Pharmacological inhibition of epidermal growth factor receptor prevents intracranial aneurysm rupture by reducing endoplasmic reticulum stress

Abstract

BACKGROUND:

Multiple pathways and factors are involved in the rupture of intracranial aneurysms. The epidermal growth factor receptor (EGFR) has been shown to mediate inflammatory vascular diseases, including atherosclerosis and aortic aneurysm. However, the role of EGFR in mediating intracranial aneurysm rupture and its underlying mechanisms have yet to be determined. Emerging evidence indicates that endoplasmic reticulum (ER) stress might be the link between EGFR activation and the resultant inflammation. ER stress is strongly implicated in inflammation and apoptosis of vascular smooth muscle cells, both of which are key components of the pathophysiology of aneurysm rupture. Therefore, we hypothesized that EGFR activation promotes aneurysmal rupture by inducing ER stress.

METHODS:

Using a preclinical mouse model of intracranial aneurysm, we examined the potential roles of EGFR and ER stress in developing aneurysmal rupture.

RESULTS:

Pharmacological inhibition of EGFR markedly decreased the rupture rate of intracranial aneurysms without altering the formation rate. EGFR inhibition also significantly reduced the mRNA expression levels of ER stress markers and inflammatory cytokines in cerebral arteries. Similarly, ER stress inhibition also significantly decreased the rupture rate. In contrast, ER stress induction nullified the protective effect of EGFR inhibition on aneurysm rupture.

CONCLUSIONS:

Our data suggest that EGFR activation is an upstream event that contributes to aneurysm rupture via the induction of ER stress. Pharmacological inhibition of EGFR or downstream ER stress may be a promising therapeutic strategy for preventing aneurysm rupture and subarachnoid hemorrhage.

Graphical Abstract

Introduction

Intracranial aneurysm rupture causes subarachnoid hemorrhage, resulting in severe mortality and morbidity.¹ Currently, available therapies for the prevention of aneurysm rupture are limited to invasive treatments such as surgical clipping and endovascular coiling.¹ Although these invasive therapies are well established, the adverse outcome rates from these procedures still present significant procedural risks.^1,2 Therefore, the pharmacological prevention of aneurysmal rupture is emerging as a potential alternative approach for patients with unruptured aneurysms.

Inflammation is emerging as a key component of the pathophysiology of intracranial aneurysms.^3–6 Given its potential as a therapeutic target, a better understanding of molecular pathways involved in this inflammatory process may contribute to the development of medical therapies for the prevention of aneurysm rupture and subsequent subarachnoid hemorrhage.

Previous studies have consistently shown that the activation of the renin-angiotensin II system in the vascular walls of intracranial aneurysms induces aneurysm rupture, which is independent of hypertension.^7–9 Upon activation by angiotensin II, the epidermal growth factor receptor (EGFR) in arterial walls has been found to play a critical role in mediating inflammatory responses and promoting vascular damage.^10–12 EGFR, referring to a family of receptor tyrosine kinases, is expressed in vascular smooth muscle cells (VSMCs). Activation of EGFR has been found to stimulate phenotypic modulation of contractile VSMCs and promote the proliferation and migration of VSMCs.^13–15 EGFR was shown to mediate vascular diseases that involve an inflammatory component, including atherosclerosis and aortic aneurysm.^16–18 However, the role of EGFR in mediating intracranial aneurysm rupture and its underlying mechanisms remained to be determined.

Emerging evidence indicates that endoplasmic reticulum (ER) stress might be the link between EGFR activation and the resultant inflammation in vascular diseases.^19–22 The endoplasmic reticulum is responsible for cellular protein synthesis and folding. Cellular stimuli that perturb ER homeostasis create a stress condition termed “ER stress.”^23,24 Under ER stress, unfolded protein aggregation and proteotoxicity induce pro-inflammatory responses via three unfolded protein response (UPR) pathways, IRE1, PERK, and ATF6.^23,24 Maladaptive UPR accompanied by excessive ER stress causes oxidative stress and inflammatory cell infiltration.^25,26 Literature suggests that ER stress and ensuing UPR are strongly implicated in the pathophysiology of vascular diseases involving VSMC inflammation and apoptosis.^23,27,28

Taking this information together, we hypothesized that EGFR activation promotes the development of aneurysmal rupture by inducing ER stress. In this study, we examined the potential roles of EGFR and ER stress in developing aneurysmal rupture using a mouse model.

Methods

Experiments were conducted following guidelines approved by the Institutional Animal Care and Use Committee. Details of experimental methods are available in Supplemental Materials.^{5,7,8,19,29–45} Fisher’s exact test was used to analyze the incidences of aneurysm formation and subarachnoid hemorrhage. Log-rank (Mantel-Cox) test was used for the analysis of the survival rate. RT-PCR data were analyzed by the Mann-Whitney U test.

Results

Inhibition of EGFR prevented the development of intracranial aneurysm rupture

As a first step to test the potential role of EGFR activation in the development of aneurysm rupture, we used an EGFR-specific inhibitor, erlotinib, in our well-established mouse model of intracranial aneurysm (Figure 1).^29,30 We employed two treatment schemes. One was to give erlotinib one day before aneurysm induction (pre-treatment), and the other was to administer erlotinib six days after aneurysm induction (post-treatment). This was because we have previously observed that aneurysms were formed in the first week in our model.^29,30 Therefore, the pre-treatment scheme would confirm whether inhibition of EGFR would have any effect on aneurysm formation. The post-treatment scheme was designed to test the effect of EGFR inhibition on aneurysm rupture after aneurysms formed.

Figure 1. Representative images of unruptured and ruptured aneurysms.

Circle of Willis in mouse brains were perfused with bromophenol blue dye. (A), No aneurysm. (B), Unruptured aneurysm. (C), Ruptured aneurysm with subarachnoid hemorrhage. Arrowheads indicate intracranial aneurysms.

Pre-treatment with erlotinib did not change the incidence of aneurysms (Figure 2A, vehicle vs. erlotinib, 67% vs. 54%, n = 27 vs. 28, P = 0.41). In contrast, pre-treatment with erlotinib significantly decreased the rupture rate compared to the vehicle treatment (Figure 2B, vehicle vs. erlotinib, 72% vs. 33%, n = 18 vs. 15, P < 0.05). Mice treated with erlotinib also had a significantly better symptom-free survival rate than vehicle-treated mice (Figure 2C, P < 0.05). There was no significant difference in blood pressure between the two groups (Table S2).

Figure 2. Inhibition of EGFR activation (pre-treatment) reduced the rate of aneurysm rupture.

Pre-treatment of the EGFR inhibitor, erlotinib, significantly decreased the rupture rate of intracranial aneurysms without altering the aneurysm formation rate (A and B). A significantly increased symptom-free survival rate (C) was seen in erlotinib-treated mice compared to vehicle-treated mice. Fisher’s exact test was used to analyze the rupture rate of aneurysms (B). Log-rank (Mantel-Cox) test was used for the analysis of the survival rate (C). ∗ P < 0.05.

Similar to the results in the pre-treatment scheme, post-treatment with erlotinib did not change the incidence of aneurysms (Figure 3A, vehicle vs. erlotinib, 61% vs. 43%, n = 18 vs. 16, P = 0.49). However, post-treatment with erlotinib significantly decreased the rupture rate compared to the vehicle treatment (Figure 3B, vehicle vs. erlotinib, 72% vs. 14%, n = 11 vs. 7, P < 0.05). Mice treated with erlotinib also had a significantly better symptom-free survival rate than vehicle-treated mice (Figure 3C, P < 0.05). There was no significant difference in blood pressure between the two groups (Table S3).

Figure 3. Inhibition of EGFR activation (post-treatment) reduced the rate of aneurysm rupture.

Post-treatment of the EGFR inhibitor significantly decreased the aneurysm rupture rate without altering the formation rate (A and B). Compared to mice in the control group, a significantly increased symptom-free survival rate was observed in mice treated with erlotinib (C). Fisher’s exact test was used to analyze the rupture rate of aneurysms (B). Log-rank (Mantel-Cox) test was used for the analysis of the survival rate (C). ∗ P < 0.05.

These results indicate that inhibition of EGFR activation prevents aneurysm rupture but has minimal effect on aneurysm formation.

Inhibition of EGFR reduced gene and protein expression of ER-stress markers and pro-inflammatory cytokines

To further test our hypothesis that EGFR activation contributes to the development of intracranial aneurysm rupture through induction of ER stress, we tested the effect of the EGFR inhibitor, erlotinib, on the gene and protein expression of ER stress markers in mice with induced aneurysms. We measured the levels of mRNA expression of five ER stress markers (GRP78, CHOP, ATF4, sXBP1, and HERPUD) in the cerebral arteries of mice treated with erlotinib or vehicle (Figure 4, top panel). Erlotinib treatment significantly decreased the mRNA expression levels of ER stress markers, namely GRP78, CHOP, sXBP1 and HERPUD, as compared to the vehicle treatment (vehicle vs. erlotinib, GRP78: 1.0 ± 0.48 vs. 0.70 ± 0.39, P < 0.05; CHOP: 1.0 ± 0.38 vs. 0.71 ± 0.19, P < 0.05; sXBP1: 1.0 ± 0.38 vs. 0.71 ± 0.34, P < 0.05; HERPUD: 1.0 ± 0.43 vs. 0.72 ± 0.43, P < 0.05). Additionally, there was a significantly decreased expression level of the oxidative stress marker HIF-1α in mice treated with erlotinib compared to the vehicle control (1.0 ± 0.22 vs. 0.69 ± 0.22, P < 0.05). We did not find a significant difference between erlotinib and vehicle treatment on the expression levels of ATF4, catalase, iNOS, SOD-1, and NF-κB (Figure S1).

Figure 4. Inhibition of EGFR activation decreased mRNA expression of ER stress markers and pro-inflammatory cytokines in cerebral arteries.

EGFR inhibitor- treated mice had significantly reduced mRNA expression of ER stress markers GRP78, CHOP, sXBP1, and HERPUD compared to vehicle-treated controls. Erlotinib-treated mice also had significantly decreased mRNA expression of pro-inflammatory cytokines MMP9, TNF-α, IL-1β, and IL-6, compared to vehicle-treated controls. Data are expressed as mean ± standard deviation, the Mann-Whitney test, ∗ P < 0.05.

In a separate set of experiments, using immunofluorescence staining, we confirmed that EGFR is highly expressed in smooth muscle cells in mice with induced aneurysms (Figure S2). Aneurysm induction caused notable apoptosis of smooth muscle cells in mice treated with vehicle. In contrast, erlotinib significantly reduced the apoptosis of smooth muscle cells (Figure S3). Erlotinib also reduced the infiltration of macrophages (Figure S4). Using immunohistochemical staining, we evaluated the protein expression levels of ER stress markers in the vessels of the circle of Willis from aneurysm-induced mice. Compared to vehicle, erlotinib treatment significantly reduced the protein expression of CHOP (Figure S5) and sXBP1 (Figure S6). In addition, Erlotinib also significantly reduced the protein expression of the proinflammatory proteinase MMP9 (Figure S7).

To determine whether the resultant ER stress from EGFR activation confers inflammation that is known to lead to the rupture of aneurysms, we tested the effect of EGFR inhibition on gene and protein expression of pro-inflammatory cytokines in our mouse model (Figure 4, lower panel). Erlotinib treatment significantly decreased the expression levels of MMP9, TNF-α, IL-1β, and IL-6 as compared to the vehicle treatment (vehicle vs. erlotinib, MMP9: 1.0 ± 0.34 vs. 0.62 ± 0.57, P < 0.05; TNF-α: 1.0 ± 0.47 vs. 0.49 ± 0.51, P < 0.05; IL-1β: 1.0 ± 0.81 vs. 0.61 ± 0.66, P < 0.05; IL-6: 1.0 ± 0.82 vs. 0.44 ± 0.59, P < 0.05).

In a separate set of experiments, we assessed the protein levels of TNFα, IL-1β, and IL-6 in tissue homogenate of the Circle of Willis from mice treated with vehicle or erlotinib using enzyme-linked immunosorbent assay (ELISA). Erlotinib treatment significantly decreased the protein levels of TNF-α, IL-1β, and IL-6 as compared to the vehicle treatment (Figure S8, vehicle vs. erlotinib, TNF-α: 1.0 ± 0.24 (n = 11) vs. 0.65 ± 0.32 (n = 12), P < 0.05; IL-1β: 1.0 ± 0.16 (n = 5) vs. 0.37 ± 0.12 (n = 6), P < 0.05; IL-6: 1.0 ± 0.24 (n = 5) vs. 0.47 ± 0.13 (n = 6), P < 0.05).

Inhibition of ER stress prevented the development of intracranial aneurysm rupture

To establish the direct link between ER stress and aneurysmal rupture, we used a well-established ER stress reducer, 4-phenyl butyric acid (4-PBA).⁴⁵ Following the same protocol for erlotinib pre-treatment, we started the treatment with vehicle or 4-PBA one day before aneurysm induction. The ER stress reducer did not affect the formation of aneurysms, as indicated by the lack of difference in the overall incidence of aneurysms (Figure 5A, vehicle vs. 4-PBA, 94% vs. 87%, n = 17 vs. 23, P = 0.62). However, the ER stress reducer significantly decreased the rupture rate (Figure 5B, vehicle vs. 4-PBA, 88% vs. 50%, n = 16 vs. 20, P < 0.05). Mice treated with 4-PBA also had a significantly better symptom-free survival rate than mice treated with the vehicle (Figure 5C, P < 0.05). There was no significant difference in blood pressure between the two groups (Table S4).

Figure 5. Inhibition of ER stress decreased the rate of aneurysm rupture.

ER stress reduction with 4-PBA significantly decreased the aneurysm rupture rate without altering the formation rate (A and B). A significantly increased symptom-free survival rate was found in mice treated with 4-PBA compared to control mice (C). Fisher’s exact test was used to analyze the rupture rate of aneurysms (B). Log-rank (Mantel-Cox) test was used for the analysis of the survival rate (C). ∗ P < 0.05.

Induction of ER stress nullified the protective effect of EGFR inhibition on aneurysm rupture

Finally, to establish a link between EGFR inhibition and ER stress reduction in the prevention of aneurysm rupture, we treated mice with the EGFR inhibitor erlotinib alone or in combination with tunicamycin, the ER stress inducer. The rationale was that if EGFR activation directly induces ER stress, then treatment of tunicamycin would nullify the protective effects of erlotinib on aneurysm rupture. As expected, there was no significant difference in aneurysm formation rate between erlotinib treatment with or without the addition of tunicamycin (Figure 6A, erlotinib vs. tunicamycin, 92% vs. 93%, n = 13 vs. 14, P = 0.96). Erlotinib reduced the aneurysm rupture rate (in comparison, the rupture rate was 69% for the vehicle control); however, this effect was completely abolished by the tunicamycin treatment (Figure 6B, erlotinib vs. tunicamycin, 45% vs. 92%, n = 11 vs. 13, P < 0.05, Fisher’s exact test). The addition of the ER stress inducer decreased the symptom-free survival rate to a near-significant level compared to the EGFR inhibitor treatment alone (Figure 6C, P = 0.055). There was no significant difference in blood pressure among these groups (Table S5).

Figure 6. The induction of ER stress nullified the protective effect of EGFR inhibition on aneurysm rupture.

The reduced aneurysmal rupture rate afforded by EGFR inhibitor treatment (in comparison, 69% for vehicle control) was completely abolished by ER stress induction using tunicamycin (A and B). The addition of the ER stress inducer decreased the symptom-free survival rate to a near-significant level compared to the erlotinib treatment alone (C). Fisher’s exact test was used to analyze the rupture rate of aneurysms (B). Log-rank (Mantel-Cox) test was used for the analysis of the survival rate (C). ∗ P < 0.05.

Discussion

In this study, using a well-established mouse model, we showed that both inhibition of EGFR activation and ER stress significantly reduced the rupture rate of intracranial aneurysms. Furthermore, the inhibition of EGFR activation reduced the gene and protein expression levels of ER stress markers and pro-inflammatory cytokines. These data suggest that EGFR activation is an upstream event that contributes to aneurysm rupture via the induction of ER stress. We confirmed this notion directly with experiments showing that pharmacological induction of ER stress abolished the protective effect of EGFR on aneurysmal rupture.

Although limited information has been available about the role of ER stress in intracranial aneurysm rupture, previous studies have suggested a potential mechanistic link between ER stress and subsequent arterial wall disruption. Prolonged ER stress can trigger apoptotic cell death, which is mediated by a caspase-12 dependent pathway and/or by transcriptional induction of C/EBP homologous protein (CHOP), and/or by activation of c-Jun N-terminal kinase.^46–48 VSMC apoptosis has been shown to induce medial expansion associated with increased elastic lamina breaks and abnormal matrix deposition in humans.^28,49 VSMC apoptosis, a potential sign and consequence of the maladaptive UPR, has been observed both in human and animal models of intracranial aneurysm.^50–55 Furthermore, ER stress and all three pathways of UPR are responsible for the phenotypic modulation of vascular smooth muscle cells (VSMCs).^13,14 The remodeling of SMCs at the aneurysm wall has been shown to be associated with aneurysm rupture in humans.^56,57 These studies support ER stress and subsequent UPR being directly involved in intracranial aneurysm rupture.

Our data showed that EGFR inhibition via erlotinib not only significantly decreased the rupture rate of the intracranial aneurysm but also decreased gene expression levels of ER stress markers of GRP78, sXBP1, HERPUD, and CHOP. GRP78 is a master regulator of ER stress that modulates downstream UPR pathways. sXBP1 is upregulated by IRE1 arm activation, and HERPUD is upregulated by ATF6 arm activation, respectively.^25,26 CHOP is the molecule at the converging point of PERK/ATF4 and ATF6 UPR pathways and is mainly related to ER stress-induced apoptotic cell death.^26,58 These results indicate that EGFR activation induces ER stress through activating all three UPR pathways. The data on ER stress induction increasing the aneurysm rupture rate further reaffirms that ER stress plays a critical role in the rupture of intracranial aneurysm. The nullification of EGFR’s protective effect on aneurysm rupture by ER stress induction suggests that ER stress is a downstream event of EGFR activation in our mouse model.

In addition to pro-inflammatory cytokines, our data also showed a significant decrease in HIF-1α due to EGFR inhibition. HIF-1α is a marker of oxidative stress. Activation of vascular EGFR produces reactive oxygen species (ROS) through Rac activation, causing oxidative stress.¹⁰ EGFR activation was also shown to produce HIF-1α in VSMCs, and HIF-1α can trigger ER stress and CHOP-mediated apoptosis.¹⁹ These are in agreement with our current findings. Therefore, as a known risk factor,^59–62 oxidative stress might compound with overall EGFR activation and ER stress induction on the aneurysm site, eventually contributing to the rupture of aneurysms.

This study has several limitations. First, the animal model may not completely replicate all biological events that lead to aneurysm rupture, as aneurysms were induced rather than spontaneously formed. Vascular inflammation is known to play a key role in the pathophysiology of intracranial aneurysms in both humans and animals. There may be significant differences in the triggering factors of vascular inflammation between human aneurysms and this model. However, the phenotypes of intracranial aneurysms in the model closely mimic that of intracranial aneurysms in humans.^5,30 More importantly, this model shares the end phenotypes, aneurysmal rupture, and associated neurological symptoms with human aneurysms, indicating its similarity of the underlying biological processes to human intracranial aneurysms.^7,30

Second, we used only male mice in this study, though we have previously examined the sex differences in intracranial aneurysms in our model.^32,63 To fully model human aneurysms, the experimental protocol that utilizes aged female mice, especially reproductively senescent female mice or female mice with long-term estrogen depletion, may be desirable in the future.^64–66 However, at this point, the inclusion of these aged female mice will make the study scope too expansive and too ambitious.

Another limitation of this study is that, although our data support the notion of EGFR activation inducing ER stress in our model, we can only postulate the pathways of ER stress involved based on mRNA expression data. Future studies using ER stress pathway-specific blockers/promoters or transgenic mice may further map out the specific pathways of ER stress responsible for promoting aneurysm rupture.

Perspectives

Our findings suggest that EGFR activation promotes intracranial aneurysm rupture by inducing ER stress in vascular walls. Future clinical studies will need to validate these findings to confirm the relationship between aneurysm rupture and EGFR-ER stress pathways.

Conclusion

This study showed the potential role of EGFR and ER stress in the development of intracranial aneurysm rupture. Pharmacological inhibition of EGFR or downstream ER stress may be a promising strategy for preventing aneurysm rupture and subarachnoid hemorrhage.

Novelty and Relevance.

1.). What Is New?

• We discovered that the activation of the epidermal growth factor receptor (EGFR) promotes intracranial aneurysm rupture in a mouse model of intracranial aneurysm.

• We further confirmed that the effect of EGFR is through inducing endoplasmic reticulum (ER) stress and subsequent inflammatory responses in vascular walls.

2.). What Is Relevant?

• High blood pressure has been shown to play an important role in intracranial aneurysm rupture.

• Currently available therapies for this devastating disease are limited to invasive treatments.

3.). Clinical/Pathophysiological implications?

Our study firmly established that pharmacological inhibition of both EGFR and downstream ER stress was protective against intracranial aneurysm rupture. These pharmacological treatments may be a promising strategy for preventing aneurysm rupture and subarachnoid hemorrhage in a clinical setting.

Nonstandard abbreviations and acronyms

4-PBA

4-Phenylbutyric acid

ATF4

activating transcription factor 4

CHOP

C/EBP homologous protein

EGFR

epidermal growth factor receptor

ELISA

enzyme-linked immunosorbent assay

ER

endoplasmic reticulum

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GRP78

glucose-regulated protein 78

HERPUD1

homocysteine-responsive endoplasmic reticulum-resident ubiquitin-like domain member 1

HIF1

hypoxia-inducible factor-1

IL-1β

interleukin-1 beta

IL-6

interleukin-6

iNOS

inducible nitric oxide synthase

MMP-9

matrix metalloproteinase-9

NF-κB

nuclear factor-kappa B

PBS

phosphate-buffered saline

ROS

reactive oxygen species

RT-PCR

R eal-time polymerase chain reaction

TNF-α

tumor necrosis factor-alpha

UPR

unfolded protein response

VSMCs

vascular smooth muscle cells

sXBP1

spliced X-Box binding protein 1