05, GP1a versus vehicle; #P o0.05,
AM630 versus Vehicle. (C) Protein expressions of IL-6, IL-1β, IFN-γ, and VEGF-A. N ¼6. *P o0.05, GP1a versus Vehicle; #Po 0.05, AM630 versus Vehicle.
132 L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136

Fig. 4. Fibrogenesis alteration after GP1a/AM630 treatment. (A) Representative immunofluorescence stains of CD45/pro-col I/α-SMA in the wound sections at 7 d and 10 d
post-injury. Enlarged areas show the fibrocytes (arrows). Bar ¼ 100 mm. (B) Cell counting of fibroblasts, myofibroblasts, and fibrocytes in the wounds. N ¼ 6. *Po 0.05, GP1a
versus Vehicle; #Po 0.05, AM630 versus Vehicle. (C and D) Representative immunoblotting results and quantitative analysis of the expression of pro-collagen I at different
posttraumatic intervals. N¼ 6. *Po 0.05, GP1a versus Vehicle; #P o0.05, AM630 versus Vehicle.

ratios of fibroblast-to-myofibroblast differentiation (generally de- 3.4. CB2 receptor activation promotes re-epithelialization in vivo
fined by the ratios of myofibroblasts to fibroblasts) were decreased
(23.5 75.3% at 5 d; 38.9 712.2% at 7 d; 47.67 5.3% at 10 d) as In the present study, the cross-sectional area of epithelial sheet
compared with vehicle (31.675.1% at 5 d; 63.4 77.2% at 7 d; in GP1a group was larger than vehicle at 5–7 d post-injury. Fur-
62.7 78.6% at 10 d). Correspondingly, the myofibroblast content thermore, the epithelial sheet was extended to a higher degree
and even resulted in decreased epidermal thickness (Fig. 1C, D).
was much more reduced in Gp1a group (Fig. 4B). In AM630 group,
These morphologies implied that both the proliferation and mi-
the fibroblast and myofibroblast contents were not significantly
gration of keratinocyte were enhanced. During skin wound heal-
changed as compared with vehicle group.
ing, keratinocytes at the wound margin undergo partial epithelial
Fibrocytes are bone marrow-derived progenitor cells that co-
to mesenchymal transition (EMT), losing intercellular adhesions
express hematopoietic and mesenchymal markers, possess in- with each other and acquiring mesenchymal property that facil-
flammatory properties, and have potential to differentiate into itate their migration (Barriere et al., 2015; Hudson et al., 2009;
myofibroblasts (Bucala et al., 1994; Reilkoff et al., 2011). There is a Lamouille et al., 2014). Therefore, besides proliferation marker
controversy on whether or not fibrocytes participate in skin PCNA, the classic EMT markers E-cadherin and vimentin were
wound repair. A few experiments revealed that no fibrocyte was immunostained to further reveal the alteration of keratinocyte
detected during skin wound healing (Barisic-Dujmovic et al., 2010; biology.
Higashiyama et al., 2011). However, many others certified the re- In GP1a group, keratinocytes adjacent to the wound margin
cruitment of fibrocytes after skin injury (Ishii et al., 2005; Mori were less-stratified with PCNA expressed throughout the epithe-
et al., 2005; Ou et al., 2015; Suga et al., 2014). The present study lium. The PCNA þ cell number per unit area was increased at 5–7 d
post-injury (Fig. 5A, B). Expression of E-cadherin was down-
was in support of the latter opinion. In vehicle group, fibrocytes
regulated. Keratinocytes over the basal layer were stretched and
were detected in the wound cavity at 3–17 d post-injury (Fig. 4A,
orientated to the wound center, and their expression of vimentin
B). After GP1a treatment, fibrocyte recruitment was greatly in-
were up-regulated (Fig. 5A, B). In AM630 group, keratinocytes
hibited at 5–10 d. No significant difference was observed in AM630 adjacent to wound margin were more-stratified with PCNA ex-
group as compared with vehicle. In accordance with the reduced pression restricted to the basal layer. The PCNA þ cell number per
contents of fibroblasts, myofibroblasts and fibrocytes, protein ex- unit area was decreased at 5–7 d post-injury (Fig. 5B). The ex-
pression of pro-collagen I was decreased after GP1a treatment pression of E-cadherin was up-regulated, and vimentin down-
(Fig. 4C, D). regulated. These morphologies further certified that GP1a
L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136 133

Fig. 5. Morphology changes of the epithelial sheet after GP1a/AM630 treatment. (A) Representative microphotographs of H&E stained, as well as PCNA, E-cadherin and
vimentin immunostained epithelium adjacent to wound margin at 7 d post-injury. Red dotted lines indicate roughly the boundary between epidermis and dermis.
Bar ¼ 100 mm. (B) Analyses of PCNA þ cell number per unit area, and the expressions of E-cadherin and vimentin at 3 d, 5 d, and 7 d post-injury. N ¼6. *Po 0.05, GP1a versus
Vehicle; #P o0.05, AM630 versus Vehicle. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

promoted, whereas AM630 inhibited keratinocyte proliferation cells changed from a typical cobblestone-like to a spindle-like
and migration. shape, and stretched out pseudopodia. Decreased expression of
E-cadherin and increased expression of vimentin were re-
3.5. CB2 receptor activation promotes re-epithelialization in vitro confirmed by immunofluorescence staining (Fig. 6D). 25 μM GP1a
treatment caused more obvious cell morphology alteration. Lots of
To testify the direct effect of CB2 receptor on keratinocyte cells detached from the cell colony and migrated out with pseu-
biology, further experiments were carried out in HaCaT cells in dopodia (Fig. S3A). However, epithelial cell growth and survival
vitro. Cell proliferation was assessed by CCK-8 assay (Fig. 6A). As critically rely on cell-matrix and cell-cell adhesion, and severe
compared with vehicle group, cell viability was not significantly disruption of these interactions could induce cell programmed
elevated after GP1a treatment at concentrations of 40 nM to 5 μM, death by anoikis (Bretland et al., 2001; Frisch and Francis, 1994). It
while was slightly decreased at the concentration of 25 μM is likely in this manner some detached cells were induced to
(95.0 72.5%). AM630 significantly decreased cell viability at con- death. This effect was supported by the slight increase of cleaved
centrations of 5 μM (94.0 73.0%) and 25 μM (86.7 73.6%). Sub- caspase 3 expression. Meanwhile, it also explained the decrease of
sequently, HaCaT cells were treated respectively with GP1a and PCNA expression (Fig. 6B, C). AM630 dose-dependently increased
AM630 at concentrations of 0.2 μM, 1 μM, 5 μM, and 25 μM. Cell the expressions of involucrin, E-cadherin, and reduced the ex-
morphology alteration and expressions of relevant proteins were pression of PCNA (Fig. 6B, C). The cell morphology was not ob-
detected. viously changed after 5 μM AM630 treatment (Fig. 6D). However,
GP1a dose-dependently down-regulated the expressions of 25 μM AM630 treatment increased the expression of cleaved
involucrin and E-cadherin, while up-regulated the expression of caspase 3 (Fig. 6B, C), and induced cell apoptosis (Fig. S3A). Cor-
vimentin (Fig. 6B, C). Expression of PCNA was not significantly responding results were observed in wound scratch assay. GP1a
changed except for a slight decrease after 25 μM GP1a treatment. accelerated wound closure at concentrations of 1 μM and 5 μM,
Morphologically, after 5 μM GP1a treatment for 24 h, lots of HaCaT while retarded it at the concentration of 25 μM (Fig. 6E).
134 L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136

Fig. 6. HaCaT cell biology alteration after GP1a/AM630 treatment. (A) The cell viability alteration determined by CCK-8 assay. (B and C) Representative immunoblotting
results and quantitative analyses of involucrin, E-cadherin, vimentin, PCNA, and cleaved caspase 3 expressions in HaCaT cells. N ¼ 5. *P o0.05, GP1a versus vehicle; #Po 0.05,
AM630 versus vehicle. (D) Representative phase-contrast microphotographs of HaCaT cells, and immunofluorescent stains of E-cadherin and vimentin. (E) Representative
wound-scratch assay and quantitative analyses of migration ratios at 24 h. Bar ¼ 0.5 mm. N ¼ 5. *Po 0.05, GP1a versus vehicle; #P o0.05, AM630 versus vehicle.

Microscopically, 25 μM GP1a treatment induced cells at the activation of CB2 receptor could ameliorate skin wound healing by
wound margin to migrate out of the cell colony quickly. However, attenuating inflammation and fibrogenesis, and promoting re-
the separated HaCaT lost interactions with each other, and were epithelialization.
induced to anoikis subsequently (Fig. S3B). AM630 at concentra- CB2 receptor has been recognized as a novel therapeutic per-
tions of 5 μM and 25 μM inhibited wound closure (Fig. 6E). spective for various inflammatory diseases (Biro et al., 2009; Klein,
These in vitro results, taken together, indicated that activating 2005; Maccarrone et al., 2015). In this study, we further confirmed
CB2 receptor could inhibit keratinocyte differentiation (involucrin its anti-inflammation property during skin wound healing. After
was decreased) and induce EMT (E-cadherin was decreased and CB2 receptor agonist treatment, neutrophil content was decreased
vimentin was increased), but not promote cell proliferation (PCNA only to a small extent. However, macrophage content was about
was not increased). Inhibiting CB2 receptor could enhance kerati- half declined. This change was accompanied with decreased ex-
nocyte adhesion (E-cadherin was increased), promote differentia-
pression of MCP-1, which is mainly secreted by macrophages and
tion (involucrin was increased), and inhibit proliferation (PCNA
acts as their powerful chemokine in turn (Deshmane et al., 2009).
was decreased).
Macrophages are key-regulators of wound repair that phagocytize
4. Discussion pathogens and necrotic tissue, and serve as important sources of
numerous cytokines and growth factors (Brancato and Albina,
In the present study, we demonstrated that pharmacological 2011; Kondo and Ishida, 2010). In accordance with their reduction,
L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136 135

expressions of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α appearances in vivo, that the wound epithelium was less-stratified
were decreased, thus further attenuated inflammation. Mean- and more extended after CB2 receptor activation.
while, expression of VEGF-A, an essential factor for angiogenesis, During skin wound healing, EGF and KGF are potent promoters
was decreased. This down-regulation did not alter the neovascular of keratinocyte proliferation and migration, while TGF-β1 could
density, but reduced granulation tissue content in the wound reverse activated keratinocytes to the basal cell phenotype and
cavity (Fig. S4). Expression of IFN-γ was relatively increased after inhibit their proliferation (Gurtner et al., 2008; Pastar et al., 2014).
CB2 receptor activation. Since IFN-γ could active macrophages, In the present study, the expressions of EGF and KGF were not
promoting phagocytosis and cytokines production (Farrar and significantly changed in the early stage. Thus, it might be pre-
Schreiber, 1993; Gordon and Taylor, 2005), this effect might act as sumed that activating CB2 receptor accelerated re-epithelialization
a feedback mechanism to enhance macrophage activity in the indirectly by decreasing TGF-β1 expression and directly by pro-
circumstance of decreased leukocyte infiltration. moting keratinocyte migration, while inhibiting CB2 receptor at-
Fibrogenesis is a dynamic process predominantly conducted by tenuated re-epithelialization by increasing TGF-β1 expression and
fibroblasts, and to some extent regulated by inflammatory med- promoting keratinocyte differentiation. We also observed the
iators (Stramer et al., 2007; Wynn and Ramalingam, 2012). TGF-β1 epidermal hypertrophy at the late stage was ameliorated after CB2
is a key driver of fibrosis which stimulates fibroblasts to synthesize receptor agonist treatment. Most probably it was associated with
collagen, and induces fibroblast-to-myofibroblast differentiation. decreased KGF expression which reduced keratinocyte prolifera-
Myofibroblasts possess stronger collagen synthesizing ability and tion. Since fibroblasts are the major effector cells at the late stage
contribute to wound contraction (Duffield et al., 2013; Wynn and and are also important sources of KGF (Martin, 1997), this phe-
Ramalingam, 2012). IL-1β, IL-6, and TNF-α could also promote fi- nomenon might be attributed to reduced fibroblast accumulation
brogenesis by increasing fibroblast proliferation and collagen after CB2 receptor activation.
synthesis (Barnes et al., 2011; Duffield et al., 2013; Zhang et al., There is a notion that in mammals, wound repair is evolutio-
1993). IFN-γ could inhibit fibroblast proliferation, and suppresses narily optimized for speed of healing under dirty conditions,
the production and functional activity of TGF-β1 (Gurujeyalakshmi where a multiply redundant, compensating and rapid in-
and Giri, 1995; Ishida et al., 2004). Since CB2 receptor doesn’t di- flammatory response allows the wound to heal quickly without
rectly modulate collagen synthesis by fibroblasts in vitro (Akh- infection. Scar formation is the price mammalians have to pay for
metshina et al., 2009), we presumed the anti-fibrogenesis effect of their survival after wounding (Bayat et al., 2003). Nowadays, it is
CB2 receptor in the present study was achieved indirectly by increasingly accepted that reducing inflammation might speed
down-regulating IL-1β, IL-6, TNF-α, TGF-β1 expressions and up- wound healing and improve the final scar appearance (Rosique
regulating IFN-γ expression. The reduced wound contraction could et al., 2015). Our present study revealed that activating CB2 re-
be attributed to decreased fibroblast-to-myofibroblast differ- ceptor could curb inflammation and attenuate scar formation.
entiation. Activating CB2 receptor also decreased the expression of Furthermore, CB2 receptor activation could promote keratinocyte
SDF-1, an important chemokine ligand for fibrocyte recruitment migration, rendering it more beneficial for wound re-epitheliali-
(Phillips et al., 2004), which might account for the fibrocyte con- zation. From the perspective of wound therapy, it should be no-
tent reduction, and to a small extent for the fibrosis alleviation. ticed that inflammation per se is important in orchestrating
In the present study we found that CB2 receptor activation wound healing, and an excessive blockade of inflammation would
exerted obvious anti-inflammation and anti-fibrogenesis effects not be a useful strategy. Appropriate collagen deposition is also
during skin wound healing, while inhibition with antagonist necessary for the skin to maintain its mechanical property. Thus,
moderately exaggerated inflammation, and did not exhibit sig- we suggest that a moderate CB2 receptor agonist treatment in the
nificant effect on fibrogenesis. Presumably, CB2 receptor might be early stage, and gradually ceased it at the late stage (for example
activated by the endocannabinoids as endocannabinoids are pro- after decrustation) might be more favorable for skin wound
duced ‘on demand’ from their membrane lipid precursors when healing.
and where needed on pathological or physiological stimuli (Mac- In conclusion, pharmacological activation of CB2 receptor could
carrone et al., 2015). The suppressive action on CB2 receptors by ameliorate wound healing by attenuating inflammation and fi-
AM630 might be partially reversed by the endocannabinoids. brogenesis, as well as promoting re-epithelialization, suggesting
Correspondingly, the expressions of profibrotic cytokines were that CB2 receptor agonist might be a novel perspective for skin
slightly increased, which did not give rise to a significant wound therapy.
fibrogenesis.
Previous knowledge about the effect of CB2 receptor on non-
tumorigenic keratinocytes is limited and controversial. It was re- Acknowledgments
ported that mixed CB1/CB2 receptor agonist WIN-55,212-2 at a low
concentration (25 nM) (Casanova et al., 2003) and CB2 receptor The study was financially supported by Grants from projects
antagonist AM630 at a low concentration (1 μM) (Toth et al., 2011) funded by National Natural Science Foundation of China (Grant No.
were ineffective in modulating keratinocyte growth in vitro. 81273342, 81372943, 81102156), Shenyang Scientific and Techno-
However, CB2 receptor knockout mice showed improved epi- logical Plan (F12-277-1-03), and Anshan Scientific and Technolo-
dermal differentiation and decreased proliferation (Roelandt et al., gical Plan (2060499).
2012). The present study provided an interpretation to this con-
troversy, and further expanded our cognition about CB2 receptor.
CB2 receptor inhibition by low concentrations of AM630 (40 nM to Appendix A. Supplementary material
1 μM) did not affect HaCaT cell proliferation or differentiation.
However, high concentrations of AM630 treatment attenuated cell Supplementary data associated with this article can be found
proliferation and induced differentiation. These results were co- in the online version at http://dx.doi.org/10.1016/j.ejphar.2016.06.
incident with the findings in CB2 receptor knockout mice. CB2 006.
receptor agonist GP1a did not promote HaCaT cell proliferation,
but dose-dependently inhibited its differentiation. Furthermore, it References
induced keratinocytes to take place EMT and promoted cell mi-
gration. These results were coinciding with the morphological Akhmetshina, A., Dees, C., Busch, N., Beer, J., Sarter, K., Zwerina, J., Zimmer, A.,
136 L.-L. Wang et al. / European Journal of Pharmacology 786 (2016) 128–136

Distler, D., Schett, G., Distler, J.H.W., 2009. The cannabinoid receptor CB2 exerts mesenchymal transition. Nat. Rev. Mol. Cell Biol. 15, 178–196.
antifibrotic effects in experimental dermal fibrosis. Arthritis Rheumatol. 60, Maccarrone, M., Bab, I., Biro, T., Cabral, G.A., Dey, S.K., Di Marzo, V., Konje, J.C.,
1129–1136. Kunos, G., Mechoulam, R., Pacher, P., Sharkey, K.A., Zimmer, A., 2015. En-
Amenta, P.S., Jallo, J.I., Tuma, R.F., Hooper, D.C., Elliott, M.B., 2014. Cannabinoid re- docannabinoid signaling at the periphery: 50 years after THC. Trends Phar-
ceptor type-2 stimulation, blockade, and deletion alter the vascular in- macol. Sci. 36, 277–296.
flammatory responses to traumatic brain injury. J. Neuroinflamm. 11, 191. Maeda, S., Fujimoto, M., Matsushita, T., Hamaguchi, Y., Takehara, K., Hasegawa, M.,
Barisic-Dujmovic, T., Boban, I., Clark, S.H., 2010. Fibroblasts/myofibroblasts that 2011. Inducible costimulator (ICOS) and ICOS ligand signaling has pivotal roles
participate in cutaneous wound healing are not derived from circulating pro- in skin wound healing via cytokine production. Am. J. Pathol. 179, 2360–2369.
genitor cells. J. Cell. Physiol. 222, 703–712. Mahmoud, M.F., Swefy, S.E., Hasan, R.A., Ibrahim, A., 2014. Role of cannabinoid
Barnes, T.C., Anderson, M.E., Moots, R.J., 2011. The many faces of interleukin-6: the receptors in hepatic fibrosis and apoptosis associated with bile duct ligation in
role of IL-6 in inflammation, vasculopathy, and fibrosis in systemic sclerosis. Int. rats. Eur. J. Pharmacol. 742, 118–124.
J. Rheumatol. 2011, 1–6. Martin, P., 1997. Wound healing – aiming for perfect skin regeneration. Science 276,
Barriere, G., Fici, P., Gallerani, G., Fabbri, F., Rigaud, M., 2015. Epithelial mesenchy- 75–81.
mal transition: a double-edged sword. Clin. Transl. Med. 4, 14. Martin, P., Nunan, R., 2015. Cellular and molecular mechanisms of repair in acute
Battista, N., Di Tommaso, M., Bari, M., Maccarrone, M., 2012. The endocannabinoid and chronic wound healing. Br. J. Dermatol. 173, 370–378.
system: an overview. Front. Behav. Neurosci. 6, 9. Michler, T., Storr, M., Kramer, J., Ochs, S., Malo, A., Reu, S., Goke, B., Schafer, C., 2013.
Bayat, A., McGrouther, D.A., Ferguson, M.W., 2003. Skin scarring. BMJ 326, 88–92. Activation of cannabinoid receptor 2 reduces inflammation in acute experi-
Birch, M., Tomlinson, A., Ferguson, M.W., 2005. Animal models for adult dermal mental pancreatitis via intra-acinar activation of p38 and MK2-dependent
wound healing. Methods Mol. Med. 117, 223–235. mechanisms. Am. J. Physiol. Gastrointest. Liver Physiol. 304, G181–G192.
Biro, T., Toth, B.I., Hasko, G., Paus, R., Pacher, P., 2009. The endocannabinoid system Mori, L., Bellini, A., Stacey, M., Schmidt, M., Mattoli, S., 2005. Fibrocytes contribute
of the skin in health and disease: novel perspectives and therapeutic oppor- to the myofibroblast population in wounded skin and originate from the bone
tunities. Trends Pharmacol. Sci. 30, 411–420. marrow. Exp. Cell Res. 304, 81–90.
Brancato, S.K., Albina, J.E., 2011. Wound macrophages as key regulators of repair. Munoz-Luque, J., Ros, J., Fernandez-Varo, G., Tugues, S., Morales-Ruiz, M., Alvarez, C.
Am. J. Pathol. 178, 19–25. E., Friedman, S.L., Arroyo, V., Jimenez, W., 2008. Regression of fibrosis after
Bretland, A.J., Lawry, J., Sharrard, R.M., 2001. A study of death by anoikis in cultured chronic stimulation of cannabinoid CB2 receptor in cirrhotic rats. J. Pharmacol.
epithelial cells. Cell Prolif. 34, 199–210. Exp. Ther. 324, 475–483.
Bucala, R., Spiegel, L.A., Chesney, J., Hogan, M., Cerami, A., 1994. Circulating fi- Munro, S., Thomas, K.L., Abu-Shaar, M., 1993. Molecular characterization of a per-
brocytes define a new leukocyte subpopulation that mediates tissue repair. ipheral receptor for cannabinoids. Nature 365, 61–65.
Mol. Med. 1, 71–81. Ou, L., Shi, Y., Dong, W., Liu, C., Schmidt, T.J., Nagarkatti, P., Nagarkatti, M., Fan, D., Ai,
Casanova, M.L., Blázquez, C., Martínez-Palacio, J., Villanueva, C., Fernández-Aceñero, W., 2015. Kruppel-like factor KLF4 facilitates cutaneous wound healing by
M.J., Huffman, J.W., Jorcano, J.L., Guzmán, M., 2003. Inhibition of skin tumor promoting fibrocyte generation from myeloid-derived suppressor cells. J. In-
growth and angiogenesis in vivo by activation of cannabinoid receptors. J. Clin. vestig. Dermatol. 135, 1425–1434.
Investig. 111, 43–50. Pastar, I., Stojadinovic, O., Yin, N.C., Ramirez, H., Nusbaum, A.G., Sawaya, A., Patel, S.
Deshmane, S.L., Kremlev, S., Amini, S., Sawaya, B.E., 2009. Monocyte chemoat- B., Khalid, L., Isseroff, R.R., Tomic-Canic, M., 2014. Epithelialization in wound
tractant protein-1 (MCP-1): an overview. J. Interferon Cytokine Res. 29, healing: a comprehensive review. Adv. Wound Care 3, 445–464.
313–326. Phillips, R.J., Burdick, M.D., Hong, K., Lutz, M.A., Murray, L.A., Xue, Y.Y., Belperio, J.A.,
Duffield, J.S., Lupher, M., Thannickal, V.J., Wynn, T.A., 2013. Host responses in tissue Keane, M.P., Strieter, R.M., 2004. Circulating fibrocytes traffic to the lungs in
repair and fibrosis. Annu. Rev. Pathol. 8, 241–276. response to CXCL12 and mediate fibrosis. J. Clin. Investig. 114, 438–446.
Farrar, M.A., Schreiber, R.D., 1993. The molecular cell biology of interferon-gamma Reilkoff, R.A., Bucala, R., Herzog, E.L., 2011. Fibrocytes: emerging effector cells in
and its receptor. Annu. Rev. Immunol. 11, 571–611. chronic inflammation. Nat. Rev. Immunol. 11, 427–435.
Frisch, S.M., Francis, H., 1994. Disruption of epithelial cell-matrix interactions in- Roelandt, T., Heughebaert, C., Bredif, S., Giddelo, C., Baudouin, C., Msika, P., Ro-
duces apoptosis. J. Cell Biol. 124, 619–626. seeuw, D., Uchida, Y., Elias, P.M., Hachem, J., 2012. Cannabinoid receptors 1 and
Gallant-Behm, C.L., Du, P., Lin, S.M., Marucha, P.T., DiPietro, L.A., Mustoe, T.A., 2011. 2 oppositely regulate epidermal permeability barrier status and differentiation.
Epithelial regulation of mesenchymal tissue behavior. J. Investig. Dermatol. 131, Exp. Dermatol. 21, 688–693.
892–899. Rosique, R.G., Rosique, M.J., Farina, J.J., 2015. Curbing inflammation in skin wound
Gordon, S., Taylor, P.R., 2005. Monocyte and macrophage heterogeneity. Nat. Rev. healing: a review. Int. J. Inflamm. 2015, 316235.
Immunol. 5, 953–964. Stramer, B.M., Mori, R., Martin, P., 2007. The inflammation–fibrosis link? A jekyll
Guillot, A., Hamdaoui, N., Bizy, A., Zoltani, K., Souktani, R., Zafrani, E.S., Mallat, A., and hyde role for blood cells during wound repair. J. Investig. Dermatol. 127,
Lotersztajn, S., Lafdil, F., 2014. Cannabinoid receptor 2 counteracts interleukin- 1009–1017.
17-induced immune and fibrogenic responses in mouse liver. Hepatology 59, Suga, H., Rennert, R.C., Rodrigues, M., Sorkin, M., Glotzbach, J.P., Januszyk, M., Fu-
296–306. jiwara, T., Longaker, M.T., Gurtner, G.C., 2014. Tracking the elusive fibrocyte:
Gurtner, G.C., Werner, S., Barrandon, Y., Longaker, M.T., 2008. Wound repair and identification and characterization of collagen-producing hematopoietic line-
regeneration. Nature 453, 314–321. age cells during murine wound healing. Stem Cells 32, 1347–1360.
Gurujeyalakshmi, G., Giri, S.N., 1995. Molecular mechanisms of antifibrotic effect of Toth, B.I., Dobrosi, N., Dajnoki, A., Czifra, G., Olah, A., Szollosi, A.G., Juhasz, I., Su-
interferon gamma in bleomycin-mouse model of lung fibrosis: downregulation gawara, K., Paus, R., Biro, T., 2011. Endocannabinoids modulate human epi-
of TGF-beta and procollagen I and III gene expression. Exp. Lung Res. 21, dermal keratinocyte proliferation and survival via the sequential engagement
791–808. of cannabinoid receptor-1 and transient receptor potential vanilloid-1. J. In-
Higashiyama, R., Nakao, S., Shibusawa, Y., Ishikawa, O., Moro, T., Mikami, K., Fu- vestig. Dermatol. 131, 1095–1104.
kumitsu, H., Ueda, Y., Minakawa, K., Tabata, Y., Bou-Gharios, G., Inagaki, Y., 2011. Tschop, J., Kasten, K.R., Nogueiras, R., Goetzman, H.S., Cave, C.M., England, L.G.,
Differential contribution of dermal resident and bone marrow-derived cells to Dattilo, J., Lentsch, A.B., Tschop, M.H., Caldwell, C.C., 2009. The cannabinoid
collagen production during wound healing and fibrogenesis in mice. J. Investig. receptor 2 is critical for the host response to sepsis. J. Immunol. 183, 499–505.
Dermatol. 131, 529–536. Wang, Z., Wang, P., Bjorling, D.E., 2014. Treatment with a cannabinoid receptor
Hudson, L.G., Newkirk, K.M., Chandler, H.L., Choi, C., Fossey, S.L., Parent, A.E., Ku- 2 agonist decreases severity of established cystitis. J. Urol. 191, 1153–1158.
sewitt, D.F., 2009. Cutaneous wound reepithelialization is compromised in mice Witkamp, R., Meijerink, J., 2014. The endocannabinoid system: an emerging key
lacking functional Slug (Snai2). J. Dermatol. Sci. 56, 19–26. player in inflammation. Curr. Opin. Clin. Nutr. Metab. Care 17, 130–138.
Ishida, Y., Kondo, T., Takayasu, T., Iwakura, Y., Mukaida, N., 2004. The essential in- Wynn, T.A., Ramalingam, T.R., 2012. Mechanisms of fibrosis: therapeutic translation
volvement of cross-talk between IFN-gamma and TGF-beta in the skin wound- for fibrotic disease. Nat. Med. 18, 1028–1040.
healing process. J. Immunol. 172, 1848–1855. Yrjölä, S., Sarparanta, M., Airaksinen, A.J., Hytti, M., Kauppinen, A., Pasonen-
Ishii, G., Sangai, T., Sugiyama, K., Ito, T., Hasebe, T., Endoh, Y., Magae, J., Ochiai, A., Seppänen, S., Adinolfi, B., Nieri, P., Manera, C., Keinänen, O., Poso, A., Nevalai-
2005. In vivo characterization of bone marrow-derived fibroblasts recruited nen, T.J., Parkkari, T., 2015. Synthesis, in vitro and in vivo evaluation of 1,3,5-
into fibrotic lesions. Stem Cells 23, 699–706. triazines as cannabinoid CB2 receptor agonists. Eur. J. Pharm. Sci. 67, 85–96.
Karsak, M., Gaffal, E., Date, R., Wang-Eckhardt, L., Rehnelt, J., Petrosino, S., Starowicz, Zhang, Y., Lee, T.C., Guillemin, B., Yu, M.C., Rom, W.N., 1993. Enhanced IL-1 beta and
K., Steuder, R., Schlicker, E., Cravatt, B., Mechoulam, R., Buettner, R., Werner, S., tumor necrosis factor-alpha release and messenger RNA expression in macro-
Di Marzo, V., Tuting, T., Zimmer, A., 2007. Attenuation of allergic contact der- phages from idiopathic pulmonary fibrosis or after asbestos exposure. J. Im-
matitis through the endocannabinoid system. Science 316, 1494–1497. munol. 150, 4188–4196.
Klein, T.W., 2005. Cannabinoid-based drugs as anti-inflammatory therapeutics. Nat. Zhao, R., Yang, B., Wang, L., Xue, P., Deng, B., Zhang, G., Jiang, S., Zhang, M., Liu, M.,
Rev. Immunol. 5, 400–411. Pi, J., Guan, D., 2013. Curcumin protects human keratinocytes against inorganic
Kondo, T., Ishida, Y., 2010. Molecular pathology of wound healing. Forensic Sci. Int. arsenite-induced acute cytotoxicity through an NRF2-dependent mechanism.
203, 93–98. Oxidative Med. Cell. Longev. 2013, 412576.
Kupczyk, P., Reich, A., Szepietowski, J.C., 2009. Cannabinoid system in the skin – a Zheng, J., Yu, T., Li, X., Fan, Y., Ma, W., Du, Y., Zhao, R., Guan, D., 2012. Cannabinoid
possible target for future therapies in dermatology. Exp. Dermatol. 18, 669–679. receptor type 2 is time-dependently expressed during skin wound healing in
Lamouille, S., Xu, J., Derynck, R., 2014. Molecular mechanisms of epithelial– mice. Int. J. Leg. Med. 126, 807–814.