ORIGINAL ARTICLE

Inhibition of EGFR Tyrosine Kinase by Erlotinib

Prevents Sclerodermatous Graft-Versus-Host Disease
in a Mouse Model
Florence Morin1,2,3, Niloufar Kavian1,2,3, Wioleta Marut1, Christiane Chéreau1, Olivier Cerles1,
Philippe Grange1, Bernard Weill1,2, Carole Nicco1,3 and Frédéric Batteux1,2,3

Chronic graft-versus-host disease (GVHD) follows allogeneic hematopoietic stem cell transplantation. It results
from alloreactive processes induced by minor histocompatibility antigen incompatibilities leading to the
activation of CD4 T cells and the development of fibrosis and inflammation of the skin and visceral organs and
autoimmunity that resemble systemic sclerosis. EGFR is a ubiquitous cell receptor deeply involved in cell
proliferation, differentiation, and motility. EGFR has recently been implicated in autoimmune and fibrotic
diseases. Therefore, we tested whether Erlotinib, an EGFR tyrosine kinase inhibitor, can prevent sclerodermatous
GVHD (Scl-GVHD). Scl-GVHD was induced in BALB/c mice by B10.D2 bone marrow and spleen cell
transplantation. Transplanted mice displayed severe clinical symptoms including alopecia, fibrosis of the skin
and visceral organs, vasculitis, and diarrhea. The symptoms were reversed in mice treated with Erlotinib. These
beneficial effects were mediated by the decreased production of activated/memory CD4+ T cells and the
reduction in T-cell infiltration of the skin and visceral organs along with a decrease in IFN-γ and IL-13 production
and autoimmune B-cell activation. The improvement provided by Erlotinib in the mouse model of Scl-GVHD
supplies a rationale for the evaluation of Erlotinib in the management of patients affected by chronic GVHD.
Journal of Investigative Dermatology (2015) 135, 2385–2393; doi:10.1038/jid.2015.174; published online 28 May 2015

INTRODUCTION and B cells from donor to induce chronic GVHD (Linhares

Chronic graft-versus-host disease (GVHD) is a side effect of
allogeneic hematopoietic stem cell transplantation, with
variable clinical presentations (Baird and Pavletic, 2006).
Chronic GVHD results from alloreactive processes induced
by minor major histocompatibility complex incompatibilities
between donor and recipient and involves donor-derived
immune cells and host cell populations. Its pathophysiology is
less well known than acute GVHD (Ferrara et al., 2009).
Chronic GVHD often mimics autoimmune diseases
(Filipovich et al., 2005). Initially, chronic GVHD was associ-
ated with a T helper type 2 (Th2)-type immune response, but
IFN and IL-17 seem implicated in its induction. Furthermore,
loss of immune tolerance by impaired thymic negative
selection allows expansion and activation of autoreactive T
1
INSERM U 1016, Institut Cochin, Université Paris Descartes Sorbonne Paris
Cité, Faculté de Médecine, Paris, France and 2Laboratoire d’Immunologie
biologique, Hôpital Cochin, AP-HP, Paris, France
Correspondence: Frédéric Batteux, Laboratoire d’Immunologie, INSERM U
1016, Institut Cochin, 8 rue Méchain, 75679 Paris Cedex 14, France.
E-mail: frederic.batteux@cch.aphp.fr
3
These authors contributed equally to this work.
Abbreviations: ALT, alanine aminotransferase; AST, aspartate
aminotransferase; AU, arbitrary unit; eGFP, enhanced green fluorescent
protein; pEGFR, phospho-EGFR; GVHD, graft-versus-host disease; Scl-GVHD,
sclerodermatous graft-versus-host disease
Received 26 August 2014; revised 3 April 2015; accepted 18 April 2015;
accepted article preview online 4 May 2015; published online 28 May 2015
et al., 2013; Nishimori et al., 2013; Socié and Ritz, 2014).
Sclerodermatous GVHD (Scl-GVHD) represents 10 to 15% of
cases of chronic GVHD (Vogelsang et al., 2003). This clinical
form, similar to scleroderma, includes fibrotic changes and
chronic inflammation of the skin, lung, and gastrointestinal
tract. Murine Scl-GVHD, produced by transplanting B10.D2
(H-2d) bone marrow and spleen cells across minor
histocompatibility loci into lethally irradiated BALB/cJ (H-2d)
recipients, displays many clinical and histologic similarities
with scleroderma (Jaffee and Claman, 1983). Syngeneic
(BALB/c into BALB/c) transplanted animals are used as
controls. This animal model summarizes the clinical features
of human Scl-GVHD including fibrosis of the skin and inner
organs and infiltrating cutaneous mononuclear cells (pre-
dominantly T cells) starting 14 days following bone marrow
and splenocyte transplantation (Yamamoto, 2010).
We have observed in a mouse model of scleroderma that
angiotensin II elicits skin and visceral fibrosis that can be
abrogated by angiotensin II receptor blockade using Irbesar-
tan (Marut et al., 2013). Angiotensin II increases fibroblast
proliferation, myofibroblasts differentiation, and the accumu-
lation of type I collagen through the phosphorylation of
Smad3, a transcription factor implicated in the activation of
the type I collagen gene, and through the activation of the
nicotinamide adenine dinucleotide phosphate oxidase and the
overproduction of reactive oxygen species (Yang et al., 2009).

© 2015 The Society for Investigative Dermatology www.jidonline.org 2385

 F Morin et al.
Erlotinib Prevents Scl-GVHD
The latter phenomenon is responsible for EGFR phos-
phorylation. The transactivation of EGFR by angiotensin
initiates the proliferation and migration of fibroblasts, a
phenomenon that can be abrogated by AT1 antagonist
(Yahata et al., 2006). This link between EGFR and fibrosis
explains the attenuation of liver fibrosis and of the develop-
ment of hepatocellular carcinoma by EGFR inhibition using
Erlotinib (Fuchs et al., 2014).
Erlotinib is a potent inhibitor of EGFR tyrosine kinase used
to treat advanced or metastatic non-small-cell lung cancer
after failure of first-line or second-line chemotherapy (Akita
and Sliwkowski, 2003; Johnson et al., 2005; Robert, 2011). In
addition to its antiproliferative properties against tumor cells,
Erlotinib inhibits T-cell proliferation and Th1-cell–mediated
immune response in vitro and in vivo through the inhibition of
the c-Raf/extracellular signal–regulated kinase cascade and
the Akt signaling pathway (Luo et al., 2011).
As GVHD involves both Th1 cell–mediated immune response
and fibrosis, we were prompted to investigate the effects of
Erlotinib on the development of chronic Scl-GVHD in mice.
RESULTS
Phospho-EGFR expression in mouse ears from GVHD animals
and modulation by Erlotinib
Levels of phospho-EGFR (pEGFR) were assessed by immuno-
histochemistry on ear pinna sections and by western blot on
lysates of ear lobes. pEGFR expression was greater in mouse
ear pinna that developed Scl-GVHD than in control mice
with a syngeneic graft (Figure 1a). This result was confirmed
by western blot. pEGFR expression was increased by 32%
compared with syngeneic control mice (P = 0.043, Figure 1b).
Erlotinib decreased the expression of phospho-EGFR com-
pared with untreated GVHD mice (0.67 ± 0.019 vs. 0.53 ±
0.025 arbitrary unit (AU), P = 0.0027, Figure 1b).
Disease severity score
The clinical score assessment started 10 days after transplan-
tation. At day 23, no difference between groups was
observed. At day 26, the score of the GVHD group reached
1.11 ± 0.3 AU, whereas the score of the syngeneic group
remained null (Po0.0001, Figure 2a). At the time of killing on
day 33, a significant improvement in the Erlotinib group was
observed versus untreated mice (2.39 ± 0.20 vs. 1.65 ± 0.10
AU, P = 0.047 at day 33, Figure 2a).
Skin involvement
At 7 days after transplantation, ear thickness was increased in
GVHD groups compared with syngeneic groups (30.0 ± 0.4
vs. 42.2 ± 0.8 μm, Po0.0001 at day 17, Figure 2b). From day
23, a decrease in earlobe thickening was observed in GVHD
mice treated with Erlotinib compared with untreated animals.
At the time of killing, the mean ear pinna thickness was lower
in Erlotinib-treated GVHD mice than in untreated mice
(45.9 ± 1.2 vs. 35.8 ± 1.4 μm, Po0.0001, Figure 2b). Sirius
red staining showed an increased collagen deposition in
GVHD mice pinna skin versus the three other groups
(Figure 2c).
2386 Journal of Investigative Dermatology (2015), Volume 135
a Syngeneic GVH
b pEGFR
β-Actin
Syn Syn Syn Syn GVH GVH GVH GVH
Erlo Erlo Erlo Erlo
pEGFR
0.8
*** *
0.7
AU
0.6
0.5
0.4
Syn Syn-Erlo GVH GVH-Erlo
Figure 1. Effects of Erlotinib on mouse ear skin following graft-versus-host
disease (GVHD) induction and oral administration of Erlotinib (Erlo).
(a) Immunohistochemical study of phospho-EGFR (pEGFR) on 7 μm ear skin
sections in syngeneic (Syn) BALB/c mice and in GVHD mice, revealed with
diaminobenzidine tetrahydrochloride. (b) Western blot of pEGFR expression in
the ear skin of BALB/c mice submitted to a syngeneic or a nonsyngeneic graft.
Photographs were taken with a Nikon Eclipse 80i microscope. Original
magnification × 50. Values in b are mean ± SEM. *P ≤ 0.05; ***P ≤ 0.001.
AU, arbitrary unit.
Lymphocytic infiltration of the skin
Cellular infiltration of the skin of Erlotinib-treated GVHD
mice was lower than in untreated animals (3.80 ± 0.55 vs.
12.86 ± 0.88, P = 0.0008, Figure 3a and b). Costaining with
phycoerythrin-labeled anti-CD3 mAb showed that green
donor cells were CD3-T lymphocytes (10.57 ± 0.48 vs.
3.00 ± 0.30, Po0.0001, Figure 3a and b). IFN-γ production
by infiltrating lymphocytes was significantly increased in
GVHD mice compared with syngeneic mice (236.5 ± 73.3 vs.
5,828 ± 896 pg ml− 1, Po 0.0001, Figure 3c). Erlotinib sig-
nificantly decreased IFN-γ secretion (3,143 ± 463.7 vs.
5,828 ± 896.0 pg ml − 1 in untreated animals, P = 0.0178,
Figure 3c). IL-13 secretion in the skin significantly increased
in GVHD mice compared with syngeneic mice (1,420 ± 241
vs. 124 ± 32 pg ml − 1, P = 0.0002, Figure 3c). Erlotinib
decreased IL-13 secretion (792 ± 180 vs. 1,420 ± 241 pg ml − 1
in untreated animals, P = 0.036, Figure 3c). These results have
been confirmed by quantitative real-time reverse-transcrip-
tase–PCR for IFN-γ and IL-13. IFN-γ mRNAs were higher in
the skin of GVHD mice than in syngeneic mice (7.69 ± 0.50-
foldvs. syngeneic mice, P = 0.0028, Figure 3d). Following
Erlotinib, IFN-γ mRNA levels in GVHD mice returned to
 F Morin et al.
Erlotinib Prevents Scl-GVHD

a 3 b 55
Syn
Syn-Erlo

Ear thickness (μm)

50

Clinical score - AU
GVH
2 GVH-Erlo 45
*
40 ***
1 35
30
0 25
0 10 20 30 40 0 10 20 30 40
Days Days

c
Syngeneic Syngeneic Erlotinib

GVH GVH Erlotinib

Figure 2. Effects of Erlotinib (Erlo) on clinical features following graft-versus-host disease (GVHD) induction and oral administration of Erlotinib. (a) Evolution
of the GVHD score in the different groups during Erlotinib treatment. (b) Evolution of ear skin thickness during Erlotinib treatment. (c) Representative Sirius
red–dyed ear skin sections of 7 μm, showing fibrosis in syngeneic (Syn) BALB/c mice and in GVHD mice, treated by Erlotinib or not. Photographs were taken
with a Nikon Eclipse 80i microscope. Original magnification × 50. Values in a and b are mean ± SEM. *P ≤ 0.05; ***P ≤ 0.001. Bars represent mean ± SEM. AU,
arbitrary unit.

initial levels (1.03 ± 0.04-fold vs. 7.69 ± 0.50-fold in the
untreated GVHD group, P = 0.0029, Figure 3d). IL-13 mRNAs
were increased in GVHD mice compared with syngeneic
(3.69 ± 0.07-fold vs. syngeneic mice, Po0.0001, Figure 3d).
This level was decreased following Erlotinib compared with
untreated mice (0.52 ± 0.04-fold vs. 3.69 ± 0.07-fold in the
untreated GVHD group, Po0.0001, Figure 3d). CCL17
mRNAs were overexpressed in the skin of GVHD mice
compared with syngeneic mice (3.34 ± 0.54-fold vs. syn-
geneic mice, P = 0.0017). Erlotinib abolished the overexpres-
sion of CCL17 in GVHD mice (0.71 ± 0.41-fold vs.
3.34 ± 0.54-fold in the untreated GVHD group, P = 0.0026,
Figure 3d). Similarly, CCL21 mRNAs were higher in GVHD
mice than in syngeneic mice (6.51 ± 0.29-fold vs. syngeneic
mice, P = 0.0016; Figure 3d). This mRNA level was decreased
in Erlotinib-treated GVHD mice compared with untreated
mice (1.23 ± 0.10-fold vs. 6.51 ± 0.29-fold in the untreated
GVHD group, P = 0.0020; Figure 3d). No difference in CCL27
mRNAs was found among the four groups (Figure 3d).
Prevention of GVHD-associated liver disease by Erlotinib
Serum alanine aminotransferase (ALT) and aspartate
aminotransferase (AST) were higher in GVHD -untreated
mice than in syngeneic groups (ALT: 184.6 ± 31.15 vs.
55.7 ± 5.85 UI l − 1, Po0.05, and AST: 464.71 ± 79.37 vs.
160.33 ± 12.38 UI l − 1, Po0.01, Figure 4b and c). Erlotinib
decreased transaminase levels in GVHD mice (ALT: 104.88 ±
12.60 vs. 184.57 ± 31.15 UI l − 1, Po0.05, and AST: 291.25 ±
17.04 vs. 464.71 ± 79.37 UI l − 1, Po0.05, Figure 4b and c).
Liver sections from untreated GVHD mice showed portal
inflammation and bile duct destruction (Figure 4a). GVHD
mice treated with Erlotinib displayed a significant reduction in
liver lesions compared with untreated GVHD animals
(0.63 ± 0.12 vs. 2.01 ± 0.47 AU, Po0.05). Mice grafted with
syngeneic bone marrow and splenocytes, treated or not with
Erlotinib, showed no liver inflammation (0.25 ± 0.07 vs.
0.19 ± 0.08 AU, P = nonsignificant).
Effects of Erlotinib on B- and T-cell responses
Anti-DNA topoisomerase I autoantibodies were increased
in GVHD mice compared with untreated syngeneic mice
(Po0.001, Figure 5a). Erlotinib significantly reduced the
levels of anti-DNA topoisomerase I autoantibodies (1.15 ±
0.23 vs. 2.65 ± 0.82 AU, Po0.001, Figure 5a). Furthermore,
Erlotinib reduced the number of B cells by 43% in syngeneic
mice compared with the untreated group (Po0.001,
Figure 5b). No difference in B-cell counts was observed
between treated and untreated GVHD mice.

www.jidonline.org 2387
 F Morin et al.
Erlotinib Prevents Scl-GVHD

a c IFN-γ
8 000
GVH GVH Erlotinib *
6 000

pg ml–1
4 000
2 000
0

lo

rlo
VH
Sy

Er

-E
G
n-

VH
Sy

G
GVH GVH Erlotinib IL-13
3 000

pg ml–1
2 000
*
1 000

lo

rlo
VH
Sy

Er

-E
G
n-

VH
Sy

G
b d qRT-PCR
Syn
eGFP Anti-CD3 10 Syn-Erlo
15 GVH
**
Mean cell numbers

15 ***
Mean cell numbers

8 GVH-Erlo
***

Fold increase
**
10 10 6
4 *** **
5 5
2
0 0 0
rlo
rlo

-γ

13

1
VH

7
VH

7
L2
N

L1

L2
-E
-E

IL
IF
G
G

C
C

C
VH
VH

C
C

C
G
G

Figure 3. Effects of Erlotinib (Erlo) on mouse skin following graft-versus-host disease (GVHD) induction and oral administration of Erlotinib. (a) Representative
5 μm frozen skin sections with enhanced green fluorescent protein (eGFP) donor cells and phycoerythrin-conjugated anti-CD3 mAb staining. (b) Scores obtained
by enumerating eGFP- and anti-CD3-positive cells in 10 areas per frozen skin section. (c) Levels of IFN-γ and IL-13 produced by skin cells and used as
markers of T helper type 1 and 2 (Th1 and Th2) responses. (d) Levels of mRNAs for IFN-γ, IL-13, CCL17, CCL21, and CCL27 in the skin. Photographs
were taken with a Nikon Eclipse 80i microscope. Original magnification: × 100. Values in b, c, and d are mean ± SEM. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001.
qRT-PCR, quantitative real-time reverse-transcriptase–PCR; Syn, syngeneic.

Similarly, CD4 T cells decreased by 40% in the Erlotinib-
treated syngeneic group compared with the untreated group
(Po0.05, Figure 5c). There was no effect on CD8 cell
numbers in treated and untreated syngeneic mice (Figure 5d).
If the levels of CD4 and CD8 T cells were significantly
decreased in untreated GVHD mice compared with untreated
syngeneic animals (93 and 87%, respectively Po0.001,
Figure 5c and d), no significant difference between the levels
of CD4 and of CD8 T cells was observed between treated and
untreated GVHD animals (Figure 5c and d ). There was an
increase in the percentages of CD4 and CD8 memory T cells
within CD4 and CD8 T-cell subsets in GVHD mice compared
with syngeneic animals (1.6 ± 1.4% vs. 83 ± 3%, Po 0.001,
Figure 5e and 50 ± 17% vs. 83 ± 7%, Po0.001, Figure 5f).
Erlotinib did not influence the percentage of CD4+ and CD8+
T cells in the syngeneic group (Figure 5e and f). Erlotinib
induced an increase in percent CD4 naive T cells in GVHD
mice compared with untreated mice (45% vs. 17 ± 1.6 ±
1.4%, Po0.001, Figure 5e), whereas it had no effect on CD8
naive T cells.
DISCUSSION
Here we show that Scl-GVHD is associated with an activation
of the EGFR pathway whose inhibition by the tyrosine kinase
inhibitor Erlotinib prevents skin and visceral lesions through
an inhibition of the allogeneic response.
Experimental chronic GVHD shares characteristic features
with scleroderma, including skin and visceral fibrosis,
inflammation with systemic sclerosis, and autoimmunity.
Erlotinib improves the clinical outcome of sublethally
irradiated BALB/c mice transplanted with B10.D2 hemato-
poietic cells. Fibrosis, alopecia, and vasculitis are clearly
reduced in treated versus untreated mice. EGFR is a
ubiquitous cell receptor involved in proliferation, differentia-
tion, and cell motility (Schneider and Wolf, 2009). This
receptor has a role in tissue fibrosis through the control of
fibroblast proliferation and migration (Pastore et al., 2008).
Moreover, EGFR signaling promotes fibrosis through a trans-
forming growth factor-β–dependent mechanism, implicating
activation of smad2/3 in various tissues like kidney or skin.
This could explain why EGFR inhibition by Erlotinib attenu-
ates liver fibrosis and the development of hepatocellular
carcinoma (Fuchs et al., 2014).
Liver involvement is a common feature in human GVHD
that also occurs in the Scl-GVHD model (Jaffee and Claman,
1983; Lee et al., 2002). Indeed, in GVHD mice, we found
a periportal lymphocytic infiltration associated with an
increase in serum transaminase levels. Erlotinib decreased

2388 Journal of Investigative Dermatology (2015), Volume 135

 F Morin et al.
Erlotinib Prevents Scl-GVHD

a Syngeneic Syngeneic Erlotinib

GVH GVH Erlotinib

b 250 c
600 *
*
200

AST (UI I–1)

ALT (UI I–1)

400
150

100
200
50

0 0
Syn Syn-Erlo GVH GVH-Erlo Syn Syn-Erlo GVH GVH-Erlo
Figure 4. Effects of Erlotinib (Erlo) on liver involvement following graft-versus-host disease (GVHD) induction and oral administration of Erlotinib.
(a) Hematoxylin–eosin staining of liver sections. GVH and GVH Erlo groups showed mixed lymphomonocytic infiltrations around the portal vein with a major
reduction in the GVH Erlo group. (b) Serum aspartate aminotransferase (AST) levels in various experimental groups. (c) Serum alanine aminotransferase (ALT)
levels in various experimental groups. Photographs were taken using a Nikon Eclipse 80i microscope. The arrows showed mixed lymphocytic infiltrations around
the portal vein. Original magnification × 100. Values in b and c are mean ± SEM. *P ≤ 0.05. Syn, syngeneic.

transaminase levels and reduced the periportal lymphocytic
infiltrate. Thus, in addition to its direct antifibrotic properties
(Fuchs et al, 2014), Erlotinib inhibits the EGFR pathway and,
consecutively, abrogates tissue infiltration and inflammation
of the liver.
The pathophysiology of chronic GVHD is still poorly
understood. However, evidence suggests that, unlike acute
GVHD that is CD8+ T-cell dependent, the events observed in
chronic GVHD depend on the activation of donor CD4+
T cells specific of minor histocompatibility antigen (Korngold
and Sprent, 1978; Kansu, 2004; Shlomchik, 2007; Chu and
Gress, 2008). After transplantation of spleen and bone
marrow cells, naive CD4+ T cells from B10.D2 initiate the
disease in irradiated BALB/c mice. Activated donor CD4+
T cells infiltrate the skin and induce fibrosis (Korngold and
Sprent, 1978; Kansu, 2004; Chu and Gress, 2008). In our
hands, the beneficial effect of Erlotinib was associated with a
reduction in the percentages of effector memory CD4+ T cells
(CD69 −CD4+CD44highCD62Llow) but not of CD8+ T cells in
GVHD mice.
This reduction is associated with a diminution in infiltrating
T cells in the skin and with a profound decrease in CCL17 and
CCL21 expression. Increased CCL17 mRNAs have already
been shown by Zhou et al. (2007) in murine Scl-GVHD.
CCL17 is the ligand of CCR4 on Th2 cells, basophils, and
natural killer cells. CCL17 produced by dendritic cells,
keratinocytes, fibroblasts, and endothelial cells (Saeki and
Tamaki, 2006) has been implicated in various skin diseases
such as atopic dermatitis and bullous pemphigoid (Furudate
et al., 2014; Hamilton et al., 2014). CCL21 is the ligand of
CCR7 expressed by T cells, B cells, and dendritic cells
(Comerford et al., 2013). It has been implicated in contact
allergic dermatitis (Eberhard et al., 2004).
In addition to its antiproliferative properties against tumor
cells, Erlotinib exerts inhibitory effects on T-cell proliferation.
Erlotinib causes G0/G1 arrest and inhibits the secretion of IL-2
and the phosphorylation of c-Raf, extracellular signal–
regulated kinase, and Akt (Luo et al., 2011). Erlotinib acts as
other antiproliferative molecules like steroids, cyclosporine,
sirolimus, or mycophenolate that are treatments for chronic
GVHD (Wolff et al., 2010, 2011 ). Erlotinib impairs Th1-
mediated immune response as it inhibits the secretion of IFN-γ
by activated T cells in vitro and ameliorates in vivo picryl
chloride–induced ear contact dermatitis (Luo et al., 2011).
IFN-γ is involved in the pathophysiology of chronic GVHD
(Korholz et al., 1997; Bruggen et al., 2014). It induces an
overexpression of HLA and adhesion molecules in target
tissues and favors the recruitment of macrophages. Therefore,
its inhibition by Erlotinib can participate in the improvement
of GVHD. Recently, the immunomodulating role of Erlotinib

www.jidonline.org 2389
 F Morin et al.
Erlotinib Prevents Scl-GVHD

a Anti-DNA Topoisomerase b Chronic GVHD is associated with the production of

I Abs B cells
4 20 autoantibodies consecutive to the production of Th2 cyto-
*** *** kines (Ruzek et al., 2004; Kavian et al., 2012). In line with

×106 B cells
3 15
previous reports, we observed a production of anti-DNA-
AU

2 NS 10
topoisomerase I antibodies in Scl-GVHD mice that was
1 5
NS decreased by Erlotinib. We conclude that Erlotinib targets
0 rlo 0 CD4+ T cells, blocks T- and B-cell cooperation in Scl-GVHD
n

lo

rlo
lo
VH

mice, and prevents the development of autoimmunity in

vH
Sy

Sy
Er

Er
-E

-E
G
G
n-

n-
VH

vH
our model.
Sy

Sy

G
G

c d This work highlights the role of the EGF receptor pathway

CD4 T cells CD8 T cells
1 500 400 in the pathophysiology of Scl-GVHD. The inhibition of EGFR
NS
* phosphorylation by Erlotinib reduces tissue inflammation and
×106 T cells

×106 T cells

300
1 000
200 fibrosis, limits the production of specific autoantibodies, and
500 NS prevents the production of CD4+ memory T cells. Inhibition of
NS 100
the EGFR pathway acts on the three main components of Scl-
0
0
GVHD: fibrosis, inflammation, and autoimmunity. These
n
n

lo

rlo

rlo
lo
vH

vH
Sy
Sy

Er

Er

mechanisms explain the beneficial effects of Erlotinib

-E

-E
G

G
n-

n-
vH

vH
Sy

Sy

observed and evidence the possible usefulness of this

G

e CD4 naive/memory+naive f CD8 naive/memory+naive molecule for the treatment of chronic GVHD in humans.
T cells T cells
100 NS 100 NS
80 80
MATERIALS AND METHODS
NS Animals, cells, and chemicals
60 *** 60
%
%

40 40
Six-week-old female BALB/c mice (Janvier Laboratory, Le Genest
20 20
Saint Isle, France) and male B10.D2-enhanced green fluorescent
0 0
protein (eGFP) mice (gift from Colette Kanellopoulos-Langevin,
CDTA–CNRS–Orléans, France) were used in all experiments.
rlo

rlo
lo
n

rlo
vH

vH
Sy

Sy
Er

-E

-E
-E
G

G
n-

Animals received humane care according to the guidelines of our

vH

vH
vH
Sy

G
G

institution (Inserm and Université Paris Descartes—CEEA34 Ethics

Figure 5. Inhibition of the production of autoantibodies in mice with graft- committee). All mice were housed in ventilated cages with sterile
versus-host disease (GVHD) systemic sclerosis treated in vivo with Erlotinib
food and water ad libitum. All cells were cultured as reported
(Erlo). Inhibition of the production of T cells with sclerodermatous GVHD (Scl-
previously (Kavian et al., 2010). All chemicals were obtained
GVHD) treated in vivo with Erlotinib. (a) Levels of anti-topoisomerase I
antibodies (Abs), as measured by ELISA. (b) Splenic B-cell numbers, as from Sigma-Aldrich (St Louis, MO), except for Erlotinib (Roche
assessed by flow cytometry. (c) Splenic CD4 T-cell numbers, as assessed by Registration, Meylan, France).
flow cytometry. (d) Splenic CD8 T-cell numbers, as assessed by flow
cytometry. (e) Ratio between naive and memory CD4 cells. (f) Ratio between Experimental procedure for induction of GVHD in BALB/c mice
naive and memory CD8 cells. Values in a–f are mean ± SEM. *P ≤ 0.05; GVHD following splenocyte and bone marrow transplantation was
***P ≤ 0.001. AU, arbitrary unit; NS, nonsignificant; Syn, syngeneic.
induced in BALB/c mice (H-2d; Janvier Laboratory) by grafting cells
from 7- to 8-week-old male B10.D2-eGFP mice (H-2d; Janvier
Laboratory), as previously described (Kavian et al., 2012). Briefly,
was observed following the treatment of head and neck recipient mice were lethally irradiated with 750 cGy from a
squamous cell carcinoma by an EGFR inhibitor that induces Gammacel [137Cs] source. After 3 hours, they were injected
the production of transforming growth factor-β and intravenously with donor spleen cells (5.106 per recipient) and
prostaglandin E2 by the tumor cells that attenuate antitumor bone marrow cells (2.106 per recipient), previously treated with a
immune responses (Kumai et al., 2014). Therefore, the hypotonic solution of potassium acetate for red blood cell lysis and
immunomodulating effects of Erlotinib associate intrinsic suspended in RPMI-1640. A control group of BALB/c recipient mice
effects on T-cell signaling and extrinsic effects through received syngeneic BALB/c spleen and bone marrow cells (syngeneic
paracrine immunomodulating signals from another target bone marrow transplantation, referred to as control animals).
cell population. In addition to IFN-γ, IL-13 is also highly
expressed in the skin of GVHD mice. Erlotinib decreases the Treatment of Scl-GVHD mice with Erlotinib
production of IL-13 in their skin. Following the graft of B10. Scl-GVHD and control mice were randomized and treated per os
D2 lymphoid cells into irradiated BALB/c mice, an increase in 5 days a week for 4 weeks with either Erlotinib (50 mg kg − 1 per day)
the expression of type 2 cytokines in the skin of Scl-GVHD or vehicle (methylcellulose 1%) alone, beginning on day 7 after bone
mice has already been observed (Zhou et al., 2007). marrow transplantation (9 mice per group). Animals were killed by
Moreover, in other chronic GVHD models, induction of cervical dislocation 7 weeks after transplantation.
GVHD fibrosis of the skin and of visceral organs requires type
2 immune responses. Consequently, the reduced production Disease severity score
of IL-13 in our model probably contributes to the amelioration To determine the incidence and severity of disease, we assigned
of skin fibrosis. a score to each mouse using the following criteria: 0: no external

2390 Journal of Investigative Dermatology (2015), Volume 135


sign; 1: piloerection on back and underside, 1: hunched posture
or lethargy; 0.5: alopecia o1 cm2; 1: alopecia 41 cm2; 1: vasculitis
(one or more purpural lesions on ears or tail); and 1: eyelid sclerosis
(blepharophymosis). The severity score is the sum of these values and
ranges from 0 (unaffected) to a maximum of 6. The incidence and
severity score were recorded every week by two blinded scientists.
Skin thickness
Skin thickness of the right and left ears lobes of mice was measured
each week with a digital calliper and expressed in mm.
Histopathological analysis
Liver and skin pieces fixed in formol were embedded in paraffin.
Then, 5-mm-thick tissue sections were stained with hematoxylin and
eosin or Picrosirius Red. Slides were examined by standard bright-
field microscopy (Olympus BX60, Rungis, France) by a pathologist
who was blinded to the experimental group assignment.
Skin collagen content
Skin pieces were dilacerated using a scalpel, put into tubes, thawed,
and mixed with pepsin (1:10 weight ratio) and 0.5 M acetic acid
overnight at room temperature under stirring. The assay of collagen
content was based on the quantitative dye-binding Sircol method
(Biocolor, Belfast, Ireland).
Immunohistochemistry of pEGFR proteins in mouse ear skin
Sections of ear pinna, 7 μm thick, were placed onto slides,
deparaffinized, and then rehydrated. Antigen retrieval, endogenous
peroxidase blocking, and nonspecific binding blocking were
performed as previously described (Marut et al., 2013). Sections
were then incubated with 1:50 rabbit polyclonal anti-pEGFR (Y1092)
(ab40815, Abcam, Paris, France) overnight at 4 °C. Incubation with
horseradish peroxidase–labeled secondary goat anti-rabbit and
staining with diaminobenzidine tetrahydrochloride were performed
as previously described (Marut et al., 2013). Pictures were taken with
a digital camera on a BX60 microscope (Olympus), using DP
Manager Software, at × 100 magnification. Immunostaining was
analyzed using ImageJ software (version 1.36; National Institutes of
Health, Bethesda, MD) for quantitative assessment of staining
positivity. Results are expressed as percentages of positive skin area
analyzed with ImageJ software.
Western blot of pEGFR proteins in mouse ears skin
Skin fragments of earlobes were incubated with 250 μl of RIPA buffer
for western blot. Protein extracts (30 μg total protein) were resolved as
previously described (Marut et al., 2013). Blots were incubated
overnight at 4 °C with a 1:500 rabbit polyclonal anti-pEGFR (Y1092)
(ab40815, Abcam). Revelation and analysis were performed as
previously described (Marut et al., 2013). Results are expressed as
AU, defined as the ratio between the densitometric intensity of the
bands of the proteins tested and bands labeled with mouse anti-β-
actin mAb.
Liver inflammation
At the time of killing, livers were removed from all mice in both
groups and kept in formol. Fixed liver pieces were embedded in
paraffin and stained with hematoxylin and eosin by standard
procedure. Slides were examined by standard bright-field
F Morin et al.
Erlotinib Prevents Scl-GVHD
microscopy (Olympus BX60) by a pathologist who was blinded to
the experimental group assignment. Degree of inflammation of
coded microscopic sections of liver was graded from 0 to 4, as
previously described (Nagler et al., 2000).
Serum AST and ALT activities
Serum activity of AST and ALT were used as markers of hepatocyte
cytolysis. AST and ALT activities were quantified using a standard
clinical automatic analyser (Modular PP, Roche Diagnostics, Meylan,
France).
Immunohistochemistry of skin tissue sections
Frozen skin tissue sections of 5 μm thickness were stained for 1 hour
at room temperature with 1:50 phycoerythrin-labeled hamster
anti-mouse CD3 (clone 145-2C11; BD Bioscience, San Jose, CA) or
with the respective phycoerythrin-labeled control isotype (BD
Bioscience). After extensive washing in phosphate-buffered saline,
slides were viewed with an Olympus microscope equipped with an
epifluorescence system to view eGFP- and phycoerythrin-labeled
cells. Images were collected using a digital camera with × 400
magnification objective. Hematopoietic donor cells appear in green
(B10.D2-eGFP) and in red for CD3+ T cells.
Determination of IFN-γ and IL-13 production in the skin
Punch biopsies (6 mm diameter) from the skin of each mouse were
incubated for 72 hours at 37 °C in complete RPMI medium with
5 μg ml − 1 concanavalin A. Supernatants were collected and frozen
at − 80 °C. Cytokine concentrations from 1:4 diluted supernatants
were measured by ELISA. “Mouse IFN gamma ELISA Ready-SET-Go!”
and “Mouse IL-13 ELISA Ready-SET- Go!” kits were purchased from
eBioscience (San Diego, CA). Concentrations were calculated from a
standard curve according to the manufacturer’s protocol. Results are
expressed in pg ml − 1.
Determination of IL-13, IFN-γ, CCL17, CCL21, and CCL27
mRNA levels in the skin
RNA isolation was realized as described by Grange et al. (2009).
Amplification was achieved with 50 ng of total RNA for each sample.
Real-time PCR was carried out with iTaq Universal SYBR Green
One-Step Kit (Bio-Rad, Hercules, CA). We used the 5′-GCTTGCC
TTGGTGGTCTCGCC-3′ and 5′-GGGCTACACAGAACCCGCCA-3′
primers for IL-13, the 5′-CAGCAACAGCAAGGCGAAAAAGG-3′ and
5′-TTTCCGCTTCCTGAGGCTGGAT-3′ primers for IFN-γ, the 5′-GG
TACCATGAGGTCACTTCAGA-3′ and 5′-CACTCTCGGCCTACATT
GGT-3′ primers for CCL17, the 5′-GGGAAACAAAGCCCCGG-3′
and 5′-GCTGTGTCTGTTCAGTTCTCTTG-3′ primers for CCL21, the
5′-CTGCTGAGGAGGATTGTC-3′ and 5′-CACGACAGCCTGGAG
GTGA-3′ primers for CCL27, and the 5′-TGCCGAGGATTTGG
AAAAAG-3′ and 5′-CCCCCCCTTGAGCACACAG-3′ primers for
HRPT that was used as a control.
Serum anti-DNA topoisomerase I autoantibodies
Levels of anti-DNA topoisomerase I IgG antibodies were assayed by
ELISA using purified calf thymus DNA topoisomerase I bound to the
wells of a microtiter plate (Abnova, Taipei City, Taiwan). Mouse
serum (1:4) and 1:1,000 anti-murine Ig horseradish peroxidase
(Dako, Glostrup, Denmark) secondary antibody were used as
previously described (Servettaz et al., 2009; Kavian et al., 2012).
www.jidonline.org 2391
 F Morin et al.
Erlotinib Prevents Scl-GVHD
Flow cytometric analysis of spleen cell subsets
Cell suspensions from spleens were prepared after hypotonic lysis of
erythrocytes in potassium acetate solution. Cells were incubated with
the appropriate labeled antibody at 4 °C for 30 minutes in phosphate-
buffered saline with 2% normal fetal calf serum. Flow cytometry was
performed using a FACS Fortessa flow cytometer (BD Biosciences),
according to standard techniques. The mAbs used in this study
were anti-B220-PECy7, anti-CD11b-biotin and Streptavidin-PerCP,
anti-CD11c-PE on one hand and anti-CD69-BV421, anti-CD44-PE,
anti-CD62L-APC, anti-CD4-PerCP, and anti-CD8-PE-Cy7 (BD
Biosciences) on the other hand. Data were analyzed with FlowJo
software (Tree Star, Ashland, OR).
Statistical analysis
All quantitative data are expressed as mean ± SEM. Data were
compared using one-way analysis of variance followed by
Dunn’s multiple comparison test. Student’s unpaired t-test was used
when comparing two groups. All analyses were carried out using the
GraphPad Prism 5.0 statistical software package (San Diego, CA),
and P-values of o0.05 were considered significant.
CONFLICT OF INTEREST
The authors state no conflict of interest.
ACKNOWLEDGMENTS
We thank A Colle for typing the manuscript and N Saidu for reviewing the
manuscript. This work was supported by grants from University Paris
Descartes.
REFERENCES
Akita RW, Sliwkowski MX (2003) Preclinical studies with Erlotinib (Tarceva).
Semin Oncol 30:15–24
Baird K, Pavletic SZ (2006) Chronic graft versus host disease. Curr Opin
Hematol 13:426–35
Bruggen MC, Klein I, Greinix H et al. (2014) Diverse T-cell responses
characterize the different manifestations of cutaneous graft-versus-host
disease. Blood 123:290–9
Chu Y-W, Gress RE (2008) Murine models of chronic graft-versus-host disease:
insights and unresolved issues. Biol Blood Marrow Transplant J Am Soc
Blood Marrow Transplant 14:365–78
Comerford I, Harata-Lee Y, Bunting MD et al. (2013) A myriad of functions
and complex regulation of the CCR7/CCL19/CCL21 chemokine axis
in the adaptive immune system. Cytokine Growth Factor Rev 24:
269–83
Eberhard Y, Ortiz S, Ruiz Lascano A et al. (2004) Up-regulation of the
chemokine CCL21 in the skin of subjects exposed to irritants. BMC
Immunol 5:7
Ferrara JLM, Levine JE, Reddy P et al. (2009) Graft-versus-host disease. Lancet
373:1550–61
Filipovich AH, Weisdorf D, Pavletic S et al. (2005) National Institutes of Health
consensus development project on criteria for clinical trials in chronic
graft-versus-host disease: I. Diagnosis and staging working group report.
Biol Blood Marrow Transplant J Am Soc Blood Marrow Transplant 11:
945–56
Fuchs BC, Hoshida Y, Fujii T et al. (2014) Epidermal growth factor receptor
inhibition attenuates liver fibrosis and development of hepatocellular
carcinoma. Hepatol Baltim Md 59:1577–90
Furudate S, Fujimura T, Kambayashi Y et al. (2014) Comparison of CD163+
CD206+ M2 macrophages in the lesional skin of bullous pemphigoid and
pemphigus vulgaris: the possible pathogenesis of bullous pemphigoid.
Dermatology 229:369–78
2392 Journal of Investigative Dermatology (2015), Volume 135
Grange PA, Raingeaud J, Calvez V et al. (2009) Nicotinamide inhibits
Propionibacterium acnes-induced IL-8 production in keratinocytes
through the NF-kappaB and MAPK pathways. J Dermatol Sci 56:
106–12
Hamilton JD, Suárez-Fariñas M, Dhingra N et al. (2014) Dupilumab improves
the molecular signature in skin of patients with moderate-to-severe atopic
dermatitis. J Allergy Clin Immunol 134:1293–300
Jaffee BD, Claman HN (1983) Chronic graft-versus-host disease (GVHD) as a
model for scleroderma. I. Description of model systems. Cell Immunol 77:
1–12
Johnson JR, Cohen M, Sridhara R et al. (2005) Approval summary for erlotinib
for treatment of patients with locally advanced or metastatic non-small
cell lung cancer after failure of at least one prior chemotherapy regimen.
Clin Cancer Res Off J Am Assoc Cancer Res 11:6414–21
Kansu E (2004) The pathophysiology of chronic graft-versus-host disease. Int J
Hematol 79:209–15
Kavian N, Servettaz A, Mongaret C et al. (2010) Targeting ADAM-17/notch
signaling abrogates the development of systemic sclerosis in a
murine model. Arthritis Rheum 62:3477–87
Kavian N, Marut W, Servettaz A et al. (2012) Arsenic trioxide prevents murine
sclerodermatous graft-versus-host disease. J Immunol 188:5142–9
Korholz D, Kunst D, Hempel L et al. (1997) Decreased interleukin 10 and
increased interferon-gamma production in patients with chronic graft-
versus-host disease after allogeneic bone marrow transplantation. Bone
Marrow Transplant 19:691–5
Korngold R, Sprent J (1978) Lethal graft-versus-host disease after bone
marrow transplantation across minor histocompatibility barriers in mice.
Prevention by removing mature T cells from marrow. J Exp Med 148:
1687–98
Kumai T, Oikawa K, Aoki N et al. (2014) Tumor-derived TGF-β and
prostaglandin E2 attenuate anti-tumor immune responses in head and
neck squamous cell carcinoma treated with EGFR inhibitor. J Transl Med
12:265
Lee SJ, Klein JP, Barrett AJ et al. (2002) Severity of chronic graft-versus-host
disease: association with treatment-related mortality and relapse. Blood
100:406–14
Linhares YPL, Pavletic S, Gale RP (2013) Chronic GVHD: Where are we?
Where do we want to be? Will immunomodulatory drugs help? Bone
Marrow Transplant 48:203–9
Luo Q, Gu Y, Zheng W et al. (2011) Erlotinib inhibits T-cell-mediated immune
response via down-regulation of the c-Raf/ERK cascade and Akt signaling
pathway. Toxicol Appl Pharmacol 251:130–6
Marut W, Kavian N, Servettaz A et al. (2013) Amelioration of systemic
fibrosis in mice by angiotensin II receptor blockade. Arthritis Rheum 65:
1367–77
Nagler A, Pines M, Abadi U et al. (2000) Oral tolerization ameliorates liver
disorders associated with chronic graft versus host disease in mice.
Hepatology 31:641–8
Nishimori H, Maeda Y, Tanimoto M (2013) Chronic graft-versus-host disease:
disease biology and novel therapeutic strategies. Acta Med Okayama 67:
1–8
Pastore S, Mascia F, Mariani V et al. (2008) The epidermal growth factor
receptor system in skin repair and inflammation. J Invest Dermatol 128:
1365–74
Robert J (2011) [Tyrosine kinase inhibitors]. Bull Cancer (Paris) 98:1321–34
Ruzek MC, Jha S, Ledbetter S et al. (2004) A modified model of graft-versus-
host-induced systemic sclerosis (scleroderma) exhibits all major aspects of
the human disease. Arthritis Rheum 50:1319–31
Saeki H, Tamaki K (2006) Thymus and activation regulated chemokine (TARC)/
CCL17 and skin diseases. J Dermatol Sci 43:75–84
Schneider MR, Wolf E (2009) The epidermal growth factor receptor ligands at
a glance. J Cell Physiol 218:460–6
Servettaz A, Goulvestre C, Kavian N et al. (2009) Selective oxidation of DNA
topoisomerase 1 induces systemic sclerosis in the mouse. J Immunol 182:
5855–64
Shlomchik WD (2007) Graft-versus-host disease. Nat Rev Immunol 7:340–52

Socié G, Ritz J (2014) Current issues in chronic graft-versus-host disease. Blood
124:374–84
Vogelsang GB, Lee L, Bensen-Kennedy DM (2003) Pathogenesis and treatment
of graft-versus-host disease after bone marrow transplant. Annu Rev Med
54:29–52
Wolff D, Gerbitz A, Ayuk F et al. (2010) Consensus conference on clinical
practice in chronic graft-versus-host disease (GVHD): first-line and
topical treatment of chronic GVHD. Biol Blood Marrow Transplant 16:
1611–28
Wolff D, Schleuning M, von Harsdorf S et al. (2011) Consensus
Conference on Clinical Practice in Chronic GVHD: second-line treatment
of chronic graft-versus-host disease. Biol Blood Marrow Transplant 17:
1–17
F Morin et al.
Erlotinib Prevents Scl-GVHD
Yahata Y, Shirakata Y, Tokumaru S et al. (2006) A novel function of angiotensin
II in skin wound healing. Induction of fibroblast and keratinocyte migration
by angiotensin II via heparin-binding epidermal growth factor (EGF)-like
growth factor-mediated EGF receptor transactivation. J Biol Chem 281:
13209–16
Yamamoto T (2010) Animal model of systemic sclerosis. J Dermatol 37:26–41
Yang F, Chung ACK, Huang XR et al. (2009) Angiotensin II induces connective
tissue growth factor and collagen I expression via transforming growth
factor-beta-dependent and -independent Smad pathways: the role
of Smad3. Hypertension 54:877–84
Zhou L, Askew D, Wu C et al. (2007) Cutaneous gene expression by DNA
microarray in murine sclerodermatous graft-versus-host disease, a model
for human scleroderma. J Invest Dermatol 127:281–92
www.jidonline.org 2393