Inhibition of epidermal growth factor receptor (EGFR) activation protected mice from abdominal aortic aneurysm (AAA), which is probably due to suppression of vascular endoplasmic reticulum (ER) stress. Since the signal was confirmed in humans, EGFR may provide a novel therapeutic target for aortic aneurysm.
No pharmacological treatment has been established that can inhibit or slow AAA development and rupture. We have shown that EGFR ‘trans’-activation is a critical signal transduction of Ang II in VSMC.
In the present study, we demonstrated that a pharmacological inhibition of EGFR by erlotinib prevented AAA development and rupture but not hypertension induced by Ang II plus BAPN treatment. The prevention was associated with inhibition of ER stress and inflammatory responses. Activation of EGFR was confirmed in human AAA.
Our finding of the prevention of AAA by a clinically utilized EGFR inhibitor provides a novel therapeutic target in this disease and will affect the field seeking a drug for AAA.
The renin–angiotensin II (Ang II) system has been implicated in the development of abdominal aortic aneurysm (AAA). Although a detailed molecular mechanism by which Ang II promotes AAA remains unclear, it seems to involve multiple cell types [leucocytes, vascular smooth muscle cells (VSMC) and endothelial cells] and signalling responses such as induction of oxidative stress, inflammatory cytokines and matrix metalloproteases (MMP) . We have demonstrated the requirement of a disintegrin and metalloproteinase domain-containing protein 17 (ADAM17) and subsequent epidermal growth factor receptor (EGFR) activation in Ang II-induced VSMC hypertrophy [2,3]. The EGFR activation by Ang II also mediates generation of reactive oxygen species in VSMC . ADAM17 expression is enhanced in AAA and ADAM17-silenced mice did not develop CaCl2-induced AAA . In VSMC, both EGFR and ADAM17 co-localize at caveolae  and AAA formation induced by Ang II plus β-aminopropionitrile (BAPN), a lysyl oxidase (Lox) inhibitor, was attenuated in caveolin-1 deficient mice . The attenuation of AAA formation was associated with suppression of ADAM17 induction, EGFR activation, endoplasmic reticulum (ER) stress, oxidative stress and induction of interleukin-6 and MMP-2 .
Lox cross-links collagen and elastin fibres thus stabilizing the vessel wall and de-regulation of Lox has been implicated in AAA formation and rupture . Decreased Lox expression and activity is associated with several AAA models [8,9] including apoE−/− mouse infused with Ang II . Ehlers Danlos Syndrome VI patients carry a mutation in the Lox gene and some of the patients develop aortic dilatation or arterial rupture . Deletion of Lox gene in mice causes aortic aneurysm and rupture . However, pharmacological inhibition of Lox by BAPN in normolipidaemic mice was insufficient to develop AAA and requires a co-treatment with Ang II [13,14]. This model reproducibly produces AAA associated with hypertension with morphological and histological characteristics similar to human AAA, but without atherosclerosis observed in other Ang II-dependent AAA models in hyperlipidaemic mice .
Erlotinib (Tarceva) is a small molecule kinase domain inhibitor of the EGFR, which has been approved by the FDA (U.S. Food and Drug Administration) for treatment of advanced non-small cell lung cancer and pancreatic cancer . In the present study, we have tested our hypothesis that inhibition of EGFR by erlotinib prevents AAA formation induced by Ang II plus BAPN.
MATERIALS AND METHODS
All animal procedures were performed with the prior approval of the Temple University Institutional Animal Care and Use Committee and in accordance with National Institutes of Health Guide for the Care and Use of Laboratory Animals. Male C57BL/6J mice were obtained from the Jackson Laboratory. Standard sterilized laboratory diet and water were available ad libitum. Eight-week-old mice were infused with Ang II (1 μg/kg/min) with or without co-infusion of erlotinib (7.5 mg/kg/day) for 4 weeks through osmotic-pump (Alzet, Durect Corp)  and received BAPN in drinking water (1 mg/ml) for the first 2 weeks . The control mice (n=8) were sham-operated at 8 weeks. Note that due to an expected mortality (50%–60%) in C57BL/6J mice with the Ang II plus BAPN treatment  and a prior pilot data with erlotinib (5–10 mg/kg/day) treatment demonstrating nearly 100% survival, 18 mice were co-treated with Ang II and BAPN, whereas eight mice were further treated with erlotinib. The high-resolution 2D imaging (B mode) to measure luminal diameters of abdominal aortas at the maximal dilation was performed with high frequency ultrasound (VisualSonics Velvo2100) on day 0, 14, 21 and 28. Blood pressure and heart rate were evaluated in the conscious state at day 28 by telemetry (DSI equipped with ADInstrument 6 software) via carotid catheter (PA-C10 transmitter). To prepare samples for histological analysis, mice were perfused with saline followed by 10% paraformaldehyde at 100 mmHg. Aortas were dissected, cleaned of extraneous tissue and then photographed. AAA was defined as a localized dilation of the abdominal aortic wall with maximal outside diameter 50% greater (> 1.7 mm) than the maximum outside diameter of sham-operated mice (1.1 mm).
We obtained surgical specimens from individuals with AAA. The Temple University Institutional Review Board approved the protocol. Formalin-fixed control human aortas were obtained from Advanced Tissue Services.
Serial cross-sections (5 μm thick) from abdominal aortas (each n=4, randomly selected from each group) were de-paraffinized and blocked in 5% goat serum and 1% BSA for 1 h at room temperature, incubated with primary antibody in PBS containing 1% BSA and 0.1% Tween 20 for 18 h at 4°C, followed by biotinylated secondary antibody for 90 min at room temperature. Slides were incubated with avidin–biotin peroxidase complex for 30 min at room temperature and staining was visualized with the substrate diaminobenzidine (Vector) producing a brown colour and counterstained with haematoxylin. An equal concentration of control IgG was used side-by-side with each antibody to ensure staining specificity . Quantification of the antibody staining was performed as reported previously with subtraction of the IgG background staining . All images were visualized on a Photometrics CoolSNAP HQ digital camera and were acquired with SPOT 4.7 Basic software using the same exposure time. Images were loaded into the ImageJ program (http://rsb.info.nih.gov/ij) for analysis. A region of interest was drawn around the entire aorta with the freehand selection tool. Adventitia was excluded from the quantification, since the adventitial areas were quite limited in aortas, except those with AAA. All images were set to the same hue, saturation and brightness. The area and intensity (absorbance) in the region of interest were then measured and analysed. Data were obtained from four mice in each group with 3–4 non-overlapping high-power fields for each antibody.
Quantitative real-time PCR
Abdominal aortas were homogenized using BioMasher and total RNA was extracted using TRIzol reagent (Invitrogen). cDNA was synthesized RevertAid First Strand cDNA Synthesis Kit (Thermo). Quantitative real-time PCR (qPCR) was performed with SYBR Green qPCR Master Mix (Fermentas) as described previously . mRNA abundance was calculated by normalization to ribosome 18S. The primers used are ADAM17: forward GGC GCG GGA GGG AGA AGT TT, reverse CGC CGC CTC ATG TTC CCG TC; ribosome 18S: forward AGT TCC AGC ACA TTT TGC GAG, reverse TCA TCC TCC GTG AGT TCT CCA.
VSMC were prepared from abdominal aorta of male Sprague–Dawley rats (∼350 g) by the explant method as described previously . Rats were killed by exsanguination under anaesthesia [ketamine 100 mg/kg and xylazine 5 mg/kg, i.p. (intraperitoneally)]. VSMC were sub-cultured in DMEM (Dulbecco's modified Eagle's medium) containing 10% FBS, penicillin and streptomycin. Cells from passage 3–10 at 80% ∼ 90% confluences in culture wells were made quiescent by incubation with serum-free medium for 2–3 days. To avoid any potential phenotypic alteration, VSMC were renewed every 2–3 months and VSMC from frozen stock were not used in the present study. The results were confirmed in at least two distinct cell lines.
Immunoblotting was performed as previously described . Quiescent VSMCs grown on six-well plates were stimulated for specified durations. The reaction was terminated by the replacement of medium with 100 μl of 1× SDS sample buffer. 40 μl of the cell lysates were subjected to SDS/PAGE and electrophoretically transferred to a nitrocellulose membrane. The membranes were then exposed to primary antibodies overnight at 4°C. After incubation with the peroxidase-linked secondary antibody for 1 h at room temperature, immunoreactive proteins were visualized using a chemiluminescence reaction kit. The results were quantified by densitometry in the linear range of film exposure using CanoScan N670U (Canon) and Un-Scan-It Gel 5.3 software (Silk Scientific). An example of data supporting the linearity has been demonstrated .
Ang II was purchased from Bachem. BAPN was purchased from TCI. Erlotinib (OSI Pharmaceuticals) was provided by Genentech. Antibody against Tyr1068-phosphorylated EGFR was purchased from Cell Signaling. Antibody against an ER stress marker, C/EBP-homolog protein-10/a growth arrest and DNA damage-inducible gene-153 (CHOP-10/GADD-153), tissue growth factor (TGF)-β and EGFR were purchased from Santa Cruz Biotechnology. Antibodies against ADAM17 and MMP-2 were purchased from Abcam. Antibody against Lys-Asp-Glu-Leu (KDEL) for detection of ER stress markers, glucose-regulated protein-78 and -94, (GRP78 and GRP94) was purchased from Enzo Life Sciences. Antibody against an oxidative stress marker, nitro-tyrosine was purchased from Millipore. Antibody against interleukin-6 was purchased from Bioss.
Kaplan–Meier survival curves were constructed and analysed using log-rank (Mantel–Cox) test. Fisher's exact test was used to analyse categorical data. Differences between multiple groups were analysed by two-way ANOVA, followed by the Tukey–Kramer post-hoc test. Data are presented as mean ± S.E.M. Statistical significance was taken at P< 0.05.
There was a significant difference in survival rates between mice treated with an EGFR inhibitor, erlotinib (100%) and non-treated mice (44.4%) during Ang II plus BAPN treatment to induce AAA. Among the 10 mice that died before 4 weeks, we were able to perform necropsy on seven mice and ruptured aortae were recognized in all of these mice (three at thoracic level and four at abdominal level) suggesting that the major cause of death in these mice was rupture. All surviving mice without erlotinib treatment had AAA with significant enlargement of maximum diameters of the abdominal aortas. One mouse also had a mild ascending aortic aneurysm (diameter 1.65 compared with 1.05±0.13 mm saline control). In contrast, erlotinib-treated mice had far less AAA incidence (one type I AAA among eight mice) and no statistically significant enhancement of the diameters. No thoracic aortic aneurysm (TAA) was observed. However, both erlotinib-treated and non-treated mice developed hypertension assessed by telemetry (Figure 1).
Prevention of mortality and AAA development but not hypertension in a mouse model of Ang II-dependent AAA
The AAA induced by Ang II plus BAPN was associated with vascular fibrosis/matrix deposition, enhanced EGFR activation, induction of ADAM17, MMP-2 and intereukin-6 and ER/oxidative stress. These AAA-associated responses were attenuated in mice treated with erlotinib (Figure 2A; Supplementary Figure S1). EGFR activation, AAA-associated protein induction and enhanced ER stress were also observed in medial area of human AAA (Figure 2B).
Enhancement of EGFR-linked signal transduction in mouse and human AAA
Expression of a fibrosis-associated cytokine, TGF-β, was enhanced in the AAA but not in erlotinib-treated abdominal aortas (Supplementary Figure S2). In addition, erlotinib suppressed ADAM17 mRNA induction seen in the AAA samples and EGFR activation induced by Ang II in cultured abdominal aortic smooth muscle cells (Supplementary Figure S3).
In the present study, we found a significant prevention of AAA in mice treated with erlotinib, an EGFR inhibitor. EGFR transactivation has been implicated in Ang II pathophysiology. Several findings support this concept in cardiac hypertrophy  and renal fibrosis . Our findings add new information to the field seeking a pharmacological treatment for AAA. The involvement of EGFR is supported by the aforementioned role of ADAM17, a critical EGFR activator, in AAA. Although the cell type responsible for linking EGFR activation to AAA was not determined in the present study, our past  and present data suggest the involvement of VSMC. ADAM17 gene was identified among the central genes associated with AAA development in mice . In the present study, erlotinib attenuated the ADAM17 induction in AAA suggesting a potential feed-forward loop of ADAM17 induction via the EGFR .
The exact mechanism of aortic rupture in this model remains unknown. In the original manuscript reporting this mouse AAA model , thoracic aortic rupture rate was even higher than abdominal aortic rupture whereas incidence of TAA was less than AAA. This tendency was confirmed in our past  and present studies. However, a related yet distinct protocol resulted in 100% incidence of thoracic aortic dissection in mice pre-treated with BAPN followed by Ang II infusion. Treatment with an MMP inhibitor was able to prevent the dissection . Therefore, erlotinib might prevent aortic rupture in the present study by inhibiting MMP induction. It has been shown that MMP inhibition also prevents AAA development but not hypertension in hyperlipidaemic mice infused with Ang II . Although contribution of hypertension to the development of AAA in this model has been suggested , our past  and present study also demonstrated that hypertension did not participate in AAA development in this model of AAA.
Upon activation by Ang II, EGFR mediates various downstream responses in VSMC, including oxidative stress  and intereukin-6 induction . This cascade was also evident in the present study and probably contributed to AAA development, according to the literature [25,26]. TGF-β is a critical factor for tissue fibrosis . TGF-β appears to be induced by Ang II through EGFR activation and mediate renal fibrosis in mice . Our data may confirm its role in vascular fibrosis associated with AAA. However, the role of TGF-β in mediating AAA is highly controversial .
We are aware that the incidence of AAA in C57BL/6 mice with Ang II alone is very low [13,14] even though the infusion is capable of activating EGFR in vasculatures . We interpret from this that EGFR activation is required to advance AAA but insufficient to initiate AAA, which requires an additional signal or condition such as priming by BAPN or hyperlipidaemia (Supplementary Figure S4). The limitations of the present study include lack of additional quantitative supports by immunoblotting and lack of confirmation in other AAA models. Future research is desired to identify the synergistic mechanism sufficient for AAA initiation as well as to address these limitations.
abdominal aortic aneurysm
a disintegrin and metalloproteinase domain-containing protein 17
II, angiotensin II
epidermal growth factor receptor
quantitative real-time PCR
thoracic aortic aneurysm
tissue growth factor
vascular smooth muscle cells
Takashi Obama and Toshiyuki Tsuji performed animal experiments and some of the immunohistochemical experiments and contributed to the discussion. Tomonori Kobayashi, Yamato Fukuda, Takehiko Takayanagi and Yoshinori Taro performed some of the animal studies, cell culture and immunohistochemical experiments. Katherine J. Elliott edited the paper. Eric Choi and Alan Daugherty provided expertise in human AAA and mouse AAA models respectively and contributed to the discussion. Satoru Eguchi and Victor Rizzo conceived, designed and coordinated the research plan and wrote/edited the paper. Steven J. Forrester and Tatsuo Kawai performed the immunohistochemical experiments for revision.
This work was supported by the National Institute of Health [grants numbers HL076770 (to S.E.) and HL086551 (to V.R.)]; and the American Heart Association [grant number 13GRNT17060036 (to S.E.)].
These authors contributed equally.