VCE-004.3, a cannabidiol aminoquinone derivative, prevents bleomycin-induced skin fibrosis and inflammation through PPARγ- and CB2 receptor-dependent pathways

Article abstract of DOI:10.1111/bph.14450

Background and Purpose: The endocannabinoid system and PPARγ are important targets for the development of novel compounds against fibrotic diseases such as systemic sclerosis (SSc), also called scleroderma. The aim of this study was to characterize VCE-004.3, a novel cannabidiol derivative, and study its anti-inflammatory and anti-fibrotic activities. Experimental Approach: The binding of VCE-004.3 to CB₁ and CB₂ receptors and PPARγ and its effect on their functional activities were studied in vitro and in silico. Anti-fibrotic effects of VCE-004.3 were investigated in NIH-3T3 fibroblasts and human dermal fibroblasts. To assess its anti-inflammatory and anti-fibrotic efficacy in vivo, we used two complementary models of bleomycin-induced fibrosis. Its effect on ERK1/2 phosphorylation induced by IgG from SSc patients and PDGF was also investigated. Key Results: VCE-004.3 bound to and activated PPARγ and CB₂ receptors and antagonized CB₁ receptors. VCE-004.3 bound to an alternative site at the PPARγ ligand binding pocket. VCE-004.3 inhibited collagen gene transcription and synthesis and prevented TGFβ-induced fibroblast migration and differentiation to myofibroblasts. It prevented skin fibrosis, myofibroblast differentiation and ERK1/2 phosphorylation in bleomycin-induced skin fibrosis. Furthermore, it reduced mast cell degranulation, macrophage activation, T-lymphocyte infiltration, and the expression of inflammatory and profibrotic factors. Topical application of VCE-004.3 also alleviated skin fibrosis. Finally, VCE-004.3 inhibited PDGF-BB- and SSc IgG-induced ERK1/2 activation in fibroblasts. Conclusions and Implications: VCE-004.3 is a novel semisynthetic cannabidiol derivative that behaves as a dual PPARγ/CB₂ agonist and CB₁ receptor modulator that could be considered for the development of novel therapies against different forms of scleroderma.

Full text of DOI:10.1111/bph.14450

VCE-004.3, A CANNABIDIOL AMINOQUINONE DERIVATIVE, PREVENTS
BLEOMYCIN-INDUCED SKIN FIBROSIS AND INFLAMMATION TROUGH
PPAR- AND CB
2
-DEPENDENT PATHWAYS
Carmen del Rio1-3*, Irene Cantarero1-3*, Belén Palomares , María Gómez-Cañas , Javier
1-3
4-6
4
-6
7
8
8,9
Fernández-Ruiz , Carolina Pavicic , Adela García-Martín , M. Luz Bellido , Rafaela
1
-3
1-3
1-3
10
Ortega-Castro , Carlos Pérez-Sánchez , Chary López-Pedrera , Giovanni Appendino ,
1
-3
1-3,#
Marco A. Calzado , and Eduardo Muñoz
*
Both authors contributed equally to this study
1
2
Maimonides Biomedical Research Institute of Córdoba, Spain; Departament of Cellular
3
Biology, Physiology and Immunology, University of Cordoba, Spain; University Hospital
4
Reina Sofía, Córdoba, Spain; Instituto Universitario de Investigación en Neuroquímica,
Departamento de Bioquímica y Biología Molecular, Facultad de Medicina, Universidad
5
Complutense, Madrid, Spain; Centro de Investigación Biomédica en Red de Enfermedades
6
Neurodegenerativas (CIBERNED), Madrid, Spain; Instituto Ramón y Cajal de Investigación
7
Innohealth Group, Madrid, Spain; 8Vivacell
Sanitaria (IRYCIS), Madrid, Spain;
9
Biotechnology, Córdoba, Spain; Emerald Health Pharmaceuticals. San Diego, CA, USA;
10Dipartimento di Scienze del Farmaco, Università del Piemonte Orientale, Novara, Italy.
Running head: VCE-004.3 prevents skin fibrosis and inflammation
#
Corresponding author: Eduardo Muñoz. Maimonides Biomedical Research Institute of
Córdoba, University of Córdoba, Avda Menéndez Pidal s/n. 14004. Córdoba, Spain.
Phone: + 34 957213766. Email: e.munoz@uco.es
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/bph.14450
This article is protected by copyright. All rights reserved.

ABSTRACT
Background and Purpose: The endocannabinoid system (ECS) as well as PPARγ are
relevant targets for the development of novel compounds against fibrotic diseases such as
Systemic Sclerosis (SSc), also called Scleroderma. The aim of this study was to characterize
VCE-004.3, a novel cannabidiol derivative, and to study it anti-inflammatory and anti-fibrotic
activities.
Experimental Approach: CB
1
, CB and PPAR binding and functional activities were
2
studied in vitro and in silico. The anti-fibrotic effects of VCE-004.3 were investigated in
NIH-3T3 fibroblasts and in human dermal fibroblasts. To assess the anti-inflammatory and
anti-fibrotic efficacy of VCE-004.3 in vivo we used two complementary models of
bleomycin-induced fibrosis. The effect of VCE-004.3 on ERK1/2 phosphorylation induced
by IgG from SSc patients and PDGF was also investigated.
Key Results: VCE-004.3 binds and activates PPARγ and CB
2
receptors and antagonizes CB
1
receptor. VCE-004.3 binds to an alternative site at the PPARγ ligand-binding pocket (LBP).
VCE-004.3 inhibits collagen gene transcription and synthesis and prevents TGF-induced
fibroblast migration and differentiation to myofibroblasts. VCE-004.3 prevents skin fibrosis,
myofibroblast differentiation and ERK1/2 phosphorylation in bleomycin-induced skin
fibrosis. In addition, VCE-004.3 reduces mast cell degranulation, macrophage activation, T
lymphocytes infiltration, and the expression of inflammatory and profibrotic factors. Topical
application of VCE-004.3 also alleviates skin fibrosis. Finally, VCE-004.3 inhibits PDGF-
BB- and SSc IgG-induced ERK1/2 activation in fibroblasts.
Conclusion and Implications: VCE-004.3 is a novel semi-synthetic cannabidiol derivative
behaving as a dual PPAR/CB
2
agonist and CB receptor modulator that could be considered
1
for the development of novel therapies against different forms of Scleroderma.
This article is protected by copyright. All rights reserved.

ABBREVIATIONS
α-SMA: α-Smooth Muscle Actin
BLM: bleomycin
CB
1
2
: Cannabinoid Receptor Type I
: Cannabinoid Receptor Type II
CB
CBD: Cannabidiol
ECS: Endocannabinoid System
LBP: Ligand binding pocket
NHDFs: Normal Human Dermal Fibroblasts
PPARγ: Peroxisome proliferator-activated receptor-γ
RGZ: Rosiglitazone
SSc: Systemic Sclerosis
TZDs: Thiazolidinediones
This article is protected by copyright. All rights reserved.

INTRODUCTION
Systemic sclerosis (SSc) or scleroderma is a chronic multiorgan autoimmune disease of
unknown etiology characterized by vascular, immunological and fibrotic abnormalities. The
disease is complex and dynamic, and the interrelationship among the three main hallmark of
SSc results in a wide spectrum of clinical presentations ranging from limited skin
involvement (limited cutaneous SSc; lcSSc) to widespread internal organ fibrosis (diffuse
cutaneous SSc; dcSSc). The pathogenesis of SSc not fully understood, and it is believed that
inflammation as well as vascular injury in the initial stages drive the autoimmune response
and precede fibrosis in the course of the disease (Varga and Abraham, 2007; Pattanaik et al.,
2
015).
Several lines of evidence showed that the endocannabinoid system (ECS) may play a role in
the physiopathology of SSc. Cannabinoids exert a broad range of biological effects including
anti-inflammatory activity (Katchan et al., 2016), endothelial function (Bouchard et al., 2003)
and cell death regulation (Garcia-Gonzalez et al., 2009). The ECS is composed of
cannabinoid receptors CB
anandamide (AEA) and 2-arachidonoylgylcerol (2-AG) and the enzymes responsible for their
degradation (Pacher et al., 2006). CB and CB are overexpressed in SSc fibroblasts (Garcia-
Gonzalez et al., 2009) and the anti-inflammatory and anti-fibrotic actions of cannabinoids
have been widely demonstrated both in vitro and in vivo. Thus, inactivation of CB decreased
the number of infiltrating T cells and macrophages in lesioned skin (Marquart et al., 2010).
By the contrary, CB activation limits leukocyte infiltration and tissue fibrosis (Akhmetshina
et al., 2009). Thereby, structurally different CB agonists such as ajulemic acid, JHW-133 and
VCE-004.8 have been shown to alleviate skin fibrosis and inflammation in experimental SSc.
Gonzalez et al., 2012), (Akhmetshina et al., 2009) (del Rio et al., 2016).
1
and CB , their endogenous lipid ligands represented by
2
1
2
1
2
2
(
In addition to binding to classical CB
1
and CB receptors some natural and semisynthetic
2
cannabinoids also bind and activate the nuclear peroxisome proliferator activated receptor
gamma (PPARγ) (Pistis and O'Sullivan, 2017). PPARγ was initially identified through its
role in the regulation of glucose and lipid metabolism and cell differentiation, but it is well
established now that PPARγ ligand agonists are also endowed with anti-inflammatory and
anti-fibrotic activities (Clark, 2002) (Dantas et al., 2015). In this sense, PPARγ is a regulator
of connective tissue homeostasis, and different experimental approaches have shown that
PPAR-γ ligands attenuate hepatic (Galli et al., 2002), renal (Bae et al., 2017) and pulmonary
This article is protected by copyright. All rights reserved.

fibrosis (Milam et al., 2008) as well as bleomycin-induced skin fibrosis (Gonzalez et al.,
012; del Rio et al., 2016; Wei et al., 2014; Ruzehaji et al., 2016).
2
Different serum autoantibodies against multiple intracellular antigens have been detected and
proposed as biomarkers for early and precise diagnosis in SSc (Kayser and Fritzler, 2015).
However, how autoantibodies contribute to SSc prognosis is not well known. Autoantibodies
directed to non-nuclear autoantigens including anti-platelet derived growth factor (PDGF)
receptor, anti-endothelial cells, anti-fibroblast and anti-angiotensin type 1-receptor
autoantibodies with possible pathogenic properties have been also described (Baroni et al.,
2
006) (Mihai and Tervaert, 2010) (Fineschi et al., 2008) (Kayser and Fritzler, 2015; Ho and
Reveille, 2003).
In this sense, targeting the ECS and PPARγ, which are considered potential therapeutic
targets in SSc, represents a viable approach to prevent or attenuate inflammation and fibrotic
process in SSc. In the present study we provide evidence of the anti-inflammatory and anti-
fibrotic activities of a novel cannabidiol (CBD) aminoquinone derivative VCE-004.3, which
targets PPARγ and CB
2
receptors, but also antagonize CB receptor.
1
METHODS
Cell lines and reagents.
HEK-293T, HEK-293T-CB1, HEK-293T-CB2, NIH-3T3 and normal human dermal
fibroblasts (NHDFs) were cultured in DMEM supplemented with 10% fetal bovine serum
(FBS), 2 mM L-glutamine and 1% penicillin/streptomycin. All cells were maintained at 37
°
C and 5% CO in a humidified atmosphere. Rosiglitazone was from Cayman Chem (Ann
2
Arbor, MI, USA), TGFβ1 was obtained from Immunotools GmbH (Friesoythe, Germany)
and non-immunized human IgG was from Jackson Immunoresearch (West Grove, PA, USA).
All other reagents were purchased from Sigma Co (St Louis, MO, USA).
VCE-004.3 synthesis.
VCE-004.3 was synthesized by oxidizing CBD with the hypervalent iodine reagent SIBX as
previously described ((del Rio et al., 2016), and then trapping it with n-pentylamine (4 molar
equivalents) in ethanol solution (ca. 6 mL/mmol of quinone). The reaction mixture was
stirred at room temperature for 18 h, and then worked up by dilution with water H O (1:5),
2
This article is protected by copyright. All rights reserved.

acidification to pH = 2 with HCl (10% aqueous solution), and extraction with CH
mL). The organic layer was dried (Na SO ), filtered and concentrated, and the residue was
purified by reverse phase chromatography (30@100% CH CN and H O) to afford (1'R,6'R)-
-(pentylamino)-6-hydroxy-3'-methyl-4-pentyl--6'-(prop-1-en-2-yl)-[1,1'-bi(cyclohexane)]-
2
Cl (30
2
2
4
3
2
3
2
1
',3,6-triene-2,5-dione as a purple-colored powder (ca. 70% yield). H NMR (CDCl , 300
3
MHz) d ppm: 6.43 (bs, 1H), 5.14 (s, 1H), 4.55 (s, 2H), 3.62 (m, 1H), 3.46 (c, J = 6.6 Hz, 2H),
2
1
.72 (m, 1H), 2.48 (t, J = 7.7 Hz, 2H), 2.31-1.72 (m, 4H), 1.68 (s, 3H), 1.64 (s, 3H), 1.48-
.24 (m, 12H), 0.90 (m, 6H). MW: 413.61
Binding assay.
PPAR binding activity was measured by using PolarScreen PPAR Competitor Assay kit
Life Technologies, Carlsbad, CA, USA) following manufacturer’s specifications. IC50
TM
(
values were calculated using GraphPad Prism. Binding affinities to cannabinoid receptors
were tested by competition studies using commercial membranes containing HEK293 EBNA
cells stably expressing human CB
1
or CB receptors (Perkin-Elmer Life and Analytical
2
3
Sciences, Boston, MA, USA). Inhibition of [ H]CP55940 radioligand binding was measured
following a procedure previously described (Ragusa et al., 2015). Competition binding data
were analysed by using GraphPad Prism® version 5.01 (GraphPad Software Inc., CA, USA).
Docking analysis.
Docking analysis was performing using AutoDock4 (Morris et al., 2009) and Vina softwares
(Trott and Olson, 2010) with the virtual screening tool PyRx (Wolff, 2009) and PyMOL
(Baugh et al., 2011). The receptors model used was the PDB reference (RCSB Protein Data
Bank accession code) 3B0R (Hughes et al., 2014), and 5U5L (Rajapaksha et al., 2017). The
search space for the docking was set according to previous findings about several binding
sites for different ligands. After analysis AutoDock Vina provides the estimated binding
affinity value, which is the sum of the intermolecular energy and the torsional free-energy
penalty. A negative value indicates that bond is thermodynamically stable, whereas a positive
value means instability. Search space for the docking was set around the binding sites
described previously (Hughes et al., 2014).
This article is protected by copyright. All rights reserved.

Cytotoxicity assay.
3
NIH-3T3 (1x10 cells) were seeded in 96 well plates and after 24 h incubated with VCE-
0
04.3 for another 6 h in the presence of 0.1 µM YOYO-1 (Life Technologies). Untreated
cells were considered as 100 % survival. Fluorescence was measured using an Incucyte FLR
imaging system (Essen BioScience, Ann Arbor, MI, USA).
Transfections and luciferase assays.
4
HEK-293T (8x10 ) cells stably expressing human cannabinoid CB
1
or CB receptors were
2
seeded in 24-well plates and after 24 h were transiently transfected with the pCRE-Luc
©
plasmid (0.2 µg/well) using Roti -Fect (Carl Roth, Karlsruhe, Germany) following
manufacturer´s specifications. The CRE luciferase reporter contains the firefly luciferase
gene under the control of multimerized cAMP response element (CRE) located upstream of a
minimal promoter. Twenty-four hours after transfection, CB
1
-HEK-293T cells were
pretreated for 30 min with VCE-004.3 and stimulated with WIN55,212 (1µM) for 6 h. CB
2
-
HEK-293T cells were pretreated for 30 min with VCE-004.3 or WIN55,212 as positive
4
control and stimulated with Forskolin (FSK) (10 M) for 6 h. NIH-3T3 (5x10 ) cells were
seeded in 24-well plates and after 24 h were transiently transfected with the indicated
constructs (PPARγ-GAL4, 0.2 µg; PPARα-GAL4, 0.2 µg; PPARδ-GAL4, 0.2 µg; GAL4-luc,
0
.8 µg; or CAGA-Luc, 0.5 µg). NIH-3T3 stably expressing Col1A2-Luc were seeded in 24-
4
well plates (5x10 cells/well) and treated as indicated. After treatments, luciferase activity
was measured using Dual-Luciferase Assay (Promega, Madison, WI, USA).
Measurement of collagen secretion and deposition.
4
NHDFs were seeded in 24 well plates (5x10 cells/well), pre-treated with VCE-004.3 for 1 h
and stimulated with TGFβ1 for the following 24 h. Deposition of collagen was determined by
measuring absorbance at 540 nm and 605 nm in Genesis 10 UV spectrophotometer after cells
were stained for 30 min with a mixture solution of 0.1% Sirius Red and 0.1% Fast Green in
saturated picric acid. Collagen secretion was studied by collecting culture medium and
measuring soluble collagen using the Sircol Soluble Collagen Assay (Biocolor, County
Antrim, UK) following manufacturer’s instructions.
This article is protected by copyright. All rights reserved.

Western blot.
5
5
Cells were seeded in 6 well plates (1.5x10 cells) or in 60 mm plates (3x10 cells). After
treatments, cells were washed with PBS and proteins extracted in 50 μl of lysis buffer (50
mM Tris–HCl pH 7.5, 150 mM NaCl, 10% glycerol and 1% NP-40) with 10 mM NaF, 1 mM
Na3VO4, 10 μg/ml leupeptine, 1 μg/ml pepstatin and aprotinin and 1 μl/ml saturated PMSF.
Thirty μg of proteins were boiled at 95 °C in Laemmli buffer and electrophoresed in 10%
SDS/PAGE gels. Separated proteins were transferred (20V for 30 min) to PVDF membranes
and blocked in 0.1% Tween 20 in TBS solution containing 5% non-fat dry milk for 1 h at
room temperature. Membranes were incubated with primary antibody against pSMAD2
(1:500; AB3849, Merck Millipore, Billerica, MA, USA), SMAD2 (1:1000; 5339, Cell
Signaling, Danvers, MA, USA), p-ERK 1/2 (1:500; sc-7383, Santa Cruz, Dallas, TX, USA)
and ERK 1/2 (1:2000; M-5670, Sigma) overnight at 4 ºC. Membranes were washed and
incubated with the appropriate horseradish peroxidase-conjugated secondary antibody for 1 h
at room temperature and detected by chemiluminescence system (GE Healthcare Europe
GmbH, Freiburg, Germany). Blots were reproved with an antibody against -tubulin or β-
actin (1:10.000; DM-1A and AC-74, Sigma). Densitometric analysis was performed with
Image J software (NIH, Bethesda, MD, USA).
Cell migration assay.
3
NHDFs (2x10 cells) were seeded in complete DMEM in a 96-well Essen ImageLock plates
(Essen BioScience). When cells reached confluence, wound scratches were made using the
9
6-pin WoundMaker (Essen BioScience). Then, culture medium was replaced by DMEM
supplemented with 1% antibiotics and with mitomycin C (10 ng/ml) to block cell
proliferation. VCE-004.3 and TGF1 10 ng/ml were added in parallel. Wound healing
activity was analysed using IncuCyte HD system by taking images every 3 h for 48 h and the
percentage of wound confluence calculated with the IncuCyte software.
Fibroblast to myofibroblast differentiation.
4
NIH-3T3 cells (1.5x10 ) were seeded onto glass coverslips or directly in 6 well plates. Cells
were serum starved (1% FBS) for 24 h, pre-stimulated with rosiglitazone (RGZ) or VCE-
0
04.3 for 1 h and incubated with 10 ng/ml TGFβ1 for 24 h. Then, cells were washed with
PBS for further western blot analysis or fixed with 4% formaldehyde for 10 min at 37 °C.
After fixation, cells were washed twice with PBS and blocked in IHC select blocking reagent
This article is protected by copyright. All rights reserved.

(Merck Millipore) for 30 min. Cells were incubated overnight at 4 °C with the primary
antibody α-SMA (1:50 in blocking solution, sc-32251, Santa Cruz), washed three times with
PBS containing 0.1% Tween 20 and incubated with the secondary antibody goat anti-rat
coupled to Alexa 488 (1:1000, Merck Millipore) for 3 h at room temperature. Coverslips
were mounted with Vectashield Mounting Medium supplemented with 4',6-diamidino-2-
phenylindole (DAPI) for nuclear staining. Observations were carried out using a Leica
DM2500 microscope and Leica DFC420c camera. Images were acquired as sets of colour-
images and prepared using Image J software.
Compliance with requirements for studies using animals.
Animal studies are reported in compliance with the ARRIVE guidelines (McGrath and Lilley,
2
015). A total of 90 BALB/C female mice (6-8 weeks old) weighing between 17-20 g
(Envigo, Valencia, Spain) were used in the studies. Experiments were performed in
accordance with European Union guideline and approved by the Animal Research Ethic
Committee of Córdoba University and the Andalusian Committee for Animal
Experimentation (2014PI/016). Mice were housed in groups of 9 animals in polycarbonate
cages (300 mm x 180 mm x 150 mm) with free access to food and water in the animal
facilities of the University of Córdoba and maintained in a 12 h light/dark cycle with
controlled temperature (20±2 ºC) and relative humidity (40-50%) (McGrath and Lilley,
2
015).
Animals and Experimental protocols.
Different administration schedules of subcutaneous bleomycin (BLM) have been widely used
as SSc models. Repetitive injections of subcutaneous BLM induce skin fibrosis localized to
the injection area and lead systemic manifestations such as lung fibrosis (Beyer et al., 2010).
Two different BLM-induced models were used to assess the effects of VCE-004.3 in vivo. (a)
The inflammatory model of SSc was used to study VCE-004.3 preventive effects. Fibrosis
was induced by daily subcutaneous administration of BLM (50 µg/mice; 100 µl) (Mylan,
Barcelona, Spain) for 3 weeks. Treatments were administered in parallel and consisted of
daily intraperitoneal (i.p.) injections of VCE-004.3 (20 mg/kg), RGZ (5 mg/kg) or vehicle
(4% DMSO, 6.2% Tween20, saline; 100 µl). Subcutaneous injections of 0.9% NaCl served as
a control. (b) Pre-established fibrosis model of SSc was induced to study VCE-004.3 curative
effect. Fibrosis was caused using daily subcutaneous BLM (20 µg/mice; 100 µl) for 6 weeks.
After 3 weeks of BLM administration animals were treated in parallel with a daily dose of i.p.
This article is protected by copyright. All rights reserved.

injections of VCE-004.3 (20 mg/kg), RGZ (5 mg/kg) or vehicle or were given topical
application of VCE-004.3 (250 µM) or vehicle (7:3 Polypropylene glycol: Ethanol; 100 µl)
for the remaining 3 weeks. Mice were acclimated to manipulators for a week previously to
the experiments in order to reduce stress and allow the detection of subtle changes in
behaviour by researchers. Mice weight was evaluated weakly and no significant changes
were observed between experimental groups. No behavioural changes were observed during
the protocols. Neither piloerection nor hunched appearances were appreciated during the
experimental procedures. Mice were sacrificed by cervical dislocation and the back skin was
removed and processed for histological examination or was frozen for further analysis.
Macroscopic evaluation of internal organs did not reveal pathological changes.
Blinding and randomization
Laboratory animals were randomly assigned to experimental groups and treatments were
assessed by a blinded investigator. The order of treatment administration was also
randomized. All animal samples were studied, and analysis was carried out in a blinded
manner. For immunohistochemical studies three random fields of each skin biopsy were
photographed and evaluated by two independent observers.
Histochemical analysis.
Skin sections (5 micrometre-thick) were stained with either Masson's trichrome, picrosirius
red, or toluidine blue. Myofibroblasts were detected in skin sections by immunofluorescent
labelling with α-SMA (1:50, sc-32251, Santa Cruz) and Alexa 488 (A-11008; Life
Technologies). For quantification, images were taken with a LSM 5 Exciter confocal
microscope (Zeiss, Jena, Germany) and quantified in 10 to 15 randomly chosen fields at 250
magnifications using ImageJ software (http://rsb.info.nih.gov/.ij). For immunohistochemical
detection of macrophages, T-cells and ERK1/2 activation, we used F4/80 (1:50; MCA497,
Bio-Rad), CD3 (1:100, sc20047, Santa Cruz) and p-ERK 1/2 (1:100; sc-7383, Santa Cruz)
antibodies respectively. Slides were developed with diaminobenzidine chromogen (Merck
Millipore) and counterstained in Harris haematoxylin. Slides were photographed, digitalized
using a Leica DFC420c camera and analysed using Image J software.
This article is protected by copyright. All rights reserved.

Real-time PCR.
Total RNA was isolated from mice frozen skin tissue using QIAzol lysis reagent (Qiagen,
Hilden, Germany) and purified with RNeasy mini kit (Qiagen). Total RNA (1 μg) was
retrotranscribed using the iScriptTM cDNA Synthesis Kit (Bio-Rad), and the cDNA generated
was analysed by real-time PCR using the iQTM SYBR Green Supermix (Bio-Rad) using a
CFX96 Real-Time PCR Detection System (Bio-Rad). The oligonucleotide primers used are
listed in Supplementary Table 1. Gene expression were normalized to GADPH mRNA levels
in each sample and expressed using the 2-ΔΔCt method.
Patients and IgG antibodies purification
In total, 12 patients, 6 with lSSc and 6 with dSSc were included in the study after ethics
committee approval. All subjects provided written informed consent. Samples were obtained
from peripheral venous blood. IgGs were purified using a pool of 6 patients with lcSSc and 6
patients with dSSc using high affinity chromatography of G-Sepharose protein (MabTrap, GE
Healthcare Bio-sciences AB, Uppsala, Sweeden) following the manufacturer's
recommendations.
Data and statistical analysis.
The data and statistical analysis comply with the recommendations on experimental design
and analysis in pharmacology (Curtis et al., 2015). In vitro results are shown as mean ± S.D.
and in vivo data as mean ± SEM. Statistical analysis were carried out using the SPSS v.19
software for Windows (IBM Corporation, New York, USA). Results were tested for normal
distribution using Kolmogorov-Smirnov normality test. Then, data were analysed using one-
way ANOVA followed by Tukey post hoc test when F achieved was P < 0.05, and there was
no significant variance in homogeneity. Some results were normalized to control to avoid
unwanted sources of variation. Differences between not normally distributed results were
studied by the Kruskall-Wallis followed by Dunns post-hoc test using the GraphPad Prism
v.5 for Windows. Minimal statistical significance was set at P < 0.05.
Nomenclature of targets and ligands.
Key protein targets and ligands in this article are hyperlinked to corresponding entries in
http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS
Guide to PHARMACOLOGY (Southan et al., 2016) and are permanently archived in the
This article is protected by copyright. All rights reserved.

Concise Guide to PHARMACOLOGY 2015/16 (Alexander et al., 2015a; Alexander et al.,
015b; Alexander et al., 2015c).
2
RESULTS
VCE-004.3 is a CBD quinone derivative with selective PPARγ and CB agonism and
2
CB
We have previously shown that VCE-004.8, a dual PPARγ/CB
fibrosis through PPARγ and CB -dependent pathways in a mice model of SSc (del Rio et al.,
016). As part of our program of chemical synthesis searching for CBD quinone derivatives
1
antagonism.
2
agonist, alleviates dermal
2
2
we have selected the novel compound VCE-004.3 (Fig. 1A), and have investigated its
mechanism of action and efficacy in dermal inflammation and fibrotic events. First, we
evaluated the binding affinity to PPARγ receptor and found that VCE-004.3 was able to bind
PPARγ with an IC50 of 3.5 M, which is a significant lower affinity than the binding of
rosiglitazone (RGZ) (Fig. 1B). VCE-004.3 was not cytotoxic and induced PPARγ
transcriptional activity in a concentration dependent manner in NIH-3T3 cells transiently co-
transfected with GAL4-luc and PPARγ-GAL4 plasmids (Fig. 1C). However, VCE-004.3 did
not activate PPARα (Fig. 1D) neither PPARδ (Fig. 1E) transcription, indicating that VCE-
0
04.3 is a selective agonist of the PPARγ isoform.
To further characterize the effects of VCE-004.3 on PPARγ transcriptional activities, we
performed experiments to study the behavior of VCE-004.3 in the presence of RGZ. To this
end, GAL4-PPARγ/GAL4-luc transfected NIH-3T3 cells were pre-incubated with increasing
concentrations of VCE-004.3 and then treated with 1μM RGZ. VCE-004.3 decreased the
RGZ-induced PPARγ transactivation (Fig. 2A), suggesting that VCE-004.3 could prevent
RGZ binding by competing with the same binding site on PPARγ. Moreover, the effect on
PPARγ activation was abrogated after removal of VCE-004.3 by washing cells with PBS,
suggesting that VCE-004.3 binds to PPARγ in a reversible and non-covalent manner (Fig.
2
B).
It was previously reported that synthetic PPARγ ligands bind to the canonical ligand binding
pocket (LBP) and also to an alternate site on this nuclear receptor, an effect that cannot be
This article is protected by copyright. All rights reserved.

blocked by covalently bound synthetic antagonists, evidencing non-overlapping binding with
the canonical pocket (Hughes et al., 2014). Therefore, we performed docking experiments
studies using crystal structures 3B0R and 5U5L deposited in the Protein Data Bank (PDB).
As depicted in Fig. 2C molecular docking with 3B0R indicated that VCE-004.3 binds Leu340
in Ω-loop Beta 3 and Tyrosine 373 in Helix 1/2 (alternative site) with a predicted Ki of
1
08,38 nM. In the presence of GW9662 in the canonical site VCE-004.3 binds to Ser342 in
the Ω-loop β3-β4 and Leu340 in β3 with a predicted Ki of 569,80 nM. In the case of 5U5L,
VCE-004.3 was found to bind Ser342 in the Ω-loop β3-β4 and Gly-346 in β4 with a predicted
Ki of 1,12 M and in the presence of GW9662 we found that VCE-004.3 binds to Leu340 in
β3 and Ser342 in Ω-loop β3-β4 with a predicted Ki of 17,72 nM.
To confirm that VCE-004.3 is functional through the alternative binding site we studied
VCE-004.3-mediated PPARγ transcriptional activity in the presence of T0070907, an
irreversible PPARγ antagonist that covalently binds to Cys285 into the PPAR LBP
canonical binding site. As expected, T0070907 was very effective to block RGZ-induced
PPAR transactivation (Fig. 2D), which is consistent with the fact that RGZ activates PPAR
by acting mainly through the canonical binding site (Hughes, 2014). By the contrary,
T0070907 did not block the transcriptional activity of VCE-004.3 (Fig. 2E). Altogether, our
results strongly suggest that, in contrast to RGZ, VCE-004.3 is a PPARγ modulator (m-
PPARγ) acting preferentially through the PPAR LBP alternative binding site. In this sense,
human Mesenchymal Stem Cells treated with VCE-004.3 showed less and smaller lipid
droplets and lower expression levels of adipogenic differentiation markers such as PPARγ,
ADIPOQ and CEBPA compared to cells treated with the PPARγ full-agonist (PPARγ-fa)
RGZ (Supplementary Fig. S1).
Next, we studied the effect of VCE-004.3 on CB
bind CB receptor with a pKi value of 5.61 ± 0.15 (Fig. 3A) and antagonized the activity of
WIN55,212 in HEK293T-CB cells transfected with the pCRE-luc reporter plasmid (Fig.
B). VCE-004.3 also bound to CB containing membranes with a pK value of 6.69 ± 0.10
Fig. 3C) behaving as CB agonist since it inhibited forskolin-induced luciferase activity in
cells transfected with the pCRE-luc reporter plasmid (Fig. 3D). The effects of
antagonists AM630
1
and CB
2
receptors. VCE-004.3 was able to
1
1
3
2
i
(
2
HEK293T-CB
2
WIN55,212 and VCE-004.3 on CB2 activation was prevented by the CB
2
This article is protected by copyright. All rights reserved.

and SR144528 (data not shown). In summary VCE-004.3 is a dual PPARγ/CB
2
agonist with
CB antagonistic activity.
1
VCE-004.3 prevents prevented collagen transcription and synthesis induced by TGFβ.
It has been shown that PPARγ ligands inhibit TGFβ-induced collagen gene transcription by
targeting the p300 transcriptional coactivator (Ghosh et al., 2009). Thus, we studied the effect
of VCE-004.3 on collagen transcriptional regulation in NIH-3T3 cells stably expressing the
Col1A2-luc plasmid, which encodes the gene promoter of the pro-alpha2 chain component of
type I collagen fused to the luciferase gene. Pre-treatment with either RGZ or VCE-004.3
resulted in a significant inhibition of collagen transcription induced by TGFβ1 stimulation
(Fig. 4A). The binding of TGFβ to its receptor entails the activation of SMAD2/SMAD3 by
phosphorylation, induction of its transcriptional activity and stimulation of the Col1A2
promoter (Shi and Massague, 2003). Pre-incubation of NIH-3T3 cells transfected with
CAGA-luc plasmid with VCE.004.3 significantly reduced SMADs transcriptional activity
(
Fig. 4B). To further investigate the effect of VCE-004.3 on TGFβ1 signaling, we studied
both the canonical and the ERK1/2 non-canonical pathways. As depicted in Fig. 4C, VCE-
04.3 did not interfere with TGFβ1-induced SMAD2, but it inhibited the phosphorylation of
0
ERK1/2 proteins in NIH-3T3 cells. Similar results were found with RGZ. Next, to correlate
the effect of VCE-004.3 on Col1A2 transcriptional activity with collagen secretion and
deposition, NHDFs were preincubated with increasing concentrations of VCE-004.3 and
stimulated with TGFβ1 for 24 h. VCE-004.3 significantly inhibited TGFβ1-induced collagen
release and deposition in NHDFs (Fig. 4D and 4E).
VCE-004.3 prevents prevented BLM-induced dermal inflammation and fibrosis in vivo.
The BLM mice model is widely used to evaluate potential anti-inflammatory and anti-fibrotic
therapies for SSc (Gonzalez et al., 2012; del Rio et al., 2016; Wu et al., 2009; Ruzehaji et al.,
2
016). BLM causes inflammation in a short period of time, characterized by elevation of
proinflammatory cytokines and growth factors that peak around day 14 of BLM exposure
Avouac, 2014). In the inflammatory model, BLM administration for 3 weeks resulted in
increased dermal thickness versus control group, which was attenuated by RGZ and VCE-
04.3 treatments. However, neither RGZ nor VCE-004.3 were able to recover the loss of
(
0
adipose tissue layer induced by BLM administration (Fig. 5A). In addition, RGZ as well as
VCE-004.3 were able to reduce collagen accumulation in the skin after BLM administration
(Fig 5B). Next, we studied the expression of cytokines related with inflammatory and fibrotic
This article is protected by copyright. All rights reserved.

processes. As expected, BLM-treated mice had significantly higher expression levels of Il-4,
Il-6 and Ccl-2 in the skin compared with control mice. Such upregulation was significantly
reduced in mice treated either with RGZ or VCE-004.3 (Fig. 5C). Despite no significant
changes on the expression levels of Il-1β, Tgfβ and Il-13 were observed after BLM challenge
these cytokines were less expressed in mice treated with RGZ or VCE-004.3 (Fig. 5C).
Mast cells degranulation as well as macrophages and lymphocyte accumulation are the
hallmark of BLM-induced fibrosis, and play important roles during the inflammatory phase
of SSc (Yamamoto et al., 1999). Toluidine blue staining showed an increased number of
degranulated mast cells in the lesioned skin of the BLM group. Although we did not observe
any effect in the RGZ-treated group, treatment with VCE-004.3 significantly decreased mast
cell degranulation to levels comparable to the control group. Moreover,
immunohistochemical examination demonstrated that VCE-004.3 significantly reduced the
+
+
infiltration of F4/80 macrophages and CD3 lymphocytes compared with BLM untreated
mice (Fig. 6).
It is well known that TGFβ drives the conversion of fibroblasts into myofibroblasts, and the
accumulation of myofibroblasts cells in fibrotic skin is a hallmark of SSc (Marangoni et al.,
2
015; Gilbane et al., 2013). Myofibroblasts are characterized by the novo synthesis of α-
smooth muscle actin (α-SMA) fibres, and we found a substantial reduction of spindle-shaped
+
α-SMA cells populating the fibrotic dermis with RGZ and VCE-004.3 treatments (Fig. 7A).
Moreover, VCE-004.3, as well as RGZ, prevented cell morphology changes of cellular
hypertrophy and cell shape variations from stellate to bipolar cells associated with TGFβ1-
induced myofibroblast differentiation (Fig. 7B). This is accompanied by a reduction of the
steady levels of α-SMA protein level becoming statistically significant at the dose of 10 µM
(Fig. 7C) and indicated that PPARγ ligands such as VCE-004.3 attenuate TGFβ-induced
myofibroblast differentiation.
This article is protected by copyright. All rights reserved.

It has been previously shown that disruption of the MAPK/ERK pathway reduces
inflammation and fibrosis in vivo (Galuppo et al., 2011). Therefore, we evaluated status of
phosphorylated ERK1/2 in the skin after BLM administration. Mice subjected to BLM
administration showed a significant increase of ERK1/2 phosphorylation while treatment
with VCE-004.3 as well as RGZ resulted in a significant decrease of p-ERK1/2
immunostaining (Fig. 8).
VCE-004.3 alleviated established fibrosis in mice.
To evaluate the efficacy of VCE-004.3 in a model of stablished skin fibrosis mice were
challenged with BLM for 6 weeks and the pharmacological treatments were carried out
during the last 3 weeks of BLM injections. Animals were treated with BLM alone, BLM plus
RGZ (5 mg/kg i.p), BLM plus VCE-004.3 (20 mg/kg i.p.), or BLM plus VCE-004.3 (250 nM
topically). Consistent with previous studies, dermal thickness increased more than two-fold
after 6 weeks of BLM challenge, and paralleled with an strong reduction in the adipose
tissue. Treatment for 3 weeks with VCE-004.3 in the absence of BLM did not influence the
structures of the skin. In contrast, treatment with either RGZ or VCE-004.3 (i.p. or topical) of
BLM-challenged mice significantly reversed dermal thickness, although they could not
recover adipose layer thickness to the control level (Fig. 9A). According to this reduction of
dermal thickness, RGZ and VCE-004.3 prevented collagen accumulation in the skin (Fig.
9
B). In a similar experiment, co-treatment with the CB antagonist AM630 abrogated the
2
effect of VCE-004.3 to a certain extent. However, the PPARγ antagonist T0070907 could not
block the effect of the compound, since not significant changes in neither dermal thickness
nor collagen accumulation were observed when compared with VCE-004.3 treated animals
(Supplementary Fig. S2). This result confirmed that the biological activity of VCE-004.3 in
vivo is mediated by targeting PPARγ at the alternative binding site as well as to CB
2
.
The mRNA levels of Il-4, Il-6 and Tgfβ were also investigated in this model. Expression
levels of Il6 were significantly reduced with all the pharmacological treatments. Il-6 mRNA
was increased in the fibrotic model, but to a much lesser extent than in the inflammatory
model, indicating that the cytokine signature is different in the two approaches. Accordingly,
Il-4 mRNA in the skin was not significantly increased after six weeks of BLM challenge but,
in any case, the expression levels were reduced in both RGZ and VCE-004.3 treated mice.
The expression of Tgfβ was strongly upregulated in the BLM group, partially inhibited in
This article is protected by copyright. All rights reserved.

VCE-004.3 (i.p.) treated mice, and significantly reduced after RGZ and topical VCE-004.3
treatments (Fig. 9C).
Effect of VCE-004.3 on ERK1+2 activation and fibroblast migration.
It is well accepted that autoantibodies could play a key role in the SSc pathogenesis.
Stimulatory autoantibodies targeting PDGF receptor have been detected in 100% of SSc
patients and demonstrated to trigger an intracellular cascade that involves ROS, Ha-Ras and
ERK1/2 activation, eventually leading to fibroblast activation and increased collagen-gene
expression (Baroni et al., 2006; Luchetti et al., 2016). In addition, different studies
demonstrated that MAPK are activated in response to fibrogenic stimuli and contribute to
myofibroblast differentiation and function (Leask, 2012). To study the effect of VCE-004.3
on ERK1/2 phosphorylation, NHDFs were preincubated with the compound under
investigation, and then stimulated with human recombinant PDGF-BB, or with IgG purified
from patients diagnosed with limited cutaneous SSc (lcSSc) or diffuse cutaneous SSc
(dcSSc), with the steady state levels for total and phosphorylated ERK1/2 protein being
analysed by western blot. VCE-004.3 clearly prevented ERK1/2 phosphorylation in all the
condition tested (Fig. 10 A-C).
Migration of skin fibroblast plays a crucial role in SSs (Gilbane et al., 2013), and we tested if
the anti-fibrotic effect of VCE-004.3 could be associated with reduced fibroblast migration.
Thus, NHDFs cell monolayers were scratched, and the effect of VCE-004.3 on TGFβ1-
induced wound healing was analysed over the time. Our results show that VCE-004.3
strongly inhibited fibroblasts migration induced by TGFβ1 (Fig. 10D).
DISCUSSION
Others and we have previously investigated the anti-fibrotic effects of synthetic cannabinoids
activating both PPARγ and CB
012; del Rio et al., 2016). In this study, we report for the first time the anti-inflammatory
and anti-fibrotic effects of a novel CBD derivative, VCE-004.3, which is a dual PPARγ/CB
agonist that also behaves as a CB antagonist.
2
receptors in animal models of skin fibrosis (Gonzalez et al.,
2
2
1
Glitazones like RGZ are PPARγ full agonists (PPARγ-fa) widely used in the clinical practice
for glycemic control. Nevertheless, this kind of PPARγ-fa may cause several adverse effects,
This article is protected by copyright. All rights reserved.

including fluid retention, congestive heart failure, and bladder cancer (Ciudin et al., 2012).
Accordingly, the search for PPARγ modulators (PPARγ-m) as safer alternatives to PPARγ
full agonists (PPARγ-fa) has grown, with the expectation of separating mechanism-associated
side effects from the insulin-sensitizing activity. It has been described that PPARγ activation
level is directly related with adipogenesis but not to insulin sensitivity (Doshi et al., 2010).
PPARγ-m can bind the receptor in a different manner than PPARγ-fa, leading conformational
changes that alter co-activator recruitment (Doshi et al., 2010). In this context, VCE-004.3
can be considered as a PPARγ-m, since it activates PPARγ with lower potency than RGZ,
and is less adipogenic. Here, we show that VCE-004.3 activates PPARγ transcriptional
activity and competes for RGZ-binding pocket, which is located within the PPARγ LBP.
However, VCE-004.3 effect of PPARγ transactivation could not be blocked by T0070907
showing that both compounds bind simultaneously to distinct PPARγ binding sites, as
confirmed by docking studies. These results are consistent with previous studies that reported
that covalent antagonists do not block ligands that interact with the alternate binding site
(Hughes et al., 2014). Moreover, synthetic PPARγ agonists that binds to the alternate site
induce PPARγ-driven anti-diabetic activities (Hughes et al., 2014). Thus, it is not unexpected
that T0070907 did not prevent the anti-fibrotic response of VCE-004.3 in our in vivo studies.
Since PPARγ ligands that bind to canonical binding site are also effective in BLM-induced
fibrosis (Wu et al., 2009; Wei et al., 2014), we can assume that, due to versatility of LBP
PPARγ, PPARγ-m (partial agonist) are also effective to prevent inflammation and fibrosis.
It has been previously shown that the canonical pathway activated by TGFβ drive SMADs
phosphorylation that is upstream of the fibrogenic process (Walton et al., 2017). Our results
show that VCE-004.3 reduces TGFβ-induced SMAD2/3 transcriptional activation and
collagen synthesis but does not affect SMAD2/3 phosphorylation. Since ligand-mediated
PPARγ activation does not interfere with SMADs phosphorylation, it is likely that VCE-
0
04.3 compete with SMADs proteins for the interaction with p300 co-activator, thus
preventing collagen synthesis induced by TGFβ (Ghosh et al., 2009). However, other TGFβ
dependent signaling pathways such as the ras/MEK/ERK cascade have been also suggested to
contribute to SSc pathogenesis (Leask, 2012). VCE-004.3 also downregulates ERK1/2
signaling, a major pathway controlling cell growth, proliferation, migration, apoptosis and
myofibroblast differentiation (McCubrey et al., 2007). Indeed, hyperactivation of ERK1/2
was observed in fibroblast from SSc patients, and MEK/ERK signaling has been required for
α-SMA stress fiber assembly (Chen et al., 2011). Interestingly, a recent report has shown that
This article is protected by copyright. All rights reserved.

skin fibroblasts express CB
conditions (Bort et al., 2017). It is therefore possible that VCE-004.3 inhibit ERK1+2
activation both in vitro and in vivo by acting directly through CB receptor. Accordingly, the
2
receptor that is further upregulated under inflammatory
2
anti-fibrotic activity of VCE.004.3 in vivo was partially prevented by the CB antagonist
2
AM630.
The earliest events of SSc include obliterative vasculopathy, which induces inflammation and
autoimmunity. Indeed, in early dcSSc, the dermis exhibits inflammatory infiltrates mainly
composed by monocytes and activated lymphocytes (Varga and Abraham, 2007).
Accordingly, we used a model characterized by early inflammation induced by 3 weeks of
BLM administration in which treatment with the compound is carried out in parallel with
bleomycin injections (Gonzalez et al., 2012). Our data indicate that VCE-004.3 prevents
dermal thickening as well as inflammatory cells infiltration, and pro-inflammatory and pro-
fibrotic cytokines expression. Previous studies have shown that PPARγ activation by RGZ
mediates immunomodulatory effects that prevent bleomycin-induced skin fibrosis (Wu et al.,
2
009). Moreover, macrophages activation as well as T cells proliferation are also regulated
by PPARγ activation (Clark et al., 2000; Harris and Phipps, 2001). Indeed PPAR and CB
ligand activators inhibited M1 polarization and influence macrophage M2 differentiation
Bouhlel et al., 2007; Guillot et al., 2014). Therefore, it is conceivable that PPARγ activation
2
(
and modulation of the cannabinoid system by VCE-004.3 could limit leukocyte infiltration
and macrophage activation in the skin after bleomycin administration (Akhmetshina et al.,
2
009; Marquart et al., 2010).
The BLM-induced model of SSc is also useful to evaluate the progression from an
inflammatory process to a fibrotic state. In this model, skin fibrosis is measured by dermal
thickness evaluation, myofibroblasts count, and collagen content assessed by hydroxyproline
assay. Alternatively, the content of collagen in skin biopsies can be also evaluated by
picrosirius staining, a semi-quantitative method, since a highly significant correlation
between picrosirius staining and hydroxyproline assays to measure skin collagen have been
found recently (Caetano et al, 2016). Progressive fibrosis of the skin and other internal organs
caused by excessive accumulation of collagen and other ECM is one of the three main
hallmarks of SSc (Pattanaik et al., 2015). Many compounds that have been tested using the
BLM model were investigated in a preventive schedule if an anti-inflammatory agent can
This article is protected by copyright. All rights reserved.

hinder the subsequent fibrosis. In the event, treatment with VCE-004.3 (topical and i.p.)
could reduce dermal thickness and collagen content as well as cytokines involved in collagen
production after fibrosis has been already developed, alleviating the previously established
damage. CB
of skin fibrosis. While CB
pharmacological inactivation of this receptor alleviates skin fibrosis. In addition, CB
1
and CB
2
receptors have shown to play an opposite role on experimental models
1
activation is detrimental for the disease, the genetic or
2
agonists have been shown to prevent BLM-induced fibrosis (Akhmetshina et al., 2009;
Marquart et al., 2010) (Gonzalez et al., 2012; del Rio et al., 2016). Thus, VCE-004.3 may
have a beneficial effect on skin fibrosis by acting as dual CB
While the CB agonism effect on BLM-induced skin fibrosis has been confirmed using
AM630, the potential CB antagonism requires further research. We found that VCE-004.3
binds to CB with relative low affinity but inhibits CB -dependent cAMP signalling with
potency similar to other CB
1
antagonist and CB
2
agonism.
2
1
1
1
1
antagonists. This Janus-behavior (Dhopeshwarkar et al, 2017)
3
could, in principle, be due to either to binding to a different site than [ H] CP55940 behaving
as a potential negative allosteric modulator. Experiments are in course to elucidate the
mechanism of action of VCE-004.3 on CB functionality.
1
As in many autoimmune diseases, B cell activation and generation of autoantibodies against
different targets are also present in SSc patients, and this plays a critical role in the
pathogenesis of the disease. Among others, agonistic PDGF receptor autoantibodies showed
to change fibroblast phenotype into SSc features through Ha-Ras-ERK1/2 and ROS cascades
(Baroni et al., 2006). In addition, the inhibition of MAPKs activation has been shown to
suppress fibrosis progression (Galuppo et al., 2011). Treatment with VCE-004.3 prevented
ERK1/2 activation in normal fibroblasts after exposition to stimulatory autoantibodies
isolated from lcSSc and dcSSc patients, indicating that the anti-fibrotic effects of VCE-004.3
may be also mediated by ERK1/2. Interestingly, the presence of autoantibodies has been also
described in the BLM model (Yamamoto, 2006), and it is therefore possible that VCE-004.3
inhibits ERK1/2 activation in vivo by targeting both TGF-induced and autoantibodies-
induced ERK1/2 phosphorylation.
Finally, transdermal penetration of cannabinoids is well established (Valiveti et al., 2004;
Lodzki et al., 2003) and in this study we have shown that topical application of VCE-004.3
This article is protected by copyright. All rights reserved.

also inhibited BLM-induced skin fibrosis and therefore this compound could be also
formulated for the treatment of localized forms of cutaneous fibrosis.
Cannabinoids like CBD are multitarget compounds, and semisynthetic derivatives have been
designed to improve the bioactivities and druggability of the natural products. In this sense,
VCE-004.3 is a novel cannabinoid derivative that acts as a dual PPAR/CB2 agonist and as a
negative modulator of CB1 receptor. Because of this, it qualifies as a candidate for the
development of novel therapies against different forms of Scleroderma.
Acknowledgments
This work was partially supported by the MINECO grant RTC-2015-3364 cofounded by the
European Development Regional Fund in the Framework of the Operative Program
“Reinforcement of research, technological development and innovation”. EM was also
supported by the MINECO grant SAF2014-53763-P. BP is a predoctoral fellow supported by
the i-PFIS program, Instituto de Salud Carlos III (IFI15/00022; European Social Fund
“investing in your future”)
Author contributions
CDR, IC and EM contribute to the conception and design of the study. CDR, IC, BP, CN,
AGM and MLB performed in vitro and in vivo experiments. CDR and BP conducted the
statistical analysis. MGC and JFR performed binding studies. ROC, CPS and CLP selected
patients and isolated IgG from plasma. CP and MAC performed in silico analysis. GA
supervised the chemical synthesis of VCE-004.3. CDR and EM wrote the manuscript. All the
authors contributed to the analysis and interpretation of data, critically reviewed and
approved the manuscript.
Conflict of interest
The authors state no conflict of interest.
This article is protected by copyright. All rights reserved.

REFERENCES
Akhmetshina A, Dees C, Busch N, Beer J, Sarter K, Zwerina J, et al. (2009). The cannabinoid
receptor CB2 exerts antifibrotic effects in experimental dermal fibrosis. Arthritis
Rheum, 60, 1129-36.
Alexander S P, Cidlowski J A, Kelly E, Marrion N, Peters J A, Benson H E, et al. (2015a).
The Concise Guide to PHARMACOLOGY 2015/16: Nuclear hormone receptors. Br J
Pharmacol, 172, 5956-78.
Alexander S P, Davenport A P, Kelly E, Marrion N, Peters J A, Benson H E, et al. (2015b).
The Concise Guide to PHARMACOLOGY 2015/16: G protein-coupled receptors. Br
J Pharmacol, 172, 5744-869.
Alexander S P, Fabbro D, Kelly E, Marrion N, Peters J A, Benson H E, et al. (2015c). The
Concise Guide to PHARMACOLOGY 2015/16: Enzymes. Br J Pharmacol, 172,
6
024-109.
Avouac J. (2014). Mouse model of experimental dermal fibrosis: the bleomycin-induced
dermal fibrosis. Methods Mol Biol, 1142, 91-8.
Bae K H, Seo J B, Jung Y A, Seo H Y, Kang S H, Jeon H J, et al. (2017). Lobeglitazone, a
Novel Peroxisome Proliferator-Activated Receptor gamma Agonist, Attenuates Renal
Fibrosis Caused by Unilateral Ureteral Obstruction in Mice. Endocrinol Metab
(Seoul), 32, 115-123.
Baroni S S, Santillo M, Bevilacqua F, Luchetti M, Spadoni T, Mancini M, et al. (2006).
Stimulatory autoantibodies to the PDGF receptor in systemic sclerosis. N Engl J Med,
3
54, 2667-76.
Baugh E H, Lyskov S, Weitzner B D & Gray J J. (2011). Real-time PyMOL visualization for
Rosetta and PyRosetta. PLoS One, 6, e21931.
Beyer C, Schett G, Distler O & Distler J H. (2010). Animal models of systemic sclerosis:
prospects and limitations. Arthritis Rheum, 62, 2831-44.
Bort A, Alvarado-Vazquez P A, Moracho-Vilrriales C, Virga K G, Gumina G, Romero-
Sandoval A, et al. (2017). Effects of JWH015 in cytokine secretion in primary human
keratinocytes and fibroblasts and its suitability for topical/transdermal delivery. Mol
Pain, 13, 1744806916688220.
Bouchard J F, Lepicier P & Lamontagne D. (2003). Contribution of endocannabinoids in the
endothelial protection afforded by ischemic preconditioning in the isolated rat heart.
Life Sci, 72, 1859-70.
Bouhlel M A, Derudas B, Rigamonti E, Dievart R, Brozek J, Haulon S, et al. (2007).
PPARgamma activation primes human monocytes into alternative M2 macrophages
with anti-inflammatory properties. Cell Metab, 6, 137-43.
Caetano GF, Fronza M, Leite MN, Gomes A, Frade MA (2016). Comparison of collagen
content in skin wounds evaluated by biochemical assay and by computer-aided
histomorphometric analysis. Pharm Biol. 54, 2555-2559.
Chen Y, Leask A, Abraham D J, Kennedy L, Shi-Wen X, Denton C P, et al. (2011).
Thrombospondin 1 is a key mediator of transforming growth factor beta-mediated cell
contractility in systemic sclerosis via a mitogen-activated protein kinase kinase
(MEK)/extracellular signal-regulated kinase (ERK)-dependent mechanism.
Fibrogenesis Tissue Repair, 4, 9.
Ciudin A, Hernandez C & Simo R. (2012). Update on cardiovascular safety of PPARgamma
agonists and relevance to medicinal chemistry and clinical pharmacology. Curr Top
Med Chem, 12, 585-604.
Clark R B. (2002). The role of PPARs in inflammation and immunity. J Leukoc Biol, 71,
3
88-400.
This article is protected by copyright. All rights reserved.

Clark R B, Bishop-Bailey D, Estrada-Hernandez T, Hla T, Puddington L & Padula S J.
2000). The nuclear receptor PPAR gamma and immunoregulation: PPAR gamma
(
mediates inhibition of helper T cell responses. J Immunol, 164, 1364-71.
Curtis M J, Bond R A, Spina D, Ahluwalia A, Alexander S P, Giembycz M A, et al. (2015).
Experimental design and analysis and their reporting: new guidance for publication in
BJP. Br J Pharmacol, 172, 3461-71.
Dantas A T, Pereira M C, De Melo Rego M J, Da Rocha L F, Jr., Pitta Ida R, Marques C D,
et al. (2015). The Role of PPAR Gamma in Systemic Sclerosis. PPAR Res, 2015,
1
24624.
Del Rio C, Navarrete C, Collado J A, Bellido M L, Gomez-Canas M, Pazos M R, et al.
2016). The cannabinoid quinol VCE-004.8 alleviates bleomycin-induced
(
scleroderma and exerts potent antifibrotic effects through peroxisome proliferator-
activated receptor-gamma and CB2 pathways. Sci Rep, 6, 21703.
Dhopeshwarkar A, Murataeva N, Makriyannis A, Straiker A & Mackie K. (2017) Two Janus
Cannabinoids That Are Both CB2 Agonists and CB1 Antagonists.). J Pharmacol Exp
Ther. 360, 300.
Doshi L S, Brahma M K, Bahirat U A, Dixit A V & Nemmani K V. (2010). Discovery and
development of selective PPAR gamma modulators as safe and effective antidiabetic
agents. Expert Opin Investig Drugs, 19, 489-512.
Fineschi S, Goffin L, Rezzonico R, Cozzi F, Dayer J M, Meroni P L, et al. (2008).
Antifibroblast antibodies in systemic sclerosis induce fibroblasts to produce
profibrotic chemokines, with partial exploitation of toll-like receptor 4. Arthritis
Rheum, 58, 3913-23.
Galli A, Crabb D W, Ceni E, Salzano R, Mello T, Svegliati-Baroni G, et al. (2002).
Antidiabetic thiazolidinediones inhibit collagen synthesis and hepatic stellate cell
activation in vivo and in vitro. Gastroenterology, 122, 1924-40.
Galuppo M, Esposito E, Mazzon E, Di Paola R, Paterniti I, Impellizzeri D, et al. (2011).
MEK inhibition suppresses the development of lung fibrosis in the bleomycin model.
Naunyn Schmiedebergs Arch Pharmacol, 384, 21-37.
Garcia-Gonzalez E, Selvi E, Balistreri E, Lorenzini S, Maggio R, Natale M R, et al. (2009).
Cannabinoids inhibit fibrogenesis in diffuse systemic sclerosis fibroblasts.
Rheumatology (Oxford), 48, 1050-6.
Ghosh A K, Bhattacharyya S, Wei J, Kim S, Barak Y, Mori Y, et al. (2009). Peroxisome
proliferator-activated receptor-gamma abrogates Smad-dependent collagen
stimulation by targeting the p300 transcriptional coactivator. FASEB J, 23, 2968-77.
Gilbane A J, Denton C P & Holmes A M. (2013). Scleroderma pathogenesis: a pivotal role
for fibroblasts as effector cells. Arthritis Res Ther, 15, 215.
Gonzalez E G, Selvi E, Balistreri E, Akhmetshina A, Palumbo K, Lorenzini S, et al. (2012).
Synthetic cannabinoid ajulemic acid exerts potent antifibrotic effects in experimental
models of systemic sclerosis. Ann Rheum Dis, 71, 1545-51.
Guillot A, Hamdaoui N, Bizy A, Zoltani K, Souktani R, Zafrani E S, et al. (2014).
Cannabinoid receptor 2 counteracts interleukin-17-induced immune and fibrogenic
responses in mouse liver. Hepatology, 59, 296-306.
Harris S G & Phipps R P. (2001). The nuclear receptor PPAR gamma is expressed by mouse
T lymphocytes and PPAR gamma agonists induce apoptosis. Eur J Immunol, 31,
1
098-105.
Ho K T & Reveille J D. (2003). The clinical relevance of autoantibodies in scleroderma.
Arthritis Res Ther, 5, 80-93.
Hughes T S, Giri P K, De Vera I M, Marciano D P, Kuruvilla D S, Shin Y, et al. (2014). An
alternate binding site for PPARgamma ligands. Nat Commun, 5, 3571.
This article is protected by copyright. All rights reserved.

Katchan V, David P & Shoenfeld Y. (2016). Cannabinoids and autoimmune diseases: A
systematic review. Autoimmun Rev, 15, 513-28.
Kayser C & Fritzler M J. (2015). Autoantibodies in systemic sclerosis: unanswered questions.
Front Immunol, 6, 167.
Leask A. (2012). MEK/ERK inhibitors: proof-of-concept studies in lung fibrosis. J Cell
Commun Signal, 6, 59-60.
Lodzki M, Godin B, Rakou L, Mechoulam R, Gallily R & Touitou E. (2003). Cannabidiol-
transdermal delivery and anti-inflammatory effect in a murine model. J Control
Release, 93, 377-87.
Luchetti M M, Moroncini G, Jose Escamez M, Svegliati Baroni S, Spadoni T, Grieco A, et al.
(2016). Induction of Scleroderma Fibrosis in Skin-Humanized Mice by
Administration of Anti-Platelet-Derived Growth Factor Receptor Agonistic
Autoantibodies. Arthritis Rheumatol, 68, 2263-73.
Marangoni R G, Korman B D, Wei J, Wood T A, Graham L V, Whitfield M L, et al. (2015).
Myofibroblasts in murine cutaneous fibrosis originate from adiponectin-positive
intradermal progenitors. Arthritis Rheumatol, 67, 1062-73.
Marquart S, Zerr P, Akhmetshina A, Palumbo K, Reich N, Tomcik M, et al. (2010).
Inactivation of the cannabinoid receptor CB1 prevents leukocyte infiltration and
experimental fibrosis. Arthritis Rheum, 62, 3467-76.
Mccubrey J A, Steelman L S, Chappell W H, Abrams S L, Wong E W, Chang F, et al.
(2007). Roles of the Raf/MEK/ERK pathway in cell growth, malignant transformation
and drug resistance. Biochim Biophys Acta, 1773, 1263-84.
Mcgrath J C & Lilley E. (2015). Implementing guidelines on reporting research using
animals (ARRIVE etc.): new requirements for publication in BJP. Br J Pharmacol,
1
72, 3189-93.
Mihai C & Tervaert J W. (2010). Anti-endothelial cell antibodies in systemic sclerosis. Ann
Rheum Dis, 69, 319-24.
Milam J E, Keshamouni V G, Phan S H, Hu B, Gangireddy S R, Hogaboam C M, et al.
(2008). PPAR-gamma agonists inhibit profibrotic phenotypes in human lung
fibroblasts and bleomycin-induced pulmonary fibrosis. Am J Physiol Lung Cell Mol
Physiol, 294, L891-901.
Morris G M, Huey R, Lindstrom W, Sanner M F, Belew R K, Goodsell D S, et al. (2009).
AutoDock4 and AutoDockTools4: Automated docking with selective receptor
flexibility. J Comput Chem, 30, 2785-91.
Pacher P, Batkai S & Kunos G. (2006). The endocannabinoid system as an emerging target of
pharmacotherapy. Pharmacol Rev, 58, 389-462.
Pattanaik D, Brown M, Postlethwaite B C & Postlethwaite A E. (2015). Pathogenesis of
Systemic Sclerosis. Front Immunol, 6, 272.
Pistis M & O'sullivan S E. (2017). The Role of Nuclear Hormone Receptors in Cannabinoid
Function. Adv Pharmacol, 80, 291-328.
Ragusa G, Gomez-Canas M, Morales P, Hurst D P, Deligia F, Pazos R, et al. (2015).
Synthesis, pharmacological evaluation and docking studies of pyrrole structure-based
CB2 receptor antagonists. Eur J Med Chem, 101, 651-67.
Rajapaksha H, Bhatia H, Wegener K, Petrovsky N & Bruning J B. (2017). X-ray crystal
structure of rivoglitazone bound to PPARgamma and PPAR subtype selectivity of
TZDs. Biochim Biophys Acta, 1861, 1981-1991.
Ruzehaji N, Frantz C, Ponsoye M, Avouac J, Pezet S, Guilbert T, et al. (2016). Pan PPAR
agonist IVA337 is effective in prevention and treatment of experimental skin fibrosis.
Ann Rheum Dis, 75, 2175-2183.
This article is protected by copyright. All rights reserved.

Shi Y & Massague J. (2003). Mechanisms of TGF-beta signaling from cell membrane to the
nucleus. Cell, 113, 685-700.
Southan C, Sharman J L, Benson H E, Faccenda E, Pawson A J, Alexander S P, et al. (2016).
The IUPHAR/BPS Guide to PHARMACOLOGY in 2016: towards curated
quantitative interactions between 1300 protein targets and 6000 ligands. Nucleic Acids
Res, 44, D1054-68.
Trott O & Olson A J. (2010). AutoDock Vina: improving the speed and accuracy of docking
with a new scoring function, efficient optimization, and multithreading. J Comput
Chem, 31, 455-61.
Valiveti S, Hammell D C, Earles D C & Stinchcomb A L. (2004). Transdermal delivery of
the synthetic cannabinoid WIN 55,212-2: in vitro/in vivo correlation. Pharm Res, 21,
1
137-45.
Varga J & Abraham D. (2007). Systemic sclerosis: a prototypic multisystem fibrotic disorder.
J Clin Invest, 117, 557-67.
Walton K L, Johnson K E and Harrison C A. (2017). Targeting TGF-β mediated SMAD
signaling for the prevention of fibrosis. Front Pharmacol, 8, 461.
Wei J, Zhu H, Komura K, Lord G, Tomcik M, Wang W, et al. (2014). A synthetic PPAR-
gamma agonist triterpenoid ameliorates experimental fibrosis: PPAR-gamma-
independent suppression of fibrotic responses. Ann Rheum Dis, 73, 446-54.
Wolf L K. (2009). New software and Websites for the Chemical Enterprise. Chemical &
Engineering News, 87
Wu M, Melichian D S, Chang E, Warner-Blankenship M, Ghosh A K & Varga J. (2009).
Rosiglitazone abrogates bleomycin-induced scleroderma and blocks profibrotic
responses through peroxisome proliferator-activated receptor-gamma. Am J Pathol,
1
74, 519-33.
Yamamoto T. (2006). The bleomycin-induced scleroderma model: what have we learned for
scleroderma pathogenesis? Arch Dermatol Res, 297, 333-44.
Yamamoto T, Takagawa S, Katayama I, Yamazaki K, Hamazaki Y, Shinkai H, et al. (1999).
Animal model of sclerotic skin. I: Local injections of bleomycin induce sclerotic skin
mimicking scleroderma. J Invest Dermatol, 112, 456-62.
This article is protected by copyright. All rights reserved.

Figure 1. VCE-004.3 is a selective PPARγ agonist. (A) Schematic synthesis of VCE-004.3.
B) VCE-004.3 and RGZ binding affinity to PPARγ receptor. Results are represented in a
(
logarithmic scale and IC50 values were calculated as the 50% inhibition of percentage of
fluorescence polarization (n=5). (C) VCE-004.3 PPARγ transcriptional activation and VCE-
0
04.3 cytotoxicity in NIH-3T3 fibroblasts. Cells were co-transfected with PPARγ-GAL4 and
GAL4-luc constructs, stimulated with VCE-004.3 or RGZ as positive control for 6 h and
lysed for luciferase activity. Results are represented in a bar chart and shown as fold
induction referred to control (n=10). To test cytotoxicity cells were treated with the
compound for 6 h in the presence of YOYO-1 and fluorescence intensity was measured and
represented in a dot chart (n=5). (D-E) HEK-293T cells were co-transfected with PPARα-
GAL4 (D) or PPARδ-GAL4 (E) and GAL4-luc, treated with VCE-for 6 h and lysed for
*
luciferase activity (n=5). Results are shown as mean ± S.D. p<0.05 versus control.
This article is protected by copyright. All rights reserved.

Figure 2. VCE-004.3 competed with RGZ for the canonical pocket and binds to an
alternate site in PPARγ. NIH-3T3 cells were transfected with PPARγ-GAL4 plus GAL4-
luc. (A) Cells were pre-treated with VCE-004.3 for 1 h and then incubated for 6 h in the
presence of RGZ (n=5). (B) Cells were pre-treated with VCE-004.3 for 1 h and then washed
with PBS and stimulated with RGZ for 6 h (n=5). (C) PPAR LBD structures 3B0R and
5
U5L bound to VCE-004.3 (blue) with and without of GW9662 (orange). (D-E) NIH-3T3
cells were transfected with PPARγ-GAL4 plus GAL4-luc, pre-treated with T0070907 (5 µM)
for 30 min and stimulated with RGZ (D) or VCE-004.3 (E) for 6 h (n=5). Cells were lysed
*
and tested for luciferase activity. Results are shown as mean ± S.D. p<0.05 versus untreated
cells.
This article is protected by copyright. All rights reserved.

Figure 3. VCE-004.3 is a functional CB
binding affinity to CB receptor membranes. Results are expressed as mean ± SEM. (n=5).
B) CB receptor antagonism. HEK293-T-CB cells transfected with pCRE-luc plasmid were
1
antagonist and CB
2
agonist. (A) VCE-004.3
1
(
1
1
pretreated for 30 min with VCE-004.3, stimulated with WIN55,212 (1µM) for 6 h and lysed
for luciferase activity. Results are expressed as mean ± S.D. (n=5) (C) Binding affinity
activity of VCE-004.3 to CB
2
receptor membranes. Results are represented as mean ± SD.
n=5). (E) VCE-004.3 is a functional CB receptor agonist. HEK 293-T cells stably
receptor were transfected with pCRE-luc plasmid and pretreated for 30 min
(
2
expressing CB
2
with VCE-004.3 or WIN55,212 as positive control. Then, cells were stimulated with FSK (10
*

M) for 6 h and lysed for luciferase activity. Results are shown as mean ± S.D. (n=5) p<0.05
#
versus untreated cells p<0.05 versus positive control.
This article is protected by copyright. All rights reserved.

Figure 4. Effect of VCE-004.3 on collagen gene transcription and synthesis in vitro. (A)
NIH-3T3-Col1a2-luc cells were pre-treated with VCE-004.3 for 1 h and stimulated with
TGFβ1 for the following 24 h (n=5). (B) NIH-3T3 cells were transfected with CAGA-luc
plasmid, pre-treated with VCE-004.3 for 1 h and stimulated with TGFβ1 for 6 h (n=5). Cells
were tested for luciferase activity. (C) Serum-starved NIH-3T3 cells were preincubated with
the compound for 1 h and stimulated with TGFβ1 for 2 h. Protein expression was studied by
Western Blot and values under images represent mean fold induction of signal intensities
after β-actin normalization (n=5) (D-E) NHDFs were serum starved (1% FBS) for 24 h.
Then, cells were pre-treated with VCE-004.3 for 1 h and stimulated with TGFβ1 for 24 h.
Soluble collagen in the culture medium was measured using the Sircol assay (D), and
collagen deposition was studied by the Sirius Red method (E). Results are represented as
*
mean percentage of inhibition ± S.D. taking TGFβ1 as 100%. p<0.05 versus TGFβ1.
This article is protected by copyright. All rights reserved.

Figure 5. VCE-004.3 alleviated skin inflammation and collagen accumulation induced
by BLM. Mice were injected with BLM for three weeks and treated in parallel with i.p.
injections of RGZ, VCE-004.3 or vehicle. (A) Representative images of Masson's trichrome
staining of skin sections and their respective measurement of dermal and subcutaneous
adipose layer thickness. (B) Representative images of collagen staining by picrosirius red dye
and their quantification. (C) Gene expression of profibrotic and proinflammatory genes
including Il-6, Tgfβ, Il-4, Ccl-2, Il1β and Il-13 were measured by q-RT-PCR. Results are
*
represented as mean ± SEM referred to control group (n=9 animals per group). p<0.05
#
versus control; p<0.05 versus BLM-treated mice.
This article is protected by copyright. All rights reserved.