Selective EGF-Receptor Inhibition in CD4+ T Cells Induces Anergy and Limits Atherosclerosis

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 71, NO. 2, 2018

ª 2018 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER

https://doi.org/10.1016/j.jacc.2017.10.084

Selective EGF-Receptor Inhibition in CD4þ T Cells Induces Anergy and Limits Atherosclerosis Lynda Zeboudj, MS,a,b Mikael Maître, MS,a,b Lea Guyonnet, PHD,a,b Ludivine Laurans, MS,a,b Jeremie Joffre, MD, PHD,a,b Jeremie Lemarie, MD, PHD,a,b Simon Bourcier, MD,a,b Wared Nour-Eldine, MS,a,b Coralie Guérin, PHD,c Jonas Friard, MS,d Abdelilah Wakkach, PHD,d Elizabeth Fabre, MD,e Alain Tedgui, PHD,a,b Ziad Mallat, MD, PHD,a,b,f Pierre-Louis Tharaux, MD, PHD,a,b Hafid Ait-Oufella, MD, PHDa,b,g

ABSTRACT BACKGROUND Several epidermal growth factor receptor (EGFR) inhibitors have been successfully developed for the treatment of cancer, limiting tumor growth and metastasis. EGFR is also expressed by leukocytes, but little is known about its role in the modulation of the immune response. OBJECTIVES The aim of this study was to determine whether EGFR expressed on CD4þ T cells is functional and to address the consequences of EGFR inhibition in atherosclerosis, a T cell–mediated vascular chronic inflammatory disease. METHODS The authors used EGFR tyrosine kinase inhibitors (AG-1478, erlotinib) and chimeric Ldlr-/-Cd4-Cre/Egfrlox/lox mouse with a specific deletion of EGFR in CD4þ T cells. RESULTS Mouse CD4þ T cells expressed EGFR, and the EGFR tyrosine kinase inhibitor AG-1478 blocked in vitro T cell proliferation and Th1/Th2 cytokine production. In vivo, treatment of Ldlr–/– mice with the EGFR inhibitor erlotinib induced T cell anergy, reduced T cell infiltration within atherosclerotic lesions, and protected against atherosclerosis development and progression. Selective deletion of EGFR in CD4þ T cells resulted in decreased T cell proliferation and activation both in vitro and in vivo, as well as reduced interferon-g, interleukin-4, and interleukin-2 production. Atherosclerotic lesion size was reduced by 2-fold in irradiated Ldlr–/– mice reconstituted with bone marrow from Cd4-Cre/Egfrlox/lox mice, compared to Cd4-Cre/Egfrþ/þ chimeric mice, after 4, 6, and 12 weeks of high-fat diet, associated with marked reduction in T cell infiltration in atherosclerotic plaques. Human blood T cells expressed EGFR and EGFR inhibition reduced T cell proliferation both in vitro and in vivo. CONCLUSIONS EGFR blockade induced T cell anergy in vitro and in vivo and reduced atherosclerosis development. Targeting EGFR may be a novel strategy to combat atherosclerosis. (J Am Coll Cardiol 2018;71:160–72) © 2018 by the American College of Cardiology Foundation.

E

pidermal growth factor receptor (EGFR) is a

autocrine ligand stimulation, or constitutively active

cell membrane–bound receptor with tyrosine

receptor mutants (1,2) can lead to dysregulation of

kinase activity involved in the control of

this fine-tuned signaling system, resulting in a vari-

major signaling pathways, including cell survival,

ety of pathophysiological disorders and promoting

proliferation, and migration. EGFR overexpression,

cancer development. Six EGFR ligands have been

From aInserm U970, Paris Cardiovascular Research Center, Paris, France; bUniversité René Descartes, Paris, France; cLuxembourg Listen to this manuscript’s

Institute of Health, Department of Infection and Immunity, Strassen, Luxembourg; dCNRS, LP2M, UMR 7370, Faculté de Méde-

audio summary by

cine, Université de Nice Sophia Antipolis, Nice, France; eDepartment of Medical Oncology, Hôpital Europeen G. Pompidou, AP-HP,

JACC Editor-in-Chief

Paris, France; fDivision of Cardiovascular Medicine, Department of Medicine, University of Cambridge, Cambridge, United

Dr. Valentin Fuster.

Kingdom; and the gService de Réanimation Médicale, Hôpital Saint-Antoine, AP-HP, Université Pierre-et-Marie Curie, Paris, France. This work was supported by Institut National de Santé et de la Recherche Médicale, research grants ANR-08-EBIO-003 (to Dr. Tharaux) from l’Agence Nationale de la Recherche, and the British Heart Foundation (to Dr. Mallat). The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Zeboudj and Maître contributed equally to this work. Drs. Ait-Oufella and Tharaux contributed equally to this work. Manuscript received April 26, 2017; revised manuscript received October 15, 2017, accepted October 30, 2017.

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

161

EGFR Blockade in T Cells Reduces Atherosclerosis

described, including epidermal growth factor (EGF),

body irradiation (9.5 Gy). The mice were

ABBREVIATIONS

Heparin Binding-EGF, amphiregulin, and transform-

repopulated with an intravenous injection of

AND ACRONYMS

ing growth factor– a . Extracellular ligand binding

bone marrow cells isolated from femurs and

causes

becomes

tibias of sex-matched C57BL/6 Cd4Cre Egfr þ/þ

autophosphorylated at distinct tyrosine residues. In

mice or Cd4Cre Egfr lox/lox littermates. After

addition, EGFR could be transactivated in the

4 weeks of recovery, mice were fed a proa-

absence of a specific ligand through G protein–

therogenic diet containing 15% fat, 1.25%

coupled receptor activation (3).

cholesterol, and 0% cholate for 4, 6, and

dimerization

of

EGFR,

which

EGFR has been extensively explored in cancer.

EGFR = epidermal growth factor receptor

IFN = interferon IL = interleukin IQR = interquartile range

12 weeks. SEE PAGE 173

EGF = epidermal growth factor

EXTENT AND COMPOSITION OF ATHEROSCLEROTIC

LESIONS. Plasma

cholesterol

Human and experimental studies have shown that

was measured using a commercial cholesterol

EGFR activation on tumor cells ultimately leads to

kit (Biomérieux, Marcy-l’Étoile, France). The

LDL = low-density lipoprotein Th = T helper TKI = tyrosine kinase inhibitor Treg = regulatory T

cell proliferation, invasion, and migration, as well as

heart was removed, and successive 10- m m transversal

promoting angiogenesis and inhibiting apoptosis (4).

sections of aortic sinus were obtained. Lipids were

Targeting of EGFR by either neutralizing monoclonal

detected using Red Oil staining. The presence of

antibodies or small-molecule tyrosine kinase in-

T cells was studied using specific antibodies as pre-

hibitors (TKIs) have been shown during the past 10

viously described (polyclonal anti-CD3, Agilent, Santa

years to be a successful therapeutic strategy in cancer

Clara, California) (10). Egfr was detected in cells and

setting (5,6). However, EGFR expression and function

lesions using rabbit polyclonal anti-phospho-Egfr

have been poorly investigated in nontumoral cells.

(Cell Signaling, Boston, Massachusetts). For human

Some investigators have described expression in

staining, an anti-EGFR antibody (clone 31G7, AbCys,

circulating leukocytes (7,8), but little is known about

Paris, France) was used. At least 4 sections per mouse

EGFR’s role in modulation of the immune response.

were

examined

for

each

immunostaining,

and

Atherosclerosis is an inflammatory disease driven

appropriate negative controls were used. Morpho-

by innate and adaptive immunity, in which CD4þ T

metric studies were performed using HistoLab soft-

cells play a pathogenic role. Interestingly, EGFR

ware (Microvisions, Evry, France) (10).

ligands, including heparin-binding EGF, have been

SPLEEN

detected in human atherosclerotic plaques (9). The

Spleen cells were purified according to standard

aim of this study was to ascertain the expression of

protocols as follows. CD4 þ T cells were negatively

EGFR in human and mouse CD4 þ T cells and to

selected using a cocktail of antibody-coated magnetic

investigate the effects of EGFR blockade on CD4 þ T

beads from Miltenyi Biotec (Bergisch Gladbach, Ger-

cell functions using pharmacological inhibitors and

many) (anti-CD8a, anti-CD11b, anti-CD45R, anti-DX5,

cell-specific genetic deletion in mouse models of

and anti-ter 119), according to the manufacturer’s

atherosclerosis.

METHODS ANIMALS. Experiments were conducted according to

the guidelines formulated by the European Community for experimental animal use (L358-86/609EEC) and were approved by the ethics committee of

CELL

RECOVERY

AND

PURIFICATION.

instructions, yielding CD4 þ cells with >95% purity. CD11c þ cells were positively selected with biotinconjugated anti-CD11c monoclonal antibody (7D4, BD Pharmingen, Franklin Lakes, New Jersey), streptavidin microbeads (Miltenyi Biotec), followed by 2 consecutive magnetic cell separations using LS columns (Miltenyi Biotec), yielding CD11c þ cells with

INSERM and the French Ministry of Agriculture

>80% purity.

(agreement A75-15-32). To generate a cell-specific

CD4 D T CELL CULTURE AND CYTOKINE ASSAYS.

knockout of Egfr in CD4 þ T cells, we crossbred mice

Cells were cultured in RPMI-1640 supplemented with

carrying a Cd4Cre allele with mice carrying a floxed

GlutaMAX (Thermo Fisher Scientific, Waltham, Mas-

Egfr allele. All animals have been backcrossed more

sachusetts), 10% fetal calf serum, 0.02 mmol/l

than 10 generations on C57bl/6 background. Ten-

b-mercaptoethanol, and antibiotics. For cytokine

week-old male C57BL/6 Ldlr / mice were put on a

measurements, CD4 þ T cells were cultured at 1  105

high-fat diet for 8 weeks and were treated orally

cells/well for 48 h in anti-CD3-coated microplates

(daily gavage) with the specific EGFR TKI erlotinib

(10 m g/ml) or with concanavalin A (10 m g/ml;

(15 mg/kg/day). For bone marrow transplantation

Sigma-Aldrich, St. Louis, Missouri). In some experi-

experiments, 10-week-old male C57bl/6 Ldlr / mice

ments, CD4 þ T cells were stimulated with purified

were subjected to medullar aplasia by lethal total

soluble

CD3-specific

antibody

(1

mg/ml;

BD

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 1 EGFR Receptor Expression and Activation in Mouse T Cells

A

D Phospho-ERK (MFI)

700

*

650 600

B

550 500 400

C

0 Anti-CD3

Baseline CTR

E

AG 1478

Intracellular Calcium Increase 2.0 1.8

(ΔRatio 340/380nm)

162

**

*

***

1.6 ** 1.4

***

*** *** ***

***

1.2

***

1.0 0.8 60 120 180 240 300 360 420 480 540 600

0

Time (Seconds) Untreated

1μM AG1478

10μM AG1478

(A) Epidermal growth factor receptor (EGFR) is expressed by splenocytes (green) and colocalizes with CD4þ T cells (red). (B) EGFR is expressed within mouse atherosclerotic lesions (red) and colocalizes with CD4þ T cells (green). Staining was performed on atherosclerotic plaques (aortic sinus) from Ldlr/ mice fed a high-fat diet for 6 weeks. (C) Stimulation of purified CD4þ T cells by coated anti-CD3 for 60 min induced focal EGFR phosphorylation that was abolished by AG 1478 (1 mmol/l). (D) Flow cytometric quantification of Erk phosphorylation gated on CD3þCD4þ T cells after 30 min of anti-CD3 (5 mg/ml) or concanavalin A (10 mg/ml) stimulation (n ¼ 4/group). (E) Naive CD4þ T cells were stimulated with a-CD3/CD28 antibodies for 72 h in the presence of 1 or 10 mmol/l AG-1478 before measuring intracellular free calcium concentration (Fura-2-AM fluorescent probe). Each trace is the fluorescence mean (340/380 nm) of 3 independent experiments (10
Pharmingen) in the presence of antigen-presenting

between values were evaluated using the nonpara-

cells purified on CD11c-coated magnetic beads (Milte-

metric Mann-Whitney U test or the Kruskal-Wallis

nyi Biotec). Interleukin (IL)–2, IL-4, IL-10, and inter-

test. A p value of <0.05 was considered to indicate

feron (IFN)–g productions in the supernatants were

statistical significance. All of these analyses were

measured using specific enzyme-linked immunosor-

performed using GraphPad Prism version 5.0b for

bent assays (R&D Systems, Minneapolis, Minnesota).

Mac (GraphPad Software, La Jolla, California). No

Other methods (cell culture, proliferation assays, cytosolic calcium recording, flow cytometry, and western blot) are available in the Online Appendix.

adjustments

were

made

for

multiple

pairwise

comparisons. On the basis of preliminary experiments, we

STATISTICAL ANALYSIS. Values are expressed as

assumed that EGFR blockade induces a 40% reduc-

median

tion of atherosclerosis. With a standard deviation of

(interquartile

range

[IQR]).

Differences

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 2 Effects of EGFR Pharmacological Inhibition on Mouse T Cell Functions

A

B Ag Specific Stimulation (OVA)

Stimulated with sCD3 & APC

CD4 T cell proliferation (X 103 cpm) 30 25 20 15

*

10 #

5

$

0 0

0.5

0.1

1.0

CD4 T cell proliferation (X 103 cpm) 8 6 #

4 2

$ 0 0

C

0 05

7-AAD

0

0.1μM

Q1 2.26

Q2 41.4

Q1 3.61

Q3 33.3

Q4 25.6

0.1

0.5

1.0

AG 1478 Concentration (μM)

AG 1478 Concentration (μM)

0.5μM Q2 39.8

Q1 5.41

Q3 31.0

Q4 24.7

1μM Q2 38.1

Q1 4.87

Q3 31.8

Q4 20.2

Q2 38.2

4

03 0 03

Q4 23.1 -10

3

0

10

3

10

4

10

5

-10

3

10

0

3

10

4

10

5

3

-10

0

10

3

4

10

10

5

-10

3

Q3 36.7 0

10

3

10

4

105

Annexin V

D

E

F

Concavalin A Stimulated

0.3 #

0.2

$ 0.1 0.0 0

0.1

0.5

1.00 0.75 0.50 0.25

$

0.00 0

1.0

0.5

1

6 #

4 2

$ 0 0

0.1

0.5

1.0

AG 1478 Concentration (μM)

Stimulated with Coated CD3

Stimulated with Concavalin

H Ifn-γ (X103 pg/ml)

Ifn-γ (X103 pg/ml) 4 3 2 *

1

# 0 0

AG 1478 Concentration (μM)

AG 1478 Concentration (μM)

G

0.1

0.1

0.5

$ 1.0

AG 1478 Concentration (μM)

I IL-4 (pg/ml) 300 250 *

200 150

#

100 50

$

0 0

0.1

0.5

1.0

AG 1478 Concentration (μM)

Stimulated with Coated CD3

Coated CD3 Stimulated

% Ifn-γ+ CD4 T cells

Stimulated with Coated CD3

% Ifn-γ+ CD4+ T cells

+

IL-10 (pg/ml)

400 300 200 100 0 0

0.1

0.5

1.0

AG 1478 Concentration (μM)

(A) In vitro AG 1478 induced a dose-dependent reduction of CD4þ T cell proliferation in response to soluble anti-CD3 stimulation, in the presence of CD11cþ dendritic cells. (B) In vitro AG 1478 induced a dose-dependent reduction of OT-II CD4þ T cell proliferation after ovalbumin (OVA) stimulation in the presence of CD11cþ dendritic cells. (C) Effects of different concentrations of AG 1478 on T cell apoptosis, defined as annexin Vpos 7-aminoactinomycin D (7-AAD)neg cells. Purified CD4þ T cells were stimulated by coated anti-CD3 (5 mg/ml) or concanavalin A (10 mg/ml), and cytokine production was evaluated by flow cytometry (D, E) or by enzyme-linked immunosorbent assay in the supernatants (F–I). Data are expressed as median (5th percentile to 95th percentile). APC ¼ antigen-presenting cells; Ifn ¼ interferon; other abbreviations as in Figure 1.

163

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 3 EGFR Pharmacological Inhibition Limited T Cell Activation In Vivo and Reduced Atherosclerosis

A

B

Cholesterolemia (g/l) 25 20

Control or Erlotinib High-fat diet 8 weeks

8-week old Ldlr–/– mice

15 10 5 0 Control

D

Splenocyte number (106)

100

CD4 T cell proliferation (X103 cpm)

30

*

150

*

80 20

60 40

10

20 0

0 Control

F

Erlotinib

Erlotinib

E ELISA (pg/ml)

C

Control Erlotinib

*

100 50

*

0 Control

IFN-γ

Erlotinib

Adv

Adv

2.5

IL-4

IL-10

T cell infiltration (%) **

2.0 1.5

CD3

1.0 0.5 0.0 Control

G

500

Erlotinib

Plaque size (X103 μm2) *

400 300

Red Oil

200 100

Control

Erlotinib

0 Control

Erlotinib

H High-fat diet 8 weeks 6-week old Ldlr–/– mice

I

Baseline

164

30

Control or Erlotinib High-fat diet 8 weeks

Plaque size (% thoracic area) NS ** *

20

10

0 Baseline Control Erlotinib

Baseline

Control

Erlotinib

Continued on the next page

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

plaque size estimated at 30%, the inclusion of 8 mice

CD4 þ T cells, showing that EGFR inhibition signifi-

per group was sufficient to detect a significant dif-

cantly reduced the production of IFN-g (Figures 2F

ference between groups with 80% power.

and 2G) and reduced the production of IL-2 (Online

RESULTS

dose-dependent manner but had no effect on IL-10

Figure 3), as well as that of IL-4 (Figure 2H), in a

EGFR IN MOUSE CD4 D T CELLS: EXPRESSION AND

production, which was very low (Figure 2I). To investigate the in vivo consequences of EGFR

SIGNALING PATHWAYS. Using immunocytostaining,

pharmacological inhibition, Ldlr/ male mice were

we detected EGFR expression by splenocytes and

put on a high-fat diet for 8 weeks and were orally

found that EGFR colocalized, nonexclusively, with

treated with erlotinib (15 mg/kg/day) (Figure 3A). At

CD4þ T cells (Figure 1A). In addition, EGFR was present

in

mouse

atherosclerotic

lesions

and

sacrifice, animal weight (Online Figure 4A) and plasma cholesterol level (Figure 3B) were not different

colocalized with CD4þ T cells (Figure 1B). In vitro,

between

anti-CD3-induced

reductions in spleen size (Online Figure 4B) and

and

concanavalin

A–induced

activation of purified CD4 þ T cells caused EGFR phosphorylation

after

60

min

of

groups,

but

there

were

significant

splenocyte number (Figure 3C) in the erlotinib-treated

stimulation

group. Splenic CD4þ T cells from erlotinib-treated

(Figure 1C). AG-1478, a pharmacological inhibitor of

group were characterized by significant decreases in

tyrosine kinase, blocked EGFR phosphorylation in

CD25 (Online Figure 5A) and CD44 high expression

stimulated T cells (Figure 1C) and significantly reduced

(Online Figure 5B), suggesting reduced in vivo acti-

ERK1/2 phosphorylation (Figure 1D, Online Figure 1)

vation.

but had no effect on AKT (Online Figure 1). In addition,

CD4 þCD25 high Foxp3 þ regulatory T cell population

we found that EGFR inhibition blocked cytoplasmic

(Online Figure 5C). Splenic purified CD4 þ T cells from

calcium increase following anti-CD3/CD28 stimulation

the erlotinib-treated group were characterized by a

(Figure 1E). In summary, anti-CD3 stimulation of CD4 þ

Erlotinib

had

no

effect

on

the

reduction of proliferation ex vivo (Figure 3D) and a

T cells led to EGFR transactivation that was efficiently

reduction of T helper (Th) 1 (IFN-g ) and Th2 (IL-4)

inhibited by AG-1478.

cytokine

EFFECTS OF PHARMACOLOGICAL INHIBITION OF

(Figure 3E). Immunohistochemistry analysis revealed

EGFR ON ATHEROSCLEROSIS. To address the role of

that erlotinib treatment induced a 70% reduction of T

EGFR on T cell functions, splenic CD4 þ T cells were

cell

purified and stimulated in vitro. EGFR activity was

(p ¼ 0.01) (Figure 3F), which was associated with a

inhibited using AG-1478 at different concentrations.

37% reduction in atherosclerotic lesion size in the

EGFR inhibition significantly reduced T cell prolifera-

aortic sinus (median 167 [IQR: 91 to 190]  103 in

tion following non-antigen-specific (Figure 2A) and

treated mice vs. 267 [IQR: 180 to 333]  103 mm 2 in

antigen-specific (Figure 2B) stimulation in a dose-

control mice, p < 0.05) (Figure 3G).

production,

infiltration

with

within

no

effect

on

atherosclerotic

IL-10

lesions

dependent manner, without any effect on T cell

In the vast majority of cases, patients in need of

apoptosis (Figure 2C) or nucleus-cytoplasm organiza-

antiatherosclerotic therapy already have established

tion (Online Figure 2). Pharmacological inhibition of

atherosclerotic plaques. Thus, we examined the ef-

EGFR significantly reduced intracellular IFN-g pro-

fects of EGFR blockade on the progression of estab-

duction by CD4 þ T cells in response to anti-CD3

lished atherosclerotic plaques in mice. Six-week-old

(Figure 2D) and concanavalin stimulation (Figure 2E).

Ldlr/ female mice were put on a high-fat diet for

This was confirmed by enzyme-linked immunosorbent

8 weeks and then were orally treated with a placebo

assay

or erlotinib (15 mg/kg/day) for 8 weeks (Figure 3H).

in

the

supernatants

of

cultured

splenic

F I G U R E 3 Continued

(A) Protocol of pharmacological blockade of EGFR in male Ldlr/ mice (erlotinib 15 mg/kg/day, orally; n ¼ 8 or 9/group). Cholesterolemia (B) and splenocyte number (C) at sacrifice. Proliferation (D) and cytokine production (enzyme-linked immunosorbent assay [ELISA]) (E) of purified splenic CD4þ T cells from control of erlotinib-treated animals after 48 h of coated anti-CD3 stimulation. (F) Quantification and representative photomicrographs of T cell infiltration in atherosclerotic lesions from Ldlr/ animals treated with phosphate-buffered saline (PBS) or erlotinib (n ¼ 8 or 9/group). (G) Quantification and representative photomicrographs of atherosclerotic lesion size in the aortic sinus from Ldlr/ animals treated with PBS or erlotinib (n ¼ 8 or 9/group). (H) Protocol of pharmacological blockade of EGFR in Ldlr/ female mice after an 8-week period of high-fat diet (erlotinib 15 mg/kg/day, orally; n ¼ 7/group). (I) Representative photomicrographs and quantification of atherosclerotic lesion size along the thoracic aorta (%) from Ldlr/ animals orally treated with PBS or erlotinib. *p < 0.05 and **p < 0.01. Data are expressed as median (5th percentile to 95th percentile). Adv ¼ adventitia; other abbreviations as in Figures 1 and 2.

165

166

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

At sacrifice, animal weight (data not shown) and

reduced. There was no difference in IL-10 production

plasma cholesterol levels (Online Figure 6C) were not

(Figure 5D). Similar reduction in cell proliferation and

different between groups. In the erlotinib-treated

cytokine

group, we observed a marked reduction in athero-

EGFR-deficient CD4 þ T cells were coincubated with

sclerotic lesion size both in the aortic sinus (me-

CD11c þ antigen–presenting cells (Figures 5A to 5D).

3

production

was

observed

when

dian 176 [IQR: 159 to 218]  10 vs. 269 [IQR: 193 to

Intracellular staining by flow cytometry confirmed

343]  103 m m 2, p < 0.05) (Online Figures 6A and 6B)

the specific reduction of IFN-g production by CD4þ

and along the thoracic aorta (median 14.0% [IQR:

(Figure 5E) but not by CD8þ T cells. Apoptosis sus-

12.4% to 14.9%] vs. 20.7% [IQR: 15.2% to 22.3%],

ceptibility was significantly lower in EGFR-deficient

p < 0.05) (Figure 3I). More important, atherosclerosis

CD4 þ T cells (Online Figure 8). To address the

plaque size increased between baseline 14-week-old

in vivo role of EGFR in CD4þ T cell proliferation,

mice and 22-week-old mice in the placebo group but

we transferred into Apoe //Rag2 / mice 20.106

did not progress in the group receiving erlotinib

CD4 þ T cell–depleted splenocytes resupplemented

(Figure 3I, Online Figure 6), suggesting that erlotinib

with 8.106 purified carboxyfluorescein succinimidyl

treatment blocked atherosclerosis progression.

ester–labeled Cd4-Cre/Egfr

þ/þ

CD4þ or

T

cells

from

Cd4-Cre/Egfrlox/lox

either

mice.

The

EGFR IN HUMAN CD4 D T CELLS: EXPRESSION AND

proliferation of adoptively transferred cells was

FUNCTIONS. To evaluate the clinical relevance of our

visualized by flow cytometric analysis of carboxy-

findings, we investigated the expression of EGFR in

fluorescein succinimidyl ester–labeled CD4 þ T cells.

human blood T cells. We performed immunocytos-

At day 10 after transfer, we found that CD4 þ T specific

þ

taining of purified blood CD4 T cells and found that

deletion of EGFR limited T cell proliferation in the

EGFR was expressed and clustered in the membrane

spleen and lymph nodes (Figure 5F).

of CD4 þ T cells after concanavalin-induced activation (Figure 4A). To investigate the functions of human EGFR, we purified circulating CD4 þ T cells from

IMPACT OF CD4 D T CELL–SPECIFIC INVALIDATION OF EGFR ON ATHEROSCLEROSIS. To address the

consequences of these findings in the context of

healthy donors and performed in vitro proliferation

atherosclerosis, we performed bone marrow trans-

tests. CD3/CD28-coated beads stimulation induced T

plantation experiments using either Cd4-Cre/Egfr þ/þ

cell proliferation and EGFR inhibition using AG-1478

or Cd4-Cre/Egfr lox/lox littermate bone marrow to

or cetuximab, an EGFR-neutralizing monoclonal

repopulate lethally irradiated Ldlr / mice. After

antibody, significantly decreased T cell proliferation

4 weeks of recovery and additional 4 weeks on a

(Figures 4B and 4C). This result was confirmed in vivo

high-fat diet, animals were sacrificed. We did not

in patients with lung cancer. Circulating T cells were

observe any difference in animal or spleen weights,

isolated from 3 patients before and 1 month after oral

but a 32% reduction was seen in the number of sple-

erlotinib treatment. Interestingly, T cell proliferation

nocytes in the chimeric Cd4-Cre/Egfrlox/lox group

was lower after erlotinib treatment (Figure 4D).

(p ¼ 0.05). Leukocyte populations (neutrophils,

CELL-SPECIFIC GENETIC INVALIDATION OF EGFR IN

monocytes, B cells, CD8þ T cells) were not different

CD4

D

T CELLS. Erlotinib administration induced

between groups, either in the blood or in the spleen

T cell anergy and reduced atherosclerosis develop-

(data not shown). Splenic CD4 þ T cell subset was not

ment. As the expression of EGFR is ubiquitous, we

different between chimeric Cd4-Cre/Egfrþ/þ/Ldlr /

next assessed the specific role of EGFR activation in

and Cd4-Cre/Egfrlox/lox/Ldlr/ mice (Figure 6A), but

T cells in erlotinib-induced atheroprotection.

the CD4 þCD25 highFoxP3 þ regulatory T (Treg) cell

We bred mice carrying a Cd4-Cre allele with mice

population was slightly reduced in the group with

CD4

þ

carrying

a

floxed

Egfr

allele

and

generated

Cd4-Cre/Egfr lox/lox mice. The deletion of EGFR specifically in CD4

þ

T cells was confirmed by immuno-

cytostaining (Online Figure 7). We purified splenic CD4

þ

T cells from control Cd4-Cre/Egfr

þ/þ

CD4-specific deletion of EGFR (14%, p < 0.05) (Figure

6B).

Cd4-Cre/Egfr

Splenic

lox/lox

/Ldlr

/

CD4 þ

T

cells

from

were characterized by a

significant decrease in CD69 (Figure 6C) and CD44 high

and

expression (Figure 6D), suggesting reduced in vivo

Cd4-Cre/Egfr lox/lox mice and performed functional

activation. Functional tests were performed ex vivo

tests. In vitro, in agreement with experiments using

in purified splenic CD4 þ T cells. The suppressive

AG-1478, proliferation of CD4

þ

T cells from Cd4Cre

function

of

Treg

cells

was

preserved

in

Egfr lox/lox mice was significantly decreased compared

Cd4-Cre/Egfrlox/lox/Ldlr/ mice (Figure 6E), but the

with wild-type cells (Figure 5A), and their production

proliferation of CD4 þ T cells in response to CD3

of Th1 (Figure 5B) and Th2 (Figure 5C) cytokines was

stimulation was reduced compared with cells from

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 4 EGFR Expression and Functions in Human T Cells

A

B

C In vitro T cell proliferation

50000

D 8000

*

* *

*

6000 Cpm

30000 20000

4000 2000

10000

6000 Cpm

40000 Cpm

Ex vivo T Cell Proliferation

In Vitro T Cell Proliferation

8000

4000 2000

0

0 No Bead

1 Bead/ 2 Cells Control

0 No Bead

1 Bead/ 1 Cell AG-1478

1 Bead/2 Cells 1 Bead/1 Cell

Control

T0

α EGF-R

Patient 3

T1

Patient 2

Patient 1

(A) Blood human CD4þ T cells were purified and stimulated with concanavalin (10 mg/ml) for 24 h. Epidermal growth factor receptor (EGFR) is expressed in CD4þ T cells after activation (fluorescent staining, green). EGFR inhibition using tyrosine kinase inhibitor (TKI) (B) or neutralizing antibody (C) decreased T cell proliferation after 48 h of culture at different concentrations of CD3/CD28-coated beads. Proliferation was measured by incorporation of methyl-[3H] thymidine (n ¼ 4). (D) Pharmacological inhibition of EGFR using erlotinib, an EGFR tyrosine kinase inhibitor, blocked in vivo T cell proliferation. T cells were isolated from 3 patients with lung cancer just before and 1 month after initiation of erlotinib treatment. Cells were stimulated with a ratio of 2 beads to 1 cell, and proliferation was measured by methyl-[3H]thymidine incorporation. *p < 0.05. Bar scale, 10 mm. Data are expressed as median (5th percentile to 95th percentile).

the control group (Figure 6F). Finally, we also observed

We finally analyzed T cell infiltration within the

a >2-fold decrease in IFN-g and IL-4 production in the

lesions. We found a decrease in T cell number in lesions

supernatant of anti-CD3-stimulated CD4þ T cells from

of Cd4Cre Egfrlox/lox / Ldlr / compared with Cd4Cre

Cd4-Cre/Egfr lox/lox/Ldlr /

Egfr þ/þ / Ldlr / mice (Figure 6B) but no difference in

mice

compared

with

Cd4-Cre/Egfr þ/þ/Ldlr/ mice (Figure 6G).

macrophage infiltration (Online Figure 10).

As shown in Figure 7A, after 4 weeks on a high-fat diet, Egfr invalidation in CD4 þ T cells led to a 36%

DISCUSSION

decrease in atherosclerotic lesion size in the aortic sinus compared with controls (median 25 [IQR: 16 to

Using several complementary approaches, we iden-

30]  10 3 vs. 34 [IQR: 31 to 57]  103 m m 2 in Cd4-Cre/

tified a critical role of EGFR in CD4 þ T cell homeo-

lox/lox

/

þ/þ

/

,

stasis, in both mice and humans. EGFR genetic

respectively, p ¼ 0.03). The reduction of atheroscle-

invalidation or pharmacological blockade impaired T

rosis was confirmed after 6 weeks (39%, p ¼ 0.02)

cell activation, proliferation, cytokine production and

and 12 weeks (43%, p ¼ 0.02) of a high-fat diet

reduced

(Figure 7A). There were no significant differences in

Illustration).

Egfr

/Ldlr

and

Cd4-Cre

Egfr

/Ldlr

plasma cholesterol levels between the groups (Online Figure 9).

Using

atherosclerosis

development

immunocytostaining

and

(Central

immunohisto-

chemistry, we found that splenic CD4 þ T cells and

167

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 5 Selective EGFR Deletion in CD4 þ T Cells Induced Anergy

B

C

**

30 20 ** 10 0

Cd4Cre Egfr+/+

10

2

**

0 No APCs

4.38

0 No APCs

Cd4Cre Egfr+/+

2.52

2

6.97

Cd4Cre Egfr

Cd4Cre Egfr+/+

+ CFSE-labeled CD4Cre Egfr+/+ CD4+ T cells -/-

IV transfer of CD4-depleted splenocytes + CFSE-labeled CD4Cre Egfr

lox/lox

CD4+ T cells

Cd4Cre Egfr

0

100

0

Cd4Cre Egfrlox/lox 100

80

80

60

60

Fat diet during 10 days

40

40

And

20

20

Analysis of CD4+ T cell proliferation by Flow cytometry

0

0 0

100

3

10

10

4

100

100

80

80

60

60

40

40

20

20

0

3

10

61.1

10

10

3

10

4

8000 6000 4000 2000 CD4

4

*

0

10

3

10

4

CD8

Lymph Nodes

38.9

0 0

*

0 0

0

42.1

CD8+ T cells

Spleen 10000

57.9

+/+

Lox/lox

*

5

CD4+ T cells

100

IV transfer of CD4-depleted splenocytes

APCs

10

CD8

F

Males apoe/Rag2

0

% Ifn-γ + among T cells

0

CD4

50

No APCs

2

0

+/+

Lox/lox

100

APCs

3

10

Cd4Cre Egfr Cd4Cre Egfr

150

Cd4Cre Egfrlox/lox

8.88

4

3

Ifn-γ

50

5

4

10

100

APCs

Cd4Cre Egfrlox/lox

5

**

Geometric MFI

10

4

IL-10 (pg/ml) 200

Geometric MFI

E

**

APCs

No APCs

IL-4 (pg/ml) 150

Stimulated with Coated CD3

Stimulated with Coated CD3

40

D

3

Ifn-γ (X10 pg/ml) 6

Stimulated with PMA/Iono

3

CD4 T Cell Proliferation (X10 cpm)

Stimulated with Coated CD3

A Stimulated with Coated CD3

168

12000

Cd4Cre Egfr Cd4Cre Egfr

8000

+/+

Lox/lox

600 400 200 0 CD4

CFSE

CD8

(A) In vitro, purified CD4þ T cells from control Cd4-Cre/Egfrþ/þ (blue) or Cd4-Cre/Egfrlox/lox mice (orange) were stimulated by soluble anti-CD3 with or without coincubation with CD11cþ dendritic cells (antigen-presenting cells [APCs]). Genetic invalidation of Egfr reduced T cell proliferation (A), IFN-g (B), and IL-4 (C) production but had no effect on IL-10 (D) (ELISA in the supernatant). (E) Representative examples and quantitative analysis of intracellular IFN-g staining of isolated splenocytes from control Cd4-Cre/Egfrþ/þ (blue) or Cd4-Cre/Egfrlox/lox mice. Plots are gated on CD4þ and CD8þ T cells. Numbers in each quadrant indicate percentages of cells. (F) Representative examples of CD4þ T cell in vivo proliferation. Cd4-Cre/Egfrþ/þ and Cd4-Cre/ Egfrlox/lox purified CD4þ T cells from pooled spleens and lymph nodes were labeled with carboxyfluorescein succinimidyl ester (CFSE) fluorescent dye. Apoe//Rag2/ mice received 20.106 CD4þ T cell–depleted splenocytes resupplemented with 8.106 purified CD4þ T cells from either Cd4-Cre/Egfrþ/þ or Cd4-CreEgfrlox/lox animals. The proliferation of adoptively transferred cells was visualized by flow cytometric analysis of CFSE-labeled CD4þ T cells at day 10 after transfer. N ¼ 5/group; *p < 0.05 and **p < 0.01. Data are expressed as median (5th percentile to 95th percentile). Iono ¼ ionomycin; IV ¼ intravenous; MFI ¼ mean fluorescence intensity; PMA ¼ phorbol 12–myristate 13–acetate; other abbreviations as in Figures 1 to 3.

blood human T cells express EGFR, especially in

Our findings indicate that EGFR signaling is crucial

response to CD3 or concanavalin A stimulation.

in CD4þ T cell homeostasis in both humans and mice.

Immunofluorescent staining showed, after anti-CD3

Pharmacological inhibition using TKIs (AG-1478 or

stimulation, that EGFR clustered on the cell mem-

erlotinib), neutralizing antibodies, or CD4 þ T cell–

brane and was phosphorylated. In addition, tissue T

specific

cells within atherosclerotic plaques express EGFR.

reduced in vitro and in vivo cell proliferation and

Our results are in line with those of Zaiss et al. (8)

Th1/Th2/Th17 cytokine production. A similar obser-

who detected both Egfr messenger ribonucleic acid

vation has been reported in a mouse model of graft

expression and EGFR protein in purified CD4 þ T

versus host disease, with a reduction of Th1 and Th2

cells.

cytokine production in erlotinib-treated animals (11).

genetic

invalidation

of

Egfr

markedly

169

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 6 Selective EGFR Deletion in CD4þ T Cells in a Chimeric Ldlr / Mouse Model of Atherosclerosis

A

Cd4Cre Egfr

+/+

/Ldlr

-/-

Cd4Cre Egfr

lox/lox

/Ldlr

B

-/-

Cd4Cre Egfr

% CD4+ T Cells Among Splenocytes

10

25 10

20

% Tregs Among CD4 T cells

5

10

3

CD3+ cd4+ 16.4

10

10

10

* 10

4

15

CD3+ cd4+ 17.6

CD3

0 /Ldlr

lox/lox

Egfr

/Ldlr

-/-

3

0

10

3

10

4

10

5

3

–10

0

10

3

10

4

Egfr

CD4

+/+

/Ldlr

-/-

D % CD44

high

10

10

Egfr

lox/lox

3

/Ldlr

–10

10

10

3

10

4 CD4

–10

Egfr

E

/Ldlr

-/-

lox/lox

Egfr

/Ldlr

0

3

10

10

4

10

5

–10

3

0

3

10

10

4

10

Suppression of Effector T Cell Proliferation (%)

100 80

+/+

/Ldlr

4

3

3

0

10

Cd4Cre Egfr

10

4

10

5

+/+

60

3

-/-

Egfr

lox/lox

/Ldlr

3

0

10

10

4

5

10

3

–10

0

10

3

10

4

10

5

CD44high

CD62L

-/-

lox/lox

0 –10

CD4Cre

Cd4Cre Egfr 40 20

3

G

3

CD4+ T Proliferation (X10 cpm)

100

–10

80

3

CD4Cre Egfr

F

5

5

–10

0

5

CD69

-/-

10

0

3 3

4

8

cd69+ 7.73

0

–10

10

*

4

CD4Cre

10

-

10

CD4

3

0

CD62L Among

12

cd69+ 11.3

Foxp3+, CD25+ 8.25

Foxp3

+

10 5

Foxp3+ CD25+ 8.97

3

-/-

CD4 T Cells

5

*

+/+

-/-

5

CD4Cre

CD4Cre

5

10

% CD69+ among CD4+ T cells

CD4Cre

/Ldlr

–10

0 –10

C 15

lox/lox

3

CD25

3

CD4Cre -/-

Cd4Cre Egfr

4

0

–10 +/+

-/-

Count

5

Egfr

/Ldlr

10

5

0

CD4Cre

+/+

+

2

Cytokine Production by CD4+ T cells (10 pg/ml) 6

*

*

75 4

60 50 40

2 25

20

* 0

0 1:1

1:2

1:4

1:8

1:16

-/-

+/+

-/-

Lox/lox

Ldlr /CD4Cre Egfr

CD4Cre Egfr

Teffector/Treg Ratio Ldlr /CD4Cre Egfr

0 +/+

/Ldlr

Ifn-Y

CD4Cre -/-

lox/lox

Egfr

/Ldlr

IL-4

IL-10

-/-

CD4Cre Egfr CD4Cre Egfr

+/+

/Ldlr

lox/lox

-/-

/Ldlr

-/-

(A) Representative examples and fluorescence-activated cell sorting (FACS) quantification of CD4þ T subset. (B) Representative examples and FACS quantification of CD4þCD25þFoxP3þ regulatory T cells. Representative examples and quantification of CD69 expression by CD4þ T cell (C) and CD44high expression by CD62LCD4þ T cells by flow cytometry (D). (E) In vitro suppressive tests of effector CD25 T cell proliferation by coculture with CD4þCD25high regulatory T cells isolated from chimeric Cd4-Cre/Egfrþ/þ/Ldlr/ or chimeric Cd4-Cre/Egfrlox/lox/Ldlr/ mice. Purified CD4þ T cells from chimeric mice were stimulated by soluble anti-CD3. Genetic invalidation of Egfr reduced T cell proliferation (F), IFN-g and IL-4 production but had no effect on IL-10 (G) ELISA in the supernatant). N ¼ 8 or 9/group; *p < 0.05. Data are expressed as median (5th percentile to 95th percentile). Abbreviations as in Figures 1 to 3.

In addition, we observed that EGFR inhibition and

migration of normal and malignant epithelial cells

invalidation

within

(13). Recently, Tai et al. (14) showed that EGFR/

atherosclerotic lesions, suggesting a modulation of

Src-signaling triggers the tyrosine phosphorylation

T cell migration. This might account for reduced

of b 4 integrin, which in turn activated focal adhesion

chemokine production, as described in a model of

kinase, which is involved in cytoskeleton reorgani-

skin inflammation, in which TKI treatment decreased

zation. In our study, we did not observe any side

Ccl-17, Ccl-21, and Ccl-27 production (11). Reduction

effect due to EGFR blockade, including survival of

of T cell infiltration in the vascular wall of Cd4Cre

erlotinib-treated

Egfr lox/lox animals might also be due to altered cell

addition, we did not find any difference in weight

motility resulting from impairment of cytoskeleton

or infection susceptibility between groups. These

reorganization (12). A large body of evidence in

observations suggest that EGFR inhibition modulates

cancer highlighted the role of EGFR signaling in

the

epithelial-mesenchymal transition and invasion or

immunosuppression.

reduced

T

cell

infiltration

immune

or

Cd4-Cre

response

but

Egfrlox/lox

does

mice.

cause

In

full

170

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

F I G U R E 7 Selective EGFR Deletion in CD4 þ T Cells Reduced Experimental Atherosclerosis Development

A

Lox/Lox /Ldlr-/Cd4Cre Egfr+/+/Ldlr-/- Cd4Cre Egfr

500 4 weeks of fat diet

Plaque Size (Aortic Sinus, X103μm2) *

400 300 *

200 100

12 weeks of fat diet

*

0 4 Weeks CD4Cre Egfr

B Cd4Cre Egfr+/+/Ldlr-/- Cd4Cre EgfrLox/Lox/Ldlr-/-

+/+

6 Weeks -/-

/Ldlr

12 Weeks

CD4Cre Egfrlox/lox /Ldlr-/-

T Cell Inf iltration (% Plaque Area) 8

*

6 Anti-CD3 staining

4 2 0 CD4Cre CD4Cre Egfr+/+ /Ldlr-/- Egfrlox/lox /Ldlr-/-

(A) Representative photomicrographs of Oil Red staining and quantitative analysis of atherosclerotic lesion size in the aortic root of irradiated Ldlr/ mice reconstituted with bone marrow from either Cd4-Cre/Egfrþ/þ or Cd4-Cre/Egfrlox/lox mice and put on a high-fat diet for 4 (n ¼ 5 or 6/group), 6 (n ¼ 9/group), and 12 (n ¼ 9 or 10/group) weeks. (B) Representative photomicrographs and quantitative analysis of T cells (CD3 staining) in atherosclerotic lesion of irradiated Ldlr/ mice reconstituted with bone marrow from either Cd4-Cre/Egfrþ/þ or Cd4-Cre/Egfrlox/lox mice and put on a high-fat diet for 4 weeks. *p < 0.05. EGFR ¼ epidermal growth factor receptor.

EGFR inhibition and invalidation did not affect

suppressive functions. In mouse models of athero-

cell death susceptibility but induced a global CD4þ

sclerosis, Treg deficiency, obtained by Foxp3, Cd28,

T cell anergy. The mechanisms of anergy induced by

or Cd80/86 genetic invalidation, increased T cell

EGFR

kinase

activation and accelerated vascular disease (17–19).

signaling pathway as suggested by reduced Erk

However, in our study, the genetic invalidation of

inhibition

likely

MAP

with

Egfr in Ldlr/ chimeric mouse model had no effect on

AG-1478. This is in agreement with studies by Luo

the suppressive function of Treg cells. We observed

et al. (15) showing that erlotinib caused G0/G1 arrest

only a slight reduction in the Treg pool in chimeric

and suppressed the phosphorylation of c-Raf and Erk

Cd4-Cre/Egfrlox/lox/Ldlr/ compared with Cd4-Cre/

in activated T cells. We also showed that EGFR

Egfrþ/þ/Ldlr / mice. Our results show that EGFR in-

pharmacological blockade negatively affected intra-

hibition had a predominant effect on T cell anergy and

cellular calcium signaling in T cells, confirming

reduced atherosclerosis development. In vitro, we

previous reports on cancer context. Bryant et al. (16)

challenged T cells with anti-CD3 antibody or conca-

showed on glioma tumor cell lines that TKIs

navalin A and observed EGFR phosphorylation, prob-

including erlotinib and gefitinib limited the acute

ably through transactivation. Transactivation of EGFR

cytoplasmic release of calcium from the endoplasmic

is well documented for a G protein–coupled receptor

reticulum in response to EGF.

such as ATR-1, the receptor for angiotensin II (20). This

phosphorylation

in

CD4 þ

involved T

cells

treated

A recent study suggested that EGFR and 1 of its

transactivation is mediated by metalloproteinase-

ligands, amphiregulin, play a specific role in Treg

dependent release of EGFR ligands, including EGF,

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

C ENTR AL I LL U STRA T I O N EGFR Blockade Limits Proatherogenic Activity of CD4 þ T Cells

Zeboudj, L. et al. J Am Coll Cardiol. 2018;71(2):160–72. Epidermal growth factor receptor (EGFR) engagement orchestrates CD4þ T cell proliferation, cytokine production, and recruitment within atherosclerotic plaques. Pharmacological blockade of EGFR signaling may constitute an attractive novel approach for the treatment of atherosclerosis. IFN ¼ interferon; IL ¼ interleukin; Th ¼ T helper.

transforming growth factor–a , and heparin-binding

engagement is important for macrophage proathero-

EGF, from their cell membrane–bound precursors

genic activity, as its pharmacological inhibition

and intermediary signaling molecules, including

reduced proinflammatory cytokine production, lipid

intracellular Ca 2þ, protein kinase C, and cytosolic

uptake, and oxidative stress. In the present study, we

tyrosine kinases such as Src kinases (21).

report that global EGFR inhibition and CD4þ T cell–

The expression of EGFR and its ligands has been

specific deletion of EGFR reduced both atheroscle-

detected in experimental and human atherosclerosis

rosis development and progression and induced a less

(9,22). Wang et al. (23) recently reported that EGFR

inflammatory plaque phenotype.

171

172

Zeboudj et al.

JACC VOL. 71, NO. 2, 2018 JANUARY 16, 2018:160–72

EGFR Blockade in T Cells Reduces Atherosclerosis

Our findings indicate that EGFR inhibitors, widely used in patients with cancer, are unlikely to worsen

ADDRESS

the risk for cardiovascular disease and further suggest

Ait-Oufella, Inserm U970, Paris Cardiovascular Research

that EGFR may constitute a novel therapeutic target

Center, Université René Descartes, 56, rue Leblanc, Paris

in

75012, France. E-mail: hafi[email protected].

atherosclerotic

disease.

The

recent

positive

FOR

CORRESPONDENCE:

Dr.

Hafid

results of CANTOS (Canakinumab Anti-Inflammatory Thrombosis Outcomes Study) (24) highlighted that

PERSPECTIVES

modulating the immune system could be a promising approach to treat atherosclerosis-related cardiovascular diseases.

COMPETENCY IN MEDICAL KNOWLEDGE: EGFRs are critical regulators of CD4þ T cell

STUDY LIMITATIONS. In this study, we showed that

activity in both mice and humans. Pharmacolog-

EGFR inhibition reduced both atherosclerosis devel-

ical EGFR blockade by erlotinib or selective

opment and progression in mice. Further studies are

invalidation in CD4þ T cells limits the develop-

required to address the cardiovascular effects of

ment and progression of atherosclerosis in an

EGFR tyrosine kinase inhibitors in cancer patients.

experimental model.

CONCLUSIONS

TRANSLATIONAL OUTLOOK: More studies are

EGFR is expressed in human and mouse CD4 þ T cells. EGFR pharmacological blockade or CD4þ T cell– specific invalidation induced T cell anergy and reduced both atherosclerosis development and pro-

needed to assess the potential therapeutic utility of pharmacological EGFR inhibition as an immunomodulatory approach to preventing complications of atherosclerosis.

gression in mice.

REFERENCES 1. Onn A, Isobe T, Wu W, et al. Epidermal growth factor receptor tyrosine kinase inhibitor does not improve paclitaxel effect in an orthotopic mouse model of lung cancer. Clin Cancer Res 2004;10: 8613–9. 2. Gridelli C, Rossi A, Maione P, et al. Erlotinib in non-small-cell lung cancer. Expert Opin Pharmacother 2007;8:2579–92. 3. Guo G, Gong K, Wohlfeld B, Hatanpaa KJ, Zhao D, Habib AA. Ligand-independent EGFR signaling. Cancer Res 2015;75:3436–41. 4. Krakstad C, Chekenya M. Survival signalling and apoptosis resistance in glioblastomas: opportunities for targeted therapeutics. Mol Cancer 2010;9:135.

atherosclerotic lesion formation in mice. Circ Res 2003;93:884–8.

dendritic cells in atherosclerosis. Circ Res 2014; 114:1640–60.

11. Morin F, Kavian N, Marut W, et al. Inhibition of EGFR tyrosine kinase by erlotinib prevents sclerodermatous graft-versus-host disease in a mouse

19. Ait-Oufella H, Salomon BL, Potteaux S, et al. Natural regulatory T cells control the development of atherosclerosis in mice. Nat Med 2006;12:178–80.

model. J Invest Dermatol 2015;135:2385–93.

20. Mehta PK, Griendling KK. Angiotensin II cell signaling: physiological and pathological effects in the cardiovascular system. Am J Physiol Cell Physiol 2007;292:C82–97.

12. Dudu V, Able RA Jr., Rotari V, Kong Q, Vazquez M. Role of epidermal growth factor-triggered PI3K/Akt signaling in the migration of medulloblastomaderived cells. Cell Mol Bioeng 2012;5:502–13. 13. Camorani S, Crescenzi E, Colecchia D, et al. Aptamer targeting EGFRvIII mutant hampers its constitutive autophosphorylation and affects migration, invasion and proliferation of glioblastoma cells. Oncotarget 2015;6:37570–87.

5. Bareschino MA, Schettino C, Troiani T, Martinelli E, Morgillo F, Ciardiello F. Erlotinib in cancer treatment. Ann Oncol 2007;18 Suppl 6:vi35–41. 6. Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N Engl J Med 2008;358:1160–74. 7. Chan G, Nogalski MT, Yurochko AD. Activation of EGFR on monocytes is required for human cytomegalovirus entry and mediates cellular motility. Proc Natl Acad Sci U S A 2009;106:22369–74. 8. Zaiss DM, van Loosdregt J, Gorlani A, et al. Amphiregulin enhances regulatory T cellsuppressive function via the epidermal growth factor receptor. Immunity 2013;38:275–84. 9. Dreux AC, Lamb DJ, Modjtahedi H, Ferns GA. The epidermal growth factor receptors and their family of ligands: their putative role in atherogenesis. Atherosclerosis 2006;186:38–53. 10. Mallat Z, Gojova A, Sauzeau V, et al. Rhoassociated protein kinase contributes to early

14. Tai YL, Chu PY, Lai IR, et al. An EGFR/Srcdependent beta4 integrin/FAK complex contributes to malignancy of breast cancer. Sci Rep 2015;5:16408. 15. Luo Q, Gu Y, Zheng W, et al. Erlotinib inhibits T-cell-mediated immune response via downregulation of the c-Raf/ERK cascade and Akt signaling pathway. Toxicol Appl Pharmacol 2011; 251:130–6. 16. Bryant JA, Finn RS, Slamon DJ, Cloughesy TF, Charles AC. EGF activates intracellular and intercellular calcium signaling by distinct pathways in tumor cells. Cancer Biol Ther 2004;3:1243–9. 17. Subramanian M, Thorp E, Hansson GK, Tabas I. Treg-mediated suppression of atherosclerosis requires MYD88 signaling in DCs. J Clin Invest 2013;123:179–88. 18. Ait-Oufella H, Sage AP, Mallat Z, Tedgui A. Adaptive (T and B cells) immunity and control by

21. Higuchi S, Ohtsu H, Suzuki H, Shirai H, Frank GD, Eguchi S. Angiotensin II signal transduction through the AT1 receptor: novel insights into mechanisms and pathophysiology. Clin Sci (Lond) 2007;112:417–28. 22. Nakata A, Miyagawa J, Yamashita S, et al. Localization of heparin-binding epidermal growth factor-like growth factor in human coronary arteries. Possible roles of HB-EGF in the formation of coronary atherosclerosis. Circulation 1996;94:2778–86. 23. Wang L, Huang Z, Huang W, et al. Inhibition of epidermal growth factor receptor attenuates atherosclerosis via decreasing inflammation and oxidative stress. Sci Rep 2017;8:45917. 24. Ridker PM, Everett BM, Thuren T, et al. Antiinflammatory therapy with canakinumab for atherosclerotic disease. N Engl J Med 2017;377:1199–231.

KEY WORDS atherosclerosis, immunity, inflammation, lymphocyte

A PPE NDI X For a supplemental Methods section as well as figures, please see the online version of this article.